Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Nov 2007; 27(22): 7856–7864.
Published online Sep 17, 2007. doi:  10.1128/MCB.00801-07
PMCID: PMC2169161

Histone H3 K4 Demethylation during Activation and Attenuation of GAL1 Transcription in Saccharomyces cerevisiae[down-pointing small open triangle]


In mammalian cells, histone lysine demethylation is carried out by two classes of enzymes, the LSD1/BHC110 class and the jumonji class. The enzymes of the jumonji class in the yeast Saccharomyces cerevisiae have recently also been shown to have lysine demethylation activity. Here we report that the protein encoded by YJR119c (termed KDM5), coding for one of five predicted jumonji domain proteins in yeast, specifically demethylates trimethylated histone H3 lysine 4 (H3K4me3), H3K4me2, and H3K4me1 in vitro. We found that loss of KDM5 increased mono-, di-, and trimethylation of lysine 4 during activation of the GAL1 gene. Interestingly, cells deleted of KDM5 also displayed a delayed reduction of K4me3 upon reestablishment of GAL1 repression. These results indicate that K4 demethylation has two roles at GAL1, first to establish appropriate levels of K4 methylation during gene activation and second to remove K4 trimethylation during the attenuation phase of transcription. Thus, analysis of lysine demethylation in yeast provides new insight into the physiological roles of jumonji demethylase enzymes.

Histone posttranslational modifications are centrally involved in genome regulation. Histone methylation occurs at both lysine (K) and arginine residues and plays a key role in chromatin organization and transcriptional regulation (18, 26, 27). In particular, methylation at lysine residues of histone H3 and H4 during gene regulation has been characterized extensively and linked to either gene activation (e.g., H3K4, H3K36, and H3K79) or repression (e.g., H3K9, H3K27, and H4K20). Histone H3 lysine methylation in the yeast Saccharomyces cerevisiae has been reported to occur at three residues (K4, K36, and K79) (27, 39), and each correlates with gene activation. The methylation reaction is catalyzed either by many lysine-specific methyltransferases within the large SET domain family (all known sites except H3K79) or by the unrelated Dot1 (H3K79) (38).

Until recently, histone lysine methylation was speculated to be irreversible and therefore possibly a true epigenetic modification persisting through cell division. However, two families of demethylation enzymes capable of targeting methylated lysine residues have now been identified (4, 6, 9, 16, 22, 36, 42, 44-46). LSD1, a catalytic engine of multicomponent corepressor complexes, was the first demonstrated lysine demethylase, reversing dimethylated histone H3 lysine 4 (H3K4me2) (22, 36, 37). However, this enzyme is unable to remove methyl groups from trimethylated lysine because of the inherent limitation of its enzymatic reaction mechanism. In contrast, a second class of histone lysine demethylases, consisting of the JmjC domain-containing proteins, are able to demethylate not only di- but also trimethylated histone marks (4, 9, 16, 42, 46). Only recently have demethylases targeting the trimethylated form of H3K4 been characterized. The four members of the JARID family in mammalian cells contain a JmjC domain which is critical for their H3K4 demethylation activity (3, 14, 17, 21). These enzymes have been shown to have in vivo effects on H3K4 methylation status, both on a global scale and at specific genes where they also affect the transcription levels. Studies of H3K4 demethylases in lower organisms are not as far advanced as those of mammalian counterparts. Lid, the only homolog of the JARID family in Drosophila melanogaster, has been shown to have H3K4 demethylation activity in vivo on a global scale (7, 23, 34). The same is true for JMJ2, the fission yeast homolog (13), and YJR119c, the budding yeast homolog (24, 35, 43) of the JARID family. However, it remains to be established whether lysine demethylation has gene-specific regulatory functions in lower eukaryotes.

Here we perform a detailed analysis of the demethylation activity of S. cerevisiae Kdm5 both in vitro and in vivo. (Note that YJR119c has been termed Jhd2 in recent literature, but we will refer to this protein as Kdm5 in accordance with a new nomenclature system for histone-modifying enzymes [Tony Kouzarides, personal communication]). We investigate whether Kdm5 is involved in regulating H3K4 methylation levels at specific genes. Our results provide evidence for a role of the enzyme in both modulating the level of K4 methylation during the peak of transcription and removing the modification during attenuation as full gene repression is reestablished.


Yeast strains and plasmids.

Yeast strains used in this study are listed in Table S1 in the supplemental material. Gene deletions were performed as described previously (25). Recombinant Kdm5 was created by cloning the KDM5 open reading frame (ORF) using a TOPO TA cloning kit (Invitrogen). An Expand high-fidelity PCR system (Roche) was used to amplify and subclone the cDNA of KDM5 into a modified pFAST Bac HTA baculovirus vector containing N-terminal FLAG and six-His tags. The KDM5 ORF, with a 500-nucleotide upstream sequence and a 250-nucleotide downstream sequence, was similarly cloned using a TOPO TA cloning kit (Invitrogen) and subcloned into pRS314 using an Expand high-fidelity PCR system (Roche). FLAG tagging and amino acid substitutions were made using a QuickChange kit (Stratagene).

Affinity purification of recombinant Kdm5.

