Page 1 of 21
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
High Resolution Parallel Reaction Monitoring with Electron Transfer Dissociation for MiddleDown Proteomics
Michael J. Sweredoski1, Annie Moradian1, Matthias Raedle1,2, Catarina Franco1,3 and Sonja Hess1,* 1
Proteome Exploration Laboratory, Division of Biology, Beckman Institute, California Institute of Technology, Pasadena, CA 91125, USA
2
Hochschule Weihenstephan-Triesdorf, University of Applied Sciences, Faculty of Biotechnology and Bioinformatic, Am Hofgarten 4, 85354 Freising, Germany
3
Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, Av. da República, 2780157 Oeiras, Portugal
[email protected] RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required according to the journal that you are submitting your paper to) *
Corresponding author: email: [email protected], phone: 626-395-2339, fax: 626-449-4159
ACS Paragon Plus Environment
1
Analytical Chemistry
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 2 of 21
ABSTRACT
In recent years, middle-down proteomics has emerged as a popular technique for the characterization and quantification of proteins not readily amenable to typical bottom-up approaches. So far, all high resolution middle-down approaches are done in data-dependent acquisition mode, using both, collisioninduced dissociation or electron capture/transfer dissociation techniques.
Here, we explore middle-down proteomics with electron transfer dissociation using a targeted acquisition mode, parallel reaction monitoring (PRM), on an Orbitrap Fusion. As an example of a highly modified protein, we used histone H3 fractions from untreated and DMSO-treated Murine ErythroLeukemia (MEL) cells. We first determined optimized instrument parameters to obtain high sequence coverage using a synthetic standard peptide. We then setup a combined method of both MS1 scans and PRM scans of the twenty most abundant combinations of methylation and acetylation of the +10 charge state of the N-terminal tail of H3. Weak cation exchange hydrophilic interaction chromatography was used to separate the N-terminal H3 tail primarily by its acetylation, and to a secondary degree, by its methylation status, which aided in the interpretation of the results. After deconvolution of the highly charged ions, peaks were annotated to a minimum set of 254 H3 proteoforms in the untreated and treated samples. Upon DMSO treatment global quantitation changes from the MS1 level show a relative decrease of 2, 3, 4, and 5 acetylations and an increase of 0 and 1 acetylations. A fragment ion map was developed to visualize specific differences between treated and untreated samples. Taken together the data presented here show that middle-down proteomics with electron transfer dissociation using PRM is a novel, attractive method for the effective analysis and quantification of large and highly modified peptides.
ACS Paragon Plus Environment
2
Page 3 of 21
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
INTRODUCTION
In recent years, middle-down proteomics has emerged as a popular technique for the characterization and quantification of proteins not readily amenable to typical bottom-up approaches such as antibodies, ubiquitin chains and histones.1-8 Thus, middle-down approaches excel where high protein sequence coverage is needed and a typical tryptic digestion would yield peptides too small for a successful LCMS/MS analysis preventing the full characterization of a given protein.9,10 Nevertheless, these approaches cannot yet be considered routine applications since the bioinformatics pipelines developed for bottom-up or top-down analyses do not readily apply for middle-down approaches. Middle-down approaches have been used with collision-induced dissociation (CID)7,11,12 and electron capture/transfer dissociation (ECD/ETD) techniques.4,13-16 Since middle-down approaches produce predominantly higher charged species, ECD/ETD techniques generally yield more extensive fragmentation across the peptide when compared to CID. As a non-ergodic process, ECD/ETD fragmentation is also favorable for the characterization of post-translationally modified peptides.17-20 Middle-down data-dependent acquisition (DDA) approaches for histones have been successfully used in combination with ECD and ETD.4,13,21-24
However, a limitation of data-dependent analyses regardless of fragmentation technique is the relatively low run-to-run reproducibility because of the stochastic nature of peak picking, which may pick a low abundance peak in one analysis and miss it in the next one. In addition, one rarely obtains multiple measurements across the entire peak. This makes detection of those low abundance peaks sporadic and error-prone for quantification. Targeted analyses such as multiple reaction monitoring (MRM)
25-27
achieve much higher reproducibility because the proteotypic peptides are analyzed in every analysis. MRM is typically done on a low-resolution triple quadrupole or QTRAP instrument, which limits these applications to peptides smaller than 2500 Da. SWATHTM and all-ion-fragmentation (AIF) are similar concepts but instead of targeting specific proteotypic peptides, they aim to fragment all ions in a given retention time and/or precursor window or swath.28-30 Data interpretation of the resulting mixed peptides ACS Paragon Plus Environment
3
Analytical Chemistry
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 4 of 21
is intensive and a matter of continuous further development.31 In 2012, Coon proposed to use parallel reaction monitoring (PRM) on a Q-Exactive to accurately quantify tryptic peptides by measuring fragment ion intensities in high-resolution MS/MS spectra.32 Here, we explore using PRM of highly charged peptides together with ETD fragmentation on the Orbitrap Fusion, a quadrupole-equipped Orbitrap mass analyzer with a front-end ETD option. In order to establish this workflow, several key innovations were necessary. a) optimization of parameters on the Orbitrap Fusion, b) correct deconvolution of the highly charged MS and MS/MS peptide ions, c) assignment of peaks to ETD fragmented ions for data interpretation, d) quantitation and e) integration of this workflow into a bioinformatics pipeline. We show that a minimum set of 254 proteoforms is present in the untreated and treated H3 fractions. Additionally, we observe global acetylation changes upon DMSO treatment. Furthermore, we developed a fragment ion map to visualize specific differences between treated and untreated samples. Taken together the data presented here show that middle-down proteomics with electron transfer dissociation using PRM is a novel, attractive method for the effective analysis and quantification of large and highly modified peptides.
EXPERIMENTAL SECTION Sample Preparation N-terminal
H3
mouse
standard
peptide
ARTKQTARKSTGGKAPRKQLATKAARKSAPATGGVKKPHRYRPGTVALRE
Acetylwas
purchased
from JPT (Berlin, Germany). H3 histones from MEL cells were digested with Glu-C. Detailed description of the sample preparation is available in the Supporting Information. Please note that the Nterminal tails of H3, H3.1 and H3.2 are identical. For simplicity we refer to them as H3. Optimization Studies using Direct Infusion of Synthetic H3 Histone Standards H3 Histone standard was directly infused into the Orbitrap Fusion mass spectrometer with the following variable parameters. ETD reaction times of 5, 10, 50, 100 and 200 ms were tested with or without ACS Paragon Plus Environment
4
Page 5 of 21
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
supplemental CID or HCD activation. Normalized collision energies of 10%, 20%, 30% and 40% for the CID and HCD supplemental activation were used. Nanoflow Liquid Chromatography Parallel Reaction Monitoring Electron Transfer Dissociation Tandem Mass Spectrometry (NanoLC-PRM-ETD -MS/MS) All other experiments were performed on a nanoflow liquid chromatography system, EASY-nLC coupled to an Orbitrap Fusion mass spectrometer, equipped with a nanoelectrospray ion source (all Thermo Fisher Scientific). For the EASY-nLC II system, solvent A consisted of 70% ACN, 30% water, and 20 mM propionic acid, pH 6, and solvent B was 25% ACN and 75% water, pH 2.5. For the LC-PRM-ETD-MS/MS experiments 1.0 µg (5 µL) of fractionated H3 histone peptides were directly loaded at a flow rate of 500 nl/min onto a 16 cm analytical WCX-HILIC column (75 µm ID) packed in-house with PolyCat A resin (3 µm, 1500 Å pore size, PolyLC). The column was enclosed in a column heater operating at 50°C. After 30 min of loading time, the peptides were separated with a 146 min gradient at a flow rate of 350 nL/min. The gradient was as follows: 0%-50% Solvent B (10 min), 50%-80% B (120 min), 80-100% B (1 min) and 100% B (15 min). The Orbitrap Fusion was operated in PRM acquisition mode to automatically alternate between a MS1 scan (m/z range 350-1000) and subsequent ETD MS/MS PRM scans for the +10 ions of 20 methyl equivalents of the H3 N-terminal tail (i.e.; m/z = 535.11, 536.52, 537.92, 539.32, 540.72, 542.12, 543.52, 544.92, 546.32, 547.72, 549.12, 550.53, 551.93, 553.33, 554.73, 556.13, 557.53, 558.93, 560.33, 561.73). MS2 scans were acquired with an m/z range of 150-2000. Both the precursor and product ions were analyzed in the Orbitrap analyzer. Full MS spectra were acquired with a resolution of 120,000 at m/z 200 with an automatic gain control (AGC) target value of 2 × 105 and maximum ion injection time of 100 ms. Ions for MS/MS fragmentation were isolated in the quadrupole with an isolation window of 1.4 m/z and analyzed in the Orbitrap analyzer. The AGC target value was set at 2 × 105 and the maximum ion injection time at 25 ms. The maximum ETD reaction injection time was 10 ms at an ETD reagent target of 200,000 ions. Supplemental collision energy of 40% with EThcD was used. Each ETD spectrum comprised 3 microscans, resulting in a duty cycle of 10 sec. ETD mass spectra were acquired with a resolution of ACS Paragon Plus Environment
5
Analytical Chemistry
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 21
30,000 at m/z 200. Conditions tested in preliminary analysis are summarized in Supporting Information (SupplTable S1). MS and MS/MS spectra were acquired in profile mode. Data acquisition was controlled by Xcalibur 3.0.63 and Tune Plus 1.1.982 software (Thermo Fisher Scientific). Data Analysis All mass spectrometric raw data files were converted to mzxml files using MSConvert (ProteoWizard v. 3.0.4006).33 For PRM spectra, in-house scripts were used to identify chromatographic peak boundaries for the different number of acetylated and methylated proteoforms. In these scripts, the XICs and mass error of each of the 7 most abundant isotopes of each 20 methyl-equivalent proteoforms were extracted from the mzxml file. The chromatographic peak boundaries for each combination of 0 to 5 acetylations and 0 to 7 methylations up to 20 methyl-equivalents were identified by the longest duration where the median mass error across the isotopes was less than 5 ppm 80% of the time. Since the difference between 1 acetylation and 3 methylations was greater than 7 ppm at the precursor mass and the typical mass error of the instrument was less than 2 ppm, we were able to identify which chromatographic peak belonged to each proteoform based on mass error alone. For fragment ion analysis, MS2 spectra related to each proteoform were summed and centroided within the calculated peak boundaries. All MS2 spectra (both from direct injection and PRM) were deconvoluted with MS-Deconv.34 These deconvoluted spectra were then annotated using MS-Product in ProteinProspector.35 For heatmap generation, YADA36 was used to deconvolute MS1 spectra. For MS1-based quantitation, Skyline 37 was used to extract areas under the extracted ion chromatograms (XICs) of the top seven most abundant isotopes of each of the modified precursors using the calculated peak boundaries. Peak areas where then normalized to the sum of all H3 N-terminal peptide peak areas. Statistical significance was assessed using the Student’s t-test.
