Vps75, A New Yeast Member of the NAP Histone Chaperone Family*

Homologues of nucleosome assembly protein 1 (NAP1) are found throughout eukaryotes. Here we identify and characterize a new NAP family histone chaperone from budding yeast, named Vps75. Purified Vps75 preferentially binds histone H3/H4 tetramers and is capable of assembling nucleosomes in vitro. In vivo, Vps75 is associated with the chromatin of both active and inactive genes and telomeres. Others have previously reported that Vps75 forms a complex with Rtt109, required for acetylation of histone H3 lysine 56 (H3 Lys-56). Cells lacking RTT109 are sensitive to hydroxyurea, pointing to a role in replication. We show that VPS75 is not required for H3 Lys-56 acetylation and that vps75Δ cells are insensitive to hydroxyurea, suggesting that although Rtt109 and Vps75 associate and are likely to be functionally connected, they also have separate roles.

Homologues of nucleosome assembly protein 1 (NAP1) are found throughout eukaryotes. Here we identify and characterize a new NAP family histone chaperone from budding yeast, named Vps75. Purified Vps75 preferentially binds histone H3/H4 tetramers and is capable of assembling nucleosomes in vitro. In vivo, Vps75 is associated with the chromatin of both active and inactive genes and telomeres. Others have previously reported that Vps75 forms a complex with Rtt109, required for acetylation of histone H3 lysine 56 (H3 Lys-56). Cells lacking RTT109 are sensitive to hydroxyurea, pointing to a role in replication. We show that VPS75 is not required for H3 Lys-56 acetylation and that vps75⌬ cells are insensitive to hydroxyurea, suggesting that although Rtt109 and Vps75 associate and are likely to be functionally connected, they also have separate roles.
In eukaryotic cells, DNA is present in a highly compacted form called chromatin. The repeating unit of chromatin is the nucleosome, formed from two histone H2A-H2B dimers and one histone H3-H4 tetramer around which 147 bp of DNA is wrapped (1). Despite the tight packaging of its structure, chromatin is highly dynamic, a characteristic that is vital in regulating nuclear processes, such as transcription and replication, which require access to DNA. Processes that positively influence chromatin fluidity include post-translational modifications of histones, incorporation of histone variants, and histone exchange by dedicated histone chaperones (for review, see Ref. 2).
Histone chaperones are proteins that regulate the interaction of histones with other proteins and DNA and also act to prevent the highly basic histones from forming inappropriate complexes. In addition to playing an important role in histone exchange during nuclear processes, histone chaperones function in nucleocytoplasmic shuttling of histones, histone storage, nucleosome assembly, and as a link between chromatin remodeling factors and histones (for reviews, see Refs. 3 and 4). Given the diversity of these roles, it is not surprising that many different histone chaperones have been identified, including nucleoplasmin, ASF1, HIR proteins, Spt6, FACT, ACF, CAF1, and NAP1.
Members of the NAP 2 family of proteins, found from yeast to mammals, possess particularly diverse roles in histone metabolism (4). These proteins are characterized by a conserved central domain (the NAP domain), a non-conserved N-terminal region of variable length, and in general, a highly acidic C-terminal region. The NAP domain is required and sufficient for histone binding, although the acidic domain may enhance this activity (5). The best characterized member of the NAP family is NAP1, which has been found in yeast, plants, flies, and mammals. Other NAPs include SET (also known as TAF-I␤), ClNAP, TSPY, and a steadily expanding group of NAP1-like proteins (NAP1L1-NAP1L5) (4).
In this study, we describe the isolation and characterization of a novel NAP family member from yeast. This protein, Vps75, was previously identified in genomic screens for vacuolar protein sorting genes and for factors that affect telomere length (6,7). Our data indicate that Vps75 is a bona fide histone chaperone with a preference for H3-H4 tetramers and that it may have a general role in chromatin assembly and rearrangement since it is found in both transcribed and non-transcribed regions in vivo.
