Candida albicans is an important fungal pathogen of immunocompromised patients, causing superficial as well as invasive systemic disease. Virulence factors include adhesion to epithelial and endothelial cells, production of extracellular proteinases, and the ability to switch between yeast-form and filamentous morphologies (Odds, 1994). The study of the molecular mechanisms of these processes has been hindered by the fact that C. albicans reads the CUG codon as serine rather than leucine, rendering heterologous gene expression largely unsuccessful (De Backer et al., 2000).
Green fluorescent protein (GFP) is widely recognized as a powerful tool in cell biology and serves as an important reporter for monitoring localization and expression of proteins in many organisms (Chalfie et al., 1994; Niedenthal et al., 1996; Valdivia et al., 1996). The GFP variants yellow fluorescent protein (YFP) and cyan fluorescent protein (CFP) were engineered to improve brightness, provide additional wavelengths of emission, and facilitate multicolour imaging of differential gene expression and protein localization (Ellenberg et al., 1998; Heim et al., 1994; Heim and Tsien, 1996).
GFP sequences engineered with codons optimized for expression in C. albicans have been expressed either from a plasmid or from an integrated chromosomal locus (Cormack et al., 1997; Morschhauser et al., 1998). Both of these constructs employed conventional cloning techniques to generate a promoter–GFP fusion that could be transformed into C. albicans. In C. albicans, plasmids are highly unstable, variable in copy number from cell to cell, and rapidly lost in the absence of selective pressure (Cannon et al., 1992; Kurtz et al., 1987; Pla et al., 1995). Thus, integrated single copy reporters are desirable. For example, a chromosomally integrated GFP construct expressed from the SAP2 promoter was expressed only under conditions that promote Sap2 proteinase expression (Morschhauser et al., 1998).
In C. albicans, gene deletion by homologous recombination is now a standard technique (Fonzi and Irwin, 1993). Recently, PCR-mediated strategies that allow rapid, efficient gene disruption or deletion have been developed (Wilson et al., 2000, 1999). These methods use PCR primers with 5′-ends that correspond to the desired target gene sequences, and 3′-ends that direct amplification of the selectable marker gene. The amplified DNA is transformed directly into C. albicans, and recombinants that carry the inserted marker at the locus of interest are identified.
In this work, we combined PCR-mediated gene insertion, as developed for C. albicans, along with site-directed mutagenesis of GFP to construct codon-optimized GFP variants. We generated a set of plasmid cassettes that direct one-step construction of fluorescent protein fusions in C. albicans. We also demonstrate the use of these cassettes to generate strains expressing two distinguishable proteins in the same C. albicans cells.
Materials and methods
Strains, growth conditions and DNA methods
All strains were derived from BWP17 (ura3::imm434/ura3::imm434 his1::hisG/his1::hisG arg4::hisG/arg4::hisG) (Wilson et al., 1999) and were grown at 30°C in rich (YEPD) medium, synthetic complete medium, or synthetic complete medium lacking specific nutrients (Sherman, 1991). Escherichia coli strain XL1-blue (Stratagene, La Jolla, CA) and standard media and methods (Ausubel et al., 1995) were used for plasmid manipulations. Yeast genomic DNA was isolated according to the method of Hoffman and Winston (1987). Polyacrylamide gel electrophoresis-purified oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA).
DNA for plasmid constructions and yeast transformation was generated by PCR using the Qiagen Taq DNA polymerase kit (Qiagen). The 50 µl reaction mix contained 3.5 mM MgCl2, 5 µl 10× kit buffer, 0.4 mM each dNTP, 10 µg BSA, 0.6 µM each primer, 2.5 U Taq DNA polymerase, and 0.1 µg plasmid template. Reactions were run for 1 cycle of 4 min at 94°C, 25 cycles of 1 min at 94°C, 1 min at 55°C, and 1 min/kb of the desired product at 72°C.
Transformants were screened by PCR as above, except using 2 µM each of primer and 1 µl (∼0.5 µM) of yeast genomic DNA.
Construction of GFP variants YFP and CFP
pMG871, containing the C. albicans codon-optimized GFP sequence in pYEGFP3 (Cormack et al., 1997) ligated to HindIII/BamHI-digested pUC119, was used to produce the template for site-directed mutagenesis (Kunkel, 1985; Kunkel et al., 1987). The oligonucleotide primers used togenerate pMG1416, which encodes YFP, were 5′-TCTAGCAAAACATTGTAAACCATAACCGAAAGT-3′ (V68L) and 5′-GGATAAGGCAGATTGGTAGGATAAGTAATGGTTGTC-3′ (T203Y). The oligonucleotides used to generate pMG1683, which encodes CFP, were 5′-AACACCCCAACCCAAAGTAGTGACTAA-3′ (F64L), 5′-ACATTGAACACCCCAACCGAAAGTAGT-3′ (Y66W), 5′-CATTGTGAGAAATATAGTTGTATTCC-3′ (N146I),5′-GTTTGTCAGCAGTGATGTAAACATTG-3′ (M153T), 5′-CTAATTTTGAAGTTAGCTTTGATACCATTC-3′ (V163A) and 5′-TCTAATTTTGAAGTGAGCTTTGATACC-3′ (N164H).
