Evolution of GPI-Aspartyl Proteinases (Yapsines) of Candida spp

The Candida genus is a polyphyletic genus with at least 150 species. Nine are recognized opportunistic pathogens of humans and animals. C. albicans is the species most frequently isolated from human infections, followed by Candida non-Candida species (CNCA), as C. glabrata, C. tropicalis, C. dubliniensis, C. parapsilosis, C. guilliermondii, C. lusitaniae, C. kefyr and C. krusei (Mean et al. 2008; Pfaller & Diekema, 2007; Almirante et al. 2005; Manzano-Gayosso et al. 2000). Some works describe the phylogenetic relationships of Candida genus and illustrate the limited relationship between the pathogenic Candida spp. The genus has been divided into: the CTG clade, which includes yeast that encodes CTG as serine instead of leucine (C. albicans, C. dubliniensis, C. tropicalis, C. parapsilosis and C. lusitaniae); and the WGD clade, which includes yeast that has undergone a genome duplication event (Saccharomyces spp., Kluyveromyces spp. and C. glabrata). Evidently, C. glabrata is more related to non-pathogenic yeasts, as Saccharomyces cerevisiae, than to the other pathogenic species (Scannell et al. 2007). C. albicans is a normal microorganism in humans, and colonise up to 70% of skin, mucoses, and faeces of individuals with no apparent detriment to health. However, in some circumstances, either through environmental factors or a weakening of the host immune system, a proliferation and infection by C. albicans arise inducing candidosis (Wei et al. 2011). Biofilm formation, adhesion, cavitation, phenotypic switching, dimorphism, interaction with the host immune system, invasion and tissue damage are virulence virulence factors for C. albicans. All these factors are related to the secreted aspartyl proteases (Sap) family, which is considered an important virulence factor and is studied as a possible target for therapeutic drug design (Naglik et al. 2004; Chaffin et al. 1998; Hube, 1998; Naglik et al. 2003, 2004, 2008). The topic of this chapter is to understand the molecular characteristics, evolution and putative functions of glycosylphosphatidylinositol (GPI)-linked aspartyl proteases (Yps), a protein superfamily distributed among all pathogenic Candida species. Cell location motifs,


