Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

In Addition to SEC11, a Newly Identified Gene,SPC3, Is Essential for Signal Peptidase Activity in the Yeast Endoplasmic Reticulum

1997; Elsevier BV; Volume: 272; Issue: 20 Linguagem: Inglês

10.1074/jbc.272.20.13152

ISSN

1083-351X

Autores

Hong Fang, Chris Mullins, Neil Green,

Tópico(s)

Cellular transport and secretion

Resumo

Among the three characterized subunits comprising the signal peptidase complex of the yeast Saccharomyces cerevisiae (Sec11p, Spc1p, and Spc2p), only Sec11p is essential for cell growth, signal peptide cleavage, and signal peptidase-dependent protein degradation. Here we report the cloning of the SPC3 gene encoding the homolog to mammalian signal peptidase subunit SPC22/23. We find that Spc3p is also required for cell growth and signal peptidase activity within the yeast endoplasmic reticulum. Among the three characterized subunits comprising the signal peptidase complex of the yeast Saccharomyces cerevisiae (Sec11p, Spc1p, and Spc2p), only Sec11p is essential for cell growth, signal peptide cleavage, and signal peptidase-dependent protein degradation. Here we report the cloning of the SPC3 gene encoding the homolog to mammalian signal peptidase subunit SPC22/23. We find that Spc3p is also required for cell growth and signal peptidase activity within the yeast endoplasmic reticulum. Amino-terminal signal sequences of proteins targeted to the endoplasmic reticulum (ER)1 are cleaved by a membrane-bound endoprotease termed signal peptidase (1Milstein C. Brownlee G.G. Harrison T.M. Matthews M.B. Nat. New Biol. 1972; 239: 117-120Crossref PubMed Scopus (374) Google Scholar, 2Blobel G. Dobberstein B. J. Cell Biol. 1975; 67: 835-851Crossref PubMed Scopus (1826) Google Scholar). This enzyme has also been shown to catalyze the proteolytic fragmentation of some abnormal membrane proteins, thereby leading to their degradation (3Yuk M.H. Lodish H.F. J. Cell Biol. 1993; 123: 1735-1749Crossref PubMed Scopus (38) Google Scholar, 4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar). Isolated from the yeast Saccharomyces cerevisiae, ER signal peptidase (SP) contains four nonidentical protein subunits (5YaDeau J.T. Klein C. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 517-521Crossref PubMed Scopus (44) Google Scholar, 6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). Three of the subunits of this signal peptidase complex (SPC) have been functionally examined. Sec11p (17 kDa) is required for signal peptide cleavage (8Bohni P.C. Deshaies R.J. Schekman R.W. J. Cell Biol. 1988; 106: 1035-1042Crossref PubMed Scopus (105) Google Scholar) and signal peptidase-dependent protein degradation (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar). Sec11p is related to two subunits of the mammalian SPC (SPC18 and SPC21) (9Greenburg G. Shelness G.S. Blobel G. J. Biol. Chem. 1989; 264: 15762-15765Abstract Full Text PDF PubMed Google Scholar, 10Shelness G.S. Blobel G. J. Biol. Chem. 1990; 265: 9512-9519Abstract Full Text PDF PubMed Google Scholar). In contrast to Sec11p, the Spc1p (11 kDa) and Spc2p (18 kDa) subunits of the yeast SPC are nonessential for cell growth and enzyme activity (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). Spc1p and Spc2p, however, perform auxiliary and nonredundant roles. Spc1p facilitates signal peptide cleavages in cells burdened with high levels of a membrane protein substrate of the signal peptidase-dependent protein degradation pathway (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar). Spc2p is important for enzyme activity and cell viability at elevated temperatures (7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). Spc1p and Spc2p are homologous to mammalian subunits SPC12 and SPC25, respectively (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar,11Greenburg G. Blobel G. J. Biol. Chem. 1994; 269: 25354-25358Abstract Full Text PDF PubMed Google Scholar, 12Kalies K.