A REPRODUCIBLE METHOD FOR ONLINE RP/RP 2D nanoLC/MS FOR THE ANALYSIS OF PROTEOMIC SAMPLES
James Murphy; Martha Stapels; Keith Fadgen; Scott Geromanos
Waters Corporation, Milford, MA
INTRODUCTION
(A)
Intra- and Inter-System Reproducibility:
45%
(A)
2D
10 fraction
(778)
An improved method for performing twodimensional chromatography with mass
spectrometry for the analysis of proteomic
samples was developed. Peptides were
separated by RP chromatography at high pH in
the first dimension and low pH in the second
dimension1 prior to data-independent analysis
with a quadrupole time-of-flight mass
spectrometer. Reproducible protein
identification and absolute quantitation results
were obtained from E. coli samples
1D
(365)
21%
2D
5 fraction
(695)
17%
standard deviation retention time was 28 seconds for the
majority of peptides are eluting in only one fraction and at
very similar retention times.
11%
To asses inter-system reproducibility, replicate 2D-5 fractions
studies were performed at Waters Corporation and the Duke
Proteomics Core Facility (http://www.genome.duke.edu/
proteomics/). Site to site protein replication was 86% (Figure
Eluent A: 0.1% formic acid in water
Eluent B: 0.1% formic acid in acetonitrile
Sample Preparation and Loading: Waters MassPREP™ E.
Coli digestion standard was utilized for single dimension (1D)
and multi dimension (2D) comparisons. Sample loading for
1D, 2D-5 fraction, and 2D-10 fraction was 750 ng, 2.4 µg, and
4.8 µg respectively. In each case, three replicate injections
were performed.
Peptides
Identified peptides to a selected protein show that distribution
7961
three experiments (Figure 6). The intensity and number of
peptides increased for the 2D experiments due to the fact that
3874
1D vs. 2D Separations:
more material was loaded compared to the 1D analysis.
Figure 2 depicts a 2D-5 fraction separation of the E. coli
sample. Organic concentrations for each step were selected to
ensure nearly equivalent MS intensity for each second
778
695
365
1600000
dimension run. Replicate injections of a 1D, 2D-5 fraction, and
1D
1D
a 2D-10 fraction study were performed. Protein and peptide
3 injections. 3874 peptides from 265 proteins, 7961 peptides
from 695 proteins, and 9415 peptides from 778 proteins were
2D (5)
1400000
2D (10)
Figure 3. Protein and peptide counts for the 1D, 2D-5 fraction, and 2D-10
fraction experiments. The numbers depicted are for proteins and
peptides that replicated in 2 of 3 experiments.
organic-containing fractions, an aqueous flow was delivered
with the 2nd dimension pump, and mixed with the eluted
fraction prior to trapping.
respectively (Figure 3). Improved sequence coverage was
found for the proteins common to all three analyses (Figure 4).
This trend continues for proteins that were observed in the 2D5 and 10 fraction analyses.
Comparing identifications from one analysis to another results
E
MS Data processing : MS data from the individual chromatograms were processed separately and then merged into one
file prior to database searching with ProteinLynx Global Server
E
(PLGS 2.3) with Identity informatics. An E. coli database with
an equal number of random sequences concatenated onto it
was used for the searches to limit the false positive rate to 4%
in a high degree of intersection (Fig. 5A). For example, 97%
of protein identifications obtained in the 1D analysis were
observed in the 2D studies. Absolute quantitation2 of the
identified proteins shows that the 2D strategy samples the less
40
Average Sequence Coverage
on the second dimension trap column (Fig. 1B) from the
1
0.1
0.01
35
30
Replicate 2
70%
Replicate 3
60%
40%
20%
•
0%
1
0
•
•
15%
•
10%
•
5%
•
0%
5
10
15
20
25
30
35
40
45
50
55
60
Stdev Retention Time (sec)
25
Proteins found in
2D (10) analysis
Figure 7. (A) Observed peptide splitting for the replicate 2D-5 fraction
analyses. (B) Retention time standard deviation for peptides in the 2D—5
fraction analyses.
