globalassessmentofcombinatorialpost-translational ... · dem mass spectrometry of intact histones...

Global Assessment of Combinatorial Post-translationalModification of Core Histones in Yeast Using ContemporaryMass SpectrometryLYS4 TRIMETHYLATION CORRELATES WITH DEGREE OF ACETYLATION ON THESAME H3 TAIL*□S

Received for publication, May 22, 2007, and in revised form, July 17, 2007 Published, JBC Papers in Press, July 24, 2007, DOI 10.1074/jbc.M704194200

Lihua Jiang‡§1, Jonell N. Smith‡, Shannon L. Anderson§¶, Ping Ma§�, Craig A. Mizzen§¶2, and Neil L. Kelleher‡§3

From the ‡Department of Chemistry, ¶Department of Cell and Developmental Biology, �Department of Statistics,and §Institute of Genomic Biology, University of Illinois, Urbana, Illinois 61801

A global view of all core histones in yeast is provided by tan-dem mass spectrometry of intact histones H2A, H2B, H4, andH3. This allowed detailed characterization of >50 distinct his-tone forms and their semiquantitative assessment in the dele-tion mutants gcn5�, spt7�, ahc1�, and rtg2�, affecting thechromatin remodeling complexes SAGA, SLIK, and ADA. The“top down” mass spectrometry approach detected dramaticdecreases in acetylation on H3 and H2B in gcn5� cells versuswild type. For H3 in wild type cells, tandem mass spectrometryrevealed adirect correlationbetween increases of Lys4 trimethy-lation and the 0, 1, 2, and 3 acetylation states of histone H3. Theresults show a wide swing from 10 to 80% Lys4 trimethylationlevels on those H3 tails harboring 0 or 3 acetylations, respec-tively. Reciprocity between these chromatin marks was appar-ent, since gcn5� cells showed a 30% decrease in trimethylationlevels on Lys4 in addition to a decrease of acetylation levels onH3 in bulk chromatin. Deletion of Set1, the Lys4 methyltrans-ferase, was associated with the linked disappearance of bothLys4methylation andLys14 andLys18 or Lys23 acetylation onH3.In sum, we have defined the “basis set” of histone forms presentin yeast chromatin using a currentmass spectrometric approachthat both quickly profiles global changes and directly probes theconnectivity of modifications on the same histone.

In eukaryotic cells, genetic information is complexed withhistone proteins in a highly organized structure known as chro-matin. The nucleosome, the fundamental structural unit ofchromatin, contains an octamer formed by two copies of each

of the core histones H2A, H2B, H3, and H4, about which iswrapped 146 bp of DNA (1). The dynamics of chromatin struc-ture directly influence the accessibility of DNA, which in turnaffects gene transcription, replication, and recombination (2).Core histone proteins are evolutionarily conserved and haveflexible N-terminal tails protruding from the nucleosome.These tails are subject to a variety of post-translational modifi-cations (PTMs)4 such as phosphorylation, ubiquitination,methylation, and acetylation (3). Such PTMs change the inter-actions of histone proteins with DNA and chromatin remodel-ing complexes, which are responsible for diverse functionswithin the cell (4). Different combinations of covalent modifi-cations form the essence of what has been termed the histonecode (5).Acetylation is perhaps the best characterized PTM on his-

tones and generally is associated with increased DNA acces-sibility not only by the neutralization of the positive chargeson lysine residues but also as interaction sites capable ofrecruiting chromatin remodeling complexes (6). Previousstudies have shown that particular histones and particularsites within histones become acetylated under distinct phys-iological conditions (7). The selection of acetylation sitescould be accomplished either by a set of different histoneacetyltransferases (HATs) or by the fine tuning of HAT spec-ificity by additional factors. So far, a variety of proteins withHAT activity have been discovered in Saccharomyces cerevi-siae (6, 8), but their in vivo substrate specificities have onlybeen partially investigated by antibodies against some of theknown acetylation sites (9). It is still a major challenge todefine the global impact on each histone’s PTMpattern uponcellular perturbation (e.g. a HAT knock-out). A study ofHAT deletion effects using a method with less bias (i.e. onethat can visualize all abundant PTMs simultaneously in anunbiased fashion) and still provide site-specific PTM infor-mation will help our understanding of how HATs worktogether in vivo and illuminate the interplay between acety-lation and other PTMs.

* The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked “advertise-ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

□S The on-line version of this article (available at http://www.jbc.org) con-tains supplemental Figs. S1–S7.

1 Supported by a Chia-chen Chu Kang fellowship through the School ofChemical Sciences at the University of Illinois.

2 The laboratory of this author was supported by Roy J. Carver CharitableTrust Grant 04-76 and March of Dimes Basil O’Connor Award 5-FY05-1232.

3 The laboratory of this author was supported by National Institutes ofHealth Grant GM 067193, the Packard Foundation, the Alfred P. SloanFoundation, and a 2004 Camille Dreyfus Teacher-Scholar Award fromthe Dreyfus Foundation. To whom correspondence should beaddressed: Dept. of Chemistry, University of Illinois, 600 S. Mathews, 53Rodger Adams Laboratory, Urbana, IL 61801. Tel.: 217-244-3927; Fax:217-244-8068; E-mail: [email protected].

4 The abbreviations used are: PTM, post-translational modification; HAT, his-tone acetyltransferase; ECD, electron capture dissociation; RPLC, reversephase liquid chromatography; ESI, electrospray ionization; FTMS, Fouriertransform mass spectrometry; WT, wild type; HPLC, high pressure liquidchromatography; aa, amino acids; ChIP, chromatin immunoprecipitation;SAGA, Spt-Ada-Gcn5-acetyltransferase.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 282, NO. 38, pp. 27923–27934, September 21, 2007© 2007 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

SEPTEMBER 21, 2007 • VOLUME 282 • NUMBER 38 JOURNAL OF BIOLOGICAL CHEMISTRY 27923

by guest on July 7, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Gcn5 was first shown to be required for transcriptional acti-vation and later recognized to have HAT activity (10). To date,it is one of the best characterized HATs both structurally andfunctionally. Recombinant Gcn5 acetylates free H3 at Lys14preferentially andH4 at Lys8 and Lys16weakly in vitro, confirm-ing the selectivity of Gcn5 for certain lysine residues withinspecific histones (11). For histone H3, this specificity isextended to other sites (Lys9, Lys14, Lys18, and Lys23) whenGcn5-containing complexes, such as SAGA and ADA, are usedwith nucleosome substrates in vitro (12), indicating that itsHAT activity is regulated by interacting proteins. Later, theGrunstein laboratory used highly specific antibodies to reexam-ine the substrate specificity of Gcn5 in vivo (9). Their studydemonstrated that Gcn5 is required for acetylation at Lys11 andLys16 of H2B and Lys9, Lys18, Lys23, and Lys27 of H3 with theexception of Lys14. In the course of these studies, almost noeffect on H4 (9, 13) was found, whereas others support thatGcn5 deletion can affect acetylation of H4 in addition to H3(14). These results are informative but do not reveal if the acety-lation sites affected by Gcn5 are on the same histone moleculeor whether they are affected equivalently, because antibodiescan only detect one PTMat a time and different antibodies havedifferent affinities. Even when antibodies showed no change ofmodification at one site, there could be change of combinationsof modifications. Due to the limitations of antibodies, a moreinformative and precise method is necessary to fully under-stand the substrate specificity of Gcn5.Previous studies have shown a direct correlation between

