UNCORRECTED PROOF

UNCORRECTED PROOF
Book_Wang_9781597454636_Proof1_December 31, 2007
9
01
02
05
06
07
08
Difference In-Gel Electrophoresis:
A High-Resolution Protein Biomarker
Research Tool
RO
O
04
F
03
09
10
11
David S. Gibson, David Bramwell, and Caitriona Scaife
DP
12
13
14
15
Summary
16
19
20
21
22
23
TE
18
Difference in-gel electrophoresis (DIGE) is a recent adaptation of conventional twodimensional gel electrophoresis (2-DE) that incorporates novel fluorescent labels, has
multiplex attributes, and boasts software-assisted image analysis. Combined, these characteristics offer significant benefits in accuracy and reproducibility to quantify differential protein
expression levels between biological samples. The DIGE technique and materials required
to perform it are described in detail within. The principles behind consistent gel image
acquisition and reliable image analysis are also considered. Within the context of biomarker
and drug target discovery, this method simplifies analysis, increases sample throughput, and
represents a reliable 2-DE platform.
RR
EC
17
24
25
26
Key Words: Biomarker; Cy dye; DIGE; fluorescent difference in-gel electrophoresis;
proteomics; two-dimensional gel electrophoresis
27
28
29
1. INTRODUCTION TO DIFFERENCE IN-GEL
ELECTROPHORESIS
30
33
34
35
36
37
38
39
Two-dimensional gel electrophoresis (2-DE) is an established platform
that facilitates the analysis of complex protein mixtures. O’Farrell was first
to introduce high-resolution two-dimensional electrophoresis by resolving
proteins to individual isoelectric point and molecular weight coordinates (1).
The main asset of this method is that it provides a global view of the state of
proteins within a sample. In theory, thousands of proteins can be visualized
at once, giving a unique qualitative “map” or “fingerprint” of changes
CO
32
UN
31
40
From: Methods in Pharmacology and Toxicology: Biomarker Methods in Drug Discovery and Development
Edited by: F. Wang © Humana Press, Totowa, NJ
41
189
Book_Wang_9781597454636_Proof1_December 31, 2007
190
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
F
RO
O
06
DP
05
TE
04
RR
EC
03
between given samples. Though many developments, such as standardized
immobilized pH gradients, have led to vast improvement in inter-run
consistency, deficiencies in sensitivity and spot matching have necessitated further adaptation using fluorescent stains. Comparison between large
groups of conventionally (silver or Coomassie) stained gels is complicated
by spot to spot warping, caused by variations in sodium dodecyl sulfate–
polyacrylamide gel electrophoresis (SDS-PAGE) gel casting, electric and
pH fields, and thermal fluctuations during electrophoresis. This leads to
problems in spot matching and necessitates multiple gel replicates to prevent
assumptions on mismatched proteins. In other words, gel to gel heterogeneity makes it difficult to distinguish with confidence between variations
in the technique and those of genuine induced biological change, such as
in disease states (2). Difference in-gel electrophoresis (DIGE) addresses a
number of these issues in that two to three samples can be subjected to
exactly the same running conditions within a single gel. Unlu et al. developed
DIGE to allow a more direct and reproducible comparison between protein
samples, differentiated by prelabeling with spectrally resolvable fluorescent
cyanine, or Cy, dyes (3). The Cy dyes are charge matched with the residues
they bind to within the proteins of a given sample and have similar
molecular weights (0.5 kDa), thus result in only slight gel shifts. The Cy
dyes are based on extended organic ring structures and hence are highly
hydrophobic. Concerns with protein precipitation prior to electrophoresis
have been surmounted by using a “minimal labeling” strategy, whereby
binding is limited to only 1% to 2% of lysine residues available within a
sample (4).
Excitation of each fluor allows the creation of a digital image of each
individually labeled sample. These dyes give additional validity to the twodimensional technique in the form of higher sensitivity, wider dynamic
range, and linearity of detection. Detection limits of 0.025 ng are possible,
with a dynamic range around five orders of magnitude. One of the strongest
features of the technique, however, is the ability to include an internal pooled
standard, which is loaded on all gels within an experiment (5). The internal
standard permits the linking of all gels in an experiment, thus offering
more reliable and intuitive software-assisted comparisons. The accuracy of
protein quantification between samples is increased dramatically, and much
smaller changes in protein expression can be studied with greater confidence. Evaluations of DIGE alongside traditional and more recent proteomic
methods using isotope-coded or isobaric tags (cICAT and iTRAQ) reveal
that it remains competitive in sensitivity and can be used with confidence
as a platform for drug discovery and development (4,6).
CO
02
UN
01
Gibson et al.
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
05
06
07
08
09
10
11
12
13
14
15
16
17
F
04
Because three cyanine dyes are available, up to three separate protein
samples can be labeled per gel. A pairwise analysis and organization could
also be used (akin to gene chip analysis), where control versus drug-treated
samples are labelled with Cy3 and Cy5 only. When normalization of expression
levels is desired across a number of different experiments and within the
one experiment, adequate quantities of each sample should be available to
create a common pooled internal standard. The internal standard can be distinguished from experimental samples by labeling with Cy2 dye. Anomalies in
spot intensity due to preferential labeling can be eliminated by randomized or
reciprocal labeling, in which half of each experimental group is labeled with
Cy3 and the other with Cy5 (7). In order to distinguish intrinsic, interindividual biological variation from genuine changes in protein expression,
biological replicates should be included in each experimental group. A recent
study focused on the DIGE technique has shown that a minimum of four
replicate gels is required to maintain a 95% chance of avoiding false
negatives, when a twofold change in expression is considered significant (8).
RO
O
03
2. DIGE EXPERIMENT CONSIDERATIONS
DP
02
18
19
3. SAMPLE PREPARATION
20
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
RR
EC
23
CO
22
Plasma and synovial fluid are used in this chapter to illustrate and describe
the steps required for the purification and minimal fluorescent labeling of
body fluid samples. For details of how to prepare cell lysates, with both
minimal and saturation types of labeling, one can refer to the Ettan DIGE
system user manual (9). The following reagents and conditions have been
used in our laboratory to produce reliable data with clinical relevance to
patient outcome but could also be applied to prospective drug trial to monitor
therapeutic response. Sample preparation should be consistent and kept as
simple as possible to reduce inter-run inconsistencies. Protein modifications
during sample preparation must be prevented, particularly degradation due
to endogenous proteolytic enzymes. Such changes in samples analyzed by
gel-based approaches can translate into misleading artifact spots with novel
molecular weights.