RESULTS AND DISCUSSION Optimization of parameters on the Orbitrap Fusion
ACS Paragon Plus Environment
6
Page 7 of 21
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
To optimize parameters on the Orbitrap Fusion for middle-down ECD, the synthetic N-terminal histone H3
peptide
fragment
Acetyl-
ARTKQTARKSTGGKAPRKQLATKAARKSAPATGGVKKPHRYRPGTVALRE was infused at a nominal concentration of 2 pmol/µL. First, we compared the +7, +8, +9, +10 charge states and found that the most intense +10 ion generally resulted in more product ions. Thus, we targeted the +10 ion at m/z = 535.11 for this analysis. We next explored whether better fragmentation is observed with standard or intact protein pressure settings because polypeptides with 50 amino acids could conceivably fall in either a small protein or a large peptide category. When we enumerated the number of fragment ions that we observed for standard and intact protein pressure settings, standard pressure settings gave superior results in all cases. Next, to determine what ETD reaction time would be ideal, we tested reaction times from 5 ms to 200 ms. We found that shorter ETD reaction times (10-20 ms) gave better results than longer reaction times (200 ms) or ultrashort reaction times of 5 ms. Simultaneously, we tested whether to apply supplemental CID or HCD activation and if so at what energy level. We found best fragmentation to occur with 20 ms ETD reaction time with supplemental CID activation collision energy of 20% and 10 ms ETD reaction time with HCD activation collision energy of 40%. The optimization of ETD fragmentation is important in order to achieve high sequence coverage and to precisely localize potential PTMs. Conversion of raw files and deconvolution of the highly charged MS/MS peptide ions To further process the acquired data, raw files need to be converted into mzxml format. After conversion to mzxml, MS-Deconv was used to deconvolute the MS/MS spectra.34 We compared both ReAdW and MSConvert38 in combination with MS-Deconv and found no significant difference. In all following experiments we used MSConvert in combination with MS-Deconv. Assignment of peaks to fragment ions for data interpretation and quantitation
ACS Paragon Plus Environment
7
Analytical Chemistry
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 8 of 21
Figure 1 shows an example of the annotated spectrum of the synthetic N-terminal histone H3 peptide standard after conversion of the raw file with MSConvert and deconvolution with MS-Deconv. As can be seen, almost complete sequence coverage was obtained. Supplemental HCD activation produced some b- and y-ions that further contributed to the overall sequence coverage. High sequence coverage is a prerequisite for successful PTM characterization. [[insert Figure 1 around here]]
PRM method After we established what fragmentation criteria were best suited for characterization of multiply charged peptides, we next setup a PRM method (Fig. 2A). The principal advantage of a targeted approach such as PRM is illustrated in Fig. 2B. A DDA method would stochastically choose the most abundant ions in a given m/z range. This is particularly disadvantageous in a middle-down approach where the most abundant species would be present in multiple charge states and resampled multiple times. All other proteoforms would remain undersampled. In contrast, the PRM method would be targeted to only one (predetermined) charge state and allow an in-depth analysis of all targeted species present in the sample. To evaluate the PRM strategy for middle-down analysis, we applied it to the fraction of H3 histones from MEL cells. Upon treatment with DMSO, MEL cells are known to produce adult hemoglobin, which served as a model for differentiation39.
[[insert Figure 2 around here]]
We previously determined that a GluC-digestion produces suitable middle-down peptides.4,13 For this, histones were extracted from MEL cells, fractionated by HPLC and the H3 fraction was collected prior to GluC digestion. To improve the analysis of the possible theoretical proteoforms, we adopted the weak cation exchange hydrophilic interaction chromatography (WCX-HILIC) separation introduced by ACS Paragon Plus Environment
8
Page 9 of 21
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
Young, which separates the H3 N-termini primarily by acetylation and, to a secondary degree, by methylation status.4,23,40 A resulting heat map of the deconvoluted data is displayed in Figure 3A showing all methyl equivalents separated on the x axis (mass) and acetylations being separated on the y-axis (retention time). An added advantage of this separation strategy is that it simplifies the data analysis. The peak corresponding to the peptide with two acetylations and three methylations can be separated from the peak with the same nominal mass with one acetylation and five methylations. This limits the theoretical possible assignments and greatly reduces search time, while improving assignment accuracy. [[insert Figure 3 A and B around here]] Repeatability and Reproducibility Targeted methods have the general advantage that they have better repeatability than untargeted, datadependent approaches25-30. To determine whether our middle-down ETD approach is repeatable, we plotted two technical replicates of the dimethylated H3 N-terminal peptide in a butterfly plot shown in Figure 3B. All fragment ions shown in red had an equivalent measurement in the two technical replicates within 10 ppm; the few ions that were not matching are shown in black. As can be seen the fragment ions are highly repeatable with Pearson’s correlations of intensities between the technical replicates of 0.88. In addition to the repeatability, we assessed reproducibility by analyzing intensity of representative fragment ions of the standard H3 peptide in 5 replicates. As shown in Figure 3C, the fragment ion intensity was highly reproducible with a CV between 2.2% and 8.2%. Annotating Fragment Ions and Enumerating Proteoforms To determine whether we could assign the peaks to ETD fragment ions of multiple proteoforms, we selected the peak in the heat map that corresponded to 0 acetylations and 2 methylations (i.e. 5366 Da) of H3, summed the spectra up and deconvoluted them before annotation. Figure 4A shows the complete ACS Paragon Plus Environment
9
Analytical Chemistry
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 10 of 21
annotation of the di-methylation H3 spectrum. By expanding the 1300-1400 m/z region, we see evidence for multiple proteoforms, represented by mono- and dimethylated z11 ions and unmodified and mono-methylated c13 ions. From this data, we can conclude based on the fragment ion intensities that an equal number of di-methylated H3 N-terminal tails have one or two methylations in the 11 C-terminal residues and twice as many unmodified as mono-methylations in the 13 N-terminal residues. In Figure 4B, the minimum set of proteoforms needed to explain all fragment ions for the dimethylated H3 Nterminal proteoform are shown. Additionally, Supplemental Table S2 shows a potential minimal set of H3 N-terminal proteoforms that explain all the fragment ions observed for all combinations of methylation and acetylation from the untreated sample shown in Figure 3A. Supplemental Table S3 shows the equivalent proteoforms for the DMSO-treated sample. These results indicate that our PRM middle-down strategy with ETD enables a thorough characterization of histone peptides and the localization of their PTMs. Please note that the combinatorial presence of multiple PTMs is largely preserved in this approach. Such information would be readily lost in a typical bottom-up approach. It is worth noting that additional modified forms may be present since this is a minimal set.9 Should a higher degree of proteoform characterization be needed, an additional fractionation step may be beneficial. [[insert Figure 4 around here]] Quantitation A specific advantage of our PRM approach is the fact that we monitor both, MS1 and defined MS2 data. For a bird’s eye view we can use the MS1 data to visualize global changes. As an example, we show in Figure 5 how acetylation patterns vary between three technical replicates each of untreated and DMSO treated H3 N-terminal tails. We observed a dynamic pattern: DMSO treated histones showed more 0 and 1 acetylated forms and fewer 2, 3, 4 and 5 acetylated forms with an average number of acetylations dropping from 1.32 in the untreated sample to 1.02 in the DMSO treated sample (t-test p-value of 0.02). [[insert Figure 5 around here]]
ACS Paragon Plus Environment
10
Page 11 of 21
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
Displaying multiple proteoforms A specific challenge associated with histones is the display of its multiple proteoforms in a humanly interpretable fashion. As shown in Figure 4, multiple proteoforms are present in the di-methylated H3 spectrum. To this end, we created a differential fragment ion map that indicates the methylation states of each fragment ion identified for the di-methylated H3 N-terminal peptide of both untreated and treated samples (Figure 6 A and B). This representation of the fragment ions observed is similar to a sequence cleavage figure, but allows us to visualize which fragment ions were observed in which samples and how many methylations were present on each fragment ion. For instance, from the c-ion series we can deduce that a portion of the H3 proteoforms detected with two methylations did not have a modification at the N-terminal ARTKQTARK because c- and b-fragment ions of the dimethylated precursor ions were observed without any modifications on these residues. At the same time, a portion of the H3 proteoforms with two methylations must be dimethylated at or before K9, since this is the first fragment ion observed for this modification. The z- and y-ions confirm this interpretation and reveal that an additional monomethylated form at K9 must have been present. This interpretation is in line with the minimum set of proteoforms needed to explain all observed fragment ions for the dimethylated precursor ion shown in Figure 4B.
