| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
Department of Bioactive Molecules, National Institute of Infectious Diseases, Tokyo 162-8640, Japan
Received 25 May 2006/ Accepted 28 August 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
disruptant showed filamentous growth under conditions inducing the yeast form, as is true of the Catup1 mutant. Consistent with this result, the
common set of hypha-specific genes was negatively regulated by both TCC1 and CaTUP1. These observations will provide new insights into CaTup1p-dependent transcriptional gene regulation in C. albicans. |
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ScTup1p and
ScSsn6p form a protein complex to act as a global repressor in S.
cerevisiae. This complex is targeted to promoters by DNA-binding
proteins specific for the different classes of repressed genes
(26). ScTUP1 was
first identified as a mutant that was able to incorporate
deoxythymidine (32).
Subsequently, a number of distinct phenotypes of the Sctup1
mutant have been observed, including slow growth, flocculation, loss of
mating in alpha strains, poor sporulation, and loss of some aspects of
glucose repression. Scssn6 was first identified as
a suppressor mutation of the snf1 mutant: Snf1p is required to
derepress the expression of many glucose-repressible genes, including
the SUC2 invertase gene, and the Scssn6 mutation
causes constitutive invertase synthesis
(8). The Scssn6
mutations are allelic to the cyc8 mutation
(8), which causes
increased production of iso-2-cytochrome c
(23). Deletion of the
ScSSN6 gene results in many phenotypes, most of which are
identical to those of the Sctup1 mutant. From the viewpoint of
protein structure, ScTup1p contains seven copies of a WD40 repeat,
named after two amino acids, tryptophan and aspartic acid, commonly
found in the repeat and its length. The seven repeats fold into a
propeller-like structure, which is hypothesized to bind the homeodomain
protein 
2 (17).
ScSsn6p includes 10 copies of the tetratricopeptide repeat (TPR),
comprising the 34 amino acids that make up the basic repeat
(10), which is related to
the interaction of ScSsn6p-ScTup1p
(29) or
ScSsn6p-2 (27).
Generally, TPR motifs have been found in a wide variety of
proteins from all organisms, from humans to prokaryotes. They mediate
molecular recognition and protein-protein interactions. While 22
proteins containing the TPR motif have been found encoded in the yeast
genome, only three proteins are involved in transcriptional regulation:
Ctr9p, Tfc4p, and ScSsn6p
(10). Of the 10 copies of
TPRs in ScSsn6p, the first to the third TPR motifs are known to be
responsible for ScTup1 binding, whereas combinations of the other TPRs
mediate interactions with different repressor proteins specific for
each gene family regulated by the ScTup1p-ScSsn6p complex
(29).
Recently, studies of Tup1-dependent gene repression in Candida albicans have been undertaken by many scientists. C. albicans is an opportunistic fungal pathogen in humans and can cause either systemic or mucosal infection. In immunocompromised patients, infection with this organism can progress to severe systemic invasion, leading to life-threatening circumstances (20, 21). C. albicans is a polymorphic fungus capable of converting its cell shape from budding yeast to a filamentous form, including pseudohyphae and true hyphae. This morphological transition has been strongly associated with pathogenicity (6).
The C.
albicans TUP1 gene was first isolated and disrupted by Braun and
Johnson (4). Since then,
several research groups have reported that Candida Tup1p
represses hypha-specific genes (HSGs) under conditions inducing the
yeast form, as suggested by the exclusive filamentation of the gene
disruptant. CaTup1p may require the DNA-binding protein CaNrg1p for the
repression of hypha-specific genes in a pathway that promotes yeast
form growth, because the Canrg1 mutant displays
constitutive filamentation similar to that of the
Catup1 mutant
(5,
19). However, whether
C. albicans Tup1p and CaNrg1p interact directly with each
other remains unknown. The binding partner of CaTup1 has also been
thought to be an Ssn6p homolog in C. albicans, analogous to
the S. cerevisiae paradigm. However, the phenotypes of the
CaSSN6 and CaTUP1 gene disruptants are definitely
different (12,
14). A recent, excellent
study based on DNA microarray analysis by Garcia-Sanchez et al.
(12) has shown that
almost no hypha-specific genes that were induced by Catup1
deletion overlapped with any genes that were upregulated by
Cassn6 deletion, implying the existence of a CaTup1p-binding
partner other than CaSsn6p with respect to morphogenesis
regulation.
In this report, we identified a novel protein interacting with Tup1p in C. albicans by using tandem affinity purification (TAP) technology. The protein, termed Tcc1p, a Tup1p complex component, formed a protein complex with CaTup1p independently of the CaSsn6p-CaTup1p complex. Deletion of the TCC1 gene resulted in pseudohyphal morphology under conditions inducing yeast form and attenuated virulence, similar to the phenotype of the Catup1 deletion mutant. These observations will give new insights into Tup1p-dependent transcriptional gene regulation in C. albicans.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Northern hybridization and quantitative real-time reverse transcription (RT)-PCR were performed as previously described (13). All primers used for the amplification of Northern probes and real-time PCR are listed in Table S1 in the supplemental material.
Immunostaining was performed as described previously (30). Microscopic observation was performed using a conventional fluorescence microscope (model IX81; Olympus, Japan) equipped with a model DP70 digital camera (Olympus, Japan).
Animal experiments were performed as described previously (30). For each group, five male CD-1 (ICR) mice aged 4 weeks (Charles River, Japan) were inoculated with 106 CFU by intravenous injection. Kaplan-Meier survival curves were compared using the log rank test. A P value of <0.05 was considered significant.
Plasmid and strain construction. All primers used in this study are listed in Table S1 in the supplemental material.
(i) p3HA-ARG4 vector.
