Characterization of a multicopy family of genes encoding a surface-expressed serine endoprotease in rat Pneumocystis carinii.

A unique family of genes encoding serine endoproteases related to the Saccharomyces cerevisiae serine endoprotease kexin was identified in Pneumocystis carinii. Unlike previously described serine endoprotease genes that are single copies, multiple copies of the P. carinii serine endoprotease are distributed throughout the genome. The proteins predicted by these variant genes demonstrate sequence variability, but they retain the conserved active sites associated with endoprotease activity. The serine endoprotease was localized to the organism surface by immunohistochemical and immunofluorescence studies and to the electron lucent layer of the cyst wall by immunoelectron microscopy. The findings of multiple copies of the serine endoprotease gene in the P. carinii genome, and its localization to the cell surface, suggest that this protease plays an important role in the biology of P. carinii and that antigenic variation of the surface-expressed serine endoprotease may be a strategy for immune evasion. P. carinii serine endoprotease provides a novel target for chemotherapeutic and immune-based approaches to the treatment of P. carinii pneumonia.

The Cryptococcus neoformans STE12alpha gene: a putative Saccharomyces cerevisiae STE12 homologue that is mating type specific.

Cryptococcus neoformans possesses two mating types, MATalpha and MATa. Alpha-Cells are more virulent than a-cells and are also, unlike a-cells, capable of producing extensive hyphae in the haploid phase. The molecular analysis of hyphae production in C. neoformans has resulted in the identification of a gene which displays substantial similarity to other fungal STE12 genes, including the presence of a highly conserved homeodomain. Overexpression of the C. neoformans gene resulted in poor growth, altered morphology and the presence of hyphal projections, phenotypes reported in similar studies of the Saccharomyces cerevisiae STE12 gene. Overexpression was also found to induce MFalpha, a pheromone, and CNLAC1, a confirmed C. neoformans virulence gene. The C. neoformans STE12alpha gene, however, has one striking difference from other fungal STE12 genes; it is found only in alpha-cells. The existence of STE12alpha in C. neoformans suggests that this fungus has elements of a conserved MAP kinase cascade, which may be organized in a novel manner.

A single expression site with a conserved leader sequence regulates variation of expression of the Pneumocystis carinii family of major surface glycoprotein genes.

The major surface glycoprotein (MSG) of Pneumocystis carinii is encoded by a family of related but distinct genes distributed throughout the P. carinii genome. Previous reports of the genomic and mRNA MSG structure suggested that there was a highly conserved 5'-untranslated region and a highly variable translated region. In the current study, we demonstrate that there is a single expression site for MSG expression and that different MSG genes are located downstream of this expression site. Isolation of a genomic clone containing the putative 5'-untranslated region has demonstrated that there was a single base sequencing error in what was considered to be the untranslated region. The corrected sequence reveals an extended open reading frame encoding a constant amino-terminal leader domain, with a typical signal peptide, for the MSG protein family. Since this constant amino-terminal domain is encoded by a single copy genomic sequence, a recombination/gene conversion-mediated antigenic switching event is required to effect the known variability in expressed MSG sequences. Therefore, like some bacterial and protozoan pathogens, the opportunistic fungal pathogen P. carinii contains a constant genomic site dedicated to MSG expression and a switchable downstream region for the variable part of the MSG gene family.

Dimorphism and haploid fruiting in Cryptococcus neoformans: association with the alpha-mating type.

Cryptococcus neoformans is a major opportunistic fungal pathogen in AIDS and other immunosuppressed patients. We have shown that wild-type haploid C. neoformans can develop an extensive hyphal phase under appropriate conditions. Hyphae produced under these conditions are monokaryotic, possess unfused clamp connections, and develop basidia with viable basidiospores. The ability to undergo this transition is determined by the presence of the alpha-mating type locus and is independent of serotype. The association of the hyphal phase with the alpha-mating type may explain the preponderance of this mating type in the environment and the nature of the infectious propagule of C. neoformans.

The Cryptococcus neoformans GAL7 gene and its use as an inducible promoter.