To generate pMG1506, which includes GFP and URA3 sequences, the BamHI–PvuI fragment of plasmid pMG871 was ligated to the BclI–PvuI URA3 fragment of pVEC (gift of P. and B. Magee, University of MN). To introduce the C.albicans ADH1 terminator after GFP, the ADH1 terminator in pYPB1–ADHp1 (Bertram et al., 1996) was isolated as a PstI–EcoRI fragment and ligated to PstI–EcoRI-cut pMG1506, between the GFP and URA3 sequences, to generate pGFP–URA3 (Figure 1).
To generate pGFP–HIS1, the C. albicans HIS1 sequence in pGEM–HIS1 (Wilson et al., 1999) was amplified by PCR using the following primers, which were engineered to contain EcoRI sites (shown in bold): forward, 5′-GGACCGGAATTCCGGGGATCCTGGAGGATGAGGAGA-3′; reverse,5′-GGACCGGAATTCCGGAATATTTATGAGAAACTATCA-3′ The resulting PCR product was digested with EcoRI and ligated to EcoRI-digested pGFP–URA3 to generate pGFP–HIS1 (Figure 1).
Using analogous approaches, the YFP and CFP sequences contained in pMG1416 and pMG1683, respectively, were used to generate pYFP–URA3, pYFP–HIS1, pCFP–URA3 and pCFP–HIS1 (Figure 1).
Transformation of C. albicans and identification of integration events
PCR was performed using one of the cassettes shown in Figure 1 as the template, and the appropriate target-gene-specific primer pairs were designed as indicated in Figure 1 and Table 1. The products from 10 PCR reactions were pooled, extracted once with chloroform, precipitated with ethanol, resuspended in 20 µl of water and used to transform C. albicans strain BWP17 (Wilson et al., 1999). Transformants were selected by plating the transformation mix on the appropriate selective medium. To identify transformants in which the cassette had correctly integrated into the target gene sequence, genomic DNA was prepared and used as the template in PCR reactions, using one primer that annealed within the transformation module and a second primer that annealed to the target gene locus outside the altered region.
Table 1. PCR primers used to amplify the transformation cassettes
Primer combinations, orientations and locations relative to the plasmid templates are indicated in Figure 1. The gene-specific sequences included in the primers used in this study were approximately 70 nucleotides in length. To tag full-length proteins, the gene-specific sequences of the forward primer end just upstream of the stop codon, preserving the reading frame of the tag, whereas those of the reverse primer are chosen to end just downstream of the stop codon.
The reading frame is indicated by spacing within the sequence; the sequence encoding the glycine linker is shown in bold.
5′-ATAATGAAAAACCCAGACCTTTGTAATTAAAAAAATTTAAACATTAGCAACAAAGTAAGAACACGATCAA-3′+R1 or R2
Gene-specific sequences used to generate tag cassettes for CDC3 and TUB1
To tag CDC3 with GFP, YFP and CFP, CDC3-specific sequence was added to the universal primer sequences as described in Table 1. To tag TUB1 with GFP, YFP, and CFP, TUB1-specific sequence was added to the universal primer sequences as described in Table 1.
Differential interference contrast microscopy and epifluorescence microscopy were performed using a Nikon Eclipse E800 photomicroscope equipped with a 100 W mercury lamp, and epifluorescence illumination with green fluorescent protein (GFP) (excitation filter 470–490 nm, barrier 520–580 nm), blue fluorescent protein/cyan fluorescent protein (CFP) (excitation filter 380–400 nm, barrier 435–485 nm) and yellow fluorescent protein (YFP) (excitation filter 490–510 nm, barrier 520–550 nm) filter sets (Chroma Technology Corp., Brattleboro, VT). Digital images were collected using a CoolCam liquid-cooled, three-chip colour CCD camera (Cool Camera Company, Decatur, GA) and captured to a Pentium II 300 MHz computer, using Image Pro Plus version 4.1 software (Media Cybernetics, Silver Springs, MD). Images were processed using Adobe Photoshop version 5.5 (Adobe Systems Corp., San Jose, CA).
Results and discussion
PCR template plasmids for gene tagging
Cormack et al. (1997) developed a GFP sequence in which the 239 codons of the Aequorea victoria GFP gene (Chalfie et al., 1994) were optimized for expression in C. albicans. In addition to the codon optimization, two additional mutations were made (S65G and S72A) to enhance the brightness of the expressed protein. We modified this codon-optimized GFP sequence using the mutations necessary to effect colour changes, from green to yellow or from green to cyan, that were developed and described for GFP variants used in mammalian expression systems (Ellenberg et al., 1998; Heim et al., 1994; Heim and Tsien, 1996). Thus, we introduced the same mutations, by site-directed mutagenesis of the codon-optimized GFP sequence, to generate the C. albicans codon-optimized versions of YFP and CFP.