-U. Hartmann E. J. Biol. Chem. 1996; 271: 3925-3929Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). Despite the fact that a multisubunit signal peptidase has been purified from yeast and mammalian cells, enzymes exhibiting less subunit complexity have been identified in other systems. Leader peptidase from the inner membrane of E. coli consists of a single polypeptide chain (13Zwizinski C. Wickner W. J. Biol. Chem. 1980; 255: 7973-7977Abstract Full Text PDF PubMed Google Scholar). This protein exhibits limited homology to Sec11p and to mammalian SPC18 and SPC21 (14Dalbey R.E. von Heijne G. Trends Biochem. Sci. 1992; 17: 474-478Abstract Full Text PDF PubMed Scopus (177) Google Scholar, 15van Dijl J.M. de Jong A. Vehmaanpera J. Venema G. Bron S. EMBO J. 1992; 11: 2819-2828Crossref PubMed Scopus (131) Google Scholar). Signal peptidase purified from hen oviduct contains two subunits, one related to Sec11p and a second related to SPC22/23 of the mammalian SPC (16Baker R.K. Lively M.O. Biochemistry. 1987; 26: 8561-8567Crossref PubMed Scopus (38) Google Scholar, 17Shelness G.S. Kanwar Y.S. Blobel G. J. Biol. Chem. 1988; 263: 17063-17070Abstract Full Text PDF PubMed Google Scholar, 18Newsome A. McLean J.W. Lively M.O. Biochem. J. 1992; 282: 447-452Crossref PubMed Scopus (14) Google Scholar, 19Lively M.O. Shelness G.S. von Heijne G. Signal Peptidases. R. G. Landes Co., Austin, TX1994: 59-71Google Scholar). In the present study, we have cloned and characterized theSPC3 gene encoding the yeast homolog to mammalian SPC22/23. We find that, as with Sec11p, Spc3p is essential for signal peptide cleavage and signal peptidase-dependent protein degradation. Our data are thus in agreement with in vitrostudies, using an avian system, which show that a two-subunit complex functions to cleave signal peptides. The idea that both Sec11p and Spc3p are related to the prokaryotic signal peptidase is also discussed. Plasmid pSPC3 bearing SPC3 was isolated from a high copy (2 μm) plasmid library marked with LEU2 (American Type Culture Collection ATCC No. 37323). Among 5,500 transformants of strain CMYD1 (sec11–7) (genotypes of yeast strains are listed in Table I), pSPC3 was isolated from one colony that grew at 32 °C. The SPC3 gene was amplified from pSPC3 by a polymerase chain reaction (PCR) using the forward oligonucleotide primer CGGGATCCACACGTGAATACTACC, which is located 204 bp upstream of the SPC3 start codon, and the reverse primer CGGAATTCAATAAATGGGAACAG, which is located 189 bp downstream of theSPC3 stop codon. The amplified fragment was restricted withBamHI and EcoRI and inserted into low-copy (CEN) plasmid pRS314 (TRP1) (21Sikorski R.S. Hieter P. Genetics. 1989; 122: 19-27Crossref PubMed Google Scholar). The resulting plasmid was named pHF332. A 1.5-kb HindIII-XbaI restriction fragment containing the SPC3 gene was excised from plasmid pSPC3 and inserted into 2 μm plasmid pRS426 (URA3) (22Christianson T.W. Sikorski R.S. Dante M. Shero J.H. Hieter P. Gene (Amst.). 1992; 110: 119-122Crossref PubMed Scopus (1429) Google Scholar). The resulting plasmid was named pHF331.Table IYeast strains used in this studyStrainGenotypeReferenceSEY6210.5MATα/MATa ura3–52/ ura3–52 leu2–3,112/ leu2–3,112 his3-Δ200/his3-Δ200 trp1-Δ901/ trp1-Δ901 suc2-Δ9/suc2-Δ9 ADE2/ade2–101 LYS2/lys2–8020Herman P.K. Emr S.D. Mol. Cell. Biol. 1990; 10: 6742-6754Crossref PubMed Scopus (354) Google ScholarSEY6210MATαura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 suc2-Δ9 lys2–8020Herman P.K. Emr S.D. Mol. Cell. Biol. 1990; 10: 6742-6754Crossref PubMed Scopus (354) Google ScholarSEY6211MATa ura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 suc2-Δ9 ade2–10120Herman P.K. Emr S.D. Mol. Cell. Biol. 1990; 10: 6742-6754Crossref PubMed Scopus (354) Google ScholarCMYD1MATa sec11–7 ura3–52 leu2–3 his4–6196Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google ScholarCMY195MATαspc2Δ1::URA3 ura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 lys2–80This studyHFY401MATαspc1Δ1::TRP1 ura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 suc2-Δ9 ade2–101 lys2–806Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google ScholarHFY405MATa spc3Δ1::LEU2 ura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 suc2-Δ9This studyHFY406MATαspc3–4 ura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 suc2-Δ9 lys2–80This studyHFY407MATa spc3–4 ura3–52 leu2–3,112 his3-Δ200 trp1-Δ901 suc2-Δ9 ade2–101This study Open table in a new tab The 1.5-kbHindIII-XbaI restriction fragment containing theSPC3 gene was inserted into the polylinker of pUC19 (23Yanisch-Perron C. Vieira J. Messing J. Gene (Amst.). 