Loss of FLWR-1 induces defects in neuro-transmission fol-lowing intense stimulation.

(A) Alignment of the amino acid sequences of FLWR-1 to hFwe4 (Homo sapiens) and Fwe-Ubi (Drosophila melanogaster). Shading depicts evolutionary conservation of amino acid residues (identity – black; homology – grey). (B) Schematic representation of the flwr-1/F20D1.1 gene locus and the size of the ok3128 deletion. Bars represent exons and connecting lines introns. (C) Mean (± SEM) swimming cycles of animals expressing ChR2(H134R) in cholinergic motor neurons (unc-17 promoter). All animals were treated with ATR. A 90 s light pulse (470 nm, 1 mW/mm2) was applied after 30 s as indicated by the blue shade. Number of animals accumulated from N = 3 biological replicates: wild type = 80 – 88, flwr-1 = 80 – 91, FLWR-1 rescue = 62 – 75. (D) Statistical analysis of swimming speed at different time points as depicted in (B). Mean (± SEM). Each dot represents a single animal. Mixed-effects model analysis with Tukey’s correction. (E) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb. N = 4 biological replicates. Two-way ANOVA with Tukey’s correction. ns not significant, *** p < 0.001.

FLWR-1 is expressed in neuronal cells and localizes to synaptic vesicles.

(A - D) Confocal micrographs (maximum projection of z-stacks or single plane) of animals co-expressing pflwr-1::GFP::FLWR-1 and psnb-1::mCherry::SNB-1. (A) Overview of GFP::FLWR-1 expression. Arrows indicate dorsal (DNC) and ventral (VNC) nerve cords. Scale bar, 100 µm. (B) GFP::FLWR-1 in head neurons and pharynx. Scale bar, 20 µm. (C) Animal depicted in (B), single plane showing neck muscle cells. Arrows indicate GFP::FLWR-1 localization to the plasma membrane. Scale bar, 20 µm. (D) GFP and mCherry fluorescence in the DNC. Scale bar, 10 µm. (E) Line scan analysis of colocalization of GFP::FLWR-1 and mCherry::SNB-1 along the DNC as represented in (D). R2 as determined by Pearson correlation. a.u. = arbitrary units of fluorescence intensity. (F) Micrograph depicting GFP::FLWR-1 and mCherry::SNB-1 fluorescence in sublateral nerve cords and commissures. Arrowheads indicate synaptic puncta. Arrow points towards SV precursor travelling along commissure as shown in Supplementary Movie 1. Scale bar, 10 µm. (G) Kymograph representing the SV precursor indicated in (F) travelling along commissures. Scale bar, 2 µm. (H) Comparison of the ratio of DNC to VNC fluorescence of GFP::FLWR-1 and mCherry::SNB-1 in wild type and unc-104(e1265) mutant background. Mean (± SEM). Each dot represents a single animal. Two-way ANOVA with Šídák’s correction. *** p < 0.001. Number of animals imaged in N = 3 biological replicates: wild type = 33, unc-104 = 29.

GABAergic signaling is increased in flwr-1 knockout mutants.

(A) Representative confocal micrographs of the DNCs in animals co-expressing pflwr-1::GFP::FLWR-1and TagRFP::ELKS-1 in cholinergic motor neurons (unc-17(short) promoter). Scale bar, 5 µm. (B) Representative confocal micrographs of the DNCs in animals co-expressing pflwr-1::GFP::FLWR-1and TagRFP::ELKS-1 in GABAergic neurons (unc-47 promoter). Scale bar, 5 µm. (C) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb with cholinergic (punc-17) and GABAergic (punc-47) expression of FLWR-1 in flwr-1(ok3128) mutant background. N = 3 - 8 biological replicates. (D) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb with unc-47(e307) and unc-47(e407)/flwr-1(ok3128) double mutants. N = 3 biological replicates. (C + D) Two-way ANOVA with Tukey’s correction. ns not significant, *** p < 0.001.

Loss of FLWR-1 induces increased neuronal excitability.