Introduction
The Candida genus is a polyphyletic genus with at least 150 species.Nine are recognized opportunistic pathogens of humans and animals.C. albicans is the species most frequently isolated from human infections, followed by Candida non-Candida species (CNCA), as C. glabrata, C. tropicalis, C. dubliniensis, C. parapsilosis, C. guilliermondii, C. lusitaniae, C. kefyr and C. krusei (Méan et al. 2008;Pfaller & Diekema, 2007;Almirante et al. 2005;Manzano-Gayosso et al. 2000).Some works describe the phylogenetic relationships of Candida genus and illustrate the limited relationship between the pathogenic Candida spp.The genus has been divided into: the CTG clade, which includes yeast that encodes CTG as serine instead of leucine (C. albicans, C. dubliniensis, C. tropicalis, C. parapsilosis and C. lusitaniae); and the WGD clade, which includes yeast that has undergone a genome duplication event (Saccharomyces spp., Kluyveromyces spp.and C. glabrata).Evidently, C. glabrata is more related to non-pathogenic yeasts, as Saccharomyces cerevisiae, than to the other pathogenic species (Scannell et al. 2007).C. albicans is a normal microorganism in humans, and colonise up to 70% of skin, mucoses, and faeces of individuals with no apparent detriment to health.However, in some circumstances, either through environmental factors or a weakening of the host immune system, a proliferation and infection by C. albicans arise inducing candidosis (Wei et al. 2011).Biofilm formation, adhesion, cavitation, phenotypic switching, dimorphism, interaction with the host immune system, invasion and tissue damage are virulence virulence factors for C. albicans.All these factors are related to the secreted aspartyl proteases (Sap) family, which is considered an important virulence factor and is studied as a possible target for therapeutic drug design (Naglik et al. 2004;Chaffin et al. 1998;Hube, 1998;Naglik et al. 2003Naglik et al. , 2004Naglik et al. , 2008)).The topic of this chapter is to understand the molecular characteristics, evolution and putative functions of glycosylphosphatidylinositol (GPI)-linked aspartyl proteases (Yps), a protein superfamily distributed among all pathogenic Candida species.Cell location motifs, The phylogenetic analyses were performed in the MEGA4 program (Tamura et al. 2007) using Maximun Parsimony evolution.A similitude and identity matrix were computed with the MatGAT4.50.2 software (Campanella et al. 2003).The phylogenetic reconstruction and similarity of PrA reproduce the phylogenetic tree topologies of Candida spp.obtained with other genes, suggesting a common ancestral gene (Fig. 2   C. glabrata is an opportunistic haploid yeast that suffered evident and extensive reductive evolutionary events.A lot of genes involved in nitrogen metabolism, carbohydrate assimilation (saccharose, galactose, etc.), as well as sulfur, phosphor, thiamine, pyridoxine and nicotinic acid biosynthesis have been lost from the genome (Byrne & Wolfe, 2005;Wolfe, 2006).This species produces between 15-20% of reported systemic yeast infections (Almirante et al. 2005;Manzano-Gayosso et al. 2000;Trick et al. 2002;Méan et al. 2008).C. glabrata is the most common yeast species isolated from patients with cancer, organ transplantation and fluconazole therapy (Safdar et al. 2001;Bodey et al. 2002).The mortality associated with C. glabrata in systemic infections of cancer patients is 50% and almost 100% in transplant patients (Anaissie et al. 1992;Goodman et al. 1992;Krcmery et al. 1998).This scenario is related to indiscriminate antifungal use, and to the innate resistance of C. glabrata (Sobel, 2006).
According to The transcription profile of 11 CgYPS was studied when yeast were ingested by macrophages.Apparently, CgYPS are important in survival and virulence of the the yeast in macrophages, damage to mousses, Epa1 protein processing, and cell wall integrity, as occur in S. cerevisiae, which possesses 5 ScYPS (ScYPS1-ScYPS3, ScYPS6 and ScYPS7) (Kaur et al. 2007;Krysan et al. 2005).They are important to cell wall synthesis and glucan homeostasis, mainly ScYPS1 and ScYPS7.It seems that ScYPS3 does not have functions associated with the cell wall (Krysan et al. 2005).
C. albicans SAP9 and C. glabrata CgYPS1 genes complement the defects in the cell wall provoked by yps1 of S. cerevisiae.One important difference is that SAP9 complement yps1 only when SAP9 is under a heterologous and constitutive promoter from S. cerevisiae, while CgYPS1 complements the mutation, using its promoter (Krysan et al. 2005), evidence that supports the othologous status proposed above for these gene pairs.As happened with ScYPS1, SAP9 gene expression increases during the stationary phase and damage of the cell wall (Monod et al. 1998;Copping et al. 2005), and protects the yeast from caspofungin (an inhibitor of 1,3-glucan synthesis) (Lesage et al. 2004).Also, inhibitors of ScYps1p disable the specificity of both proteins, ScYps1p and Sap9 (Cawley et al. 2003).Distribution of the SAP gene superfamily among C. albicans strains is universal (Gilfillan et al. 1998;Bautista et al. 2003;Parra et al. 2009), although one study concludes that the distribution of SAP genes in clinical strains depends on infection associated with isolation (Kalkanci et al. 2005).Given the number of CgYPS in C. glabrata and their potential role in pathogenesis, it is important to establish the universality of CgYPS in C. glabrata populations.2007b).

Factor
Our group explored the CgYPS gene distribution among clinical isolates (n=52) and type strains CBS138 and BG6 (N=2) by an original multiplex PCR procedure (Table 5).The yeasts were routinely grown on YPD broth and DNA was extracted using a previously reported protocol (Hoffman & Winston 1987).PCR was performed in a DNA thermal cycler 9600 (Applied Biosystems, Foster City, CA).Amplification reactions (25 μL) were performed using a buffer containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 2 mM MgCl 2 , 0.2 mM of each deoxynucleoside triphosphate, 0.6 μM each primers, 4 ng/μL of genomic DNA, and 1.5 U/μL of Taq polymerase (Invitrogen).The PCR conditions included a denaturation step for 3 min at 94°C, followed by 38 amplification cycles consisting of 1 min at 94°C, 1 min annealing temperature and 1 min at 72°C.A final extension step was performed for 7 min at 72°C.Fig. 3 shows the amplification products of CgYPS gene fragments of some representative C. glabrata clinical strains electrophoresed in 1% agarose gels.Similar PCR conditions were used to study the universal distribution of PrA.