1985; 33: 103-119Crossref PubMed Scopus (11449) Google Scholar). The resulting plasmid was restricted with NheI, which cuts 30 bp upstream of the SPC3 open reading frame, andSpeI, which cuts 3 bp downstream of the stop codon ofSPC3, and the SPC3 gene was replaced with a 1.6-kb NheI DNA fragment containing the LEU2gene. This NheI fragment was obtained from a modified YDp-L plasmid (24Berben G. Dumont J. Gilliquet V. Bolle P.-A. Hilger F. Yeast. 1991; 7: 475-477Crossref PubMed Scopus (317) Google Scholar) that contains LEU2 flanked by twoNheI sites (a NheI linker was inserted at theSmaI site of YDp-L). This construct containing a replacement of SPC3 with the LEU2 gene was restricted withHindIII and XbaI and transformed into diploid strain SEY6210.5 with selection for leucine prototrophs. Transformants bearing a replacement of the SPC3 gene with theLEU2 gene were confirmed by genomic PCR using oligonucleotide primers described above. For construction of a frameshift mutation in SPC3, the single BglII site of pHF331 (located at codons 80–82 of SPC3) was restricted and then filled in with the Klenow fragment of DNA polymerase and ligated. For construction of temperature-sensitive mutations, the plasmid pSPC3 was used as a template for PCR under conditions that favor misincorporation of dNTPs by Taq polymerase (25Leung D.W. Chen E. Goeddel D.V. Technique ( Phila. ). 1989; 1: 11-15Google Scholar). The above described oligonucleotide primers located upsteam and downstream of theSPC3 gene were used. After 30 cycles of DNA amplification, the PCR product was restricted with BamHI andEcoRI and cloned into the polylinker of plasmid pRS314 (CEN6 TRP1). The resulting plasmids carrying a pool of mutagenized SPC3 genes were transformed into strain HFY405 (Δspc3)/pHF331 (2 μm SPC3 URA3), and transformants were subjected to the plasmid shuffle technique (26Sikorski R.S. Boeke J.D. Methods Enzymol. 1991; 194: 302-318Crossref PubMed Scopus (492) Google Scholar). From 200 transformants, four temperature-sensitive mutants containing plasmids pHF333 (CEN6 spc3–1 TRP1), pHF334 (CEN6 spc3–2 TRP1), pHF335 (CEN6 spc3–3 TRP1), and pHF336 (CEN6 spc3–4 TRP1) were isolated. To confirm that the temperature-sensitive phenotype was plasmid linked, each of the plasmids was purified and transformed into strain HFY405/pHF331, using plasmid shuffling to replace pHF331 with each of the transformed plasmids. The spc3–4 mutation was integrated into the SPC3chromosomal locus using methods described previously (27Scherer S. Davis R.W. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 4951-4955Crossref PubMed Scopus (482) Google Scholar). TheBamHI-EcoRI fragment containing thespc3–4 gene from pHF336 was inserted into plasmid pRS306 (URA3) (22Christianson T.W. Sikorski R.S. Dante M. Shero J.H. Hieter P. Gene (Amst.). 1992; 110: 119-122Crossref PubMed Scopus (1429) Google Scholar). The resulting construct was restricted withNheI. Linear DNA was transformed into haploid strains SEY6210 and SEY6211. A Ura+ transformant of each strain was placed on a minimal agar plate supplemented appropriately (28Green N. Walter P. Mol. Cell. Biol. 1992; 12: 276-282Crossref PubMed Scopus (21) Google Scholar) and containing 5-fluoroorotic acid (1 mg/ml). After several days of incubation at 23 °C, colonies were picked and tested for growth at 37 °C. From this analysis, temperature-sensitive strains were detected at a frequency of ∼50%. Plasmid pSPC3 was introduced into representative temperature-sensitive strains to test for complementation of the temperature-sensitive growth defect. The resulting temperature-sensitive strains used in this study were named HFY406 (MATα spc3–4) and HFY407 (MATa spc3–4). In a previous study, we observed that overexpression of Spc1p suppressed a mutational defect in Sec11p (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar). A similar approach was thus employed in a search for the yeast homolog to mammalian SPC subunit SPC22/23 (Experimental Procedures). In this analysis, pSPC3 was isolated from a high-copy (2 μm) plasmid library through the ability of this plasmid to suppress the growth defect of strain CMYD1 (sec11–7) (Fig. 1 A). As shown below, the plasmid pSPC3 bears the gene encoding the yeast homolog of SPC22/23. Fig. 1 A also depicts the growth of strain CMYD1 carrying high copy plasmid pSPC1, which contains the SPC1gene (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar). The data show that pSPC1 suppressed the sec11mutation at a higher temperature than pSPC3. DNA sequencing at one end of the chromosomal fragment contained within pSPC3 revealed a match to a sequence on chromosome XII. The identified sequence was that of PET309, which encodes a protein required for stability and translation of COX1 mitochondrial mRNA (29Manthey G.M. McEwen J.E. EMBO J. 1995; 14: 4031-4043Crossref PubMed Scopus (179) Google Scholar). Among the open reading frames present nearPET309, we identified an uncharacterized gene that we have named SPC3 (accession no.Z73238). 