[[insert Figure 6 around here]]
While great advances have been made in middle-down proteomics, big challenges still remain, particularly when it comes to highly modified proteins such as histones, where high sequence coverage, preservation of the combinatorial PTMs and high run-to-run reproducibility is required. All previous middle-down approaches have used DDA modes. Here, we have explored to study PRM of ETDfragmented large peptides using a robust ETD ion source and the front end quadrupole for precursor ion ACS Paragon Plus Environment
11
Analytical Chemistry
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 12 of 21
selection. The high mass accuracy achieved in the Orbitrap enabled the correct assignment of the deconvoluted peaks. After optimization of the Orbitrap parameters, we have developed an analysis strategy that allowed us to deconvolute the highly charged MS and MS/MS peptide ions. By design, the PRM method produced hybrid or chimeric spectra of isobaric peptides that require extensive characterization. We were able to assign the resulting peaks to ETD fragment ions and interpreted a minimum set of 254 proteoforms for the H3 N-terminus. This interpretation was aided by the use of WCX-HILIC, which allowed us to reduce the possible number of assignments based on the specific methylation and acetylation status of the proteoforms. Additionally, our method can take a bird’s eye view and quantify changes at the MS1 level. As an example, we showed acetylation changes upon DMSO treatment. In particular, a relative increase in 0 and 1 acetylations and decrease in 2-4 acetylations in the DMSO-treated MEL cells was observed. These may be responsible for the induction of adult hemoglobin after DMSO treatment. Similar to SWATH and AIF experiments28-30, the MS1 level could also be used as an archive to check later whether a hitherto unknown PTM could also have been present in our sample. The flexibility of our approach also allows us to check individual changes within each proteoform. Finally, we show that the run-to-run reproducibility is high. CONCLUSIONS We have successfully established a PRM workflow for middle-down ETD for the analysis of peptides larger than 5 kDa for the first time. More specifically, we used histone H3 N-terminal fragments for this analysis. We demonstrate that PRM combined with middle-down ETD is a superior method for the reproducible quantitation of highly modified peptides. We obtained high peptide sequence coverage and identified combinatorial PTMs on 254 histone H3 N-terminal fragments. While we have shown the proof of principle on the histone H3 N-terminal tail one can easily envision expanding this analysis to all histones and other proteins. Taken together the data presented here show that PRM middle-down nano-LC-ETD MS/MS is a novel, attractive method for the effective analysis and quantification of large and highly modified peptides.
ACS Paragon Plus Environment
12
Page 13 of 21
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
SUPPORTING INFORMATION Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org ACKNOWLEDGMENTS The Proteome Exploration Laboratory is supported by the Gordon and Betty Moore Foundation through grant GBMF775 and the Beckman Institute and the HHMI (Orbitrap Fusion instrument). CF was funded by Fulbright, PROLAB (ASBMB) through a Wood-Whelan fellowship from IUBMB and from Fundação para a Ciência e Tecnologia, Portugal, with both a project grant (PTDC/MARBIO/2174/2012) and postdoctoral fellowship to CF (SFRH/BPD/79271/2011). The authors thank Dr. Andrew Alpert of PolyLC Inc. for technical advice. Conflict of Interest Statement There is no conflict of interest. REFERENCES (1) Srzentic, K.; Fornelli, L.; Laskay, U. A.; Monod, M.; Beck, A.; Ayoub, D.; Tsybin, Y. O. Analytical Chemistry 2014, 86, 9945-9953. (2) Sidoli, S.; Schwammle, V.; Ruminowicz, C.; Hansen, T. A.; Wu, X.; Helin, K.; Jensen, O. N. Proteomics 2014, 14, 2200-2211. (3) Valkevich, E. M.; Sanchez, N. A.; Ge, Y.; Strieter, E. R. Biochemistry 2014, 53, 4979-4989. (4) Moradian, A.; Kalli, A.; Sweredoski, M. J.; Hess, S. Proteomics 2014, 14, 489-497. (5) Cannon, J. R.; Edwards, N. J.; Fenselau, C. Journal of Mass Spectrometry : JMS 2013, 48, 340-343. (6) Wu, C.; Tran, J. C.; Zamdborg, L.; Durbin, K. R.; Li, M.; Ahlf, D. R.; Early, B. P.; Thomas, P. M.; Sweedler, J. V.; Kelleher, N. L. Nature Methods 2012, 9, 822-824. (7) Cannon, J.; Lohnes, K.; Wynne, C.; Wang, Y.; Edwards, N.; Fenselau, C. Journal of Proteome Research 2010, 9, 3886-3890. (8) Plazas-Mayorca, M. D.; Bloom, J. S.; Zeissler, U.; Leroy, G.; Young, N. L.; DiMaggio, P. A.; Krugylak, L.; Schneider, R.; Garcia, B. A. Molecular bioSystems 2010, 6, 1719-1729. (9) Arnaudo, A. M.; Molden, R. C.; Garcia, B. A. Crit. Rev. Biochem. Mol. 2011, 46, 284-294. (10) Garcia, B. A.; Mollah, S.; Ueberheide, B. M.; Busby, S. A.; Muratore, T. L.; Shabanowitz, J.; Hunt, D. F. Nature Protoc. 2007, 2, 933-938. (11) Boyne, M. T.; Garcia, B. A.; Li, M. X.; Zamdborg, L.; Wenger, C. D.; Babai, S.; Kelleher, N. L. J. Proteome Res. 2009, 8, 374-379. (12) Xu, P.; Peng, J. Analytical Chemistry 2008, 80, 3438-3444. (13) Kalli, A.; Sweredoski, M. J.; Hess, S. Analytical Chemistry 2013, 85, 3501-3507. ACS Paragon Plus Environment
13
Analytical Chemistry
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 14 of 21
(14) Windwarder, M.; Altmann, F. Journal of Proteomics 2014, 108, 258-268. (15) Fornelli, L.; Ayoub, D.; Aizikov, K.; Beck, A.; Tsybin, Y. O. Analytical Chemistry 2014, 86, 30053012. (16) Nardelli, S. C.; Che, F. Y.; Silmon de Monerri, N. C.; Xiao, H.; Nieves, E.; Madrid-Aliste, C.; Angel, S. O.; Sullivan, W. J., Jr.; Angeletti, R. H.; Kim, K.; Weiss, L. M. mBio 2013, 4, e00922-00913. (17) Mikesh, L. M.; Ueberheide, B.; Chi, A.; Coon, J. J.; Syka, J. E. P.; Shabanowitz, J.; Hunt, D. F. Biochim. Biophys. Acta 2006, 1764, 1811-1822. (18) Zubarev, R. A.; Zubarev, A. R.; Savitski, M. M. Journal of the American Society for Mass Spectrometry 2008, 19, 753-761. (19) Bonenfant, D.; Coulot, M.; Towbin, H.; Schindler, P.; van Oostrum, J. Molecular & Cellular Proteomics : MCP 2006, 5, 541-552. (20) Nicklay, J. J.; Shechter, D.; Chitta, R. K.; Garcia, B. A.; Shabanowitz, J.; Allis, C. D.; Hunt, D. F. The Journal of Biological Chemistry 2009, 284, 1075-1085. (21) Garcia, B. A.; Siuti, N.; Thomas, C. E.; Mizzen, C. A.; Kelleher, N. L. Int. J. Mass Spectrom. 