Isolation of Histone from Sorghum Leaf Tissue for Top Down Mass Spectrometry Profiling of Potential Epigenetic Markers

1. Preparing sorghum leaf material

NOTE: The sorghum plants were grown in soil in the field in Parlier, CA.

1. Collect sorghum leaves from plants into 50 mL centrifuge tubes and immediately freeze the tube in liquid nitrogen. Collect leaf tissue by tearing off the third and fourth fully emerged leaf from the primary tiller.
   NOTE: More details of field condition, sample growth, and collection can be found in the published report¹⁸.
2. Grind leaves with liquid nitrogen and immediately transfer to a centrifuge tube.
3. Store the ground leaf at -80 °C until use. Take about 4 g of cryo-ground leaf powder for histone analysis of each sample.

2. Preparing buffers and materials (3–4 h)

NOTE: The high concentration stock solutions can be made ahead of time and stored until use. But all working buffers must be made fresh on the day of the extraction (by dilution from stock and mixing with other contents) and to be placed on ice during the process. The whole experiment should be performed at 4 °C unless recommended otherwise.

1. Prepare 2.5 M sucrose by dissolving 42.8 g of sucrose (342.30 g/mol) in 15 mL of sterile water on heat plate in a glass container with continuous stirring. Bring up the volume to 50 mL once the sucrose has dissolved completely. Store the sucrose in 4 °C until use.
2. Prepare 1 M Tris pH 8 by dissolving 1.576 g of Tris HCl in 10 mL of H₂O in a 15 mL centrifuge tube. Adjust pH with NaOH to 8 and check with pH paper. Store it at 4 °C until use.
3. Prepare 1 M Dithiothreitol (DTT) by weighing 231 mg of DTT (154.25 g/mol) and dissolving it in 1.5 mL sterile water. DTT must be made fresh or use stored frozen aliquots.
4. (Optional) Prepare the additional inhibitors by mixing three different salts. Prepare 18.38 mg of sodium orthovanadate (183.91 g/mol) in 1 mL of sterile water, then prepare separately sodium butyrate by adding 11.008 mg of sodium butyrate (110.09 g/mol) in 1 mL of sterile water. Prepare the final salt by adding 4.199 mg of sodium fluoride (41.99 g/mol) in 1 mL of water. Mix the three salt solutions together in equal volume as stock solution for “additional inhibitors” (33 mM of each of the three chemicals).
   NOTE: Sodium vanadate polymerizes at concentrations higher than 0.1 mM under neutral pH. It is advised to activate sodium vanadate to depolymerize it for maximum efficacy following published protocols¹⁹. Alternatively, activated sodium vanadate is commercially available. Herein, sodium vanadate was not activated intentionally, so the efficacy does not get reduced. Activated sodium vanadate has not been tested for this protocol yet.
5. Prepare 1 M of MgCl₂ by dissolving 0.952 g of anhydrous magnesium chloride (95.2 g/mol) in 10 mL of H₂O in a 15 mL centrifuge tube. Store 1 M MgCl₂ at 4 °C until use.
6. Prepare 10% (v/v) Triton X-100 by mixing 53.5 g of Triton X-100 with 35 mL of sterile water, bring up to 50 mL with water and store it at room temperature.
7. Prepare 5% Guanidine buffer pH7 (referred as “Gdn buffer”) that will be used to condition the resin at least overnight – prepare 0.1 M potassium hydrogen phosphate dibasic (K₂HPO₄) by weighing 870 mg of K₂HPO₄ and dissolving in 50 mL of sterile water and store at 4 °C.
8. Weigh 0.7 g of guanidine hydrochloride and dissolve in 0.1 M K₂HPO₄ to a final volume of 14 mL. Adjust pH to 7 by checking with pH paper.
9. Soak the dry weak cation exchange (WCX) resin in 5% Guanidine buffer pH 7 overnight. Remove the supernatant and refill with fresh 5% Gdn buffer and soak it again overnight to let the resin fully equilibrate (until the supernatant has the same pH as the original buffer).
10. Before starting the experiment in the next section, mix the reagents to make EB1, EB2A, and EB2B buffer based on Table 1. Add all inhibitors and DTT fresh just before use.