Expression and Purification of Recombinant Proteins-To purify Vps75 from yeast, HTM-tagged Vps75 cells were harvested, resuspended in an equal volume (w/v) of lysis buffer (40 mM Hepes (pH 7.5), 250 mM potassium acetate, 20% glycerol, 5 mM dithiothreitol, 0.1% Nonidet P-40, protease inhibitors), and lysed using a freezer mill. Lysate was clarified at 12,000 ϫ g for 10 min and again in an ultracentrifuge using a Beckman Ti45 rotor (40,000 rpm, 1 h, 4°C). The supernatant was incubated with 9E10 affinity resin for 90 min at 4°C, and the resin was washed with 40 bed volumes lysis buffer and 10 volumes elution buffer (40 mM Tris-HCl (pH 7.5), 250 mM potassium acetate, 10% glycerol). Vps75-His was eluted twice by incubating with 1 volume elution buffer containing 40 g of TEV protease for 2 h at 12°C and further purified by MonoQ chromatography. GST-Vps75 was expressed in Escherichia coli BL21 cells from pGEX-6P-1 and purified on glutathione-Sepharose using standard techniques.
Histone expression vectors were a kind gift from Brad Cairns. Individual yeast histones were purified from E. coli, and histone * This work was supported by a generous in-house grant (to J. Q. S.) and by a European Molecular Biology Organization (EMBO) long-term fellowship (to L. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed. complexes (H2A-H2B dimers, H3-H4 tetramers, and octamers) were prepared as described (8). HeLa core histones used in nucleosome assembly assays were prepared as described (9). Mass Spectrometry-Protein samples were analyzed by matrix-assisted laser desorption/ionization-time of flight using an Applied Biosystems 4700 proteomics analyzer.
Binding Reactions and GST Pull-down Assays-To assess Vps75 binding to core histones, 9 pmol of Vps75-His purified from yeast was incubated with 20 l of core histones immobilized on 4% agarose (Sigma H3889) or empty 4% agarose for 2 h at 4°C in 500 l of buffer A (20 mM Tris-HCl (pH 7.5), 10% glycerol, 50 g/ml bovine serum albumin, 0.1% Nonidet P-40, 1 mM EDTA, 1 mM dithiothreitol) containing 200 mM NaCl. Resin was washed in buffer A, and bound protein was subjected to SDS-PAGE and Western blotting using an antibody to the His tag (Sigma).
GST pull-down assays were done by incubating GST-Vps75 (or GST) and yeast histone complexes with 20 l of glutathione-Sepharose (Amersham Biosciences) in 250 l of buffer A containing 100 mM NaCl for 3 h at 10°C. After extensive washing, bound proteins were subjected to SDS-PAGE and visualized by silver staining.
Nucleosome Assembly Assays-Vps75 nucleosome assembly activity was measured essentially as described (5). 60 fmol of pRS316 plasmid was relaxed by incubation with 3 units of calf thymus topoisomerase I (Invitrogen) in assay buffer (10 mM Tris-HCl (pH 8.0), 150 mM potassium acetate, 1 mM EDTA, 0.1 mg/ml bovine serum albumin) for 30 min at 37°C. HeLa core histones (0.5-2 pmol) and Vps75 (from yeast; 4.5-18 pmol) were preincubated in assay buffer for 30 min at 37°C. Samples were then mixed with 60 fmol of the relaxed pRS316 and further incubated at 37°C for 45 min. The reactions were analyzed by agarose gel electrophoresis as described (7).
Histone Association and Chromatin Immunoprecipitation (ChIP) Assays-Qualitative histone association assays were carried out as described (10). Chromatin was immunoprecipitated with an antibody specific for the C terminus of histone H3 (a kind gift of Alain Verreault), and Vps75 was detected using a monoclonal antibody (9E10) directed against the Myc tag. ChIP assays were carried out essentially as described (11) but with minor modifications; DNA co-precipitated with 9E10 was analyzed by quantitative real-time PCR using a Bio-Rad MyIQ iCy-cler in standard conditions and ABsolute TM QPCR SYBR Green reagents (ABgene). Primer sequences are available on request. Immunoprecipitation efficiency was determined by dividing the amount of precipitated DNA by the amount in the input sample.
