Cryptosporidium uses CSpV1 to activate host type I interferon and attenuate antiparasitic defenses

Abstract

Cryptosporidium infects gastrointestinal epithelium and is a leading cause of infectious diarrhea and diarrheal-related death in children worldwide. There are no vaccines and no fully effective therapy available for the infection. Type II and III interferon (IFN) responses are important determinants of susceptibility to infection but the role for type I IFN response remains obscure. Cryptosporidium parvum virus 1 (CSpV1) is a double-stranded RNA (dsRNA) virus harbored by Cryptosporidium spp. Here we show that intestinal epithelial conditional Ifnar1^-/- mice (deficient in type I IFN receptor) are resistant to C. parvum infection. CSpV1-dsRNAs are delivered into host cells and trigger type I IFN response in infected cells. Whereas C. parvum infection attenuates epithelial response to IFN-γ, loss of type I IFN signaling or inhibition of CSpV1-dsRNA delivery can restore IFN-γ-mediated protective response. Our findings demonstrate that type I IFN signaling in intestinal epithelial cells is detrimental to intestinal anti-C. parvum defense and Cryptosporidium uses CSpV1 to activate type I IFN signaling to evade epithelial antiparasitic response.

Fig. 1. C. parvum infection in conditional intestinal epithelial Ifnar1 knockout mice and IEC4.1 cell cultures.

a HE morphological features of small intestine of Ifnar1^fl/fl and Villin.Ifnar1^−/− neonates following infection. Neonatal mice (5 days old) were orally inoculated with C. parvum oocysts (10⁶ oocysts per animal) and ileal epithelium (4 cm of small intestine tissue from the ileocecal junction) was collected (72 h p.i.) Representative HE images from four independent experiments are shown. Insets are higher magnification of the boxed regions showing the parasites (arrows). Bars = 50 µm. b C. parvum intestinal infection burden in Ifnar1^fl/fl and Villin.Ifnar1^−/− mice. Neonatal mice were inoculated with C. parvum oocysts (10⁶ oocysts per animal) for 48 h and 72 h. Infection burden was evaluated by RT-qPCR of cp18S gene in the isolated intestinal epithelium (fold changes to Ifnar1^fl/fl normalized to host Gapdh) and parasite counting of intestine sections after immunofluorescent staining of C. parvum (average number/60X field). Data are presented as mean values ± SD, compiled from 4 independent experiments and the dots represent the mean value of each experiment with 6 mice in each group. Statistical significance was determined by two-tailed unpaired Student’s t-test. c Immunofluorescent staining of C. parvum, and PCNA and EpCAM staining of small intestine of Ifnar1^fl/fl and Villin.Ifnar1^−/− neonates with and without C. parvum infection (at 72 h p.i.). Representative images from 4 independent experiments are shown. Insets are higher magnification of the boxed regions showing the parasites revealed by immunofluorescent staining for parasite counting (arrowheads). Blue: DAPI (DNA), green: PCNA or EpCAM, red: C. parvum (arrows). Bars = 50 µm. d No significant difference in infection burden between IEC4.1 and IEC4.1-Ifnar1^−/− cells following C. parvum infection in vitro. IEC4.1 and IEC4.1-Ifnar1^−/− cells were infected with C. parvum for 2 h and 24 h. Infection burden was evaluated by RT-qPCR (cp18S, fold changes to IEC4.1 normalized to Gapdh). The dots represent data from three biological replicates. Data are presented as mean values ± SD. Blue: DAPI (DNA), red: C. parvum (arrows). Bars = 5 µm. Source data are provided as a Source Data file.

Fig. 2. Gene expression profiles in infected intestinal epithelium and cell cultures.