2Ellen Beasley (Stanford University) independently registered the SPC3 gene in theSaccharomyces genome data base under the nameSPY2. To determine whetherSPC3 suppressed sec11–7, a frameshift mutation was constructed at codon 82 of the SPC3 gene (see "Experimental Procedures"). The high copy plasmid bearing frame-shifted SPC3 did not suppress the growth defect of strain CMYD1 (sec11–7) at 32 °C. The control plasmid containing wild-type SPC3 suppressed the sec11–7mutation. To determine if SPC3 also suppresses the signal peptidase defect seen in sec11 mutant cells, strains CMYD1 (sec11) and CMYD1/pSPC3 were subjected to a pulse-chase analysis using methods described previously (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 28Green N. Walter P. Mol. Cell. Biol. 1992; 12: 276-282Crossref PubMed Scopus (21) Google Scholar). Cells were grown to log phase at 23 °C, shifted to 32 °C for 30 min, pulse labeled for 5 min with 35S-EXPRESS, and chased for 30 min in the presence of excess unlabeled methionine and cysteine, and proteins were precipitated from cell extracts with antibodies directed against ER resident Kar2p (30Normington K. Kohno K. Kozutsumi Y. Gething M.-J. Sambrook J. Cell. 1989; 57: 1223-1242Abstract Full Text PDF PubMed Scopus (304) Google Scholar, 31Rose M. Misra L.M. Vogel J.P. Cell. 1989; 57: 1211-1221Abstract Full Text PDF PubMed Scopus (526) Google Scholar). As shown in Fig. 1 B, the precursor of Kar2p was converted to mature Kar2p to a greater extent insec11 mutant cells containing pSPC3 than in sec11cells lacking pSPC3 (compare lanes 1 and 3). Overexpression of the SPC3 gene thus partially corrected the signal peptide cleavage defect in the sec11 mutant. Primary amino acid sequence analysis revealed that the product of theSPC3 gene, Spc3p, exhibited 55% similarity and 22% identity to SPC22/23 of the mammalian SPC (Fig. 2). These measurements are similar to the homologies shared between Spc1p and Spc2p and their mammalian counterparts (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). Spc3p is also related to the SPC22/23 homolog (18Newsome A. McLean J.W. Lively M.O. Biochem. J. 1992; 282: 447-452Crossref PubMed Scopus (14) Google Scholar) present in the two-subunit avian SPC (54% similarity, 24% identity). Spc3p is nonhomologous, however, to Sec11p, Spc1p, and Spc2p. Hydropathy analysis of Spc3p and SPC22/23 (32Kyte J. Doolittle R.F. J. Mol. Biol. 1982; 157: 105-132Crossref PubMed Scopus (17095) Google Scholar) identified a single hydrophobic stretch of amino acids located near the amino termini of the respective proteins. This region in SPC22/23 has, indeed, been shown to span the ER membrane (33Shelness G.S. Lin L. Nicchitta C.V. J. Biol. Chem. 1993; 268: 5201-5208Abstract Full Text PDF PubMed Google Scholar). The calculated molecular masses of SPC22/23 (20.3 kDa) and Spc3p (21.3 kDa) as well as the isoelectric points, 8.8 for SPC22/23 and 8.6 for Spc3p, are similar. Heterozygous disruption of the SPC3 gene was accomplished in diploid strain SEY6210.5 through replacement of SPC3 with the LEU2 marker (see "Experimental Procedures"). Genomic PCR analysis was used to confirm the heterozygous disruption of SPC3 in Leu+ transformants of strain SEY6210.5 (data not shown). The heterozygote was sporulated, and the spores generated were examined using tetrad analysis. Dissection of 10 tetrads revealed only two viable spores from each tetrad. All recovered strains were leucine auxotrophs, suggesting cells bearing the spc3 null mutation were inviable. The requirement of SPC3 for cell growth was confirmed by introducing plasmid pHF331 (SPC3 URA3) into above described strain SEY6210.5 containing a heterozygous disruption of theSPC3 gene. Tetrad analysis of this transformant yielded four viable spores from each of five tetrads examined. Two spores from each tetrad were Leu+ Ura+. Colonies were not recovered upon placement of the Leu+ Ura+ cells on agar plates containing 5-fluoroorotic acid, a chemical used to select for yeast cells cured of plasmids bearing the URA3gene (34Boeke J.D. Trueheart J. Natsoulis G. Fink G.R. Methods Enzymol. 