Downloaded from genesdev.cshlp.

org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

The S. cerevisiae SET3 complex includes

two histone deacetylases, Hos2 and Hst1,
and is a meiotic-specific repressor
of the sporulation gene program
W.W.M. Pim Pijnappel,1,6 Daniel Schaft,1,6 Assen Roguev,1 Anna Shevchenko,2 Hille Tekotte,4
Matthias Wilm,2 Guillaume Rigaut,2 Bertrand Séraphin,5 Rein Aasland,3 and A. Francis Stewart1,7
1
Gene Expression Program, European Molecular Biology Laboratory, 69117 Heidelberg, Germany and Genomics, Technische
Universitaet Dresden, 01307 Dresden, Germany; 2Instrumentation Program, European Molecular Biology Laboratory, 69117
Heidelberg, Germany; 3Department of Molecular Biology, University of Bergen, N-5020 Bergen, Norway

Set3 is one of two proteins in the yeast Saccharomyces cerevisiae that, like Drosophila Trithorax, contains
both SET and PHD domains. We found that Set3 forms a single complex, Set3C, with Snt1, YIL112w, Sif2,
Cpr1, and two putative histone deacetylases, Hos2 and NAD-dependent Hst1. Set3C includes NAD-dependent
and independent deacetylase activities when assayed in vitro. Homology searches suggest that Set3C is the
yeast analog of the mammalian HDAC3/SMRT complex. Set3C represses genes in early/middle of the yeast
sporulation program, including the key meiotic regulators ime2 and ndt80. Whereas Hos2 is only found in
Set3C, Hst1 is also present in a complex with Sum1, supporting previous characterizations of Hst1 and Sum1
as repressors of middle sporulation genes during vegetative growth. However, Hst1 is not required for meiotic
repression by Set3C, thus implying that Set3C (−Hst1) and not Hst1–Sum1, is the meiotic-specific repressor of
early/middle sporulation genes.
[Key Words: SET domain; repression; sporulation; Hos2; Hst1; deacetylase]
Received May 7, 2001; revised version accepted September 17, 2001.

Genetic screens in Drosophila have identified several this yeast (Ekwall et al. 1996). The S. cerevisiae Set1
classes of chromatin regulatory proteins including fac- protein influences silencing at telomeres, and both Set1
tors involved in position effect variegation (PEV), and and Clr4 influence silencing in mating-type switching in
members of the Polycomb and trithorax Groups (Pc-G, their respective yeasts (Nislow et al. 1997; Ivanova et al.
trx-G) (Pirotta 1998; Wakimoto 1998). Clues as to the 1998). Drosophila Trx and a mammalian homolog, Mll,
underlying mechanisms used by these genetically de- play crucial roles in maintenance of homeotic complex
fined factors have arisen from the identification of sev- gene expression patterns in development (Yu et al. 1995;
eral characteristic protein motifs including the SET do- Ingham 1998).
main. The SET domain was first identified by its occur- The SET domains in SU(VAR) 3–9 and Clr4 have been
rence in Su(var)3–9 (suppressor of position effect identified recently as histone methyltransferases acting
variegation 3–9), E(z) (Enhancer of zeste), and Trx (Tri- on lysine 9 of histone H3 (Rea et al. 2000). In the same
thorax) (Jones and Gelbart 1993; Tschiersch et al. 1994; assay, the SET domains of E(z) or Trx homologs did not
Stassen et al. 1995). Functional data on SET domain pro- show histone methyltransferase activity, suggesting that
teins relate to chromatin regulation and, in certain cases, they may act as methyltransferases on other substrates.
epigenetic mechanisms. Most prominently, Clr4, the Other roles for SET domains remain possible. The SET
Schizosaccharomyces pombe homolog of Su(var)3–9, is domains of Trx and/or MLL have been shown by two-
involved in epigenetic maintenance of centromeres in hybrid and coimmunoprecipitation analyses to partici-
pate in several protein–protein interactions including
Present addresses: 4Wellcome Trust Centre for Cell Biology, ICMB, interaction with (1) the SET domain of another trx-G
King’s Buildings, University of Edinburgh, Edinburgh EH9 3JR, Scotland;
5
member, Ash1 (Rozovskaia et al. 2000); (2) SBF1, a myo-
CGM-Centre National de la Recherche Scientifique, Cedex 91198,
France.
tubularin related protein (Cui et al. 1998); and (3) the
6
These authors contributed equally to this work. trx-G-related Swi–Snf complex proteins, INI1 and SNR1
7
Corresponding author. (Rozenblatt-Rozen et al. 1998). Consequently, current data
E-MAIL [email protected]; FAX 49-351-210-1409.
Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/ suggests that the SET domain encompasses at least two
gad.207401. roles, including histone methyltransferase activity and

GENES & DEVELOPMENT 15:2991–3004 © 2001 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/01 $5.00; www.genesdev.org 2991
 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

