C. elegans ciliated sensory neurons release extracellular vesicles that function in animal communication

Abstract

Cells release extracellular vesicles (ECVs) that play important roles in intercellular communication and may mediate a broad range of physiological and pathological processes. Many fundamental aspects of ECV biogenesis and signaling have yet to be determined, with ECV detection being a challenge and obstacle due to the small size (100 nm) of the ECVs. We developed an in vivo system to visualize the dynamic release of GFP-labeled ECVs. We show here that specific Caenorhabdidits elegans ciliated sensory neurons shed and release ECVs containing GFP-tagged polycystins LOV-1 and PKD-2. These ECVs are also abundant in the lumen surrounding the cilium. Electron tomography and genetic analysis indicate that ECV biogenesis occurs via budding from the plasma membrane at the ciliary base and not via fusion of multivesicular bodies. Intraflagellar transport and kinesin-3 KLP-6 are required for environmental release of PKD-2::GFP-containing ECVs. ECVs isolated from wild-type animals induce male tail-chasing behavior, while ECVs isolated from klp-6 animals and lacking PKD-2::GFP do not. We conclude that environmentally released ECVs play a role in animal communication and mating-related behaviors.

Figure 1

IL2 and male-specific B-type ciliated neurons release GFP-labeled ECVs. Top panel, cartoon of six IL2 and B-type sensory neurons in adult C. elegans male (in the head, four CEM neurons, and in the tail, one HOB and 16 RnB neurons). (A, B) Male head and tail images of LOV-1::GFP reporter (N-terminal extracellular domain of LOV-1 (1–991 aa) fused to GFP). In all panels, red arrows point to ECVs surrounding the head and the tail. Insets show framed area zoomed to 4X, and increased brightness to show GFP-labeled ECVs. (C, D) Male head and tail images of PKD-2::GFP reporter. Green arrow head points to a CEM cilium with PKD-2::GFP enriched at the tip. Yellow arrows point to the cuticular pore of the ray neurons and PKD-2::GFP release around the pore. (E, F) CWP-1::GFP release from the head and the tail. (G–J) Negatively stained ECVs. (G) ECVs with no primary antibody control. (H, I and J) Different images of LOV-1 antibody labeling endogenous LOV-1 on ECVs purified from wild-type hermaphrodites and males without a GFP transgene, detection by 0.8 nm ultrasmall gold, followed by silver enhancement. Scale bar of A–F is 10 μm, of G–J is 100 nm. (K) Model based on electron tomography (ET) of the distal end of the CEM neuron and its surroundings. The glial sheath cell and socket cell form a continuous lumen surrounding the CEM neuron cilium, which is exposed to the environment directly through a cuticular opening. The lumen is shared by CEM and CEP neurons. The CEM neuron is more centrally located in the lumen, while the CEP neuron is closer to the side of the lumen. ECVs are observed in the lumen (161.8 ± 82.8 nm, Average ±SD of vesicles in each lumen).

Figure 2

ECV release is constitutive, independent of ESCRT-0 and -I components, and dependent on IFT and the kinesin-3 klp-6. (A) Wild-type L4 larvae, virgin males and mated males equally shed and release PKD-2::GFP ECVs. (B) PKD-2::GFP ECV release is impaired in daf-10 (IFT-A), osm-5 (IFT-B), osm-3;klp-11 (redundant anterograde kinesin-2 motors), klp-6 (kinesin-3) and che-3 (dynein heavy chain, retrograde motor). osm-3, klp-11, and bbs-7 and MVB biogenesis pathway single mutants are not defective in PKD-2::GFP release. Error bar indicates 95% confident intervals, * Different from wild type, Fisher’s exact test, Bonferroni-Holm corrected P < 0.05). (C–D) Cross sections of the cephalic sensillum at CEM transition zone level in wild type and klp-6 mutant. Yellow arrows point to the CEP neuron; red arrows point to the CEM neuron; red arrowheads point to the ECVs. Scale bar is 200 nm. (E) Quantification of the lumen cross-sectional area in wild-type and klp-6 mutant males. (F) klp-6 mutants accumulate PKD-2::GFP at the ciliary region, and the ratio of total fluorescence intensity of cilium to soma is significantly greater than wild type (in E–F, error bar indicates Stdev, *** p<0.001 by Mann-Whitney test).

Figure 3

Purified ECVs promote adult male-specific behaviors. (A) Diagram of choice assay plate. Three spots of M9 (negative control), wild-type ECV suspension, and klp-6 ECV suspension were placed equidistance apart on a 5 cm seeded NGM plate. A male was placed in the center of the plate and his behavior recorded for five minutes. (B) The number of times the male visited each test spot is not significantly different. (C) Males exhibit more reversals on either the wild-type or the klp-6 ECV spot than on the M9 buffer (minus ECV) control spot (p<0.001). The number of reversals is not different between the wild-type and klp-6 ECV samples. Statistical analysis was done by Kruskal-Wallis test, six independent trials with 10 assays per trials. (D) To measure tail chasing behavior, the male was placed in a bigger spot comprised of four smaller drops of the ECV preparation, with small food lawn to restrict the male from wandering too far away from the spot. (E) Males displayed tail chasing behavior only on wild-type ECVs (Supplemental Movie 3) but not on klp-6 ECVs or M9 buffer control. Six independent trials with 20, 21, and 25 individual male assays for wild-type ECV, klp-6 ECV and M9 buffer control respectively were performed. * Different from wild type, Fisher’s exact test, Bonferroni-Holm corrected P < 0.05).