1987; 154: 164-175Crossref PubMed Scopus (1071) Google Scholar). The requirement of SPC3 for cell growth was circumvented in further biochemical studies through the generation of temperature-sensitive (ts) mutations in the SPC3 gene. Using a PCR-based method of nucleotide misincorporation (see "Experimental Procedures"), mutant alleles spc3–1, spc3–2, spc3–3, and spc3–4 were generated. The spc3–3 mutant failed to grow at 32 °C, whereas the nonpermissive temperature for the remaining mutants was 37 °C. We screened cells containing these mutations in an attempt to identify at least one allele that rapidly inhibited Spc3p function following a shift of mutant cells to their nonpermissive temperatures. Strain HFY405 (Δspc3) bearing a series of plasmids containing these mutations (plasmids are described in "Experimental Procedures") was grown to log phase at 23 °C, then shifted to 37 °C for the time intervals indicated in Fig.3. Cell extracts were examined by Western blotting using anti-Kar2p antibodies. Significant amounts of preKar2p accumulated in the spc3–3 and spc3–4 mutants after a 30-min incubation at their nonpermissive temperature (lanes 8–9). Increasing levels of preKar2p appeared in these cells after 1 h (lanes 12 and 13) and 2 h (lanes 16 and 17). In contrast, the spc3–1 andspc3–2 mutants exhibited a modest defect in the cleavage of preKar2p even after 2 h at the nonpermissive temperature (lanes 14 and 15). Furthermore, the nonpermissive temperature for the spc3–4 mutant (37 °C) was higher than the sec11–7 mutant (32 °C), a difference that made it possible to follow both mutations in genetic crosses. For these reasons, the spc3–4 mutation was chosen for further study. A pulse labeling analysis (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 28Green N. Walter P. Mol. Cell. Biol. 1992; 12: 276-282Crossref PubMed Scopus (21) Google Scholar) was employed to more accurately measure signal peptide cleavage in the spc3–4 mutant. Strain HFY405 (Δspc3) bearing pHF336 (CEN6 spc3–4 TRP1) or control plasmid pHF332 (CEN6 SPC3 TRP1) was grown to log phase at 23 °C and then shifted to 37 °C for 5 min. Cells were pulse labeled, and proteins were precipitated from cell extracts with anti-Kar2p antibodies. The sec11 mutant was examined for use as a control. The data in Fig. 4demonstrate that cells bearing the wild-type SPC3 gene displayed efficient cleavage of preKar2p (lane 1). In contrast, the spc3 (lane 2) and sec11(lane 3) mutations strongly inhibited preKar2p cleavage at the nonpermissive temperature. The fact that this defect was present in the spc3 mutant after shifting cells to 37 °C for only 5 min argues in favor of a direct effect of the spc3 mutation on the signal peptidase reaction. The results presented thus far do not eliminate the possibility that the spc3–4 mutation inhibited protein translocation, thereby resulting in an accumulation of preKar2p in the cytoplasm. To test this possibility, strain HFY405 (Δspc3)/pHF336 (spc3–4) was grown to log phase at the permissive temperature and shifted to 37 °C for 1.5 h, and membranes were prepared using methods described previously (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar). The membrane fraction was treated with proteinase K in the presence and absence of detergent and subjected to Western blotting with anti-Kar2p antibodies. The results depicted in Fig. 5 A demonstrated that preKar2p was protected from proteolytic digestion in extracts lacking detergent (lanes 3–7) but exposed to proteolytic attack in extracts incubated with detergent (lanes 8–12). A proteolytic fragment, Kar2f, was also present in this analysis. Kar2f has been seen previously in proteinase K digests of Kar2p (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar). Together, these data indicate that the spc3–4 mutation did not inhibit protein translocation. To corroborate this result, vacuolar protein carboxypeptidase Y (CPY) (35Stevens T.B. Esmon B. Schekman R. Cell. 1982; 30: 443-448Abstract Full Text PDF Scopus (371) Google Scholar) was examined. CPY was chosen for