The adenosine transporter of Toxoplasma gondii. Identification by insertional mutagenesis, cloning, and recombinant expression.

Purine transport into the protozoan parasite Toxoplasma gondii plays an indispensable nutritional function for this pathogen. To facilitate genetic and biochemical characterization of the adenosine transporter of the parasite, T. gondii tachyzoites were transfected with an insertional mutagenesis vector, and clonal mutants were selected for resistance to the cytotoxic adenosine analog adenine arabinoside (Ara-A). Whereas some Ara-A-resistant clones exhibited disruption of the adenosine kinase (AK) locus, others displayed normal AK activity, suggesting that a second locus had been tagged by the insertional mutagenesis plasmid. These Ara-A(r) AK+ mutants displayed reduced adenosine uptake capability, implying a defect in adenosine transport. Sequences flanking the transgene integration point in one mutant were rescued from a genomic library of Ara-A(r) AK+ DNA, and Southern blot analysis revealed that all Ara-A(r) AK+ mutants were disrupted at the same locus. Probes derived from this locus, designated TgAT, were employed to isolate genomic and cDNA clones from wild-type libraries. Conceptual translation of the TgAT cDNA open reading frame predicts a 462 amino acid protein containing 11 transmembrane domains, a primary structure and membrane topology similar to members of the mammalian equilibrative nucleoside transporter family. Expression of TgAT cRNA in Xenopus laevis oocytes increased adenosine uptake capacity in a saturable manner, with an apparent K(m) value of 114 microM. Uptake was inhibited by various nucleosides, nucleoside analogs, hypoxanthine, guanine, and dipyridamole. The combination of genetic and biochemical studies demonstrates that TgAT is the sole functional adenosine transporter in T. gondii and a rational target for therapeutic intervention.

Insertional tagging of at least two loci associated with resistance to adenine arabinoside in Toxoplasma gondii, and cloning of the adenosine kinase locus.

A genetic approach has been exploited to investigate adenylate salvage pathways in the protozoan parasite Toxoplasma gondii, a purine auxotroph. Using a new insertional mutagenesis vector designed to facilitate the rescue of tagged loci even when multiple plasmids integrate as a tandem array, 15 independent clonal lines resistant to the toxic nucleoside analog adenine arabinoside (AraA) were generated. Approximately two-thirds of these clones lack adenosine kinase (AK) activity. Parallel studies identified an expressed sequence tag (EST) exhibiting a small region of weak similarity to human AK, and this locus was tagged in several AK-deficient insertional mutants. Library screening yielded full-length cDNA and genomic clones. The T. gondii AK gene contains five exons spanning a approximately 3 kb locus, and the predicted coding sequence was employed to identify additional AK genes and cDNAs in the GenBank and dbEST databases. A genomic construct lacking essential coding sequence was used to create defined genetic knock-outs at the T. gondii AK locus, and AK activity was restored using a cDNA-derived minigene. Hybridization analysis of DNA from 13 AraA-resistant insertional mutants reveals three distinct classes: (i) AK-mutants tagged at the AK locus; (ii) AK- mutants not tagged at the AK locus, suggesting the possibility that another locus may be involved in regulating AK expression; and (iii) mutants with normal AK activity (potential transport mutants).

Origin, targeting, and function of the apicomplexan plastid.

The discovery of a plastid in Plasmodium, Toxoplasma and related protozoan parasites provides a satisfying resolution to several long-standing mysteries: the mechanism of action for various surprisingly effective antibiotics; the subcellular location of an enigmatic 35 kb episomal DNA; and the nature of an unusual intracellular structure containing multiple membranes. The apicomplexan plastid highlights the importance of lateral genetic transfer in evolution and provides an accessible system for the investigation of protein targeting to secondary endosymbiotic organelles. Combining molecular genetic identification of targeting signals with whole genome analysis promises to yield a complete picture of organellar metabolic pathways and new targets for drug design.

Transport and trafficking: Toxoplasma as a model for Plasmodium.