Table 1: Composition for extraction buffers (EBs).

11. Make the Nuclei Lysis Buffer (NLB) based on Table 2. Prepare NLB in advance and store at 4 °C until use. Add PI tablets fresh just before use at 1x (0.5 tablet per 5 mL). See Table 2 for specific volumes.

Table 2: Composition for the nuclei lysis buffer (NLB).

3. Nuclei isolation procedure

NOTE: It is recommended to perform steps 3.1–3.3 of the first day (2–3 h), save the nuclei in NLB buffer at -80 °C and resume the following day (or later) for protein purification (4 h). The nuclei isolation steps in this protocol were adapted from a sorghum ChIP-seq protocol being used at the Joint Genome Institute. Additional washes and sucrose gradient separation may be required to ensure nuclei purity for ChIP-seq applications.

1. Filtration of debris (~0.5 h)
   1. Weigh ground leaf powder ~4 g, ensuring it remains frozen by placing on dry ice or liquid nitrogen until ready to use.
   2. Add protease inhibitor tablets to EB1 to a final concentration of 0.2x (0.5 tablet for 25 mL per sample). Use a miniature plastic pestle or a pipette tip to pre-crush tablets in a microcentrifuge tube prior to adding to buffers to aid in dissolution of the tablet in the buffer. To prevent material loss, add the PI tablet and sonicate the buffer to dissolve the tablet.
   3. Add 20 mL of EB1 to the frozen ground leaf powder, gently vortex and mix them until the powder is completely suspended. Keep mixing gently for ~10 min.
   4. Filter through mesh 100, rinsing the filtered material twice with 2 mL of EB1 each time.
      NOTE: Both the filtrate and the filtered debris should be green. If tracking using a microscope, one should be able to see intact nuclei and intact chloroplasts in the filtrate at this point. Majority of large debris should be absent/depleted. Mix dyes such as methylene blue with sample. Nuclei are easily observable as ~3–5 µm diameter dark blue/aquamarine spheres when visualized using a 20x, 40x, and/or 100x objective. Relative to nuclei, chloroplasts are similar in size, but greenish in color and often more oval in shape. Vacuoles are also similar to nuclei in size and shape, but they will not readily take up the Methylene blue dye.
   5. Centrifuge the combined filtrate at 3,000 x g for 10 min at 4 °C in a swinging bucket rotor to pellet debris and large subcellular organelles, including nuclei and chloroplasts.
      NOTE: It is recommended to prepare EB2A during this spin (see step 3.2.1).
   6. Decant the supernatant, being careful to not disturb the pellet.
      NOTE: As no detergent has yet been added, the pellet should remain intense green and the supernatant should be, at most, pale green/yellow.
2. Lysis of non-target organelles (~0.5 h)
   1. Prepare EB2A by adding protease inhibitors to a final concentration of 0.4x (0.5 tablet per 12.5 mL EB2A).
   2. Resuspend the pellet from step 3.1.6 in 5 mL of EB2A and incubate on ice for 10 min with gentle mixing.
      NOTE: The detergent concentration needs to be optimized to preferentially lyse intact cells and chloroplasts but not nuclei. The amount required can vary among organisms. It is recommended to check for lysis of chloroplasts and retention of intact nuclei under microscope.
   3. Centrifuge at 2,100 x g for 15 min at 4 °C in a swinging bucket rotor to pellet debris and nuclei.
      NOTE: At this stage, the supernatant should be intensely green, and the pellet should be much less green than observed in the previous stages due to the lysis of chloroplasts and chlorophyll release into the cytosol.
   4. Decant the supernatant, being careful to not disturb the pellet.
3. Isolation of nuclei from remaining cytoplasmic contaminants (~0.5 h)
   1. Prepare EB2B by adding protease inhibitors to a final concentration of 1x (0.5 tablet per 5 mL EB2B).
   2. Resuspend crude nuclear pellet from step 3.2.3 in 2 mL of EB2B.
      NOTE: EB2B does not contain Triton X-100, so no additional lysis should occur at this point.
   3. Centrifuge at 2,100 x g for 15 min at 4 °C in a swinging bucket rotor to pellet debris and nuclei.