2007, 259, 184-196. (22) Garcia, B. A.; Thomas, C. E.; Kelleher, N. L.; Mizzen, C. A. J. Proteome Res. 2008, 7, 4225-4236. (23) Jung, H. R.; Sidoli, S.; Haldbo, S.; Sprenger, R. R.; Schwammle, V.; Pasini, D.; Helin, K.; Jensen, O. N. Analytical Chemistry 2013, 85, 8232-8239. (24) Sidoli, S.; Lin, S.; Karch, K. R.; Garcia, B. A. Analytical Chemistry 2015, 87, 3129-3133. (25) Ebhardt, H. A.; Sabido, E.; Huttenhain, R.; Collins, B.; Aebersold, R. Proteomics 2012, 12, 11851193. (26) Maiolica, A.; Junger, M. A.; Ezkurdia, I.; Aebersold, R. Journal of Proteomics 2012, 75, 34953513. (27) Picotti, P.; Aebersold, R. Nature methods 2012, 9, 555-566. (28) Gillet, L. C.; Navarro, P.; Tate, S.; Rost, H.; Selevsek, N.; Reiter, L.; Bonner, R.; Aebersold, R. Molecular & cellular proteomics : MCP 2012, 11, O111 016717. (29) Geiger, T.; Cox, J.; Mann, M. Molecular & Cellular Proteomics : MCP 2010, 9, 2252-2261. (30) Venable, J. D.; Dong, M. Q.; Wohlschlegel, J.; Dillin, A.; Yates, J. R. Nature Methods 2004, 1, 3945. (31) Keller, A.; Bader, S. L.; Shteynberg, D.; Hood, L.; Moritz, R. L. Molecular & Cellular Proteomics : MCP 2015, 14, 1411-1418. (32) Peterson, A. C.; Russell, J. D.; Bailey, D. J.; Westphall, M. S.; Coon, J. J. Mol. Cell. Proteomics 2012, 11, 1475-1488. (33) Kessner, D.; Chambers, M.; Burke, R.; Agus, D.; Mallick, P. Bioinformatics 2008, 24, 2534-2536. (34) Liu, X.; Inbar, Y.; Dorrestein, P. C.; Wynne, C.; Edwards, N.; Souda, P.; Whitelegge, J. P.; Bafna, V.; Pevzner, P. A. Mol. Cell. Proteomics 2010, 9, 2772-2782. (35) Chalkley, R. J.; Baker, P. R.; Huang, L.; Hansen, K. C.; Allen, N. P.; Rexach, M.; Burlingame, A. L. Molecular & Cellular Proteomics : MCP 2005, 4, 1194-1204. (36) Carvalho, P. C.; Xu, T.; Han, X.; Cociorva, D.; Barbosa, V. C.; Yates, J. R., 3rd. Bioinformatics 2009, 25, 2734-2736. (37) MacLean, B.; Tomazela, D. M.; Shulman, N.; Chambers, M.; Finney, G. L.; Frewen, B.; Kern, R.; Tabb, D. L.; Liebler, D. C.; MacCoss, M. J. Bioinformatics 2010, 26, 966-968. (38) French, W. R.; Zimmerman, L. J.; Schilling, B.; Gibson, B. W.; Miller, C. A.; Townsend, R. R.; Sherrod, S. D.; Goodwin, C. R.; McLean, J. A.; Tabb, D. L. Journal of proteome research 2015, 14, 1299-1307. (39) Friend, C. The Journal of Experimental Medicine 1957, 105, 307-318. (40) Young, N. L.; DiMaggio, P. A.; Garcia, B. A. Cell. Mol. Life Sci. 2010, 67, 3983-4000.
ACS Paragon Plus Environment
14
Page 15 of 21
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
FIGURES
Figure 1: MS/MS spectrum of the synthetic N-terminal histone H3 peptide standard AcetylARTKQTARKSTGGKAPRKQLATKAARKSAPATGGVKKPHRYRPGTVALRE
with
assigned
fragment ions. Note that the spectrum is split in m/z range from (top) 400-3000 and (bottom) 3000-5500 to better display the fragment ion assignments.
ACS Paragon Plus Environment
15
Analytical Chemistry
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 16 of 21
Figure 2: (A) PRM method for 20 methyl equivalents of the +10 charge of the N-terminal histone H3 fragment. (B) Illustration of the principal differences between a DDA and PRM method (* indicate ions selected for MS/MS based on intensity and ▼ indicate targeted ions by PRM). Especially, in middledown approaches multiple charge states of the same proteoform exist leading to multiple sampling of the same proteoforms rather than an in-depth analysis of all species present. In contrast, a PRM approach targets only the (predetermined) ions of interest, with much greater depths. In addition, when the targets are very low abundance and the charge state cannot be easily determined, the PRM method still allows for systematic monitoring of the proteoforms of interest.
ACS Paragon Plus Environment
16
Page 17 of 21
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
Figure 3A: Heat map of the retention time vs. deconvoluted mass of the GluC-digested N-terminus of Histone H3 spectra separated by WCX-HILIC. Methylation status is shown on the x-axis and acetylation status is shown on the y-axis. Figure 3B: Butterfly plot of two technical replicates of the dimethylated H3 N-terminal peptide. Ions shown in red are matching between the two technical replicates within 10 ppm; the few ions that were not matching are shown in black. Pearsons’s correlation of intensities between the two replicates was 0.88 using a m/z bin width of 0.05. Figure 3C: Reproducibility of the intensity of representative ETD fragment ions from five replicates (Rep1-5). CV varies between 2.2% and 8.2%.
ACS Paragon Plus Environment
17
Analytical Chemistry
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 18 of 21
Figure 4A: Annotation of the di-methylated H3 spectrum. When expanding the 1300-1400 m/z region, we see evidence of multiple proteoform (i.e., one or two methylations in the 11 C-terminal residues and zero or one methylations in the 13 N-terminal residues). Figure 4B: A minimum set of four proteoforms were needed to explain the fragment ions observed for the di-methylated H3 N-terminal peptide of the untreated sample. Violet squares represent monomethylations and green squares represent dimethylations. Annotations for all other proteoforms are listed in the Supplemental Table S2 (untreated) and S3 (DMSO-treated).
ACS Paragon Plus Environment
18
Page 19 of 21
50%
Percent Sumed XIC
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
45%
Untreated
40%
DMSO
35% 30% 25% 20% 15% 10% 5% 0% 0Ac
1Ac
2Ac
3Ac
4Ac
Figure 5: Quantitation on the MS1 level. DMSO treatment leads to a slight relative increase in 0 and 1 acetylations and a decrease in 2, 3, 4 and 5 acetylations across the three replicates. Error bars here represent the standard error. XIC areas were summed across all number of methylations. Summed XICs were normalized to the total of all summed XICs in each replicate. The overall average number of acetylations decreased from 1.32 in the untreated samples to 1.02 in the DMSO treated sample with a ttest p-value of 0.02.
ACS Paragon Plus Environment
19
Analytical Chemistry
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 20 of 21
Figure 6: Fragment ion map for the di-methylated H3 N-terminal peptide for the untreated (A, top) and DMSO-treated (B, bottom) sample. Gray squares represent fragment ions observed. Figures 6 A and B also display the differences for the di-methylated H3 N-terminal peptide between the (A) untreated and (B) treated sample.
