HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 279, Issue 23, 24323-24333, June 4, 2004

This Article Abstract Full Text (PDF) All Versions of this Article: 279/23/24323    most recent M401981200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Van Dyke, M. W. Articles by Mehta, D. V. Search for Related Content PubMed PubMed Citation Articles by Van Dyke, M. W. Articles by Mehta, D. V. Social Bookmarking                      What's this?

Stm1p, a G₄ Quadruplex and Purine Motif Triplex Nucleic Acid-binding Protein, Interacts with Ribosomes and Subtelomeric Y' DNA in Saccharomyces cerevisiae^*

Michael W. Van Dyke, Laura D. Nelson, Rodney G. Weilbaecher, and Dakshesh V. Mehta

From the Department of Molecular and Cellular Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030

Received for publication, February 23, 2004

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
The Saccharomyces cerevisiae protein Stm1 was originally identified as a G4 quadruplex and purine motif triplex nucleic acid-binding protein. However, more recent studies have suggested a role for Stm1p in processes ranging from antiapoptosis to telomere maintenance. To better understand the biological role of Stm1p and its potential for G^*G multiplex binding, we used epitope-tagged protein and immunological methods to identify the subcellular localization and protein and nucleic acid partners of Stm1p in vivo. Indirect immunofluorescence microscopy indicated that Stm1p is primarily a cytoplasmic protein, although a small percentage is also present in the nucleus. Conventional immunoprecipitation found that Stm1p is associated with ribosomal proteins and rRNA. This association was verified by rate zonal separation through sucrose gradients, which showed that Stm1p binds exclusively to mature 80 S ribosomes and polysomes. Chromatin immunoprecipitation experiments found that Stm1p preferentially binds telomere-proximal Y' element DNA sequences. Taken together, our data suggest that Stm1p is primarily a ribosome-associated protein, but one that can also interact with DNA, especially subtelomeric sequences. We discuss the implications of our findings in relation to prior genetic, genomic, and proteomic studies that have identified STM1 and/or Stm1p as well as the possible biological role of Stm1p.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
It has long been recognized that certain G-rich nucleic acids can adopt structures other than Watson-Crick base-paired duplexes (1). Examples include purine motif triple helical DNA (Pu triplex)¹ and four-stranded quadruplexes (G4 DNA or RNA) (2, 3). These nucleic acids are characterized by Hoogsteen hydrogen bonding between guanines (G^*G) and very high negative charge densities due to the presence of multiple phosphate backbones, hence the simpler designation "G^*G multiplex." Both Pu triplexes and G4 structures can form under physiological conditions and are both extremely stable species once formed.

Although a considerable amount of information is available regarding the properties of G^*G multiplex structures in vitro, little is known regarding their existence and biological roles in vivo (4, 5). Sequences capable of forming these structures are present in all eukaryotic organisms. Examples include telomeric sequences, long oligopurine tracts within several gene promoters, and certain neuronal mRNAs (6–8). G^*G multiplexes have been hypothesized as functional intermediates in many biological processes, including chromosome condensation, recombination, telomere maintenance, and transcriptional and translational control (8–14). By contrast, potentially deleterious G^*G structures can also form in DNA or RNA during replication or transcription, with their removal being critical for normal cellular function (15, 16).

One line of evidence used to support the hypothesis that G^*G multiplexes do exist in vivo is the existence of cellular proteins that specifically act upon such structures. Proteins have been identified in eukaryotes ranging from yeast to humans that specifically and avidly bind Pu triplexes and G4 DNAs and RNAs in vitro (8, 10, 17–28, 30–44). In addition, helicases have also been identified that efficiently unwind these G^*G multiplex structures in vitro (16, 44–47).²

Obviously, biochemical demonstrations of protein-G^*G multiplex interactions in vitro do not necessarily prove that such phenomena actually occur in vivo, let alone that such species truly exist. To address this important question, we sought to identify those nucleic acids bound by G^*G multiplex-binding proteins within living cells. As a model system, we chose the Saccharomyces cerevisiae protein Stm1p, which had previously been shown to be an avid and specific binder of G4 and Pu triplex DNAs (22, 32). Stm1p is a moderately abundant (35,000 copies/cell), 30-kDa basic (pI = 10.5) protein that has been implicated in biological processes ranging from apoptosis to telomere maintenance (48, 49). We constructed epitope-tagged STM1 strains to determine the cellular localization and in vivo binding partners of this protein. Here we report that Stm1p is primarily a ribosome-associated protein, although it can also be found specifically associated with subtelomeric Y' elements in yeast chromosomes.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Yeast Strains—Standard yeast genetic techniques were used (50). PCR-based homologous recombination (51) was used to construct strains in which the chromosomal STM1 gene is tagged with three tandem HA epitopes at either the N or C terminus. PCR screening and Western blot analysis of protein expression was used to verify strain construction. Strain K699 (MATa ade2-1 can1–100 his3–11,-15 leu2-3,-112 ssd1-d trp1-1 ura3) is the isogenic parent of strain STM1-HA₃ (DM28). STM1-HA₃ (DM28) yeast express Stm1p with a C-terminal HA₃ epitope (Stm1p-HA₃) under the control of the endogenous STM1 promoter. The DNA used to construct strain STM1-HA₃ (DM28) was amplified from plasmid pFA6a-3HA-kanMX6 with primers HAC-F2 (5'-GAA CCG TAA CAT TGA CGT TTC TAA CTT GCC ATC TTT GGC TCG GAT CCC CGG GTT AAT TAA-3') and HAC-R1 (5'-TTA TTG GAT TCT TTC AGT TGG AAT TAT TCA TAT ATA AGG CGA ATT CGA GCT CGT TTA AAC-3'). Strain JB811 (MATa leu2-3,112 trp1 ura3–52 prb1 prc pep4-3) is the isogenic parent of strain HA₃-STM1 (LN12). HA₃-STM1 (LN12) yeast express Stm1p with a N-terminal HA₃ epitope (HA₃-Stm1p) under the control of the GAL1 promoter. The DNA used to construct strain HA₃-STM1 (LN12) was amplified from plasmid pFA6a-kanMX6-PGAL1–3HA with primers HAN-F4 (5'-TTT CTT TGC AAA TTT CTC TTC CCC CCA CAG TAT TCT TTT AGA ATT CGA GCT CGT TTA AAC-3') and HAN-R3 (5'-CGT CTT CGA CGT CGT TAC CTA ACA AAT CAA ATG GGT TGG AGC ACT GAG CAG CGT AAT CTG-3'. The stm1- strain A1454 (MATa aro7 ade8 his3 leu2 ura3 stm1::ADE8) has been described previously (52). The YKU80-myc₁₈ strain (TVL323) expresses a C-terminal myc₁₈ epitope-tagged Yku80p from the endogenous YKU80 promoter. This strain and its untagged isogenic parent used in the chromatin immunoprecipitation experiment are haploids derived from AVL49 (MAT a/ ura3–52/ura3–52 lys2–801/lys2–801 ade2–101/ade2–101 trp1-1/trp1- 1 his3- 200/his3- 200 leu2- 1/leu2- 1 CF+/- [TRP1-SUP11-CEN4 D8B]). The YKU80-myc18 strain (TVL323) was constructed and provided by A. Bertuch and V. Lundblad (Baylor College of Medicine, Houston, TX).