protein–protein interactions with other factors involved Identification of Set3 protein–protein interactions
in chromatin regulation.
A total of six genes containing SET domains were The TAP method (Rigaut et al. 1999) was used to de-
found in searches of the S. cerevisiae genome, including termine protein–protein interactions with Set3. This
the previously studied set1 gene, and now termed set1 method relies on fusion of the TAP tag, a double-speci-
through set6 (Fig. 1). Interestingly, two of these six, set3 ficity tag, to the protein of interest by knock-in to the
and set4, also encode a PHD finger. The PHD finger is endogenous coding sequence, followed by two rounds of
another domain that was initially identified by its occur- affinity chromatography. All purified proteins were iden-
rence in Pc-G and trx-G members (Aasland et al. 1995) tified by mass spectrometry (MS; Shevchenko et al.
and is also found in a variety of chromatin regulators and 1996a,b). Purification of C-terminally tagged Set3 (Set3–
transcriptional cofactors including CBP/p300, ATRX, TAP) resulted in specific copurification of seven pro-
DNMT3, and the TIF1/KAP-1 family (for review, see teins, now termed Set3C, which were not observed in
Pascual et al. 2000; Capili et al. 2001 for references and purifications of wild-type or unrelated TAP-tagged
the recent determinations of the PHD finger structure). strains (Fig. 2A; data not shown). In addition to Set3–
Furthermore, the coincidence of SET domains and PHD TAP, the other six identified proteins were Snt1 (pro-
fingers is a characteristic displayed by the trx-G proteins, posed name for YCR033w on the basis of the presence of
Trx and Ash1 (Stassen et al. 1995; Tripoulas et al. 1996) two SANT domains; Aasland et al. 1996), YIL112w, Sif2
and their mammalian homologs and analogs, including (SIR4 interacting factor 2; Cockell et al. 1998), Hos2
the NSD/WHSC1 family (Huang et al. 1998; Stec et al. (Rundlett et al. 1996), Hst1 (Brachmann et al. 1995), and
1998; Angrand et al. 2001). Because there are only 14 Cpr1 (cyclophilin A; for review, see Dolinski and Heit-
genes in S. cerevisiae that encode PHD fingers (http:// man 1997) as indicated in Figure 2A. The heat-shock
www.uib.no/aasland/set/), the coincidence of the SET proteins Ssa1 and Ssb1 were also present as minor bands,
and PHD domains in set3 and set4 was provocative and however, we consistently find these highly abundant
initiated the work described here on set3. proteins in wild-type control and unrelated purifications
After analyzing set3 in several conventional ways under the conditions used in this study, indicating that
without significant insight, we applied proteomic tech- they are unspecific contaminants (data not shown). On
niques to Set3 (Rigaut et al. 1999; Shevchenko et al. the basis of Coomassie staining (Fig. 2A), we estimate
1999). The identified Set3 complex (Set3C) includes two relative stoicheiometries (indicated in brackets) as fol-
potential histone deacetylases, Hos2 and Hst1. Hos2 is a lows: Snt1 (2); YIL112w (1); Set3 (1); Sif2 (2); Hos2 (1);
class I histone deacetylase like Rpd3 and the mammalian Hst1 (<1/4); and Cpr1 (1).
HDACs 1–3 (Rundlett et al. 1996). Hst1 is a member of To verify the Set3 complex (Set3C) and to analyze fur-
the recently identified Sir2 class of NAD-dependent ther the interacting proteins of the other six Set3C com-
deacetylases (Imai et al. 2000; Landry et al. 2000; Smith ponents, the TAP tag was fused to the coding regions of
et al. 2000) and has been linked previously to repression each and the same protocols of purification and MS
of sporulation genes (Xie et al. 1999; Lindgren et al. analysis were applied. The purifications of Snt1, YIL112,
2000). Here we identify a role for Set3C in repression of Sif2, and Hos2 yielded very similar results as the Set3–
the sporulation gene program. TAP purification (Fig. 2B). Along with additional obser-
vations, the results shown in Figure 2B are summarized
as follows and establish that Set3C is composed of seven
Results proteins: (1) Snt1–TAP purification yielded MS identifi-
cations of all proteins of Set3C. Relative stoicheiom-
Phenotypic analyses of Set3 disruption in haploid cells
etries were similar as in Set3–TAP, and no new copuri-
Disruptions in haploid cells of set3, set4, or both to- fying proteins could be detected, showing that endog-
gether, resulted in normal viability with no obvious phe- enous Snt1 is present in Set3C and not in other
notype (data not shown). The previous finding that set1 complexes or as free protein. (2) YIL112w–TAP yielded
influences silencing at telomere and mating type loci all seven proteins of Set3C without any new copurified
(Nislow et al. 1997) prompted similar tests of set3 dis- protein. Thus, YIL112w protein–protein interactions are
ruptions on these processes. A moderate reduction in limited to Set3C by this experimental criterion. Whereas
silencing of a telomeric ura3 (10-fold reduced growth on the stoicheiometry of Set3:Sif2:Hos2:Hst1:Cpr1 was
5-fluoro-orotic acid plates), and no effect on silencing of again estimated as 1:2:1:<1/4:1, comigration of YIL112w–
a lacZ reporter gene integrated at HML, was found in TAP with Snt1 obscured the relative stoicheiometry of
⌬set3 strains (data not shown). We also screened ⌬set3 these two Set3C components. (3) TAP purification of
strains for sensitivity to the alkylating agents MMS or Sif2 yielded the entire Set3C without any new copurify-
EMS, UV irradiation, bleomycin, hydroxyurea, and heat, ing proteins, but in this case, Sif2–TAP was present in
and found no viability differences with wild-type yeast clear excess over other members of Set3C. This indicates
(data not shown). Therefore, it is unlikely that Set3 has a that endogenous Sif2 occurs as free protein as well as in
direct role in silencing at telomeres and mating-type the Set3C. Sif2 has been identified as a Sir4-interacting
loci, DNA repair, or stress responses. To approach set3 factor because of an interaction in a yeast two-hybrid
function from a different angle, we examined the protein assay (Cockell et al. 1998). We failed to detect the pres-
interaction partners of Set3. ence of Sir4 in Sif2–TAP purifications (Fig. 2B) or the

2992 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Meiotic repression by SET3 deacetylase complex

Figure 1. SET domain proteins in S. cerevisiae. (A) Cartoons of the six yeast SET domain proteins identified. Each sequence was
retrieved from SGD (http://genome-www.stanford.edu/Saccharomyces/) by the systematic yeast protein names given. The length of
each protein and the position in each protein of the PHD fingers and SET domains are also indicated. Set5 and Set6 have large
insertions (173 and 192 amino acids, respectively) indicated as light green boxes in the SET icon. E-values obtained when using the Set1
SET domain as a probe in a PSI-BLAST search (four iterations; Altschul et al. 1997) against the yeast proteome, are indicated. (B) A
multiple alignment of the six SET domains in yeast. The alignment was generated with CLUSTAL_X (Thompson et al. 1997) and
subsequently manually refined. The alignment is color coded to highlight conserved features using the color scheme implemented in
CLUSTAL_X. The two positions in Set5 and Set6 with large insertions are indicated. Below the alignment is shown a consensus
featuring positions conserved in at least five of the six sequences. The six positions in the alignment that correspond to the motif
common to SET domains and methyltransferases (Rea et al. 2000) are indicated with red dots. The alignment includes the C-terminal
SET domain-associated cysteine cluster, postSET (position 203–217 in the alignment).

GENES & DEVELOPMENT 2993

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

presence of Sif2 in Sir4–TAP purifications (data not volved in folding Hos2 or the Hos2–TAP fusion protein.
shown), suggesting that Sif2–Sir4 interaction might be (5) Importantly, TAP purification of the substoicheio-
transient and/or too lowly abundant to be detected by metric component Hst1 yielded all seven proteins of
this method. (4) Hos2–TAP yielded the entire Set3C. In Set3C (Fig. 2C), confirming the presence of Hst1 in
addition, the complete TRiC complex (Leroux and Hartl Set3C. However, about fourfold more Hst1–TAP was re-
2000) was identified. This may indicate that Hos2 is pre- trieved than Set3C components and two additional pro-
sent in two complexes, Set3C and TRiC. More likely, the teins were copurified. Sum1 was copurified in approxi-
chaperone function of the TRiC complex may be in- mately equal abundance to Hst1 along with the pioneer

(Figure 2 continued on facing page)

2994 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Meiotic repression by SET3 deacetylase complex