3.1. Sample Purification and Assay
UN
21
TE
01
191
Cellular or particulate material should be removed from the body fluid
by centrifugation prior to any further purification steps. This circumvents
contamination by sub-proteomes other than that of the body fluid that is
to be analyzed. Endogenous protease activity should be inactivated for
reasons already eluded to above. A number of approaches are possible, with
Book_Wang_9781597454636_Proof1_December 31, 2007
192
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
F
RO
O
06
AQ2
DP
05
AQ1
TE
04
RR
EC
03
varying consequences to the resulting sample integrity. Protease inhibitor
cocktails (such as the Complete Protease Inhibitor Cocktail Tablets; Roche)
can be used to inactivate a wide variety of degradative enzyme classes
including cysteine, matrix metallo and serine proteases. This remains our
preferred method of body fluid stabilization, and an adaptation is now
also available, in the form of a blood tube with proprietary inhibitors for
immediate and convenient sample protection (BD P100). Some authors,
however, caution against their use in certain applications, as artifacts can
result from modified protein charge, or peptide-based inhibitors such as
leupeptin may interfere with mass spectrometry analysis (10,11). Proteases
may also be inactivated by high or low pH extremes with Tris buffer or
trichloroacetic acid (TCA), respectively, or alternatively total protein can
be precipitated by TCA/acetone. In balance though, protein yield may be
diminished by incomplete precipitation or resolubilization. Once stabilized,
salts can be removed from protein samples (if higher than 10 mM) by
dialysis with low-molecular-weight cut-off membranes, though if analysis
of small peptides is desired, precipitation could be implemented. Other
macromolecules such as lipids, polysaccharides, and nucleic acids should be
removed by organic solvent, unless present at low concentrations (as with
plasma). The sample can be lyophilized if concentration is necessary (5 to
10 mg/mL is an ideal protein concentration, though labeling of 1 mg/mL is
possible) and resuspended in a minimal quantity of DIGE-compatible lysis
buffer [DLB; 30 mM Tris, 7 M urea, 2 M thiourea, 4% (w/v) CHAPS, pH
8.5]. Ampholytes and dithiothreitol (DTT) are omitted from the lysis buffer
prior to the labeling reaction as both primary amines and thiol groups will
compete with the proteins for the available Cy dye. The pH of the sample
to be labeled is also critical to the reaction, so check that the sample pH is
8.5 by spotting on a pH indicator strip and, if necessary, make drop-wise
adjustments with dilute sodium hydroxide. The concentration of protein in
each sample should be assayed either by Bradford reagent or using the
proprietary Ettan 2D Quant kit (GE Healthcare).
CO
02
3.2. Sample Labeling
Aside from the pH and protein concentration recommendations already
made, the efficiency of minimal dye labeling is dependent on the ratio of
dye to protein (400 pmol Cy dye to 50 μg protein is cited in the DIGE
user manual; GE Healthcare). The Cy dye fluors should be reconstituted
in anhydrous dimethyl formamide under the manufacturer’s guidelines to
create a 1 mM stock solution. Each has a characteristic deep color as follows:
Cy3, red; Cy5, blue; and Cy2, yellow (as shown in Fig. 1). The sample and
dye quantities required for a six-gel, three-dye pilot experiment are shown
UN
01
Gibson et al.
AQ3
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
193
01
02
F
03
04
RO
O
05
06
07
08
09
10
11
DP
12
13
14
15
16
TE
17
18
19
20
RR
EC
21
22
23
24
25
Fig. 1. Schematic representation of the laboratory procedures involved in a
typical DIGE experiment.
26
29
30
31
32
33
34
35
36
37
38
39
40
41
CO
28
in the following worked example (Table 1). Paired plasma and synovial
fluid samples from six patients (A–F) are labeled with Cy3 and Cy5. An
experimental design incorporating randomization of sample labeling and
loading across gels is demonstrated to avoid systematic errors. An internal
pooled standard is generated by combining equal amounts of all matched
plasma and synovial fluid samples, followed by Cy2 dye labeling. Sufficient
pooled internal standard is prepared to allow enough aliquots for each gel
in the experiment. It is also prudent to create a slight excess (10% to 20%)
of each dye reaction to ensure a complete aliquot is loaded on each gel.
Thus for the individual plasma and samples, 60 μg is labeled with Cy3
or Cy5, but only 50 μg will be loaded of each. A single internal standard
is therefore prepared, which comprises 30 μg of each of the 12 samples
(6 plasma and 6 synovial fluid) and labeled with Cy2 dye. Before labeling, it
is recommended that all sample concentrations are normalized to 10 μg/μL,
to make subsequent pipetting easier (Table 2).
UN
27
Book_Wang_9781597454636_Proof1_December 31, 2007
194
02
03
04
Table 1
The Sample and Dye Quantities Required for a Six-Gel, Three-Dye Pilot
Experiment Are Shown for a Three-Dye Run Analyzing Plasma (PL) and
Synovial Fluid (SF) from Six Patients (Anonymized as A, B, C, D, E,
and F)
F
01
Gibson et al.
08
09
Gel
1
2
10
11
3
12
13
4
14
15
5
16
17
18
6
Cy2 pooled standard
50 μg (4.17 μg of each
A–F and PL A–F)
50 μg (4.17 μg of each
A–F and PL A–F)
50 μg (4.17 μg of each
A–F and PL A–F)
50 μg (4.17 μg of each
A–F and PL A–F)
50 μg (4.17 μg of each
A–F and PL A–F)
50 μg (4.17 μg of each
A–F and PL A–F)
Cy5
Cy3
sample SF
50 μg of PL C
50 μg of SF D
sample SF
50 μg of SF A
50 μg of PL F
sample SF
50 μg of PL B
50 μg of SF E
sample SF
sample SF
sample SF
19
20
Beforehand: Label Reaction Tubes
21
23
24
25
30
31
32
33
34
35
36
37
38
39
40
41
50 μg of PL D
50 μg of SF B
50 μg of SF F
50 μg of PL E
4. Aliquot volumes of each sample equivalent to 100 μg into each of the first
set of SF or PL individually labeled tubes. Adjust protein concentrations of all
samples to 10 μg/μL by addition of DIGE compatible lysis buffer (DLB), as
shown in Table 1.