Gel Electrophoretic Analysis of Stm1p—Yeast cells were grown to midlogarithmic phase (A₆₀₀ = 1.0) in YPD (1% yeast extract, 2% peptone, and 1% dextrose) medium. When required for HA₃-Stm1p expression, the final 6 h of growth was performed in YPA-galactose (1% yeast extract, 2% peptone, 0.003% adenine sulfate, and 0.2% galactose) medium. Cells were disrupted by vortexing with an equal volume of acid-washed 0.5-mm Glasperlen glass beads (B. Braun Biotech, Allentown, PA) in cold LY buffer (50 mM Tris-HCl, pH 7.5, 150 mM KCl, 6 mM MgCl₂, 2 mM EDTA, 10% glycerol, 0.1% Triton X-100, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 µM leupeptin, 1 µM pepstatin, and 2.5 µg/ml antipain). Cell lysates were clarified by centrifugation at 12,000 x g for 10 min at 4 °C, and the protein concentration was determined using a Bradford assay (Bio-Rad).

To analyze Stm1p using Western blotting, 15 µg of total protein from the clarified whole cell extract was heat-denatured in Laemmli sample buffer and loaded onto an SDS-10% polyacrylamide gel (53). After electrophoretic separation, the proteins were electroblotted onto nitrocellulose membranes (Schleicher & Schuell). Blocked membranes were probed with either the anti-HA.11 epitope-specific mouse mAb 16B12 (Covance Research Products, Berkeley, CA) or a rabbit polyclonal antibody (University of Texas M.D. Anderson Cancer Center Department of Veterinary Medicine and Surgery, Bastrop, TX) generated against bacterial recombinant Stm1p,³ followed by a sheep anti-mouse or sheep anti-rabbit IgG horseradish peroxidase-conjugated secondary antibody (Amersham Biosciences), depending on the primary antibody used. Antibodies were visualized using an Amersham ECL chemiluminescence kit and Hyperfilm following the manufacturer's instructions. Western blots were analyzed by densitometry using an AlphaImager 3400 imaging system and AlphaEase FC software (Alpha Innotech, San Leandro, CA).

To analyze Stm1p binding to triplex DNA by EMSA, 10 µg of total protein from the clarified whole cell extract was incubated with 1 nM radiolabeled X-Pu-T19 probe for 30 min at 24 °C in a 10-µl volume containing 25 mM HEPES-Na⁺, pH 7.9, 50 mM KCl, 1 mM MgCl₂, 10% glycerol, 0.5 mM dithiothreitol, and 2 µg poly(dI-dC) carrier DNA as previously described (32). The resulting protein-probe complexes were resolved by nondenaturing PAGE, visualized by autoradiography, and quantitated using a Storm 840 PhosphoImager (Amersham Biosciences).

Immunofluorescence Microscopy—Immunofluorescence microscopy was performed essentially as previously described (48). Yeast cells grown to midlogarithmic phase (A₆₀₀ 0.5–1.0) were harvested and fixed in 50 mM potassium phosphate (pH 7.0) and 3.7% formaldehyde. Spheroplasts were made by treating these cells with 5 µg/ml Zymolyase 100T (Sigma) and 0.1% -mercaptoethanol in SP buffer (1.2 M sorbitol and 50 mM potassium phosphate, pH 7). Resulting spheroplasts were adhered onto poly-L-lysine-coated slides (Sigma). These were treated with HA epitope-specific mouse mAb 16B12 (1:200), followed by 5 µg/ml Cy3-conjugated affinity-purified goat anti-mouse IgG antibodies (Jackson Laboratory, Bar Harbor, ME) in phosphate-buffered saline and 0.1% bovine serum albumin. The bound cells were finally treated with the DNA-binding fluorescent dye 4',6-diamidino-2-phenylindole (1 ng/ml in phosphate-buffered saline) and one drop of ProLong antifade (Molecular Probes, Eugene, OR) before mounting and sealing. The cells were visualized by differential interference contrast microscopy or by fluorescence using a Zeiss Axioplan 2 microscope, and the images were processed with AxioVision version 2.05 software (Carl Zeiss Microimaging, Thornwood, NY).

Immunoprecipitation—Immunoprecipitations were performed essentially according to published protocols but with minor modification (54). Clarified yeast extract (1 mg of total protein), as described above, was incubated with 5 µg of anti-HA mAb 16B12 in a total volume of 0.5 ml of IPM buffer (50 mM Tris-HCl, pH 7.5, 150 mM KCl, 6 mM MgCl₂,2mM EDTA, 0.1% Triton X-100, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 µM leupeptin, 1 µM pepstatin, and 2.5 µg/ml antipain) for 3 h at 4 °C with continual inversion. Suspensions were clarified by centrifugation (12,000 x g for 10 min at 4 °C), and the supernatants were added to an equal volume of IPM buffer containing 50 µl of a protein G-agarose bead suspension (Roche Applied Science). These were incubated for 3 h at 4 °C with continual inversion to effect binding and isolated by centrifugation.