Flag-His-Kdm5 was expressed in 500 ml of baculovirus-infected Sf21 insect cells and harvested at 72 h (Wistar Institute Protein Expression Core). Cells were lysed with 50 ml insect lysis buffer KCl250 (50 mM HEPES, pH 8.0, 250 mM KCl, 10% glycerol, 0.1% NP-40, 1 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride [PMSF], protease inhibitors), sonicated, and incubated overnight at 4°C with 200 μl mouse anti-Flag (α-Flag) agarose beads (Sigma). Beads were washed five times with 20× volume insect wash buffer KCl500 (50 mM HEPES, pH 8.0, 500 mM KCl, 10% glycerol, 0.1% NP-40, 1 mM EDTA, 0.5 mM PMSF, protease inhibitors) and once with 20× volume demethylation buffer (50 mM HEPES, pH 8.0, 5% glycerol). The recombinant protein was eluted with 0.5 μg/μl 1× Flag peptide in demethylation buffer.

In vitro demethylation reaction and Western analysis.

Bulk calf thymus histones (4 μg; Sigma) or H3 peptides (100 ng; Upstate) were incubated with the indicated amounts of recombinant proteins in histone demethylase assay buffer [5% glycerol, 50 mM HEPES K, pH 8.0, 2 mM ascorbate, 1 mM α-ketoglutarate, 100 μM Fe(II), 0.2 mM PMSF, protease inhibitors] in a final volume of 11 μl at 37°C for 5 h. Incubation products were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting. Mixtures from the reactions that used histones and peptides as substrates were electrophoresed on 4 to 20% Tris-glycine and 12% NuPAGE gels, respectively (11, 20), transferred to PVDF membranes, and probed with methyl-specific antibodies (10). An anti-H3 antibody served as a loading control. The following antibodies were used for the Western blotting: H3K4me1 (ab8895; Abcam), H3K4me2 (ab7766; Abcam), H3K4me3 (07-473; Upstate), H3 (ab1791; Abcam), H3K36me3 (ab9050; Abcam), H3K79me2 (ab3594; Abcam), and FLAG-horseradish peroxidase (A-8592; Sigma). For Western analysis on global protein levels, whole-cell extracts were electrophoresed on 4 to 20% Tris-glycine gels, transferred to 0.2-μm nitrocellulose membranes, and probed with the following antibodies: anti-Set1 (from A. Shilatifard, Stowers Institute, MO), anti-FLAG-horseradish peroxidase (M2; Sigma), and anti-H3 (ab1791; Abcam).

ChIP and GAL1 and SUC2 inductions.

Chromatin immunoprecipitation (ChIP) experiments were carried out as described before (2, 12). For GAL1 inductions, cells were grown in yeast extract-peptone-dextrose (YPD) overnight, diluted in yeast extract-peptone-raffinose to mid-log phase, and grown for 2 to 3 h. Galactose was added to the media to a final concentration of 2%, and 2 h later glucose was added to a final concentration of 2%. For SUC2 inductions, cells were grown in YPD overnight, diluted to mid-log phase in the morning, and grown for 2 h. Cells were spun down, washed in yeast extract-peptone plus 0.05% glucose, and resuspended in the low-glucose media. Two hours later, glucose was added to the media for a final concentration of 2%. Cells were cross-linked with 1% formaldehyde for 5 min (histone ChIP) or 20 min (factor ChIP). Protein (1 to 2 mg) was immunoprecipitated as described previously (12). Antibodies used for ChIP were anti-H3 (ab1791; Abcam), anti-H3 monomethyl K4 (ab8895; Abcam), anti-H3 dimethyl K4 (ab7766; Abcam), anti-H3 trimethyl K4 (ab8580; Abcam), and anti-Set1 (from A. Shilatifard, Stowers Institute, MO). Inputs (1:100 dilutions) and eluates were amplified using an ABI 7900HT fast thermal cycler (Applied Biosystems) and the primer pairs described in Table S2 in the supplemental material. Each PCR consisted of 10 μl SYBR green dye I (Applied Biosystems), 0.1 μl forward primer (10 μM), 0.1 μl reverse primer (10 μM), 4.8 μl H2O, and 5 μl sample. The PCRs went through a program consisting of 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min. Each sample was run in triplicate, and average values of eluates were normalized to average values of inputs (relative IP) and then further normalized to the relative IP of an untranscribed region of DNA (the IntV region).

RNA analysis.

Cells were grown to mid-log phase and 5 to 10 ml of cultures harvested for RNA analysis. Total RNA was prepared by hot acidic phenol extraction as described before (5). RNA was reverse transcribed with TaqMan reverse transcription reagents (Applied Biosystems) by using random hexamer primers according to the manufacturer's instructions. The resulting cDNA was amplified in real time using an ABI 7900HT fast thermal cycler (Applied Biosystems) with the primer pairs described in Table S2 in the supplemental material. Each PCR consisted of 10 μl SYBR green dye I (Applied Biosystems), 0.1 μl forward primer (10 μM), 0.1 μl reverse primer (10 μM), 4.8 μl H2O, and 5 μl sample. The PCRs went through a program consisting of 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min. The data were normalized to amplification of 18S ribosomal cDNA.

Antibody production.