The p3HA-ARG4
plasmid was designed to fuse a protein with three tandem repeats of the
hemagglutinin (3xHA) tag at the C terminus and to
contain the ARG4 marker. A KpnI-SacI fragment digested from
plasmid pRS-Arg4SpeI
(33) was cloned into the
KpnI-SacI sites of pUC18 to yield pUC18-ARG4. Then, a 3.8-kb
HindIII-PvuI fragment containing the ARG4 marker of pUC18-ARG4
and a 3.2-kb HindIII-PvuI fragment containing the 3xHA tag and
the ACT1 terminator of p3HA-ACT1 were ligated to yield
p3HA-ARG4.
(ii) pMyc-SAT1 and p3HA-SAT1 vectors. The p3Myc-SAT1 plasmid was designed to facilitate the constitutive expression of a protein fused with a three-tandem repeat of the Myc (3xMyc) tag at the C terminus and to contain a nourseothricin resistance marker, SAT1. To generate the DNA fragment 3xMyc-A, which encodes the N-terminal portion of the tag sequence, two oligonucleotides, 3xMyc-1 and 3xMyc-4, were annealed by boiling them for 3 min and then allowing them to cool to room temperature. To generate the DNA fragment 3xMyc-B, which encodes the C-terminal portion of the tag sequence, two oligonucleotides, 3xMyc-2 and 3xMyc-3, were annealed by the above-described method. The two DNA fragments 3xMyc-A and 3xMyc-B were then ligated, gel purified, and inserted into the XhoI-SphI sites of pFLAG-ACT1 (31) to yield p3Myc-ACT1 (30). A HindIII-PstI fragment digested from plasmid pSFS1A (22) was cloned into the HindIII-PstI sites of pUC18 to yield pUC18-SAT1. Then, a 2.9-kb HindIII-PvuI fragment containing an SAT1 marker of pUC18-SAT1 and a 3.2-kb HindIII-PvuI fragment containing a 3xMyc tag and an ACT1 terminator of p3Myc-ACT1 were ligated to yield p3Myc-SAT1. p3HA-ACT1 was used instead of p3Myc-ACT1 to yield p3HA-SAT1.
(iii) p6HF-Met3 vector. The p6HF-Met3 plasmid was designed to facilitate the conditional expression of a protein fused with a six-histidine-FLAG (HF) tag at the C terminus. A 5.6-kb XhoI-EcoRI fragment containing the MET3 promoter of p3HA-MET3 (30) and a 1-kb XhoI-EcoRI fragment containing the HF tag and the ACT1 terminator of p6HF-ACT1 (16) were ligated to yield p6HF-Met3.
(iv) TCC1 disruption. Gene disruption of TCC1 was performed using a method similar to that described previously (13). Briefly, two fragments, disTCC1-A and disTCC1-B, were amplified using primers disTCC1-1 and -2 and disTCC1-3 and -4, respectively, and used as a flanking homology region for a gene disruption cassette. The PCR-amplified disruption cassette containing an hph200-URA3-hph200 or ARG4 marker was transformed into the TUA4 arg4– ura3– strain (hph200 represents a 200-bp portion of hph). Finally, both alleles of the TCC1 locus were replaced with hph200 and ARG4, yielding strain TCC103. Direct-colony PCR and genomic PCR were performed to verify the strain construction in each step.
(v) TCC1 revertant. For a complementation test, the DNA fragment TCC1comp was amplified from TUA4 chromosomal DNA using primers disTCC1-1 and TCC1comp-C. A mixture of the amplified DNA fragments TCC1comp and fragTCC1-HF (used for HF tagging, as described below) was introduced into TCC103 to generate TCC107, in which one allele has the wild-type open reading frame tagged with His6-FLAG at the C terminus and the other allele is replaced by an ARG4 marker. At first, as a negative control, the DNA fragment TCC1comp-nega-C was amplified using primers TCC1comp-nega5' and TCC1comp-C and TUA4 chromosomal DNA as a template. Next, the DNA fragment TCC1comp-nega was amplified using primers disTCC1-1 and TCC1comp-C and DNA fragments disTCC1-A and TCC1comp-nega-C. A mixture of the resultant DNA fragments TCC1comp-nega and fragTCC1-HF was introduced into TCC103 to generate TCC106, in which one allele has an incomplete TCC1 open reading frame with the region between methionine-1 and proline-704 deleted and the other allele is the ARG4 marker. Direct-colony PCR and genomic PCR were performed to verify the strain construction in each step.
(vi) TUP1, SSN6, and NRG1 disruptions. Gene disruption of TUP1, SSN6, and NRG1 was performed in a manner similar to that described above. A primer set consisting of disTUP1-1, -2, -3, and -4 or disNRG1-1, -2, -3, and -4 for the TUP1 or the NRG1 gene, respectively, was used for the creation of a disruption cassette. A disruption cassette for SSN6 was amplified using 120-mer-long primers, disSSN6-5' and iSSN6-3'. Both alleles of each of the TUP1, SSN6, or NRG1 genes were replaced with the ARG4 marker and a Ura-blaster cassette, hph200-URA3-hph200 (for SSN6) or hisG200-URA3-hisG200 (for TUP1 and NRG1), in strain TUA4, to generate the homozygous mutant TUP102, SSN602, or NRG102, respectively. SSN602 cells were plated on 5-fluoroorotic acid-containing medium to isolate the ura– segregants (SSN603). Direct-colony PCR and genomic PCR were performed to verify the strain construction in each step.
(vii) HGC1 disruption.
