The Human Papillomavirus Type 16 E5 Protein Alters Vacuolar H+-ATPase Function and Stability in Saccharomyces cerevisiae

Virology 280, 169–175 (2001) doi:10.1006/viro.2000.0783, available online at http://www.idealibrary.com on

The Human Papillomavirus Type 16 E5 Protein Alters Vacuolar H ⫹-ATPase Function and Stability in Saccharomyces cerevisiae M. W. Briggs,* ,1 J. L. Adam,* and D. J. McCance* ,† ,2 *Department of Microbiology and Immunology and †the Cancer Center, University of Rochester, Rochester, New York 14642 Received October 10, 2000; returned to author for revision November 27, 2000; accepted November 30, 2000 The human papillomavirus 16 (HPV-16) E5 oncoprotein is a small integral membrane protein that binds to the 16-kDa subunit of the vacuolar H ⫹-ATPase (v-ATPase). Conservation within the family of v-ATPases prompted us to look to Saccharomyces cerevisiae as a potential model organism for E5 study. The E5 open reading frame, driven by a galactoseinducible promoter, was integrated into the yeast genome, and the resulting strain demonstrated a nearly complete growth arrest at neutral pH, consistent with defects associated with yeast v-ATPase mutants. Furthermore, this strain demonstrated a severe reduction in pH-dependent and v-ATPase-dependent vacuolar localization of fluorescent markers. Overexpression of the yeast 16-kDa subunit homolog partially suppressed E5-associated growth defects. E5 expression was correlated with a disassociation of the integral (V o) and peripheral (V i) v-ATPase sub-complexes, as well as a dramatic reduction of the steady-state levels of one mature V o subunit and the concomitant accumulation of its major proteolytic fragment, with unchanged levels of two V i subunits. Similar analyses of selected E5 mutants in yeast demonstrated a correlation between E5 biology and v-ATPase disruption. Our observations suggest that wild-type HPV-16 E5 acts during the assembly of the v-ATPase to inhibit, either directly or indirectly, V o stability and complex formation. © 2001 Academic Press Key Words: Papillomavirus; HPV-16 E5 protein; Saccharomyces; vacuolar H ⫹-ATPase.

INTRODUCTION

and mucosa, thus a conventional in vitro system for the propagation of papillomavirus does not exist. A stratified squamous epithelial model system has been used to show that the HPV-16 and -18 genomes can immortalize and inhibit the differentiation of human keratinocytes (McCance et al., 1988; Pirisi et al., 1987). Keratinocyte immortalization can be accomplished solely by the combined activities of two HPV-16 early gene products, those of E6 and E7. However, it has been demonstrated that the E5 gene product can act in cis to enhance such immortalization ⱕ10-fold (Stoppler et al., 1996), perhaps by enhancing the cellular response to growth factor stimulation (Gu and Matlashewski, 1995). It appears, therefore, that these three oncogenes act in concert to maintain the host cell in a proliferative state and, in conjunction with as yet unknown host cell dysfunctions, can lead to the development of malignant disease. The HPV-16 E5 protein is a small (83 amino acid), highly hydrophobic protein, and as such, localizes to intracellular membranes (Bubb et al., 1988). E5 has been demonstrated to bind to the 16-kDa pore-forming subunit of the v-ATPase (Adam et al., 2000; Conrad et al., 1993) and to inhibit endosomal acidification in human keratinocytes (Straight et al., 1995). Previous results from our laboratory have shown that exogenous E5 expression in murine fibroblasts leads to a transformed phenotype that is synergized by EGF (Straight et al., 1993). Furthermore, the transient expression of E5 in human keratinocytes was shown to inhibit degradation of internalized EGFR

Papillomaviruses are small DNA viruses that produce proliferative lesions in a number of mammalian hosts. Human papillomavirus (HPV) infections cause a variety of benign proliferations including condylomata acuminata (genital warts), epithelial cysts, cervical intraepithelial neoplasias (CIN), papillomas, and other types of hyperkeratoses. Their involvement in the development of some major human cancers is a particular health concern. Certain high-risk HPV types (the best-studied examples being HPV-16 and -18) have been identified as being causally associated with ⱖ90% of cancers of the cervix as well as with ⱖ50% of other anogenital cancers (zur Hausen, 1996). All papillomaviruses contain a genome of ⬃8000 bp, containing six “early” region open reading frames (ORFs) and two “late” ORFs. The late ORFs encode the major and minor structural capsid proteins; the early ORFs encode proteins involved in viral DNA replication and viral and cellular transcription. Viral DNA replication, as well as translation of late viral proteins and viral particle assembly, is restricted to the differentiating layers of skin

1

Present Address: Department of Pathology, Harvard Medical School, Brigham & Women’s Hospital, 75 Francis Street, Boston, MA 02115. 2 To whom reprint requests should be addressed. Fax: (716) 4759576. E-mail: [email protected]. 169

