Ciliary intrinsic mechanisms regulate dynamic ciliary extracellular vesicle release from sensory neurons

Summary

Extracellular vesicles (EVs) are submicron membranous structures and key mediators of intercellular communication^{1, 2}. Recent research has highlighted roles for cilia-derived EVs in signal transduction, underscoring their importance as bioactive extracellular organelles containing conserved ciliary signaling proteins^{3, 4}. Members of the TRP channel Polycystin-2 (PKD-2) family are found in ciliary EVs of the green algae Chlamydomonas and the nematode Caenorhabditis elegans ^{5, 6} and in EVs in the mouse embryonic node and isolated from human urine^{7, 8}. In C. elegans, PKD-2 is expressed in male-specific EV releasing sensory neurons, which extend ciliary tips to ciliary pore and directly release EVs into the environment^{6, 9}. Males release EVs in a mechanically-stimulated manner, regulate EV cargo content in response to mating partners, and deposit PKD-2::GFP-labeled EVs on the vulval cuticle of hermaphrodites during mating^{9, 10}. Combined, our findings suggest that ciliary EV release is a dynamic process. Herein, we identify mechanisms controlling dynamic EV shedding using time-lapse imaging. Cilia can sustain the release of PKD-2-labeled EVs for two hours. This extended release doesnť require neuronal transmission. Instead, ciliary intrinsic mechanisms regulate PKD-2 ciliary membrane replenishment and dynamic EV release. The kinesin-3 motor KLP-6 is necessary for initial and extended EV release, while the transition zone protein NPHP-4 is required only for sustained EV release. The dynamic replenishment of PKD-2 at the ciliary tip is key to sustained EV release. Our study provides a comprehensive portrait of real-time ciliary EV release and mechanisms supporting cilia as proficient EV release platforms.

In brief

Wang et al. provide a comprehensive and mechanistic portrait of real-time ciliary EV release. C. elegans sensory neurons continuously release EVs from cilia, a process independent of neuronal transmission. Instead, the ciliary kinesin-3 KLP-6 and transition zone protein NPHP-4 maintain a dynamic ciliary tip pool of PKD-2 for sustained EV release.

Graphical Abstract

Results and discussion

We discovered that cilia are capable of releasing PKD-2::GFP-labeled EVs for a duration of up to two hours. Surprisingly, synaptic transmission is not required for the extended PKD-2 ciliary EV release from sensory neurons. Rather, the ciliary transport machinery involving the kinesin-3 motor KLP-6 and the transition zone protein NPHP-4 are necessary for the replenishment of PKD-2 in the ciliary tip and this prolonged release of EVs. These findings emphasize the critical role of intrinsic ciliary mechanisms in governing the dynamic release of ciliary EVs in a living organism and offer a glimpse into the process by which ciliated sensory neurons release EVs.

Extended release of PKD-2::GFP labeled ciliary EVs for two hours

PKD-2-laden ciliary EVs are generated from both the ciliary base and tip, with the latter being directly released into the surrounding environment and mediating inter-animal communication^{6, 9, 11}. C. elegans males release PKD-2::GFP-labeled ciliary EVs in response to contact with a coverslip while the animals are mounted on glass slides for microscopic imaging ¹⁰. To gain insight into the dynamics of ciliary EV release, we conducted a series of time-lapse imaging experiments. We found that C. elegans males consistently released PKD-2::GFP-labeled ciliary EVs over a span of two hours (Figure 1). We observed environmental EV accumulation at release sites at the head and tail, attributed to the immobilization of animals through anesthesia (Figure 1A). In the head region, EV counts gradually increased, with statistical significance achieved at the two-hour time point (Figure 1B). In the tail region, we observed a more consistent increase in EV numbers, culminating in statistically significant maximum at the one-hour time point (Figure 1B). Further analysis of EV number trajectories of individual animals revealed that, although both the head and tail displayed extended EV release patterns, the tail exhibited a more consistent EV number increase (Figure 1C). We therefore used the more consistent male tail assay for subsequent analysis of EV dynamic release.

Figure 1. Extended release of PKD-2::GFP-labeled extracellular vesicles (EVs) by C. elegans male sensory neurons.

(A) Representative images capturing the head and tail regions of C. elegans males at 0, 1, and 2 hours following mounting on a glass slide.

Blue lines indicate the outline of the male head or tail. Orange lines indicate the outline of the EV clouds released by the head and the tail. Orange arrowheads point to EVs that are released in a string-like pattern.

