Protein translocation through the endoplasmic reticulum (ER) membrane is a complex phenomenon involving a number of related activities tightly coordinated with previous and subsequent events, providing a link between protein synthesis and maturation (Johnson and van Waes, 1999; Schnell and Hebert, 2003). The translocation process, which can occur either co-translationally or post-translationally (Kalies and Hartmann, 1998), is triggered by N-terminal signal sequences (or signal peptides) that direct precursors of secretory and membrane proteins to the translocation sites (or translocons). The central component of a translocon is the Sec61 complex, which consists of three subunits: an α-subunit, which spans the membrane 10 times and appears to form the channel walls, and β- and γ-subunits, which each span the membrane a single time and play an as-yet undetermined role in the translocation. In yeast, the co-translational translocation takes place through two different types of translocons that diverge in their α-subunits (Sec61p and Ssh1p) and β-subunits (Sbh1p and Sbh2p, respectively), but share the same γ-subunit (Sss1p) (Deshaies and Schekman, 1987; Panzner et al., 1995; Finke et al., 1996). Besides, post-translational translocation in yeast occurs by means of another protein complex in which the Sec61p complex is also involved (Panzner et al., 1995).
The signal peptide of most secretory and many membrane proteins is cleaved by the signal peptidase, a protein complex with endoproteolytic activity located close to the translocation channel at the lumenal side of the ER membrane (Johnson and van Waes, 1999). In yeast, the signal peptidase complex (SPC) consists of four polypeptides (YaDeau et al., 1991). Two of them, Spc1p and Spc2p, are non-essential for growth or signal-peptide cleavage activity and are homologues of the mammalian signal peptidase subunits SPC12 and SPC25, respectively (Fang et al., 1996; Mullins et al., 1996). The other two components, Spc3p and Sec11p, are essential for cell viability and the depletion of any one of them leads to the accumulation of unprocessed precursors of secretory proteins (Böhni et al., 1988; Meyer and Hartmann, 1997). Moreover, Spc3p and Sec11p interact functionally, since the overexpression of Spc3p suppresses a sec11 mutation, and spc3 and sec11 mutations are synthetically lethal (Fang et al., 1997). However, Sec11p has been shown to be the only catalytic subunit of the Saccharomyces cerevisiae SPC (YaDeau et al., 1991; VanValkenburgh et al., 1999) and is functionally a homologue of the two catalytic subunits of the mammalian signal peptidase, SPC18 and SPC25 (Liang et al., 2003). The essential character of Spc3p (the yeast homologue of the mammalian SPC subunit SPC22/23) seems to be that it is required to stabilize Sec11p (Meyer and Hartmann, 1997; VanValkenburgh et al., 1999).
Not much attention has been paid to the secretory process of proteins in the clinically relevant yeast Candida albicans, despite the fact that enzymes secreted into the extracellular medium and cellular surface components are some of the factors that seem to be related to the ability of this fungal pathogen to colonize, survive within and infect the human host (Chaffin et al., 1998; Monod et al., 2002). To contribute to the knowledge of the elements that participate in the first stages of the secretory process in C. albicans, in this study we report the isolation and characterization of the C. albicans orthologue of the SPC3 gene that encodes one of the two essential subunits of the signal peptidase complex in S. cerevisiae.
Materials and methods
Microorganisms and growth conditions
Candida albicans, Saccharomyces cerevisiae and Escherichia coli strains used in this study are listed in Table 1. E. coli cells were grown at 37 °C in LB medium (1% tryptone, 0.5% yeast extract, 1% NaCl), supplemented with ampicillin at a final concentration of 100 µ g/ml when necessary for plasmid selection. Yeast strains were routinely grown at 28 °C in YPD medium (2% peptone, 1% yeast extract, 2% glucose) or in SD medium (0.67% yeast nitrogen base without amino acids, 2% glucose) with appropriate supplements as required. Solid media were supplemented with 2% agar. Uridine (50 µ g/ml) was added to the media for C. albicans Ura− strains. Selection of C. albicans uracil auxotrophs was done by growing on medium containing 1 mg/ml 5′-fluoroorotic acid (5′-FOA), as previously described (Boeke et al., 1984; Fonzi and Irwin, 1993). When needed, S. cerevisiae cells were grown at the restrictive temperature of 32 °C, 37 °C or 38.5 °C to verify the complementation of the thermosensitive phenotypes of sec11, spc3 or sec61 mutants, respectively.