(A) Mean (± SEM) normalized DNC fluorescence of animals expressing GCaMP3 and ChrimsonSA in GABAergic neurons (unc-25 promoter). All animals were supplemented with ATR. A 10 s light pulse (590 nm, 40 µW/mm2) was applied after 5 s as indicated by red shade. (B) Mean (± SEM) normalized fluorescence during stimulation (seconds 6 to 14) as depicted in (A). Each dot indicates a single animal. Unpaired t-test. ** p < 0.01. (A + B) Number of animals imaged in N = 5 biological replicates: wild type = 40, flwr-1 = 41. Outliers were removed from both datasets as detected by the iterative Grubb’s method (GraphPad Prism). (C) Mean (± SEM) body length of animals expressing ChR2(H134R) in GABAergic neurons (unc-47 promoter) in the unc-29(e1072) mutant background, normalized to the average before stimulation. All animals were supplemented with ATR. A 20 s light pulse (470 nm, 100 µW/mm2) was applied after 5 s as indicated by blue shade. (D) Mean (± SEM) relative body length during stimulation (seconds 6 to 24) as depicted in (C). Each dot indicates a single animal. Unpaired t-test. * p < 0.05. Number of animals measured in N = 4 biological replicates: wild type = 51, flwr-1 = 49. (E) Mean (± SEM) normalized DNC fluorescence of animals expressing GCaMP6f and ChrimsonSA in cholinergic motor neurons (unc-17b promoter). All animals were supplemented with ATR. A 10 s light pulse (590 nm, 40 µW/mm2) was applied after 5 s as indicated by red shade. (F) Median (with IQR) normalized fluorescence during stimulation (seconds 6 to 14) as depicted in (E). Each dot indicates a single animal. Mann-Whitney test. ** p < 0.01. (E + F) Number of animals imaged in N = 6 biological replicates: wild type = 41, flwr-1 = 43. Outliers were removed from both datasets as in (B). (G) Schematic representation of motor neuron innervation of BWMs. Arrows indicate putative increased (green) or decreased (red) neurotransmission/excitation of the involved cell types in flwr-1 mutants compared to wildtype.

FLWR-1 facilitates endocytosis in non-neuronal and neuronal cells.

(A) Exemplary images of coelomocytes in wild type and flwr-1(ok3128) animals expressing secreted ssGFP in BWMs (myo-3 promoter). Scale bar, 10 µm. (B) Mean (± SEM) background corrected fluorescence of coelomocytes. Each dot indicates the mean coelomocyte fluorescence per animal. a.u. = arbitrary units of fluorescence intensity. Mann-Whitney test. *** p < 0.001. Number of animals imaged in N = 3 biological replicates: wild type = 30, flwr-1 = 29. (C) Mean (± SEM) normalized DNC fluorescence of animals expressing SNG-1::pHluorin and ChrimsonSA in cholinergic neurons (unc-17 promoter). All animals were supplemented with ATR. A 30 s light pulse (590 nm, 40 µW/mm2) was applied after 10 s as indicated by the red shade. (D) Mean (± SEM) pHluorin fluorescence as depicted in (C) but additionally normalized to the maximum value of each dataset. (E) Mean (± SEM) normalized fluorescence during stimulation (seconds 15 to 35) as depicted in (C). Each dot indicates a single animal. Unpaired t-test. (F) Mean (± SEM) calculated exponential decay constants of fluorescence decline after stimulation. Each dot indicates a single animal. Unpaired t-test. (C – F) Number of animals imaged in N = 5 biological replicates: wild type = 27, flwr-1 = 32. * p < 0.05. (G) Representative voltage-clamp recordings of currents detected in BWMs. Animals express ChR2(H134R) in cholinergic motor neurons (unc-17 promoter, transgene zxIs6) and have been treated with ATR. A 30 s light stimulus (470 nm, 8 mW/mm2) was applied as indicated by blue bars. (H) Normalized mPSC frequency in BWMs. All animals were treated with ATR. A 30 s light pulse (470 nm, 8 mW/mm2) was applied as indicated by blue shade. Dashed lines indicate one-phase exponential regression analysis fitted to the mean mPSC frequencies during stimulation. Calculated time constants of decay are shown. Two-way ANOVA with Šidák’s correction. All significant differences to wild type are depicted. (I) mPSC amplitude in BWMs of animals measured in (G + H). (J) Mean (± SEM) mPSC amplitude during light stimulation as indicated in (I). Unpaired t-test. * p < 0.05. (G - J) Number of animals: wild type = 9, flwr-1 = 8.

Ultra-structural analysis reveals defective recycling of SVs after stimulation in flwr-1 mutants.

