Fig 1.
Gram-negative bacteria found in the human gut affect C. elegans proteostasis and enhance protein aggregation in the intestine.
A) Age-dependent aggregation of intestinal polyQs (polyQ44) colonized by E. coli OP50. Data are represented as the number of aggregates per 20–30 intestines quantified in animals at day 1–5 post-hatching. B) The effect of select enteric bacteria on polyQ aggregation in the C. elegans intestine. Data are represented as the average number of aggregates per C. elegans intestine normalized to E. coli OP50 control strain (lower dotted line). Each data point is an average of a minimum of three independent experiments with a total of at least 100 animals. Black circles represent control bacteria and bacteria that did not have any significant effect on polyglutamine aggregation upon colonization of the C. elegans intestine. Red solid circles represent bacteria that significantly enhanced aggregation by <3-fold. Red open circles represent bacteria that significantly enhanced aggregation by >3-fold (arbitrary threshold). Arrows represent bacteria that were chosen for follow-up experiments. C) The effect of select commensal bacteria on polyQ aggregation in the C. elegans intestine. Data are represented as the average number of aggregates per C. elegans intestine normalized to E. coli OP50 control strain. Each data point is an average of a minimum of three independent experiments with a total of at least 100 animals. D) PCR confirmation of E. coli MG1655 curli mutant strains. WT, csgA::kan (left two bands) amplified with primers flanking the ΔcsgA locus. WT, csgD::kan (right two bans) amplified with primers flanking the ΔcsgD locus. E) Phenotypic confirmation of the curli-deficient ΔcsgA and ΔcsgD mutant strains using Congo Red plate assay. F) ProteoStat staining of total aggregates produced by E. coli MG1655 wild-type (WT), MG1655 ΔcsgA, and MG1655 ΔcsgD. Data are represented as the average fluorescent signal per bacterial strain stained with ProteoStat. Each data point is an average of two independent experiments with three replicates per run. G) The effect of E. coli MG1655 curli mutants on polyQ (polyQ44) aggregation in the C. elegans intestine. Data are represented as the average number of aggregates per C. elegans intestine. Each data point is an average of three independent experiments with a total of 90 animals. Error bars represent standard error of the mean (SEM). Statistical significance was calculated using one-way analysis of variance (ANOVA) followed by multiple comparison Dunnett’s post-hoc test (****p< 0.0001).
Fig 2.
Colonization of C. elegans expressing intestinal polyQ with human enteric pathogens influences motility.
A) A cartoon depicting time-off-pick (TOP) phenotype measure as the time (seconds) it takes a worm to crawl off a pick when pickup up by the midbody section. B) Age-dependent changes of motility in animals expressing intestinal polyQ (polyQ33 and polyQ44) grown on E. coli OP50 control strain. Higher motility measured in TOP seconds indicates a higher motility defect. Data are represented as the average TOP per worm. Each data point represents 20 worms. C) Motility of animals expressing intestinal polyQ grown on K. pneumoniae KP182 and P. aeruginosa PAO1. Data are represented as the average TOP per 60 worms over three independent experiments, normalized to E. coli OP50 control strain. Error bars represent SEM. Statistical significance was calculated using one-way ANOVA followed by multiple comparison Dunnett’s post-hoc test (**p<0.01, ****p <0.0001).
Fig 3.
Colonization of the C. elegans intestine with human enteric pathogens influences proteostasis in a tissue non-autonomous manner.
A) The effect of select bacteria on protein folding in the muscle. Data are represented as the average number of aggregates per worm normalized to E. coli OP50 control strain. Each bar is an average of three independent experiments with a total of 100 animals. B) Age-dependent decline in motility assessed by enumerating body bends per 30 seconds in muscle-specific polyQ35 and N2 control worms. The data are represented as the average number of body bends quantified on day 3–5 post-hatching in a total of 15 animals per each day. The effect of bacteria on the motility of C) control animals and D) animals expressing muscle-specific polyQ35. Data are represented as the average number of body bends per worm normalized to worms fed E. coli OP50 control strain. Each bar is an average of three independent experiments with a total of 45 animals. E) Age-dependent decline in motility assessed by increased TOP. The effect of bacteria on the motility of F) control animals and G) animals expressing neuron-specific polyQ40. Data are represented as the average TOP per worm normalized to animals fed E. coli OP50 control strain. Each bar is an average of three independent experiments with a total of 60 animals. H) A cartoon depicting experiments which demonstrated that bacterial colonization of the C. elegans intestine affects the F1 generation. I) Quantification of intestinal aggregates in the F1 and F2 generations from parental animals that were fed select test and control bacteria. Each bar represents the average number of aggregates per intestine normalized to the control (E. coli OP50). Data are representative of three (F1) and one (F2) independent experiments with a total of 100 and 30 animals, respectively. Error bars represent SEM. Statistical significance was calculated using one-way ANOVA followed by multiple comparison Dunnett’s post-hoc test (*p<0.05, **p<0.01, ****p<0.0001).
Fig 4.
The effect of butyrate on bacteria-induced aggregation in the intestine.
Data are represented as the average number of aggregates per intestine normalized to the control (0 mM butyrate). Each bar is an average of three independent experiments with a total of 100 animals. Error bars represent SEM. Statistical significance was calculated using one-way ANOVA followed by multiple comparison Dunnett’s post-hoc test (*p<0.05, **p<0.01, ****p<0.0001).