(B) Quantification of EV counts from the head and tail regions at 0, 1, and 2 hours. The scatter plot with lines indicates the mean ± SEM. Each data point represents the total EV count released by an individual C. elegans male, either from its head (left) or tail (right). n = 16 and the data were scored over 4 days. Statistical analysis was performed using one-way ANOVA with Bonferroni correction. ns denotes not significant (p ≥ 0.05), and * denotes p < 0.05.

(C) Individual trajectories depicting the EV release pattern from the head and tail regions of C. elegans males. Each trajectory represents EV counts from a single animal. The EV count at new time points includes newly released EVs in addition to previously released EVs that remain visible post-photobleaching. Photobleaching explains the decline in EV counts among certain animals at the 2-hour mark, where newly released EVs are fewer than the photobleached older EVs (photobleached twice).

See also Figure S1, Video S1 - S2.

To further characterize the dynamics of EV release during the first hour, we imaged EV release every 10 minutes (Figure S1, Video S1). The number of EVs consistently increased over the course of an hour and reached statistical significance at the 30-minute mark (Figure S1A). Analysis of individual EV trajectories in animals revealed that 5 out of 8 males continuously released EVs over the course of an hour, whereas in 3 out of 8 animals, EV numbers plateaued at the 30-minute mark (Figure S1B). This observation may explain the variability in EV release during the two-hour assay. To visualize real-time EV release, we performed time-lapse imaging within one-minute intervals (Video S2, 39 seconds per frame). As the cilium was deflected and moved along the coverslip, the ciliary tip released EVs that appeared as strings and that formed clouds (Figure 1A, Video S2). Our time-lapse imaging confirms that these continuously released EVs originate from the cilia tips. Collectively, these findings demonstrate the capacity of cilia to release EVs for up to two hours from sensory neurons within living animals.

Synaptic transmission is not required for dynamic PKD-2 ciliary EV release from sensory neurons

Sensory neurons release EVs from cilia when mechanically stimulated and during mating¹⁰. Therefore, we asked whether synaptic transmission and communication from other neurons was necessary for the dynamic PKD-2 ciliary EV release. We examined mutants defective in docking synaptic vesicles (unc-13) and in dense core vesicle exocytosis (unc-31)^12–14, (Figure 2). Neither unc-13 nor unc-31 mutant males displayed deficiencies in PKD-2 ciliary EV release (see Figure 2A–C). At the initial time point (0 hours), both unc-13 and unc-31 mutants exhibited PKD-2 EV release at wild-type levels. The cumulative EV release over one hour was comparable between unc-13, unc-31, and wild-type males (Figure 2). This data indicates that synaptic vesicle and dense core vesicle exocytosis are not required for PKD-2 initial or extended ciliary EV release.

Figure 2. Neuronal transmission-independent release of PKD-2::GFP-labeled ciliary EVs.

(A) Representative images showing EV release from the tail at the initial imaging (0 hr) and the second imaging (1 hr) while the animals were mounted on slides, in wild-type, unc-13, and unc-31 males.

(B) Schematic diagram depicts the functional roles of UNC-13 and UNC-31 in synaptic vesicle- and dense core vesicle-mediated neuronal transmission in the axon.

(C) Quantification of EV release from the tail at the initial imaging (0 hr) and second imaging (1 hr) in wild-type, unc-13, and unc-31 males. The scatter plot with lines indicates the mean ± SEM. Each data point represents the total EV count released by an individual C. elegans male. 12–15 animals were imaged for each genotype. Statistical analysis was performed by two-way ANOVA with Bonferroni correction. ns denotes not significant, * denotes p < 0.05, and ** denotes p < 0.01.

Synaptic transmission is not required for dynamic PKD-2 ciliary EV release. This finding is consistent with a previous report showing that touch-induced calcium responses in these ray RnB neurons do not rely on synaptic transmission¹⁵. This ciliary EV shedding, independent of synaptic transmission, aligns with the finding that ciliary EVs carrying polycystin-2 are released from single-celled Chlamydomonas⁵. Our results are consistent with a ciliary intrinsic mechanism mediating the dynamic release of PKD-2 ciliary EVs.