Gcn5 HAT activity and activation of specific model genes,including HIS3, PHO5, and HO (15–17). A more comprehen-sive understanding of Gcn5 HAT activity utilized microarrayanalysis to show expression differences for over 60% of genes inyeast (18, 19). A combination of serial analysis of gene expres-sion and chromatin immunoprecipitation methods has shownthat Gcn5 plays a global role in regulating histone H3 and H4acetylation (20). However, direct measurements of the effect ofGcn5 on the global levels of PTMs on all core histone proteinshave not been described.Here, we used top downmass spectrometry (21, 22) to exam-

ine the global effects of Gcn5 and eight other mutants on theexpression and modification of all four core histone proteins.We first established the “basis set” of all abundant PTM com-binations on intact histones in wild type cells, including site-specific PTM localization, using electron capture dissociation(ECD) for tandem mass spectrometry (23, 24). Recently, thisgeneral approach has been demonstrated on diverse forms ofhuman histones (25–27), with quantitative determination ofspecific combinations of modifications and PTM occupancyat specific sites (including the ability to resolve “positionalisomers” of acetylation for H4) (28). Besides characterizingall core histones, ECD analysis of the H3 tail from wild-typechromatin also revealed a correlation between methylationand acetylation; there is a wide swing from 10 to 80% Lys4trimethylation levels on the species with 0 and 3 acetylationstates, respectively.With the top down perspective, global alteration of histone

modification in gcn5� cells was striking and readily apparent;levels of hyperacetylated forms of H2B and H3 were greatly

diminished, whereas acetylation of H4 and H2A was onlyslightly affected. gcn5� cells showed reduced acetylation at spe-cific sites inH3 and a 30% decrease in the overall trimethylationlevel on Lys4, underscoring the reciprocity between these twochromatin marks. Set1 mutants also showed the loss of acety-lation at Lys14 and Lys18 or Lys23 on the same molecules whereLys4 trimethylation was also lost. Results from eight othermutants are also presented andhighlight the potential to profilea large number of conditions for their effects on bulkchromatin.

EXPERIMENTAL PROCEDURES

Yeast Cell Culture—S288C S. cerevisiae cells (MATa his3�leu2� met15� ura3�) with the interruption of Gcn5, Ahc1,Rtg2, Ada1, Spt7, Ard1, Nat1, or Nat4, respectively, were pur-chased fromOpen Biosystems. All of the strains were grown inYPD (1% yeast extract, 2% bactopeptone, 2% glucose, pH 7) at30 °C to midlog phase until A600 � 0.6 units. Growth curves ofwild type and gcn5� cells were established by recording celldensity at A600 every 1.5 h. Cells were harvested by centrifuga-tion at 4 °C for 5 min at 1000 � g and washed twice with sterilewater before storage of the cell pellets at �80 °C.Set1 mutant MBY1217 (MAT� his3�200 ade2::hisG leu2�

ura2�0 met15�0 trp1�63 Ty1his3AI-236r, Ty1ade2AI515set1�::TRP1) and the parental strain were obtained from thelaboratory of Scott Briggs (Purdue University). Construction ofplasmids expressing full length Set1-(1–1080) and theN-termi-nal truncated Set1-(829–1080) were described in previousstudies (29, 30). Cells were grown on SC medium (0.67% (w/v)yeast nitrogen base supplemented with amino acids, 2% glu-cose) lacking uracil to midlog phase until an A600 of 0.6 units.Cells were harvested as described above.Histone Extraction and Separation—Frozen cell pellets were

thawed on ice and extracted three times with 2.5 volumes ofyeast protein purification reagent (Yper; Pierce), 0.1 M dithio-threitol, 5 nMmicrocysteine (Sigma), 500�M4-(2-aminoethyl)-benzenesulfonyl fluoride (Roche Applied Science) and 100 mMsodium butyrate. Briefly, resuspended cells were incubated onice for about 30 min, lysates were clarified by 5 min of centrif-ugation at 18,000� g at 4 °C, and the supernatantwas removed.Following two additional repetitions, the cytoplasm-reducedcell pellet was then washed three times with sterile water toremove excess Yper. Histone extraction was performed twiceby incubation of the cell pelletwith 2.5 volumes of 8Mdeionizedurea and 0.4 N sulfuric acid for 30 min on ice. The sample wasspun down again, and clarified supernatants were pooled andimmediately loaded onto an equilibrated disposable C4 column(J. T. Baker Inc. BAKERBOND sptTM wide pore butyl (C4) dis-posable extraction column). After washing the column with 10volumes of buffer A (0.1% trifluoroacetic acid in distilled H2O),histones were eluted with buffer B (60% acetonitrile, 0.1% trif-luoroacetic acid). Acetonitrile was then removed from the col-lected fractions using a SpeedVac.Before further separation by RPLC, the sample was oxidized

by incubating at room temperature in 3% formic acid andH2O2for 4 h (31). Oxidized protein was loaded onto a 4.6-mm �250-mm C8 or C18 Vydac column. Proteins were fractionatedin 100 min from 30 to 60% B (A, 5% CH3CN with 0.1% triflu-

Characterization of Yeast Histones by Top Down MS

27924 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 38 • SEPTEMBER 21, 2007


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

oroacetic acid; B, 90%CH3CNwith 0.094% trifluoroacetic acid)and detected by absorbance at 214 nm. Histone proteins wereeluted between 55 and 95 min (Fig. 1A). Every fraction wassubsequently concentrated in a SpeedVac.Different acetylation states of H3 were further separated by

acid urea gel electrophoresis using gels composed of 15%TritonX-100, acetic acid, urea (typically 1.5-mm thick and 13 � 14cm2) according toMullen et al. (32), except that gels were run at250 V for 20 h to obtain better resolution of H3. Proteins werevisualized by Coomassie Blue staining. A duplicate gel wastransferred to polyvinylidene difluoride and probed with anti-Lys4 trimethylation antibodies (Abcam).H3Glu-CDigestion andRPLCSeparation—H3purified from

RPLC was dried down and resuspended in 50 �l of 25 mMNH4HCO3 buffer (pH 4). Glu-C (Roche Applied Science) wasadded at a ratio of Glu-C/H3 of 1:10 (w/w) and incubated atroom temperature for 4–5 h. The reaction was stopped byfreezing the sample. H3 digests were loaded to a 4.6 mm � 250mm C18 Vydac column pre-equilibrated with 0.2% trifluoro-acetic acid. Peptides were fractionated in 60 min from 0 to 60%