ORF protein, YOR279. Physical interaction between 2001). Consequently, we knocked the TAP tag into
Hst1 and Sum1 has been proposed on the basis of similar sum1, and retrieved Hst1 and YOR279, but not any other
behavior in a genetic screen for repressors of mid sporu- member of Set3C, or any other specific copurifying pro-
lation genes (Xie et al. 1999), and Hst1 and Sum1 have teins (Fig. 2C). As opposed to approximately equal quan-
recently been coimmunoprecipitated (Rusché and Rine tities of Sum1 and Hst1–TAP, Sum1–TAP was retrieved
in greater abundance than Hst1. Also a specific shorter,
120-kD version of Sum1–TAP was observed that was not
observed in the Hst1–TAP purification. These observa-
tions indicate that Sum1 exists partly complexed with
Hst1 and partly as free protein, and suggests that an N-
terminally shorter version of Sum1 does not associate
with Hst1. We also conclude that Hst1 is present in two
distinct complexes, Set3C and the Hst1–Sum1 complex.
(6) TAP purification of Cpr1 yielded very high levels of
Cpr1–TAP without apparent copurifying proteins (data
not shown). As opposed to other members of Set3C, Cpr1
is a highly abundant protein. Despite the absence of
other Set3C members in Cpr1–TAP purifications, the
consistent presence of Cpr1 in TAP purifications of
SET3, Snt1, YIL112w, Sif2, Hos2, and Hst1, but not wild-
type control, Sum1, or unrelated purifications (Fig. 2B,C;
data not shown) strongly implicates Cpr1 as a bona fide
member of Set3C. Furthermore, Cpr1 is required for
function of the Hos2-related histone deacetylase, Rpd3
(Arevalo-Rodriguez et al. 2000), suggesting a functional
relationship between cyclophilins and Rpd3/Hos2-type
histone deacetylases.
In summary, sequential TAP purification identified (1)
Figure 2. The Set3 protein complex. (A) Set3–TAP purifica- the Set3C consisting of the following seven proteins:
tion. Coomassie-stained 7%–25% polyacrylamide SDS gel Snt1, YIL112, Set3, Sif2, Hos2, Hst1, and Cpr1; with all
showing the purified complex (Set3–TAP). Cartoons of MS- available data indicating an approximate stoicheiometric
identified proteins are shown with full-length protein sizes in compostion of 2:1:1:2:1:<1/4:1, respectively; (2) the
amino acids (aa). Protein domains are also indicated in amino Hst1–Sum1 complex consisting of Hst1, Sum1, and
acids. The tagged protein, Set3, is indicated in bold together YOR279; (3) free, uncomplexed Sif2; (4) free, uncom-
with its C-terminal remnant of the TAP tag. Protein size mark-
plexed Sum1; (5) free, uncomplexed Cpr1.
ers (broad range, NEB) are indicated in kilodaltons. Ssa1 and
Ssb1 are heat-shock proteins also found in mock purifications,
and therefore represent unspecific contaminants. (B) Verifica- Requirements for Set3C integrity
tion of the complex. As in A, showing the complexes obtained
by TAP purification of Snt1, YIL112w, Sif2, or Hos2, as indi- To dissect protein–protein interactions within Set3C, we
cated. The tagged proteins with remaining TAP fusions are in- constructed strains in which one complex member car-
dicated in bold. TRiC complex in Hos2–TAP: 8, Cct6; 9, Cct8; ried the TAP tag, whereas a part of the complex was
10, Cct2; 11, Cct3; 12, Cct4; 13, Cct7. (C) Defining a proteomic deleted. Sif2–TAP purification in a ⌬set3 strain yielded
border of Set3C. As in A and B, showing TAP purification of Sif2–TAP and Snt1 without the other Set3C members
Hst1 and Sum1, as indicated. The presence of many bands, (Fig. 3, left). Snt1 appeared partially degraded, indicating
marked by asterisks, was later found to be due to the batches of
instability outside intact Set3C. This shows that Sif2
calmodulin beads used. All the bands marked by an asterisk
were identified by MS, and are nonspecific contaminants, also
and Snt1 interact directly and require Set3 to interact
visible in Figure 3. (They are, from top to bottom, in the Hst1– with Hos2, YIL112w, Hst1, and Cpr1. Furthermore, TAP
TAP panel: J04692; Ssa2: Ssb1: HSP60; ef1-␣; enolase2; phos- purification of a Set3–TAP allele in which the SET do-
phoglycerate kinase; 60s ribosomal protein [rp] l4-B; fructose- main had been precisely deleted from the rest of the pro-
biphosphate aldolase; glyceraldehyde 3-phosphate dehydroge- tein failed to retrieve anything. However, Western detec-
nase; 60s rp l2; mixture of 40s rp s1-B, 60s rp l8-A, 40s rp s4; tion of the TAP tag on this Set3–⌬SET domain–TAP
mixture of 60s rp l1, triosephosphate isomerase; mixture of 60 protein in total cell lysates yielded detectable but very
ns rp l7-A, 60 s rp l10; mixture of 40s rp s7-A, 60s rp l13-A; 60 low levels of expression (data not shown).
s rp l20; and in the SUM1–TAP panel: mixture of sum1, ef2; YIL112w–TAP purification in a ⌬hos2 background
hsp82; jo4692; sum1; ssa2; ssb2; pyruvate kinase 1; mixture of
yielded YIL112w–TAP and Hst1, but no other Set3C
sum1, pyruvate decarboxylase isozyme; ef1␣; mixture of eno-
lase 2, 60s rp l3; 60s rp l4-A; glyceraldehyde3-phosphate dehy-
members (Fig. 3, middle). This shows that YIL112w and
drogenase; 60s rpl2; mixture of 40s rp s1-B, 40s rp s4, 60s rp l8-B; Hst1 interact directly and require Hos2 to interact with
mixture of 60s rp l7-B, 60s rp l10, 60s rp l1; mixture of 60s l13-B, Set3, Sif2, Snt1, and Cpr1.
40s rp s7-A; 60s rp l20; mixture of 40s rp s24, 40s rp s17-A, 60s In contrast, deletion of hst1 did not disturb Set3C pu-
rpl25, 40s rp s19-B). rified from Sif2–TAP (Fig. 3, right). Thus, Hst1 is periph-

GENES & DEVELOPMENT 2995

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

Figure 3. Set3 and Hos2 are central components of

the Set3C but Hst1 is not. Coomassie-stained poly-
acrylamide SDS gels showing TAP purifications of
Sif2–TAP in a ⌬set3 background (left), YIL112–TAP
in a ⌬hos2 background (middle), and Sif2–TAP in
⌬hst1 background (right). The tagged proteins with
remaining TAP fusion are indicated in bold.

eral to Set3C. These results define Set3 and Hos2 as cru- Hos2 in Set3C and Hst1 in both Set3C and the Hst1–
cial members of Set3C required for complex integrity. Sum1 complexes are active histone deacetylases.

Enzyme activities in Set3C Set3C is a repressor of meiosis

The presence of a SET domain in Set3C suggested that it Diploid ⌬set3 and ⌬hos2 strains were tested for meiotic
may include a histone methyltransferase activity. We landmarks during sporulation. Figure 5A shows that ki-
failed to detect any such activity in Set3C in several netics of premeiotic DNA replication, as measured by
assays, although TAP–Clr4 (expressed from a CEN plas- FACS (fluorescence activated cell sorter) analysis, were
mid and purified by the same methodology) and Set1, in unchanged in ⌬set3 and ⌬hos2 strains as compared with
the context of its complex, Set1C, proved to be very ac- wild type. Therefore, Set3C is unlikely to be involved in
tive (Roguev et al. 2001). Nevertheless, Set3C may be a
methyltransferase when tested under different condi-
tions.
Set3C includes two predicted histone deacetylase en-
zyme activities. Hos2 is one of the five S. cerevisiae
members of the first identified histone deacetylase fam-
ily that includes Rpd3 and Hda1 (Rundlett et al. 1996).
Hst1 is one of the five S. cerevisiae members of the sec-
ond, NAD-dependent, histone deacetylase family that
includes Sir2 (Imai et al. 2000; Landry et al. 2000; Smith
et al. 2000). To determine whether Set3C contains his-
tone deacetylase activity, we tested deacetylation of par-
tially purified Set3–TAP or Hst1–TAP complexes with
3
H-acetylated histone H4 peptide with and without
NAD. Consistent with the relative stoicheiometries in
Set3C (indicated in brackets) of Hos2 (1) and Hst1 (<1/4),
the predominant deacetylase activity in the Set3–TAP
eluate was NAD independent, and was moderately
Figure 4. The Set3 and Hst1 complexes contain both NAD-
stimulated by NAD (Fig. 4).
independent and NAD-dependent histone deacetylase activi-
Application of the same assay to Hst1–TAP eluate, ties. IgG Sepharose eluates from the strains indicated were as-
which contains both Set3C and the Hst1–Sum1 com- sayed for histone deacetylase activity as described in Materials
plex, showed, as expected, lower levels of NAD-indepen- and Methods. The presence or absence of 0.5 mM NAD or
dent and higher levels of NAD-stimulated histone NADH in the reaction is also indicated. Data represent 3H re-
deacetylase activity (Fig. 4). These results indicate that lease as a percentage of input, and are means ± SEM.