0042-6822/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

170

BRIGGS, ADAM, AND MCCANCE

and enhance the recycling of EGFR to the surface, thereby amplifying growth factor signaling to the cell. These observations are consistent with a role for HPV-16 E5 in inhibiting the activity of the v-ATPase through its binding to the 16-kDa subunit. Moreover, such a role for HPV-16 E5 would parallel the current understanding of the bovine papillomavirus type 1 (BPV-1) E5 homolog (Goldstein and Schlegel, 1990; Martin et al., 1989; Petti et al., 1991; Schiller et al., 1986). The eukaryotic vacuole is an organelle central to a number of essential cellular processes, including: receptor:ligand internalization, turnover, and recycling (Galloway et al., 1988; Johnson et al., 1993; Schandel and Jenness, 1994); sorting of newly translated proteins from the endoplasmic reticulum (van Weert et al., 1995); and amino acid and ion homeostasis (Anraku et al., 1989; Harrison et al., 1994; Sze et al., 1992). The activity of the v-ATPase in acidifying the vacuolar lumen is necessary to achieve each of these functions efficiently. Therefore disruption of the v-ATPase by HPV-16 E5 could have dramatic effects on the cell in terms of growth, response to extracellular signals, and production and processing of newly formed proteins. This contention is exemplified by the description of a mutant form of the 16-kDa subunit of the ATPase that at once disrupts binding to E5 and also converts the mutant protein into an oncoprotein that induces anchorage-independent growth of NIH-3T3 cells (Andresson et al., 1995). This observation highlights the important role for the ATPase in regulating cell proliferation and implies that HPV-16 E5 manipulates the vATPase to benefit the propagation of the virus. The v-ATPase is a structurally and functionally conserved protein complex, demonstrating remarkable homology among eukaryotes from fungi to mammals (Kane, 1999; Nelson and Klionsky, 1996). Structurally, all v-ATPases contain distinct peripheral catalytic portions and hydrophobic membrane portions (Kane and Stevens, 1992), which can assemble together or independently of each other (Kane, 1992; Kane et al., 1999). The ability to target individual gene disruptions in the yeast system has allowed for the greatest advancement in our understanding of the ordered events during v-ATPase biosynthesis and assembly and the consequences of disrupted v-ATPase activity (Doherty and Kane, 1993; Yamashiro et al., 1990). This level of understanding has been hastened by the development of antibodies to various subunits of the ATPase in yeast. Therefore, the yeast system is ideally suited to dissect the structural and functional relationship between HPV-16 E5 and the evolutionarily conserved 16-kDa subunit of the v-ATPase. Using Saccharomyces as a register for E5 activity, we have shown that the expression of this viral protein in yeast leads to growth and metabolite mobilization defects consistent with alterations in the activity of the v-ATPase. Moreover, E5 expression disrupts the multisubunit v-ATPase, perhaps during assembly in the endoplasmic reticulum.

FIG. 1. Northern analysis of E5 mRNA induction. Total RNA (10 ␮g) prepared from indicated strains, either uninduced or partially induced with 0.1% glucose/2% galactose, was separated by formaldehyde agarose gel electrophoresis and processed for Northern blotting. Blots were probed with an antisense E5 riboprobe and visualized by PhosphorImaging. Full-length wild-type and mutant E5 mRNAs are indicated by the arrow.

RESULTS AND DISCUSSION Growth and physiology of E5-expressing yeast cells Mutations in genes encoding various yeast v-ATPase subunits have been described that lead to a number of common phenotypes, including an inability of the mutant cells to grow at neutral pH, while retaining viability in acidic media (Anraku et al., 1989). We therefore measured the growth rates of logarithmically growing (⬃5 ⫻ 10 6 cells/ml) IO and IE5 cells shifted from glucose to inducing glucose/galactose media buffered to pH 5 or 7.5. Recent data from our laboratory (Adam et al., 2000) revealed that E5 expression led to a defect in growth under all conditions examined. Northern analyses demonstrated the moderate accumulation of full-length E5 mRNA (indicated in Fig. 1), as well as a prematurely terminated species. The latter is due to the cleavage and polyadenylation of cryptic yeast mRNA 3⬘ end formation signals in the AT-rich E5 DNA sequence (see Materials and Methods). Long exposure of Northerns revealed slight leakiness of the GAL10 promoter (data not shown), so it was not surprising that E5 caused a growth defect even with its promoter in an “uninduced” state. Significantly, E5-expressing yeast exhibited the most severe growth defect compared to wild type in inducing media buffered to pH 7.5, causing a near complete growth arrest (Adam et al., 2000). This observation paralleled those described for v-ATPase mutants and suggested that E5 could be acting through this complex. The inability to grow at high pH suggested that E5 cells were unable to acidify their vacuoles and maintain homeostasis with the extracellular environment. We used the weakly basic fluorophore quinacrine to examine the vacuolar acidity of wild-type and E5-expressing yeast. When added to the growth media, quinacrine passively diffuses into acidic subcellular compartments such as the vacuole (Weisman et al., 1987) but does not accumulate in yeast cells carrying lesions in v-ATPase subunits (Anraku et al., 1989). Since IE5 cells demonstrated slow growth even under non-inducing conditions,