5. Aliquot 6 μL of each normalized sample (60 μg) from the above, into each of
the second fresh set of SF or PL individually labeled tubes and 3 μL (30 μg)
of each sample into the one PS labeled tube. This gives a total of 36 μL (or
360 μg protein) in the pooled standard tube (PS).
CO
29
Sample Preparation
Sample Labeling
UN
28
50 μg of PL A
1. Label 12 Microfuge tubes (0.5 mL) as Table 2 (SF A-F, PL A-F) for preparation
of samples prior to labeling.
2. Label a second fresh set of 12 Microfuge tubes as above, for the labeling
reaction.
3. Label one Microfuge tube as PS for the pooled standard.
26
27
50 μg of SF C
RR
EC
22
DP
07
TE
06
RO
O
05
6. The Cy dyes are diluted from 1 mM stock concentration to a working concentration of 400 pmol/μL with DMF, and 1.2 μL of the Cy3 and Cy5 dyes is
added to individual samples in a randomized fashion as described in Table 1.
An aliquot of 7.2 μL of 400 pmol/μL Cy2 is added to the pooled standard tube.
(Note: Only reconstitute minimal quantities of Cy dye working dilutions for
AQ4
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
CO
06
UN
05
4.8
3.4
5.7
8.5
5.7
3.6
2.5
2.8
4.3
3.8
4.2
3.2
[Protein] μg/μL
21.0
29.2
17.6
11.8
17.4
27.6
39.4
35.5
23.2
26.5
23.6
31.4
Sample
SFA
SF B
SF C
SF D
SF E
SF F
PL A
PL B
PL C
PL D
PL E
PL F
Volume of DLB
for 10 μg/μL (μL)
*Sample concentrations are normalized to 10 μg/μL.
Total for pooled internal standard
36
PS
RO
O
DP
TE
6
6
6
6
6
6
6
6
6
6
6
6
04
F
SF A
SF B
SF C
SF D
SF E
SF F
PL A
PL B
PL C
PL D
PL E
PL F
Labeling tube
03
3
3
3
3
3
3
3
3
3
3
3
3
Volume add to
individual labeling
tubes (μL)
02
5.2
6.6
3.3
1.5
4.3
6.4
7.5
7.2
5.7
6.2
5.8
6.8
Volume add to PS
labeling tube (μL)
01
RR
EC
Volume required
for 100 μg (μL)
Table 2
The Volumes of Samples and Diluent Required for Individual and Pooled Dye Reactions Are Shown*
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
195
Book_Wang_9781597454636_Proof1_December 31, 2007
196
03
04
05
06
07
08
09
10
11
the experiment as any remaining stock can be kept for future work at –20°C
for 3 months.)
7. All labeling reaction tubes should be mixed thoroughly by pipette and vortex
and pulse-centrifuged to collect the mixture at the bottom. Incubate the labeling
reaction tubes on ice in the dark for 30 min. (Note: Subsequent exposure of all
dye reactions to ambient light, whether in IPG strip or gel, should be minimized
to prevent degradation/bleaching of the fluorophore.)
8. Add 1.2 μL of 10 mM lysine to each of the Cy3 and Cy5 dye reactions and
7.2 μL to the Cy2 dye reaction to stop the labeling. Again, mix and centrifuge
briefly before incubating for a further 10 min on ice in the dark. The labeling
reaction is now complete, and labeled samples can be stored for up to 3 months
at –70°C in a light protected container, if not used immediately.
F
02
RO
O
01
Gibson et al.
15
16
17
18
19
20
21
22
23
4. FIRST-DIMENSION ISOELECTRIC FOCUSING (IEF)
The ampholytes and DTT that had been omitted prior to the labeling
are added at this point in the form of a 2X sample buffer. The sample
is denatured with dithiothreitol (DTT) and a volume equivalent to 50 μg
of each individual Cy3 and Cy5 labeled sample are then combined with
50 μg of the Cy2 pooled internal standard. The mixture is subsequently
rehydrated onto 24-cm immobilized pH gradient (IPG) strips for highest
resolution (sample in-gel rehydration). Samples with larger quantities of
high molecular weight proteins, alkaline proteins, or hydrophobic proteins
are likely to be poorly absorbed into the IPG strip gel matrix and would
benefit substantially from cup loading detailed elsewhere (12).
TE
14
RR
EC
13
DP
12
24
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Beforehand: Prepare Buffers
1. Prepare 2X sample buffer [8 M urea, 130 mM DTT, 4% (w/v) CHAPS, 2%
(v/v) IEF ampholytes 4–7] and rehydration buffer (8 M urea, 13 mM DTT, 4%
(w/v) CHAPS, 1% (v/v) IEF ampholytes 4–7].
2. Remove pH 4–7 IPG strips from freezer to thaw on bench and ensure
rehydration tray is level. Note strip numbers and label six fresh microcentrifuge
tubes as Table 1 (1–6).
CO
26
Pooling Samples, Strip Rehydration, and Isoelectric Focusing
1. Add equal volumes of 2X sample buffer to individual Cy3 and Cy5 labeled
samples (8.4 μL each) and to the Cy2 labeled pooled standard (50.4 μL), mix
and leave on ice for 10 min (each tube now has 50 μg labeled protein in 14 μL).
2. Aliquot 14 μL of each Cy3, Cy5, and Cy2 labeled samples to be focused on
the same IPG strip into the tubes as indicated in Table 1. Add 408 μL of
rehydration buffer to each tube, mix, and centrifuge briefly.
3. Pipette each mixture into a separate channel of the rehydration tray. Peel off
the protective cover from the IPG strip and carefully lower it gel side down
into the rehydration buffer-sample; remove any air bubbles with a pipette tip.