Single or
double disruptions of HGC1 and/or TCC1 were performed
using a method similar to that described above. Briefly, for the first
allele, the two fragments disHGC1-A and disHGC1-B were amplified using
primers disHGC1-1 and -2 and disHGC1-3 and -4, respectively, and used
as flanking homology regions for a gene disruption cassette with
pC220-URA3 (30). The
PCR-amplified disruption cassette containing the C220-URA3
marker was transformed into the TUA4
arg4– ura3– strain or TCC103 (tcc1), yielding HGC101 or HGC111, respectively.
Then, strain HGC101 and HGC111 were plated on a medium containing
5-fluoroorotic acid to isolate HGC102 and HGC112, respectively. For
another allele of HGC1, the DNA fragment disHGC1-C was
amplified using primers disHGC1-5 and disHGC1-6; we used disHGC1-A and
disHGC1-C as a gene disruption cassette with pUC19-URA3KX
(30). A second
transformation using the amplified DNA led to the isolation of an
hgc1 single mutant (HGC104) or an
hgc1 tcc1 double mutant (HGC114).
Direct-colony PCR and genomic PCR were performed to verify the strain
construction in each step.
(viii) TCC1 deletion series. To generate plasmids expressing the full length of Tcc1p, a DNA fragment was amplified by primers TCC1-FL-5' and TCC1-FL-3', using the TUA4 chromosome as a template, digested with PstI and XhoI, and cloned into the PstI-XhoI sites of p6HF-MET3, yielding p6HF-MET3-TCC1. To generate plasmids expressing a deletion series of Tcc1p, a DNA fragment corresponding to amino acid positions 1 to 475 or 1 to 250 of Tcc1p was amplified by primers TCC1-FL-5' and TCC-N1-3' or by TCC1-FL-5' and TCC1-N2-3', using the p6HF-MET3-TCC1 plasmid as a template, digested with PstI and XhoI, and cloned into the PstI-XhoI sites of p6HF-ACT1, yielding p6HF-ACT1-TCC1-N1 or p6HF-ACT1-TCC1-N2, respectively. A DNA fragment corresponding to amino acid positions 251 to 736 or 476 to 736 of Tcc1p was amplified by primers TCC1-C1-5' and TCC-FL-3' or by TCC1-C2-5' and TCC1-FL-3', using the p6HF-MET3-TCC1 plasmid as a template, digested with PstI and XhoI, and cloned into the PstI-XhoI sites of p6HF-MET3, yielding p6HF-MET3-TCC1-C1 or p6HF-MET3-TCC1-C2, respectively.
To generate Ura-auxotrophic strain iTUP1-HA-ARG4 as a host for the TCC1 deletion series, a DNA fragment encoding a 3xHA tag and an ARG4 marker was amplified by primers iTUP1-5' and iTUP1-3', using p3HA-ARG4 as a template, and introduced into TUA4. To verify the strain construction, direct-colony PCR was performed, after which the nucleotide sequence of the PCR fragment was confirmed. Plasmid p6HF-ACT1, p6HF-MET3-TCC1, p6HF-ACT1-TCC1-N1, p6HF-ACT1-TCC1-N2, p6HF-MET3-TCC1-C1, or p6HF-MET3-TCC1-C2 was introduced into iTUP1-3HA-ARG4 to yield DelTCC1VEC, DelTCC1-W, DelTCC1-N1, DelTCC1-N2, DelTCC1-C1, or DelTCC1-C2, respectively.
(ix) Epitope tagging. To tag CaTup1p with the HF epitope in the genomic locus, a DNA fragment containing the 3' region of TUP1, the HF tag sequence, the ACT1 terminator, the URA3 marker, and the downstream region of TUP1 was amplified by PCR with primers iTUP1-5' and iTUP1-3', using p6HF-ACT1 (16) as a template, and introduced into TUA4 to generate iTUP1-HF. Similarly, in order to tag Tcc1p or CaSsn6p with the HF epitope, primers iTCC1-5' and iTCC1-3' or primers iSSN6-5' and iSSN6-3' were used for the amplification of fragTCC1-HF or fragSSN6-HF to yield the iTCC1-HF or iSSN6-HF DNA cassette, respectively.
To generate strain iTCC1-HA, a DNA fragment containing a 3xHA tag, an ACT1 terminator, and a URA3 marker was amplified by primers iTCC1-5' and iTCC1-3', using p3HA-ACT1 as a template, and introduced into TUA4.
To generate strain iSSN6-HA, a DNA fragment containing a 3xHA tag, an ACT1 terminator, and a URA3 marker was amplified by primers iSSN6-5' and iSSN6-3', using p3HA-ACT1 as a template, and introduced into TUA4.
To generate strain iSSN6-Myc, a DNA fragment containing a 3xMyc tag, an ACT1 terminator, and an SAT1 marker was amplified by primers iSSN6-5' and iSSN6-3', using p3Myc-SAT1 as a template, and introduced into TUA6. Nourseothricin-resistant clones were selected on YPD medium containing 200 µg/ml clonNAT (WERNER BioAgents, Germany).
Strain 3TAG-TTS was designed to simultaneously express three tagged proteins, CaTup1p-HF, Tcc1p-HA, and CaSsn6p-Myc. A DNA fragment containing a 3xHA tag and an ARG4 marker was amplified by primers iTCC1-5' and iTCC1-3', using p3HA-ARG4 as a template, and introduced into iTUP1-HF to yield 2TAG-TT. Next, a DNA fragment containing a 3xMyc tag and an SAT1 marker was amplified by primers iSSN6-5' and iSSN6-3', using p3Myc-SAT1 as a template, and introduced into 2TAG-TT to yield 3TAG-TTS.