To investigate immunoprecipitated proteins, thoroughly washed beads from the above immunoprecipitation were resuspended in Laemmli sample buffer, boiled for 5 min to release bound proteins, and then clarified by centrifugation before being loaded onto SDS-10% polyacrylamide gels. After electrophoretic separation, the proteins were visualized by staining with Coomassie Brilliant Blue R250. Molecular mass analysis and protein quantitation were performed on the dried gel using an AlphaImager 3400 imaging system and AlphaEase FC software. Individual protein bands from these gels were excised and sent to the Protein Chemistry Core Facility at Columbia University for matrix-assisted laser desorption ionization mass spectrometry peptide mass mapping. Peptide masses were searched against the National Center for Biotechnology Information and/or GenBank^TM GenPept databases using either the ProFound or the MS-Fit program (55, 56).

To investigate immunoprecipitated nucleic acids, the above mentioned immunocomplexed beads were resuspended in 0.3 ml of E buffer (50 mM sodium acetate, pH 5.3, 10 mM EDTA, and 1% SDS) and incubated for 5 min at 65 °C to release the bound nucleic acids. These nucleic acids were purified by phenol extraction and ethanol precipitation, and the purified nucleic acids were electrophoresed through either 1.25% agarose TAE or 5% acrylamide/0.13% bisacrylamide TBE gels. Nucleic acids were visualized by staining with ethidium bromide, and their analysis and quantitation were performed using an AlphaImager. To verify the identities of the immunoprecipitated RNAs, Northern blotting was performed, essentially as described previously (57, 58). Formaldehyde-treated RNAs were electrophoresed on either a 1% agarose MOPS, 3.7% formaldehyde gel or a 7% acrylamide, 0.18% bisacrylamide/50% urea gel, depending on the RNA size. The RNAs were transferred to a nitrocellulose membrane by either capillary action (agarose) or electroblotting (acrylamide) and UV cross-linked (Stratalinker 2400, Stratagene, La Jolla, CA). Membrane-bound RNAs were hybridized for 18 h at 42 °C with 5 x 10⁶ cpm of radiolabeled oligonucleotide probes specific for S. cerevisiae ribosomal RNAs, including 25 S (5'-CTC TAA TCA TTC GCT TTA CC-3'), 18 S (5'-CAT GGC TTA ATC TTT GAG AC-3'), 5.8 S (5'-CGC ATT TCG CTG CGT TCT TCA TCG ATG-3'), and 5 S (5'-GGT CAC CCA CTA CAC TAC TCG G-3'). Hybridized RNAs were visualized by autoradiography and analyzed using a Storm PhosphorImager (Amersham Biosciences).

Ribosome Purification and Analysis—Analysis of S. cerevisiae ribosomes, ribosomal subunits, and polysomes by sucrose gradient ultracentrifugation followed published procedures with minor modifications (59). Briefly, yeast cells were grown to midlogarithmic phase in YPD medium at 30 °C with mild agitation (200 rpm) and then harvested by centrifugation. For polysome analysis, 50 µg/ml cycloheximide was added immediately before harvesting; for monosome and ribosome subunit analysis, it was not. Yeast extracts were prepared as previously described, with the exception that breaking buffer A contained 50 µg/ml cycloheximide for polysome analysis. After processing, five A₂₆₀ units of clarified extract was layered onto a 12-ml linear 7–47% sucrose gradient containing 50 mM Tris acetate, pH 7.0, 50 mM NH₄Cl, 12 mM MgCl₂, and 1 mM dithiothreitol. These gradients were centrifuged in an SW-40 rotor (Beckman Coulter, Fullerton, CA) at 38,500 rpm for 2.5 h. The gradients were recovered using an Auto Densi-Flow IIC gradient fractionator (Labconco, Kansas City, MO), and 0.6-ml fractions were collected. During fraction collection, absorbance at 254 nm was recorded using a UA-5 absorbance/fluorescence detector (Isco, Lincoln, NE). Proteins in these fractions were analyzed using SDS-PAGE, EMSA, and Western blotting, as described above.

Chromatin Immunoprecipitation and Analysis—Chromatin immunoprecipitations (ChIPs) were performed essentially as described previously (60). Yeast cultures were grown in 200 ml of YPA-galactose to a density of 10⁷ cells/ml and fixed with 1% formaldehyde. Cell pellets were resuspended in 1 ml of lysis buffer and lysed using a BeadBeater cell disrupter (BioSpec Products, Bartlesville, OK) three times for 1 min at 4 °C. Lysates were sonicated (Branson Ultrasonics, Danbury, CT) three times for 20 s (constant output, 1.5 duty cycle) to an average DNA length of 500 bp, which was confirmed by agarose gel electrophoresis and ethidium staining. Ten percent of the clarified cell lysate was set aside and designated the "input." Immunoprecipitations were performed for 3 h at 4 °C with 500 µl of clarified lysate, 30 µl of protein A/G-Sepharose (Sigma), and either 5 µg of anti-MYC 9E10 (Covance) or 1 µg of anti-HA 16B12 mAbs. Immunoprecipitates were washed, and the cross-links were reversed at 65 °C for 6 h. The recovered DNA was further purified with QiaQuick (Qiagen, Valencia, CA).

ChIP samples and their corresponding input dilutions were denatured by heating to 95 °C for 10 min in the presence of 0.4 M NaOH and 10 mM EDTA. Afterward, an equal volume of 2 M ammonium acetate, pH 7.0, was added to neutralize the solution. Denatured DNA was applied to a Hybond N+ membrane (Amersham Biosciences) using a Bio Dot microfiltration apparatus (Bio-Rad). Wells were rinsed twice with 2x SSC (300 mM NaCl and 30 mM sodium citrate, pH 7.0). The blotted membrane was rinsed once more with 2x SSC for 5 min, and the DNA was UV cross-linked to the membrane.