Gene Primer Location
Expected amplified fragment (bp) Table 5. Primer pairs used for conventional multiplex PCR of C. glabrata YPS genes.Bp, Base pair; Tm, Melting temperature.
The universality of the 12 CgYPS genes among all C. glabrata clinical isolates and type strains was confirmed (Fig. 3), which suggests that all CgYPS are important to yeast life cycle as pathogen or commensal, and probably are differentially regulated according to each environmental condition, as occurs with C. albicans SAP.  6).Prediction of motif sequences was performed with PROSITE (http://www.expasy.org)(Falquet et al. 2002).Some of the proteins possess a typical molecular structure of aspartyl proteases, but others have some differences in composition (Fig. 4; Table 6).Some of them possess high Ser/Thr content in the amino terminal, suggesting that this zone is exposed at the surface of the protein.The presence of Ser/Thr in the carboxyl terminal in almost all YPS is postulated to be heavily O-glycosylated.The exact function of this Ser/Thr-rich domain in yapsins has not been investigated.However, O-mannosylation is important for proper cellwall biogenesis and integrity.It has also been proposed that clustered O-glycans create rigid stalks that keep protein domains away from membranes or wall surfaces (Lipke & Ovalle, 1998).The presence of a GPI attachment site, a characteristic feature of the yapsin family, was determined with big-PI predictor (http://mendel.imp.univie.ac.at/gpi/gpi_server.html),and GPI-SOM.GPI-anchor signals were identified by a Kohonen Self Organizing Map (http://gpi.unibe.ch/).A total of 36 protein sequences were analyzed, but GPI sites were recognized only in 21 proteins.GPI sites were not detected in ScYps2 and CgYps2, although both proteins have been previously confirmed as Yps proteins.The software programs must be enhanced, but an experimental approach to confirm the cell location is necessary.PSORTII (http://www.psort.org/)and Softberry (http://www.softberry.com)programs were used to predict subcellular localization.All proteins detected seem to be extracellular, which could be because of the presence of a signal peptide in the amino terminal extreme.Nevertheless during their synthesis, yapsins are cotranslocated and modified by the addition of GPI to the lumen of the endoplasmic reticulum (ER).Then proteins are glycosylated in Golgi apparatus, associated to membrane vesicles and sent to plasma membrane or the cell wall (Mayor & Rieaman, 2004;Caro et al. 1997).Softberry program was also used to find exons, which were absent in all genes studied.A search was made for internal protein sequence repeats to detect possible internal duplication events, but none were detected by TRUST (Szklarczyk & Heringa, 2004) even though it is likely were not detected by TRUST (Szklarczyk & Heringa, 2004) even when it is likely that the Yps and Sap superfamilies have duplicated aspartyl protease motifs.The analysis of possible evolutive and molecular events that has given place to the presence of different numbers of YPS in each pathogenic Candida species was made to establish the COGs between Yps.Phylogenetic analysis was performed by an alignment of YPS homologues identified in silico and those of the previously characterized.The alignment was carried out using MUSCLE in SeaView 2.4 program (Galtier et al. 1996) with default alignment parameter adjustments.The phylogenetic analyses were performed in the MEGA4 program (Tamura et al. 2007) using minimum evolution computed with the Poisson correction.A similitude and identity matrix were computed with the MatGAT4.50.2 software (Campanella et al. 2003).To corroborate support for the branches on trees, bootstrap analysis (1,000 replicates) was performed.Synteny analysis was made to recognize the putative COGs (Fig. 5).CgYPS1, CgYPS7, ScYPS1, ScYPS3, CaYPS7, CaSAP98, CaSAP99 and BAR1 are GPI anchored aspartyl proteases; APC2 and CAGL0M04235g, subunit of the anaphase-promoting; APT1, acylprotein thioesterase; ATP22, mitochondrial inner membrane protein; CAGL0M04147g, similar to low affinity vacuolar membrane, is a localized monovalent cation/H+ antiporter protein; CAGL0M04169g, similar to cell wall glycoprotein involved in beta-glucan assembly; CDH1, cell-cycle regulated activator of the anaphase-promoting complex/cyclosome (APC/C); CLF1, crooked neck-like factor; DOP1, protein essential for viability; EKL1, ethanolamine kinase; FAF1, protein required for pre-rRNA processing and 40S ribosomal subunit assembly; HAT, histone acetyltransferase; HXT3, low affinity glucose transporter of the major facilitator superfamily; LDG3 and LDG4, leucine, aspartic acid, glycine rich; MNN42, putative positive regulator of mannosylphosphate transferase; MNT3, alpha-1,3mannosyltransferase; MTQ2, S-adenosylmethionine-dependent methyltransferase; MRP1, mitochondrial ribosomal protein of the small subunit; NOP16, constituent of 66S preribosomal particles; NTA1, amidase; orf19.2088,shared subunit of DNA polymerase (II) epsilon and of ISW2/yCHRAC chromatin accessibility complex; PDR11, ATP-binding cassette transporter, PEX7, peroxisomal signal receptor; PFK27, 6-phosphofructo-2-kinase; PHHB, transposon mutation affects filamentous growth; PMI40, mannose-6-phosphate isomerase; POM152, nuclear pore membrane glycoprotein; RPL2B, protein component of the large ribosomal subunit; SAN1, ubiquitin-protein-ligase; SBE2, protein involved in the transport of cell wall components from the Golgi to the cell surface; SEC1, Sm-like protein involved in docking and fusion of exocytic vesicles through binding to assembled SNARE complexes at the membrane; SNL1, putative protein involved in nuclear pore complex biogenesis and maintenance; SRN2, component of the ESCRT-I complex; SVF1, protein with a potential role in cell survival pathways; SW15, transcription factor that activates transcription of genes expressed at the M/G1 phase boundary and in G1 phase; TAF12, subunit (61/68 kDa) of TFIID and SAGA complexes; TIR3, cell wall mannoprotein of the Srp1p/Tip1p family of serine-alanine-rich proteins; TMA20, protein associated with ribosomes with a putative RNA binding domain; UGA11: gamma-aminobutyrate transaminase (4-aminobutyrate aminotransferase); VID28, protein involved in proteasomedependent catabolite degradation of fructose-1,6-bisphosphatase (FBPase); v-SNARE, component of the vacuolar SNARE complex involved in vesicle fusion; YCF1, putative glutathione S-conjugate transporter; YLR126C, protein with similarity to glutamine amidotransferase proteins; YMD8, putative nucleotide sugar transporter; ORF, APM2, BSC6, CAGL0M04125g, Cd36_72050, Cd36_72080, FM02, IFK2, orf19.6482, RTC1, tRNA-Glu, YDR352W, YDR348C and YLR125W and ORF, unknown predicted open reading frame.