ACS Paragon Plus Environment
20
Page 21 of 21
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
For TOC only
ACS Paragon Plus Environment
21
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
High Resolution Parallel Reaction Monitoring with Electron Transfer Dissociation for MiddleDown Proteomics
Michael J. Sweredoski1, Annie Moradian1, Matthias Raedle1,2, Catarina Franco1,3 and Sonja Hess1,* 1
Proteome Exploration Laboratory, Division of Biology, Beckman Institute, California Institute of Technology, Pasadena, CA 91125, USA
2
Hochschule Weihenstephan-Triesdorf, University of Applied Sciences, Faculty of Biotechnology and Bioinformatic, Am Hofgarten 4, 85354 Freising, Germany
3
Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, Av. da República, 2780157 Oeiras, Portugal
[email protected] RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required according to the journal that you are submitting your paper to) *
Corresponding author: email: [email protected], phone: 626-395-2339, fax: 626-449-4159
ACS Paragon Plus Environment
1
Analytical Chemistry
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 2 of 21
ABSTRACT
In recent years, middle-down proteomics has emerged as a popular technique for the characterization and quantification of proteins not readily amenable to typical bottom-up approaches. So far, all high resolution middle-down approaches are done in data-dependent acquisition mode, using both, collisioninduced dissociation or electron capture/transfer dissociation techniques.
Here, we explore middle-down proteomics with electron transfer dissociation using a targeted acquisition mode, parallel reaction monitoring (PRM), on an Orbitrap Fusion. As an example of a highly modified protein, we used histone H3 fractions from untreated and DMSO-treated Murine ErythroLeukemia (MEL) cells. We first determined optimized instrument parameters to obtain high sequence coverage using a synthetic standard peptide. We then setup a combined method of both MS1 scans and PRM scans of the twenty most abundant combinations of methylation and acetylation of the +10 charge state of the N-terminal tail of H3. Weak cation exchange hydrophilic interaction chromatography was used to separate the N-terminal H3 tail primarily by its acetylation, and to a secondary degree, by its methylation status, which aided in the interpretation of the results. After deconvolution of the highly charged ions, peaks were annotated to a minimum set of 254 H3 proteoforms in the untreated and treated samples. Upon DMSO treatment global quantitation changes from the MS1 level show a relative decrease of 2, 3, 4, and 5 acetylations and an increase of 0 and 1 acetylations. A fragment ion map was developed to visualize specific differences between treated and untreated samples. Taken together the data presented here show that middle-down proteomics with electron transfer dissociation using PRM is a novel, attractive method for the effective analysis and quantification of large and highly modified peptides.
ACS Paragon Plus Environment
2
Page 3 of 21
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
INTRODUCTION
In recent years, middle-down proteomics has emerged as a popular technique for the characterization and quantification of proteins not readily amenable to typical bottom-up approaches such as antibodies, ubiquitin chains and histones.1-8 Thus, middle-down approaches excel where high protein sequence coverage is needed and a typical tryptic digestion would yield peptides too small for a successful LCMS/MS analysis preventing the full characterization of a given protein.9,10 Nevertheless, these approaches cannot yet be considered routine applications since the bioinformatics pipelines developed for bottom-up or top-down analyses do not readily apply for middle-down approaches. Middle-down approaches have been used with collision-induced dissociation (CID)7,11,12 and electron capture/transfer dissociation (ECD/ETD) techniques.4,13-16 Since middle-down approaches produce predominantly higher charged species, ECD/ETD techniques generally yield more extensive fragmentation across the peptide when compared to CID. As a non-ergodic process, ECD/ETD fragmentation is also favorable for the characterization of post-translationally modified peptides.17-20 Middle-down data-dependent acquisition (DDA) approaches for histones have been successfully used in combination with ECD and ETD.4,13,21-24
However, a limitation of data-dependent analyses regardless of fragmentation technique is the relatively low run-to-run reproducibility because of the stochastic nature of peak picking, which may pick a low abundance peak in one analysis and miss it in the next one. In addition, one rarely obtains multiple measurements across the entire peak. This makes detection of those low abundance peaks sporadic and error-prone for quantification. Targeted analyses such as multiple reaction monitoring (MRM)
25-27
achieve much higher reproducibility because the proteotypic peptides are analyzed in every analysis. MRM is typically done on a low-resolution triple quadrupole or QTRAP instrument, which limits these applications to peptides smaller than 2500 Da. SWATHTM and all-ion-fragmentation (AIF) are similar concepts but instead of targeting specific proteotypic peptides, they aim to fragment all ions in a given retention time and/or precursor window or swath.28-30 Data interpretation of the resulting mixed peptides ACS Paragon Plus Environment
3
Analytical Chemistry
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 4 of 21
is intensive and a matter of continuous further development.31 In 2012, Coon proposed to use parallel reaction monitoring (PRM) on a Q-Exactive to accurately quantify tryptic peptides by measuring fragment ion intensities in high-resolution MS/MS spectra.32 Here, we explore using PRM of highly charged peptides together with ETD fragmentation on the Orbitrap Fusion, a quadrupole-equipped Orbitrap mass analyzer with a front-end ETD option. In order to establish this workflow, several key innovations were necessary. a) optimization of parameters on the Orbitrap Fusion, b) correct deconvolution of the highly charged MS and MS/MS peptide ions, c) assignment of peaks to ETD fragmented ions for data interpretation, d) quantitation and e) integration of this workflow into a bioinformatics pipeline. We show that a minimum set of 254 proteoforms is present in the untreated and treated H3 fractions. Additionally, we observe global acetylation changes upon DMSO treatment. Furthermore, we developed a fragment ion map to visualize specific differences between treated and untreated samples. Taken together the data presented here show that middle-down proteomics with electron transfer dissociation using PRM is a novel, attractive method for the effective analysis and quantification of large and highly modified peptides.
EXPERIMENTAL SECTION Sample Preparation N-terminal
H3
mouse
standard
peptide
ARTKQTARKSTGGKAPRKQLATKAARKSAPATGGVKKPHRYRPGTVALRE
Acetylwas
purchased
from JPT (Berlin, Germany). H3 histones from MEL cells were digested with Glu-C. Detailed description of the sample preparation is available in the Supporting Information. Please note that the Nterminal tails of H3, H3.1 and H3.2 are identical. For simplicity we refer to them as H3. Optimization Studies using Direct Infusion of Synthetic H3 Histone Standards H3 Histone standard was directly infused into the Orbitrap Fusion mass spectrometer with the following variable parameters. ETD reaction times of 5, 10, 50, 100 and 200 ms were tested with or without ACS Paragon Plus Environment
4
Page 5 of 21
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
supplemental CID or HCD activation. Normalized collision energies of 10%, 20%, 30% and 40% for the CID and HCD supplemental activation were used. Nanoflow Liquid Chromatography Parallel Reaction Monitoring Electron Transfer Dissociation Tandem Mass Spectrometry (NanoLC-PRM-ETD -MS/MS) All other experiments were performed on a nanoflow liquid chromatography system, EASY-nLC coupled to an Orbitrap Fusion mass spectrometer, equipped with a nanoelectrospray ion source (all Thermo Fisher Scientific). For the EASY-nLC II system, solvent A consisted of 70% ACN, 30% water, and 20 mM propionic acid, pH 6, and solvent B was 25% ACN and 75% water, pH 2.5. For the LC-PRM-ETD-MS/MS experiments 1.0 µg (5 µL) of fractionated H3 histone peptides were directly loaded at a flow rate of 500 nl/min onto a 16 cm analytical WCX-HILIC column (75 µm ID) packed in-house with PolyCat A resin (3 µm, 1500 Å pore size, PolyLC). The column was enclosed in a column heater operating at 50°C. After 30 min of loading time, the peptides were separated with a 146 min gradient at a flow rate of 350 nL/min. The gradient was as follows: 0%-50% Solvent B (10 min), 50%-80% B (120 min), 80-100% B (1 min) and 100% B (15 min). The Orbitrap Fusion was operated in PRM acquisition mode to automatically alternate between a MS1 scan (m/z range 350-1000) and subsequent ETD MS/MS PRM scans for the +10 ions of 20 methyl equivalents of the H3 N-terminal tail (i.e.; m/z = 535.11, 536.52, 537.92, 539.32, 540.72, 542.12, 543.52, 544.92, 546.32, 547.72, 549.12, 550.53, 551.93, 553.33, 554.73, 556.13, 557.53, 558.93, 560.33, 561.73). MS2 scans were acquired with an m/z range of 150-2000. Both the precursor and product ions were analyzed in the Orbitrap analyzer. Full MS spectra were acquired with a resolution of 120,000 at m/z 200 with an automatic gain control (AGC) target value of 2 × 105 and maximum ion injection time of 100 ms. Ions for MS/MS fragmentation were isolated in the quadrupole with an isolation window of 1.4 m/z and analyzed in the Orbitrap analyzer. The AGC target value was set at 2 × 105 and the maximum ion injection time at 25 ms. The maximum ETD reaction injection time was 10 ms at an ETD reagent target of 200,000 ions. Supplemental collision energy of 40% with EThcD was used. Each ETD spectrum comprised 3 microscans, resulting in a duty cycle of 10 sec. ETD mass spectra were acquired with a resolution of ACS Paragon Plus Environment
5
Analytical Chemistry
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 21
30,000 at m/z 200. Conditions tested in preliminary analysis are summarized in Supporting Information (SupplTable S1). MS and MS/MS spectra were acquired in profile mode. Data acquisition was controlled by Xcalibur 3.0.63 and Tune Plus 1.1.982 software (Thermo Fisher Scientific). Data Analysis All mass spectrometric raw data files were converted to mzxml files using MSConvert (ProteoWizard v. 3.0.4006).33 For PRM spectra, in-house scripts were used to identify chromatographic peak boundaries for the different number of acetylated and methylated proteoforms. In these scripts, the XICs and mass error of each of the 7 most abundant isotopes of each 20 methyl-equivalent proteoforms were extracted from the mzxml file. The chromatographic peak boundaries for each combination of 0 to 5 acetylations and 0 to 7 methylations up to 20 methyl-equivalents were identified by the longest duration where the median mass error across the isotopes was less than 5 ppm 80% of the time. Since the difference between 1 acetylation and 3 methylations was greater than 7 ppm at the precursor mass and the typical mass error of the instrument was less than 2 ppm, we were able to identify which chromatographic peak belonged to each proteoform based on mass error alone. For fragment ion analysis, MS2 spectra related to each proteoform were summed and centroided within the calculated peak boundaries. All MS2 spectra (both from direct injection and PRM) were deconvoluted with MS-Deconv.34 These deconvoluted spectra were then annotated using MS-Product in ProteinProspector.35 For heatmap generation, YADA36 was used to deconvolute MS1 spectra. For MS1-based quantitation, Skyline 37 was used to extract areas under the extracted ion chromatograms (XICs) of the top seven most abundant isotopes of each of the modified precursors using the calculated peak boundaries. Peak areas where then normalized to the sum of all H3 N-terminal peptide peak areas. Statistical significance was assessed using the Student’s t-test.