Random-prime labeling (Megaprime; Amersham Bioscience) reactions with [-³²P]dCTP and [-³²P]dGTP was used to label all probes. The telomere probe and the random genomic DNA probe were generated by random primed labeling of the alternating copolymer poly(dG-dT/dC-dA) (Amersham Biosciences) or total yeast genomic DNA, respectively. The DNAs for all the other probes were generated by PCR from yeast genomic DNA. The amplified DNAs, ranging from 275 to 450 bp in length, were gel-purified and extracted with QiaQuick prior to random prime labeling. The telomere probe was hybridized for >10 h at 65 °C, whereas all other probes were hybridized for >10 h at 60 °C. Blots were washed three times and exposed to phosphoscreens and analyzed with ImageQuant software. The five regions of the rDNA repeat used individually and collectively to probe blots corresponded to a nontranscribed sequence (NTS), an early transcribed sequence (ETS), a sequence within the 5 S rRNA gene (5S), and two sequences within the 25 S rRNA gene (25S/1 and 25S/2). The primer sequences used to amplify these five rDNA regions from genomic DNA have been described previously (61). The sequences of primer pairs used to PCR-amplify various ARS and subtelomeric regions for probe preparation included ARS305 (5'-CTC CGT TTT TAG CCC CCC GTG-3';5'-GAT TGA GGC CAC AGC AAG ACC G-3'), ARS1 (5'-GGT GAA ATG GTA AAA GTC AAC CCC CTG CG-3'; 5'-GCT GGT GGA CTG ACG CCA GAA AAT GTT-3'), ARS306 (5'-GCA AGC ATC TTG TTT GTA ACG CGA-3'; 5'-CCT CAG CGA TGT GCT CGC TC-3'), ARS603 (5'-CTC TTT CCC AGA TGA TAT CTA GAT GG-3'; 5'-CGA GGC TAA ATT AGA ATT TTG AAG TC-3'), ARS501 (5'-ATG CCT GAA AAA GAA TGT GTC T-3'; 5'-GCG TTA GGT AAT GCT CAA TTT T-3'), subtelomere Y' element, telomere-proximal (5'-AAG CTG CAT TTA GCA GGC AT-3'; 5'-TCC AAC TTC TCT GCT CGA ATC-3'), subtelomere Y' element, telomere-distal (5'-CGT TGT TTC AAG TGC ATA CTT T-3'; 5'-TGG GCA TAG ATC ACG CTT CA-3'), and subtelomere X element (5'-GAA GCA TGG TTT AGC GTG TT-3'; 5'-AAT GTT TCT GTG TTC TGG TGC G-3').

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Generation and Characterization of Yeast Expressing Epitope-tagged Stm1p—Immunological assays, including Western blotting, immunofluorescence microscopy, and immunoprecipitation, provide a powerful means of investigating proteins and determining their cellular localization and associated macromolecules. In yeast, the primary means of immunological protein surveillance involves using epitope-tagged proteins and well characterized monoclonal antibodies (62). Using the modules of Longtine et al. (51), we constructed strains with the STM1 gene tagged with three copies of the HA epitope on either its N or C terminus (HA₃-STM1 or STM1-HA₃) and located at its normal chromosomal location. Expression of the epitope-tagged Stm1p was driven from the endogenous STM1 promoter in STM1-HA₃ yeast and from the inducible GAL1 promoter in HA₃-STM1 yeast.

To characterize Stm1p expression in these strains, we performed Western blot analysis with anti-Stm1p polyclonal antibodies (Fig. 1A). We observed the expected expression of Stm1 proteins in our strains: e.g. galactose-dependent expression of HA₃-Stm1p (lanes 2 and 3), constitutive expression of Stm1-HA₃ (lane 6), and an absence of untagged Stm1p in both epitope-tagged strains (compare with lanes 1 and 4). More importantly, the absolute levels of epitope-tagged Stm1p present in the strains were comparable with those of untagged, wild-type Stm1p in the parental strains, with HA₃-Stm1p present at 170% and Stm1-HA₃p present at 54% of the wild-type levels. Western blot analysis was also performed with the anti-HA mAb (Fig. 1B). These data confirmed the presence of the HA₃ epitope tag as well as the specificity of this antibody and its minimal cross-reactivity with other yeast proteins, as has been observed by others (62).

View larger version (28K): [in this window] [in a new window]   FIG. 1. Physical and functional characterization of epitope-tagged Stm1p in yeast. Stm1 protein amounts and integrity were characterized through Western blotting of SDS-PAGE-resolved whole cell yeast extract proteins with either rabbit polyclonal anti-Stm1p polyclonal antibodies (A) or anti-HA.11 monoclonal antibody (B). Equivalent amounts of whole cell extracts were loaded to compare the HA3-STM1 strain (lanes 2 and 3) with its untagged parent strain (lane 1) and to compare the STM1-HA3 strain (lane 6) with its untagged parent (lane 5) as well as with an stm1- strain (lane 4). Stm1 proteins were expressed from the endogenous, constitutive STM1 promoter in all strains except HA3-STM1, which was under the control of the inducible GAL1 promoter. Induction by galactose is indicated by a plus sign. Apparent molecular masses were determined by comparison with protein molecular weight markers and are indicated at the left. Stm1p and other yeast Pu triplex-binding protein-DNA interactions were characterized through an EMSA with the radiolabeled purine motif triple-helical DNA probe X-PuT19 (C). Lanes were identical to those in A and B, with the addition of a probe alone control (P, lane 0). Locations of the gel well (W), the different yeast protein-triplex complexes (C0, C1*, C1, C2), the unbound triplex (T), and the unbound duplex (D) are indicated at the left. C1* and C1 correspond to the X-Pu complexes containing Stm1-HA3 and Stm1 proteins, respectively.