Preparation of DNA, restriction digestions, agarose-gel electrophoresis, and cloning of DNA fragments were done as previously described (Sambrook et al., 1989). Polymerase chain reaction (PCR) was carried out in a Perkin-Elmer Cetus DNA 9600 thermal cycler. The conditions of amplification varied with the primers used.
S. cerevisiae and C. albicans transformations were carried out using the lithium acetate method (Gietz et al., 1992). Alternatively, the one-step method of Chen et al. (1992) was also used for S. cerevisiae transformations.
A C. albicans genomic DNA library constructed on plasmid p1041 (Goshorn et al., 1992) was used for the transformation of the sec61-2 thermosensitive mutant S. cerevisiae CSY100. After 24 h at 28 °C, the temperature was shifted to 38.5 °C to select clones able to grow at this temperature (Böhni et al., 1988).
DNA sequencing and homology analysis
The nucleotide sequence of a 2005 bp NsiI fragment of C. albicans genomic DNA, including the SPC3 coding region, was determined on both strands by the dideoxy-chain termination method, with the modified T7 polymerase (Amersham) and either universal sequencing primers or custom-synthesized oligonucleotide primers. Sequence analyses were carried out with the Gene Runner 3.0 program (Hasting Software, Inc., 1994). Homology searches of GenBank and EMBL data bases were done using the FASTA algorithm (Pearson and Lipman, 1988). Multiple-sequence alignments were performed with CLUSTAL W 1.6 (Thompson et al., 1994). Transmembrane domains were identified using TMpred software (http://www.chembnet.org/software/TMPRED_form.html).
Southern blot analysis
C. albicans genomic DNA was prepared by the method of Hoffman and Winston (1987). Samples of genomic DNA (5 µ g approximately) were digested with restriction enzymes and the resulting fragments were transferred onto a positively-charged nylon membrane (Boehringer-Mannheim) by capillarity using the alkaline transfer method (Chomczynski, 1992), after separation of fragments by electrophoresis in 0.8% agarose gels in TAE buffer. Southern hybridizations were carried out using the Non-Radioactive Labeling and Detection Kit (Boehringer-Mannheim) according to the manufacturer's instruction. The Southern probe for SPC3 consisted of a DIG-labelled PCR product obtained by amplification of C. albicans genomic DNA with primers Sp1 and Sp2 (see below).
Vectors, plasmids and strain constructions
Complementation analyses of spc3 and sec61 mutations were carried out, respectively, with plasmid pHF331, a multicopy vector derived from pRS34 by insertion of S. cerevisiae SPC3 gene (Fang et al., 1997) and plasmid pCS23, a single-copy, URA3-ScSEC61 plasmid derived from YCp50, generously provided by C. Stirling (University of Manchester, UK).
Plasmid pCA47 was obtained from a clone isolated during screening for suppressors of the sec61-2 mutation in the CSY100 strain. A 2 kb NsiI fragment, containing the whole CaSPC3 gene, was cloned in the PstI site of the multicopy vector YEp352 to create the plasmid pASPC3. A 1.3 kb HindIII fragment from the 5′ region of the CaSPC3 ORF, immediately upstream from the translation initiation codon, was replaced in pASPC3 by a 410 bp HindIII fragment from the yeast integrating vector YIp128A1, including the promoter of the S. cerevisiae ADH1 gene, to generate the plasmid pPAASPC3.