Like Plasmodium, the protozoan parasite Toxoplasma gondii is a member of the phylum Apicomplexa, and an obligate intracellular pathogen. Unlike Plasmodium, however, Toxoplasma is highly amenable to experimental manipulation in the laboratory. The development of molecular transformation protocols for T. gondii has provided both scientific precedent and practical selectable markers for Plasmodium. Beyond the feasibility of molecular biological experimentation now possible in both systems, the high frequency of stable transformation in Toxoplasma allows this parasite to be used for molecular genetic analysis. The ability to control homologous vs. non-homologous recombination in T. gondii permits gene knockouts/allelic replacements at previously cloned loci, and saturation insertional mutagenesis of the entire parasite genome (and cloning of the tagged loci). T. gondii also exhibits unusual ultrastructural clarity, facilitating cell biological analysis. The accessibility of Toxoplasma as an experimental system allows this parasite to be used as a surrogate for asking many questions that cannot easily be addressed in Plasmodium itself. T. gondii also serves as a model system for genetic exploration of parasite biology and host-parasite interactions. Success stories include: biochemical analysis of antifolate resistance mechanisms; pharmacological studies on the mechanisms of macrolide activity; genetic identification of nucleobase/nucleoside transporters and metabolic pathways; and cell biological characterization of the apicomplexan plastid. As with any model system, not all questions of interest to malariologists can be addressed in Toxoplasma; differentiating between sensible and foolish questions requires familiarity with the biological similarities and differences of these systems.

Nuclear-encoded proteins target to the plastid in Toxoplasma gondii and Plasmodium falciparum.

A vestigial, nonphotosynthetic plastid has been identified recently in protozoan parasites of the phylum Apicomplexa. The apicomplexan plastid, or "apicoplast," is indispensable, but the complete sequence of both the Plasmodium falciparum and Toxoplasma gondii apicoplast genomes has offered no clue as to what essential metabolic function(s) this organelle might perform in parasites. To investigate possible functions of the apicoplast, we sought to identify nuclear-encoded genes whose products are targeted to the apicoplast in Plasmodium and Toxoplasma. We describe here nuclear genes encoding ribosomal proteins S9 and L28 and the fatty acid biosynthetic enzymes acyl carrier protein (ACP), beta-ketoacyl-ACP synthase III (FabH), and beta-hydroxyacyl-ACP dehydratase (FabZ). These genes show high similarity to plastid homologues, and immunolocalization of S9 and ACP verifies that the proteins accumulate in the plastid. All the putatively apicoplast-targeted proteins bear N-terminal presequences consistent with plastid targeting, and the ACP presequence is shown to be sufficient to target a recombinant green fluorescent protein reporter to the apicoplast in transgenic T. gondii. Localization of ACP, and very probably FabH and FabZ, in the apicoplast implicates fatty acid biosynthesis as a likely function of the apicoplast. Moreover, inhibition of P. falciparum growth by thiolactomycin, an inhibitor of FabH, indicates a vital role for apicoplast fatty acid biosynthesis. Because the fatty acid biosynthesis genes identified here are of a plastid/bacterial type, and distinct from those of the equivalent pathway in animals, fatty acid biosynthesis is potentially an excellent target for therapeutics directed against malaria, toxoplasmosis, and other apicomplexan-mediated diseases.

Gene knock-outs and allelic replacements in Toxoplasma gondii: HXGPRT as a selectable marker for hit-and-run mutagenesis.

The hypoxanthine-xanthine-guanine phosphoribosyl transferase (HXGPRT) gene of the protozoan parasite Toxoplasma gondii encodes a safe, practical genetic marker suitable for both positive and negative selection. Taking advantage of the ability to control homologous versus nonhomologous recombination in haploid T. gondii tachyzoites by manipulating the length of homologous DNA sequence, we have explored the possibility of 'hit-and-run' mutagenesis to introduce gene knock-outs (or allelic replacements) at loci for which no known selection or screen is available. Using the uracil phosphoribosyl transferase (UPRT) locus as a target, a genomic clone containing approximately 8 kb encompassing the UPRT gene (but lacking essential coding sequence) was fused to a cDNA-derived HXGPRT 'minigene', which lacks sufficient contiguous genomic sequence for homologous recombination. After transfection of circular plasmid DNA, positive selection for HXGPRT activity identified stable transformants, > 30% of which were found to have integrated at the UPRT locus as 'pseudodiploids' (produced by single-site homologous recombination between the circular plasmid and genomic DNA). Upon removal of mycophenolic acid, resolution of pseudodiploids by spontaneous intrachromosomal homologous recombination was selected using 6-thioxanthine, yielding a 1:1 ratio of UPRT knock-out parasites to wild-type revertants, at frequencies of approximately 10(-6) per parasite doubling. Applications of 'hit-and-run' technology relative to other gene targeting strategies are discussed.

