K36 Modifications on

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Quantitative Profiling of Combinational K27/K36 Modifications on Histone H3 Variants in Mouse Organs Yanyan Yu, Jiajia Chen, Yuan Gao, Jun Gao, Rijing Liao, Yi Wang, counde oyang, En Li, Chenhui Zeng, Shaolian Zhou, Pengyuan Yang, Hong Jin, and Wei Yi J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.5b01164 • Publication Date (Web): 22 Jan 2016 Downloaded from http://pubs.acs.org on January 25, 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.

Journal of Proteome Research 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 32

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

Journal of Proteome Research

Quantitative Profiling of Combinational K27/K36 Modifications on Histone H3 Variants in Mouse Organs Yanyan Yu,†,‡ Jiajia Chen,§,‡ Yuan Gao,† Jun Gao,† Rijing Liao,† Yi Wang, § Counde Oyang,† En Li,† Chenhui Zeng,† Shaolian Zhou,† Pengyuan Yang,*,§ Hong Jin,*,§ Wei Yi*,† §

Department of Chemistry & Institutes of Biomedical Sciences, Fudan University, 220 Handan

Road, Shanghai 200433, China †

China Novartis Institutes for BioMedical Research Co. Ltd., Building 8, Lane 898 Halei Road,

Shanghai 201203, China ABSTRACT: The co-existing post-translational modifications (PTMs) on histone H3 Nterminal tails were known to crosstalk between each other, indicating their interdependency in the epigenetic regulation pathways. H3K36 methylation, an important activating mark, was recently reported to antagonize with PRC2 mediated H3K27 methylation with possible crosstalk mechanism during transcription regulation process.(1) Based on our previous studies, we further integrated RP/HILIC liquid chromatography with MRM mass spectrometry to quantify histone PTMs from various mouse organs, especially the combinatorial K27/K36 marks for all three major histone H3 variants. Despite their subtle difference in physicochemical properties, we successfully obtained decent separation and high detection sensitivity for both histone H3.3 specific peptides and histone H3.1/3.2 specific peptides. In addition, the overall abundance of H3.3 can be

ACS Paragon Plus Environment

1


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 32

quantified simultaneously. We applied this method to investigate the pattern of the combinatorial K27/K36 marks for all three major histone H3 variants across five mouse organs. Intriguing distribution differences were observed not only between different H3 variants, but also between different organs. Our data shed the new insights on histone codes functions in epigenetic regulation during cell differentiation and developmental process. KEY WORDS: histone H3 variants, post-translational modifications, quantification, mouse organs.

INTRODUCTION Histone is the fundamental building block of nucleosomes that are responsible for genomic DNA packing. Canonical histone proteins are extensively decorated with covalent modifications on their N-terminal tails,(2) which are critical in the regulation of gene expression, gene stability, and nuclear architecture.(3-6) Accumulative evidence indicates that histone variants are involved in marking distinct chromatin states, orchestrating transcription, as well as regulating elongation and splicing of surrounding DNA.(7-15) Previous studies have shown that histone H3.3 displays distinct functions in terms of the timing of expression and mode of incorporation into chromatin as compared to canonical histone H3.(11, 16-20) For example, H3.3 is deposited into transcriptionally active regions to replace displaced nucleosomes throughout the cell cycle, whereas its canonical counterparts H3.1 and


2

Page 3 of 32

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


H3.2 are deposited in a replication-dependent manner.(12, 21) Recently, it is highlighted that the disruption of the specific histone codes, combinatorial K27/K36 PTMs, by frequent oncogenic mutations is crucial for the pathogenesis of malignant pediatric brain tumor.(22-24) These mutations include the mutation of histone H3.3 itself (K27M and G34R/V), and the mutations of IDH1 and SETD2. All these mutations lead to methylation level changes at either K27 or K36 residue. In addition, the crosstalk between K27 and K36 methylations is well documented in cellular studies, which indicates the feedback mechanism in epigenetic regulation.(1) To study the physiological role of H3.3 variant, Lin and his coworkers have shown that knockdown of histone H3.3 leads to embryonic developmental arrest, chromosome mis-segregation, and chromatin condensation.(25) Taken together, these recent studies provided new insight into previously unknown aspects of endogenous H3.3 in the genome. And the new questions are also raised to investigate the global epigenetic landscape of H3.3 and the contribution of its K27/K36 combinatorial codes to its unique functions in specific organs such as brain. Therefore, it is in urgent needs to develop an in-depth method to precisely characterize and quantify histone H3.3 PTMs, in order to decipher the epigenetic regulation mechanism of H3.3. However, it is challenging to develop the robust and accurate methods to quantify histone H3 variants respectively due to the extremely high homology in primary sequence between different variants. In the last decades, a number of immunological techniques were developed to study histone modifications, however, the application of these techniques in epigenetic studies are