HPV-16 E5 EXPRESSION IN YEAST

FIG. 2. (A) Vacuolar labeling with quinacrine. The weakly basic fluorophore quinacrine was allowed to passively diffuse into acidic compartments of wild-type or E5-expressing cells (indicated on the left) and photographed with phase-contrast (left) or fluorescence (right) illumination at ⫻40 magnification. (B) Detection of phosphoribosylaminoamidazole (AIR). IO cells (left) or IE5 cells (right) were analyzed by overlapping phase-contrast and fluorescence microscopy at 100⫻ magnification. The origin of the cells photographed (from stationary liquid culture or from plates) is indicated on the left. These data have been summarized in Adam et al., 2000.

we analyzed quinacrine uptake in the two strains in non-buffered glucose media. As evidenced by Fig. 2A, while IO cells demonstrated prominent staining of their vacuoles with quinacrine, IE5 cells were unable to accumulate this marker. These results suggested that E5 is a potent inhibitor of vacuolar acidification and were consistent with an E5-associated defect in the activity of the v-ATPase. The parent yeast strain of the IO and IE5 strains carries a lesion in the ADE2 gene and produces a pink pigment in culture and on plates. The pigment is due to the active (dependent upon the v-ATPase) (Weisman et al., 1987) and stable accumulation of a fluorescent intermediate of the adenine biosynthetic pathway (phospho-

171

ribosylaminoimidazole, or “AIR”) in the vacuole. Furthermore, this fluorophore fails to accumulate in strains carrying defects in selected v-ATPase subunits (Foury, 1990). We grew strains IO and IE5 in glucose media, and when both cultures reached stationary phase, we analyzed them by overlapping phase contrast and fluorescent microscopy, to highlight the vacuoles (Fig. 2B, top). IO cells accumulated the ade2 intermediate to reveal the prominent vacuole, whereas in IE5 cells, vacuoles were unlabeled. After an extended period of storage of strain IE5 on plates at 4°C (ca. 1 month), the normally white colonies began to turn pink, indicating that E5 expression does not simply interrupt the adenine biosynthetic pathway. We examined cells from such plates by microscopy (Fig. 2B, bottom) and found that IE5 cells were able to accumulate the AIR intermediate but not in the vacuolar compartment, resulting in diffuse staining throughout the cell. Since E5 did not inhibit the biosynthesis nor accumulation of AIR, but merely its transport into the vacuole, this observation further implied that E5 has a negative impact on the activity of the v-ATPase. A particular advantage to using yeast as a model organism is the well-characterized system developed for genetic analysis. We attempted to confirm the above-described phenotypic association of E5 with the v-ATPase by overexpressing the VMA3 gene, which encodes the 17-kDa subunit (yeast homolog to the mammalian 16 kDa). VMA3, controlled by its endogenous promoter, was engineered onto a high copy plasmid and transformed into the IO and IE5 strains, and growth was measured with cells grown in synthetic media for plasmid maintenance. We observed partial suppression of the E5-associated growth defects (Adam et al., 2000). In non-inducing or inducing acidic media, the presence of increased 17-kDa levels did not affect IE5 cell growth. However, at high pH, when IE5 cells were severely inhibited for growth, overexpression of the 17 kDa protein restored growth to near wild-type levels. These observations provided support for our belief that the v-ATPase 17-kDa subunit is being affected by expression of E5. E5 alters the ultrastructure of the yeast v-ATPase To make a more convincing argument for a relationship between E5 expression and the v-ATPase, we analyzed yeast v-ATPase structure by immunoprecipitation (IP) of the complex. The ATPase antibodies used were specific for either a peripheral ATPase subunit of 69 kDa or an integral ATPase subunit of 100 kDa. Following mild lysis of yeast spheroplasts in the presence of crosslinker, the antibodies to the peripheral subunit reproducibly immunoprecipitated v-ATPase subunits from the entire complex (Fig. 3, ␣-69 kDa panel) as demonstrated previously (Doherty and Kane, 1993). Also as previously shown (Kane et al., 1992), the 100-kDa antibody precipitated only the integral membrane portion of

