the ess1 prolyl isomerase is required for growth …the ess1 prolyl-isomerase might be a useful...

Copyright 2002 by the Genetics Society of America

The Ess1 Prolyl Isomerase Is Required for Growth and MorphogeneticSwitching in Candida albicans

Gina Devasahayam,*,‡ Vishnu Chaturvedi*,†,‡ and Steven D. Hanes*,‡,1

*Molecular Genetics Program, †Mycology Laboratory, Wadsworth Center, New York State Department of Health, Albany, New York 12208and ‡Department of Biomedical Sciences, School of Public Health, State University of New York, Albany, New York 12208

Manuscript received July 26, 2001Accepted for publication October 8, 2001

ABSTRACTProlyl-isomerases (PPIases) are found in all organisms and are important for the folding and activity

of many proteins. Of the 13 PPIases in Saccharomyces cerevisiae only Ess1, a parvulin-class PPIase, is essentialfor growth. Ess1 is required to complete mitosis, and Ess1 and its mammalian homolog, Pin1, interactdirectly with RNA polymerase II. Here, we isolate the ESS1 gene from the pathogenic fungus Candidaalbicans and show that it is functionally homologous to the S. cerevisiae ESS1. We generate conditional-lethal (ts) alleles of C. albicans ESS1 and use these mutations to demonstrate that ESS1 is essential for growthin C. albicans. We also show that reducing the dosage or activity of ESS1 blocks morphogenetic switchingfrom the yeast to the hyphal and pseudohyphal forms under certain conditions. Analysis of double mutantsof ESS1 and TUP1 or CPH1, two genes known to be involved in morphogenetic switching, suggests thatESS1 functions in the same pathway as CPH1 and upstream of or in parallel to TUP1. Given that switchingis important for virulence of C. albicans, inhibitors of Ess1 might be useful as antifungal agents.

PEPTIDYL-PROLYL cis-/trans-isomerases (PPIases) tial for growth in the budding yeast Saccharomyces cerevis-are enzymes that catalyze the cis-/trans-isomerization iae (Dolinski et al. 1997). However, the single parvulin-

of the peptide bond preceding the amino acid proline. class PPIase in S. cerevisiae, Ess1, is essential (HanesDuring translation, the ribosome is thought to synthe- 1988; Hanes et al. 1989). Yeast cells depleted of Ess1size all peptide bonds in the trans-isomer form. However, arrest late in mitosis and undergo nuclear fragmenta-structural studies show that �10% of peptidyl-prolyl tion (Lu et al. 1996; Wu et al. 2000). Homologs of thebonds are in the cis-isomer form (Stewart et al. 1990). ESS1 gene have been isolated from a variety of organ-Although isomerization of the X-Pro bonds from trans- isms, including Aspergillus nidulans (Crenshaw et al.to cis- can occur spontaneously in nascent proteins, it 1998), Neurospora crassa (Kops et al. 1998), Drosophilais thought to be facilitated by the action of PPIases. melanogaster (Maleszka et al. 1996), Xenopus laevis (Wink-PPIases also bind to mature proteins in different cellular ler et al. 2000), and humans (Lu et al. 1996). The flycompartments to control their activity, subunit assembly, homolog dodo and the human homolog PIN1 both com-and transport (Rutherford and Zuker 1994; Hunter plement ess1 mutants in budding yeast (Lu et al. 1996;1998; Gothel and Marahiel 1999). Maleszka et al. 1996), indicating a highly conserved mech-

There are three families of PPIases: cyclophilins, anism of action. However, dodo is not essential in flies,FK506-binding proteins (FKBPs), and parvulins; mem- nor is PIN1 essential in mice, suggesting the presencebers of each family are conserved in both prokaryotes of redundant or alternative pathways in metazoansand eukaryotes (Dolinski and Heitman 1997). Cyclo- (Maleszka et al. 1996, 1997; Fujimori et al. 1999). Inphilins and FKBPs are best known because they bind Xenopus extracts, Pin1 has been shown to be importantthe immunosuppressive drugs cyclosporin A and FK506, for the DNA replication checkpoint (Winkler et al. 2000).respectively (Kunz and Hall 1993; Fruman et al. 1994).

Among the eukaryotic parvulins, only the Ess1 homo-FKBPs also bind rapamycin, and all three compounds

logs have an amino-terminal WW domain, in additionexhibit antifungal activity (Hemenway and Heitmanto their carboxy-terminal PPIase domain. The WW do-1993). Parvulins are structurally distinct from both themain is a protein-protein interaction module that recog-cyclophilins and FKBPs and do not bind these com-nizes proline-rich sequences (Einbond and Sudol 1996;pounds (Rahfeld et al. 1994a,b).Sudol 1996). The WW domains of Pin1 and Ess1 bindNone of the eight cyclophilins or four FKBPs is essen-with high affinity to phospho-serine-proline motifs (Yaffeet al. 1997; Lu et al. 1999a). Human Pin1 has been shownto bind mitotic phosphoproteins that contain this motif

1Corresponding author: Wadsworth Center, New York State Depart- (Shen et al. 1998) and to hyperphosphorylated tau pro-ment of Health, 120 New Scotland Ave., Albany, NY 12208.E-mail: [email protected] tein, which is a component of fibroid tangles found

Genetics 160: 37–48 ( January 2002)

38 G. Devasahayam, V. Chaturvedi and S. D. Hanes

MATERIALS AND METHODSin the brains of Alzheimer’s disease patients (Lu et al.1999b). Media and culture conditions: C. albicans strain SC5314 was

In yeast, genetic studies have shown that Ess1 is impor- grown in rich medium (YEPD) and the ura3� derivative strainCAI4 was grown in YEPD � 25 �g/ml uridine. Filamentation-tant for transcription by RNA polymerase II (RNA polII)inducing media were prepared as described: Spider mediumand that the cell-cycle defect in ess1 mutants is probably(Liu et al. 1994), liquid Lee’s medium (Lee et al. 1975), solidan indirect effect of misregulation of genes requiredmodified Lee’s (Liu et al. 1994), and milk-tween agar and

for mitosis (Wu et al. 2000). Biochemical and structural cornmeal-tween agar ( Jitsurong et al. 1993). For serum-con-studies have shown that Ess1 and Pin1 bind directly to taining media, the liquid contained YEPD � 10% fetal bovine

serum (FBS; Sigma, St. Louis), and the solid contained 4%the carboxy-terminal domain (CTD) of RNA pol II,FBS and 2% agar. For growth rate experiments, cells werewhich contains multiple phospho-Ser-Pro motifs, andgrown to midlog phase and were used to inoculate culturesto components of the Sin3/Rpd3 histone deacetylaseto a starting OD600 of 0.1. For plating efficiency experiments,

complex (Morris et al. 1999; Arevalo-Rodrıguez et al. cells were grown in liquid YEPD at 30� to midlog phase prior2000; Verdecia et al. 2000; Wu et al. 2000). Our current to plating on solid YEPD medium. C. albicans cells were made

competent and transformed by a standard lithium acetatemodel is that Ess1 isomerizes the CTD and regulatesprocedure (Ito et al. 1983).the sequential binding of proteins to the RNA pol II

