Mihaela Caprioara Doktorarbeit 2007

Mihaela Caprioara Doktorarbeit 2007
Inaugural-Dissertation
zur Erlangung der Doktorwürde
der
Naturwissenschaftlich-Mathematischen Gesamtfakultät
der
Ruprecht-Karls Universität
Heidelberg
vorgelegt von
M. Sc. - Dipl. Chem. Mihaela Caprioara
aus Piatra Neamt/Rumänien
Tag der mündlichen Prüfung: 26.11.2007
DNA-Based Ligands for Use in Asymmetric Catalysis and
Development of Metallo-(deoxy)Ribozymes
Gutachter:
Prof. Dr. Andres Jäschke
Prof. Dr. Nils Metzler-Nolte
Acknowledgments
I thank all my friends and colleagues from the University of Heidelberg and from other
research groups who contributed to the work described in this thesis.
I express my sincere gratitude to my supervisor, Prof. Andres Jäschke, for giving me the
opportunity to work in a very stimulating research group, on a highly interesting project,
as well as for his continuous support, encouragement, and confidence at all stages of my
work. Thank you for all suggestions that helped me to improve the quality of this thesis.
I would like to thank Prof. Nils Metzler-Nolte who agreed to review and co-examine
this thesis.
I want to acknowledge Dr. Roberto Fiammengo and express my recognition for his
guidance, expertise, determination, and understanding, which considerably added to my
experience throughout all these years. He provided me with direction, stimulated me in
being critical with my results, and had always prompt answers to my questions. I
appreciate his work and patience to read the preliminary versions of my writings,
including this thesis, and the assistance he provided through the most difficult
correction steps.
I would like to acknowledge Prof. Nils Metzler-Nolte and Dr. Srecko Kirin who kindly
provided the N,N-bis(2-picolyl)amine derivative, as well as Prof. Lutz Gade for
generously providing the PYRPHOS ligand. I also thank Prof. Roland Krämer and Dr.
Andriy Mokhir for helpful discussion and technical support in performing the MALDI
TOF measurements. I am grateful to Dr. Marianne Engeser, University of Bonn, for our
fruitful collaboration, for recording the electrospray mass spectra of the Rh(I)PYRPHOS complex and the DNA-PYRPHOS conjugate, as well as for her help with
the interpretation of the data.
I thank Dr. Mark Helm, Pierre Fournier, Stephanie Pfander and Anna Wiesmayr for
generously offering to proof-read sections of my thesis. I appreciate their support, their
rigorous and constructive corrections. I am grateful to Stephanie Pfander and Alexander
Nierth for the linguistic improvements of the “Zusammenfassung” of my thesis.
I want to acknowledge Pierre Fournier with whom I shared the work in the lab, for the
many helpful discussions about catalysis and for his enthusiasm in the times when
difficult tasks came up.
Special thanks go to Dr. Richard Wombacher, Columbia University, for his continuous
interest in my project and for our scientific debates, knowledge exchange, and venting
of frustration during our breaks. I also thank Dr. Barbara-Sylvia Weigand, Stephanie
Pfander, Markus Petermeier for being supportive and understanding, for their friendship
and encouragement over the last four years. I am grateful to all former and current
colleagues for the stimulating working environment and for their contributions to the
lab research standards.
This work would have been incomplete without the help of Sandra Suhm, Heiko Rudy,
Tobias Timmerman, Besarta Nezaj, Marina Silbereis. I want to thank them for the great
technical assistance in synthesis, mass spectrometry and NMR measurements, as well as
for their excellent job to fulfil the needs of the lab.
My special thanks go to the office staff, in particular to Mrs. Viola Funk and Mrs. Karin
Weiß, for all the instances in which their assistance helped me along the way and for all
their support and advices at times of critical need.
Thanks to my family for their sincere encouragement and support they provided me
through my entire life. In particular, I would like to thank my fiancé and my best friend,
Razvan, for his patient support, encouragement, and for helping me finish this thesis.
This work was financially supported by Deutsche Forschungsgemeinschaft (SFB 623).
To my family
Summary
Caprioara, Mihaela; M. Sc. Chem.
Title: DNA-Based Ligands for Use in Asymmetric Catalysis and Development of
Metallo-(deoxy)Ribozymes
1. Gutachter: Prof. Dr. Andres Jäschke
2. Gutachter: Prof. Dr. Nils Metzler-Nolte
The fascinating way nature relies on biomolecules, mostly proteins and sometimes RNA, to
carry out sophisticated chemical processes led to more and more efforts to use the concepts of
biology for preparing efficient chiral catalysts. The “hybrid catalyst” approach that combines
the steric information derived from a protein scaffold with the catalytic activity of transition
metal complexes offers a resourceful means of developing semisynthetic metalloenzymes for
enantioselective applications. Since the discovery of nucleic acids with enzyme-like functions,
the catalytic potential of nucleic acids is being revealed by in vitro selection and evolution of
novel ribozymes and DNAzymes. Nucleic acids, especially RNA, appear to be versatile
catalysts capable of accelerating a broad range of reactions and exquisitely discriminating
between chiral targets. However, while proteins dominated the construction of hybrid catalysts,
the application of DNA and RNA in asymmetric catalysis has hardly been explored.
This work aimed at exploring the chirality of nucleic acids and generating hybrid catalysts
based on DNA and RNA. Towards the development of metallo-(deoxy)ribozymes assisted by
combinatorial strategies (e.g., SELEX), a straightforward synthetic way of embedding transition
metal complexes in nucleic acids folds was established. DNA sequences carrying mono- and
bidentate phosphine ligands as well as P,N-ligands were successfully prepared starting from
amino-modified oligonucleotide precursors. The optimized “convertible nucleoside” approach
allowed the parallel, high-yielding synthesis of various alkylamino-DNA conjugates differing in
length and structure of the spacer. Coupling of amino-oligonucleotides with PYRPHOS, BINAP
and PHOX ligands equipped with a carboxyl group led to the incorporation of phosphine
moieties at predetermined internal sites. Moreover, the stability of the DNA-tethered BINAP
and PHOX was reasonably high, which makes them attractive candidates for the development
of transition metal-containing oligonucleotides. To this end, systematic studies on the behavior
of phosphine- and PHOX-metal complexes in aqueous medium - a prerequisite of nucleic acid
catalysts - were carried out. Two model organometallic transformations were selected that were
compatible with the structure and chemistry of nucleic acids. The rhodium(I)-catalyzed 1,4addition of phenyl boronic acid to 2-cyclohexen-1-one and iridium(I)-catalyzed allylic
amination of the branched phenyl allyl acetate, respectively, proceeded efficiently in the
presence of phosphorus-based ligands, in aqueous medium, at room temperature and low
catalyst concentration. For the first model reaction, the best conversion (80%) was achieved
with the isolated [Rh(nbd)BINAP]BF4 complex, in 6:1 dioxane/water, and TEA additive. On the
basis of these data, a suitable system for assessing the catalytic potential of the DNA-BINAP
ligand was implemented. In the second chosen reaction the in situ formed Ir(I)-PHOX
complexes (0.05-0.1 mM) gave rise to racemic, branched allylic amination products in good
yields (33-75%), in 3:7 dioxane/water. Kinetic resolution of the racemic substrate was then
attempted by employing catalysts generated from the [Ir(cod)Cl]2 precursor and single- and
double-stranded DNA-PHOX conjugates. Good conversions were obtained in the presence of
G-poor DNA/DNA and RNA/DNA hybrids bearing the PHOX moiety, indicating a potential
role of the G-N7 site in the first coordination sphere. With all tested DNA-PHOX conjugates,
the levels of enantioselectivity remained modest. The results described in this work provide
useful information for understanding the influence of nucleic acid sequence and covalent
tethering on the reaction outcome. These are the first reported applications of DNA-based
ligands in organometallic catalysis and they build the fundamentals for further development of
selective nucleic acid catalysts, by means of rational design and in vitro selection approaches.
Zusammenfassung
Die faszinierende Art und Weise, in der die Natur auf Biomoleküle - meist Proteine und
teilweise RNA - zurückgreift um anspruchsvolle chemische Prozesse auszuführen, hat zu immer
mehr Bemühungen geführt, die Konzepte der Biologie für die Herstellung effizienter chiraler
Katalysatoren zu nutzen. Die Verbindung dreidimensionaler Proteinstrukturen mit der
katalytischen Aktivität von Übergangsmetallkomplexen ist eine interessante Herangehensweise
in der Synthese sogenannter Hybrid-Katalysatoren für enantioselektiven Anwendungen. Seit der
Entdeckung von Nukleinsäuren mit enzym-ähnlichen Funktionen wurde deren katalytisches
Potential durch in vitro Selektion und Evolution neuer Ribozyme und DNAzyme deutlich
gezeigt. Nukleinsäuren, insbesondere RNA, sind demnach vielseitige Katalysatoren, die in der
Lage sind eine breite Palette an Reaktionen zu beschleunigen und außerordentlich gut zwischen
chiralen Zielmolekülen zu unterscheiden. Während Proteine die Entwicklung von HybridKatalysatoren jedoch weitgehend beherrschen, wurde die Anwendung von DNA oder RNA in
der asymmetrischen Katalyse bisher kaum untersucht.
Das Ziel dieser Arbeit war die Synthese von Hybrid-Katalysatoren auf Basis eines DNA und
RNA Gerüstes. Für die kombinatorisch-gestützte Entwicklung von Metallo-Ribozymen und Deoxyribozymen (z. B. mittels SELEX) wurde ein direkter Syntheseweg zum Einbau von
Übergangsmetall-Komplexen in Nukleinsäurestrukturen etabliert. DNA Sequenzen welche einund zweizähnige Phosphin-Liganden sowie P,N-Liganden tragen, wurden ausgehend von
amino-modifizierten Oligonukleotid Vorstufen erfolgreich synthetisiert. Der optimierte Ansatz
mittels sogenannter „convertable nucleosides“ erlaubt die parallele Synthese verschiedener
alkylamino-DNA Konjugate, welche sich in Länge und Struktur der Spacer unterscheiden. Die
Kopplung der Amino-Oligonukleotide mit PYRPHOS-, BINAP- und PHOX-Liganden, welche
mit einer Carboxylgruppe ausgestattet sind, führt zum Einbau der Phosphin Bausteine an einer
zuvor festgelegten Stelle im Nukleotidstrang. Ferner macht die hohe Stabilität der DNAgebundenen BINAP und PHOX Liganden diese zu attraktiven Kandidaten für die Entwicklung
von Oligonukleotiden, die Übergangsmetall-Komplexe enthalten. Zu diesem Zweck wurden
systematische Studien zum Verhalten von Phosphin- und PHOX-Metallkomplexen im
wässrigen Medium durchgeführt - eine Voraussetzung für Katalysatoren auf Nukleinsäurebasis.
Zwei metallorganische Transformationen, die mit der Struktur und den chemischen
Eigenschaften von Nukleinsäuren kompatibel sind, wurden als Modellreaktionen ausgewählt.
Die Rhodium(I)-katalysierte 1,4-Addition von Phenylborsäure zu 2-Cyclohexen-1-on und die
Iridium(I)-katalysierte allylische Aminierung des verzweigten Phenylallylacetats verliefen
erfolgreich in Anwesenheit von phosphorbasierten Liganden, in wässrigem Medium,
Raumtemperatur und niedriger Katalysatorkonzentration. Für die erste Modellreaktion wurde
die beste Umsetzung mit dem isolierten [Rh(nbd)BINAP]BF4 Komplex in 6:1 Dioxan/Wasser
und unter TEA Zugabe erzielt (80 %). Auf Grundlage dieser Daten wurde ein geeignetes
System erstellt, um das katalytische Potential von DNA-BINAP Liganden zu beurteilen. Bei der
zweiten Modellreaktion führte der in situ gebildete Ir(I)-PHOX Komplex (0.05-0.1 mM) in 3:7
Dioxan/Wasser zu guten Ausbeuten (33-75 %) der racemischen, verzweigten
Aminierungsprodukte. Bei der kinetischen Auflösung racemischer Substrate wurden
Katalysatoren verwendet, die aus dem [Ir(cod)Cl]2-Vorstufe und einfach- und doppelsträngigen
DNA-PHOX Konjugaten hergestellt wurden. Gute Umsätze wurden in Anwesenheit von Garmen DNA/DNA und RNA/DNA Hybriden, die eine PHOX Gruppe tragen, erzielt, was auf
eine mögliche Funktion der G-N7 Position in der ersten Koordinationsspähre hindeutet. Die
Enatioselektivität blieb jedoch bei allen getesteten DNA-PHOX Konjugaten gering. Die
Ergebnisse dieser Arbeit bieten hilfreiche Informationen für das Verständnis darüber, welchen
Einfluss die Nukleinsäuresequenz und die kovalente Verknüpfung auf den Ausgang der
Reaktion haben. Dies ist der erste Bericht über die Anwendungen von DNA-basierten Liganden
in der metallorganischen Katalyse und setzt den Grundstein für die weitere Entwicklung
selektiver Nukleinsäurekatalysatoren mittels der Methodik des rationalen Designs und der in
vitro Selektion.
INDEX
1
INTRODUCTION .............................................................................. 17
1.1
SYNTHESIS OF CHIRAL ORGANIC MOLECULES ........................................................................ 17
1.1.1
Asymmetric Catalysis with Phosphorus Ligands ............................................................... 19
1.1.1.1
PYRPHOS Ligands................................................................................................................. 19
1.1.1.2
BINAP Ligands....................................................................................................................... 21
1.1.1.3
PHOX Ligands........................................................................................................................ 24
1.1.1.4
Phosphite and Phosphoramidite Ligands ................................................................................ 26
1.1.2
Biocatalysis ........................................................................................................................ 27
1.1.2.1
Artificial Enantioselective Enzymes ....................................................................................... 28
1.1.2.2
Hybrid Catalysts...................................................................................................................... 30
1.2
NUCLEIC ACID ENZYMES ........................................................................................................ 33
1.2.1
Chirality in the Structure of Nucleic Acids ........................................................................ 33
1.2.2
In vitro Selection of Nucleic Acid-Enzymes ....................................................................... 38
1.2.3
DNA-zymes......................................................................................................................... 40
1.2.4
Ribozymes .......................................................................................................................... 41
1.2.5
DNA-Based Hybrid Catalysts ............................................................................................ 43
1.3
TRANSITION METAL-DNA CONJUGATES ................................................................................ 46
1.3.1
Non-covalent Interactions.................................................................................................. 46
1.3.2
Covalent Attachment.......................................................................................................... 47
1.3.2.1
Post-synthetic Functionalization ............................................................................................. 47
1.3.2.2
Automated Solid-phase Synthesis........................................................................................... 49
2
OBJECTIVES..................................................................................... 53
3
RESULTS AND DISCUSSIONS ...................................................... 55
3.1
INCORPORATION OF METAL COMPLEXES INTO NUCLEIC ACIDS .............................................. 55
3.1.1
Functionalization of DNA with Phosphoramidite Ligand Precursors ............................... 58
3.1.2
DNA-Phosphine Ligands.................................................................................................... 62
3.1.2.1
Amino-Functionalized Oligonucleotides ................................................................................ 62
3.1.2.2
Duplex Stability of Amino-Tethered Oligonucleotides........................................................... 65
3.1.2.3
Reactivity of Amino-Modified Oligonucleotides.................................................................... 68
3.1.2.4
Post-synthetic Functionalization of Amino-modified DNA with Phosphine Ligands............. 69
3.1.2.5
Duplex Stability of Bisphosphine-Tethered DNA................................................................... 76
3.2
ORGANOMETALLIC TRANSFORMATIONS IN WATER................................................................. 78
3.2.1
Phosphine- and Phosphinooxazoline-Metal Complexes .................................................... 79
3.2.2
Rhodium(I)-Catalyzed 1,4-Addition................................................................................... 87
3.2.3
Iridium(I)-Catalyzed Allylic Amination ............................................................................. 92
3.2.3.1
Preparation of Allylic Substrates and Products. Analytical Methods...................................... 95
3.2.3.2
Preliminary Results of Catalysis with Ir-PHOX Complexes................................................... 98
3.2.3.3
Allylic Amination with DNA-Appended Phosphinooxazoline Ligands................................ 104
3.2.3.4
Allylic Amination with Double-stranded DNA-appended Phosphinooxazoline Ligands ..... 110
4
CONCLUSIONS AND OUTLOOK ............................................... 117
5
MATERIALS AND METHODS .................................................... 127
5.1
STANDARD METHODS AND REAGENTS .................................................................................. 127
5.2
SYNTHESIS OF PHOSPHORUS LIGANDS AND THEIR TRANSITION METAL COMPLEXES ............ 128
5.2.1
Synthesis of Phosphoramidite Ligands ............................................................................ 128
5.2.2
Synthesis of PHOX Ligands ............................................................................................. 130
5.2.3
Synthesis of Palladium(II)- and Platinum(II)-Phosphine Complexes .............................. 133
5.2.4
Synthesis of Platinum(II)-, Palladium(II)- and Rhodium(I)-PYRPHOS Complexes ........ 134
5.2.5
Synthesis of Rhodium(I)- and Iridium(I)-PHOX Complexes............................................ 135
5.3
OLIGONUCLEOTIDES .............................................................................................................. 137
5.3.1
Automated Solid-Phase Synthesis .................................................................................... 138
5.3.2
General Procedure for the Synthesis of Amino-Modified ODNs ..................................... 140
5.3.3
General Procedure for the Synthesis of 7-deaza-G -containing Amino-Modified ODNs .141
5.3.4
General Procedure for the Synthesis of Complementary DNA ........................................ 143
5.4
SYNTHESIS OF DNA-BASED LIGANDS ................................................................................... 144
5.4.1
Incorporation of Phosphite Moiety by DNA Solid-phase Synthesis ................................. 144
5.4.2
Synthesis of DNA-Phosphine Conjugates ........................................................................ 146
5.4.3
Synthesis of DNA-Appended N,N-bis(2-picolyl)amine Conjugates.................................. 148
5.5
HIGH-PRESSURE LIQUID CHROMATOGRAPHY ....................................................................... 150
5.5.1
Reversed-Phase HPLC Purification of Oligodeoxynucleotides ....................................... 150
5.6
MASS SPECTROMETRY ANALYSIS OF OLIGONUCLEOTIDES ................................................... 152
5.7
5’-RADIOACTIVE LABELLING OF OLIGONUCLEOTIDES .......................................................... 153
5.8
ANALYSIS AND QUANTIFICATION OF DNA............................................................................ 154
5.8.1
Quantification of Oligonucleotides by UV Absorbance ................................................... 154
5.8.2
Analysis of DNA Duplexes by Thermal Melting Curves .................................................. 155
5.8.3
Polyacrylamide Gel Electrophoresis (PAGE).................................................................. 156
5.9
TRANSITION METAL-CATALYZED REACTIONS ...................................................................... 157
5.9.1
Conjugate Addition .......................................................................................................... 157
5.9.1.1
Synthesis of 3-Phenyl-1-cyclohexanone ............................................................................... 157
5.9.1.2
General Procedure for 1,4-Addition of Phenylboronic Acid to 2-Cyclohexen-1-one ........... 158
5.9.2
6
7
Allylic Amination ............................................................................................................. 161
5.9.2.1
Synthesis and Stability of the Allylic Substrate .................................................................... 161
5.9.2.2
Synthesis of Linear and Branched Allylic Amines ............................................................... 162
5.9.2.3
Allylic Amination Catalyzed by Ir-PHOX Complexes ......................................................... 164
APPENDICES .................................................................................. 169
6.1
LIST OF ABBREVIATIONS ....................................................................................................... 169
6.2
INSTRUMENTS AND SPECIAL MATERIALS .............................................................................. 172
REFERENCES ................................................................................. 175
1
Introduction
1.1
Synthesis of Chiral Organic Molecules
“How would you like to live in a looking-glass house, Kitty? ... Perhaps looking-glass
milk isn’t good to drink.” (Lewis Caroll, Through the Looking-Glass)
„Chirality ... is an intrinsic universal feature of various levels of matter.“[1]
Living world depends on molecular chirality, in that crucial biopolymers associated
with life are made up of chiral monomers (L-amino acids in proteins, D-sugars in RNA
and DNA).[2, 3] Chirality was first described in 1848 by Louis Pasteur who demonstrated
that tartrate enantiomers rotated the plane of the polarized light in different ways, and
only the right-handed enantiomer was present in wine lees.
In living organisms, all chemical transformations, recognition or information processing
involve chiral molecules, such as enzymes or hormones.[4] Most physiological processes
are based on precise molecular interactions between the chiral host molecules and the
two enantiomeric guest molecules. Despite the structural similarity, two enantiomers
behave and are recognized differently in a chiral environment. Biological chiral
receptors, for example, interact mostly with drug molecules having the proper absolute
configuration, leading to distinct biological activities of the two enantiomers. The
importance of the relationship between pharmacological activity and chirality was
demonstrated in the early 1960s, by the tragic administration to pregnant women of
thalidomide (Figure 1.1), in the racemate form. The R enantiomer has the desired
sedative
properties,
while
(S)-thalidomide
is
teratogenic
and
led
to
fetal
malformations.[1] However, even in 2000, only 40% of synthetic chiral drugs were sold
in single enantiomer dosage form.[5]
1.1. Synthesis of Chiral Organic Molecules
18
Figure 1.1. Thalidomide enantiomers and their biological properties.
The fascinating way nature is producing substances with chirality and the preference of
one enantiomer over the other in the living matter have attracted considerable interest
for developing synthetic routes to enantiomerically pure compounds, with the major
goal of gaining significant clinical, scientific and industrial benefits. Three main
strategies have been established: (i) enantiomer separation; (ii) transformation of a
chiral precursor and (iii) enantioselective reactions. Enantiomerically pure substances
were earlier obtained by the classical resolution of a racemate or chemical conversion of
naturally occurring chiral compounds, such as tartaric and lactic acids, amino acids,
carbohydrates, terpens, or alkaloids. Nevertheless, the use of enantioselective catalysts
became over years the most efficient and attractive approach.
For many years, the practical access to pure enantiomers relied on biochemical or
biological methods. Nature utilizes enzymes for this purpose and relies on
configurational and conformational optimization by structural variation of the chiral
building blocks. The complex biological structures nature offers us have often modest
applicability in chemical synthesis due to their limited substrate scope and operational
stability.[6] However, with the help of protein engineering techniques, such as mutation,
selection, and directed evolution, enzymes have been successfully improved and
biocatalysts with novel properties developed.[7]
In parallel to this field, metal-catalyzed enantioselective transformations have received
much attention. Transition metal catalysts possess great potential in synthesis of
enantiopure compounds from achiral/prochiral precursors, namely through an
asymmetric reaction[8], or from racemic mixtures. Such metal catalysts usually consist
1.1. Synthesis of Chiral Organic Molecules
19
of a metal centre and a ligand carrying the stereochemical information in order to ensure
the catalysis to proceed in a stereoselective manner.[9] In 2001, Noyori, Sharpless and
Knowles were rewarded with the Nobel Prize for their achievements in the field of
asymmetric organometallic catalysis.[1, 10, 11]
1.1.1
Asymmetric Catalysis with Phosphorus Ligands
The search for valuable asymmetric catalytic systems is one of the most active research
areas in organic and bioorganic chemistry. A wide range of man-made catalysts has
been constructed over the past four decades, and is mainly based on chiral metal
complexes. The chiral information in the products prepared by enatioselective catalysis
derives from the optically active ligands bound to the transition metal. Therefore, the
proper design of the chiral ligands is the most important requirement for achieving high
efficiency.[12, 13] Much effort has been dedicated to the synthesis of new chiral ligands or
tuning of existing ligands to obtain high selectivity and reactivity.
After
the
discovery
of
Wilkinson’s
homogeneous
hydrogenation
catalyst
[14]
[RhCl(PPh3)3]
, phosphorus ligands have attracted considerable interest. A second
development in the mid-1960s was made by Knowles and Horner who replaced the
triphenylphosphine of the Wilkinson’s catalysts with chiral monophosphines. These
ligands were used in hydrogenation reactions, albeit with poor enantioselectivities. Few
years later, monophosphines were successfully replaced due to the rapid development of
chiral bisphosphorus ligands. Some of the most relevant phosphorus-based ligands will
be reviewed here.
1.1.1.1 PYRPHOS Ligands
A large number of chelating bisphosphine ligands bearing a chiral carbon backbone has
been synthesized and mainly employed in Rh-catalyzed asymmetric hydrogenation of
dehydroamino acids. The discovery of 1,2-bisphosphine PYRPHOS-type (3,4-bisdiphenylphosphino-pyrrolidine) ligands[15] enlarged the spectrum of substrates that
could be processed with chiral bisphosphines in Rh-catalyzed hydrogenations (e.g. N(acylamino)acrylates, enamides, enol acylates, and itaconic acids.[16-18] A few selected
examples of such ligands are illustrated in Figure 1.2.
1.1. Synthesis of Chiral Organic Molecules
20
Figure 1.2. PYRPHOS ligands.
The majority of bisphosphine ligands consisting of a chiral carbon-backbone contain
two aryl substituents attached to each phosphorus atom, which exert strong steric and
electronic influences in their transition metal complexes. The orientation of the phenyl
substituents has been reported to control the stereoselectivity and, therefore, the
sterically congested phosphines possess remarkable catalytic properties. For example,
PYRPHOS ligands, which contain a symmetrically placed nitrogen atom, generate rigid
five-membered chelate rings with transition metals beneficial for the optimal transfer of
chirality to the reaction centre. With Rh-PYRPHOS as catalyst, hydrogenation of N(acetylamino)cinnamic acid to (S)-N-acetyl-phenylalanine proceeded with >99% yield
[15, 19]
and over 95% ee.
An important advantage of these ligands is that they can be
readily tuned by variations of the N-substituent. Moreover, they can be easily attached
to a linker through an amide bond and therefore facilitate the further immobilization of
their transition metal complexes on polyethylene glycol supports,[20] silica gel,[15] gold
surfaces,[21] or at the end of dendrimers.[18,
22]
In general, the stereoselectivity of the
immobilized systems was comparable to those obtained using analogue homogeneous
transition metal catalysts. Interesting examples of PYRPHOS application have been
reported by Gade and co-workers in early 2000s. Upon fixation of the ligand in the
densely packed environment of a dendrimer, the resulting polynuclear complexes
induced enantioselectivities of 74-93% in Rh(I)-catalyzed hydrogenation of Z-methylN-(acetylamino)cinnamate and dimethyl itaconate,[18] and 69% in Pd(0)-catalyzed
allylic amination of 1,3-diphenyl-1-acetoxypropene,[22] respectively.
1.1. Synthesis of Chiral Organic Molecules
21
1.1.1.2 BINAP Ligands
Numerous studies have generally demonstrated that ligands with C2 symmetry elements
perform excellent stereochemical control,[12] most likely due the reduced number of
conformations the ligand can adopt in the coordination sphere of the metal. Noyori’s
pioneering work on atropisomeric C2-symmetric bisphosphine ligand BINAP (2,2'bis(diphenylphosphino)-1,1'-binaphthyl)[23] (Figure 1.3) and its successful applications
in Rh-catalyzed hydrogenation of α-(acylamino)acrylic acids and esters[24] (Table 1.1,
entry 1) opened the way to a new class of fully aromatic chiral phosphorus ligands. The
BINAP skeleton is conformationally flexible and can accommodate a large variety of
transition metals, generating seven-membered rings that contain only sp²-hybridized
carbon atoms. The rotational freedom around the donor atom-metal bond in the
resulting chelate structures is therefore restricted. This feature, responsible for the
chirality transfer to the metal coordination sites involved in the catalytic transformation,
explains the high chiral recognition ability of BINAP ligands in numerous catalytic
reactions.
Figure 1.3. BINAP ligands.
Transition metal-BINAP complexes have been extensively employed to reduce
prochiral C=O, C=N, or C=C bonds. Enantioselective hydrogenation of olefins with
chiral rhodium or ruthenium catalysts is one of the most established methods in
asymmetric
catalysis. Examples
of
substrates
processed
with
BINAP-based
hydrogenation catalysts include functionalised olefins and ketones (e.g. αdehydroamino acid derivatives,[24, 25] (β-acylamino)acrylates,[26] cyclic enamides,[27, 28]
α,β-unsaturated acids,[29] allylic alcohols,[30] itaconic acids, α or β-keto esters,[31, 32] and
imines.[33]
1.1. Synthesis of Chiral Organic Molecules
22
Sato’s preliminary work on Pd-catalyzed Heck reactions with BINAP ligands,[34]
followed by several studies on substrates variation,[35-38] clearly demonstrated the
potential of such chiral bisphosphines in enantioselective C-C bond forming processes.
Although BINAP commonly induced modest regio- and stereoselectivity in asymmetric
Heck reactions,[39] its high catalytic performance has been confirmed in Rh-catalyzed
conjugate 1,4-additions.
Table 1.1. Rh(I)-BINAP catalyzed asymmetric addition
Entry
Olefin
[40]
1
O
Ar
% ee
PhB(OH)2
93-97
PhB(OH)2
92-97
ArB(OH)2
87-95
n
n = 0, 1, 2
[40]
2
O
R1
R1 = i-Pr, Am
[41, 42]
3
O
R1
O
R2
Ar = Ph, 3-MeOC6H4
R1 = Me, n-Pr
R2 = Me, Et, t-Bu, CH2Ph
4[43]
P(OR2)2
O
(ArBO)3
91-96
Ar = C6H5, 4-CH3C6H5, 3-ClC6H5
R2 = Me, Et, Ph
In 1998, Miyaura and Hayashi described the conjugate addition of arylboronic acids to
α,β-unsaturated ketones in the presence of a Rh-phosphine complex (Table 1.2).[44]
Under optimized reaction conditions, high yields and stereoselectivities were achieved
for a wide range of substrates using BINAP as chiral ligand.[40, 45] The stereochemical
outcome of the enantioselective additions using BINAP is dictated by the formation of a
chiral pocket containing the vacant coordination site in the highly distorted structure of
the Rh complex.[46] For this reason, BINAP ligands remain the most efficient
bisphosphines examined in Rh-catalyzed conjugate addition,[41] affording high yields
and over 90% enantioselectivity with various types of cyclic or linear enones and
1.1. Synthesis of Chiral Organic Molecules
23
organoboronic acids.[47] Several examples of electron-deficient olefins, including α,βunsaturated ketones or esters, and phosphonates, as well as organoboron/boroxine
reagents employed in Rh-BINAP catalyzed asymmetric 1,4-additions, are summarized
in Table 1.1.
A common aspect of Rh-catalyzed 1,4-additions is that they are usually accelerated by
the presence of water as cosolvent. Therefore, the development of water-soluble
organometallic catalysts has attracted considerable attention. A typical example is
introduction of water-soluble functional groups onto BINAP. This strategy has been
used to generate BINAP analogues (i.e., diguanidinium-BINAP,[48] the bromohydrate
form of the 6,6’-dimethylamino-BINAP[49]) for Rh- or Ru-mediated hydrogenation
reactions, while aqueous 1,4-addition employs a resin-bound Rh-BINAP variant. The
connection of BINAP with the amphiphilic polymer having an amino group was done
via amide bond formation, using a BINAP-carboxylic acid derivative (Figure 1.3).[50]
The resulting polymer-supported BINAP-Rh catalyst afforded high yield (71-95%) and
high enantioselectivity (91-97%) in water for the 1,4-addition of phenylboronic acid to
α,β-unsaturated ketones.
An interesting family of heterobidentate chiral P,S-ligands, BINAP(S) (Figure 1.3),
generated by conversion of BINAP to mono-sulfides, was reported in the early
2000s.[51] With such ligand systems, highly selective Pd-catalyzed racemate
resolution,[52,
53]
as well as asymmetric amination of allylic substrates was obtained
(Figure 1.4).[53, 54] The observed regioselectivity in case of allylic substrates containing
rather small substituents was in contrast to the traditional regioselectivity of Pd systems,
and branched products were predominantly formed. These intriguing results, together
with the induced high regio- and stereoselectivities, made BINAP(S) a ligand of choice
for further development in the field of allylic substitution reactions, where usually chiral
P,N-ligands are used.
Figure 1.4. Pd-catalyzed allylic substitution with BINAP(S) ligand.
1.1. Synthesis of Chiral Organic Molecules
24
1.1.1.3 PHOX Ligands
A powerful class of chiral bidentate P,N-ligands,[55] namely phosphinooxazolines
(PHOX) (Figure 1.5), structurally similar to Crabtree’s catalysts combines a hard, N
(oxazoline) with a soft, P (phosphine) donor and has been introduced by Helmchen in
1993.[56]
PHOX chirality derives from an asymmetric centre placed in the oxazoline ring directly
connected to an aromatic phosphine moiety. Upon transition metal coordination, PHOX
ligands form six-membered chelates that bind the substrate in a stereoselective manner.
It has been assumed that such nonsymmetrical ligands would allow more enantiocontrol
than C2-symmetric ligands[57] that dominated for long time asymmetric catalysis.
Chiral PHOX ligands were initially used in Pd-catalyzed allylic alkylation of
symmetrically substituted allyl acetates,[56] yielding complete substrate conversion and
high enantioselectivity. Additionally, they induced excellent enantioselectivities in Ircatalyzed asymmetric hydrogenation of unfunctionalized olefins[58-60] for which
classical Rh or Ru-BINAP catalysts were not efficient. Promising results were also
obtained with olefins lacking a polar group,[61] and imines,[62] as substrates for
hydrogenation.
Remarkable contributions from the groups of Helmchen and Pfaltz demonstrated the
great catalytic properties of PHOX ligands in transition metal-catalyzed C-C and C-N
bond formation, including allylic substitutions and Heck reactions. Due to their modular
structure, a number of very effective PHOX ligands were readily generated by varying
the substituent at the stereogenic centre of the oxazoline moiety, responsible for catalyst
reactivity.[63, 64] Commonly, high conversions were obtained with bulky groups located
next to the metal centre. PHOX ligands with W, Mo, Pd, and Ir precursors have been
tested for various allyl systems, either symmetrically substituted,[56,
65,
66]
or
monosubstituted linear substrates[67, 68] in reaction with nucleophiles, such as dimethyl
malonate,[57] amines,[65] nitro compounds,[69] and sulfinates.[70] High enantiomeric
excess and metal-dependent regioselectivity were mainly achieved with arylallyl
derivatives.[57] Furthermore, interesting results were obtained in intramolecular
aminations,[71] in spite of long reaction times generally required in PHOX-Ir catalyzed
1.1. Synthesis of Chiral Organic Molecules
25
allylic substitutions with amines.[72] So far, conversion of racemic monosubstituted allyl
compounds to a single enantiomer through kinetic resolution in the presence of chiral
PHOX ligands has not been reported.
Figure 1.5. (A) Enantioselective hydrogenation in the presence of PHOX-Ir complexes. (B) Transition
metal-catalyzed allylic alkylations and Heck reactions using PHOX ligands.
1.1. Synthesis of Chiral Organic Molecules
26
Although impressive enantioselectivities have been also observed in Pd-catalyzed Heck
reactions, the use of PHOX ligands in such transformations appeared to be limited due
to their low reactivity.[63] Figure 1.5 illustrates several applications of PHOX complexes
in asymmetric catalysis.
1.1.1.4 Phosphite and Phosphoramidite Ligands
Traditionally, chelating phosphine ligands with a -PPh2 moiety, appeared to be ideal in
many asymmetric transformations, but P-O and P-N containing monodentate
phosphorus ligands, such as monophosphites and phosphoramidites, were often found to
be as effective. This type of ligands benefits of straightforward preparation, modular
construction for tuning of properties, high π-acidity, and, in contrast to phosphine
analogues, resistance to oxidation.[73, 74] Although the relative instability, especially in
protic solvents, noticeably limited their utility, within the last decade several new
classes of phosphite and phosphoramidite ligands of high stability have been developed.
Such ligands are capable of rate acceleration and stereoselectivity in a number of
essential organic transformations, such as allylic substitution, hydrogenation and
conjugate addition.[74]
Some of the most simple and, at the same time, selective monophosphite ligands are the
biaryl-derived species (Figure 1.6). In combination with an iridium catalyst precursor,
triphenylphosphites were found to induce high rate acceleration and regioselectivity in
favour of the branched product for both aryl and alkyl substrates in allylic substitution
reactions.[75-77] Chiral biaryl-derived phosphites gave excellent results in Ir-catalyzed
transformations of unsymmetrical allylic substrates.[78, 79] Structurally related phosphitephosphine combinations have been successfully applied to Pd-catalyzed allylic
transformations of symmetrically disubstituted substrates.[80]
Figure 1.6. BINOL and biphenyl-derived monophosphite ligands.
1.1. Synthesis of Chiral Organic Molecules
27
The use of BINOL-derived structures appeared to be important for achieving high
enantioselectivity. For example, chiral BINOL-derived monophosphites afforded high
levels of enantioselectivity in Rh-catalyzed hydrogenation of itaconic acid.[81, 82]
Furthermore, P-O/N bond containing ligands have been demonstrated to be very good
chiral sources for use in asymmetric conjugate addition reactions. After the successful
applications of Cu(II)-phosphoramidite complexes based on the BINOL scaffold,[83-88]
structurally related phosphite ligands containing a biphenolic unit and a chiral P-bonded
alcohol were also screened in Rh-catalyzed 1,4-additions of arylboronic acids to enone.
In aqueous media and under basic conditions, phosphites afforded excellent yields but
only
moderate
enantioselectivities,
albeit
higher
than
the
corresponding
phosphoramidites.[89]
1.1.2
Biocatalysis
An attractive alternative to enantioselective metal catalysts is offered by biological
systems. Natural enzymes catalyze biological transformations with remarkable
specificity and efficiency, using a limited number of functional groups in the protein
structure. Given that enzymes are products of evolution, they mostly function with high
selectivity only with polar, polyfunctional natural substrates (e.g., carbohydrates, acid
derivatives, and biopolymers), and under physiological conditions. Furthermore, the
single-handed orientation and lock-and-key specificity of the proteins significantly
reduce the substrate pool of catalysis, and might often cause severe product inhibition.
By this means, natural enzymes are generally not optimal catalysts.
One example of highly versatile class of natural enzymes used in synthesis of optically
pure compounds consists of hydrolases, namely lipases and esterases.[90-92] Many of
them were found to perform well even in organic solvents. Although lipases/esterases
can generally accommodate various synthetic substrates and maintain their chiral
recognition properties, they do not always exhibit satisfying catalytic performance, in
terms of activity, stability and most importantly, enantioselectivity.
1.1. Synthesis of Chiral Organic Molecules
28
1.1.2.1 Artificial Enantioselective Enzymes
Protein engineering has been employed to generate artificial enzymes that are modified
to be compatible with a desired chemical process. Two main approaches, namely (i)
rational design for fine-tuning existing biocatalysts, and (ii) combinatorial techniques
based on libraries and suitable selection methods[93] have been developed. In the first
approach, one or few amino acids in the enzyme are rationally either replaced with the
remaining natural amino acids by using site-directed mutagenesis techniques,[94, 95] or
chemically modified.[96] However, this method is rather challenging as it requires a
thorough understanding of the structure and mechanism of the targeted enzymes. An
efficient alternative to rational protein design was developed, namely the directed
evolution[95, 97, 98] (Figure 1.7). Large libraries were generated by random mutagenesis
(e.g., error-prone PCR)[99] of the gene encoding the catalyst, or recombinative methods
of gene fragments (e.g., DNA-shuffling).[100] The resulting mutants were then subjected
to gene expression and high-throughput screening methods[101, 102] in order to identify
improved variants. The mutant gene of the optimal enzyme variant can be resubmitted
to mutagenesis/expression/screening cycles until biocatalysts with improved properties
are obtained.
The first example of an in vitro evolved enantioselective enzyme was reported by Reetz
in 1997.[103] The isolated lipase afforded >90% ee in hydrolytic kinetic resolution of
chiral esters, while the wild-type enzyme from Pseudomonas aeruginosa gave only
2%.[103,
104]
Several enantioselective enzymes, such as lipases,[105,
106]
esterases,[107]
aldolases,[108] or oxygenases,[109] were optimized by in vitro evolution and mainly
employed in biological transformations.
1.1. Synthesis of Chiral Organic Molecules
Figure 1.7. Directed evolution of (enantioselective) enzymes.
29
1.1. Synthesis of Chiral Organic Molecules
30
1.1.2.2 Hybrid Catalysts
The applications of the artificial enantioselective enzymes in chemical processes were
expanded by the development of “hybrid catalysts” for use in transition-metal catalyzed
organic reactions. This term was introduced by Reetz in 2002,[110, 111] and consists in
embedding ligands and metal complexes thereof at a specific site in a protein. This
approach was inspired by Whitesides’ work, who showed that asymmetric catalytic
hydrogenations of α-acetamido-acrylic acid could be performed by anchoring an achiral
diphosphinerhodium(I) complex, via a biotin carrier, in a chiral cavity of the protein
avidin[112] (Figure 1.8). A similar system based on interaction of enantiopure
biotinylated PYRPHOS-rhodium(I) complex with the host protein avidin, was used as
catalyst in asymmetric hydrogenation of itaconic acid.[113] In both cases, the chirality of
the protein induced modest levels of enantioselectivity, with observed ee values ranging
between 33 and 44%, but definitely proved the principle of hybrid catalysts.
Figure 1.8. Asymmetric hydrogenation with avidin containing biotinylated phosphine-rhodium (I)
complex.
The enantioselectivity was later significantly improved by chemical tuning and rational
protein design. Ward introduced different spacers between biotin and the achiral
diphosphinerhodium(I) complexes, replaced avidin with streptavidin - a similar protein
with a deeper binding pocket - and, upon site-directed mutagenesis, obtained 96%
enantioselectivity in hydrogenation of α-acetamido-acrylic acid.[114, 115] Using the same
chemo-genetic optimization procedure,[96] enantioselective hydrogenases for Rucatalyzed reduction of prochiral ketones have been developed (97% ee).[116]
1.1. Synthesis of Chiral Organic Molecules
31
Figure 1.9. (A) Non-covalent anchoring of an active catalyst within a chiral host (hybrid catalyst). (B)
Examples of artificial metalloenzymes in asymmetric catalysis.
Transport proteins, such as serum albumins, are another class of efficient host proteins
able to strongly bind hydrophobic guests, for example porphyrins. Albumin-conjugated
manganese(III) and copper(II) complexes were found to catalyze sulfoxidation reactions
1.1. Synthesis of Chiral Organic Molecules
and,
respectively,
Diels-Alder
enantioselectivities.[117-119]
32
cycloadditions,
Non-covalent
insertion
with
of
moderate
to
good
chromium(III)-salophen
complexes into apomyoglobin mutants yielded metalloenzymes with low catalytic
efficiency in asymmetric sulfoxidation (27-83% yield, ee <13%).[120]
Reetz and co-workers applied the directed evolution approach for tuning the
enantioselectivity of hybrid catalysts, using the Whitesides system and streptavidin.[110,
111, 121]
A library of mutant hybrid catalysts was produced via random mutagenesis, and
posttranslational, non-covalent modification with metal complexes. Iterative cycles of
mutagenesis coupled with enantioselective screening procedure led to an improved
protein mutant, showing 65% ee.[121] A few examples of evolved artificial
metalloenzymes generated by non-covalent anchoring to protein cavities are depicted in
Figure 1.9 B.
In parallel to non-covalent anchoring strategies, covalent incorporation of transition
metal catalysts ensures an unambiguous localization of the metal centre in the host
protein. For this purpose, proteins with a single accessible reactive amino acid residue,
typically cysteine or serine, are site-specific functionalized with appropriately modified
ligand moieties. Di Stefano reported an artificial metalloenzyme obtained by attachment
of 1,10-phenanthroline to the cysteine residue of the adipocyte lipid-binding protein for
use in Cu-catalyzed enantioselective hydrolysis of amides and esters (31-86% ee).[122]
Using this approach, Reetz introduced salen and dipyridine moieties in the binding site
of papain[111, 123] (Figure 1.10). Preliminary studies showed that hybrid manganese-salen
and rhodium-dipyridine catalysts were active in epoxidation and hydrogenation, albeit
with low enantioselectivities (ee <10%).
De Vries described the covalent anchoring of a monophosphite to the cysteine residue
of papain (Figure 1.10), that yielded 100% racemic product in Rh-catalyzed reduction
of methyl-acetamidoacrylate.[124]
1.2 Nucleic Acid Enzymes
33
Figure 1.10. Catalysis with transition metal complexes covalently attached to proteins. Rhodiumphosphite- (top) and manganese-salen-functionalized papain (bottom).
Alternatively, diphosphine ligands carrying a p-nitrophosphonate moiety were
covalently linked to the serine residue in the active site of lipase. However, this
functionalization
turned out to be reversible, and undesired hydrolysis of the
phosphonate moiety was observed.[111]
1.2
Nucleic Acid Enzymes
1.2.1
Chirality in the Structure of Nucleic Acids
Nucleic acids are polymers composed of a polar, negatively charged sugar-phosphate
backbone and hydrophobic nucleobases (Figure 1.11). This amphiphilic nature, together
with the hydrogen bonding and stacking potential of nucleosides, determines the
assembly and maintenance of secondary and tertiary structures within nucleic acids.
The asymmetric D-ribose and D-2-deoxyribose sugars contain stereogenic centres,
whose pucker configuration is important for the overall DNA or RNA structure.
Typically, DNA adopts a double helical, antiparallel structure, via Watson-Crick base
pairing (Figure 1.12), whereas RNA exists mainly in a single-stranded form. However,
double helix elements are also a common feature of RNA structure and are fundamental
in biological functions of RNA.
1.2 Nucleic Acid Enzymes
34
Figure 1.11. Constitution of DNA and RNA, with the name of the monomeric nucleoside units.
The most common DNA conformations are the B- and the A-forms (Figure 1.13), both
right-handed, but with different sugar conformation (C2’-endo for B-DNA and C3’endo for A-DNA, Figure 1.14). In addition, each form displays distinctive helical
parameters, such as diameter, pitch, twist and tilt angels.[125, 126] The B-DNA appears as
a compact cylinder with a hydrophobic interior of base pairs stacked nearly
perpendicular to the helix axis at 3.4 Å intervals, achieving a complete rotation after 10
base pairs.[127,
128]
An important feature of the B-form is the presence of two
distinguishable minor and major grooves providing selective surfaces for the binding of
ligands such as proteins or small molecules. The A-DNA is wider and shorter than the
B-helix, and its bases are tilted to the helix axis (Figure 1.13). This form is
characterized by a complete turn after eleven base pairs and a reduced rise per base pair
of 2.6-3.3 Å.[127, 129] In this channel-type arrangement the minor groove is smaller, while
the major grove becomes deeper and narrower. This A-type helical orientation is
preferred by double-stranded regions of RNA (as in hairpins), RNA-DNA hybrids, as
well as by DNA-DNA duplexes containing one or more ribose units. A third
conformation type, although not very common in nature, is the left-handed Z-DNA. It is
generated by alternating conformations of the ribose rings (C2’-endo and C3’-endo) and
of the nucleobases (syn and anti) (Figure 1.14), and contains only one deep helical
groove.
1.2 Nucleic Acid Enzymes
35
Figure 1.12. A) Watson-Crick base pairs in DNA (top) and RNA (bottom). B) Hoogsteen base pairs in
DNA.
Figure 1.13. Different DNA helices.
1.2 Nucleic Acid Enzymes
36
Figure 1.14. A) C2’-endo or C3’-endo conformations of the ribose ring, B) anti and syn base orientation
(exemplified by adenine).
Beside the standard double helix form, DNA can adopt a number of different, more
complex structures, such as triplexes (through Hoogsteen base-pairing (Figure 1.12 B),
at low pH), quadruplexes (by folding of a guanosine-rich single chain), and Holliday
junctions (of four DNA strands),[126,
130, 131]
important for interaction with biological
components, such as proteins. By comparison to DNA, RNA possesses higher structural
and dynamic flexibility and has a higher the propensity to fold and form 3D higherordered structures that alternate helices and single-stranded regions or loops.
These structural features confer distinctive properties to nucleic acids and provide
numerous discriminatory intermolecular contacts with target molecules. In particular,
chirality plays a crucial role on the interactions of nucleic acids with chemical species,
such as drugs, or metal complexes,[125] determining the DNA/RNA molecular
recognition, binding affinity, and, if applicable, enantiodiscrimination.
For example, nucleic acids are able of forming precise binding pockets for the specific
recognition of substrates and cofactors. Therewith, combinatorial chemistry has been
used to identify nucleic acid sequences, namely aptamers, which recognize and bind
targets ranging from simple ions and small molecules, to peptides and single proteins. A
large number of small-molecule RNA aptamers have been isolated that can interact with
nucleotides and free nucleobases, amino acids, cofactors, basic antibiotics, and
transition-state analogues.[132] Single-stranded DNA can also recognize a variety of
small molecules, including ATP, organic dyes, porphyrins, and arginine.[133]
1.2 Nucleic Acid Enzymes
37
Figure 1.15. Binding pocket of the theophylline-RNA aptamer complex.[134, 135]
Structural studies have revealed that, upon contacting the ligand and conformational
change, the aptamers are able of forming precise, highly-ordered pockets, consisting
mainly of purine-rich loops. These elements are highly conserved and, by interactions
with spatially close nucleotides, often engaged in forming triplex, quadruplex, junctions
or pseudoknot structures.[136] Since planarity of the target molecules, presence of Hbond donors and acceptors, and positively charged groups appear to be the main factors
involved in molecular recognition, molecules bind with different affinities depending on
their geometry, hydrophobicity and overall charge. Aminoglycoside antibiotics with
multiple primary amino groups,[137,
138]
as well as the nucleotides,[139] and
nucleobases[140] are among the high-affinity ligands. In addition, many RNA aptamers
show high substrate specificity and can differentiate among closely related molecules.
The theophylline aptamer (Figure 1.15), for example, discriminates against caffeine,
which has only one additional methyl group, by 104-fold.[141]
Some aptamers can discriminate enantiomers of the target molecules, such as amino
acids and synthetic drugs, and bind them with high enantioselectivity. Examples of
small ligand enantiomers include L-arginine,[142] L-histidine,[143] or (R)-thalidomide.[144]
Small molecules can interact with the minor groove of B-DNA (e.g., polyamides[145]),
or intercalate between the base pairs. DNA intercalators contain a planar aromatic
heterocyclic functionality (achiral) which can insert and stack between the base pairs of
helical DNA. Their conjugation with chiral moieties can lead to a stereochemical
preference for interaction. Beside small chiral molecule recognition, nucleic acids
structures offer potential chiral environment,[146, 147] or chiral template for asymmetric
1.2 Nucleic Acid Enzymes
38
synthesis.[148, 149] However, this field of research is practically unexplored. Nevertheless,
exploring the chirality of nucleic acids in combination with in vitro evolution and
selection methodologies is a promising approach and may lead to a new generation of
bio-inspired functional molecules.