Strain D2T06ACT1 was designed to
simultaneously express two tagged proteins, CaTup1p-HF and Tcc1p-HA, in
an ssn6 genetic background. The same PCR-amplified
DNA cassette as that used for the iTUP1-HF construction was introduced
into SSN603 to generate D2T05. The p6HF-TCC1 plasmid was digested with
PstI and XhoI and cloned into the PstI-XhoI sites of p3HA-SAT1 to yield
p3HA-SAT1-TCC1. The StuI-digested p3HA-SAT1-TCC1 plasmid was
transformed into D2T05 to yield D2T06ACT1, in which Tcc1p-HA is
expressed from the ACT1 promoter.
Strain TUP112 was designed to express Tcc1p-HA protein in a strain in which the CaTUP1 gene can be repressed in the presence of methionine and cysteine. To generate Ura-auxotrophic strain iTCC1-HA-ARG4 as a host for CaTUP1 depletion, a DNA fragment encoding a 3xHA tag and an ARG4 marker was amplified by primers iTCC1-5' and iTCC1-3', using p3HA-ARG4 as a template, and introduced into TUA4. Gene replacement of the first allele of CaTUP1 was performed in a manner similar to that described above. A primer set consisting of disTUP1-1, -2, -3, and -4 was used to create a disruption cassette, which was amplified using pUC18-SAT1 as a template. The first allele was replaced with the SAT1 marker in strain iTCC1-HA-ARG4 to yield TUP111. To construct pCaDis-TUP1, DNA fragment TUP1-A was amplified with primers TUP1-N and TUP1-SacI-3', DNA fragment TUP1-B was amplified with primers TUP1-SacI-5' and TUP1-400-3', and each DNA fragment was gel purified. A 400-bp portion of CaTUP1 containing an artificial SacI site was amplified by mixing the two DNA fragments (TUP1-A and TUP1-B) and the two primers (TUP1-N and TUP1-400-3'), digested by BamHI and SphI, and then cloned into the BamHI-SphI sites of pCaDis (7). The SacI-digested pCaDis-TUP1 was transformed into a heterozygous strain, TUP111, to generate TUP112.
To verify the strain construction, direct-colony PCR was performed, after which the nucleotide sequence of the PCR fragment was confirmed.
Yeast two-hybrid assay. The MATCHMAKER two-hybrid system 3 (Clontech) was used for the yeast two-hybrid assay.
A DNA fragment corresponding to amino acid positions 1 to 130 of CaTup1p was amplified by primers THTUP1-5' and THTUP1-3', using TUA4 chromosomal DNA as a template, digested with BamHI and PstI, and cloned into the BamHI-PstI sites of pGADT7 (Clontech), yielding pGADT7-TUP1-N.
A DNA fragment corresponding to amino acid positions 1 to 475 of the Tcc1p codon, optimized for S. cerevisiae, where two CUG codons were changed to TCG, was obtained by a three-step PCR. First, three DNA fragments were amplified using p6HF-TCC1-N as a template with primers THTCC1-5' and THTCC1-mut1-3', primers THTCC1-mut1-5' and THTCC1-mut2-3', and primers THTCC1-mut2-5' and THTCC1-3' to generate THTCC1-NT-A, THTCC1-NT-B, and THTCC1-NT-C, respectively. Each fragment was purified by agarose gel electrophoresis to prevent contamination of the template DNA. A second PCR was performed with two DNA fragments, THTCC1-NT-A and THTCC1-NT-B, and two primers, THTCC1-5' and THTCC1-mut2-3', in the same tube to generate THTCC1-NT-AB. A third PCR was performed with two DNA fragments, THTCC1-NT-AB and THTCC1-NT-C, and two primers, THTCC1-5' and THTCC1-3'. A resultant DNA fragment containing the substitutions at Ser182 and Ser258 (changed from CTG to TCG) was digested with EcoRI and PstI and cloned into the EcoRI-PstI sites of pGBKT7 (Clontech) to generate pGBKT7-TCC1-N.
As shown in
Fig. 4, both pGBKT7 and
pGADT7 derivative plasmids were simultaneously introduced into S.
cerevisiae AH109 (Clontech), and the transformants were checked
for the expression of MEL1, which encodes
-galactosidase, according to the manufacturer's protocol.
Plasmids pGADT7-T and pGBKT7-53 were supplied as positive
controls in the MATCHMAKER two-hybrid system 3
(Clontech).
|
Subcellular fractionation. Cells were harvested and treated to obtain spheroplasts with Zymolyase 100T in Zymolyase buffer (50 mM Tris HCl [pH 7.5], 10 mM MgCl2, 1 M sorbitol, 1 mM dithiothreitol) at 30°C for 40 min with mild shaking. The spheroplast suspension was introduced drop by drop into a beaker containing Ficoll buffer (18% [wt/vol] Ficoll-400, 10 mM Tris HCl [pH 7.5], 20 mM KCl, 5 mM MgCl2, 3 mM dithiothreitol, 1 mM EDTA). The diluted solution was centrifuged at 20,000 x g for 20 min at 4°C, and the supernatants were used for Western analysis as a cytoplasmic fraction. The resultant pellets were resuspended in Ficoll buffer in a volume equal to the supernatant and used for Western analysis as a nuclear fraction.
Nucleotide sequence accession number. The newly determined sequence for TCC1 was deposited in GenBank under accession number AB252688 [GenBank] .