Expression, purification, and characterization of uracil phosphoribosyltransferase from Toxoplasma gondii.

The coding region derived from a full-length CDNA spanning the uracil phosphoribosyltransferase (UPRT) gene of Toxoplasma gondii has been ligated into a bacterial expression vector and overexpressed in E. coli. Recombinant UPRT protein migrated with a molecular mass of 27 kDa on SDS polyacrylamide gels and was purified to homogeneity by conventional protein purification techniques. In solution, UPRT behaved as a monomer and exhibited K(m)app values of 3.5 microM for uracil and 243 microM for phosphoribosylpyrophosphate, respectively. Other naturally occurring pyrimidine or purine bases were not recognized as substrates. [14C]Uracil phosphoribosylation was inhibited by 5-fluorouracil with a Ki value of 25 microM and was not activated by GTP. Ample quantities of recombinant enzyme are now available for biochemical and structural studies, facilitating evaluation of UPRT as a possible therapeutic target.

The barley stripe mosaic virus 58-kilodalton beta(b) protein is a multifunctional RNA binding protein.

The barley stripe mosaic virus (BSMV) beta(b) gene product is the major viral nonstructural protein synthesized during early stages of the infection cycle and is required for systemic movement of the virus. To examine the biochemical properties of beta(b), a histidine tag was engineered at the amino terminus and the protein was purified from BSMV-infected barley tissue by metal affinity chromatography. The beta(b) protein bound ATPs in vitro, with a preference for ATP over dATP, and also exhibited ATPase activity. In addition, beta(b) bound RNA without detectable sequence specificity. However, binding was selective, as the beta(b) protein had a strong affinity for both single-stranded (ss) and double-stranded (ds) RNAs but not for tRNA or DNA substrates. Mutational analyses of beta(b) purified from Escherichia coli indicated that the protein has multiple RNA binding sites. These sites appear to contribute differently, because mutants that were altered in their binding affinities for ss and ds RNA substrates were recovered.

Tagging genes and trapping promoters in Toxoplasma gondii by insertional mutagenesis.

Plasmid vectors that incorporate sequence elements from the dehydrofolate reductase-thymidylate synthase (DHFR-TS) locus of Toxoplasma gondii integrate into the parasite genome with remarkably high frequency (>1% of transfected parasites). These vectors may-but need not-include mutant DHFR-TS alleles that confer pyrimethamine resistance to transgenic parasites. Large genomic constructs integrate at the endogenous locus by homologous recombination, but cDNA-derived sequences lacking long stretches of contiguous genomic DNA (due to intron excision) typically integrate into chromosomal DNA by nonhomologous recombination. Nonhomologous integration occurs effectively at random; and coupled with the high frequency of transformation, this allows a large fraction of the parasite genome to be tagged in a single electroporation cuvette. Genomic tagging permits insertional mutagenesis studies conceptually analogous to transposon mutagenesis in bacteria, yeast, Drosophila, etc. In theory (and, thus far, in practice), this allows identification of any gene whose inactivation is not lethal to the haploid tachyzoite form of T. gondii and for which a suitable selection or screen is available. Transformation vectors can be engineered to facilitate rescue of the tagged locus and to include a variety of reporters or selectable markers. Genetic strategies are also possible, using reporters whose function can be assayed by metabolic, visual, or immunological screens to "trap" genes that are activated (or inactivated) under various conditions of interest.

Insertional tagging, cloning, and expression of the Toxoplasma gondii hypoxanthine-xanthine-guanine phosphoribosyltransferase gene. Use as a selectable marker for stable transformation.