Cloning of C. albicans ESS1 by complementation: Tempera-complex that are required for transcription initiation,ture-sensitive S. cerevisiae strain ess1L94P was transformed with a

elongation, termination, and mRNA processing (Wu et C. albicans genomic DNA library and ura� prototrophs wereal. 2000). In addition, Ess1 negatively regulates the Sin3/ selected. The library was made in a high-copy episomal plas-

mid, YEp352 (2�, URA3; Navarro-Garcia et al. 1995). TheRpd3 histone deacetylase complex (Arevalo-Rodrıgueztransformed cells were incubated at 30� overnight in liquidet al. 2000).synthetic complete medium lacking uracil before being platedTo further understand the function of Ess1 and toand incubated at 37� for 4–5 days. Of an estimated 2 � 106

assess its potential as an antifungal drug target, we are transformants plated, 28 grew repeatedly at 37� upon sequen-studying homologs from human fungal pathogens, in- tial passaging. Plasmids were rescued from all 28 transformants

and retransformed into S. cerevisiaie ess1L94P. Of the 28, 5 werecluding C. albicans. C. albicans exists as either a commen-again able to rescue the no-growth phenotype of ess1L94P atsal or an opportunistic pathogen in humans and causes37�. These five were also 5-fluoroorotic acid (5-FOA) sensitive,life-threatening systemic infections in immunocomprom-suggesting that the complementation was plasmid linked. On

ised individuals (Odds 1988; Kao et al. 2000). Systemic the basis of restriction mapping and size of insert, the clonesC. albicans infections are also a serious problem for were divided into two classes. The first class (insert size �3.5

kb) contained RPB7 (GenBank accession no. AF224269; Wupatients undergoing cancer chemotherapy or organ trans-et al. 2000) and the second class (insert size �8 kb) containedplantation (Anaissie et al. 1998). C. albicans is dimorphicESS1 (GenBank accession no. AF224270), with the putativein that it undergoes morphogenetic switching between start codon being 255 bp from one end of the insert. A 1-kb

a yeast form and a hyphal (and pseudohyphal) form fragment containing ESS1 was subcloned, and the resulting(Odds 1988; Brown and Gow 1999). This switching plasmid (pGD-CaESS1) also complemented the S. cerevisiae

ess1L94P mutant at 37�.is important for virulence (Odds 1988; Cutler 1991;Plasmid construction:pGD-CaESS1 was constructed in pRS426Madhani and Fink 1998) and is initiated by external

(Sikorski and Hieter 1989) by insertion of an EcoRI-BamHIsignals that are transduced to transcriptional regulatory fragment from the original library clone containing ESS1. Theproteins, such as Cph1 and Tup1 that activate or repress EcoRI site is in the polylinker of YEp352 and BamHI is 200 bpdownstream target genes, respectively (Liu et al. 1994; downstream of the stop codon. pGD-CaRPB7 was also con-

structed in pRS426, except that the 1-kb fragment containingBraun and Johnson 1997). Given the role of Ess1 inRPB7 was amplified from the original library clone using prim-transcription and its requirement for growth in buddingers with EcoRI and BamHI sites. A construct to delete ESS1

yeast, we examined its importance for the hyphal transi- from C. albicans, pGD-Caess1�, was constructed in pUC19 intion in C. albicans. several steps. First, the 255-bp region upstream of ESS1 was

amplified using primers with KpnI and BamHI sites. This frag-Here, we describe the isolation of the C. albicans homo-ment was cloned into KpnI and BamHI sites of pUC19 to givelog of ESS1 and demonstrate that it is essential forpGD-1. Next, the �4.4-kb hisG -CaURA3-hisG cassette (BglII-growth in this organism. This was done by traditionalBamHI) from pCUB-6 (Fonzi and Irwin 1993) was cloned

gene knockout experiments and by generating a C. albi- into the BamHI site of pGD-1 to give pGD-2. Finally, the �2.3-cans strain bearing a conditional-lethal ess1ts mutation. kb BamHI fragment that begins 200 bp downstream of the

ESS1 stop codon was inserted in the correct orientation intoWork described here suggests a method by which condi-the BamHI site of pGD-2 to generate pGD-Caess1�. Mutanttional alleles of C. albicans genes can be generated usingalleles of C. albicans ESS1 were generated by site-directed PCRS. cerevisiae as a surrogate host and that such alleles can mutagenesis (Horton et al. 1990): CAT � AGA for H171R

be used to prove essentiality and to study gene function and TCA � CCA for S129P. The PCR fragments containingin C. albicans. We also show that C. albicans ESS1 is the mutant alleles were cloned as EcoRI-BamHI fragments into

pRS426, yielding pGD-ess1H171R and pGD-ess1S129P. Replacementimportant for morphogenetic switching. Our resultsconstructs pGD-3 and pGD-4, designed to introduce the tssuggest that some, but not all, of the functions of Ess1alleles into C. albicans, were similar to the deletion construct,in budding yeast are conserved in C. albicans and that pGD-Caess1�, except that ESS1 sequences were derived from

the Ess1 prolyl-isomerase might be a useful target for pGD-ess1 H171R and pGD-ess1S129P, the mutant alleles being theleft flanking region.the development of antifungal drugs.

39A PPIase Required for Filamentous Growth in Candida

Strain construction: To generate a heterozygous ess1�::hisG-URA3-hisG/ESS1 (CaGD1) mutant strain, a �7-kb insert (SacI-SphI) was excised from pGD-Caess1� and used to transformC. albicans CAI4 and uracil prototrophs were selected. Toselect for ura3� derivatives (ess1�::hisG/ESS1), in which theURA3 gene is looped out by recombination between the hisGrepeats, the mutant strain CaGD1 was grown in rich medium(YEPD) for 1–2 days, plated on 5-FOA-containing medium,and incubated at 30�. The resulting strain was CaGD2. Toattempt to generate homozygous ess1 deletion mutants, themutant strain CaGD2 was transformed again with SacI-SphIdigested pGD-Caess1�.

Temperature sensitivity of the C. albicans mutant allelesess1H171R and ess1S129P was tested using S. cerevisiae ess1� strains.For ess1H171R, diploid strain YSH-55 (ess1�::HIS3/ESS1) wastransformed with pGD-ess1H171R, cells were induced to sporu-late, and tetrads were dissected. Segregants that were HIS�

and URA� were tested for temperature sensitivity by replicaplating to 30� and 37�. For ess1S129P, haploid strain YXW 2.1(ess1�::TRP1) containing human PIN1 on a 2-�m LEU2 plas-mid was transformed with pGD-ess1S129P and grown in the ab-sence of selection for the PIN1 plasmid in liquid medium Figure 1.—Cloning of the ESS1 gene from C. albicans bycontaining leucine for 2–3 days at 25�. A total of 0.2% of complementation. Complementation and suppression of S.colonies analyzed lost PIN1. They were tested for temperature cerevisiae ess1L94P by C. albicans ESS1 and RPB7, respectively, aresensitivity by replica plating at 25� and 37�, and all were found shown at the restrictive temperature (37�). ESS1 and RPB7to be ts. To construct C. albicans ess1ts strains, the replacement were expressed from subclones pGD-CaESS1 and pGD-CaRPB7,constructs pGD-3 and pGD-4 were digested with SacI-XhoI and respectively. Control plasmids were YEpESS1 (Hanes et al.transformed into the heterozygous ess1�::hisG/ESS1 mutant 1989) or an empty vector pRS426 (Sikorski and Hieter(CaGD2) and uracil prototrophs were selected. They were 1989). Cells were streaked onto synthetic complete mediumtested for temperature sensitivity by replica plating at 25�, lacking uracil and incubated at 30� or 37� for 3 days.30�, 37�, and 42�. Multiple isolates of the heterozygous strain(CaGD2) were used to generate the ess1�::hisG/ess1H171R:hisG-URA3-hisG (CaGD3) mutant strains.

considered filamentous and included cells with both hyphaeCPH1 was disrupted in three strain backgrounds: CAI4, twoand pseudohyphae. To view colony morphology, a Zeiss SV6isolates of CaGD2, and two isolates of CaGD4. The strains werestereo microscope was used at 20� magnification.tranformed with a XhoI-SacI digest of the disruption construct

(pHL156) containing the hisG-URA3-hisG cassette inserted atthe NarI site of the open reading frame (Liu et al. 1994). To

RESULTSgenerate homozygous CPH1 deletions, the transformation wasrepeated after reversion of the URA3 marker to ura-minus