B (90% acetonitrile with 0.188% tri-fluoroacetic acid) and detected byabsorbance at 214 nm (Fig. 3A).Mass Spectrometry: ESI/Quadru-

pole-FTMS—Purified and dried his-tone sampleswere resuspended in 50�l of ESI solution (50% acetonitrile,49% H2O, and 1% formic acid), and10 �l was loaded to a nanosprayrobot (Advion BioSciences, Ithaca,NY). The system used in this studywas a custom 8.5 T quadrupole-FTMS hybrid described elsewhere(24). Ions from ESI were directedthrough a heated metal capillaryand multiple ion guides into the ioncell (�10�10 torr) of the FTMS sys-tem. Theoretical isotopic distribu-tions were generated using IsoProversion 3.0 and fit to experimentaldata by least squares to assign themost abundant isotopic peak. Typi-cally, 10–50 scans were used to col-lect broadband data. Samples werealso analyzed using a quadrupole-based selection strategy. In short,targeted peaks were selected by aquadrupole filter before accumula-tion (1–2 s) in an external octupoleand transfer to the ion cell. Ionswere fragmented by ECD (23). Mostspectra were stored as 512,000 datapoints and processed using zero orone truncation, no zero fill, andHanning apodization. MS/MS datawere analyzed by amodified versionof the THRASH algorithm. Individ-ual histone proteins were identified

and characterized by uploading data onto the prosightptmWorld Wide Web server to search the yeast data base.Statistical Comparison with ChIP-on-chip Data—To cross-

validate our findings, we compared our results with publiclyavailable data. In a recently published paper, genome-wide his-tone acetylation levels of H3 at Lys9 and Lys14 and methylationlevels of H3 at Lys4 of wild type yeast in rich medium (YPD)were measured using ChIP-on-chip (33). The average expres-sion ratios of immunoprecipitatedDNA to control DNA acrossexperimental replicates were downloaded from the WorldWideWeb. After a log2 transformation of Lys9 and Lys14 acety-lation level and Lys4 trimethylation level, a plot comparing theacetylation (Lys9 and Lys14) and methylation (Lys4) levels at3,581 intergenic regions was obtained for subsequent analysis(34).

RESULTS

Histone Extraction and Purification—Extraction of histonesfrom yeast cells is widely known to be more problematic thanfrom mammalian cells. Waterborg and colleagues (35) have

FIGURE 1. HPLC separation and MS detection of intact histone H4 and H2A from WT and gcn5� cells.A, chromatogram of WT whole cell histone extract. Proteins were fractionated by reverse phase HPLC, and elutionwas monitored at absorbance of 214 nm. B and C, broadband MS data of WT H4 and H2A. D and E, broadband MSdata of gcn5� H4 and H2A. The asterisk indicates an artificial phosphate adduct (�98 Da) of the 1 Ac form. 1 Ac isN-terminal acetylation.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

described a quick method for yeast histone preparation, but inour hands the yield is less than 50% of expected (data notshown). To address this limitation, we designed a simplifiedprocedure that relies on the observation that yeast chromatin isnot readily solubilized when the cell wall is intact. Thus, weused a mixture of nonionic detergents to deplete cells of thenonhistone proteins and then used conventional acid extrac-tion in the presence of 8 M urea to prepare crude histone fromthe residual cell pellet. By using this method, more than 95%histone recovery histone was confirmed through Western blotanalysis using an antibody to hyperacetylated H4 (data notshown). Oxidation of samples before RPLC separation helpedto resolve each histone to a single chromatographic peak (Fig.1A), with H3 giving two peaks later during the RPLC run (seebelow). The PTMprofiles obtained from intact histones did notchange appreciably upon repeated sampling, confirming thatdifferentially modified forms of intact histones co-elute duringreversed phase separation (35); this enables quantitative deter-mination of PTMprofiles using top downMS (25–27). ForWTand mutant cells described below, results are presented forintact profiles first, followed by site-specific PTM localizationusing MS/MS.Effects of Gcn5 Deletion on H4 Acetylation—To date, there

have been inconsistent results when assessing the effect ofGcn5 onhistoneH4. By using antibodies specific for each acety-lated site, Roth and co-workers (14) showed that deletion ofGcn5 caused a diminution of acetylation at each of the fouracetylation sites in H4. Other studies have shown that Gcn5does not acetylate H4 in vivo and in vitro (9, 13). In order toclarify these issues, we compared H4 modification forms fromwild type and gcn5� cells. To begin, we acquired high resolu-tion, intact mass spectra of histone H4 (Fig. 1B). The H4 mod-ification profile showed that themajority ofH4 (�85%) is acety-lated with di- and triacetylated forms as the dominantmodification forms, which is consistent with a much largerfraction of the yeast genome actively being transcribed versusmammalian histoneH4 (22).Gcn5 deletion results in very slightreduction in global H4 acetylation (Fig. 1, B versus D). Thisresult is consistently reproducible (n � 5) and appears to benear the limit of the FTMS-based assay to reliably detectchanges in PTM patterns. These results indicate that Gcn5deletion has only a marginal effect on the global H4 acetylationpattern in vivo.To determine which lysine sites were acetylated, we per-

formed ECD on each acetylation state (states 1–5) from WTand gcn5� cells. In WT cells, the 1 Ac form was N-terminallyacetylated (data not shown). Modifications on the 2 Ac formwere localized to the N terminus and Lys16. For the 3 Ac form,Lys12 was acetylated in addition to the initial two acetylations(supplemental Fig. S1, A and B). The 4 Ac form was acetylatedat Lys8 in addition to acetylation sites in the 3 Ac form (supple-mental Fig. S1C). In the 5 Ac form, N terminus, Lys16, Lys12,Lys8, and Lys5 were all acetylated (data not shown). All acetyla-tion forms were found to be pure species; no acetylation iso-mers were detected (28). The directionality of acetylation isrigorously a C3/N-terminal zipper (i.e. Lys163/Lys123/Lys83/Lys5) in both WT and gcn5� cells. We conclude that Gcn5deletion only slightly reduces the relative abundance of each

acetylation formonH4 and does not affect residue specificity ofacetylation.Gcn5 Deletion: Effect on H2A—There are two H2A gene

products, H2A.1 and H2A.2, and one H2A.Z variant in yeastcells. H2A.1 and H2A.2 have exactly the same mass and thesame protein sequence, apart from a C-terminal switch of anAla-Thr in H2A.1 to a Thr-Ala in H2A.2. Combining the

FIGURE 2. Mass spectra of intact H2B from WT and gcn5� cells. A andB, broadband MS of H2B modification patterns in WT and gcn5� cells, respec-tively. C and D, graphical fragment maps for ECD MS/MS results of the 1 Acform of H2B.1 and H2B.2, respectively. All of the z ions are unmodified, and cions have a mass difference of �42 Da from the theoretical c ions, indicatingN-terminal acetylation. E, reduced representation of H2B N-terminalacetylation.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

intact protein profile (Fig. 1C) and ECD data of each formreveals that the most abundant form is N-terminally acety-lated, followed by the form with N-terminal and Lys7 acety-lation. The formwith three acetylations is present at very lowabundance, and after digestion with Glu-C, the site waslocalized to Lys4 (data not shown). Most studies suggest thatGcn5 has no effect on H2A acetylation, but others have sug-gested that Gcn5 may have some role in H2A acetylation (9).Our data show no detectable changes in the acetylation levelof H2A upon Gcn5 disruption (Fig. 1, C versus E). H2A.Z wasshown to be present in two adjacent nucleosomes in mostyeast gene promoters (36), but H2A.Z levels are too low fordetection by theMS approach employed here. However,Mil-lar et al. have shown that Gcn5 affects acetylation on Lys14 ofH2A.Z (37).H2B from WT and gcn5� Cells—There are two H2B forms

present in yeast, H2B.1 and H2B.2, with a mass difference of 15Da. Previous studies revealed that 6 lysines can be acetylated oneach gene product, with the specific HATs responsible for indi-vidual acetylations not yet all known. Using top downMS, up tofive acetylations were observed on H2B.1 and H2B.2 (Fig. 2A).Due to the low abundance of the forms with six acetylations,they were not obvious in the “broadband” FT mass spectrum.However, after signal enhancement using selected ion accumu-lation, which increases the dynamic range by�50-fold (21, 38),these species were detectable (data not shown). For bothH2B.1and H2B.2, monoacetylation was the most abundant form, andthe abundance of other forms decreased with increasing acety-lation.Using top downMS tomonitor the relative abundance ofall forms at once, a spectrum of H2B from gcn5� cells showedstrikingly that all forms with �1 acetylation were undetectable