3


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 32

limited by the specificity of antibodies, the interference from structurally similar modifications and the low throughput of the method.(26-28) Moreover, only a few antibodies are developed to distinguish endogenous histone variants, and, to our current knowledge, no antibodies are commercially available to recognize the variant-specific PTMs on the histone H3 N-termini. Mass spectrometry emerged as a powerful tool for the characterization of histone variants as well as their PTMs in a quantitative manner.(26,

29-31)

Leveraging the power of multiple reaction

monitoring (MRM)-based LC-MS/MS method, we have already demonstrated that the feasibility to quantify modified histone peptides with high sensitivity, robustness and throughput. Based on our previous studies,

(30, 32)

we developed a new MRM assay combining with RPLC and HILIC

technology. It contains a peptide library consisting of 69 stable isotope labeled peptides (SIS) according to the N-terminal sequences of all three major histone H3 variants. We are able to quantify and distinguish the PTMs on K27 and K36 between histone H3.1/H3.2 and H3.3 variants, even the sequence difference between H3.1/3.2 and H3.3 is only a single amino acid (Ala31 in H3.1/3.2 and Ser31 in H3.3). This newly developed method could measure the relative abundance of each major histone H3 variant together with its combinatorial K27/K36 modifications in a dynamic range of more than 104. Furthermore, we applied the method to study histone variants modifications in different mouse organs, and the significant difference in the expression level of histone H3.3 protein and its unique combinatorial modification patterns between mouse organs, such as cerebrum, heart, kidney, liver and spleen, were discovered.


4

Page 5 of 32

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


MATERIALS AND METHODS Chemicals and Reagents. All reagents and solvents used in this study were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless noted otherwise. Trypsin (sequencing grademodified) was purchased from Promega (Madison, WI, USA). Acetonitrile was ordered from Fisher Scientific (Fair Lawn, NJ, USA). Propionic acid N‑hydroxysuccinimide ester (NHS) was prepared in our lab.(32) Pure water was produced by Milli-Q gradient A10 system (Millipore, Bedford, MA). The stable isotope labeled histone peptides containing various PTMs and chemical derivatization (purity > 95%) were purchased from New England Peptide LLC (Gardner, MA, USA). Preparation of Standard Curve Using Stable Isotope Labeled Histone Peptides for Validation of Linearity and Specificity. The stock solution, containing all 69 stable isotope labeled histone peptides, was prepared by dissolving accurate amounts of reference standards in acetonitrile: water (3:7, v/v) at concentrations of 100 µM for each peptide (Table S1). For the calibration curve measurement, a series of working standard solutions were prepared by mixing the stock solution into cellular matrix to yield final concentrations of 0.1, 0.5, 1, 5, 10, 50, 100, 500 and 1000 nM. For those peptides with high abundance in biological samples, the concentration for calibration would be increased to 2500 nM. For method specificity test, the stable isotope labeled histone H3.l/H3.2 peptides were mixed with H3.3 peptides at different