a Expression of IFNs in intestinal epithelium from infected Ifnar1^fl/fl neonates. Mice (5 days old) were orally inoculated with C. parvum (10⁶ oocysts/animal) and ileal epithelium (4 cm from the ileocecal junction) was isolated at 48 h p.i. IFN gene expression was evaluated by RT-qPCR and presented as fold change to uninfected control. Dots represent data from experiment with six mice in each group and presented as mean values ± SD. b Expression of IFNs in infected IEC4.1 cells. IFN gene expression (24 h p.i.) was evaluated by RT-qPCR. Ifn-β1 protein content in the supernatants (24 h and 48 h p.i.) was measured by ELISA. Data are from three biological replicates and presented as mean values ± SD. p values were determined by two-tailed unpaired Student’s t-test (in a and b). c Gene expression profiles in intestinal epithelium from infected neonates (48 h and 72 h p.i.) by RNA-Seq. The numbers of genes with an altered expression level following infection are listed. d Gene set enrichment analysis (GSEA) in intestinal epithelium from infected Ifnar1^fl/fl (48 h p.i.) compared with that in uninfected Ifnar1^fl/fl or infected Villin.Ifnar1^−/− neonates (48 h p.i.). p values were calculated based on Kolmogorov–Smirnov test and adjusted by Benjamini–Hochberg method. Normalized enrichment scores and adjusted p values for the signaling pathways are shown. e Selected gene groups revealed by RNA-Seq in intestinal epithelium from infected Ifnar1^fl/fl and Villin.Ifnar1^−/− neonates (48 h. p.i.). Representative genes for type I/II/III IFN signaling and misc. upregulated genes (expression levels further increased in infected Villin.Ifnar1^−/− neonates) are shown. f Volcano plots depicting the differentially upregulated genes between infected Villin.Ifnar1^−/− and Ifnar1^fl/fl animals (48 h p.i.). Two-tailed Wald tests were performed for statistical analysis and p value was adjusted by Benjamini–Hochberg method. Dashed line indicates a false discovery rate cutoff of 0.05. Data are from three animals for each group (c–f). g Gene expression profiles in IEC4.1 following infection. Numbers of genes whose expression levels were significantly altered in infected IEC4.1 and IEC4.1-Ifnar1^−/− cells (24 h p.i.) by RNA-Seq are listed. h GSEA of gene expression profiles in infected IEC4.1-Ifnar1^−/− cells (24 h p.i.) compared with that in infected IEC4.1 cells (24 h p.i.). Data are from three biological replicates (3 RNA-seq replicates each group) in g and h. p values (in h) were calculated based on Kolmogorov–Smirnov test and adjusted by Benjamini–Hochberg method. Normalized enrichment scores and adjusted p values for the signaling pathways are shown. Source data are provided as a Source Data file.

Fig. 3. Delivery of CSpV1-dsRNAs into the cytoplasmic region of intestinal epithelial cells following C. parvum infection.

a Detection of CSpV1-dsRNAs in the cytoplasm of infected cells. IEC4.1 cells were exposed to C. parvum for 24 h and cytoplasmic fractions were isolated. CSpV1-dsRdRp, CspV1-dsCA, and several parasite RNAs (e.g., Cgd7_4540) in the cytoplasmic fractions were measured by RT-qPCR. Data are presented as the fold change to control normalized to host Gapdh. Cytoplasmic extract from cells collected from cell suspension mixed with the same amount of parasites was used as control (Ctrl). Data are from three biological replicates and presented as mean values ± SD. p values were determined by two-tailed unpaired Student’s t-test. b In situ hybridization of CSpV1-dsRdRp in C. parvum-infected cell. IEC4.1 cells were exposed to C. parvum for 24 h and CSpV1-dsRdRp was detected in the cytoplasm (arrows) using DIG-labeled CSpV1-dsRdRp probes. Representative images from three independent experiments are shown. Blue: DAPI (DNA), green: CSpV1-dsRdRp (arrows), C. parvum: arrowheads. Bars = 1 µm. c Immunofluorescent staining of CSpV1-dsRNAs using J-2 antibody in IEC4.1 cells following C. parvum infection (24 h) or in cells transfected with CSpV1-dsRdRp, CSPv1-dsCA, or a non-specific double-stranded siRNA (control RNA) for 24 h. Representative images from three independent experiments are shown. Blue: DAPI (DNA), red: C. parvum, green: CSpV1-dsRNAs (positive for J2 antibody, arrows). Bars = 1 µm. d–f Dot blots using J2 antibody to detect CSpV1-dsRNAs. Detection of in vitro-transcribed CSpV1-dsCA and CSpV1-dsRdRp (template plasmid DNA for in vitro transcription as control) (d); detection of CSpV1-dsRNAs in different doses of C. parvum sporozoite lysate (e); detection of CSpV1-dsRNAs in untreated IEC4.1 cells and cells infected by inactivated C. parvum or infective C. parvum or transfected with CSpV1-dsRNAs (f). Representative dot blots from three independent experiments are shown. g RIP assay for CSpV1-dsRNAs from IEC4.1 cells following C. parvum infection or CSpV1-dsCA or CSpV1-dsRdRp transfection. Cells were exposed to C. parvum or transfected with CSpV1-dsCA or CSpV1-dsRdRp for 24 h. CSpV1-dsRNAs were pulled down using J2 antibody. Isotype IgG was used as negative control. Data are from three biological replicates and presented as mean values ± SD. p values were determined by two-tailed unpaired Student’s t-test. Source data are provided as a Source Data file.