this analysis because untranslocated preproCPY (ppCPY) migrates differently from the translocated form on SDS-PAGE gels due to core glycosylation of this protein in the ER lumen. The Western blot shown in Fig. 5 Breveals significant amounts of ppCPY in the temperature-sensitivesec61 mutant (lane 5), which is known to exhibit a defect in protein translocation (36Stirling C.J. Rothblatt J. Hosobuchi M. Deshaies R. Schekman R. Mol. Biol. Cell. 1992; 3: 129-142Crossref PubMed Scopus (277) Google Scholar). In contrast, mutant strain HFY405 (Δspc3)/pHF336 (spc3–4) exhibited ppCPY levels that were similar to the levels present in cells containing wild-type SPC3 (compare lanes 1 and2). A distinct form of CPY (designated CPY*) was present in the spc3–4 mutant at the nonpermissive temperature (lane 1). CPY* comigrated with a form of CPY found in thesec11 mutant (lane 3). In addition, CPY* migrated similarly to a form of CPY found in the sec23 mutant (lane 4). The sec11 (8Bohni P.C. Deshaies R.J. Schekman R.W. J. Cell Biol. 1988; 106: 1035-1042Crossref PubMed Scopus (105) Google Scholar) and sec23 (37Novick P. Field C. Schekman R. Cell. 1980; 21: 205-215Abstract Full Text PDF PubMed Scopus (1256) Google Scholar) mutations inhibit signal peptide cleavage and protein transport to the Golgi apparatus, respectively. CPY*, therefore, probably represents aberrant glycosylation of ppCPY resulting from a defect in signal peptidase activity. We have previously demonstrated that chimeric membrane protein AHDK2 is fragmented in a Sec11p-dependent manner and that the fragmentation of AHDK2 by signal peptidase is necessary for its proteolytic elimination (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar). AHDK2 is anchored to the ER membrane by a transmembrane segment from arginine permease. Fused to the luminal side of this transmembrane segment is a fragment of histidinol dehydrogenase (HD), which is followed by the Kar2p moiety. The presence of Kar2p at the C terminus of AHDK2 permits detection of proteolytic intermediate f1, which is produced through a cleavage near the transmembrane segment of AHDK2. To determine whether Spc3p is important for cleavage at the f1 site of AHDK2, expression plasmid pAHDK2 (URA3) (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar) was introduced into strain HFY405 (Δspc3)/pHF336 (CEN6 spc3–4 TRP1). This strain and control strain HFY405 (Δspc3) bearing plasmids pAHDK2 and pHF332 (CEN6 SPC3 TRP1) were grown to log phase at 23 °C, shifted to 37 °C for 30 min, then analyzed by pulse-chase using anti-HD antibodies. As expected, cells containing the wild-typeSPC3 gene displayed AHDK2 and proteolytic fragment f1 during the pulse-chase analysis (Fig. 6, lanes 1–3). In contrast, no apparent f1 was produced in spc3mutant cells (lanes 4–6). The sec11–7 mutation displays a genetic interaction, termed synthetic lethality, with the Δspc1 and Δspc2 mutations (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). Specifically, the growth of cells containing the sec11mutation and either the spc1 or spc2 null mutation is inhibited more severely than the growth of cells containing these mutations individually. We therefore asked whether thesec11–7 mutation is synthetically lethal with thespc3–4 mutation. To avoid the problem of loss of plasmid pHF336 (spc3–4) in the analysis of yeast tetrads, the chromosomal SPC3 gene locus was replaced by thespc3–4 mutation (see "Experimental Procedures"). Analysis of a spc3 sec11 double mutant was initiated by crossing strains HFY406 (spc3–4) and CMYD1 (sec11–7). These mutations were followed in the analysis of tetrads by the ability of the spc3–4 andsec11–7 mutations to inhibit cell growth at 37 and 32 °C, respectively. The distribution of progeny in 18 tetrads that were examined at 23 °C was as follows: 14 tetratype (onespc3 mutant spore, one sec11 mutant spore, one wild-type spore, and one inviable spore), three nonparental ditype (two wild-type and two inviable spores), and one parental ditype (twosec11 and two spc3 mutant spores). An additional three tetrads containing only two viable spores were produced. Phenotypic analysis of these spores was inconsistent with the nonparental ditype. Therefore, some of the spores in these tetrads probably succumbed to random spore death. The data derived from this analysis demonstrated that the spc3–4 sec11–7 