UN
25
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
03
04
Table 3
Isoelectric Focusing Conditions Appropriate to Proteins Soluble Within
the Acidic Range pH 4–7 for 24-cm IPG Strips
Voltage mode
Voltage(V)
Duration (h:min)
3500
1000
8000
100
—
0:10
1:00
>24 h
05
06
07
08
09
Step
Gradient
Step
Step and hold
10
11
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
DP
15
4. Overlay each strip with ∼2 mL of IPG strip cover fluid to prevent evaporation
and slide on the plastic cover. Cover with aluminum foil and incubate overnight
(or at least 10 h) at room temperature. Low voltage applied during rehydration
can improve entry of high molecular weight protein (13).
5. After adequate rehydration, use clean forceps to remove the IPG strips from
the rehydration tray. Blot off excess fluid and carefully position the IPG strips
(gel side up) within the ceramic manifold of the IPGphor isoelectric focusing
unit (GE Healthcare). Ensure that the positive (anodic) end of each strip is
oriented toward the anode of unit.
6. Apply one filter paper electrode pad, which has been moistened with de-ionized
water (remove excess fluid with blotting paper), to each end of the IPG strip
in such a way as to overlap the gel by approximately 3 to 5 mm. Place the
respective electrodes over the filter paper pads and clip firmly in place and
overlay each strip with cover fluid to prevent dehydration (108 mL for whole
manifold).
7. Close the IPGphor IEF unit safety lid and the instrument running conditions can
be programmed. We suggest conditions appropriate to proteins soluble within
the acidic range pH 4–7, for 24-cm IPG strips (Table 3). Current should be
limited to 50 μA per IPG strip. Other run conditions for the various strip sizes
and pH ranges are available [Ettan IPGphor manifold user manual (80-6499-52
edition AO); GE Healthcare].
8. Once the strips have been focused, they can be equilibrated straight away for
the second dimension or stored for several months at –80°C between plastic
sheets to prevent damage to the brittle frozen strips.
TE
14
75
—
—
—
RR
EC
13
Volt-hours (kVh)
CO
12
5. SECOND-DIMENSION SDS-PAGE (2-DE)
UN
AQ5
1.
2.
3.
4.
F
02
RO
O
01
197
In order to improve protein migration from focused IPG strip to the
second-dimension separation gel, it is important to equilibrate the strips in a
buffer containing sodium dodecyl sulfate (SDS), urea, and glycerol. In the
first equilibration step, DTT is added to completely reduce any remaining
disulfide bonds, whereas iodoacetamide is added in the second step to
Book_Wang_9781597454636_Proof1_December 31, 2007
198
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
F
RO
O
06
AQ6
AQ7
DP
05
Beforehand: Prepare Equilibration Buffer and Gels
TE
04
1. Prepare the stock SDS Equilibration buffer [50 mM Tris pH 8.8, 6 M urea,
30% (v/v) glycerol, 2% SDS (w/v)] (10 mL per strip will be required, excess
can be aliquoted and stored at –20°C for future use).
2. Prepare the working-strength Gel Running buffer [25 mM Tris base, 192 mM
glycine, 0.1% SDS (w/v)] (up to 25 L is required to fill the Dodeca gel tank).
3. Prepare 600 mL of gel casting solution (sufficient for six 1.0-mm, 12% gels;
100 mL per gel) [240 mL 30% (w/v) acrylamide/methylene bisacrylamide
solution (37.5:1 ratio), 156 mL 4X Laemmli Resolving Gel buffer (0.375 TrisHCl pH 8.8, 0.1% SDS final concentration, 197.4 mL de-ionized water, 240 μL
TEMED, 2.4 mL of 10% (w/v) ammonium persulfate (APS)]. Pass the solution
through an 0.2-μm filter and degas under vacuum, prior to the addition of APS.
4. Apply waterproof adhesive tape to the sides of assembled glass plates and
spacers, place into the gel caster with plastic sheet spacers, and gradually pour
in the prepared mixture. Carefully layer ∼2 mL of water-saturated butanol on
top of each gel to remove bubbles and create a level surface. Flush the butanol
off the gels with double de-ionized water after approximately 1 h and cover
the casting unit with tin foil. Ideally, gels should be cast the day before use to
ensure complete polymerization.
5. Prepare a 1.0% agarose solution in Gel Running buffer with 50 mg bromophenol
blue incorporated per 100 mL. Dissolve the agarose by a short incubation in a
microwave on low-medium power.
RR
EC
03
alkylate the resultant sulfydryl groups thereby preventing reoxidation,
which may complicate downstream mass spectrometry identification. In
addition, the iodoacetamide “mops up” any free DTT, which may otherwise
cause point-streaking artifacts, apparent in silver-stained gels. The Protean
Plus Dodeca multiple gel unit (Biorad) is used in the worked example,
as it can accommodate up to 12 large-format slab gels, has an in-built
buffer recirculation and cooling system and plate electrodes for consistent
resolution. A recipe for a 12% homogenous polyacrylamide Tris-glycine
gel is also given using stabilized, high-purity Protogel reagents (National
Diagnostics), though gradient gels and other buffer systems may give
better resolution for select protein molecular weight ranges (14,15). It
should be noted that the nature of the Cy dyes used in the DIGE technique
necessitate the use of specialized low-fluorescence glass plates. The glass
plates require thorough cleaning with dilute alcohol and de-ionized water
using lint-free tissues to remove any dried salts or gel fragments prior to
casting.
CO
02
UN
01
Gibson et al.
IPG Strip Equilibration
1. Prepare Equilibration buffer A by dissolving 100 mg of DTT per 10 mL of
stock SDS Equilibration buffer (5 mL needed per IPG strip). Dispense a 5-mL
AQ8
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
03
04
05
06
07
08
09
aliquot of buffer A into equilibration tubes and place the IPG strips carefully
into each. (Disposable plastic 10-mL pipettes with the conical tip broken off
are sufficiently long; reseal with parafilm between incubations.) Incubate the
strips at room temperature with gentle rocking for 15 min, then decant the
buffer.
2. Prepare buffer B by dissolving 250 mg of iodoacetamide per 10 mL of stock
SDS Equilibration buffer (5 mL needed per IPG strip). Dispense a 5-mL aliquot
of buffer B into each equilibration tube and reseal. Equilibrate the strips for a
further 15 min at room temperature with gentle rocking, decant the buffer, and
proceed to electrophoresis section.