The recombinant N-terminally truncated Set1 protein (rSet1) was expressed in Escherichia coli. The pMCSG7 plasmid vector was used for protein expression (40) to produce a fusion protein with polyhistidine tags along with the Set1 protein coding sequence. The construction of the expression vector was carried out by PCR using primers 5′ TACTTCCAATTCAATGCTatgAGCACATATACTCCTACCGTCA 3′ and 5′ TTATCCACTTCCAATGtcaGTTCAAGAAACCTTTACAATTAGGTG 3′ (lowercase letters represent the positions of the start and stop codons). One milligram of purified rSet1 protein was injected into rabbits and guinea pigs (Pocono Rabbit Farm & Laboratory) (protocol for antibody production at http://www.prfal.com/protocol.php).


Kdm5 has specific demethylase activity towards all methylated states of H3K4.

The five S. cerevisiae predicted ORFs bearing homology to JmjC catalytic domains were compared to Jarid1d, a histone demethylase in mammals that targets H3K4me3 and H3K4me2 (21). Within the JmjC domain and over the entire protein, Kdm5 (YJR119c) has the highest identity to Jarid1d (Fig. (Fig.1A;1A; see also Fig. S1 in the supplemental material). Further, Kdm5 (like Jarid1d) has JmjN, BRIGHT, and PHD domains in colinear arrangement with Jarid1d, which is not the case for the other yeast jumonji proteins (Fig. (Fig.1A;1A; see also Fig. S1 in the supplemental material). These similarities prompted us to test whether Kdm5, like Jarid1d, is a histone H3K4 demethylase.

FIG. 1.
Kdm5 demethylates H3K4 mono-, di-, and trimethyl. (A) Schematic representation of JARID1d and the five closest S. cerevisiae homologues, including Kdm5. Sequence identities of the JmjC domains of the five proteins to the JARID1d JmjC domain are indicated. ...

FLAG-tagged Kdm5 was expressed via baculovirus infection of Sf21 insect cells. After 72 h of infection, cells were lysed and recombinant protein purified on FLAG antibody beads, followed by FLAG peptide elution, gel electrophoresis, and Colloidal Blue staining. The results showed one band of expected size for Kdm5 (Fig. (Fig.1B),1B), which was confirmed by Western blotting with FLAG antibody (data not shown).

The protein was tested in a demethylase assay using calf thymus histones as substrates, under conditions that were used previously to test Jarid1d (21). The assay with the yeast protein resulted in a substantial reduction in H3K4me3 and H3K4me2 levels in a dose-dependent manner and a slight reduction in H3K4me1 levels (Fig. (Fig.1C).1C). However, no changes in K36me3 and K79me2 levels were observed (Fig. (Fig.1C),1C), which is in accordance with other reports showing no activity of this enzyme towards H3 methylated on K36 and K79 (24). This profile of activity was similar to that of Jarid1d, used as a control for these reactions (Fig. (Fig.1C1C).

In order to investigate the importance of the different domains of Kdm5, we created point mutations at conserved residues in either the PHD domain (D254A) or the JmjC domain (H427A). Recombinant Kdm5 carrying either of these mutations was expressed in insect cells and yielded amounts of protein equal to that of the wild-type protein (Fig. (Fig.1B).1B). When tested in the demethylase assay, the recombinant protein carrying a mutation in the PHD domain (D254A) yielded results similar to those for the wild-type protein (Fig. (Fig.1D).1D). However, the JmjC domain mutant (the H427A mutant) showed no activity towards the histone substrates methylated on H3K4 (Fig. (Fig.1D),1D), suggesting that the enzymatic activity of Kdm5 is indeed dependent on the conserved sequence of the JmjC domain.

Although these assays showed only a minor reduction in H3K4me1 levels, the full activity of the enzyme towards this substrate could be masked by the conversion of H3K4me3 and H3K4me2 into the H3K4me1 form. Therefore, we performed the demethylase assay with histone H3 peptides containing the three forms of methylated K4. The reactions where the substrate was either H3K4me3 or H3K4me2 resulted in an accumulation of H3K4me1, but the reaction with H3K4me1 as a substrate resulted in an almost complete reduction in me1 signal (Fig. (Fig.1E).1E). Again, the recombinant protein with the H427A mutation did not show any activity towards the methylated peptides (Fig. (Fig.1E1E).

Kdm5 affects H3K4 methylation levels during active transcription of the GAL1 gene and has modest effects on GAL1 expression.

We analyzed the function of Kdm5 in vivo. A strain bearing a deletion of the gene was generated and tested for effects on H3K4 methylation. Histones were prepared by acid extraction from the parental and deletion strains and analyzed by quantitative Western blotting using K4 methylation-specific antibodies. No clear differences in the levels of mono-, di-, or trimethylation were detected, and an antibody for unmodified histone was used as a comparison (data not shown).