To facilitate the use of these sequences in tagging multiple genes differentially at the carboxy-terminus of their native genomic loci, we developed a set of PCR template cassettes, each containing one of the fluorescent proteins and one of two selectable markers, URA3 or HIS1 (Figure 1). Thus, by using a C. albicans strain with auxotrophies for ura3 and his1, two genes can be tagged within the same strain.
Validation of the system
We wanted to construct GFP, YFP and CFP fusion proteins to study gene expression and protein localization in C. albicans. To show that these fluorescent proteins are efficiently expressed and visualized in C. albicans, we used the cassettes described in Figure 1 to tag C. albicans CDC3 and TUB1. CDC3 encodes the homologue of the S.cerevisiae septin Cdc3, a protein that localizes to the mother-bud neck (Field and Kellogg, 1999; Longtine et al., 1996). TUB1 encodes the homologue of S. cerevisiae tubulin, a cytoskeletal protein that localizes to the mitotic spindle as well as to cytoplasmic microtubules (Schatz et al., 1988).
Cdc3–GFP was generated using the pGFP–URA3 cassette and the CDC3 forward and reverse primers listed in Table 1. C. albicans strain BWP17 was transformed with the PCR product and Ura+ transformants were verified by PCR. Fluorescence microscopy revealed that Cdc3–GFP localized at the mother-bud necks of yeast cells (Figure 2A) in a cell cycle-dependent manner, similar to the localization pattern reported for S.cerevisiae Cdc3. Of note, Cdc3–GFP was observed in 100% of cells withsmall or medium-sized buds. Similarly, when pYFP–URA3 or pCFP–URA3 were used as templates with CDC3 primers, we observed yellow or cyan fluorescent signal at the mother-bud neck of cells (Figure 2A). In addition, an abundant non-cell cycle-regulated protein was expressed and visualized in >96% of cells (data not shown).
We used GFP, YFP and CFP cassettes to also tag TUB1 in C. albicans, and observed the characteristic localization of mitotic spindles for all three tubulin-fluorescent protein fusions (Figure 2B and data not shown). Of note, the efficiency of homologous integration into the gene of interest, determined by PCR screening, was approximately 30–40% of the original transformants. Thus, the codon-optimized GFP, as previously shown by Cormack et al. (1997), as well as the codon-optimized YFP and CFP, are efficiently expressed and visualized in C. albicans when expressed within the genomic context of a particular gene.
To determine whether YFP and CFP fusion proteins could be localized simultaneously within the same cell, we generated a strain carrying both CDC3–CFP and TUB1–YFP by sequential transformation. Fluorescence microscopy using cyan- and yellow-specific filter sets detected a cyan signal at the mother-bud neck corresponding to Cdc3, and either a punctate or a linear yellow signal within the cell cytoplasm corresponding to the mitotic spindle prior to or during anaphase, respectively (Figure 2B). Thus, our codon-optimized YFP and CFP can be used simultaneously within the same cell to assess the expression of two proteins and/or the degree of co-localization of two proteins.
We have described a set of plasmids containing fluorescent protein cassettes that are useful as templates for PCR-mediated gene tagging in C.albicans. Use of these cassettes results in the generation of a protein tagged at the carboxy-terminus that can be visualized in vivo in C. albicans cells. Furthermore, the GFP epitope may provide a more efficient way to isolate and visualize proteins in C. albicans cell lysates using commercially available anti-GFP antibodies.
The codon-optimized GFP variants YFP and CFP are important tools that will facilitate the study of protein co-localization and co-expression in C. albicans. In addition, similar to what has been found for gene disruption strategies (Wilson et al., 2000, 1999), we found that PCR-mediated gene tagging is a faster and more efficient way to generate fusion proteins in C. albicans than using conventional cloning techniques. Our method of tagging genes at the native genomic locus offers the opportunity to do time-lapse studies of protein localization in vivo, and to analyse the expression of proteins from their native promoters. The ability to do time-lapse studies of protein localization and expression in C. albicans is a significant technological advance that will aid our efforts to understand the biological changes that are required for the process of morphogenesis.
We thank Brendan Cormack for providing the GFP sequence optimized for use in C. albicans, Aaron Mitchell and Pete and Bebe Magee for providing strains and/or plasmids, and Mark S. Longtine for helpful discussions. This work was supported by Grant No. 0677 from the Burroughs-Wellcome Fund (J.B.), NIH AI-25827 (J.B.), a March of Dimes Basil O'Connor Award (C.G.), and National Institutes of Health Grant No. 1 K08 AI01712-01 (C.G.).