Isolation of the C. albicans homolog of ESS1: We usedusing 5-FOA. TUP1 was also deleted in CAI4, two isolates ofthe S. cerevisiae ESS1 gene as a probe in low-stringencyCaGD2, and two isolates of CaGD4. The strains were trans-

formed with an SphI digest of the deletion construct (p383c) Southern hybridization against C. albicans genomic DNA.containing the hisG-URA3-hisG cassette that replaces the TUP1 The results suggested the existence of a single-copy ho-open reading frame and 330 bp of upstream sequence (Braun molog of ESS1 (data not shown). To clone the gene,and Johnson 1997). Homozygous TUP1 deletions were gener-

we took advantage of a temperature-sensitive strain ofated as described above for CPH1.S. cerevisiae (ess1L94P; Wu et al. 2000). We transformed aSouthern hybridization: Genomic DNA from C. albicans wasC. albicans genomic DNA library into this strain to searchprepared as described (Adams et al. 1997). High-stringency

Southern hybridization was done at 65�. The probes used were for C. albicans genes capable of complementing the no-ESS1 (a 539-bp PCR fragment containing 255 bp upstream of growth phenotype at the restrictive temperature (37�).the start codon and 284 bp of ORF sequence), hisG, and Of the 2 � 106 transformants screened, we obtainedURA3. Labeling of the probes was done by random priming

five complementing clones representing two different(Boehringer Mannheim, Indianapolis). Genescreen Plus hy-genes (Figure 1).bridization transfer membranes (New England Nuclear Life

Science, Boston) were used. Filters were stripped in boiling Three of these clones contained a high-copy suppres-1% SDS and 0.1� SSC (0.015 m sodium chloride and 0.0015 m sor called RPB7 that is the C. albicans homolog of thesodium citrate) prior to reuse in subsequent hybridizations. RNA polymerase II seventh subunit (McKune et al. 1993;Microscopy: For 4�,6-diamidino-2-phenylindole (DAPI) stain-

Khazak et al. 1995). We think that this gene suppressesing, cells were grown at 30� and 40�, sonicated, fixed in 70%by a mechanism that involves rescue of an RNA pol IIethanol, resuspended in mounting medium (1 mg/ml pheny-

lenediamine in 2� PBS/80% glycerol) containing 50 ng/ml transcription deficiency (Wu et al. 2000). The two otherDAPI (Sigma), and examined and photographed on a Nikon clones contained a gene that encodes a predicted 177-Optiphot equipped with a Quad Fluor epifluorescence attach- amino-acid protein with 42% identity to S. cerevisiae Ess1,ment and a 60� 1.4 NA Planapo objective. Filamentous cells

47% to A. nidulans Pin1, 41% to Drosophila Dodo, andwere treated with Triton X-100 (2%) and briefly sonicated43% to human Pin1 (Figure 2). The overall structureand fixed in 3.7% formaldehyde before being visualized. Cells

with morphologies other than the yeast form (rounded) were of the encoded protein is conserved with that of all


Figure 2.—Alignment ofC. albicans Ess1 (accessionno. AAK00626) with othermembers of the Ess1 familyof parvulin-class prolyl-iso-merases: S. cerevisiae Ess1 (S52-764), A. nidulans PINA (AAC-49984), D. melanogaster Dodo(P54353), and Homo sapiensPin1 (XP_00 9024). Thealignment was done usingClustalW multiple sequencealignment program, version1.7 (Thompson et al. 1994).Dashes indicate gaps. Shad-ed areas indicate regions ofidentity with the C. albicansprotein or identity amongothers. The signature tryp-tophans of the WW domainare indicated by dots. Theunderlined residues in thePPIase domain are residuesthat were changed to intro-duce the indicated tempera-ture-sensitive mutations.

known Ess1/Pin1 homologs. It has an N-terminal WW the wild type, and it therefore may have been a favoredtarget.domain and a C-terminal PPIase domain, and the four

active-site residues (H157, C113, H59, and S154 in Pin1) Since C. albicans is diploid and asexual, proving theessentiality of genes is problematic and, although thethat are predicted on the basis of the crystal structure

of Pin1 (Ranganathan et al. 1997) are conserved. The results described above suggest that ESS1 is essential,they are not conclusive. We therefore adopted an inde-C. albicans gene complemented a complete deletion of

the ESS1 gene (ess1�) in S. cerevisiae (data not shown). pendent approach. We replaced the remaining wild-type allele in a heterozygous strain with a conditionallyThese results show that the C. albicans gene is the func-

tional homolog of budding yeast ESS1. lethal allele. In this way, if transformants were generatedunder permissive conditions, replacement of the wild-The ESS1 gene is essential in C. albicans: C. albicans

is a diploid organism. Therefore, to determine whether type allele should no longer be deleterious. To generateconditional alleles we engineered substitutions in theESS1 is essential in C. albicans, we attempted to delete

both alleles by sequential gene deletion. To do this we C. albicans gene on the basis of mutations in the S.cerevisiae gene that render it temperature sensitive (Wuused the modified Ura-blaster approach (Fonzi and

Irwin 1993; Figure 3A). The first allele was deleted et al. 2000). The mutations were H171R and S129P(H164R and S122P in S. cerevisiae Ess1). They were cho-successfully in strain CAI4, as demonstrated by Southern

hybridization (Figure 4, lanes 2 and 3) and PCR (data sen because in S. cerevisiae both mutant proteins arestable at permissive and restrictive temperatures andnot shown). In the resulting heterozygous strains

(CaGD1�/�; Table 1) no obvious growth defect was allow normal growth at permissive temperature, andone of them (H171R) is in a predicted active-site residueobserved. In these strains the URA3 marker was excised,

and the resulting strains were transformed with the same (Wu et al. 2000).To show that the C. albicans ESS1 mutant alleles weredeletion construct, in an attempt to delete the second

allele. Of the 60 transformants analyzed, none showed temperature sensitive, we introduced them into an S.cerevisiae strain in which ESS1 was deleted (see materi-a deletion of the remaining wild-type allele, as deter-

mined by PCR and Southern hybridization (data not als and methods). As shown in Figure 5, ess1� strainscarrying C. albicans ess1H171R or ess1S129P showed tempera-shown). The inability to recover homozygous deletions

(�/�) suggested that ESS1 is an essential gene. How- ture-sensitive growth. Interestingly, each allele conferreda distinct temperature-sensitive profile. Both strainsever, it is also possible that deletion of the second allele

did not occur due to heterozygosity at the second locus, grew at 25� and did not grow at 37�; but the ess1S129P

strain also failed to grow at intermediate temperatureswhich might have inhibited homologous recombination(Yesland and Fonzi 2000). Also, the deletion construct (30� and 35�), indicating a lower threshold of tempera-

ture sensitivity. These results show that the C. albicansshared more homology with the deleted allele than with


Figure 4.—Southern blot analysis of genomic DNA pre-pared from isogenic C. albicans strains containing wild-type ormutant alleles of ESS1. The same blot was probed sequentially

Figure 3.—Strategy for deletion and replacement of C. with three different probes: CaESS1, CaURA3, and hisG. Allalbicans ESS1. (A) Deletion of ESS1. The deletion cassette lanes contain 1–2 �g of HindIII-digested genomic DNA fromrecombines with one of the alleles by homologous recombina- the following strains: (1) wild-type C. albicans CAI4; (2 and 3)tion and this event is identified by selecting for uracil proto- two Ura� isolates of CAI4 generated by transformation withtrophs. The deletion cassette can be reused to delete the the SacI-SphI-digested deletion construct pGD-Caess1�; (4 andremaining wild-type allele, after the URA3 gene is removed 5) 5-FOA-resistant, ura� revertants of strains equivalent tofrom the heterozygote by 5-FOA selection (Alani et al. 1987; those in lanes 2 and 3; (6–8) heterozygous strains (ess1�/Fonzi and Irwin 1993). (B) Generating an ess1ts strain. The ESS1), as shown in lanes 4 and 5 transformed with a SacI-XhoI-replacement cassette consists of the ess1ts allele as the left digested replacement construct, pGD-3 containing ess1H171R;flanking region and a 2.3-kb BamHI fragment, starting 200 and (9) FOA-resistant, ura� revertant of the ess1H171R/� strainbp downstream of the stop codon, as the right flanking region. shown in lane 7. The banding pattern from each of the threeThe desired replacement events are selected for as uracil pro- probes confirms the expected genotypes (Table 1).totrophs that are temperature sensitive.