Although global methylation levels are not altered substantially by deletion of KDM5, methylation may be affected on specific genes. We examined the GAL1 gene under conditions where RNA is not transcribed (raffinose) or transcribed at high levels (galactose) (Fig. (Fig.2A).2A). ChIP analysis using highly specific antibodies (see Fig. S2 in the supplemental material) was carried out for H3K4 methylation at the GAL1 gene under these conditions. We examined mono-, di-, and trimethylation levels relative to the IntV region and normalized to unmodified H3 levels. In the wild-type strain, K4 monomethylation increases twofold at the 5′ end of the ORF in galactose, whereas, as expected (30), both di- and trimethylation strongly increase in galactose (approximately 10-fold) (Fig. (Fig.2B).2B). We examined K4 methylation levels in the KDM5 deletion relative to the wild type in galactose and found substantial increases for mono- and dimethylation (more than fourfold) and lower increases for trimethylation (less than twofold) (Fig. (Fig.2B).2B). We analyzed the 3′ end of the ORF and observed no increases for mono, di-, or trimethylation in the wild-type strain in galactose but increases in all three states in the KDM5 deletion strain relative to the wild type (see Fig. Fig.5A5A).

FIG. 2.
RNA and histone H3 K4 methylation levels during activation of GAL1. (A) GAL1 RNA levels in wild-type strain during galactose induction and return to growth in glucose. RNA was extracted from cell pellets collected after 2 h of growth in raffinose (RAFF)-containing ...
FIG. 5.
Histone H3 K4 methylation and Set1 levels at GAL1 3′ ORF. (A) Histone H3 K4 methyl ChIP in wild-type and KDM5 deletion strains during activation of GAL1. Samples were taken after 2 h of growth in raffinose (RAFF) and 2 h of growth in galactose ...

We examined GAL1 RNA levels in the KDM5 deletion strain in galactose to determine whether the increase in methylation is reflected in higher transcription. RNA levels are slightly higher in the absence of Kdm5 (Fig. (Fig.2C).2C). We speculated that there might be an additive effect on RNA levels among the S. cerevisiae jumonji domain proteins. To test this, we prepared a double disruption of KDM2 (JHD1) and KDM5. We chose KDM2 because it was shown previously to have demethylase activity towards H3 methylated on K36 (8, 42), a distinct activation-linked H3 methylation (12). We found that the double disruption showed a greater increase in RNA levels than either single disruption (Fig. (Fig.2C),2C), an effect that was observed at various times of galactose induction (data not shown).

Kdm5 affects Set1 recruitment during active transcription of the GAL1 gene.

The Set1 enzyme carries out all H3K4 methylation in S. cerevisiae, including mono-, di-, and trimethylation (1, 19, 29, 31, 32). Set1 is a component of the protein complex COMPASS, composed of intrinsic components that regulate its relative abilities to mono-, di-, and trimethylate K4 (28, 29, 31, 33). Higher levels of methylation at GAL1 in the kdm5 mutant led us to examine the level of the Set1 enzyme at the gene. We used an antibody highly specific for Set1 as shown by Western analysis of whole-cell yeast extracts and ChIP assay (Fig. (Fig.3A).3A). ChIP assays showed that Set1 levels increased more than twofold in the KDM5 deletion strain compared to the wild type in galactose, both at the 5′ ORF (Fig. (Fig.3B)3B) and the 3′ ORF (see Fig. Fig.5C,5C, left) of GAL1. To test whether the lower Set1 recruitment in the wild-type strain was dependent on the demethylase activity of Kdm5, we performed the same ChIP experiments with two additional strains consisting of the KDM5 deletion strain reconstituted with either a wild-type copy of KDM5 or a copy of KDM5 carrying a mutant JmjC domain. Both the wild-type and the mutant copy were FLAG tagged and were found to express at equal levels judging by FLAG Western blot analysis of whole-cell extracts (Fig. (Fig.3C).3C). The FLAG-tagged wild-type copy complemented the deletion strain, as it resulted in the same low levels of Set1 recruitment during galactose induction as the wild-type parental strain (Fig. (Fig.3D).3D). However, the deletion strain containing the mutant JmjC domain protein showed as high levels of Set1 as the deletion strain alone (Fig. (Fig.3D),3D), suggesting that the demethylase activity of Kdm5 is involved in the regulation of Set1 recruitment during GAL1 activation. This effect is not due to global changes in Set1 protein levels as they were determined to be equal in all strains used for these ChIP experiments (Fig. (Fig.3E3E).

FIG. 3.
Set1 levels during activation of GAL1. (A) Specificity of α-Set1 antibody. (Top) Western blot analysis of increasing amounts of whole-cell extracts from either a wild-type (WT) strain or a SET1 deletion strain. (Bottom) Set1 ChIP with two negative ...

Kdm5 affects H3K4 methylation levels during shutdown of GAL1 transcription.