Fig. 4. Delivery of CSpV1-dsRNAs triggers activation of type I IFN signaling in host cells.

a Upregulation of type I IFN and type I IFN-stimulated genes in IEC4.1 cells transfected with CSpV1-dsCA or CSpV1-dsRdRp for 24 h as revealed by RT-qPCR. b Slight activation of type I IFN signaling in IEC4.1 cells transfected with CSpV1-ssRdRp-F, CSpV1-ssRdRp-R, CSpV1-ssCA-F and CSpV1-ssCA-R, respectively, for 24 h. Data (in a and b) are from three biological replicates and presented as mean values ± SD. p values were determined by one-way ANOVA followed by Tukey’s HSD test. c Heatmap of gene expression profiles in IEC4.1 cells following C. parvum infection or cells transfected with CSpV1-dsCA or CSpV1-dsRdRp or in combination. Cells were infected for 24 h or transfected with CSpV1-dsRNAs for 24 h. The heatmap representing differentially expressed genes (log10 fold changes) for each biological replicate (3 RNA-seq replicates each group). d Venn Diagrams demonstrating genes differentially expressed (with adjusted p values < 0.05) in IEC4.1 cells following infection (24 h p.i.) or transfected with CSpV1-dsCA or CSpV1-dsRdRp or their combination (for 24 h). e Heatmap representing expression of type I IFN-stimulated genes in infected IEC4.1 cells (24 h p.i.) or cells transfection of CSpV1-dsCA or CSpV1-dsRdRp or their combination (24 h). Data (in c–e) are from three RNA-Seq biological replicates for each group. f Interference with CSpV1-RNA delivery through transfection of host cells with siRNA combinations (siPOOLs) designed to target CSpV1-dsRNAs. IEC4.1 cells were transfected with the siPOOLs for 24 h, followed by infection for 24 h. Levels of CSpV1-dsRdRp were measured by RT-qPCR. A non-specific scrambled siRNA (siControl) was used for control. g Infection burden in IEC4.1 cells transfected with the siPOOLs or siControl for 24 h, followed by infection for 24 h. Levels of cpHsp70 were measured by RT-qPCR. h Attenuation of C. parvum-triggered Ifnb1 transcription in cells treated with siPOOLs. IEC4.1 cells were transfected with the siPOOLs or siControl for 24 h, followed by infection for 24 h, and Ifnb1 levels were measured. Data (in f–h) are from three biological replicates and presented as mean values ± SD. p values were determined by two-way ANOVA followed by Tukey’s HSD test (in f and h). Source data are provided as a Source Data file.

Fig. 5. CSpV1-dsRNAs trigger type I IFN signaling through activation of the Pkr and Rig-I/Mavs/Sting signaling pathways.

a Stable IEC4.1 cells deficient in Ifih1, Ddx58, Mavs, or Eif2ak2 generated using the CRISPR/Cas9 approach as confirmed by Western blot. Representative gel images from three independent experiments are shown. b Knockdown of Tlr3 with an siRNA in IEC4.1 cells as confirmed by RT-qPCR. Data are from three biological replicates and presented as mean values ± SD. p values were determined by two-tailed unpaired Student’s t-test. c C. parvum- and CSpV1-dsRNA-induced Ifnb1 expression in IEC4.1 cells deficient in Ifih1, Ddx58, Mavs, or Eif2ak2 and in cells treated with the Tlr3-siRNA. Cells were exposed to C. parvum or transfected with CSpV1-dsRNAs for 24 h. To knockdown Tlr3, cells were first treated with the Tlr3-siRNA for 24 h and then exposed to infection or CSpV1-dsRNA transfection. Ifnb1 expression was measured by RT-qPCR. d Inhibition of Sting signaling and Pkr signaling on C. parvum-induced expression of type I IFNs and type I IFN-controlled genes. IEC4.1 cells were exposed to C. parvum infection for 8 h and cultured for additional 16 h in the presence or absence of the inhibitor C-178 (to Sting signaling) or C-16 (to Pkr signaling). Usp18, Ifi44, Oas1g, Isg15, and Ifnb1 RNA levels were measured by RT-qPCR. p values were determined by one-way or two-way ANOVA followed by Tukey’s HSD test (in c, d). e, f Interaction between CSpV1-dsRNAs and components in the Rig-I/Mavs/Pkr pathways in IEC4.1 cells following infection or CSpV1-dsRNA transfection. Cells were exposed to C. parvum or transfected with CSpV1-dsRNAs for 24 h. Cell lysates were collected for RNA-protein interaction measurement using RNA immunoprecipitation assay. For infected cells, a parasite RNA, Cgd7_Flc_0990, was used for control. p values were determined by two-tailed unpaired Student’s t-test (in e, f). Data (in c–f) are from three biological replicates and presented as mean values ± SD. g Model of CSpV1-dsRNAs to trigger type I IFN signaling in host cells. CSpV1-dsRNAs can be recognized by RIG-I and PKR in infected cells, resulting in activation of the Pkr and Rig-I/Mdv5/Sting signaling pathways and subsequently, activation of type I IFN signaling. Source data are provided as a Source Data file.

