The eukaryotic genome is packaged around histone proteins, which are subject to a myriad of post-translational modifications. By controlling DNA accessibility and the recruitment of protein complexes that mediate chromatin-related processes, these modifications constitute a key mechanism of epigenetic regulation. Since mass spectrometry can easily distinguish between these different modifications, it has become an essential technique in deciphering the histone code. Although robust LC-MS/MS methods are available to analyze modifications on the histone N-terminal tails, routine methods for characterizing ubiquitin marks on histone C-terminal regions, especially H2AK119ub, are less robust. Here we report the development of a simple workflow for the detection and improved quantification of the canonical histone ubiquitination marks H2AK119ub and H2BK120ub. The method entails a fully tryptic digestion of acid-extracted histones followed by derivatization with heavy or light propionic anhydride. A pooled sample is then spiked into oppositely labeled single samples as a reference channel for relative quantification, and data is acquired using PRM-based nanoLC-MS/MS. We validated our approach with synthetic peptides as well as treatments known to modulate the levels of H2AK119ub and H2BK120ub. This new method complements existing histone workflows, largely focused on the lysine-rich N-terminal regions, by extending modification analysis to other sequence contexts.

CELL CULTURE AND INHIBITOR EXPERIMENTS. HEK293, 293T, and 10T1/2 cells were cultured in DMEM with 10% FBS, 1X GlutaMAX, and 1X penicillin/streptomycin in a humidified incubator at 37°C and 5% CO2. Parental and RING1A/B-deficient 10T1/2 cells were generously provided by D. Weinberg and C. David Allis. For inhibitor treatments, HEK293T cells were seeded at 5E5-1E6 per well in 6 well plates in a final volume of 3 ml of media. On the following day, inhibitors were added at the indicated concentrations for either 2 hrs (actinomycin D) or 24 hrs (etoposide, mitomycin C, panobinostat, Ei1). Cells were then collected by trypsinization, washed in PBS, snap frozen, and stored at -80°C until histone extraction. HISTONE EXTRACTION. Acid-extracted histones were prepared from frozen cell pellets as previously described (Lund PJ, et al. Methods Mol Biol. 2019; 1977: 43–70. PMID: 30980322. PMCID: PMC6543536). Pellets were first resuspended in 200 μl of Nuclear Isolation Buffer (NIB: 15 mM Tris pH 7.5, 15 mM NaCl, 60 mM KCl, 5 mM MgCl2, 1 mM CaCl2, 250 mM sucrose) supplemented with 0.2% NP-40, 1 mM DTT, 500 μM AEBSF, 5 nM microcystin, and 10 mM sodium butyrate. After incubation on ice for 10 mins, the nuclei were collected by centrifugation at 500 x g for 5 mins at 4°C. Nuclei were washed twice in NIB without NP-40 and centrifuged as above. To extract histones, nuclei were incubated in 200 μl of 0.4 N sulfuric acid at 4°C for approximately 2 hrs. After centrifugation at 3,400 x g for 5 mins at 4°C, the histone-containing supernatant was precipitated by adding trichloroacetic acid to a final concentration of 25% and incubating overnight on ice in the cold room. The precipitated histones were pelleted by centrifugation at 3,400 x g for 5 mins at 4°C and then washed with acidified (0.1% HCl) acetone followed by pure acetone. The washed pellets were air dried to remove residual acetone. Histones were resuspended in 25 μl of 0.1 M ammonium bicarbonate, and insoluble debris was removed by centrifugation at 17,000 x g for 5 mins at room temperature. HISTONE DIGESTION, DERIVATIZATION, AND POOLING. For digestion, 10-20 μl of histone extract, containing approximately 10 μg of protein, was incubated overnight at room temperature with 0.5 μg of sequencing-grade trypsin (Promega) in 0.1 M ammonium bicarbonate in a final volume of 20 μl. Unlike our previous approaches, no derivatization