Transcription of GAL1 shuts down as the cells are shifted from galactose media into repressive glucose (Fig. (Fig.2A).2A). We tested whether demethylation of H3K4 correlates with this attenuation phase of gene expression. When the wild-type strain is switched from galactose growth to glucose, trimethylation is reduced, dimethylation remains constant, and monomethylation increases slightly at the 5′ end of the gene (Fig. (Fig.4A).4A). Interestingly, the reduction in dimethylation and especially in trimethylation is significantly delayed in the absence of Kdm5 (Fig. (Fig.4A).4A). We found that the wild-type and KDM5 deletion strains grew at the same rates (data not shown); hence, the methylation difference detected in the two strains during return to the repressive state is not the result of altered replication. We examined Set1 levels in glucose conditions and found that the levels decreased in the absence of Kdm5, just as in the wild-type strain (Fig. (Fig.4B).4B). The 3′ end of the gene also showed an increase in methylation but no increase in Set1 levels (Fig. 5B and C, right). Thus, persistence of high methylation without Kdm5 is not due to Set1 remaining bound to the GAL1 gene. To determine whether the effects on methylation levels were specifically due to Kdm5 enzymatic activity, we performed these ChIP experiments with the two additional strains used previously for the Set1 ChIP, containing either a wild-type copy of KDM5 or a copy of KDM5 carrying a mutant JmjC domain. All four strains were grown during galactose-inducing conditions followed by glucose-repressive conditions, and ChIP analysis was carried out as described above. When focused on repressive conditions for GAL1, where trimethylation of H3K4 is reduced in a wild-type strain, we saw similar trimethylation levels in the KDM5 deletion strain carrying a FLAG-tagged wild-type copy of KDM5 (Fig. (Fig.4C).4C). However, the mutant copy of KDM5 resulted in higher trimethylation levels, comparable to the levels seen in the deletion strain (Fig. (Fig.4C).4C). These results add further support to our hypothesis that Kdm5 has demethylase activity in vivo and that this activity is involved directly in gene regulation.

FIG. 4.
Histone H3 K4 methylation and Set1 levels during repression of GAL1. (A) Histone H3 K4 methyl ChIP in wild-type (WT) and KDM5 deletion strains, followed by quantitative PCR analysis at the GAL1 5′ ORF. Samples were taken after 75 min and 280 min ...

Kdm5 lowers H3K4 trimethylation levels during shutdown of SUC2 transcription.

We examined H3K4 methylation levels at other genes in the KDM5 deletion strain. First, we tested constitutive genes, such as PMA1 and ADH1, but could not detect any differences in methylation levels between the wild-type and the deletion strains (data not shown). Next we examined another inducible gene, SUC2. SUC2 RNA levels increased during the induction, but there was no significant difference in transcription between the wild-type and the deletion strains (data not shown). ChIP analysis of H3K4 trimethylation levels at the 5′ end of the gene showed a substantial decrease in the wild-type strain from the induced state back to the repressed state of SUC2 transcription, whereas the KDM5 deletion strain showed only a very slight decrease (Fig. (Fig.6).6). These results show that Kdm5 effects on K4 methylation levels are not restricted to the GAL1 gene.

FIG. 6.
Histone H3 K4 methylation levels during activation and repression of SUC2. Histone H3 K4 trimethyl ChIP in wild-type (WT) and KDM5 deletion strains, followed by quantitative PCR analysis at the SUC2 5′ ORF. Samples were taken after 2 h of growth ...


Here we report that the protein encoded by KDM5 is a histone demethylase and shows substrate specificity for histone H3K4 in vitro. Kdm5 specifically reverses all methylated states of K4 but not other H3 methylation states (K36me3 and K79me2) (Fig. 1C and D). A mutation in the JmjC domain of Kdm5 abrogates this demethylase activity towards methylated H3K4, whereas a mutation in the PHD domain does not (Fig. (Fig.1D1D).

We have observed Kdm5-dependent H3K4 demethylation in vivo within the GAL1 and SUC2 ORFs. K4 is methylated by the Set1 component of the COMPASS complex, and the relative levels of mono-, di-, and trimethylation are established by regulation of COMPASS (33). While monomethylation does not increase during gene activation, both di- and trimethylation increase when GAL1 is induced by growth in galactose media. The trimethylated form of H3K4 is known to be focused specifically at the 5′ end of genes (32), and, via PHD domain interactions, it may help to recruit complexes involved in gene activation (41).

We observed three consequences of deleting Kdm5, manifesting in altered amounts and locations of K4 methylation within the GAL1 ORF. One normal function of Kdm5 is to prevent high levels of Set1 recruitment during maximally active gene transcription (Fig. (Fig.3B).3B). The elevated levels of Set1 recruitment in the KDM5 deletion strain result in abnormally high levels of mono- and dimethylated H3K4 (Fig. (Fig.2B).2B). The result is an altered balance of the methylation states, where the relative increase in trimethylation is reduced compared to the increases in mono- and dimethylation. This function of Kdm5 is dependent upon its enzymatic activity, as the catalytic mutant version of Kdm5 shows the same high levels of Set1 recruitment at the GAL1 5′ ORF region as the KDM5 deletion strain (Fig. (Fig.3D).3D). One possible explanation for this effect might be that the loss of Kdm5 activity results in the recruitment of a specific form of COMPASS, one that produces primarily di- and monomethylated H3K4. For instance, it has been shown that a strain lacking the COMPASS subunit Cps60 or Cps40 results in normal global levels of H3K4 di- and monomethylation but almost nonexistent levels of H3K4 trimethylation (33).