Amino
The lack of SAP genes and the expansion of 12 CgYPS genes in C. glabrata, and the extended family of SAP genes in C. albicans support the hypothesis that both protein superfamilies are an example of convergent evolution.Although more research is necessary to reach definite conclusions, apparently YPS of C. glabrata and SAP of C. albicans have developed some equivalent physiological functions and roles in virulence.The rest of pathogenic Candida species are less virulent, and, curiously, harbour less genes in their genomes than C. albicans.These facts lead to the supposition that SAP and YPS have evolved in an independent way for at least 700 million years.However, more SAP duplication events have happened in C. albicans (Parra et al. 2009).Phylogenetic analyses of Yps deduced protein sequences of Candida spp.and S. cerevisiae allow for the definition of 8 Yps families, A-H (Fig. 5).In particular, CgYps1-12 proteins of C. glabrata were clustered in four families.Family A was constituted exclusively of nine Yps of C. glabrata  encoded in chromosome E.With exception of CgYps2, all codifying genes of these proteins are organized in tandem, and possibly derived from at least eight recent duplication events that occurred exclusively in the C. glabrata genome.Apparently these recent duplications led to the emergence of a paralogous gene family with novel or slightly different functions.No pseudogenes were detected in CgYPS1-11 genes, but in their deduced proteins a moderate amino acid similitude (48-53%) and identity (36-38%) were retained.Frequently, very high similitudes are maintained by concerted evolution in paralogous members of some multigene families (László, 1999).However, in CgYPS genes, this evolutive phenomenon is not evident.Previously, CgYPS4 and CgYPS11 were recognized as GPI anchored aspartyl proteases (Kaur et al., 2007), but comparative studies of the regulatory region and expression of each CgYPS genes are necessary to clearly define the physiological role and orthology relationships of each gene.Family B was formed by a set of Yps proteins, detected exclusively in S. cerevisiae (ScYps2-3 and ScYps6), and a highly similar putative orthologous pair (ScYPS1/CgYPS1) (Fig. 6A).Also, the partial synteny observed between the ScYPS2/CgYPS2 gene pair supports the hypothesis that those protein-coding genes are probable orthologous (Fig. 6B).Family C was integrated by CgYps12 and ScBar1 of S. cerevisiae, a putative orthologous pair with low similitude synteny, but with a clear ancestor-descendant relationship (Fig. 6G).Finally, family E was formed by a representative of each Candida spp.Yps, CgYps7 and ScYps7.This family forms a sub tree with the same topology as those phylogenies constructed with ribosomal and other protein sequences (Diezman et al., 2004).The CgYPS7 and ScYPS7 genes exhibited an extensive synteny (Fig. 6C), but no synteny with CaYPS (orf19.6481)and CdYPS (Cd36_72090) was observed (Fig. 6D).In C. albicans and C. dubliniensis genome databases these YPS are described as ScYPS7 orthologous genes (Schaefer et al., 2007).Nevertheless, both YPS exhibited low similarity with ScYPS7 (37.2-38.7%)and no-synteny.The final decision to consider family E as an orthologous family will depend on comparative analyses of functional features not yet performed.Families C, F, G and H have not any C. glabrata or S. cerevisiae Yps representative protein.
Families C and H were formed only by one ClYps gene of C. lusitaniae and seven CpYps genes of C. parapsilosis, respectively (Fig. 5).Curiously, C. lusitaniae is the species that harbours the fewest ClYPS (n=1) and SAP (n=3) genes, and its isolation frequency from clinical samples ,as well as its virulence, are lower than the other Candida species (Abi-Said et al. 1997).This evidence supports a hypothesis of relevance of aspartyl proteases in virulence.That is, species with numerous aspartyl proteases in virulence; species with broad aspartyl proteases are more virulent than those with a limited number of these proteins.Family F harboured C. albicans, C. dubliniensis