was performed prior to digestion. After digestion, one half (10 μl) of each sample was transferred to a new tube to create a pooled mix. This pooled mix was then divided into two equal aliquots. Likewise, the remainder of each individual sample was also divided into two equal aliquots, and 0.1 M ammonium bicarbonate was added to each aliquot to obtain a final volume of 20 μl. Subsequently, one set of the individual and pooled samples was derivatized with light propionic anhydride (D0, Millipore Sigma) while the second set was derivatized with heavy propionic anhydride (D10, 98%, Cambridge Isotope Laboratories). The derivatization was performed by adding 1 volume of reagent (25% propionic anhydride and 75% isopropanol) to 2 volumes of sample or pool. A small scoop of ammonium bicarbonate salt was added to the reaction with a beveled P1000 pipette tip for additional buffering capacity. The reaction was incubated at 37°C for 15 mins and dried in a speed-vac. Samples and pools were then resuspended in 20 μl of 0.1 M ammonium bicarbonate for a second round of propionylation. Afterwards, the derivatized samples and pools were resuspended in 10% trifluoroacetic acid at roughly 0.1 μg/μl (note: exercise caution as samples will bubble as the bicarbonate salt is released as CO2). A small volume (~1 μl) of sample was applied to pH paper to confirm acidification (pH < 3), and additional TFA was added if necessary to decompose any remaining ammonium bicarbonate and decrease the pH. A constant volume of the heavy-labeled pool was then mixed with each of the light-labeled samples and vice versa (light-labeled pool to heavy-labeled samples) to obtain roughly a 1:1 heavy:light ratio by mass (e.g., 2 μg of sample and 2 μg of pool). In this manner, the pool serves as a standard reference channel across all samples in both forward (heavy pool + light sample) and reverse (light pool + heavy sample) labeling schemes. Finally, the mixtures were desalted with C18 stage tips, eluted with 0.1% formic acid in 50% acetonitrile, dried in a speed-vac, and resuspended in 20 μl of 0.1% formic acid for LC-MS/MS analysis. SYNTHETIC PEPTIDE EXPERIMENTS. Synthetic peptides modeling the endogenous H2AK119ub and H2BK120ub peptides were purchased from GenScript and dissolved in water. Initial experiments utilized Protein-AQUA histone peptides from Cell Signaling Tech. To assess the tryptic digestion patterns of the sequences surrounding H2AK119ub and H2BK120ub, 10 pmol of synthetic AQUA peptides were spiked into 50 ug of acid-extracted histones from HEK293 cells, either before or after digestion with 1 μg of trypsin overnight at room temperature. Digests were then acidified, desalted, dried, and resuspended in 0.1% formic acid for LC-MS/MS. To examine whether propionylation facilitated the detection of H2AK119ub, 10 pmol of synthetic peptides were diluted into 75 mM ammonium bicarbonate and derivatized as above. To simulate changes in H2Bub independent of total H2B levels, increasing amounts (0, 20, 40, 80, 160, 320, 640 fmol) of synthetic H2BK120ub peptides (AVTK[gg]YTSSK and AVTK[gg]YTSAK, GenScript) were spiked into a constant amount (2000 fmol) of an unmodified synthetic peptide (HAVSEGTK) from H2B in a final volume of 30 μl buffered with 0.1 M ammonium bicarbonate. The molar amounts of the H2BK120ub peptides represent the sum of both sequences, which were mixed at a ratio of 3 parts (AVTK[gg]YTSSK) to 1 part (AVTK[gg]YTSAK) to account for the higher abundance of the former. In another titration series, the AVTK[gg]YTSAK peptide was held constant at 1% molar abundance (20 fmol) while the AVTK[gg]YTSSK peptide was serially doubled from 20 to 640 fmol (1% to 32%). Samples were then derivatized with heavy or light propionic anhydride as above. After labeling and acidification with 5% TFA, sample pools were created by combining equal volumes of all