Differential Recovery of Peptides from Sample Tubes and the Reproducibility of Quantitative Proteomic Data Steven J. Bark* and Vivian Hook The Skaggs School of Pharmacy and Pharmaceutical Sciences, The University of California, San Diego, 9500 Gilman Drive, MC 0744, La Jolla, California 92093-0744 Received May 17, 2007
Abstract: Differential recovery of peptides due to nonspecific adsorption can seriously compromise reproducibility and quality of proteomic data for peptide analyses by liquid chromatography-mass spectrometry (LC-MS). This study demonstrates large variations in reproducibility and quantitation of LC-MS data for peptides derived from tryptic digests of BSA upon storage in different sample tubes. Notably, we show that highly improved consistency and lower errors in quantitation of BSA tryptic peptides in replicate measurements is achieved with low-retention tubes compared to regular eppendorf tubes. Furthermore, qualitative differences in peptides detected by LC-MS were observed in the two types of storage tubes. These results illustrate the necessity for careful evaluation of storage vessels and conditions to minimize variability in sample quality for LC-MS experiments. Keywords: peptides • LC-MS • recovery • reproducibility • nonspecific adsorption • quantitation • proteomics
Introduction Variable recovery of peptides due to nonspecific adsorption in storage vessels is a well-known phenomenon that impacts proteomic data. Although the fundamental objective of proteomics is to define the identity and quantity of all peptide species within a biological sample, differential recovery due to adsorptive processes can distort the data by reduction and changes in their apparent abundance, and may even eliminate observation of particular peptides or proteins. Despite the critical importance of this process and its effects, there are few studies that have extensively characterized the extent of variability and reproducibility introduced by storage tubes on peptide detection by liquid chromatography-mass spectrometry (LC-MS) methods in proteomic experiments.1 Because individual peptides2-5 show nonspecific adsorption to storage vessels, examination of the extent of this phenomenon to simultaneous recovery of multiple peptides in proteomic studies is necessary for evaluation of the quality of proteomic data. Quantitative proteomic comparisons among different samples is desirable to understand alterations in systems of peptides * To whom correspondence should be addressed. Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego, 9500 Gilman Drive, MC-0744, La Jolla, CA 92093, E-mail,
[email protected]. 10.1021/pr070294o CCC: $37.00
2007 American Chemical Society
and proteins in biological systems.6-15 Such proteomic studies demand preparation and separation of peptide or protein components of experimental samples prior to LC-MS analyses. There is an underlying assumption that the samples processed with identical procedures experience the same conditions and can be compared directly. However, variability in nonspecific adsorption and recovery during sample preparation compromises this assumption because (1) the outcomes of peptide detection and quantitation by LC-MS may vary although samples have been prepared under identical conditions, which leads to (2) variability in peptide measurements that may not reflect actual abundance in the biological sample. Here in this study, we report data demonstrating that the reproducibility of proteomic data is highly dependent on recovery of peptides that varies tremendously with use of different sample tubes. This step occurs during sample preparation prior to LC-MS analyses. We demonstrate significant variations in detection and quantitation of BSA tryptic peptides (test sample) resulting from sample preparation in regular compared to low-retention eppendorf tubes. These findings demonstrate the critical importance of variability in peptide recoveries from sample tubes that impacts biological conclusions derived from proteomic studies.