Second, during the attenuation phase following induced transcription, the level of trimethylated K4 is reduced, finally falling to levels characteristic of the fully repressed gene. It appears that the loss of methylation is an active process mediated by Kdm5, because deletion of the demethylase results in delayed reduction of the methylation (Fig. (Fig.4A).4A). This is further supported by analysis of a strain where Kdm5 bears a mutant JmjC domain, which shows methylation levels comparable to those for the KDM5 deletion strain (Fig. (Fig.4C).4C). In contrast to the events that occur during GAL1 induction, during repression of this gene, the Set1 recruitment in the KDM5 deletion strain is not higher than the recruitment levels in the wild-type strain. This suggests that Kdm5 is actively demethylating H3K4 at the GAL1 5′ ORF during the transcriptional repression phase. Another inducible gene, SUC2, also shows a delayed reduction in trimethylation levels in the deletion strain during attenuation. It may be that Kdm5 functions primarily at inducible genes, since none of the constitutively active genes tested showed differences in methylation levels between the KDM5 deletion and wild-type strains. In mammalian cells, low methylation levels in repressed genes have been shown to be maintained by JARID demethylases (3, 17, 21), but it was not known previously that there is active demethylase-dependent reduction of H3K4 trimethylation levels during shutdown of transcription. However, we cannot conclude that Kdm5 is directly responsible for this demethylation event since we were not able to detect Kdm5 protein at the GAL1 ORF by ChIP. The S. cerevisiae and the Schizosaccharomyces pombe JmjC proteins appear to be refractory to ChIP, as other groups report similar difficulty (13, 15). This may be due to transient association of the enzymes with chromatin or to the enzymes functioning on histones as octamers while not tightly associated with DNA.

A third function of Kdm5 is observed at the 3′ end of the gene. Normally, there is no apparent increase in any methylation state or Set1 levels during gene activation, whereas in the absence of Kdm5 there is an increase in Set1 and in all methylation states relative to those for the wild type (Fig. (Fig.55).

Our analysis thus reveals that Kdm5 has multiple roles in regulating K4 methylation. At the 5′ end of the GAL1 ORF, Kdm5 establishes relative methylation levels, limits methylation levels during activation, and functions to reduce methylation levels during attenuation of transcription. At the 3′ end of the gene, Kdm5 prevents methylation. It thus appears that reversible lysine methylation of histones is conserved through evolution, and this investigation of S. cerevisiae reveals novel gene-specific regulatory functions of demethylation.

Supplementary Material

[Supplemental material]


Research was supported by grants from NIH (GM55360) and NSF (MCB-9604208) to S.L.B.


[down-pointing small open triangle]Published ahead of print on 17 September 2007.

Supplemental material for this article may be found at http://mcb.asm.org/.