Growth Sensitivity Assays-The effect of hydroxyurea on growth was tested as described (12).

RESULTS
Sequence Analysis of Vps75-Although several NAP family members have been described in human and mouse cells, only one, NAP1, has been identified in yeast. However, homology searches revealed that yeast contains at least one other likely NAP protein, Vps75. A BLAST query of the Vps75 protein sequence indicated that a region encompassing amino acids 12-229 shares homology with the NAP domain. The structure of the NAP domain of yeast NAP1 (yNAP1) at 3.0 Å resolution was recently solved (13). It comprises two distinct domains (I and II), each of which can be further delineated into two subdomains (Fig. 1A). Sequence alignment of Vps75 with various other NAP family members revealed that it is more closely related to the Drosophila and human SET/TAF-I and human TSPY proteins than yNAP1 (data not shown). This is supported by analysis of its domain architecture (Fig. 1A). Like SET and TSPY, Vps75 completely lacks subdomain B, termed the accessory domain (AD; shown in yellow in yNAP1). The AD has been implicated in regulating access to a functional nuclear export signal (NES; shown as a blue/white stripe) embedded in A (shown in blue) (13). Vps75, SET, and TSPY also lack a recognizable NES (data not shown), suggesting an evolutionary link between the conservation of the NES and the AD and supporting the idea that this domain plays a role in regulating access to the NES. These observations raise the possibility that unlike yNAP1 (14), Vps75 does not function in nucleocytoplasmic shuttling of histones but possesses a primarily nuclear role. This theory is consistent with the observations that Vps75 is found in the nucleus (15) and that it lacks the phosphorylation sites implicated in translocation of Drosophila NAP1 between the cytoplasm and the nucleus (16).
In contrast to the apparent differences observed between NAP domain I of Vps75 and yNAP1, domain II appears to be structurally conserved, apart from a small acidic region found near the start of subdomain D in yNAP1 that is lacking in Vps75. However, Vps75 contains a highly acidic region near its C terminus (Fig. 1A), a characteristic of most NAP family proteins, and has a theoretical pI of 4.47, which is also similar to other members of the family.
Purification of Vps75-To experimentally address the possibility that Vps75 is a NAP family histone chaperone, we created a yeast strain expressing Vps75 tagged at the C terminus with HTM from the endogenous VPS75 promoter. Crude extract was passed over a 9E10 (anti-Myc) affinity column, and Vps75- His was eluted by TEV-mediated cleavage (Fig. 1B, lane 1). Copurifying proteins were identified by mass spectrometry and included Ssb1, Rtt109, and Rlp24. These proteins were previously found to interact with Vps75 in a global screen of yeast protein complexes (17). Using anion-exchange chromatography, we were subsequently able to purify Vps75 to virtual homogeneity (Fig. 1B, lane 2). The absence of co-purifying proteins after MonoQ chromatography could indicate that these were contaminants and not truly associated with Vps75 in vivo. However, Vps75 interacts strongly with Q-Sepharose, and weakly associated proteins might be displaced by washing. As such, it is possible that at least Rtt109 exists in a complex with Vps75 in physiological salt conditions. Supporting this idea, Krogan et al. (17) found a stoichiometric complex of Vps75 and Rtt109 when either protein was purified by tandem affinity purification. Our own efforts indicate that Vps75 and Rtt109 form a complex that can be dissociated by high salt concentrations during isolation (data not shown).
Vps75 Preferentially Binds H3-H4 Tetramers in Vitro-The defining feature of a histone chaperone is its ability to bind histones. To determine whether Vps75 possesses such an activity, we incubated it with resin-immobilized core histones.
Vps75-His bound efficiently to this matrix but not to the control matrix lacking histones ( Fig. 2A).