A Cryptococcus neoformans galactose auxotroph was created by ultraviolet light mutagenesis and complemented with a C. neoformans genomic library. The translated sequence of the complementing DNA revealed a high degree of similarity to a number of UDP glucose-D-galactose-1-phosphate uridylyltransferases. Expression of C. neoformans GAL7 mRNA followed a pattern similar to Saccharomyces cerevisiae expression; it was first observed within 2.5 min of induction and fully induced by 30 min. The gene was completely repressed in the presence of glucose. The GAL7 promoter was isolated and used to construct a promoter cassette. Two genes were tested in this cassette for galactose regulation by creating GAL7 promoter fusions with their coding regions. MF alpha, which encodes a pheromone, was found to produce filaments only in transformants that were induced by galactose. A second gene, beta-glucuronidase (gusA), which is a commonly used reporter gene, was tested and also found to be expressed. When the GAL7p::GUS fusion was used to quantify inducibility of the GAL7 promoter, the level of enzyme activity was at least 500-fold greater for cells grown in galactose than for cells grown in glucose. The GAL7 promoter is the first inducible promoter characterized in C. neoformans and the GUS gene is the first heterologous gene shown to be expressed in this yeast pathogen.

Gene for an extracellular matrix receptor protein from Pneumocystis carinii.

An initial and crucial step in the establishment of many microbial infections is the attachment of the pathogen to the host cells. Thus, adherence of Pneumocystis carinii (Pc) to type I pneumocytes is believed to be important in the induction of Pc pneumonia. Little is known about the nature of the attachment of Pc to type I cells, although extracellular matrix (ECM) proteins, such as fibronectin and laminin, have been implicated in the process. We report here the isolation of a Pc gene encoding a receptor protein that binds both fibronectin and laminin in vitro. A cDNA clone encoding the Pc ECM receptor was isolated from a Pc cDNA library and identified on the basis of sequence homology to the human colon carcinoma laminin receptor. Southern blot analysis of Pc genomic DNA confirmed that the cDNA was of Pc origin. Northern blot analysis of Pc total RNA showed a predominant mRNA of approximately 1400 nucleotides that hybridized to the ECM receptor gene. The ECM receptor predicted from the cDNA sequence is 295 amino acid residues long, with a molecular mass of 32.8 kDa. The C-terminal third of the polypeptide is highly negatively charged, whereas the N-terminal two-thirds contains hydrophobic segments that may play a role in membrane association. Sequence analysis and alignment of the N terminus with the laminin receptor cDNA sequence of human colon carcinoma support the conclusion that the Pc ECM receptor cDNA clone is a full-length clone. A Western blot of the overexpressed ECM receptor protein bound both laminin and fibronectin in vitro. Antibodies raised to the overexpressed receptor protein interacted with a 33-kDa protein in total Pc cell lysates. These findings raise the possibility that the Pc ECM receptor protein may mediate the organism's attachment to type I pneumocytes and, thus, may play a crucial role in Pc pathogenesis.

Cloning, sequence analysis and expression of the gene encoding imidazole glycerol phosphate dehydratase in Cryptococcus neoformans.

A cDNA from Cryptococcus neoformans, encoding imidazole glycerol phosphate dehydratase (IGPD), was isolated by complementation of a his3 mutant strain of Saccharomyces cerevisiae. The C. neoformans HIS3 cDNA encodes an approx. 22-kDa protein with a high degree of amino-acid sequence similarity to IGPDs from ten other microorganisms, as well as Arabidopsis thaliana. Most striking are two conserved HHXXE regions and several conserved His, Asp and Glu residues. The cDNA was engineered for expression in Escherichia coli and an approx. 26-kDa protein was identified by SDS-PAGE. DNA and N-terminal sequence analyses confirmed that this protein was C. neoformans IGPD. Furthermore, IGPD assays of crude extracts from IGPD-producing E. coli cells demonstrated that the C. neoformans protein was catalytically active.

Melanin-deficient mutants of Cryptococcus neoformans.

Cryptococcus neoformans is a significant fungal pathogen in immunocompromised patients. The ability of C. neoformans to produce melanin has been correlated with virulence. The role of melanin in promoting virulence is unclear, although an anti-oxidant function has been suggested. To begin to define the genetic mechanisms responsible for melanin production in C. neoformans, we describe the isolation of seven melanin-deficient mutant classes. Some of the mutants can be suppressed by addition of Cu2+ to media, suggesting that the phenoloxidase of C. neoformans, like other fungal phenoloxidases, contains copper. Other mutants display a recessive sterile phenotype. A genetic and phenotypic characterisation of these mutants is presented.

A receptor tyrosine kinase found in breast carcinoma cells has an extracellular discoidin I-like domain.