1.2.2
In vitro Selection of Nucleic Acid-Enzymes
Protein enzymes have dominated for a long time the field of biocatalysis. The 20 amino
acid components, in addition to the hydrogen bonding ability of the polyamide
backbone ensure substantial chemical diversity and structural versatility in enzymatic
catalysis. In contrast to proteins, nucleic acids with just four monomers and few
functional groups are restricted to hydrogen bonding, π-stacking, and metal-ion
coordination (Figure 1.16) for folding and interactions with potential substrates,[150] and
therefore catalytically limited. Only a few nucleic acid enzymes are found in nature, all
of them being RNA enzymes (ribozymes) (e.g., hammerhead ribozyme, hepatitis delta
virus ribozyme, group I self-splicing introns, and the ribosome).[151-153] They mediate
phosphodiester bond cleavage/formation and are responsible for peptide bond formation
in protein biosynthesis.[154-156] DNA is less catalytically competent than RNA, in part
since it lacks the 2’-hydroxyl group that can engage in hydrogen bonding as both donor
and acceptor (Figure 1.16). Moreover, the DNA double-helical form restricts the
structural flexibility and the potential of folding permitted by single-stranded
confomations and possibly required in catalysis.
Figure 1.16. Interactions occurring in DNA and RNA structures and that can contribute to catalysis.
1.2 Nucleic Acid Enzymes
39
Directed evolution[157] and in vitro selection strategies have led to the development of
DNA and RNA molecules with specific molecular recognition properties and catalytic
abilities. Various DNA enzymes (DNAzymes) and an even larger number of ribozymes
able of catalyzing a broad range of chemical reactions have been isolated using SELEX
techniques (Systematic Evolution of Ligands by EXponential enrichment).[151, 158-162]
Figure 1.17. In vitro selection of nucleic acids. The enriched DNA pool re-enters the selection cycle.
A general in vitro selection approach is shown in Figure 1.17. SELEX involves
screening of combinatorial libraries containing random sequences. For a given reaction,
one substrate (bond forming) or product molecule (bond cleavage) is attached to the
population of potential catalysts. A chemical tag (e.g., biotin) is appended to the other
substrate or to the product, so that a bond-forming reaction results in joining the tag to
the catalyst, whereas a bond-breaking reaction results in releasing the tag from the
catalyst. The tag is typically captured by affinity chromatography (e.g., streptavidincoated support). The applied selection procedure is tag specific, retaining tagged
molecules in case of a bond-forming reaction or rejecting tagged molecules in case of
1.2 Nucleic Acid Enzymes
40
bond-cleavage reaction. The active species are isolated, forming an enriched library,
which is then amplified. After iterated selection-amplification cycles, the individual
catalytically active species are identified by cloning and sequencing protocols, and
further optimized by rational design.
1.2.3
DNA-zymes
Deoxyribozymes isolated from pools of random-sequence DNAs catalyze the Pb2+dependent cleavage of RNA[163] and the oxidative Cu2+-mediated cleavage of DNA,[164]
facilitate the 3’,5’-linkage between two chemically activated DNA sequences in the
presence of Zn2+ or Cu2+,[165] formation of 3’-5’ and 2’-5’ junctions and of linear
(Figure 1.18) and branched RNA,[166] promote the metallation of porphyrin rings,[167]
and display peroxidase activity upon binding to hemin (Fe(III)-protoporphyrinIX).[168]
Many other reactions involving nucleic acid covalent modification are catalyzed by
DNA: ATP-dependent self-phosphorylation,[169] DNA adenylation (capping),[170] and
site-specific deglycosylation (depurination).[171] Overall, deoxyribozymes appeared to
generate rate enhancements similar to that of typical RNA enzymes, albeit inferior to
their protein counterparts.[172]
Figure 1.18. RNA ligation catalyzed by deoxyribozymes. Formation of 2’-5’ and 3’-5’ junctions between
readily available termini results in a linear RNA product.
Nearly all DNAzymes require metals for catalysis. Monovalent, divalent and even
lanthanide metal ions[173] can assist the optimal folding of DNA to form complex
tertiary structures. Moreover, divalent metal ions and lanthanides behave as Lewis acid
catalysts or general acid/base catalysts and trigger the reaction at the active site of the
enzyme.[173, 174]
To compensate for the lack of chemical moieties present in proteins, additional amino
acid cofactors have been incorporated into DNA. For example, an L-histidine dependent
deoxyribozyme has been reported to catalyze RNA transesterification in the absence of
1.2 Nucleic Acid Enzymes
41
divalent metal ions.[175] Histidine might act as general base catalyst to assist in
deprotonation of the target-site 2’-hydroxyl group.
Several approaches make use of modified nucleotides to expand the array of chemical
functionality of DNA. For example, imidazole and primary amino groups have been
incorporated into DNA (Figure 1.19) as surrogates for histidine and lysine. These
modified DNAzymes are able to catalyze RNA hydrolysis independently of a divalent
cation.[176, 177] The same amino-modified deoxyribozyme was found to effect scission of
DNA containing abasic sites and display apurinic/apyrimidinic lyase-endonucleaseactivity.[178]
Figure 1.19. A) RNA-cleaving modified deoxyribozyme (black, modified nucleotides are shown in blue)
and its target sequence (green). B) Structures of the modified nucleotides: histaminyldeoxyadenosine (A)
and aminoallyl-deoxyuridine (U).
1.2.4
Ribozymes
Since the discovery of catalytic properties of natural RNAs 25 years ago[179, 180], a large
number of ribozymes with novel catalytic properties has been developed by means of in
vitro selection. The chemical transformations catalyzed by RNA range from classical
reactions such as RNA hydrolysis and ligation to reactions including redox catalysis,[181]
urea synthesis,[182] glycosidic bond formation and nucleotide synthesis,[183,
184]
RNA
polymerisation,[185, 186] and aminoacylation of tRNA with aminoacids.[187, 188] Moreover,
it has been demonstrated that C-C bond forming reactions could be also accelerated by
RNAs. Examples include Diels-Alder reaction,[189-191] Michael addition,[192] and aldol
condensation.[193]
1.2 Nucleic Acid Enzymes
42
The impressive catalytic potential of RNA comes from its ability to fold into 3D
structures and form binding cavities for various substrates and metal ions.[194] In the
presence of divalent ions, RNAs can properly fold in very stable and rigid
conformations. In some cases, metal ions are involved directly in catalysis, by
stabilizing leaving groups or transition states.
Interesting examples of nucleic acids interactions with metal ions in aid of activity
regulation are revealed by the metal-binding allosteric ribozymes. Breaker described
several hammerhead ribozymes that are triggered and regulated selectively by binding
of certain metal ions (Figure 1.20), such as Cd2+, Co2+, Mn2+, Ni2+, Zn2+, Fe2+.[195]
Figure 1.20. Selection scheme for the isolation of cation-dependent ribozymes. The ribozyme core
contains a 40 nucleotide random-sequence domain (40nt). The RNA population is prepared by T7
transcription (1) and submitted to negative (no metal effectors) selection (2) and positive (with metal
effectors) (3) selection. The RNA species enriched for allosteric function are amplified by RT-PCR. Mg2+
ions (steps 2 and 3) are included in the selection cycle to promote high catalytic activity of the
hammerhead ribozyme.
The important role of metal ions as cofactors has been demonstrated not only in
reactions involving phosphodiester chemistry, as in the Pb2+-dependent 2’-O-mediated
RNA self cleavage or the 2’,3’-cyclic phosphate hydrolysis. For example, ribozymes
showing alcohol dehydrogenase activity,[181] or catalyzing racemic aldol reactions[193]
were selected in the presence of Zn2+. The catalytic activity of the Diels-Alder ribozyme
evolved in Eaton’s group was dependent on the presence of Cu2+and 4-pyridyl modified
uracil residues. Likely, Cu2+ and Ni2+ play a key role either in the structure of RNA or
in catalysis by providing Lewis acid sites upon coordination to the pyridyl moieties.[196,
1.2 Nucleic Acid Enzymes
197]
43
In contrast, the Diels-Alder ribozyme isolated by Jäschke’s group showed fast
multiple turnovers without requiring transition metals or replacement of the natural
nucleotides. In this case, hydrophobic interactions, electronic and proximity effects
were responsible for achieving catalysis. The crystal structure revealed the presence of a
preformed catalytic pocket almost perfectly complementary to the reaction product.[198]
Furthermore, this is the single reported example where the chiral binding cavity of a
selected RNAzyme directed the reaction towards one enantiomer of the chiral product,
resulting in an enantiomeric excess of 89%[147] (Figure 1.21).
Figure 1.21. A) Diels-Alder ribozyme crystal structure. B) Diels-Alder reaction between oligo(ethylene
glycol)anthracene derivatives and N-pentylmaleimide catalyzed by the ribozyme.[147]
The observation that RNA is capable of stereodiscrimination was also supported by
Eaton’s work on in vitro selection of RNA urea synthase. This ribozyme, that promotes
the formation of a urea bond between peptide phosphonate substrates and the exocyclic
amino group of the 3’-terminal cytidine, can stereoselectively recognize peptide
substrates for catalysis.[182]
1.2.5
DNA-Based Hybrid Catalysts
While proteins proved to be suitable chiral scaffolds to form hybrid catalysts and induce
enantioselectivity in asymmetric catalysis, attempts of employing nucleic acids in a
similar context have been only recently described. In 2005, Feringa and co-workers
reported on a supramolecular catalyst generated by intercalation of a copper-based
Lewis acid in salmon testes DNA (Figure 1.22). The double helical DNA provides then
enantioselectivity in Lewis acid catalyzed Diels-Alder cycloadditon of aza-chalcone to
cyclopentadiene, in water.[146, 199] Feringa’s work provided the first example of the use
1.2 Nucleic Acid Enzymes
44
of DNA as source of chirality in asymmetric catalysis.
Figure 1.22. Diels-Alder reaction of cyclopentadiene with aza-chalcone catalyzed by copper complexes
in the presence of DNA.
In this approach two classes of catalysts have been investigated. In the first case, achiral
bidentate pyridine ligands for Cu2+ coordination were attached via short spacers to 9amino acridine, a DNA intercalating moiety (Figure 1.22). As a result, the active Cu2+
centre is brought into close proximity to the DNA chiral environment and allows for
transferring the chirality from DNA to the reaction product. With such catalysts,
moderate to good enantiomeric excesses were achieved: 53% for the major isomer
(endo) and up to 90% for the minor (exo) isomer.[146] Interestingly, the design of the
metal binding ligand and the distance between the metal complex and the DNA helix
considerably affected the enantioselectivity. Thereby, aryl- and naphthyl-containing
ligands induced preference for the synthesis of opposite enantiomers. Elongation of the
spacers (n=3) resulted in change of the enantiopreference observed with a short linker
(n=2), while longer linkers (n=5) gave unsatisfactory results.
Optimization studies led to a second class of DNA-based catalysts (Figure 1.22), by
replacing the intercalator-spacer-ligand system with a bipyridine-containing moiety,
which behaves both as intercalator and bidentate ligand. In this system, the catalytic
metal centre is accommodated much closer to the DNA backbone. The obtained endoselectivities and enantioselectivities were dependent on the size and DNA binding
1.2 Nucleic Acid Enzymes
45
strength of the aromatic ligands. The best results (>99% endo isomer, >99% ee) were
given by smallest ligands, indicating that a shorter distance between Cu2+ and DNA is
beneficial for chirality transfer. In addition, the most active catalysts contained medium
DNA binders, suggesting that some flexibility in the binding of the complex favours a
preferred orientation of the reaction product.[199, 200]
The chirality transfer is explained by two possible mechanisms. In a one step
mechanism, the chiral DNA environment directs the orientation of the diene towards
one of the prochiral faces of the copper-bound dienophile. This pathway might
correspond to catalysts where Cu2+ is positioned very close to the DNA, which is
achieved by binding to intercalating ligands. Alternatively, in a two step mechanism, the
DNA
chirality
could
be
transmitted
to
the
achiral
ligand,
leading
to
enantiodiscrimination and different DNA binding affinities of the resulting chiral metal
complex. A preferred chiral conformation of the catalyst would then translate into
enantioselectivity in the catalyzed reaction.
However, in these systems the exact position of the metal complexes within the DNA is
not defined, making a thorough understanding of the role of DNA difficult. Towards
this end, a well-defined positioning of the metal complex and a precise control of the
coordination environment are essential. This prerequisite has been addressed by Kamer
who, at the beginning of 2007, reported on site-specific incorporation of
monophosphine
ligands
into
DNA
trimers.[201]
Solid-phase
bound
synthetic
oligonucleotides containing internal 5-iodo-2’-deoxyuridine have been reacted with
diphenylphosphine under Pd-catalyzed cross-coupling conditions. The resulting
trinucleotide-phosphine ligands have been tested in Pd-catalyzed allylic substitution, in
25% aqueous medium, giving <83% conversion and <12% stereoselectivity. In these
systems, the stereocontrol comes from the ribose moiety and not from the DNA folding.
Nevertheless, in the absence of more elaborate systems with well-defined secondary
structures, the application of nucleic acids as catalysts or scaffolds for transition metal
catalysts remains rather limited.
1.3 Transition Metal-DNA Conjugates
1.3
46
Transition Metal-DNA Conjugates
Although it became clear that DNA and RNA could optimally fit substrates or transition
states in a binding pocket and induce enantioselectivity, the catalytic potential of nucleic
acids in asymmetric catalysis remained practically unexplored. So far, the incorporation
of transition metal complexes into DNA and RNA was only considered for the
development of functional biomolecules with potential applications as therapeutics,
artificial nucleases, and as nanotechnology construction material. Metal complexes can
bind to nucleic acids via both noncovalent interactions and covalent attachment.
1.3.1
Non-covalent Interactions
Metal complexes are a very interesting class of reagents, which can site-specifically
target double-stranded DNA and RNA. Therefore, they found useful applications as
luminescent probes for DNA, mismatch recognition tools and structural probes for
RNA.
A labile ligand of the transition metal complex can be substituted by a nucleophile in
DNA, leading to formation of metal-DNA adducts. Nucleobases or phosphate groups
are available for direct coordination to the metal centre. Certain highly reactive metal
complexes are known to possess therapeutic effects due to irreversible binding to DNA.
One of the best examples is cis-platinum (cis-PtCl2(NH3)2), a square planar complex,
which is a very effective anticancer drug. Cis-platinum targets the nuclear DNA,
forming a critical lesion by cross-linking two adjacent guanines or an adenine and a
guanine on the same strand, through coordination of the platinum ion to the N7
nitrogen.[125, 202] Furthermore, antitumor activity and pronounced metastatic properties
were observed with ruthenium analogues, such as cis- and trans-RuIICl2(DMSO)4,
RuII(bpy)2Cl2, RuIII(tpy)Cl3, and [RuII(NH3)5Cl]Cl.[203, 204]
On the other hand, metal complexes interact with DNA via electrostatic binding, surface
binding to the minor or major groove, or intercalation of planar aromatic ligand into the
stacked base pairs.[205] In this category, coordinatively saturated octahedral complexes
of Ru2+ and Rh3+ containing phenanthroline units have been extensively employed as
1.3 Transition Metal-DNA Conjugates
47
luminescent reporters,[205] DNA cleaving,[206] or cross-linking agents,[207] and for the
study of long range energy and electron-transfer processes through DNA.[208]
Complexes
in
which
one
phenanthroline
moiety
is
replaced
by
9,10-
phenanthrenequinone diimine (phi) or extended by two aromatic rings such as in
dipyridophenazine (dppz) ligand are among the most studied major groove metallointercalators. Rhodium, ruthenium, and osmium complexes containing phi ligands
promote cleavage of DNA and RNA sites upon photoirradiation, and can be used as
probes for nucleic acids structure.[205] [Ru((phen)2dppz)]2+ possesses interesting
photophysical
properties,
intercalation.[209]
Minor
phenanthroline)copper(I),[210,
and
induces
groove
211]
a
binding
“light-switch”
molecules,
effect,
such
upon
as
DNA
bis(1,10-
Fe(II)-bleomycin,[212] and metal-porphyrins, display
DNA strand scission without irradiation.[213]
The coordination chirality of octahedral complexes gives rise to different binding
constants and recognition properties for the two enantiomers of the same metal
complex. The enantioselectivity in DNA binding was clearly established by Barton et
al. An interesting example refers to tris(2,7-diphenylphenanthroline)ruthenium(II),
whose enantiomeric forms specifically target right-handed B-DNA, and left-handed ZDNA, respectively, suggesting a correlation between the handedness of the complex
and that of the host DNA.[214]
1.3.2
Covalent Attachment
The requisite for metal complexes to target nucleic acids in a sequence-specific fashion
has led to the development of synthetic strategies for precise incorporation. The most
attractive way of achieving this goal involves tethering metal complexes to nucleic
acids via covalent attachment. In principle, metal binders can be appended to
oligonucleotide sequences either at the 5’- or 3’-termini or internally, at the nucleobase
residue or at the ribose 2’-position (Figure 1.23).
1.3.2.1 Post-synthetic Functionalization
Covalent attachment has been traditionally accomplished by post-synthetic strategies,
namely conjugation of a functionalized oligonucleotide with either a metal-chelator,
1.3 Transition Metal-DNA Conjugates
48
followed by metal complexation (1), or directly with a metal complex (2) (Figure 1.23).
In both cases, the post-synthetic derivatization has been commonly addressed by
reacting oligonucleotides containing 3’ or 5’-terminal amines or amine-tethered
nucleosides, with activated esters. These approaches afford nucleic acids that carry for
example metal-based cleavage reagents,[215] luminescent probes and redox-active
species.[216-218] Several examples are briefly described here (Figure 1.23).
Figure 1.23. Functionalization of DNA with transition metal complexes: 1) ligand attachment followed
by metallation, and 2) conjugation with transition metal complex. The DNA sites for attachment are
shown in blue. tap = 1,4,5,8-tetraazaphenanthrene, tpy = 2,2’:6’,2”-terpyridine.
Sigman reported on 5’-terminal attachment of 1,10-phenanthroline ligand to DNA
(Figure 1.23 left), which upon hybridization with an RNA target, induced site-directed
Cu2+-mediated hydrolysis of RNA and DNA.[219,
220]
Internal modifications of DNA
with metal binding moieties has been described by Telser et al. N-hydroxysuccinimidyl
(NHS) esters of bipyridine ligands for ruthenium coordination (Figure 1.23 left) have
been attached to DNA sequences containing 4- or 5-amino modified cytidine,
respectively deoxyuridine residues.[216] Recently, Liu et al. developed a method for
introduction of monophosphines into 3’- or 5’-amino tethered oligonucleotides.[221, 222]
Chemically stable metal complexes (e.g., ferrocene, Ru(II) complexes) (Figure 1.23
right) have been incorporated into DNA by postsynthetic derivatization of appropriate
amino-modified oligonucleotide precursors at position 5 of a thymine,[217,
termini,[218, 224] or at internucleotide positions.[218]
223]
3’/5’-
1.3 Transition Metal-DNA Conjugates
49
Post-synthetic solid phase strategies proved to be very efficient for attachment of
sensitive metalating species, due to the fact that all manipulations could be performed in
organic solvents. Oligonucleotides equipped with a 3’- or/and 5’-alkylamino
functionality
were
successfully
derivatized
with
[Rh(phi)2(bpy’)]3+
and
[Os(phen)(byp’)(Me2-dppz)]2+ complexes (phi = 9,10-phenanthrene quinonediimine,
bpy’ = 4-butyric acid-4’-methyl bipyridyl; phen = 1,10-phenanthroline, Me2-dppz = 7,8dimethyldipyridophenazine), while still attached onto the solid support[225] (Figure
1.24).
Figure 1.24. Solid phase synthesis of DNA-tethered rhodium and osmium complexes: 1) coupling of 3’amino modified DNA with the NHS ester of [Rh(phi)2(bpy’)]3+ complex, 2) 5’-amino functionalization of
the resin bound DNA, 3) coupling of the 5’-amino modified oligonucleotide with [Os(phen)(byp’)(Me2dppz)]2+ complex. DMT = 4,4’-dimethoxytrityl.
A similar approach was very recently reported by the group of Kamer. Resin-bound
trinucleotide DNA containing 5-iodo-2’-deoxyuridine was functionalized with
triphenylphosphine under Pd(0)-catalyzed cross-coupling conditions.[201]
1.3.2.2 Automated Solid-phase Synthesis
Interesting approaches of “on-column” derivatization have been established by Grinstaff
et al., in which metal complexes were incorporated into DNA during automated solid
phase synthesis. These methods couple an alkyne functionalized-ferrocene,[226, 227] or Ru(bpy)32+,[228] to a solid phase-bound oligonucleotide, containing 5-iodo-deoxyuridine
(Figure 1.25).
1.3 Transition Metal-DNA Conjugates
50
Figure 1.25. Conjugation of ferrocene and Ru(bpy)32+ complexes (B = A, C, G or T): 1) incorporation of
5-iodo-2’-deoxyuridine phosphoramidite during standard DNA synthesis, 2) Pd(0) cross-coupling of
alkyne functionalized metal complex and the resin-bound 5-iodo-2’-deoxyuridine, 3) normal
oligonucleotide synthesis is resumed.
Alternatively, solid-phase methodologies have utilized ligand-tethered or metallated
nucleoside analogues or metal-coordinating nucleoside mimics for subsequent use in
automated DNA assembly. Non-nucleoside based moieties such as 2,2’-bipyridine,[229]
phenanthroline or terpyridine[230] ligands, for Ru(II) and Cu(II) coordination, have been
converted into phosphoramidite building blocks and introduced via automated synthesis
at internal positions into DNA sequences. In other approaches, conjugation of EDTA
with C5-amino-modified thymidine via amide bond formation,[231] and coupling of
phenanthroline to the N2 position of deoxyguanosine,[232] followed by standard solid-
1.3 Transition Metal-DNA Conjugates
51
phase phosphoramidite chemistry, afforded oligonucleotides to promote Fe(II)-mediated
sequence-specific cleavage of DNA. Bipyridine ligands have been attached to
nucleobases (e.g., C5-iodo-deoxyuridine) by Sonogashira coupling, resulting in ethynyllinked conjugates.[233] A similar strategy has been employed by Tor’s group for sitespecific incorporation of Ru(II) donor and Os(II) acceptor polypyridine complexes as
tools to study photoinduced energy transfer in DNA duplexes.[234] Under Pd-catalyzed
cross-coupling conditions, ferrocene was tethered to 5-iodo-deoxyuridine and
incorporated into DNA using automated synthesis techniques.[227, 235]
Beside metallated phosphoramidite monomers that can be incorporated during solidphase synthesis, customized solid supports containing metallonucleosides, such as 2’Ru(bpy)2-deoxyuridine, have instead been prepared to initiate DNA synthesis, yielding
3’-metallated oligonucleotides.[236]
Figure 1.26. Modified oligonucleotides containing pyridine (A), 2,2’-bipyridine (B) and salen (C)
ligandosides coordinated to copper and manganese ions. D) The assembly of ten metal-salen base pairs
inside a DNA duplex.
Recently, the groups of Shionoya, Schultz, Sheppard, Tor and Carell reported metalcoordinating nucleic acids consisting of nucleoside mimics, called ligandosides, where
the heterocyclic base is replaced by a strong chelator. Moreover, such metal-binding
1.3 Transition Metal-DNA Conjugates
52
nucleosides could pair through metal coordination and replace the natural hydrogenbonded base-pairs. In particular, pyridines,[237] bipyridines,[237, 238] and salicylaldehyde
derivatives, precursors of salen ligands,[239-241] were coupled to the ribose units,
converted into phosphoramidites, and finally assembled into DNA sequences by solid
phase synthesis (Figure 1.26). Metal-mediated (e.g., Pd2+, Cu2+, Ni2+, Zn2+, and Mn2+)
ligandosides base pairing were then formed, affording stable DNA assemblies. In these
structures the metal is located inside the duplex structure. In addition, polynuclear metal
complexes could be formed in a predictable manner by incorporation of consecutive
metal-base pairs, thereby creating a double helix DNA where five copper,[242] or ten
manganese[243] (Figure 1.26 D) ions are stacked on top of each other. Such assemblies
are presumably precursors of molecular devices, such as molecular magnets and wires.
Despite several advantages of the solid-phase synthesis methodology, including
versatility, high yields of metal incorporation, and routine product isolation, the success
of this approach depends on the synthesis of individual metallated monomers,
compatible with automated DNA synthesis. A severe limitation is the requirement for
stable ligands that can survive the conditions used during synthesis, like deprotection,
oxidation, capping, or isolation. Therefore, the known repertoire of metal-binding
functionalities is rather scarce and consists mainly of nitrogen- and oxygen-donor
ligands. Moreover, the majority of metal complexes conjugated to oligonucleotides are
kinetically inert or without catalytic activity, except for their Lewis acidity.
The development of an efficient and flexible synthetic strategy for the incorporation of
other interesting classes of ligands would therefore be beneficial for the further progress
of the field and facilitate the generation of novel metalloribozymes and -DNAzymes.
2 Objectives
2
53
Objectives
The continuous interest in isolation of RNA and DNA molecules with novel catalytic
activities, in particular in chemical processes not existing in the biological world, and
the success of semisynthetic metalloenzymes in asymmetric catalysis prompted us to
become interested in the design of nucleic acid-based hybrid catalysts for
organometallic transformations.
This project aims at introducing metal-binding ligands into RNA and DNA folds and
developing transition metal complexes in which the activity is primarily dictated by the
organometallic catalyst precursor, while the selectivity is governed by the chiral cavity
created in the host nucleic acid molecule. The molecular recognition power of nucleic
acids, combined with the catalytic properties of transition metal complexes, is assumed
to facilitate catalytic reactions for which no enzymes or ribozymes are known. Since in
this case, a rational design approach could not span all possible structures, the
application of combinatorial methods is expected to generate artificial metalloDNAzymes and -ribozymes with the desired activity and selectivity. The combinatorial
selection of RNA-based hybrid catalysts is, however, a long term goal, which requires a
well-matched selection scheme. Unlike the known ribozymes, the in vitro selection of
RNA-hybrid catalysts needs an overall system that combines structural and functional
information from both nucleic acids and organometallic chemistry. For this, the
following major subjects have to be challenged:
•
well defined positioning of the metal complex in the RNA molecules
•
suitable reactions compatible with the structure and chemistry of nucleic acids
•
low stereoselectivity provided by the transition metal catalyst in the absence of
nucleic acids.
Towards this end, several specific questions should be addressed. For the selection of
RNA-transition metal catalysts the use of relatively short DNA/RNA hybrids is
envisioned as a way to provide the system with the necessary ligand for a transition
metal ion. The main focus of this work is aimed at the development of versatile methods
for the site-specific incorporation of metal-binding functionalities into nucleic acids.
2 Objectives
54
These approaches will be applied to covalently attach phosphites, mono- and bidentate
phosphines, as well as P,N-ligands either at the termini or at specific internal positions
of oligonucleotides and also in combination with various structural parameters. This
will allow expanding the repertoire of DNA sequences specifically interacting with
transition metals and afford attractive precursors for the development of metallo(deoxy)ribozymes. Studies on metal complex formation with phosphine and
phosphinooxazoline ligands in aqueous mixtures are carried out in order to explore the
suitability of phosphorus ligands for nucleic acid-based transition metal catalysis.
Moreover, a major target is the selection of suitable model reactions, namely transition
metal-catalyzed transformations that can be performed in water, and subsequently in
combination with DNA- and RNA-based ligands. In addition, rigorous analytical
methods need to be established that allow the detection of activity and selectivity with
pmol amount of catalyst. Since the steric course of the reaction is expected to be
influenced by the nucleic acid fold, transition metal complexes are chosen/designed that
catalyze the background reaction with modest or, preferably, no stereoselectivity.
Towards this end, the synthesis, characterization, and evaluation of the ligand systems
and transition metal complexes thereof, as well as of the reaction substrates will be
carried out and discussed in detail.
Implementation of achiral ligands in these systems and therefore the exploitation of the
nucleic acid scaffold as the only source of chirality are severely restricted by the
covalent attachment of the ligand to the biopolymer. Therefore, this work aims at
assessing to what extent the stereogenic information carried by a chiral ligand will be
complemented by that of the nucleic acid part. On the way to the goal of creating
nucleic acid-based catalysts, the influence of DNA and RNA on the activity and
selectivity of the tethered metal complexes will be investigated by employing rationally
designed model compounds. Due to the inherent chirality of the DNA backbone, DNAbased ligands may effect transfer of chiral information to the chemical reaction.
Additionally, from the design and synthesis standpoint such DNA conjugates are
attractive systems to work with. Finally, exploring the DNA sequence design, ligand
tethering and nucleic acid helix properties is anticipated to aid in gaining insights into
the structural basis of DNA-transition metal interactions and to provide tools for
designing DNA-based catalysts with improved properties.
3.1 Incorporation of Metal Complexes into Nucleic Acids
3
Results and Discussions
3.1
Incorporation of Metal Complexes into Nucleic Acids
55
Design of RNA-transition metal system
In the particular case of RNA-hybrid catalysts, the selection procedure requires an
additional step for embedding transition metal complexes within RNA sequences
(Figure 3.1). Furthermore, a well-defined localization of the transition metal in the
nucleic acid scaffold is essential for a clear understanding and, later on, for
manipulating the RNA’s role in catalysis.
Figure 3.1. General scheme for in vitro selection of RNA-based hybrid catalysts.
Therefore, we aim at creating precise metal binding sites in RNA by careful placement
of ligands (preferably achiral) within the context of the overall tertiary structure. For
3.1 Incorporation of Metal Complexes into Nucleic Acids
56
that, two possible approaches were considered: 1) incorporation of metal-binders at a
particular nucleotide site within the random region of each sequence of the RNA pool
(Figure 3.1 and 3.2 A); 2) co-selection in the presence of a stoichiometric amount of
DNA functionalized with a ligand for metal coordination (Figure 3.1 B).
Figure 3.2. Site-specific incorporation of transition metal complexes into RNA and DNA.
The first approach requires either ligand-tethered nucleotides or modified nucleotides
bearing functionalizable groups for post-transcriptional modification with metal
chelators. For that, one must take into account the factors affecting the synthetic
accessibility of the unnatural nucleotides in the form of triphosphates,[244-246] their
compatibility with the known polymerases[247] involved in the selection cycle, and the
stability of metal complexes during enzymatic manipulation.[248] Additionally,
substituting an unnatural base in the DNA or RNA template involves a novel hydrogenbonding pattern in order to retain the high-fidelity in transcription, PCR and reversetranscription reactions and to afford the site-specific introduction of its corresponding
complement. Therefore, establishing a unique, convenient base-pairing system to be
reproducibly incorporated at a site-specific occurrence during each selection cycle
appears very challenging and time-consuming.[249]
As an alternative to achieve precise positioning of the transition metal complex in the
RNA fold, a double-stranded DNA/RNA hybrid based on a short modified DNA
fragment that matches the 3’-end constant region (cDNA priming site) of the original
random-sequence population was chosen (Figure 3.2 B). With the help of a transition
metal-DNA carrier a new selection scheme could be designed. In this case, the
localization of the DNA-tethered metal complex to the RNA sequence is primarily
directed by proximity and Watson-Crick base-pairing. However, these hybridization
methods are necessarily limited to reaction conditions compatible with DNA/RNA
3.1 Incorporation of Metal Complexes into Nucleic Acids
57
duplex formation, such as aqueous environment, precluding a large number of potential
chemical transformations. Therefore, the DNA/RNA hybrid approach requires the study
of a model reaction, in which proper conditions for RNA folding, proximity effects and
tertiary interactions, important for generating local binding sites and catalytic pockets,
are maintained.
Criteria of ligand choice
The design of the ligand should produce a structure which can be synthesized fairly
readily and appropriately for incorporation into DNA/RNA. Furthermore, the choice of
the ligand must take into account the factors affecting the stability of metal complexes
in close location to nucleophilic sites in nucleic acids. Nucleobases or phosphate groups
in nucleic acids are available for direct coordination to the metal centre. Thus, they can
substitute and prevent the ligands from binding, and finally dramatically influence the
reactivity of transition metal catalysts. It is anticipated that strong metal binders might
overcome this problem by reducing the lability of metal complexes with respect to
simple substitution, and dominate the control on metal coordination environment by
virtue of the electron-rich nature and chelate effects.
An efficient approach to the modification of DNA with transition metal chelators needs
to fulfill the following requirements: (1) generality - the incorporation of ligands for
metal coordination at any position along the DNA should be possible; (2) structural
stability - the modification should ensure minimal structural perturbation of the DNA
duplexes; (3) versatility and tunability - various metal binders as well as tethers for
attachement to the DNA should be accessible; (4) simplicity - synthetic approach
compatible with the chosen ligands; and (5) high yielding and regioselective DNA
functionalization.
Phosphorus ligands possess high affinity for transition metals and are among the most
efficient and extensively used ligands in transition metal catalysis. Attracted by the
broad applicability of this class of ligands, we attempted the preformation of phosphitecontaining oligonucleotides (ODN) and also prepared DNA conjugates carrying monoand bidentate phosphine and phosphinooxazoline ligands as precursors for introducing
metal centres at well-defined positions in DNA and RNA sequences.
3.1 Incorporation of Metal Complexes into Nucleic Acids
3.1.1
58
Functionalization of DNA with Phosphoramidite Ligand
Precursors
Cyclic phosphites were initially chosen as target ligands for incorporation into DNA
employing
solid-phase
synthesis.
This
approach
involves
the
synthesis
of
phosphoramidites analogues P1-3 (Scheme 3.1) and their sequence-specific
incorporation into oligonucleotides using the phosphoramidite chemistry. The choice of
such ligands originated from the following criteria: (1) low sensitivity to oxidation
compared to phosphines and (2) straightforward synthesis from phosphoramidite
precursor.
Cyclic phosphoramidites P1-3 were obtained by direct phosphitylation of appropriate
diols with neat phosphorus trichloride, resulting in a chlorophosphite, followed by
displacement with amines (Scheme 3.1). The phosphoramidites were purified by flash
chromatography over silicagel. These building blocks were thus obtained in high purity
(important for high coupling efficiency during oligonucleotides synthesis) and
characterized by 1H and 31P NMR.
Scheme 3.1. Synthesis of phosphoramidites P1-3.
As model systems we have chosen to covalently attach the aromatic phosphoramidite
P2 at the 5’-end of pentanucleotide DNA sequences using standard solid phase
phosphoramidite DNA synthesis conditions and to generate a phosphite-type linkage
(Scheme 3.2). Attachment of an achiral phosphoramidite precursor (P1) or chiral but
configurationally fluxional P-bonded biphenol unit (P2,3)[82,
89]
would result in chiral
DNA-based ligands in which the chirality is exclusively dominated by the nucleic acid
3.1 Incorporation of Metal Complexes into Nucleic Acids
59
structure.
Incorporation of phosphite units required several adaptations in the automated synthesis
of oligonucleotides, especially omission of the iodine-oxidation and deblocking step in
this particular coupling cycle. Despite the stability of these ligands, the phosphoramidite
unit is not likely to tolerate the acidic treatment required by removal of trityl groups.
Phosphoramidite P2 was initially coupled to sODN1 (Chapter 5.4.1) as the last residue,
using the trityl-on synthesis mode. The deprotection and cleavage of the completed
oligonucleotides from the solid support was achieved by overnight incubation with
concentrated ammonium hydroxide at room temperature, affording the crude
oligonucleotide. The product was analyzed by reversed-phase HPLC (Figure 3.3 A),
and identified as the 5’-OH unmodified sODN1 by MALDI-TOF MS.
Scheme 3.2. Attempted solid-phase synthesis of phosphite-containing oligonucleotides sODN1-3.
Reaction conditions: (a) detritylation (TCA, dichloromethane), P2 phosphoramidite coupling (BTT,
acetonitrile), (b) I2 oxidation, (c) deprotection and cleavage from the solid support with 28% NH4OH, rt,
overnight.
Further attempts to couple either P1 or P2 to 5mer oligonucleotides containing nonstandard phosphoramidite building blocks (e.g., a decaethyleneglycol spacer molecule
S[250] as for sODN2 and sODN3, Scheme 3.2) gave in all cases unsatisfactory results.
However, for proving the assembling of P2, the synthesis of the phosphitefunctionalized DNA conjugate was re-attempted carrying out the same solid-phase
protocol under usual conditions, without excluding the oxidation step of automated
3.1 Incorporation of Metal Complexes into Nucleic Acids
60
oligonucleotide synthesis. Upon iodine-oxidation the presumably formed phosphite
linkage would result in a highly stable phosphate ester-type functionality (Scheme 3.2),
unproblematic for post-synthetic manipulations.
As a result, removal of the DNA from the solid support using concentrated ammonium
hydroxide, followed by overnight incubation at room temperature yielded a single
product as observed in the HPLC chromatogram (Figure 3.3 B). Owing to the
hydrophobicity added by the phosphite moiety, the modified DNA conjugate eluted
with later retention time (tR = 29.9 min) compared to the unmodified DNA (tR = 18.0
min). The isolated product was then analyzed by mass spectrometry and, corresponded
indeed to the biphenyl-phosphate containing DNA conjugate sODN1-P2(O) (overall
yield: 45%; calculated [M-H]-: 1691, measured: 1710).
Figure 3.3. HPLC chromatograms of the attempted DNA solid-phase derivatization (no iodine oxidation)
with P2 (A), the phosphate-containing DNA conjugate sODN1-P2(O) (iodine oxidation) (B), and the
crude DNA product obtained after treatment of resin-bound DNA-phosphite sODN1-P2 with
[Rh(cod)Cl]2, followed by ammonium hydroxide deprotection. ■ Failure DNA sequences obtained by
automated solid-phase synthesis and 5’-OH unmodified sODN1; ● protected oligonucleotide and residual
organic molecules.
3.1 Incorporation of Metal Complexes into Nucleic Acids
61
Having demonstrated the coupling of P2 phosphoramidite, we were still confronted
with stability and isolation of DNA-phosphite conjugates. At this point, we assumed
that long time exposure to basic conditions required by post-synthetic DNA cleavage
and deprotection might lead to decomposition of the DNA-appended phosphite.
It is known that compounds containing P–O/N bonds have proved to some extent
unstable, being able to undergo hydrolysis in protic solvents and lead to either cyclic Hphosphonates or ring-cleavage products[251] due to the tendency of phosphorus to form
P=O bonds. For example, Feringa reported on stability of phosphoramidite ligands
which after heating to 100 °C for 5 hours in dioxane/H2O 10:1 were completely
hydrolized, whereas their corresponding rhodium(I) complexes remained unchanged
despite the drastic conditions.[252] This observation was also confirmed in the case of
phosphite ligands that appeared to be fully stable in protic solvents upon coordinating
transition metals.[253]
Prompted by these findings, we made use of the phosphites ability to bind transition
metals and form stable complexes, as a way to overcome the problematic isolation of
phosphite-containing DNA from aqueous mixtures. Complex formation between the
DNA-appended phosphite and 1,5-cyclooctadienerhodium(I) chloride dimer was
attempted by simply combining the solution of [Rh(cod)Cl]2 precursor in acetonitrile
with the CPG beads coated with the DNA conjugate and stirring the resulting
suspension. In this approach, the phosphate moieties of the DNA backbone were all
protected as cyanoethyl esters so the oligonucleotide was uncharged and well solvated
by organic solvents. The crude DNA was liberated from the bead and the protecting
groups removed by treatment with aqueous ammonia. This final step was monitored by
reversed-phase HPLC under the same conditions employed for analysis of unmodified
oligonucleotides. Figure 3.3 C shows the chromatogram of the crude DNA product after
30 minutes incubation at 65°C with ammonium hydroxide. Several early-eluting side
products that did not contain the hydrophobic biphenolic moiety, likely failed sequences
in the DNA synthesis, and the 5’-OH unmodified sODN1 oligonucleotide were
obtained. Additional DNA products with higher retention times might correspond to
protected oligonucleotide likely due to insufficient deprotection time. The single DNA
conjugate (tR = 27.8 min) not belonging to any of these two categories was isolated in
18% yield, liophylized and characterized by MALDI-TOF mass spectrometry. The mass
3.1 Incorporation of Metal Complexes into Nucleic Acids
62
spectrum confirmed that the isolated species was the pure sODN1-P2 phosphite-DNA
conjugate (sODN1-P2: calc. [M-H]- 1675, found 1685) and no trace of decomposition
products could be detected.
These results indicate that the Rh complex was formed, in a certain extent, and was
fairly stable under the conditions employed in the deprotection and cleavage protocol.
However, the moderate yield does not exclude the presence of free DNA- phosphite in
the crude reaction mixture because free phosphite ligand hydrolyses in water, leading to
5’-OH unmodified sODN1, which was observed during HPLC analysis. Furthermore,
the isolation of rhodium complexes on reversed-phase chromatography appears difficult
and generally results in displacement of the transiton metal ion from the complex due to
the hemilability of P-Rh bond.
All these observations directed our attention to phosphine ligands, known to be more
stable against hydrolysis, albeit highly sensitive to oxidiation, as suitable metal-binding
moieties for DNA functionalization.
3.1.2
DNA-Phosphine Ligands
Obviously, phosphine ligands need to be incorporated after DNA synthesis to avoid
exposure to the oxidation step during automated solid-phase synthesis. We envisioned
the post-synthetic modification of oligonucleotides as the most suitable and versatile
approach for the preparation of phosphine-carrying DNA conjugates. Various
aminoalkyl-modified oligonucleotides have been successfully reacted at predetermined
internal sites with carboxylate derivatives of PYRPHOS, BINAP and PHOX ligands
(Figure 2.3), affording the first examples of DNA sequences carrying mono- and
bidentate phosphine ligands as well as P,N-ligands.[254]
3.1.2.1 Amino-Functionalized Oligonucleotides
For the preparation of amino-modified oligonucleotides, we employed the “convertible
nucleoside” approach developed by Verdine[255] and Swann.[256] The oligonucleotide
chain is elongated using a building block that is a precursor of the amino-functionalized
nucleoside, a so-called “convertible nucleoside”. This precursor may be transformed
into a range of differently modified nucleosides in the final steps of the synthesis[257-259]
3.1 Incorporation of Metal Complexes into Nucleic Acids
63
(Scheme 3.3). This strategy applies to the modification of exocyclic positions, such as
the 4-position of deoxycytidine, a biologically important site as it is directly involved in
DNA Watson-Crick base pairings.[255, 256, 260, 261]
Scheme 3.3. Synthesis of amino-modified oligonucleotides using the “convertible nucleoside” approach
(convertible nucleoside = 4-triazolyl-2’-deoxyuridine). PG = protecting group, = CPG.
The “convertible nucleoside” approach (Scheme 3.3) involves three steps: (1)
preparation of a versatile monomer containing a suitable leaving group on the
nucleobase, such as 4-triazolyl-deoxyuridine; (2) incorporation of the monomer into
oligomers; (3) replacement of the leaving group after DNA synthesis with diamines.
This approach offers the advantage that the triazolyl leaving group is stable during DNA
synthesis and subsequent deblocking step and only post-synthetically convertible to
introduce the required alkylamino group on the base.
In our case, the “convertible nucleoside” strategy was adapted and optimized.
Oligomers containing 4-triazolyl-deoxyuridine were prepared by automated DNA
synthesis, on 1 µmol scale. No modification was made for the incorporation of the
modified monomer compared to the natural ones, for which a satisfactory coupling
yield, mostly over 95%, was obtained. All oligonucleotides were synthesized with
retained 5’-terminal trityl group that simplifies the purification of the desired full-length
products.
3.1 Incorporation of Metal Complexes into Nucleic Acids
64
In order to facilitate the removal of the protecting groups from normal bases, we chose
to use base-labile monomers, in which dA, dG and dC are protected with the t-butylphenoxyacetyl (TAC) group. This protecting group can be readily removed by
concentrated aqueous ammonia solution within 2 h at room temperature or 15 min at
55°C (Proligo’s protocol). Together with other non-standard phosphoramidite building
blocks (e.g., a decaethyleneglycol spacer molecule S as for ODN3), or 7-deazaguanosine monomer (ODN5a,b), this protocol allowed moderate to high-yielding
syntheses of long ODNs with the convertible nucleotide at varying internal positions.
After synthesis, the ODNs were treated with suitable diamines (Table 3.1):
ethylenediamine,
1,4-butanediamine,
and
1,13-diamino-4,7,10-trioxatridecane,
affording conversion of the 4-triazolyl-dU to different 4-alkylamino-dC derivatives that
can base-pair like a normal cytidine nucleotide.[260] At the same time, the ODN is
cleaved from the support and deprotected.
Table 3.1. Preparation of amino-modified ODN1-5.[a]
ODN
Sequence
ODN1a
ODN1b
H2 N
5'-GC AGT GAA GGCR TGA GCT CC-3'
ODN1c
ODN2
ODN3[c]
H2N
5'-GC AGT GAA GGC TGA GCT CCT ACRC-3'
ODN5b[d]
O
2O
30
H2 N
25
H2N
R
5'-GC AGC GAT AAC TAA GCG CT-3'
5'-GC AGT GAA XXCR TXA GCT CC-3'
21
H2 N
H2N
O
H2N
H2 N
40
32
H2 N
5'-GC AGT GAA GGC TGA GCT CCS CRC-3'
ODN4c
ODN5a[d]
42
H2N
ODN4a
ODN4b
Yield[b]
[%]
35
R
2O
22
14[e]
5[e]
[a] Reaction conditions: 5 M aqueous solution of 1,4-diaminobutane or ethylenediamine, r.t., 4 h or neat
1,13-diamino-4,7,10-trioxatridecane, r.t., 4 h (followed by additional treatment with water, 5 h). [b]
Isolated yields after solid phase synthesis (1 µmol), conversion and purification. [c] A decaethylene
glycol unit S was incorporated during solid phase synthesis. [d] X denotes 7-deaza-guanosine nucleotide.
[e] Moderate yields due to low coupling efficiency of the X monomer.
3.1 Incorporation of Metal Complexes into Nucleic Acids
65
To investigate the reaction time required for complete deprotection, cleavage and
replacement of the triazolyl group, the resin-bound DMT-on ODN1 was treated with
1,4-butanediamine at room temperature and monitored by reversed-phase HPLC (Figure
3.4). The desired amino-modified oligomer as confirmed by MALDI-TOF mass
spectrometry was obtained in 4 hours and longer reaction times did not improve the
yields of deprotection, cleavage and conversion.
Compared to reported methods, this mild one-pot conversion, deprotection and cleavage
procedure gives consistently high yields of amino-modified DNA sequences in short
reaction times.
Figure 3.4. Reversed-phase HPLC chromatograms of the crude tritylated ODN1a at different reaction
times.
Furthermore, our synthetic approach allows the parallel synthesis of various conjugates
differing in length and structure of the spacer, which may be of particular relevance in
determining the interaction between the transition metal complex and the
biopolymer.[146]
3.1.2.2 Duplex Stability of Amino-Tethered Oligonucleotides
It is well known that in B-DNA form, the non-base-paired substituent on the N4position of deoxycytidine projects directly into the central space of the major groove
(Figure 3.5) and provides an excellent location for the attachment of DNA-interacting
ligands.[255] At the same time, it corresponds to one of the least sterically demanding
positions available on a DNA base.
3.1 Incorporation of Metal Complexes into Nucleic Acids
66
Figure 3.5. N-alkyl tethers attached to the exocyclic amine of dC, allowing the major groove to be
targeted without interfering with Watson-Crick base-pairing.
Therefore, attachment of a tether at this site should induce little steric perturbation of
duplex DNA structure as demonstrated by X-ray crystallographic studies on duplex
DNA containing the related N6-methyladenine.[262] Figure 3.6 illustrates the positioning
of the tether in a 4-thioethyl-dC containing B-DNA model.
Figure 3.6. Location of N4-ethylthiol in dC-tethered B-DNA.[255]
However, it has been shown that N4-alkylamine and -alkylthiol substitutents (Figure
3.6) do destabilize duplex DNA, in some extent, on the basis of electronic factors. For
example, a destabilizing effect of 6-7°C has been observed for a 15mer double-stranded
DNA, each strand containing either a N4-butylamino- or ethylamino-modified 2’deoxycytidine residue. Therefore, in many applications the effects of single unnatural
bases of this type on DNA duplex stability have been considered relatively modest.
3.1 Incorporation of Metal Complexes into Nucleic Acids
67
Moreover, it has been assumed that further derivatization and subsequent covalent
attachment of DNA-interacting moieties would not exert a strong effect on duplex
stability.
These systems resemble very well our amino-modified oligonucleotides ODN1a and
ODN1b, respectively. Therefore, we assumed that also in our case the effect of
alkylamine tethers on DNA stability would be low in aqueous solutions containing salts.
Since we aim at performing organometallic transformations in the presence of DNA
hybrids, we were uncertain what amount of organic solvent would be optimal for our
system. The amino-modified oligonucleotide ODN1a precursor (Table 3.1) which
contains an intermediate C4-linker was used to study the effect of organic solvents on
the double-helical DNA conformation. We then prepared the complementary DNA
strand cDNA1 (5’-GG AGC TCA GCC TTC ACT GC- 3’) using standard solid-phase
synthesis and deprotection procedures (Chapter 5.3.4). The duplexes were formed by
combining equimolar amounts (2 nmol) of each strand in Hepes buffer (15 mM, pH
7.5)[263] in the presence of dioxane (0-30% v/v), annealing them together by heating to
90 °C and gradually cooling to room temperature. The results are depicted in Figure 3.7.
Figure 3.7. A) Melting profiles of the ODN1a/cDNA1 duplex in the presence of various concentrations
of dioxane. B) Plot of melting temperature versus dioxane concentration.