Nucleosome Assembly Activity of Vps75-A number of NAP family members are capable of assembling chromatin on naked DNA in vitro (4). The plasmid supercoiling assay, which measures the assembly activity on the basis of the introduction of negative supercoils into a relaxed DNA template (18), was used to test whether Vps75 also possesses such activity. Vps75-His introduced negative supercoils on plasmid DNA, an effect that was dose-dependent (Fig.  2C, lanes 3-5) and that required histones (lanes [7][8][9]. This is indicative of nucleosome assembly and suggests that Vps75 may function in chromatin dynamics in vivo. Association of Vps75 with Chromatin in Vivo-To determine whether Vps75 associates with histones and/or chromatin in vivo, we first utilized a histone association assay (10). Protein-DNA complexes in yeast cells producing HTM-tagged Vps75 were cross-linked by formaldehyde treatment, and the resulting chromatin was sheared by extensive sonication to yield pieces averaging 500 bp in length. Immunoprecipitation with a histone H3-specific antibody was carried out, and following reversal of the cross-links, co-precipitated Vps75 was detected by immunoblotting. A proportion of cellular Vps75 was indeed found to be associated with histone H3 (Fig. 3A). In theory, this association could be mediated by DNA or via a direct interaction between Vps75 and histone H3. Because of the resistance of cross-linked chromatin to nuclease treatment (10), we were unable to distinguish between these possibilities. Furthermore, this assay was unable to reveal whether Vps75 is associated with histones in chromatin or with free histones in the cytoplasm or nucleoplasm. To try and gain a more complete understanding of the putative interaction between Vps75 and chromatin, we carried out ChIP experiments. Vps75 was detectable at the pro-  Vps75 binds H3-H4 tetramers and assembles nucleosomes in vitro. A, interaction of Vps75 with core histones immobilized on an agarose matrix. Agarose beads were used as a negative control (Empty). 50% of the input is shown, suggesting strong histone binding. B, GST-Vps75 pull-down assays of H2A-H2B dimers (lanes 1-3), H3-H4 tetramers ((H3-H4) 2 ; lanes 4 -6), or a mixture of the two (lanes 7-9). Purified histone dimers (35 pmol), tetramers (19 pmol), or both were incubated with 35 pmol of recombinant GST-Vps75 or 70 pmol of GST alone. Bound proteins were separated by 4 -12% SDS-PAGE and visualized by silver staining. 50% of each histone input is shown. Size markers are shown on the right. Yeast histone H2A and H3 are not resolved in these gels. Asterisks denote contaminants or degradation products. Note that histone H2A is stained more efficiently with silver than is H2B; however, the H2A-H2B dimer is stoichiometric as determined by gel filtration and Coomassie Blue staining (not shown). C, nucleosome assembly activity of Vps75 by a plasmid-supercoiling assay. Vps75-His (purified from yeast) and HeLa core histones were incubated together and mixed with plasmid DNA relaxed with topoisomerase I (lanes 3-9). The formation of nucleosomes, coupled to the introduction of negative supercoils in the plasmids, was examined by electrophoresis in a 1% agarose gel and staining with ethidium bromide. Negatively supercoiled plasmids and relaxed are shown as controls (lanes 1 and 2, respectively). moter and within the body of a repressed gene, GAL10, and an active housekeeping gene, YEF3. It was also present in the middle of two other constitutive genes tested (the highly expressed FBA1 and the lowly expressed MRS5 genes) and at an open reading frame-free region at the telomere of chromosome VI, TEL6R (Fig. 3B).
Vps75 Is Not Required for Histone 3 Lysine 56 Acetylation-A recent study demonstrated that Rtt109 is required for acetylation of histone 3 at lysine 56 (H3 Lys-56) in yeast (12). Given that Vps75 co-purifies with Rtt109 and that it binds histones in vitro, we analyzed H3 Lys-56 acetylation in a vps75⌬ strain. In contrast to strains lacking RTT109 or ASF1, H3 Lys-56 acetylation could still clearly be detected in vps75⌬ (Fig. 4A). In addition, the vps75⌬ strain showed no sensitivity to hydroxyurea, a drug that can be used as an indicator for defects in H3 Lys-56 acetylation (Fig. 4B). Together, these data suggest that Vps75 is not required for H3 Lys-56 acetylation, although we cannot rule out the possibility that it is involved in regulating this histone mark at a subset of regions in the genome.