(A) Representative TEM micrographs of cholinergic en-passant synapses in wild type and flwr-1(ok3128) animals expressing ChR2(H134R) in cholinergic motor neurons (unc-17 promoter). Animals were optionally treated with ATR as indicated. Dense projections (DP), endosomes (E), dense core vesicles (blue arrows), SVs (black arrowheads), docked vesicles (white arrowheads) and large vesicles (LV) are indicated. Scale bars, 100 nm. (B) Violin plot depicting the number of SVs counted per synaptic profile. (C) Violin plot depicting the number of docked vesicles observed per synaptic profile. (D) Violin plot depicting the number of endosomes per synapse. (B - D) Bold line represents the median, and the dashed lines the IQR. Kruskal-Wallis test. Only statistically significant differences are depicted. * p < 0.05,** p < 0.01, *** p < 0.001. Number of synaptic profiles imaged: wild type (-ATR) = 56, wild type (-ATR) = 51, flwr-1 (-ATR) = 55, flwr-1 (+ATR) = 59.

Increased excitability in flwr-1 mutants may be caused by negative regulation of MCA-3.

(A) Schematic representation of the mca-3 gene locus including exon structure of isoforms mca-3a and mca-3b. Bars represent exons and connecting lines introns. The size of the ok2048 deletion as well as the putative calmodulin-binding domain are indicated. (B) Mean (± SEM) normalized DNC fluorescence of animals expressing GCaMP6f and ChrimsonSA in cholinergic motor neurons (unc-17b promoter). All animals were supplemented with ATR. A 10 s light pulse (590 nm, 40 µW/mm2) was applied after 5 s as indicated by red shade. (C) Median (with IQR) normalized fluorescence during stimulation (seconds 6 to 14) as depicted in (B). Each dot indicates a single animal. Kruskall-Wallis test. Only statistically significant differences are depicted. ** p < 0.01, *** p < 0.001. (B + C) Number of animals imaged: wild type = 65, flwr-1 = 43, mca-3 = 35, mca-3; flwr-1 = 28. Outliers were removed from all datasets as detected by iterative Grubb’s method (GraphPad Prism). (D) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb. Two-way ANOVA with Tukey’s correction. ns not significant, *** p < 0.001.

Basic amino acid residues on the intracellular surface of FLWR-1 may be involved in PI(4,5)P2 lipid binding.

(A) Partial alignment of the amino acid sequences of FLWR-1, hFwe4 (Homo sapiens), and Fwe-Ubi (Drosophila melanogaster). Shading depicts evolutionary conservation of amino acid residues (black – identity; grey - homology). (B) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb. A mutated FLWR-1 in which glutamate 74 was changed to glutamine was expressed in flwr-1(ok3128) mutant background under its endogenous promoter. Two-way ANOVA with Tukey’s correction. N = 3 biological replicates. *** p < 0.001. (C) As in (A), highlighting the N-terminal sequences. (D) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb. A mutated FLWR-1 in which arginine 27 and lysine 31 were altered to alanines, was expressed in flwr-1(ok3128) mutant background under its endogenous promoter. Two-way ANOVA with Tukey’s correction. N = 3 biological replicates. (E) Model summarizing findings made in this study. For details, see discussion.

Loss of FLWR-1 does not change body length but reduces number of living progeny.

(A) Mean (± SEM) body length of wild type and flwr-1(ok3128) mutants in mm. Unpaired t-test. ns not significant. Number of animals accumulated from N = 4 biological replicates: wild type = 29, flwr-1 = 27. (B) Mean (± SEM) number of living progeny per animal. Unpaired t-test. ** p < 0.01. Number of animals accumulated from N = 4 biological replicates: wild type = 20, flwr-1 = 18.

FLWR-1 is predicted to be a tetraspan transmembrane protein and is transported by UNC-104 kinesin.

(A) Graphical representation of the results of DeepTMHMM prediction of membrane orientation of the FLWR-1 protein based on its amino acid sequence. The red shapes indicate the presence of four transmembrane domains. (B) Alphafold3 prediction of FLWR-1 protein structure. The coloring represents the calculated predicted local distance difference test (plDDT) as shown below the structure. Higher plDDT values indicate a higher confidence of correct prediction. (C) Heatmap plot depicting flwr-1(F20D1.1) single-cell RNAseq data generated by the CeNGEN database. Coloring indicates transcripts per million (TPM) per tissue normalized to the average expression as indicated in the legend. The size of the dots represents the percentage of cells of this cell type expressing the gene. (D) Example images depicting GFP::FLWR-1 fluorescence in nerve ring and nerve cords in wild type and unc-104(e1265) mutants. Scale bar, 20µm.

FLWR-1 expression in body wall muscle cells partially rescues aldicarb resistance of flwr-1 mutants; FLWR-1 localizes to cholinergic and GABAergic active zones.