and C. tropicalis yapsins organized congruently according to the ribosomal phylogenetic tree.The C. albicans CaBar1 (orf19.2082)and C. dubliniensis CdBar1 (Cd36_15430) gene, found in family F, has been described as orthologous to S. cerevisiae BAR1 (Schaefer et al., 2007) found in family C.In both species, C. albicans and S. cerevisiae, the protein is involved in alpha pheromone degradation and secreted to the periplasmic space of mating alpha-type cells.These proteins help cells find mating partners by cleaving and inactivating the alpha factor, which allows cells to recover from alpha-factor-induced cell cycle arrest (Mackay et al., 1988).The in silico analysis performed in this work established that these proteins and the Bar1 from C. dubliniensis are extracellular, but anchored to the cell wall or cell membrane.Also, phylogenetic analysis shows that Bar1 from C. albicans and C. dubliniensis belongs to the Yps superfamily, with a similarity of 40%, and are not grouped with CgYps12 of C. glabrata (CgYps12 or CgBar1) and Bar1 of S. cerevisiae.The reason for which an aspartyl protease, that apparently is secreted, is groupedwith the yapsines superfamily could be a mistake in the cell location method because almost all software use the signal peptide, transmembranal regions, and the GPI site in the C-terminal, to predict the cell location.In C. albicans it has been detected that aspartyl proteases are associated with the plasmatic membrane, or to both the plasmatic membrane and cell wall.This makes the experimental corroboration of the cell location necessary.The Bar1 protein of C. albicans has been described as a protein with three domains: 2 aspartyl protease domains and another unidentified.Apparently, this GPI-membrane anchored domain determines that Bar proteins are not secreted, but anchored to cellular membranes, and their two actives sites are oriented to cellular membranes, and their two actives sites are oriented to the exterior to inactivate alpha pheromone, which is secreted by Mat-alpha cells.In C. albicans, the degradation of secreted alpha pheromone is not exclusive to Bar1.CaYPS7 (orf19.6481) of family E also encodes for this function with lesser efficiency (Schaefer et al., 2007).This physiological redundancy has not been demonstrated in S. cerevisiae ScYps7.C. albicans can mate under some in vitro and in vivo conditions when alpha pheromone is degraded (Hull et al., 2000;Magee & Magee, 2000) and C. glabrata harbours homologous genes of S. cerevisiae that control the mating (Srikantha et al., 2003).Nevertheless, in C. glabrata a cell cycle has not been demonstrated, and the participation of CgYps7 of C. glabrata in alpha pheromone inactivation has not been demonstrated.No possible gene orthologous to possible gene orthologous to ScBar1 was detected in C. guilliermondii, C. lusitaniae, C. parapsilosis, C. tropicalis, C. guilliermondii or C. lusitaniae.All these yeasts have a heterothallic sex cycle (cross-mating only), but C. parapsilosis and C. tropicalis mating has never been observed (Butler et al., 2009).Family G is formed by two C. albicans/C.dubliniensis Yps protein pairs with high similitude (>88%), located in tandem in chromosome 2 and with very similar synteny.All this data is evidence from the recent speciation of both species (Fig. 6E).According to the Candida genome database (http://www.candidagenome.org/cgi-bin/locus.pl?locus=orf19.852)Cal orf19.852and Cdu Cd36_18370 sequences are described as CaSAP98 and CdSAP98 genes, respectively, and have their best hits with PEP4 of S. cerevisiae (Pra protein).S. cerevisiae PrA is a vacuolar protease, and clearly C. albicans/C.dubliniensis Yps are not phylogenetically grouped with PrA.In our opinion no orthology relationship among these proteins exists.Cal orf19.853and Cdu Cd36_18360 formed a second pair, described as CaSAP99 and CdSAP99 genes, which had their best hits with ScYPS3 of S. Cerevisiae.Similarly, it is clear that CaSAP99 has no synteny, phylogenetic relationship, or possible common physiological role with ScYPS3.