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Tcc1p and CaSsn6p interact independently with CaTup1p. To confirm that CaTup1p and Tcc1p bind to each other and to analyze the relationships among CaTup1p, Tcc1p, and CaSsn6p, we constructed a strain in which CaTup1p, Tcc1p, and CaSsn6p were tagged with disparate epitope tags (His6-FLAG, 3xHA, and 3xMyc, respectively) and performed immunoprecipitation, followed by Western blotting. We also constructed three strains, in each of which His6-FLAG-tagged CaTup1p (iTup1p-HF), 3xHA-tagged Tcc1p (iTcc1p-HA), or 3xMyc-tagged CaSsn6p (iSsn6p-Myc) was expressed under the control of each native promoter. Western analysis confirmed that each protein was expressed with each epitope tag as expected (Fig. 2A, lanes 1 to 4). Cell extracts from each strain were then subjected to tandem affinity purification. Western blotting using anti-FLAG, anti-HA, and anti-Myc antibodies with the immunoprecipitant from each strain demonstrated that Tcc1p-HA and CaSsn6p-Myc were detected in the fraction of strain 3TAG-TTS (Fig. 2A, lane 5), indicating that the CaTup1p binds to Tcc1p and CaSsn6p. When a similar immunoprecipitation using anti-HA antibody agarose was performed, only CaTup1p-HF and Tcc1p-HA were detected and CaSsn6p-Myc was below the detectable level (Fig. 2A, lane 9). Similarly, an immunoprecipitation experiment using anti-Myc antibody agarose showed interaction between CaTup1p-HF and CaSsn6p-Myc and no interaction of Tcc1p-HA (Fig. 2A, lane 13). These reciprocal experiments indicated that CaTup1p-Tcc1p and CaTup1p-CaSsn6p were independent complexes.
|
cells expressing Tcc1p-HA and CaTup1p-HF (Fig.
2B). Western blotting
using anti-HA antibody revealed that Tcc1p-HA was copurified with
CaTup1p-HF, even in the absence of CaSSN6 (Fig.
2B, lane 3), suggesting
that CaSsn6p might have no effect on CaTup1p-Tcc1p complex
formation.
The TPR domain of Tcc1p contributes to interaction with CaTup1p.
To determine which region in Tcc1p is
necessary for CaTup1p binding, we used an immunoprecipitation
experiment and a yeast two-hybrid system. For immunoprecipitation, a
series of His6-FLAG-tagged deletion mutants of the Tcc1p
protein were expressed from the ACT1 or
MET3 promoter in the C. albicans cells that contain
HA-tagged CaTup1p. By performing TAP, the Tcc1p mutant that contains
TPR motifs corresponding to the amino acid positions 1 to 475 (Tcc1-N1)
or 1 to 250 (Tcc1-N2) was detected as a protein interacting with
HA-tagged CaTup1p (Fig.
3B, lanes 3 and 4, respectively). CaTup1p-HA was not
detected in the immunoprecipitated complex with the C-terminal domains
of Tcc1p, corresponding to amino acid positions 251 to 736 (Tcc1-C1)
and 476 to 736 (Tcc1-C2), although only a small amount of the Tcc1p
mutant was present (Fig.
3B, lanes 5 and 6). Even
when three times as much crude extract as indicated in Fig.
3B was used for
purification, no signals of CaTup1p-HA were detected (data not shown).
The faint signals of the Tcc1-N2, Tcc1-C1, and Tcc1-C2 deletion mutants
could possibly be due to protein instability. These results suggest
that the first two TPR domains of Tcc1p might serve an important role
in binding to CaTup1p, although the possibility that the C-terminal
glutamine-rich region can bind to CaTup1p could not be denied. To
confirm the importance of the TPR domain in the CaTup1p-Tcc1p
interaction in vivo, we designed a yeast two-hybrid system. The effect
of the repressive activity of CaTup1p was avoided by using the portion
of CaTup1p comprising amino acids 1 to 130, which contains no WD
repeats, because the full-length CaTup1p fused to the Gal4 DNA-binding
domain itself could repress the reporter activity. The CaTup1p mutant
without this N-terminal portion could not bind to Tcc1p in the
above-described immunoprecipitation experiment (our unpublished
results). It is important to note that TCC1 contained two CUG
codons, which encode serine in C. albicans but leucine in
S. cerevisiae
(25), in the
sequence coding for amino acid positions 1 to 475. To express
TCC1 functionally in S. cerevisiae, we changed both
these codons into TCG by PCR-mediated mutagenesis, and the N-terminal
portion of codon-optimized Tcc1p was fused to the Gal4-activating
domain. When the Gal4-binding-domain-fused N-terminal portion of
CaTup1p (amino acids 1 to 130) and the Gal4-activation domain-fused
N-terminal portion of Tcc1p (amino acids 1 to 475) were simultaneously
expressed in a host strain, AH109, that possesses an
-galactosidase MEL1 gene as a reporter, the
constructed strain took on a blue color on agar medium
containing X--Gal
(5-bromo-4-chloro-3-indolyl--D-galactopyranoside)
(Fig. 3C), indicating an
in vivo protein-protein interaction. Integrating the results of
immunoprecipitation and the yeast two-hybrid assay suggests that the
TPR domain of Tcc1p and the N-terminal glutamine-rich domain of CaTup1p
contribute to Tcc1p-CaTup1p
binding.