Both UNC-13 and UNC-31 are conserved bridge molecules between fusing vesicles and target membranes. Neither unc-13 nor unc-31 is required for PKD-2 ciliary EV release. However, we do not rule out that unc-13 and unc-31 regulate MVB-mediated exosome biogenesis in C. elegans. The Drosophila UNC-13 homolog, stac, localizes to the multivesicular body (MVB) and is required for MVB-mediated EV release and tracheal cell fusion in embryos¹⁶. The human ortholog of UNC-13, Munc13–4, is required for MVB maturation and exosome release in cancer cells¹⁷. The UNC-31 ortholog, CAPS1, overexpression promotes exosome-regulated cancer cell migration¹⁸.

Ciliary intrinsic mechanisms regulate continuous release of PKD-2 carrying EVs

The ciliary kinesin-3 motor KLP-6 (kinesin like protein 6) is essential for PKD-2 environmental EV release at the ciliary tip. In klp-6 mutant animals, PKD-2 is not released in environmental EVs, leading to excessive shedding of EVs at the ciliary base, into the glial lumen surrounding the ciliary base⁶, (Figure S2A–D). To explore the potential contribution of the ciliary base EV reservoir to the sustained release of PKD-2 EVs, we conducted a time-lapse analysis of PKD-2 EV release in klp-6 mutant males (Figure 3). klp-6 mutant animals exhibited a defect in environmental EV release at both the initial and the one-hour time point (Figure 3A, B, D). These data show that ciliary base EVs are not released into the environment during the prolonged one-hour time-lapse assay, and that KLP-6 kinesin motor-facilitated enrichment of EV cargos at the ciliary tip is necessary for the dynamic release of PKD-2 ciliary EVs.

Figure 3. Continued PKD-2 ciliary EV release relies on ciliary transport by the kinesin 3 protein KLP-6 and the transition zone protein NPHP-4.

(A-C) Representative images of PKD-2::GFP ciliary EV release in male tail of wild-type, klp-6, and nphp-4 males.

(D) Quantification of EV counts from wild-type and klp-6 mutant males in the tail regions at 0 and 1 hour. The scatter plot with lines indicates the mean ± SEM. Each data point represents the total EV count released by an individual C. elegans male.

(E) Quantification of EV counts from wild-type and nphp-4 mutant males in the tail regions at 0 and 1 hour. The scatter plot with lines indicates the mean ± SEM. Each data point represents the total EV count released by an individual C. elegans male.

For D-E, statistical analysis was performed by two-way ANOVA with Bonferroni correction. n = 12–15 animals per genotype. ns denotes not significant, * denotes p < 0.05, and ** denotes p < 0.01., **** p<0.0001.

(F) Model of PKD-2 ciliary EV dynamics. The sustained release of PKD-2 ciliary EVs over an hour necessitates the dynamic ciliary replenishment of PKD-2. KLP-6 controls ciliary PKD-2 EV release by concentrating PKD-2 at the ciliary tip. Meanwhile, the transition zone protein NPHP-4 regulates the replenishment of the PKD-2::GFP ciliary tip pool.

See also Figure S2–3, Table S1.

To further test the hypothesis that continuous EV release originates from the ciliary tip, we reasoned that a mutant defective in dynamic PKD-2 ciliary membrane replenishment would also be defective in sustained EV release. In cultured mammalian cells, the KLP-6 homolog KIF13B physically interacts with the transition zone protein NPHP4 (nephronophthisis gene 4) to regulate ciliary membrane composition and Sonic hedgehog signaling¹⁹. Therefore, we determined whether NPHP-4 was important for PKD-2 ciliary EV release. We previously found that the nphp-4 mutant does not grossly affect PKD-2 ciliary localization²⁰. At the initial time point, the nphp-4 mutant did not show defects in PKD-2::GFP ciliary EV release. However, at the one-hour time point, nphp-4 mutant males failed to release additional EVs (Figure 3 C, E, Figure S3A–B). Using transmission electron microscopy (TEM), we found that nphp-4 mutants displayed excessive EV accumulation surrounding the ciliary base resembling the klp-6 mutant phenotype in the CEM cilia in male head (Figure S2). Therefore, we hypothesize that the continuous release of EVs requires dynamic PKD-2 transport, a process that may be compromised in nphp-4 mutants, thereby impairing the cilia's capability for continuous EV release (Figure 3F). To test this hypothesis, we characterized the dynamics of PKD-2 ciliary distribution through time-lapse imaging.