Stm1p Is Primarily a Cytoplasmic Protein—Given its very high binding affinity for G4 quadruplex and Pu triplex DNAs in vitro, Stm1p was initially thought to be a yeast nuclear protein (22, 32). This localization was supported by the presence of two consensus nuclear localization domains within the Stm1p protein sequence (64, 65). However, nuclear localization was not found with other predictive methods (66, 67). To understand why, we sought to empirically determine the cellular localization of Stm1-HA₃p by immunofluorescence microscopy. Using the anti-HA mAb 16B12 and Cy3-conjugated anti-mouse IgG polyclonal antibodies, Stm1-HA₃ protein was observed as a diffuse, granular fluorescence throughout formaldehyde-fixed yeast spheroplasts (Fig. 2, top panels, indicated in red). Most of this fluorescence was distributed within the cell cytoplasm, although a significant portion was also present in the nucleus, as evidenced by the overlap between Stm1-HA₃ fluorescence (indicated in red) and the fluorescence of a DNA-binding dye, 4',6-diamidino-2-phenylindole (indicated in blue). No substantial fluorescence was observed in either untagged wild-type or stm1- strains (data not shown). Also, no substantial changes in distribution were noted in the Stm1-HA₃ fluorescence pattern for yeast in different phases of the cell cycle (e.g. see Fig. 2, right panels, budding yeast in late mitosis).

View larger version (51K): [in this window] [in a new window]   FIG. 2. Stm1p resides primarily in the yeast cytoplasm. STM-HA3 yeast were examined by immunofluorescence microscopy. Stm1-HA3p, indicated by red fluorescence, was visualized using the anti-HA.11 primary antibody followed by Cy3-conjugated secondary antibody (-HA). Nuclear chromosomal DNA, indicated by blue fluorescence, was visualized by its binding to 4',6-diamidino-2-phenylindole (DAPI). DIC, differential interference contrast microscopy.

Using SDS-PAGE and Coomassie Blue staining, we observed more than 20 protein bands that coimmunoprecipitate with the Stm1-HA₃ protein (Fig. 3, lane 4). These bands varied in intensity over a 5-fold range, from 50 to 250 ng of protein, and corresponded to proteins with apparent molecular masses ranging from 50 kDa to less than 12.4 kDa. No protein bands were observed in a control reaction lacking the anti-HA antibody (lane 3), and only the heavy and light chains of the anti-HA antibody were observed in immunoprecipitates from wild-type yeast extracts (lane 2). All protein bands other than those from the antibody and Stm1-HA₃p were completely lost when the immunoprecipitates were washed with 500 mM NaCl (lane 5) and were partially lost when 15 mM EDTA was present during immunoprecipitation (lane 7). These additional proteins were also lost when RNase A was present in the immunoprecipitation buffer (lane 8) but not when DNase I was present (lane 6).

View larger version (92K): [in this window] [in a new window]   FIG. 3. Stm1p is part of a multiprotein complex. Proteins associating with Stm1-HA3 were examined by immunoprecipitation with the anti-HA.11 mAb (-HA) and SDS-PAGE. Shown is a photograph of a Coomassie Blue-stained gel. Immunoprecipitations were performed under a variety of conditions, including substitution of a yeast extract from the untagged parent strain (wt) and incubations in the presence of high ionic strength (500 Na), a divalent cation chelator (EDTA), and the nonspecific endonucleases DNase I and RNase A. Molecular masses of the marker proteins are indicated at the left. The identities of the IgG heavy and light chains and Stm1-HA3 protein are indicated at the right.

View larger version (47K): [in this window] [in a new window]   FIG. 4. Stm1p is associated with multiple ribosomal proteins. Proteins immunoprecipitated with Stm1-HA3 were resolved by SDS-PAGE, and their identities were determined by peptide fingerprinting. Molecular masses of the marker proteins are indicated at the left. The 2 x 8-mm gel slices used for peptide fingerprinting are indicated by boxes. The identities of the IgG heavy (h) and light (l) chains, Stm1-HA3, and the Stm1-HA3-associated proteins are indicated at the right. Rpl, ribosomal protein, large subunit; Rps, ribosomal protein, small subunit.

View larger version (87K): [in this window] [in a new window]   FIG. 5. Stm1p is associated with multiple RNAs. RNAs associating with Stm1-HA3 were examined by immunoprecipitation with the anti-HA.11 mAb (-HA) and either nondenaturing agarose (A) or denaturing polyacrylamide (B) gel electrophoresis. Shown are photographs of ethidium bromide-stained gels. Immunoprecipitations were performed under a variety of conditions, including substitution of extract from the untagged parent strain (wt) and incubations in the presence of high ionic strength (500 Na), a divalent cation chelator (EDTA), and the nonspecific endonucleases DNase I and RNase A. Lengths of duplex DNA markers in A and single-stranded RNA markers in B are indicated at the left. A titration of yeast RNAs (lanes 10–13) are included for quantitation purposes. The identities of the ribosomal RNAs (25S, 18S, and 5S) and tRNA are indicated at the right.

View larger version (31K): [in this window] [in a new window]   FIG. 6. Stm1p is associated with multiple ribosomal RNAs. RNAs that coimmunoprecipitate with Stm1-HA3 were resolved by either denaturing agarose (A) or polyacrylamide gel electrophoresis (B), and their identities were determined by Northern blotting. RNA-specific probes used are indicated at the left of each autoradiogram. Input material includes RNAs that coimmunoprecipitate in the presence of a Mg2+-containing buffer (M) or EDTA-containing buffer (E) buffer and a titration of total yeast RNA for comparison.

View larger version (31K): [in this window] [in a new window]   FIG. 7. Stm1p is associated with 80S ribosomes and polysomes. Yeast were treated either with (left panels) or without (right panels) the translation inhibitor cycloheximide (CHX), and their ribosomes were fractionated on a 7–47% sucrose gradient and scanned for absorbance at 254 nm (A254). A, UV absorbance trace of the fractionated gradient. Locations of different ribosomal species (40 and 60 S subunits, 80 S monosomes, and polysomes) are indicated above each trace. Fraction numbers are indicated below. B, Western blots of SDS-PAGE-separated Stm1-HA3p present in each fraction probed with anti-HA.11 mAb. C, EMSA analysis of triplex binding activities present in sucrose gradient fractions from untreated yeast. Fraction numbers are indicated below. Locations of the gel well (W), the different yeast protein-triplex complexes (C0, C0*, C1*, and C2), the unbound triplex (T), and the unbound duplex (D) are indicated at the left. C1* corresponds to the X-PuT19 complex containing Stm1-HA3p.