Fig. 6. Impaired cellular response to IFN-γ stimulation in infected epithelial cells and its association with C. parvum-induced type I IFN signaling activation.

a Pre-treatment of IEC4.1 cells with IFN-γ decreased C. parvum infection burden. Cells were first treated with IFN-γ (10 ng/ml) for 8 h, exposed to C. parvum for 24 h, and cp18S was quantified by RT-qPCR. b Impaired cellular response to IFN-γ stimulation in infected IEC4.1 cells associated with type I IFN signaling. IEC4.1 and Ifnar1^−/− IEC4.1 cells were first exposed to C. parvum infection for 24 h followed by IFN-γ stimulation for additional 2 h. Cellular response to IFN-γ stimulation was reflected by the expression level of the Ido1 gene. c Knockdown of CSpV1-dsRNA delivery in infected IEC4.1 cells through transfection of the specific siRNA combinations (siPOOLs) partially restored the induction of Ido1 gene expression in response to IFN-γ stimulation. A non-specific control siRNA (siControl) was used for control. Data (in a–c) are from three biological replicates and presented as mean values ± SD. p values were determined by two-way ANOVA followed by Tukey’s HSD test (in a–c). d Heatmaps representing altered expression levels of selected genes for the key components associated with IFN-γ signaling in infected IEC4.1 cells or neonatal intestinal epithelium. IEC4.1 cells (24 h p.i.) or intestinal epithelium from Ifnar1^fl/fl and Villin.Ifnar1^−/− neonates (5 days old) after infection (48 h and 72 h p.i.) were collected, followed by RNA-Seq. Expression levels of genes for key components associated with IFN-γ signaling are shown. Data are from three RNA-Seq biological replicates for each group. e Protein content of the Ifngr1 subunit of IFN-γ receptor in IEC4.1 and Ifnar1^−/− IEC4.1 cells, intestinal epithelium from Ifnar1^fl/fl and Villin.Ifnar1^−/− neonates as assessed by Western. Gapdh was used as a loading control. Representative gels from three independent experiments are shown. f Intestinal epithelium of Ifnar1^fl/fl and Villin.Ifnar1^−/− neonatal mice (either uninfected or 72h-infected animals) showed a similar staining for Ifngr1 protein by immunofluorescence. Representative images from 4 independent experiments are shown. Blue: DAPI (DNA), green: Ifngr1, red: C. parvum (arrows). Bars = 50 µm. Source data are provided as a Source Data file.

Fig. 7. Effects of anti-viral and anti-parasitic drugs on C. parvum infection of intestinal epithelium.

a Effects of anti-viral and anti-parasitic drugs on C. parvum infection of cultured HCT-8 cells. Cells were exposed to C. parvum infection for 24 h in the presence or absence of anti-viral and anti-parasitic drugs or their combination, followed by RT-qPCR of cpHsp70. b Schematic diagram of human 3D enteroids and 2D intestinal epithelial monolayer cultures. c Brightfield microscopy images of 3D enteroids (4 days culture from isolated human crypts) and 2D monolayers derived from 3D enteroids are shown. Representative images from three independent experiments are shown. Bars = 50 µm. d Expression of type I/III IFN and type I/III IFN-stimulated genes in human 2D intestinal epithelial monolayers following infection. Human 2D intestinal epithelial monolayer cultures were exposed to C. parvum infection for 24 h. Expression levels of USP18, IFI44, OAS1, IFNB1, IFNL1, and IFNL2/3 were measured by RT-qPCR. e Effects of anti-viral and anti-parasitic drugs on C. parvum infection of human 2D intestinal epithelial monolayers. 2D monolayers were exposed to C. parvum infection for 24 h, in the presence or absence of anti-viral and anti-parasitic drugs or their combination, followed by RT-qPCR measurement of cpHsp70. f, g Effects of antiviral and antiparasitic drugs on C. parvum intestinal infection in neonates. Five-day-old C57BL/6 neonates were orally administered C. parvum oocysts; antiviral and antiparasitic drugs or their combination were given at 2 h p.i. followed by daily for 3 days. Intestinal tissues were collected and infection burden was measured by RT-qPCR of cpHsp70 (f) and immunofluorescent staining of C. parvum (g). Blue: DAPI (DNA), red: C. parvum (arrows). A higher magnification of the boxed region to visualize C. parvum is shown. Bars = 50 µm. Data represent three biological replicates (a, d, e) or 6 mice in each group (f, g) and presented as mean values ± SD. p values were determined by one-way ANOVA followed by Tukey’s HSD test (in a, e, f) or by two-tailed unpaired Student’s t-test (in d). Source data are provided as a Source Data file.