(Fig. 2B) except for a minor amountof 2 Ac forms, even when selectedion accumulation was used (datanot shown).Previous studies have shown that

Lys6, Lys11, Lys16, Lys17, Lys21,Lys22, and the N terminus of H2Bcan be acetylated. Tandem massspectrometry (MS/MS) using elec-trons for ion dissociation was per-formed on the mono-, di-, tri-, andtetra-acetylated forms. ECD data ofthe monoacetylation forms fromWT and gcn5� cells showed thatboth were acetylated exclusively atthe N terminus (Fig. 2, C and D).ECD of the diacetylation form fromthe WT sample indicated the pres-ence of acetylation “isomers” withpartial occupancy on Lys16 andLys16 (supplemental Fig. S2A). Thepercentage of each isomer wasdetermined by comparing the abun-dance of fragment ions that arisefrom backbone cleavages betweenthese sites (28). The data show thatthe second acetylation was split,

with 60% occupancy on Lys22 and 40% occupancy on Lys16(with the first acetylation 100% on the N terminus). The samemethod applied to triacetylated H2B showed that the N termi-nus and Lys16 were acetylated at 100% occupancy, whereas thethird acetylation was split 60% at Lys11 and 40% at Lys22 (Sup-plemental Fig. S2B). In the tetra-acetylated forms, all of theLys11, Lys16, and Lys22 were acetylated (supplemental Fig. S2C).Due to the low abundance of the 5 Ac and 6 Ac forms, reliableECDdata have not been acquired to date. These results indicatethat acetylation of Lys6, Lys11, Lys16, Lys17, and Lys22 inH2B arelost upon deletion of Gcn5.WT H3 and Effects of Gcn5 Deletion—For WT S. cerevisiae

extracted with our new procedure, two H3 peaks were consis-tently observed in all of our RPLC chromatograms (Fig. 1A, farright), whereas other histone extractionmethods only yield oneH3 peak (35). FT mass spectra showed that the modificationprofiles of these two HPLC peaks are different (supplementalFig. S3). The reason for the elution differences between thesetwo peaks is unknown at the current time. Considering that thenominal mass of three methylations is the same as one acetyla-tion (�42 Da), these peaks could differ in the levels of differentmodifications and/or their site isomers. Thus, to simplify sam-ple comparisons, we combined these two H3 HPLC peaks (dis-played separately in supplemental Fig. S3) for all sample com-parisons reported below.As shown in Fig. 3, A and B, the modification profiles for H3

in gcn5� cells showed that the relative abundance of peaks with�m values of �126, �168, and �210 Da above the unmodifiedform decreased detectably. Consistent with a large body of lit-erature on Gcn5 deletion, it seemed likely that these decreaseswere attributable primarily to H3 hypoacetylation. However,

FIGURE 3. Change of modification profiles of H3 assessed by MS, acid-urea electrophoresis, and Westernblotting. A and B, MS of intact H3 modification profiles in WT and gcn5� cells. C, acid urea gel of intact histoneH3 from WT and gcn5�. D, Western results from using an anti-Lys4 trimethylation antibody to visualize ananalogous acid urea gel analogous to that in C.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

recently Zhang et al. (39) showed that histone methylation isalso correlated with acetylation, suggesting that the changecould be caused by decreased methylation on Lys4 or Lys79 aswell. In order to assess this with an orthogonal method, wecompared these MS profiles with the resolution achieved onacid urea gels, well known for the ability to resolve histonesaccording to acetylation state (32). Based on the Coomassie-stained gel of Fig. 3C, H3 from gcn5� cells clearly has less acety-lation versus that fromWT cells. To correlate these acetylation

states with trimethylation at Lys4(K4me3), the anti-K4me3 antibodywas used to probe a transfer from ananalogous gel. This suggested thatless K4me3 occurred in gcn5� cells(Fig. 3D), but extracting quantita-tive information about the specificacetylation states was challenging.In order to quantify the relationshipbetween acetylation and methyla-tion states, we performed ECD onthe bulkH3 from bothwild type andgcn5� cells.

With direct ECD fragmentationof all intact H3 isoforms en masse,each acetylation site is difficult toassess in isolation from one another(27). Therefore, we used endopro-teinase Glu-C to liberate the H3peptides corresponding to the 1–50aa (40), 51–59 aa, and 74–94 aa seg-ments, which collectively containalmost all knownmodification sites.Mass spectrometry identified onemajor peak in the chromatogramcorresponding to H3 1–50 aa inbothWT and gcn5�H3 digests (Fig.4A). In addition to the major HPLCpeak at �30 min, two “satellite”peaks at 30.5 and 31.5 min wereidentified to be heavily acetylatedforms of the 1–50 aa piece of WTH3 (see Fig. 4B). In Glu-C digests ofH3 from gcn5� cells, both satellitepeaks were typically missing, withthe 30.5 min peak observable only ifwe doubled our conventional load-ing. Although these two satellitepeaks are low in abundance (�10milliabsorbance units at 214 nm),the data are reproducible. To fur-ther characterize the modificationson those heavily modified species,we combined peaks 2 and 3 and per-formed bulk ECD on all of theseforms together. Bulk ECD results onthe hypermodified forms in thesefractions showed that 55% of Lys4and 20% of Lys36 were trimethyl-

ated, 75% of Lys18 was acetylated, and all of the other lysineresidues (Lys9, Lys14, Lys23, and Lys27) were 100% modified(supplemental Fig. S5).Fig. 4C shows that the change of modification profiles on the

1–50 aa piece are similar to those observed for intact H3 uponGcn5 deletion (Fig. 3,A andB). The relative abundance of peakswith �84, �126, and �168 Da were decreased dramatically ingcn5� cells, whereas the �42 Da peak remains much the same.ECD of the �42 Da peak of WT and gcn5� cells showed that it

FIGURE 4. MS results of 1–50 aa of H3 from WT and gcn5� cells. A, HPLC profiles of Glu-C digested H3 fromWT and gcn5� cells. The top trace is from WT H3, and the bottom one is from gcn5� H3. Peaks 1–3 contain the1–50 aa of H3. Peak 3 was not observed in the gcn5� cells. B, MS of three different HPLC peaks that contain 1–50aa of Glu-C-digested WT H3. C, MS results 1–50 aa of pooled H3 HPLC fractions from WT and gcn5� cells. D, thechange of Lys4 trimethylation level at 0, 1, 2, and 3 acetylation states of WT H3. Modification level on Lys4, Lys9,Lys14, Lys18, and Lys23 from 0 Ac to 3 Ac state are listed in Table 1. These results are based on pooled H3 HPLCfractions 1–3. Asterisks, noise spikes.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

is amixture of 20% of trimethylation on Lys4 and 80% trimethy-lation on Lys36. We further characterized the �84, �126, and�168 Da peaks in WT cells by ECD. The results showed thatthere is amixture of “PTM isomers” in each peak. From the�42Da peak to the �168 Da species, we observed that Lys4 trim-ethylation increased from 10 to 80% (Fig. 4D). Specifically, theMS/MS-based assay quantified the amount of Lys4 trimethyla-tion on the 0, 1, 2, and 3 acetylation states as 10, 55, 75, and 80%,respectively (Table 1 ). In going from 0 to 3 acetylations, thedistribution of acetylation occupancies for each discrete statewas semiquantitatively determined (Table 1). These resultsrevealed that higher degrees of trimethylation level on Lys4 areassociated with higher degrees of acetylation.