172

BRIGGS, ADAM, AND MCCANCE

Mutant E5 phenotypes correlate with v-ATPase disruption

FIG. 3. Coimmunoprecipitations of whole-cell extracts from 35S-Cysteine-labeled IO and IE5 cells. The extracts used were as follows: (A) IO glucose, (B) IO glucose/galactose, (C) IE5 glucose, (D) IE5 glucose/ galactose. Either 10% of each labeled whole-cell extract used in immunoprecipitations (IPs) or pellets from IPs using mouse IgG, ␣-vATPase 69-kDa or ␣-v-ATPase 100-kDa were loaded as indicated. Products were separated by SDS–PAGE on a 12% gel and the gel was dried and exposed to a phosphorimager cassette for analysis. Selected v-ATPase subunits are indicated at the right. The ⬃75-kDa protein is indicated by asterisk.

the v-ATPase that was not associated with the peripheral complex (Fig. 3, ␣-100 kDa panel). In E5-expressing cells, we observed that the 69- and 100-kDa antibodies precipitated reduced levels of indicated v-ATPase subunits, and most noticeably when E5 was induced (Fig. 3, lanes c and d). These reductions were consistent with previously described effects of yeast v-ATPase subunit mutations that interfered with the assembly of the complex (Ho et al., 1993; Kane, 1992). Interestingly, IPs also revealed a protein band of ⬃75 kDa, again most prominently when E5 was induced. A protein band of this size has been shown to accumulate in various v-ATPase mutants or during prolonged vacuolar purification protocols and represents the major proteolytic fragment of the 100-kDa protein, Vph1p (Ho et al., 1993; Kane et al., 1992). We therefore performed Western analyses of crude vacuolar extracts to confirm this identity (Fig. 4). While IO extracts revealed predominantly full-length 100-kDa Vph1p, IE5 extracts from cells grown in inducing media showed a dramatic reduction of fulllength 100-kDa protein, with almost all of the Vph1p accumulating as the 75-kDa proteolytic fragment. Furthermore, analyses using anti-V i antibodies to the 60-kDa subunit demonstrated that E5 expression did not dramatically reduce the steady-state levels of this subunit, just the amount precipitated as a complex (cf. Fig. 3). Taken together, these E5-associated phenotypes are consistent with v-ATPase mutants that disrupt the integral portion, and therefore the assembly, of the v-ATPase complex.

In a recent report (Adam et al., 2000), we described the construction of a number of E5 mutant proteins analyzed for 16-kDa protein binding in COS-1 cells and growth defects in yeast. Our work revealed that while all point mutants bound the pore subunit, certain mutations abrogated the ability of E5 to affect yeast growth (cf. Adam et al., 2000), despite similar levels of expression (cf. Fig. 1). We therefore analyzed a representative mutant of each type in our yeast system by IP of metabolically labeled cells grown in inducing media (Fig. 5). As evidenced in Fig. 5B, we found that the M5 mutation, which completely abolished E5-associated growth defects in yeast (Adam et al., 2000), did not cause reduced ATPase subunit precipitation with the 100-kDa antibody nor the accumulation of the 75-kDa Vph1p fragment. Conversely, the M2 mutation, which had no effect on E5 activity, appears identical in IP pattern to wild-type E5. These observations imply that there is a direct relationship between E5 growth defects and v-ATPase destabilization. We have demonstrated that in Saccharomyces cerevisiae, the expression of HPV-16 E5 leads to a severe growth defect at neutral pH, an inhibition of the acidification of the vacuolar lumen, abrogation of v-ATPasedependent compartmentalization of a fluorescent marker, and destabilization of the v-ATPase complex. These results are consistent with our model that E5 inhibits v-ATPase function by altering the structure of the v-ATPase through interactions with the 17-kDa poreforming subunit. This effect on v-ATPase structure and/or assembly by E5 is novel and provides important insights into the action of the E5 protein. The v-ATPase is one of

FIG. 4. Western blot analyses of v-ATPase subunits from crude v-ATPase extracts from IO and IE5 grown in inducing media. Unlabeled v-ATPase extracts (1 ␮g) from IO- and IE5-induced cells were separated by SDS–PAGE on a 10% gel and transferred to nitrocellulose for sequential Western blot analyses using antibodies directed to the 100and 60-kDa v-ATPase subunits. Respective subunits, and the ⬃75-kDa proteolytic fragment of Vph1p, are indicated on the left.

HPV-16 E5 EXPRESSION IN YEAST

FIG. 5. Coimmunoprecipitations of whole-cell extracts from 35S-Cysteine-labeled IO, IE5, and E5 mutant cells grown in inducing media. Origins of mutant strains used are discussed under Materials and Methods. (A) 10% of each labeled whole-cell extract used in immunoprecipitations (IPs) or (B) pellets from IPs using ␣-v-ATPase 100 kDa. Products were separated by SDS–PAGE on a 12% gel, and the gel was dried and exposed to a phosphorimager cassette for analysis. Selected v-ATPase subunits are indicated. The ⬃75-kDa Vph1p proteolytic fragment is indicated by asterisk.