5


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 32

ratios of 100:1, 20:1, 10:1, 1:1, 1:10, 1:20 and 1:100 respectively. These solutions were, then, mixed with cellular matrix (1:1, v/v) to serve as the specificity test solutions. Extraction and Purification of Histone H3 Proteins from Mouse Organs. All histone proteins were extracted freshly from three individual 14-week old female B6.Cg mice (purchased from Shanghai Organism Science & Technology Development Co.,Ltd., Shanghai, China) from the following organs: the left hemisphere of the cerebrum (cerebrum-left), the right hemisphere of the cerebrum (cerebrum-right), heart, the left kidney, the right kidney, liver and spleen. And all mouse surgical procedures were approved and conducted in accordance with the guide for the care and use of laboratory animals at the shanghai research center for model organisms. Briefly, mice were perfused with PBS to remove blood, and then dissected to obtain the organs. The organs were then minced into small cubes with clippers before being homogenized with a nucleus isolation buffer (15 mM Tris-HCl (pH7.5), 60 mM KCl, 11 mM CaCl2, 5 mM NaCl, 5 mM MgCl2, 250 mM sucrose, 1 mM DTT, 10 mM sodium butyrate, 0.3% NP-40 and protease inhibitor cocktail (Thermo Scientific, USA). Nuclei were pelleted with a centrifugation of 500 × g, then washed one time with nucleus isolation buffer without NP-40, and then extracted with 0.4 M HCl overnight at 4 °C

(33)

. After a buffer exchange to ddH2O with Amicon ultra 0.5 mL

centrifugal filters (MWCO 10 kDa, Millipore, Bedford, MA), the extracted histone H3 proteins were further purified with a reverse phase C8 column (300Å, 4.6×150 mm, Agilent, USA) on an Agilent 1200 system (Waldbronn, Germany). The flow rate was 1 mL/min and the gradient was


6

Page 7 of 32

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


as following: starting at 5% B for 5 mins, followed by 30 mins from 5% to 35% mobile phase B, then maintaining at 35% B for 8 mins, increasing to 55% B in 42 min (mobile phase A: 0.1% TFA in ddH2O; and mobile phase B: 0.1% TFA in acetonitrile). The fractions containing histone H3 were collected and concentrated with Speed Vacuum before further derivatization. Derivatization and Digestion of Histone H3 Proteins. All the experiments were performed as described previously.(32) In brief, histone H3 proteins were dissolved with 50 mM ammonium bicarbonate. After adding the same volume of 200 mM propionic acid N‑hydroxysuccinimide ester (NHS) in acetonitrile, the mixture was incubated at 50°C for 30 mins. The reacted solutions were dried by Speed Vacuum, and the derivatized proteins were then dissolved with 25mM NH4HCO3 (pH10). The final pH was adjusted to 8.5. Then trypsin was added at the ratio of protein to trypsin = 20:1(w/w). The digestion was carried out at 37 °C overnight. After digestion, a second round derivatization was processed; the resulting derivatized histone peptides were then dried thoroughly and resolved in 30 µL 0.1% formic acid/30% acetonitrile. Since the histone concentration was rather high and varies in different mouse organs, in order to reach similar ion intensity in the following LC-MS analysis, the derivatized histone peptides were diluted in series ratio of 1:10, 1:10, 1:2, 1:10, 1:10, 1:8 and 1:20 for samples prepared from left cerebrum, right cerebrum, heart, left kidney, right kidney, liver and spleen respectively, RP/HILIC-MRM Analysis for Histone H3 Markers in Different Variants. Prior to LC-MS analysis, the equal volume of internal standard peptides was mixed with the diluted derivatized


7


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 32

tissue samples. In the final solution, the internal standards concentrations were 50 nM for all stable isotope labeled histone peptides, except K27 acetylation, K4 methylation and K79 methylation containing peptides. Due to the low abundance of these three types of peptides in our samples, we instead used 5 nM internal standards for these peptides in the final solution. Eventually, a 2 uL aliquot of final solution was injected into the LC-MS/MS system for analysis and three injections for technical replicates were performed. Peptides were separated using Waters AcquityTM UPLC system (USA). RPLC separation was performed on a reverse phase C18 column (2.5 µm, 2.1×100 mm, XselectTM HSS T3, Waters, USA) at a flow rate of 0.35 mL/min. The chromatographic gradient was 5-13% B from 0 to 8 min, 18-25% B from 20 to 23 min, 25-90% B from 23 to 25 min, and maintaining 90% B for 3.5 mins (mobile phase A: 0.1% formic acid in ddH2O ; mobile phase B: 0.1% formic acid in acetonitrile). HILIC separation was performed with a CORTECSTM HILIC column (2.7 µm, 2.1×100 mm, Waters, USA) at a flow rate of 0.2 mL/min. The chromatographic gradient was 85% B from 0 to 5 min, 85-60% B from 5 to 18 min, 60-45% B from 18 to 23 min, then 45% B for another 3 mins, followed by a 0.5 min linear gradient from 45% to 85% B and a quick re-equilibration with 85% B at a flow rate of 0.3 mL/min for 5.5 mins (mobile phase A: 0.2% formic acid and 10mM NH4COOH in ddH2O; mobile phase B: 0.2% formic acid and 10mM NH4COOH in 10% methanol and 60% acetonitrile). After LC separation, the peptides were directly analyzed online by an Applied Biosystems Sciex Qtrap 5500 mass spectrometer (Applied Biosystems, USA) for all LC-MRM analysis. All parameters for mass spectrometers are as following: ion spray voltage = 4500 V,