Stm1p Associates Specifically with Y' Element Sequences Proximal to Yeast Telomeres—Although most Stm1p is apparently located in the cytoplasm (Fig. 2) and is associated with ribosomes (Fig. 7), there also is a fraction of Stm1p visible in the nucleus (Fig. 2). Stm1p has been shown to specifically and avidly interact with both RNA- and DNA-containing G^*G multiplex structures in vitro (22, 32). Thus, to address whether Stm1p interacts with genomic DNA sequences in vivo, we performed ChIPs with HA₃-STM1 yeast. In this technique, DNA-binding proteins can be cross-linked to their cognate DNA binding sites during a brief exposure of live yeast to formaldehyde (72). The cross-link between protein and DNA facilitates recovery of the DNA site during immunoprecipitation of the DNA-binding protein. PCR or probe hybridization is typically used to analyze the recovered DNA sequences. For our purpose, probe hybridization was preferable, since most of the G/C-rich genomic regions we were initially interested in examining reside in naturally repetitive sequences, which are more difficult to analyze by PCR.

The specificity of our ChIP and probe hybridization analysis was first determined by ChIP using the control yeast strain YKU80-myc₁₈ or its isogenic untagged parent (Fig. 8, right column, and Table I). Yku80p and Hdf1p comprise the yeast Ku70/Ku80 heterodimer, a well known DNA end-binding complex present at the yeast telomere (73). We observed that Yku80-myc₁₈ preferentially associates with TG_1–3 telomere tracts (Fig. 8A) and with telomere-proximal regions of the subtelomeric Y' element (Table I). "-Fold enhanced association" values were calculated by normalizing specific hybridization signals against nonspecific background hybridization signal. In the first of two controls to assess nonspecific hybridization, primary antibody (-myc) was not added during immunoprecipitation of Yku80-myc₁₈ extracts. In the second background control, ChIP was performed on the untagged isogenic parent strain to assess dependence on the epitope-tagged protein. Higher backgrounds were consistently observed in the mock ChIP experiments lacking primary antibody. Consequently, all -fold enhanced association values were calculated normalizing to this background control. Thus, Yku80-myc₁₈ was found to exhibit 30- and 22-fold enhanced association for telomere and telomere-proximal Y' element sequences. Probing sequences progressively further from the telomere, we did not detect a substantial association of Yku80-myc₁₈ at the telomere-distal end of the Y' element or at the subtelomeric X element. As expected, we did not observe any Yku80-myc₁₈ association with ARS or with randomly distributed sequences throughout the yeast genome (Table I). A somewhat unexpected finding was the association of Yku80-myc₁₈ with certain rDNA sequences (Fig. 8B). However, taken as a whole, our analysis of the control YKU80-myc₁₈ strain suggests that the specificity of our ChIP and probe hybridization analysis was quite reasonable.

View larger version (49K): [in this window] [in a new window]   FIG. 8. Stm1p associates with subtelomeric Y' DNA. ChIPs were performed with HA3-STM1 yeast and its untagged parental strain using anti-HA.11 antibody (-HA, left column of spots). For comparison, a control ChIP was performed with YKU80-myc18 yeast and its untagged parental strain using anti-myc 9E11 antibody (-myc, right column of spots). The ChIP samples and dilutions representing 1.0 and 0.1% of the of the corresponding cell extract used for immunoprecipitation ("input") were applied to a nitrocellulose membrane and hybridized with several different probes, including a telomere probe (A) and an rDNA probe (B) (25S/1, 25S/2, 5S, ETS, and NTS, as described under "Experimental Procedures." To ascertain a relative measure of -fold enhanced chromatin immunoprecipitation, two controls were included to evaluate specificity and background levels. First, we performed a mock ChIP (ChIP -Ab) in which the appropriate primary antibody was omitted from an otherwise standard chromatin immunoprecipitation using extract prepared from the epitope-tagged strains. Second, we performed a standard chromatin immunoprecipitation with primary antibody on extracts of the untagged isogenic parent strains. In all cases, higher background hybridizations were observed with the former, ChIP -Ab control experiment. Therefore, we used this control as the background reference point to calculate the -fold enhanced chromatin immunoprecipitations as shown in Table I.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Stm1p is a moderately abundant (35,000–46,800 proteins/cell) S. cerevisiae protein that has been investigated for a variety of reasons over the last 10 years. Originally identified by biochemical means as one of the primary G4 quadruplex nucleic acid-binding proteins in yeast whole cell extracts, Stm1p or its corresponding gene has been identified in a number of genetic or biochemical screens. Although these studies have implicated Stm1p in several biological processes, ranging from apoptosis to telomere maintenance, they have not conclusively determined the molecular functions of Stm1p or its biological roles. With the development of multiple high throughput technologies and their applications in studies of S. cerevisiae, there now exists a wealth of information regarding almost every yeast protein, including Stm1p. Thus, we are at a point where the actual role of Stm1p can now be better determined.

Using immunofluorescence microscopy, we found that Stm1-HA₃p was primarily localized in the yeast cell cytoplasm, exhibiting a granular pattern of immunofluorescence, with a small percentage of the Stm1-HA₃p also present in the cell nucleus. Our results concur nicely with those obtained in a genome-wide transposon-tagging survey, in which a C-terminal truncated Stm1(178–273)p with a 93-amino acid HAT tag yielded a cytoplasmic/nuclear granular staining pattern of immunofluorescence with moderate intensity (75). Similar results were obtained using direct fluorescence microscopy with a full-length Stm1p bearing a C-terminal Aequorea victoria green fluorescent protein, which showed a primarily cytoplasmic localization (76). These data, however, contrast with those of Ligr et al. (48), who found a predominantly perinuclear localization of Stm1p C-terminally tagged with a single IRS epitope. They further determined through chromosome spreading experiments that Stm1p may also be directly associated with the chromatin that is present on the periphery of nucleolids. The reasons for this difference in the subcellular location between the work of Ligr et al. and others is not known but may reflect the choice of epitope tag or yeast strain used. In all cases, these tagged proteins were expressed from the endogenous STM1 promoter and from integrants at the normal chromosome XII location. Regardless, the consensus is that Stm1p is primarily a cytoplasmic protein, although some nuclear involvement cannot be completely ruled out.