2996 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Meiotic repression by SET3 deacetylase complex

the regulation of early meiotic events such as initiation tions of Set3 and all associated factors to define the pro-
of meiosis or premeiotic DNA synthesis. In contrast, teomic environment of Set3C (Figs. 1–3; schematized in
both ⌬set3 and ⌬hos2 strains underwent faster progres- Fig. 6). Set3 occurs in a single complex, Set3C, consisting
sion through meiosis I, meiosis II, and ascus formation of Snt1 (2), YIL112w (1), Set3 (1), Sif2 (2), Hos2 (1), Hst1
than wild type, and resulted in reduced levels of viable (<1/4), and Cpr1 (1) (estimated stoicheiometries in brack-
ascus formation, as determined by morphological and ets). The relatively low amount of Hst1 in this complex
tetrad analyses (Fig. 5B; data not shown). Thus, ⌬set3 and suggests that it is transiently interacting with Set3C.
⌬hos2 phenotypes are very similar, consistent with the Furthermore, Hst1 is also present in a complex with
proteomic analysis of Set3C, and indicate a role for Sum1 and YOR279 (Hst1–Sum1 complex). TAP purifica-
Set3C in middle sporulation. tion of Sum1 did not copurify Set3C. Vice versa, Sum1
DNA microarray analysis has led to the clustering of was not detected in TAP purifications of core Set3C
genes induced during sporulation based on the timing of members. Thus, Set3C and the Hst1–Sum1 complex are
their expression (Chu et al. 1998; Primig et al. 2000). To distinct entities. However, we cannot exclude the possi-
determine whether disruption of Set3C affected sporula- bility that these complexes interact, and that a Sum1–
tion-specific gene expression, we performed Northern Set3C interaction is too weak to be detected by the ap-
blot analysis of representative members of these clusters proach used. We also conclude that Snt1, YIL112w, Set3,
and of known key regulators of sporulation during a and Hos2 are predominantly included in Set3C and are
sporulation time course (Fig. 5C). Genes in clusters 2 not present as free proteins or in other stable protein
(acs1) and 3 (dal7), which have amino acid and other complexes, whereas Sif2, Cpr1, and Sum1 do exist as free
metabolic functions, had unchanged expression levels in proteins in addition to their involvement in the observed
⌬set3 cells. Expression of the cluster 4 gene rfa1, which complexes.
functions in DNA synthesis, repair and recombination, In addition to the deacetylase domains and the SET
was only slightly increased in ⌬set3 cells. In contrast, domain and PHD finger of Set3, Set3C includes several
expression of spo11, which also belongs to cluster 4 but other identifiable protein motifs. The SANT domain was
functions in chromosome synapsis and segregation, was identified by its occurrence in chromatin proteins in-
precociously induced in ⌬set3, reaching higher levels cluding Swi3, Ada2, NcoR, and TFIIIB⬘⬘ (Aasland et al.
than in wild-type cells. The cluster 5 genes spr3, sps2, 1996). It is distantly related to the DNA-binding domain
and sps18, which function in cytokinesis and differen- of c-myb and may act in nonsequence-specific DNA
tiation, were all expressed earlier and at higher levels in binding by TFIIIB⬘⬘ (Kumar et al. 1997). Thus, Snt1 may
⌬set3. be a DNA-binding component of Set3C. Sif2 is a seven
Analysis of key regulators of sporulation revealed that WD40-repeat propeller protein. Such proteins participate
ime1, which is required for early meiotic gene expression in many cellular functions, however, TBL1 [transducin
including ime2 and spo11 (Mitchell et al. 1990), was ex- (␤)-like I], Groucho, and TUP1 have been shown to bind
pressed at slightly higher levels in ⌬set3 cells. In con- histones (Edmonson et al. 1996; Palaparti et al. 1997;
trast, ime2, which is involved in both early and middle/ Guenther et al. 2000). Hence, Sif2 may contribute
late gene expression including spo11 and sps2 (Mitchell nucleosomal binding activity to Set3C. It is interesting
et al. 1990), was expressed earlier and at higher levels in to note that Snt1 and Sif2 interact (Fig. 3) and are the two
⌬set3, as was ndt80, a regulator of middle/late genes in- members of Set3C that are present twice each. Hence,
cluding itself, spr3, sps2, and sps18 (Chu and Herskowitz they may present a tetrameric chromatin interaction
1998). Notably, none of the genes repressed by Set3C in module for Set3C (indicated as a box in Fig. 6). YIL112w
sporulation were up-regulated at the zero time point, contains four ankyrin repeats, which have been shown
which is representative of gene expression during veg- previously to participate in a variety of protein–protein
etative growth. Hence, we identify Set3C as a meiotic interactions (Sedgwick and Smerdon 1999). The ankyrin
specific repressor of cluster 4 and 5 genes of the sporu- repeats of YIL112w and/or its coiled-coil region may be
lation program. involved in the direct interaction with Hst1 and/or the
Deletion of hst1 did not relieve repression of ime2 or implied interaction with Hos2.
ndt80, whereas deletion of hos2 in the absence of or pres-
ence of hst1 did (Fig. 5D; data not shown). These results
Biochemical activities of Set3C and probable
parallel the effects on Set3C of set3, hos2, and hst1 de-
conservation in eukaryotes
letions. Both Set3 and Hos2 are required for Set3C integ-
rity, whereas Hst1 is not (Fig. 3). Thus, in early sporula- Although certain SET domain proteins including Su-
tion, Set3C represses two key regulators of the sporula- (var)3–9 have histone methyltransferase activity (Rea et
tion gene program, ime2 and ndt80, and Hst1 is not al. 2000), we were unable to detect this activity in Set3C.
required for Set3C formation or function. However, Set3 is required for Set3C (Fig. 3), indicating a
structural role in complex integrity.
Discussion Set3C combines both NAD-dependent and inde-
pendent histone deacetylase activities (Fig. 4). In con-
Protein interactions of Set3
cordance with other studies on histone deacetylases
A proteomic approach based on sequential TAP tagging (Rundlett et al. 1996; Hassig et al. 1998), we find that the
and purification was used to define the protein interac- activities associated with Set3C and Hst1–Sum1 com-

GENES & DEVELOPMENT 2997

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

(Figure 5 continued on facing page)

2998 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Meiotic repression by SET3 deacetylase complex

(Figure 5 continued) deacetylase activities in the same complex remains to be

determined.
Set3C shows interesting similarities to the partially
characterized mammalian HDAC3 complexes (Guen-
ther et al. 2000; Li et al. 2000). In addition to the fact that
Hos2 is the closest yeast homolog of HDAC3, complexes
from HeLa cells include the corepressors SMRT and
NCoR, which like Snt1 are SANT domain proteins. The
HeLa complexes also include the seven WD40-repeat
protein, TBL1. Among the many seven WD40-repeat pro-
teins in mammals and yeast, Sif2 is clearly the yeast
homolog of TBL1. We speculate that further character-
izations of HDAC3 complexes may present more simi-
larities, including, possibly, the inclusion in HDAC3
complexes of an NAD-dependent histone deacetylase
and the uncharacterized SET and PHD domain protein
predicted by EST 095038, which is the closest human
homolog of Set3.