The ‘ura-blaster’ procedure of Fonzi and Irwin (1993) was followed to attempt to construct a C. albicans spc3 null mutant. A fragment from the 5′ region of the CaSPC3 ORF, including nucleotides in positions −658 to +57 (considering position 1 to be that of the adenine of the first ATG codon), was PCR-amplified from C. albicans genomic DNA using primers Sp1 (5′-GCATGCTTCTGTGCGTAGTTTAG) and Sp2 (5′-GTCGACGGATGAAGTAAATGCCA), designed to introduce restriction sites (underlined) for endonucleases SphI and SalI, respectively. The resulting PCR product was cloned into the pGEM®-T Easy vector (Promega). Then, the SphI–SalI fragment was released and directionally ligated into the corresponding sites of plasmid pMB7 (generously provided by W. Fonzi, Georgetown University, Washington, DC, USA) to produce plasmid p5′DelCaSPC3. Following sequence verification, a second PCR product, including nucleotides in positions +522 to +1157 at the 3′ region of the CaSPC3 ORF, was amplified as above using primers Sp3 (5′-AGATCTCGGTGAATTCCAATTTG) and Sp4 (5′-GAGCTCCTTATGGTTTCAAATCG) to introduce restriction sites (underlined) for BglII and SacI, cloned into the pGEM®-T Easy vector, digested with the same endonucleases and subcloned into the corresponding sites of p5′DelCaSPC3 to generate plasmid pDelCaSPC3. Digestion of pDelSPC3 with SphI and SacI released a 5.3 kb SPC3 construction, flanked by the sequences corresponding to primers Sp1 and Sp4, in which an internal 463 bp portion of the SPC3 coding region, encompassing nucleotides +58 to +521, was replaced by the hisG–URA3–hisG cassette. This 5.3 kb fragment was used to try to sequentially disrupt both CaSPC3 alleles in strain CAI4 by following the previously described procedure (Fonzi and Irwin, 1993), except that transformation was performed by the lithium acetate method. Alternatively, after disruption of the first CaSPC3 allele, a Ura− clone (JMCA2) was recovered by growing cells of a previously selected uridine prototroph (JMCA1) in the presence of 5′-FOA and used to place the second copy of CaSPC3 under the control of the methionine-repressible MET3-promoter, by following the procedure of Care et al. (1999). A portion of the 5′ coding region of CaSPC3, including nucleotides −10 to +378, was PCR-amplified using primers Sp5 (5′-GGATCCAAGCTTCAACAATG) and Sp6 (5′-CTGCAGCATCTTCTTTATTAG) to introduce, respectively, the restriction sites (underlined) BamHI, on the 5′-side, and PstI, on the 3′ side, and then cloned in the pGEM®-T Easy vector. The 399 bp cloned fragment was released by double digestion with BamHI and PstI and then subcloned into the same restriction sites of the pCaDIS vector (Care et al., 1999), immediately after the MET3 promoter. The resulting plasmid, pDISCaSPC3, was linearized at the unique XcmI site within the CaSPC3 sequence to target the integration of this fragment at the chromosomal locus of CaSPC3, after transformation of the JMCA2 strain. The genotype of the selected hemizygous Ura+ transformant (JMCA3), carrying the MET3p-driven CaSPC3 gene, was verified by Southern blot analysis.
SDS–PAGE and Western blot analysis
Total yeast protein extracts were prepared using glass beads, according to a standard procedure (Ausubel et al., 1995). Quantified proteins (as determined using the BCA Protein Determination Kit; Pierce, Rockford, IL) were separated by SDS–polyacrylamide gel electrophoresis (SDS–PAGE) in 9% polyacrylamide gels, following the method of Laemmli (1970), and transferred onto nitrocellulose membranes (Millipore) as described by Towbin et al. (1979). The membranes were blocked for 1 h in TRIS-buffered saline containing 0.1% Tween-20 (TBST) and 10% non-fat milk. The blocked membranes were washed three times in TBST and incubated for 1 h in TBST containing either anti-carboxypeptidase Y (CPY) antiserum (1 : 7500) or anti-Kar2p antiserum (1 : 10 000). After TBST washes, the membranes were incubated with anti-rabbit IgG coupled to horseradish peroxidase (1 : 3000) and, finally, antibody binding was visualized on X-ray film using ECL (Amersham).