RESULTS AND DISCUSSION Optimization of parameters on the Orbitrap Fusion
ACS Paragon Plus Environment
6
Page 7 of 21
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
To optimize parameters on the Orbitrap Fusion for middle-down ECD, the synthetic N-terminal histone H3
peptide
fragment
Acetyl-
ARTKQTARKSTGGKAPRKQLATKAARKSAPATGGVKKPHRYRPGTVALRE was infused at a nominal concentration of 2 pmol/µL. First, we compared the +7, +8, +9, +10 charge states and found that the most intense +10 ion generally resulted in more product ions. Thus, we targeted the +10 ion at m/z = 535.11 for this analysis. We next explored whether better fragmentation is observed with standard or intact protein pressure settings because polypeptides with 50 amino acids could conceivably fall in either a small protein or a large peptide category. When we enumerated the number of fragment ions that we observed for standard and intact protein pressure settings, standard pressure settings gave superior results in all cases. Next, to determine what ETD reaction time would be ideal, we tested reaction times from 5 ms to 200 ms. We found that shorter ETD reaction times (10-20 ms) gave better results than longer reaction times (200 ms) or ultrashort reaction times of 5 ms. Simultaneously, we tested whether to apply supplemental CID or HCD activation and if so at what energy level. We found best fragmentation to occur with 20 ms ETD reaction time with supplemental CID activation collision energy of 20% and 10 ms ETD reaction time with HCD activation collision energy of 40%. The optimization of ETD fragmentation is important in order to achieve high sequence coverage and to precisely localize potential PTMs. Conversion of raw files and deconvolution of the highly charged MS/MS peptide ions To further process the acquired data, raw files need to be converted into mzxml format. After conversion to mzxml, MS-Deconv was used to deconvolute the MS/MS spectra.34 We compared both ReAdW and MSConvert38 in combination with MS-Deconv and found no significant difference. In all following experiments we used MSConvert in combination with MS-Deconv. Assignment of peaks to fragment ions for data interpretation and quantitation
ACS Paragon Plus Environment
7
Analytical Chemistry
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 8 of 21
Figure 1 shows an example of the annotated spectrum of the synthetic N-terminal histone H3 peptide standard after conversion of the raw file with MSConvert and deconvolution with MS-Deconv. As can be seen, almost complete sequence coverage was obtained. Supplemental HCD activation produced some b- and y-ions that further contributed to the overall sequence coverage. High sequence coverage is a prerequisite for successful PTM characterization. [[insert Figure 1 around here]]
PRM method After we established what fragmentation criteria were best suited for characterization of multiply charged peptides, we next setup a PRM method (Fig. 2A). The principal advantage of a targeted approach such as PRM is illustrated in Fig. 2B. A DDA method would stochastically choose the most abundant ions in a given m/z range. This is particularly disadvantageous in a middle-down approach where the most abundant species would be present in multiple charge states and resampled multiple times. All other proteoforms would remain undersampled. In contrast, the PRM method would be targeted to only one (predetermined) charge state and allow an in-depth analysis of all targeted species present in the sample. To evaluate the PRM strategy for middle-down analysis, we applied it to the fraction of H3 histones from MEL cells. Upon treatment with DMSO, MEL cells are known to produce adult hemoglobin, which served as a model for differentiation39.
[[insert Figure 2 around here]]
We previously determined that a GluC-digestion produces suitable middle-down peptides.4,13 For this, histones were extracted from MEL cells, fractionated by HPLC and the H3 fraction was collected prior to GluC digestion. To improve the analysis of the possible theoretical proteoforms, we adopted the weak cation exchange hydrophilic interaction chromatography (WCX-HILIC) separation introduced by ACS Paragon Plus Environment
8
Page 9 of 21
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
Young, which separates the H3 N-termini primarily by acetylation and, to a secondary degree, by methylation status.4,23,40 A resulting heat map of the deconvoluted data is displayed in Figure 3A showing all methyl equivalents separated on the x axis (mass) and acetylations being separated on the y-axis (retention time). An added advantage of this separation strategy is that it simplifies the data analysis. The peak corresponding to the peptide with two acetylations and three methylations can be separated from the peak with the same nominal mass with one acetylation and five methylations. This limits the theoretical possible assignments and greatly reduces search time, while improving assignment accuracy. [[insert Figure 3 A and B around here]] Repeatability and Reproducibility Targeted methods have the general advantage that they have better repeatability than untargeted, datadependent approaches25-30. To determine whether our middle-down ETD approach is repeatable, we plotted two technical replicates of the dimethylated H3 N-terminal peptide in a butterfly plot shown in Figure 3B. All fragment ions shown in red had an equivalent measurement in the two technical replicates within 10 ppm; the few ions that were not matching are shown in black. As can be seen the fragment ions are highly repeatable with Pearson’s correlations of intensities between the technical replicates of 0.88. In addition to the repeatability, we assessed reproducibility by analyzing intensity of representative fragment ions of the standard H3 peptide in 5 replicates. As shown in Figure 3C, the fragment ion intensity was highly reproducible with a CV between 2.2% and 8.2%. Annotating Fragment Ions and Enumerating Proteoforms To determine whether we could assign the peaks to ETD fragment ions of multiple proteoforms, we selected the peak in the heat map that corresponded to 0 acetylations and 2 methylations (i.e. 5366 Da) of H3, summed the spectra up and deconvoluted them before annotation. Figure 4A shows the complete ACS Paragon Plus Environment
9
Analytical Chemistry
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 10 of 21
annotation of the di-methylation H3 spectrum. By expanding the 1300-1400 m/z region, we see evidence for multiple proteoforms, represented by mono- and dimethylated z11 ions and unmodified and mono-methylated c13 ions. From this data, we can conclude based on the fragment ion intensities that an equal number of di-methylated H3 N-terminal tails have one or two methylations in the 11 C-terminal residues and twice as many unmodified as mono-methylations in the 13 N-terminal residues. In Figure 4B, the minimum set of proteoforms needed to explain all fragment ions for the dimethylated H3 Nterminal proteoform are shown. Additionally, Supplemental Table S2 shows a potential minimal set of H3 N-terminal proteoforms that explain all the fragment ions observed for all combinations of methylation and acetylation from the untreated sample shown in Figure 3A. Supplemental Table S3 shows the equivalent proteoforms for the DMSO-treated sample. These results indicate that our PRM middle-down strategy with ETD enables a thorough characterization of histone peptides and the localization of their PTMs. Please note that the combinatorial presence of multiple PTMs is largely preserved in this approach. Such information would be readily lost in a typical bottom-up approach. It is worth noting that additional modified forms may be present since this is a minimal set.9 Should a higher degree of proteoform characterization be needed, an additional fractionation step may be beneficial. [[insert Figure 4 around here]] Quantitation A specific advantage of our PRM approach is the fact that we monitor both, MS1 and defined MS2 data. For a bird’s eye view we can use the MS1 data to visualize global changes. As an example, we show in Figure 5 how acetylation patterns vary between three technical replicates each of untreated and DMSO treated H3 N-terminal tails. We observed a dynamic pattern: DMSO treated histones showed more 0 and 1 acetylated forms and fewer 2, 3, 4 and 5 acetylated forms with an average number of acetylations dropping from 1.32 in the untreated sample to 1.02 in the DMSO treated sample (t-test p-value of 0.02). [[insert Figure 5 around here]]
ACS Paragon Plus Environment
10
Page 11 of 21
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
Displaying multiple proteoforms A specific challenge associated with histones is the display of its multiple proteoforms in a humanly interpretable fashion. As shown in Figure 4, multiple proteoforms are present in the di-methylated H3 spectrum. To this end, we created a differential fragment ion map that indicates the methylation states of each fragment ion identified for the di-methylated H3 N-terminal peptide of both untreated and treated samples (Figure 6 A and B). This representation of the fragment ions observed is similar to a sequence cleavage figure, but allows us to visualize which fragment ions were observed in which samples and how many methylations were present on each fragment ion. For instance, from the c-ion series we can deduce that a portion of the H3 proteoforms detected with two methylations did not have a modification at the N-terminal ARTKQTARK because c- and b-fragment ions of the dimethylated precursor ions were observed without any modifications on these residues. At the same time, a portion of the H3 proteoforms with two methylations must be dimethylated at or before K9, since this is the first fragment ion observed for this modification. The z- and y-ions confirm this interpretation and reveal that an additional monomethylated form at K9 must have been present. This interpretation is in line with the minimum set of proteoforms needed to explain all observed fragment ions for the dimethylated precursor ion shown in Figure 4B.