Function of the Set3C and the Hst1–Sum1 complex

Examination of the kinetics of sporulation landmarks
revealed that ⌬hos2 or ⌬set3 strains underwent normal
premeiotic DNA synthesis but showed a faster progres-
sion through meiosis. Deletion of hos2 has been reported
to reduce ascus formation dramatically (Bisland et al.
1998). We found only a slight reduction in ⌬hos2 strains,
both in the strain of Bisland et al. (1998) and the strain
used throughout our studies under various conditions
(Fig. 5; data not shown). Presumably, loss of hos2 has a
Figure 5. Set3C is a repressor of meiosis. Wild type, ⌬set3, or
more profound impact on ascus production under other
⌬hos2 diploid strains were sporulated and assayed for meiotic
landmarks at the times indicated (in hours). (A) Premeiotic conditions. In concordance with the fast meiotic pheno-
DNA synthesis as measured by FACS analysis. X-axis, fluores- types shown by ⌬hos2 or ⌬set3, loss of either protein
cence intensity; Y-axis, number of cells. (Left peaks) 2n DNA; disrupts Set3C, leaving subcomponents (Fig. 3) that did
(right peaks) 4n DNA (as confirmed by microscopic analysis of not retain Set3C function. Hence, repression by Set3C
sorted cells). (Blue graphs) wild type; (orange graphs) ⌬set3; may require both the essential components and the in-
(green graphs) ⌬hos2. (B) Cells completing meiosis I, meiosis II, tegrity of the complex.
and ascus formation (Asci) were counted using a fluorescence Northern analysis showed that Set3C function is di-
microscope. At least 300 cells per time point per strain were rected toward repression of clusters 4 and 5 of the sporu-
counted. (C) Northern blot analysis of meiotic genes during
lation gene program, including the key regulators ime2
sporulation of wild-type (wt) or ⌬set3 cells. Data were obtained
and ndt80. Set3C repression of ndt80 could be indirect,
from a separate experiment as in A and B, and the absolute time
of sporulation onset is therefore slightly different. Relative dif- as ime2 is required for expression of ndt80, but not vice
ferences between wild type and ⌬set3 are similar to those in B. versa (Hepworth et al. 1998). Hence, the fast meiotic
Genes examined are indicated next to the expression data, as are phenotypes shown by ⌬hos2 or ⌬set3, and the precocious
corresponding gene clusters (Primig et al. 2000) and gene func- induction of ndt80 and other later genes, could be solely
tions. The mRNA coding for the ribosomal protein Tcm1 was due to earlier Set3C repression of ime2 and other cluster
used as control for RNA. Decreased expression of this gene very 4 genes.
early during sporulation, followed by stable expression levels, is Consequently, an attractive explanation of repression
as reported (Chu and Herskowitz 1998). Expression of asc1 and by Set3C action involves Hos2 deacetylation of nucleo-
dal7 confirmed the quality of the RNA at these early time
somes near the ime2 gene. This presumes that a change
points. (D) As in C, except RNA was harvested from wild-type,
early after induction of sporulation, which could include
⌬hst1, and ⌬hst1,⌬hos2 strains.
altered acetylation of these nucleosomes, potentially ac-
tivates ime2 gene expression. However, action by Set3C
delays the onset of expression so that the regulatory cas-
plex deacetylate several histone substrates in vitro with cade controlled by Ime2 is delayed. This delay may serve
no apparent substrate specificity (data not shown). Fur- to improve sporulation efficiency, and/or to integrate a
thermore, Hst1 is dispensible for Set3C integrity and further regulatory checkpoint. This model requires a
function in meiotic repression of early/middle sporula- mechanism to abrogate Set3C repression in middle
tion genes (Figs. 3C, 5D). Consequently, the functional sporulation. Studies to test aspects of this model are un-
significance of combining the two types of histone derway, however, biochemical analysis of Set3C during

GENES & DEVELOPMENT 2999

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

Figure 6. The proteomic environment of Set3. The proteomic environment of Set3, associated factors, and their regulation of part of
the sporulation program is presented. The proteins involved are shown as colored objects below the regulatory interactions. Question
marks indicate uncertainties in the regulation circuitry. Note that Set3C repression applies in meiosis and Hst1–Sum1 repression in
vegetative growth. (MSE) Middle sporulation element. The Set3C includes two molecules of each Snt1 and Sif2 in a tetramer. Protein
interactions in Set3C are shown as black boxes, probable interactions as gray boxes. The +/− beside Hst1 and YOR279 indicates the
likelihood that these proteins are not always present in their respective complexes. Those proteins that are also found free are
indicated, as is the relationship between Hst1 and NAD.

sporulation showed no apparent changes in Set3C com- tinized all available two-hybrid data, including global
position between vegetative and sporulation conditions, analyses (Uetz et al. 2000; Ito et al. 2001) for interactions
indicating that regulation of Set3C activity does not in- between any of the factors encompassed here. Only an
volve loss or gain of proteins bound in Set3C (data not interaction between Sif2 and YIL112w was reported (in
shown). both studies; Uetz et al. 2000; Ito et al. 2001). This in-
Although Hst1 is a part of Set3C, it is not required for teraction was not detected by us with either TAP-tagged
the function of Set3C described here. Rather, Hst1 re- Sif2 or YIL112 in the absence of Set3 or Hos2, respec-
presses middle sporulation genes in vegetative growth, tively (Fig. 3), suggesting that the published two-hybrid
probably through the Hst1–Sum1 complex (Xie et al. results relied on the presence of endongenous Set3 and
1999; Fig. 2C). Hst1/Sum1 repression operates through Hos2. Ito et al. (2001) also reported interaction between
Sum1 binding the MSE (middle sporulation element) Sif2–YJR141w, which could not be confirmed by our bio-
DNA sequence element, which is present in the cluster chemical analysis. Furthermore, we failed to detect any
5 genes spr3, sps2, sps18, and the key regulator ndt80 evidence for the yeast two-hybrid interaction between
(Chu and Herskowitz 1998). Why Hst1 is also present in Sif2 and Sir4 (Cockell et al. 1998). The lack of concur-
Set3C is not clear. Because both Set3C and the Hst1– rence between our results and yeast two-hybrid results
Sum1 complex act as repressors, appear to include his- provokes questions about the fidelity and utility of two-
tone deacetylase activities (Fig. 4), and repress overlap- hybrid analyses (Seol et al. 2001). We think that coherent
ping sets of sporulation-specific genes, it is very likely documentation and understanding of proteomes must be
that the presence of Hst1 in Set3C has functional signifi- based on core data sets composed of strong, biochemi-
cance. As with the questions regarding targets and cally documented, interactions. Mapping of two-hybrid
mechanism of Set3C repression discussed above, further data onto these core data sets may sort out true from
work is required to define the events involved in regula- false positives and point to higher-order regulatory cir-
tion of Set3C and Hst1/Sum1 repression. cuits.

Methodological approaches in proteomics Materials and methods

Ongoing discussions about proteomics have focused on
the methodological options available (Rigaut et al. 1999; Sequence analysis
Shevchenko et al. 1999; Vidal 2001), with global yeast The yeast genome was searched for SET domains and PHD fin-
two-hybrid approaches assuming prominence. We scru- gers by BLASTP searches (Altschul et al. 1997) and Profilesearch

3000 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Meiotic repression by SET3 deacetylase complex