|
To examine the cellular localization of Tcc1p, CaTup1p, or CaSsn6p, each protein was tagged with three tandem repeats of HA at the C terminus, and cells expressing HA-tagged proteins were immunostained with anti-HA antibody. CaTup1p-HA and CaSsn6p-HA localized in the nucleus as expected (Fig. 4B). Immunostaining with anti-HA antibody also demonstrated nuclear localization of Tcc1p (Fig. 4B). Under conditions that induce hyphae, while obvious nuclear localization of CaTup1p-HA and CaSsn6p-HA was observed, no signal of Tcc1p-HA was detected (data not shown), suggesting that the accumulation of Tcc1p in the nucleus might be dependent on morphology. To confirm the nuclear localization biochemically, subcellular fractionation was performed. A successful fractionation was verified by Western blotting probed with anti-histone H4 antibody. Tcc1p-HA was detected in both the nuclear fraction and the cytoplasmic fraction (Fig. 4C). To investigate whether the accumulation of Tcc1p in the nucleus depends on CaTup1p, HA-tagged Tcc1p was expressed in a CaTUP1 conditional mutant, in which CaTUP1 lies under the regulation of the MET3 promoter, and subcellular fractionation was performed. The repression of CaTUP1 in the presence of methionine and cysteine was verified by microscopic observation and quantitative RT-PCR of CaTUP1 and ECE1 mRNA (data not shown). In the presence of methionine and cysteine (with the MET3 promoter off), CaTUP1 expression was 3% of that of the wild type and, as a consequence, a hypha-specific gene, HYR1, was elevated significantly. Even if the expression of CaTup1p was depressed, Tcc1p-HA was detected in both the nucleus and the cytoplasm (data not shown), indicating that the nuclear localization of Tcc1p may not depend on CaTup1p.
tcc1 disruptant shows cell elongation phenotype.
To investigate the cellular functions
of Tcc1p in C. albicans, we deleted both copies. If Tcc1p
functions as a CaTup1p-mediated transcriptional repressor, phenotypes
of the tcc1 mutant would at least partly overlap
those of the Catup1 mutant. The two copies of
TCC1 were sequentially replaced with ARG4 and a
200-bp portion of hph (hph200) in strain TUA4. The
ability to generate a viable tcc1/tcc1 null mutant
strain indicates that TCC1 is not an essential gene in C.
albicans. There are no significant differences between the wild
type and the tcc1 mutant in their susceptibilities to
fluconazole, calcofluor white, sodium chloride, or hydrogen peroxide
(data not shown), indicating that TCC1 might not play an
important role in drug or stress resistance. To confirm that the loss
of the TCC1 function was responsible for any of the observed
phenotypes, a PCR-amplified fragment containing a TCC1
complete open reading frame was used to replace the hph200
locus of the tcc1 mutant TCC103 to generate the
reconstituted strain TCC107, in which Tcc1p is tagged with
His6-FLAG at the C terminus. As a negative control strain, a
PCR-amplified fragment containing only a 100-bp C-terminal portion of
the TCC1 gene was used to replace the hph200 locus of
TCC103 to generate null mutant TCC106. Both the null mutant TCC106 and
the reconstituted strain TCC107, which have a single copy
of URA3 at their respective TCC1 loci, were used for
the following experiments to compare phenotypes of morphology and
virulence.
The effect of the TCC1 deletion on the
phenotype of the tcc1 mutant was studied under
conditions that promote both yeast and hyphal growth in C.
albicans (Fig.
5A). On a serum agar medium that induces hyphal growth, there were no
significant differences among the wild type, the tcc1
null mutant, and the revertant. On a YPD agar medium adjusted to pH 5,
under conditions in which the wild type and the revertant exhibit
smooth colonies and the Catup1 and
Canrg1 mutants should show filamentous growth, as
reported previously (4,
5,
19), the TCC1
disruptant exhibited rough colonies, indicating more filamentous growth
in the disruptant. The filamentous phenotype of the
tcc1 mutant was enhanced by an alkaline condition. In
addition, on Spider medium that induces filamentation, the growth zone
of the tcc1 mutant that indicates filamentation was
larger than that of the wild type. These observations suggest that the
deletion of TCC1 may enhance the filamentation of C.
albicans under the conditions examined in this
study.
|
cells, whereas the length of
tcc1 cells was shorter than the previously reported
lengths of Catup1 and Canrg1 cells
(4,
5,
19). Calculation of the
axial growth rates (length/width) supported the fact that
tcc1 cells exhibited the cell elongation phenotype
during yeast growth (TUA6, 1.18 ± 0.31 [n
= 204]; TCC106, 2.47 ± 0.85 [n = 221];
and TCC107, 1.44 ± 0.26 [n = 206]).
Additionally, tcc1 cells did not exhibit a
Catup1-like constitutive filamentation without
constriction. There were no significant differences between the wild
type and the tcc1 mutant grown in Spider medium or in
serum medium, which induces hyphal growth. The phenotypes described
above were restored in the reconstituted strain TCC107, indicating that
filamentous phenotypes were caused solely by the deletion of the
TCC1 gene.
Effects of TCC1 disruption on the transcription of hypha-specific genes.
To determine whether
the transcription of HSGs is induced under yeast growth conditions by
TCC1 disruption as well as by CaTUP1 or
CaNRG1 disruption, we compared the expression levels of HSGs
such as HWP1
(28), ECE1
(2), HYR1
(1), and HGC1
(34) by Northern
hybridization and quantitative real-time RT-PCR. If Tcc1p served as a
transcriptional repressor for HSGs, as does CaTup1p, HSGs would be
upregulated by Tcc1p depletion. The wild type, the disruptant, and the
reconstituted strain were cultured under yeast or hyphal growth
conditions and subjected to Northern blotting analysis (Fig.
6A). A comparison of yeast growth (Fig.
6A, lane 1) and hyphal
growth (Fig. 6A, lane 2 or
3) confirmed the successful detection of HSGs. There were no
significant differences between the wild type and the
tcc1 disruptant in the HWP1, the
ECE1, or the HGC1 gene expression levels, based on
Northern analysis. The expression of HYR1 was decreased by
TCC1 deletion under hyphal growth conditions, consistent with
previous data indicating that HYR1 mRNA in the
Catup1 mutant under such conditions was lower than
that in the wild type
(15). This implies that
the CaTup1p-Tcc1p complex might regulate the activation or repression
of a transcriptional repressor in the HYR1 gene during
filamentous growth. However, Northern analysis indicated that the yeast
growth conditions did not seem to induce any significant HSGs.
|
mutant was less than that of the
Catup1 or Canrg1 mutant (Fig.
