Mass spectrometry analysis of Arabidopsis histone H3 reveals distinct combinations of post-translational modifications

Abstract

Chromatin is regulated at many different levels, from higher-order packing to individual nucleosome placement. Recent studies have shown that individual histone modifications, and combinations thereof, play a key role in modulating chromatin structure and gene activity. Reported here is an analysis of Arabidopsis histone H3 modifications by nanoflow-HPLC coupled to electrospray ionization on a hybrid linear ion trap-Fourier transform mass spectrometer (LTQ/FTMS). We find that the sites of acetylation and methylation, in general, correlate well with other plants and animals. Two well-studied modifications, dimethylation of Lys-9 (correlated with silencing) and acetylation of Lys-14 (correlated with active chromatin) while abundant by themselves were rarely found on the same histone H3 tail. In contrast, dimethylation at Lys-27 and monomethylation at Lys-36 were commonly found together. Interestingly, acetylation at Lys-9 was found only in a low percentage of histones while acetylation of Lys-14 was very abundant. The two histone H3 variants, H3.1 and H3.2, also differ in the abundance of silencing and activating marks confirming other studies showing that the replication-independent histone H3 is enriched in active chromatin.

Figure 1

Cleavage of histone H3 for analysis by mass spectrometry. Schematic representation of the protocol used to generate tryptic peptides from residues 1–50 of histone H3. The resulting fragments contained residues 3–8, 9–17, 18–26, 27–40 and 41–50. Circles and squares above the line represent hypothetical sites of methylation and acetylation, respectively. Diamonds indicate that peptide N-termini and lysine epsilon amino groups have been derivatized with a C₂H₅CO-moiety (propionylation). Monomethyl lysines are also propionylated, but dimethyl-, trimethyl- and acetylated lysines are not.

Figure 2

Analysis of peptide isoforms containing residues 9–17 of histone H3. (A) Percent relative abundances observed for eight different isoforms of the tryptic peptide containing residues 9–17 of histone H3. The amino acid sequence for residues 9–17 is shown below the graph with K9 and K14 labeled. The eight isoforms include the unmodified sequence plus peptides containing mono-, di- and tri-methylated K9, acetyl-K14, acetyl-K9 plus acetyl-K14, dimethyl-K9 plus acetyl-K14 and monomethyl-K9 plus acetyl-K14. Percent relative abundance data was generated by mass spectrometry and is calculated by dividing the ion current (peak area) observed for each peptide by the total ion current (peak areas) for all fragments containing residues 9–17. ^*Peptide isoforms containing a single site of acetylation at K14 or K9 could not be separated by HPLC and co-eluted into the mass spectrometer. Analysis of the MS/MS spectra recorded on the mixture indicated that the two peptides were present in the following ratio; AcK14/AcK9 = 9/1. (B) Dot blots determining specificity of α-AcK9 antibodies using AcK9 peptide (1–12), AcK14 peptide (8–19), and Ler nuclear extract using either AcK9 or AcK14 peptide as a competitor. (C) Western blot of Ler nuclear extract probed with either α-AcK9 antibodies or α-AcK9 antibodies preincubated with AcK9 peptide competitor.

Figure 3

Analysis of peptide isoforms containing residues 18–26 of histone H3. Percent relative ion abundances observed for five different isoforms of the tryptic peptide containing residues 18–26, KQLATKAAR. Low ion abundance for the monomethylated peptide isoform made it impossible to assign the site of modification as either 1MK18 or 1MK23.

Figure 4

Analysis of peptide isoforms containing residues 27–40 of histone H3.1 and H3.2. Percent relative ion abundances observed for nine peptide isoforms of the tryptic peptide containing residues 27–40, (A) KSAPATGGVKKPHR from histone H3.1 and (B) KSAPTTGGVKKPHR from histone H3.2.

Figure 5

Comparative analysis H3 peptide isoforms (residues 9–17) derived from wild-type (Ler) and mutant (kyp) Arabidopsis tissues. (A) Diagram of the method employed to derivatize and isotopically label peptides isolated from the above two sources. H3 peptides from the kyp mutant were converted to d₅-ethyl esters and those from Ler (wild-type) were converted to d₀-ethyl esters. A mixture containing equal amounts of these two samples was analyzed by mass spectrometry. (B) Total ion chromatogram and individual mass spectra that highlight the ion abundances observed for the 2MK9 and 41–50 peptides isolated from the two plant tissues. Theoretical and experimentally measured masses for the doubly charged 2MK9 and 41–50 peptides obtained from the Ler sample are also shown. (C) Plot of ion abundance ratios (kyp/Ler) observed for peptide isoforms containing residues 9–17.