F
02
RO
O
01
199
10
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
DP
15
TE
14
1. Rinse the equilibrated IPG strips in Gel Running buffer and using forceps place
each strip across the top of a gel, such that the plastic backing of the strip
makes contact with the back glass plate and the anodic end of the strip is at the
top left of the gel. A thin spatula can then be used to maneuver the strip down
into the well, with care so no bubbles are introduced between strip and gel.
A good tip is to push one end of the IPG strip down, between the glass plates,
so it makes contact with the top of the gel and gradually push the opposite end
down so the strip sits level on the surface.
2. Layer approximately 2 mL of premelted 1% agarose sealing solution carefully
across the strip, again try to eliminate any air bubbles trapped between the strip
and gel. Allow to cool and solidify for 5 min and repeat the procedure for the
outstanding IPG strips.
3. Insert the prepared gel cassettes into position in the electrophoresis unit (this
can be made easier by prewetting each cassette by dipping it briefly into the
Gel Running buffer). Ensure the buffer chamber is filled to the manufacturer’s
recommended level and place the safety lid on paying attention to the orientation
of the electrodes.
4. Start the electrophoresis with 2 W per gel for 45 min followed by 17 W per
gel for 4 h, both at 20°C. Alternately, for convenience, the gels can be run
overnight at 0.75 W per gel. (Such an overnight run can take anything from 18
to 21 h.) The run should be terminated when the bromophenol blue dye front
reaches the bottom of the gel.
5. In preparation for scanning the gels, the scanner instrument should be switched
on and left to warm up for 30 min and the glass plate should be thoroughly
cleaned with lint-free tissues. As DIGE gels are scanned while they are in the
glass cassettes, these too should be cleaned carefully to remove any residual
running buffer, gel smears, and so forth, from the surface of the plates. If
not scanned immediately, gels can be individually wrapped in cling film and
stored at 4°C but should be imaged as soon as possible to avoid signal loss and
dissipation of focused spots. Gels stored at 4°C should also be allowed to come
up to room temperature before scanning to avoid any condensation forming on
the surface of the glass.
RR
EC
13
CO
12
SDS-PAGE
UN
11
Book_Wang_9781597454636_Proof1_December 31, 2007
200
04
05
06
07
08
09
10
11
12
The capture of gel image data is a critical stage in the whole analysis
workflow that can often present a source of easily prevented experimental
noise. It is possible to make careless choices at this stage that can result in
a 200-fold drop in sensitivity. A variety of DIGE-compatible gel scanning
instruments based on charge-coupled device (CCD) or photomultiplier
tube (PMT) detection technologies are available (GE Healthcare, Typhoon;
Syngene, ChemiGenius; Fuji, FLA 1500; Biorad, FX Molecular Imager).
Depending on the make of scanner, there are several parameters to optimize;
the spatial resolution, the dynamic range, the scan content, and background
offsets. These features are therefore briefly discussed and recommendations
made in each case to improve the process of image acquisition.
F
03
6. IMAGE ACQUISITION
RO
O
02
DP
01
Gibson et al.
13
14
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Image (or spatial) resolution relates to the number of pixels (picture
elements) displayed per unit length of a digital image and is often measured
in dots per inch (dpi) or in micrometers (the size of the area each pixel
represents). Images with a higher spatial resolution are composed of a greater
number of pixels and have more image detail that those of lower spatial
resolution. It is important to be aware that variations in spatial resolution will
not only affect the final appearance of the image but will also impinge on
the quality of spot detection and the accuracy of any subsequent quantitative
measurements.
At low resolutions, there are fewer pixels available to represent each spot,
and as a result, spot detection and quantitative accuracy will be compromised. Image (spatial) resolution is illustrated in more detail in Fig. 2.
Higher resolution means that more pixels, and hence more data, are available
for the analysis: the spot highlighted in red in Fig. 3 is represented by 63
pixels at a relatively low 100 dpi resolution (Fig. 2A), compared with 485
pixels at a higher 300 dpi resolution (Fig. 2B).
There is, however, a maximum resolution, which once exceeded produces
minimal additional information. Once resolution is sufficient to adequately
represent the smallest features, any further increases in spatial resolution
simply increase the ability to represent the system “noise.” In addition,
every doubling in spatial resolution quadruples the amount of data that has
to be processed, which can cause problems in processing speed and file and
memory management. For example, a typical 20 × 20 cm DIGE single dye
image captured at 100 dpi versus 300 dpi would result in diverse file sizes
of 1.2 Mb and 10.6 Mb, respectively (both at 16-bit depth).
TE
18
RR
EC
17
6.1. Spatial Resolution
CO
16
UN
15
AQ9
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
201
01
02
F
03
04
RO
O
05
06
07
08
09
10
11
DP
12
13
14
15
16
18
19
Fig. 2. Portions of two-dimensional gel image scans showing typical image
quality, spot outline, and three-dimensional spot intensity (top to bottom) at 100
dpi (8-bit) and 300 dpi (16-bit) resolutions.
TE
17
20
21
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
RR
EC
23
To summarize, in most situations, 300 dpi or 100 μm will provide an
image that is large enough for accurate analysis and small enough for
efficient processing. However, if your gels are small (e.g., minigels), then
you may need to increase the resolution to achieve this. As a rule of thumb,
the active area of the gel (i.e., the area of spot material) should fall in
the range 1000 to 1800 pixels in both horizontal and vertical directions.
This range provides a good trade-off in information content and analysis
performance.
6.2. Dynamic Range (Bit Depth)
CO
22
Also referred to as color depth, bit depth or pixel depth is the number of
bits used to represent the color (grayscale, intensity levels) of each pixel in an
image. Greater bit depth allows a greater range of colors or shades of gray to
be represented by a pixel. If possible, scan at 16-bit rather than 8-bit. The bit
depth of a 16-bit image (65,536 levels of grayscale) compared with an 8-bit
image (256 levels of grayscale) results in enhanced sensitivity and accuracy
of quantification for less-abundant proteins. The possible grayscale levels
available along with the resultant dynamic range (orders of magnitude) for
the types of images commonly used in two-dimensional gel image analysis
are indicated in Table 4. In reality, the images displayed on the computer
UN
AQ10
Book_Wang_9781597454636_Proof1_December 31, 2007
202
Gibson et al.