Monitoring of invertase activity
Invertase activity was assayed in whole cells according to the procedure of Goldstein and Lampen (1975). One unit of invertase is the amount of enzyme that hydrolyses sucrose to yield 1 µ mol glucose/min at 30 °C and pH 4.5. To derepress invertase synthesis, cells from pre-cultures grown overnight in SD with 4% glucose were harvested by centrifugation, washed with sterile distilled water at 28 °C, and used to inoculate (at an initial OD600 of 0.3) 50 ml of the same medium with 0.3% glucose as carbon source and cultivated under the same conditions. After 1 h, half of the culture was shifted to 38.5 °C and samples were collected at intervals, harvested by centrifugation, washed twice with distilled water and resuspended in acetate buffer (0.1 M, pH 4.5). Aliquots were used to determine cell density before assaying periplasmic invertase activity.
Nucleotide sequence Accession No
The nucleotide sequence data reported in this study have been submitted to the EMBL Nucleotide Sequence Database under Accession No. AJ417660.
Results and discussion
Detection and cloning of the C. albicans SPC3 gene
The CaSPC3 gene was unexpectedly identified when sequencing the C. albicans DNA inserted in one of the clones selected in a screening of a genomic library, carried out with the aim of searching for genes encoding products involved in protein translocation across the C. albicans ER membrane. The selected clone (pCA47) was able to complement the thermosensitive (ts) phenotype of the sec61-2 mutant strain S. cerevisiae CSY100. Since mutations in the SEC61 gene cause the accumulation in the cytoplasm of precursors of secretory proteins (Deshaies and Schekman, 1987; Stirling et al., 1992), the isolated transformant was also analysed to check whether, besides the restoration of the ts phenotype, the protein secretion was unblocked. For this purpose, two secretory proteins were chosen: periplasmic invertase and vacuolar carboxypeptidase Y (CPY). As can be observed in the time-course experiment shown in Figure 1A, at the restrictive temperature the transformant cells quickly reached invertase activity levels much higher than those exhibited by the mutant strain and similar to those of the mutant strain transformed with a SEC61 gene-bearing plasmid.
On the other hand, Western blot analysis of cell extracts using anti-CPY polyclonal antibodies showed that the secretion of CPY was partially unblocked in the transformant at the restrictive temperature. In comparison with the mutant strain, there was a lower accumulation of precursors in the transformant shown by a decrease in the intensity of the band corresponding to prepro- and pre-CPY, which correlated with an increase in the intensity of the band corresponding to the glycosylated forms (p1 and p2) of the enzyme (Figure 1B). Similar results were obtained by using anti-invertase antiserum in Western blots of extracts from yeast cells grown under conditions which allowed synthesis of this enzyme (data not shown).
A 2 kbp NsiI fragment from pCA47, which retained the ability to complement the sec61 mutation, was subcloned into the PstI site of the multicopy vector YEp352 to generate the pASPC3 plasmid and then sequenced on both strands. The sequence analysis revealed a complete open reading frame (ORF), encompassing nucleotides 1285–1863, which could encode a polypeptide of 192 amino acid residues with a deduced molecular weight of 21.8 kDa. This ORF was identical to orf6.5910 of the C. albicans genome sequence assembly 6 available at CandidaDB (http://www.pasteur.fr/Galar_Fungail/CandidaDB/) developed by the Galar Fungail European Consortium. Analysis of the encoded polypeptide with the PSORT II program (http://psort.nibb.ac.jp/) predicted a type II integral membrane protein, probably located at the ER with its N-terminus inside the cytoplasm. The predicted protein showed significant homology with the proteins encoded by S. cerevisiae SPC3 (45% identity) and human SPC22/23 (37% identity) genes (Figure 2A). In consequence, the cloned gene was designated CaSPC3.
Despite the limited identity in the amino acid sequences of C. albicans and S. cerevisiae Spc3 proteins, the overall great structural similarity between the two proteins was shown when their hydropathy profiles were compared (Figure 2B). The single hydrophobic stretch of amino acids identified near the amino-terminus of CaSpc3p was similar to that found in an equivalent position in the S. cerevisiae Spc3p, which has been proposed to span the ER membrane (Fang et al., 1997; Meyer and Hartmann, 1997).