double mutant was inviable at 23 °C, temperature permissive for thespc3–4 and sec11–7 single mutants. We next questioned whether the spc3–4 mutation was synthetically lethal with the Δspc1 and Δspc2mutations. Random spore analysis of diploid cells derived from a cross between strains HFY407 (spc3–4) and HFY401 (Δspc1::TRP1) produced viable spores that were both tryptophan prototrophs and temperature-sensitive (frequency of 4/14). This indicated the spc3–4 Δspc1 double mutant grew at 23 °C. The double mutant also grew well at 32 °C, indicating no apparent genetic interaction between thespc3–4 and Δspc1 mutations under the tested conditions. In contrast, analysis of diploid cells derived from strains HFY407 (spc3–4) and CMY195 (Δspc2::URA3) yielded progeny with the following distribution: 11 tetratype (one Ura+, one Ura−, one ts/Ura−, and one inviable spore), two nonparental ditype (two Ura− and two inviable spores), and two parental ditype (two ts/Ura−, and two Ura+ spores). Random spore death probably led to one tetrad containing one Ura+, one Ura− and two inviable spores and a second tetrad containing 2 ts/Ura−, one Ura+, and one inviable spore. Importantly, the absence of any spore from this dissection that was both ts and Ura+indicates that the spc3–4 Δspc2 double mutant was inviable at 23 °C. These results thus make a distinction between Spc1p and Spc2p, both of which are nonessential for signal peptidase activity (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). The Δspc2 mutation but not the Δspc1 mutation is synthetically lethal with the spc3–4 mutation. In addition, these data suggest a difference may exist between Sec11p and Spc3p, both of which are essential for signal peptidase activity. Thesec11–7 mutation is synthetically lethal with the Δspc1 and Δspc2 mutations, whereas thespc3–4 mutation is synthetically lethal with Δspc2 but not Δspc1. One interpretation of these results is that Spc3p may physically interact with Spc2p but not Spc1p; however, it is also plausible that allele-specific mutations in the SPC3 gene may be found that synthetically interact with the Δspc1 mutation. Sec11p has been shown previously to function directly in the cleavage of signal peptides and the fragmentation of abnormal membrane proteins in the yeast Saccharomyces cerevisiae (4Mullins C. Lu Y. Campbell A. Fang H. Green N. J. Biol. Chem. 1995; 270: 17139-17147Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 8Bohni P.C. Deshaies R.J. Schekman R.W. J. Cell Biol. 1988; 106: 1035-1042Crossref PubMed Scopus (105) Google Scholar). Results described here indicate that Spc3p is also required for these enzymatic activities. Moreover, Spc3p and Sec11p interact genetically in that overexpression of Spc3p suppresses the sec11–7mutation, and the spc3–4 and sec11–7 mutations are synthetically lethal. Importantly, the fact that the conditionalspc3–4 mutation inhibits signal peptidase activity after a brief shift to its nonpermissive temperature supports the notion that Spc3p also directly participates in the signal peptide cleavage reaction. In contrast to Spc3p and Sec11p, Spc1p and Spc2p are nonessential for enzyme activity (6Fang H. Panzner S. Mullins C. Hartmann E. Green N. J. Biol. Chem. 1996; 271: 16460-16465Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 7Mullins C. Meyer H.-A. Hartmann E. Green N. Fang H. J. Biol. Chem. 1996; 271: 29094-29099Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). The mammalian homologs of Spc1p and Spc2p contain two transmembrane segments, an orientation that positions the majority of their sequences on the opposite side of the ER membrane from the active site of signal peptidase (12Kalies K.-U. Hartmann E. J. Biol. Chem. 1996; 271: 3925-3929Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). Conversely, the mammalian homologs of Sec11p and Spc3p are type II single spanning membrane proteins containing a large luminal domain (33Shelness G.S. Lin L. Nicchitta C.V. J. Biol. Chem. 1993; 268: 5201-5208Abstract Full Text PDF PubMed Google Scholar). Since Spc3p and Sec11p function directly in the signal peptide cleavage reaction, it seems likely that the SPC subunits predominately exposed to the lumen comprise a two-subunit core enzyme that is required for signal peptidase activity. This conclusion is supported by the fact that a two-subunit avian signal peptidase containing homologs to Spc3p and Sec11p has been shown to function in the signal peptide cleavage reaction in vitro (16Baker R.K. Lively M.O. Biochemistry. 