Fig 1.
Morphological changes of activated S. carpocapsae nematodes.
(A) The head region of a non-activated IJ by scanning electron microscopy (SEM). (B) The head region of an activated nematode by SEM. The mouth is marked by a red arrow and the excretory pore by a black arrowhead in (A) and (B). (C) A non-activated IJ by light microscopy (400x). (D) An enlarged view of the boxed region in (C). (E) A partially activated nematode (400x). (F) An enlarged view of the boxed region in (E) where the partially expanded terminal pharyngeal bulb (arrow) is located. (G) A fully activated nematode (400x). (H) An enlarged view of the boxed region in (G). (I) A fully activated nematode whose gut is wide open (400x). (J) An enlarged view of the boxed region in (I). The black arrowhead in (A) and (B) point to the secretory/excretory pore. The black arrows in (F), (H) and (J) point to the pharyngeal bulbs. The red star in (J) marks the open gut.
Fig 2.
Activated S. carpocapsae ESPs contain lethal proteins.
(A) In vitro activation rates of IJs over a time course. Error bars represent standard errors. (B) A silver-stained protein gel of ESPs from nematodes that have been activated for different lengths of time, and ESPs from axenic nematodes. The left-hand side of the gel shows the secreted proteins from symbiont-associated IJs activated for different amounts of time. The right-hand side of the gel shows the proteins secreted from symbiont-associated (S) and axenic (A) IJs that were exposed to waxworm homogenate for 12 h. Each lane contains 1% of the total ESP volume. The 0h sample was collected from non-activated IJs. L, protein ladder; S, symbiotic IJs; A, axenic IJs. (C) Survival curves within 72 hrs of Drosophila injected with 20 ng of ESPs from nematodes having been activated for different lengths of time. The 30 hr curve (blue) mostly overlaps with the 6 hr curve (red). (D) Survival curves over 40 days of Drosophila injected with 20 ng of ESPs from nematodes having been activated for different lengths of time. (E) The average amount of ESPs secreted by individual IJs in 3 hours. (F) Survival curves of 2nd instar Bombyx mori larvae injected with 650 ng of ESPs from axenic nematodes that were activated for 12 h. (G) Phenotype curves of last instar Galleria mellonella larvae injected with 4 μg of ESPs from axenic nematodes that were activated for 12 h. The dotted line shows the number of larvae that were either killed or paralyzed after the injection. Paralyzed waxworms recovered over time.
Fig 3.
Gene expression dynamics during in vivo IJ activation.
(A) Gene expression correlation matrix comparing the Spearman’s rank correlation coefficients between activated and non-activated IJs. Yellow colored cells indicate high correlations while blue cells indicate low correlations. Gene expression was transformed (log2(counts+1)) prior to calculating the correlation coefficients. Samples were K-means clustered by their Spearman’s rank correlation coefficients. (B) Activated and non-activated IJ transcriptomes plotted in the space of the first two principal components (which comprises 46% of the variation). A short description of what each principal component (PC) explains is included next to the PCs. The green line show the trajectory of IJ activation. (C) Scatterplots comparing the expression of all genes between non-activated IJs and 9 hr, 12 hr, 15 hr in vivo activated IJs. Genes labeled in orange and blue are differentially expressed (FDR < 0.05 and fold change > 2) between the stages compared, while grey genes are not. Expression for each gene was averaged across the biological replicates. (D) maSigPro plots showing the average expression of clusters of genes that are differentially expressed over the in vivo activation time course. The GO terms are representative of these expression clusters.
Fig 4.
The gene expression profile of the 12 hour in vitro activated nematodes is most similar to that of 15 hour in vivo activated nematodes.
(A) Scatterplot comparing the average gene expression counts of non-activated IJs and 12 hr in vitro activated IJs. Genes colored in orange and blue are differentially expressed (FDR < 0.05 and fold change > 2) and have higher expression in non-activated IJs and 12 hr in vitro activated IJs respectively. (B) Scatterplot comparing the average gene expression counts between 12 hr in vitro and 15 hr in vivo activated IJs. Genes colored in orange and blue are differentially expressed (FDR < 0.05 and fold change > 2) and have higher expression in 12 hr in vitro and 15 hr in vivo activated IJs respectively. (C) Venn diagrams showing genes that are upregulated and downregulated in the 12 hr in vitro and 15 hr in vivo activated IJs relative to non-activated IJs. The total number of upregulated and downregulated genes and their genome percentages are shown in the upward and downward block arrows, respectively. Representative GO terms and p-values for each category are listed on the right. Categories that do not have GO terms had no significant enrichment results.
Fig 5.
Expression of venom protein genes during IJ activation.
(A) A heatmap showing the mean-centered gene expression counts of 472 venom protein genes across activated and non-activated IJs. Protein products of these genes were detected with mass spectrometry in the venom from 12 hr in vitro activated IJs. DE, differentially expressed. The green and red colored rows in the DE column represent the upregulated and downregulated genes in activated IJs, which are used in (B). (B) Venn diagrams showing the venom protein genes that are transcriptionally upregulated or downregulated in activated IJs relative to non-activated IJs, and their breakdown by activation method (12 hr in vitro and 15 hr in vivo). (C) maSigPro plots showing venom protein genes that are differentially expressed over the 15 hr in vivo activation time course.
Fig 6.
Comparative analysis of venom proteins from activated IJs.
(A) Top BLAST hits of the 472 venom proteins in non-Steinernema organisms. (B) The top 16 Pfam domains in the venom proteins. (C) Merops breakdown of proteases and protease inhibitors found in the venom. (D) A neighbor-joining gene tree of the Shk domain-containing proteins found in the venom. Gray boxes highlight S. carpocapsae proteins with high similarity to Shk domain-containing proteins of vertebrate-parasitic nematodes. Bootstrap values of more than 80% (from 1000 replicates) are indicated at nodes. Bar, 10% divergence. (E) A pie chart showing the number of venom proteins that are conserved with at least one vertebrate-parasitic nematode, conserved with at least one other species of Steinernema, or are specific to S. carpocapsae. (F) A pie chart showing the percentage of molecules in the venom that belong to each of the categories above (conserved with at least one vertebrate-parasitic nematode, conserved with at least one other species of Steinernema, or specific to S. carpocapsae).
Table 1.
A large proportion of some protein families are secreted venom proteins.