Complementation analysis of spc3, sec11 and sec61 mutants by CaSPC3
That the cloned gene was the true orthologue in C. albicans of SPC3 was confirmed by the fact that the pASPC3 plasmid, carrying the CaSPC3 gene, complemented the thermosensitive growth of the spc3 mutant strain HFY407 as efficiently as the S. cerevisiae SPC3 gene (Figure 3 A). An apparent partial inhibitory effect on growth at 28 °C was observed in transformants carrying either S. cerevisiae or C. albicans SPC3 gene, when compared to the untransformed spc3 mutant or to the transformant carrying the ScSEC61 gene. This effect could be a consequence of an excess of Spc3p in the transformant cells, due to overexpression of either one of the SPC3 orthologues included in multicopy vectors, at the permissive temperature. Additionally, as a further confirmation that the product encoded by CaSPC3 was genuinely active as an essential component of signal peptidase, a Western blot analysis of cell extracts was performed using antibodies directed against the ER-resident protein encoded by the KAR2 gene (Rose et al., 1989). The signal-peptide-containing precursor of Kar2p accumulates in the S. cerevisiae spc3 mutant when it is incubated at restrictive temperature (37 °C), while the presence of the wild-type SPC3 gene in the cells allowed efficient cleavage of preKar2p (Fang et al., 1997). Figure 4 shows that the accumulation of unprocessed preKar2p at the restrictive temperature was also unblocked in the transformant carrying the CaSPC3, in a similar fashion to what was observed in transformants bearing the S. cerevisiae gene.
It has been previously reported that in S. cerevisiae the SPC3 gene, when cloned into a high-copy-number plasmid, is able to suppress a mutational defect in the Sec11p, the catalytic subunit of the signal peptidase complex (Fang et al., 1997). However, this complementation does not work reciprocally, since the overexpression of Sec11p is not able to rescue Δspc3 cells (Meyer and Hartmann, 1997). To test whether the C. albicans orthologue of SPC3 was also able to rescue the ts phenotype of a S. cerevisiae sec11 mutant, the multicopy plasmid pASPC3 was introduced by transformation in strain CYMD1. As shown in Figure 3B, the selected transformant was unable to grow at the restrictive temperature of 32 °C. This lack of complementation could be easily explained if the expression level of CaSPC3 in the transformants were much lower than that of the S. cerevisiae orthologue, and a greater amount of CaSpc3p would then be required to complement the sec11 mutation. However, the fact that CaSPC3 was also unable to complement when its expression occurred under control of the S. cerevisiae ADH1 gene promoter seems to rule out that possibility. The ability of Spc3p to suppress the sec11 temperature-sensitive phenotype in S. cerevisiae has been interpreted as an indirect evidence of a kind of interaction between Spc3p and Sec11p in the ER membrane, in which Spc3p could act as a stabilizing agent, providing a backbone for Sec11p to fold correctly (Meyer and Hartmann, 1997; VanValkenburgh et al., 1999). Conversely, it would also be possible to speculate that the lack of complementation of the S. cerevisiae sec11 mutant by CaSPC3 could be the result of the inability of CaSpc3p to interact with ScSec11p, due to differences in the amino acid sequences between the C. albicans and S. cerevisiae SPC3 gene products, which might change the structure of the Spc3p domain(s) involved in the interaction with Sec11p.
A similar kind of protein interaction could explain why CaSPC3 was able to rescue the ts phenotype of a S. cerevisiae sec61 mutant to a similar extent to the ScSEC61 gene (Figure 3C). Furthermore, the S. cerevisiae SPC3 was also able to complement the sec61 mutation, although not as efficiently as the C. albicans orthologue. It is possible, however, that the apparent poorer growth of transformants carrying the ScSPC3 gene could be the result of an overproduction of Spc3p at 38.5 °C that could be partially inhibitory for growth, in a similar way to what was previously discussed for the phenomenon observed in Figure 3A. A more reduced level of expression under the same conditions would explain the optimum growth of transformants carrying CaSPC3. However, this hypothesis would be in contradiction with that pointed out to explain the apparent inhibition of growth of the spc3 mutant transformed with any of the SPC3 orthologues, since in that case the supposed inhibitory overproduction of Spc3p would occur at 28 °C, and not at 38.5 °C as in the transformants of the sec61 mutant. Unfortunately, none of these hypotheses could be experimentally tested, since a method for direct measuring of Spc3p levels was not available.