To streamline analysis of H3 1–50 aa from gcn5� cells, wecombined all HPLC peaks of 1–50 aa and did bulk ECD. Thisshowed that the acetylation levels on Lys9 (Fig. 5A) and Lys14(supplemental Fig. S4A) each decreased by �40%. The acetyla-tion levels on Lys9 and Lys14 were determined through analysesof the c10 and c17 fragment ions, respectively. Although themass changes due to trimethylation and acetylation are veryclose, they are readily distinguished by accurate mass measure-ments on the high resolution FTMS. As seen in Fig. 5B, the�42Da peak of c10 showed peak splitting. The lowm/z peak repre-sents acetylation from Lys9, and the high m/z peak representstrimethylation from Lys4. Overall, there is �10% acetylationoccupancy for Lys18 and Lys23 (supplemental Fig. S4, B and C)inWTcells, but neither of themwere observed in gcn5� cells. Achange in the level of methylation at Lys4 and Lys79 was alsoobserved. Trimethylation on Lys4 decreased about 30% (Fig.5C), whereas trimethylation on Lys79 increased about 25% (Fig.5D). Acetylation on Lys56 (Fig. 5E) and methylation on Lys36and Lys27 remained the same in gcn5� cells (supplemental Fig.S4,D and E). The quantitative change of acetylation and meth-ylation levels on each lysine residue is summarized in Fig. 6.SinceGcn5 is aHAT, the decrease of overall acetylation level onH3 is expected in �gcn5 cells. Here, the observed decrease ofLys4 trimethylation level also demonstrates that Lys4 methyla-

tion and acetylation levels onH3 areregulated in reciprocal fashion invivo.Set1 Mutants—To further char-

acterize the relationship betweentrimethylation at Lys4 and acetyla-tion on the same H3 molecule, weexplored the changes in PTM pat-terning on H3 from different Set1mutant cells. Set1 has been shownto be the sole methyltransferaseresponsible for H3 Lys4 mono-, di-,and trimethylation in yeast (29,41–44). Its N-terminal truncatedform (829–1080 aa) lacks trimethy-lation ability on Lys4 but still main-tains normal levels of mono- and di-methylation (30).Fig. 7, A and B, shows mass spec-

tra of intact H3 and Glu-C-digestedH3 (1–50 aa), respectively, from theparental strain and Set1 mutants.Although the broadband MSshowed dramatic changes of modi-fications on H3 for the mutants, ionfragmentation was required tolocalize the changes to specific sites.Bulk ECD was performed on the1–50 aa piece of H3 as describedabove, and the results from signa-ture fragment ions reporting onlysines 4, 9, 14, 18, and 23 are sum-marized graphically in Fig. 7D andsupplemental Fig. S6. The c4 frag-

FIGURE 5. Comparison of the modifications at different lysine residues of H3 in WT and gcn5� cells.A, change of modification on fragment ion c10, which contains modification information on Lys4 and Lys9.B, peak splitting of the c10 �42 Da peak in A due to the presence of both acetylation (42.0106 Da) andtrimethylation (42.047 Da). The low m/z peak represents acetylation from Lys9, and the high m/z peakrepresents trimethylation from Lys4. C–E, change of modification on Lys4, Lys79, and Lys56, respectively.

TABLE 1Modification level at different lysine residues on H3 from 0 Ac to 3 Acstate

Modifications �42 Da(0 Ac)

�84 Da(1 Ac)

�126 Da(2 Ac)

�168 Da(3 Ac)

% % % %K4me3 10 55 75 80Lys9 Ac 0 20 35 45Lys14 Ac 0 40 50 100Lys18 Ac 0 10 10 35Lys23 Ac 0 15 60 100




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

ment ion from ECD reporting on the Lys4 modification (Fig.7C) showed that in Set1-(829–1080) cells, Lys4 mono- and di-methylation still existed, but trimethylation totally disappeared(Fig. 7C,middle). In set1� cells, only unmodified forms of Lys4were observed; all of themethylated formswere absent (Fig. 7C,bottom). Fragment ion c10 (Fig. S6A) in combination with c4provides information on Lys9 and showed that Lys9 was acety-lated. Fragment ions c17 and c18 (Fig. S6,B andC) contain infor-mation on Lys14 and Lys18, respectively. Fig. S6B showed thatthe�84Dapeakwas not observed in Set1-(829–1080) cells anddecreased more than 50% in set1� cells. Through a procedurefor internal calibration to boost mass accuracy, the �84 Dapeak was shown to be a combination of Lys9 and Lys14 acetyla-tion in set1� cells, whereas it is mainly a combination of Lys4trimethylation and Lys14 acetylation in the WT cells. Fig. S6Cshowed that the �126 Da peak was missing in both Set1mutants, whereas it was observed in WT cells. The �126 Dapeak contains Lys4 trimethylation and acetylation at Lys14 andLys18. These results showed that Lys4 trimethylation is requiredfor the subsequent Lys14 and Lys18 acetylation. Since this par-ticular modification form that has Lys4 trimethylation andLys14 and Lys18 acetylation was not observed in both Set1mutants, it can be concluded that it is trimethylation and notmono- or dimethylation that predominantly engages the acety-lation machinery to place activating acetylation marks at Lys14and Lys18. No significant change of overall modification levelon Lys23, Lys27, and Lys36 was observed (supplemental Fig. S6).

