Genetic crosses within and between species of Cryptosporidium

Abstract

Parasites and their hosts are engaged in reciprocal coevolution that balances competing mechanisms of virulence, resistance, and evasion. This often leads to host specificity, but genomic reassortment between different strains can enable parasites to jump host barriers and conquer new niches. In the apicomplexan parasite Cryptosporidium, genetic exchange has been hypothesized to play a prominent role in adaptation to humans. The sexual lifecycle of the parasite provides a potential mechanism for such exchange; however, the boundaries of Cryptosporidium sex are currently undefined. To explore this experimentally, we established a model for genetic crosses. Drug resistance was engineered using a mutated phenylalanyl tRNA synthetase gene and marking strains with this and the previously used Neo transgene enabled selection of recombinant progeny. This is highly efficient, and genomic recombination is evident and can be continuously monitored in real time by drug resistance, flow cytometry, and PCR mapping. Using this approach, multiple loci can now be modified with ease. We demonstrate that essential genes can be ablated by crossing a Cre recombinase driver strain with floxed strains. We further find that genetic crosses are also feasible between species. Crossing Cryptosporidium parvum, a parasite of cattle and humans, and Cryptosporidium tyzzeri a mouse parasite resulted in progeny with a recombinant genome derived from both species that continues to vigorously replicate sexually. These experiments have important fundamental and translational implications for the evolution of Cryptosporidium and open the door to reverse- and forward-genetic analysis of parasite biology and host specificity.

Keywords: Apicomplexa; diarrheal disease; genetic cross; parasite; sex.

Fig. 1.

PheRS can be used as a selection marker for stable transgenesis. (A) Map of the pheRS locus targeted by insertion of a construct that includes the single base mutation that confers resistance to BRD7929, nano luciferase (Nluc), and the fluorescent protein reporter tdNeonGreen. (B) Ifnγ^−/− mice were infected with transfected parasites and burden was monitored by fecal luciferase activity. (C) Oocyst were purified from the feces and subjected to flow cytometry and transgenic parasites were highly fluorescent (gray: Cp WT; green: Cp-pheRS^r-Nluc-tdNG).

Fig. 2.

Using pheRS as selection marker in trans. (A) Map of the TK locus targeted by insertion of a construct that includes the pheRS gene (the last 113 bp were recodonized and carry the resistance mutation) and nano luciferase (Nluc). (B) Fecal luciferase activity from mice infected with parasites transfected using the transgene shown in A. (C) Parasite growth in HCT-8 cells assessed by measuring luminescence in the presence of the indicated concentrations of BRD7929. Means and SD of 3 biological replicates are shown; the entire experiment was conducted twice with similar results. (D and E) Ifnγ^−/− mice were infected with pheRS^r-TK-KO (D) or pheRS^WT (E), treated with the indicated doses of BRD7929 on days 4 to 8, and fecal nano luciferase activity was measured. (F) Parasites were transfected with construct shown in A. Infected Ifnγ^−/− mice were treated with 5 mpk BRD7929, and fecal nano luciferase activity was measured. (G) PCR mapping using genomic DNA from transgenic parasites selected in F demonstrating integration into the TK locus.

Fig. 3.

Establishment of highly efficient crosses in Cryptosporidium. (A) Maps of the loci used to mark each of the parental C. parvum lines. (B) Experimental set-up of the crossing experiment between pheRS^r-Nluc-tdNG which is susceptible to BRD7929 and expresses tdNeonGreen and Paro^r -tdTom which is susceptible to paromomycin and expresses tdTomato. (C–E) Ifnγ^−/− mice were infected with the indicated parasite strains, subjected to dual drug treatment, and parasite burden was measured by fecal luciferase assay. Note that single infected mice are cured while co-infected mice remain infected. BRD7929 treatment was stopped on day 8, paromomycin treatment was continued until the end of the experiment. (F–H) Flow cytometric analysis of oocysts isolated from E at the days indicated (days 11 and 14 were indistinguishable from ref. , see SI Appendix, Fig. S6 A and B). F shows a mixture of both parental strains. (I) PCR mapping of the marked TK and pheRS loci in the parental lines and the cross (numbers indicate primer pairs highlighted in A). Note that the progeny of the cross is recombined and inherited markers and reporters from both parents.

Fig. 4.

Conditional gene ablation in crossed progeny. (A) Illustration of the experimental set-up crossing a Cre driver with floxed strains and the respective phenotypic assays for gene excision. (B–D) Progeny of the loxP-tdTom-loxP-tdNeon-Nluc-Paro^r color switch cross was used to infect HCT-8 cells and cultures were treated with rapamycin (Rapa). (B) Genomic DNA was extracted from oocysts (cross) or from cultures infected for 3 d, and the floxed gene was amplified by PCR. Amplicons are 2,076 bp for the floxed, and 467 bp for the excised locus, respectively. (C) Representative micrographs of HCT-8 cells infected with cross progeny after 48 h with or without rapamycin. (Scale bar, 5 μm.) (D) HCT-8 cells infected with cross progeny were analyzed by flow cytometry after 72 h of rapamycin treatment and relative transgene expression is plotted as their portions of total fluorescent cells. Means and SDs of three biological replicates are shown. (E and F) Ifnγ^−/− mice were infected with progeny of the cross between the DiCre and Cp-loxP-Paro^r -loxP strains. (E) Ifnγ^−/− mice infected with the progeny of a cross between the DiCre and Cp-loxP-Paro^r-loxP lines, treated with rapamycin or vehicle, and genomic DNA was extracted from the feces at the indicated times. PCR analysis revealed amplicons for the floxed (1,269 bp) and excised locus (432 bp), respectively. Note strong rapamycin dependent induction. (F) Fecal nano luciferase measurements from these mice revealed paromomycin susceptibility as a consequence of in vivo Cre-mediated excision induced by rapamycin.

Fig. 5.

Testing the species boundaries of genetic exchange in Cryptosporidium. (A) Ifnγ^−/− mice were infected with C. parvum pheRS^r-Nluc-tdNG and C. tyzzeri Paro^r-tdTom and subjected to dual drug treatment over the time indicated by shading in yellow. Parasite burden was measured by the fecal nano luciferase assay. (B) Flow cytometry of a mixture of both parental oocysts, and (C), oocyst isolated on day 21 (arrow in A) following coinfection showing uniform double positive fluorescence. (D) The indicated loci were amplified by PCR from day 21 progeny oocysts, and amplicons were subjected to Sanger sequencing. Amplicons were genotyped using species-specific SNPs. C. tyzzeri SNPS are shown in red, those specific for C. parvum in green. Agarose gel shows the fragments that were sent in for sequencing. Red and green lines depict SNPs between C. parvum and C. tyzzeri. (E) Coverage of nanopore reads of the cross. The red line indicates median coverage for each chromosome. (F) Calculation of the common C. tyzzeri SNP position and frequency obtained from bulk sequencing of the crossed progeny plotted across all 8 chromosomes. (G) Progeny was also subjected to single-oocyst sequencing and a recombinant locus on chromosome 8 is depicted. The colors in the read coverage plot represent the nucleotide of SNPs, with adenine in green, cytosine in blue, guanine in yellow, and thymine in red (see SI Appendix, Fig. S9 for a multiple sequence alignment of long reads at this locus) and SI Appendix alignment of 50 randomly selected reads.