tained for the H171R transformation, 18 were tempera-ess1ts alleles are temperature sensitive in an S. cerevisiaeture sensitive. For 3 of these, the chromosomal locushost, suggesting that they will be temperature sensitivewas amplified and the DNA sequence determined; allin C. albicans. Moreover, these results suggest that S.contained the H171R mutation. In addition, Southerncerevisiae can be used as a surrogate host to generateanalysis demonstrated that the integrations occurred atnew ts allelles in C. albicans genes or genes of otherthe correct locus and yielded the expected recombi-pathogenic fungi (see discussion).nants (Figure 4, lanes 6–8). In contrast, we were unableNext, we used the ess1S129P and ess1H171R alleles to re-to generate an ess1S129P/ess1� strain. Of 958 transformantsplace wild-type ESS1 in a C. albicans heterozygous strainobtained for the S129P transformation (at 25� and 30�),(�/�). The replacement strategy is depicted in Figurenone were temperature sensitive, suggesting that the3B. Linearized replacement constructs containing eachmutation was lethal over a deletion (ts/�).of the mutant alleles were transformed into three inde-

pendent heterozygous strains. Of 295 transformants ob- To test whether ESS1 is essential in C. albicans, growth


TABLE 1

Strains used in this study

Strain name Parent Genotype Reference

S. cerevisiaeYSH-55 W303-1A MATa/MAT ade2-1/ade2-1 can1-100/can1-100 his3-11,15/his 3-11,15 Wu et al. (2000)

leu2-3,112/LEU2 ura3-1/ura3-1 ESS1/ess1�::HIS3YXW-2.1 W303-1A MATa ura3-1 leu2-3,112 trp1-1 can1-100 ade2-1 his3-11,15 [phi�] Wu et al. (2000)

ess1�::TRP1 pTPI-PIN1 (LEU2)YGD-ts22W W303-1A MATa ura3-1 leu2-3,112 trp1-1 can1-100 ade2-1 his3-11,15 [phi�] Wu et al. (2000)

ess1-H164RYGD-ts45W W303-1A MATa ura3-1 leu2-3,112 trp1-1 can1-100 ade2-1 his3-11,15 [phi�] Wu et al. (2000)

ess1-L94P

C. albicansSC5314 Wild type Fonzi and Irwin (1993)CAI4 SC5314 ura3::imm434/ura3::imm434 Fonzi and Irwin (1993)CaGD1 CAI4 ESS1/ess1�::hisG-URA3-hisG ura3::imm434/ura3::imm434 This studyCaGD2 CaGD1 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 This studyCaGD3 CaGD2 ess1 H171R:hisG-URA3-hisG/ess1�::hisG ura3::imm434/ura3::imm434 This studyCaGD4 CaGD3 ess1 H171R:hisG/ess1�::hisG ura3::imm434/ura3::imm434 This study

CPH1 mutantsCaGD5 CaGD1 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 CPH1/cph1� ::hisG-URA3-hisG This studyCaGD6 CaGD5 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 CPH1/cph1� ::hisG This studyCaGD7 CaGD6 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 cph1� ::hisG-URA3-hisG/ This study

cph1�::hisGCaGD8 CaGD4 ess1 H171R:hisG/ess1�::hisG ura3::imm434/ura3::imm434 CPH1/cph1� ::hisG- This study

URA3-hisGCaGD9 CaGD8 ess1 H171R/ess1�::hisG ura3::imm434/ura3::imm434 CPH1/cph1� ::hisG This studyCaGD10 CaGD9 ess1 H171R:hisG/ess1�::hisG ura3::imm434/ura3::imm434 cph1� ::hisG-URA3- This study

hisG/cph1� ::hisG

TUP1 mutantsCaGD11 CaGD1 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 TUP1/tup1� ::hisG-URA3-hisG This studyCaGD12 CaGD11 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 TUP1/tup1� ::hisG This studyCaGD13 CaGD12 ESS1/ess1�::hisG ura3::imm434/ura3::imm434 tup1� ::hisG-URA3-hisG/ This study

tup1� ::hisGCaGD14 CaGD4 ess1 H171R:hisG/ess1�::hisG ura3::imm434/ura3::imm434 TUP1/tup1� ::hisG- This study

URA3-hisGCaGD15 CaGD14 ess1 H171R:hisG/ess1�::hisG ura3::imm434/ura3::imm434 TUP1/tup1� ::hisG This studyCaGD16 CaGD15 ess1 H171:hisG/ess1�::hisG ura3::imm434/ura3::imm434 tup1� ::hisG-URA3- This study

hisG/tup1� ::hisG

of the ess1H171R/ess1� strain (ess1ts) was compared at per- the mutant cells. Note that the size of the ess1ts colonies issmaller than wild type, indicating a slight growth defectmissive and nonpermissive temperatures. Mutant cells

streaked on solid media grew normally at 30� but not even at permissive temperature.Microscopic examination of mutant cells shows that,at 37� or 42�, whereas wild-type and heterozygous cells

grew at all temperatures (Figure 6A). In liquid media, as with S. cerevisiae ess1ts mutants, the C. albicans ess1ts

mutant cells undergo uniform arrest late in mitosis (Fig-ess1ts cells grew well at 30�, albeit slightly slower thanwild type. However, at 37�, ess1ts cells grew very poorly ure 6D). DAPI staining indicates that the nuclear divi-

sion is completed prior to arrest (Figure 6D). However,(data not shown), and at 40� growth was arrested (Figure6B). To further demonstrate that ESS1 is required for we did not observe nuclear fragmentation as in S. cerevis-

iae (Lu et al. 1996; Wu et al. 2000). Mutant cells thatgrowth, we tested the plating efficiency of wild-type vs.ess1ts cells at different temperatures. As shown in Figure were growth arrested at high temperature (40� and 42�)

retained viability for up to 18 hr (data not shown),6C, ess1ts cells form colonies at 30� but are unable to formcolonies at 40�, whereas wild-type cells form equivalent perhaps because they activated a checkpoint control.

At low temperature (25�) in YEPD, we observed thatnumbers of colonies at both temperatures. At 40�, theplating efficiency of wild-type cells was �75% vs. 0% for some cells (�20–40%) displayed a filamentous pheno-


impaired for germ-tube formation in Lee’s medium andto a lesser extent in Spider medium. A minor reductionwas observed in serum-inducing medium. The ess1ts mu-tant showed similar defects; germ-tube formation wasseverely reduced in Lee’s medium and to a lesser extentin Spider and serum-inducing media. At this tempera-ture, however, growth of the ts strain slows (but doesnot stop) and cells begin to accumulate in mitosis; thusit is not clear whether the failure to form germ tubesin Lee’s medium is a direct effect or is due to the growthimpairment. Comparison of hyphal induction in liquidmedium at permissive temperature (30�) was not feasi-ble, since even the wild type failed to form filamentswith high frequency. The mutant strains were also testedin milk-tween agar, cornmeal agar, and Medium 199(pH 7.0), and similar results were obtained; filamenta-Figure 5.—C. albicans ess1ts mutant alleles are temperaturetion was reduced in the heterozygote and the ts mutantsensitive in S. cerevisiae. S. cerevisiae deletion strains (YSH-55(data not shown). These results suggest that, at least inhaploids or YXW2.1; see materials and methods) carrying