01
02
F
03
04
RO
O
05
06
07
08
09
10
11
DP
12
13
14
15
16
TE
17
18
19
20
RR
EC
21
22
23
24
25
26
27
Fig. 3. Schematic representation of the processes involved in DIGE image
analysis. (see Color Plate 4, following p. x)
28
29
32
33
34
35
36
37
38
39
40
41
screen will only be represented in 256 shades of gray, and so an 8-bit
image will look identical to a 16-bit image by eye. However, image analysis
software can distinguish between the different levels of gray. As a rule,
the more levels of gray represented in an image, the better the ability to
differentiate low-abundance spots from background, and the greater the
quantitative accuracy. This is further illustrated in Fig. 2, comparing spot
detection in an identical area on the same two-dimensional gel, captured at
8-bit and 16-bit, respectively.
The dynamic range can be adjusted in CCD camera systems by altering
the exposure time and in laser-based systems by fine tuning the voltage of
the PMT detector. The dynamic range should be optimized to maximize
the use of available grayscale values. Aim for the maximum gray levels
CO
31
UN
30
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
03
04
Table 4
The Resultant Grayscale Levels and Dynamic Range (Orders of
Magnitude) Are Shown for a Range of Image Bit Depths
Commonly Used to Analyze Two-dimensional Gels
05
06
Bit depth
Intensity (grayscale)
levels
Orders of
magnitude
Percent of
16-bit scan
256 (28 )
1024 (210 )
4096 (212 )
65,536 (216 )
2.4
3.01
3.6
4.8
0.39
1.56
6.25
100
07
08
09
10
8
10
12
16
11
DP
12
13
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
TE
17
6.3. Scan Content
RR
EC
16
Wherever possible, it is best to try to keep the area scanned as valid gel
area. It is common to see scans where there are lots of scanner bed, labels,
and so forth, in the captured image. Some scanners automatically adjust the
scan settings based on what they “see” so a significant proportion of the
dynamic range is lost by representing scanner bed or labels. It may be best
to switch off auto gain control features or, at least, outline the areas you
are interested in so the scanner optimizes only the region of interest. Extra
“non gel” areas provide no useful information and should be cropped prior
to image analysis. These scanning artifacts can skew the image statistics,
“steal” dynamic range, increase storage requirements, and cause extra work
in manual stages of analysis. Another good tip for consistency (and to save
time later) is to always scan gel images using the same orientation and with
the same settings.
Postscan processing of two-dimensional gel images using Adobe
Photoshop or other general image processing software should be avoided, as
CO
15
in the image to be 5% to 10% less than those available. Scanner response
curves can be nonlinear, and inconsistent settings can cause issues. For
DIGE experiments without a Cy2 internal standard, chose settings that
optimize the dynamic range for each stain and keep these consistent. When
optimizing the dynamic range, it is important to avoid saturation effects.
Saturation occurs when gray levels exceed the maximum available. When a
spot becomes saturated, any differences in high pixel intensities cannot be
resolved, and the spot appears truncated when viewed in three dimensions.
No reliable quantitative data can be generated from a saturated spot, and
saturated spots may also have an overall effect on normalization if included
in later analysis.
UN
14
F
02
RO
O
01
203
Book_Wang_9781597454636_Proof1_December 31, 2007
204
04
05
06
07
08
09
10
11
12
13
14
15
F
03
these do not maintain the integrity of your original data, and any calibration
information contained in the image file will be lost. The manipulations
may make the images “look better” to the human eye but are simply transforming the original data. If the images look bad in the first place, then you
should try to optimize the scanning not manipulate the images digitally. If
possible, use GEL or IMG/INF files formats rather than generating TIFF
files. The former often contain additional grayscale calibration information,
which will not be included in the TIFF version. In any case, do not use
JPEG file storage format as it is a standard for “lossy” image compression,
which is optimized to allow the loss of information that is least noticeable
to the human eye. This does not mean it does not affect measures made by
computers. Lossy compression throws information away and manipulates
the image data. Converting a JPEG image back to a TIFF is not a solution;
once the image has been compressed in this way, the data has been lost and
cannot be retrieved.
RO
O
02
DP
01
Gibson et al.
16
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
TE
20
Image analysis of conventional silver or Coomassie stained twodimensional gels of individual samples can be highly subjective and very
time consuming, due to inherent unpredictable distortions between gels.
These inconsistencies prevent the perfect alignment and matching of spots
between gels. On the other hand, DIGE-derived images from the same gel are
precisely superimposable, and gel to gel spot matching is uniquely assisted
by the internal standard (Cy2 labeled). So by virtue of the prelabeling and
comigration of differentially labeled samples, variation in spot “coordinates” and intensity are accounted for, and gel images analysis is much
more efficient (16). It is important to understand the principles that traditional software applications apply, highlight their limitations, and provide a
recommended analysis workflow (depicted in Fig. 3 and see Color Plate 4,
following p. x).
RR
EC
19
7. IMAGE ANALYSIS: SOFTWARE PRINCIPLES
AND WORKFLOWS
CO
18
7.1. Image Warping and Spot Matching
The biggest problem in the image analysis of a gel-based experiment is
data alignment. This is easy to see by comparing the state of the art statistical
analysis for array-based experiments and the sorts of analysis performed
on gel-based systems. The key difference between the two workflows is
that the exact locations of data points are known in the array scenario. In
traditional two-dimensional gel analysis, the alignment issue is tackled by a
between gel spot matching stage. Most traditional analysis strategies follow
UN
17
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
F
RO
O
06
DP
05
TE
04
RR
EC
03
these key steps; detect spots on each gel individually, attempt to match the
spots across the experiment (possibly with the assistance of whole image gel
warping), measure spots, apply some form of normalization, apply statistical
analysis (17). The recommended workflow (Fig. 3) is distinguished by
initially warping gels, then matching and delineating spots, and imposing
the same spot outlines across all gels prior to quantification.