173

complex and may further underscore the importance of E5 to the biology and pathology of HPV-16, at least as evidenced by the yeast system. Further experimentation with our yeast system will be required to clarify this issue. In another report (Adam et al., 2000), our yeast system has allowed us to determine the potential biological relevance of E5::v-ATPase binding to vacuolar acidification. Those studies indicated that 16-kDa binding is not sufficient for the biological effect of E5, and our data in this report complement these findings by implying that a primary activity of E5 is to disrupt the stability and assembly of the v-ATPase. Taken together, our observations indicate that E5 does not simply act by blocking the pore of the v-ATPase but instead leads to the destabilization of the complex. We have shown that in human keratinocytes E5 expression inhibits endosomal acidification (Straight et al., 1995) resulting in reduced epidermal growth factor receptor degradation and a proliferative response. In yeast, the fact that disruption of the acidification of the vacuole, which is central to yeast growth and metabolism, leads to growth inhibition reflects the specialization of the endosome/lysosome organelle in mammalian cells. However, we believe that yeast system will allow the study of E5/v-ATPase interactions and help to determine functions that can then be tested in the less tractable human keratinocyte. MATERIALS AND METHODS

a large family of energizing cellular complexes that are evolutionarily ancient. The 16-kDa pore-forming subunit, an important target of E5, is 67% identical between mammals and yeast (17-kDa homolog) and 100% identical in the fourth transmembrane helix shown to be essential for BPV-1 E5 (Andresson et al., 1995) and HPV-16 E5 binding (Adam et al., 2000). Therefore, our observations may lead to a greater understanding of the functional significance of the relationship between E5 and the mammalian vATPase 16-kDa subunit. A comparison of the effects of E5 expression on yeast growth to those of deletions of various v-ATPase subunits demonstrates that E5 has associated defects that are more severe than simple disruption of the complex (Anraku et al., 1989; Kane, 1992). While vma deletion mutants (deletion of the 17-kDa yeast subunit) show similarities to E5-expressing cells in their slow growth and high-pH lethality, E5 expression led to arrest (under all conditions) so severe that only partial induction of the protein allowed us to perform the experiments described in this manuscript. This observation could reflect a dominant-negative relationship between E5 and the 17-kDa protein, such that the presence of E5 in the cell leads to an altered function of the pore-forming subunit. Conversely, it could indicate that E5 affects cellular activities beyond its interaction with the non-essential v-ATPase

Strains, plasmids, and culture conditions Yeast strain SC725 (MAT ␣ , leu2-2,112, ade2-101, trp1 ⌬1, his3 ⌬200, lys2-801, ura3-52) (a gift of Dr. Eric Phizicky) was the parent used for the experiments described in this paper. SC725H was derived from SC725 by integration of a BamHI fragment from plasmid pJJ215 (Jones and Prakash, 1990) containing the entire HIS3 locus to repair the chromosomal lesion. Plasmid YEp51 was a gift from Dr. Patricia Kane (Broach et al., 1983). The E5 open reading frame was cloned into this Sal1/Bcl1-cut plasmid by PCR using oligonucleotides oE5Sal5⬘ (5⬘-ATTTGTCGAC ATATGACAAA TCTTGATACT GC-3⬘) and oDE5-329.2 (5⬘-GCTGATCATT ATGTAATCAG AAAGCGTGCA TGTGTATGTA TCAAGAACAA TGGTATGTAA ACAAATATGA TG3⬘) to produce plasmid YGDE5. The extended length of oDE5-329.2 was necessary for the incorporation of single-base changes within the E5 open reading frame (indicated in bold). The AT-rich 3⬘ portion of the wild-type E5 open reading frame signaled premature mRNA 3⬘ end formation in yeast, so point mutations were generated that did not alter E5 amino acid sequence, but yielded detectable full-length E5 mRNA (cf. Fig. 2). YEp51 and YGDE5 were subjected to PCR using oligonucleotides oLEU2int (5⬘-GAACGGATCT CCAGATCATC G-3⬘) and oFLP3⬘ (5⬘-GAGCGCTTCC GAAAATGCAA CGCG-3⬘) and