C. albicans ESS1 (wild-type or ess1ts alleles) were spotted onto certain inducing media, Ess1 is required for efficientplates as serial 1:5 dilutions from an original concentration morphogenetic switching in C. albicans.of �106 cells/ml and incubated for 3–4 days at the indicated Epistasis analysis between ESS1 and TUP1 and CPH1temperatures.

mutations: Several distinct genetic pathways have beenshown to control morphogenetic switching in C. albicans(reviewed in Ernst 2000; Lengeler et al. 2000). Totype (Figure 7). This suggests that ESS1 not only is re-

quired for growth but also may be involved in hyphal determine whether ESS1 could be placed within one ofthe known pathways, we carried out epistasis analysismorphogenesis.

ess1H171R mutants are defective in hyphal morphogene- with two genes, TUP1 and CPH1, known to be involvedin switching. TUP1 is a homolog of S. cerevisiae TUP1sis: During infection of the mammalian host, C. albicans

undergoes morphogenetic switching from the yeast to and in C. albicans is a repressor of genes required forfilamentation (Braun and Johnson 1997). CPH1 is athe pseudohypal and hyphal forms. This process of fil-

amentation can also be induced in culture using various homolog of the S. cerevisiae STE12 transcription activator(Liu et al. 1994). In both organisms, CPH1/STE12 aremedia such as Lee’s medium, serum-containing me-

dium, and Spider medium (Lee et al. 1975; Liu et al. targets of mitogen-activated protein (MAP) kinase path-ways (Liu et al. 1994; Kohler and Fink 1996; Leberer1994). We tested the ability of ess1�/ESS1 and ess1H171R/

ess1� strains to undergo filamentation in these media. et al. 1996; Madhani and Fink 1997).We generated double-mutant combinations betweenCells were plated on solid-inducing media and incu-

bated at 30� for 7 days. Examination of colony morphol- ESS1 and TUP1 and between ESS1 and CPH1. We alsoattempted to generate double mutants with EFG1 (Stoldtogies revealed abundant filaments for the wild-type

strain (SC5314); however, far fewer, if any, filaments et al. 1997), but were unsuccessful, perhaps owing toa synthetic-lethality phenotype (data not shown). Onewere formed in either the heterozyous or ts-mutant

strains (Figure 8). The ess1H171R/ess1� strain grew more or both copies of TUP1 were deleted from wild-type,ess1�/�, and ess1ts/� cells to generate six new geno-slowly at 30� on all media tested. Inductions were also

done at the normal induction temperature (37�) and types. Similarly, we deleted one or both copies of CPH1in the same three ESS1 backgrounds. We then scoredagain the heterozygous strain and ess1H171R/ess1� mutants

failed to form filaments in Lee’s and Spider media, the ability of the double mutants to undergo morphoge-netic switching in noninducing media (YEPD) or underalthough they did form some in serum-containing me-

dium (data not shown). However, at 37� the ts mutant various hyphal-inducing conditions using liquid media(Table 2).grows slowly, and the resulting colonies were much

smaller than wild type. Double-mutant analysis revealed that null mutationsin TUP1 are epistatic to hypomorphic mutations in ESS1.We also examined filamentation in liquid-inducing

media. At 37�, germ-tube formation (the initial stage of As shown in Figures 8 and 9, mutants with reduced ESS1gene dosage (ess1�/�) and a ts mutant (ess1H171R/�) failhyphae formation) occurs rapidly, within the first few

hours of induction. We compared the ability of wild- to form filaments under certain inducing conditions. Incontrast, tup1 mutations (tup1�/tup1�) form filamentstype, heterozygous (ess1�/ESS1), and ess1ts mutant strains

to form germ tubes. As expected, wild-type cells formed constitutively, even in rich media (Braun and Johnson1997). We found that tup1 ess1 double mutants formedabundant germ tubes in all three inducing media (Fig-

ure 9). In contrast, the ess1�/ESS1 strain was severely filaments consitutively, regardless of the ESS1 genotype


Figure 6.—Growth prop-erties and arrest phenotypeof C. albicans ess1 mutants.(A) YEPD � uridine platesat 30�, 37�, and 42� contain-ing wild-type CAI4, ess1�/�,and ess1H171R/� mutants. (B)Growth curves of wild-typeCAI4, ess1�/�, and ess1H171R/� mutants in liquid YEPDmedia at 30� and 40�. (C)Plating efficiency of wild-type vs. ess1H171R/� mutantcells at 30� and 40�. Equiva-lent numbers of log-phasecells grown at 30�, based onOD600, were plated on YEPDmedium and allowed togrow for 4 days at the indi-cated temperature. (D)DAPI staining and Nomar-ski (DIC) optics of ess1H171R/� mutants at permissive(30�) and nonpermissive(40�) temperatures. Cellswere shifted to nonpermis-sive temperature and incu-bated overnight.

(Figure 10A). Filamentation was also observed under mutant, suggesting that ESS1 acts genetically upstreamof, or in a different pathway from, TUP1.inducing conditions using Lee’s, serum-containing, and

Spider media (Figure 10A, Table 2). As expected, control Results with CPH1 suggest that ESS1 and CPH1 mightfunction in a common pathway. Mutations in CPH1heterozygous tup1 mutants (tup1�/�) did not form fil-

aments in rich medium (YEPD), regardless of the ESS1 block morphogenetic switching in Spider medium, butthey do not block switching in response to serum-con-genotype (Table 2). Thus, in all cases the tup1 ess1 dou-

ble mutants displayed the phenotype of the tup1 single taining media (Liu et al. 1994). In our experiments,cph1 mutants were also reduced for filamentation inLee’s medium. Mutations in ESS1 behaved similarly,with a switching defect most apparent in Lee’s and Spi-der, but less so in serum-containing media (Figure 9).Results with cph1 ess1 double-mutants are shown in Fig-ure 10B. In Lee’s medium, double mutant cells arecompletely defective in filamentation. This phenotypeis similar to that seen in the individual cph1 or ess1 singlemutants, although the penetrance appears to be higher,with more cells remaining exclusively in the yeast form.In serum-containing medium, cells appear partiallycompetent to form filaments, similar to ess1 single mu-

Figure 7.—ess1H171R/� mutants filament spontaneously at tants. In Spider medium, double-mutant cells were un-low temperature (25�). Wild-type SC5314, ess1�/�, and able to form filaments, similar to cph1 single mutants.ess1H171R/� mutants were grown on YEPD agar for 4 days at

Thus, the induction response of ess1 cph1 double-mutant25�; cells removed from the plate were resuspended in water,cells was, in general, similar to that of each of the singlesonicated briefly, and then fixed in 3.7% formaldehyde, before

being photographed. Bar, 10 �m. mutants, perhaps with some degree of increased pene-


TABLE 2

Summary of filamentation phenotypes of tup1 ess1 and cph1ess1 double mutants

TUP1 genotypeInducing ESS1medium genotype �/� �/� �/�

Lee’s �/� �� /� � ��ts/� � ��

Spider �/� �� /� � ��ts/� � ��

Serum �/� �� /� �� ts/� ��

YEPD �/� � � ��Noninducing �/� � � ��

ts/� � � ��Figure 8.—ESS1 mutants are defective in filamentation on

CPH1 genotypesolid hyphal-inducing media. Colony morphologies of wild-Inducing ESS1type SC5314, ess1�/�, and ess1H171R/� mutants grown on solidmedium genotype �/� �/� �/�Lee’s, solid serum-containing, and solid Spider media at 30�

for 7 days are shown. Colonies were photographed at the sameLee’s �/� �� magnification (20�).