8

Page 9 of 32

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


curtain gas = 40 p.s.i., interface heater temperature = 550°C, collision gas =high, and ion source gas 1 and 2 = 65 p.s.i. MRM acquisition methods were constructed by 276 MRM ion pairs (140 ion pairs in RP-MRM and 136 in HILIC-MRM method) with peptide-specific tuned declustering potential (DP), collision energy (CE) voltages, and retention time constraints, and in each LC-MRM method four period acquisitions were included (Supplementary table S1). All MRM data were processed through Skyline software (Washington University at St. Louis, St Louis, MO, USA). All data were manually inspected to ensure correct peak detection and accurate integration according to four criteria: 1) the correct m/z is selected for both the heavy and light trace of each peptide; 2) the peak shapes are Gaussian like and do not show excessively jagged appearance; 3) the retention time of certain peptide is similar between different runs; and 4) the relative contribution of each transition to the total signal is similar between each sample. Statistics was also performed using a one-tails homoscedastic t-test. Data Validation with Western Blot. Tissue extracted histones were mixed with Laemmli Sample Buffer and boiled at 95 °C for 10 min. After separation on SDS-PAGE gels, histones were transferred to PVDF membranes. Membranes were blocked in 5% milk with a dilution of 1:2000 for the rabbit polyclonal antibody against histone H3 (Cell Signaling Technology, UK), and 1:600 for the rabbit polyclonal antibody against histone H3.3 (Millipore, US). Bands were visualized using an ECL plus detection system (Thermo fisher scientific, USA) and scanned by


9


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 32

Las-3000 Luminescent Image Analyzer (Fujifilm, Japan). The validation experiment was repeated three times.

RESULTS AND DISCUSSION Based on our previous studies,

(30, 32)

we set up an improved quantitative approach with

integrated RP/HILIC-MRM technologies for accurately measuring PTMs on histone H3 variants. In current approach, we incorporate additional 20 stable isotope labeling histone H3.3 K27-R40 peptides containing a C-terminal [13C6] and [15N4] Arg with propionylation derivatization in order to develop the MRM-based LC-MS/MS methods for the in-depth analysis of the PTMs on K27 and K36 of all three major histone H3 variants. Challenges of LC-MS Method for Quantifying PTMs on Histone H3 Variants. Human histone H3 proteins have three major variants, histone H3.1, H3.2 and H3.3. Histone H3.3 appears to be less abundant as compared to canonical histone H3.1/3.2 in most mammalian tissues. And its primary sequence only differs in 4 and 5 amino acid residues from H3.1 and H3.2 variants respectively. Especially, on the N termini of histone H3 that carries most prevalent PTMs, there is only one single amino acid difference at the 31th residue between H3.3 (Serine) and H3.1/H3.2 (Alanine) (Shown in Figure 1A). The molecular weight of non-modified K27R40 peptide of histone H3.1/H3.2 is 16 Da (an oxygen atom) smaller than that of histone H3.3, which allows us to detect and quantify K27-R40 peptide from histone H3.1/H3.2 and H3.3 respectively with MRM approach in principle. However, considering the PTMs occurred on K27