The black curve in Figure 3.7 A shows the change in absorbance at 260 nm of the
ODN1a/cDNA1 duplex dissolved in 100% buffer containing 150 mM NaClO4 and 7.5
mM Mg(ClO4)2 when subjected to heating from 15 to 90°C at 5°C per min (control
experiment). The shape of the curve resembles a typical DNA denaturation curve.[264]
Thermal denaturation involves unstacking of the bases, which gives rise to an increase
in absorbance. The Tm value determined for this curve in Figure 3.7 B is 57.6±0.6°C.
3.1 Incorporation of Metal Complexes into Nucleic Acids
68
When the DNA duplex pre-formed in 5-30% dioxane was subjected to the same
heating/cooling annealing procedure as in water, it yielded the same sigmoidal
dependence as the black curve in Figure 3.7 A, indicating that annealings in neat water
and in dioxane/water mixtures show the same behaviour.[265, 266] Under these conditions
the melting profiles of double-stranded amino-modified DNA were substantially shifted
with increasing concentration of organic solvent. However, the hyperchromicities
remained unchanged, indicating that the organic solvent did not alter the DNA integrity
significantly at room temperature.
The melting temperatures were plotted as a function of the volume percentage of
dioxane. Figure 3.7 B shows an almost linear dependence of the Tm values on the
dioxane concentration, in agreement with literature data that supported a linear decrease
of the thermal stabilities of DNA/DNA and RNA/DNA duplexes with increasing
concentration of formamide.[267]
Our results demonstrated that ODN1a/cDNA1 still preserved its double-helical
conformation in 30% aqueous dioxane, at high salt concentration, although its thermal
stability was lower than in 100% water, as reflected by a 21.4°C reduction in the Tm
value (Figure 3.7 B). We estimate that structurally related DNA/DNA and DNA/RNA
duplexes, for example functionalized with metal-chelating moieties, would show similar
behaviour in aqueous-dioxane mixtures as their amino-tethered double-stranded DNA
precurors.
3.1.2.3 Reactivity of Amino-Modified Oligonucleotides
Before studying the phosphine systems and their attachment to DNA, which were
expected to require anaerobic conditions, the coupling reaction of amino-modified
ODNs with carboxylic acid groups and amide bond formation were initially investigated
using the N,N’-bis(2-picolyl)amine derivative bpa[268], highly stable against oxidation.
Thus, ODN1a, ODN2 and ODN3 were reacted with the in situ formed Nhydroxysuccinimide (NHS) ester (100-200 equiv) of bpa in 66.7% DMF, pH = 8.5, for
48 hours, at room temperature (Scheme 3.4). Precautions were taken in the pH
adjustment of the reaction mixture, because at high pH competing hydrolysis of the
NHS ester bond occurs. Under our conditions, the coupling of bpa to ODN2 and ODN3
proceeded to completion, as demonstrated by the PAGE. In the case of coupling to
3.1 Incorporation of Metal Complexes into Nucleic Acids
69
ODN1, the reaction conversion could not be estimated due to co-migration of excess
bpa activated ester with the amino-modified DNA starting material.
Figure 3.8. UV-shadowing of the 18% PAGE gel (λ = 254 nm) used for analysis of the DNA-appeneded
bpa conjugates ODN14-16. Lane 1 - control amino-modified ODN1a; lane 2 - control amino-modified
ODN2; lane 3 - control amino-modified ODN3 (see Table 3.1 for abbreviations). Coupling reaction of
bpa after 48 hours incubation, at room temperature with amino-modified ODN1a - lane 4, ODN2 - lane
5, and ODN3 - lane 6, respectively. The arrows indicate excess bpa NHS-activated ester and hydrolyzed
ester, respectively.
Figure 3.8 depicts an analytical denaturing polyacrylamide gel comparing the retarded
bands corresponding to N,N’-bis(2-picolyl)amine-containing DNA conjugates ODN1416 to the starting amino-modified ODN1a. Due to the increased mass, the
electrophoretic mobility of ODN14-16 is lower than of the oligonucleotide precursor.
3.1.2.4 Post-synthetic Functionalization of Amino-modified DNA with
Phosphine Ligands
Having demonstrated the reactivity of amino-modified ODNs, amide bond formation
between ODN1a and commercially available 4-(diphenylphosphino)-benzoic acid L1
was chosen as the model reaction to investigate the coupling of phosphine-based ligands
(Scheme 3.4). Phosphine L1 was first activated by using N-(3-dimethylaminopropyl)N′-ethylcarbodiimide (EDC) in the presence of N-hydroxysuccinimide (NHS), and the
in situ generated active ester was directly added to the ODN1a solution.
3.1 Incorporation of Metal Complexes into Nucleic Acids
70
The coupling reaction was analyzed by reversed-phase HPLC and proceeded to
completion, affording 60% of the desired DNA-phosphine conjugate ODN6. Not
surprisingly, a fraction (≤20%) of ODN6 was oxidized to the corresponding phosphine
oxide ODN6(O), presumably during workup. Small amounts of byproducts were
observed that did not carry a phosphine moiety, suggesting slight degradation of the
starting material (Figure 3.9).
The desired phosphine-containing oligonucleotide could be easily isolated by HPLC
and was stable under the purification conditions. While HPLC purification as described
here does not cause oxidation of the phosphine-DNA, sample preparation, collection
and manipulation should be performed in oxygen free conditions. MALDI-TOF mass
spectrometry of the HPLC purified ODN6, however, gave only the mass of the oxidized
product ODN6(O) (Table 3.2).
Scheme 3.4. Post-synthetic functionalization of amino-modified ODN with N,N’-bis(2-picolyl)amine
bpa, phosphines L1-3 and phosphinooxazoline L6. C* = N4-alkylamino-modified 2’-deoxycytidine. S =
decaethylene glycol unit.
3.1 Incorporation of Metal Complexes into Nucleic Acids
71
Figure 3.9. HPLC chromatogram of ODN6 pre-purified by chloroform extraction and ethanol
precipitation to remove the excess of coupling reagents. Trace amounts of degradation products elute
between 20 and 24 min, similarly to the starting material ODN1a (tR = 21.9 min) (for HPLC conditions
see Chapter 5.5.1).
To prove the identitity of the oligonucleotide eluting with tR = 39.0 min as phosphineDNA conjugate ODN4a, the HPLC eluate was treated with sulfur[269] to yield the airstable phosphine sulfide analogue ODN6(S) with tR = 38.0 min. The MALDI mass
spectrum clearly confirmed that the isolated species was the pure phosphine-DNA
ODN6 and no trace of the oxide ODN6(O) was detected.
Amino-modified oligonucleotides ODN2 and ODN3 were also reacted with phosphine
L1. The coupling reactions proceeded consistently well, affording 65% and 68% of
ODN7 and ODN8 respectively (Table 3.2).
Table 3.2. Isolated yields and MALDI-TOF analysis of ODN6-8.[a]
Entry
Conversion
[%]
Isolated yield
[%]
ODN(O)[b]
-
ODN(S)[c]
[M-H]
calcd
obsd
calcd
obsd
>99
60
6228
6234
6244
6249
ODN6
96
65
7424
7430
7440
7447
ODN7
>99
68
7328
7332
7344
7350
ODN8
[a] ODN7 and ODN8 are the coupling products of L1 with ODN2 and ODN3, respectively. [b] ODN(O):
DNA-phosphine oxide. [c] ODN(S): DNA-phosphine sulfide.
Having established the optimal conditions for coupling the monophosphine derivative
L1 to DNA, we then studied the reaction of L2,3 and L6 with ODN1a (Scheme 3.4).
Bisphosphines L2[19] and L3[50] are derivatives of the well-known ligands PYRPHOS
and BINAP, respectively, extensively used in organometallic catalysis,[13,
18, 21, 22, 40]
3.1 Incorporation of Metal Complexes into Nucleic Acids
72
while L6 belongs to the family of PHOX ligands with applications in allylic
substitution, hydrogenation and asymmetric Heck reactions.[57, 63]
Compound L6 was synthesized starting either from commercially available 2(diphenylphosphino)-benzoic acid 2a (Scheme 3.5 A) or 2-iodo-benzoic acid 2b
(Scheme 3.5 B) and L-serine methyl ester hydrochloride 3 (H-L-Ser-OMe), followed by
oxazoline ring closure in the presence of Burgess’s reagent. In the second synthetic
approach, the PPh2-group was introduced by palladium-catalyzed P-C cross coupling
reaction with diphenylphosphine 7.[270] The phosphinooxazoline was isolated as sodium
salt L6, since acidification results in oxazoline ring opening.[271]
Scheme 3.5. Synthesis of phosphinooxazoline L6.
The coupling reactions of L2, L3 and L6 to ODN1a were monitored by reversed-phase
HPLC and proceeded with 98% and 95% conversion for L2 and L6, respectively. The
amounts of oxidized species (mono- and bisoxide for ODN9 and monoxide for
ODN11a) were below 10%. In case of L3, the observed conversion was lower (55%),
most probably due to the limited solubility of the BINAP derivative in the aqueous
reaction mixture (Table 3.3). Oxidation products (mono- and bisoxide of ODN10) were
found to be formed in <7% yield. Conjugates ODN9, ODN10 and ODN11a were
purified and isolated by reversed-phased HPLC. Figure 3.10 A illustrates typical HPLC
chromatograms obtained for the coupling of L2, L3 and L6. The isolated products were
analyzed by mass spectrometry (Table 3.3) in the form of the corresponding phosphine
sulfide analogues. All other byproducts generated by full or partial oxidation were also
isolated and characterized: phosphine bisoxides (for L2 and L3) and phosphine
3.1 Incorporation of Metal Complexes into Nucleic Acids
73
monoxides (for L6; and for L2 and L3 characterized as monoxide-monosulfide).
Table 3.3. Isolated yields and MALDI-TOF MS analysis of ODN9-11a.
ODN(O)n[a]
ODN(O)(S)
ODN(S)n[a]
[b]
Entry
Coupled Isolated yield
m/z
ligand
[%]
calcd
obsd
calcd
obsd
calcd
obsd
74
6489
6491
6508
6510
6524
6528
ODN9
L2
38
6605
6610
6621
6626
6637
6637
ODN10
L3
78
6297
6296
6313
6314
ODN11a
L6
[a] n = 2 for L2 and L3, and n = 1 for L6. [b] ODN9 and ODN11a detected in negative mode ([M-H]-),
ODN10 in positive mode ([M+H]+). DNA sequence: 5’-GC AGT GAA GGC* TGA GCT CC-3’, C* =
N4-PHOX-appended 2’-deoxycytidine.
While MALDI mass spectrometry was found unsuitable for the direct detection of
phosphine conjugates, ESI-MS analysis gave in the only one case attempted (ODN9)
the main peak corresponding to the non-oxidized phosphine, indicating that this
technique might be suitable for the characterization of phosphine-DNA species without
the need of sulfur treatment (Figure 3.11).
DNA-phosphine conjugates ODN6-11a are generally air sensitive and must be
manipulated under oxygen-free conditions, as commonly done with phosphine ligands.
Nevertheless, the observed rates of oxidation are notably different, depending on the
attached ligand. The relative stabilities and the conditions under which these conjugates
could be handled were investigated by an HPLC assay. Oligonucleotides ODN6,
ODN9, ODN10 and ODN11a were isolated by HPLC, the eluates stored at room
temperature for 1 h under argon, and then re-analyzed by HPLC. This allowed
measuring the extent of oxidation caused by oxygen dissolved in the HPLC solvents
from the very moment after their isolation. Oligomers ODN6 and ODN9 showed
disappointingly low stabilities, yielding large amounts of fully oxidized species (60 and
90%, respectively). In contrast, ODN10 and ODN11a were found to be stable under
these conditions, giving <10% of oxidized product in case of ODN10 and no detectable
amount for ODN11a (Figure 3.10 B). These results demonstrate that the stability of the
DNA-appended BINAP and PHOX conjugates, ODN10 and ODN11a, respectively, is
high enough to allow manipulations of such conjugates even under suboptimal
conditions, e.g., outside a glove box.
3.1 Incorporation of Metal Complexes into Nucleic Acids
74
Figure 3.10. HPLC chromatograms of the phosphine and phosphinoxazoline derivatized DNA conjugates
ODN6, ODN9,10 and ODN11a. A) Crude product after chloroform extraction and ethanol precipitation.
■ Trace impurity from the starting material ODN1a. B) HPLC purified product, after 1 h at r.t.
3.1 Incorporation of Metal Complexes into Nucleic Acids
75
Figure 3.11. Mass spectrometry analysis of ODN9. A) MALDI-TOF spectrum of ODN9(O)2 (HPLC
isolated ODN9. B) MALDI-TOF spectrum of ODN9(S)2. C) ESI MS spectrum (part of the deconvoluted
spectrum) of ODN9 (measured: 6457.28, calculated: 6457.27).
The stability of DNA-tethered BINAP and PHOX ligands makes them attractive
precursors for the development of metal-containing oligonucleotides. Moreover, we
chose to generate a series of DNA-based phosphinooxazoline ligands with variable
spacer length to ensure a reasonably large spectrum of DNA-transition metal
interactions.
Beside the four-carbon linker (ODN1a) employed so far for DNA ligand attachment,
two more flexible linkers were chosen, namely a short two-carbon tether (ODN1b) and
a 13-atom spacer (ODN1c), respectively (Table 3.1). Amino-modified oligonucleotides
ODN1b and ODN1c were then reacted with the phosphinooxazoline derivative L6
(Scheme 3.4), yielding 58% and 75% ODN11b and ODN11c respectively, and below
10% oxidized products (Figure 3.12). In case of ODN11b, the observed conversion was
lower (71%) compared to ODN11c (94%), likely due certain steric hindrance caused by
the shorter linker used for functionalization.
3.1 Incorporation of Metal Complexes into Nucleic Acids
76
Figure 3.12. HPLC chromatogram of ODN11b and ODN11c pre-purified by chloroform extraction and
ethanol precipitation to remove the excess of coupling reagents. ■ Trace impurity from the starting
material ODN1b and ODN1c respectively. ● Trace amounts of unreacted starting material (tR = 8.9 min)
and degradation products (for HPLC conditions see Materials and Methods, Chapter 5.5.1)
It is reasonable to think that the most useful tethers for attaching metal complexes are
relatively short since a short tether permits more stereoselective control of the DNA
scaffold. In our amino-modified oligonucleotides strategy, the two-carbon tether was
the shortest tether that could be introduced to provide reactive amine functional groups
for subsequent derivatization. A long tether might also prove beneficial for achieving
enough flexibility to reach catalytic pockets in applications involving DNA/RNA (pool)
hybrids (Figure 3.2) and in vitro selection. However, for a rational design approach, it
would be rather difficult to predict the influence of long tethers on the transfer of
chirality from the DNA to the metal centre. For example, long spacers may provide too
little interaction between the DNA and the transition metal complex appended onto the
tether. In addition, a long tether may introduce too much flexibility into the DNA,
thereby precluding the goal of a structural constraint.
3.1.2.5 Duplex Stability of Bisphosphine-Tethered DNA
Although it has been generally assumed that further derivatization of N4-alkyl-dC
residues does not interfere with formation of B-form duplex DNA, several factors, such
as steric effects, may still influence the relative stability of the modified
oligonucleotides. For example, the presence of large aromatic ligands in the major
groove might cause a destabilization effect by repelling water molecules and bound
3.1 Incorporation of Metal Complexes into Nucleic Acids
77
small cations. However, the introduction of a positively charged metal complex is
expected to cancel this effect and to electrostatically stabilize the duplex.[234]
The PYRPHOS-appended DNA ODN9 was tested in its ability to form duplexes with
complementary cDNA1 and cRNA1, as it would be later required according to its use in
the selection scheme (Figure 3.1).
Figure 3.13. Native PAGE showing the formation of duplexes between PYRPHOS-appended DNA
ODN1a and complementary cDNA1 or cRNA1. Asterisks indicate the presence of a radioactive 32P label
at the 5’-end of the oligonucleotides. From left to right: Lane 1 - ODN1a*; lane 2 - ODN1a*/cDNA1;
lane 3 - ODN1a*/cRNA1; lane 4 - cRNA1*; lane 5 - ODN1a/cRNA1*; lane 6 - ODN9(O)2*; lane 7 ODN9(O)2*/cDNA1; lane 8 - ODN9(O)2*/cRNA1; lane 9 - ODN9(S)2*; lane 10 - ODN9(S)2*/cDNA1;
lane 11 - ODN9(S)2*/cRNA1; lane 12 - ODN9(O)(S)*; lane 13 - ODN9(O)(S)*/cDNA1; lane 14 ODN9(O)(S)*/cRNA1; lane 15 - ODN1a*; lane 16 - cDNA1*; lane 17 - ODN1a*/cDNA1.
Hybridization buffer: 100 mM HEPES pH 7.5, 200 mM NaCl, 1 mM EDTA.
The hybridization experiments were conducted with only one of the two
oligonucleotides 5’-32P-labelled, using standard denaturation/reannealing cycles (see
Materials and Methods, Chapter 5.8.3). As reference, the amino-modified ODN1a was
used. Due to the fact that all manipulations had to be carried out in air, the bisoxide,
bisulfide
and
monoxide-monosulfide
analogues,
ODN9(O)2,
ODN9(S)2,
and
ODN9(O)(S) respectively, were used as replacement for ODN9 to avoid mixtures of
non-oxidized, partially and fully oxidized products. The formation of duplexes was
3.2 Organometallic Transformations in Water
78
controlled by native 16% PAGE (Figure 3.13).
The results of these experiments clearly show that attachment of the ligand does not
prevent the formation of duplexes. In all cases the hybridization appears to be
quantitative under the conditions employed.
3.2
Organometallic Transformations in Water
Together with synthetic approaches for the incorporation of chelating functionalities
and, subsequently, of transition metal centres at well-defined positions in DNA or RNA
sequences, appropriate model reactions for the use of nucleic acid-based ligands are also
required.
Most organometallic transformations have been traditionally conducted in polar or
nonpolar organic solvents. Water has been less used either because of sensitivity of
organometallic catalysts to water, or because most organic compounds are not easily
soluble in water. The observations of reaction rate enhancement as well as the
possibility of achieving new selectivities in water led to an increased interest in
exploring the properties of water in organometallic-catalyzed reactions.[272-274] A
number of homogeneous transition metal-catalyzed reactions, such as rhodium(I)catalyzed hydrogenation, palladium(II)-catalyzed amination, rhodium(I)-catalyzed 1,4addition, copper(II)-catalyzed Diels-Alder cycloaddition, and aldol condensation have
been successfully carried out in aqueous mixtures, and mainly in the presence of water
soluble phosphorus ligands.[274] However, a clear understanding of the nature of the
interactions occurring in aqueous media has still to be worked out.
Scheme 3.6. A) Rhodium(I)-catalyzed 1,4-conjugate addition of phenylboronic acid to α,β-unsaturated
ketones. B) Iridium(I)-catalyzed allylic amination.
3.2 Organometallic Transformations in Water
79
The reactions we selected as model systems for the development of nucleic acid based
hybrid catalyst were: rhodium(I)-catalyzed 1,4-addition of boronic acids to α,βunsaturated carbonyl compounds and iridium(I)-catalyzed allylic amination (Scheme
3.6), because of their compatibility with the aqueous environment. Preliminary studies
were devised using these model reactions to study the catalytic competence of the new
DNA-phosphine and phosphinooxazoline ligands ODN9-11.
3.2.1
Phosphine- and Phosphinooxazoline-Metal Complexes
Phosphine and phosphinooxazoline-metal complexes were prepared using various
transition metal precursors and standard synthetic procedures. Preliminary assays
concerning the synthesis and behaviour of such complexes in the presence of water
were conducted to demonstrate whether such catalysts could be formed and were stable
enough under these conditions. In addition, maintaining anaerobic conditions during
handling air-sensitive phosphine ligands and metal complexes thereof in small reaction
volumes (as required by the use of nucleic acids), in an adequate reaction setup and
outside a glove box is not trivial. The standard Schlenk techniques and degassing of
reaction mixtures were found sufficient to avoid air contamination even in reduced scale
reactions (e.g., <100 μL).
The first control experiments involved the monophosphine ligand L1. Reaction of
palladium(II) PdCl2(PhCN)2 with L1 in neat acetonitrile led to the formation of complex
PdCl2(L1)2 (8) (Scheme 3.7), showing a singlet in the 31P NMR spectrum (DMSO-d6, δ
= 24.6 ppm). Also, a single complex (9) was obtained when L1 was reacted with
[Pt(cod)Cl]2 in neat dichloromethane (Scheme 3.7), showing a singlet in the
31
P NMR
spectrum at δ = 14.3 ppm (CD3OD), with platinum-phosphine coupling (satellite JPt,P =
1852 Hz) characteristic of a trans complex.[201] When the same complex was formed in
aqueous mixture 93:7 acetonitrile/H2O - likely suitable conditions to prepare DNAmetal complexes - the 31P NMR spectrum showed a singlet at δ = 18.4 ppm (DMSO-d6)
indicative for phosphorus-platinum coordination.
The palladium and platinum complexes were formed quantitatively and no oxidized
products were observed. Moreover, the interaction between the phosphine L1 and the
solvent appeared to remain unchanged in water. The platinum complex 9 could be
3.2 Organometallic Transformations in Water
80
easily prepared in water and was stable under the conditions employed.
Scheme 3.7. Synthesis of palladium(II)- and platinum(II)-L1 complexes.
The PYRPHOS ligand L2 was first treated with one equivalent of PdCl2(PhCN)2 or half
equivalent of [Pt(cod)Cl]2 per bisphosphine unit, at ambient temperature in neat
dichloromethane to afford palladium(II) or platinum(II) complexes 10 and 11 (Scheme
3.8). The complex formation was monitored by 31P NMR spectroscopy. The
31
P NMR
spectra demonstrated the absence of unreacted PYRPHOS after the complete
conversion and displayed a single coordination-shifted resonance at δ = 42.6 ppm
(DMSO-d6) and 26.4 ppm (CD3OD) for 10 and 11 repectively (δ = -11.0 to -12.2 for the
non-metallated phosphine.[18] For complex 11 the platinum-phosphorus coupling
constant JPt,P = 1163 Hz corresponded to the expected range of such phosphine-platinum
compounds. The absence of 1,5-cyclooctadiene and benzonitrile resonances in the 1H
NMR spectrum indicated that in both isolated complexes the coordination sites were
likely occupied by chlorine atoms.
The binding capability of PYRPHOS ligand L2 was also investigated in aqueous
solvent. Stirring L2 with [Rh(cod)Cl]2 (one equivalent bisphosphine per Rh) at room
temperature, in 40% water with acetonitrile cosolvent, gave the metallated complex 12
(Scheme 3.8). The complete reaction was confirmed by the
31
P NMR spectrum. The
resonance of the free phosphine unit was shifted to lower field (ABX system[18];
CD3OD, δA = 38.2 ppm, δB = 36.8 ppm) due to the metal complexation. The signal
splitting was attributed to Rh(I = 1/2)/P coupling. The JRh,P constants of 150 and 153 Hz
are within the expected range for such phosphinerhodium compounds.
To further investigate the stability of rhodium(I)-PYRPHOS complex towards air and
3.2 Organometallic Transformations in Water
81
aqueous media, complex 13 was prepared using Rh(nbd)BF4 as metal precursor. This
precursor was generated from [Rh(nbd)Cl]2 and AgBF4 in acetone (Scheme 3.8)
followed
by
removal
of
the
precipitated
[Rh(L2)(nbd)]+BF4- was analyzed by 1H and
31
AgCl.
The
resulting
complex
P NMR. The NMR spectra confirmed
the desired metal complex (Acetone-d6, P resonances: δA = 36.9 ppm, δB = 35.6 ppm).
Scheme 3.8. Synthesis of palladium(II), platinum(II) and rhodium(I)-L2 complexes.
The complex 13 was immediately dissolved in acetonitrile and analyzed by ESI mass
spectrometry. In this case, the base peak corresponds to a complex without
norbornadiene ligand (Figure 3.14 A). Instead, two solvent molecules seem to
coordinate the metal. An additional peak could be assigned to oxidized species. Other
signals could not be attributed so far. A certain level of decomposition of the Rh(I)
complex 13 was observed after one week storage in acetonitrile, at room temperature
and air. However, under these conditions, the main compound still formed the most
abundant signal (Figure 3.14 B).
3.2 Organometallic Transformations in Water
82
Figure 3.14. ESI MS spectrum of complex 13: A) in acetonitrile, immediately sprayed; B) sprayed after
one week storage in acetonitrile at ambient temperature and air.
To verify the hypothesis of solvent molecules as ligands instead of norbornadiene, two
other solvents were tested. Firstly, the complex was dissolved in acetone. After storage
for one week at room temperature and air, the solution was analyzed by ESI MS (Figure
3.15 A). In this case, the base peak is due to a Rh-complex similar to the one observed
in acetonitrile: Rh(L2)(acetone)2+ (m/z: calcd. 772.18, obsd. 772.17). Three other
signals could be readily assigned: Rh(L2)(acetone)+ (m/z: calcd. 714.14, obsd. 714.13),
Rh(L2)+ (m/z: calcd. 656.10, obsd. 656.09) and Rh(L2)(water)+ (m/z calcd. 674.11,
obsd. 674.10). Unlike the spectrum recorded in acetonitrile (inhomogeneous solution),
the impurities containing the oxidized species showed higher intensities. When the Rhcomplex was dissolved in methanol and immediately sprayed, the results were similar to
the acetone solution, the coordinated solvent molecules being now methanol (Figure
3.15 B).
3.2 Organometallic Transformations in Water
83
Figure 3.15. ESI MS spectrum of complex 12: A) sprayed after one week storage in acetone at ambient
temperature and air; B) in methanol, immediately sprayed.
Finally, the CID-MS/MS of the isolated ion m/z 772.17 sprayed from acetone was
recorded to give additional evidence for the formation of Rh(I)-PYRPHOS complex,
and, implicitly, Rh(L2)(acetone)2+ species. The observed pattern perfectly matched the
expectations and two acetone molecules were sequentially lost (Figure 3.16).
Figure 3.16. CID-MS/MS spectrum of Rh(L2)(acetone)2+.
In conclusion, the rhodium(I)-PYRPHOS complex can be analyzed by ESI-MS in
acetone and in acetonitrile. The proposed cation structure Rh(L2)(nbd)+could not be
3.2 Organometallic Transformations in Water
84
confirmed. No norbornadiene ligand was found attached to the rhodium centre. Instead,
up to two solvent molecules, either acetone or acetonitrile or methanol or water were
coordinated. Norbornadiene as ligand was observed only in complexes with partially
oxidized L2. These results could indicate that L2 binds to rhodium not only via the
phosphorus atoms, but also with one or two carbonyl groups. In all cases, significant
amounts of degradation were observed after one week in solution at room temperature
and in air, but in each case, the original compound was stable enough to still be
detectable without difficulties.
It was also interesting to investigate the stability of phosphinooxazoline-based
complexes against oxidation. Control experiments in organic solvent were initially
carried out. For solubility reasons, we chose to evaluate metallation of
phosphinooxazoline L5 which contains a methylester functionality (Scheme 3.9),
instead of the L6 carboxylate analogue. Reaction of the complex [Rh(nbd)Cl]2 with
AgBF4 in acetone proceeded with cleavage of the chloride bridges, followed by
exchange of the chloride ion with BF4- to give the monomeric precursor Rh(nbd)BF4.
Reaction with one equivalent L5 yielded the compound [Rh(L5)(nbd)]+BF4- 14
(Scheme 3.9) as confirmed by ESI mass spectrometry (m/z: calcd. 584.09, obsd.
584.09), and no additional (oxidation) byproducts were observed. Also according to 31P
NMR of the isolated complex, only one product was formed. The coordination of the
phosphorus donor atom to the rhodium was evident due to the characteristic downfield
shift of the phosphorus resonance δ = 31.9 compared to -4.8 ppm (CDCl3) of the free
L5. The iridium(I) complex 15 was prepared in a similar way starting from [Ir(cod)Cl]2
(Scheme 3.9). Also in this case, the desired complex [Ir(L5)(cod)]+BF4- was obtained as
demonstrated by
31
P NMR anlysis (CDCl3 δ = 14.8). Both complexes were stable
against air and moisture and could be easily handled in the laboratory atmosphere. The
isolated solid products are fine powders, and our attempts to grow crystals have been
fruitless to date.
3.2 Organometallic Transformations in Water
85
Scheme 3.9. Synthesis of rhodium(I) and iridium(I)-L4/L5 complexes.
We next attempted the formation of iridium(I) complexes in aqueous media, using the
well-described, commercially available compound L4 in comparison with our ligand L5
(Scheme 3.9). Control experiments were also performed in neat organic solvent. Both
ligands were dissolved in either neat dioxane or 3:7 dioxane/water. After addition of
[Ir(cod]Cl]2 (0.1 mM final concentration of Ir-complex, as later used in catalytic
attempts), the metallation was complete within seconds, as was evident from the change
of color from orange-yellow to dark red, and monitored by
31
P NMR spectroscopy. In
most of the cases the 31P NMR spectrum indicated the presence of single species. The
observed phosphorus resonances were in good agreement with those reported in
3.2 Organometallic Transformations in Water
86
literature for similar Ir(I)-PHOX complexes in neat organic solvents:[275, 276] 10.3 ppm
(dioxane, 10% CDCl3) and 15.7, 15.6 ppm (3:7 dioxane/water, 10% D2O) for L4, and
8.8 ppm (dioxane, 10% CDCl3) and 15.2 ppm (3:7 dioxane/water, 10% D2O) for L5,
respectively (Figure 3.17).
15.73
15.64
︶
-6.97
10.27
B
A
︶
4
L
r
I
4
L
r
I
4
L
50
0
50
ppm (t1)
D
C
︶
8.83
︶
5
L
r
I
5
L
r
I
ppm (t1)
50
0
15.16
ppm (t1)
0
50
ppm (t1)
0
31
Figure 3.17. P NMR spectrums of iridium(I)-phosphinooxazoline complexes showing phosphorus
metallation: A) Ir-L4 in dioxane, B) Ir-L4 in 3:7 dioxane/water, C) Ir-L5 in dioxane, D) Ir-L5 in 3:7
dioxane/water.
Phosphinooxazoline complexes of Ir(I) have not been so far characterized in the
presence of water. Moreover, catalytic applications of such complexes in aqueous
environment have not been reported until now. Our preliminary analyses provide useful
information about the metallation process in water, although additional work has to be
done to elucidate the role of the solvent in coordination and interaction with the metal
centre.
These
studies
concerning
the
synthesis
and
stability
of
phosphine
and
phosphinooxazoline complexes in aqueous media supports the hypothesis that such
complexes are reasonably stable and suitable for organometallic asymmetric reactions
performed in the presence of water.
3.2 Organometallic Transformations in Water
3.2.2
87
Rhodium(I)-Catalyzed 1,4-Addition
In 1997 Miyaura reported the first non-asymmetric 1,4-addition of aryl- and alkenylboronic acids to α,β-unsaturated ketones using a phosphine-Rh(I) catalyst in 15%
aqueous mixture.[44] In 2003, under similar reaction conditions (10:1 dioxane / water),
Feringa demonstrated that monodentate phosphoramidites could be also used as ligands
in the rhodium-catalyzed asymmetric conjugate addition of boronic acids, offering the
advantage of straightforward fine-tuning the ligand for selectivity improvement.[252, 277]
In most of 1,4-addition studies, the amount of water used in combination with the
organic cosolvent (often dioxane) was in the range of 9-15%, while with an
immobilized Rh-BINAP complex (Figure 1.3), the reaction could be performed in pure
water.[50] In addition, in most of the cases high temperatures (50-100°C) and catalyst
loading (1-3%) were used.
Miyaura and coworkers observed great rate acceleration in conjugate addition of
phenylboronic acid to cyclohexenone when using a Rh(I)-BINAP complex in the
presence of inorganic or organic bases.[278] This discovery allowed them to perform the
reaction at room temperature with quantitative conversion while, in the absence of base,
only trace amounts of product were obtained. The best performing base was found to be
triethylamine which could be used in 0.1-1.0 equiv in respect to the substrate. Recently,
Piarulli employed similar conditions (1.0 equiv KOH instead of triethylamine) with
combinations of biphenolic phosphoramidite and phosphite ligands, and obtained high
yields and stereoselectivities.[89]
Cyclic enones are the most commonly investigated substrates due to their high
reactivity. Phenylboronic acid 16 and 2-cyclohexen-1-one 17 are commercially
available. Reference product 3-phenyl-1-cyclohexanone 18 was prepared according to
Scheme 3.10, in racemic form, and the analytical methods for monitoring the reaction
and determining the enantioselectivity have been established (Figure 3.18).
Scheme 3.10. Synthesis of 3-phenyl-1-cyclohexanone 18.
3.2 Organometallic Transformations in Water
88
Figure 3.18. A) Reversed-phase HPLC analysis of 1,4-addition of phenylboronic acid 16 to 2cyclohexen-1-one 17 (Elution: 30% water, 70% acetonitrile; tR(17) = 3.0 min, tR(16) = 3.3 min, tR(18) =
5.1 min). B) Calibration curve with 3-phenyl-1-cyclohexanone product 18.
Initial studies were focused on the optimization of reaction conditions, including
temperature, solvent, phosphorus ligands and Rh-complexes thereof (Figure 3.19), for
carrying out addition of phenylboronic acid 16 to 2-cyclohexen-1-one 17 to obtain 3phenyl-1-cyclohexanone 18.
Figure 3.19. Phosphorus-based ligands and Rh(I) complexes screened for activity in 1,4-addition
reaction.
The phosphoramidites P1 and P2 were screened initially, under standard reaction
conditions,[252] in 10:1 dioxane/water, using 1.5 mol% [Rh(cod)Cl]2 and 7.5 mol% of
ligands (Rh/L= 1:2.5), without basic additives. The reaction was also performed under
increased concentration of water at room temperature for 24-72 hours. All experiments
3.2 Organometallic Transformations in Water
89
were carried out under argon atmosphere without pre-degassing the solvents and
reagents. A few selected results are presented in Table 3.4.
Table 3.4. Rh-catalyzed conjugate addition of arylboronic acid 16 to 2-cyclohexen-1-one 17 with
phosphoramidite P1 and P2 ligands.
Entry
Ligand
Dioxane/water
Temp.
Time
Conversion[c]
[%]
[°C]
[h]
1
10:1
60
2
80%
P1
2
1:5
60
24
70%
P1
3
1:10
rt
24
<5%
4
1:10
rt
72
20%
P1
5
1:10
rt
24
<5%
P2
1:10
rt
72
<5%
6[b]
P1
[a] Reaction conditions: 0.08 mmol 17, 0.25 mmol 16 (3.0 equiv - excess due to competitive hydrolysis),
3 mol% catalyst loading, 2.2 mL reaction volume. [b] 0.5 mol% [Rh] catalyst loading. [c] The
conversions were estimated by 1H NMR analysis with i-propanol as internal standard (entries 1-2 and 4-6)
and by thin-layer chromatography (entry 3).
In general, the catalysts were efficient when the reaction was carried out at 60°C in 10:1
or 1:5 dioxane/water (70-80% conversion) (Table 3.4, entries 1-2), while at room
temperature and 1:10 dioxane/water the conversion dropped dramatically even after
long reaction times (Table 3.4, entries 4-5). Moreover, the catalyst precursor
[Rh(cod)Cl]2 was found almost inactive (Table 3.4, entry 3), although, under the
conditions employed by Miyaura et al., the same catalyst showed high reactivity.[278] In
a parallel test reaction, the Rh catalyst loading was reduced to 0.5 mol%, and after 72
hours reaction time only trace amount of product was formed (Table 3.4, entry 6).
Surprisingly, biphenolic phosphoramidite P2 seemed to be less active than the P1
ligand, although structurally related ligands have been reported as efficient systems in
similar transformations, albeit in 10:1 dioxane/water.[89] The reduced yields could be
explained by the fact that the Rh-phosphoramidite catalysts were generated in situ in
1:10 dioxane/water mixtures, where ligand hydrolysis might compete in a higher extent
with metal complex formation. In addition, under these conditions, the system became
heterogeneous, making the results unreproducible. However, in the absence of the base,
high temperature appeared to be absolutely necessary for achieving good conversions.
Prolonging the reaction time did not improve the yields due to the competitive
hydrolytic deboronation of arylboronic acids with water (Table 3.4, entry 4 versus 3 and
5).
For homogeneity reasons we chose to attempt 1,4-addition with water-soluble
phosphine ligands.[279] Rhodium complex [Rh(cod)Cl]2
was combined with the
3.2 Organometallic Transformations in Water
90
commercially available phosphine ligand TPPDS L7 (0.38 mol % catalyst) 1:10
dioxane/water mixture and reacted with 17, 2.5 equivalents of phenylboronic acid 16,
and 2.1 equivalents of K2CO3 at 37°C for 48 hours. To prevent hydrolytic deboronation
of 16 to unreactive benzene, a phase transfer reagent, i.e. sodiumdodecyl sulfate (SDS,
0.5 equivalent), was added to the reaction mixture.[279] As alternative to aqueous
solution of K2CO3, Tris buffer (20 mM, pH 8.0) was used instead. The same reaction
conditions were also employed only with [Rh(cod)Cl]2 precursor. In parallel, 1,4addition reaction was conducted with monophosphine L1 ligand, under reduced
concentration of organic solvent, i.e. 1:1 dioxane/water and 1:1 methanol/water, at 50°
and with 3.85 mol% catalyst loading. The results are illustrated in Table 3.5.
Table 3.5. Rh-catalyzed conjugate addition of arylboronic acid 16 to 2-cyclohexen-1-one 17 with
monophosphines L1 and L7.[a]
Entry Ligand
Solvent
Base
Phase transfer
Temp.
Time
Conversion[c]
reagent
[°C]
[h]
[%]
1
37
48
<1
dioxane/water 1:10
L7
2
SDS
37
48
<1
L7
SDS
37
48
10
3
K2CO3
L7
4
Tris
SDS
37
48
41
5
Tris
SDS
37
48
7
L7
dioxane/water 1:1
SDS
50
19
3
6[b]
L1
methanol/water 1:1
SDS
50
19
1
7[b]
L1
[a] Reaction conditions: 0.52 mmol 17, 1.3 mmol 16 (2.5 equiv), 0.38 mol% catalyst loading, 2.2 mL
reaction volume. [b] 3.85 mol% [Rh] catalyst loading. [c] The conversions were determined by reversedphase HPLC (elution: 30% water, 70% acetonitrile), using the calibration curve showed in Figure 3.18 B.
The best results were obtained with [Rh(cod)Cl]2 which gave 18 in 41% yield (Table
3.5, entry 4). The fairly high activity of the Rh catalyst in 91% water might be explained
by the basic conditions provided by K2CO3 and also by reduced competivive hydrolytic
deboronation due to the presence of SDS. However these additives apperead to be
inefficient in combination with Rh-phosphine complexes. Although the Rh-TPPDS
complex was highly reactive in 1,4-addition reactions in neat water,[279] in our case, the
addition of organic cosolvent (9-50%) with the sulfonated phosphine ligand L7 or with
triphenylphosphine L1 resulted in almost no reaction after 48 hours (Table 3.5, entries 4
and 5-7).
Literature data on Rh(I)-catalyzed addition of phenylboronic acids to olefins in the
presence of sulfonated ligands or triphenylphoshine showed that the use of cosolvents
or neat organic solvents with such ligands resulted in very low conversions.[279] In these
systems, the reactivity could be restored by changing to 100% aqueous environment.
3.2 Organometallic Transformations in Water
91
However, such conditions would not be appropriate for our system due to the poor
solubility of the product and substrates.
Finally, we attempted 1,4-addition reactions with Rh-bisphosphine complexes 13, 14,
and [Rh(BINAP)(nbd)]+BF4-[250] (Figure 3.19) at 3 mol% catalysts loading, in 6:1
dioxane/water mixture. Using either [Rh(cod)Cl]2 precursor or the synthesized
[Rh(BINAP)(nbd)]+BF4- complex, the observation of high activity in the presence of
trietylamine was again confirmed by our experiments. Good conversions, 77 and 80%
respectively, were obtained after 6 hours in the presence of one equivalent of
triethylamine, while in its absence the product was formed only in 47% (Table 3.6,
entries 2 and 6 versus 1). The [Rh(cod)Cl]2 precursor showed again high activity
without additional ligand (Table 3.6, entries 1-2).
Table 3.6. Effect of oxygen-free conditions[a], ligand and Rh-complex on the 1,4-addition of
phenylboronic acid 16 to 2-cyclohexen-1-one 17.[b]
Entry
Rh(I)
Isolated
Base[c]
Temp Time Conv[d]
Solvent
L
precursor
Rh(I)-complex
[h]
diox/H2O [°C]
[%]
1
[Rh(cod]Cl]2
6:1
37
16
47
2
[Rh(cod]Cl]2
NaHCO3
37
16
47
3
TEA
rt
6
0
13
rt
6
0
4
[Rh(L8)(nbd)]+BF4TEA
rt
6
0
5
[Rh(L8)(nbd)]+BF4TEA
6:1
rt
6
80
6
[Rh(L8)(nbd)]+BF47
[Rh(nbd]Cl]2
TEA
rt
19
12
L8
TEA
3:7
rt
4
0
8[e]
14
9[e]
[Rh(C2H4]Cl]2 L5
TEA
rt
4
0
[a] Reactions 1-5 were carried out under argon atmosphere with undegassed solvents and reagents.
Reactions 6-9 were conducted under oxygen-free conditions. [b] Reaction conditions: 1.0 mmol 17, 1.5
mmol 16 (1.5 equiv), 3 mol% catalyst loading, 3.5 mL reaction volume, unless otherwise stated. [c] 0.1
equiv NaHCO3 or 1.0 equiv TEA. [d] Determined by reversed-phase HPLC analysis (elution: 30% water,
70% acetonitrile, tR(18) = 5.1 min) with the calibration curve showed in Figure 3.18 B, unless otherwise
stated. [e] Reaction conditions: 41.0 μmol 17, 61.5 μmol 16 (1.5 equiv), 2.4 mol% catalyst loading, 1.0
mL reaction volume. Conversion determined by reversed-phase HPLC analysis (elution: 50% water, 50%
acetonitrile, tR(18) = 11.0 min) in the presence of internal standard.[250]
It was also found that the isolated [Rh(BINAP)(nbd)]+BF4- complex was superior to the
rhodium complex in situ generated by mixing [Rh(nbd)Cl]2 with 1.5 equivalents of (S)BINAP L8 per rhodium, in 6:1 dioxane/water and 1.0 equiv TEA, at room temperature,
which gave only 12% conversion (Table 3.6, entry 6 versus 7). It was also possible to
show that it was certainly necessary to run these reactions in an absolutely oxygen-free
environment (Table 3.6, entry 5 versus 6). Unlike the [Rh(cod)Cl]2 precursor (Table 3.6,
entries 1 and 2), the Rh-PYRPHOS (13) and -BINAP complexes were completely
inactive in the presence of oxygen, independently of TEA additive (Table 3.6, entries 3-
3.2 Organometallic Transformations in Water
92
5). Some of the intermediates formed during the catalytic cycle[280] must therefore be
very
sensitive
to
+
oxidation,
since
[Rh(PYRPHOS)(nbd)]+BF4-
(13)
and
-
[Rh(BINAP)(nbd)] BF4 complexes have been shown to remain reasonably unchanged
in solution in the presence of air and at room temperature for at least one week (Figure
3.14 B).[250]
Other bidentate ligands, such as PHOX ligand L5, did not catalyze the 1,4-addition, no
matter whether its isolated Rh(I)-complex 14 or in situ generated complex from
[Rh(C2H4)Cl]2 and 1.1 equivalents L5 per rhodium were used under similar conditions
(1.0 equiv TEA, 2.4 mol% catalyst loading), albeit in 70% water (Table 3.6, entries 8
and 9).
The synthesized Rh-BINAP complex afforded so far the best yield of the addition
product under oxygen-free conditions, at 3 mol% catalyst loading (10 mM) and with
TEA additive, while in the presence of free BINAP L8 low conversion was observed.
These results represent a solid starting point for studies involving DNA-BINAP
conjugates, such as ODN10 synthesized from amino-DNA ODN1a and the BINAPcarboxylic acid derivative L3 (Scheme 3.4). Based on all informations, it seems that the
best way of preparing a DNA-appended Rh(I)-BINAP complex will consist in the pretreatment of [Rh(cod)Cl]2 or [Rh(nbd)Cl]2 precursor with AgBF4 and then addition to
the DNA-phosphine ligand. The concentration of dioxane could not be reduced lower
than 30% due to the limited solubility of reagents. However, this value seems to be well
tolerated by double-stranded nucleic acids, as demonstrated in our studies on DNA
duplex stability in the presence of organic solvents (Chapter 3.1.2.2, Figure 3.7).
3.2.3
Iridium(I)-Catalyzed Allylic Amination
Transition metal catalyzed asymmetric allylic substitutions are among the most
important carbon-carbon and carbon-heteroatom bond forming reactions in organic
synthesis.[281] Two classes of allylic compounds have been enantioselectively
transformed with chiral catalysts: (1) symmetrically substituted racemic and (2)
monosubstituted linear or branched (racemic) allylic substrates (Scheme 3.11).
While Pd(0)-catalysts have been typically used in the former case, the monosubstituted
allylic substrates were less often employed with such systems because of
3.2 Organometallic Transformations in Water
93
regioselectivity in favour of linear achiral products. The regioselectivity control of
allylic substitution reactions is mainly governed by the choice of the transition metal
ion. For example, with Pd(0)-catalysts, linear products were generally produced
(Scheme 3.11 A, while Mo-[72, 282-284] or W-based catalysts[285] yielded chiral branched
products from monosubstituted linear substrates, giving therefore asymmetric induction
(Scheme 3.11 B). Rh,[286] Fe,[287] or Ru[288,
289]
complexes were commonly used to
catalyze substitutions of enantiomerically enriched branched substrates, yielding
branched products with retention of configuration and a high degree of conservation of
enantiomeric excess (“memory effect”) (Scheme 3.11 C).
Scheme 3.11. Transition-metal catalyzed allylic substitution of A) symmetrically substituted, B)
monosubstituted linear and C) monosubtituted branched substrates.
In 1997, Takeuchi reported the first use of Ir-catalysts in allylic substitution, combining
[Ir(cod)Cl]2
precursor
with
triphenylphosphite
ligand
to
achieve
excellent
regioselectivities in favour of the branched product from linear allylic substrates.[290] In
the same year, Helmchen reported the first asymmetric version of allylic substitution
using chiral phosphinooxazoline ligands.[67] Achiral linear aryl acetates were
transformed into branched chiral products (Scheme 3.11 B), with high regioselectivity
and enantiopurity. However, the reaction was slower in comparison to the reaction
catalyzed by the [Ir(cod)Cl]2/P(OPh)3 system or even [Ir(cod)Cl]2 precursor alone.[76,
3.2 Organometallic Transformations in Water
291]
94
Since then, Ir-catalysts have received much attention in asymmetric allylic
substitution.
A large number of catalytic systems consisting of Ir precursors and chiral monodentate
phosphoramidite
[292-297]
or achiral phosphite[75,
76]
ligands have been developed. In
contrast, further investigations on Ir-PHOX complexes gave disappointing results
especially in transformations involving alkyl-substituted allylic substrates: aminations
were generally slow[72] and interesting results could only be achieved in intramolecular
aminations.[71]
Although not many examples of Ir-PHOX-catalyzed allylic aminations have been
published so far, we chose to study this system with DNA-based catalysts for the
following reasons: (1) these reactions require polar solvents,[76] (2) bidentate PHOX
ligands should reduce competition of nucleic acids donor groups in binding the
transition metal due to their strong chelating properties, (3) the modest catalytic
performance of the Ir-PHOX systems in allylic amination might be enhanced in
combination with nucleic acid properties, as finally aimed by our hybrid catalyst
approach.
Scheme 3.12. Iridium(I)-catalyzed allylic amination of linear 20 and branched 21 phenyl-allyl acetates
with morpholine 22 and glycine ethyl ester 23, using chiral phosphinooxazoline ligands L4 and L5.
3.2 Organometallic Transformations in Water
95
As test reactions for probing the catalytic performance of PHOX ligands L4 and L5,
iridium-catalyzed allylic substitutions of monosubstituted phenyl-allyl acetates 20 and
21 with amine nucleophiles morpholine 22 and glycine ethyl ester hydrochloride 23
were selected (Scheme 3.12).
When monosubstituted allyl substrates (e.g. 20 and 21) are used in allylic amination
reactions, the possibility of two regioisomeric products arises: the linear isomer and the
branched isomer. With iridium(I)-catalysts the branched-to-linear-ratio can be shifted to
the formation of the branched isomer as major product, affording the chance to observe
asymmetric induction from achiral substrates. We were also interested in the ability of
Ir-PHOX complexes to effect kinetic resolution[298] of racemic branched substrate, e.g.
compound 21, and to yield enantiomerically enriched product and substrate,
respectively. The general principle of kinetic resolution, namely achievement of partial
or complete resolution by virtue of unequal rates of reaction of the enantiomers in a
racemate with a chiral catalyst, is illustrated in Figure 3.20. The maximum theoretical
yield is 50% due to the consumption of only one enantiomer.
Figure 3.20. Principle of the classic kinetic resolution.
3.2.3.1 Preparation of Allylic Substrates and Products. Analytical
Methods
The branched substrate 21 was prepared by esterification of the commercially available
α-benzylvinyl alcohol 19 (Scheme 3.13). Reference products 24-26 were prepared as
described in Scheme 3.12. Branched product 24 was obtained as racemic mixture from
21 via amination with morpholine 22, in ethanol, using a catalyst in situ generated from
[Rh(cod)Cl]2 and triphenyl phosphite. Achiral linear product 25 was prepared in a
similar way, involving amination of the commercially available cinnamyl acetate 20
3.2 Organometallic Transformations in Water
96
with 22 in dry THF, with Pd(PPh3)4 catalyst. The synthesis of the branched product 26
was performed starting from the racemic allylic acetate 21 and glycine ethyl ester 23
(hydrochloride form), in 1:1 acetonitrile/water with 2.0 equiv NaHCO3, using
[Rh(cod)Cl]2 and trimethyl phosphite ligand.
Scheme 3.13. Synthesis of phenyl-allyl substrate 21 and of amination products 24-26.