�/� � � �ts/� � � �

Spider �/� �� trance. The simplest interpretation is that ESS1 and�/� �

CPH1 act in the same pathway or in redundant pathways. ts/� �Serum �/� ��

�/� �� DISCUSSION ts/� �� ND ��

The results presented here show that the Ess1 prolyl- Cells were grown at 30� in YEPD and then shifted to theisomerase is conserved in C. albicans and suggest that indicated liquid-inducing medium at 37 � for 4 hr. ��, a

population in which almost all cells are hyphal and/or pseu-its function is similar to that in budding yeast. Using adohyphal; ��, a population in which �50% are hyphaland/or pseudohyphal; �, a population in which �50% arehyphal and/or pseudohyphal; , a population in which thereare few hyphal and/or pseudohyphal; �, a population inwhich no filamentous cells were observed; ND, not deter-mined.

novel ts-mutant approach, we showed that ESS1 is essen-tial for growth in C. albicans and that, as in buddingyeast, loss-of-function mutants arrest in mitosis. How-ever, these cells do not undergo nuclear fragmentationas in S. cerevisiae, suggesting that the mitotic arrest andnuclear fragmentation are genetically separable, at leastin C. albicans.

S. cerevisiae as a surrogate host for generating ts allelesof C. albicans genes: Demonstrating that genes are essen-tial in an obligate diploid such as C. albicans has beenproblematic. Although there are some new approaches(e.g., Enloe et al. 2000; Lee et al. 2001), the use ofconditional alleles would, as we have shown, offer majoradvantages. Not only do they provide a way to prove

Figure 9.—ESS1 mutants are defective in filamentation in essentiality of a gene, but they also allow the study ofliquid hyphal-inducing media. Cell morphologies of wild-type terminal loss-of-function phenotypes and provide a waySC5314, ess1�/�, and ess1H171R/� mutants in liquid Lee’s, se-

to search for second-site suppressors. In our case, therum-containing, and Spider media are shown. Cells werets alleles previously identified in S. cerevisiae ESS1 pro-grown in YEPD media at 30� and then shifted to the indicated

inducing medium for 4 hr at 37�. Bar, 10 �m. vided the basis for engineering equivalent mutations in


et al. 1996; Fujimori et al. 1999). It is not known whyEss1 is essential in some organisms, but not others. Thisdifference is not simply a matter of the presence ofduplicated genes, as no closely related genes are foundin these organisms. It is possible that there exist naturalsuppressor pathways, similar to those identified by Wuet al. (2000) that affect gene transcription. In fact, someess1-deletion strains of S. cerevisiae are able to grow atreduced temperature (25�), albeit very slowly (H. Huangand T. Hunter, personal communication). WhetherEss1 is essential in other human fungal pathogens re-mains to be determined.

Ess1 is required for hyphal induction: Under condi-tions in which Ess1 function is compromised, e.g., in theess1H171R ts mutant at permissive temperature, C. albicanscells were defective in filamentation when shifted toinducing medium. Likewise, when the gene dosage wasreduced, as in heterozygous mutant cells (ess1�/ESS1),filamentation was severely compromised. Surprisingly,C. albicans ess1H171R seems to filament spontaneously atreduced temperature (25�), even in the absence of in-ducing signals. This cold-sensitive phenotype might bedue to an increase in inappropriate protein-protein in-teractions by the defective protein at low temperatures(Jarvick and Botstein 1973). Taken together, thesedata suggest that Ess1 function is involved in morphoge-netic switching, perhaps via a role in transcription (seebelow). Since switching is important for virulence, it

Figure 10.—Phenotypes of tup1 ess1 and cph1 ess1 double will be interesting to determine whether mutations inmutants. Cell morphologies of (A) tup1 deletion mutants and ESS1 also affect pathogenicity of C. albicans in animal(B) cph1 deletion mutants in wild-type, ess1�/�, and ess1H171R/�

models.backgrounds after hyphal induction in liquid Lee’s, serum-Morphogenetic switching, and in particular the transi-containing, and Spider media are shown. Cells were grown in

YEPD media at 30� and then shifted to the indicated inducing tion from the yeast to hyphal forms, can be induced bymedia for 4 hr at 37�. Bars, 10 �m. different environmental stimuli ranging from starva-

tion, to pH changes, to the presence of N-acetylglucos-amine and proline in the medium (Ernst 2000). These

the C. albicans gene. However, this site-directed strategy signals are transduced by several independent signal-might not succeed in all cases. For example, a mutation transduction pathways, including a MAP-kinase pathwaygenerated in C. albicans NMT1 based on a ts allele in S. and a cAMP-dependent pathway. As in S. cerevisiae, acerevisiae showed a loss-of-function phenotype, but this common theme is that the downstream targets of thesephenotype was not temperature sensitive in C. albicans, pathways appear to be transcription regulatory proteins.although the enzyme activity did show a ts effect in vitro Ess1 has been linked to transcription regulation (Wu et(Weinberg et al. 1995). al. 2000); thus, it is not surprising that Ess1 is important

Our use of S. cerevisiae as a surrogate host to prescreen for the yeast-to-hyphal transition. It is possible that thefor a conditional-mutant phenotype suggests it should expression of genes required for filamentation is alteredbe possible to generate de novo conditional alleles in any in ess1 mutant cells.C. albicans gene that has a phenotype when expressed in Epistasis analysis with the transcription repressorS. cerevisiae. This could be accomplished using gapped- TUP1 revealed that mutations in TUP1 are epistatic toplasmid mutagenesis methods (Muhlrad et al. 1992). those in ESS1. It is unclear from these experimentsThe generation and use of such conditional alleles will whether Ess1 functions upstream of Tup1 or in a parallelbe an important addition to the repertoire of genetic pathway. It is possible, for example, that Ess1 is requiredtechniques available for pathogenic fungi. for the transcription of the TUP1 gene. In the absence

Ess1 is essential in some organisms but not others: of TUP1, filamentation genes are derepressed regardlessWhile Ess1 is essential in both S. cerevisiae and C. albicans, of whether Ess1 is functional. Alternatively, Ess1 mightits homologs (Pin1/Dodo) do not appear to be essential function in a different pathway. Two findings favor thisin Schizosaccharomyces pombe (H. Huang and T. Hunter, latter possibility. First, previous work has suggested that

TUP1 is likely to have a function in the serum-inductionpersonal communication) or in metazoans (Maleszka


pathway (Braun and Johnson 2000). In contrast, ESS1 LITERATURE CITEDdoes not appear to function in the serum-induction Adams, A., D. E. Gottschling, C. A. Kaiser and T. Stearns, 1997pathway, as suggested by the fact that ess1 mutants, de- Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course

Manual. Cold Spring Harbor Laboratory Press, Cold Spring Har-spite having filamentation defects in Lee’s and Spiderbor, NY.media, are capable of forming filaments in serum-con- Alani, E., L. Cao and N. Kleckner, 1987 A method for gene disrup-

taining medium (Figure 9). tion that allows repeated use of URA3 selection in the construc-tion of multiply disrupted yeast strains. Genetics 116: 541–545.Second, ess1 mutant strains show defects similar to

Anaissie, E. J., J. H. Rex, O. Uzun and S. Vartivarian, 1998 Pre-cph1 mutants, suggesting that they are in the same path- dictors of adverse outcome in cancer patients with candidemia.way. CPH1 is a member of a MAP kinase cascade (Liu Am. J. Med. 104: 238–245.