Using conventional immunoprecipitation, we found that Stm1p was primarily associated with ribosomal proteins and rRNAs. This result was confirmed by fractionation of soluble yeast proteins using sucrose gradient ultracentrifugation and Western blotting, which found almost all of the Stm1-HA₃p to be primarily associated with 80 S monosomes and polysomes. Such results were not observed in a systematic analysis of S. cerevisiae protein complexes using tandem affinity purification, which found Stm1p present in three complexes obtained with TAP-tagged Sec31p, Las17p, or Hek2p (77). Interestingly, both the TAP-tagged Sec31p and Hek2p complexes contained some ribosomal proteins and/or translation initiation proteins. However, these observations most likely reflect the focus of these investigations, which were directed toward identifying proteins in 100–900-kDa complexes (78). Stm1p has been identified as an abundant, salt-labile ribosome-associated protein using an overexpressed C-terminal FLAG-His₆ epitope-tagged ribosomal protein Rpl25p and immunoaffinity purification (79). However, these authors did not determine which ribosomal species were associated with Stm1p or that most soluble Stm1p is ribosome-associated. Likewise, an Stm1p-related protein, Ray38p, was found to be released from ribosomes isolated from cycloheximide-treated wild-type Candida maltosa yeast (80). The association of Ray38p with ribosomes was verified by size exclusion chromatography and Western blotting, although the identity of the exact ribosomal species was not determined. Note that we also examined the association of Stm1p with ribosomes isolated from cycloheximide-treated S. cerevisiae cells and found that although polysome integrity was maintained as expected, little to no Stm1p was released.

Using chromatin immunoprecipitation, we found that HA₃-Stm1p preferentially associated with telomere-proximal Y' element sequences. Some prior evidence had suggested that Stm1p might interact with the ends of chromosomes. The aforementioned chromosome spreading data of Ligr et al. (48) is consistent with a telomeric localization for Stm1p, given that telomeres are overrepresented on the periphery of nucleolids (81). Ito et al. (82) documented a two-hybrid interaction between a MEC3 bait and an STM1 prey in a comprehensive two-hybrid analysis of yeast protein-protein interactions. Mec3p is a DNA-binding protein involved in DNA damage-dependent checkpoints, telomere silencing, and telomere length maintenance (83). Further evidence was obtained in a directed two-hybrid screen using CDC13 as bait, which isolated a single STM1 clone out of 17,000 total colonies (49). Cdc13p is a telomere-binding protein specific for (TG_1–3)_n single-stranded DNA and is directly involved in regulating the replication of telomeres and protecting the integrity of telomeres (84). The physical interaction between the N-terminal portion of Cdc13p and Stm1p was verified using a glutathione S-transferase pull-down experiment (49). In addition, STM1 in multicopy partially suppressed the temperature-sensitive growth and long telomere phenotypes of cdc13-1 yeast but failed to suppress the elevated level of single-stranded G-rich tails. Interestingly, a fusion between STM1 and the N terminus of CDC13, which lacks the DNA binding domain required for Cdc13p function (85), complemented the viability of cdc13- yeast, albeit growth was weak (49). Last, the partial suppression of cdc13-1 temperature sensitivity by STM1 multicopy expression was abolished by multicopy expression of Sgs1p, a primarily nucleolar RecQ family helicase that is involved in maintaining genome stability and is of special interest because of its ability to unwind G-quadruplex and Pu-triplex DNAs (16).²

Genetic analyses have proven a highly powerful means of elucidating the biological functions of proteins in S. cerevisiae, and STM1 has appeared in several genetic screens. STM1 has been identified as a multicopy suppressor of temperature-sensitive tom1 and htr1 mutants (63), staurosporine-sensitive pop2, ccr4, and pkc1 mutants, and a caffeine-sensitive mpt5 mutant (52). None of these genes have an obvious direct relationship with either ribosome or telomere function. STM1 has more recently been identified in a screen for yeast cDNAs whose overexpression caused growth arrest in proteasome-defective pre1-1 pre4-1 cells in a fashion consistent with an apoptosis-like cell death (48). STM1 deletion has been reported to have a minimal phenotype (52, 63), although subsequent investigators found significant growth defects for stm1-1 yeast when grown in the presence of caffeine, following UV irradiation, or when grown at elevated temperatures in the presence of the radiomimetic drug bleomycin, but not when grown with staurosporine or the alkylating agent methyl methanesulfonate (48). Most recently, STM1 was identified in a large scale synthetic genetic array analysis with deletions of the query genes CTF4 and KRE9, genes that are involved in sister chromatid cohesion and cell surface (16)--glucan assembly, respectively (87). Taken together, these diverse genetic data suggest that Stm1p is involved in multiple, seemingly unrelated, biological processes. Alternatively, they may also reflect the involvement of Stm1p in a very general process (e.g. transcription or translation) that impinges on many more specific biological processes.

Complete sequencing of the S. cerevisiae genome, coupled with advances in microarray technology, has allowed the facile determination of genome-wide changes in gene expression as a function of different genotypic changes or environmental stimuli. Changes in Stm1 mRNA levels have been observed in many of these (88). Potentially more useful is the ability to identify yeast genes whose expression pattern mirrors that of STM1 for a given stimulus, since such coordinate regulation can be indicative of a shared function. Only a few of the microarray studies are easily amenable to comparative analysis, and of these only a subset demonstrated significant changes in STM1 expression (89). However, among these studies, including expression changes in response to DNA-damaging agents (90), environmental changes (91), varying zinc levels (92), histone depletion (93), the diauxic shift (94), and sporulation (29), 62 of the 113 genes reported as having expression profiles similar to that of STM1 were known structural components of the ribosome, with the next most commonly represented genes being either those of unknown function (9 of 113) or those involved in translation (6 of 113). This coordinate regulation was most aptly demonstrated in two studies, response to DNA-damaging agents and environmental changes, which reported 32 of the 34 genes that responded like STM1 as those encoding ribosomal proteins and the remaining genes being either involved in translation or of unknown function. Taken together, these data strongly suggest that Stm1p is involved in either ribosome structure or function.