20
1D
2D (5)
2D (10)
Figure 4. Average sequence coverage for the proteins observed in the
replicate analyses. Sequence coverage for proteins observed in all three
analyses is in red, for proteins observed in the 2D analyses is in green,
and proteins only observed in the 2D—10 fraction analysis is in blue.
TO DOWNLOAD A COPY OF THIS POSTER, VISIT WWW.WATERS.COM/POSTERS
5
20%
0
abundant proteins when compared to 1D alone (Fig. 5B).
2
3
4
Number of Fractions
1000000
600000
1
10
CONCLUSION
30%
25%
800000
0.1
Waters Corporation (ng/µg)
Figure 8. (A) Protein intersection for identical 2D-5 fraction studies
performed at Waters Corporation and the Duke Proteomics Core Facility
(http://www.genome.duke.edu/proteomics/). (B) Absolute quantitation
for the proteins that replicated between each site.
50%
1200000
200000
Proteins found
in 2D analyses
80%
10%
2D (5)
400000
Proteins found
in all three
analyses
Replicate 1
90%
2D (10)
identified for the 1D, 2D-5 fraction, and 2D-10 fraction studies,
Online Dilution with RP/RP: To maximize sample recovery
R2 0.973
0.01
100%
of intensity of peptides was relatively consistent across the
RESULTS
identifications were considered valid if they were found in 2 of
10
9415
Observed Peptides
Gradient: 5-40% B for 90 min at 300 nL/min
Proteins
for all replicating proteins.
Observed Peptides
Column: 75 µm x 15 cm BEH C18 (1.7 µm)
Figure 1. Fluidic layout for the 2D nanoACQUITY for high/low pH RP/RP
analysis with online dilution. Two gradient pumps (nanoBSM) are utilized
for sample injection and fractionation (A) as well as analytical gradient
delivery (B).
Figure 5. (A) Protein intersection for 1D, 2D-5 fraction, and 2D-10 fraction
studies; (B) Absolute quantitation for proteins observed.
Peptide Intensity
Second Dimension (Fig. 1B):
(Figure 8B). The site to site reproducibility is very good. The
standard deviation in absolute quantitation was less than 20%
C18 (5 µm)
Online dilution flow rate: 10 µL/min aqueous
at the protein level.
8A). Absolute quantitation of proteins was also evaluated
(B)
Duke Proteomics Core Facility
replicate studies (Figure 7B). For replicate injections, the
14%
(B)
First Dimension (Fig. 1A):
Eluent B: acetonitrile
86% in
common
was compared for the 2D-5 fraction injections. The median
Figure 2. 2D-5 fraction E. coli digest separation. Percentage organic used
for each fractional step is shown. Total sample loading was 2.4 µg.
Eluent A: 20 mM ammonium formate pH 10.0
Waters Corporation
695 in
2 of 3 replicates
RPLC in the first dimension. Elution in the second dimension
(B)
LC/MS System: 2D nanoACQUITY UPLC®/ SYNAPT™ HDMS™
Gradient formation: discontinuous step gradient at 1 µL/min
higher value of ~19%. In either case, greater than 95% of all
Duke Proteomics Facility
596 in
2 of 3 replicates
peptides were observed in only 1 or 2 fractions. This is a good
Retention Time (min)
Column: 150 µm x 10 cm XBridge
~13% (Figure 7A). The 2D-10 fraction replicate had a slightly
(A)
indication of the quality of separation obtained using high pH
METHODS
™
Peptide splitting for the replicate 2D-5 fraction analysis was
RP/RP provides a high resolution separation in both
dimensions
Replicate studies reveal peptides are identified in the
same fractions at the same retention times
Over 90% of proteins identified with 1D were
replicated in 2D separations
Fractionation provides extended coverage of the
proteome
Fidelity of absolute concentration is maintained when
using 2D strategies
Intra- and inter-system reproducibility of the 2D RP/
RP methodology is excellent
References
1. M. Gilar, et.al., J Sep Sci 2005, 28, 1694-1703
2.
J. Silva, et.al., Mol. Cell Proteomics, 2006, 144-156
Figure 6. Peptide intensity for Aconitate hydratase 2 for 1D, 2D-5 fraction,
and 2D-10 fraction analyses.
720002679EN
©2008 Waters Corporation