DISCUSSION
In this report, the identification and characterization of a new NAP family member from yeast, Vps75, is described. Using a GST pull-down assay, we demonstrated that Vps75 has a preference for (H3-H4) 2 tetramers over (H2A-H2B) dimers. Chromatin assembly occurs by a two-step process in which a tet-ramer is deposited first followed by two dimers, and it is generally thought that different histone chaperones catalyze each step (4). Our experiments suggest that Vps75 belongs to the H3/H4 family of histone chaperones. This contrasts with many in vivo and in vitro studies that have generally described NAP family members as chaperones for H2A and H2B (for review, see Ref. 4), although a recent study reported that yNAP1 preferentially binds (H3-H4) 2 tetramers in vitro (19). We are currently conducting further experiments to elucidate the histone preference of Vps75 in vivo.
Using ChIP, we found Vps75 at all genomic regions tested. Our data suggest that the presence of Vps75 is not strictly correlated with transcription since we detected it at a repressed gene (GAL10) and at high levels at the telomere of chromosome VI. It is important to point out, however, that a role in transcription cannot be ruled out on the basis of these experiments. However, the fact that Vps75 is detected at telomeres is particularly interesting in light of the finding that vps75⌬ mutants display shortened telomeres (6). Thus, our data might suggest a role for Vps75 in assembly or maintenance of telomere chromatin. It is clear, however, that more work needs to be done to define a role for Vps75 in specific chromatin-related processes.
We and others have found that Vps75 co-purifies with Rtt109, a protein recently implicated in H3 Lys-56 acetylation, and thus processes such as DNA repair, replication, and transcription (12,20,21). Two lines of evidence from our work indicate that Vps75 is not required for deposition of this histone mark. First, a vps75⌬ strain did not exhibit the dramatic reduction in bulk 43 Lys-56 acetylation observed in rtt109⌬ and asf1⌬. Second, the vps75⌬ strain was also not sensitive to hydroxyurea, which is toxic to strains with defects in H3 Lys-56 acetylation. Preliminary ChIP experiments with vps75⌬ cells have also failed to uncover significant differences in the level of histone H3 Lys-56 acetylation in chromatin (data not shown). A, detection of histone H3-associated Vps75. After cross-linking cells with formaldehyde, chromatin was fragmented by sonication and immunoprecipitated with a histone H3 antibody (H3C) or no antibody (Empty). Co-precipitated Vps75 was detected by immunoblotting with anti-Myc antibody. 2% of the input chromatin is shown. B, detection of Vps75 at specific genomic regions by ChIP. A yeast strain producing C-terminally tagged Vps75 (Vps75-HTM) or the untagged control strain (W303) was grown in YPD, cross-linked, and subjected to ChIP using an antibody against the Myc epitope. Real-time PCR was used to detect co-precipitated DNA from across the GAL10 and YEF3 genes, from the middle (mid) of the FBA1 and MRS5 genes, and from an open reading frame-free region at the telomere of chromosome VI (TEL6R). Data are expressed as the immunoprecipitation efficiency of the tagged strain divided by the immunoprecipitation efficiency of the untagged strain (shown as 1) and represent the average of four independent experiments, with standard errors shown. prom, promoter. Although we cannot rule out the possibility that Vps75 is important for H3 Lys-56 acetylation at a certain subset of genes or genomic locations, it is equally plausible that the Vps75/ Rtt109 complex possesses a distinct function. One possibility is that Rtt109 modulates the interaction of Vps75 with histones, thereby serving to regulate the chromatin assembly/disassembly activity of Vps75. Rtt109 is a positively charged protein (theoretical pI 9.41) with some highly basic regions, which could conceivably compete with histones for binding to the acidic Vps75 protein. This is presently under investigation.
In conclusion, we describe here the identification of Vps75 as a new member of the NAP domain family. This information should prove useful in unraveling the complexities of histone metabolism and important associated processes, such as DNA replication and transcription.