Dynamic replenishment of PKD-2 at the ciliary tip underlies sustained PKD-2 carrying ciliary EV release

To visualize ciliary distribution of PKD-2::GFP, we labeled the cilium with the β-tubulin marker TBB-4::tdTomato. PKD-2::GFP is enriched at the ciliary tip and is localized to both the cilium and the ciliary base in the male tail RnB neurons (Figures 4A, Figure S4A–B), consistent with PKD-2::GFP localization in cephalic male (CEM) cilia (Wang, 2021) In wild type, we observed an overall increase in the PKD-2::GFP signal over an hour (Figure 4A,C). In the nphp-4 mutant, the PKD-2::GFP signal decreased at the ciliary tip (Figure 4B, D). These data suggest that cilia actively maintain the PKD-2::GFP enrichment at the ciliary tip, a process that requires NPHP-4. The fluorescence intensity profiling indicated that PKD-2 distributed into three pools: the tip, proximal cilium, and the ciliary base (Figure 4C–D). We quantified the dynamic distribution of PKD-2 along the three pools at the 0 and 1-hour time points. In wild type, the ciliary tip pool of PKD-2 was maintained at a consistent level (Figure 4E). At the 1-hour mark, the ciliary tip PKD-2 level in the nphp-4 mutant was significantly lower than wild type (Figure 4E). Within the proximal cilium, the PKD-2::GFP level remained constant in both wild type and the nphp-4 mutant (Figure 4F). In contrast, the ciliary base pool of PKD-2 in the nphp-4 mutant was lower than wild type at both time points (Figure 4G). We also observed that the PKD-2::GFP fluorescence intensity within the ciliary EVs was slightly lower in the nphp-4 mutant compared to wild type (Figure S4D), suggesting that ciliary EV content might be altered, In Chlamydomonas, transition zone proteins regulate ciliary and ciliary EV composition²¹. We conclude that continuous EV release is facilitated by the dynamic replenishment of PKD-2 at the ciliary tip, and that this process requires the transition zone protein NPHP-4. The diminished ciliary base pool of PKD-2 in the nphp-4 mutant suggests that replenishment of the ciliary tip pool is sourced from the ciliary base, supporting a transport mechanism for PKD-2 tip pool replenishment (Figure 4H).

Figure 4. Maintaining the PKD-2 pool at the ciliary tip ensures continuous EV release.

(A-B) Schematics and representative images of RnB cilia in wild-type and nphp-4 mutant animals at 0-hour and 1-hour time points. PKD-2::GFP localizes to the ciliary base, ciliary membrane, and is enriched at the ciliary tip in both wild-type and nphp-4 mutant animals at the initial 0 hr time point. The ciliary tip pool of PKD-2::GFP is maintained at the one-hour time point in the wild type but not in the nphp-4 mutant. Overall, the nphp-4 mutant displays normal ciliary morphology as visualized by TBB-4::tdTomato. However, the localization of PKD-2::GFP at the ciliary base is reduced in the nphp-4 mutant.

(C-D) Representative profiling graph of ciliary PKD-2::GFP corresponding to the cilium shown in panels (A-B). In wild type, PKD-2 is replenished at the ciliary tip, but not in the nphp-4 mutant. In wild type, the PKD-2 ciliary tip peak is higher at the 1-hour time point than at the 0-hour time point. In the nphp-4 mutant, the PKD-2 ciliary tip peak at the 0-hour time point is comparable to that of the wild type. However, in the nphp-4 mutant the enrichment at the ciliary tip is lower at the 1-hour time point than at the 0-hour time point in the nphp-4 mutant. Although the PKD-2 ciliary membrane fluorescence intensity is comparable between the wild type and the nphp-4 mutant, the fluorescence intensity of PKD-2::GFP at the ciliary base is lower in the nphp-4 mutant than in the wild type.

(E-G) Quantification of PKD-2::GFP dynamic distribution along cilia in wild-type and nphp-4 mutant males. The box plots represent the mean ± SEM. The relative fluorescence intensity of PKD-2 is calculated by normalizing to the distal ciliary PKD-2 fluorescence in the wild type (see methods for details). The sample sizes were 39 cilia from 7 animals for the wild type and 38 cilia from 7 animals for the nphp-4 mutant, respectively. Statistical analyses were conducted using the Kruskal-Wallis test with Dunn's correction. The term 'ns' denotes not significant, while *, **, and *** indicate p-values of < 0.05, < 0.01, and < 0.001, respectively. (E) The ciliary tip pool of PKD-2 is maintained at the 1-hour time point in the wild type but not in the nphp-4 mutant. (F) The ciliary membrane pool of PKD-2 is maintained at a comparable level in both the wild type and the nphp-4 mutant. (G) Overall, PKD-2 is reduced at the ciliary base in the nphp-4 mutant.