Although in WT type cells, Lys23 is 100% acetylated when Lys4is highly trimethylated (Table 1), we did not obtain enoughinformation about the correlation between them from thesebulk ECD results because of the excessive complexity of thedata. In order to increase the dynamic range for the detection ofminor forms, we compared the ECD results of the �126 Dapeak of 1–50 aa of H3 from WT and set1� cells. As shown insupplemental Fig. S7, in set1� cells, the peak with Lys4 trim-ethylation plus Lys14 and Lys23 acetylation disappeared. Thisresult showed that with the loss of Lys4 trimethylation, Lys23cannot be added to this particular modification form.Gcn5-containing Protein Complexes—In vitro studies have

shown that Gcn5 alone does not acetylate H2B; however, in thecontext of protein complexes such as SAGA, itmight be capableof directly acetylating H2B or affecting the function of otherHATs, which could acetylateH2B. In addition to SAGA, severalother protein complexes, such asADAandSLIK, have also beendiscovered to contain Gcn5 (45). Gcn5 may have different oroverlapping functions in those protein complexes through pro-tein interactions. In order to verify that, we compared thechange of histonemodification profiles upon disruption of eachprotein complex. Previous studies have shown that Spt7 is anessential component of SAGA (46). Other proteins, such asAhc1 and Rtg2, are found to be required for the integrity ofADA and SLIK protein complexes (47). Using the methoddescribed here, histone profiles were obtained from each ofthese mutants. Fig. 8 reveals that although the relative abun-dance of different modification forms was altered comparedwithWT, H2B and H3 were still highly modified after knock-ing out one protein complex. This result suggests that con-siderable redundancy exists between Gcn5-containing pro-tein complexes or possibly other types of HATs that cancompensate for the loss of HAT function of these specifictypes of Gcn5 complexes.N-terminal Acetyltransferases—With onlyN-terminal acety-

lation of H2B left unaffected by Gcn5 knock-out, we investi-gated the effect of different N-terminal acetyltransferases onhistones. Previous studies have shown that Ard1 and Nat1 areH2B N-terminal acetyltransferases, with Nat4 known to acety-late H4 and H2A (32, 48). Top down MS data of histones fromnat1� cells showed no change in acetylation levels on H2B andother histones (data not shown). MS data from ard1� cellsrevealed a dramatic increase of unmodified H2B over themonoacetylated form (Fig. 8A, second row from top). MS/MSanalysis of this monoacetylated form of H2B by ECD showedresidual N-terminal acetylation exclusively (data not shown).This indicates that Ard1 is the major H2B N-terminal acetyl-transferase, and there exists one or more unknown N-terminalacetyltransferases. The H3 modification pattern was also dra-matically changed in ard1� cells (Fig. 8B, second row from top).MS data of histones from nat4� showed the increase of H4 andH2A unacetylated forms, consistent with prior studies (32, 48).

DISCUSSION

Histone Detection and MS/MS with Minimal Bias—This isthe first implementation of high resolution “top down” massspectrometry on yeast histones and allowed characterization ofmore than 50 distinct molecular forms arising from multiple

FIGURE 6. Change of acetylation (A) and methylation (B) levels at differ-ent lysine residues of histone H3 in WT and gcn5� cells.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

modifications of the yeast core histones from wild type andmutant cell lines. In addition to establishing a definitive “basisset” of expressed yeast histoneswithminimal bias (i.e. detectionof intact histone forms generally), this global view of bulk chro-matin allows quantitative assessment of all major forms of eachhistone upon mutation (or other perturbation), with MS/MSallowing direct connection of intact forms with a quantitativereadout of PTM occupancy at specific sites (28). Since this par-ticular approach determines both the changes of abundance formultiply modified forms and the localization of each modifica-tion to a specific amino acid residue, this study overcomes someof the limitations of antibodies by using MS of intact histones(and the N-terminal 1–50 aa piece of histone H3). Antibodiesare good for PTMs in isolation, but they generally cannot accu-rately account for changes involving combinations ofmodifica-tions. In addition, since different antibodies have differentaffinities, accurate quantitation of changes in the relative abun-dance of different PTMs is not possible. Although top downmass spectrometry can overcome these problems, one limita-tion relative to antibodies is that of limit of detection. Themethodology employed here has the ability to detect species at

0.1–1% relative abundance in bulkchromatin, with the quantitativepicture of histones precise to �5%using MS/MS data of moderatequality (28). For the direct compar-ison of histone H3 from WT andgcn5� cells, the MS results qualita-tively match those from acid ureagels and Western analysis ofselected PTMcombinations (Fig. 3).Effects of Gcn5 on All Core

Histones—Due to the inherent limi-tations of antibodies discussedabove, previous studies that weremostly based on antibodies did notshow a comprehensive view of theeffects of Gcn5 on all core histones.By using top down strategiesdescribed in this paper, we reexam-ined its effect on all core histoneswith minimal bias. The resultsshowed that knock-out of Gcn5 hadslight effects on H4 and H2A butdramatically decreased acetylationlevels on histone H2B and H3. Fur-ther investigation using tandemmass spectrometry localized thechange ofmodification to individualamino acid residues. The resultsdemonstrated that for H2B, Gcn5 isthemajorHAT that affects all acety-lation sites except for the N termi-nus. Ard1 was shown to be themajor N-terminal acetyltransferasefor H2B. On H3, not only acetyla-tion on Lys9, Lys14, Lys18, and Lys23were affected by Gcn5 but also the

methylation levels on Lys4 and Lys79. The correlation of meth-ylation and acetylation on H3 is discussed further below.Since Gcn5 exists in different protein complexes, such as

SAGA, SLIK, and ADA, it may have different effects on histoneproteins in different protein complexes. However, disruption ofeach individual protein complex did not show a dramaticchange of modification on H2B and H3. This may due to theredundancy of function among different Gcn5-containingcomplexes or possibly other HATs. Here, a “global” view pre-vents us from determining whether PTM changes are confinedto only a portion of the genome. However, with Gcn5 knockedout, HAT function of all of the protein complexes was lost,resulting in a dramatic decrease of acetylation on H2B and H3.Linking Methylation with Acetylation on the Same H3 Tail—

There are three lines of evidence that reveal the interplay ofacetylation and Lys4 trimethylation on the H3 tail, with thequantitation of this reciprocal relationship afforded by ourapproach a novel feature of this work. ForWTyeast histoneH3,the FTMS-based assay quantified the direct link betweenincreases of Lys4 trimethylation and the 0, 1, 2, and 3 acetylationstates of histone H3. The results reveal a swing from 10, 55, 75

FIGURE 7. Comparison of H3 modification at intact, Glu-C-digested and fragment level from WT, Set1-(829 –1080), and set1� cells. A and B, change of modification pattern on intact H3 and Glu-C-digested H3,respectively. C, change of Lys4 methylation in WT and Set1 mutants. D, representation of the change of mod-ifications on Lys4, Lys9, Lys14, Lys18, and Lys23 detected in MS/MS experiments from WT and set1� cells. In set1�cells, mono-, di-, and trimethylation on Lys4 disappeared. The forms that contain Lys4 trimethylation, Lys14

acetylation, and Lys18 or Lys23 acetylation were not observed in set1� cells.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

to 80% trimethylation levels on the0, 1, 2, and 3 acetylation states,respectively. This work thereforemeshes well with assertions inrecent reviews (49–52) and otherwork (39, 53) proposing a high cor-relation between Lys4 trimethyla-tion and hyperacetylation on H3.Our results from deletion of

gcn5� and set1� further character-ize the interplay between acetyla-tion and Lys4 trimethylation on H3and are largely consistent with pre-vious studies. In gcn5� cells, theexpected decrease of acetylationlevels (primarily on Lys9, Lys14,Lys18, and Lys23 of H3) were accom-panied by an overall drop of 30% intrimethylation level on Lys4. Zhanget al. (39) has shown that a higherlevel of methylation of Lys4 andLys79 is associated with hyperacety-lation on H3. Since the absence ofGcn5 results in decreased acetyla-tion, lower methylation levels onLys4 and Lys79 are expected. How-ever, our observation of a 20%decrease of trimethylation on Lys79in gcn5� cells is inconsistent withthe study of Zhang et al. (39). Theresults from Set1 mutants alsoshowed that loss of Lys4 trimethyla-tion, not mono- or dimethylation,caused the loss of detectable Lys14and Lys18 or Lys23 acetylation on thesameH3 tail. Interestingly, previousstudies have shown that some of the