In a large number of studies, allylic carbonates are usually the substrates of choice,
while allylic acetates are less often used, owing to their lower reactivity and
selectivity.[292]
Preliminary
attempts
in
our
group
showed
that
branched
methylcarbonate derivatives could undergo isomerisation and cleavage of the carbonate
moiety in aqueous environment.[250] These observations prompted us to use acetate
substrates instead. Since we aim at carrying out allylic aminations in aqueous solvent, at
basic pH, the stability of substrate 21 against saponification under these conditions had
to be initially investigated. Solutions of 21 in 1:1 water/acetonitrile and 0.1 M aqueous
3.2 Organometallic Transformations in Water
97
NaHCO3/acetonitrile were incubated at room temperature for 9 hours and systematically
analyzed by reversed-phase HPLC (Figure 3.21).
Figure 3.21. Stability of allylic acetate 21 in A) 1:1 water/acetonitrile and B) 0.1 M aqueous
NaHCO3/acetonitrile. Gradient: 50% water and 50% acetonitrile, tR(21) = 14.4 min. ■ Impurity.
Figure 3.22. A) Reversed-phase HPLC analysis of branched amination product 24 with naphthalene as
internal standard. B) Gas chromatography analysis of mixture of branched 21 and linear 20 allylic
substrates and branched 24 and linear 25 amination products with dodecane as internal standard. C)
HPLC chiral separation of racemic branched product 24.
3.2 Organometallic Transformations in Water
98
The HPLC chromatograms show similar behaviour of allylic acetate 21 in both water
(Figure 3.21 A) and NaHCO3 mixtures (Figure 3.21 B). Even after long incubation
times, the allylic acetate remained reasonably stable, making it a suitable substrate for
carrying out aminations in aqueous solvent, under basic conditions.
The reference substrates and products were then employed to establish analytical
methods for following the reaction and determining the enantioselectivity: 1) reversedphase HPLC, elution with 50:50 acetonitrile/water, tR = 11.8 min (24), 24.0 min
(naphthalene as internal standard) (Figure 3.22 A); 2) gas chromatography, gradient: 2
min at 150°C, increase to 230°C with 15°C/min; tR = 3.3 min (dodecane as internal
standard), 3.9 min (21), 5.5 min (20), 6.4 min (24) and 8.0 min (25) (Figure 3.22 B); 3)
HPLC chiral separation, elution with n-hexane/i-propanol 99:1, tR(24) = 10.8, 12.4 min
(Figure 3.22 C).
3.2.3.2 Preliminary Results of Catalysis with Ir-PHOX Complexes
The development of allylic substitutions in water has generally received little attention.
The main reason for that is likely the hydrolysis of the electrophile reactant that may
compete with the desired nucleophilic substitution.[299] Only few reports of palladiumcatalyzed allylic substitutions (aminations) in water have been published. The best
results were obtained with a heterogeneous system based on immobilized phosphine or
P,N-chelate ligands.[300-302] Uemura and coworkers reported on a homogeneous version
of palladium-catalyzed allylic substitutions in water, in which a phosphinite-oxazoline
ligand afforded moderate to high yields and good enantiomeric excess (85%).[303]
However, this system proved to be more efficient in acetonitrile alone (92% ee) rather
than in water or water/acetonitrile mixtures.
Our first studies were conducted to determine if Ir-PHOX catalysts impart activity and
stereoselectivity in allylic aminations carried out in aqueous media. Allylic aminations
of phenyl-allyl acetates 20 and 21 catalyzed by [Rh(cod)Cl]2, [Ir(cod)Cl]2 and
combinations of these complexes with L5 were initially investigated. The
phosphinooxazoline L5 is the precursor of L6 derivative used for attachment to the
DNA.
3.2 Organometallic Transformations in Water
99
Table 3.7. Effect of transition metal precursor and catalyst concentration on amination of branched allylic
substrate 21 with amines 22 and 23.[a]
Conv[b]
Catalyst
Ligand
Nucleophile
Cat.conc.
Solvent
precursor
[mM]
ACN/H2O
[%]
1
[Rh(cod)Cl]2
10
1:1
20
22
2[c]
Rh(nbd)BF4
10
1:1
40
L5
22
10
1:1
>95
3
[Ir(cod)Cl]2
22
10
1:1
>95
4
[Ir(cod)Cl]2
L5
22
[Ir(cod)Cl]2
10
1:1
50[e]
5[d]
L5
23
1.0
1:1
>95
6
[Ir(cod)Cl]2
L5
22
1.0
3:7
>95
7
[Ir(cod)Cl]2
L5
22
0.5
3:7
>95
8
[Ir(cod)Cl]2
L5
22
[a] Reaction conditions: 0.05 mmol 21, 0.07 mmol 22 (1.5 equiv), 2 mol% [Ir(cod]Cl]2, 4.2 mol% L5, 1.0
mL reaction volume, r.t., 14 hours, unless otherwise stated. [b] Conversion estimated by thin-layer
chromatography, unless otherwise stated. [c] Isolated Rh(I) complex 14 from Rh(nbd)BF4 and L5. [d]
0.07 mmol 23 (hydrochloride form) (1.5 equiv), 0.07 mmol NaHCO3 (1.5 equiv). [e] Conversion
determined by reversed-phase HPLC, elution with 50% water and 50% acetonitrile, tR(26) = 12.4 min.
Entry
In the first set of experiments, substrate 21 was reacted with morpholine 22 and glycine
ethyl ester 23. The results, summarized in Table 3.7, clearly show that the complex
formed in situ by combining [Ir(cod)Cl]2 and L5 is highly active in the presence of 5070% water (entries 4 and 6-8). Moreover, complete conversion was achieved with
morpholine 22 (3.0 equiv) as nucleophile even at 0.1 mol% catalyst loading, that
corresponds to a [Ir] catalyst concentration of 0.5 mM (Table 3.7, entry 7). The reaction
rate was significantly higher with morpholine 22 than with glycine ethyl ester 23 (Table
3.7, entry 4 versus 5).
High conversion was also induced by [Ir(cod)Cl]2 without additional ligand (Table 3.7,
entry 3). In addition, the [Ir(cod)Cl]2 precursor was superior to [Rh(cod)Cl]2 which
under the same conditions gave only 20% product formation (Table 3.7, entry 3 versus
1). With the isolated cationic complex [Rh(L5)(nbd)]+BF4- 14 (Table 3.7, entry 2) the
results were distinctly better than with the [Rh(cod)Cl]2 catalyst precursor, slight rate
acceleration being observed (entry 1).
Because of the low reactivity of glycine ethyl ester 23, the next studies were focused on
aminations with morpholine nucleophile 22.
Following the results obtained with the branched substrate 21, we next attempted the
amination of linear allyl acetate 20, using the same Ir(I) catalyst prepared in situ from
[Ir(cod)Cl]2 and L5. Test reactions were carried out in both standard conditions (neat
organic solvent), and 70% aqueous solvent as in our early experiments. The results are
shown in Table 3.8.
3.2 Organometallic Transformations in Water
100
Table 3.8. Allylic amination of linear (20) and branched (21) substrates with morpholine 22 in neat
organic solvent and 70% water, with [Ir(cod]Cl]2 and ligand L5.[a]
Entry
Allylic
Ligand
Solvent
Temp.
Conv[b]
substrate
[°C]
[%]
1
acetonitrile
rt
<1
20
2
acetonitrile
rt
<1
20
L5
3
dioxane
rt
<1
20
4
dioxane
rt
<1
20
L5
3:7 dioxane/water
rt
<1
5[c]
20
3:7 dioxane/water
rt
<1
6[c]
20
L5
3:7 dioxane/water
rt
96
7[c]
21
3:7 dioxane/water
rt
98
8[c]
21
L5
3:7 dioxane/water
50
5
9[c]
20
L5
[a] Reaction conditions: 0.05 mmol 20 or 21, 0.15 mmol 22 (3.0 equiv), 1.4 mol% L5, 0.5 mM [Ir]
catalyst, 1.0 mL reaction volume, r.t., 14 hours, unless otherwise stated. [b] Conversion determined by
gas chromatography. [c] 6 hours reaction time.
The reaction rate was significantly higher for the branched substrate 21, and complete
conversion was achieved in 70% aqueous solvent, in 6 hours (Table 3.8, entries 7 and
8), while the isomeric linear substrate was found in all cases unreacted even after 14
hours reaction time (entries 1-6). The attempt to enhance the reaction rate at elevated
temperature failed: the amination of the linear substrate remained sluggish, even at
50°C, leading to only 5% conversion (entry 9).
Scheme 3.14. Ir(I)-catalyzed allylic amination of branched and linear substrates via σ- and π-allyl
intermediates.
Differences in reaction rates between branched and linear substrates have been
generally observed in Ir(I)-[77] and Rh(I)[286]-catalyzed allylic substitutions. They have
been attributed to the SN2’ mechanism occurring in the formation of the π-allyl
intermediate during the catalytic cycle proposed by Helmchen et al.[67] (Scheme 3.14).
Therefore, the oxidative addition of the substrate to Ir(I) and Rh(I) species is expected
to be faster in the case of the branched isomer due to a less sterically congested
3.2 Organometallic Transformations in Water
101
environment imposed by the substituents at the end of the allylic system.[286]
Our results were in good agreement with the only one reported example of amination
reactions with Ir-PHOX complexes. Helmchen et al. observed that the Ir-catalyst
generated from [Ir(cod]Cl]2 and the phosphinooxazoline ligand (S)-i-Pr-PHOX (L4) was
ineffective in the intramolecular amination of linear allyl acetates, whereas high yield
and moderate to high enantioselectivity were obtained with the branched homologues
(albeit after long reaction times: 4-6 days).[71]
These observations prompted us to evaluate the above-described catalyst mixture of
[Ir(cod)Cl]2 and classical PHOX ligand L4,[67] extensively used in Ir(I)-catalyzed
asymmetric allylic substitutions, versus the structurally-related L5. The aminations of
the branched allyl acetate 21 were performed in neat dioxane or 3:7 dioxane/water
mixture containing 100 mM NaClO4 and 5 mM Mg(ClO4)2 (the presence of salts is later
required by the use of nucleic acid-based systems). In addition, test reactions were
carried out in order to assess the reactivity of Ir-PHOX system at low catalyst
concentration that is an essential condition for creating DNA-based catalysts. The
results are presented in Table 3.9.
Table 3.9. Allylic amination of branched substrate 21 with morpholine 22 according to Scheme 3.12,
using PHOX ligands (S)-L4 and (S)-L5.[a]
Time
Conv[c]
Entry
PHOX Ligand
[Ir]
Solvent[b]
Ee[d]
[h]
[mM]
[%]
[%]
1
1.0
dioxane
1.5
<2
n.d.
L4
2
1.0
3:7 dioxane/water
1.5
99
n.d.
L4
3
1.0
3:7 dioxane/water
1.5
99
6
L5
1.0
3:7 dioxane/water
13
50
3
4[e]
L4
5
0.1
3:7 dioxane/water
13
73
6
0.1
3:7 dioxane/water
13
71
5
L4
7
0.1
3:7 dioxane/water
13
67
1
L5
[a] Reaction conditions: 0.05 mmol 21, 0.15 mmol 22 (3.0 equiv), 2.5 mol% L5 (entries 1-4) and 0.25
mol% L5 (entries 5-7), 1.0 mL reaction volume, r.t., unless otherwise stated. [b] 100 mM NaClO4, 5 mM
Mg(ClO4)2 aqueous solution. [c] Conversion determined by gas-chromatography with dodecane as
internal standard. [d] Determined by HPLC, tR(24) = 10.8, 12.4 min. [e] Kinetic resolution conditions: 0.5
equiv 22.
Although dioxane is a common solvent in allylic substitutions, reaction in neat dioxane
with the standard Ir-L4 catalyst gave almost no product (entry 1). At the same catalyst
concentration, reaction in 3:7 dioxane/water mixture proceeded smoothly, affording
complete conversion with both L4 and L5 ligands, in only 1.5 hours (entries 2 and 3).
Under these conditions, the amination proceeded efficiently even at lower catalyst
concentration (0.1 mM), affording 67-73% conversion (entries 5-7). However, no effect
3.2 Organometallic Transformations in Water
102
on the catalyst activity due to L4 or L5 ligand could be observed, since similar
conversion was obtained also with only [Ir(cod)Cl]2 precursor (entry 6).
The isolated branched amine product 24 obtained with chiral phosphinooxazolines L4
and L5 was submitted to chiral separation. The very low observed enantioselectivities
(Table 3.9, entries 3 and 6,7) are consistent with a strong memory effect of the Ircatalyst obtained from a racemic starting material, as previously observed in similar
transformations with malonate nucleophile.[298] We therefore attempted to use such
memory effects in a kinetic resolution reaction, either by allowing the reaction to
proceed only to 50% conversion or using half equivalent of amine nucleophile.
Disappointingly, treatment of the racemic starting material 21 with half equivalent of
morpholine 22 gave the branched amine product 24 in only 3% ee (Table 3.9, entry 4).
As already mentioned in the introduction of this chapter, allylic aminations can be
accelerated by polar solvents (e.g. alcohols, acetonitrile). It was argued that such
solvents were involved in the stabilization of transition states of the oxidative addition
to Ir(I) species and the nucleophilic attack of the amine.[76] Because of these solvent
effects, both oxidative addition and nucleophilic attack might be enhanced. Intriguingly,
Ir(I)-catalyzed intramolecular aminations with L4 PHOX ligand became slower and
also less selective when polar solvents (e.g., dimethylformamide or acetonitrile) were
used as alternative to toluene.[71]
The results collected in Table 3.9 clearly show that in our system water is a better
solvent than dioxane, and likely a “participating” solvent in catalysis.[299] The unique
solvating properties of water, and its potential contribution to the electronic properties
as well as the steric environment of the catalytic system, exerted through direct binding
to iridium ion, or/and second coordination sphere interactions, might account for the
outcome of the allylic amination. However, the precise reason for the effect of water on
the steric course of the reaction remains unclear.
Beside the solvent effect, the lack of stereoselectivity observed in our catalytic attempts
(Table 3.9, entries 3,4 and 6,7) might be attributed to ligand, or/and amine[76]
nucleophile. Although in the majority of asymmetric allylic substitutions with PHOX
ligands, the relative bulkyness of the oxazoline substituent at the chirality centre appears
one of the decisive factors for achieveing selective catalysts, with L4-based complex
being one of them,[57,
304]
in our case the enantioselectivity induced by L4 was
3.2 Organometallic Transformations in Water
103
unsatisfactory. The structurally related chiral PHOX ligand L5 also displayed weak
selective properties. The steric effect of the amine nucleophile might additionally
influence the stereocontrol of the reaction, as previously observed by Takeuchi:[76] the
stereoselectivity decreases as the steric bulk of the amine decreases.
Unlike most transition metal catalyzed processes, allylic aminations do not exclusively
rely on a single mechanism as a source of asymmetry. A possible catalytic cycle for the
Ir(I)-catalyzed allylic amination is illustrated in Figure 3.23. The following general
steps have been proposed by Helmchen et al.[67]: 1) in situ formation of the Ir(I) catalyst
(complex A), 2) oxidative addition of the substrate to the metal centre and formation of
π-allyl complexes (complex B/B’), and c) attack of the nucleophile (Nu), trans to
phosphorus. It has also been proposed that these reactions proceed with double
inversion, via σ-allyl or π-allyl complexes, which undergo σ-π-σ isomerisation
[57, 77]
(Scheme 3.15).
Figure 3.23. Proposed mechanism for the allylic amination of a branched racemic allyl substrate using
Ir(I)-L5 complex.
Although the preference for one of the allylic substrate faces (kinetic control) (Scheme
3.15 I) could be one possible mechanism,[57] the source of the enantioselectivity is
complicated by the possibility that one or more steps in the catalytic cycle may be the
enantiodiscriminating step(s). Enantioselection can also derive from a certain degree of
isomerisation between the π-allyl intermediates (thermodynamic control), or/and a
stereospecific nucleophilic attack (kinetic control) (Scheme 3.15 II and III,
respectively).
3.2 Organometallic Transformations in Water
104
Scheme 3.15. Enantiocontrol in Ir(I)-catalyzed allylic amination.
Our attempt to reduce the rate of the nucleophilic attack by lowering the nucleophile
concentration (Table 3.9, entry 7), and thereby install the stereocontrol likely promoted
by the enantioface preference of our catalyst, was so far unsuccessful. At this stage, the
mechanism by which the Ir-PHOX catalyst imposes its chirality upon the branched
amination product is difficult to understand.
However, our results lead to the following conclusions: 1) the Ir(I)-catalyzed allylic
amination is compatible with the use of nucleic acids; 2) the in situ formed Ir(I)-L5
complex is active in 70% aqueous mixtures, in the presence of salts, at room
temperature and at low catalyst concentration (100 μM); 2) the Ir(I)-catalyst generated
with a chiral PHOX ligand, either L4 or L5, does not provide enantioselectivity in
amination of racemic branched alyllic substrate with morpholine. Nevertheless, these
promising findings represent a convenient starting point for the development of DNAor RNA-based asymmetric catalysts, in which the nucleic acid fold can possibly
contribute to the stereoselectivity of the process.
3.2.3.3 Allylic Amination with DNA-Appended Phosphinooxazoline
Ligands
Iridium(I)-phosphinooxazoline complexes are among the most powerful catalysts
employed in allylic substitutions, albeit never used in combination with DNA or RNA
scaffolds.
The promising results obtained with the catalyst formed with [Ir(cod)Cl]2 and ligand L5
led us to investigate the ability of DNA-appended PHOX conjugates to generate
enantioselective catalysts for amination of racemic branched allyl acetate with
3.2 Organometallic Transformations in Water
105
morpholine, assuming that the DNA chirality would provide the stereocontrol of the
reaction.
To limit the non-specific binding of [Ir(cod)Cl]2 to the DNA, maximum one equivalent
of iridium precursor has to be used for complex formation with DNA-based PHOX
ligands. However, since the DNA strands are anionic, non-specific electrostatic
attractions between these strands and cationic species, including the iridium ion, might
occur. More importantly though, the DNA heterocyclic bases have a variety of nitrogen
and oxygen donor atoms which can coordinate iridium and influence its catalytic
properties. It has been reported that adenine derivatives, for example, can form
rhodium-complexes with N1, N6 and N7 as binding sites.[305]
Control experiments were conducted to determine if in our case DNA interferes with
iridium binding and leads to catalytically inactive species. Amination reactions were
performed in the presence of 1.1-1.4 equivalents of synthetic unmodified DNA (23 mer,
cDNA2: 5’-GG AGC TCA CAA GTC CTT CAC TGC-3’), which was either firstly
combined with the L5 ligand, followed by addition of [Ir(cod)Cl]2 precursor, or directly
added to the pre-formed Ir-L5 complex. The experiments were carried out on 100 μL
scale and the catalyst concentration was maintained in the range of 50-100 μM. Early
assays showed that the solution of metal precursor [Ir(cod)Cl]2 must be freshly prepared
each time. A stock solution of metal precursor [Ir(cod)Cl]2 in the presence of
phosphinooxazoline ligand L5, in dioxane, could be instead used for at least one month
when stored at -20° C without noticing any difference in catalytic activity. All reactions
were performed in water/dioxane 7:3, and/or in the presence of 100 mM NaClO4 and 5
mM Mg(ClO4)2. This solvent mixture ressembles the conditions used in the stability
assays of double-stranded DNA (Chapter 3.1.2.2), the presence of the salts being
important for maintaining DNA/DNA or DNA/RNA constructs in helical conformation.
Moreover, since the amount of mono- and divalent cations is relatively high, we assume
that the charged phosphate connecting units are compensated by metal ions. Localizing
metal ions along the DNA phosphodiester backbone may increase the chance that DNA
will be shielded from the iridium ion. In this case, the potential interactions of the
iridium ion with the DNA would be restricted to nucleobase coordination.
3.2 Organometallic Transformations in Water
106
Table 3.10. Effect of DNA on Ir(I)-catalyzed amination of branched allylic substrate 21 with morpholine
22.[a]
[Ir]
[DNA][b]
Salts
Time
Conversion[c]
[μM]
[μM]
[h]
[%]
1
50
19
20
L5
2
50
53.3
19
24
L5
3
50
+
16.5
37
L5
4
50
53.3
+
16.5
33
L5
5
100
+
6
75
6
100
135
+
6
67
7
100
+
6
67
L5
100
135
+
6
72
8[d]
L5
[a] Reaction conditions: 5 μmol 21, 15 μmol 22 (3.0 equiv), 100 μL reaction volume, r.t., unless
otherwise stated. [b] Pre-formed Ir-L5 catalyst was added to the DNA (1.1-1.4 equiv) solution.
Experiments performed at least in duplicate. [c] Conversion determined by gas chromatography with
dodecane as internal standard. [d] [Ir(cod]Cl]2 was added to a 1:1 PHOX/DNA solution.
Entry
Ligand
The results shown in Table 3.10 confirmed that the catalytic activity of both [Ir(cod)Cl]2
precursor and pre-formed Ir-L5 complex was preserved in the presence of DNA (entries
2, 4, and 6). Interestingly, the presence of salts in the reaction milieu led to a slight
increase in rate acceleration (entry 3 versus 1). However, no significant salt-dependent
effect on the reaction conversion could be observed when the DNA was added to the
reaction mixture (entry 4 versus 2). This finding indicated that the DNA phosphate
groups, apparently available for interactions with the iridium ion, did not trigger
formation of catalytically inactive species. Furthermore, it appeared that the catalytic
complex in situ generated from [Ir(cod)Cl]2 and L5 was formed in the presence of one
equivalent of DNA per PHOX ligand, since no decrease in conversion was observed
(entry 8 versus 7). On the other hand, this result can also be argued by the high catalytic
activity of the [Ir(cod)Cl]2 precursor without additional ligand (shown in the previous
chapter: Table 3.9, entry 5 versus 6 and 7), which anyway is not disturbed by the
presence of DNA (Table 3.10, entry 5 versus 6). Therefore, under these conditions, the
3.2 Organometallic Transformations in Water
107
ability of the DNA coordinating sites to compete with the PHOX ligand in iridiumbinding could not be entirely excluded, although a bidentate phosphinooxazoline is
expected to be a superior ligand relative to the nucleobases and the phosphate groups
are likely to be shielded by the mono- and divalent ions present in the reaction mixture.
Importantly, our data clearly demonstrated that the interactions, if any, between the
umodified synthetic 23mer DNA and the [Ir(cod)Cl]2 precursor or the pre-formed Ir-L5
complex, were negligible since they did not alter the catalytic properties of the system.
Encouraged by these results, we then tested the DNA-based PHOX ligands in our model
reaction with the branched allyl acetate 21 and morpholine nucleophile 22. It has been
assumed that the DNA-appended PHOX ligand would provide chelating control on the
iridium ion, and favour its precise positioning into the DNA chiral environment. We
screened DNA-PHOX constructs containing diverse linker units, four-carbon, twocarbon, and 13-atom spacer (ODN11a-c) (Chapter 3.1.2.4), and attempted to study the
influence of the linker length on the catalytic properties of the resulting DNA-tethered
iridium complex as well as on the transfer of chirality from the DNA scaffold.[146, 199]
The Ir(I) complexes were prepared by mixing 2.2-2.6 equivalents of HPLC purified
ODN11a-c conjugates with [Ir(cod)Cl]2 in a degassed aqueous solution of 143.0 mM
NaClO4 and 7.0 mM Mg(ClO4)2 (see Materials and Methods, Chapter 5.9.2.3). In order
to rule out decomposition of DNA-PHOX ligands caused by the oxidation of the
phosphinooxazoline moiety during the reaction, as well as possible side reactions of
DNA with components of the reaction mixture, the DNA-appended Ir(I) complexes
were incubated with the allyl substrate 21 and morpholine 22, under the same
conditions used for the amination reaction, and then analyzed by HPLC. The
presumably formed Ir(I)-PHOX complex is expected to be stable towards oxidation.
Figure 3.24 illustrates the HPLC chromatograms obtained for the ODN11a under the
above-described conditions. The oxidation proceeded slowly, from <10% (A)
immediately after addition of [Ir(cod)Cl]2 to the DNA to 54% after overnight incubation
in the presence of the substrates (B). Similarly, slight amount of oxidized DNA-PHOX
conjugate (<10%) was also obtained in the case of ODN11b and ODN11c soon after
treatment with [Ir(cod)Cl]2, while high levels of oxidation, 50% and 70%, respectively,
were observed after carrying out the amination reaction. However, the amount of the
observed oxide ODN11a-c(O) in the HPLC assays is probably overestimated. Beside
3.2 Organometallic Transformations in Water
108
the long incubation time, a certain level of oxidation that might occur during HPLC
sample preparation (outside a glove-box) and due to the oxygen dissolved in the HPLC
solvents must be also considered.
Figure 3.24. HPLC chromatograms of the phosphinooxazoline-derivatized DNA conjugate ODN11a. A)
HPLC purified product immediately after redissolving in aqueous salt solution and addition of 0.5
equivalent [Ir(cod)Cl]2. B) Overnight incubation at room temperature with allylic substrate and
morpholine. DNA sequence: 5’-GC AGT GAA GGC* TGA GCT CC-3’, C* = N4-PHOX-appended 2’deoxycytidine.
We assumed that in all cases the remaining amount of non-oxidized DNA-PHOX
conjugate (approx. 50%) was high enough to ensure appropriate conditions for carrying
out allylic amination reactions in the presence of ODN11a-c. The reactions were
performed on a 50 or 100 μL scale, in 7:3 water/dioxane, at room temperature. The
concentration of DNA-appended PHOX ligand was maintained between 66 and 130
µM, considering 1.1-1.3 equivalents of ligand per iridium and only 90% purity of the
DNA-conjugate due to inevitable oxidation. The results are shown in Table 3.11.
Disappointingly, in all cases when the DNA-tethered PHOX ligand was used, the
catalyst activity was highly reduced (<28% conversion) compared to both [Ir(cod)Cl]2
precursor (entry 1) and pre-formed Ir-L5 complex
(entry 2) and additionally, no
enantioselectivity was observed (entries 3, 4, and 6-8). The best conversion (28%) was
obtained with the DNA conjugate containing the longest spacer between the PHOX
moiety and the DNA scaffold, ODN11c (entry 8). This result reflects the higher
flexibility introduced by the Ir(I)-PHOX-tethered linker, apparently important for
preventing coordination of the iridium ion with the DNA electron donor atoms, and
finally preserving the catalytic activity.
3.2 Organometallic Transformations in Water
109
Table 3.11. Iridium(I)-catalyzed allylic amination of branched allyl acetate 21 with morpholine 22 with
DNA-appended PHOX ligand ODN11a-c.[a]
[Ligand]
[Ir]
Conversion[b]
Ee[c]
[μM]
[μM]
[%]
[%]
100
73
2[e]
140
100
82
<1
L5
3
130
100
12
5
ODN11b
4
130
100
<1
n.d.
ODN11a
5
75
60
84
n.d.
6
78
60
10
2
ODN11b
7
78
60
9
<1
ODN11a
8
78
60
28
<1
ODN11c
[a] Reaction conditions: 5 μmol 21, 5.5 μmol 22 (1.1 equiv), 100 mM NaClO4, 5 mM Mg(ClO4)2, 100 μL
reaction volume, r.t., 16 hours. [b] Conversion determined by gas chromatography with dodecane as
internal standard. [c] Determined by HPLC, detection wavelength λ = 254 nm, tR(24) = 10.8, 12.4 min.
[e] 0.5 mol% L5.
Entry
Ligand
These results were inconsistent with our previous findings that clearly supported the
assumption that (unmodified) DNA does not participate in iridium coordination (Table
3.10), or at least the existing interactions are not detrimental for catalytic activity. To
explain the results obtained with PHOX-carrying DNA sequences, one could consider
that the formation of the DNA-appended Ir(I)-PHOX complex would result in close
localization of the metal centre relative to the DNA and, consequently, in proximity to a
plethora of ligands, such as the nonbridging phosphoryl oxygens and the 19 hydroxyl
groups of the backbone, as well as the nitrogens and oxygens of the purine or
pyrimidine bases. In this case, new iridium-DNA interactions, that haven’t been
observed so far, might emerge. Based on reasons discussed before, it was reasonable to
assume that such interactions would preferentially involve the nucleobases. In order to
prevent undesirable coordination, we chose to use double-stranded DNA constructs,
3.2 Organometallic Transformations in Water
110
where the complementary strand sequesters most of the heteroatoms of the nucleobases
through Watson-Crick base-pairing, making them no more available for metal
coordination.
3.2.3.4 Allylic
Amination
with
Double-stranded
DNA-appended
Phosphinooxazoline Ligands
We envisaged the Watson-Crick base-pairing of DNA as a tool to provide a particular
steric environment for the transition metal. In this case, the flexibility of the singlestranded DNA given by a substantial degree of bond rotation occuring in the
phosphodiester backbone linkages is considerably reduced. As a result, the structural
constraints generated upon duplex formation are expected to facilitate the transfer of
chirality from the DNA helix to the catalytic centre.
A series of oligonucleotides (19mers) containing the phosphinooxazoline ligand moiety
attached to the deoxycytidine-19 residue (direction 5’-3’) via three different spacers
(ODN11a-c) were hybridized with the complementary DNA strand cDNA1 (Figure
3.25 A). Two additional complementary oligonucleotides cDNA2 and cDNA3 were
chosen, that upon hybridization generated small bulges (3-4 nt) on either the unmodified
(cDNA2) or the PHOX-tethered DNA strand (ODN11a-c), in close vicinity to the
ligand attachment site (Figure 3.25 B and C). This design based on introducing elements
of flexibility of particular size and location within the duplex was anticipated to bring
about shape changes and provide more complex structures for catalysis.
Confident to our previous observations regarding the B-DNA duplex stability in
mixtures containing 30% water-miscible organic cosolvents in aqueous buffers (Figure
3.7), all double-stranded DNAs were used in Ir(I)-catalyzed allylic amination of
branched acetate substrate 21 with morpholine 22, in 70% aqueous solvent. Treatment
of DNA-PHOX conjugates with equimolar amounts of complementary strands
(typically 2.0-4.0 nmol) at room temperature, in aqueous salt solution, results in
spontaneous assembling of DNA duplexes. After complexation with [Ir(cod)Cl]2
precursor and in situ formation of Ir(I)-catalyst (1.1-1.3 equivalents of DNA-appended
PHOX ligand per iridium), the amination reactions were started by addition of acetate
substrate 21 and morpholine nucleophile 22. Parallel test reactions with single-stranded
3.2 Organometallic Transformations in Water
111
DNA-PHOX conjugates were also performed, as previously described, to allow a more
accurate comparison of the catalytic systems. Some results are shown in Table 3.12.
Figure 3.25. Double-stranded DNA-appended Ir(I)-phosphinooxazoline complexes. L denotes solvent
molecule, chloride ion (from the [Ir(cod)Cl]2 precursor), or other coordinating species.
Contrary to our expectations, the catalytic activity of DNA-bound Ir-PHOX complexes
could not be restored by blocking the oxygens and nitrogens on the Watson-Crick edge
of the purine or pyrimidine bases by DNA duplex formation (entries 6-14 vs 3 and 4). It
seems that all designed double-stranded DNA constructs still provide coordinating
moieties responsible for formation of catalytically inactive species (entries 6-14 vs 1
3.2 Organometallic Transformations in Water
112
and 2). At first glance, we assumed that the only available donor atoms after WatsonCrick base pairing are the N7 nitrogen atoms of deoxy-adenosine and -guanosine
residues. Moreover, these sites are localized at the floor of the major groove of the BDNA duplex (one example of dG=dC base-pair positioning towards the B-DNA major
and minor grooves is shown in Figure 3.5). It is known that a functional tether attached
at the N4-exocyclic amine of deoxycytidine residue protrudes out into the major groove
space.[255,
306, 307]
Therefore we expect that the phosphinooxazoline moiety and its
resulting iridium-complex would be specifically positioned in the major groove in close
proximity to the presumably coordinating N7 atoms of adjacent dG and dA
nucleobases[305] (Figure 3.26 A).
Table 3.12. Iridium(I)-catalyzed allylic amination of branched allyl acetate 21 with morpholine with
double-stranded DNA-appended PHOX ligand.[a]
Entry
PHOX
DNA-PHOX
Complementary DNA
Conversion[b]
Ee[c]
Ligand
Ligand
[%]
[%]
cDNA1
cDNA2
cDNA3
1
44
n.d.
2
65
<1
L4
3
8
3
ODN11a
4
14
2
ODN11b
5
23
<1
ODN11c
6
+
8
<1
ODN11a
7
+
3
3
8
+
7
2
9
+
22
<1
ODN11b
10
+
7
7
11
+
15
<1
12
+
18
<1
ODN11c
13
+
3
6
14
+
6
n.d.
[a] Reaction conditions: 0.25 μmol 21, 0.63 μmol 22 (1.1 equiv), 100 mM NaClO4, 5 mM Mg(ClO4)2, 50
μL reaction volume, 0.5 mol% [Ir(cod]Cl]2, 1.2 mol% L4 (entry 2), 1.4 mol% ODN11a-c (entries 3-14),
r.t., 19 hours. [b] Conversion determined by gas-chromatography with dodecane as internal standard. [c]
Determined by HPLC, detection wavelength λ = 220 nm, tR(24) = 10.8, 12.4 min.
Molecular modelling was further used to gain insight into possible intramolecular
interactions within these molecules. Theoretical models based on quantum mechanical
calculations have been proposed for the amino tether-functionalized Ir-L5 complexes.
The energy-minimized structures, that fairly resemble the published X-ray crystal
structure of homologues π-allyl Ir(III)-complexes,[68] were then used to estimate the
distance between the iridium atom and the carbon C1 of the linker, directly attached at
the dC nucleobase (Figure 3.26 B). This measurement gives the theoretical, linkerdependent distance of the metal centre within the DNA duplex, and consequently the
3.2 Organometallic Transformations in Water
113
maximal interaction sphere with nucleobases. Based on these calculations, we assumed
that the iridium ion can reach either the deoxyguanosine residues present above and
below the modified deoxycytidine (5’ → 3’, dG = dC base pairs 9, 10, and 13) or the dG
directly involved in Watson-Crick pair with the PHOX-tethered dC (dG = dC base pair
11) (Figure 3.26 A).
Figure 3.26. A) Localization of the Ir(I)-L5 complex inside the B-DNA duplex (left). B) Theoretical
model of the dC-tethered-Ir(I)-L5 complex within the DNA. The model of the Ir-PHOX complex was
constructed in Chem3DDraw and MM2+ minimized (two coordinating chlorine atoms were chosen for
simplicity of the model). The theoretical maximal free rotation (r) allowed by the spacer was estimated in
Chem3DDraw using energy-minimized structures.
To confirm this hypothesis, we focused on applying specific structural changes in the
nucleotide sequence of the DNA-based PHOX ligands. Two new sets of DNA-PHOX
conjugates were prepared, in which the above-mentioned potentially coordinating
residues located on the sense strand were replaced by either the weakly coordinating 2’deoxyadenosine or the non-coordinating 7-deaza-riboguanosine. Moreover, the base
composition was slightly changed in the former case, in order to ensure a dG/dC content
and a subsequent thermal stability comparable to the initial duplexes. For the second
type of substitution, the remaining DNA sequence was conserved.
3.2 Organometallic Transformations in Water
114
The aminoalkyl-modified ODN precursors ODN4a-c and ODN5a,b were prepared
using the convertible nucleoside strategy, followed by deprotection, cleavage and
substitution with aliphatic diamines, as described in section 3.1.2.1. Carbodiimide
coupling with the carboxylate PHOX-derivative L6 in the presence of Nhydroxysuccinimide, according to Scheme 3.4, afforded the DNA-PHOX conjugates
ODN12a-c and ODN13 in 42-78% yield (Table 3.13).
Table 3.13. Post-synthetic functionalization of amino-modified ODN4a-c and ODN5a,b with
phosphinooxazoline L6.[a]
ODN
Sequence
Ligand
Linker R
Yield
[%]
ODN12a
ODN12b
5'-GC AGC GAT AACR TAA GCG CT-3'
L6
42
H2N
65
R
5'-GC AGC GAT AAC TAA GCG CT-3'
R
ODN12c
5'-GC AGC GAT AAC TAA GCG CT-3'
ODN13
5'-GC AGT GAA XXCR TXA GCT CC-3'
H2N
H2N
O
H2N
2O
78
47
[a] Reaction conditions: 39-80 μM amino-modified ODN, 33.3 mM L6, 0.1 mM NaCO3 1:2 DMF/H2O,
r.t., overnight.
The DNA-PHOX conjugates were then tested in allylic amination reactions using the
previously described procedure. To introduce more structural variation, DNA/RNA
hybrids were also prepared, in addition to DNA/DNA duplexes. Such DNA/RNA
duplexes are known to adopt an A-DNA structure that results in a deep and narrow
major groove and a very shallow and wide minor groove (Figure 1.13, Chapter 1.2.1).
Such structural features might induce more structural constraints and thus afford more
selective nucleic acid domains for catalysis. Moreover, these DNA/RNA hybrids can be
seen as appropriate models to approximate the system designed for the in vitro selection
of RNA-based hybrid catalysts.
The DNA-Ir(I) complexes were prepared in situ by combining [Ir(cod)Cl]2 with a
degassed solution of single- or double-stranded DNA/DNA or RNA/DNA-appended
PHOX ligand (1.3 equivalents of DNA-PHOX conjugate per iridium, typically 1.3-2.6
nmol). In all cases, the final concentration of Ir(I) catalyst was maintained between 20-
3.2 Organometallic Transformations in Water
115
40 μM and the reactions were carried out on a 50 μL scale. The results of the catalysis
attempts are given in Table 3.14.
Table 3.14. Iridium(I)-catalyzed allylic amination of branched allyl acetate 21 with morpholine 22 with
single- and double-stranded DNA-appended PHOX ligand ODN12a-c and ODN13.[a]
Entry DNA-PHOX
[Ir]
Complementary DNA or RNA[b]
Conversion[c]
Ligand
[μM]
[%]
cDNA3
cDNA4
cRNA
1
40
34
2
40
30
ODN12a
3
40
24
ODN12b
4
40
+
20
ODN12b
5
40
+
18
ODN12b
20
+
16
6[d]
ODN12a
20
+
11
7[d]
ODN12a
23
+
8
8[d]
ODN12c
23
+
18
9[d]
ODN12c
30
+
36
10[e]
ODN13
[a] Reaction conditions: 0.25 μmol 21, 0.63 μmol 22 (1.1 equiv), 100 mM NaClO4, 5 mM Mg(ClO4)2, 50
μL reaction volume, r.t., 19 hours, 1.0 mol% ODN12a,b, unless stated otherwise. [b] cDNA3: 5’-GG
AGC TCC TTC ACT GC-3’; cDNA4: 5’-AG CGC TTA GTT ATC GCT GC-3’; cRNA: 5’-AG CGC
UUA GUU AUC GCU GC-3’. [c] Conversion determined by gas-chromatography with dodecane as
internal standard. [d] 0.5 mol% ODN12a,c. [e] 0.8 mol% ODN13.
The data show that by substituting the dG nucleotides on the sense strand in close
proximity to the metal complex attachment site, the catalytic activity of the DNAtethered Ir-PHOX complexes attached to DNA could be generally restored.
Nevertheless, it appears that the dA residue exhibits weak iridium binding strength
through its N7 site, as indicated by the slightly lower conversion obtained with the
ODN12a,b conjugates (entries 2-7 versus 1). The DNA domain controlled by the long
tether carrying the phosphinooxazoline moiety in ODN12c includes several remote dG
residues that, beside those involved in the dG=dC base-pairs 9, 10, and 13, are
apparently in charge of interactions with the iridium ion, leading to reduced catalytic
activity, as observed with the ODN12c/cDNA4 duplex (entry 8). However, the
DNA/RNA analogue, generated from ODN12c and its complementary cRNA apperead
to behave differently and higher reaction conversion was attained (entry 9). Since these
two constructs adopt distinct helical forms, one can speculate that the specific
accommodation of the attached Ir-L5 complex in the A-DNA duplex rules out to some
extent the unfavourable interactions with the DNA coordinating sites. Notably, the best
result was achieved with the 7-deaza-riboguanosine-containing DNA construct
ODN13/cDNA3 (entry 10). In this case, the two-carbon linker-appended Ir-PHOX
complex positions itself into the bulge, approaching only the nucleobases that can not
3.2 Organometallic Transformations in Water
116
compete for metal coordination: 7-deaza-G above and dT below the tethering site. This
results in no observable inhibition effect on the catalytic activity. All these studies need
to be reconciled in order to fully understand whether all three 9dG,
10
dG, and
13
dG
nucleobases or combinations of them are responsible for disrupting the distinct binding
motif provided by the bidentate DNA-appended phosphinooxazoline ligand.
The enantioselectivity was in all cases very poor (<3%). The length of the linker
carrying the iridium complex seems to play an important role in the stereoselectivity of
the catalytic species.[146, 200] At this point, we believe that all three chosen tethers are too
long and they presumably mimimize the interactions between the DNA and the bound
iridium(I) catalyst, yielding a weak transfer of chirality from the DNA backbone to the
metal active site.
4 Conclusions and Oulook
4
117
Conclusions and Outlook
The incorporation of transition metal complexes into DNA and RNA is an important
objective for the development of functional biomolecules with potential applications as
therapeutics, artificial nucleases, and as nanotechnology construction material. Inspired
by the seminal work of Whitesides, who showed that asymmetric catalytic
hydrogenations could be performed by anchoring an achiral RhI complex in a chiral
cavity of the protein avidin, we aimed at embedding transition metal complexes in
nucleic acids folds in order to generate nucleic acid-based hybrid catalysts. In addition,
the use of combinatorial methods in the later stages of the project is expected to assist
the quick development of artificial metallo-DNAzymes and -ribozymes with the desired
activity and selectivity.
Figure 4.1. In vitro selection of RNA-based hybrid catalysts with DNA-appended transition metal
complexes.
In this respect, a SELEX-type strategy devised for screening combinatorial libraries of
up to 1015 metal-binding nucleic acids, followed by PCR amplification (Figure 4.1)
(and, in some cases, diversification), would be a valuable tool towards hybrid catalysts,
exclusive of the high-throughput-screening systems required by the evolutive
4 Conclusions and Oulook
118
development of enantioselective enzymes. In vitro selection of RNA-hybrid catalysts
remains probably a difficult approach since it requires a proper combination of
structural and functional information from both nucleic acids and organometallic
chemistry. In order to establish a robust selection scheme, several strategies were sought
to allow for a precise positioning of the metal complex within the RNA fold. As
discussed in the first part of Chapter 3.1, the most suitable way to provide the system
with the necessary ligand for transition metal coordination consists of employing short
DNA/RNA hybrids, with the DNA acting as a metal chelator-carrier oligonucleotide.
This is the key step in the selection cycle, where a modified RNA library is creating by
simply carrying out a hybridization step with a suitably functionalized DNA strand
(Figure 4.1).
At the moment we started this work, there was no precedent for nucleic acid-based
hybrid catalysts. The major challenges in the development of hybrid catalysts based on
DNA or RNA and transition metal complexes are in the field of asymmetric catalytic
carbon-carbon and carbon-heteroatom bond forming reactions. Since the most powerful
catalysts for these reactions are based on phosphorus ligands, we attempted to
covalently attach achiral phosphite units at the 5’-terminus of the DNA sequences.
Despite the successful assembling of such DNA conjugates using solid-phase synthesis
and phosphoramidite ligand precursors, the desired DNA-appended phosphite could not
be isolated due to its very low stability under the conditions used in purification.
However, the problematic isolation of these conjugates could be overcome by the preformation of the corresponding rhodium(I) complex that was reasonably stable during
the deprotection and cleavage of the resin-bound, functionalized oligonucleotide
(Chapter 3.1.1)
The derivatization of DNA with a second class of phosphorus based ligands, namely
phosphines, has been then envisioned as a suitable approach to modifying
oligonucleotides with versatile ligands for transition metal coordination. An efficient
post-synthetic strategy for the site-specific incorporation of mono- and bisphosphine, as
well as phosphinooxazoline ligands into DNA sequences has been established. Parallel
synthesis of various DNA precursors bearing a primary alkylamino functionality that
can be selectively addressed was achieved by the one-pot conversion, deprotection and
cleavage of convertible nucleoside-containing oligomers with diamines (Chapter
4 Conclusions and Oulook
119
3.1.2.1). The subsequent coupling of amino-modified oligonucleotides with PYRPHOS,
BINAP and PHOX ligands equipped with a carboxyl group allowed the attachment of
the phosphine moieties at defined predetermined internal sites (Figure 4.2) (Chapter
3.1.2). Phosphine-containing oligonucleotides and their phosphine sulfide analogues
were characterized by mass spectrometry (MALDI-TOF and FT-ICR-ESI) and their
stability after purification and isolation systematically investigated. While the DNAappended PYRPHOS ligand was quickly oxidized, BINAP and PHOX conjugates
showed high stabilities, making them useful precursors for the development of metalcontaining oligonucleotides. The approach described here provides new chelating
functionalities for introducing metal centres at well-defined positions in DNA or RNA
sequences and a unique collection of DNA-based phosphine ligands for creating
efficient catalysts (Chapter 3.2.3). In addition, the combination with spacers differing in
length and structure (Figure 4.2) might be relevant for the interactions between the
transition metal complex and the biopolymer backbone.
Figure 4.2. Site-specific incorporation of phosphine ligands into DNA sequences.
In addition to the synthetic challenges for incorporation of transition metal complexes,
one barrier to expanding the applications of DNA and RNA in asymmetric
organometallic catalysis stems from finding suitable target reactions that proceed in the
presence of water and are compatible with the use of nucleic acids. Although an
increasing number of transition metal-catalyzed transformations in aqueous mixtures are
being reported, several reaction parameters and conditions had to be systematically
examined and revised in order to achieve proper model systems compatible with the
structure and properties of nucleic acids. Our studies on the preparation and stability of
4 Conclusions and Oulook
120
mono- and bisphosphine- and phosphinooxazoline-based complexes with transition
metals (e.g., palladium, platinum, rhodium, iridium) demonstrated that these systems
tolerated well water as cosolvent and could be easily handled in the laboratory
atmosphere (Chapter 3.2.1). These findings together with the high affinity for transition
metals make phosphorus ligands certainly the entities of choice for embedding
transition metal complexes in DNA and RNA folds and creating metal-based catalytic
nucleic acids.
From the wide range of homogeneous processes that phosphine- and phosphinooxazoline-transition metal complexes can accelerate, commonly in neat organic solvent,
two model reactions were selected. The rhodium(I)-catalyzed 1,4-addition and
iridium(I)-catalyzed allylic amination proceeded efficiently in aqueous medium (e.g., up
to 70% water in allylic aminations), at room temperature, even with low catalyst
concentration (e.g. 0.1 mM concentration of iridium catalyst). These findings were a
successful event in our long-standing efforts to establish optimal systems, which
wellmatch nucleic acids properties. In our hands, the ligand of choice for the 1,4addition of phenyl boronic acid to 2-cyclohexen-1-one (Scheme 3.10) appeared to be
the BINAP ligand L8. The best conversion was obtained with the isolated
[Rh(nbd)BINAP]BF4 catalyst which was found superior to the related complex
generated from the PYRPHOS ligand (Chapter 3.2.2). Our preliminary results using
mono- and biphosphine ligands stimulate the applications of the DNA-appended
BINAP in rhodium-catalyzed conjugate additions. For the convenient test of the DNAbased systems, additional optimization of the reaction conditions (concentration of the
catalyst and substrates, water content) and investigation of the degree of
enantioselectivity induced by the BINAP itself have still to be done.
In the second model reaction employed in this work, the in situ prepared Ir(I)-PHOX
complexes afforded amination products in good yields, starting from the branched
racemic phenyl allyl acetate, while the catalyst was found ineffective with the isomeric
linear substrate (Scheme 3.12). Significant rate accelaration was observed when 70%
water was used as cosolvent in aminations of the branched substrate with morpholine.
Under the same conditions, the linear acetate remained unreactive, albeit slight
enhancement of the reaction rate was observed at elevated temperature. Although the
achieved conversion is very modest by conventional catalysis’ standards, this system
4 Conclusions and Oulook
121
introduces an interesting approach towards in vitro evolution of nucleic acid hybrid
catalysts on the basis of stability and activity at high temperature. More in detail,
SELEX-type techniques could aid to the isolation of thermostable nucleic acid
sequences that, upon recruiting the Ir(I)-PHOX complex, are capable of complementing
its catalytic activity and finally make unreactive substrates accessible.
The very good conversions (>98%) afforded by Ir(I) complexes in the aminations of the
branched starting material, under environmetally friendly conditions, and at low catalyst
loading (0.5 mM) (Chapter 3.2.3.2) contribute to the continued efforts of expanding the
scope of biocatalysis. In addition, the novel chiral PHOX derivatives L5 and L6
reported in this work are attractive candidates for applications in organometallic
catalysis, such as hydrogenations, Heck reactions or hydrosilylations. The ligand L6
possesses suitable functionality for further derivatization, such as dendrimer fixation,
and finally generation of metallodendrimer catalysts. By the virtue of the so called
“dendrimer effect”, high levels of selectivity in asymmetric transformations can be
enforced. On the other hand, this PHOX derivative can undergo functionalization with
solid supports and assist formation of immobilized catalysts. This approach would allow
new applications of PHOX ligands in heterogeneous catalysis.
Importantly for our purposes, the Ir(I)-PHOX(L5) complex (0.05-0.10 mM) remained
highly active in the allylic amination reaction in the presence of unmodified DNA and
high concentration of salts. This observation led to the conclusion that no undesired
interactions between the transition metal ion and the DNA coordinating sites occured
that could shut down the reaction. Interestingly, no kinetic resolution on the branched
racemic substrate and, implicitly, enantioselectivity could be achieved, despite the
stereogenic nature of the PHOX-based catalysts. Since allylic amination reactions do
not exclusively rely on a single mechanism as source of asymmetry, the origin of
stereoselectivity remains difficult to understand at this stage. The mechanism by which
the ligand can transfer its chirality with high fidelity to the amination product, or by
which the steric effects of the amine nucleophile contribute to enantiodiscrimination
make definitely the object of interesting upcoming studies.
Nevertheless these data open the way for exploring the catalytic potential of nucleic
acids. As already outlined in the “Objectives” section of this work, the low levels of
enantioselectivity induced by conventional transition metal catalysts could be surpassed
4 Conclusions and Oulook
122
by making use of biopolymers. Two possibilities may arise: 1) the biopolymer exerts
exclusive stereocontrol on the reaction outcome, or 2) the steric information carried by a
chiral ligand is enriched by that of the biopolymeric part.