5). Therefore, since it
remains a possibility that derepression by TCC1 deletion might
be weaker than that by CaTUP1 or CaNRG1 deletion, we
attempted to detect the expression of HSGs by quantitative real-time
RT-PCR, which has a higher sensitivity than Northern analysis. Compared
to the results reported so far, the HWP1 and ECE1
mRNA were derepressed in Catup1 and
Canrg1 cells but not in the Cassn6 mutant
(Fig. 6B), consistent with
the report by Garcia-Sanchez et al.
(12). Interestingly,
HYR1 mRNA and HGC1 mRNA were derepressed in
Catup1 and Canrg1 cells and even in
the Cassn6 cells. However, Garcia-Sanchez et al.
(12) reported that
HYR1 mRNA was not elevated in Canrg1 and
Cassn6 cells and that HGC1 mRNA was not
elevated in Cassn6 cells. These discrepancies are
probably caused by an experimental difference between microarray
analysis and quantitative RT-PCR. These results indicate that
HWP1/ECE1 and HYR1/HGC1 might be
transcriptionally regulated in a different manner and that CaSsn6p may
regulate HYR1 and HGC1 negatively. Moreover,
all HSGs tested were derepressed by the deletion of
TCC1, but the expression level of the TCC1 mutant was
lower than that of the Catup1 or
Canrg1 mutant (Fig.
6B). Therefore, these
results further support that Tcc1p, accompanied by CaTup1p, might
function as a transcriptional repressor involved in the morphological
transition of C. albicans.
The filamentous phenotype of the tcc1 mutant is restored by HGC1 disruption.
In the previous paragraph, we
demonstrated that Tcc1p might repress the expression of the
hypha-specific G1 cyclin Hgc1p under conditions that induce
yeast growth. The concept that HGC1 acts downstream of CaTup1p
regulation has been supported by the previous report that constitutive
filamentation is blocked by the deletion of HGC1
(34). To examine whether
Tcc1p is also linked to the function of HGC1, we deleted
HGC1 together with TCC1. The observation that the
hgc1 single mutant was deficient in hyphal formation
(Fig.
7) was consistent with a previous report
(34). The double mutant
did not show filamentation on a serum medium (Fig.
7), in that the deletion
of TCC1 did not induce filamentation of hgc1
cells and the deletion of HGC1 reduced filamentation of
tcc1 cells. On a YPD agar medium, the deletion of
HGC1 eliminated the filamentous phenotype of
tcc1 cells as well as that of
Catup1 cells. This reinforced the idea that CaTup1p
and Tcc1p might coordinate in regulating the function of HGC1,
followed by the repression of hyphal
formation.
|
disruptant.
CaTUP1 or
CaNRG1 disruption causes reduced virulence in a mouse model of
systemic candidiasis
(19). To determine
whether Tcc1p plays an important role in pathogenicity, mice were
intravenously injected with the wild type (TUA6), the
tcc1 null mutant (TCC106), and the revertant (TCC107)
and monitored for survival. We found that mice infected with
106 CFU of the tcc1 mutant did not die
until 31 days postinfection, with a sustained-survival
curve, whereas the same-size inocula of the wild-type or the
reconstructed strain killed all infected mice within 11 days (Fig.
8). The mean survival times for the mice infected with the wild type, the
null mutant, and the revertant were 7.5 ± 3.78 days, 21.2
± 8.07 days, and 9 ± 2.35 days, respectively (TCC106
versus TUA6, P < 0.01; TCC106 versus TCC107,
P < 0.02). This demonstrates that TCC1 is
involved in C. albicans virulence and supports, in part, the
concept that Tcc1p might function with CaTup1p in C.
albicans.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To identify a binding partner that
actually regulates the morphology of C. albicans together with
CaTup1p, we purified a protein complex including CaTup1p. One of the
components contained in the CaTup1p complex was identified as
TCC1, a Tup1p-complex component. In
addition, we demonstrated that CaTup1p independently interacted with
Tcc1p or CaSsn6p. However, Tcc1p and CaSsn6p share common properties
and behaviors as described below. Their first similarity is the
existence of a TPR domain: Tcc1p contains four copies of the TPR motif,
and nine TPRs are located within the C. albicans Ssn6p
polypeptide. Since both Tcc1p and CaSsn6p possess the TPR motif, Tcc1p
probably interacts with CaTup1p in a mode similar to that of CaSsn6p.
Actually, we demonstrated that the TPR domains of Tcc1p are responsible
for CaTup1p binding by using immunoprecipitation and yeast two-hybrid
analysis (Fig. 3). Their
second behavior in common is their expression profiles. The expression
levels of Tcc1p and CaSsn6p, which were tagged with different epitopes
in one cell, were altered in response to cell cycle toxins and glucose
depletion: the expression levels reached a peak when cells were
arrested at the G2/M phase and decreased to an undetectable
level in the unbudded G1 cells (Fig.