Experimental Procedures Two standard BSA trypsin digests, one in a regular eppendorf tube, non-siliconized (Eppendorf Safe-Loc Biopur, Fisher Scientific), and the other in a low-retention eppendorf tube (Eppendorf Protein Lo-Bind, Fisher Scientific), were performed with reduction and alkylation to provide a standard peptide mixture for LC-MS experiments. Briefly, a 1.65 µL sample of BSA (3.3 µg) (Albumin Standard, 2 mg/mL in 0.9% aqueous NaCl and Sodium Azide, Pierce) was lyophilized and redissolved in 8 µL of 20% acetonitrile. The sample was reduced (100 nmol Tris(2-carboxyethyl) phosphine hydrochloride (TCEP, Pierce), 1 µL of a 100 mM solution) at 55 °C for 10 min followed by alkylation in darkness (100 nmole iodoacetamide, 1 µL of a 100 mM solution, Sigma Aldrich) at room temperature for 20 min. Then, 36 uL of 25 mM ammonium bicarbonate, pH ) 7.0, 2 µL of 100 mM CaCl2, and 2 µL of trypsin (40 ng, Promega Sequencing Grade) were added for a final volume of 50 µL (0.066 µg/µL or 1.0 pmol/µL final protein concentration), and the proteolysis was allowed to proceed for 4 h at 37 °C. In the first experiment, samples from each digest were subjected to LC-MS analysis immediately after trypsin digestion (within 30 min). In the second experiment, a sample from the digest Journal of Proteome Research 2007, 6, 4511-4516
4511
Published on Web 09/12/2007
technical notes
Differential Recovery of Peptides
Table 1. Quantitative Differences in Peak Intensities of BSA Tryptic Peptides by Baseline Peak Chromatogram Analyses of Samples Prepared in Low-Retention Compared to Regular Tubesa Peak Intensities low-retention tubes retention time (minutes)
6.4 6.7 7.9 8.5 8.6 9.0 10.7 10.8 10.8
Figure 1. Comparison of baseline peak chromatograms (BPC) of BSA tryptic peptides prepared in low-retention and regular tubes. BSA trypsin digest samples (100 fmol/µL BSA) were divided into 3 low-retention and 3 regular eppendorf sample tubes. After incubation at room temperature for 15 min, samples were transferred to injection vials for LC-MS analysis. One 30 fmol injection was made from each sample tube. Overlays from triplicate runs for samples in low-retention tubes (A), and from triplicate runs for samples in regular tubes (B) are shown. Peptide samples prepared in the low-retention tubes showed improved reproducibility; note that the peak eluting at 12 min from siliconized tubes (A) was not appreciably detected in any samples from regular tubes.
performed in the low-retention eppendorf tube was diluted to 100 fmol/µL and aliquots were placed in 3 regular eppendorf tubes and 3 low-retention eppendorf tubes (10 µL in each tube). The aliquots were incubated at room temperature for 15 min prior to transfer to glass injection vials for LC-MS analysis. Injections were sequential for each triplicate analysis over a 4-hour time frame for all injections. In addition, the original digests from experiment 1 were frozen at -20 °C for 14 days. Peptide recovery and instrument variability was tested by sequential injections of 100 fmole/µL BSA digest over a timecourse consistent with our experiments. The variability observed was consistent with sample incubation experiments with low-retention tubes. All LC-MS analyses were performed on an Agilent XCT Ultra ion trap mass spectrometer coupled to an Agilent 1100 nanoHPLC system fitted with a HPLC-Chip injection system. The LC separation was performed on an Agilent C18 analytical HPLC chip (Agilent Zorbax C18 Chip, 43 mm × 75 µm, 40 nL trap) and utilized solvent B (acetonitrile with 0.1% formic acid) in solvent A (water with 0.1% formic acid), and the gradient progressed from 3 to 45% B in 18 min followed by an increase to 75% B in 5 min. A longer gradient used the same 3-45% B progression over 50 min followed by a ramp to 95% B over 10 min. The mass spectrometer was set for data-dependent scanning in MS/MS mode on the two most abundant ions present in the MS scan. The exclusion time was set to 0.2 min, isolation window set to 4 amu, and voltages set to -1850 V (capillary), -500 V (counterelectrode), and 1.30 V (fragmenta4512
Journal of Proteome Research • Vol. 6, No. 11, 2007
X ( SD (×106)
3.37 ( 0.12 4.26 ( 0.25 6.36 ( 0.99 14.83 ( 1.05 0.99 ( 0.18 5.05 ( 0.82 12.70 ( 2.98 2.30 ( 0.30 21.23 ( 1.83
rSD (% of X) (×106)
regular tubes X ( SD (×106)
3.56 6.06 ( 2.59 5.87 5.75 ( 0.89 15.57 7.80 ( 2.04 7.08 15.93 ( 3.41 18.18 1.28 ( 0.38 16.24 4.06 ( 2.52 23.65 7.62 ( 3.70 13.04 3.32 ( 0.59 8.64 17.31 ( 7.24 Avg. rSD, %X: 12.43
rSD (% of X) (×106)
42.74 15.48 26.15 21.41 28.91 62.07 48.56 17.77 41.87 Avg. rSD, %X: 34.11
a Tryptic peptides derived from BSA were analyzed by BPC (baseline peak chromatogram) and relative peak intensities were calculated as X ( SD. Standard deviation (SD) is reported as absolute numbers with relative SD (rSD) of the mean (X) reported separately. The average rSD for all analyses is reported at the bottom. Peak intensities are shown as relative units. Only peaks observed in all experimental replicate measurements for both lowretention and regular samples were used for analysis.
tion). Smart ion target was set to 500 000 to correct for background ions. The maximum injection time was set to 100 ms. All other default settings were used and left unaltered in all experiments. Comparison between baseline peak chromatograms (BPC) and peak-intensity spectra were performed by automated peak-finding and peak-height quantitation in Chemstation (Agilent Technologies, Santa Clara, CA) followed by manual examination and alignment of peaks. Quantitative comparisons of peak intensities were reported for peaks present in all triplicate runs and p-value statistical analysis was performed using the statistical functions in KoalaCalc (Macropod Software). Data searches were processed using SpectrumMill (Agilent Technologies) with default parameters for data extraction. Database searches used 1.4 amu for precursor mass tolerance with all other default parameters retained. The Bovine search database was extracted and indexed from the 9/2006 update of the NCBI NR database using the Spectrum Mill toolkit extraction program.
Results The analysis of peptide recovery was performed using the primary trypsin digest of BSA performed in the low-retention sample tube. An aliquot of this digest was diluted to 100 fmole/ µL (1/10) to generate 6 aliquots of 10 µL each. These aliquots were incubated for 15 min at room temperature with 3 samples in regular and 3 in low-retention eppendorf tubes. After 15 min, the samples were transferred to injection vials for LC-MS analysis. Clear differences in reproducibility of LC data were observed for peptides prepared in regular compared to lowretention tubes, as shown by Base Peak Chromatograms (BPC) of the two samples (Figure 1). The BPC data showed much better reproducibility for peak intensities in the low-retention versus the regular eppendorf tubes obtained from triplicate LC-MS runs (Table 1). The relative standard deviations (rSD) for average peak intensities of peptides in the regular tubes
technical notes
Bark and Hook
Table 2. Peptide Analyses by Mass Spectrometry: Differences in Average Peak Intensities and Standard Deviations (SD) from Samples Prepared in Low-Retention Compared to Regular Tubesa Peak Intensities low-retention tubes Peak m/z
511.9 538.1 547.6 654.3 722.8 755.8 801.5 820.4 850.3 879.2 906.8
X ( SD (×106)
0.81 ( 0.09 0.32 ( 0.04 2.38 ( 0.24 1.04 ( 0.23 0.71 ( 0.03 0.78 ( 0.32 3.29 ( 0.87 0.69 ( 0.03 2.77 ( 1.03 4.47 ( 0.23 0.97 ( 0.14
rSD (% of X) (×106)
11.11 12.50 10.08 22.12 4.23 41.03 26.44 4.35 37.18 5.15 14.43 Avg. rSD, %X: 17.15
regular tubes X ( SD (×106)
1.08 ( 0.24 0.43 ( 0.08 2.38 ( 0.24 1.26 ( 0.43 1.13 ( 0.41 0.80 ( 0.55 3.13 ( 0.72 0.90 ( 0.20 1.43 ( 0.60 3.87 ( 1.81 0.74 ( 0.50
rSD (% of X) (×106)
22.22 18.60 10.08 34.13 36.28 68.75 23.00 22.22 41.96 46.77 67.57 Avg. rSD, %X: 35.60
a Tryptic peptides derived from BSA were analyzed by LC-MS, and the relative intensities of peptide ions with indicated m/z were analyzed for X ( SD peak intensities. Standard deviation (SD) information is reported as absolute numbers with relative SD (rSD) of the mean reported separately. The average rSD for all analyses is reported at the bottom. Ion intensities are shown as relative units. Only masses observed in all experimental replicate measurements for both low-retention and regular samples were used for analysis.
exceeded the errors in the low-retention tubes by 2-10-fold (Table 1). For example, the peptide with retention of 6.4 min showed a rSD of 42.7% of the mean intensity in regular tubes, but showed a rSD of 3.56% of the mean in low-retention tubes. The number of peaks exhibiting high variability was more prevalent in the regular tubes than in the low-retention tubes. Overall, the average standard deviation of peak intensities from the regular eppendorf tubes was 2-3 fold greater than that from the low-retention tubes (with rSD of 34.1 versus 12.4% of the mean intensities, in BPC mode, Table 1). Proteomic quantitation of peptide species often focuses specifically on a particular ion species observed in the mass spectrometry analyses, rather than a gross ion intensity profile obtained from BPC (or TIC) data. Therefore, we compared 11 specific peptide ion species present in all samples in both experimental data sets for regular and low-retention tubes (Table 2). Confirming the initial BPC analysis, large differences in relative intensities of specific peptide ions were observed. Comparisons of the relative peak intensities of peptide ions with m/z of approximately 500-900 in triplicate LC-MS runs from regular and low-retention tubes shows high variability in rSD for ions obtained from the regular tubes. For example, the peak intensities of the peptide ions of 511.9 and 722.8 m/z from regular tubes showed a rSD value that was about twice and nine times greater than the rSD value for these same ions obtained from the low-retention tubes (Figure 2). The average rSD in the regular eppendorf data set was significantly greater than that in the low-retention eppendorf data set (rSD of 35.6 versus 17.2% of the mean, respectively, Table 2). The statistical significance of the deviations observed in this data was evaluated by paired t-test analysis for both the BPC and ion intensity data. BPC data calculated p ) 0.0049 whereas the ion intensity data calculated p ) 0.0024. In both cases, the
Figure 2. Mass spectrometry of peptide ions prepared from BSA digests in low-retention and regular eppendorf tubes. BPC analyses of peptides was followed by evaluation of specific peptide ions by liquid-chromatography-mass spectrometry (LCMS). Representative BPC and ion intensity analyses are illustrated for the peaks with retention times of 6.3 and 6.7 min for triplicate LC runs of peptide samples from low-retention tubes (A) and from regular tubes (B). MS analyses of the peaks from lowretention tubes and regular tubes detected peptide ions with m/z of 722.8 for the peak at 6.3 min and 512.0 for the peak at 6.7 min.
null hypothesis is inconsistent with the observed data distributions indicating the differences between the data are not random. System variability was tested directly by sequential injections of a single sample over a similar timecourse as the performed incubation experiments (∼4 h). Two identical Agilent ion-trap mass spectrometers were tested. The variability average rSD for each system was measured at 14.5 and 16.6%, which is consistent with the low-retention incubation experiments. Significantly, after storage of tryptic peptides for 14 days, highly variable and poor recovery of peptides were observed by LC in the BPC data for regular tubes (Figure 3). After 14 days of storage at -20 °C, a 30 femtomole sample from each BSA digest stored in regular or low-retention tubes were subjected to LC-MS/MS analysis using the longer 60 min gradient (see Experimental Procedures). Loss of peak intensity signals for samples stored in regular tubes exceeded 95% for most peptides; moreover, searches on MS/MS data for these samples demonstrated minimal identification to BSA. In contrast, depletion of peak intensity signals for samples stored in low-retention eppendorf tubes was much less prevalent; MS/ MS analyses confirmed identification of peptides derived from BSA digests. The low-retention tubes clearly provided greater reproducibility of proteomic data compared to the regular tubes. Journal of Proteome Research • Vol. 6, No. 11, 2007 4513
Differential Recovery of Peptides
technical notes
Figure 3. Comparison of baseline peak chromatogram (BPC) for BSA tryptic digests after storage in low-retention tubes and regular tubes. BSA tryptic digests were performed at 1 pmol/µL in low-retention tubes (A) or in regular eppendorf tubes (B) and all samples were stored for 2 weeks prior to LC-MS analyses. After storage, 30 fmol of the BSA digest from each sample tube was analyzed by LC-MS/MS. The chromatograms for the BSA digests in low-retention (C) and regular sample tubes (D) after completion of digests are displayed. The larger number of apparent peaks in the BSA digest in regular sample tube (D) is reflected in the higher intensity of several peaks compared to the BSA digest in the low-retention sample tube (C). Most peaks are found in digests from both sample tubes. 4514
Journal of Proteome Research • Vol. 6, No. 11, 2007
technical notes Discussion Data presented from this study demonstrate that different sample tubes can introduce significant variability in LC-MS analyses of a standard peptide mixture. This variability is manifested by increased standard deviations in both BPC and specific ion intensity profiles. It is likely that the difference in nonspecific adsorption of peptides to regular tubes compared to low-retention tubes accounts for the high variability observed in recovery of peptides. In our experiments, triplicate runs of peptides were prepared from the same primary BSA trypsin digest and incubated for 15 min in either regular or low-retention eppendorf sample tubes. BPC and ion intensity data show that the average relative standard deviation (rSD) of peak intensities in the regular eppendorf data set was significantly higher than that in the low-retention eppendorf data set. Some individual peaks showed increased rSD values in regular tubes that were up to 10-fold greater than that in low-retention tubes. It is also interesting to note that the variability observed in many proteomic studies is within some of the relative standard deviations observed in our experiments. It is important to note that the absolute value of intensities for peptides in either BPC or for specific ions apparently favored regular sample tubes in initial analysis. Multiple experiments indicate that this is highly variable depending on the particular regular sample tube while the low-retention tubes are more consistent. However, longer storage for 2 weeks showed strikingly deleterious effects on recovery for peptides stored in regular tubes, with nearly complete loss of detected peptides. Clearly, these results demonstrate high variation in reproducibility and losses of peptide peak intensities obtained from LCMS data for samples prepared in regular tubes, and reduced variability with improved peptide recoveries for samples prepared in low-retention tubes. Comparison of the data was not peak-to-peak as would be used in biological experiments, because our objective was to define and compare variability in rSD across all identifiable species present in all runs. Therefore, comparison was only made for peaks and ion species present in all experiments to allow reliable determinations for rSD’s. Our triplicate experiments generated comparisons of 9 rSD’s in the BPC dataset and 11 rSD’s in the specific ion species data. It is attributable to these larger datasets that our calculated p-values are well below the statistical cutoff of p ) 0.05 (BPC data calculated p ) 0.0049 whereas the ion intensity data calculated p ) 0.0024). An underlying assumption in quantitative proteomic comparisons is that the observed variation reflects the biological state of the sample. However, this assumption may not be valid if there is significant variability from sample handling or storage. Labeling reactions or sample storage necessarily requires separate tubes for individual samples. The variable recoveries and high error (rSD) for quantitation of peptides from different sample tubes occurred after only a 15-minute period subsequent to dilution of the BSA-derived peptides. Therefore, it is likely that sample-handling and storage processes can magnify variability present in recovery and adsorption processes because of the longer time scale in contact with sample tubes. In fact, this study observed nearly complete losses of peptide signals upon long-term (14 days) storage in regular tubes. Advantageously, the use of low-retention tubes reduced variability in these proteomic experiments.
Bark and Hook
A recent study has linked adsorption losses and recoveries of peptides in sample vials and to injector apparatus to variability in proteomic data.1 It should be noted that our experiments did not indicate high variability from injection vials or the injector itself during the time-course consistent with our experiments. It must be realized that sample concentration is a major factor in peptide adsorptive losses and recovery. Concentration of samples has a direct bearing on observed losses from adsorption. The concentrations used by van Midwoud et al.,1 range from 1.67 to 16.7 fmole/uL (1.67-16.7 nM). The concentrations employed in our experiments were significantly higher at 100 fmol/µL and 1 pmol/µL (100 nM and 1 µM). Our use of higher concentrations could mask the effects of adsorption to other components of the analytical system. This is consistent with the report of an adsorption plateau for Cetrorelix at about 300 nM.5 The variability we observed in our tests indicates the contribution from sample tubes was of primary significance at the higher concentrations. It should be noted that adsorption losses are additive and likely occur both in the sample tubes and the vials and injection system. To reduce these effects, it is common practice in proteomics to work at the highest concentrations that can be practically achieved. While this can mitigate possible losses and variability, many important peptide and protein components of authentic biological samples will likely be present at low concentrations that may require enrichment for quality LC-MS analyses. It must be noted that data presented in this study includes only peptides found in all experimental runs. The variability of the data from the regular eppendorf tubes is even more significant when all peptide data including those data not present in single or multiple LC-MS analyses are considered (Supporting Information, Tables 3A/B and 4A/B). Use of replicate measurements represents an important means to reduce the inherent errors for quantitative comparisons between data sets. It should be emphasized that such an approach, though correct and essential, only quantitates variation within the observed data. It does not evaluate the components producing this variability. To adequately address the problems associated with variability (such as the adsorption processes described here), specific detailed experiments are required to evaluate the sources and magnitude of error and mitigate their effects. Although these experiments were designed to answer the simple question of reproducibility of proteomic data for peptide samples prepared in different tubes, there are profound implications from these results. At the very least, these studies should provide a note of caution about the effects of reagents and vessels used in proteomic studies. Quantitation in proteomics, especially where relative quantitative comparison between data sets is anticipated, requires rigorous analysis of variances in data sets consisting of replicate experiments. Assignment of statistical significance cannot be reliably undertaken without defining this variation because the differences observed must be greater than the magnitude of the variances observed in each data set.16 Therefore, the improved reproducibility obtained by using low-retention sample tubes from the earliest steps in sample preparation, with appropriate methods to minimize adsorption and maximize recovery, will enhance observation of important differences between biological samples in quantitative proteomic studies.
Acknowledgment. Grant support for this research from the National Institutes of Health to V.H. is appreciated. Journal of Proteome Research • Vol. 6, No. 11, 2007 4515
technical notes
Differential Recovery of Peptides
Supporting Information Available: Tables 3A, 3B, 4A, and 4B. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) van Midwoud, P. M.; Rieux, L.; Bischoff, R.; Verpoorte, E.; Niederlander, H. A. G. Improvement of Recovery and Repeatability in Liquid Chromatography-Mass Spectrometry Analysis of Peptides. J. Proteome Res. 2007, 6, 781-791. (2) Hyenstrand, P.; Metcalf, J. S.; Beattie, K. A.; Codd, G. A. Effects of adsorption to plastics and solvent conditions in the analysis of the cyanobacterial toxin microcystin-LR by high performance liquid chromatography. Toxicon 2001, 39, 589. (3) Law, S. L.; Shih, C. L. Adsorption of calcitonin to glass. Drug Dev. Ind. Pharm. 1999, 25, 253. (4) Bauer, H. H.; Muller, M.; Goette, J.; Merkle, H. P.; Fringeli, U. P. Interfacial adsorption and aggregation associated changes in secondary structure of human calcitonin monitored by ATR-FTIR spectroscopy. Biochem. 1994, 33, 12276. (5) Grohganz, H.; Rischer, M.; Brandl, M. Adsorption of decapeptide Cetrorelix depends both on the composition of the dissolution medium and the type of solid surface. Eur. J. Pharm. Sci. 2004, 21, 191. (6) Fu, Q.; Garnham, C. P.; Elliott, S. T.; Bovenkamp, D. E.; Van Eyk, J. E. A robust, streamlined and reproducible method for proteomic analysis of serum by delimpidation, albumin and IgG depletion and two-dimensional gel electrophoresis. Proteomics 2005, 5, 2656-2664. (7) Baseski, H. M.; Watson, C. J.; Cellar, N. A.; Shackman, J. G.; Kennedy, R. T. Capillary liquid chromatography with MS3 for the determination of enkephalins in microdialysis samples from the striatum of anesthetized and freely-moving rats. J. Mass Spectrom. 2005, 40, 146-153.
4516
Journal of Proteome Research • Vol. 6, No. 11, 2007
(8) Yuan, X.; Desiderio, D. M. Human cerebrospinal fluid peptidomics. J. Mass Spectrom. 2005, 40, 176-181. (9) Sinnaeve, B. A.; Storme, M. L.; Van Bocxlaer, J. F.; Capillary liquid chromatography and tandem mass spectrometry for the quantification of enkephalins in cerebrospinal fluid. J. Sep. Sci. 2005, 28, 1779-1784. (10) Wu, C. C.; MacCoss, M. J.; Howell, K.; Matthews, D. E.; Yates, J. R., III Metabolic Labeling of Mammalian Organisms with Stable Isotopes for Quantitative Proteomic Analysis. Anal. Chem. 2004, 76, 4951-4959. (11) Liu, H.; Sadygov, R. G.; Yates, J. R., III A Model for Random Sampling and Estimation of Relative Protein Abundance in Shotgun Proteomics. Anal. Chem. 2004, 76, 4193-4201. (12) Wang, H.; Qian, W.-J.; Chin, M. H.; Petyuk, V. A.; Barry, R. C.; Liu, T.; Gritsenko, M. A.; Mottaz, H. M.; Moore, R. J.; Camp, D. G., II; Khan, A.; H.; Smith, D. J.; Smith, R. D. Characterization of the Mouse Brain Proteome Using Global Proteomic Analysis Complemented with Cysteinyl-Peptide Enrichment. J. Proteome Res. 2006, 5, 361-369. (13) Che, F-Y.; Fricker, L. D. Quantiative peptidomics of mouse pituitary: comparison of different stable isotopic tags. J. Mass Spectrom. 2005, 40, 238-249. (14) Gerber, S. A.; Rush, J.; Stemman, O.; Kirschner, M. W.; Gygi, S. P. Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS. Proc. Natl. Acad. Sci. 2003, 100 69406945. (15) Oda, Y.; Huang, K.; Cross, F. R.; Cowburn, D.; Chait, B. T. Accurate quantitation of protein expression and site-specific phosphorylation. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6591-6596. (16) Zybailov, B.; Mosley, A. L.; Sardiu, M. E.; Coleman, M. K.; Florens, L.; Washburn, M. P. Statistical Analysis of Membrane Proteome Expression Changes in Saccharomyces cerevisiae. J. Proteome Res. 2006, 5, 2339-2347.
PR070294O