We attempted to assess to what extent synthetic 19mer DNA sequences may asssist the
transfer of chirality in allylic amination reactions. We considered that the DNA-based
ligands were attractive scaffolds for transition metal catalysts due to the fact that they
could be easily engineered and well-defined secondary structures based on WatsonCrick base pairing could be designed. Single-stranded DNA-PHOX conjugates were
initially tested in the allylic amination of the branched racemic allyl acetate with
morpholine, in 70% water, as model reaction. Preliminary attempts, in which DNA
sequences containing a 2’-deoxyguanosine-rich domain in close proximity to the ligand
attachment site were employed, yielded lower conversions compared to those obtained
when the non-bound iridium(I)-PHOX complex was used instead (Chapters 3.2.3.2 and
3.2.3.3). A similar trend was observed for all tethered DNA-PHOX systems,
independently of the length of the spacer carrying the ligand. These results evoked
several key questions: 1) Can the iridium(I) complex be selectively formed with the
DNA-appended bidentate PHOX ligand? 2) Can other binding sites for the transition
metal be found in the DNA molecule? 3) Can the phosphate units and nucleobase
heteroatoms contribute to the first coordination sphere? One possible answer might
come from our early observations discussed above, on conserved catalytic activity of
both [Ir(cod)Cl]2 precursor and pre-formed Ir(I)-PHOX complex in the presence of
unmodified DNA. According to these results, one could assume that complex formation
between the iridium(I) precursor and the DNA-appended PHOX ligand rendered a more
precise and closer localization of the metal ion next to potential coordinating sites of the
DNA backbone. As a result, the adjacent phosphate units or/and the nucleobase
heteroatoms might directly participate in coordination and drastically influence the
catalytic properties of the transition metal complex. Since mono- and divalent cations
were used in large excess in the catalytic reactions, it has been assumed that the
phosphodiester linkages were apparently shielded. In this case, it is mainly the
nucleobase nitrogen and oxygen donor atoms to compete for the free metal coordination
sites. To support this hypothesis, we have selected DNA/DNA and DNA/RNA hybrid
systems (Chapter 3.2.3.4) in which the complementary strand sequesters most of the
4 Conclusions and Oulook
123
nucleobase heteroatoms through Watson-Crick base-pairing, making them inaccessible
for metal coordination. In addition, structural diversity has been introduced in the DNA
systems and oligonucleotide sequences have been designed that upon simple
hybridization induced the formation of three or four nucleotides bulge motifs. In these
constructs, the transition metal complex is embedded in the double helix structure, and
additionally located within the bulge in the sense strand or flanked by the bulge of the
antisense strand (Figure 3.25). In all cases, no observable catalytic effect of the DNAanchored iridium(I)-PHOX complex was obtained (Chapter 3.2.3.4). Based on these
results we concluded that the N7 nitrogen atoms from the non-Watson Crick edge of the
purines might be the single sites prone to direct coordination and formation of
catalytically inactive species. To shed light on these observations, we evaluated the
putative interaction sphere of the iridium ion with the nucleobases within the DNA
helical model, based on the translational and rotational mobility induced by the covalent
tethering of the metal complex (Figure 3.26). Indeed, several 2’-deoxyguanosine
residues are directly located within this domain, in close proximity to the ligand
attachment site. Notably, by exchanging particular neighbouring dGs for either the 7deaza-riboguanosine or 2’-deoxyadenosine, the catalytic performance of the Ir-PHOX
complexes anchored to the DNA could be restored (Chapter 3.2.3.4). Interestingly, the
dG nucleotide directly involved in Watson-Crick base-pairing with the complementary
PHOX-appended dC is apparently accepted for maintaining efficient catalytic
properties. However, in all cases, no chiral induction was obtained starting from
racemic substrates, probably owing to the lack of structural constraints and also intimate
contact between the DNA helix and the catalytic metal centre. This might be explained
by the flexibility and rather long distance introduced by the linkers carrying the metal
complex, although a two-carbon spacer was expected to confer stereodiscriminating
abilities on the system.
The results presented here are the first examples on applications of oligonucleotidesbased ligands in organometallic catalysis and they contribute to the fundamentals of
exploring the potential of nucleic acids in asymmetric transformations. Several
possibilities could therefore be considered for the future. Firstly, the catalytic
performance of the DNA-based ligands can be improved by rationally designing the
nucleotide sequence, the modification position, the tether length and the secondary
4 Conclusions and Oulook
124
structure elements. For example, DNA constructs ranging from simple, structurally
distinct DNA/DNA and DNA/RNA hybrids to nucleic acid systems containing highlystructured motifs (e.g. branched DNA, hairpins, internal loops, three- and four-way
junctions) (Figure 4.3) offer access to a large variety of chiral, readily available targets
for covalent anchoring of transition metal complexes.
Figure 4.3. Proposed structures for embedding transition metal complexes and construction of DNAbased catalysts. A) DNA/DNA or DNA/RNA duplexes. B) Double helix structure with internal loop. C)
Three-stem junction DNA structure. D) Four-strem junction DNA structure. E) DNA G-Quadruplex. F)
Hydrogen bonding pattern of the G-tetrad.
Our modular synthetic strategy of DNA-ligand assembling from convertible nucleotide,
diamine and functionalized ligand components could be further tuned for introducing
shorter or conformationally constrained tethers and, finally creating selective DNAbased catalysts. Moreover, the data described above provide valuable information
regarding the nucleobase requirements for adequate DNA sequence design and
minimization of the critical first sphere coordination interactions that lead to inhibition
4 Conclusions and Oulook
125
of the catalytic activity. Nevertheless, the distinct structural features of the Gquadruplexes (structural elements found in the telomere ends of the chromosomes)
(Figure 4.3) might offer an intriguing alternative approach to the design of G-poor
stretches as a tool of preventing purine coordination. More in detail, one poly(G) strand
can assemble in a three-dimensional structure containing two (or more) G-quartets (or
G-tetrads), the strand contributing four G residues to each G-tetrad. The guanine tetrads
can stack upon each other to form four-stranded structures with a guanine tetrad core.
This type of intermolecular arrangement may adopt a basket-like confomation, in which
a distinct loop connecting diagonally related strands is formed (Figure 4.3). This
particular structural element could be engineered to accomodate a phosphine ligandtethered dC residue and to allow for subsequent location of the transition metal ion in a
precise, resourceful cavity. Since each G residue of the core is involved in both WatsonCrick and Hoogsteen base-pairing (Figure 4.3), one might assume that a G-quadruplex
would also provide a favourable coordination environment. The excellent recognition
properties of G-quadruplexes could then be used for effecting transfer of chirality to the
metal-catalyzed reaction.
The de novo synthesis of nucleic acid catalysts entirely based on rational design,
combined with the rather time-consuming screening for activity and selectivity can be a
difficult task. Combinatorial techniques, often called “shotgun” techniques, offer a
valuable alternative based on the probability that the desired catalyst is represented in a
library of randomly synthesized molecules. The experimenter’s efforts required by the
conventional screening of rationally designed sequences are eliminated in a SELEXtype strategy, by the use of “column chromatography” methods to isolate the
catalytically active species, as described at the beginning of this chapter. Hence,
conducting in vitro selection of RNA hybrid catalysts, assisted by the ligand-DNA
carrier and the transition metal precursor, is anticipated to facilitate the discovery of
novel catalysts for organometallic transformations. The capacity of RNA to fold,
provide particular coordination cavities and binding pockets and thereby to exquisitely
tailor the first- as well the second coordination sphere of the active site is expected to be
explored through combinatorial strategies. The optimal transfer of chiral information
from the RNA scaffold to the chemical reaction can be then achieved. The RNA catalyst
could be subsequently evolved to a more selective/active species to generate artificial
4 Conclusions and Oulook
126
ribozymes with custom-made properties. The SELEX scheme proposed in this work
implies selection of RNA species exclusively based on their ability of accelerating the
desired transformation. The isolated catalysts will be then submitted to screening
systems for determining the level of stereoselectivity. Ideally, besides effecting control
on the electronic properties, the selected RNA molecules will also be able to impart the
desired steric information. Alternatively, a methodology that allows for simultaneous
selection for stereoselectivity has to be implemented.
With only little precedent in the field of nucleic acid-based hybrid catalysts, the results
reported here represent a step forward in the development of metallo-ribozymes and deoxyribozymes and allow new research at the interface between the fields of transition
metal catalysis and biocatalysis.
5.1 Standard Methods and Reagents
127
5
Materials and Methods
5.1
Standard Methods and Reagents
Standard methods, such as DNA/RNA ethanol precipitation, polyacrylamide gel
electrophoresis (Rotiphorese DNA sequencing system), NAP G25 - gel filtration,
spectrophotometric quantification of oligonucleotides, gel elution of nucleic acids, UVshadowing, were carried out according to published protocols.[308, 309] All reagents were
purchased from Aldrich, Fluka, Acros Organics or Proligo (for oligonucleotide
synthesis) and used without further purification. DMF and THF were purchased from
Fluka in septum sealed bottles and kept under inert atmosphere (dry solvents over
molecular sieves).
Reactions with air-sensitive compounds were performed under argon atmosphere using
standard Schlenk techniques. Degassing of solvents and reaction mixtures containing
O2-sensitive phosphines was achieved through a minimum of three successive freezepump-thaw cycles.
TLC analyses were carried out using silica gel plates Polygram® Sil G/UV254
(40×80 mm) from Macherey-Nagel. Flash chromatography was carried out on silica gel
40 μm from J.T. Baker. NMR spectra were recorded on Mercury Plus 300, Varian
VNMR S 500, Bruker AC-300, or DRX-300 spectrometers. 1H and
13
C{1H} NMR
spectra were calibrated to TMS on the basis of the relative chemical shift of the solvent
as an internal standard.
31
P{1H} NMR spectra were calibrated to an external standard
(85% H3PO4). Abbreviations used are as follows: s = singlet, d = doublet, t = triplet, m
= multiplet, bs = broad singlet, bd = broad doublet. FAB and EI mass spectra were
recorded on a JEOL JMS-700 sector field mass spectrometer. MALDI-TOF mass
spectra were recorded on a Bruker BIFLEX III spectrometer. ESI MS analysis for small
compounds
was
performed
on
a
Finnigan
MAT
TSQ
700
spectrometer.
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
5.2
128
Synthesis of Phosphorus Ligands and their Transition
Metal Complexes
(2-Diphenylphosphine)-4-benzoic
phenyl)phenylphosphine
acid
(TPPDS)
L7
L1
and
disodium
were
purchased
from
bis(4-sulfonatoSigma-Aldrich.
Bisphosphines L2[18, 19] and L3[50, 310] are derivatives of the well-known ligands pyrphos
and BINAP, and were prepared according to literature procedures. Compound L2 was
synthesized starting from (R,R)-3,4-bis(diphenylphosphano)pyrrolidine. For L3 and L8
preparation, the commercially available (S)-2,2’-dihydroxy-1,1’-binaphthalene was
used.[250]
Phosphinooxazoline
ligand
(S)-2-[2-(diphenylphosphino)phenyl]-4-(1-
methylethyl)-4,5-dihydrooxazole) (L4), [Pt(cod)Cl]2, PdCl2(PhCN)2, [Rh(cod)Cl]2,
[Rh(nbd)Cl]2, [Rh(C2H4)2Cl]2 and [Ir(cod)Cl]2 were purchased from Strem Chemicals.
Rhodium(I)-complex [RhL8(nbd)]+BF4- was prepared in our group.[250] Compounds L14 and L7-8 are shown in Figure 5.1.
Figure 5.1. Phosphines L1 and L7, bisphosphines L2-3 and L8, and phosphinooxazoline L4 ligands.
5.2.1
Synthesis of Phosphoramidite Ligands
The synthesis of phosphoramidite ligands P1-3 (Figure 5.2) was accomplished
according to published procedures,[311,
312]
by heating the commercially available
neopentyl glycol (for P1) or 2,2´-biphenol (for P2 and P3) with phosphorus trichloride,
followed
by
treatment
with
appropriate
amines
(diethylamine,
or
N,N’-
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
129
diisopropylamine) according to Scheme 3.1.
Figure 5.2. Phosphoramidites P1-3.
General procedure for the synthesis of phosphoramidites P1-3. A solution of diol
(5.1 mmol), neopentyl glycol for P1 (1a), or 2,2´-biphenol for P2,3 (1b) in PCl3 (4 mL,
45 mmol, 8.8 equiv) was refluxed for 4 h. The excess of PCl3 was removed by
distillation and the residual foam consisting of the phosphoryl chloride of the diol was
diluted with toluene and concentrated (3 x 5 mL) to remove the excess of PCl3. The
resulting yellow oil was dissolved in dry toluene (4 mL) and added to a solution of
diamine (6.6 mmol, 1.3 equiv DEA (680 µL) for P1,2 or DIPA (920 µL) for P3) and
TEA (3 ml, 21 mmol, 4.1 equiv) in 5 mL dry THF. After being stirred for 16 h at room
temperature, the reaction mixture was concentrated under vacuum and the crude product
was purified by flash chromatography (column preconditioned with the eluent
containing 1% TEA).
1,3-Propanediol-2,2-dimethyl-N,N’-diethylphophoramidite (P1) The crude product
was purified by flash chromatography (elution with EA/n-hex 3:97). Yield: 32%.
Colorless oil. 1H NMR (300 MHz; acetone-d6) δ 3.81 (bd, 2H), 3.62 (dd, 2H), 3.13 (m,
4H), 1.21 (s, 3H), 1.07 (t, 6H), 0.75 (s, 3H). 31P NMR (122 MHz; acetone-d6) δ 147.62.
O,O’-(1,1’-Biphenyl-2,2-diyl)-N,N’-diethylphophoramidite (P2) The crude was
purified by flash chromatography (elution with EA/n-hex 4:96). Yield: 42%. White
solid. 1H NMR (300 MHz; DMSO-d6) δ 7.07 (d, 2H), 6.95 (t, 2H), 6.82 (t, 2H), 6.73 (d,
2H), 2.50 (m, 4H), 0.54 (t, 6H). 31P NMR (122 MHz; DMSO-d6) δ 149.67.
O,O’-(1,1’-Biphenyl-2,2-diyl)-N,N’-diisopropylphophoramidite (P3). The crude
product was purified by flash chromatography (elution with EA/n-hex 5:95). Yield:
46%. White solid. 1H NMR (300 MHz; acetone-d6) δ 7.58 (d, 2H), 7.44 (m, 2H), 7.28
(dd, 4H), 3.58 (m, 2H), 1.25 (d, 12H). 31P NMR (122 MHz; acetone-d6) δ 152.12.
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
5.2.2
130
Synthesis of PHOX Ligands
Phosphinooxazolines L5,6 (Figure 5.3) were synthesized starting from commercially
available
2-(diphenylphosphino)-benzoic
acid
2a
and
L-serine
methyl
ester
hydrochloride (H-L-Ser-OMe) 3, followed by oxazoline ring formation in the presence
of Burgess’s reagent (Route A). An alternative synthetic route was also employed,
starting from commercially available 2-iodo-benzoic acid 2b and H-L-Ser-OMe (Route
B), followed by oxazoline ring formation and palladium-catalyzed P-C cross coupling
reaction with diphenylphosphine,[270] according to Scheme 3.5.
Figure 5.3. Phosphinooxazolines L5-6.
Route A
Preparation of (S)-N-(2-Hydroxy-1-carboxymethyl-ethyl)-2-(diphenylphosphino)benzamide (4). To a stirred solution of 2-(diphenylphosphane)benzoic acid 2a (1.5 g,
4.9 mmol, 1.1 equiv) and L-serine methyl ester hydrochloride 3 (0.693 g, 4.45 mmol) in
CH2Cl2 (40 mL) were added TEA (0.68 mL, 4.9 mmol, 1.1 equiv) and EDC (0.94 g, 4.9
mmol, 1.1 equiv). The reaction mixture was stirred for 4 h at r.t., until the starting
material was consumed according to TLC (EA/n-hex 1:1). The mixture was diluted with
100 ml CH2Cl2, washed with 5% NaHCO3 (50 mL), 1 M HCl (50 mL) and brine (2 × 50
mL), dried over Na2SO4, filtered and concentrated under reduced pressure. The product
was chromatographed on silica gel eluting with EA/n-hex 1:1 to yield the amide 4 (1.6
g, 3.92 mmol, 80%) as a white, amorphous solid. 1H NMR (500 MHz; CDCl3) δ 7.65
(dd, J = 7.4, 3.7 Hz, 1H), 7.42-7.29 (m, 12H), 7.01 (dd, J = 7.6, 4.3 Hz, 1H), 6.87 (bd, J
= 7.0, 1H), 4.71 (m, 1H), 3.88 (m, 2H), 3.74 (s, 3H), 2.82 (bs, 1H).
13
C NMR (126
MHz; CDCl3) δ 170.62, 168.88, 140.79 (d, JC,P = 25.3 Hz), 136.38 (d, JC,P = 18.9 Hz),
136.31 (d, JC,P = 19.1 Hz), 135.58 (d, JC,P = 18.3 Hz), 134.23, 133.98, 133.82, 133.66,
130.52, 129.14, 129.03, 128.97, 128.79, 128.74, 128.66, 128.61, 127.87, 127.83, 62.86,
55.35, 52.73.
31
P NMR (202 MHz; CDCl3) δ -10.51. FAB MS: m/z 408.1 [M]+ (calcd
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
131
for [C23H22NO4P]+ 408.13).
Preparation of (S)-Methyl-2-(2-diphenylphosphino-phenyl)-4,5-dihydrooxazolo-4carboxylate (L5). To a stirred solution of 4 (1.0 g, 2.5 mmol) in dry THF (20 mL) was
added
(methoxycarbonylsulfamoyl)triethylammonium
hydroxide,
inner
salt[313]
(Burgess’ reagent, 0.703 g, 2.95 mmol, 1.2 equiv). After being refluxed for 4 hours
(TLC control: EA/n-hex 3:7), the reaction mixture was allowed to cool down to room
temperature and diluted with 200 mL EA. The resulting solution was washed with water
(2 × 100 mL) and brine (100 mL), and dried over Na2SO4. Removal of the solvent under
reduced pressure afforded the crude product as brownish oil. Purification by flash
chromatography (elution with EA/n-hex 3:7, column preconditioned with the eluent
containing 1% TEA) gave phosphinooxazoline L5 as colorless oil (0.401 g, 1.13 mmol,
42%). 1H NMR (500 MHz; CDCl3) δ 7.92 (ddd, J = 7.5, 3.5, 1.5 Hz, 1H), 7.36-7.30 (m,
12H), 6.91 (ddd, J = 7.6, 4.3, 1.0 Hz, 1H), 4.69 (dd, J = 10.6, 8.2 Hz, 1H), 4.38 (t, J =
8.4 Hz, 1H), 4.26 (dd, J = 10.5, 8.6 Hz, 1H), 3.68 (s, 3H). 13C NMR (126 MHz; CDCl3)
δ 171.08, 166.45, 139.30 (d, JC,P = 25.9 Hz), 137.69 (d, JC,P = 11.9 Hz), 137.50 (d, JC,P
= 10.4 Hz), 134.28, 134.11, 134.02, 133.85, 133.67 (d, JC,P = 1.9 Hz), 130.95 (d, JC,P =
19.0 Hz), 130.94, 130.36 (d, JC,P = 2.8 Hz), 128.74, 128.61, 128.52, 128.46, 128.41,
128.35, 128.00, 69.10, 68.45, 52.50. 31P NMR (202 MHz; CDCl3) δ -4.80. EI MS: m/z
389.0 [M]+ (calcd for [C23H20NO3P]+ 389.12).
Preparation
of
(S)-2-(2-diphenylphosphino-phenyl)-4,5-dihydrooxazolo-4-
carboxylic acid sodium salt (L6). Compound L5 (0.35 g, 0.90 mmol) was stirred in a
0.5 M solution of NaOH (3 mL) 6 hours at r.t. The reaction mixture was diluted with
water (3 mL) and the product was precipitated by slow addition of acetone. After
filtration and drying under vacuum, 0.34 g of the sodium salt L6 was recovered (0.85
mmol, 95%, white solid). 1H NMR (500 MHz; D2O) δ 7.66 (dd, J = 7.4, 3.0 Hz, 1H),
7.10 (t, J = 7.5 Hz, 1H), 6.98-6.78 (m, 10H), 6.70 (t, J = 7.5 Hz, 1H), 6.55-6.50 (m,
1H), 4.17-4.11 (m, 1H), 3.97 (t, J = 8.3 Hz, 1H), 3.88-3.81 (m, 1H).
13
C NMR (126
MHz; D2O) δ 178.16, 166.89, 137.24 (d, JC,P = 19.7 Hz), 136.11 (d, JC,P = 7.9 Hz),
135.87 (d, JC,P = 7.6 Hz), 133.88, 133.75, 133.72, 133.59, 133.15, 131.73, 131.39,
131.23, 130.91, 130.24, 129.06, 128.73, 128.61, 128.59, 128.56, 71.20, 69.56. 31P NMR
(202 MHz; D2O) δ -6.57. ESI MS: m/z 374.2 [M-Na]- (calcd for [C22H17NO3P]374.10). Acidification results in oxazoline ring opening.[271]
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
132
Route B
Preparation of (S)-N-(2-Hydroxy-1-carboxymethyl-ethyl)-2-iodo-benzamide (5). To
a stirred solution of 2-iodo-benzoic acid 2b (1.5 g, 6.05 mmol, 1.1 equiv) and L-serine
methyl ester hydrochloride 3 (0.856 g, 5.50 mmol) in CH2Cl2 (40 mL) were added TEA
(0.84 mL, 6.05 mmol, 1.1 equiv) and EDC (1.16 g, 6.05 mmol, 1.1 equiv). The reaction
mixture was stirred 2 h at r.t., until the starting material was consumed according to
TLC (EA/n-hex 3:7). The mixture was diluted with 100 ml CH2Cl2, washed with 5%
NaHCO3 (50 mL), 1 M HCl (50 mL) and brine (2 × 50 mL), dried over Na2SO4, filtered
and concentrated under reduced pressure. The product was chromatographed on silica
gel eluting with EA/n-hex 3:7 to yield the amide 5 (1.8 g, 5.16 mmol, 93%) as a white,
amorphous solid. 1H NMR (300 MHz; CDCl3) δ 7.87 (m, 1H), 7.46-7.36 (m, 2H), 7.23
(ddd, J = 8.0, 7.2, 1.9 Hz, 1H), 6.77 (bd, 1H), 4.86 (td, J = 7.2, 3.4, 1H), 4.11 (d, J =
3.2, 2H), 3.83 (s, 3H), 2.51 (bs, 1H).
13
C NMR (126 MHz; CDCl3) δ 170.52, 169.31,
144.18, 139.93, 131.49, 128.37, 128.24, 92.41, 63.11, 55.13, 52.90. FAB MS: m/z 349.9
[M]+ (calcd for [C11H12INO4]+ 349.98).
Preparation of (S)-Methyl-2-(2-iodo-phenyl)-4,5-dihydrooxazolo-4-carboxylate (6).
To a stirred solution of 5 (1.5 g, 4.3 mmol) in dry THF (25 mL) was added
(methoxycarbonylsulfamoyl)triethylammonium hydroxide, inner salt (Burgess’ reagent,
1.23 g, 5.16 mmol, 1.2 equiv). After being refluxed for 4 hours (TLC control: EA/n-hex
1:4), the reaction mixture was allowed to cool down to room temperature and diluted
with 200 mL EA. The resulting solution was washed with water (2 × 100 mL) and brine
(100 mL), and dried over Na2SO4. Removal of the solvent under reduced pressure
afforded the crude product as brownish oil. Purification by flash chromatography
(elution with EA/n-hex 1:4, column preconditioned with the eluent containing 1% TEA)
gave oxazoline 6 as colorless oil (0.852 g, 1.13 mmol, 60%). 1H NMR (300 MHz;
CDCl3) δ 7.93 (dd, J = 8.0, 1.2 Hz, 1H), 7.66 (dd, J = 7.7, 1.7 Hz 1H), 7.38 (dt, J = 7.6,
7.6, 1.2 Hz, 1H), 7.09-7.15 (m, 1H), 5.00 (dd, J = 10.6 Hz, 8.0 Hz, 1H), 4.62 (dd, J =
10.6, 8.8 Hz, 1H), 3.83 (s, 3H). 13C NMR (126 MHz; CDCl3) δ 171.22, 166.87, 140.49,
132.79, 132.01, 131.02, 127.78, 94.49, 69.79, 68.67, 52.72. EI MS: m/z 331.0 [M]+
(calcd for [C11H10INO3]+ 330.97).
Preparation of (S)-Methyl-2-(2-diphenylphosphino-phenyl)-4,5-dihydrooxazolo-4-
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
133
carboxylate (L5). In a Schlenk flask, 6 (800 mg, 2.42 mmol), dry DMF (4 mL) and
TEA (370 µL, 2.66 mmol, 1.1 equiv) were charged together and the resulting solution
degassed. After addition of diphenylphosphine 7 (500 µL, 2.90 mmol, 1.2 equiv), the
solution was heated up to 80°C. A solution of Pd(OAc)2 (10.87 mg, 48.4 µmol) in 2 mL
DMF was separately prepared and degassed. 1 mL from this solution (1 mol% final
concentration of Pd catalyst) was added to the reaction mixture. The deep purple
solution was further heated at 80°C until completion (monitored by TLC: EA/n-hex 3:7;
4 h). The reaction mixture was allowed to cool down at room temperature and diluted
with 100 mL EA. The resulting solution was washed with brine (2 x 50 mL), the
organic phase transferred via a stainless steel cannula in a Schlenk flask, and dried over
Na2SO4. The solvent was removed under vacuum, and the residue loaded onto a
silicagel column (elution with EA/n-hex 3:7, column preconditioned with the eluent
containing 1% TEA) to obtain phosphinooxazoline L5 a colorless oil (1.77 mmol, 688
mg, 73%).
5.2.3
Synthesis
of
Palladium(II)-
and
Platinum(II)-Phosphine
Complexes
Dichlorobis[(4-carboxphenyl)diphenylphosphine]-palladium(II) (8). Reaction of L1
(49.0 mg, 0.16 mmol, 2.0 equiv) with PdCl2(PhCN)2 (30.7 mg, 0.08 mmol) in 4 mL
acetonitrile, overnight, at room temperature, gave a yellow powder. The solid was
filtered, washed with diethylether and dried, yielding 46.0 mg (0.06 mmol, 73%)
Pd(L1)2Cl2 complex 8. 1H NMR (300 MHz; DMSO-d6) δ 13.71 (bs, 2H), 7.99-7.97 (d,
4H), 7.76-7.31 (m, 24H). 31P NMR (122 MHz; DMSO-d6) δ 24.58.
Dichlorobis[(4-carboxphenyl)diphenylphosphine]-platinum(II) (9). The platinum
complex was synthesized according to published procedure,[314] using [Pt(cod)Cl]2 and
monophosphine L1. 1H NMR (300 MHz; CD3OD) δ 7.80-7.73 (bd, 4H), 7.60-7.57 (m,
8H), 7.47-7.40 (m, 8H), 7.32-7.29 (m, 8H).
31
P NMR (122 MHz; CD3OD) δ 14.34
(satellite due to 34% 195Pt, JPt,P = 1852.2 Hz).
A similar procedure was used to prepare Pt(L1)2Cl2 complex 9 in 9:1 acetonitrile/H2O.
[Pt(cod)Cl]2 (29.9 mg, 0.08 mmol) was dissolved in 0.6 ml acetonitrile and then treated
with 3.4 mL aqueous solution (1:0.7 acetonitirile/100 mM TEAA in 80% acetonitrile
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
134
(buffer B)) of L1 (98.0 mg, 0.32 mmol, 4.0 equiv) overnight, at room temperature, until
a yellow precipitate was formed. The mixture was filtered, the resulting pale yellow
solid washed with diethylether and dried, yielding 82.5 mg (0.10 mmol, 62%)
Pt(L1)2Cl2 complex 9. 1H NMR (300 MHz; DMSO-d6) δ 8.04-6.55 (m, aromatic).
31
P
NMR (122 MHz; DMSO-d6) δ 18.44.
5.2.4
Synthesis of Platinum(II)-, Palladium(II)- and Rhodium(I)-
PYRPHOS Complexes
Dichlorobis[(R,R)-N-(4-carboxylbutanoyl)-3,4-bis(diphenylphosphino)pyrrolidine]palladium(II) (10). To a solution of Pd(PhCN)Cl2 (38.4 mg, 0.1 mmol) in
dichloromethane (2 mL), under argon, was added solid PYRPHOS L2 (55.3 mg, 0.1
mmol) (1.0 equiv bisphoshine unit per Rh). The resulting yellow-orange solution was
stirred at room temperature for 18 h. After filtration over Celite (2 cm), the solution was
concentrated to 0.5 mL under reduced pressure. Upon addition of diethyl ether, an
orange solid precipitated which was then filtered, washed with diethyl ether, and dried
under vacuum (10: yield 78%, 56 mg, 0.07 mmol). 1H NMR (250 MHz; DMSO-d6) δ
7.96-7.48 (m, 20H), 3.46-3.31 (m, 4H), 2.90 (m, 1H), 2.65 (m, 1H), 2.12-1.88 (m, 4H),
1.52 (m, 2H). 31P NMR (101 MHz; DMSO-d6) δ 42.45.
Dichlorobis[(R,R)-N-(4-carboxylbutanoyl)-3,4-bis(diphenylphosphino)pyrrolidine]platinum(II) (11). To a solution of [Pt(cod)Cl]2 (37.4 mg, 0.10 mmol) in
dicholoromethane (2 mL), under argon, was added solid PYRPHOS L2 (110.6 mg, 0.20
mmol). The resulting yellow mixture was stirred at room temperature for 18 h and then
passed through Celite (2 cm). The solvent was partially removed under reduced pressure
and an identical amount of diethyl ether was added. The resulting yellow solid was
filtered, washed with ether and dried under vacuum (11: yield 83%, 68 mg, 0.08 mol).
1
H NMR (250 MHz; CD3OD) δ 7.77-7.39 (m, 20H), 3.61-3.39 (m, 4H), 2.88 (m, 1H),
2.68 (m, 1H), 2.09-1.89 (m, 4H), 1.61-1.46 (m, 2H). 31P NMR (101 MHz; CD3OD) δ
26.38 (satellite JPt,P = 1162.5 Hz).
Dichlorobis[(R,R)-N-(4-carboxylbutanoyl)-3,4-bis(diphenylphosphino)pyrrolidine]rhodium(I) (12). To a degassed solution of [Rh(cod)Cl]2 (24.6 mg, 0.05 mmol) in
acetonitirile (1 mL) was added a solution of L2 (55.3 mg, 0.1 mmol, 1.0 equiv
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
135
bisphosphine unit per Rh) in degassed acetonitirile/water (6:4, 1 mL). The resulting
solution was stirred overnight, at room temperature. An orange solid precipitated which
was then filtered, washed with diethylether and dried, yielding 52.0 mg (0.07 mmol,
72%) Rh(L2)2Cl2 complex 12. 1H NMR (250 MHz; CD3OD) δ 7.66-7.21 (m, 20H),
3.43-3.38 (m, 4H), 2.82 (t, 1H), 2.64 (t, 1h), 2.08-1.91 (m, 4H), 1.60-1.54 (m, 2H). 31P
NMR (101 MHz; CD3OD) δ 38.16 (m, 1P), 36.84 (m, 1P).
[(Bicyclo[2.2.1]hepta-2,5-diene)-[(R,R)-N-(4-carboxylbutanoyl)-3,4-bis(diphenylphosphino)pyrrolidine]-rhodium(I) (13). [Rh(nbd)Cl]2 (20.8 mg, 45.2 µmol) and
AgBF4 (17.6 mg, 90.3 µmol, 2.0 equiv) were dissolved in freshly degassed acetone (5.5
mL) and stirred for 45 min at room temperature. The precipitated AgCl was filtered off
(G4) and the filtrate was immediately degassed. To the resulting yellow solution was
added solid PYRPHOS L2 (50.0 mg, 90.3 µmol, 2.0 equiv). The solution turned
immediately deep orange. After stirring for 2 h at room temperature, the mixture was
concentrated under reduced pressure and treated with diethyl ether until weak turbidity
appeared. After overnight storage at -20°C, an orange solid was formed. The precipitate
was filtered, washed with ether, and dried under vacuum. The desired [Rh(nbd)L2]+BF4complex 13 was obtained as yellow-orange powder, in 72% yield (60 mg, 71 µmol). 1H
NMR (300 MHz; Acetone-d6) δ 10.43 (bs, 1H), 8.07-7.24 (m, 20H), 5.66 (bs, 1H, nbd),
5.11 (bs, 1H, nbd), 4.22-3.86 (m, 2H, nbd), 3.40 (dd, 4H), 3.05 (bs, 1H), 2.83 (bs, 2H,
nbd), 2.61 (bs, 1H), 2.25-2.18 (m, 4H), 1.95-1.64 (m, 2H), 1.19 (m, 2H, nbd). 31P NMR
(122 MHz; Acetone-d6) δ 36.92 (m, 1P), 35.63 (m, 1P). ESI MS (sample dissolved in
acetonitrile): m/z 738.15 [M]+ (calcd for Rh(L2)(acetonitrile)2+ [C37H39N3O3P2Rh]+
738.16).
5.2.5
Synthesis of Rhodium(I)- and Iridium(I)-PHOX Complexes
[(Bicyclo[2.2.1]hepta-2,5-diene)-[(S)-methyl-2-(2-diphenylphosphino-phenyl)-4,5dihydrooxazolo-4-carboxylate]rhodium(I)]-tetrafluoroborate (14). In a Schlenk
flask (under argon), 2,5-norbornadiene-rhodium(I) chloride ٛ dimer, [Rh(nbd)Cl]2
(115.3 mg, 0.25 mmol), was dissolved in degassed acetone (20 mL). After addition of
AgBF4 (97.84 mg, 0.5 mmol, 2.0 equiv), the mixture was stirred at room temperature for
1 hour and then AgCl was filtered off. The resulting solution of [Rh(nbd)Solv2]+BF4-
5.2 Synthesis of Phosphorus Ligands and their Transition Metal Complexes
136
was immediately degassed. Separately, PHOX ligand L5 (101 mg, 0.26 mmol, 1.04
equiv) was dissolved in degassed acetone (2 mL) and pre-formed [Rh(nbd)Solv2]+BF4in acetone (10 mL, 0.25 mmol) added. The resulting mixture was stirred at room
temperature, for 14 hours, and then concentrated to small volume under reduced
pressure.
Figure 5.4. Phosphinooxazoline-rhodium(I) and -iridium(I) complexes.
Addition of diethyl ether resulted in precipitation of an orange solid that was filtered in
air, without protection against oxygen, washed with ether and dried under vacuum,
affording [Rh(nbd)L5]+BF4- complex 14 in 65% yield (95 mg, 0.16 mmol). 1H NMR
(300 MHz; CDCl3) δ 8.20 (m, 1H), 7.83 (m, 1H), 7.34-7.63 (m, 11H), 7.12 (m, 1H),
5.92 (bs, 1H, nbd), 5.73 (bd, 1H, nbd), 5.06 (t, 1H), 4.95 (m, 1H), 4.70 (m, 1H), 3.96
(bs, 2H, nbd), 3.58 (bs, 1H, nbd), 3.44 (m, 1H, nbd), 1.50 (m, 2H, nbd). 31P NMR (122
MHz; CDCl3) δ 31.91 (d, JP,Rh = 168.9 Hz). ESI MS: m/z 584.09 [M]+ (calcd for
[C30H28NO3PRh]+ 584.09).
[(1,5-cycloctadiene)-[(S)-methyl-2-(2-diphenylphosphino-phenyl)-4,5-dihydrooxazolo-4-carboxylate]iridium(I)]-tetrafluoroborate (15). Similar procedure as for
the preparation of the complex 14: [Ir(cod)Cl]2 (215.3 mg, 0.32 mmol) and AgBF4
(250.1 mg, 0.64 mmol, 2.0 equiv) dissolved together in acetone (28 mL), then stirring
for 1 hour at room temperature. After filtration of AgCl, the resulting solution of
[Rh(nbd)Solv2]+BF4- (23 mL) was added to L5 (300 mg, 0.77 mmol, 1.2 equiv per
iridium ion). The complex formation was carried out at room temperature, for 1 hour.
The workup was performed as described (see complex 14), affording [Ir(cod)L5]+BF4complex 15 as red solid, in 78% yield (384.1 mg, 0.50 mmol). 1H NMR (500 MHz;
DMSO-d6) δ 8.70-8.68 (m, 1H), 8.28-8.26 (m, 2H), 8.12-7.92 (m, 9H), 7.54-7.50 (m,
2H), 5.60 (dd, J = 10.1, 4.1 Hz, 1H), 5.42 (dd, J = 9.3, 4.1 Hz, 1H), 5.24 (m, 1H), 3.97
5.3 Oligonucleotides
137
(s, 2H, cod), 3.75 (bs, cod), 2.92 (m, 4H, cod). 31P NMR (122 MHz; DMSO-d6) δ 14.95.
General procedure for the in situ preparation of (phosphinooxazoline)iridium(I)
complexes in 100% dioxane and 3:7 dioxane/H2O . Under argon atmosphere, 2.5
µmol [Ir(cod]Cl]2 (6.7 mg) were added to a solution of 5.5 µmol phosphinooxazoline
ligand (8.2 mg L4 or 8.5 mg L5, 2.2 equiv) in 0.5 ml degassed dioxane or 3:7
dioxane/H2O mixture. The redish purple resulting solution was stirred for 30 min, at
room temperature and then directly subjected to 31P NMR analysis.
Phosphinooxazoline(L4)iridium (I) complex. 31P NMR (202 MHz) δ 10.27 (dioxane,
10% CDCl3); 15.64 (3:7 dioxane/H2O, 10% D2O).
Phosphinooxazoline(L5)iridium (I) complex.
31
P NMR (202 MHz) δ 8.84 (dioxane,
10% CDCl3); 15.16 (3:7 dioxane/H2O, 10% D2O).
5.3
Oligonucleotides
DNA and RNA sequences employed in this work for preparation of single and doublestranded constructs carrying metal chelating moieties are shown in Table 5.1, in 5’ to 3’
orientation. Oligodeoxynucleotides ODN1-5 (19mer) and their complementary strands
cDNA1-3 (19, 23, and 16mers) were prepared by standard automated solid-phase
synthesis. Unmodified complementary 19mer sequences cDNA4 (ε = 195700 L·mol1
·cm-1), cRNA1 (ε = 173 800 L·mol-1·cm-1), and cRNA (ε = 181900 L·mol-1·cm-1) were
obtained from IBA in 1 µmol scale synthesis, as double HPLC purified solution.
Table 5.1. Oligonucleotide sequences.
ODN
Sequence
ODN1a
ODN1b
R
H2N
5'-GC AGT GAA GGCR TGA GCT CC-3'
H2N
ODN1c
ODN2
ODN3[a]
H2N
2O
R
5'-GC AGT GAA GGC TGA GCT CCT AC C-3'
H2N
5'-GC AGT GAA GGC TGA GCT CCS CRC-3'
H2N
ODN4a
ODN4b
O
H2N
5'-GC AGC GAT AACR TAA GCG CT-3'
ODN4c
H2N
H2N
ODN5a[b]
5'-GC AGT GAA XXCR TXA GCT CC-3'
ODN5b[b]
5'-GC AGT GAA XXCR TXA GCT CC-3'
cDNA1
5'-GG AGC TCA GCC TTC ACT GC-3'
O
2O
H2N
H2N
-
5.3 Oligonucleotides
cDNA2
cDNA3
cDNA4
cRNA1
cRNA
138
-
5'-GG AGC TCA CAA GTC CTT CAC TGC-3'
5'-GG AGC TCC TTC ACT GC-3'
5'-AG CGC TTA GTT ATC GCT GC-3'
5'-GG AGC UCA GCC UUC UCA GC-3'
5'-AG CGC UUA GUU AUC GCU GC-3'
[a] A decaethylene glycol spacer S was incorporated during solid phase synthesis. [b] X = 7-deaza-riboG
5.3.1
Automated Solid-Phase Synthesis
Solid-phase DNA synthesis was performed on an ExpediteTM 8909 automated
synthesizer using the conventional phosphoramidite chemistry,[315] dC or dG (t-butylphenoxyacetyl, TAC) controlled pore glass support (40 μmol/g, 500Å) and βcyanoethyl-phosphoramidites containing base-labile TAC-protecting groups (Proligo).
The decaethyleneglycol phosphoramidite S was prepared in our lab.[250] 4-Triazolyldeoxyuridine phosphoramidite was purchased from Glen Research and 7-deazaguanosine phosphoramidite was obtained from ChemGenes. The exocyclic amine in 7deaza-guanosine phosphoramidite was protected with the standard iso-butyryl group.
The 2’-hydroxyl group was protected as a t-butyldimethylsilylether. Standard reagents
employed in DNA solid-phase synthesis (deblocking reagent - dichloroacetic acid in
dichloromethane, activator - dicyanoimidazole, oxidizing reagent - iodine in THF/H2O,
and capping reagent - t-butyl-phenoxyacetanhydride in acetonitrile), as well as
acetonitrile (water content ≤10 ppm) were purchased from Proligo and Sigma Aldrich
Fine Chemicals.
Solid-phase synthesis of ODNs was performed on 1 µmol or 15 µmol scale synthesis,
usually
leaving
the
terminal
4,4’-dimethoxytrityl
(DMT)
group
on.
The
phosphoramidites were used as 0.067 M (DNA monomers), and 0.1 M (RNA monomer)
acetonitrile solutions. The standard protocols provided by Applied Biosystems were
optimized (Tables 5.2 and 5.3).
5.3 Oligonucleotides
139
Table 5.2. Protocol for 1 µmol scale solid-phase DNA synthesis (dA cycle).[a]
Step
Function
Mode[b]
Deblocking
144 /*Index Fract. Coll.
0 /*Default
38 /*Diverted Wsh A
141 /*Trityl Mon. On/Off
16 /*Dblk
0 /*Default
16 /*Dblk
38 /*Diverted Wsh A
141 /*Trityl Mon. On/Off
38 /*Diverted Wsh A
144 /*Index Fract. Coll.
1 /*Wsh
2 /*Act
18 /*A + Act
18 /*A + Act
2 /*Act
18 /*A + Act
2 /*Act
0 /*Default
1 /*Wsh
1 /*Wsh
Capping
NA
WAIT
PULSE
NA
PULSE
WAIT
PULSE
PULSE
NA
PULSE
NA
PULSE
PULSE
PULSE
PULSE
PULSE
PULSE
PULSE
WAIT
PULSE
PULSE
Amount
(pulse)
1
0
15
1
20
0
40
60
0
20
2
8
5
5
3
3
2
3
0
7
21
Time
(sec)
0
1.5
0
1
0
20
40
0
1
0
0
0
0
0
24
24
16
24
20
56
0
“Event out ON”
“Wait”
“Flush system with Wsh A”
“START data collection”
“Dblk to column”
“Default”
“Deblock”
“Flush system with Wsh A”
“STOP data collection”
“Flush system with Wsh A”
“Event out OFF”
“Flush system with Wsh”
“Flush system with Act”
“Monomer + Act to column”
“Couple monomer”
“Couple monomer”
“Couple monomer”
“Couple monomer”
“Default”
“Couple monomer”
“Flush system with Wsh”
12 /*Wsh A
13 /*Caps
12 /*Wsh A
12 /*Wsh A
PULSE
PULSE
PULSE
PULSE
20
8
9
21
0
0
23
0
“Flush system with Wsh A”
“Caps to column”
“Cap”
“Flush system with Wsh A”
Oxidizing
15 /*Ox
0 /*Default
12 /*Wsh A
PULSE
WAIT
PULSE
35
0
60
0
20
0
“Ox to column”
“Default”
“Flush system with Wsh A”
Capping
13 /*Caps
12 /*Wsh A
PULSE
PULSE
7
45
0
0
“Caps to column”
“End of cycle wash”
Coupling
Description
[a] Debloking reagent = dblk, acetonitrile = Wsh, WshA, activator = act, capping regents = Caps, oxidizer
= Ox. [b] 1 PULSE = 16 µL.
Table 5.3. Protocol for 15 µmol scale solid-phase DNA synthesis (dA cycle).[a]
Step
Function
Mode[b]
Amount
(pulse)
Time
(sec)
Deblocking
144 /*Index Fract. Coll.
0 /*Default
141 /*Trityl Mon. On/Off
38 /*Diverted Wsh A
16 /*Dblk
0 /*Default
16 /*Dblk
0 /*Default
38 /*Diverted Wsh A
141 /*Trityl Mon. On/Off
144 /*Index Fract. Coll.
12 /* Wsh A
1 /*Wsh
2 /*Act
41 /*Gas B
NA
WAIT
NA
PULSE
PULSE
WAIT
PULSE
WAIT
PULSE
NA
NA
PULSE
PULSE
PULSE
PULSE
1
0
1
50
500
0
500
0
50
0
2
400
40
35
1
0
1.5
1
0
0
20
0
20
0
1
0
0
0
0
20
Coupling
Description
“Event out ON”
“Wait”
“START data collection”
“Flush system with Wsh A”
“Dblk to column”
“Default”
“Dblk to column”
“Default”
“Flush system with Wsh A”
“STOP data collection”
“Event out OFF”
“Flush system with Wsh A”
“Flush system with Wsh”
“Flush system with Act”
“Gas B”
5.3 Oligonucleotides
140
18 /*A + Act
0 /*Default
18 /*A + Act
0 /*Default
2 /*Act
1 /*Wsh
41 /*Gas B
18 /*A + Act
0 /*Default
2 /*Act
1 /*Wsh
PULSE
WAIT
PULSE
WAIT
PULSE
PULSE
PULSE
PULSE
WAIT
PULSE
PULSE
25
0
25
0
20
40
1
8
0
20
100
0
60
0
60
30
0
20
0
60
30
0
“Monomer + Act to column”
“Couple monomer”
“Monomer + Act to column”
“Couple monomer”
“Couple monomer”
“Flush system with Wsh”
“Gas B”
“Monomer + Act to column”
“Couple monomer”
“Couple monomer”
“Flush system with Wsh”
Capping
12 /*Wsh A
13 /*Caps
13 /*Caps
12 /*Wsh A
12 /*Wsh A
PULSE
PULSE
PULSE
PULSE
PULSE
100
75
25
15
100
0
0
15
40
0
“Flush system with Wsh A”
“Caps to column”
“Cap”
“Cap”
“Flush system with Wsh A”
Oxidizing
15 /*Ox
0 /*Default
12 /*Wsh A
PULSE
WAIT
PULSE
125
0
100
0
20
0
“Ox to column”
“Default”
“Flush system with Wsh A”
Capping
13 /*Caps
12 /*Wsh A
PULSE
PULSE
50
340
0
0
“Caps to column”
“End of cycle wash”
[a] Debloking reagent = dblk, acetonitrile = Wsh and WshA, argon = Gas B, activator = act, capping
regents = Caps, oxidizer = Ox. [b] 1 PULSE = 16 µL.
5.3.2
General Procedure for the Synthesis of Amino-Modified ODNs
For the preparation of amino-modified oligonucleotides (ODN), the “convertible
nucleoside approach”[255,
256]
was adapted and optimized. 4-Triazolyl-deoxyuridine
phosphoramidite was assembled at varying internal positions on DNA during automated
solid phase synthesis, in combination with other non-standard phosphoramidite building
blocks (e.g., a decaethyleneglycol spacer molecule S as for ODN3). Base-labile TAC (tbutyl-phenoxyacetyl) protecting groups were used for all natural nucleoside monomers.
Treatment of the fully protected, resin-bound ODN with1 mL aqueous solution of
ethylenediamine (5 M) or 1,4-butanediamine (5 M) at room temperature for 4 h,
afforded the one-pot cleavage from support, deprotection and conversion of the 4triazolyl-dU to different 4-alkylamino-dC. In the case of 1,13-diamino-4,7,10trioxatridecane, the treatment with 0.8 mL neat amine (4 h) was followed by additional
stirring in the presence of 0.5 mL water (5 h). The cleaved products were filtered (0.22
μm membrane filter) and the CPG washed with H2O (3 × 0.5 mL). The resulting
fractions (~2.5 mL) were combined, and after CHCl3 extraction (3 × 1 mL) the DNA
material was passed through a Sephadex G-25 NAP column (Amersham Biosciences)
5.3 Oligonucleotides
141
for removal of remaining organic residues, using water as eluent. The crude ODN was
then purified by reversed-phase HPLC. Fractions containing the tritylated ODN were
collected and lyophilized. The terminal DMT group was removed by treatment with 2%
v/v TFA (1 mL) for 2 min at room temperature. After quenching the acid with NaHCO3,
the ODNs were ethanol precipitated. Desalting on Sephadex G-25 column afforded pure
fully detritylated ODNs (>95% purity), as confirmed by analytical reversed-phase
HPLC. The amount of DNA was quantified by UV spectroscopy (λmax = 260 nm,
εODN1a-c = 181300 L·mol-1·cm-1, εODN2 = 220600 L·mol-1·cm-1, εODN3 = 195700 L·mol1
·cm-1, εODN4a-c = 186000 L·mol-1·cm-1), resulting in overall yields of 21-42%.
Table 5.4. Isolated yields and MALDI-TOF analysis of ODN1-4.
m/z[a]
Entry
Isolated yield
(%)
calcd
obsd
35
5921
5928
ODN1a
42
5895
5898
ODN1b
40
6055
6059
ODN1c
32
7120
7125
ODN2
33
7024
7029
ODN3
25
5866
5856
ODN4a
21
5894
5887
ODN4b
22
6026
6019
ODN4c
[a] ODN1a, ODN2 and ODN3 detected in negative mode ([M-H]-), ODN1b-c and ODN4a-c in positive
mode ([M+H]+).
5.3.3
General Procedure for the Synthesis of 7-deaza-riboG -
containing Amino-Modified ODNs
Chimeric DNA sequences were synthesized using extended coupling times (10 min) for
for more efficient incorporation of the 7-deaza-riboG monomer. The 7-deaza-riboG
coupling cycle performed for 1 µmol scale synthesis is shown in Table 5.5.