4A). In addition, the
complex formation with CaTup1p in response to cell cycle toxins was no
different between Tcc1p and CaSsn6p (data not shown). Their third
similarity is localization to the nucleus. Immunostaining with anti-HA
antibody demonstrated that Tcc1p and CaSsn6p localize to the nucleus
(Fig. 4B), although both
were predicted to contain no nuclear localization signals. Whereas
subcellular fractionation further supports the nuclear localization of
Tcc1p, Tcc1p was also detected in the cytoplasmic fraction. Tcc1p
localization to the nucleus was not altered in a
CaTUP1-depleted cell, suggesting that Tcc1p localization may
not depend on CaTup1p. Since the nuclear localization of Tcc1p was not
observed in hyphal cells, shuttling between the cytoplasm and the
nucleus would be regulated by some signaling associated with
morphology. Despite these common characteristics, Tcc1p and CaSsn6p
seem to function differently: the phenotypes of the mutants and the
transcriptional control of HWP1 and ECE1 were
contradictory between Tcc1p and CaSsn6p. One of the reasons that the
functions of Tcc1p and CaSsn6p have different effects is assumed to be
because they might require disparate sequence-specific DNA-binding
proteins. Nevertheless, derepression of HYR1 and HGC1
mRNA in tcc1 and Cassn6 cells might
indicate the existence of a common DNA-binding protein. Future global
transcriptional analysis using a DNA microarray might give clues for
solving the question of why the depletion of each protein resulted in
opposing phenotypes, despite the common binding partner, the similar
expression profiles, and the nuclear localization. Also, study of a
double deletion of TCC1 and CaSSN6 would explain how
the two products share roles.
The
Catup1 null mutant demonstrates constitutive
filamentation even under conditions that induce the yeast form
(4). The
Cassn6 homozygous mutant shows a higher rate for
phenotype switching (12)
and no filamentation on serum or Spider agar medium (data not shown).
When both alleles of TCC1 identified in this study were
deleted, the disruptant grew in a pseudohyphal form under conditions
that induce the yeast form. Obviously, the morphological phenotype of
tcc1 is more similar to that
of Catup1 than to that of
Cassn6 from the viewpoint of the cell elongation
phenotype. Taken together with the results of the
immunoprecipitation experiment and morphological
observation, it is highly possible that the protein complex consisting
of CaTup1p and Tcc1p might behave independently of
CaTup1p-CaSsn6p. However, the filamentous phenotype of the
tcc1 mutant was not as severe as that of the
Catup1 or Canrg1 mutant, and the
tcc1 mutant showed a relatively mild cell elongation.
This reduced severity is probably associated with the degrees of HSG
repression. For example, the elevation of derepressed HSG transcription
by the deletion of TCC1 was not stronger than that in the
Catup1 or Canrg1 mutant (Fig.
6). In addition, although
we did not compare them directly, the virulence demonstrated by the
tcc1, Catup1, or
Canrg1 mutant indicates a decrease or attenuation:
all tcc1 mutant-infected mice were killed within 31
days (Fig. 8), while
almost no mice inoculated with the strain from which CaTUP1 or
CaNRG1 was deleted were dead within the time period examined
(4,
5,
19). These results
suggest that the degree of cell elongation, the increased quantity of
HSG transcription, and the attenuated virulence, which are caused by
gene disruption, are closely connected to each other. While the
Catup1 mutant was demonstrated to be more sensitive
to hydrogen peroxide than the wild type
(18), the
tcc1 mutant did not show significant differences in
drug or stress resistance (data not shown), enhancing the significance
of the effect of Tcc1p on morphogenesis. Moreover, deletion of the
CaTup1p-binding partner gene, TCC1, did not result in more
severe filamentation than did deletion of TUP1, suggesting
that a relationship of CaTup1p and Tcc1p is not as essential for
transcriptional repression but that Tcc1p would play a supportive role
for the CaTup1p function.
We have identified Tcc1p as
a novel factor that interacts with CaTup1p. No homolog of Tcc1p could
be identified in S. cerevisiae, suggesting the possibility
that Tcc1p has C. albicans-specific functions. Considering the
phenotypes of the tcc1 mutant, it would not be
unreasonable to think about a relationship with CaNrg1p, a DNA-binding
protein that represses hypha-specific genes during the yeast phase.
Evidence that proves direct binding between Tup1p and Nrg1p in C.
albicans has not yet been reported. Based on genes regulated by
TCC1, a hypothesis is proposed that DNA-binding protein
CaNrg1p recruits a Tcc1p-CaTup1p protein complex to promoter regions of
hypha-specific genes, including HGC1, to repress transcription
of these genes. A mechanical relationship among CaTup1p, Tcc1p, and
CaNrg1p remains unproven. We are now investigating the relationship
between CaNrg1p and CaTup1p-Tcc1p, although it is difficult to deal
with the CaNrg1p protein because of its low-level expression and
fragility (our unpublished results).
To conclude, we
have presented a deletion analysis of TCC1, a novel C.
albicans gene encoding a protein with TPR motifs. Tcc1p forms a
protein complex with CaTup1p, although direct binding between the two
proteins remains unproven. Many genes have been reported to date whose
deletion results in constitutive filamentation similar to that in the
Catup1 mutant. An analysis of detailed relationships,
including the transcriptional regulation of, or the
protein-protein interaction resulting from, these genes that maintain
the yeast form, including Tcc1p, will provide a clue to the elucidation
of the mechanisms of the yeast-hypha transition and virulence in C.
albicans.
|
ACKNOWLEDGMENTS |
|---|
A. Kaneko was supported by the Cooperative System for Supporting Priority Research, Japan Science and Technology Corporation. This work was supported in part by grant KH53315 from the Japan Health Sciences Foundation. It was also supported by Health Science Research Grants for Research on Emerging and Re-emerging Infectious Diseases of the Ministry of Health, Labor and Welfare of Japan.
|
FOOTNOTES |
|---|
Published ahead of print on 22 September 2006.
Supplemental
material for this article may be found at http://ec.asm.org/.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2.Mol. Cell. Biol.
17:6023-6028.[Abstract]