Similarly to what has been previously described for the absence of reciprocity of SEC11 and SPC3 to complement each other, there was not a reciprocal complementation of the spc3 mutant by the SEC61 gene (Figure 3 A, lower row). These results suggest that even though the products of SEC61 and SPC3 genes are not mutually interchangeable to support cell viability, the Spc3p could be involved in the functional interaction, at the ER membrane, between the translocon and the signal peptidase complex. Physical and functional interactions between these two ER-membrane-associated protein complexes have been previously reported in mammals, where the interaction between the signal peptidase subunit SPC25 with the Sec61 complex, via the Sec61β subunit, has been proved by cross-linking experiments (Kalies et al., 1998; Nilsson et al., 2002; Abell et al., 2003). Similarly, in yeast cells the interaction between Spc2p with Sbh1p and Sbh2p (the β-subunits of the Sec61p and Ssh1p complexes, respectively) has been proposed to form the interface for the association between the SPC and the Sec61 complexes (Antonin et al., 2000). It is then possible that the complementation of the sec61 mutant by SPC3 is the result of either a direct interaction between Spc3p and Sec61p, or of an indirect action of the former on the latter mediated by Spc2p, in the SPC, and Sbh1p, in the translocon.
CaSPC3 is required for cell growth
The S. cerevisiae SPC3 gene has been shown to be essential not only for signal peptidase activity but also for cell viability (Fang et al., 1997; Meyer and Hartmann, 1997). To test whether CaSPC3 was also an essential gene, attempts were done to obtain a C. albicans strain depleted of Spc3p. First, heterozygous disruption of the CaSPC3 gene was accomplished in the Ura− strain CAI4 by the ‘ura-blaster’ procedure of Fonzi and Irwin (1993), through the almost-complete deletion of one of the CaSPC3 alleles and replacement with a hisG–URA3–hisG cassette (Figure 5). The resulting uridine prototroph strain, JMCA1, was grown in the presence of 5′-fluoroorotic acid to recover the uracil auxotrophy after excision of the URA3 gene. Attempts to delete the remaining wild-type CaSPC3 allele in the resulting heterozygous strain (JMCA2) by transformation with the same ura-blaster cassette always gave negative results, suggesting the essential character of CaSPC3 for viability. To prove that cell growth depended on CaSPC3, the promoter region of the wild-type CaSPC3 allele remaining in strain JMCA2 was replaced by the methionine-repressible MET3-promoter, following the procedure of Care et al. (1999). The presence of cysteine and methionine at equal final concentrations of 0.5 or 1.5 mM in the culture medium completely blocked growth of the resulting strain JMCA3, while had no detectable effect on growth of the heterozygous strain JMCA2 (Figure 6). The clear effect of the repression of the MET3-promoter on growth, together with the failure to obtain a viable double-deletion spc3 mutant through the ura-blaster procedure, strongly suggests that CaSPC3 is essential for cell viability like its S. cerevisiae counterpart.
We thank Hong Fang (Vanderbilt University, USA), William Fonzi (Georgetown University, USA), Colin Stirling (University of Manchester, UK) and Peter Sudbery (University of Sheffield, UK) for generous gifts of strains and plasmids. We also thank Dieter H. Wolf (University of Stuttgart, Germany) and Mark Rose (Princeton University, USA) for anti-CPY and anti-Kar2p antibodies, respectively. This work was partially financed by Grant PI2001/041 from Dirección General de Universidades e Investigación (Gobierno de Canarias), and by predoctoral fellowships from the Ministerio de Educación Cultura y Deporte to J.M.R. and F.G.