Deprotection, cleavage from solid-support and conversion of the triazolyl group were
carried out by treatment with 1 mL aqueous solution of ethylenediamine (5 M) or 1,4butanediamine (5 M) at room temperature, overnight.
5.3 Oligonucleotides
142
Table 5.5. Coupling cycle of 7-deaza-riboG used in 1 µmol scale DNA synthesis.
Step
Function
Mode
NA
WAIT
PULSE
NA
PULSE
WAIT
PULSE
PULSE
NA
PULSE
NA
PULSE
PULSE
PULSE
PULSE
PULSE
PULSE
PULSE
WAIT
PULSE
PULSE
Amount
(pulse)
1
0
15
1
20
0
40
60
0
20
2
8
5
5
5
8
2
3
0
7
21
Time
(sec)
0
1.5
0
1
0
20
40
0
1
0
0
0
0
0
150
150
120
120
20
40
0
Deblocking
144 /*Index Fract. Coll.
0 /*Default
38 /*Diverted Wsh A
141 /*Trityl Mon. On/Off
16 /*Dblk
0 /*Default
16 /*Dblk
38 /*Diverted Wsh A
141 /*Trityl Mon. On/Off
38 /*Diverted Wsh A
144 /*Index Fract. Coll.
1 /*Wsh
2 /*Act
23 /*6 + Act
23 /*6 + Act
2 /*Act
23 /*6 + Act
2 /*Act
0 /*Default
1 /*Wsh
1 /*Wsh
“Event out ON”
“Wait”
“Flush system with Wsh A”
“START data collection”
“Dblk to column”
“Default”
“Deblock”
“Flush system with Wsh A”
“STOP data collection”
“Flush system with Wsh A”
“Event out OFF”
“Flush system with Wsh”
“Flush system with Act”
“Monomer + Act to column”
“Couple monomer”
“Couple monomer”
“Couple monomer”
“Couple monomer”
“Default”
“Couple monomer”
“Flush system with Wsh”
Capping
12 /*Wsh A
13 /*Caps
12 /*Wsh A
12 /*Wsh A
PULSE
PULSE
PULSE
PULSE
20
8
9
21
0
0
23
0
“Flush system with Wsh A”
“Caps to column”
“Cap”
“Flush system with Wsh A”
Oxidizing
15 /*Ox
0 /*Default
12 /*Wsh A
PULSE
WAIT
PULSE
35
0
60
0
20
0
“Ox to column”
“Default”
“Flush system with Wsh A”
Capping
13 /*Caps
12 /*Wsh A
PULSE
PULSE
7
45
0
0
“Caps to column”
“End of cycle wash”
Coupling[a]
Description
[a] In the coupling step, 6 stands for 7-deaza-riboG monomer.
The cleaved products were filtered (0.22 μm membrane filter) and the CPG washed with
acetonitrile/ethanol/H2O 3:1:1 (3 × 0.5 mL).
The resulting fractions (~2.5 mL) were combined and passed through a Sephadex G-25
NAP-25 column for removal of remaining organic residues, using water as eluent. After
lyophilization, the 2´-O-TBDMS was removed by treatment with 1 ml of 0.1 M
tetrabutylammonium floride in THF for 24 h at room temperature. The solution of the
crude oligomer was diluted with 1.5 ml water, desalted on a Sephadex G-25 NAP
column, and purified by reversed-phase HPLC. Tritylated oligonucleotides were
collected and lyophilized. The terminal DMT group was removed using the protocol
previously described. The purity of the detritylated oligomers was confirmed by
analytical reversed-phase HPLC. The amount of DNA was quantified by UV
5.3 Oligonucleotides
143
spectroscopy (λmax = 260 nm, εODN5a,b = 181300 L·mol-1·cm-1), resulting in overall yields
of 14% ODN5a and 5% ODN5b, respectively. The MALDI-TOF MS analysis (positive
mode [M+H]+) of the isolated oligomers is shown in Table 5.6.
Table 5.6. Isolated yields and MALDI-TOF analysis of ODN5a,b.
Entry
Isolated yield
(%)
calcd
14
5945
ODN5a
5973
ODN5b
5
5.3.4
[M+H]+
obsd
5945
5980
General Procedure for the Synthesis of Complementary DNA
Unmodified ODNs were prepared by 1 µmol scale synthesis. Treatment of the fully
protected, resin-bound ODN with 1 mL ammonium hydroxide 28% for 1 h at room
temperature afforded the cleavage from solid-support and removal of the protection
groups. The cleaved products were filtered (0.22 μm membrane filter) and the CPG
washed with H2O (3 × 0.5 mL). The resulting fractions (~2.5 mL) were combined, and
after CHCl3 extraction (3 × 1 mL) the DNA material was passed through a Sephadex G25 NAP column for removal of remaining organic residues, using water as eluent. The
crude ODN was then purified by reversed-phased HPLC. Fractions containing the
tritylated ODN were collected and lyophilized. The terminal DMT group was removed
by treatment with 2% v/v TFA (1 mL) for 2 min at room temperature. After quenching
the acid with NaHCO3, the ODNs were ethanol precipitated. Desalting on Sephadex G25 column afforded pure fully detritylated ODNs (>95% purity), as confirmed by
analytical reversed-phase HPLC. The amount of DNA was quantified by UV
spectroscopy (λmax = 260 nm, εcDNA1 = 170200 L·mol-1·cm-1, εcDNA2 = 211700 L·mol1
·cm-1, εcDNA3 = 140700 L·mol-1·cm-1), resulting in overall yields of 20-37%.
Table 5.7. Isolated yields and MALDI-TOF analysis of cDNA1-3.
Entry
Isolated yield
(%)
calcd
20
5765
cDNA1
37
6986
cDNA2
25
4834
cDNA3
[M+H]+
obsd
5767
6982
4835
5.4 Synthesis of DNA-based Ligands
144
5.4
Synthesis of DNA-Based Ligands
5.4.1
Incorporation of Phosphite Moiety by DNA Solid-phase
Synthesis
General procedure for solid-phase synthesis of DNA-phosphite conjugates.
Synthesis of modified oligodeoxynucleotides sODN1-3 (Table 5.8) carrying a terminal
phosphite moiety (Scheme 3.2) was attempted by the phosphoramidite approach on the
Expedite synthesizer, on 0.2 µmole scale synthesis, using nucleotide precursor
phosphoramidites with standard protecting groups (acetyl for dG, benzoyl for dA and
dC). Phosphoramidite P2 precursor (0.1 M solution in acetonitrile) was assembled at the
5’ terminus on short DNA sequences, using a modified coupling protocol. The coupling
time was extended and the standard dicyanoimidazole activator was replaced with 5benzylthio-(1H)-tetrazole (BTT, 0.25 M solution in acetonitrile), and the oxidation step
was omitted. Final deblocking was not necessary in this case.
Table 5.8. 5’-Functionalization of ODN1-3 with phosphoramidite moieties.[a]
ODN
sODN1
Sequence
5'-TA CGC-3'
Coupling of R moiety to
the 5’-end
Spacer S
-
O
P
O
O
O
sODN2
5'-STA CGC-3'
O
5'-TSA CGC-3'
O
O
O
sODN3
P
10
P
O
O
O
P
O
10
[a] Attempted solid-phase coupling of P1 or P2 to oligonucleotides didn’t lead to the desired conjugates,
and only 5’-OH unmodified sODN1-3 were obtained. sODN1-P2 could be isolated as DNA-phosphate
conjugate sODN1-P2(O).
5.4 Synthesis of DNA-based Ligands
145
After the synthesis, the oligonucleotides were deprotected and cleaved from the solid
support by treatment with 1 mL of 28% aqueous ammonia overnight at room
temperature.
Table 5.9. Isolated yields and MALDI-TOF analysis of sODN1-3.
sODN[a]
Entry
Isolated yield
sODN-P2(O)[b]
(%)
[M-H]
calcd
obsd
calcd
obsd
43
1675
1462
1691
1710
sODN1
28
2195
1982
sODN2
46
2195
1982
sODN3
[a] Conjugates sODN1-3 were isolated as 5’-OH unmodified oligonucleotides, when no oxidation was
performed after P2 coupling during solid phase synthesis. [b] sODN(O) was obtained by carrying out
complete coupling cycle (including oxidation) for P2. Isolated yield refers to the DNA product obtained
after solid phase synthesis and HPLC purification.
The cleaved products were filtered (0.22 μm membrane filter) and the CPG washed with
H2O (3 × 0.5 mL). The ammonia solution was removed by evaporation in a speedvac.
The crude oligonucleotide was redissolved in 0.5 mL H2O and purified by reversedphase HPLC. The fractions containing the major peak were collected and lyophilized.
The isolated DNA products corresponded to the 5’-unmodified oligonucleotides, as
confirmed by MALDI-TOF mass spectrometry. The amount of DNA was quantified by
UV spectroscopy (λmax = 260 nm, εsODN1-3 = 45900 L·mol-1·cm-1), resulting in overall
yields of 28-46% of sODN1-3.
Synthesis of DNA-appended biphenyl-phosphate ester conjugate sODN1-P2(O).
The incorporation of P2 residue at the 5’-end sODN1 by solid phase synthesis was reattempted by carrying out the complete P2 coupling cycle, including oxidation step.
After ammonia deprotection and cleavage from the solid support, the crude
oligonucleotide was purified by reversed-phase HPLC. The fractions containing the
major peak were collected, lyophilized, resuspended in H2O, and analyzed by MALDITOF mass spectrometry. The isolated DNA conjugate corresponded to 5’-biphenylphosphotriester-containing oligonucleotide sODN1-P2(O) (see Table 5.9). The amount
of DNA was quantified by UV spectroscopy (λmax = 260 nm, εsODN1 = 45900 L·mol1
·cm-1), resulting in an overall yield of 48%.
Attempted solid-phase synthesis of sODN1-P2-rhodium(I) complex. DNA-phosphite
oligonucleotide sODN1-P2 was prepared by standard solid-phase synthesis on a 0.2
µmol scale as previously described. The phosphoramidite P2 was assembled on the
solid support-bound DNA by automated coupling, no oxidation being performed. The
5.4 Synthesis of DNA-based Ligands
146
DNA-coated beads were transferred from the synthesis column to an Eppendorf tube
and a solution of [Rh(cod)Cl]2 (9.8 mg, 20.0 nmol) in acetonitrile (0.2 mL) was added.
After vigorously mixing the resulting suspension for 30 minutes at room temperature,
the solution of metal complex was removed, the resin washed with acetonitrile (3 × 0.5
mL) and dried. The beads-supported DNA was then combined with 1.0 mL 28%
ammonium hydroxide. After 30 minutes of incubation at 65°C, an aliquot (50 µL) from
the deprotection mixture was removed, cooled down to room temperature, filtered and
analyzed by reversed-phase HPLC. Beside several peaks eluting in the time range of the
unmodified DNA (tR = 18.0 min) and high-eluting organic residues (tR > 36.0min)
released in the deprotection step, a distinct DNA peak was observed (tR = 27.8 min).
The fraction corresponding to this DNA product was collected and liophylized.
MALDI-TOF mass spectrometry confirmed the formation of the sODN1-P2 conjugate
(calc. [M-H]- 1675, found 1685).
5.4.2
Synthesis of DNA-Phosphine Conjugates
General procedure for the functionalization of amino-ODNs with phosphine
ligands. The phosphine derivatives L1-3 and L6 (1.0 equiv) were converted to the
corresponding activated esters in degassed DMF in 45-60 min at room temperature by
reaction with NHS (1.0 equiv) in the presence of EDC (1.2 equiv). In parallel, ODN1ac, ODN2,3, ODN4a,b, and ODN5a were dissolved in NaHCO3 (0.1 M, pH 8.3) and the
resulting solutions were degassed. The coupling reactions were performed by
combining the solutions of the in situ generated NHS-ester (200-600 equiv) and the
amino-modified DNA (final DMF/H2O ratio 2:1 v/v) to achieve final ODN1a, ODN2,
and ODN3 concentration of 115, 103, and 104 μM, respectively, for coupling with L1
(22 mM), and final ODN1a concentration of 35, 45, and 45 μM, respectively, for
coupling with L2, L3 and L6 (17 mM in all cases).
5.4 Synthesis of DNA-based Ligands
147
Table 5.10. Post-synthetic functionalization of amino-modified ODNs with phosphines L1-3 and
phosphinooxazoline L6.
ODN
Sequence
Ligand
Linker R
L1
ODN6
5'-GC AGT GAA GGCR TGA GCT CC-3'
H N
2
ODN7
ODN8[a]
5'-GC AGT GAA GGC TGA GCT CCT ACRC-3'
L1
5'-GC AGT GAA GGC TGA GCT CCS CRC-3'
L1
ODN9
5'-GC AGT GAA GGC TGA GCT CC-3'
L2
ODN10
5'-GC AGT GAA GGCR TGA GCT CC-3'
L3
ODN11a
5'-GC AGT GAA GGCR TGA GCT CC-3'
L6
ODN11b
5'-GC AGT GAA GGCR TGA GCT CC-3'
L6
ODN11c
5'-GC AGT GAA GGCR TGA GCT CC-3'
L6
ODN12a
5'-GC AGC GAT AACR TAA GCG CT-3'
L6
ODN12b
5'-GC AGC GAT AACR TAA GCG CT-3'
L6
ODN12c
5'-GC AGC GAT AACR TAA GCG CT-3'
L6
ODN13[b]
R
R
5'-GC AGT GAA XXC TXA GCT CC-3'
L6
H2N
H2N
H2N
H2N
H2N
H2N
H2N
O
2O
H2N
H2N
H2N
O
2O
H2N
[a] S = decaethylene glycol spacer; [b] X = 7-deaza-riboG
The coupling of L6 with ODN1b,c, ODN4a-c, and ODN5a was carried out using 33.3
mM ligand and 56 (ODN1b), 53 (ODN1c), 54 (ODN4a), 39 (ODN4b), 50 (ODN4c)
and 80 (ODN5a) µM, respectively, oligonucleotide final concentrations. After stirring
overnight, at room temperature, the reaction mixtures were diluted with water, extracted
with chloroform (3 × 2 mL), and the crude products isolated by ethanol precipitation.
The phosphine-DNA conjugates ODN6-13 were purified by reversed-phase HPLC,
lyophilized, redissolved in degassed water and used immediately in catalytic
experiments. Conversions were estimated by comparing the amount of conjugated
oligonucleotide to the amount of unreacted ODN1-5 as shown in the chromatograms
(52-98%) and isolated yields were calculated based on UV measurements (38-78%).
ε260 for ODN6-13 were approximated to the ones of the corresponding starting
materials.
Table 5.11. Isolated yields and MALDI-TOF MS analysis of ODN6-11a.
ODN(O)n[a]
ODN(O)(S)
Entry
Isolated yield
m/z
[%]
calcd
obsd
calcd
obsd
60
6228
6234
ODN6
65
7424
7430
ODN7
68
7328
7332
ODN8
74
6489
6491
6508
6510
ODN9
38
6605
6610
6621
6626
ODN10
78
6297
6296
ODN11a
[a] n = 2 for ODN9,10, and n = 1 for ODN6-8 and ODN11a. [b] ODN6-9 and
negative mode ([M-H]-), ODN10 in positive mode ([M+H]+).
ODN(S)n[a]
calcd
obsd
6244
6249
7440
7447
7344
7350
6524
6528
6637
6637
6313
6314
ODN11a detected in
5.4 Synthesis of DNA-based Ligands
148
MALDI-TOF mass spectrometry of the HPLC purified ODN6-11a, however, gave in
general only the mass of the oxidized products ODN6-8(O), ODN9,10(O)2, and
ODN11a(O). While MALDI mass spectrometry was found unsuitable for the direct
detection of phosphine conjugates, ESI-MS gave in the only attempted case (ODN9) the
main peak corresponding to the non-oxidized phosphine.
Sulphide-protection of the DNA-phosphine conjugates. To prove the identitity of the
HPLC high-eluting oligonucleotides as phosphine-DNA conjugates ODN6-11a, the
HPLC eluate was collected and immediately treated with elemental sulfur. After
incubating the resulting mixture for 1h at room temperature, the solution was filtered
(0.22 μm membrane filter), lyophilized and redissolved in water (20 µL), yielding the
air-stable phosphine sulfide analogues ODN6-8(S), ODN9,10(S)2, and ODN11a(S) as
confirmed by MALDI-TOF MS (Table 5.11).
In case of bisphoshines L2 and L3 coupling, HPLC chromatograms showed the
formation of a second oligonucleotide peak (<10%), eluting earlier than the main
product. The fractions containing these oligonucleotides were also isolated and reacted
with sulfur. They corresponded to byroducts generated by partial oxidation and were
characterized by MALDI mass spectrometry as monoxide-monosulfide (Table 5.11).
5.4.3
Synthesis
of
DNA-Appended
N,N-bis(2-picolyl)amine
Conjugates
Chelating nitrogen ligand derived from N,N-bis(2-picolyl)amine, namely (PyCH2)2NCH2-p-C6H4-COOH (bpa) (Figure 5.5), obtained from Dr. Srecko Kirin, Prof. Dr. Nils
Metzler-Nolte,[268] was reacted with alkylamino-modified oligonucleotides (ODN1a,
ODN2 and ODN3).
Figure 5.5. (PyCH2)2N-CH2-p-C6H4-COOH Ligand.
5.4 Synthesis of DNA-based Ligands
149
Bpa (1.0 equiv) was first dissolved in DMF and activated with NHS (1.0 equiv) in the
presence of EDC (1.2 equiv.), for 1 h, at room temeperature. The in situ generated
active ester (100-250 equiv) was then directly added to an ODN solution in NaHCO3
(100 mM, pH 8.3) to achieve final DMF/H2O ratio 2:1 v/v and DNA concentrations of
56.4 (ODN1a), 46.3 (ODN2), and 119.7 µM (ODN3) respectively. The final
concentration of bipyridine ligand was maintained in all cases 11.11 mM. After slow
shaking for 16 h at room temperature, the coupling solution was mixed with an equal
amount of formamide-loading buffer and analyzed by 18% polyacrylamide gel
electrophoresis (Figure 3.8, Chapter 3.1.2.3). The coupling reaction proceeded to
completion, leading to only one band with lower electrophoretic mobility compared
with the amino-modified oligonucleotide, as observed by UV illumination (λ = 254 nm)
of the gel (UV-Transilluminator CAMAC Reprostar II).
Table 5.12. Sequences and MALDI-TOF analysis of ODN14-16.
Entry
Sequence
O
ODN14
Bpa
NH
4
5'-GC AGT GAA GGC TGA GCT CC-3'
O
ODN15
ODN16[a]
Bpa
NH
4
5'-GC AGT GAA GGC TGA GCT CCT ACC-3'
O
Bpa
NH
4
5'-GC AGT GAA GGC TGA GCT CCS CC-3'
[M-H]calcd
obsd
6235
6242
7434
7440
7338
7341
[a] S = decaethylene glycol spacer
The bands were excised and the DNA recovered by elution with ammonium acetate (0.5
M, 0.5 mL) overnight, at room temperature. After ethanol precipitation, the DNA
conjugates ODN14-16 (25% isolated yield) were redissolved in water (50 µL) and
analyzed by MALDI-TOF mass spectrometry (Table 5.12).
5.5 High-Pressure Liquid Chromatography
5.5
High-Pressure Liquid Chromatography
5.5.1
Reversed-Phase HPLC Purification of Oligodeoxynucleotides
150
Oligonucleotides purification was performed on reversed-phase HPLC. By purifying
with the DMT group still attached at the 5’-terminus of the synthetic oligonucleotide,
failure sequences that contain no DMT groups are weakly bound to the column and
easily separated from the product which is more strongly retained and eluted later. The
buffers used in reversed-phase HPLC technique are volatile and the purified product can
be rapidly recovered by lyophilization of the volatile solvent.
HPLC analyses were performed on an Agilent 1100 Series HPLC system equipped with
an diode array detector using a Phenomenex® Luna 5 μm C18 column (4.6 × 250 mm)
and eluting with a gradient of 100 mM triethylammonium acetate (TEAA) pH 7.0
(buffer A) and 100 mM TEAA in 80% acetonitrile (buffer B) at 1 mL/min flow-rate.
Preparative HPLC runs were carried out using Phenomenex® Luna 5 μm C18 column
(15.0 × 250 mm) and eluting with a gradient of buffer A and buffer B, followed by a
gradient of water and acetonitrile, at 6 mL/min flow-rate.
Reversed-phase HPLC analysis of ODN1 conversion, deprotection and cleavage
(DMT-on). Treatment of the DMT-on resin-bound ODN1a with a 5 M aqueous solution
of diaminobutane was monitored by reversed-phase HPLC (Gradient: 3 min at 15% B,
increase to 29% B over 11 min, isocratic for 29 min; detection at 260 nm). Reaction
times longer than 4 hours did not improve the yields of deprotection, cleavage and
conversion (see Figure 3.4, Chapter 3.1.2.1).
Reversed-phase HPLC purification of amino-modified ODN1a-c, ODN2,3,
ODN4a-c, ODN5a,b, and their complementary strands cDNA1-3 (DMT-on).
Gradient: 3 min at 2% B, increase to 29% B over 8 min, isocratic for 7 min; change
elution system to water / acetonitrile: change to 23% acetonitrile over 1min, increase
from 23% acetonitrile to 30% acetonitrile over 6 min, isocratic for 3 min; detection at
260 nm; 6 mL/min flow-rate; 55°C.
Characterization of ODN1-5, and cDNA1-3 (DMT-off). The purity of detritylated
amino-modified oligonucleotides ODN1-5 and of their complementary, unmodified
DNA sequences cDNA1-3 was confirmed by analytical HPLC, using a gradient of
5.5 High-Pressure Liquid Chromatography
151
buffer B from 1% to 25% within 40 min. The observed retention times are shown in
Table 5.13.
Table 5.13. Retention times of ODN1-5 and cODN1-3.
ODN
Retention time (DMT-on)
[min]
23.7
ODN1a
23.5
ODN1b
24.0
ODN1c
ODN2
ODN3
23.7
ODN4a
23.6
ODN4b
23.5
ODN4c
23.9
ODN5a
24.1
ODN5b
26.5
cDNA1
26.4
cDNA2
23.4
cDNA3
Retention time (DMT-off)
[min]
25.5
24.7
25.9
26.5
32.8
24.8
24.9
26.4
22.8
22.9
24.3
25.1
27.2
Reversed-phase HPLC purification of phosphine and phosphinooxazolinefunctionalized ODN6-13. Gradients used in the HPLC purification were as follows:
1) increase from 5% B to 15% B over 20 min, increase from 15% B to 25% B over 10
min, increase from 25% B to 40% over 5 min, increase from 40% B to 100% B over 10
min (ODN6-8);
2) increase from 10% B to 62% B over 52 min (ODN9,10);
3) increase from 1% B to 75% B over 40 min (ODN11-13); detection at 260 nm, 1
mL/min flow-rate, 45°C column oven.
Retention times of ODN6-13 and their oxidized species are reported in Table 5.14.
Table 5.14. Retention times of phosphine- and phosphinooxazoline-DNA conjugates ODN6-11.
Retention time [min]
Entry
ODN(O)
ODN
ODN(O)2
34.4
39.0
ODN6
35.1
39.1
ODN7
35.7
39.2
ODN8
22.9
30.4
37.5
ODN9
32.2
43.5
50.7
ODN10
20.3
27.9
ODN11a
18.3
24.6
ODN11b
23.1
32.2
ODN11c
19.3
26.2
ODN12a
18.3
27.6
ODN12b
25.4
31.4
ODN12c
21.7
29.0
ODN13
5.6 Mass Spectrometry Analysis of Oligonucleotides
152
Reversed-phase HPLC analysis of the attempted coupling reaction between P1 or
P2 phosphoramidites and sODN1-3. Gradient: increase from 1%B to 15% B over 20
min, increase from 15% B to 40% B within 10 min; detection at 260 nm).
Table 5.15. Retention times of sODN1-3.
Entry
5’-HO-ODN
18.0
sODN1
26.5
sODN2
26.2
sODN3
[a] sODN1-P2 was isolated in the oxidized form.
5.6
Retention time [min]
sODN-P2(O)[a]
27.9
-
Mass Spectrometry Analysis of Oligonucleotides
Conditions for MALDI-TOF MS analysis. Oligonucleotides were dissolved in water
to a final concentration of 10 μM (commonly 100 pmol, 10 µL) and desalted by using
C18ZipTips (Millipore Corporation, Bedford, MA, USA). The C18 resin was wetted
using 50% aqueous acetonitrile solution (2 x 10 µL) and then equilibrated by washing
with 0.1 M TEAA (3 x 10 µL). For binding of the oligonucleotide to the resin, the
sample was aspirated and dispensed (approx. 5-10 times). The salts were removed by
washing with 0.1 M TEAA. The desalted DNA was eluted by aspirating and dispensing
about 5 times 5 µL 50% acteonitrile/water in a separate vial.
The samples for analysis were prepared using the dried droplet method with the
following matrix solutions: 1) 6-aza-2-thiothymine/diammonium hydrogen citrate in 1:2
v/v
water/acetonitrile
(detection
in
negative
mode);
2)
3-hydroxy-picolinic
acid/diammonium hydrogen citrate in 1.2:1 v/v water/acetonitrile (detection in positive
mode).
Conditions for ESI MS analysis of ODN9. ESI mass spectra were recorded in the
negative mode on a Bruker APEX IV Fourier-transform ion cyclotron resonance (FTICR) mass spectrometer with a 7.05 T magnet and an Apollo electrospray (ESI) ion
source equipped with an off-axis 70o stainless steel spray needle. Typically, 50 μM
analyte solutions (ACN/H2O 1:1) were introduced into the ion source with a syringe
pump (Cole-Parmers Instruments, Series 74900) at flow rates of 3 to 4 µL/min. Ion
transfer into the first of three differential pump stages in the ion source occurred
through a glass capillary with 0.5 mm inner diameter and nickel coatings at both ends.
5.7 5’-Radioactive Labelling of Oligonucleotides
153
Ionization parameters were adjusted as follows: capillary voltage: 4.1 kV; end plate
voltage 3.6 kV; capexit voltage: -280 V; skimmer voltages: -5 to -7.5 V; temperature of
drying gas: 40 oC. Nitrogen was used as ٛ tirٛ ssive (25 psi) and drying gas (5 psi). The
ions were accumulated in the instruments n-hexapole for 1-1.5 s, introduced into the
FT-ICR cell which was operated at pressures below 10-10 mbar, and detected by a
standard excitation and detection sequence. For each measurement, up to 128 scans
were averaged to improve the signal-to-noise ratio.
5.7
5’-Radioactive Labelling of Oligonucleotides
Synthetic oligonucleotides are labelled in phosphorylation reactions catalyzed by
bacteriophage T4 polynucleotide kinase (PNK). The γ-phosphate is transferred from
ATP to the free 5’-hydroxyl group of the target oligonucleotide, affording the
radioactive labelling of DNA. Standard 5’-kinase labelling reaction included the DNA
to be labelled, [γ-32P]-ATP, T4 PNK, and buffer.
Table 5.16. Standard 5’-radioactive labelling of oligonucleotides.
DNA
10xBuffer
γ-32P-ATP
T4-PNK
H 2O
Amount
(pmol)
Volume
(µL)
Observations
6-7.5
2-3
1.5
5
1
to 15 µL
From chemical synthesis
Fermentas
3000 Ci/mmol, 10 mCi/mL (Amersham)
10 U/µL (Fermentas)
After incubation at 37°C for 2 h, the labelled oligonucleotide was purified by 20%
polyacrylamide gel electrophoresis, under denaturing conditions, using a sequencing gel
apparatus. The reaction mixture was mixed with an equal volume of formamide-loading
buffer and loaded to the gel. After electrophoresis, the gel was exposed to
autoradiographic film. The gel band corresponding to the desired product was excised,
transferred to an eppendorf tube and incubated with ammonium acetate (pH 7, 0.5 M,
0.4 mL) overnight, at 37°C. The resulting solution containing the oligonucleotide was
then filtered (spin filters, 0.22 μm cellulose acetate membrane, 3 min centrifugation at
12000 rpm), and the labelled oligonucleotide isolated by ethanol precipitation. After
rinsing with 70% ethanol, the DNA pellet was resuspeded in water (0.4 mL) and the
incorporated
radioactivity
estimated
by
scintillation
counter
measurements.
5.8 Analysis and Quantification of DNA
154
5.8
Analysis and Quantification of DNA
5.8.1
Quantification of Oligonucleotides by UV Absorbance
The nucleobases in DNA and RNA absorb light with a maximum absorbance of 260
nm. Oligonucleotides were the most accurately and conveniently quantified, after
synthesis, by measuring their absorbance at 260 nm in a spectrophotometer.
The DNA samples were measured with a Shimadzu UV-160A UV-spectrophotometer,
in quartz cuvettes (Quarzglas HELLMA) or with NanoDrop ND-100 Specrophotometer
(PeqLab Biotechnologie GmbH) and blanked with the same solution used to dissolve
the oligonucleotide (usually water). The absorbance of a DNA sample at 260 nm was
used to calculate the DNA concentration when the extinction coefficient ε was known.
The molar extinction coefficient describes the amount of absorbance at 260nm (A260) of
1 mol·L-1 DNA solution measured in 1 cm path-length cuvette. This definition is
derived from the Beer-Lambert law showed in the following equation:
A = log(I 0 I ) = ε ⋅ c ⋅ l
Equation 5.1
where A is the absorbance, I0 and I are the intensities of incident and transmitted light,
respectively, c is the molar concentration of the oligonucleotide (mol·L-1), l is the length
of the light path trough the sample (cm), and ε is the molar extinction coefficient of the
molecule (L·mol-1·cm-1).
The extinction coefficient is a physical constant that is unique for each DNA sequence,
since each nucleotide constituent has a different absorbance at 260 nm. The extinction
coefficient was calculated for each oligonucleotide using an equation that incorporates
the contribution of each base (Equation 5.2):
ε = A(15.2 ) + C (7.05) + G (12.01) + T (8.4 )
Equation 5.2
at pH 8, where A, C, G, T are the numbers of dAs, dCs, dG, and dTs, respectively, and
the numbers in parantheses are the molar extinction coefficients for each nucleotide.
5.8 Analysis and Quantification of DNA
5.8.2
155
Analysis of DNA Duplexes by Thermal Melting Curves
The stability of a DNA-DNA duplex was measured by thermal denaturation
experiments on a UV-visible spectrophotometer, by recording the absorbance at 260 nm
as a function of temperature. Heating a DNA sample results in a change in absorbance
properties, which reflects a conformational change of the molecule in solution, and
allows the determination of DNA secondary structure stability. Duplex denaturation
leads to a hyperchromism of 15-20%. Cooling the sample leads to a renaturation of the
structure.
A thermal denaturation experiment of duplex DNA yields the melting temperature value
I, which corresponds to the temperature at which half of the sample is base-paired
(double-helical state), and half is unwinded. Tm deteremination implies the
measurement of the absorbance properties of the folded and unfolded forms as a
function of temperature.
Such an assay was used to study hybridization between an amino-modified
oligonucleotide (ODN1a) and its complementary target (cDNA1). The stability of the
resulting double-stranded structure was investigated in aqueous solutions containing
various concentrations of dioxane. A final DNA concentration of 2 µM (strand
concentration) and an optical pathlength of 1 cm were used in order to obtain an
absorbance value around 0.6 (in the linearity range of the instrument).
The samples were prepared by mixing equimolar amounts of the two DNA strands (2
nmol) in Hepes buffer (15 mM, 0.64 mL, pH 7.5 at 25°C) containing 150 mM NaClO4
and 7.5 mM Mg(ClO4)2, followed by addition of water or/and dioxane up to 0, 5, 10, 20,
and 30% final concentration) and 1 mL final volume. The samples were degassed by
sonication (15 min) in order to remove air bubbles eventually formed in solution, which
might alter the absorbance measurements. The solutions were then placed in cuvettes
and sealed by carefully adding a thin layer of mineral oil on the top of solution to
prevent partial evaporation of the analyte solution occurring at high temeperature. The
multisample cell holder accommodated 6 cuvettes. For each sample, a melting curve
experiment consisted of one fast reversible heating/cooling cycle (15°C → 90°C →
15°C with a thermal gradient of 5°C/min, maintaining 90°C for 5 min and, at the end,
15°C for 5 min), followed by two cycles performed with 0.5°C/min thermal gradient..
5.8 Analysis and Quantification of DNA
156
The heating (dissociation state) and cooling (initial state) profiles are superimposable,
indicating that the transition is kinetically reversible.
Melting temperatures were observed by following the change in UV absorbance as the
temperature was increased and determined by computer fit, followed by calculation of
the maximum of the first derivative of the absorbance signal (dA/dT). Uncertainty in Tm
values was estimated to be ±0.6-0.8.
5.8.3
Polyacrylamide Gel Electrophoresis (PAGE)
Denaturing polyacrylamide gels were used for purification and separation of singlestranded DNA. The gel was polymerized in the presence of urea as denaturing agent.
Denaturing PAGE was used to purify
32
P-radioactive labelled 19mer DNAs, cDNA1,
ODN1a, ODN9(O)2, ODN9(O)(S) and ODN9(S)2 and to assay the synthesis of
bipyridine-DNA conjugates ODN14-16. A 20% gel solution was prepared by diluting a
stock solution of acrylamide: bisacrylamide (19:1 (% w/v), 120 mL) with 10 x TBE
buffer (boric acid 0.89 M, Na2EDTA 0.02 M, TRIS 0.89 M, 15 mL), followed by
addition of urea solution (7 M, 15 mL). The gel polymerization was initiated by
addition of N,N,N’,N’-tetramethylene diamine (TEMED) (50 µL) and ammonium
persulfate (APS) (10% in water, 0.75 mL) and was completed in 30-60 minutes at room
temperature. The DNA samples (30 µL / well) were loaded together with gel tracking
dyes (xylene cyanol and bromphenol blue) on the gel which has been pre-run for 15-20
min at 240 V. The PAGE electrophoresis was carried out in TBE buffer, using a
sequencing gel apparatus, at 1200 V, for 8 h, until the bromophenyl blue indicator dye
that comigrates with the DNA sequence was about ¾ of the way to the bottom. The
oligonucleotides were than visualized using the PhosphorImager instrument for 32P-5’radioactive ٛ labelled or by UV shadowing (shining 254 nm UV light from a hand held
lamp).
Nondenaturing polyacrylamide gel was used for analysis of double-stranded DNAs
formed between ODN1a, ODN9(O)2, ODN9(O)(S) and ODN9(S)2 and the
complementary cDNA1 and cRNA1, respectively. The duplexes were prepared by
combining trace amounts of 32P-radioactive labeled 19mer DNA or RNA (100.000 cpm)
with excess of the corresponding complementary strand (typically 100 pmol), and
dissolving them in buffer (100 mM Hepes pH 7.5, 200 mM NaCl, and 1 mM EDTA).
5.9 Transition Metal-Catalyzed Reactions
157
The final volume was adjusted to 20 µL by addition of water (5 μM final
concentration). The oligonucleotides were heated for 10 min at 75°C, and allowed to
hybridize by slowly cooling down first to 55°C and strirring for 20 min, followed by
cooling down to 37°C within 1 h. After mixing with an equal volume of glycerol 20%
containing traking dye, the duplex solutions were loaded on nondenaturing 16%
polyacrylamide gel (60 mL solution acrylamide: bisacrylamide (19:1 (% w/v), 90 mL 1
x TBE buffer, 50 µL TEMED, 0.75 mL APS, no urea). The gel was run at low voltage
(400 V) to prevent heating that might cause melting of the strands, for 16 h, until
complete migration of the bromphenolblue dye.
5.9
Transition Metal-Catalyzed Reactions
5.9.1
Conjugate Addition
5.9.1.1 Synthesis of 3-Phenyl-1-cyclohexanone
In a Schlenk flask, under argon, [Rh(cod)Cl]2 (60.4 mg, 0.12 mmol, 3 mol%) and
trimethylphosphite (80.0 µL, 0.60 mmol, 2.8 equiv per Rh) were dissolved in 20 mL of
dioxane. After addition of water (2 mL), the resulting mixture was stirred for 10 min at
room temperature. Phenylboronic acid 16 (3.0 g, 24.0 mmol, 3.0 equiv) was added to
the solution. The mixture was heated at 100°C and 2-cyclohexen-1-one 17 (0.8 g, 8.0
mmol) was added. The resulting solution was additionally stirred for 2 hours at 100°C,
then allowed to cool down to room temperature, quenched with saturated NaHCO3 (15
mL) and extracted with diethyl ether (2 × 25 mL). The organic phase was then washed
with brine (2 × 20 mL) and dried over Na2SO4. Removal of the solvent under reduced
pressure afforded the crude product as brownish oil. Purification by flash
chromatography (elution with EA/n-hex 1:9) gave compound 18 as colorless oil (0.9
mg, 5.6 mmol, 70%). 1H NMR (300 MHz; CDCl3) δ 7.38-7.22 (m, 5H), 3.04 (m, 1H),
2.64-2.33 (m, 4H), 2.02-2.21 (m, 2H), 2.06-1.74 (m, 2H).
5.9 Transition Metal-Catalyzed Reactions
158
5.9.1.2 General Procedure for 1,4-Addition of Phenylboronic Acid to 2Cyclohexen-1-one
•
Phosphoramidite ligands. For each catalytic experiment, [Rh(cod)Cl]2 (6.0 mg,
12.0 µmol, 3 mol%) was added to a solution of phosphoramidite P1 (12.3 mg, 60.0
µmol, 2.5 equiv per Rh) or P2 (17.3 mg, 60.0 µmol, 2.5 equiv per Rh) in 2.2 mL
dioxane/water (10:1, 1:5 or 1:10) placed in a Schlenk flask containing, if the case, a
reflux condenser. After being stirred for 10 min at room temperature, phenylboronic
acid 16 (300.0 mg, 2.4 mmol, 3.0 equiv) was added. The resulting mixture was heated at
60°C or stirred for 10 min at room temperature before addition of the enone substrate.
The flask was then charged with 2-cyclohexen-1-one 17 (77.0 mg, 0.8 mmol). The
progress of the reaction was monitored by TLC (elution with EA/n-hex 1:9). The
reaction mixture was stirred for 24-72 hours at 60°C or at room temperature (36.4 mM
final substrate concentration and 10.9 mM final [Rh] catalyst concentration) and then
quenched by addition of saturated NaHCO3 (2 mL). The reaction product was extracted
with diethyl ether (2 × 5 mL), washed with brine (2 × 5mL) and dried over Na2SO4. The
residual solvent was evaporated under reduced pressure. The yields were estimated by
1
H NMR spectroscopy (CDCl3) with isopropanol as internal standard (5-80%
conversion).
Catalytic experiments using lower Rh catalyst loading, [Rh(cod)Cl]2 (1.2 mg, 2.0 µmol,
0.5 mol%), were preformed in the presence of P1 ligand (2.1 mg, 10.0 µmol, 2.5 equiv
per Rh) in 1:10 dioxane/water (2.2 mL), following the previously described procedure.
The reaction mixture was stirred for 24 hours at room temperature. In this case, the final
[Rh] catalyst concentration was 1.8 mM, while the enone substrate 17 was maintained at
36.4 mM concentration.
Duplicated control experiments of 1,4-addition in the presence of only [Rh(cod)Cl]2
precursor were also carried out. In a Schlenk flask, uder argon, [Rh(cod)Cl]2 (6.0 mg,
12.0 µmol, 3 mol%) was dissolved in dioxane (0.2 mL). Upon addition of water (2 mL)
and phenylboronic acid 16 (300.0 mg, 2.4 mmol, 3.0 equiv), the resulting solution was
stirred for 10 min at room temperature. 2-Cyclohexen-1-one 17 (77.0 mg, 0.8 mmol)
was then added and the reaction mixture stirred for 24 hours at room temperature. The
product 18 formation was monitored by TLC (elution with EA/n-hex 1:9) (5%
5.9 Transition Metal-Catalyzed Reactions
159
conversion).
•
Monophosphine ligands. A solution of [Rh(cod)Cl]2 (0.5 mg, 1.0 μmol) in
dioxane (0.2 mL) was added to a solution of commercially available phosphine TPPDS
L7 (2.0 mg, 4.0 μmol, 4.0 equiv) in water (2.0 mL), containing K2CO3 (150.6 mg, 1.1
mmol, 2.1 equiv), or Tris buffer (2.0 mL, 20 mM, pH 8.0) to achieve final [Rh]
concentration of 9.1, 0.91 and 0.23 mM, respectively. The resulting mixture was stirred
under argon for 30 min, at room temperature. After addition of sodium dodecyl sulfate
SDS surfactant (78.0 mg, 0.27 mmol, 0.5 equiv), phenylboronic acid 16 (158.5 mg, 1.3
mmol, 2.5 equiv) and 2-cyclohexen-1-one 17 (49.6 mg, 0.52 mmol), the reaction
mixture was heated at 37°C and the stirring continued for 48 hours.
Control experiment with [Rh(cod)Cl]2 catalyst and no ligand was also carried out.
[Rh(cod)Cl]2 (6.2 mg, 12.5 µmol) was dissolved in dioxane (8 mL). From this solution,
0.2 mL was diluted with Tris buffer (2.0 mL, 20 mM, pH 8.0) to achieve 0.23 mM final
[Rh] concentration. After addition of SDS (78.0 mg, 0.27 mmol, 0.5 equiv),
phenylboronic acid (158.5 mg, 1.3 mmol, 3.0 equiv) and 3-phenyl-1-cyclohexanone 18
(49.6 mg, 0.52 mmol), the resulting mixture was stirred for 48 hours at 37°C.
When phosphine L1 was used, the 1,4-addition reaction was performed in 50% water.
[Rh(cod)Cl]2 (4.9 mg, 10.0 µmol, 3.85 mol%) was added to a solution L1 (12.2 mg,
40.0 µmol, 2.0 equiv per Rh) in dioxane or methanol (1.1 mL). SDS (78.0 mg, 0.27
mmol, 0.5 equiv) and water (1.1 mL) were added and the whole mixture stirred for 30
min at room temperature. To the flask were then added successively phenylboronic acid
16 (158.5 mg, 1.3 mmol, 3.0 equiv) and 2-cyclohexen-1-one 17 (49.6 mg, 0.52 mmol).
The reaction mixture was heated at 50°C and then stirred at 50°C for 19 hours. After
extraction with diethyl ether (2 × 2 mL), the organic phase was filtered through a short
silicagel plug (20 × 6 mm, in a 6 mm pipette). After evaporation of the solvent under
reduced pressure, the residue was dissolved in 2:1 acetonitrile/water (3 mL). The yields
were determined by reversed-phase HPLC analysis (C18 column (4.6×250 mm), elution
with 70% acetonitrile and 30% water, detection wavelength 254 nm; tR = 5.6 min (3phenyl-cyclohexanone)). A calibration curve was obtained by plotting the integrated
peak area versus various product concentrations (2.25, 4.50, 9.00, 13.50 and 18 mM
concentration 3-phenyl-1-cyclohexanone; cP = 0.0036·Area + 0.0929, R2 = 0.9902).
Conversion: 7-42%.
5.9 Transition Metal-Catalyzed Reactions
•
160
Bisphosphine L2 and L8 ligands. To a degassed solution of [RhL2(nbd)]+BF4-
complex 13 (27.1 mg, 0.03 mmol) or [RhL8(nbd)]+BF4- (24.0 mg, 0.03 mmol) in
dioxane (3.0 mL), phenylboronic acid 16 (183.0 mg, 1.5 mmol, 1.5 equiv) and water
(0.5 mL) were added. The resulting mixture was further degassed by two freeze-thaw
cycles. After stirring for 30 min at room temperature, TEA (0.14 mL, 1.0 mmol, 1.0
equiv) was added and the reaction started by addition of 2-cyclohexen-1-one 17 (97.0
µL, 1.0 mmol). The resulting mixture was stirred for 6 hours at room temperature. The
final concentration of Rh complex was in all cases 10 mM, corresponding to 3 mol%
catalyst loading.
Test reactions were setup and preformed also in air. Control experiments only with
[Rh(cod)Cl]2 pre-catalyst were carried out in the absence and presence of base (0.1
equiv NaHCO3) under the conditions described above. Reactions with the in situ
generated Rh complex were performed by addition of [Rh(nbd)Cl]2 (6.9 mg, 0.015
mmol) to a degassed solution of BINAP ligand L8 (28.0 mg, 0.045 mmol, 1.5 equiv per
Rh) in 6:1 dioxane/water (3 mL) and then similarly to the other catalytic reactions.
In all cases, after filtration through a syringe filter (0.2 µm, PTFE), the crude product
was diluted with water to 3.5 mL and analyzed by reversed-phase HPLC (C18 column
(4.6 × 250 mm), elution with 50% water, 50% acetonitrile, detection wavelength 260
nm, tR(18) = 11.0 min). Conversions: 0-80%.
•
PHOX L5 ligand. 1,4-Addition with the isolated Rh-PHOX complex 14 or with
the in situ prepared complex was carried out in 3:7 dioxane/water. The following stock
solutions in dioxane were used: 1) solution of 0.41 M 2-cyclohexen-1-one, 0.40 M TEA
and 0.02 M benzophenone as internal standard, 2) 0.60 M phenylboronic acid 16, and 3)
0.01 M [RhL5(nbd)]+BF4- complex 14. Additionaly, Rh(I) complex was in situ prepared
by weighing [Rh(C2H4)2Cl]2 pre-catalyst (3.9 mg, 0.010 mmol) and ligand L5 (8.6 mg,
0.022 mmol, 1.1 equiv ligand per Rh) into a Schlenk flask, under argon atmosphere.
Degassed dioxane was then added (2.0 mL) to generate 10 mM stock solution of
catalyst. The resulting mixture turned immediately deep red and was used in catalytic
reactions without further purification.
For each catalytic experiment, a Schlenk flask was charged with degassed water (0.7
mL). To the flask were added ٛ successively degassed solutions of [RhL5(nbd)]+BF4(0.01 mL, 1.0 µmol), or the in situ prepared Rh(I) catalyst (0.1 mL, 1.0 µmol), 2-
5.9 Transition Metal-Catalyzed Reactions
161
cyclohexen-1-one 17 and TEA (0.1 mL, 41.0 µmol enone substrate, 20.0 µmol internal
standard and 40.0 µmol base, respectively), and finally phenlyboronic acid 16 (0.1 mL,
60.0 µmol, 1.5 equiv), yielding 1.0 mM final catalyst concentration and 2.4 mol%
catalyst loading, respectively. The reaction mixture was stirred for 4 hours at room
temperature. After extraction with diethyl ether (2 × 1 mL), the organic phase was
passed through a short silica gel plug (10 × 6 mm, in a 6 mm pipette). The eluate was
then evaporated to dryness under reduced pressure. The residue was dissolved in 1:1
water/acetonitrile and analyzed by reversed-phase HPLC (C18 column (4.6×250 mm),
elution with 50% water and 50% acetonitrile, detection wavelength 260 nm, tR(18) =
11.0 min).
5.9.2
Allylic Amination
5.9.2.1 Synthesis and Stability of the Allylic Substrate
The branched monosubstituted allylic substrate, racemic mixture, was prepared by
esterification of 1-phenylprop-2-en-1-ol 19. The linear allylic substrate, cinnamyl
acetate 20, is commercially available.
Synthesis of 1-phenyl-2-propenyl acetate (21). A solution of commercially available
1-phenylprop-2-en-1-ol 19 (2.0 g, 14.9 mmol) and DMAP (0.18 g, 1.5 mmol, 0.1 equiv)
at 0°C was slowly treated with acetic anhydride (7.1 mL, 74.5 mmol, 5 equiv) and then
allowed to warm up to 25°C overnight. The conversion of allylic alcohol to allylic
acetate was complete (TLC control 1:9 EA/n-hex). After quenching with saturated
solution of NaHCO3 (60 mL), the resulting solution was further stirred for additional 30
min at room temperature. The mixture was extracted with diethylether (2 × 60 mL) and
the combined organic fractions were washed with 5% NaHCO3 (100 mL), 1 M HCl
(100 mL) and brine (100 mL) before being dried (Na2SO4) and concentrated in vacuo.
Chromatography on silica gel (column preconditioned with the eluent containing 1%
TEA, elution with 5:95 EA/n-hex) afforded compound 21 as colorless oil (1.8 g, 10.4
mmol, 82%). 1H NMR (300 MHz; CDCl3) δ 7.28-7.22 (m, 5H), 6.18 (td, J = 5.8, 1.3,
1H), 5.92 (ddd, J = 17.2, 10.4, 5.9 Hz, 1H), 5.19 (m, 2H), 2.02 (s, 3H). 13C NMR (126
MHz; CDCl3) δ 169.84, 138.81, 136.21, 128.47, 128.08, 127.06, 116.80, 76.09, 21.15.
EI MS: m/z 176.1 [M]+ (calcd for [C11H12O2]+ 176.08).
5.9 Transition Metal-Catalyzed Reactions
162
Stability of the allylic substrate in aqueous (basic) environment. The stability of
branched allyic acetate 21 in water and under basic conditions was investigated. Stock
solutions of the two substrates were prepared by weighing 1-phenyl-2-propenyl acetate
21 (35.2 mg, 0.2 mmol) into a vial and adding 10 mL acetonitrile. Samples from the
resulting solution were diluted with an equal volume of water or aqueous 0.1 M
NaHCO3 solution, affording 10.0 mM final concentration of 21. The resulting solutions
were stored at room temperature for 9 hours. Aliquots (20 µL) were withdrawn at
regular time intervals and analyzed by reversed-phase HPLC (C18 column (4.6×250
mm), elution with 50% water and 50% acetonitrile, detection wavelength 260 nm, tR
(21) = 14.4 min) (Figure 3.21, Chapter 3.2.3.1).