What is the actual role of Stm1p in yeast, and does it involve interaction with G^*G multiplex nucleic acids? Stm1p association with the ribosome strongly suggests a role in translation or an associated process (e.g. protein folding or export). In addition, the large ribosomal RNAs contain multiple G-rich stretches that have the potential for G^*G multiplex formation. Telomere-proximal Y' element DNA sequences include a region of 36-bp repeats, an ARS consensus sequence, and multiple Tbf1p-binding sites (86). Many of these sequences are G/T-rich and have the potential for interesting DNA structures. The exact role of these sequences is unknown. Nonetheless, the interaction of Stm1p with both RNA- and DNA-containing macromolecular complexes is intriguing and suggests a possible type of communication between protein synthesis and various DNA-dependent phenomena. We hypothesize that under normal growth conditions, Stm1p is primarily associated with functional ribosomes, where it plays an ancillary role in the process of translation. A small fraction of the cellular Stm1p is normally associated with chromosomal DNA, particularly subtelomeric sequences, where it may be involved in some aspect of telomere biosynthesis. However, under circumstances of stress, especially stresses that compromise ribosome function, the Stm1p binding equilibrium will be shifted toward higher concentrations of free protein. If not degraded through the proteosomal pathway, these free Stm1p can then promiscuously interact with DNA and impede various DNA template-dependent processes. Given effective checkpoint controls, these impeded processes could result in the delay of cell cycle progression. Such a delay could be used to rectify conditions of stress and/or to respond to their consequences (e.g. through postreplicative repair or sporulation, both of which have been shown to involve G^*G multiplex-acting proteins) (31, 68).³ However, under circumstances where this delay cannot be achieved, or if sufficient concentrations of Stm1p are present, events may then transpire resulting in an apoptosis-like death. Determining the precise Stm1p-binding sites on both ribosomes and subtelomeric Y' DNA as well as experiments directed toward elucidating the functions of Stm1p in translation and DNA-dependent processes under normal and stress conditions should ultimately aid in understanding of the biological roles of Stm1p.

	FOOTNOTES

 
^* This work was supported by Robert A. Welch Foundation Grant G-1199 (to M. W. V. D.). 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.

Present address: Aurigene Discovery Technology Ltd., Electronic City Phase II, Hosur Rd., 39-40 KIABD Industrial Area, Bangalore 560 100, India.

To whom correspondence should be addressed: Dept. of Molecular and Cellular Oncology, The University of Texas M. D. Anderson Cancer Center, Unit 79, 1515 Holcombe Blvd., Houston, TX 77030-4009. Tel.: 713-792-8954; Fax: 713-794-0209; E-mail: mvandyke{at}mdanderson.org.

¹ The abbreviations used are: Pu triplex, purine motif triple helical DNA; ARS, autonomous replicating sequence(s); ChIP, chromatin immunoprecipitation; EMSA, electrophoretic mobility shift assay; G4, G-quartet; G^*G, Hoogsteen hydrogen-bonded guanines; HA, influenza hemagglutinin protein epitope; mAb, monoclonal antibody; NTS, nontranscribed sequence; ETS, early transcribed sequence.

² R. D. Hunt and M. W. Van Dyke, unpublished observations.

³ D. V. Mehta, unpublished observations.

	ACKNOWLEDGMENTS

 
We thank John Pringle for the pFA6a series of plasmids used in making the deletion and epitope-tagged STM1 strains, Christian Koch for S. cerevisiae strain K699, and A. Bertuch and V. Lundblad for strain YKU80-myc₁₈. We also thank Mary Ann Gawinowicz, Natasha Van Dyke, and Miles Wilkerson for assistance and comments and the Saccharomyces Genome Database for its wealth of freely accessible information.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Wells, R. D., Collier, D. A., Hanvey, J. C., Shimizu, M., and Wohlrab, F. (1988) FASEB J. 2, 2939-2949[Abstract]
2. Frank-Kamenetskii, M. D., and Mirkin, S. M. (1995) Annu. Rev. Biochem. 64, 65-95[Medline] [Order article via Infotrieve]
3. Williamson, J. R. (1994) Annu. Rev. Biophys. Biomol. Struct. 23, 541-576[Medline] [Order article via Infotrieve]
4. Zain, R., and Sun, J. S. (2003) Cell. Mol. Life Sci. 60, 862-870[Medline] [Order article via Infotrieve]
5. Arthanari, H., and Bolton, P. H. (2001) Chem. Biol. 8, 221-230[CrossRef][Medline] [Order article via Infotrieve]
6. Blackburn, E. H. (1991) Nature 350, 569-573[CrossRef][Medline] [Order article via Infotrieve]
7. Behe M. J. (1995) Nucleic Acids Res. 23, 689-695[Abstract/Free Full Text]
8. Darnell, J. C., Jensen, K. B., Jin, P., Brown, V., Warren, S. T., and Darnell, R. B. (2001) Cell 107, 489-499[CrossRef][Medline] [Order article via Infotrieve]
9. Agazie, Y. M., Burkholder, G. D., and Lee, J. S. (1996) Biochem. J. 316, 461-466[Medline] [Order article via Infotrieve]
10. Dempsey, L. A., Sun, H., Hanakahi, L. A., and Maizels, N. (1999) J. Biol. Chem. 274, 1066-1071[Abstract/Free Full Text]
11. Rooney, S. M., and Moore, P. D. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 2141-2144[Abstract/Free Full Text]
12. Zahler, A. M., Williamson, J. R., and Cech, T. R. (1991) Nature 350, 718-720[CrossRef][Medline] [Order article via Infotrieve]
13. Kohwi, Y., and Kohwi-Shigamatsu, T. (1991) Genes Dev. 5, 2547-2554[Abstract/Free Full Text]
14. Simonsson, T., Pecinka, P., and Kubista, M. (1998) Nucleic Acids Res. 26, 1167-1172[Abstract/Free Full Text]
15. Kopel, V., Pozner, A., Baran, N., and Manor, H. (1996) Nucleic Acids Res. 24, 330-335[Abstract/Free Full Text]
16. Sun, H., Bennett, R. J., and Maizels, N. (1999) Nucleic Acids Res. 27, 1978-1984[Abstract/Free Full Text]
17. Weisman-Shomer, P., and Fry,