We have identified a breast carcinoma tyrosine phosphoprotein, discoidin domain receptor (DDR), that defines an unusual class of receptor tyrosine kinases. The DDR cDNA predicts a C-terminal tyrosine kinase domain and an N-terminal domain similar to the Dictyostelium discoideum lectin discoidin I. These domains are connected by an extraordinary hydrophilic proline/glycine-rich domain, which is interrupted by a predicted transmembrane sequence. This extended proline/glycine-rich region may be required for an unusual geometry of interaction with ligand or substrates. Discoidin I domains are also found in other proteins, including coagulation factors V and VIII, and may represent a class of domains that interact with specific cell surface molecules.

Para-aminobenzoate synthase gene of Saccharomyces cerevisiae encodes a bifunctional enzyme.

The Saccharomyces cerevisiae gene for para-aminobenzoate (PABA) synthase has been identified based upon its ability to confer sulfonamide resistance when present on a multicopy episomal vector. The 3840 bp DNA sequence fragment reported here contains a 2199 bp open reading frame encoding a 733 amino acid protein with similarity to the two components of PABA synthase described for prokaryotes (Escherichia coli PabA and PabB), suggesting that PABA synthase is bifunctional in yeast. The cloned sequence was confirmed to be PABA synthase by gene disruption. Chromosome gel analysis places the gene for PABA synthase on chromosome XIV.

Cloning, expression, and characterization of Cryptococcus neoformans dihydrofolate reductase.

The Cryptococcus neoformans dihydrofolate reductase (DHFR) gene has been isolated from cDNA and genomic DNA libraries. The 690-base pair coding sequence codes for a 25,152-Da protein, which is the largest monofunctional DHFR yet reported. The gene contains two introns, and several putative regulatory sequences have been identified. The coding sequence was placed in a pUC-based expression vector, which expresses C. neoformans DHFR in Escherichia coli at a level of about 5% of the total soluble extract. The expressed DHFR was purified to homogeneity by methotrexate-Sepharose affinity chromatography, followed by anion exchange chromatography on Q-Sepharose. On SDS-polyacrylamide gel electrophoresis, the purified enzyme migrates as a single protein with apparent mass of 28 kDa. The molecular weight, as determined by electrospray mass spectral analysis, and the amino-terminal sequence are in accord with what was predicted from the DNA sequence. Steady state kinetic parameters, effects of pH, salts, and inhibition constants of several anti-folates have been determined.

Multiple genes encode the major surface glycoprotein of Pneumocystis carinii.

The major surface antigen of Pneumocystis carinii, a life-threatening opportunistic pathogen in human immunodeficiency virus-infected patients, is an abundant glycoprotein that functions in host-organism interactions. A monoclonal antibody to this antigen is protective in animals, and thus this antigen is a good candidate for development as a vaccine to prevent or control P. carinii infection. We have cloned and sequenced seven related but unique genes encoding the major surface glycoprotein of rat P. carinii. Partial amino acid sequencing confirmed the identity of these genes. Based on Southern blot studies using chromosomal or restricted DNA, the major surface glycoproteins are the products of a multicopy family of genes. The predicted protein has an M(r) of approximately 123,000, is relatively rich in cysteine residues (5.5%) that are very strongly conserved, and contains a well conserved hydrophobic region at the carboxyl terminus. The presence of multiple related msg genes encoding the major surface glycoprotein of P. carinii suggests that antigenic variation is a possible mechanism for evading host defenses. Further characterization of this family of genes should allow the development of novel approaches to the control of this pathogen.

The alpha-mating type locus of Cryptococcus neoformans contains a peptide pheromone gene.

The opportunistic fungal pathogen Cryptococcus neoformans has two mating types, MATa and MAT alpha. The MAT alpha strains are more virulent. Mating of opposite mating type haploid yeast cells results in the production of a filamentous hyphal phase. The MAT alpha locus has been isolated in this study in order to identify the genetic differences between mating types and their contribution to virulence. A 138-bp fragment of MAT alpha-specific DNA which cosegregates with alpha-mating type was isolated by using a difference cloning method. Overlapping phage and cosmid clones spanning the entire MAT alpha locus were isolated by using this MAT alpha-specific fragment as a probe. Mapping of these clones physically defined the MAT alpha locus to a 35- to 45-kb region which is present only in MAT alpha strains. Transformation studies with fragments of the MAT alpha locus identified a 2.1-kb XbaI-HindIII fragment that directs starvation-induced filament formation in MATa cells but not in MAT alpha cells. This 2.1-kb fragment contains a gene, MF alpha, with a small open reading frame encoding a pheromone precursor similar to the lipoprotein mating factors found in Saccharomyces cerevisiae, Ustilago maydis, and Schizosaccharomyces pombe. The ability of the MATa cells to express, process, and secrete the MAT alpha pheromone in response to starvation suggests similar mechanisms for these processes in both cell types. These results also suggest that the production of pheromone is under a type of nutritional control shared by the two cell types.