A nonhomologous integration vector was used to identify the Toxoplasma gondii hypoxanthine-xanthine-guanine phosphoribosyl transferase (HXGPRT) gene by insertional mutagenesis. Parasite mutants resistant to 6-thioxanthine arose at a frequency of approximately3 x 10(-7). Genomic DNA flanking the insertion sites was retrieved by marker rescue and used to identify molecular clones exhibiting unambiguous homology to H(X)GPRT genes from other species. Sequence analysis of vector/genome junction sites reveals that integration of the linearized vector occurred with minimal rearrangement of either vector or target sequences, although the addition of filler DNA and small duplications or deletions of genomic sequences at the transgene termini was observed. Two differentially spliced classes of cDNA clones were identified, both of which complement hpt and gpt mutations in Escherichia coli. Kinetic analysis of purified recombinant enzyme revealed no significant differences between the two isoforms. Internally deleted clones spanning the genomic locus were used to create "knock-out" parasites, which lack all detectable HXGPRT activity. Complete activity could be restored to these knock-out mutants by transient transformation with either genomic DNA or cDNA-derived minigenes encoding both enzyme isoforms. Stable HXGPRT+ transformants were isolated under selection with mycophenolic acid, demonstrating the feasibility of HXGPRT as both a positive and negative selectable marker for stable transformation of T. gondii.

RNA-binding activities of barley stripe mosaic virus gamma b fusion proteins.

The barley stripe mosaic virus (BSMV) gamma-b gene encodes a 17 kDa cysteine-rich protein known to affect virulence and to have a role in regulating viral gene expression. We have constructed recombinant gamma-b-glutathione S-transferase fusion proteins in Escherichia coli and have determined the ability of the purified fusion proteins and various mutant derivatives to bind nucleic acids in vitro. Gel-shift analyses revealed that the wild-type gamma-b-fusion protein is able to bind RNA cooperatively. The binding affinity is highly selective for single-stranded RNA because double-stranded RNA, single-stranded and double-stranded DNA, and transfer RNA were unable to compete for binding with the labelled RNA probes. However, BSMV-specific sequence binding was not observed since a chloroplast RNA competed for binding with 32P-labelled transcripts derived from the BSMV genome. The first 44 amino acids of the 152 amino acid gamma-b fusion protein encompassing one of two cysteine-rich 'zinc finger-like' motifs, and a basic region separating the finger-like motifs are required for RNA binding. Site- specific amino acid substitutions within two groups of lysine and arginine residues located in the basic motif reduced the binding affinity of the fusion protein greatly, but cysteine and histidine substitutions designed to disrupt the finger-like motifs failed to have appreciable effects on binding. These findings indicate that the regulatory properties of gamma-b may be mediated in part by RNA binding activities.

Heterologous expression and characterization of the bifunctional dihydrofolate reductase-thymidylate synthase enzyme of Toxoplasma gondii.

We have expressed catalytically active Toxoplasma gondii dihydrofolate-thymidylate synthase (DHFR-TS) and the individual TS and DHFR domains in Escherichia coli using the T7 promoter of pET-15b. DHFR-TS constituted approximately 10% of the total soluble cell protein and was purified using methotrexate-Sepharose chromatography to yield 10 mg of homogeneous DHFR-TS per liter of culture. The DHFR domain was recovered as insoluble inclusion bodies which could be unfolded and refolded to recover soluble, active enzyme. The TS domain was overexpressed as a soluble protein by growing the cells at 24 degrees C; this is the first report of the expression of an active TS domain from a bifunctional enzyme. The kcat and K(m) values for DHFR-TS are similar to those of other previously characterized protozoan DHFRs and TSs. The antimicrobial antifolates, TMP and Pyr, inhibit DHFR activity of the bifunctional protein in accord with their effects in crude enzyme preparations and in vivo systems. Kinetic parameters and Ki values for TMP and Pyr with the isolated DHFR domain were identical to the values for DHFR in the bifunctional enzyme. Evidence of kinetic channeling of the dihydrofolate product of TS to the DHFR domain in the bifunctional enzyme was obtained by kinetic and inhibition studies. Properties such as yield, stability, and activities of the recombinant T. gondii DHFR-TS provide clear advantages over other bifunctional DHFR-TSs as a model for future studies.

Insertional mutagenesis and marker rescue in a protozoan parasite: cloning of the uracil phosphoribosyltransferase locus from Toxoplasma gondii.