      NOTE: Small organelles and cytoplasmic components should not pellet, so they should remain in the supernatant.
   4. Decant the supernatant, being careful to not disturb the pellet.
   5. Resuspend the pellet using 250 µL of NLB (add 0.5 protease inhibitor tablet fresh for 5 mL).
      NOTE: The goal is to resuspend the nuclei in a minimum amount of NLB without significant material loss. Because NLB is very viscous and the pellets contain a large amount of insoluble debris, it is very difficult to pipette and tends to cling to the inside of pipette tips. For this reason, it is recommended to reuse the same pipette tip whenever possible. If concerned with residual material in a pipette tip, simply hang the pipette from a shelf or rack for ~1 min to allow gravity to collect material at the opening of the tip. Do not aggressively pipette to resuspend the pellets. Instead, use the pipette tip as a stir rod until the pelleted material can be aspirated into the pipette tip. i.e., it is perfectly fine for large pellet clumps to stay at this stage so long as it can be drawn into a pipette tip.
   6. Vortex 15 s at max to homogenize and partially resuspend the material. Sonicate for 5 min at 4 °C, then store at -80 °C.
      NOTE: For subsequent steps, keep in mind that the total amount of NLB added is 250 µL, but the total apparent volume of the sample can be up to twice as much due to insoluble debris. The sample is frozen and thawed to assist in the lysis of nuclei.
4. Nuclei lysis and histone extraction (~4 h)
   1. Add 750 µL of 5% Gdn buffer to the thawed sample. Sonicate for 15 min at 4 °C.
   2. Transfer sample into a single 2 mL tube and spin 10,000 x g for 10 min at 4 °C.
      NOTE: The supernatant will likely look green. The following chromatography steps should remove most of the pigments from the protein.
   3. While waiting on step 3.4.1 and 3.4.2, prepare the column for ion exchange chromatography clean up. Rinse the chromatography column with 2 mL of acetonitrile and 4 mL of water to minimize contamination on surface.
   4. Load 200~300 µL of WCX resin (pre-conditioned with 5% Gdn buffer) onto the chromatography column. Let the resin settle. Wash four times with 1 mL of 5% Gdn buffer. Keep the tube and column on ice for the rest of the purification steps.
   5. Put the chromatography column on a 2 mL collection tube. Load the supernatant from step 3.4.2 slowly onto the resin bed without disrupting the resin (try to slowly drop from the side of the tubes). Let the solution flow through by gravity. As the solution is flowing through, load the eluent back to the top of the column 6–8 times to allow maximum binding to the resin. Then, discard the eluent.
   6. Load 2 mL of 5% Gdn buffer to wash non-histone proteins off the column. Discard the eluent.
   7. Elute histones with 1 mL 20% Gdn buffer. Collect the eluent, which contains histone proteins.
   8. Use 3 kDa molecular weight cut off (MWCO) spin filter (0.5 mL) to desalt the eluent from step 3.4.6. Before use, load 500 µL wash solvent (0.2% formic acid in 3% ACN) and spin it down twice to clean the filter.
      NOTE: It is recommended to start washing the MWCO filter while performing the resin chromatography steps to save time. The following spin filter steps take ~3–4 h.
   9. First load 500 µL of histone sample, spin at 14,000 x g for ~25 min to reduce volume down to ~100 µL. Then load another 400 µL of sample and spin at 14,000 x g again for ~25 min. Load the final 100 µL of sample, rinse the sample tube with 300 µL wash solvent and load the solvent into the filter. Spin at 14,000 x g again for ~25 min.
   10. Load 400 µL wash solvent, spin at 14, 000 x g for ~25 min to reduce volume to ~100 µL or less. Each cycle reduces the salt concentration by one-fifth. Repeat for another three cycles to bring guanidine concentration to ~0.01%. Reverse the filter into a clean collection tube and spin at 1,000 x g for 2 min. Save the purified histone sample at -20 °C or -80 °C for analysis.
       NOTE: It is recommended to spin longer (30–40 min) at the last step to minimize sample volume to obtain higher concentration. The volume should be able to go down to 50–70 µL.