Lys4 methyl readers reside within different chromatin remod-eling complexes that also have acetyltransferase activities.For example, Chd1 and Yng1, twomajor trimethyl Lys4 bind-ing proteins in yeast, are components of SAGA/SLIK andNuA3, respectively (54, 55). Loss of Lys4 trimethylation maycause the loss of binding sites for those protein complexes,which can acetylate Lys14 and Lys18 or Lys23 to further acti-vate transcription. These results are consistent with a morepotent role for Lys4 trimethylation (versusmono- or dimeth-ylation) in transcriptional activation.Integrating Diverse Data Types: Linking K4me3 with Acetyla-

tion at Thousands of Loci—Correlation of recent ChIP-on-chipresults (32) with our global histone modification readouts bymass spectrometry was initiated by creating the scatter plotbetween genome-wide acetylation levels at Lys14 and trimethy-lation levels at Lys4 of H3 in 3,581 intergenic regions of WTyeast cells (Fig. 9). The acetylation level of Lys14 and trimethy-lation level of Lys4 on H3 exhibit a strong positive linear rela-tionship (Pearson correlation coefficient 0.72) after taking log2transformation of their values. From our studies of acetylationsite usage, we found that Lys14 and Lys23 are the most occupied

FIGURE 8. Comparison of all core histone modification patterns in WT, ard1�, spt7�, rtg2�, and ahc1�cells. Columns A, B, C, and D compare H2B, H3, H4, and H2A, respectively, in all of the mutants.

FIGURE 9. The genome-wide acetylation levels of H3 at Lys14 versus trim-ethylation levels of H3 at Lys4 in 3,581 intergenic regions of WT yeastcells. Each data point represents an intergenic region. The x axis is the log2transformed trimethylation level of Lys4, and the y axis is the log2 transformedacetylation level of Lys14. The acetylation and methylation levels are obtainedfrom ChIP-on-chip results.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

acetylation sites and correlate tightly with the increase in Lys4trimethylation level (Table 1). The results showed that Lys9acetylation correlates less tightly with Lys4 trimethylation(Table 1); indeed, an analogous plot to Fig. 9 using Lys9 acety-lation gives a Pearson correlation coefficient at 0.42. Since Lys23acetylation has not been studied in these ChIP-on-chip exper-iments, the statistical results are unavailable, but fromourmassspectrometry results, we believe that it should also correlatewell with Lys4 trimethylation. The reciprocal relationshipbetweenLys14 acetylation andLys4 trimethylation has also beenshown in the gcn5� and set1� cells (discussed above). Onenovel feature of our work is that data in Fig. 7D and supplemen-tal Fig. S7 showed that Lys14 acetylation and Lys4 trimethyla-tion co-exist on the same H3 tail and that the loss of Lys4 trim-ethylation caused the linked disappearance of acetylation onLys14.Yeast provides an excellent system for studying the nature of

transcriptionally active chromatin, as a large percentage of thechromatin is in a transcriptional competent conformation (86%of the genome encodes genes). This makes it especially inter-esting to examine the levels of histone acetylations. In yeastcells besides Gcn5, there are many other HATs which alsoexhibit site and histone specificity. Currently, because of theadvantage of yeast genetics, the knock-out library makes itconvenient to study the effects of all yeast histone modifyingenzymes. In combination with the mass spectrometry andhistone extraction methods described here, it enables us toinvestigate the substrate specificity of all the known HATson histones, which will unravel how HATs work together toproduce specific acetylation patterns. The acetylation levelsare maintained by the balance of HATs and histone deacety-lases. However, until now the interplay of HATs and histonedeacetylases has remained unclear. The study of histonedeacetylase substrate specificity and the changes of acetyla-tion patterns after double mutation of certain HATs andhistone deacetylases will elucidate the correlations betweenthem.

Acknowledgment—We thank Dr. Scott Briggs for Set1 mutants.

REFERENCES1. Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F., and Richmond,

T. J. (1997) Nature 389, 251–2602. Margueron, R., Trojer, P., and Reinberg, D. (2005)Curr. Opin. Genet. Dev.

15, 163–1763. Zhang, Y., and Reinberg, D. (2001) Genes Dev. 15, 2343–23604. Jenuwein, T., and Allis, C. D. (2001) Science 293, 1074–10805. Strahl, B. D., and Allis, C. D. (2000) Nature 403, 41–456. Roth, S. Y., Denu, J. M., and Allis, C. D. (2001) Annu. Rev. Biochem. 70,

81–1207. Chicoine, L. G., Schulman, I. G., Richman, R., Cook, R. G., and Allis, C. D.

(1986) J. Biol. Chem. 261, 1071–10768. Sterner, D. E., and Berger, S. L. (2000) Microbiol. Mol. Biol. Rev. 64,

435–4599. Suka, N., Suka, Y., Carmen, A. A., Wu, J., and Grunstein, M. (2001) Mol.

Cell 8, 473–47910. Brownell, J. E., Zhou, J., Ranalli, T., Kobayashi, R., Edmondson,D.G., Roth,

S. Y., and Allis, C. D. (1996) Cell 84, 843–85111. Kuo, M. H., Brownell, J. E., Sobel, R. E., Ranalli, T. A., Cook, R. G., Ed-

mondson, D. G., Roth, S. Y., and Allis, C. D. (1996) Nature 383, 269–272

12. Grant, P. A., Duggan, L., Cote, J., Roberts, S.M., Brownell, J. E., Candau, R.,Ohba, R., Owen-Hughes, T., Allis, C. D., Winston, F., Berger, S. L., andWorkman, J. L. (1997) Genes Dev. 11, 1640–1650

13. Grant, P. A., Eberharter, A., John, S., Cook, R. G., Turner, B. M., andWorkman, J. L. (1999) J. Biol. Chem. 274, 5895–5900

14. Zhang,W. Z., Bone, J. R., Edmondson, D. G., Turner, B.M., and Roth, S. Y.(1998) EMBO J. 17, 3155–3167

15. Kuo, M. H., Zhou, J. X., Jambeck, P., Churchill, M. E. A., and Allis, C. D.(1998) Genes Dev. 12, 627–639

16. Gregory, P. D., Schmid, A., Zavari, M., Lui, L., Berger, S. L., and Horz, W.(1998)Mol. Cell 1, 495–505

17. Vogelauer, M., Wu, J. S., Suka, N., and Grunstein, M. (2000) Nature 408,495–498

18. Kurdistani, S. K., Tavazoie, S., andGrunstein,M. (2004)Cell 117, 721–73319. Huisinga, K. L., and Pugh, B. F. (2004)Mol. Cell 13, 573–58520. Roh, T. Y., Ngau, W. C., Cui, K. R., Landsman, D., and Zhao, K. J. (2004)

Nat. Biotechnol. 22, 1013–101621. Meng, F. Y., Cargile, B. J., Patrie, S. M., Johnson, J. R., McLoughlin, S. M.,

and Kelleher, N. L. (2002) Anal. Chem. 74, 2923–292922. Pesavento, J. J., Kim, Y. B., Taylor, G. K., and Kelleher, N. L. (2004) J. Am.