Nonhomologous integration vectors have been used to demonstrate the feasibility of insertional mutagenesis in haploid tachyzoites of the protozoan parasite Toxoplasma gondii. Mutant clones resistant to 5-fluorouracil were identified at a frequency of approximately 10(-6) (approximately 2 x 10(-5) of the stable transformants). Four independent mutants were isolated, all of which were shown to lack uracil phosphoribosyl-transferase (UPRT) activity and harbor transgenes integrated at closely linked loci, suggesting inactivation of the UPRT-encoding gene. Genomic DNA flanking the insertion point (along with the integrated vector) was readily recovered by bacterial transformation with restriction-digested, self-ligated total genomic DNA. Screening of genomic libraries with the recovered fragment identified sequences exhibiting high homology to known UPRT-encoding genes from other species, and cDNA clones were isolated that contain a single open reading frame predicted to encode the 244-amino acid enzyme. Homologous recombination vectors were exploited to create genetic knock-outs at the UPRT locus, which are deficient in enzyme activity but can be complemented by transient transformation with wild-type sequences--formally confirming identification of the functional UPRT gene. Mapping of transgene insertion points indicates that multiple independent mutants arose from integration at distinct sites within the UPRT gene, suggesting that nonhomologous integration is sufficiently random to permit tagging of the entire parasite genome in a single transformation.

Inhibition of cytoplasmic and organellar protein synthesis in Toxoplasma gondii. Implications for the target of macrolide antibiotics.

We investigated potential targets for the activity of protein synthesis inhibitors against the protozoan parasite Toxoplasma gondii. Although nanomolar concentrations of azithromycin and clindamycin prevent replication of T. gondii in both cell culture and in vivo assays, no inhibition of protein labeling was observed in either extracellular or intracellular parasites treated with up to 100 microM drug for up to 24 h. Quantitative analysis of > 300 individual spots on two-dimensional gels revealed no proteins selectively depleted by 100 microM azithromycin. In contrast, cycloheximide inhibited protein synthesis in a dose-dependent manner. Nucleotide sequence analysis of the peptidyl transferase region from genes encoding the large subunit of the parasite's ribosomal RNA predict that the cytoplasmic ribosomes of T. gondii, like other eukaryotic ribosomes, should be resistant to macrolide antibiotics. Combining cycloheximide treatment with two-dimensional gel analysis revealed a small subset of parasite proteins likely to be synthesized on mitochondrial ribosomes. Synthesis of these proteins was inhibited by 100 microM tetracycline, but not by 100 microM azithromycin or clindamycin. Ribosomal DNA sequences believed to be derived from the T. gondii mitochondrial genome predict macrolide/lincosamide resistance. PCR amplification of total T. gondii DNA identified an additional class of prokaryotic-type ribosomal genes, similar to the plastid-like ribosomal genes of the Plasmodium falciparum. Ribosomes encoded by these genes are predicted to be sensitive to the lincosamide/macrolide class of antibiotics, and may serve as the functional target for azithromycin, clindamycin, and other protein synthesis inhibitors in Toxoplasma and related parasites.

The barley stripe mosaic virus gamma b gene encodes a multifunctional cysteine-rich protein that affects pathogenesis.

Barley stripe mosaic virus contains seven genes, one of which specifies a 17-kD cysteine-rich protein, gamma b, that is known to affect virulence. To further characterize the role of gamma b in pathogenesis, we mutagenized sequences encoding amino acids within two clusters of cysteine and histidine residues in the cysteine-rich domain and a group of basic amino acids located between the clusters and determined the effects of these mutations on the symptom phenotype in barley. Three single amino acid substitutions in cluster 1 and two amino acid exchanges in the basic region caused bleached symptoms associated with pronounced elevations in accumulation of gamma b protein. In contrast, three single amino acid substitutions in cluster 2 and a mutation in the basic motif resulted in attenuated ("null") symptoms typical of those produced when the gamma b gene is deleted. Tissue infected with these "null" mutants accumulated slightly elevated amounts of the gamma b protein but significantly lower levels of coat protein and the putative movement protein beta b. Genetic complementation tests revealed that cluster 1 mutations are dominant over the wild-type gamma b gene, whereas those in cluster 2 are recessive. These results highlight the pivotal role of gamma b in pathogenesis and suggest that the two cysteine-rich clusters are functionally distinct and that they affect different aspects of disease development.