The main problem with the traditional workflow is that it is not able
to provide data of sufficient quality required to perform advanced statistical analysis. The core issue is focused on missing values; that is, a data
point (usually a spot) that is not available across all samples. A recent
study measured 42% missing values (i.e., not experimentally induced data
omission) for a 16-gel experiment (18). It has also been shown that the
number of missing values increases with the number of replicates (19). This
produces the predicament of reducing the extent and quality of data, as
the investigator endeavors to improve statistical power by increasing the
number of replicates. In the traditional analysis workflow, missing values
arise from two main sources: (1) the same measurements not being taken
from each gel (usually due to spot detection) and (2) the measurements not
being correctly matched between gels.
The same measurement not being taken from each gel has two main
effects. The first is down to the fact that in traditional analysis scenarios,
each gel has spots detected in isolation. This can lead to inconsistent results
because essentially the pattern is determined from a single instance and
as such is more prone to technical variance. The second issue is usually
attributed to experimental conditions where a spot has zero expression in
one or more of the groups. In this case, a spot will not be detected on an
individual gel basis, and we are left with a hole in the data. Strangely, the
proteomics community tends to differ with standard scientific practice with
zero expression spots preferring to have no measurement or “unmatched”
rather than measuring the value “zero.” This is analogous to measuring
air temperature and saying “there was no temperature” when it hits zero
instead of measuring and recording zero. This particular stance also forces
a multiple instead of a unified statistical framework approach to analysis,
which is laborious, prone to bias, and still may be suboptimal.
CO
02
UN
01
205
7.2. Spot Delineation/Outline
Geometric correction alone does not solve all of the issues as we would
still be prone to threshold of detection issues on a per gel basis and also
not matching spots that were not expressed in certain groups. The obvious
next step is to stabilize and standardize the spot detection across all of the
Book_Wang_9781597454636_Proof1_December 31, 2007
206
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
F
RO
O
06
DP
05
TE
04
RR
EC
03
images within an experiment. The geometric correction combined with an
“ensemble,” experiment-wide, spot pattern would go a large way to not only
removing the bulk of missing values but also removing a large amount of
manual intervention. We would also gain most in the low-expression spots
that are harder to detect consistently on a per gel basis. If we apply geometric
correction and derive a suitable outline for spots that are not undergoing
experimentally induced variation, then the software can handle more than
half (usually more) of the spots in an experiment fully automatically. This
is essentially reverse logic where it is more efficient to bias the analysis to
discount what is not changing rather than to optimize for what may be. This
also removes the situation where you have to spend a lot of time editing
a series of spots to find “they don’t change.” Considering the efficiency
of overall analysis, applying the same spot outlines across geometrically
corrected images is advantageous, and the benefits increase as the number
of spots subject to experimentally induced expression changes decreases.
A restriction imposed by the same spot outlines is that the area considered
to contain a spot is consistent across images. This means that there are only
going to be issues if the expression change is so large that it cannot be
represented by the same area across the gels. For spots in isolation, this is
not an issue as a larger area can be used (any extra pixels included in the
larger boundary will be zero in the smaller spots). Tight clusters of spots
that vary slightly in location and potentially patterns of spots that overlap
when considered across all groupings may however prove problematic and
require additional editing. Interestingly, posttranslational modification shifts
of proteins are better handled by the same outlines workflow than the
traditional approach, as one can apply a robust statistical framework to
discovering them. Therefore, contrary to current practice and “gut feel,”
it appears to be more favorable to apply the same spot outline pattern to
geometrically corrected gels and robustly analyze the bulk of the spots
within the experiment in a “first pass” with minimal manual intervention
and hence bias.
Once the same outlines analysis has been completed, one will have a list
of “interesting” areas. These will either be completely satisfactory with the
automatic outlines or it may be deemed that editing should be considered. It
is advisable that the initial analysis be completed fully without intervention
and this analysis saved. From this, a list of areas for further examination
should be made and these explored in a mode that adds to the information
and not replaces it. At this point, we may consider editing outlines. The
editing process is less biased when one uses the same outline across all
gels within the experiment. This is because the volume of a spot is affected
by the number of pixels chosen to include in its boundary; if imposed
CO
02
UN
01
Gibson et al.
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
04
05
06
07
08
09
10
11
12
13
14
15
16
F
03
inconsistently between groups, it is easy to introduce unintentional bias.
Because there is a much reduced requirement on manual intervention, one
can spend a lot more time on the areas that matter. Feedback from multiple
labs has shown that differential editing is rarely necessary at all and, if it
is, tends not to be a major factor in the results of the experiment. Areas
where this is deemed necessary should always be treated with caution and
may be a target area for future experiments where the zoom gels are used or
alterations to the sample preparation improve the data quality in these areas.
In summary, the suggested workflow is
1.
2.
3.
4.
5.
Geometrically correct your gels.
Create a spot pattern representative of all of the spots within your experiment.
Analyze the spots completely.
Complete statistical analysis.
Create a copy of the experiment and edit/apply differential outlines on areas
where it is deemed strictly necessary.
6. Complete statistical analysis.
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
RR
EC
21
Because DIGE offers superlative sensitivity to quantify minute differences in protein expression levels, relative to conventional staining
techniques, sample load is reduced accordingly (from as low as 75 μg to
200 μg per gel). Once spots of interest have been identified, it is then
necessary to run a “preparative” gel with a higher sample loading (0.5 mg to
2.0 mg) ensuring sufficient protein in these spots to obtain reliable identifications by mass spectrometry (20). A spot “pick list” can be generated, such
that isoelectric point and molecular weight coordinate data derived from
the DIGE “analytical” gels can be transposed onto a Coomassie or silver
stained gel.
9. CONCLUSION
CO
20
8. PREPARATIVE GELS FOR SPOT PICKING
Whereas the concept of DIGE is relatively novel, it has unique attributes
that make it particularly suitable in the drug and biomarker discovery
process. Because the technique is not dependent on antibody-protein
or oligonucleotide-RNA/DNA affinity, it does not preclude the target
identity, therefore represents changes in the sample without preemptive bias.
Although extended multiplex capability and an increased dynamic range are
desirable, it remains the most sensitive validated gel-based proteomic tool
with direct relevance to innovation in clinical and pharmaceutical research.