5.9.2.2 Synthesis of Linear and Branched Allylic Amines
Synthesis of 4-(1-phenyl-2-propenyl)-morpholine (24). Allyl substrate 1-phenyl-2propenyl acetate 21 (0.5 g, 2.8 mmol) was weighed into a Schlenk flask equipped with a
reflux condenser, dissolved in ethanol (10 mL) and the resulting solution degassed (3
cycles). [Rh(cod)Cl]2 (28.1 mg, 0.06 mmol, 4.3 mol%) and triphenyl phosphite (63.0
µL, 0.24 mmol, 2.0 equiv/Rh) were then added and the mixture immediately degassed
(one cycle). After addition of morpholine 22 (0.75 mL, 8.5 mmol, 3.0 equiv), the
reaction mixture was heated at 40°C and stirred overnight until the ester was fully
converted to the amine (TLC control 1:9 EA/n-hex). After cooling the solution at room
temperature, the crude was extracted with diethyl ether (20 mL), and the organic layer
washed with saturated aqueous NaHCO3 solution (ca 20 mL) until the pH of the organic
layer was 8.0. The organic layer was then dried over Na2SO4, the solvent was
evaporated under reduced pressure and the product purified by column chromatography
(eluent 1:9 EA/n-hex). The allylic amine 24 was obtained as colorless oil (0.55 g, 2.7
mmol, 96%). 1H NMR (300 MHz; CDCl3) δ 7.27-7.13 (m, 5H), 5.81 (ddd, J = 17.1,
10.1, 8.8 Hz, 1H), 5.14 (ddd, J = 17.1, 1.6, 0.7 Hz, 1H), 5.00 (dd, J = 10.1, 1.8, 1H),
3.59 (m, 4H), 3.53 (d, J = 8.8 Hz, 1H), 2.43-2.36 (m, 2H), 2.27-2.20 (m, 2H). 13C NMR
(76 MHz; CDCl3) δ 141.48, 139.66, 128.47, 127.83, 127.13, 116.48, 75.39, 67.02,
51.88. EI MS: m/z 203.2 [M]+ (calcd for [C13H17NO]+ 203.13).
Synthesis of 4-(3-phenyl-2-propenyl)-morpholine (25). In a Schlenk flask, Pd(PPh3)4
(0.11 mmol, 0.13 g, 2 mol%) was dissolved in degassed dry THF (15 mL). Cinnamyl
5.9 Transition Metal-Catalyzed Reactions
163
acetate 20 (1.0 g, 5.7 mmol) and morpholine 22 (1.5 mL, 17.1 mmol, 3.0 equiv) were
then added to the reaction mixture. The resulting solution was heated to 50°C and
allowed to ٛ tir at 50°C for 6 hours until complete conversion of the allyic substrate
(TLC control 3:7 EA/n-hex). After cooling down to room temperature, the crude
product was extracted with diethyl ether (20 mL) and washed 5% aqueous NaHCO3
solution (20 mL), and brine (20 mL). The organic phase was dried over Na2SO4,
concentrated under reduced pressure and the product purified by flash column
chromatography (elution 1:4 EA/n-hex) to give linear allylic amine 25 as colorless oil
(0.97 g, 4.7 mmol, 84%). 1H NMR (300 MHz; CDCl3) δ 7.62-7.11 (m, 5H), 6.44 (dd, J
= 15.9, 1H), 6.16 (td, J = 15.9, 6.8, 1H), 3.64 (m, 4H), 3.05 (dd, J = 6.8, 1.3, 2H), 2.40
(m, 4H). 13C NMR (76 MHz; CDCl3) δ 136.65, 133.21, 128.43, 127.42, 126.17, 125.93,
66.83, 61.33, 53.56. EI MS: m/z 203.3 [M]+ (calcd for [C13H17NO]+ 203.13).
Synthesis of N-(1-phenyl-2-propenyl) glycine ethyl ester (26). In a Schlenk flask, 1phenyl-2-propenyl acetate 21 (0.5 g, 2.8 mmol) was dissolved in acetonitirile (4 mL)
and the solution degassed. To this solution were added [Rh(cod)Cl]2 (56.0 mg, 0.11
mmol) and trimethylphosphite (70.5 mg, 71.0 µL, 0.57 mmol, 2.6 equiv per Rh).
Separately, glycine ethyl ester hydrochloride 23 (0.6 g, 4.0 mmol, 1.4 equiv) was
charged into a Schlenk flask. After addition of water (4 mL) and NaHCO3 (0.47 g, 5.6
mmol, 2.0 equiv), the resulting solution was degassed, transferred to the reaction
mixture and stirred at room temperature overnight (TLC control 3:7 EA/n-hex). The
crude product was extracted with diethyl ether (2 × 20 mL) and the combined extracts
washed with 1 M HCl (2 × 20 mL), 5% NaHCO3 (20 mL) and brine (20 mL). After
adjusting the pH to 7 with 5% NaHCO3 (15 mL), the aqueous layer was one more time
extracted with diethyl ether (40 mL). The combined organic phase was dried over
Na2SO4. The solvent was evaporated and the residue chromatographed on silica gel
(elution 1:9 EA/n-hex), affording amine 26 as colorless oil (0.26 mg, 1.2 mmol, 41%).
1
H NMR (300 MHz; CDCl3) δ 7.31-7.17 (m, 5H), 5.83 (m, 1H), 5.18 (d, J = 16.2 Hz,
1H), 5.06 (m, 1H), 4.14 (m, 2H), 3.29 (s, 2H), 1.19 (t, 3H). 13C NMR (76 MHz; CDCl3)
δ 172.52, 142.06, 140.24, 128.61, 127.45, 127.36, 126.37, 115.69, 65.45, 60.75, 48.54,
14.22.
5.9 Transition Metal-Catalyzed Reactions
164
5.9.2.3 Allylic Amination Catalyzed by Ir-PHOX Complexes
Stock solutions of [Ir(cod)Cl]2 (50 mM), and phosphinooxazolines L4 and L5 (6.3 mM)
were prepared by weighing [Ir(cod)Cl]2 (33.6 mg, 0.05 mmol) into a volumetric flask
and adding 1 mL dioxane, and L4 (4.5 mg, 12.0 µmol) and L5 (4.7 mg, 12.0 µmol) into
Schlenk flasks and adding to each 1.8 mL degassed dioxane, respectively. Ir-catalyst
was preparing by adding with a syringe 0.2 mL stock solution [Ir(cod)Cl]2 to each
Schlenk flask containing the ligand (1.25 equiv ligand per Ir), affording a final volume
of 2 mL. The resulting solutions turned immediately deep red indicating the in situ
formation of Ir-L4 and Ir-L5 complexes (10.0 mM). From each solution, 0.2 and 0.3
mL were transferred into separate Schlenk flasks and diluted with degassed dioxane (up
to 2.0 and 5.0 mL, respectively) to give 1.0 mM and 0.6 mM Ir-L4 and Ir-L5 stock
solutions. The Ir-PHOX stock solutions were stored at -20°C and used for multiple
catalytic experiments.
Stock solutions of [Ir(cod)Cl]2 were prepared by weighing 4.9 mg (12.5 µmol) in
volumetric flasks by adding the appropriate amount of dioxane to achieve a range of
5.0, 0.5, 0.3, 0.25, and 0.1 mM final concentrations. [Ir(cod)Cl]2 stock solutions were
each time newly prepared, degassed (by at least three freeze-thaw cycles) and used for
single set of experiments.
Allylic substrate stock solution was prepared by weighing dodecane (489.4 mg, 2.88
mmol), as internal standard, together with 1-phenyl-2-propenyl acetate 21 (885.6 mg,
5.0 mmol) or cinnamyl acetate 20 (885.6 mg, 5.0 mmol) into two vials and adding 10.0
mL dioxane to each vial to make the final concentration of 0.50 M allylic substrate and
0.28 M internal standard.
Two aqueous stock solutions of morpholine 22 were prepared by weighing each time
163.4 mg (1.88 mmol) into a volumetric flask and adding 1:7 dioxane/water or
dioxane/aqueous 125 mM NaClO4, 6.25 mM Mg(ClO4)2 solution (10 mL), and yielding
0.18 M final morpholine concentration. Stock solution of morpholine in neat dioxane
was also prepared by dissolving 22 (480.0 mg, 5.5 mmol) in 10 mL dioxane to a final
concentration of 0.55 M.
Stock solution of glycine ethyl ester hydrochloride 23 containing NaHCO3 was prepared
by weighing the aminoacid ester (195.4 mg, 1.4 mmol) and the base (176.4 mg, 2.1
5.9 Transition Metal-Catalyzed Reactions
165
mmol, 1.5 equiv) in a volumetric flask and adding water (10 mL) to give 0.14 M final
concentration of glycine ethyl ester 23 in 0.21 M aqueous NaHCO3 solution.
All stock solutions were then placed in Schlenk flasks, stored under argon (at -20°C)
and degassed every time befor using in allylic amination reactions.
Gas chromatography analysis. The course of the catalytic allylic aminations was
monitored by gas chromatography performed with a Shimadzu GC-2014 instrument.
Capillary column FS-Supreme-5, 30 m × 0.38 mm AD × 0.25 mm, carrier gas helium.
GC-method: Tinjector = 200°C, Tdetector = 250°C, 1.03 mL/min flow rate, 120.0 kPa
column pressure, 1 µL injected volume, split ratio 40.0. Temperature program: 2 min at
150°C, increase to 230°C with 15°C/min; tR = 3.3 min (dodecane), 3.9 min (21), 5.5
min (20), 6.4 min (24) and 8.0 min (25). The products and the substrates were
quantified using internal standard. Internal standard (dodecane) concentration was
maintained 1.44 mM in all cases, while concentrations of substrates and products were
raging between 0.028-0.690 mM (21, fitted equation: A/Astandard = 0.8958·(c/cstandard) +
0.0005, R2 = 0.9978), 0.051-2.537 mM (20, fitted equation: A/Astandard =
0.8139·(c/cstandard) + 0.0127, R2 = 0.9976), 2.509-0.050 mM (24, fitted equation:
A/Astandard = 0.9346·(c/cstandard) + 0.0122, R2 = 0.9977) and 0.625-0.025 mM (25, fitted
equation: A/Astandard = 0.9825·(c/cstandard) – 0.0070, R2 = 0.9976).
HPLC analysis. The determination of the enantiomeric excesses of 4-(1-phenyl-2propenyl)-morpholine 24 was effected by Agilent 1100 Series HPLC system equipped
with an diode array detector, using a Diacel CHIRALCEL OJ-H (0.46 cm × 25 cm) and
eluting with n-hexane/i-propanol = 99:1 at 0.7 mL/min flow-rate (detection wavelength
= 220, 254 nm; tR (24)= 10.8, 12.5 min).
General procedure for iridium(I)-catalyzed allylic amination reaction. All reactions
were performed in the presence of 1.0 and 0.05-0.1 mM Ir catalyst, on 1.0 mL scale.
Stock solution of Ir catalyst (0.1 mL) was added to 0.1 mL degassed stock solution of
branched 21 or linear allylic substrate 20 (0.05 mmol, 8.8 mg) in a Schlenk flask
containing a stirring bar. Aqueous solution of morpholine 22 (0.8 mL, 13.1 mg, 0.15
mmol, 3.0 equiv) was added with a syringe, and the reaction was stirred at various
temperatures (25, 37, 50°C) for 1-16 hours (TLC control 3:7 EA/n-hex). The crude
product was extracted with diethyl ether (3 × 1 mL) and the combined organic layers
were dried over Na2SO4. After partial removal of the solvent under reduced pressure,
5.9 Transition Metal-Catalyzed Reactions
166
the mixture was filtered over a layer of silica gel (20 × 6 mm, in a 6 mm pipette). The
solvent was then evaporated under light vacuum and the residue was dissolved in 1,2dichloroethane (2 mL) and analyzed by gas chromatography. The crude reaction
mixture was then purified by preparative TLC (elution 3:7 EA/n-hex) and the desired
product was subjected to determination of enantiomeric excess by HPLC.
The same general procedure was also used with 0.1 mL of the stock solution of Ir-L5
complex (10.0 mM, in acetonitrile), 0.1 mL stock solution of branched allylic acetate 21
(0.5 M, acetonitrile), 0.3 mL acetonitrile and 0.5 mL of the stock solution of glycine
ethyl ester 23 (0.14 M, in 0.14 M aqueous NaHCO3). The amination reaction was
conducted at room temperature for 16 hours. The crude reaction mixture was diluted to
5 mL by addition of 1:1 water/acetonitrile, filtered (0.22 µm, PTFE) and analyzed by
reversed-phase HPLC (injected volume 20 µL, C18 column (4.6×250 mm), elution with
50% water and 50% acetonitrile, 1 mL/min flow rate, 25°C column temperature,
detection wavelength 260 nm, tR(26) = 12.4 min).
Irdium(I)-catalyzed allylic amination in the presence of unmodified DNA. All
control reactions were carried out at 0.05-0.1 mM Ir catalyst, in 3:7 dioxane/water (100
μL) and in the presence of 23mer cDNA2. Typically, in a PCR or Eppendorf tube
placed under argon and containing a stirring bar (Figure 5.6), DNA (5.33-13.3 nmol)
was dissolved in aqueous solution of morpholine 22 (80 μL, 15.0 μmol), containing
NaClO4 and Mg(ClO4)2 (see preparation of morpholine stock solutions). The resulting
solution was then mixed with 5-10 μL stock solution of [Ir(cod)Cl]2 (0.5 mM) or
preformed catalyst Ir-L5 (1.0 mM) in a volume of 90 μL, yielding 1.1-1.3 equiv DNA
per Ir. After stirring the resulting solution for 30 min at room temperature, the reaction
was started with the addition of allylic substrate 21 (10 μL, 5.0 μmol) and mixed by
continuous stirring for 14-18 hours at room temperature. After addition of diethyl ether
(200 μL), the reaction mixture was passed over a silica gel plug (20 × 6 mm, in a 6 mm
Pasteur pipette) and the crude product eluted with diethyl ether. The solvent was
removed under light vacuum, the residue dissolved in 1,2-dichloroethane (0.5-0.7 mL)
and analyzed by gas-chromatography for estimating the reaction conversion. The
product was then purified by preparative TLC (elution 3:7 EA/n-hex, product recovery
in ethyl acetate) and the enantiomeric excess determined by chiral HPLC.
5.9 Transition Metal-Catalyzed Reactions
167
Figure 5.6. Reaction set-up for Ir(I)-catalyzed amination using DNA-based PHOX ligands.
Iridium(I)-catalyzed allylic amination with single-stranded DNA-PHOX ligands
ODN11a-c and ODN12a,b. The HPLC fraction containing the desired DNA-PHOX
conjugate was collected in a 25 mL Greiner tube, under argon atmosphere and
immediately degassed. A known volume of eluate, typically 0.6-0.8 mL, was transferred
in an Eppendorf tube (1.5 mL), purged several times with argon and lyophilized
overnight. The remaining eluate was used in UV measurements to determine the amount
of isolated DNA-PHOX conjugate and implicitly the amount of DNA-ligand to be used
for the catalytic experiment (typically 1.5-5.2 nmol). The lyophilized DNA was then
redissolved in thouroughly degassed 143.0 mM NaClO4, 7.0 mM Mg(ClO4)2 aqueous
solution (35-70 µL) and combined with the corresponding amount of 0.5 mM, 0.3 mM,
or 0.25 mM [Ir(cod)Cl]2 stock solution (5.0-10.0 µL, 1.5-5.0 nmol) to generate in situ
the DNA-Ir catalyst (1.1-1.3 equiv DNA-PHOX/Ir). After stirring for 20-30 min under
argon, an aliquot of stock solution of allylic acetate substrate 21 (5.0-10.0 µL) was
added to the DNA solution to a final concentration of 50 mM and the stirring continued
for 10 min. Finally, degassed solution of morpholine 22 (0.55 M, 5.0-10.0 µL, 1.1 equiv
amine to the allylic substrate) was added and the amination reaction conducted for 1619 hours, at room temperature, with a final Ir catalysts concentration ranging between
20 and 100 µM. The reaction mixture was then diluted with diethyl ether (0.2 mL) and
the reaction vial washed with diethyl ether (2 × 0.2 mL). The resulting solution was
filtered trough a short silica gel plug that was thouroughly washed with diethyl ether,
5.9 Transition Metal-Catalyzed Reactions
168
concentrated under light vacuum and subjected to GC and chiral HPLC analysis as
described in the paragraph “General procedure for Ir-catalyzed allylic amination
reaction”.
In parallel, the stability of the DNA-PHOX conjugates upon addition of Ir precatalyst
and under the conditions employed in allylic amination reaction was investigated. To a
ODN11a-c solution (1.1-1.3 equiv per metal ion) was added stock solution of
[Ir(cod)Cl]2 to a final volume of 0.1 mL 3:7 dioxane/water and 0.1 mM final Ir(I)
catalyst concentration. The resulting mixture was sampled (approx. 25 µL) at various
time points and analyzed by reversed-phase HPLC (gradient: increase from 1% B to
75% B over 40 min; detection at 260 nm, 1 mL/min flow-rate, 45°C column oven).
Iridium(I)-catalyzed allylic amination with double-stranded DNA-PHOX ligands.
Catalytic experiments using DNA/DNA or DNA/RNA duplexes were carried at room
temperature, in 3:7 dioxane/water and 0.05-0.1 mL reaction scale. Concentrations of
Ir(I) catalysts were maintained between 20 and 50 µM, unless otherwise stated. The
double-stranded constructs were prepared by mixing equimolar quantities of DNAPHOX conjugate ODN11-13 and complementary cDNA1-4 or cRNA sequence in an
aqueous solution containing 143.0 mM NaClO4 and 7.0 mM Mg(ClO4)2, at room
temperature. Complementary DNA and RNA strands (2.0-4.0 nmol) were lyophilized,
resuspended in degassed water with salts (35.0-70.0 µL) and added to an Eppendorf
tube (Figure 5.6) containing the freshly lyophilized DNA-PHOX ligand. The resulting
solution was immediately purged with argon. The two nucleic acid strands were then
allowed to anneal at room temperature over 30-45 min and then treated with degassed
[Ir(cod)Cl]2 solution (5.0-10.0 µL, 1.1-1.3 equiv DNA-PHOX per iridium ion). After
stirring for 30 min at room temperature, stock solutions of allylic acetate 21 and
respectively morpholine 22 substrates (5.0-10.0 µL each, final concentrations 50 mM
and 55 mM respectively) were added and stirring continued for 16-19 hours, under
argon atmosphere. Work-up of the crude reaction mixture followed the procedure
described in the preceding paragraph.
6.1 List of Abbreviations
6
Appendices
6.1
List of Abbreviations
δ
Chemical shift
ε
Molar extinction coefficient
λ
Wavelength
A
Adenosine; Peak area
A260
Absorbance at 260 nm
Ac
Acetyl
ACN
Acetonitrile
ATP
Adenosine triphosphate
bd
Broad doublet
BINAP
2,2’-Bis(diphenylphosphino)-1,1’-binaphthyl
BINOL
1,1’-Bi-2-naphthol
bpa
N,N’-Bis(2-picolyl)amine
bpy
2,2’-Bipyridine
bs
Broad singlet
BTT
5-Benzylthio-(1H)-tetrazole
c
Concentration
C
Cytidine
cDNA
Complementary DNA
Ci
Curie; 1Ci = 37 MBq
CID
Collision-induced dissociation
cod
1,5-cyclooctadiene
CPG
Controlled pore glass
d
Doublet
dA
2’-Deoxy-adenosine
dC
2’-Deoxycytidine
DCM
Dichloromethane
169
6.1 List of Abbreviations
DEA
Diethylamine
dG
2’-Deoxy-guanosine
DIPA
Diisopropylamine
DMAP
4-(Dimethylamino)pyridine
DMF
N,N’-Dimethylformamide
DMSO
Dimethyl sulfoxide
DMT
4,4’-Dimethoxytrytil
DNA
Deoxyribonucleic acid
dppz
Dipyridophenazine
EA
Ethylacetate
EDC
N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride
EDTA
Ethylenediamine tetraacetate
ee
Enantiomeric excess
EI
Electron impact
equiv
Equivalent
ESI
Electrospray ionization
EtOH
Ethanol
FAB
Fast atomic bombardment
FT-ICR
Fourier-transform ion cyclotron resonance
g
Gram
G
Guanosine
h
Hour
Hepes
4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid
hex
Hexane
HPLC
High Pressure Liquid Chromatography
I
Spin quantum number
I
Light intensity
J
Coupling constant
K
Reaction rate constant
l
Length of the light path
L
Liter; Ligand
LG
Leaving group
170
6.1 List of Abbreviations
m
Meter; Multiplet
M
Mol/L; Molar; Transition metal
MALDI-TOF
Matrix assisted laser desorption ionization time-of-flight
Me
Methyl
Me2-dppz
7,8-Dimethyldipyridophenazine
min
Minute
MS
Mass spectrometry
nbd
Bicyclo[2.2.1]hepta-2,5-diene; norbornadiene
NHS
N-hydroxysuccinimide, N-hydroxysuccinimidyl
nm
Nanometer
NMR
Nuclear magnetic resonance
Nu
Nucleophile
ODN
Oligodeoxynucleotide
Pa
Pascal
PAGE
Polyacrylamide gel electrophoresis
PCR
Polymerase chain reaction
PG
Protecting group
Ph
Phenyl
PhCN
Benzonitrile
phen
1,10-Phenanthroline
phi
9,10-Phenanthrenequinone diimine
PHOX
2-(2-Diphenylphosphino-phenyl)-4,5-dihydrooxazole
PNK
Polynucleotide kinase
ppm
Parts per milion
Pr
Propyl
PTFE
Polytetrafluoroethane
PYRPHOS
3,4-Bis-diphenylphosphino-pyrrolidine
RNA
Ribonucleic acid
rpm
Rotations per minute
rt
Room temperature
RT
Reverse transcription
s
Singlet
171
6.2 Instruments and Special Materials
172
SDS
Sodiumdodecyl sulfate
sec
Second
SELEX
Systematic evolution of ligands by exponential enrichment
SN
Nucleophilic substitution
Solv
Solvent
t
Triplet
T
Thymidine; Temperature
TAC
(t-Butyl)phenoxyacetyl
tap
1,4,5,8-Tetraazaphenantrene
TBE
Tris-borate-EDTA buffer
TCA
Trichloroacetic acid
TEA
Triethylamine
TEAA
Triethylammonium acetate
TFA
Trifluoroacetic acid
THF
Tetrahydrofuran
TLC
Thin layer chromatography
Tm
Melting temperature
TPPDS
Bis(4-sulfonatophenyl)phenylphosphine
tpy
2,2′:6′,2″-Terpyridine
tR
Retention time
Tris
Trishydroxymethylaminomethane; 2-amino-2-hydroxymethyl-1,3propanediol
U
Uridine; Unit
UV
Ultraviolet
6.2
Instruments and Special Materials
Analytical balance
AX 204 and B3001-S Mettler Toledo
Centrifuges
Eppendorf 5804 R and Mikro 120 Hettich
Electrophoresis chamber
GIBCO BRL Sequencing System LIFE
TECHNOLOGIESTM
6.2 Instruments and Special Materials
173
Eppendorf and PCR tubes, siliconized
Biozym
Exposure cassettes
For 35 × 43 cm Kodak imaging screens
Freeze dryer
BenchTop K Series, VirTis Ismatec
Gas chromatograph
Schimadzu GC-2014
- capillary column FS-Supreme-5, 30 m ×
0.38 mm
Gel Documentation equipment
AlphaImagerTM 2200 Alpha Innotech
Greiner tubes
CellStar
HPLC
Agilent 1100 Series
HPLC Columns:
- Luna C18, 5 µm, 4.6 250 mm and 15.0 × Phenomenex®
250 mm
- Chiralcel OJ-H, 4.6 × 250 mm
Daicel
Mass Spectrometer:
- MALDI-TOF
Bruker BIFLEX III
- FAB and EI
JEOL JMS-700
- ESI
Finnigan MAT TSQ 700
- ESI FT-ICR
Bruker APEX IV
Minicentrifuges
Kiesker
NAP columns, Sephadex G-25
GE Healthcare (Amersham Biosciences)
NMR Spectrometer
Mercury Plus 300, Varian VNMR S 500,
Bruker AC-300, DRX-300
pH-Meter
MP 220 Mettler Toledo
Phosphorimager
Typhoon 9400 Amersham Biosciences
Pipettes
Abimed P2, P20, P200, P1000
Scintillation counter
Beckman LS 6500
Silica gel 40 μm
J.T. Baker
Silica gel plates Polygram® Sil G/UV254 Macherey-Nagel
40 × 80 mm
Speed vac
Univapo 100 ECH
Spin filters
Nanosep® MF Centrifugal devices, 0.2
μm PALL, Life Sciences
6.2 Instruments and Special Materials
174
Syntheziser
Applied Biosystems ExpediteTM 8909
Syringe filters
PTFE, 13 mm, 0.2 μm, Carl Roth
Thermomixer
Eppendorf, Thermomixer 5436
Ultrapure Water Purification System
Milli-Q, Millipore
UV Cuvettes
Quarzglas SUPRASIL, HELLMA
UV-Lamp 254 nm
Benda NU-8 KL
UV-Transilluminator
254 nm, 300 × 200 mm Carl Roth
UV/VIS Spectrophotometer
- Ultrospac 2100 pro
Amersham Pharmacia Biotech
- NanoDrop ND-1000
Peqlab Biotechnologie
- Cary 100 Bio
Varian
X-ray film
Fuji, Medical X-ray Film RXOG (Safety)
X-ray film cassettes
Kodak, X-OMATIC
7 References
7
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
175
References
R. Noyori, Asymmetric catalysis: science and opportunities (Nobel Lecture).
Angew. Chem., Int. Ed. Engl. 2002, 41, 2008-22.
W. A. Bonner, The origin and amplification of biomolecular chirality. Origins
Life Evol. Biosphere 1991, 21, 59-111.
W. A. Bonner, Chirality and life. Origins Life Evol. Biosphere 1995, 25, 175-90.
B. L. Feringa, R. A. Van Delden, Absolute asymmetric synthesis: the origin,
control, and amplification of chirality. Angew. Chem., Int. Ed. Engl. 1999, 38,
3419-38.
C. M. Thomas, T. R. Ward, Artificial metalloenzymes: proteins as hosts for
enantioselective catalysis. Chem. Soc. Rev. 2005, 34, 337-46.
H. E. Schoemaker, D. Mink, M. G. Wubbolts, Dispelling the Myths-Biocatalysis
in Industrial Synthesis. Science 2003, 299, 1694-98.
K. A. Powell, S. W. Ramer, S. B. Del Cardayre, W. P. C. Stemmer, M. B.
Tobin, P. F. Longchamp, G. W. Huisman, Directed evolution and biocatalysis.
Angew. Chem., Int. Ed. Engl. 2001, 40, 3948-59.
E. N. Jacobsen, A. Pfaltz, H. Yamamoto, Comprehensive Asymmetric Catalysis,
Vol. 1-3, Springer, Berlin, 1999.
K. Muniz, C. Bolm, Configurational control in stereochemically pure ligands
and metal complexes for asymmetric catalysis. Chem. Eur. J. 2000, 6, 2309-16.
W. S. Knowles, Asymmetric hydrogenations (Nobel Lecture). Angew. Chem.,
Int. Ed. Engl. 2002, 41, 1998-2007.
K. B. Sharpless, Searching for new reactivity (Nobel Lecture). Angew. Chem.,
Int. Ed. Engl. 2002, 41, 2024-32.
K. V. L. Crepy, T. Imamoto, New P-chirogenic phosphine ligands and their use
in catalytic asymmetric reactions. Top. Curr. Chem. 2003, 229, 1-40.
R. Noyori, H. Takaya, BINAP: an efficient chiral element for asymmetric
catalysis. Acc. Chem. Res. 1990, 23, 345-50.
J. A. Osborn, F. H. Jardine, J. F. Young, G. Wilkinson, Preparation and
properties of tris(triphenylphosphine)halorhodium(I) and some reactions thereof
including catalytic homogeneous hydrogenation of olefins and acetylenes and
their derivatives. J. Chem. Soc., A: Inorg. Phys. Theor. 1966, 1711-32.
U. Nagel, Asymmetric hydrogenation of a-(acetylamino)cinnamic acids with a
new rhodium complex; the concept of an optimal ligand. Angew. Chem. 1984,
7 References
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
176
96, 425-6.
M. J. Burk, C2-symmetric bis(phospholanes) and their use in highly
enantioselective hydrogenation reactions. J. Am. Chem. Soc. 1991, 113, 851819.
M. J. Burk, Modular Phospholane Ligands in Asymmetric Catalysis. Acc. Chem.
Res. 2000, 33, 363-72.
G. D. Engel, L. H. Gade, Construction and probing of multisite chiral catalysts:
dendrimer fixation of C2-symmetrical diphosphinerhodium complexes. Chem.
Eur. J. 2002, 8, 4319-29.
U. Nagel, E. Kinzel, Enantioselective catalysis. 4. Synthesis of N-substituted
(R,R)-3,4-bis(diphenylphosphino)pyrrolidines. The use of their rhodium
complexes for the asymmetric hydrogenation of a-(acylamino)acrylic acid
derivatives. Chem. Ber. 1986, 119, 3326-43.
Q. H. Fan, G. J. Deng, C. C. Lin, A. S. C. Chan, Preparation and use of MeOPEG-supported chiral diphosphine ligands: soluble polymer-supported catalysts
for asymmetric hydrogenation. Tetrahedron: Asymmetry 2001, 12, 1241-47.
T. Belser, M. Stoehr, A. Pfaltz, Immobilization of Rhodium Complexes at
Thiolate Monolayers on Gold Surfaces: Catalytic and Structural Studies. J. Am.
Chem. Soc. 2005, 127, 8720-31.
Y. Ribourdouille, G. D. Engel, M. Richard-Plouet, L. H. Gade, A strongly
positive dendrimer effect in asymmetric catalysis: allylic aminations with
Pyrphos-palladium functionalised PPI and PAMAM dendrimers. Chem.
Commun. 2003, 1228-29.
A. Miyashita, A. Yasuda, H. Takaya, K. Toriumi, T. Ito, T. Souchi, R. Noyori,
Synthesis of 2,2'-bis(diphenylphosphino)-1,1'-binaphthyl (BINAP), an
atropisomeric chiral bis(triaryl)phosphine, and its use in the rhodium(I)catalyzed asymmetric hydrogenation of a-(acylamino)acrylic acids. J. Am.
Chem. Soc. 1980, 102, 7932-4.
A. Miyashita, H. Takaya, T. Souchi, R. Noyori, 2,2'-Bis(diphenylphosphino)1,1'-binaphthyl (BINAP). A new atropisomeric bis(triaryl)phosphine. Synthesis
and its use in the rhodium(I)-catalyzed asymmetric hydrogenation of a(acylamino)acrylic acids. Tetrahedron 1984, 40, 1245-53.
U. Schmidt, J. Langner, B. Kirschbaum, C. Braun, Synthesis and
enantioselective hydrogenation of a-acyloxyacrylates. Synthesis 1994, 1138-40.
W. D. Lubell, M. Kitamura, R. Noyori, Enantioselective synthesis of b-amino
acids based on BINAP-ruthenium(II) catalyzed hydrogenation. Tetrahedron:
Asymmetry 1991, 2, 543-54.
7 References
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
177
M. Kitamura, Y. Hsiao, M. Ohta, M. Tsukamoto, T. Ohta, H. Takaya, R. Noyori,
General asymmetric synthesis of isoquinoline alkaloids. Enantioselective
hydrogenation of enamides catalyzed by BINAP-ruthenium(II) complexes. J.
Org. Chem. 1994, 59, 297-310.
D. M. Tschaen, L. Abramson, D. Cai, R. Desmond, U.-H. Dolling, L. Frey, S.
Karady, Y.-J. Shi, T. R. Verhoeven, Asymmetric Synthesis of MK-0499. J. Org.
Chem. 1995, 60, 4324-30.
T. Ohta, H. Takaya, M. Kitamura, K. Nagai, R. Noyori, Asymmetric
hydrogenation of unsaturated carboxylic acids catalyzed by BINAPruthenium(II) complexes. J. Org. Chem. 1987, 52, 3174-6.
H. Takaya, T. Ohta, N. Sayo, H. Kumobayashi, S. Akutagawa, S. Inoue, I.
Kasahara, R. Noyori, Enantioselective hydrogenation of allylic and homoallylic
alcohols. J. Am. Chem. Soc. 1987, 109, 1596-7.
R. Noyori, T. Ohkuma, M. Kitamura, H. Takaya, N. Sayo, H. Kumobayashi, S.
Akutagawa, Asymmetric hydrogenation of b-keto carboxylic esters. A practical,
purely chemical access to b-hydroxy esters in high enantiomeric purity. J. Am.
Chem. Soc. 1987, 109, 5856-8.
S. Jeulin, S. D. De Paule, V. Ratovelomanana-Vidal, J.-P. Genet, N. Champion,
P. Dellis, Chiral biphenyl diphosphines for asymmetric catalysis:
stereoelectronic design and industrial perspectives. Proc. Natl. Acad. Sci. USA
2004, 101, 5799-804.
K. Tani, J.-i. Onouchi, T. Yamagata, Y. Kataoka, Iridium(I)-catalyzed
asymmetric hydrogenation of prochiral imines; protic amines as catalyst
improvers. Chem. Lett. 1995, 955-6.
Y. Sato, M. Sodeoka, M. Shibasaki, Catalytic asymmetric carbon-carbon bond
formation: asymmetric synthesis of cis-decalin derivatives by palladiumcatalyzed cyclization of prochiral alkenyl iodides. J. Org. Chem. 1989, 54, 47389.
F. Ozawa, Y. Kobatake, A. Kubo, T. Hayashi, Palladium-catalyzed asymmetric
hydroalkenylation of norbornene. J. Chem. Soc., Chem. Commun. 1994, 1323-4.
F. Ozawa, A. Kubo, Y. Matsumoto, T. Hayashi, E. Nishioka, K. Yanagi, K.
Moriguchi, Palladium-catalyzed asymmetric arylation of 2,3-dihydrofuran with
phenyl triflate. A novel asymmetric catalysis involving a kinetic resolution
process. Organometallics 1993, 12, 4188-96.
Y. Nakamura, S. Takeuchi, S. Zhang, K. Okumura, Y. Ohgo, Preparation of a
fluorous chiral BINAP and application to an asymmetric Heck reaction.
Tetrahedron Lett. 2002, 43, 3053-56.
7 References
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
178
D. Kiely, P. J. Guiry, A comparative study of diphosphine and phosphinamine
palladium complexes on a new substrate for the intramolecular asymmetric
Heck reaction. Tetrahedron Lett. 2002, 43, 9545-47.
O. Loiseleur, P. Meier, A. Pfaltz, Chiral phosphanyldihydrooxazoles in
asymmetric catalysis: enantioselective Heck reactions. Angew. Chem., Int. Ed.
Engl. 1996, 35, 200-2.
Y. Takaya, M. Ogasawara, T. Hayashi, M. Sakai, N. Miyaura, RhodiumCatalyzed Asymmetric 1,4-Addition of Aryl- and Alkenylboronic Acids to
Enones. J. Am. Chem. Soc. 1998, 120, 5579-80.
S. Sakuma, M. Sakai, R. Itooka, N. Miyaura, Asymmetric Conjugate 1,4Addition of Arylboronic Acids to a,b-Unsaturated Esters Catalyzed by
Rhodium(I)/(S)-binap. J. Org. Chem. 2000, 65, 5951-55.
Y. Takaya, T. Senda, H. Kurushima, M. Ogasawara, T. Hayashi, Rhodiumcatalyzed asymmetric 1,4-addition of arylboron reagents to a,b-unsaturated
esters. Tetrahedron: Asymmetry 1999, 10, 4047-56.
T. Hayashi, T. Senda, Y. Takaya, M. Ogasawara, Rhodium-Catalyzed
Asymmetric 1,4-Addition to 1-Alkenylphosphonates. J. Am. Chem. Soc. 1999,
121, 11591-92.
M. Sakai, H. Hayashi, N. Miyaura, Rhodium-Catalyzed Conjugate Addition of
Aryl- or 1-Alkenylboronic Acids to Enones. Organometallics 1997, 16, 422931.
T. Hayashi, T. Senda, M. Ogasawara, Rhodium-Catalyzed Asymmetric
Conjugate Addition of Organoboronic Acids to Nitroalkenes. J. Am. Chem. Soc.
2000, 122, 10716-17.
K. Fagnou, M. Lautens, Rhodium-catalyzed carbon-carbon bond forming
reactions of organometallic compounds. Chem. Rev. 2003, 103, 169-96.
T. Hayashi, K. Yamasaki, Rhodium-Catalyzed Asymmetric 1,4-Addition and Its
Related Asymmetric Reactions. Chem. Rev. 2003, 103, 2829-44.
P. Guerreiro, V. Ratovelomanana-Vidal, J. P. Genet, P. Dellis, Recyclable
diguanidinium-BINAP and PEG-BINAP supported catalysts: syntheses and use
in Rh(I) and Ru(II) asymmetric hydrogenation reactions. Tetrahedron Lett.
2001, 42, 3423-26.
T. Lamouille, C. Saluzzo, R. ter Halle, F. Le Guyader, M. Lemaire,
Hydrogenation of ethyl acetoacetate catalyzed by hydrosoluble BINAP
derivatives. Tetrahedron Lett. 2001, 42, 663-64.
Y. Otomaru, T. Senda, T. Hayashi, Preparation of an Amphiphilic ResinSupported BINAP Ligand and Its Use for Rhodium-Catalyzed Asymmetric 1,4-
7 References
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
179
Addition of Phenylboronic Acid in Water. Org. Lett. 2004, 6, 3357-59.
C. J. Chapman, C. G. Frost, M. P. Gill-Carey, G. Kociok-Kohn, M. F. Mahon,
A. S. Weller, M. C. Willis, Diphosphine mono-sulfides: readily available chiral
monophosphines. Tetrahedron: Asymmetry 2003, 14, 705-10.
J. W. Faller, J. C. Wilt, J. Parr, Kinetic Resolution and Unusual Regioselectivity
in Palladium-Catalyzed Allylic Alkylations with a Chiral P,S Ligand. Org. Lett.
2004, 6, 1301-04.
J. W. Faller, C. Wilt Jeremy, Palladium/BINAP(S)-catalyzed asymmetric allylic
amination. Org. Lett. 2005, 7, 633-6.
J. W. Faller, J. C. Wilt, Regioselectivity in the Palladium/(S)-BINAP(S)Catalyzed Asymmetric Allylic Amination: Reaction Scope, Kinetics, and
Stereodynamics. Organometallics 2005, 24, 5076-83.
G. Helmchen, S. Kudis, P. Sennhenn, H. Steinhagen, Enantioselective catalysis
with complexes of asymmetric P,N-chelate ligands. Pure Appl. Chem. 1997, 69,
513-18.
J. Sprinz, G. Helmchen, Phosphinoaryl- and phosphinoalkyloxazolines as new
chiral ligands for enantioselective catalysis: very high enantioselectivity in
palladium catalyzed allylic substitutions. Tetrahedron Lett. 1993, 34, 1769-72.
G. Helmchen, A. Pfaltz, Phosphinooxazolines-A New Class of Versatile,
Modular P,N-Ligands for Asymmetric Catalysis. Acc. Chem. Res. 2000, 33, 33645.
A. Lightfoot, P. Schnider, A. Pfaltz, Enantioselective hydrogenation of olefins
with iridium-phosphanodihydrooxazole catalysts. Angew. Chem., Int. Ed. Engl.
1998, 37, 2897-99.
D. G. Blackmond, A. Lightfoot, A. Pfaltz, T. Rosner, P. Schnider, N.
Zimmermann,
Enantioselective
hydrogenation
of
olefins
with
phosphinooxazoline-iridium catalysts. Chirality 2000, 12, 442-49.
S. P. Smidt, N. Zimmermann, M. Studer, A. Pfaltz, Enantioselective
hydrogenation of alkenes with iridium-PHOX catalysts: A kinetic study of anion
effects. Chem. Eur. J. 2004, 10, 4685-93.
A. Pfaltz, J. Blankenstein, R. Hilgraf, E. Hormann, S. McIntyre, F. Menges, M.
Schonleber, S. P. Smidt, B. Wustenberg, N. Zimmermann, Iridium-catalyzed
enantioselective hydrogenation of olefins. Adv. Synth. Catal. 2003, 345, 33-44.
P. Schnider, G. Koch, R. Pretot, G. Wang, F. M. Bohnen, C. Kruger, A. Pfaltz,
Enantioselective
hydrogenation
of
imines
with
chiral
(phosphanodihydrooxazole)iridium catalysts. Chem. Eur. J. 1997, 3, 887-92.
O. Loiseleur, M. Hayashi, M. Keenan, N. Schmees, A. Pfaltz, Enantioselective
7 References
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
180
Heck reactions using chiral P,N-ligands. J. Organomet. Chem. 1999, 576, 16-22.
G. Helmchen, Enantioselective palladium-catalyzed allylic substitutions with
asymmetric chiral ligands. J. Organomet. Chem. 1999, 576, 203-14.
P. von Matt, O. Loiseleur, G. Koch, A. Pfaltz, C. Lefeber, T. Feucht, G.
Helmchen, Enantioselective allylic amination with chiral (phosphinooxazoline)palladium catalysts. Tetrahedron: Asymmetry 1994, 5, 573-84.
R. N. Constantine, N. Kim, R. C. Bunt, Hammett Studies of Enantiocontrol by
PHOX Ligands in Pd-Catalyzed Allylic Substitution Reactions. Org. Lett. 2003,
5, 2279-82.
J. P. Janssen, G. Helmchen, First enantioselective alkylations of
monosubstituted allylic acetates catalyzed by chiral iridium complexes.
Tetrahedron Lett. 1997, 38, 8025-26.
C. Garcia-Yebra, J. P. Janssen, F. Rominger, G. Helmchen, Asymmetric
Iridium(I)-Catalyzed Allylic Alkylation of Monosubstituted Allylic Substrates
with Phosphinooxazolines as Ligands. Isolation, Characterization, and
Reactivity of Chiral (Allyl)iridium(III) Complexes. Organometallics 2004, 23,
5459-70.
H. Rieck, G. Helmchen, Palladium complex catalyzed asymmetric allylic
substitutions with nitromethane: enantioselectivities exceeding 99.9% ee.
Angew. Chem., Int. Ed. Engl. 1996, 34, 2687-9.
H. Eichelmann, H.-J. Gais, Palladium-catalyzed asymmetric allylic
sulfonylation. Tetrahedron: Asymmetry 1995, 6, 643-6.
C. Welter, O. Koch, G. Lipowsky, G. Helmchen, First intramolecular
enantioselective iridium-catalyzed allylic aminations. Chem. Commun. 2004,
896-97.
G. Helmchen, A. Dahnz, P. Duebon, M. Schelwies, R. Weihofen, Iridiumcatalyzed asymmetric allylic substitutions. Chem. Commun. 2007, 675-91.
K. N. Gavrilov, O. G. Bondarev, A. V. Korostylev, A. I. Polosukhin, V. N.
Tsarev, N. E. Kadilnikov, S. E. Lyubimov, A. A. Shiryaev, S. V. Zheglov, H.-J.
Gais, V. A. Davankov, Novel P,N-bidentate phosphite ligands in asymmetric
catalysis. Chirality 2003, 15, S97-S103.
J. Ansell, M. Wills, Enantioselective catalysis using phosphorus-donor ligands
containing two or three P-N or P-O bonds. Chem. Soc. Rev. 2002, 31, 259-68.
R. Takeuchi, M. Kashio, Iridium Complex-Catalyzed Allylic Alkylation of
Allylic Esters and Allylic Alcohols: Unique Regio- and Stereoselectivity. J. Am.
Chem. Soc. 1998, 120, 8647-55.
R. Takeuchi, N. Ue, K. Tanabe, K. Yamashita, N. Shiga, Iridium complex-
7 References
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
181
catalyzed allylic amination of allylic esters. J. Am. Chem. Soc. 2001, 123, 952534.
B. Bartels, C. Garcia-Yebra, F. Rominger, G. Helmchen, Iridium-catalysed
allylic substitution: Stereochemical aspects and isolation of Ir(III) complexes
related to the catalytic cycle. Eur. J. Inorg. Chem. 2002, 2569-86.
B. Bartels, G. Helmchen, Ir-catalyzed allylic substitution: mechanistic aspects
and asymmetric synthesis with phosphorus amidites as ligands. Chem. Commun.
1999, 741-42.
K. Fuji, N. Kinoshita, T. Kawabata, K. Tanaka, Enantioselective allylic
substitution catalyzed by an iridium complex: remarkable effects of the counter
cation. Chem. Commun. 1999, 2289-90.
O. Pamies, M. Dieguez, G. Net, A. Ruiz, C. Claver, Highly Enantioselective RhCatalyzed Hydrogenation Based on Phosphine-Phosphite Ligands Derived from
Carbohydrates. J. Org. Chem. 2001, 66, 8364-69.
M. T. Reetz, G. Mehler, Highly enantioselective Rh-catalyzed hydrogenation
reactions based on chiral monophosphite ligands. Angew. Chem., Int. Ed. Engl.
2000, 39, 3889-90.
M. T. Reetz, J.-A. Ma, R. Goddard, BINOL-derived monodentate phosphites
and phosphoramidites with phosphorus stereogenic centers: Novel ligands for
transition-metal catalysis. Angew. Chem., Int. Ed. Engl. 2005, 44, 412-15.
B. L. Feringa, M. Pineschi, L. A. Arnold, R. Imbos, A. H. M. De Vries, Highly
enantioselective catalytic conjugate addition and tandem conjugate addition aldol reactions of organozinc reagents. Angew. Chem., Int. Ed. Engl. 1997, 36,
2620-23.
A. Alexakis, S. Rosset, J. Allamand, S. March, F. Guillen, C. Benhaim, Novel
biphenol phosphoramidite ligands for the enantioselective copper-catalyzed
conjugate addition of dialkyl zincs. Synlett 2001, 1375-78.
A. Alexakis, C. Benhaim, S. Rosset, M. Humam, Dramatic Improvement of the
Enantiomeric Excess in the Asymmetric Conjugate Addition Reaction Using
New Experimental Conditions. J. Am. Chem. Soc. 2002, 124, 5262-63.
E. Keller, J. Maurer, R. Naasz, T. Schader, A. Meetsma, B. L. Feringa,
Unexpected enhancement of enantioselectivity in copper(II) catalyzed conjugate
addition of diethylzinc to cyclic enones with novel TADDOL phosphorus
amidite ligands. Tetrahedron: Asymmetry 1998, 9, 2409-13.
R. Imbos, M. H. G. Brilman, M. Pineschi, B. L. Feringa, Highly
Enantioselective Catalytic Conjugate Additions to Cyclohexadienones. Org.
Lett. 1999, 1, 623-25.
7 References
[88]
[89]
[90]
[91]
[92]
[93]
[94]
[95]
[96]
[97]
[98]
[99]
[100]
[101]
[102]
[103]
[104]
182
L. A. Arnold, R. Imbos, A. Mandoli, A. H. M. De Vries, R. Naasz, B. L.
Feringa, Enantioselective catalytic conjugate addition of dialkylzinc reagents
using copper-phosphoramidite complexes; ligand variation and non-linear
effects. Tetrahedron 2000, 56, 2865-78.
C. Monti, C. Gennari, U. Piarulli, Rh-catalyzed enantioselective conjugate
addition of arylboronic acids with a dynamic library of chiral tropos phosphorus
ligands. Chem. Eur. J. 2007, 13, 1547-58.
U. T. Bornscheuer, Methods to increase enantioselectivity of lipases and
esterases. Curr. Opin. Biotechnol. 2002, 13, 543-47.
R. D. Schmid, R. Verger, Lipases: interfacial enzymes with attractive
applications. Angew. Chem., Int. Ed. Engl. 1998, 37, 1609-33.
M. T. Reetz, Lipases as practical biocatalysts. Curr. Op. Chem. Biol. 2002, 6,
145-50.
F. H. Arnold, Combinatorial and computational challenges for biocatalyst
design. Nature 2001, 409, 253-57.
D. Rotticci, J. C. Rotticci-Mulder, S. Denman, T. Norin, K. Hult, Improved
enantioselectivity of a lipase by rational protein engineering. ChemBioChem
2001, 2, 766-70.
U. T. Bornscheuer, M. Pohl, Improved biocatalysts by directed evolution and
rational protein design. Curr. Op. Chem. Biol. 2001, 5, 137-43.
D. Qi, C. M. Tann, D. Haring, M. D. Distefano, Generation of new enzymes via
covalent modification of existing proteins. Chem. Rev. 2001, 101, 3081-111.
M. T. Reetz, Strategies for the development of enantioselective catalysts. Pure
Appl. Chem. 1999, 71, 1503-09.
E. T. Farinas, T. Bulter, F. H. Arnold, Directed enzyme evolution. Curr. Opin.
Biotechnol. 2001, 12, 545-51.
F. H. Arnold, Design by directed evolution. Acc. Chem. Res. 1998, 31, 125-31.
P. A. Patten, R. J. Howard, W. P. C. Stemmer, Applications of DNA shuffling to
pharmaceuticals and vaccines. Curr. Opin. Biotechnol. 1997, 8, 724-33.
K.-E. Jaeger, M. T. Reetz, Directed evolution of enantioselective enzymes for
organic chemistry. Curr. Op. Chem. Biol. 2000, 4, 68-73.
M. T. Reetz, Combinatorial and Evolution-Based Methods in the Creation of
Enantioselective Catalysts. Angew. Chem., Int. Ed. Engl. 2001, 40, 284-310.
M. T. Reetz, A. Zonta, K. Schimossek, K. Liebeton, K.-E. Jaeger, Creation of
enantioselective biocatalysts for organic chemistry by in vitro evolution. Angew.
Chem., Int. Ed. Engl. 1998, 36, 2830-32.
M. T. Reetz, K.-E. Jaeger, Enantioselective enzymes for organic synthesis
7 References
[105]
[106]
[107]
[108]
[109]
[110]
[111]
[112]
[113]
[114]
[115]
[116]
183
created by directed evolution. Chem. Eur. J. 2000, 6, 407-12.
M. T. Reetz, S. Wilensek, D. Zha, K. E. Jaeger, Directed Evolution of an
Enantioselective
Enzyme
through
Combinatorial
Multiple-Cassette
Mutagenesis. Angew. Chem., Int. Ed. Engl. 2001, 40, 3589-91.
S. A. Funke, A. Eipper, M. T. Reetz, N. Otte, W. Theil, G. Van Pouderoyen, B.
W. Dijkstra, K.-E. Jaeger, T. Eggert, Directed Evolution of an Enantioselective
Bacillus subtilis Lipase. Biocat. Biotransform. 2003, 21, 67-73.