Homologous recombination and gene replacement at the dihydrofolate reductase-thymidylate synthase locus in Toxoplasma gondii.

To investigate the feasibility of genomic transgene expression and gene targeting in Toxoplasma gondii, parasites have been transfected with constructs differing in the length of contiguous genomic sequence spanning the dihydrofolate reductase-thymidylate synthase (DHFR-TS) gene. We have previously reported that vectors derived from a DHFR-TS cDNA 'minigene' containing mutations in the DHFR coding sequence confer pyrimethamine resistance to transfected parasites (Donald and Roos, 1993). Stably resistant parasite clones arise at high frequency, generally by virtue of transgene integration into parasite chromosomes at locations scattered throughout the genome. In contrast, using a vector which contains 8 kb of contiguous genomic sequence (vs. < 2 kb for the cDNA-derived vectors), approximately half of the integration events occur by homologous recombination. Homologous recombination appears to occur at even higher frequency when a 16 kb genomic clone is used. Circular plasmids were more efficient than linearized molecules at producing homologous recombination in this system, integrating by reciprocal crossing-over to produce a duplication of the DHFR-TS locus. Double crossing-over (or gene conversion) was also observed at low frequency, resulting in complete allelic replacement in this haploid stage of the parasite. The ability to produce either homologous or non-homologous recombinants, by the selection of appropriate transformation constructs, has considerable genetic potential.

Multiple genetic determinants of barley stripe mosaic virus influence lesion phenotype on Chenopodium amaranticolor.

The ND18 and Type strains of barley stripe mosaic hordeivirus (BSMV) differ in the local lesion phenotypes they elicit on Chenopodium amaranticolor. The ND18 strain produces large necrotic lesions on this host by 3 to 4 days postinoculation, whereas the Type strain is less virulent and elicits chlorotic local lesions which appear about 2 weeks after inoculation. We have used infectious in vitro transcripts derived from full-length cDNA clones of these two BSMV strains to investigate the genetic basis for their differential virulence on C. amaranticolor. Pseudorecombination of the wild-type alpha, beta, and gamma genomic RNAs of each strain revealed that the lesion forming phenotype segregated with RNA gamma. Fine mapping of the phenotypic determinants on RNA gamma was carried out by constructing deletion mutants, chimeric recombinants, and point mutants. These experiments showed that three different genetic elements in the Type strain RNA gamma contribute significantly to its attenuated virulence on C. amaranticolor. In addition, pseudorecombination experiments using mutant Type strain gamma RNAs that were more virulent than native Type RNA gamma indicated that the clean segregation of the lesion forming phenotype observed with wild-type RNA gamma is fortuitous. This lesion phenotype is dependent on both the multiple attenuating determinants in the wild-type Type strain RNA gamma and the source of genomic RNAs alpha and beta in the inoculum. The complexity of these virulence determinants clearly illustrates the limitations of classical pseudorecombination as a tool for the genetic analysis of plant viruses.

Stable molecular transformation of Toxoplasma gondii: a selectable dihydrofolate reductase-thymidylate synthase marker based on drug-resistance mutations in malaria.

To facilitate genetic analysis of the protozoan parasite Toxoplasma gondii, sequences derived from the parasite's fused dihydrofolate reductase-thymidylate synthase (DHFR-TS) gene have been used to produce vectors suitable for stable molecular transformation. Mutations introduced into the DHFR coding region by analogy with pyrimethamine-resistant malaria confer drug resistance to Toxoplasma, providing useful information on the structure of fused DHFR-TS enzymes and a powerful selectable marker for molecular genetic studies. Depending on the particular drug-resistance allele employed and the conditions of selection, stable resistance can be generated either by single copy nonhomologous insertion into chromosomal DNA or by massively amplified transgenes. Frequencies of integration are independent of selection, and transgenes are stable without continued selection. Cointegration of a reporter gene adjacent to the selectable marker (under the control of an independent promoter) shows no loss of the cointegrated sequences over many parasite generations. By bringing the full power of molecular genetic analysis to bear on Toxoplasma, these studies should greatly facilitate the development of a model genetic system for Apicomplexan parasites.