(H) Schematic representation of the dynamic replenishment of PKD-2 at the ciliary tip, underlying the cilium's capability for sustained EV release. PKD-2 is localized to three pools: the ciliary base membrane, the ciliary membrane, and the ciliary tip, the latter of which supports continuous EV release. Consequently, when the replenishment of the ciliary tip pool is compromised, the cilia release fewer EVs. The diminished PKD-2 pool at the ciliary base may be one mechanism that impairs the replenishment of the ciliary tip pool in the nphp-4 mutant.

See also Figure S4.

C. elegans males release ciliary EVs in response to mechanical stimuli and in response to mating partners^{9, 10}. Here we comprehensively characterize real-time ciliary EV release and identify components that govern its dynamic regulation. Our study offers novel insights into the dynamic regulation of EV release from cilia of sensory neurons in C. elegans. The prolonged release of PKD-2::GFP-labeled ciliary EVs for up to two hours underscores the robust capability of cilia to produce EVs. This work opens new avenues for exploring the regulators of ciliary EV biogenesis, cargo sorting, and signaling.

EV release dynamics are different between the cephalic CEM neurons and tail RnB neurons, which may be attributed to developmental, anatomical or functional differences. The CEM neurons are born during early embryogenesis, whereas the RnB neurons in the male tail develop post-embryonically^{22, 23}. The CEM cilium is wrapped by two types of glia, while the RnB cilium is ensheathed by a single structural glial cell^{22, 23}. The CEM neurons sense chemical pheromones, while distinct RnB neurons are activated by interactions with a hermaphrodite mating partner^{15, 24–27}. The CEM and RnB neurons express overlapping and distinct sets of genes^{28, 29}. Even endogenously-tagged LOV-1/polycystin-1 and PKD-2/polycystin-2 are expressed at different levels: PKD-2 is abundantly expressed in CEM and RnB neurons while LOV-1 expression is variable in CEM neurons³⁰. Polycystin-mediated male behaviors are genetically separable^{25, 28, 30, 31}. The many distinct properties of CEM and RnB cilia may also render different sensitivity to environmental factors that may influence ciliary EV release. For example, environmental stress affects the level of polyglutamylation in ciliary microtubules and requires the stress response MAPK pmk-1³². Balanced ciliary glutamylation and PMK-1 are important for PKD-2::GFP EV release and male mating behavior^{28, 33, 34}. While mechanical pressure induces EV shedding from RnB neurons¹⁰, we do not know the range of stimuli that may activate EV shedding from CEM, RnBs, or other ciliated neurons in the worm. Our sustained EV release assay provides a new opportunity to study signaling-regulated ciliary EV release.

The requirement for KLP-6 in extended EV shedding is consistent with KLP-6's essential role in EV shedding from the ciliary tip^{6, 9}. Fluorescence profiling of the dynamic distribution of PKD-2 revealed that the ciliary tip pool is renewed and required to sustain ciliary tip EV release. The nphp-4 transition zone mutant failed to replenish PKD-2 ciliary tip pool and displayed a defect in sustained ciliary tip EV release, consistent with the ciliary tip being the source of initial and sustained EV release. The nphp-4 mutant maintained a constant ciliary membrane pool of PKD-2, indicating that PKD-2 ciliary tip enrichment requires an additional step beyond ciliary membrane localization. This phenomenon mirrors that observed in Chlamydomonas, where PKD2 localizes to the proximal and distal flagella via different transport modes for distinct functions^{35, 36}. Whether mammalian primary cilia regulate polycystin subciliary localization remains to be determined.

Transition zone proteins serve as gatekeepers by orchestrating ciliary transport and separating the cilium from the rest of the cell³⁷. In the green algae Chlamydomonas, mutants of nphp-4 exhibit altered ciliary EV compositions²¹ and accumulate ectopic membrane and soluble proteins within the cilia³⁸. In C. elegans amphid and phasmid cilia, NPHP-4 genetically interacts with homodimeric kinesin-2 OSM-3³⁹, suggesting that NPHP4 may influence transport mediated by kinesin-2 and kinesin-3 ciliary motors¹⁹. The transition zone protein NPHP4 actively supports PKD-2 ciliary tip dynamics and may alter the release efficiency and composition of ciliary EVs, pointing at a role of abnormal ciliary EVs in pathogenesis of ciliopathies. Our C. elegans system enables discovery of conserved mechanisms regulating ciliary membrane renewal and ciliary ectocytosis, which provides much needed in vivo insight to EV biogenesis and signaling.