(as implemented on the Bioccelerator facility at EMBL; http:// of IgG Sepharose (Pharmacia), equilibrated in buffer E (Logie and
eta.embl-heidelberg.de:8000/) using profiles based on align- Peterson 1999) for 2 h at 4°C by use of a Bio-Rad disposable
ments of previously identified SET domains and PHD fingers column. Two to three columns (the equivalent of 4–6 L yeast
(http://www.uib.no/aasland/set/). culture at OD600 2–3) were used per purification shown. The
IgG Sepharose column was washed with 35 mL of buffer E lack-
ing proteinase inhibitors, followed by 10 mL of TEV cleavage
Strains buffer (Rigaut et al. 1999). TEV cleavage was performed using 10
Strains are listed in Table 1. Yeast transformations were per- µL of rTEV (GIBCO) in 1 mL of TEV cleavage buffer for 2 h at
formed as described (Soni et al. 1993). All haploid strains were 16°C. Calmodulin Sepharose (Stratagene) purification was per-
derived from MGD353-13D (Puig et al. 1998). The TAP tag was formed as described (Rigaut et al. 1999). Purified proteins were
introduced at the endogenous locus of the gene of interest as an concentrated as described by Wessel and Flügge (1984). After
in-frame C-terminal fusion with the coding sequence as de- separation on 7%–25% SDS–polyacrylamide gels, proteins were
scribed (Puig et al. 1998). Gene disruptions were performed as stained with Coomassie, in-gel digested with trypsin, and iden-
described (Puig et al. 1998). The diploid strains ⌬set3 and ⌬hos2 tified by MS as described (Shevchenko et al. 1996a,b).
were obtained by crossing to BSY 323 (Mat␣ genotype) followed
by tetrad analysis. Resulting haploids carrying the desired dis-
Histone deacetylase assay
ruption were mated. Correct modifications were confirmed by
PCR and Western blot analysis. The double mutant ⌬hst1 Yeast Histone H4 peptide (amino acids 2–20, Upstate) was
⌬hos2 (DS140) was obtained by integrating a K.l. ura3 disrup- chemically acetylated by use of sodium [3H]acetate (NEN, 2.8
tion cassette flanked by loxP sites. Transient expression of Cre- Ci/mmole) as described (Taunton et al. 1996), and purified by
recombinase resulted in cassette excision. The hos2 ORF was gel filtration over G15 Sepharose. IgG Sepharose eluates were
then disrupted by inserting a cassette, again using K.l. ura3 as prepared as above, and were added to 30,000 dpm 3H H4 peptide
selectable marker. in a final volume of 30 µL 10 mM Tris (pH 8.0), 150 mM NaCl,
0.1 µg/µL BSA, and 0.5 mM NAD or NADH (Sigma) as indi-
cated. After incubation at 30°C for 90 min, the reaction was
TAP purification and mass spectrometry spotted on P81 phosphocellulose paper (Whatmann). Filters
The extraction of yeast cells was performed as described for the were washed four times for 15 min in 50 mM NaHCO3 (pH 9),
yeast Swi/Snf complex (Logie and Peterson 1999). The TAP tag dried completely, and counted in 2 mL of scintillation cocktail.
consists of a calmodulin binding peptide (CBP), a TEV protease
cleavage site, and two IgG-binding units of protein A, respec-
Sporulation assays
tively, as described (Rigaut et al. 1999). TAP purification was
performed according to Rigaut et al. (1999), with the following Diploid strains were grown in YPD to OD600 of 2.0. Cells were
modifications: 10 mL of supernatant of the 43,000 rpm centrifu- harvested by centrifugation, and resuspended in sporulation
gation (Logie and Peterson 1999) was allowed to bind to 200 µL medium (1.5% KAc at pH 7.5, dilution 1:3) supplemented with

Table 1. S. cerevisiae strains used in this study

Name Genotype G Strains Reference

−
MGD-353- Mata ade2; arg4 (RV ), leu2-3, 112; trp1-289l ura3-52 Puig et al. 1998
13D
BSY320 Mata/␣ ade2/ade2; arg4 (RV−)/arg4 (RV−); leu2-3,112/leu2-3, 112; trp1-289; ura3-52/ura3-52 Puig et al. 1998
BSY323 Mat ␣ ade2; leu2-3,112; lys2;trp1-289; ura3-52 This study
BSY792 Mata set3 C-term. TAP tagged introducing K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS4 Mata yill 12w C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS6 Mata snt1 C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS8 Mata sif2 C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS9 Mata hos2 C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS22 Mata cpr1 C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
AR5 Mata hst1 C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS101 Mata sum1 C-term. TAP K.l. TRP1 ade2; arg4 (RV−); leu2-3,112; ura3-52 This study
DS16 Mata set3⬋ K.l. URA3 ade2; arg4 (RV−); leu2-3,112; trp1-289 This study
DS18 Mata hos2⬋ K.l. URA3 ade2; arg4 (RV−); leu2-3,112; trp1-289 This study
DS19 Mata yil112w C-term. TAP K.l. TRP1; hos2⬋K.l.URA3 ade2; arg4 (RV−); leu2-3,112 This study
DS21 Mata sif2 C-term. TAP K.l. TRP1; set3⬋K.l.URA3 ade2; arg4 (RV−); leu2-3,112 This study
DS169 Mata sif2 C-term. TAP K.l. TRP1; hst1⬋K.l.URA3 ade2; arg4 (RV−); leu2-3,112 This study
DS52-1 Mata/␣ set3⬋ K.l. URA3/set3⬋K.l.URA3 ade2/ade2; ARG4/arg4 (RV−); leu2-3,112/leu2-3,112; This study
lys2/LYS2; trp1-289/trp1-289
DS53-3 Mata/␣ hos2⬋ K.l. URA3/hos2⬋K.l.URA3 ade2/ade2; ARG4/arg4 (RV−); leu2-3,112/ This study
leu2-3,112; lys2/LYS2; trp1-289/trp1-289
DS129 Mata/␣ Hst1⬋floxed K.l. URA3/hst1⬋floxedK.l.URA3 ade2/ade2; ARG4/arg4 (RV−); This study
leu2-3,112/leu2-3,112; lys2/LYS2; trp1-289/trp1-289
DS140 Mata/␣ hst1⬋loxP/hst1⬋loxP hos2⬋K.l. URA3/hos2⬋K.l.URA3 ade2/ade2; ARG4/arg4 (RV−); This study
leu2-3,112/leu2-3,112; lys2/LYS2; trp1-289/trp1-289