Isolation of telomerelike sequences from Cryptococcus neoformans and their use in high-efficiency transformation.

Development of a transformation system for the fungal human pathogen Cryptococcus neoformans is an important prerequisite for the identification of genes involved in virulence. It has previously been reported that low-efficiency transformation can be achieved by using the cloned C. neoformans URA5 gene and ura5 mutants. The introduction of linearized URA5 vectors into C. neoformans resulted in unstable transformants which apparently harbored linear extrachromosomal DNA molecules. In this paper, the nature of these molecules is confirmed to be linear by exonuclease digestion. Recovery of the extrachromosomal DNA in Escherichia coli and sequence analysis demonstrates that repeats characteristic of telomeric DNA have been added to the ends of the introduced DNA. The recovered plasmids are capable of transforming at much higher efficiencies either in the supercoiled state (up to 200 transformants per microgram) or the linear state (up to 90,000 transformants per microgram).

Molecular and genetic analysis of URA5 transformants of Cryptococcus neoformans.

Cryptococcus neoformans var. neoformans ura5 mutants were transformed with linearized or circular plasmids containing the C. neoformans orotidine monophosphate pyrophosphorylase gene. Following electroporation, randomly isolated transformants were analyzed for the mitotic and meiotic stability of uracil prototrophy. All stable transformants tested showed nonspecific ectopic integration. Uracil prototrophy in these transformants was stable through meiosis. Some of the stable transformants showed integration of both URA5 and vector sequences, while others lacked any vector sequences. Unstable transformants exhibited the presence of an autonomously replicating plasmid which had undergone significant sequence rearrangement. The autonomously replicating plasmid in the transformants was observed to be the same size or smaller than the transforming plasmid, was maintained in a linear form, and had acquired a genomic sequence(s) with homology to a sequence(s) on all the chromosomes. The conservation of a 300-bp sequence at the 5' end of the URA5 gene was observed in all the rearranged plasmids. These results suggest mechanisms of plasmid maintenance in C. neoformans that are different from those reported for other yeasts. The ura5 mutant was significantly less virulent than the wild type. The transformants did not recover virulence regardless of prototrophic stability.

Genetic association of mating types and virulence in Cryptococcus neoformans.

A pair of congenic Cryptococcus neoformans var. neoformans strains, B-4476 (a mating type) and B-4500 (alpha mating type), that presumably differ only in mating type was constructed. This pair and their progeny, five alpha type and five a type, were tested for virulence in mice. In the parent strains as well as the progeny, alpha type was clearly more virulent than a type. In addition, death tended to occur earlier among the alpha-strain-infected mice that died than among the mice that died by infection caused by a strains. These data strongly suggest the genetic association of virulence with mating type in this human fungal pathogen.

Selection of ura5 and ura3 mutants from the two varieties of Cryptococcus neoformans on 5-fluoroorotic acid medium.

Spontaneous mutants requiring uracil were isolated from both varieties of Cryptococcus neoformans by plating on 5-fluoroorotic acid (5-FOA) medium. Of the 36 strains tested (18 var. neoformans and 18 var. gattii), 24 (12 of each variety) generated 5-FOA-resistant cells requiring uracil for growth. Six of the 12 C. neoformans var. gattii strains produced ura3 cells while the remaining six strains produced ura5 cells. None of the 12 strains produced both ura3 cells and ura5 cells. All 12 isolates of var. neoformans, however, produced ura5 cells and one of them produced ura3 as well as ura5 cells. A genetic lesion in the URA5 gene of an isolate of C. neoformans var. gattii was confirmed by complement with the cognate URA5 gene of C. neoformans var. neoformans. The ura3 isolates were tentatively identified by their ability to grow on a medium containing uridine but not on a medium with orotic acid or orotidine. Enzymatic assays for orotidine-5'-phosphate decarboxylase activity confirmed the isolates to be ura3 mutants. Hybridization analysis of total DNA, digested with EcoRI or StuI and probed with pURA5g2, revealed the presence of only one copy of URA5 in the strains of either variety, regardless of the prevalence of ura5 mutants. Extensive polymorphism was observed in the restriction patterns of the fragments containing the URA5 locus. The prevalence of spontaneously arising ura3 mutants among the isolates of C. neoformans var. gattii, but not among the isolates of C. neoformans var. neoformans, is one more biological difference that distinguishes the two varieties.