[[insert Figure 6 around here]]
While great advances have been made in middle-down proteomics, big challenges still remain, particularly when it comes to highly modified proteins such as histones, where high sequence coverage, preservation of the combinatorial PTMs and high run-to-run reproducibility is required. All previous middle-down approaches have used DDA modes. Here, we have explored to study PRM of ETDfragmented large peptides using a robust ETD ion source and the front end quadrupole for precursor ion ACS Paragon Plus Environment
11
Analytical Chemistry
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 12 of 21
selection. The high mass accuracy achieved in the Orbitrap enabled the correct assignment of the deconvoluted peaks. After optimization of the Orbitrap parameters, we have developed an analysis strategy that allowed us to deconvolute the highly charged MS and MS/MS peptide ions. By design, the PRM method produced hybrid or chimeric spectra of isobaric peptides that require extensive characterization. We were able to assign the resulting peaks to ETD fragment ions and interpreted a minimum set of 254 proteoforms for the H3 N-terminus. This interpretation was aided by the use of WCX-HILIC, which allowed us to reduce the possible number of assignments based on the specific methylation and acetylation status of the proteoforms. Additionally, our method can take a bird’s eye view and quantify changes at the MS1 level. As an example, we showed acetylation changes upon DMSO treatment. In particular, a relative increase in 0 and 1 acetylations and decrease in 2-4 acetylations in the DMSO-treated MEL cells was observed. These may be responsible for the induction of adult hemoglobin after DMSO treatment. Similar to SWATH and AIF experiments28-30, the MS1 level could also be used as an archive to check later whether a hitherto unknown PTM could also have been present in our sample. The flexibility of our approach also allows us to check individual changes within each proteoform. Finally, we show that the run-to-run reproducibility is high. CONCLUSIONS We have successfully established a PRM workflow for middle-down ETD for the analysis of peptides larger than 5 kDa for the first time. More specifically, we used histone H3 N-terminal fragments for this analysis. We demonstrate that PRM combined with middle-down ETD is a superior method for the reproducible quantitation of highly modified peptides. We obtained high peptide sequence coverage and identified combinatorial PTMs on 254 histone H3 N-terminal fragments. While we have shown the proof of principle on the histone H3 N-terminal tail one can easily envision expanding this analysis to all histones and other proteins. Taken together the data presented here show that PRM middle-down nano-LC-ETD MS/MS is a novel, attractive method for the effective analysis and quantification of large and highly modified peptides.
ACS Paragon Plus Environment
12
Page 13 of 21
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
SUPPORTING INFORMATION Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org ACKNOWLEDGMENTS The Proteome Exploration Laboratory is supported by the Gordon and Betty Moore Foundation through grant GBMF775 and the Beckman Institute and the HHMI (Orbitrap Fusion instrument). CF was funded by Fulbright, PROLAB (ASBMB) through a Wood-Whelan fellowship from IUBMB and from Fundação para a Ciência e Tecnologia, Portugal, with both a project grant (PTDC/MARBIO/2174/2012) and postdoctoral fellowship to CF (SFRH/BPD/79271/2011). The authors thank Dr. Andrew Alpert of PolyLC Inc. for technical advice. Conflict of Interest Statement There is no conflict of interest. REFERENCES (1) Srzentic, K.; Fornelli, L.; Laskay, U. A.; Monod, M.; Beck, A.; Ayoub, D.; Tsybin, Y. O. Analytical Chemistry 2014, 86, 9945-9953. (2) Sidoli, S.; Schwammle, V.; Ruminowicz, C.; Hansen, T. A.; Wu, X.; Helin, K.; Jensen, O. N. Proteomics 2014, 14, 2200-2211. (3) Valkevich, E. M.; Sanchez, N. A.; Ge, Y.; Strieter, E. R. Biochemistry 2014, 53, 4979-4989. (4) Moradian, A.; Kalli, A.; Sweredoski, M. J.; Hess, S. Proteomics 2014, 14, 489-497. (5) Cannon, J. R.; Edwards, N. J.; Fenselau, C. Journal of Mass Spectrometry : JMS 2013, 48, 340-343. (6) Wu, C.; Tran, J. C.; Zamdborg, L.; Durbin, K. R.; Li, M.; Ahlf, D. R.; Early, B. P.; Thomas, P. M.; Sweedler, J. V.; Kelleher, N. L. Nature Methods 2012, 9, 822-824. (7) Cannon, J.; Lohnes, K.; Wynne, C.; Wang, Y.; Edwards, N.; Fenselau, C. Journal of Proteome Research 2010, 9, 3886-3890. (8) Plazas-Mayorca, M. D.; Bloom, J. S.; Zeissler, U.; Leroy, G.; Young, N. L.; DiMaggio, P. A.; Krugylak, L.; Schneider, R.; Garcia, B. A. Molecular bioSystems 2010, 6, 1719-1729. (9) Arnaudo, A. M.; Molden, R. C.; Garcia, B. A. Crit. Rev. Biochem. Mol. 2011, 46, 284-294. (10) Garcia, B. A.; Mollah, S.; Ueberheide, B. M.; Busby, S. A.; Muratore, T. L.; Shabanowitz, J.; Hunt, D. F. Nature Protoc. 2007, 2, 933-938. (11) Boyne, M. T.; Garcia, B. A.; Li, M. X.; Zamdborg, L.; Wenger, C. D.; Babai, S.; Kelleher, N. L. J. Proteome Res. 2009, 8, 374-379. (12) Xu, P.; Peng, J. Analytical Chemistry 2008, 80, 3438-3444. (13) Kalli, A.; Sweredoski, M. J.; Hess, S. Analytical Chemistry 2013, 85, 3501-3507. ACS Paragon Plus Environment
13
Analytical Chemistry
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 14 of 21
(14) Windwarder, M.; Altmann, F. Journal of Proteomics 2014, 108, 258-268. (15) Fornelli, L.; Ayoub, D.; Aizikov, K.; Beck, A.; Tsybin, Y. O. Analytical Chemistry 2014, 86, 30053012. (16) Nardelli, S. C.; Che, F. Y.; Silmon de Monerri, N. C.; Xiao, H.; Nieves, E.; Madrid-Aliste, C.; Angel, S. O.; Sullivan, W. J., Jr.; Angeletti, R. H.; Kim, K.; Weiss, L. M. mBio 2013, 4, e00922-00913. (17) Mikesh, L. M.; Ueberheide, B.; Chi, A.; Coon, J. J.; Syka, J. E. P.; Shabanowitz, J.; Hunt, D. F. Biochim. Biophys. Acta 2006, 1764, 1811-1822. (18) Zubarev, R. A.; Zubarev, A. R.; Savitski, M. M. Journal of the American Society for Mass Spectrometry 2008, 19, 753-761. (19) Bonenfant, D.; Coulot, M.; Towbin, H.; Schindler, P.; van Oostrum, J. Molecular & Cellular Proteomics : MCP 2006, 5, 541-552. (20) Nicklay, J. J.; Shechter, D.; Chitta, R. K.; Garcia, B. A.; Shabanowitz, J.; Allis, C. D.; Hunt, D. F. The Journal of Biological Chemistry 2009, 284, 1075-1085. (21) Garcia, B. A.; Siuti, N.; Thomas, C. E.; Mizzen, C. A.; Kelleher, N. L. Int. J. Mass Spectrom. 