Improvements in dedicated analytical software have greatly increased the
throughput and confidence in data derived from DIGE images. DIGE holds
UN
19
TE
17
18
RO
O
02
DP
01
207
Book_Wang_9781597454636_Proof1_December 31, 2007
208
02
03
much promise in the challenge to uncover new molecular targets, screen
putative drug efficacy, and monitor therapeutic response for a wide range
of debilitating diseases.
F
01
Gibson et al.
04
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
RO
O
10
DP
09
TE
08
1. O’Farrell PH. High resolution two-dimensional electrophoresis of proteins.
J Biol Chem 1975;250(10):4007–4021.
2. Marouga R, David S, Hawkins E. The development of the DIGE system:
2D fluorescence difference gel analysis technology. Anal Bioanal Chem
2005;382(3):669–678.
3. Unlu M, Morgan ME, Minden JS. Difference gel electrophoresis: a single
gel method for detecting changes in protein extracts. Electrophoresis
1997;18(11):2071–2077.
4. Tonge R, Shaw J, Middleton B, et al. Validation and development of fluorescence two-dimensional differential gel electrophoresis proteomics technology.
Proteomics 2001;1(3):377–396.
5. Alban A, David SO, Bjorkesten L, et al. A novel experimental design
for comparative two-dimensional gel analysis: two-dimensional difference
gel electrophoresis incorporating a pooled internal standard. Proteomics
2003;3(1):36–44.
6. Wu WW, Wang G, Baek SJ, Shen RF. Comparative study of three proteomic
quantitative methods, DIGE, cICAT, and iTRAQ, using 2D gel- or LC-MALDI
TOF/TOF. J Proteome Res 2006;5(3):651–658.
7. Lilley KS, Friedman DB. All about DIGE: quantification technology for
differential-display 2D-gel proteomics. Expert Rev Proteomics 2004;1(4):
401–409.
8. Karp NA, Lilley KS. Maximising sensitivity for detecting changes in
protein expression: experimental design using minimal CyDyes. Proteomics
2005;5(12):3105–3115.
9. GE Healthcare. Ettan DIGE System User Manual. AB(18–1173–17). GE
Healthcare, 2006:1–112.
10. Gorg A, Drewes O, Weiss W. Separation of proteins using two-dimensional
gel electrophoresis. In: Simpson RJ, ed. Purifying Proteins for Proteomics—A
Laboratory Manual. New York: CSHL Press, 2003:391–430.
11. Gibson DS, Blelock S, Brockbank S, et al. Proteomic analysis of recurrent
joint inflammation in juvenile idiopathic arthritis. J Proteome Res 2006;5(8):
1988–1995.
12. Gorg A, Obermaier C, Boguth G, et al. The current state of twodimensional electrophoresis with immobilized pH gradients. Electrophoresis
2000;21(6):1037–1053.
13. Gorg A, Obermaier C, Boguth G, Weiss W. Recent developments in twodimensional gel electrophoresis with immobilized pH gradients: wide pH
gradients up to pH 12, longer separation distances and simplified procedures.
Electrophoresis 1999;20(4–5):712–717.
RR
EC
07
CO
06
REFERENCES
UN
05
AQ11
Book_Wang_9781597454636_Proof1_December 31, 2007
Difference In-Gel Electrophoresis
05
06
07
08
09
10
11
12
13
14
15
16
17
18
F
04
RO
O
03
14. Laemmli UK. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature 1970;227(5259):680–685.
15. Anderson NL, Anderson NG. Analytical techniques for cell fractions. XXII.
Two-dimensional analysis of serum and tissue proteins: multiple gradient-slab
gel electrophoresis. Anal Biochem 1978;85(2):341–354.
16. Corzett TH, Fodor IK, Choi MW, et al. Statistical analysis of the experimental
variation in the proteomic characterization of human plasma by two-dimensional
difference gel electrophoresis. J Proteome Res 2006;5(10):2611–2619.
17. Fodor IK, Nelson DO, Alegria-Hartman M, et al. Statistical challenges in the
analysis of two-dimensional difference gel electrophoresis experiments using
DeCyder. Bioinformatics 2005;21(19):3733–3740.
18. Grove H, Hollung K, Uhlen AK, Martens H, Faergestad EM. Challenges related
to analysis of protein spot volumes from two-dimensional gel electrophoresis
as revealed by replicate gels. J Proteome Res 2006;5(12):3399–3410.
19. Houtman R, Krijgsveld J, Kool M, et al. Lung proteome alterations in a mouse
model for nonallergic asthma. Proteomics 2003;3(10):2008–2018.
20. Mahnke RC, Corzett TH, McCutchen-Maloney SL, Chromy BA. An integrated
proteomic workflow for two-dimensional differential gel electrophoresis and
robotic spot picking. J Proteome Res 2006;5(9):2093–2097.
DP
02
TE
01
19
20
RR
EC
21
22
23
24
25
26
27
28
29
33
34
35
36
37
38
39
40
41
UN
32
CO
30
31
209
Book_Wang_9781597454636_Proof1_December 31, 2007
01
02
F
03
04
RO
O
05
06
07
08
09
10
11
DP
12
13
14
15
16
TE
17
18
19
20
RR
EC
21
22
23
24
25
26
27
28
29
32
33
34
35
36
37
38
39
40
41
UN
31
CO
30
Book_Wang_9781597454636_Proof1_December 31, 2007
01
Chapter-9
03
04
Query No. Page No.
Line No.
Query
F
02
192
02
Provide location of Roche.
AQ2
192
07
Provide name and location of manufacturer.
AQ3
192
31
Provide location of manufacturer.
AQ4
194
31
AQ5
197
Table 3
We have italicized the underlined words throughout
the text and table. Please check.
In Table 2, define acronym DLB.
11
AQ6
198
05
Provide location of manufacturer.
12
AQ7
198
10
Provide location of manufacturer.
13
AQ8
198
26
Define acronym TEMED.
14
AQ9
200
08
Provide locations for Syngene and Fuji.
AQ10
200
18
AQ11
208
28
Explain parts A and B in the legend add part labels
to Fig. 2 legend.
In Ref. 9, provide location of manufacturer as
publisher.
RO
O
AQ1
05
07
08
09
10
15
17
TE
16
18
19
20
RR
EC
21
22
23
24
25
26
27
28
29
33
34
35
36
37
38
39
40
41
UN
32
CO
30
31
DP
06
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