174

BRIGGS, ADAM, AND MCCANCE

Pfu polymerase to generate blunt-ended products. PCR products were cloned into the HIS3 locus found within plasmid pJJ217 [a gift of Dr. Eric Phizicky (Jones and Prakash, 1990)], which was cut with NdeI, filled in with Klenow and cut again with Eco47III. Sequence containing the his3 flanking regions, the selectable LEU2 marker, and the GAL10 promoter and FLP1 terminator either containing or lacking the E5 ORF was excised by digestion with PvuII. One microgram of these digests was used to transform strain SC725H to produce strains SC725H-his3::YEp51 (“IO”) and SC725H-his3::YGDE5 (“IE5”). Integrations were confirmed by Southern blotting, and inducible expression of full-length E5 mRNA was confirmed by Northern blotting. Construction of the E5 M2 (C18G, C20G) and M5 (Y39G) mutations is described in (Adam et al., 2000). Complementation of the ade2 defect in working strains was achieved by transformation with the centromeric plasmid YCpADE2 (a gift of Dr. Eric Phizicky from E. coli strain EMP987). Plasmid YEpVMA3 was constructed by excision of the VMA3 ORF from plasmid pMK9 (Kane et al., 1992) with EcoRI and PstI and insertion into similarly cut YEplac195 (Gietz and Sugino, 1988). Strains were cultured in either standard YPD or Synthetic Complete media lacking indicated amino acids (for plasmid maintenance) (Sherman et al., 1979). For analyses of growth in inducing and non-inducing media, yeast cells were grown to stationary phase in rich media (YPD) and then appropriate numbers of cells were diluted to early logarithmic phase into buffered rich or synthetic media containing either 2% dextrose or 0.1% dextrose/2% galactose [as full induction of E5 led to growth arrest (Guarente et al., 1982)]. Growth was monitored by optical density measurements at 600 nm. It is noteworthy that full induction of E5 expression (in 2% galactose) led to immediate growth arrest and was therefore abandoned to allow the phenotypic analyses described below. Quinacrine labeling has been described (Kane et al., 1989). Northern analysis Total RNA was prepared and Northern analysis carried out as described previously (Patel, 1992), with blots probed with an antisense riboprobe specific for E5 derived from plasmid pAT16. Levels of E5 mRNA were quantitated by storage PhosphorImager analysis (Molecular Dynamics) and normalized to rRNA levels, which were quantitated by FluorImager analysis (Molecular Dynamics) of ethidium bromide-stained gels. Microscopy Yeast cells were analyzed by digital microscopy (Olympus BX60 microscope and UV lamp, Optronics Engineering model DEI-750 digital camera) at either ⫻40 or ⫻100 magnification under phase-contrast or fluorescence

(535-nm excitation, fluorescein filter), as indicated in appropriate figures. Digital images were analyzed using Adobe Photoshop 5.0 on an Apple Power Macintosh 8100/80AV computer. Protein analyses Monoclonal antibodies to three of the v-ATPase subunits [gifts of Dr. Patricia Kane (Doherty and Kane, 1993)] were utilized for immunoprecipitations and Westerns as described below. Cell labeling and isolation of crude v-ATPase extracts were isolated as previously described (Kane et al., 1989). Briefly, 10 ODs of mid-log phase cells (OD 600 ⫽ 0.4–0.8) were pelleted, washed with 0.1 M Tris-SO 4, pH 9.4, 10 mM DTT, and converted to spheroplasts by zymolyase 100T (Sigma) treatment (200 ␮g zymolyase in SD-met,cys, 1.2 M sorbitol, 20 min at 30°C). Cells were washed twice with SD-met,cys, 1.2 M sorbitol, and 0.5 OD units were resuspended in SD-met,cys, 1.2 M sorbitol at 150 ␮l per immunoprecipitation, and allowed to recover at 30°C for 20 min. Cells were labeled with 50 ␮Ci of [ 35S]cysteine (NEN) for one-half of one cell doubling. Labeled spheroplasts were incubated on ice for 2 min, then pelleted and resuspended in 100 ␮l solubilization buffer per immunoprecipitation (PBS, 1% C 12E 9 (polyoxyethylene-9 lauryl ether), Sigma] plus protease inhibitors and 0.75 mM of the crosslinker DSP (dithiobis(succinimidyl-propionate) for 1 h on ice. An equal volume of 50 mM Tris–HCl, pH 7.5, 1 mM EDTA, 10% glycerol was added to quench the crosslinker for 20 min. Lysate volumes (200 ␮l) were raised to 500 ␮l with solubilization buffer (lacking C 12E 9). One microgram of mouse IgG, 5 ␮l 8B1 monoclonal antibody (recognizing the 69-kDa subunit of the v-ATPase), or 250 ␮l 10D7 monoclonal antibody (recognizing the 100-kDa subunit of the v-ATPase) was then added. Volumes were raised to 1 ml with PBS plus 20 ␮l 50% protein A/G Plus–Sepharose (Santa Cruz), and samples were rotated at 4°C overnight. Immune complexes were washed two times with 10 mM Tris–HCl pH 8.0, 1% Triton X-100, 20 mM EDTA. Samples were separated on 12% SDS polyacrylamide gels, and the gels were dried and exposed to a phosphorimager screen for analysis. For Western analyses, 1 ␮g of unlabeled crude v-ATPase extracts from IO and IE5 strains were separated by SDS–PAGE on a 10% gel and transferred to nitrocellulose. After blocking in PBS/0.1% Tween-20/5% milk, primary antibodies 8B1 (69 kDa), 13D11 (60 kDa), and 10D7 (100 kDa) were added at (1:1000, 1:2000, and 1:500, respectively) for 1 h. After four PBST washes, secondary antibodies (goat anti mouseHRP, 1:2000, Santa Cruz) were added for 1 h. The blot was washed as before, ECL reagent was added (NEN), and the blot was exposed to film for visualization. ACKNOWLEDGMENTS We thank the laboratories of Eric Phizicky and Patricia Kane for important biological reagents used in this study. This work was sup-