2007, 259, 184-196. (22) Garcia, B. A.; Thomas, C. E.; Kelleher, N. L.; Mizzen, C. A. J. Proteome Res. 2008, 7, 4225-4236. (23) Jung, H. R.; Sidoli, S.; Haldbo, S.; Sprenger, R. R.; Schwammle, V.; Pasini, D.; Helin, K.; Jensen, O. N. Analytical Chemistry 2013, 85, 8232-8239. (24) Sidoli, S.; Lin, S.; Karch, K. R.; Garcia, B. A. Analytical Chemistry 2015, 87, 3129-3133. (25) Ebhardt, H. A.; Sabido, E.; Huttenhain, R.; Collins, B.; Aebersold, R. Proteomics 2012, 12, 11851193. (26) Maiolica, A.; Junger, M. A.; Ezkurdia, I.; Aebersold, R. Journal of Proteomics 2012, 75, 34953513. (27) Picotti, P.; Aebersold, R. Nature methods 2012, 9, 555-566. (28) Gillet, L. C.; Navarro, P.; Tate, S.; Rost, H.; Selevsek, N.; Reiter, L.; Bonner, R.; Aebersold, R. Molecular & cellular proteomics : MCP 2012, 11, O111 016717. (29) Geiger, T.; Cox, J.; Mann, M. Molecular & Cellular Proteomics : MCP 2010, 9, 2252-2261. (30) Venable, J. D.; Dong, M. Q.; Wohlschlegel, J.; Dillin, A.; Yates, J. R. Nature Methods 2004, 1, 3945. (31) Keller, A.; Bader, S. L.; Shteynberg, D.; Hood, L.; Moritz, R. L. Molecular & Cellular Proteomics : MCP 2015, 14, 1411-1418. (32) Peterson, A. C.; Russell, J. D.; Bailey, D. J.; Westphall, M. S.; Coon, J. J. Mol. Cell. Proteomics 2012, 11, 1475-1488. (33) Kessner, D.; Chambers, M.; Burke, R.; Agus, D.; Mallick, P. Bioinformatics 2008, 24, 2534-2536. (34) Liu, X.; Inbar, Y.; Dorrestein, P. C.; Wynne, C.; Edwards, N.; Souda, P.; Whitelegge, J. P.; Bafna, V.; Pevzner, P. A. Mol. Cell. Proteomics 2010, 9, 2772-2782. (35) Chalkley, R. J.; Baker, P. R.; Huang, L.; Hansen, K. C.; Allen, N. P.; Rexach, M.; Burlingame, A. L. Molecular & Cellular Proteomics : MCP 2005, 4, 1194-1204. (36) Carvalho, P. C.; Xu, T.; Han, X.; Cociorva, D.; Barbosa, V. C.; Yates, J. R., 3rd. Bioinformatics 2009, 25, 2734-2736. (37) MacLean, B.; Tomazela, D. M.; Shulman, N.; Chambers, M.; Finney, G. L.; Frewen, B.; Kern, R.; Tabb, D. L.; Liebler, D. C.; MacCoss, M. J. Bioinformatics 2010, 26, 966-968. (38) French, W. R.; Zimmerman, L. J.; Schilling, B.; Gibson, B. W.; Miller, C. A.; Townsend, R. R.; Sherrod, S. D.; Goodwin, C. R.; McLean, J. A.; Tabb, D. L. Journal of proteome research 2015, 14, 1299-1307. (39) Friend, C. The Journal of Experimental Medicine 1957, 105, 307-318. (40) Young, N. L.; DiMaggio, P. A.; Garcia, B. A. Cell. Mol. Life Sci. 2010, 67, 3983-4000.
ACS Paragon Plus Environment
14
Page 15 of 21
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
FIGURES
Figure 1: MS/MS spectrum of the synthetic N-terminal histone H3 peptide standard AcetylARTKQTARKSTGGKAPRKQLATKAARKSAPATGGVKKPHRYRPGTVALRE
with
assigned
fragment ions. Note that the spectrum is split in m/z range from (top) 400-3000 and (bottom) 3000-5500 to better display the fragment ion assignments.
ACS Paragon Plus Environment
15
Analytical Chemistry
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 16 of 21
Figure 2: (A) PRM method for 20 methyl equivalents of the +10 charge of the N-terminal histone H3 fragment. (B) Illustration of the principal differences between a DDA and PRM method (* indicate ions selected for MS/MS based on intensity and ▼ indicate targeted ions by PRM). Especially, in middledown approaches multiple charge states of the same proteoform exist leading to multiple sampling of the same proteoforms rather than an in-depth analysis of all species present. In contrast, a PRM approach targets only the (predetermined) ions of interest, with much greater depths. In addition, when the targets are very low abundance and the charge state cannot be easily determined, the PRM method still allows for systematic monitoring of the proteoforms of interest.
ACS Paragon Plus Environment
16
Page 17 of 21
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
Figure 3A: Heat map of the retention time vs. deconvoluted mass of the GluC-digested N-terminus of Histone H3 spectra separated by WCX-HILIC. Methylation status is shown on the x-axis and acetylation status is shown on the y-axis. Figure 3B: Butterfly plot of two technical replicates of the dimethylated H3 N-terminal peptide. Ions shown in red are matching between the two technical replicates within 10 ppm; the few ions that were not matching are shown in black. Pearsons’s correlation of intensities between the two replicates was 0.88 using a m/z bin width of 0.05. Figure 3C: Reproducibility of the intensity of representative ETD fragment ions from five replicates (Rep1-5). CV varies between 2.2% and 8.2%.
ACS Paragon Plus Environment
17
Analytical Chemistry
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 18 of 21
Figure 4A: Annotation of the di-methylated H3 spectrum. When expanding the 1300-1400 m/z region, we see evidence of multiple proteoform (i.e., one or two methylations in the 11 C-terminal residues and zero or one methylations in the 13 N-terminal residues). Figure 4B: A minimum set of four proteoforms were needed to explain the fragment ions observed for the di-methylated H3 N-terminal peptide of the untreated sample. Violet squares represent monomethylations and green squares represent dimethylations. Annotations for all other proteoforms are listed in the Supplemental Table S2 (untreated) and S3 (DMSO-treated).
ACS Paragon Plus Environment
18
Page 19 of 21
50%
Percent Sumed XIC
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
45%
Untreated
40%
DMSO
35% 30% 25% 20% 15% 10% 5% 0% 0Ac
1Ac
2Ac
3Ac
4Ac
Figure 5: Quantitation on the MS1 level. DMSO treatment leads to a slight relative increase in 0 and 1 acetylations and a decrease in 2, 3, 4 and 5 acetylations across the three replicates. Error bars here represent the standard error. XIC areas were summed across all number of methylations. Summed XICs were normalized to the total of all summed XICs in each replicate. The overall average number of acetylations decreased from 1.32 in the untreated samples to 1.02 in the DMSO treated sample with a ttest p-value of 0.02.
ACS Paragon Plus Environment
19
Analytical Chemistry
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 20 of 21
Figure 6: Fragment ion map for the di-methylated H3 N-terminal peptide for the untreated (A, top) and DMSO-treated (B, bottom) sample. Gray squares represent fragment ions observed. Figures 6 A and B also display the differences for the di-methylated H3 N-terminal peptide between the (A) untreated and (B) treated sample.
ACS Paragon Plus Environment
20
Page 21 of 21
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
For TOC only
ACS Paragon Plus Environment
21
