Susceptibility to experimental infection of the invertebrate locusts (Schistocerca gregaria) with the apicomplexan parasite Neospora caninum

Abstract

Neuropathogenesis is a feature of Neospora caninum infection. In order to explore this in the absence of acquired host immunity to the parasite, we have tested infection in locusts (Schistocerca gregaria). We show for the first time that locusts are permissive to intra-hemocoel infection with N. caninum tachyzoites. This was characterized by alteration in body weight, fecal output, hemoparasitemia, and sickness-related behavior. Infected locusts exhibited progressive signs of sickness leading to mortality. Also, N. caninum showed neuropathogenic affinity, induced histological changes in the brain and was able to replicate in the brain of infected locusts. Fatty acid (FA) profiling analysis of the brains by gas chromatography and multi-variate prediction models discriminated with high accuracy (98%) between the FA profiles of the infected and control locusts. DNA microarray gene expression profiling distinguished infected from control S. gregaria brain tissues on the basis of distinct differentially-expressed genes. These data indicate that locusts are permissible to infection with N. caninum and that the parasite retains its tropism for neural tissues in the invertebrate host. Locusts may facilitate preclinical testing of interventional strategies to inhibit the growth of N. caninum tachyzoites. Further studies on how N. caninum brings about changes in locust brain tissue are now warranted.

Keywords: Behaviour; Host-pathogen interaction; Infection; Invertebrate model; Locusts.

Figure 1. Survival of locusts given various doses of Neospora caninum tachyzoites by the intra-hemocoel route.

Groups (G1 to G4) of locusts (n = 10) were administered doses of N. caninum of 10³ (G1), 10⁴ (G2), 10⁵ (G3), and 10⁶ (G4) per locust. Control locusts were sham-inoculated with RPMI cultured medium. An environmental control group (e-group) of non-infected locusts incubated under the same conditions as other groups was also included. Survival was monitored daily after infection. Results represent average survival curve based on three independent experiments. Control vs. G1 (p = 0.0008); control vs. G2 (p = 0.0007); control vs. G3 (p = 0.0004); control vs. G4 (p = 0.0047). Locusts inoculated with 10 or 100 tachyzoites did not exhibit any signs of sickness or mortality (data not shown).

Figure 2. Relative body weight (BW) of Neospora caninum-infected locusts compared to controls at different time points after infection.

Shown are means ± SEM of percent body weight change compared with initial body weight for surviving locusts at each time point. There was no significant change in the BW of control locusts and locusts in group 1 (infected with 10³ tachyzoites) along the course of the experiment, but in groups 2 and 4, infected with 10⁴ and 10⁶, respectively, there was significant loss in weight beginning by 2 day after infection (p = 0.0024 and 0.0012, respectively). In group 3 infected with 10⁵ the weight loss began by 3 day after infection (p < 0.0001). Not all locusts completed the course of the experiment due to associated mortality. Data was compared using paired t-test (p-value < 0.05).

Figure 3. The effect of Neospora caninum infection on locust fecal output.

Besides the environmental control (E-control) group, an additional group of locusts were inoculated with media only and considered the non-infected control. Groups 1, 2, 3, and 4 were infected as described in materials and methods. Fecal output per group was weighted daily for up to 7 days PI. Total fecal output was divided by the number of living locusts for every day. There was non-significant increase in fecal output one day after infection, followed by significant decrease until day 7 after infection, with p-value 0.03, 0.04, 0.05, and 0.03 for group 1, 2, 3, and 4, respectively. Data was compared using paired t-test (with p-value <0.05 as significant). Results are presented as means from three independent experiments.

Figure 4. Representative micrographs of Neospora caninum-infected locust brains.

Locusts were injected with 10⁶ N. caninum and their brains were dissected out at 5 days post-infection. Subsequently, the brains were sectioned and stained with haematoxylin and eosin. N. caninum triggered inflammatory response (arrows) in the brain tissue of infected locusts (A). No parasite was detected in the brain (A) or in the fat body surrounding the brain of (B). Magnification ×400.

Figure 5. PCR amplification of the Neospora caninum-specific Nc5 region (Np21/Np6).

Amplification of DNA extracts from brains of locusts experimentally infected with N. caninum showed the presence of genetic evidence of N. caninum in the brain of locusts from day 1 (d1) to day 5 (d5) PI. M: 100-bp molecular size marker; Lane 1: positive control represent DNA extracted from ∼3 × 10⁶ tachyzoites; lanes d1 to d5: N. caninum in brain d1 to d5.

Figure 6. Raman spectroscopic imaging of tachyzoites.

Bright field (A) and corresponding Raman (B) image of N. caninum tachyzoites derived from locusts’ brains. Bright field (C) and corresponding Raman (D) image of N. caninum tachyzoites derived from culture. Bar applies to all figures, 5 µm.

Figure 7. Comparative profiling of chemical structure of tachyzoites of Neospora caninum from locust-derived and original isolates using Raman spectroscopy.

(A) Comparative Raman spectra of tachyzoites of N. caninum culture-derived (blue) and locust brain-derived isolates (green) in the region from 700 to 1,700 cm^-1. (B) Principal component analysis score plots in the plane of principal components 2 vs. principal component 1 for samples tested. Each dot represents a chemical molecule and the dots are colored according to the biological group the sample belongs to. Blue dots indicate N. caninum culture-derived isolate and green dots indicate locust brain-derived isolate. Minor (green dots at the top), but non-significant differences were present between the chemical profile of each isolate. Raman spectra of locust-derived isolate are less scatter (i.e., less variation in their structures) compared to spectra of culture-derived isolate.

Figure 8. Heat map of differentially expressed lipids in locust brains.

Unsupervised two-dimensional hierarchical clustering of the 7 fatty acids that showed fold change differences between infected locust groups (n = 5 locusts) and their adjacent control daily for 5 days post infection with Neospora caninum. The heat map of differentially expressed lipids based on clustering is shown in the figure. Each column represents a lipid species and each row represents a locust group. Red colour indicates lipids that were upregulated and yellow color indicates lipids that were downregulated. Orange indicates lipids whose level is unchanged in infected locust’s brain as compared to normal. A significant discriminative power between the infected and control samples of the locust’s brain was evident. Samples are identified by a three-part code: “F/C (infected/controls)”. “Time point”. “Replicate number”. Fatty acids are reordered after applying a hierarchical clustering to their profiles. Hierarchical clustering of the rows and columns highlights groups of significantly correlated infection and lipids.

Figure 9. Volcano plot representation of the microarray data showing both significantly expressed transcripts and magnitude of change.

Negative log10 p-value on y axis indicates the significance of each gene, and the fold change (log base 2) mean expression difference on the x axis. Each gene is represented by a dot. Data are representative of three hybridizations per group.

Figure 10. Hierarchical clustering of significantly expressed genes of three infected vs. three control locusts.

The heat map shows two relatively distinct clusters of highly differentially expressed transcripts obtained from pairwise comparison between infected vs. control locust groups. Each row represents each sample tested and each column represents a single probeset (gene). On the hierarchical tree at the left side of the diagram, the upper half (red) indicates the control samples and the lower half (orange) indicates the infected samples. Relative gene expression is color represented: red is higher-level expression relative to the sample mean, blue is relatively lower-level expression, grey is no-change. The 11 probesets/genes in the upper right quadrant of the cluster map are genes that decreased upon infection relative to the control samples (shown in the lower right quadrant). The 6 probes/genes in the left upper quadrant were genes that were increased in control samples relative to infected samples (in the lower left quadrant).