Conclusion
Why have C. albicans/C.dubliniensis and C. glabrata/S.cerevisiae been suffering some genetic duplication events in their Sap and Yps superfamilies?This is something that has not been resolved, but it is clear that the decrease in virulence in null mutants, in both CaSAP and CgYPS, endorse the idea that the presence and expansion of SAP and YPS families is necessary for adaptation to the host, and therefore for survival and virulence.Also, species with broad aspartyl protease families are more virulent than those with a limited number of these proteins.C. glabrata belongs to a phylogenetic group with no pathogenic yeast, and its virulence attributes could be evolving independently from the CTG clade, where C. albicans is the main opportunistic pathogenic species.The expansion of the CgYPS gene superfamily of C. glabrata maintains a parallelism with the expansion of the SAP gene superfamily of C. albicans, and constitutes a possible example of convergent evolution.The transition from a commensally life style to a successful opportunistic pathogen could be related to gene expansion that encodes for each kind of aspartyl protease.A lot of experimental methodologies must be performed to recognize the orthologous gene families, as well as the virulence, participation and transition commensal-pathogen roles of aspartyl proteases, including Sap and Yps.

Acknowledgment
We are grateful for the financial support from CONACyT-CB-13695, CONACyT-69984, SIP201005214 and SIP20113066.BPO is a fellow of CONACyT and PIFI-IPN.Thanks to Dr. Bernard Dujon (Institut Pasteur and Université Pierre et Marie Curie) for donating strains.Thanks also to Bruce Allan Larsen for reviewing the use of English.www.intechopen.com

Table 1 .
; Table 2).In brief, C. albicans was more related to C. dubliniensis, followed by C. tropicalis, C. parapsilosis, C. guilliermondi and C. lusitaniae.Meanwhile, C. glabrata PrA was more related to S. cerevisiae PrA than other Candida species.Soluble and membrane-bound * vacuolar proteolytic system of S. cerevisiae. C.

Table 2 .
Similarity and identity (UP/down) between PrA proteins from pathogenic Candida spp.

Table 3
. Vacuolar aspartyl proteases in pathogenic Candida species.(AN): Access number in the respective genome; MM: molecular mass; IP: Isoelectric Point; C: Chromosome or Contig or supercontig.
Table 4, virulence factors of C. albicans and C. glabrata are quite different.However, an evident feature is the difference in number and kind of aspartyl proteases.A total of 12 YPS genes, but no SAP genes have been detected in C. glabrata.Contrarily, a total of 10 SAP genes, but no YPS genes have been recognized in C. albicans.Clearly, the phylogenetic trees constructed with ribosomal or other gene groups include the majority of the clinical relevant Candida species, with exception of C. glabrata, which is grouped in another cluster with non-pathogenic yeasts, as S. cerevisiae and Kluyveromyces spp.This evidence suggests that the aspartyl proteases in Candida spp.have evolved independently as virulence factors at least two times, and possibly the amplification by duplication of SAP

Table 4 .
Comparison of virulence factors of C. glabrata and C. albicans (modified from Li,

Table 6
. Aspartyl proteases GPI-linked to cell membrane in pathogenic Candida spp.AN: Access number in the respective genome; MM: molecular mass in kilodaltons (kDa); IP: Isoelectric Point; C: Chromosome/Contig or supercontig; the atypical amino acids in the PROSITE motif are shown in black (Eukaryotic and viral aspartyl protease active site).