4. Mass spectrometry of purified histones

1. Liquid chromatography mass spectrometry (LC-MS) data acquisition
   1. Estimate protein concentration by Bicinchoninic Acid (BCA) assay following the manufacturer’s protocol.
      NOTE: BCA can only provide an estimate of total protein concentration, but not the quality of histone purification. If MS instrumentation is not readily available for checking the quality of histone purification, western blot can be used. Reversed-phase LC coupled with 210 nm ultraviolet absorbance detection as described in our previous report can be also used²⁰. The chromatogram can be compared with a known standard for checking sample quality. However, different organisms can have different elution profiles. Therefore, using histone standards from similar organisms is highly recommended.
   2. Connect a C18 reversed phase (RP) analytical column (e.g., 3 µm 300 Å, column inner diameter 75 µm, outer diameter 360 µm, length 70 cm) and a C18 trap column (e.g., 3.6 µm, column inner diameter 150 µm, outer diameter 360 µm, length 5cm) to a dual-pump nanoflow liquid chromatography system (e.g., Waters NanoAcquity). The binary solvents are A: 0.1% formic acid in water, and B: 0.1% formic acid in acetonitrile.
      NOTE: The dual pump LC includes a wash pump and a gradient pump. Both pumps go through two stages in each analysis—a trapping stage followed by the analytical stage. In the trapping stage, the wash pump flows into the trap column and the gradient pump flows into the analytical column. In the analytical stage, the trap column is coupled with the analytical column, and the gradient pump flows into both columns. The wash pump then goes to the waste.
   3. Trapping stage: Set up the LC method to first load 1–2 µg of histone sample onto the trap column. Desalt the sample by the wash pump at 3 µL/min 5% solvent B for 10 min. Set the analytical pump at 0.3 µL/min 5% solvent B for equilibration.
   4. Analytical stage: Set the gradient pump (0.3 µL/min) to start at from 5% B and ramp to 30% at 15 min. Then, increase to 41% B at 100 min before a high organic wash up to 95% B at the end.
      NOTE: The gradient can be optimized depending on the different retention profiles on individual columns. Typically, full-length histones elute around 30%–40% B on the specified LC conditions. Longer gradients can be used to increase the numbers of MS2 spectra to capture more histone proteoforms.
   5. Set up data-dependent acquisition method on a high-resolution MS (e.g., Thermo Orbitrap Fusion Lumos or similar) with electron transfer dissociation (ETD) capability. Use the intact protein mode and perform all the necessary calibrations as suggested by the manufacturer. Critical parameters are described below. These will be specific to the instrument used.
      1. MS1: scan range 600–2,000 m/z, resolution 120k (at m/z 200), 4 microscans, AGC target 1E6, max injection 50 ms.
      2. MS2: resolution 120k; 1 microscan; AGC target 1E6; data dependent MS/MS: alternating ETD (25 ms reaction time, max injection time 500 ms) and higher-energy collisional dissociation (HCD, 28% normalized collision energy with ±5% stepped energy, max injection time 100 ms); isolation window of 0.6 Da; priority on highest charge states.
      3. Dynamic exclusion: 120 s time window, ±0.7 Da mass window. Exclude charge states lower than 5 and undetermined charge states.
   6. Run a few injections of peptide or histone standards on new columns to equilibrate and check the system, before running the actual samples. For running large number of samples, add short blanks or washes in between samples to minimize carry over. Let the columns equilibrate for 15–20 min at the starting condition (5% solvent B) before the next sample.
      NOTE: Longer LC gradients and higher max injection time for MS2 can improve the spectral quality for identifying more histone proteoforms.
2. LC-MS data processing and proteoform identification
   1. Obtain the (sorghum) protein sequence in FASTA format from JGI (https://genome.jgi.doe.gov) or UniProt (https://www.uniprot.org/).