GENES & DEVELOPMENT 3001

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

required amino acids. Cultures were incubated at 23°C, and program of sporulation in budding yeast. Science 282: 699–
samples were taken at the times indicated. DNA was stained 705.
with propidium iodide as described (Haase and Lew 1997). Cells Cockell, M., Renauld, H., Watt, P., and Gasser, S.M. 1998. Sif2p
were sonicated three times for 2 sec with 5-sec intervals at 20% interacts with Sir4p amino-terminal domain and antago-
power to resolve cell clumps. DNA synthesis was measured by nizes telomeric silencing in yeast. Curr. Biol. 8: 787–790.
FACS analysis. Microscopic analysis showed that this treat- Cui, X., De Vivo, I., Slany, R., Miyamoto, A., Firestein, R., and
ment did not disrupt individual cells. Progression into meiosis Cleary, M.L. 1998. Association of SET domain and myotu-
was determined by counting at least 300 cells per point using a bularin-related proteins modulates growth control. Nat.
fluorescent microscope. Northern blot analysis was performed Genet. 18: 331–337.
by use of standard procedures. Probes were prepared by random Dolinski, K. and Heitman, J. 1997. Peptidyl-prolyl isomerases
prime labeling of PCR fragments of the genes indicated. (PPIases). In Guidebook to molecular chaperones and pro-
tein folding catalysts (ed. S. Tooze), pp. 359–369. Oxford
University Press, Oxford, UK.
Acknowledgments Edmondson, D.G., Smith, M.M., and Roth, S.Y. 1996. Repres-
sion domain of the yeast global repressor Tup1 interacts di-
We thank Colin Logie, Henk Stunnenberg, Alexander Brehm,
rectly with histones H3 and H4. Genes & Dev. 10: 1247–
Peter Becker, Jeanette Rientjes, Elisabeth Bragado-Nilsson,
1259.
Wolfgang Zachariae, Anne Atzberger, Susan Gasser, and Moira
Ekwall, K., Nimmo, E.R., Javerzat, J.P., Borgstrom, B., Egel, R.,
Cockell for reagents, advice, and discussions; Joep Muyrers and
Cranston, G., and Allshire, R. 1996. Mutations in the fission
Julia Schaft for auxiliary experiments; and Michelle Meredyth
yeast silencing factors clr4+ and rik1+ disrupt the localisa-
for critical reading of the manuscript. W.P. was supported by an
tion of the chromo domain protein Swi6p and impair cen-
EU Biotechnology fellowship.
tromere function. J. Cell. Sci. 109: 2637–2648.
The publication costs of this article were defrayed in part by
Guenther, M.G., Lane, W.S., Fischle, W., Verdin, E., Lazar,
payment of page charges. This article must therefore be hereby
M.A., and Shiekhattar, R. 2000. A core SMRT corepressor
marked “advertisement” in accordance with 18 USC section
complex containing HDAC3 and TBL1, a WD40-repeat pro-
1734 solely to indicate this fact.
tein linked to deafness. Genes & Dev. 14: 1048–1057.
Haase, S.B. and Lew, D.J. 1997. Flow cytometric analysis
of DNA content in budding yeast. Methods Enzymol.
References
283: 322–332.
Aasland, R., Gibson, T.J., and Stewart, A.F. 1995. The PHD Hassig, C.A., Tong, J.K., Fleischer, T.C., Owa, T., Grable, P.G.,
finger: Implications for chromatin-mediated transcriptional Ayer, D.E., and Schreiber, S.L. 1998. A role for histone
regulation. Trends Biochem. Sci. 20: 56–59. deacetylase activity in HDAC1-mediated transcriptional re-
Aasland, R., Stewart, A.F., and Gibson, T. 1996. The SANT pression. Proc. Natl. Acad. Sci. 95: 3519–3524.
domain: A putative DNA-binding domain in the SWI-SNF Hepworth, S.R., Friesen, H., and Segall, J. 1998. NDT80 and the
and ADA complexes, the transcriptional co-repressor N-CoR meiotic recombination checkpoint regulate expression of
and TFIIIB. Trends Biochem. Sci. 21: 87–88. middle sporulation-specific genes in Saccharomyces cerevi-
Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, siae. Mol. Cell. Biol. 18: 5750–5761.
Z., Miller, W., and Lipman, D.J. 1997. Gapped BLAST and Huang, N., vom Baur, E., Garnier, J.M., Lerouge, T., Vonesch,
PSI-BLAST: A new generation of protein database search pro- J.L., Lutz, Y., Chambon, P., and Losson, R. 1998. Two dis-
grams. Nucleic Acids Res. 25: 3389–3402. tinct nuclear receptor interaction domains in NSD1, a novel
Angrand, P.-O., Apiou, F, Stewart, A.F., Dutrillaux, B., Losson, SET protein that exhibits characteristics of both corepres-
R., and Chambon, P. 2001. NSD3, a new SET domain-con- sors and coactivators. EMBO J. 17: 3398–3412.
taining gene, maps to 8p12 and is amplified in human breast Imai, S., Armstrong, C.M., Kaeberlein, M., and Guarente, L.
cancer cell lines. Genomics 74: 79–88. 2000. Transcriptional silencing and longevity protein Sir2 is
Arevalo-Rodriguez, M., Cardenas, M.E., Wu, X., Hanes, S.D., an NAD-dependent histone deacetylase. Nature 403: 795–
and Heitman, J. 2000. Cyclophilin A and Ess1 interact with 800.
and regulate silencing by the Sin3-Rpd3 histone deacetylase. Ingham, P.W. 1998. Trithorax and the regulation of homeotic
EMBO J. 19: 3739–3749. gene expression in Drosophila: A historical perspective. Int.
Bisland, E., Dahlén, M., and Sunnerhagen, P. 1998. Genomic J. Dev. Biol. 42: 423–429.
disruption of six budding yeast genes gives one drastic Ito, T., Chiba, T., Ozawa, R., Yoshida, M., Hattori, M., and
example of phenotype strain-dependence. Yeast 14: Sakaki, Y. 2001. A comprehensive two-hybrid analysis to
655–664. explore the yeast protein interactome. Proc. Natl. Acad. Sci.
Brachmann, C.B., Sherman, J.M., Devine, S.E., Cameron, E.E., 98: 4569–4574.
Pillus, L., and Boeke, J.D. 1995. The SIR2 gene family, con- Ivanova, A.V., Bonaduce, M.J., Ivanov, S.V., and Klar, A.J. 1998.
served from bacteria to humans, functions in silencing, cell The chromo and SET domains of the Clr4 protein are es-
cycle progression, and chromosome stability. Genes & Dev. sential for silencing in fission yeast. Nat. Genet. 19: 192–
9: 2888–2902. 195.
Capili, A.D., Schultz, D.C., Rauscher, I.F., and Borden, K.L. Jones, R.S. and Gelbart, W.M. 1993. The Drosophila Polycomb-
2001. Solution structure of the PHD domain from the KAP-1 group gene Enhancer of zeste contains a region with se-
corepressor: structural determinants for PHD, RING and quence similarity to trithorax. Mol. Cell. Biol. 13: 6357–
LIM zinc-binding domains. EMBO J. 20: 165–177. 6366.
Chu, S. and Herskowitz, I. 1998. Gametogenesis in yeast is regu- Kumar, A., Kassavetis, G.A., Geiduschek, E.P., Hambalko, M.,
lated by a transcriptional cascade dependent on Ndt80. Mol. and Brent, C.J. 1997. Functional dissection of the B⬘ compo-
Cell 1: 685–696. nent of RNA polymerase III transcription factor IIIB: A scaf-
Chu, S., DeRisi, J., Eisen, M., Mulholland, J., Botstein, D., folding protein with multiple roles in assembly and initia-
Brown, P.O., and Herskowitz, I. 1998. The transcriptional tion of transcription. Mol. Cell. Biol. 17: 1868–1880.

3002 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Meiotic repression by SET3 deacetylase complex