STAR★METHODS

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contacts: Maureen Barr (mmbarr@rutgers.edu).

Materials availability

Plasmids and transgenic C. elegans strains are available upon request.

Data and code availability

All data reported in this paper will be shared by the lead contact upon request.

This paper does not report original code.

Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Experimental model and subject details

C. elegans culture and genetics were performed as described⁴³. C. elegans strains were maintained at 20°C on nematode growth medium (NGM) plates seeded with Escherichia coli (OP50 strain) as a food source.

Strain identifiers and genotypes are cataloged in the Key Resources Table.

Key Resources Table.

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Bacterial Strains
E. coli, OP50 strain	http://www.cgc.umn.edu	RRID: WB-STRAIN: OP50
Chemicals, Peptides, and Recombinant Proteins
Levamisole	http://www.acros.com	Cat # 187870100
Agarose	https://www.sigmaaldrich.com/	Cat # A9539
Glutaraldehyde	http://www.emsdiasum.com	Cat # 16220
Sodium cacodylate trihydrate	http://www.emsdiasum.com	Cat # 12300–25
Osmium Tetroxide	http://www.emsdiasum.com	Cat # 19110
Uranyl acetate	https://www.emsdiasum.com	Cat # 22400
Sodium acetate	https://www.sigmaaldrich.com/	Cat # 1.06264
Hard-Plus Resin-812	https://www.emsdiasum.com	Cat # 14115
Experimental Models: Organisms/Strains
C. elegans strain name	Genotype	Reference
PT621: him-5(e1490) myIs4 [PKD-2::GFP+Punc-122::GFP]V	Bae et al (2006)41	RRID: WB-STRAIN: PT621
PT3716: unc-13(e51)I; him-5(e1490), myIs4V	this paper	RRID: WB-STRAIN: PT3716
PT3038: unc-31(e169)IV; him-5(e1490), myIs4V	This paper	RRID: WB-STRAIN: PT3038
PT443: myIs1[PKD2::GFP] pkd-2(sy606) IV; him-5(e1490)	Bae et al (2006)41	RRID: WB-STRAIN: PT443
PT1195: klp-6(my8) III; him-5 (e1490), myIs4 V	Morsci and Barr (2011)42	RRID: WB-STRAIN: PT1195
PT835: pkd-2(sy606) myIs1 IV; nphp-4(tm925) him-5(e1490) V	Jauregui et al (2008)16	RRID: WB-STRAIN: PT835
PT1435: pha-1(e2123) III; myIs1 IV; him-5(e1490) V; myEx552 [Ppkd-2::tbb-4::tdTomato]	Jauregui et al (2008)16	RRID: WB-STRAIN: PT1435
PT1439: pha-1(e2123) III; myIs1 IV; nph-4(tm925) him-5(e1490) V; myEx552 [Ppkd-2::tbb-4::tdTomato]	Jauregui et al (2008)16	RRID: WB-STRAIN: PT1439
Software and Algorithms
Zen Blue 2.0	https://www.zeiss.com	N/A
Zen Black 2.0	https://www.zeiss.com	N/A
Imaris 9.5	https://imaris.oxinst.com	N/A
Prism 9	http://www.graphpad.com	N/A
Post-hoc Power Calculator	https://clincalc.com/stats/Power.aspx	N/A

Method details

Airyscan super-resolution microscopy

Super-resolution imaging was performed on the Zeiss LSM880 confocal system equipped with Airyscan Super-Resolution Detector with 7 Single Photon Lasers (405, 458.488, 514, 561, 594, 633nm), Axio Observer 7 Motorized Inverted Microscope, Motorized X-Y Stage with Z-Piezo, T-PMT.