   2. Use MSConvert²¹ (http://proteowizard.sourceforge.net/tools.shtml) to convert the instrument raw data files (*.raw) into mzML format.
   3. Download TopPIC suite²² (http://proteomics.informatics.iupui.edu/software/toppic/) for data processing. The program can be run in either command line or through the graphical interface.
   4. Use TopFD in the TopPIC suite to deconvolute the spectra from the mzML file from step 4.2.2. The default parameters can be used. But the “precursor window” (-w) needs to be reduced to 1 m/z because a narrow isolation window is used.
   5. Use TopPIC in the TopPIC suite to identify proteoforms. Most of the default parameters can be used. Set the spectrum and proteoform cutoff type to FDR (false discovery rate) and set the cutoff value to 0.01 (1% FDR) or as desired. Set the “proteoform error tolerance” to 5 (Dalton). Load the FASTA file from step 4.2.1 and the “*_ms2.msalign” file from step 4.2.4. Then start the search.
      NOTE: The “proteoform error tolerance” setting will combine proteoforms with similar masses (± 5 Da) as one. This helps reduce redundancy in the proteoform counts. However, it should be used with caution because large tolerance will merge proteoforms with small or no mass differences. This parameter is only available in TopPIC version 1.3 or later.
   6. The identified proteoforms can be examined in the “*_proteoform.csv” file or visualized using the Topview module under the “*_html” folder of the output.
   7. The proteoforms list generated from the steps above using TopPIC annotates the histone PTMs as mass shifts. In order to localize individual PTMs, a modification list must be included. Detailed description can be found in the TopPIC manual. Alternatively, proceed to the next step to perform a complementary data analysis using the Informed-Proteomics package²³ (https://github.com/PNNL-Comp-Mass-Spec/Informed-Proteomics).
   8. Follow the instructions and use the PbfGen module to convert the instrument raw data to a PBF file. Then deconvolute the MS1 data using ProMex module to output a ms1ft file (feature list, each feature represents a unique combination of mass and retention time).
   9. Create a focused FASTA for Informed-Proteomics using the identified protein list from TopPIC in step 4.2.6.
      NOTE: Searching the entire genome using Informed-Proteomics with large number of variable PTMs can be extremely slow and may cause crashes. Therefore, it is recommended to reduce the size of FASTA by only including the target proteins.
   10. Create a targeted modification list to search for histone PTMs following the format in the example file. The common PTMs to include are: Lysine acetylation, lysine mono-methylation, lysine di-methylation, lysine tri-methylation, serine/threonine/tyrosine phosphorylation, protein N-terminal acetylation, methionine/cysteine oxidation. For sorghum, protein N-terminal mono-methylation, di-methylation, and trimethylation should be added.
       NOTE: Informed-Proteomics only looks for PTMs specified in the list. If unspecified PTMs are present, the proteoform may not be identified, or may be misidentified to other proteoforms. However, the PTM list should be kept as short as possible to minimize the search time.
   11. Execute the MSPathFinder module to identify proteoforms using the files from step 4.2.8, the focused FASTA from step 4.2.9, and the modification list from step 4.2.10. The default parameters can be used.
   12. The results can be visualized in LcMsSpectator by loading all the result files.
       NOTE: Other bioinformatics tools are available for processing and visualizing top-down data, each with its own strengths^{24,25,26,27,28}. Sorghum and many other organisms have limited known information regarding histone PTMs in the database. Use TopPIC first to identify mass shifts from PTMs. This analysis can readily discover both known and unknown PTMs. Then, the detected PTMs can be searched in a targeted fashion either by specifying a PTM list in TopPIC, or with other complementary tools.