10

Page 11 of 32

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


and K36, such as methylation, acetylation and propionylation (chemical derivatization on unand mono-methyl lysine), for example, certain K27-R40 peptides from histone H3.1/H3.2 can possess an additional CH2 group, and their molecular weights only differ by 2 Da from those of H3.3 variant (Figure 1B). In total, 41 H3 variant peptide pairs are considered to have precursor m/z similarities (as shown in Table S2) and some of them are also quite similar in the physicochemical properties in terms of chromatographic behaviors. As shown in Figure 1C and 1D, the intensities of histone H3.1 isotopic M+2 peaks which were about 30%-50% of their mono-isotopic peaks in the mass spectra, overlap with the mono-isotopic peak from certain proteoform of histone H3.3. Furthermore, once these peptides carry three or four positive charges during ESI process, the m/z differences of their precursor ions would become much smaller. Therefore these precursor ions could consequently interfere with each other in MRM channels as the Q1 filter window is commonly not narrow enough (usually ±0.7 Da) in the triple quadrupole type instruments. For example, the molecular weights of K27me2K36me3 and K27me3K36me2 peptides from histone H3.1/H3.2, and K27me2K36me2 from histone H3.3 are 1625.0 Da and 1626.9 Da respectively. With 4+ charges, their predominant precursor ions are 404.75 Da and 405.24 Da respectively. As shown in Figure 1C, an isotopic ion signal of histone H3.1/H3.2 peptides (m/z 405.25 Da) overlapped with the mono-isotopic ion signal of histone H3.3 peptide (m/z 405.24 Da), which would certainly compromise the MRM quantification accuracy under the co-elution condition. Another example is shown in Figure 1D: the third isotopic ion peak of K27me1K36me0 or K27me0K36me1 from histone H3.1/H3.2 (m/z 558.66 Da) presumably has


11


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 32

severe interference with the quantification of H3.3_K27me0K36me0 peptide (m/z 558.65 Da) in MRM analysis if decent chromatographic separation is not possible. Based on the above examples, we summarized three situations to elucidate the interferences between histone variant peptides with various modifications with LC/MS method: 1. Tri-methyl and acetyl groups are almost the same in molecular weight (me3: 42.011Da; ac: 42.047Da), which is a challenge for triple-Q mass spectrometer, unless such peptides can be prior separated by HPLC. 2. Combinatorial modifications on K27 and K36 of the same histone variant sequence will lead to isomeric peptides which are difficult to be resolved by HPLC. As shown in Figure 2A, K27me2K36me3 and K27me3K36me2 from histone H3.1 or H3.2 have the exact same m/z for their precursor ion and the product ions from y1 to y4. Unfortunately, these peptides also have similar retentions on HPLC column (Figure 2C). In this scenario, we need to select product ions such as b ions or y5 to y13 ions in order to distinguish them through MRM method (Figure 2C). 3. Interferences between different histone variants peptides pose another challenge for quantification with MRM method. As shown in Figure 2A, the m/z for precursor ions of K27me2K36me3 or K27me3K36me2 from histone H3.1/H3.2 (m/z 404.7 Da) are very close to that of K27me2K36me2 from histone H3.3 (m/z 405.2 Da). Meanwhile, for histone H3.3 peptide, its b1-b4 and y1-y9 product ions are exactly the same with those of K27me2K36me3 and K27me3K36me2 from histone H3.1/H3.2. Furthermore, the m/z of its b5-b13 and y10-y13 product ions are close to those of related histone H3.1/H3.2 peptides (∆m/z=2 Da for 1+ charge


12

Page 13 of 32

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


ion and ∆m/z=1 Da for 2+ charge ion). The tiny m/z difference in product ions could also introduce interference for quantification of these peptides due to isotopic distribution and low m/z resolution. For example, the MRM channel for H3.3_ K27me0K36me0 is 558.6717.4, and for H3.1/H3.2_ K27me0K36me1, it is 558.0716.4 (Figure 2B and 2D). We can observe two chromatogram peaks of H3.3_K27me0K36me0 in Figure 2D, where the peak on 15.2 min is considered as the interference from H3.1/H3.2_ K27me0K36me1 peptide. Combine RP and HILIC Separation for MRM Quantification. To overcome those challenges, we hence worked toward a specific and sensitive method combining RPLC-MRM and HILIC-MRM technologies. As shown in Figure 2, at least one of the three histone variant peptides, although with similar precursor m/z and HPLC retention, can be well separated by switching to either HILIC or RP chromatography. And the rest two peptides with close retention times can be further distinguished by selecting distinct MRM channels. With this complementary strategy, we are able to quantify all combinatorial K27/K36 modifications from all three different histone H3 variants accurately. Finally, total 34 peptides containing di-methylation and/or trimethylation could be selectively separated from HILIC-MRM method with high specificity (Table 1). Detailed information about their retentions and MRM transitions is shown in Table S1 and Figure S1. In fact, these 34 peptides carry less propionylation groups due to their dimethylation and/or tri-methylation modifications and they are assumed to be more compatible for HILIC separation due to less hydrophobicity. The rest 35 peptides were quantified through