Arevalo-Rodrıguez, M., M. E. Cardenas, X. Wu, S. D. Hanes andet al. 1994; Kohler and Fink 1996), which is serumJ. Heitman, 2000 Cyclophilin A and Ess1 interact with andindependent and distinct from that of TUP1. That ESS1 regulate silencing by the Sin3-Rpd3 histone deacetylase. EMBO

and CPH1 function in the same pathway is supported by J. 19: 3739–3749.Braun, B. R., and A. D. Johnson, 1997 Control of filament forma-the observation that, upon induction, cph1 ess1 double-

tion in Candida albicans by the transcriptional repressor TUP1.mutant cells show neither enhanced nor decreased fil- Science 277: 105–109.amentation relative to the single mutants. Moreover, Braun, B. R., and A. D. Johnson, 2000 TUP1, CPH1 and EFG1 make

independent contributions to filamentation in Candida albicans.the phenotypes of ess1 tup1 (Figure 10) and cph1 tup1Genetics 155: 57–67.double mutants (Braun and Johnson 1997) are similar.

Brown, A. J., and N. A. Gow, 1999 Regulatory networks controllingThus, it is possible that ESS1 functions in a MAP kinase Candida albicans morphogenesis. Trends Microbiol. 7: 333–338.

Crenshaw, D. G., J. Yang, A. R. Means and S. Kornbluth, 1998pathway, perhaps by modulating transcription of down-The mitotic peptidyl-prolyl isomerase, Pin1, interacts with Cdc25stream target genes. Interestingly, recent work in Dro-and Plx1. EMBO J. 17: 1315–1327.

sophila indicates that Dodo functions as part of a MAP Cutler, J. E., 1991 Putative virulence factors of Candida albicans.Annu. Rev. Microbiol. 45: 187–218.kinase pathway to control dorso-ventral patterning in

Dolinski, K., and J. Heitman, 1997 Peptidyl-prolyl isomerases—anthe early embryo (Hsu et al. 2001). Finally, it is possibleoverview of the cyclophilin, FKBP and parvulin families, pp. 359–

that Ess1 in C. albicans may function in multiple path- 369 in Guidebook to Molecular Chaperones and Protein Folding Cataly-ways, by isomerization of different prolyl-containing tar- sis, edited by M. J. Gething. Oxford University Press, Oxford.

Dolinski, K., S. Muir, M. Cardenas and J. Heitman, 1997 Allget proteins.cyclophilins and FK506 binding proteins are, individually andEss1 as an antifungal drug target: Our results suggest collectively, dispensable for viability in Saccharomyces cerevisiae.

that Ess1, a parvulin-class PPIase, might be a viable target Proc. Natl. Acad. Sci. USA 94: 13093–13098.Einbond, A., and M. Sudol, 1996 Towards prediction of cognatefor antifungal agents. Although the cyclophilin- and

complexes between the WW domain and proline-rich ligands.FKBP-class PPIases have been traditional targets for FEBS Lett. 384: 1–8.drug development, Ess1 offers several potential advan- Enloe, B., A. Diamond and A. P. Mitchell, 2000 A single-transfor-

mation gene function test in diploid Candida albicans. J. Bacteriol.tages. First, Ess1 inhibitors might be used to prevent82(20): 5730–5736.cell growth. Or, at lower doses that reduce (but do not Ernst, J. F., 2000 Transcription factors in Candida albicans—

eliminate) Ess1 function, such inhibitors might prevent environmental control of morphogenesis. Microbiology 146:1763–1774.morphogenetic switching, which is required for viru-

Fonzi, W. A., and M. Y. Irwin, 1993 Isogenic strain constructionlence. Furthermore, since Ess1 is an essential gene, cellsand gene mapping in Candida albicans. Genetics 134: 717–728.

cannot become resistant, as they do against cyclosporin Fruman, D. A., S. J. Burakoff and B. E. Bierer, 1994 Immuno-philins in protein folding and immunosuppression. FASEB J. 8:A or FK506, by mutations that abolish production of391–400.the respective PPIase. That is, a null mutation in ESS1

Fujimori, F., K. Takahashi, C. Uchida and T. Uchida, 1999 Micewould be lethal, and mutations that reduce Ess1 activity lacking Pin1 develop normally, but are defective in entering cellwould be switching defective. Finally, since the mamma- cycle from G(0) arrest. Biochem. Biophys. Res. Commun. 265:

658–663.lian homolog Pin1 does not appear to be essential, toxicGothel, S. F., and M. A. Marahiel, 1999 Peptidyl-prolyl cis-transside effects might be minimal. Further study will be isomerases, a superfamily of ubiquitous folding catalysts. Cell.

needed to identify high-affinity Ess1 inhibitors and to Mol. Life Sci. 55: 423–436.Hanes, S. D., 1988 Isolation, sequence and mutational analysis oftest their efficacy in preventing fungal growth and patho-

ESSI, a gene essential for growth in Saccharomyces cerevisiae. Ph.D.genicity.Thesis, Brown University, Providence, RI.

Hanes, S. D., P. R. Shank and K. A. Bostian, 1989 Sequence andWe thank Federico Navarro-Garcıa for the C. albicans genomic DNAmutational analysis of ESS1, a gene essential for growth in Saccharo-library; Gerald Fink, William Fonzi, Alexander Johnson, and Alexan-myces cerevisiae. Yeast 5: 55–72.der Tzagoloff for plasmids and strains; and the Wadsworth Center

Hemenway, C., and J. Heitman, 1993 Proline isomerases in microor-Core Facilities: Molecular Genetics, Richard Cole at Advanced Lightganisms and small eukaryotes, pp. 38–46 in ImmunosuppressiveMicroscopy and Media, and also Marisa Foehr for technical assistance;and Antiinflammatory Drugs, edited by A. C. Allison, K. J. Laf-Stephen de Waal Malefyt for a complementation experiment; Ramaferty and H. Fliri. The New York Academy of Sciences, New

Ramani and Jim Mittler for help with flow cytometry; Dilip Nag for York.comments on the manuscript; and Cathy Wilcox and Xiaoyun Wu Horton, R. M., Z. Cai, S. N. Ho and L. R. Pease, 1990 Gene splicingfor helpful discussions. G.D. acknowledges the Molecular Mycology by overlap extension: tailor-made genes using the polymerasecourse held at The Marine Biological Laboratory, Woods Hole, MA. chain reaction. Biotechniques 8: 528–535.This work was supported by a Pfizer Educational Award (V.C.) and Hsu, T., D. Mcrackan, T. S. Vincent and H. G. De Couet, 2001grants from the National Institutes of Health [AI-41968 (V.C.) and Drosophila Pin1 prolyl isomerase Dodo is a MAP kinase signal

responder during oogenesis. Nat. Cell Biol. 3: 538–543.R01-GM55108 (S.D.H.)].


Hunter, T., 1998 Prolyl isomerases and nuclear function. Cell 92: McKune, K., K. L. Richards, A. M. Edwards, R. A. Young and N. A.Woychik, 1993 RPB7, one of two dissociable subunits of yeast141–143.RNA polymerase II, is essential for cell viability. Yeast 9: 295–299.Ito, H., Y. Fukada, K. Murata and A. Kimura, 1983 Transformation

Morris, D. P., H. P. Phatnani and A. L. Greenleaf, 1999 Phospho-of intact yeast cells treated with alkali cations. J. Bacteriol. 53:carboxyl-terminal domain binding and the role of a prolyl iso-163–168.merase in pre-mRNA 3�-end formation. J. Biol. Chem. 274: 31583–Jarvick, J., and D. Botstein, 1973 A genetic method for determin-31587.ing the order of events in a biological pathway. Proc. Natl. Acad.

Muhlrad, D., R. Hunter and R. Parker, 1992 A rapid method forSci. USA 70: 2046–2050.localized mutagenesis of yeast genes. Yeast 8: 79–82.Jitsurong, S., S. Kiamsiri and N. Pattararangrong, 1993 New

Navarro-Garcia, F., M. Sanchez, J. Pla and C. Nombela, 1995milk medium for germ tube and chlamydoconidia productionFunctional characterization of the MKC1 gene of Candida albicans,by Candida albicans. Mycopathologia 123: 95–98.which encodes a mitogen-activated protein kinase homolog re-Kao, A. S., M. E. Brandt, W. R. Pruitt, L. A. Conn, B. A. Perkinslated to cell integrity. Mol. Cell. Biol. 15: 2197–2206.et al., 2000 The epidemiology of candidemia in two United

Odds, F. C., 1988 Candida and Candidosis : A Review and Bibliography,States cities: results of a population-based active surveillance. Clin.Ed. 2. Bailliere Tindall, London.Infect. Dis. 29: 1164–1170.