Chem. Soc. 126, 3386–338723. Zubarev, R. A., Horn, D. M., Fridriksson, E. K., Kelleher, N. L., Kruger,

N. A., Lewis, M. A., Carpenter, B. K., and McLafferty, F. W. (2000) Anal.Chem. 72, 563–573

24. Zubarev, R. A. (2004) Curr. Opin. Biotechnol. 15, 12–1625. Boyne, M. T., Pesavento, J. J., Mizzen, C. A., and Kelleher, N. L. (2006) J.

Proteome Res. 5, 248–25326. Siuti, N., Roth, M. J., Mizzen, C. A., Kelleher, N. L., and Pesavento, J. J.

(2006) J. Proteome Res. 5, 233–23927. Thomas, C. E., Kelleher, N. L., andMizzen, C. A. (2006) J. Proteome Res. 5,

240–24728. Pesavento, J. J., Mizzen, C. A., and Kelleher, N. L. (2006) Anal. Chem. 78,

4271–428029. Briggs, S. D., Bryk, M., Strahl, B. D., Cheung, W. L., Davie, J. K., Dent,

S. Y. R., Winston, F., and Allis, C. D. (2001) Genes Dev. 15, 3286–329530. Fingerman, I. M., Wu, C. L., Wilson, B. D., and Briggs, S. D. (2005) J. Biol.

Chem. 280, 28761–2876531. Pesavento, J. J., Garcia, B. A., Streeky, J. A., Kelleher, N. L., and Mizzen,

C. A. (2007)Mol. Cell Proteomics, in press32. Mullen, J. R., Kayne, P. S., Moerschell, R. P., Tsunasawa, S., Gribskov, M.,

Colavito-Shepanski, M., Grunstein, M., Sherman, F., and Sternglanz, R.(1989) EMBO J. 8, 2067–2075

33. Pokholok, D. K., Harbison, C. T., Levine, S., Cole, M., Hannett, N.M., Lee,T. I., Bell, G. W., Walker, K., Rolfe, P. A., Herbolsheimer, E., Zeitlinger, J.,Lewitter, F., Gifford, D. K., and Young, R. A. (2005) Cell 122, 517–527

34. Yuan, G. Z., Ma, P., Zhong, w., and Liu, J. S. (2006) Genome Biol. 7, 7035. Waterborg, J. H. (2000) J. Biol. Chem. 275, 13007–1301136. Meneghini, M. D., Wu, M., andMadhani, H. D. (2003) Cell 112, 725–73637. Millar, C. B., Xu, F., Zhang, K. L., and Grunstein, M. (2006)Genes Dev. 20,

711–72238. Senko, M. W., Canterbury, J. D., Guan, S., and Marshall, A. G. (1996)

Rapid Commun. Mass Spectrom. 10, 1839–184439. Zhang, K. L., Siino, J. S., Jones, P. R., Yau, P.M., and Bradbury, E.M. (2004)

Proteomics 4, 3765–377540. Garcia, B. A., Pesavento, J. J., Mizzen, C. A., and Kelleher, N. L. (2007)Nat.

Methods 4, 487–48941. Santos-Rosa, H., Schneider, R., Bannister, A. J., Sherriff, J., Bernstein, B. E.,

Emre, N. C. T., Schreiber, S. L., Mellor, J., and Kouzarides, T. (2002) Na-ture 419, 407–411

42. Nagy, P. L., Griesenbeck, J., Kornberg, R. D., and Cleary,M. L. (2002) Proc.Natl. Acad. Sci. U. S. A. 99, 90–94

43. Roguev, A., Schaft, D., Shevchenko, A., Pijnappel, W., Wilm, M., Aasland,R., and Stewart, A. F. (2001) EMBO J. 20, 7137–7148

44. Krogan, N. J., Dover, J., Khorrami, S., Greenblatt, J. F., Schneider, J., John-ston, M., and Shilatifard, A. (2002) J. Biol. Chem. 277, 10753–10755

45. Lee, K. K., Prochasson, P., Florens, L., Swanson, S. K., Washburn, M. P.,and Workman, J. L. (2004) Biochem. Soc. Trans. 32, 899–903

46. Wu, P. Y. J., and Winston, F. (2002)Mol. Cell. Biol. 22, 5367–5379




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

47. Pray-Grant, M. G., Schieltz, D., McMahon, S. J., Wood, J. M., Kennedy,E. L., Cook, R. G.,Workman, J. L., Yates, J. R., and Grant, P. A. (2002)Mol.Cell. Biol. 22, 8774–8786

48. Song, O. K., Wang, X., Waterborg, J. H., and Sternglanz, R. (2003) J. Biol.Chem. 278, 38109–38112

49. Sims, R. J., and Reinberg, D. (2006) Genes Dev. 20, 2779–278650. Millar, C. B., andGrunstein,M. (2006)Nat. Rev.Mol. Cell Biol. 7, 657–66651. Clayton, A. L., Hazzalin, C. A., and Mahadevan, L. C. (2006)Mol. Cell 23,

289–296

52. Ruthenburg, A. J., Allis, C. D., andWysocka, J. (2007)Mol. Cell 25, 15–3053. Taverna, S. D., Ueberheide, B. M., Liu, Y. F., Tackett, A. J., Diaz, R. L.,

Shabanowitz, J., Chait, B. T., Hunt, D. F., and Allis, C. D. (2007) Proc. Natl.Acad. Sci. U. S. A. 104, 2086–2091

54. Pray-Grant, M. G., Daniel, J. A., Schieltz, D., Yates, J. R., and Grant, P. A.(2005) Nature 433, 434–438

55. Taverna, S. D., Ilin, S., Rogers, R. S., Tanny, J. C., Lavender, H., Li, H. T.,Baker, L., Boyle, J., Blair, L. P., Chait, B. T., Patel, D. J., Aitchison, J. D.,Tackett, A. J., and Allis, C. D. (2006)Mol. Cell 24, 785–796




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

L. KelleherLihua Jiang, Jonell N. Smith, Shannon L. Anderson, Ping Ma, Craig A. Mizzen and Neil

THE SAME H3 TAILTRIMETHYLATION CORRELATES WITH DEGREE OF ACETYLATION ON

Histones in Yeast Using Contemporary Mass Spectrometry: LYS4 Global Assessment of Combinatorial Post-translational Modification of Core

doi: 10.1074/jbc.M704194200 originally published online July 24, 20072007, 282:27923-27934.J. Biol. Chem.

10.1074/jbc.M704194200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2007/07/30/M704194200.DC1

http://www.jbc.org/content/282/38/27923.full.html#ref-list-1

This article cites 54 references, 20 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M704194200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;282/38/27923&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/282/38/27923

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=282/38/27923&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/282/38/27923

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2007/07/30/M704194200.DC1

http://www.jbc.org/content/282/38/27923.full.html#ref-list-1

http://www.jbc.org/

globalassessmentofcombinatorialpost-translational ... · dem mass spectrometry of intact histones...

Documents