13


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 32

RPLC-MRM method (Table 1, Table S1 and Figure S2). In addition, in HILIC method, we observed significantly improved signals for K4me2 and K4me3 peptides comparing to RPLC method (Figure S3), illustrating the sensitivity advantage of HILIC over the RPLC method for hydrophilic peptides. Table 1. Histone H3 variant peptides with modifications selected for HILIC/RPLC separation. Modified peptides separated with RPLC

Modified peptides separated with HILIC

(35 standards)

(34 standards)

H3.1/H3.2_K27ac1K36me0

K18me1K23ac0

H3.1/H3.2_K27ac1K36me2

H3.3_K27me0K36me3

H3.1/H3.2_K27ac1K36me1

K18me1K23ac1

H3.1/H3.2_K27ac1K36me3

H3.3_K27me1K36me2

H3.1/H3.2_K27me0K36me0

K18me1K23me1

H3.1/H3.2_K27me0K36me2

H3.3_K27me1K36me3

H3.1/H3.2_K27me0K36me1

K4me0

H3.1/H3.2_K27me0K36me3

H3.3_K27me2K36me0

H3.1/H3.2_K27me1K36me0

K4me1

H3.1/H3.2_K27me1K36me2

H3.3_K27me2K36me1

H3.1/H3.2_K27me1K36me1

K79me0

H3.1/H3.2_K27me1K36me3

H3.3_K27me2K36me2

H3.3_K27ac1K36me0

K79me1

H3.1/H3.2_K27me2K36me0

H3.3_K27me2K36me3

H3.3_K27ac1K36me1

K79me2

H3.1/H3.2_K27me2K36me1

H3.3_K27me3K36me0

H3.3_K27me0K36me0

K79me3

H3.1/H3.2_K27me2K36me2

H3.3_K27me3K36me1

H3.3_K27me0K36me1

K9ac1K14ac0

H3.1/H3.2_K27me2K36me3

H3.3_K27me3K36me2

H3.3_K27me1K36me0

K9ac1K14ac1

H3.1/H3.2_K27me3K36me0

H3.3_K27me3K36me3

H3.3_K27me1K36me1

K9me0K14ac0

H3.1/H3.2_K27me3K36me1

K4me2

K18ac0K23ac0

K9me0K14ac1

H3.1/H3.2_K27me3K36me2

K4me3

K18ac0K23ac1

K9me1K14ac0

H3.1/H3.2_K27me3K36me3

K9me2K14ac0

K18ac0K23me1

K9me1K14ac1

H3.3_K27ac1K36me2

K9me2K14ac1

K18ac1K23ac0

EIR

H3.3_K27ac1K36me3

K9me3K14ac0

K18ac1K23ac1

YRPGTVALR

H3.3_K27me0K36me2

K9me3K14ac1

K18ac1K23me1

Method Validation for Specificity, Linearity and Reproducibility. Specificity: As discussed earlier, total 41 pairs of peptides from different histone H3 variants have similar m/z for precursor ions and product ions. To evaluate the method specificity to those peptides, stable isotope labeled K27-R40 peptides from either histone H3.3 or H3.1/H3.2 were mixed at various