HPV-16 E5 EXPRESSION IN YEAST ported by NIAID Grant AI-30798 and American Cancer Society Grant VM-144, and M.W.B. was supported by a postdoctoral training grant from the University of Rochester Cancer Center, NCI CA-09363D.

REFERENCES Adam, J. L., Briggs, M. W., and McCance, D. J. (2000). A mutagenic analysis of the E5 protein of Human Papillomavirus Type 16 reveals that E5 binding to the vacuolar H⫹-ATPase is not sufficient for biological activity, using mammalian and yeast expression systems. Virology 272, 315–325. Andresson, T., Sparkowski, J., Goldstein, D. J., and Schlegel, R. (1995). Vacuolar H(⫹)-ATPase mutants transform cells and define a binding site for the papillomavirus E5 oncoprotein. J. Biol. Chem. 270, 6830– 6837. Anraku, Y., Umemoto, N., Hirata, R., and Wada, Y. (1989). Structure and function of the yeast vacuolar membrane proton ATPase. J. Bioenerg. Biomembr. 21, 589–603. Broach, J. R., Li, Y.-Y., Wu, L.-C. C., and Jayaram, M. (1983). “Vectors for High-Level, Inducible Expression of Cloned Genes in Yeast,” pp. 83–117. Academic Press, New York. Bubb, V., McCance, D. J., and Schlegel, R. (1988). DNA sequence of the HPV-16 E5 ORF and the structural conservation of its encoded protein. Virology 163, 243–246. Conrad, M., Bubb, V. J., and Schlegel, R. (1993). The human papillomavirus type 6 and 16 E5 proteins are membrane-associated proteins which associate with the 16-kilodalton pore-forming protein. J. Virol. 67, 6170–6178. Doherty, R. D., and Kane, P. M. (1993). Partial assembly of the yeast vacuolar H(⫹)-ATPase in mutants lacking one subunit of the enzyme. J. Biol. Chem. 268, 16845–16851. Foury, F. (1990). The 31-kDa polypeptide is an essential subunit of the vacuolar ATPase in Saccharomyces cerevisiae. J. Biol. Chem. 265, 18554–18560. Galloway, C. J., Dean, G. E., Fuchs, R., and Mellman, I. (1988). Analysis of endosome and lysosome acidification in vitro. Methods Enzymol. 157, 601–611. Gietz, R. D., and Sugino, A. (1988). New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74, 527–534. Goldstein, D. J., and Schlegel, R. (1990). The E5 oncoprotein of bovine papillomavirus binds to a 16 kd cellular protein. EMBO J. 9, 137–145. Gu, Z., and Matlashewski, G. (1995). Effect of human papillomavirus type 16 oncogenes on MAP kinase activity. J. Virol. 69, 8051–8056. Guarente, L., Yocum, R. R., and Gifford, P. (1982). A GAL10-CYC1 hybrid yeast promoter identifies the GAL4 regulatory region as an upstream site. Proc. Natl. Acad. Sci. USA 79, 7410–7414. Harrison, M. A., Jones, P. C., Kim, Y. I., Finbow, M. E., and Findlay, J. B. (1994). Functional properties of a hybrid vacuolar H(⫹)-ATPase in Saccharomyces cells expressing the Nephrops 16-kDa proteolipid. Eur. J. Biochem. 221, 111–120. Ho, M. N., Hill, K. J., Lindorfer, M. A., and Stevens, T. H. (1993). Isolation of vacuolar membrane H(⫹)-ATPase-deficient yeast mutants; the VMA5 and VMA4 genes are essential for assembly and activity of the vacuolar H(⫹)-ATPase. J. Biol. Chem. 268, 221–227. Johnson, L. S., Dunn, K. W., Pytowski, B., and McGraw, T. E. (1993). Endosome acidification and receptor trafficking: Bafilomycin A1 slows receptor externalization by a mechanism involving the receptor’s internalization motif. Mol. Biol. Cell 4, 1251–1266. Jones, J. S., and Prakash, L. (1990). Yeast Saccharomyces cerevisiae selectable markers in pUC18 polylinkers. Yeast 6, 363–366. Kane, P. M. (1992). Biogenesis of the yeast vacuolar H(⫹)-ATPase. J. Exp. Biol. 172, 93–103. Kane, P. M. (1999). Biosynthesis and regulation of the yeast vacuolar H⫹-ATPase. J. Bioenerg. Biomembr. 31, 49–56.