Time-lapse imaging and PKD-2::GFP EVs quantification

The day before imaging, 20–30 L4 larval males were picked to NGM plates freshly seeded with E. coli OP50 to attain a synchronized population. On the day of imaging, now-adult male C. elegans were picked from the NGM plates to a coverslip containing 1 μL drops of 10 mM levamisole in water and mounted on a 10% agarose pad on a slide, and imaged in 1-minute, 10-minute or one-hour intervals. All images were taken with Airyscan super-resolution on an LSM 880 confocal microscope (Zeiss). Images were saved and processed using Zen Black software (Zeiss) to generate maximum intensity projections and Zen Blue software (Zeiss) image analysis wizard to quantify the number and fluorescence intensity of EVs. Images were processed and prepared for publication using Imaris, FIJI software, Adobe Illustrator and Adobe Photoshop. Maximum intensity projects of Z-stack images were inverted and adjusted to grayscale in Adobe Photoshop.

Profiling of PKD-2::GFP distribution along cilia and at the ciliary base

Z-stack acquisition for each male tail were performed to encompass environmental EVs, the cilium, and the ciliary base. Time-lapse images were captured using a Plan-Apochromat 63x/1.40 Oil objective with a 2x zoom acquisition area under the Airyscan super-resolution mode at initial (0 hr) and one-hour (1 hr) time points. Raw image files were processed using the Airyscan Processing program within the Zen Black 2.0 software. Maximum projection images of each Z stack were analyzed using the Profiling Program in Zen Blue software. The total fluorescence intensity and the distance from the ciliary tip were recorded in an Excel file for subsequent data analysis. The fluorescence intensity of the distal cilia was determined by measuring 0 – 1000 nm from the ciliary tip, delineating the end of PKD-2 ciliary tip enrichment. The fluorescence intensity of the proximal cilia was determined by a 1000–2000 nm distance from the ciliary tip, whereas the ciliary base fluorescence intensity was determined from 2000 nm from the ciliary tip to the end of the PKD-2::GFP signal at the ciliary base. Relative fluorescence intensity was normalized to the mean fluorescence intensity of the distal ciliary PKD-2::GFP.

Transmission Electron Microscopy (TEM)

For method details, refer to Jauregui et al (2008)²⁰. Briefly, worms were fixed in 3% glutaraldehyde in cacodylate buffer on ice. Then, heads were cut off, moved to fresh fixative, and held in fix overnight at 4°C. Animals were rinsed in cacodylate buffer and stained with 1% osmium tetroxide in cacodylate buffer for 1 hr at 4°C. After embedding in small groups in agarose, the specimens were en bloc stained with 1% uranyl acetate in sodium acetate buffer, dehydrated, and embedded in Embed812 resin according to the general procedures described by Hall (1995)⁴⁴. Thin sections were collected on a diamond knife and post-stained before being viewed on a Philips CM10 electron microscope.

Quantification and statistical analysis

The Prism software package (GraphPad Software 8) was used to carry out statistical analyses. Information about statistical tests, p values and n numbers are provided in the respective figures and figure legends. Sample size was established using G-power software to be able to detect moderate effects with 80% power at P = 0.05. G-power is calculated with Post-hoc Power Calculator (clincalc.com), at website: https://clincalc.com/stats/Power.aspx.

Supplementary Material

3

Video S1, Timelapse capture of dynamic EV release at 10-minute intervals per frame, related to Figure 1. The video presents time-lapse imaging captured at 10-minute intervals over a one-hour period, demonstrating the continuous release of EVs throughout that duration. The first frame displays a cartoon of a C. elegans male; the rectangle indicates the imaging area for the micrograph shown below. The micrograph is an overlap of bright field and a fluorescence image of the male tail, demonstrating that PKD-2::GFP is enriched at the ciliary tip and is released from the ray pore structure into the environment outside the worm. The image is adapted from Figure 1 in Wang et al (2020)¹⁰.

4

Video S2, Timelapse capture of dynamic EV release at 39 second intervals per frame, related to Figure 1. The video presents time-lapse imaging captured at 39-second intervals over five minutes demonstrating the continuous release of EVs from the ciliary tip. The left panel displays a cartoon of a C. elegans male; the rectangle indicates the imaging area for the micrograph shown on the right. The micrograph is an overlap of bright field and a fluorescence image of the male tail, demonstrating that PKD-2::GFP is enriched at the ciliary tip and is released from the ray pore structure into the environment outside the worm. The image is adapted from Figure 1 in Wang et al (2020)¹⁰.

Highlights.

C. elegans neurons dynamically release PKD-2 carrying ciliary EVs for two hours.

Sustained ciliary EV release does not rely on neuronal transmission.

Replenishment of PKD-2 at the ciliary tip is key to sustained release of ciliary EVs.

Ciliary tip replenishment requires transition zone component NPHP-4.