Fig 5.
The effect of butyrate on toxicity associated with bacteria-mediated intestinal polyQ aggregation.
Aggregation-dependent toxicity is assessed by measuring motility (TOP phenotype) in the presence of 100 mM butyrate. Data are represented as the average number of TOP seconds per worm normalized to the control (0 mM butyrate). Two roller strains, Control (AM446, no polyQ) and polyQ33 (no aggregates), are controls. Each bar is an average of three independent experiments with a total of 60 animals. Error bars represent SEM. Statistical significance between each pair was calculated using Student’s t-test (*p<0.05, ****p<0.0001).
Fig 6.
Butyrate affects bacteria-mediated protein aggregation in the C. elegans muscle.
Data are represented as the average number of aggregates of muscle-specific polyQ35 (AM140) per worm normalized to the control (0 mM butyrate). Each bar is an average of three independent experiments with a total of 100 animals and the error bars represent SEM. Statistical significance was calculated using one-way ANOVA followed by multiple comparison Dunnett’s post-hoc test (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).
Fig 7.
SKN-1 and DAF-16 are involved in butyrate-mediated suppression of aggregation.
A) The effect of hsf-1, skn-1, and daf-16 knockdown on intestine-specific polyQ44 aggregation in the presence of exogenous butyrate supplementation, compared to empty vector control (L4440). Data are represented as the average number of aggregates per C. elegans intestine. Each data point is an average of a minimum of three independent experiments with a total of at least 85 animals. Error bars represent SEM. Statistical significance was calculated using one-way ANOVA followed by multiple comparison Dunnett’s post-hoc test (*p<0.05, ****p<0.0001). B) Fluorescent and Nomarski images of worms expressing reporter constructs regulated by HSF-1, SKN-1, and DAF-16. Worms were fed E. coli expressing either the control empty vector (EV) shown in the left panels or specific RNAi shown in the right panels. To activate the reporters, animals expressing hsp70p::GFP and gcs-1p::GFP were either heat shocked (HS) or exposed to 5 mM acrylamide (AA), respectively. Functional knockdowns were confirmed by the attenuated GFP expression in animals fed RNAi versus EV. Scale bar = 200 μm.
Fig 8.
Butyrate-producing E. coli suppresses aggregation and the associated toxicity.
Animals were fed four different strains of bacteria: non-butyrogenic controls (E. coli OP50, E. coliWT, E. coliΔ) and conditional butyrogenic E. coli (E. coliBt). A) The graphs represent the average number of intestinal polyQ44 aggregates per worm normalized to the control (no arabinose). Each bar is an average of three independent experiments with a total of 100 animals. B) Intestinal aggregate-dependent toxicity normalized to the control (no arabinose) assessed with the TOP phenotype. The left panel represents roller worms (Control), the middle panel represents polyQ33 worms, and the right panel represents worms expressing polyQ44. Each bar is an average of three independent experiments with a total of 60 animals. Error bars represent SEM. Statistical significance between each pair was calculated using Student’s t-test (*p<0.05, ****p<0.0001).
Fig 9.
Colonization of the C. elegans intestine with butyrogenic E. coli suppresses aggregation and associated toxicity in distal tissues.
A) The effect of butyrogenic bacteria (E. coliBt) on the aggregation profile of worms expressing muscle-specific polyQ35. Data are represented as the average number of aggregates of muscle-specific polyQ35 (AM140) per worm normalized to the control (no arabinose). Animals were fed four different strains of bacteria: non-butyrogenic controls (E. coli OP50, E. coliWT, E. coliΔ) and conditional butyrogenic E. coli (E. coliBt). B) The effect of butyrogenic bacteria on the motility of animals expressing muscle-specific polyQ35 and C) control N2 worms. Data are represented as the average number of body bends per worm normalized to control (no arabinose). Each bar is an average of three independent experiments with a total of 45 animals. D) The effect of butyrogenic bacteria on the motility of animals expressing neuronal polyQ40 and on E) control N2 worms. Data are represented as the average TOP seconds per worm normalized to control animals (no arabinose). Error bars represent SEM. Statistical significance between each pair was calculated using Student’s t-test (*p<0.05, **p<0.01, ***p<0.0005 ****p<0.0001).
Fig 10.
Co-colonization of E. coliBt suppresses P. aeruginosa-mediated enhancement of polyQ aggregation.
A) The effect of butyrogenic bacteria (E. coliBt) on the aggregation of intestinal polyQ44 (AM738) in worms co-colonized with E. coli OP50 or P. aeruginosa in the presence and absence of 3% L-arabinose. Data are represented as the average number of aggregates per intestine. Each bar is an average of three independent experiments with a total of 100 animals. Error bars represent SEM. Statistical significance between each pair was calculated using Student’s t-test (****p<0.0001). B) Model figure. Enteropathogenic bacteria disrupt proteostasis across C. elegans tissues, including intestine, muscle, neurons, and gonads. The bacteria-mediated proteotoxicity is alleviated by butyrate and butyrogenic bacteria.
Table 1.
Reagents and resources used in this study and their associated sources and identifiers.