heavy and all light samples separately, which were then spiked back into the oppositely labeled individual samples to serve as reference standards. Mixtures were desalted and analyzed by LC-MS/MS. MASS SPECTROMETRY. Histone peptides were injected into a nano-LC system (EASY-nLC 1000, Thermo) equipped with fused silica columns (75 μm x 15-20 cm, Polymicro Tech) fabricated in house with C18 material (ReproSil-Pur 120 C18-AQ, 3 μm, Dr. Maisch GmbH). Solvents A and B were 0.1% formic acid in water and 0.1% formic acid in 80% acetonitrile, respectively. The injection volume was 2 μl. After an initial hold at 2% solvent B for 2 mins, a gradient elution was performed from 2 to 50% B over 30 mins and then 50% to 90% B over 5 mins. This was followed by a 4 min wash at 90% B and then a column re-equilibration phase. The flow rate was set at 300 nl/min. Eluting peptides were electrosprayed at 2.3 kV into a mass spectrometer (Q-Exactive or QE-HF, Thermo) operating in positive mode with the capillary set to 250°C. The instrument duty cycle consisted of one full scan, acquired at 70,000 resolution over 250-1400 m/z with the AGC target set at 1e6 and the maximum injection time set at 200 ms. This full scan was followed by PRM scans targeting both unmodified and ubiquitin (GG)-modified peptides. The core inclusion list, containing 8 pairs of heavy and light peptides, is outlined in Table 2. Typical settings for the PRM scans were a resolution of 35,000, an AGC target of 1e6, a maximum injection time of 110 ms, an isolation window of 1.6 m/z, a normalized collision energy of 30, and a loop count of 10. For initial method optimization and scouting, the instrument (Thermo Fusion) was operated in DDA mode. Settings for DDA consisted of a full scan in positive profile mode from 300-1800 m/z at 60,000 resolution in the Orbitrap with a max injection time of 50 ms and an AGC target of 1e6. Using a quadrupole isolation window of 2 m/z, precursor ions with charge states of +2 to +6 were fragmented by HCD at a collision energy of 28%. MS/MS scans were collected in the ion trap in centroid mode. The scan rate was set to rapid, the max injection time was set to 50 ms, the AGC target was set to 1e4, and dynamic exclusion was set to 10 sec. The experiment cycle time of MS and MS/MS scans was 1.5 sec. DATA ANALYSIS. Chromatographic peak areas of fragment ions of interest were extracted with Skyline and exported for further analysis in R. The areas of these quantitative fragment ions were summed for each peptide and then divided by the area of the corresponding reference peptide from the pooled spike-in with the opposite isotope label to obtain a normalized ratio (H/L for heavy sample and light pool or L/H for light sample and heavy pool). The ratios for the reciprocal labeling schemes (H/L and L/H) were averaged. To calculate the relative level of H2Aub in a sample, the H/L ratio of the H2AK119ub peptide was divided by the H/L ratio of the AGLQFPVGR peptide, representing total H2A. The same was done for H2Bub except that the H/L ratio of total H2B was taken as the average of three unmodified H2B peptides (HAVSEGTK, LLLPGELAK, QVHPDTGISSK). These relative levels of H2Aub and H2Bub were then compared across samples. For initial LC-MS/MS runs in DDA mode, the data was analyzed with ProteomeDiscoverer (v2.2) using a FASTA reference database covering canonical and variant forms of human histone H2A, H2B, H3, and H4. Parameters for Sequest searches consisted of a fully tryptic digestion pattern with up to 2 missed cleavages, precursor mass tolerance of 10 ppm, and fragment mass tolerance of 0.5 Da. Dynamic modifications included diglycine (+114.043 Da to K), propionylated diglycine (+170.069 Da to K), and propionylation (+56.026 Da to K and N-termini). The Percolator and Fixed Value PSM Validator nodes were used in parallel workflows for FDR control.