175

Kane, P. M., Kuehn, M. C., Howald-Stevenson, I., and Stevens, T. H. (1992). Assembly and targeting of peripheral and integral membrane subunits of the yeast vacuolar H(⫹)-ATPase. J. Biol. Chem. 267, 447–454. Kane, P. M., and Stevens, T. H. (1992). Subunit composition, biosynthesis, and assembly of the yeast vacuolar proton-translocating ATPase. J. Bioenerg. Biomembr. 24, 383–393. Kane, P. M., Tarsio, M., and Liu, J. (1999). Early steps in assembly of the yeast vacuolar H⫹-ATPase. J. Biol. Chem. 274, 17275–17283. Kane, P. M., Yamashiro, C. T., and Stevens, T. H. (1989). Biochemical characterization of the yeast vacuolar H(⫹)-ATPase. J. Biol. Chem. 264, 19236–19244. Martin, P., Vass, W. C., Schiller, J. T., Lowy, D. R., and Velu, T. J. (1989). The bovine papillomavirus E5 transforming protein can stimulate the transforming activity of EGF and CSF-1 receptors. Cell 59, 21–32. McCance, D. J., Kopan, R., Fuchs, E., and Laimins, L. A. (1988). Human papillomavirus type 16 alters human epithelial cell differentiation in vitro. Proc. Natl. Acad. Sci. USA 85, 7169–7173. Nelson, N., and Klionsky, D. J. (1996). Vacuolar H(⫹)-ATPase: from mammals to yeast and back. Experientia 52, 1101–1110. Patel, D., and Butler, J. S. (1992). Conditional defect in mRNA 3⬘ end processing caused by a mutation in the gene for poly(A) polymerase. Mol. Cell. Biol. 12, 3297–3304. Petti, L., Nilson, L. A., and DiMaio, D. (1991). Activation of the plateletderived growth factor receptor by the bovine papillomavirus E5 transforming protein. EMBO J. 10, 845–855. Pirisi, L., Yasumoto, S., Feller, M., Doniger, J., and DiPaolo, J. A. (1987). Transformation of human fibroblasts and keratinocytes with human papillomavirus type 16 DNA. J. Virol. 61, 1061–1066. Schandel, K. A., and Jenness, D. D. (1994). Direct evidence for ligandinduced internalization of the yeast alpha-factor pheromone receptor. Mol. Cell. Biol. 14, 7245–7255. Schiller, J. T., Vass, W. C., Vousden, K. H., and Lowy, D. R. (1986). E5 open reading frame of bovine papillomavirus type 1 encodes a transforming gene. J. Virol. 57, 1–6. Sherman, F., Fink, G. R., and Lawrence, C. W. (1979). “Methods in Yeast Genetics.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Stoppler, M. C., Straight, S. W., Tsao, G., Schlegel, R., and McCance, D. J. (1996). The E5 gene of HPV-16 enhances keratinocyte immortalization by full-length DNA. Virology 223, 251–254. Straight, S. W., Herman, B., and McCance, D. J. (1995). The E5 oncoprotein of human papillomavirus type 16 inhibits the acidification of endosomes in human keratinocytes. J. Virol. 69, 3185–3192. Straight, S. W., Hinkle, P. M., Jewers, R. J., and McCance, D. J. (1993). The E5 oncoprotein of human papillomavirus type 16 transforms fibroblasts and effects the downregulation of the epidermal growth factor receptor in keratinocytes. J. Virol. 67, 4521–4532. Sze, H., Ward, J. M., and Lai, S. (1992). Vacuolar H(⫹)-translocating ATPases from plants: Structure, function, and isoforms. J. Bioenerg. Biomembr. 24, 371–381. van Weert, A. W., Dunn, K. W., Gueze, H. J., Maxfield, F. R., and Stoorvogel, W. (1995). Transport from late endosomes to lysosomes, but not sorting of integral membrane proteins in endosomes, depends on the vacuolar proton pump. J. Cell Biol. 130, 821–834. Weisman, L. S., Bacallao, R., and Wickner, W. (1987). Multiple methods of visualizing the yeast vacuole permit evaluation of its morphology and inheritance during the cell cycle. J. Cell Biol. 105, 1539–1547. Yamashiro, C. T., Kane, P. M., Wolczyk, D. F., Preston, R. A., and Stevens, T. H. (1990). Role of vacuolar acidification in protein sorting and zymogen activation: A genetic analysis of the yeast vacuolar protontranslocating ATPase. Mol. Cell. Biol. 10, 3737–3749. zur Hausen, H. (1996). Papillomavirus infections—a major cause of human cancers. Biochim. Biophys. Acta 1288, F55–F78.