Rahfeld, J. U., K. P. Rucknagel, B. Schelbert, B. Ludwig, J.Khazak, V., P. P. Sadhale, N. A. Woychik, R. Brent and E. A.Hacker et al., 1994a Confirmation of the existence of a thirdGolemis, 1995 Human RNA polymerase II subunit hsRPB7family among peptidyl-prolyl cis/trans isomerases: amino acid se-functions in yeast and influences stress survival and cell morphol-quence and recombinant production of parvulin. FEBS Lett. 352:ogy. Mol. Biol. Cell 6: 759–775.180–184.Kohler, J. R., and G. R. Fink, 1996 Candida albicans strains heterozy-

Rahfeld, J. U., A. Schierhorn, K. Mann and G. Fischer, 1994bgous and homozygous for mutations in mitogen-activated proteinA novel peptidyl-prolyl cis/trans isomerase from Escherichia coli.kinase signaling components have defects in hyphal develop- FEBS Lett. 343: 65–69.ment. Proc. Natl. Acad. Sci. USA 93(23): 13223–13228. Ranganathan, R., K. P. Lu, T. Hunter and J. P. Noel, 1997 Struc-Kops, O., C. Eckerskorn, S. Hottenrott, G. Fischer, H. Mi et tural and functional analysis of the mitotic rotamase Pin1 suggests

al., 1998 Ssp1, a site-specific parvulin homolog from Neurospora substrate recognition is phosphorylation dependent. Cell 89:crassa active in protein folding. J. Biol. Chem. 273: 31971–31976. 875–886.

Kunz, J., and M. N. Hall, 1993 Cyclosporin A, FK506 and rapamycin: Rutherford, S. L., and C. S. Zuker, 1994 Protein folding and themore than just immunosuppression. Trends Biochem. Sci. 18: regulation of signaling pathways. Cell 79: 1129–1132.334–338. Shen, M., P. T. Stukenberg, M. W. Kirschner and K. P. Lu, 1998

Leberer, E., D. Harcus, I. D. Broadbent, K. L. Clark, D. Dignard The essential mitotic peptidyl-prolyl isomerase Pin1 binds andet al., 1996 Signal transduction through homologs of the Ste20p regulates mitosis-specific phosphoproteins. Genes Dev. 12: 706–and Ste7p protein kinases can trigger hyphal formation in the 720.pathogenic fungus Candida albicans. Proc. Natl. Acad. Sci. USA Sikorski, R. S., and P. Hieter, 1989 A system of shuttle vectors and93(23): 13217–13222. yeast host strains designed for efficient manipulation of DNA in

Lee, K. L., H. R. Buckley and C. C. Campbell, 1975 An amino acid Saccharomyces cerevisiae. Genetics 122: 19–27.liquid synthetic medium for the development of mycelial and Stewart, D. E., A. Sarkar and J. E. Wampler, 1990 Occurrenceyeast forms of Candida albicans. Sabouraudia 13: 148–153. and role of cis peptide bonds in protein structures. J. Mol. Biol.

Lee, S. A., Y. Mao, Z. Zhang and B. Wong, 2001 Overexpression 214: 253–260.of a dominant-negative allele of YPT1 inhibits growth and aspartyl Stoldt, V. R., A. Sonneborn, C. E. Leuker and J. F. Ernst, 1997protease secretion in Candida albicans. Microbiology 147: 1961– Efg1p, an essential regulator of morphogenesis of the human1970. pathogen Candida albicans, is a member of a conserved class of

Lengeler, K. B., R. C. Davidson, C. D’Souza, T. Harashima, W. bHLH proteins regulating morphogenetic processes in fungi.Shen et al., 2000 Signal transduction cascades regulating fungal EMBO J. 16: 1982–1991.development and virulence. Microbiol. Mol. Biol. Rev. 64(4): Sudol, M., 1996 Structure and function of the WW domain. Prog.746–785. Biophys. Mol. Biol. 65: 113–132.

Thompson, J. D., D. G. Higgins and T. J. Gibson, 1994 ClustalW:Liu, H., J. Kohler and G. R. Fink, 1994 Suppression of hyphalimproving the sensitivity of progressive multiple sequence align-formation in Candida albicans by mutation of a STE12 homolog.ment through sequence weighting, position specific gap penaltiesScience 266: 1723–1726.and weight matrix choice. Nucleic Acids Res. 22: 4673–4680.Lu, K. P., S. D. Hanes and T. Hunter, 1996 A human peptidyl-

Verdecia, M. A., M. E. Bowman, K. P. Lu, T. Hunter and J. P. Noel,prolyl isomerase essential for regulation of mitosis. Nature 380:2000 Structural basis for phosphoserine-proline recognition by544–547.group IV WW domains. Nat. Struct. Biol. 7: 639–643.Lu, P. J., X. Z. Zhou, M. Shen and K. P. Lu, 1999a Function of

Weinberg, R. A., C. A. McWherter, S. K. Freeman, D. C. Wood,WW domains as phosphoserine- or phosphothreonine-bindingJ. I. Gordon et al., 1995 Genetic studies reveal that myristoyl-modules. Science 283: 1325–1328.CoA:protein N-myristoyltransferase is an essential enzyme in Can-Lu, P. J., G. Wulf, X. Z. Zhou, P. Davies and K. P. Lu, 1999b Thedida albicans. Mol. Microbiol. 16: 241–250.prolyl isomerase Pin1 restores the function of Alzheimer-associ-

Winkler, K. E., K. I. Swenson, S. Kornbluth and A. R. Means, 2000ated phosphorylated tau protein. Nature 399: 784–788.Requirement of the prolyl isomerase Pin1 for the replicationMadhani, H. D., and G. R. Fink, 1997 Combinatorial control re-checkpoint. Science 287: 1644–1647.quired for the specificity of yeast MAPK signaling. Science

Wu, X., C. B. Wilcox, G. Devasahayam, R. L. Hackett, M. Arevalo-275(5304): 1314–1317.Rodrıguez et al., 2000 The Ess1 prolyl isomerase is linked toMadhani, H. D., and G. R. Fink, 1998 The control of filamentouschromatin remodeling complexes and the general transcriptiondifferentiation and virulence in fungi. Trends Cell Biol. 8: 348– machinery. EMBO J. 19: 3727–3738.353. Yaffe, M. B., M. Schutkowski, M. Shen, X. Z. Zhou, P. T. Stuken-

Maleszka, R., S. D. Hanes, R. L. Hackett, H. G. De Couet and berg et al., 1997 Sequence-specific and phosphorylation-depen-G. L. Miklos, 1996 The Drosophila melanogaster dodo (dod) gene, dent proline isomerization: a potential mitotic regulatory mecha-conserved in humans, is functionally interchangeable with the nism. Science 278: 1957–1960.ESS1 cell division gene of Saccharomyces cerevisiae. Proc. Natl. Acad. Yesland, K., and W. A. Fonzi, 2000 Allele-specific gene targetingSci. USA 93: 447–451. in Candida albicans results from heterology between alleles. Micro-

Maleszka, R., A. Lupas, S. D. Hanes and G. L. Miklos, 1997 The biology 146: 2097–2104.dodo gene family encodes a novel protein involved in signal trans-duction and protein folding. Gene 203: 89–93. Communicating editor: A. P. Mitchell

the ess1 prolyl isomerase is required for growth …the ess1 prolyl-isomerase might be a useful...

Documents