1. Briggs, S. D., M. Bryk, B. D. Strahl, W. L. Cheung, J. K. Davie, S. Y. Dent, F. Winston, and C. D. Allis. 2001. Histone H3 lysine 4 methylation is mediated by Set1 and required for cell growth and rDNA silencing in Saccharomyces cerevisiae. Genes Dev. 15:3286-3295. [PMC free article] [PubMed]
2. Burke, D., D. Dawson, and T. Stearns. 2000. Methods in yeast genetics: a Cold Spring Harbor Laboratory course manual, p. 149-154. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
3. Christensen, J., K. Agger, P. A. Cloos, D. Pasini, S. Rose, L. Sennels, J. Rappsilber, K. H. Hansen, A. E. Salcini, and K. Helin. 2007. RBP2 belongs to a family of demethylases, specific for tri-and dimethylated lysine 4 on histone 3. Cell 128:1063-1076. [PubMed]
4. Cloos, P. A., J. Christensen, K. Agger, A. Maiolica, J. Rappsilber, T. Antal, K. H. Hansen, and K. Helin. 2006. The putative oncogene GASC1 demethylates tri- and dimethylated lysine 9 on histone H3. Nature 442:307-311. [PubMed]
5. Collart, M. A., and S. Oliviero. 1996. Preparation of yeast RNA, p. 13.12.1-13.12.2. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. A. Seidman, and K. Struhl (ed.), Current protocols in molecular biology, vol. 2. John Wiley & Sons, Hoboken, NJ.
6. Cuthbert, G. L., S. Daujat, A. W. Snowden, H. Erdjument-Bromage, T. Hagiwara, M. Yamada, R. Schneider, P. D. Gregory, P. Tempst, A. J. Bannister, and T. Kouzarides. 2004. Histone deimination antagonizes arginine methylation. Cell 118:545-553. [PubMed]
7. Eissenberg, J. C., M. G. Lee, J. Schneider, A. Ilvarsonn, R. Shiekhattar, and A. Shilatifard. 2007. The trithorax-group gene in Drosophila little imaginal discs encodes a trimethylated histone H3 Lys4 demethylase. Nat. Struct. Mol. Biol. 14:344-346. [PubMed]
8. Fang, J., G. J. Hogan, G. Liang, J. D. Lieb, and Y. Zhang. 2007. The Saccharomyces cerevisiae histone demethylase Jhd1 fine-tunes the distribution of H3K36me2. Mol. Cell. Biol. 27:5055-5065. [PMC free article] [PubMed]
9. Fodor, B. D., S. Kubicek, M. Yonezawa, R. J. O'Sullivan, R. Sengupta, L. Perez-Burgos, S. Opravil, K. Mechtler, G. Schotta, and T. Jenuwein. 2006. Jmjd2b antagonizes H3K9 trimethylation at pericentric heterochromatin in mammalian cells. Genes Dev. 20:1557-1562. [PMC free article] [PubMed]
10. Gallagher, S., S. E. Winston, S. A. Fuller, and J. G. R. Hurrell. 2004. Immunoblotting and immunodetection, p. 10.8.1-10.8.17. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. A. Seidman, and K. Struhl (ed.), Current protocols in molecular biology, vol. 2. John Wiley & Sons, Hoboken, NJ.
11. Gallagher, S. R. 2006. One-dimensional SDS gel electrophoresis of proteins, p. 10.2.1-10.2.35. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. A. Seidman, and K. Struhl (ed.), Current protocols in molecular biology, vol. 2. John Wiley & Sons, Hoboken, NJ.
12. Henry, K. W., A. Wyce, W. S. Lo, L. J. Duggan, N. C. Emre, C. F. Kao, L. Pillus, A. Shilatifard, M. A. Osley, and S. L. Berger. 2003. Transcriptional activation via sequential histone H2B ubiquitylation and deubiquitylation, mediated by SAGA-associated Ubp8. Genes Dev. 17:2648-2663. [PMC free article] [PubMed]
13. Huarte, M., F. Lan, T. Kim, M. W. Vaughn, M. Zaratiegui, R. A. Martienssen, S. Buratowski, and Y. Shi. 2007. The fission yeast jmj2 reverses histone h3 lysine 4 trimethylation. J. Biol. Chem. 282:21662-21670. [PubMed]
14. Iwase, S., F. Lan, P. Bayliss, L. de la Torre-Ubieta, M. Huarte, H. H. Qi, J. R. Whetstine, A. Bonni, T. M. Roberts, and Y. Shi. 2007. The X-linked mental retardation gene SMCX/JARID1C defines a family of histone H3 lysine 4 demethylases. Cell 128:1077-1088. [PubMed]
15. Kim, T., and S. Buratowski. 2007. Two Saccharomyces cerevisiae JmjC domain proteins demethylate histone H3 Lys36 in transcribed regions to promote elongation. J. Biol. Chem. 282:20827-20835. [PubMed]
16. Klose, R. J., K. Yamane, Y. Bae, D. Zhang, H. Erdjument-Bromage, P. Tempst, J. Wong, and Y. Zhang. 2006. The transcriptional repressor JHDM3A demethylates trimethyl histone H3 lysine 9 and lysine 36. Nature 442:312-316. [PubMed]
17. Klose, R. J., Q. Yan, Z. Tothova, K. Yamane, H. Erdjument-Bromage, P. Tempst, D. G. Gilliland, Y. Zhang, and W. G. Kaelin, Jr. 2007. The retinoblastoma binding protein RBP2 is an H3K4 demethylase. Cell 128:889-900. [PubMed]
18. Kouzarides, T. 2002. Histone methylation in transcriptional control. Curr. Opin. Genet. Dev. 12:198-209. [PubMed]
19. Krogan, N. J., J. Dover, S. Khorrami, J. F. Greenblatt, J. Schneider, M. Johnston, and A. Shilatifard. 2002. COMPASS, a histone H3 (lysine 4) methyltransferase required for telomeric silencing of gene expression. J. Biol. Chem. 277:10753-10755. [PubMed]
20. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. [PubMed]
21. Lee, M. G., J. Norman, A. Shilatifard, and R. Shiekhattar. 2007. Physical and functional association of a trimethyl H3K4 demethylase and Ring6a/MBLR, a polycomb-like protein. Cell 128:877-887. [PubMed]
22. Lee, M. G., C. Wynder, N. Cooch, and R. Shiekhattar. 2005. An essential role for CoREST in nucleosomal histone 3 lysine 4 demethylation. Nature 437:432-435. [PubMed]
23. Lee, N., J. Zhang, R. J. Klose, H. Erdjument-Bromage, P. Tempst, R. S. Jones, and Y. Zhang. 2007. The trithorax-group protein Lid is a histone H3 trimethyl-Lys4 demethylase. Nat. Struct. Mol. Biol. 14:341-343. [PubMed]