1987; 26: 8561-8567Crossref PubMed Scopus (38) Google Scholar). The conservation of the signal peptide cleavage reaction throughout evolution is revealed by the fact that signal sequences are cleaved correctly when expressed in either eukaryotic or prokaryotic cells. Despite this fact, signal peptides are cleaved by a single polypeptide chain, leader peptidase, in E. coli (13Zwizinski C. Wickner W. J. Biol. Chem. 1980; 255: 7973-7977Abstract Full Text PDF PubMed Google Scholar). While the overall similarity between Sec11p and leader peptidase is rather weak, it has been noted that the periplasmic domain of leader peptidase contains three distinct regions of homology to the proposed luminal domain of Sec11p (termed Box I, II, and III) (14Dalbey R.E. von Heijne G. Trends Biochem. Sci. 1992; 17: 474-478Abstract Full Text PDF PubMed Scopus (177) Google Scholar, 15van Dijl J.M. de Jong A. Vehmaanpera J. Venema G. Bron S. EMBO J. 1992; 11: 2819-2828Crossref PubMed Scopus (131) Google Scholar). Box I contains a serine amino acid found in all proteins of the signal peptidase I family (Fig.7 A). This residue is the only serine in leader peptidase required for enzyme activity (39Sung M. Dalbey R.E. J. Biol. Chem. 1992; 267: 13154-13159Abstract Full Text PDF PubMed Google Scholar). Box II contains a lysine residue that is essential for leader peptidase function (40Tschantz W.R. Sung M. Delgado-Partin V.M. Dalbey R.E. J. Biol. Chem. 1993; 268: 27349-27354Abstract Full Text PDF PubMed Google Scholar) although the corresponding amino acid in Sec11p is histidine. These serine and lysine residues are proposed to constitue a catalytic dyad within leader peptidase, which is similar to the active site of LexA-type proteases (14Dalbey R.E. von Heijne G. Trends Biochem. Sci. 1992; 17: 474-478Abstract Full Text PDF PubMed Scopus (177) Google Scholar, 38Black M.T. Munn J.G.R. Allsop A.E. Biochem. J. 1992; 282: 539-543Crossref PubMed Scopus (60) Google Scholar, 40Tschantz W.R. Sung M. Delgado-Partin V.M. Dalbey R.E. J. Biol. Chem. 1993; 268: 27349-27354Abstract Full Text PDF PubMed Google Scholar). Only four amino acids in Box III of leader peptidase are present in Sec11p (39Sung M. Dalbey R.E. J. Biol. Chem. 1992; 267: 13154-13159Abstract Full Text PDF PubMed Google Scholar, 40Tschantz W.R. Sung M. Delgado-Partin V.M. Dalbey R.E. J. Biol. Chem. 1993; 268: 27349-27354Abstract Full Text PDF PubMed Google Scholar). Considering that Spc3p and Sec11p are nonhomologous to each other, we suggest that these subunits function together in a manner analogous to leader peptidase. We have performed a comparison of leader peptidase and Spc3p sequences, using the computer-based algorithm described previously (41Wilbur W.J. Lipman D.J. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 726-730Crossref PubMed Scopus (1052) Google Scholar, 42Pearson W.R. Lipman D.J. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 2444-244860Crossref PubMed Scopus (9363) Google Scholar). As shown in Fig. 7 B, this analysis reveals a similarity between the luminal domain of Spc3p and the carboxyl-terminal half of the periplasmic domain of leader peptidase. The Box I and Box II amino acids are located in the amino-terminal half of leader peptidase, thus suggesting the alignment of Sec11p and Spc3p sequences to distinct regions of leader peptidase. Moreover, the number of amino acids present in the luminal domain of leader peptidase (∼250 amino acids), which contains the catalytic site, is similar to the number of amino acids in the proposed luminal domains of both Sec11p (∼135 amino acids) and Spc3p (∼150 amino acids). These correlations raise the intriguing possibility that Sec11p and Spc3p evolved from a common ancestor. However, in light of the modest homologies between these proteins, we cannot exclude the possibility that leader peptidase and the SPC utilize fundamentally distinct mechanisms for cleaving signal peptides. Future experiments to test these two models are in progress. We thank Dean Ballard (Vanderbilt University) for critical reading of this manuscript and Randy Sheckman (University of California, Berkeley) for the gift of anti-CPY antibodies.

Referência(s)