Microsporidia infection alters C. elegans lipid levels

Abstract

Microsporidia are fungal-related obligate intracellular parasites that infect many types of animals. Microsporidia have exceptionally reduced genomes resulting in limited metabolic capabilities and are thought to be reliant on host metabolism to fuel their own growth. Here, we investigate the impact of microsporidia infection on host lipid metabolism using the nematode Caenorhabditis elegans along with its natural microsporidian pathogen Nematocida parisii. We show that infection causes an increase in the level of C. elegans lipid droplet associated lipase, ATGL-1, and a decrease in host fat levels. A mutation that decreases ATGL-1 activity and overexpression of ATGL-1 did not significantly change N. parisii infection levels. Using lipidomics we show that N. parisii infection decreases C. elegans triglyceride levels and results in increased ceramides that we speculate are synthesized by N. parisii. Mutations in host genes involved in ceramide synthesis did not significantly change the levels of N. parisii infection. Together these results show that microsporidia can cause changes to lipid metabolism of their hosts, but some individual mutations of C. elegans lipid enzymes do not alter microsporidian growth.

Fig 1. ATGL-1 levels increase and lipid droplets decrease in response to N. parisii infection.

(A-B) atgl-1::gfp L1 stage C. elegans were cultured for a total of 72 hours and were either left uninfected or infected for 24, 48, or 72 hours. (A) Representative fluorescence images of live adult atg-1::gfp worms. White lines indicate the boundaries of worm bodies. Scale bars are 124 μm. (B) Quantification of ATGL-1::GFP. Data is from three independent replicates of 18−20 worms each. (C-D). L1 stage Wildtype (N2), atgl-1::gfp, atgl-1 (gk176565) worms of each strain were either infected or left uninfected for 72 hours and stained with Oil Red O. (C) Representative images of Oil Red O stained worms. (D) Quantification percentage Oil Red O intensity per worm. Data is from three independent replicates of at least 20 worms each. Mean ± SD represented by horizontal bars. P-values were determined using one-way ANOVA with post hoc Šidák correction with tests between infected and uninfected samples of the same strain (A and B) and between uninfected samples of different strains (B). Significance is defined as * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, and ns means not significant.

Fig 2. Increased or reduced ATGL-1 activity does not significantly change N. parisii growth.

L1 stage wild-type (N2), atgl-1::gfp, atgl-1 (gk176565) worms were either not infected or infected with N. parisii for 48 (A) or 72 hours (D-E). Worms were then fixed and stained with a FISH probe against the N. parisii 18S RNA and DY96. (A-B) Quantitation of pathogen burden (meronts) per animal. (C) Quantification of pathogen burden (spores) per animal (D) Quantification of number of embryos per animal. (E) Number of embryos per worm, normalized to the uninfected condition for each strain. Data is from three independent replicates of at least 20 worms each. Mean ± SD represented by horizontal bars. P-values were determined using one-way ANOVA with post hoc Dunnett’s test with comparisons to N2 (A-C), Šidák correction (D and E) with tests between infected and uninfected samples of the same strain (D and E), and between infected samples of different strains (E). Significance is defined as * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, and ns means not significant.

Fig 3. N. parisii induces global changes to C. elegans lipidome.

(A-D) C. elegans were infected or left uninfected for 72 h and lipids measured using LC/MS. Heat map showing lipid metabolites that are upregulated (blue) or downregulated (yellow) upon infection at least 2-fold with a significant P-value determined by Student’s T-test (P < 0.01). (A) All classes of lipids with 2 or more lipids significantly up or down regulated. (B) Acylcarnitines. (C) Phosphatidylcholines. (D) Ceramides. Those lipids detected in Burton et al. 2017 colored in orange [55].

Fig 4. Sphingolipid biosynthetic pathway.

Schematic of sphingolipid biosynthesis pathway. Metabolites are indicated in italics, enzymes indicated in bold, and C. elegans enzyme names are indicated in purple [58]. Presence of enzymes from humans, C. elegans, yeast (Saccharomyces cerevisiae), and N. parisii was determined with BLAST, using a threshold E‐value of greater than 10⁻⁵ [59]. The presence of the enzyme is indicated by a colored circle corresponding to the legend at the left.

Fig 5. N. parisii infection is not significantly changed in C. elegans sphingolipid biosynthesis mutants.

(A-D) L1 stage N2 or indicated mutant worms were either not infected or infected with N. parisii for 72 hours. Worms were then fixed and stained with a FISH probe against the N. parisii 18S and DY96. (A) Quantitation of pathogen burden (meronts) per animal. (B) Quantification of pathogen burden (spores) per animal. (C) Quantification of number of embryos per animal. (D) Quantification of population gravidity (gravidity is defined as an animal carrying at least 1 embryo), normalized to the uninfected condition for each strain. (E-F) L1 stage wild-type (N2), asah-1 (ve779), asah-2 (ok564) worms were infected with N. parisii for 48 (E) or 72 hours (F), fixed, and stained with FISH probe and DY96. (E) Quantitation of pathogen burden (meronts) per animal. (F) Quantification of pathogen burden (spores) per animal. (G) L1 stage N2 animals were infected on control or sphingosine containing medium for 72 hours, fixed, and stained with DY96. Pathogen burden (spores) was quantified. Data is from three independent replicates of at least 20 worms each, except for (G) which had two replicates. Mean ± SD represented by horizontal bars. p-values were determined using one-way ANOVA with post hoc Dunnett’s test with comparisons to N2 and those that have a p > .05 and are not shown (A-D), Šidák correction with tests between infected and uninfected samples of the same strain (E), and Tukey’s multiple comparison between all samples (F). Significance is defined as * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, and ns means not significant.