24. Liang, G., R. J. Klose, K. E. Gardner, and Y. Zhang. 2007. Yeast Jhd2p is a histone H3 Lys4 trimethyl demethylase. Nat. Struct. Mol. Biol. 14:243-245. [PubMed]
25. Longtine, M. S., A. McKenzie III, D. J. Demarini, N. G. Shah, A. Wach, A. Brachat, P. Philippsen, and J. R. Pringle. 1998. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14:953-961. [PubMed]
26. Margueron, R., P. Trojer, and D. Reinberg. 2005. The key to development: interpreting the histone code? Curr. Opin. Genet. Dev. 15:163-176. [PubMed]
27. Martin, C., and Y. Zhang. 2005. The diverse functions of histone lysine methylation. Nat. Rev. Mol. Cell Biol. 6:838-849. [PubMed]
28. Morillon, A., N. Karabetsou, A. Nair, and J. Mellor. 2005. Dynamic lysine methylation on histone H3 defines the regulatory phase of gene transcription. Mol. Cell 18:723-734. [PubMed]
29. Nagy, P. L., J. Griesenbeck, R. D. Kornberg, and M. L. Cleary. 2002. A trithorax-group complex purified from Saccharomyces cerevisiae is required for methylation of histone H3. Proc. Natl. Acad. Sci. USA 99:90-94. [PMC free article] [PubMed]
30. Ng, H. H., F. Robert, R. A. Young, and K. Struhl. 2003. Targeted recruitment of Set1 histone methylase by elongating Pol II provides a localized mark and memory of recent transcriptional activity. Mol. Cell 11:709-719. [PubMed]
31. Roguev, A., D. Schaft, A. Shevchenko, W. W. Pijnappel, M. Wilm, R. Aasland, and A. F. Stewart. 2001. The Saccharomyces cerevisiae Set1 complex includes an Ash2 homologue and methylates histone 3 lysine 4. EMBO J. 20:7137-7148. [PMC free article] [PubMed]
32. Santos-Rosa, H., R. Schneider, A. J. Bannister, J. Sherriff, B. E. Bernstein, N. C. Emre, S. L. Schreiber, J. Mellor, and T. Kouzarides. 2002. Active genes are tri-methylated at K4 of histone H3. Nature 419:407-411. [PubMed]
33. Schneider, J., A. Wood, J. S. Lee, R. Schuster, J. Dueker, C. Maguire, S. K. Swanson, L. Florens, M. P. Washburn, and A. Shilatifard. 2005. Molecular regulation of histone H3 trimethylation by COMPASS and the regulation of gene expression. Mol. Cell 19:849-856. [PubMed]
34. Secombe, J., L. Li, L. Carlos, and R. N. Eisenman. 2007. The trithorax group protein Lid is a trimethyl histone H3K4 demethylase required for dMyc-induced cell growth. Genes Dev. 21:537-551. [PMC free article] [PubMed]
35. Seward, D. J., G. Cubberley, S. Kim, M. Schonewald, L. Zhang, B. Tripet, and D. L. Bentley. 2007. Demethylation of trimethylated histone H3 Lys4 in vivo by JARID1 JmjC proteins. Nat. Struct. Mol. Biol. 14:240-242. [PubMed]
36. Shi, Y., F. Lan, C. Matson, P. Mulligan, J. R. Whetstine, P. A. Cole, R. A. Casero, and Y. Shi. 2004. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119:941-953. [PubMed]
37. Shi, Y. J., C. Matson, F. Lan, S. Iwase, T. Baba, and Y. Shi. 2005. Regulation of LSD1 histone demethylase activity by its associated factors. Mol. Cell 19:857-864. [PubMed]
38. Shilatifard, A. 2006. Chromatin modifications by methylation and ubiquitination: implications in the regulation of gene expression. Annu. Rev. Biochem. 75:243-269. [PubMed]
39. Sims, R. J., III, K. Nishioka, and D. Reinberg. 2003. Histone lysine methylation: a signature for chromatin function. Trends Genet. 19:629-639. [PubMed]
40. Stols, L., M. Gu, L. Dieckman, R. Raffen, F. R. Collart, and M. I. Donnelley. 2002. A new vector for high throughput, ligation independent cloning encoding a TEV protease cleavage site. Protein Expr. Purif. 25:8-15. [PubMed]
41. Taverna, S. D., S. Ilin, R. S. Rogers, J. C. Tanny, H. Lavender, H. Li, L. Baker, J. Boyle, L. P. Blair, B. T. Chait, D. J. Patel, J. D. Aitchison, A. J. Tackett, and C. D. Allis. 2006. Yng1 PHD finger binding to H3 trimethylated at K4 promotes NuA3 HAT activity at K14 of H3 and transcription at a subset of targeted ORFs. Mol. Cell 24:785-796. [PubMed]
42. Tsukada, Y., J. Fang, H. Erdjument-Bromage, M. E. Warren, C. H. Borchers, P. Tempst, and Y. Zhang. 2006. Histone demethylation by a family of JmjC domain-containing proteins. Nature 439:811-816. [PubMed]
43. Tu, S., E. M. Bulloch, L. Yang, C. Ren, W. C. Huang, P. H. Hsu, C. H. Chen, C. L. Liao, H. M. Yu, W. S. Lo, M. A. Freitas, and M. D. Tsai. 2007. Identification of histone demethylases in Saccharomyces cerevisiae. J. Biol. Chem. 282:14262-14271. [PMC free article] [PubMed]
44. Wang, Y., J. Wysocka, J. Sayegh, Y. H. Lee, J. R. Perlin, L. Leonelli, L. S. Sonbuchner, C. H. McDonald, R. G. Cook, Y. Dou, R. G. Roeder, S. Clarke, M. R. Stallcup, C. D. Allis, and S. A. Coonrod. 2004. Human PAD4 regulates histone arginine methylation levels via demethylimination. Science 306:279-283. [PubMed]
45. Whetstine, J. R., A. Nottke, F. Lan, M. Huarte, S. Smolikov, Z. Chen, E. Spooner, E. Li, G. Zhang, M. Colaiacovo, and Y. Shi. 2006. Reversal of histone lysine trimethylation by the JMJD2 family of histone demethylases. Cell 125:467-481. [PubMed]
46. Yamane, K., C. Toumazou, Y. Tsukada, H. Erdjument-Bromage, P. Tempst, J. Wong, and Y. Zhang. 2006. JHDM2A, a JmjC-containing H3K9 demethylase, facilitates transcription activation by androgen receptor. Cell 125:483-495. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...