G. P. Horsman, A. M. F. Liu, E. Henke, U. T. Bornscheuer, R. J. Kazlauskas,
Mutations in distant residues moderately increase the enantioselectivity of
pseudomonas fluorescens esterase towards methyl 3-bromo-2-methylpropanoate
and ethyl 3-phenylbutyrate. Chem. Eur. J. 2003, 9, 1933-39.
G. J. Williams, S. Domann, A. Nelson, A. Berry, Modifying the stereochemistry
of an enzyme-catalyzed reaction by directed evolution. Proc. Natl. Acad. Sci.
USA 2003, 100, 3143-48.
M. T. Reetz, B. Brunner, T. Schneider, F. Schulz, C. M. Clouthier, M. M.
Kayser, Asymmetric catalysis: Directed evolution as a method to create
enantioselective cyclohexanone monooxygenases for catalysis in BaeyerVilliger reactions. Angew. Chem., Int. Ed. Engl. 2004, 43, 4075-78.
M. T. Reetz, Directed evolution of selective enzymes and hybrid catalysts.
Tetrahedron 2002, 58, 6595-602.
M. T. Reetz, M. Rentzsch, A. Pletsch, M. Maywald, Towards the directed
evolution of hybrid catalysts. Chimia 2002, 56, 721-23.
M. E. Wilson, G. M. Whitesides, Conversion of a protein to a homogeneous
asymmetric hydrogenation catalyst by site-specific modification with a
diphosphinerhodium(I) moiety. J. Am. Chem. Soc. 1978, 100, 306-07.
C.-C. Lin, C.-W. Lin, A. S. C. Chan, Catalytic hydrogenation of itaconic acid in
a biotinylated Pyrphos-rhodium(I) system in a protein cavity. Tetrahedron:
Asymmetry 1999, 10, 1887-93.
J. Collot, J. Gradinaru, N. Humbert, M. Skander, A. Zocchi, T. R. Ward,
Artificial Metalloenzymes for Enantioselective Catalysis Based on BiotinAvidin. J. Am. Chem. Soc. 2003, 125, 9030-31.
C. M. Thomas, T. R. Ward, Artificial metalloenzymes: proteins as hosts for
enantioselective catalysis. Chem. Soc. Rev. 2005, 34, 337-46.
C. Letondor, A. Pordea, N. Humbert, A. Ivanova, S. Mazurek, M. Novic, T. R.
Ward, Artificial Transfer Hydrogenases Based on the Biotin-(Strept)avidin
Technology: Fine Tuning the Selectivity by Saturation Mutagenesis of the Host
Protein. J. Am. Chem. Soc. 2006, 128, 8320-28.
7 References
184
[117] A. Mahammed, Z. Gross, Albumin-Conjugated Corrole Metal Complexes:
Extremely Simple Yet Very Efficient Biomimetic Oxidation Systems. J. Am.
Chem. Soc. 2005, 127, 2883-87.
[118] C. Letondor, T. R. Ward, Artificial metalloenzymes for enantioselective
catalysis: Recent advances. ChemBioChem 2006, 7, 1845-52.
[119] M. T. Reetz, N. Jiao, Copper-phthalocyanine conjugates of serum albumins as
enantioselective catalysts in Diels-Alder reactions. Angew. Chem., Int. Ed. Engl.
2006, 45, 2416-19.
[120] M. Ohashi, T. Koshiyama, T. Ueno, M. Yanase, H. Fujii, Y. Watanabe,
Preparation of artificial metalloenzymes by insertion of chromium(III) Schiff
base complexes into apomyoglobin mutants. Angew. Chem., Int. Ed. Engl. 2003,
42, 1005-8.
[121] M. T. Reetz, J. J. P. Peyralans, A. Maichele, Y. Fu, M. Maywald, Directed
evolution of hybrid enzymes: Evolving enantioselectivity of an achiral Rhcomplex anchored to a protein. Chem. Commun. 2006, 4318-20.
[122] R. R. Davies, M. D. Distefano, A Semisynthetic Metalloenzyme Based on a
Protein Cavity That Catalyzes the Enantioselective Hydrolysis of Ester and
Amide Substrates. J. Am. Chem. Soc. 1997, 119, 11643-52.
[123] M. T. Reetz, Controlling the enantioselectivity of enzymes by directed
evolution: practical and theoretical ramifications. Proc. Natl. Acad. Sci. USA
2004, 101, 5716-22.
[124] L. Panella, J. Broos, J. Jin, M. W. Fraaije, D. B. Janssen, M. JeronimusStratingh, B. L. Feringa, A. J. Minnaard, J. G. De Vries, Merging homogeneous
catalysis with biocatalysis; papain as hydrogenation catalyst. Chem. Commun.
2005, 5656-58.
[125] R. Corradini, S. Sforza, T. Tedeschi, R. Marchelli, Chirality as a tool in nucleic
acid recognition: principles and relevance in biotechnology and in medicinal
chemistry. Chirality 2007, 19, 269-94.
[126] S. M. Hecht, Editor, Bioorganic Chemistry: Nucleic Acids. Oxford, UK: Oxford
University Press, 1996.
[127] R. E. Dickerson, H. R. Drew, B. N. Conner, R. M. Wing, A. V. Fratini, M. L.
Kopka, The anatomy of A-, B-, and Z-DNA. Science 1982, 216, 475-85.
[128] R. Wing, H. Drew, T. Takano, C. Broka, S. Tanaka, K. Itakura, R. E. Dickerson,
Crystal structure analysis of a complete turn of B-DNA. Nature 1980, 287, 7558.
[129] C. A. Frederick, G. J. Quigley, M. K. Teng, M. Coll, G. A. Van der Marel, J. H.
Van Boom, A. Rich, A. H. Wang, Molecular structure of an A-DNA decamer
7 References
[130]
[131]
[132]
[133]
[134]
[135]
[136]
[137]
[138]
[139]
[140]
[141]
[142]
[143]
[144]
[145]
185
d(ACCGGCCGGT). Eur. J. Biochem. 1989, 181, 295-307.
D. M. J. Lilley, R. M. Clegg, S. Diekmann, N. C. Seeman, E. von Kitzing, P. J.
Hagerman, A nomenclature of junctions and branchpoints in nucleic acids.
Nucleic Acids Res. 1995, 23, 3363-4.
D. M. J. Lilley, All change at Holliday junction. Proc. Natl. Acad. Sci. USA
1997, 94, 9513-15.
D. S. Wilson, J. W. Szostak, In vitro selection of functional nucleic acids. Annu.
Rev. Biochem. 1999, 68, 611-47.
R. R. Breaker, DNA aptamers and DNA enzymes. Curr. Op. Chem. Biol. 1997,
1, 26-31.
G. R. Zimmermann, R. D. Jenison, C. L. Wick, J.-P. Simorre, A. Pardi,
Interlocking structural motifs mediate molecular discrimination by a
theophylline-binding RNA. Nat. Struct. Biol. 1997, 4, 644-49.
T. Hermann, D. J. Patel, Adaptive recognition by nucleic acid aptamers. Science
2000, 287, 820-25.
M. Famulok, Oligonucleotide aptamers that recognize small molecules. Curr.
Opin. Struct. Biol. 1999, 9, 324-29.
Y. Wang, R. R. Rando, Specific binding of aminoglycoside antibiotics to RNA.
Chem. Biol. 1995, 2, 281-90.
Y. Tor, Targeting RNA with small molecules. ChemBioChem 2003, 4, 9981007.
M. Sassanfar, J. W. Szostak, An RNA motif that binds ATP. Nature 1993, 364,
550-3.
D. Kiga, Y. Futamura, K. Sakamoto, S. Yokoyama, An RNA aptamer to the
xanthine/guanine base with a distinctive mode of purine recognition. Nucleic
Acids Res. 1998, 26, 1755-60.
R. D. Jenison, S. C. Gill, A. Pardi, B. Polisky, High-resolution molecular
discrimination by RNA. Science 1994, 263, 1425-9.
A. Geiger, P. Burgstaller, H. von der Eltz, A. Roeder, M. Famulok, RNA
aptamers that bind L-arginine with sub-micromolar dissociation constants and
high enantioselectivity. Nucleic Acids Res. 1996, 24, 1029-36.
I. Majerfeld, D. Puthenvedu, M. Yarus, RNA Affinity for Molecular LHistidine; Genetic Code Origins. J. Mol. Evol. 2005, 61, 226-35.
A. Shoji, M. Kuwahara, H. Ozaki, H. Sawai, Modified DNA Aptamer That
Binds the (R)-Isomer of a Thalidomide Derivative with High Enantioselectivity.
J. Am. Chem. Soc. 2007, 129, 1456-64.
P. B. Dervan, Molecular recognition of DNA by small molecules. Bioorg. Med.
7 References
[146]
[147]
[148]
[149]
[150]
[151]
[152]
[153]
[154]
[155]
[156]
[157]
[158]
[159]
[160]
186
Chem. 2001, 9, 2215-35.
G. Roelfes, B. L. Feringa, DNA-based asymmetric catalysis. Angew. Chem., Int.
Ed. Engl. 2005, 44, 3230-32.
R. Wombacher, S. Keiper, S. Suhm, A. Serganov, D. J. Patel, A. Jäschke,
Control of stereoselectivity in an enzymatic reaction by backdoor access.
Angew. Chem., Int. Ed. Engl. 2006, 45, 2469-72.
Liu and co-workers tethered the same phosphane derivative L1 to a 3’-amino
terminated DNA sequence and used the resulting conjugate in a DNA-templated
Wittig olefination. See: , X. Li, D. R. Liu, Stereoselectivity in DNA-templated
organic synthesis and its origins. J. Am. Chem. Soc. 2003, 125, 10188-9.
X. Li, D. R. Liu, DNA-templated organic synthesis: Nature's strategy for
controlling chemical reactivity applied to synthetic molecules. Angew. Chem.,
Int. Ed. Engl. 2004, 43, 4848-70.
S. K. Silverman, Deoxyribozymes: DNA catalysts for bioorganic chemistry. Org
Biomol Chem 2004, 2, 2701-6.
M. Famulok, A. Jenne, Catalysis based on nucleic acid structures. Top. Curr.
Chem. 1999, 202, 101-31.
C. Hammann, D. M. J. Lilley, Folding and activity of the hammerhead
ribozyme. ChemBioChem 2002, 3, 690-700.
D. M. J. Lilley, Structure, folding and mechanisms of ribozymes. Curr. Opin.
Struct. Biol. 2005, 15, 313-23.
B. T. Wimberly, D. E. Brodersen, W. M. Clemons, Jr., R. J. Morgan-Warren, A.
P. Carter, C. Vonrhein, T. Hartsch, V. Ramakrishnan, Structure of the 30S
ribosomal subunit. Nature 2000, 407, 327-39.
A. P. Carter, W. M. Clemons, D. E. Brodersen, R. J. Morgan-Warren, B. T.
Wimberly, V. Ramakrishnan, Functional insights from the structure of the 30S
ribosomal subunit and its interactions with antibiotics. Nature 2000, 407, 34048.
V. Ramakrishnan, Ribosome structure and the mechanism of translation. Cell
2002, 108, 557-72.
G. F. Joyce, Directed evolution of nucleic acid enzymes. Annu. Rev. Biochem.
2004, 73, 791-836.
R. Green, A. D. Ellington, J. W. Szostak, In vitro genetic analysis of the
Tetrahymena self-splicing intron. Nature 1990, 347, 406-8.
G. F. Joyce, In vitro evolution of nucleic acids. Curr. Opin. Struct. Biol. 1994, 4,
331-6.
A. Jäschke, Artificial ribozymes and deoxyribozymes. Curr. Opin. Struct. Biol.
7 References
[161]
[162]
[163]
[164]
[165]
[166]
[167]
[168]
[169]
[170]
[171]
[172]
[173]
[174]
[175]
[176]
187
2001, 11, 321-26.
A. Jäschke, S. Keiper, SELEX: systematic evolution of ligands by exponential
enrichment. Highlights in Bioorganic Chemistry 2004, 433-35.
R. Fiammengo, A. Jäschke, Nucleic acid enzymes. Curr. Opin. Biotechnol.
2005, 16, 614-21.
R. R. Breaker, G. F. Joyce, A DNA enzyme that cleaves RNA. Chem. Biol.
1994, 1, 223-9.
N. Carmi, L. A. Shultz, R. R. Breaker, In vitro selection of self-cleaving DNAs.
Chem. Biol. 1996, 3, 1039-46.
B. Cuenoud, J. W. Szostak, A DNA metalloenzyme with DNA ligase activity.
Nature 1995, 375, 611-14.
R. L. Coppins, S. K. Silverman, A DNA enzyme that mimics the first step of
RNA splicing. Nat. Struct. Mol. Biol. 2004, 11, 270-74.
Y. Li, D. Sen, A catalytic DNA for porphyrin metallation. Nat. Struct. Biol.
1996, 3, 743-7.
P. Travascio, A. J. Bennet, D. Y. Wang, D. Sen, A ribozyme and a catalytic
DNA with peroxidase activity: active sites versus cofactor-binding sites. Chem.
Biol. 1999, 6, 779-87.
Y. Li, R. R. Breaker, Phosphorylating DNA with DNA. Proc. Natl. Acad. Sci.
USA 1999, 96, 2746-51.
Y. Li, Y. Liu, R. R. Breaker, Capping DNA with DNA. Biochemistry 2000, 39,
3106-14.
T. L. Sheppard, P. Ordoukhanian, G. F. Joyce, A DNA enzyme with Nglycosylase activity. Proc. Natl. Acad. Sci. USA 2000, 97, 7802-07.
G. M. Emilsson, R. R. Breaker, Deoxyribozymes: new activities and new
applications. Cellular and Molecular Life Sciences 2002, 59, 596-607.
C. R. Geyer, D. Sen, Lanthanide probes for a phosphodiester-cleaving, leaddependent, DNAzyme. J. Mol. Biol. 1998, 275, 483-9.
G. J. Narlikar, D. Herschlag, Mechanistic aspects of enzymic catalysis: Lessons
from comparison of RNA and protein enzymes. Annu. Rev. Biochem. 1997, 66,
19-59.
A. Roth, R. R. Breaker, An amino acid as a cofactor for a catalytic
polynucleotide. Proc. Natl. Acad. Sci. USA 1998, 95, 6027-31.
R. Ting, J. M. Thomas, L. Lermer, D. M. Perrin, Substrate specificity and
kinetic framework of a DNAzyme with an expanded chemical repertoire: a
putative RNaseA mimic that catalyzes RNA hydrolysis independent of a
divalent metal cation. Nucleic Acids Res. 2004, 32, 6660-72.
7 References
188
[177] A. V. Sidorov, J. A. Grasby, D. M. Williams, Sequence-specific cleavage of
RNA in the absence of divalent metal ions by a DNAzyme incorporating
imidazolyl and amino functionalities. Nucleic Acids Res. 2004, 32, 1591-601.
[178] J. P. May, R. Ting, L. Lermer, J. M. Thomas, Y. Roupioz, D. M. Perrin,
Covalent Schiff Base Catalysis and Turnover by a DNAzyme: A M2+Independent AP-Endonuclease Mimic. J. Am. Chem. Soc. 2004, 126, 4145-56.
[179] K. Kruger, P. J. Grabowski, A. J. Zaug, J. Sands, D. E. Gottschling, T. R. Cech,
Self-splicing RNA: autoexcision and autocyclization of the ribosomal RNA
intervening sequence of Tetrahymena. Cell 1982, 31, 147-57.
[180] C. Guerrier-Takada, K. Gardiner, T. Marsh, N. Pace, S. Altman, The RNA
moiety of ribonuclease P is the catalytic subunit of the enzyme. Cell 1983, 35,
849-57.
[181] S. Tsukiji, S. B. Pattnaik, H. Suga, An alcohol dehydrogenase ribozyme. Nat.
Struct. Biol. 2003, 10, 713-17.
[182] D. Nieuwlandt, M. West, X. Cheng, G. Kirshenheuter, B. E. Eaton, The first
example of an RNA urea synthase: Selection through the enzyme active site of
human neutrophile elastase. ChemBioChem 2003, 4, 651-54.
[183] P. J. Unrau, D. P. Bartel, RNA-catalyzed nucleotide synthesis. Nature 1998,
395, 260-63.
[184] M. W. L. Lau, K. E. C. Cadieux, P. J. Unrau, Isolation of Fast Purine Nucleotide
Synthase Ribozymes. J. Am. Chem. Soc. 2004, 126, 15686-93.
[185] W. K. Johnston, P. J. Unrau, M. S. Lawrence, M. E. Glasner, D. P. Bartel, RNAcatalyzed RNA polymerization: accurate and general RNA-templated primer
extension. Science 2001, 292, 1319-25.
[186] M. S. Lawrence, D. P. Bartel, Processivity of ribozyme-catalyzed RNA
polymerization. Biochemistry 2003, 42, 8748-55.
[187] H. Murakami, H. Saito, H. Suga, A Versatile tRNA Aminoacylation Catalyst
Based on RNA. Chem. Biol. 2003, 10, 655-62.
[188] N. Li, F. Huang, Ribozyme-Catalyzed Aminoacylation from CoA Thioesters.
Biochemistry 2005, 44, 4582-90.
[189] T. M. Tarasow, S. L. Tarasow, B. E. Eaton, RNA-catalysed carbon-carbon bond
formation. Nature 1997, 389, 54-57.
[190] B. Seelig, A. Jäschke, A small catalytic RNA motif with Diels-Alderase activity.
Chem. Biol. 1999, 6, 167-76.
[191] B. Seelig, S. Keiper, F. Stuhlmann, A. Jäschke, Enantioselective ribozyme
catalysis of a bimolecular cycloaddition reaction. Angew. Chem., Int. Ed. Engl.
2000, 39, 4576-79.
7 References
189
[192] G. Sengle, A. Eisenfuhr, P. S. Arora, J. S. Nowick, M. Famulok, Novel RNA
catalysts for the Michael reaction. Chem. Biol. 2001, 8, 459-73.
[193] S. Fusz, A. Eisenfuehr, S. G. Srivatsan, A. Heckel, M. Famulok, A Ribozyme
for the Aldol Reaction. Chem. Biol. 2005, 12, 941-50.
[194] L. Jaeger, The new world of ribozymes. Curr. Opin. Struct. Biol. 1997, 7, 32435.
[195] M. Zivarts, Y. Liu, R. R. Breaker, Engineered allosteric ribozymes that respond
to specific divalent metal ions. Nucleic Acids Res. 2005, 33, 622-31.
[196] T. M. Tarasow, S. L. Tarasow, C. Tu, E. Kellogg, B. E. Eaton, Characteristics of
an RNA Diels-Alderase Active Site. J. Am. Chem. Soc. 1999, 121, 3614-17.
[197] T. M. Tarasow, E. Kellogg, B. L. Holley, D. Nieuwlandt, S. L. Tarasow, B. E.
Eaton, The Effect of Mutation on RNA Diels-Alderases. J. Am. Chem. Soc.
2004, 126, 11843-51.
[198] A. Serganov, S. Keiper, L. Malinina, V. Tereshko, E. Skripkin, C. Höbartner, A.
Polonskaia, T. Phan Anh, R. Wombacher, R. Micura, Z. Dauter, A. Jäschke, D.
J. Patel, Structural basis for Diels-Alder ribozyme-catalyzed carbon-carbon bond
formation. Nat. Struct. Mol. Biol. 2005, 12, 218-24.
[199] G. Roelfes, A. J. Boersma, B. L. Feringa, Highly enantioselective DNA-based
catalysis. Chem. Commun. 2006, 635-37.
[200] G. Roelfes, DNA and RNA induced enantioselectivity in chemical synthesis.
Mol. Biosyst. 2007, 3, 126-35.
[201] L. Ropartz, N. J. Meeuwenoord, G. A. van der Marel, P. W. N. M. van
Leeuwen, A. M. Z. Slawin, P. C. J. Kamer, Phosphine containing
oligonucleotides for the development of metallodeoxyribozymes. Chem.
Commun. 2007, 1556-58.
[202] J. Reedijk, Improved understanding in platinum antitumor chemistry. Chem.
Commun. 1996, 801-6.
[203] A. Anagnostopoulou, E. Moldrheim, N. Katsaros, E. Sletten, Interaction of cisand trans-RuCl2(DMSO)4 with the nucleotides GpA, d(GpA), ApG, d(ApG)
and d(CCTGGTCC): high-field NMR characterization of the reaction products.
J. Biol. Inorg. Chem. 1999, 4, 199-208.
[204] C. X. Zhang, S. J. Lippard, New metal complexes as potential therapeutics.
Curr. Op. Chem. Biol. 2003, 7, 481-89.
[205] K. E. Erkkila, D. T. Odom, J. K. Barton, Recognition and Reaction of
Metallointercalators with DNA. Chem. Rev. 1999, 99, 2777-95.
[206] M. B. Fleisher, K. C. Waterman, N. J. Turro, J. K. Barton, Light-induced
cleavage of DNA by metal complexes. Inorg. Chem. 1986, 25, 3549-51.
7 References
190
[207] L. Jacquet, J. M. Kelly, A. Kirsch-De Mesmaeker, Photoadduct between
tris(1,4,5,8-tetraazaphenanthrene)ruthenium(II) and guanosine monophosphate a model for a new mode of covalent binding of metal complexes to DNA. J.
Chem. Soc., Chem. Commun. 1995, 913-14.
[208] R. E. Molmlin, P. J. Dandliker, J. K. Barton, Charge transfer through the DNA
base stack. Angew. Chem., Int. Ed. Engl. 1998, 36, 2715-30.
[209] A. E. Friedman, J. C. Chambron, J. P. Sauvage, N. J. Turro, J. K. Barton, A
molecular light switch for DNA: Ru(bpy)2(dppz)2+. J. Am. Chem. Soc. 1990,
112, 4960-2.
[210] D. S. Sigman, Nuclease activity of 1,10-phenanthroline-copper ion. Acc. Chem.
Res. 1986, 19, 180-6.
[211] K. J. Humphreys, K. D. Karlin, S. E. Rokita, Efficient and specific strand
scission of DNA by a dinuclear copper complex: Comparative reactivity of
complexes with linked tris(2-pyridylmethyl)amine moieties. J. Am. Chem. Soc.
2002, 124, 6009-19.
[212] C. A. Claussen, E. C. Long, Nucleic acid recognition by metal complexes of
bleomycin. Chem. Rev. 1999, 99, 2797-816.
[213] M. Balaz, M. De Napoli, A. E. Holmes, A. Mammana, K. Nakanishi, N. Berova,
R. Purrello, A cationic zinc porphyrin as a chiroptical probe for Z-DNA. Angew.
Chem., Int. Ed. Engl. 2005, 44, 4006-09.
[214] J. K. Barton, L. A. Basile, A. Danishefsky, A. Alexandrescu, Chiral probes for
the
handedness
of
DNA
helixes:
enantiomers
of
tris(4,7diphenylphenanthroline)ruthenium(II). Proc. Natl. Acad. Sci. USA 1984, 81,
1961-5.
[215] I. Dubey, G. Pratviel, B. Meunier, Synthesis and DNA cleavage of 2'-O-aminolinked metalloporphyrin-oligonucleotide conjugates. J. Chem. Soc., Perkin
Trans. 1 2000, 3088-95.
[216] J. Telser, K. A. Cruickshank, K. S. Schanze, T. L. Netzel, DNA oligomers and
duplexes containing a covalently attached derivative of tris(2,2'bipyridine)ruthenium(II): synthesis and characterization by thermodynamic and
optical spectroscopic measurements. J. Am. Chem. Soc. 1989, 111, 7221-26.
[217] I. Ortmans, S. Content, N. Boutonnet, A. Kirsch-De Mesmaeker, W. Bannwarth,
J. F. Constant, E. Defrancq, J. Lhomme, Ru-labeled oligonucleotides for
photoinduced reactions on targeted DNA guanines. Chem. Eur. J. 1999, 5, 271221.
[218] D. Ossipov, S. Gohil, J. Chattopadhyaya, Synthesis of the DNA[Ru(tpy)(dppz)(CH3CN)]2+ conjugates and their photo cross-linking studies
7 References
[219]
[220]
[221]
[222]
[223]
[224]
[225]
[226]
[227]
[228]
[229]
[230]
191
with the complementary DNA strand. J. Am. Chem. Soc. 2002, 124, 13416-33.
C. H. B. Chen, D. S. Sigman, Sequence-specific scission of RNA by 1,10phenanthroline-copper linked to deoxyoligonucleotides. J. Am. Chem. Soc.
1988, 110, 6570-2.
J. Gallagher, C.-h. B. Chen, C. Q. Pan, D. M. Perrin, Y.-M. Cho, D. S. Sigman,
Optimizing the Targeted Chemical Nuclease Activity of 1,10-PhenanthrolineCopper by Ligand Modification. Bioconjugate Chem. 1996, 7, 413-20.
Liu and co-workers tethered the same phosphine derivative 2a to a 3’-amino
terminated DNA sequence and used the resulting conjugate in a DNA-templated
Wittig olefination. See:, Z. J. Gartner, M. W. Kanan, D. R. Liu, Multistep smallmolecule synthesis programmed by DNA templates. J. Am. Chem. Soc. 2002,
124, 10304-06.
K. Sakurai, T. M. Snyder, D. R. Liu, DNA-Templated Functional Group
Transformations Enable Sequence-Programmed Synthesis Using SmallMolecule Reagents. J. Am. Chem. Soc. 2005, 127, 1660-61.
O. Lentzen, E. Defrancq, J.-F. Constant, S. Schumm, D. Garcia-Fresnadillo, C.
Moucheron, P. Dumy, A. Kirsch-De Mesmaeker, Determination of DNA
guanine sites forming photo-adducts with Ru(II)-labeled oligonucleotides; DNA
polymerase inhibition by the resulting photo-crosslinking. J. Biol. Inorg. Chem.
2004, 9, 100-08.
T. Ihara, Y. Maruo, S. Takenaka, M. Takagi, Ferrocene-oligonucleotide
conjugates for electrochemical probing of DNA. Nucleic Acids Res. 1996, 24,
4273-80.
R. E. Holmlin, P. J. Dandliker, J. K. Barton, Synthesis of MetallointercalatorDNA Conjugates on a Solid Support. Bioconjugate Chem. 1999, 10, 1122-30.
A. E. Beilstein, M. W. Grinstaff, On-column derivatization of
oligodeoxynucleotides with ferrocene. Chem. Commun. 2000, 509-10.
A. E. Beilstein, M. W. Grinstaff, Synthesis and characterization of ferrocenelabeled oligodeoxynucleotides. J. Organomet. Chem. 2001, 637-639, 398-406.
S. I. Khan, M. W. Grinstaff, Palladium(0)-Catalyzed Modification of
Oligonucleotides during Automated Solid-Phase Synthesis. J. Am. Chem. Soc.
1999, 121, 4704-05.
K. Wiederholt, L. W. McLaughlin, A 2,2'-bipyridine ligand for incorporation
into oligodeoxynucleotides: synthesis, stability and fluorescence properties of
ruthenium-DNA complexes. Nucleic Acids Res. 1999, 27, 2487-93.
W. C. Putnam, A. T. Daniher, B. N. Trawick, J. K. Bashkin, Efficient new
ribozyme mimics: direct mapping of molecular design principles from small
7 References
[231]
[232]
[233]
[234]
[235]
[236]
[237]
[238]
[239]
[240]
[241]
[242]
[243]
[244]
192
molecules to macromolecular, biomimetic catalysts. Nucleic Acids Res. 2001,
29, 2199-204.
G. B. Dreyer, P. B. Dervan, Sequence-specific cleavage of single-stranded
DNA: oligodeoxynucleotide-EDTA X Fe(II). Proc. Natl. Acad. Sci. USA 1985,
82, 968-72.
D. E. Bergstrom, N. P. Gerry, Precision Sequence-Specific Cleavage of a
Nucleic Acid by a Minor-Groove-Directed Metal-Binding Ligand Linked
through N-2 of Deoxyguanosine. J. Am. Chem. Soc. 1994, 116, 12067-68.
S. T. Gaballah, C. E. Kerr, B. E. Eaton, T. L. Netzel, Synthesis of 5-(2,2'bipyridinyl and 2,2'-bipyridinediiumyl)-2'-deoxyuridine nucleosides: precursors
to metallo-DNA conjugates. Nucelosides Nucleotides Nucleic Acids 2002, 21,
547-60.
D. J. Hurley, Y. Tor, Ru(II) and Os(II) Nucleosides and Oligonucleotides:
Synthesis and Properties. J. Am. Chem. Soc. 2002, 124, 3749-62.
C. J. Yu, H. Yowanto, Y. Wan, T. J. Meade, Y. Chong, M. Strong, L. H.
Donilon, J. F. Kayyem, M. Gozin, G. F. Blackburn, Uridine-Conjugated
Ferrocene DNA Oligonucleotides: Unexpected Cyclization Reaction of the
Uridine Base. J. Am. Chem. Soc. 2000, 122, 6767-68.
E. S. Krider, J. J. Rack, N. L. Frank, T. J. Meade, Automated Synthesis of 3'Metalated Oligonucleotides. Inorg. Chem. 2001, 40, 4002-09.
E. Meggers, P. L. Holland, W. B. Tolman, F. E. Romesberg, P. G. Schultz, A
Novel Copper-Mediated DNA Base Pair. J. Am. Chem. Soc. 2000, 122, 1071415.
H. Weizman, Y. Tor, 2,2'-Bipyridine Ligandoside: A Novel Building Block for
Modifying DNA with Intra-Duplex Metal Complexes. J. Am. Chem. Soc. 2001,
123, 3375-76.
J. L. Czlapinski, T. L. Sheppard, Nucleic Acid Template-Directed Assembly of
Metallosalen-DNA Conjugates. J. Am. Chem. Soc. 2001, 123, 8618-19.
M. Shionoya, K. Tanaka, Artificial metallo-DNA: a bio-inspired approach to
metal array programming. Curr. Op. Chem. Biol. 2004, 8, 592-97.
G. H. Clever, K. Polborn, T. Carell, A highly DNA-duplex-stabilizing metalsalen base pair. Angew. Chem., Int. Ed. Engl. 2005, 44, 7204-08.
K. Tanaka, A. Tengeiji, T. Kato, N. Toyama, M. Shionoya, A Discrete SelfAssembled Metal Array in Artificial DNA. Science 2003, 299, 1212-13.
G. H. Clever, T. Carell, Controlled stacking of 10 transition-metal ions inside a
DNA duplex. Angew. Chem., Int. Ed. Engl. 2007, 46, 250-53.
J. D. Vaught, T. Dewey, B. E. Eaton, T7 RNA polymerase transcription with 5-
7 References
[245]
[246]
[247]
[248]
[249]
[250]
193
position modified UTP derivatives. J. Am. Chem. Soc. 2004, 126, 11231-7.
O. Thum, S. Jager, M. Famulok, Functionalized DNA: a new replicable
biopolymer. Angew. Chem., Int. Ed. Engl. 2001, 40, 3990-93.
S. Jaeger, G. Rasched, H. Kornreich-Leshem, M. Engeser, O. Thum, M.
Famulok, A Versatile Toolbox for Variable DNA Functionalization at High
Density. J. Am. Chem. Soc. 2005, 127, 15071-82.
H. Sawai, A. Ozaki-Nakamura, M. Mine, H. Ozaki, Synthesis of New Modified
DNAs by Hyperthermophilic DNA Polymerase: Substrate and Template
Specificity of Functionalized Thymidine Analogues Bearing an sp3-Hybridized
Carbon at the C5 a-Position for Several DNA Polymerases. Bioconjugate Chem.
2002, 13, 309-16.
H. Weizman, Y. Tor, Redox-active metal-containing nucleotides: Synthesis,
tunability, and enzymatic incorporation into DNA. J. Am. Chem. Soc. 2002, 124,
1568-69.
S. A. Benner, Understanding Nucleic Acids Using Synthetic Chemistry. Acc.
Chem. Res. 2004, 37, 784-97.
Unpublished data Dr. R. Fiammengo, University of Heidelberg, 2004-2005. The
decaethyleneglycol spacer S, the ligands L3, L8 and the complex
[Rh(BINAP)(nbd)]BF4 were kindly provided by Dr. R. Fiammengo.
[251] M. A. Said, K. C. K. Swamy, M. Veith, V. Huch, Reactivity of cyclic arsenites
and phosphites: x-ray structures of bis(5,5-dimethyl-1,3,2-dioxarsenan-2-yl)
ether and bis(2,4,8,10-tetra-tert-butyl-12H-dibenzo[d,g][1,3,2]dioxarsenocin-6yl) ether. J. Chem. Soc., Perkin Trans. 1 1995, 2945-51.
[252] J.-G. Boiteau, R. Imbos, A. J. Minnaard, B. L. Feringa, Rhodium-Catalyzed
Asymmetric Conjugate Additions of Boronic Acids Using Monodentate
Phosphoramidite Ligands. Org. Lett. 2003, 5, 681-84.
[253] C. Monti, C. Gennari, U. Piarulli, Enantioselective conjugate addition of
phenylboronic acid to enones catalyzed by a chiral tropos/atropos rhodium
complex at the coalescence temperature. Chem. Commun. 2005, 5281-83.
[254] M. Caprioara, R. Fiammengo, M. Engeser, A. Jäschke, DNA-based phosphane
ligands. Chem. Eur. J. 2007, 13, 2089-95, S89/1-S89/10.
[255] A. M. MacMillan, G. L. Verdine, Synthesis of functionally tethered
oligodeoxynucleotides by the convertible nucleoside approach. J. Org. Chem.
1990, 55, 5931-33.
[256] Y. Z. Xu, Q. Zheng, P. F. Swann, Synthesis of DNA containing modified bases
by post-synthetic substitution. Synthesis of oligomers containing 4-substituted
thymine:
O4-alkylthymine,
5-methylcytosine,
N4-dimethylamino-5-
7 References
[257]
[258]
[259]
[260]
[261]
[262]
194
methylcytosine, and 4-thiothymine. J. Org. Chem. 1992, 57, 3839-45.
N. Haginoya, A. Ono, Y. Nomura, Y. Ueno, A. Matsuda, Nucleosides and
Nucleotides. 160. Synthesis of Oligodeoxyribonucleotides Containing 5-(NAminoalkyl)carbamoyl-2'-deoxyuridines by a New Postsynthetic Modification
Method and Their Thermal Stability and Nuclease-Resistance Properties.
Bioconjugate Chem. 1997, 8, 271-80.
O. Gorchs, M. Hernandez, L. Garriga, E. Pedroso, A. Grandas, J. Farras, A New
Method for the Preparation of Modified Oligonucleotides. Org. Lett. 2002, 4,
1827-30.
R. Narukulla, D. E. G. Shuker, Y.-Z. Xu, Post-synthetic and site-specific
modification of endocyclic nitrogen atoms of purines in DNA and its potential
for biological and structural studies. Nucleic Acids Res. 2005, 33, 1767-78.
A. M. MacMillan, G. L. Verdine, Engineering tethered DNA molecules by the
convertible nucleoside approach. Tetrahedron 1991, 47, 2603-16.
R. S. Coleman, E. A. Kesicki, Synthesis and Postsynthetic Modification of
Oligodeoxynucleotides Containing 4-Thio-2'-deoxyuridine (ds4U). J. Am.
Chem. Soc. 1994, 116, 11636-42.
C. A. Frederick, G. J. Quigley, G. A. Van der Marel, J. H. Van Boom, A. H. J.
Wang, A. Rich, Methylation of the EcoRI recognition site does not alter DNA
conformation: the crystal structure of d(CGCGAm6ATTCGCG) at 2.0-.ANG.
resolution. J. Biol. Chem. 1988, 263, 17872-9.
[263] http://www.sigmaaldrich.com/catalog/search/TablePage/14572938.
[264] J.-L. Mergny, L. Lacroix, Analysis of thermal melting curves. Oligonucleotides
2003, 13, 515-37.
[265] G. Bonner, A. M. Klibanov, Structural stability of DNA in nonaqueous solvents.
Biotechnol. Bioeng. 2000, 68, 339-44.
[266] G. Lober, H. Schutz, V. Kleinwachter, Effect of organic solvents on the
properties of the complexes of DNA with proflavine and similar compounds.
Biopolymers 1972, 11, 2439-59.
[267] J. Casey, N. Davidson, Rates of formation and thermal stabilities of RNA:DNA
and DNA:DNA duplexes at high concentrations of formamide. Nucleic Acids
Res. 1977, 4, 1539-52.
[268] S. I. Kirin, P. Duebon, T. Weyhermueller, E. Bill, N. Metzler-Nolte, Amino
Acid and Peptide Bioconjugates of Copper(II) and Zinc(II) Complexes with a
Modified N,N-Bis(2-picolyl)amine Ligand. Inorg. Chem. 2005, 44, 5405-15.
[269] S. R. Gilbertson, G. Chen, M. McLoughlin, Versatile Building Block for the
Synthesis
of
Phosphine-Containing
Peptides:
The
Sulfide
of
7 References
[270]
[271]
[272]
[273]
[274]
[275]
[276]
[277]
[278]
[279]
[280]
[281]
[282]
195
Diphenylphosphinoserine. J. Am. Chem. Soc. 1994, 116, 4481-2.
O. Herd, A. Hessler, M. Hingst, M. Tepper, O. Stelzer, Water soluble
phosphines. VIII. Palladium-catalyzed P-C cross coupling reactions between
primary or secondary phosphines and functional aryl iodides - a novel synthetic
route to water soluble phosphines. J. Organomet. Chem. 1996, 522, 69-76.
Acidifying results in oxazoline ring opening. See:, A. Laaziri, J. Uziel, S. Juge,
Electrophilic ring opening of oxazolines derived from serine and threonine: a
practical entry to N(N)-protected b-halogeno a-amino esters. Tetrahedron:
Asymmetry 1998, 9, 437-47.
B. M. Novak, R. H. Grubbs, Catalytic organometallic chemistry in water: the
aqueous ring-opening metathesis polymerization of 7-oxanorbornene
derivatives. J. Am. Chem. Soc. 1988, 110, 7542-3.
B. Cornils, W. A. Herrmann, Editors, Aqueous-Phase Organometallic Catalysis:
Concepts and Applications, 1998.
D. Sinou, Asymmetric organometallic-catalyzed reactions in aqueous media.
Adv. Synth. Catal. 2002, 344, 221-37.
S. P. Smidt, Iridium-Catalysed Enantioselective Hydrogenation - New P,N
Ligands and Mechanistic Investigations. Inauguraldissertation 2003.
S. P. Smidt, A. Pfaltz, E. Martinez-Viviente, P. S. Pregosin, A. Albinati, X-ray
and NOE Studies on Trinuclear Iridium Hydride Phosphino Oxazoline (PHOX)
Complexes. Organometallics 2003, 22, 1000-09.
J.-G. Boiteau, A. J. Minnaard, B. L. Feringa, High Efficiency and
Enantioselectivity in the Rh-Catalyzed Conjugate Addition of Arylboronic
Acids Using Monodentate Phosphoramidites. J. Org. Chem. 2003, 68, 9481-84.
R. Itooka, Y. Iguchi, N. Miyaura, Rhodium-Catalyzed 1,4-Addition of
Arylboronic Acids to a,b-Unsaturated Carbonyl Compounds: Large Accelerating
Effects of Bases and Ligands. J. Org. Chem. 2003, 68, 6000-04.
M. Lautens, A. Roy, K. Fukuoka, K. Fagnou, B. Martin-Matute, RhodiumCatalyzed Coupling Reactions of Arylboronic Acids to Olefins in Aqueous
Media. J. Am. Chem. Soc. 2001, 123, 5358-59.
T. Hayashi, M. Takahashi, Y. Takaya, M. Ogasawara, Catalytic Cycle of
Rhodium-Catalyzed Asymmetric 1,4-Addition of Organoboronic Acids.
Arylrhodium, Oxa-p-allylrhodium, and Hydroxorhodium Intermediates. J. Am.
Chem. Soc. 2002, 124, 5052-58.
B. M. Trost, M. L. Crawley, Asymmetric Transition-Metal-Catalyzed Allylic
Alkylations: Applications in Total Synthesis. Chem. Rev. 2003, 103, 2921-43.
F. Glorius, A. Pfaltz, Enantioselective Molybdenum-Catalyzed Allylic
7 References
[283]
[284]
[285]
[286]
[287]
[288]
[289]
[290]
[291]
[292]
[293]
[294]
196
Alkylation Using Chiral Bisoxazoline Ligands. Org. Lett. 1999, 1, 141-44.
B. M. Trost, K. Dogra, I. Hachiya, T. Emura, D. L. Hughes, S. Krska, R. A.
Reamer, M. Palucki, N. Yasuda, P. J. Reider, Designed ligands as probes for the
catalytic binding mode in mo-catalyzed asymmetric allylic alkylation. Angew.
Chem., Int. Ed. Engl. 2002, 41, 1929-32.
O. Belda, C. Moberg, Molybdenum-Catalyzed Asymmetric Allylic Alkylations.
Acc. Chem. Res. 2004, 37, 159-67.
G. C. Lloyd-Jones, A. Pfaltz, Chiral phosphinodihydrooxazoles in asymmetric
catalysis: tungsten-catalyzed allylic substitution. Angew. Chem., Int. Ed. Engl.
1995, 34, 462-4.
P. A. Evans, J. D. Nelson, Conservation of Absolute Configuration in the
Acyclic Rhodium-Catalyzed Allylic Alkylation Reaction: Evidence for an Enyl
(s + p) Organorhodium Intermediate. J. Am. Chem. Soc. 1998, 120, 5581-82.
D. Enders, B. Jandeleit, S. von Berg, G. Raabe, J. Runsink, Iron-Mediated
Allylic Substitution Reactions with Chirality Transfer. Stereochemistry of the
Formation of Diastereo- and Enantiomerically Enriched Olefinic and Allylic
Tetracarbonyl Iron Complexes. Organometallics 2001, 20, 4312-32.
Y. Morisaki, T. Kondo, T. Mitsudo, Ruthenium-Catalyzed Allylic Substitution
of Cyclic Allyl Carbonates with Nucleophiles. Stereoselectivity and Scope of the
Reaction. Organometallics 1999, 18, 4742-46.
B. M. Trost, P. L. Fraisse, Z. T. Ball, A stereospecific ruthenium-catalyzed
allylic alkylation. Angew. Chem., Int. Ed. Engl. 2002, 41, 1059-61.
R. Takeuchi, M. Kashio, Highly selective allylic alkylation with a carbon
nucleophile at the more substituted allylic terminus catalyzed by an iridium
complex: an efficient method for constructing quaternary carbon centers. Angew.
Chem., Int. Ed. Engl. 1997, 36, 263-65.
R. Takeuchi, N. Shiga, Complete Retention of Z Geometry in Allylic
Substitution Catalyzed by an Iridium Complex. Org. Lett. 1999, 1, 265-67.
T. Ohmura, J. F. Hartwig, Regio- and Enantioselective Allylic Amination of
Achiral Allylic Esters Catalyzed by an Iridium-Phosphoramidite Complex. J.
Am. Chem. Soc. 2002, 124, 15164-65.
F. Lopez, T. Ohmura, J. F. Hartwig, Regio- and Enantioselective IridiumCatalyzed Intermolecular Allylic Etherification of Achiral Allylic Carbonates
with Phenoxides. J. Am. Chem. Soc. 2003, 125, 3426-27.
C. A. Kiener, C. Shu, C. Incarvito, J. F. Hartwig, Identification of an Activated
Catalyst in the Iridium-Catalyzed Allylic Amination and Etherification.
Increased Rates, Scope, and Selectivity. J. Am. Chem. Soc. 2003, 125, 14272-73.
7 References
197
[295] A. Leitner, S. Shekhar, M. J. Pouy, J. F. Hartwig, A Simple Iridium Catalyst
with a Single Resolved Stereocenter for Enantioselective Allylic Amination.
Catalyst Selection from Mechanistic Analysis. J. Am. Chem. Soc. 2005, 127,
15506-14.
[296] A. Leitner, C. Shu, J. F. Hartwig, Editing the stereochemical elements in an
iridium catalyst for enantioselective allylic amination. Proc. Natl. Acad. Sci.
USA 2004, 101, 5830-33.
[297] D. Polet, A. Alexakis, Kinetic Study of Various Phosphoramidite Ligands in the
Iridium-Catalyzed Allylic Substitution. Org. Lett. 2005, 7, 1621-24.
[298] D. Polet, A. Alexakis, K. Tissot-Croset, C. Corminboeuf, K. Ditrich,
Phosphoramidite ligands in iridium-catalyzed allylic substitution. Chem. Eur. J.
2006, 12, 3596-609.
[299] U. M. Lindstroem, Stereoselective Organic Reactions in Water. Chem. Rev.
2002, 102, 2751-71.
[300] Y. Uozumi, H. Danjo, T. Hayashi, Palladium-catalyzed asymmetric allylic
substitution in aqueous media using amphiphilic resin-supported MOP ligands.
Tetrahedron Lett. 1998, 39, 8303-06.
[301] S. dos Santos, Y. Tong, F. Quignard, A. Choplin, D. Sinou, J. P. Dutasta,
Supported Aqueous-Phase Palladium Catalysts for the Reaction of Allylic
Substitution: Toward an Understanding of the Catalytic System.
Organometallics 1998, 17, 78-89.
[302] M. Feuerstein, D. Laurenti, H. Doucet, M. Santelli, Palladium-tetraphosphine
catalysed allylic substitution in water. Tetrahedron Lett. 2001, 42, 2313-15.
[303] T. Hashizume, K. Yonehara, K. Ohe, S. Uemura, A Novel Amphiphilic Chiral
Ligand Derived from D-Glucosamine. Application to Palladium-Catalyzed
Asymmetric Allylic Substitution Reaction in an Aqueous or an Organic
Medium, Allowing for Catalyst Recycling. J. Org. Chem. 2000, 65, 5197-201.
[304] B. Wiese, G. Helmchen, Chiral phosphinooxazolines with a bi- or tricyclic
oxazoline moiety - applications in Pd-catalyzed allylic alkylations. Tetrahedron
Lett. 1998, 39, 5727-30.
[305] W. S. Sheldrick, B. Guenther, Synthesis and structure of the mixed bridged
diene-rhodium(I) complex [(COD)Rh(m-Cl)(m-OAc)Rh(COD)]. Reactions with
the modified purine bases N6,N6-dimethyladenine and 8-aza-9-methyladenine.
J. Organomet. Chem. 1989, 375, 233-43.
[306] J. D. Engel, P. H. von Hippel, Effects of methylation on the stability of nucleic
acid conformations: studies at the monomer level. Biochemistry 1974, 13, 414358.
7 References
198
[307] C. R. Allerson, S. L. Chen, G. L. Verdine, A Chemical Method for Site-Specific
Modification of RNA: The Convertible Nucleoside Approach. J. Am. Chem.
Soc. 1997, 119, 7423-33.
[308] J. Sambrook, D. W. Russell, Molecular Cloning. A Laboratory Manual, Vol. 13, Cold Spring Harbor, New York, 2001.
[309] F. M. Ausubel, R. Brent, R. E. Kingston, M. D. D., J. G. Seidman, J. A. Smith,
K. Struhl, Current Protocols in Molecular Biology, Vol. 1, John Wiley & Sons,
Inc., 2004.
[310] H. Hocke, Y. Uozumi, A simple synthetic approach to homochiral 6- and 6'substituted 1,1'-binaphthyl derivatives. Tetrahedron 2003, 59, 619-30.
[311] A. Rimkus, N. Sewald, First Synthesis of a b2-Homoamino Acid by
Enantioselective Catalysis. Org. Lett. 2003, 5, 79-80.
[312] J. Duran, M. Gulias, L. Castedo, J. L. Mascarenas, Ligand-Induced Acceleration
of the Intramolecular [3 + 2] Cycloaddition between Alkynes and
Alkylidenecyclopropanes. Org. Lett. 2005, 7, 5693-96.
[313] J. Hu, M. J. Miller, Total Synthesis of a Mycobactin S, a Siderophore and
Growth Promoter of Mycobacterium Smegmatis, and Determination of its
Growth Inhibitory Activity against Mycobacterium tuberculosis. J. Am. Chem.
Soc. 1997, 119, 3462-68.
[314] V. Ravindar, H. Schumann, H. Hemling, J. Blum, Synthesis and structure
determination of some platinum(II) complexes with hydrophilic carboxylated
tertiary phosphine ligands. Inorg. Chim. Acta 1995, 240, 145-52.
[315] S. L. Beaucage, R. P. Iyer, The functionalization of oligonucleotides via
phosphoramidite derivatives. Tetrahedron 1993, 49, 1925-63.
List of publications
Papers
M. Caprioara, R. Fiammengo, M. Engeser, A. Jäschke, DNA-Based Phosphane
Ligands, Chemistry - A European Journal, 2007, 13, 2089-2095
Posters
- R. Fiammengo, M. Caprioara, P. Fournier, K. Musilek, N. Sauer, A. Jäschke Towards RNA-based hybrid catalysts, Joint Workshop “Templates meet Catalysis”,
June 7-8 2004 Bonn, Germany
- M. Caprioara, R. Fiammengo, C. Kuhmann, A. Jäschke - Functionalized DNA and
RNA as Ligands for Asymmetric Catalysis, Nucleic Acid Chemical Biology (NACB)
PhD Summer School, June 19-23 2005 Odense, Denmark
- Roberto Fiammengo, Mihaela Caprioara, Pierre Fournier, Andres Jäschke - Selective
hybrid catalysts based on nucleic acids, Heidelberg Forum of Molecular Catalysis, 8th
July 2005, Heidelberg, Germany
- M. Caprioara, R. Fiammengo, P. Fournier, A. Jäschke - Functionalized DNA and
RNA as Ligands for Asymmetric Catalysis and in vitro selection of Hybrid Catalysts,
“Concepts and Advances in Modern Catalysis” Joint Workshop, University of
California, Berkley, Northwestern University, Evanston, University of Heidelberg, May
5-6 2006, Heidelberg, Germany
- M. Caprioara, P. Fournier, R. Fiammengo, A. Jäschke - DNA-Based Ligands for
Transition Metals and Use in Asymmetric Catalysis, Heidelberg Forum of Molecular
Catalysis, June 2007, Heidelberg, Germany
Oral presentations
M. Caprioara, P. Fournier, R. Fiammengo, A. Jäschke - Towards RNA-based Hybrid
Catalysts, Groningen Meets Heidelberg - Catalysis and More, 13rd June 2007,
Heidelberg, Germany
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement