Intact Protein Quantitation Using Pseudoisobaric Dimethyl Labeling

Subscriber access provided by University of Sussex Library

Article

Intact Protein Quantitation Using Pseudo-isobaric Dimethyl Labeling Houqin Fang, Kaijie Xiao, Yunhui Li, Fan Yu, Yan Liu, Bingbing Xue, and Zhixin Tian Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b01388 • Publication Date (Web): 30 Jun 2016 Downloaded from http://pubs.acs.org on July 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Intact Protein Quantitation Using Pseudo‐isobaric Dimethyl Label‐ ing Houqin Fang‡, Kaijie Xiao‡, Yunhui Li, Fan Yu, Yan Liu, Bingbing Xue, and Zhixin Tian* School of Chemical Science & Engineering and Shanghai Key Laboratory of Chemical Assess‐ ment and Sustainability, Tongji University, Shanghai 200092, China Supporting Information

ABSTRACT: Protein structural and functional studies rely on complete qualitative and quantitative information of pro‐ tein species (proteoforms); thus, it is important to quantify differentially expressed proteins at their molecular level. Here we report our development of universal pseudo‐isobaric dimethyl labeling (pIDL) of amino groups at both the N‐terminal and lysine residues for relative quantitation of intact proteins. Initial proof‐of‐principle study was conducted on standard protein myoglobin and hepatocellular proteomes (HepG2 vs. LO2). The amino groups from both the N‐terminal and ly‐ sine were dimethylated with HXHO (X=13C or C) and NaBY3CN (Y=H or D). At the standard protein level, labeling effi‐ ciency, effect of product ion size and mass resolution on quantitation accuracy were explored; and a good linear quantita‐ tion dynamic range up to 50‐fold was obtained. For the hepatocellular proteome samples, 33 proteins were quantified with RSD≤10% from one‐dimensional RPLC‐MS/MS analysis of the 1:1 mixed samples. The method in this study can be extended to quantitation of other intact proteome systems. The universal “one‐pot” dimethyl labeling of all the amino groups in a protein without the need of pre‐blocking of those on the lysine residues is made possible by protein identifica‐ tion and quantitation analysis using ProteinGoggle 2.0 with customized databases of both precursor and product ions containing heavy isotopes.

 INTRODUCTION Systematic and accurate quantitation of differentially expressed proteins in physiological and pathological con‐ ditions has always been an indispensable yet challenging goal of mass spectrometry (MS) based proteomics.1‐4 Sta‐ ble isotope labeling (SIL), either in vitro or in vivo, is among one of the most common methods in protein quantitation5,6, which so far has mostly been carried out at the peptide level. Early in 1999, Aebersold and co‐workers developed iso‐ tope‐coded affinity tags (ICATs)7, where all cysteines in enzymatic peptides were labeled with light and heavy ICATs, respectively. Relative quantitation is achieved through abundance ratio of paired peptides in the MS spectra. The advantage of this method is that it can enrich low abundance peptides using biotin on the tags, and therefore big dynamics range of quantitation is possible. Other than ICATs, proteolytic 18O labeling8, isobaric tags for relative and absolute quantitation (iTRAQ)9, tandem mass tags (TMT)10, and among others have also been of‐ ten reported. Besides aforementioned in vitro methods, in vivo metabolic labeling method, such as SILAC (stable

isotope labeling by amino acids in cell culture)11 is also popular. In SILAC, cells are grown in culture media lack‐ ing standard essential (auxotrophic) amino acid(s) but supplemented with a non‐radioactive, isotopically labeled form of that amino acid. Variant forms of SILAC including Super‐SILAC12,13, spiked‐in SILAC14,15 and Neucode SILAC16 have thereafter been reported. SIL procedure could be labor intensive and isotopic reagents are often quite ex‐ pensive. Reductive dimethylation of primary amines on lysine and N‐terminal using formaldehyde (HCHO) and sodium cyanoborohydride (NaBH3CN) has proven to be simple, yet cost‐effective and powerful.17,18 With its intro‐ duction in 1928 and thereafter evolution 19‐22, reductive dimethylation in its current form (in terms of using HCHO and NaBH3CN) was first used for peptide quanti‐ tation in 200323. In order to get isotopic pairs in the MS spectra with fixed mass difference to facilitate down‐ stream data interpretation and MS‐level quantitation, a variety of blocking strategies of the lysine amino group (such as guanidination) has been developed24,25. Hybrid labeling using dimethylation and SILAC to reach 6‐plex26 and 5‐plex isotopic dimethyl labeling27 have also been reported. MS/MS level quantitation with isobaric or 1|P a g e

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

pseudo‐isobaric labeling could be achieved through com‐ binatorial isotopic labeling of both the C‐ and N‐termini. For isobaric dimethyl labeling, several isobaric peptide terminal labeling (IPTL) strategies have been developed. Isobaric Lys‐C peptides could be obtained through cross succinylation (H4 vs. D4) and dimethylation (D4 vs. H4) of the N‐terminal and C‐terminal amino groups.28 Enzymatic 18O2 labeling of the C‐terminal carboxylate and dimethyl D4 labeling of the N‐terminal primary amine together with guanidination blocking of lysine amines provides at least 10‐fold linear quantitation and enables finding of 124 differentially expressed proteins in hepatocellular carcinoma (HCC) with more than 2‐fold differences.25 Besides duplex labeling, Thiede et al. also developed a triplex labeling using combinatorial N‐ and C‐terminal dimethyl labels of CH3+13CD3, CHD2+13CH2D, and 13CD3+CH3, respectively, which was successfully demonstrated through temporal profiling of HeLa cells incubated with the kinesin Eg5 inhibitor S‐Trityl‐L‐ cysteine.29 Besides isobaric fragment ions containing different number of heavy isotopes for relative quantitation, pseu‐ do‐isobaric fragments ions containing different type of heavy isotopes (such as 13C vs. D) could also be resolved in high‐resolution mass spectrometers and used for relative quantitation. Peptides with pseudo‐isobaric labels nor‐ mally could not be resolved in MS spectra, and thus are co‐selected for MS/MS fragmentation to produce MS/MS spectra, where small pseudo‐isobaric fragment ions could be resolved. For pseudo‐isobaric dimethyl labeling, Zhang et al. labeled two different groups of Lys‐C peptides at both the N‐terminal and C‐terminal lysine amino groups with (‐13CD2H)4 vs. (‐CD3)4, respectively. The subtle total mass difference between the labelled pseudo‐ isobaric peptide pairs is 11.688 mDa, and these two pep‐ tides are normally not resolved in their MS spectra; small a‐, b‐ and y‐ions (up to y6) with a mass difference of 5.84 mDa, however, were all resolved using 60 K resolution. A good linearity (R2=0.999) across a 100‐fold dynamic range for BSA digests and a strong correlation (R2 = 0.887) of differentially expressed proteins in Hca‐F and Hca‐P cell lines were finally obtained.30 Yates et al. applied a similar method in a MudPIT experiment and accurately meas‐ ured the protein levels of the cystic fibrosis transmem‐ brane conductance regulator.31 The bottom‐up proteomics procedure has high throughput and sensitivity in protein quantitation at the peptide level, but it may inherently not be able to quanti‐ fy a protein (gene product) with multiple proteoforms (protein species) where likely only one species is of inter‐ est under certain condition.32 Therefore, protein quantita‐ tion at the protein species level, i.e., the top‐down prote‐ omics procedure, is essential for molecular level under‐ standing of proteoforms, especially identifying combina‐ torial amino acid variation and post‐translational modifi‐ cations33. So far, a few top‐down SIL studies have been tried for intact protein quantitation. Smith and coworkers first applied SIL to top‐down quantitative proteome anal‐

Page 2 of 14

ysis of E. coli and S. cerevisiae using cell culture with iso‐ topically labeled amino acids;34‐36 protein abundance change associated with cadmium (Cd2+) stress response was precisely measured.34 With 15N‐labeling using (15NH4)2SO4 and 15N‐labeled histidine, leucine, and tryp‐ tophan in cell culture of S. cerevisiae, Kelleher, et al. suc‐ cessfully quantified 231 metabolically labeled (14N/15N) protein pairs.37 In 2007, Mann and coworkers extended their SILAC from peptide to protein quantitation;38 they showed that Grb2, a 28 kDa signaling protein, could be quantified with an average standard deviation of 6%. Hung & Tholey first demonstrated intact protein quantifi‐ cation using tandem mass tag (TMT) labeling;39 six‐plex TMT was used to quantify a standard protein mixture and nice linear range of 10‐fold (10:1 – 1:10) was achieved. De‐ spite initial success, top‐down quantitation approaches still face quite a few issues, such as isotopic envelope broadening of heavy labeled proteins due to incomplete labeling 38,40 and amino acid conversion40,41, non‐specific labeling and side reactions39. Here we report our attempt of top‐down quantitation using reductive dimethylation. All primary amino groups (N‐terminal and lysine) in intact proteins are labeled with pseudo‐isobaric (CDH2)2 and (13CH3)2 without blocking, which is made possible by our unique database search using iMEF42 and ProteinGoggle 2.043 where customized databases of both precursor and product ions with isotop‐ ic labels could be built. Our first proof‐of‐principle study of pIDL was carried out using a standard protein, myo‐ globin. Effect of mass resolution and size of product ions on quantitation accuracy as well as achievable quantita‐ tion dynamic range were explored. Besides myoglobin, the quantitative capability of the approach was also demonstrated with the intact proteomes from hepatocel‐ lular cell lines. The results show that pIDL is also a simple, cost‐effective, yet powerful quantitative strategy for top‐ down proteomics, which should make promising contri‐ bution to structural and functional study as well as clini‐ cal application of protein species.

 EXPERIMENTAL SECTION Chemicals and Reagents. Formaldehyde (HCHO) (Product # F8775, 37% w/w in H2O), sodium cyanoboro‐ hydride (Product # 42077, NaBH3CN), ammonia solution (Product # 320145, 28% v/v in H2O), sodium acetate (Product # 71185, 99%), sodium cyanoborodeuteride (NaBD3CN, 98%, 96% D, cat. no.190020) and horse myo‐ globin (M1882) were purchased from Sigma‐Aldrich (St. Louis, MO, USA). All solvents (such as acetonitrile) used were HPLC grade and also purchased from Sigma‐Aldrich (St. Louis, MO, USA). Formaldehyde (H13CHO, 20% w/w in H2O, 99% 13C, cat. no. CLM‐806‐0) was obtained from Cambridge Isotope Laboratories, Inc. (Andover, MA, USA). Ultrapure water was produced on site by Millipore Simplicity System (Billerica, MA). Preparation of the HepG2 and LO2 proteomes. The hepatocellular carcinoma (HepG2) and normal (LO2) cells 2|P a g e


Page 3 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


were bought from the American Type Culture Collection (http://www.atcc.org/) and Shanghai Biological Technolo‐ gy (Shanghai, China), respectively. HepG2 cells are main‐ tained in RPMI 1640 medium which is supplemented with 10% fetal bovine serum (FBS) and 100 ug/mL penicillin‐ streptomycin. LO2 cells are maintained in Dulbecco's modified Eagle's medium which is supplemented with 10% FBS and 100 ug/mL penicillin‐streptomycin. Both cell lines were cultured in 100 mm culture dishes that were placed in a humidified incubator maintained at 37 °C and 5% CO2. Cells were first grown to approximately 80% confluence. After centrifugation and washing with phosphate‐buffered saline (PBS), the cell pellets were re‐ suspended in the lysis buffer (50 mM Tris‐HCl, 100 mM NaCl) supplemented with protease inhibitor cocktail and sonicated for 15 min on ice. The lysate was centrifuged at 14,000g for 15 min, and the supernatant protein mixture was collected. After protein concentration measurement by the BCA method, the mixture was aliquoted into 1.5 mL centrifuge tubes and stored at ‐80 ℃. Reductive dimethylation of myoglobin and hepato‐ cellular proteomes. Stock solution of myoglobin (5 μg/μL), NaBH3CN (600 mM), NaBD3CN (600 mM), and CH3COONa (100 mM, pH 5‐6) were freshly made by dis‐ solving the corresponding powder samples in water. Soni‐ cation was used to facilitate dissolving of myoglobin. Stock solution of HCHO (4%, w/w), H13CHO (4%, w/w), and NH4OH (4%, v/v) were freshly made from the corre‐ sponding solution samples. Pseudo‐isobaric dimethylation of N‐terminal and lysine amino groups with HCHO and NaBH3CN was carried out using the same protocol as reported for peptides.18 To check the labeling efficiency of the peptide protocol for intact proteins, dimethylation of myoglobin with HCHO and NaBH3CN was first carried out. In a 0.5 mL centrifuge tube, 5.0 μL myoglobin solution was added to 100.0 μL sodium acetate buffer. After mixing, 4.0 μL freshly pre‐ pared HCHO solution (4%) was then added followed by immediate vortex for 1min. With addition of 4.0 μL stock solution of NaBH3CN, the reaction mixture was vortexed and incubated in a fume hood for 1h at room temperature (20 ℃). Then, the reaction was interrupted by addition of 4 μL stock solution of NH4OH. For dimethyl labeling with (13CH3)2 (short for 13C，Figure S1A) and (CDH2)2 (short for D, Figure S1B), the procedure was the same as above except that isotopic reagents H13CHO+NaBH3CN and HCHO+NaBD3CN were used instead of HCHO+NaBH3CN. After labeling, pseudo‐isobaric myoglobin with 13C and D labels were mixed at the ratios of 13C/D=1:100, 1:50, 1:10, 1:1, 10:1, 50:1, 100:1 and stored in a 20℃ freezer for subse‐ quent RPLC‐MS/MS analysis. Two identical aliquots of HepG2 and LO2 proteomes were labeled with (13CH3)2 and (CDH2)2, respectively, using the same reagents and procedure described above. RPLC‐MS/MS analysis of labeled myoglobin and hepatocellular proteomes. Myoglobin with dimethyla‐ tion of (CH3)2, (13CH3)2, and (CDH2)2 were first ana‐

lyzed separately using RPLC‐MS/MS to check labeling efficiency. Mixed myoglobin with dimethylation of pseu‐ do‐isobaric (13CH3)2 and (CDH2)2 was then analyzed in the same way for investigation of linear quantitation range. For all RPLC‐MS/MS analyses, labeled myoglobin or hepatocellular proteome (either individual or mixed) was first desalted and trapped on a 4 cm long trapping col‐ umn (360 μm o.d. × 200 μm i.d.) packed with C4 particles (300 Å, 5 μm). The loading buffer (4.8% ACN, 95.0% H2O, and 0.2% FA) at a flow rate of 5 μL/min for 6 minutes. The trapped protein was separated on a 65 cm long ana‐ lytical column (360 μm o.d. × 75 μm i.d.) packed with the same C4 particles as described above on a Dionex Ulti‐ mate 3000 RSLC nano‐HPLC system (Thermo Fisher Sci‐ entific). Buffer A is mixture of 95.0% H2O, 4.8% ACN and 0.2% FA; Buffer B is mixture of 95.0% ACN, 4.8% H2O and 0.2% FA. Elution at constant flow of 300 nL/min was con‐ ducted at the following gradient: 1‐15% B for 1 min, 15‐65% B for 91 min, 65‐75% B for 6 min, 75‐99% B for 5 min. Eluted proteins were detected online with ESI tandem mass spectrometry using a Q Exactive mass spectrometer (Thermo Fisher Scientific) coupled with a nano‐ESI source. MS spectra were acquired in the 600‐2000 m/z range using a mass resolution setting of 140 k (m/z 200). Fragmentation was obtained in a data‐dependent mode (Top 10) with higher‐energy collisional dissociation (HCD). For MS/MS spectra, fixed first mass of m/z 50 was utilized. Three mass resolution setting of 35 k, 70 k and 140 k was tried for finding of the optimal resolution to be used. AGC target value and maximum injection time were placed at 5e5 and 250 ms for both MS and MS/MS scans. Isolation window and dynamic exclusion were set at 10.0 m/z and 20.0 s. Normalized collision energy was set at 30.0 %. The temperature of the ion transfer capillary was set to 250 °C. The spray voltage was set to 2.6 kV. Protein identification and quantitation. Protein identification and quantitative analysis of the RPLC‐ MS/MS datasets acquired above was done with Pro‐ teinGoggle 2.0. 43 Customized precursor and product ion databases with dimethyl labels ((CH3)2, (13CH3)2, and (CDH2)2) were first built separately with the flat text files downloaded from Uniprot (http://www.uniprot.org/). ProteinGoggle has the capability of reading in heavy iso‐ topes to produce the corresponding theoretical isotopic envelopes, and Ct and D are used as element‐like symbol for 13C and D atoms, respectively. Dimethylation of all the amino groups on the N‐terminal and lysine was treated static, i.e., as fixed modification. The search parameters for protein identification are as follows. For precursor ions, IPACO, IPMD, and IPAD are 40%, 15 ppm, and 100%, respectively; for product ions the corresponding values are 20%, 15 ppm, and 50%, respectively. IPACO, IPMD, and IPAD are short names of isotopic peak abundance cutoff, isotopic peak m/z deviation, and isotopic peak abundance deviation, respectively. For datasets from analysis of mixed 13C‐ and D‐labeled myoglobin, proteins were first searched against the 13C database to find all 3|P a g e



1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

matching b and y ions. For each of these matching prod‐ uct ion, experimental m/z and abundance of all the ob‐ served isotopic peaks (including the mono‐isotopic peaks) within IPMD and IPAD tolerance are stored in a meta data file. The corresponding D‐labeled mono‐isotopic peaks for the aforementioned matching b and y ions were then searched in the metadata file and quantitative re‐ sults were output in a txt file. The relative abundance of the paired pseudo‐isobaric mono‐isotopic peaks of 13C‐ and D‐labeled product ions was used for calculation of observed ratios and relative quantitation. For each mix‐ ing ratio, the top‐10 IDs with most matching product ions and highest P Score were used to calculate average ob‐ served ratio and standard deviation.

 RESULTS AND DISCUSSION Labeling efficiency of protein reductive dimethyla‐ tion. Labeling efficiency of dimethylation of N‐terminal and lysine amino groups was first checked with non‐ isotopic reagents HCHO and NaBH3CN. There are 19 ly‐ sines across the amino acid sequence of myoglobin; to‐ gether with the N‐terminal amino groups, there are in total 20 amino groups which should all be labeled with (CH3)2. ESI MS spectrum of the fully labeled myoglobin

Page 4 of 14

is shown in Figure 1A of all charge states and the deconvo‐ luted spectrum with z=+1 was plotted in Figure 1B. As can be seen in Figure 1B, only one major peak pattern was observed which means dimethyl labeling is nearly quanti‐ tative given few percentage of impurity of the isotopic reagents. The minor peak pattern is proposed to be the NH3 loss ion of the major peak pattern on the right, be‐ cause this ion is also observed in the MS spectrum of the original myoglobin sample (Figure S2) and there is a good fingerprinting between the experimental and theoretical isotopic envelopes (Figure S3). To check whether this ma‐ jor ion contains 20 (CH3)2 labels as expected, the exper‐ imental isotopic envelope of charge state z=20+ was first fingerprinted to the corresponding theoretical isotopic envelope (Figure 1C). A good match was found as IPMDs of all the observed isotopic peaks are within 4 ppm and the relative abundance of all isotopic peaks are within 17%. The exact labeling sites were further confirmed with da‐ tabase search of the MS/MS spectrum from HCD of the 20+ precursor ion against the customized database with the 20 dimethyl labels as static modifications. A total of 35 matching b and y ions were found, and the graphical fragmentation map is shown in Figure 1D.

Figure 1. Dimethyl labeling efficiency of myoglobin using HCHO and NaBH3CN. (A) MS spectrum of labeled myoglobin with different charge states. (B) the deconvoluted MS spectrum in (A). (C) The iEF map, i.e., fingerprinting of the experimental iso‐ topic envelope on the corresponding theoretical isotopic envelope, of the charge state 20+ ion, IPMD (ppm) and IPAD (%) val‐ ues of each isotopic peak are shown. (D) The graphical fragmentation map from database search using ProteinGoggle 2.0, show‐ ing matching b and y ions as well as dimethylation of primary amino groups at the N‐terminal and lysine residues (19 in total). 4|P a g e


Page 5 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Similar quantitative labeling efficiency was obtained for dimethylation of (13CH3)2 and (CDH2)2 using isotopic reagents H13CHO+NaBH3CN and HCHO+NaBD3CN. The corresponding full MS spectrum, the deconvoluted MS spectrum, the iEF map of the 20+ precursor ion, and the graphical fragmentation map for the HCD spectrum of the 20+ precursor ion are provided in Supplemental Fig‐ ure S4 and S5, respectively. The detailed experimental and theoretical information for the matching b and y ions are presented in Table S1 and S2. To summarize, dimethyl labeling of both the N‐terminal and lysine amino groups using HCHO and NaBH3CN is almost quantitative under the current reaction conditions. Effect of product ion size on quantitation accuracy. Effect of the size of the product ions on quantitation ac‐ curacy was explored with 1:1 mixed 13C‐ and D‐labeled myoglobin. Under 140 k resolution, the pseudo‐isobaric mono‐isotopic peak pairs of six product ions (including b2‐1+, b3‐1+, b4‐1+, b5‐1+, b6‐1+, and b7‐2+) were resolved. All of these six ions contain one pseudo‐isobaric dimethyl label, either (13CH3)2 or (CDH2)2. The corresponding measured IPMDs, mass difference, and the observed rati‐ os for the paired mono‐isotopic peaks in the pseudo‐ isobaric product ions among the top 10 best MS/MS spec‐ tra with highest P Score are listed in Table 1. All the ob‐ served ratios from all MS/MS spectra are plotted in Figure 2. The IPMD values of the mono‐isotopic peaks are all within 4 ppm. However, only the first three ions (b2‐1+, b3‐1+ and b4‐1+) have better agreement between the measured and the theoretical mass difference (5.84 mDa). For the observed ratios, good average values are obtained for all the six product ions, although b5‐1+ has slightly higher deviation than the other 5 ions. Table 1. Isotopic Peak m/z Deviation (IPMD, ppm) and Mass Difference (ΔM, mDa) of the Observed Paired 13 Mono‐isotopic Peaks with Pseudo‐isobaric ( CH3)2 and (CDH2) 2 Labeling in Different Product Ions. The Mix‐ 13 ing Ratio of ( CH3)2 and (CDH2)2 Labeled Myoglobin is 1:1. The Average Values and the Corresponding Standard Deviations are Calculated from the Top 10 IDs with Highest P Scores. 13

product ion

IPMD, C (ppm)

IPMD, D (ppm)

ΔM (mDa)

Obs. ratio 13 ( C/D)

b2‐1＋

‐0.4±0.5

‐0.5±1.5

5.81±0.07

(1.1±0.1):1 (1.0±0.1):1

b3‐1＋

‐1.0±0.9

‐0.8±0.9

5.89±0.06

b4‐1＋

‐1.2±0.8

‐0.7±0.8

6.06±0.11

b5‐1＋

‐3.2±1.2

‐0.5±0.9

7.14±0.62

b6‐1＋

‐0.1±1.1

‐1.7±0.7

4.79±0.73

b7‐2＋

‐0.8±1.7

‐2.7±1.4

4.22±2.00

(1.1±0.2):1 (1.3±0.3):1 (1.0±0.1):1 (1.0±0.1):1

Effects of MS resolution on quantitation. The exact mass difference of the pseudo‐isobaric dimethyl labels between (13CH3)2 and (CDH2)2 is 5.84 mDa. The total mass difference between the 1 3 C‐ and D‐labeled

Figure 2. Effect of size of pseudo‐isobaric product ions on quantitation accuracy. The expected ratio is 1:1, i.e., differen‐ tially labeled myoglobin with (13CH3)2 and (CDH2)2 was mixed at the ratio of 1:1. The box plot contains all the ob‐ served values for each product ion.

myoglobin is 5.84*20=116.8 mDa. As the molecular weight of myoglobin is 16951.99966 Da. There is no resolution of the pseudo‐isobaric precursor ions with our MS with 140 k resolution (m/z 200). For accurate relative quantitation using abundance of pseudo‐isobaric product ions, high enough mass resolution is needed to resolve the paired mono‐isotopic peaks to provide accurate peak position and distance; on the other hand, higher resolution means longer detection time and lower signal. So, a compro‐ mised mass resolution must be found. To find optimal resolution, 35 k, 70 k, and 140 k to resolve b2‐1+ ion were investigated and compared; the measured results are shown in Figure 3. The b2‐1+ ion has one dimethyl label at the N‐terminal, and the theoretical m/z of the paired pseudo‐isobaric mono‐isotopic peaks with 13C and D la‐ bels are 201.15081and 201.15665, respectively. The theoreti‐ cal mass difference is 5.84 mDa. The paired pseudo‐ isobaric mono‐isotopic peaks are resolved in all the three mass resolution; and IPMDs of all the isotopic peaks are within 5 ppm. In addition, the measured mass difference at both 70 k and 140 k is relative good with deviation of less than 2%; while much bigger deviation (up to 16%) was observed at 35 k. Largely, mass resolution of 140 k is needed for accurate identification and quantitation. Achievable linear quantitation range with Myoglo‐ bin. After finding the optimal MS resolution to resolve the pseudo‐isobaric monoisotopic peak pair and product ion to provide accurate quantitation, we further investi‐ gate the quantitation dynamic range with pseudo‐isobaric dimethylation. Seven RPLC‐MS/MS datasets were ob‐ tained from analyses of the 7 isotopically labeled myoglo‐ bin samples, where (13CH3)2‐ and (CDH2)2‐labeled myo‐ globin were mixed in a 100‐fold range with 7 different ratios: 1:100, 1:50, 1:10, 1:1, 10:1, 50:1, and 100:1. Database search and quantitative analysis of these datasets was carried out using ProteinGoggle, and the quantitative 5|P a g e



1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 14

results were shown in Table 2; the observed log2(13C/D) ratios (average values and standard deviation) were calcu‐ lated from the paired mono‐isotopic peak abundance of the top10 IDs with highest P Scores. From the most abundant product ion b2‐1+, a good quantitation linearity (R2=1.00) was obtained in the 50‐ fold range with the five data points of 1:50, 1:10, 1:1, 10:1, and 50:1 (Figure 4); but, big deviation of the observed ra‐ tio from the corresponding expected ratios was

Figure 4. Linear quantitation dynamic range of myoglobin using the most abundant product ion, b2‐1+, in a 50‐fold 13 range. Differentially labeled myoglobin with ( CH3)2 and (CDH2)2 was mixed at the ratios of 1:50, 1:10, 1:1, 10:1 and 50:1. The average values and the corresponding standard devia‐ tions are calculated from the top 10 IDs with highest P Scores.

Figure 3. Effect of mass resolution on quantitation accuracy. Measured isotopic peak m/z deviation (IPMD, ppm) and mass difference (mDa) of the paired pseudo‐isobaric mono‐ isotopic peaks in the b2‐1+ product ion are investigated at mass resolution of 35 k, 70 k and 140 k, respectively. The mix‐ 13 ing ratio of ( CH3)2 and (CDH2)2 labeled myoglobin is 1:1. The average values and the corresponding standard devia‐ tions are calculated from the top 10 IDs with highest P Scores.

observed when the dynamic range was extended to 100‐ fold. Pseudo‐isobaric product ion pairs were also observed for b3‐1+ and b4‐1+ in the 50‐fold range; but no good line‐ arity was obtained for either of the two ions, which should be mostly due to their lower abundance as com‐ pared to that of b2‐1+. Quantitation of hepatocellular intact proteome us‐ ing pIDL. The intact proteomes from the hepatocellular HepG2 and LO2 cell lines were prepared to benchmark the applicability of pIDL for quantitation of intact protein mixture. Two identical aliquots of HepG2 and LO2 intact proteomes were separately labeled with (13CH3)2 and (CDH2)2, mixed with 1:1 ratio, and the mixed sample was analyzed with one‐dimensional RPLC‐MS/MS using HCD and three technical replicates. After qualitative and quan‐ titative data analysis using ProteinGoggle, 33 proteins were quantified at least twice out of the three replicate

Table 2. All observed paired pseudo‐isobaric product ions for relative quantitation in the 50‐fold range. Differentially labeled myoglobin with (13CH3)2 (13C) and (CDH2)2 (D) was mixed at 5 different ratios in a 50‐fold range. The average values and the corresponding standard deviations are calculated from the top 10 IDs with highest P Scores. N/A means no experimental data is available. 13

product ion

C/D

1:100

1:50

1:10

1:1

10:1

50:1

100:1

b2‐1＋

1:(34.4±12.3)

1:(47.1±9.2)

1:(9.0±0.9)

(1.1±0.1):1

(12.4±2.4):1

(49.0±1.0):1

(35.7±11.5):1

b3‐1＋

1:(30.6±6.3)

1:(32.5±6.1)

1:(9.1±0.1)

(1.0±0.0):1

(16.9±6.9):1

(33.6±3.7):1

(25.8±7.1):1

b4‐1＋

1:(25.6±7.7)

1:(25.1±5.1)

1:(11.5±1.4)

(1.1±5.1):1

(18.1±8.2):1

(23.4±6.6):1

(25.5±5.5):1

b5‐1＋

N/A

N/A

1:(20.1±5.1)

(1.3±6.3):1

N/A

N/A

N/A

b6‐1＋

N/A

N/A

N/A

(1.0±0.9):1

N/A

N/A

N/A

b7‐2＋

N/A

N/A

N/A

(1.0±1.0):1

N/A

N/A

N/A

6|P a g e


Page 7 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


runs with RSD10%. The detailed information (including protein ID, accession number, sequence length, product ion for quantitation, and average relative abundance ratio of HepG2/LO2, SD and RSD values) for these 33 proteins is provided in supplemental Table S3. More than 60% of the proteins are quantified with either b1‐1+ or b2‐1+; and the rest proteins are quantified with either b3‐1+, b3‐2+, b4‐1+, b4‐2+, y1‐1+, y2‐2+, y3‐1+ or y3‐2+. This may indi‐ cate that different proteins have both different type and different size of product ions for best quantitation, alt‐ hough further investigation with more data is essential. This observation is different from that of peptides where different product ions (a1, b1, y1, a2, b2, y2, y3, b3, y4, y6) performs equally well for relative quantitation. The un‐ derlying reason can be that a protein always is fragment‐ ed into many more product ions with much lower abun‐ dance than a peptide is, and relative quantitation accura‐ cy is strongly sensitive to the abundance of the product ion. Relative to LO2, 12 out of the 33 quantified proteins were found to be up regulated or down regulated in HePG2 with a threshold of bigger than 2‐fold. The anno‐ tated MS/MS spectrum, the pseudo‐isobaric b2‐1+ ion pair for quantitation, and the graphical fragmentation map for protein NEDD8_HUMAN are shown in Figure 5 as an example.

could be resolved under both 70 k and above and their relative abundance could be used for quantitation. A good linear quantitation range (R2=1.00) was obtained from the most abundant b2‐1+ ion in a 50‐fold range with mixing ratios of 1:50, 1:10, 1:1, 10:1, and 50:1. The pIDL method was also benchmarked with hepatocellular carcinoma HepG2 proteome relative to that of normal LO2, and 33 unique proteins were quantified within 10% RSD. Other than combination of (13CH3)2‐ vs. (CDH2)2, pseudo‐isobaric labels could also be (13CH2D)2 vs. (CHD2)2, and (13CHD2)2 vs. (CD3)2 depending on the availability and cost of the isotopic reagents of HCHO and NaBH3CN. This protein quantitation method using pseudo‐isobaric dimethyl labels on the amino groups, in principle, could be extended to any other pseudo‐isobaric labels of other functional groups on either protein termini or amino acid residues. Accurate quantitation of protein species at the protein level should facilitate structural and functional studies and eventually contribute to the related applica‐ tion, such as development of protein therapeutic drugs.

 ASSOCIATED CONTENT

Supporting Information The supporting information is available free of charge via the ACS Publications website at http://pubs.acs.org/. Supplemental figures (5) regarding labeling efficiency and observed ratios, as well as tables (3) about the matching b and y ions from 13C‐ and D‐labeled myoglobin are provided, 43 pages in total.

 AUTHOR INFORMATION Corresponding Author * [email protected]

Author Contributions

Figure 5. The annotated MS/MS spectrum (A), the pseudo‐ isobaric b2‐1+ ion pair (A, inset), and the graphical fragmen‐ tation map (B) for the quantified protein NEDD8_HUMAN. Dimethyl labelled amino acids with amino groups are high‐ lighted with solid red squares.

The quantified 33 proteins have a sequence length range of 24 (HMN13_HUMAN) to 120 (RPRML_HUMAN). Addi‐ tional dimensional pre‐fractionation as well as separation of the hepatocellular proteomes is under way for more comprehensive identification and quantification.

 CONCLUSIONS We have extended simple, cost‐effective, yet powerful pseudo‐isobaric dimethyl labeling from peptide to intact protein quantitation. In benchmarking with myoglobin as model protein, two identical aliquots of myoglobin were labeled with (13CH3)2 and (CDH2)2, and pseudo‐isobaric mono‐isotopic peak pairs of a couple of product ions

ZXT conceived the study; HQF did the experiment; KJX de‐ veloped ProteinGoggle 2.0; HQF, KJX, and FY analyzed the data; YHL, YL, and BBX maintained the instrument; HQF and ZXT wrote the manuscript. ‡These authors contributed equally.

Notes The authors declare no competing financial interest.

 ACKNOWLEDGMENT This research was financially supported by National Science Foundation of China (21575104), China State Key Basic Re‐ search Program Grant (2013CB911203), Shanghai Science and Technology Commission (14DZ2261100), the China “Youth 1000‐talents Program”.

 ABBREVIATIONS MS, mass spectrometry; SIL, stable isotope labeling; ICATs, isotope‐coded affinity tags; iTRAQ, isobaric tags for relative and absolute quantitation; TMT, tandem mass tag; SILAC, stable isotope labeling by amino acids in cell culture; iMEF, isotopic mass‐to‐charge (m/z) ratio and envelope finger‐ printing; HPLC, high performance liquid chromatograph HCD, higher‐energy collisional dissociation; AGC, automatic gain control; IPACO, isotopic peak abundance cutoff; IPMD, 7|P a g e



1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

isotopic peak mass‐to‐charge (m/z) deviation; IPAD, isotopic peak abundance deviation

 REFERENCES (1) Ong, S. E.; Mann, M. Nat. Chem. Biol. 2005, 1, 252‐262. (2) Ong, S. E.; Foster, L. J.; Mann, M. Methods 2003, 29, 124‐130. (3) Tao, W. A.; Aebersold, R. Curr. Opin. Biotech. 2003, 14, 110‐118. (4) Cox, J.; Mann, M. Annu. Rev. Biochem. 2011, 80, 273‐299. (5) Bantscheff, M.; Schirle, M.; Sweetman, G.; Rick, J.; Kuster, B. Anal. Bioanal. Chem. 2007, 389, 1017‐1031. (6) Gevaert, K.; Impens, F.; Ghesquiere, B.; Van Damme, P.; Lambrechts, A.; Vandekerckhove, J. Proteomics 2008, 8, 4873‐ 4885. (7) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aebersold, R. Nat. Biotechnol. 1999, 17, 994‐999. (8) Yao, X. D.; Freas, A.; Ramirez, J.; Demirev, P. A.; Fenselau, C. Anal. Chem. 2001, 73, 2836‐2842. (9) Ross, P. L.; Huang, Y. L. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet‐Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Mol. Cell. Proteomics 2004, 3, 1154‐ 1169. (10) Thompson, A.; Schafer, J.; Kuhn, K.; Kienle, S.; Schwarz, J.; Schmidt, G.; Neumann, T.; Hamon, C. Anal. Chem. 2003, 75, 1895‐1904. (11) Ong, S. E.; Blagoev, B.; Kratchmarova, I.; Kristensen, D. B.; Steen, H.; Pandey, A.; Mann, M. Mol. Cell. Proteomics 2002, 1, 376‐386. (12) Deeb, S. J.; D'Souza, R. C.; Cox, J.; Schmidt‐Supprian, M.; Mann, M. Mol. Cell. Proteomics 2012, 11, 77‐89. (13) Geiger, T.; Cox, J.; Ostasiewicz, P.; Wisniewski, J. R.; Mann, M. Nat. Methods 2010, 7, 383‐385. (14) Monetti, M.; Nagaraj, N.; Sharma, K.; Mann, M. Nat. Methods 2011, 8, 655‐658. (15) Geiger, T.; Wisniewski, J. R.; Cox, J.; Zanivan, S.; Kruger, M.; Ishihama, Y.; Mann, M. Nat. protocols 2011, 6, 147‐157. (16) Hebert, A. S.; Merrill, A. E.; Bailey, D. J.; Still, A. J.; Westphall, M. S.; Strieter, E. R.; Pagliarini, D. J.; Coon, J. J. Nat. Methods 2013, 10, 332‐334. (17) Kovanich, D.; Cappadona, S.; Raijmakers, R.; Mohammed, S.; Scholten, A.; Heck, A. J. R. Anal Bioanal Chem 2012, 404, 991‐ 1009. (18) Boersema, P. J.; Raijmakers, R.; Lemeer, S.; Mohammed, S.; Heck, A. J. R. Nat. protocols 2009, 4, 484‐494. (19) Means, G. E.; Feeney, R. E. Biochemistry 1968, 7, 2192‐2201. (20) Rice, R. H.; Means, G. E. J. Biol. Chem. 1971, 246, 831‐832. (21) Dottavio‐Martin, D.; Ravel, J. M. Anal. Biochem. 1978, 87, 562‐565. (22) Neil Jentoft, D. G. D. J. Biol. Chem. 1979, 254, 4359. (23) Hsu, J. L.; Huang, S. Y.; Chow, N. H.; Chen, S. H. Anal. Chem. 2003, 75, 6843‐6852. (24) Ji, C.; Guo, N.; Li, L. J. Proteome Res. 2005, 4, 2099‐2108. (25) Yang, S. J.; Nie, A. Y.; Zhang, L.; Yan, G. Q.; Yao, J.; Xie, L. Q.; Lu, H. J.; Yang, P. Y. J. Proteomics 2012, 75, 5797‐5806. (26) Wang, F.; Cheng, K.; Wei, X.; Qin, H.; Chen, R.; Liu, J.; Zou, H. Sci. Rep. 2013, 3, 1827. (27) Wu, Y.; Wang, F.; Liu, Z.; Qin, H.; Song, C.; Huang, J.; Bian, Y.; Wei, X.; Dong, J.; Zou, H. Chem. Commun. 2014, 50, 1708‐1710. (28) Koehler, C. J.; Strozynski, M.; Kozielski, F.; Treumann, A.; Thiede, B. J. Proteome Res. 2009, 8, 4333‐4341. (29) Koehler, C. J.; Arntzen, M. O.; de Souza, G. A.; Thiede, B. Anal. Chem. 2013, 85, 2478‐2485. (30) Zhou, Y.; Shan, Y. C.; Wu, Q.; Zhang, S.; Zhang, L. H.; Zhang, Y. K. Anal. Chem. 2013, 85, 10658‐10663. (31) Bamberger, C.; Pankow, S.; Park, S. K.; Yates, J. R., 3rd. J. Proteome Res. 2014, 13, 1494‐1501. (32) Jungblut, P. R. J. Proteomics 2014, 107, 98‐102.

Page 8 of 14

(33) Zhang, Z. R.; Wu, S.; Stenoien, D. L.; Pasa‐Tolic, L. Annu. Rev. Anal. Chem. 2014, 7, 427‐454. (34) Pasa‐Tolic, L.; Jensen, P. K.; Anderson, G. A.; Lipton, M. S.; Peden, K. K.; Martinovic, S.; Tolic, N.; Bruce, J. E.; Smith, R. D. J. Am. Chem. Soc. 1999, 121, 7949‐7950. (35) Veenstra, T. D.; Martinovic, S.; Anderson, G. A.; Pasa‐Tolic, L.; Smith, R. D. J. Am. Soc. Mass Spectr. 2000, 11, 78‐82. (36) Martinovic, S.; Veenstra, T. D.; Anderson, G. A.; Pasa‐Tolic, L.; Smith, R. D. J. Mass Spectrom. 2002, 37, 99‐107. (37) Parks, B. A.; Jiang, L.; Thomas, P. M.; Wenger, C. D.; Roth, M. J.; Boyne, M. T.; Burke, P. V.; Kwast, K. E.; Kelleher, N. L. Anal. Chem. 2007, 79, 7984‐7991. (38) Waanders, L. F.; Hanke, S.; Mann, M. J. Am. Soc. Mass Spectr. 2007, 18, 2058‐2064. (39) Hung, C. W.; Tholey, A. Anal. Chem. 2012, 84, 161‐170. (40) Collier, T. S.; Hawkridge, A. M.; Georgianna, D. R.; Payne, G. A.; Muddiman, D. C. Anal Chem 2008, 80, 4994‐5001. (41) Collier, T. S.; Muddiman, D. C. Amino Acids 2012, 43, 1109‐ 1117. (42) Li, L.; Tian, Z. Rapid Commun. Mass Spectrom. 2013, 27, 1267‐1277. (43) Xiao, K. J.; Yu, F.; Fang, H. Q.; Xue, B. B.; Liu, Y.; Tian, Z. X. Sci. Rep. 2015, 5, 14755.

8|P a g e


Page 9 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents / Abstract Graphics

9|P a g e



1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Dimethyl labeling efficiency of myoglobin using HCHO and NaBH3CN. (A) MS spectrum of labeled myoglobin with different charge states. (B) the deconvoluted MS spectrum in (A). (C) The iEF map, i.e., fingerprinting of the experimental isotopic envelope on the corresponding theoretical isotopic envelope, of the charge state 20+ ion, IPMD (ppm) and IPAD (%) values of each isotopic peak are shown. (D) The graphical fragmentation map from database search using ProteinGoggle 2.0, showing matching b and y ions as well as dimethylation of primary amino groups at the N-terminal and lysine residues (19 in total). Figure 1 84x54mm (300 x 300 DPI)


Page 10 of 14

Page 11 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Effect of size of pseudo-isobaric product ions on quantita-tion accuracy. The expected ratio is 1:1, i.e., differentially labeled myo-globin with −(13CH3)2 and −(CDH2)2 was mixed at the ratio of 1:1. The box plot contains all the observed values for each product ion. Figure 2 84x85mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Effect of mass resolution on quantitation accuracy. Meas-ured isotopic peak m/z deviation (IPMD, ppm) and mass difference (mDa) of the paired pseudo-isobaric mono-isotopic peaks in the b2-1+ product ion are investigated at mass resolution of 35 k, 70 k and 140 k, respectively. The mixing ratio of −(13CH3)2 and −(CDH2)2 labeled myoglobin is 1:1. The average values and the corresponding standard deviations are calculated from the top 10 IDs with highest P Scores. Figure 3 84x56mm (300 x 300 DPI)


Page 12 of 14

Page 13 of 14

1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Linear quantitation dynamic range of myoglobin using the most abundant product ion, b2-1+, in a 50-fold range. Differentially labeled myoglobin with −(13CH3)2 and −(CDH2)2 was mixed at the ratios of 1:50, 1:10, 1:1, 10:1 and 50:1. The average values and the corresponding standard deviations are calculated from the top 10 IDs with highest P Scores. Figure 4 84x62mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 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 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. The annotated MS/MS spectrum (A), the pseudo-isobaric b2-1+ ion pair (A, inset), and the graphical fragmentation map (B) for the quantified protein NEDD8_HUMAN. Dimethyl labelled amino acids with amino groups are highlighted with solid red squares. Figure 5 84x67mm (300 x 300 DPI)


Page 14 of 14

Intact Protein Quantitation Using Pseudoisobaric Dimethyl Labeling

Recommend Documents