14

Page 15 of 32

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


ratios (1:100, 1:20, 1:10, 1:1, 10:1 and 100:1). Then we examined the correlation between the peak area ratio and theoretical mixing ratio. The correlation coefficients of the calibration curves were between 0.9941 and 0.9999. This indicates that the interference is negligible among H3.1/H3.2 and H3.3 peptides in our approach (Table S3). To further assess the method specificity in sample matrix, we measured the response ratios of two selected MRM transitions for each peptide in neat solution and sample matrix respectively (Table S4). The values of stable isotope labeled peptides between these conditions are quite consistent (RE< 20%), indicating the sample matrix has minimal interference. Meanwhile, the ratios of two MRM transition signals for endogenous peptides were compared as well (as shown in column 7 in Table S4), which again showed good consistency. Furthermore, to test the influence of different biological matrixes, we measured the labeled peptides signal response in various mouse tissue samples. The peak areas of 69 labeled peptides spiked in different samples matrixes are summarized in Figure S4, which shows that the signal responses were quite consistent across all five different organs. The data strongly suggests that matrix interference on the labeled peptides is negligible among different mouse organs. Linearity and Reproducibility: Histone H3.3 is relatively less abundant than its canonical counterparts in vivo, accounting only about 10-20% of total histone H3 in somatic cells (22). In addition, modifications on histone H3 have large dynamic ranges (>103) in terms of absolute abundance, which requires analytical method with better sensitivity and larger dynamic range. In


15


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 32

the present study, the lower limits of quantifications (LLOQ) were determined as 0.1 nM or 0.5nM for these 69 stable isotope labeled peptides with the signal-to-noise ratios over ten. Stable isotope labeled peptides at different concentration levels were spiked into the real sample matrix to evaluate the linearity, which yield dynamic responses over 4 orders of magnitude. The correlation coefficients of the calibration curves ranged from 0.9906 to 0.9999 (Table S4). To access the reproducibility, all the tissue samples were analyzed three times and CV% is less than 20% (Table S6). Quantification of Histone H3.3 Protein and its PTMs in Various Mouse Organs: Although they share the same genome, the terminally-differentiated organs have distinct morphological and biochemical characteristics. Epigenetic mechanisms including DNA methylation, RNA interference and histone modifications are emerging as key factors in determining such different cell types. Thus, understanding the histone PTMs and variants from different organs will facilitate us to better define their roles in regulating early cell development and differentiation. Herein, we quantified the expression of histone H3.3 and the profile of post-translational modifications on different histone H3 variants in five mouse organs, including cerebrum, heart, kidney, liver, and spleen, among which the left and right parts of cerebrum and kidney were analyzed separately. Using the absolute quantification method, we comprehensively investigated the modification profile on the N-terminal of histone H3, especially for the combinatorial PTMs on K27 and K36


16

Page 17 of 32

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


of histone variants from different mouse organs. The relative level of a specific single-site modification was calculated based on the absolute quantification data using the following formula. For example, the relative abundance of K27Me2 = ∑ K27Me2 containing peptides / ∑ K27R40 combinatorial peptides × 100%, where ∑ K27-R40 combinatorial peptides = ∑ K27Me0 containing peptides +∑ K27Me1 containing peptides + ∑ K27Me2 containing peptides + ∑ K27Me3 containing peptides + ∑ K27Ac1containing peptides. Based on this calculation formula, the combinatorial K27/K36 modifications distribution of histone variants for 5 mice organs were calculated and shown in Figure 3. To be noted, some combinatorial modifications of K27-K36 peptides, e.g. K27Me0/K36Me2, differed between histone H3.3 and its canonical counterparts in all five organ samples, indicating that histone variants carry diverging codes for epigenetic regulations. This could be regulated by variant-specific histone methyl-transferases/de-methylase and chaperone proteins recently identified in epigenetic studies (18). The abundances of certain combinatorial codes were largely varied between five different organs (Figure 3), e.g. K27Me2K36Me2. The tissue-specific distribution suggests the specific histone code may have distinct functional roles in different organs during organ developmental stage. The relative abundance in percentage value of single-site modifications in different mice organs were summarized in Figure 4A and B. The comprehensive K27 and K36 modification landscape on histone H3.1/H3.2 and H3.3 for five organs was shown in Figure 4A. . The overall distribution of K27/K36 modification levels appeared to be similar between H3.1/H3.2 and H3.3


17


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 32

except K27 acetylation. Specifically, di-methylation level on K27 was evidently higher among all histone H3 variants in cerebrum, heart, kidney, liver and spleen samples. In contrast, unmethylation level on K27 of histone H3.1/H3.2 was only found to be much higher in spleen samples. And the level of K27me3 of H3.1/H3.2 in spleen were significantly lower than in other organs (p

K36 Modifications on

Recommend Documents