Landry, J., Sutton, A., Tafrov, S.T., Heller, R.C., Stebbins, J., RAX and ASH1 proteins. Oncogene 19: 351–357.
Pillus, L., and Sternglanz, R. 2000. The silencing protein Rundlett, S.E., Carmen, A.A., Kobayashi, R., Bavykin, S.,
SIR2 and its homologs are NAD-dependent protein deacety- Turner, B.M., and Grunstein, M. 1996. HDA1 and RPD3 are
lases. Proc. Natl. Acad. Sci. 97: 5807–5811. members of distinct yeast histone deacetylase complexes
Leroux, M.R. and Hartl, F.U. 2000. Protein folding: Versatility of that regulate silencing and transcription. Proc. Natl. Acad.
the cytosolic chaperonin TRiC/CCT. Curr. Biol. 10: R260– Sci. 93: 14503–14508.
R264. Rusché, L.N. and Rine, J. 2001. Conversion of a gene-specific
Li, J., Wang, J., Nawaz, Z., Liu, J.M., Qin, J., and Wong, J. 2000. repressor to a regional silencer. Genes & Dev. 15: 955–967.
Both corepressor proteins SMRT and N-CoR exist in large Sedgwick, S.G. and Smerdon, S.J. 1999. The ankyrin repeat: A
protein complexes containing HDAC3. EMBO J. 19: 4342– diversity of interactions on a common structural framework.
4250. Trends Biochem. Sci. 24: 311–316.
Lindgren, A., Bungard, D., Pierce, M., Xie, J., Vershon, A., and Seol, J.H., Shevchenko, A., Shevchenko, A., and Deshaies, R.J.
Winter, E. 2000. The pachytene checkpoint in Saccharomy- 2001. Skp1 forms multiple protein complexes, including
ces cerevisiae requires the Sum1 transcriptional repressor. RAVE, a regulator of V-ATPase assembly. Nat. Cell. Biol.
EMBO J. 19: 6489–6497. 3: 384–391.
Logie, C. and Peterson, C.L. 1999. Purification and biochemical Shevchenko, A., Zachariae, W., and Shevchenko, A. 1999. A
properties of yeast SWI/SNF complex. Methods Enzymol. strategy for the characterization of protein interaction net-
304: 726–741. works by mass spectrometry. Biochem. Soc. Trans. 27: 549–
Mitchell, A.P., Driscoll, S.E., and Smith, H.E. 1990. Positive 554.
control of sporulation genes by the Ime1 and Ime2 products Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. 1996a.
in Saccharomyces cerevisiae. Mol. Cell. Biol. 10: 2104–2110. Mass spectrometric sequencing of proteins silver-stained
Nislow, C., Ray, E., and Pillus, L. 1997. SET1, a yeast member polyacrylamide gels. Anal. Chem. 68: 850–858.
of the trithorax family, functions in transcriptional silenc- Shevchenko, A., Jensen, O.N., Podtelejnikov, V., Sagliocco, F.,
ing and diverse cellular processes. Mol. Biol. Cell 8: 2421– Wilm, M., Vorm, O., Mortensen, P., Shevchenko, A.,
2436. Boucherie, H., and Mann, M. 1996b. Linking genome and
Palaparti, A., Baratz, A., and Stifani, S. 1997. The Groucho/ proteome by mass spectrometry: Large-scale identification
transducin-like enhancer of split transcriptional repressors of yeast proteins from two dimensional gels. Proc. Natl.
interact with the genetically defined amino-terminal silenc- Acad. Sci. 93: 14440–14445.
ing domain of histone H3. J. Biol. Chem. 272: 26604–26610. Smith, J.S., Brachmann, C.B., Celic, I., Kenna, M.A., Muham-
Pascual, J., Martinez-Yamout, M., Dyson, H.J., and Wright, P.E. mad, S., Starai, V.J., Avalos, J.L., Escalante-Semerena, J.C.,
2000. Structure of the PHD zinc finger from human Willi- Grubmeyer, C., Wolberger, C., et al. 2000. A phylogeneti-
ams-Beuren syndrome transcription factor. J. Mol. Biol. cally conserved NAD+-dependent protein deacetylase activ-
304: 723–729. ity in the Sir2 protein family. Proc. Natl. Acad. Sci.
Pirrotta, V. 1998. Polycombing the genome: PcG, trxG, and 97: 6658–6663.
chromatin silencing. Cell 93: 333–336. Soni, R., Carmichael, J.P., and Murray, J.A. 1993. Parameters
Primig, M., Williams, R.M., Winzeler, E.A., Tevzadze, G.G., affecting lithium acetate-mediated transformation of Sac-
Conway, A.R., Hwang, S.Y., Davis, R.W., and Esposito, R.E. charomyces cerevisiae and development of a rapid and sim-
2000. The core meiotic transcriptome in budding yeasts. plified procedure. Curr. Genet. 24: 455–459.
Nat. Genet. 26: 415–423. Stassen, M.J., Bailey, D., Nelson, S., Chinwalla, V., and Harte,
Puig, O., Rutz, B., Luukkonen, B.G., Kandels-Lewis, S., Bragado- P.J. 1995. The Drosophila trithorax proteins contain a novel
Nilsson, E., and Seraphin, B. 1998. New constructs and strat-
variant of the nuclear receptor type DNA binding domain
egies for efficient PCR-based gene manipulations in yeast.
and an ancient conserved motif found in other chromosomal
Yeast 14: 1139–1146.
proteins. Mech. Dev. 52: 209–223.
Rea, S., Eisenhaber, F., O’Carroll, D., Strahl, B.D., Sun, Z.W.,
Stec, I., Wright, T.J., van Ommen, G.J., de Boer, P.A., van
Schmid, M., Opravil, S., Mechtler, K., Ponting, C.P., Allis,
Haeringen, A., Moorman, A.F., Altherr, M.R., and den Dun-
C.D., et al. 2000. Regulation of chromatin structure by site-
nen, J.T. 1998. WHSC1, a 90 kb SET domain-containing gene
specific histone H3 methyltransferases. Nature 406: 593–
maps in the Wolf-Hirschorn syndrome critical region and is
599.
fused to IgH in t(4;14) multiple myeloma. Hum. Mol. Genet.
Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., and
Seraphin, B. 1999. A generic protein purification method for 7: 1071–1082.
protein complex characterization and proteome exploration. Taunton, J., Hassig, C.A., and Schreiber, S.L. 1996. A mamma-
Nat. Biotechnol. 17: 1030–1032. lian histone deacetylase related to the yeast transcriptional
Roguev, A., Schaft, D., Shevchenko, A., Pijnappel, W.W.M., regulator Rpd3p. Science 272: 408–411.
Wilm, M., Aasland, R. and Stewart, A.F. 2001. The S. cer- Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., and
evisiae Set1 complex includes an Ash2 homolog and meth- Higgins, D.G. 1997 The CLUSTAL_X windows inter-
ylates histone 3 lysine 4. EMBO J. (in press). face: Flexible strategies for multiple sequence alignment
Rozenblatt-Rosen, O., Rozovskaia, T., Burakov, D., Sedkov, Y., aided by quality analysis tools. Nucleic Acids Res. 25: 4876–
Tillib, S., Blechman, J., Nakamura, T., Croce, C.M., Mazo, 4882.
A., and Canaani, E. 1998. The C-terminal SET domains of Tripoulas, N., LaJeunesse, D., Gildea, J., and Shearn, A. 1996.
ALL-1 and TRITHORAX interact with the INI1 and SNR1 The Drosophila ash1 gene product, which is localized at spe-
proteins, components of the SWI/SNF complex. Proc. Natl. cific sites on polytene chromosomes, contains a SET domain
Acad. Sci. 95: 4152–4157. and a PHD finger. Genetics 143: 913–928.
Rozovskaia, T., Rozenblatt-Rosen, O., Sedkov, Y., Burakov, D., Tschiersch, B., Hofmann, A., Krauss, V., Dorn, R., Korge, G.,
Yano, T., Nakamura, T., Petruck, S., Ben-Simchon, L., and Reuter, G. 1994. The protein encoded by the Drosophila
Croce, C.M., Mazo, A., et al. 2000. Self-association of the position-effect variegation suppressor gene Su(var)3–9 com-
SET domains of human ALL-1 and of Drosophila TRITHO- bines domains of antagonistic regulators of homeotic gene

GENES & DEVELOPMENT 3003

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

Pijnappel et al.

complexes. EMBO J. 13: 3822–3831.

Uetz, P., Giot, L., Cagney, G., Mansfield, T.A., Judson, R.S.,
Knight, J.R., Lockshon, D., Narayan, V., Srinivasan, M., Po-
chart, P., et al. 2000. A comprehensive analysis of protein-
protein interactions in Saccharomyces cerevisiae. Nature
403: 623–627.
Vidal, M. 2001 A biological atlas of functional maps. Cell
104: 333–339.
Wakimoto, B.T. 1998. Beyond the nucleosome: Epigenetic as-
pects of position-effect variegation in Drosophila. Cell
93: 321–324.
Wessel, D. and Flugge, U.I. 1984. A method for the quantitative
recovery of protein in dilute solution in the presence of de-
tergents and lipids. Anal. Biochem. 138: 141–143.
Xie, J., Pierce, M., Gailus-Durner, V., Wagner, M., Winter, E.,
and Vershon, A.K. 1999. Sum1 and Hst1 repress middle
sporulation-specific gene expression during mitosis in Sac-
charomyces cerevisiae. EMBO J. 18: 6448–6454.
Yu, B.D., Hess, J.L., Horning, S.E., Brown, G.A., and Korsmeyer,
S.J. 1995. Altered Hox expression and segmental identity in
Mll-mutant mice. Nature 378: 505–508.

3004 GENES & DEVELOPMENT

 Downloaded from genesdev.cshlp.org on June 8, 2016 - Published by Cold Spring Harbor Laboratory Press

The S. cerevisiae SET3 complex includes two histone deacetylases,

Hos2 and Hst1, and is a meiotic-specific repressor of the sporulation
gene program
W.W.M. Pim Pijnappel, Daniel Schaft, Assen Roguev, et al.

Genes Dev. 2001 15: 2991-3004

Access the most recent version at doi:10.1101/gad.207401

References This article cites 62 articles, 31 of which can be accessed free at:
http://genesdev.cshlp.org/content/15/22/2991.full.html#ref-list-1

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the top
Service right corner of the article or click here.

To subscribe to Genes & Development go to:

http://genesdev.cshlp.org/subscriptions

Cold Spring Harbor Laboratory Press