(A) Mean (± SEM) fraction of moving animals after exposure to 0.25 mM levamisole. N = 3 biological replicates. (B) Mean (± SEM) fraction of moving animals after exposure to 1.5 mM aldicarb with expression of FLWR-1 in BWMs (pmyo-3) in flwr-1(ok3128) mutant background. N = 5 biological replicates. (A + B) Two-way ANOVAs with Tukey’s correction. There were no significant differences in (A); *** p < 0.001. (C) Line scan analysis of colocalization of GFP::FLWR-1 and TagRFP::ELKS-1 along the DNC as represented in the uppermost micrographs in Fig. 3A. R2 as determined by Pearson correlation. a.u. = arbitrary units of fluorescence intensity. (D) Line scan analysis of colocalization of GFP::FLWR-1 and TagRFP::ELKS-1 along the DNC as represented in the uppermost micrographs in Fig. 3B. R2 as determined by Pearson correlation. a.u. = arbitrary units of fluorescence intensity. (E) Comparison of Pearson correlation coefficients of line scans along DNCs represented in Fig. 3. Mean (± SEM). Unpaired t-test.

flwr-1 mutants show defecting cholinergic neurotransmission only during continuous stimulation.

(A, B) Mean (± SEM) mPSC frequency and amplitude, respectively, before stimulation. Unpaired t-test. Ns not significant. Number of animals: wild type = 16, flwr-1 = 14. (C) Mean (± SEM) inward currents of BWMs recordings induced by 10 ms light pulses (470 nm, 8 mW/mm2) applied every 2 s (0.5 Hz). Two-way ANOVA with Šidák’s correction for multiple comparisons. ns not significant. Number of animals: wild type = 9, flwr-1 = 8. (D) As in (C), but 2Hz stimulation. ns not significant. Number of animals: wild type = 7, flwr-1 = 7. (E) Representative voltage-clamp recording of currents detected in BWMs. This wild type animal expresses ChR2(H134R) in cholinergic motor neurons (unc-17promoter) and has been treated with ATR. A 30 s light stimulus (470 nm, 8 mW/mm2) as well as a 10 ms pulse after a 15 s inter-stimulus interval (ISI) were applied as indicated by blue bars. (F) Mean (± SEM) mPSC frequency in BWMs of animals expressing ChR2(H134R) in cholinergic motor neurons (unc-17 promoter). All animals have been treated with ATR. A 30 s light pulse (470 nm, 8 mW/mm2) was applied as indicated by blue shade. (G) Mean (± SEM) mPSC frequency during 30 s stimulation. Unpaired t-test. * p < 0.05. (H) Analysis of the amplitude of the first peak during 30 s photostimulation and the second peak after 15 s ISI as indicated in (E). Two-way ANOVA with Tukey’s correction for multiple comparisons. ns not significant, *** p < 0.001. (F - H) Number of animals: wild type = 9, flwr-1 = 8.

flwr-1 mutants have a normal number of large vesicles and dense core vesicles before and after stimulation.

(A) Violin plot depicting the number of large vesicles counted per synaptic profile. (B) Violin plot depicting the number of dense core vesicles counted per synaptic profile. (A + B) Bold line represents the median and the dashed lines the IQR. Kruskal-Wallis test. Only significant statistical differences are depicted. ns not significant. Number of synaptic profiles imaged: wild type (-ATR) = 56, wild type (-ATR) = 51, flwr-1 (-ATR) = 55, flwr-1 (+ATR) = 59.

Comparison of flwr-1 and mca-3 expression by single-cell RNAseq data.

Heatmap plot depicting flwr-1(F20D1.1) and mca-3 single-cell RNAseq data generated by the CeNGEN database (https://cengen.shinyapps.io/CengenApp/). Coloring indicates transcripts per million (TPM) per tissue normalized to the average expression as indicated in the legend. The size of the dots represents the percentage of cells of this cell type expressing the gene. Coelomocyte data is highlighted by the blue box.

Structural analysis of important amino acid residues in FLWR-1 and other proteins.

(A) Top view of an Alphafold3 prediction of a tetrameric FLWR-1 structure. Glutamate 74 is indicated as colored spheres in each monomer. (B) Cryo-EM structure of the human L-type voltage-gated calcium channel Cav1.2 (PDB: 8EOG) (Chen et al., 2023). Glutamate residues in the central pore are indicated as spheres. Ca2+ ions within the pore are depicted as turquoise spheres. (C) X-Ray structure of the Kir2.2 potassium channel in a complex with PI(4,5)P2 (PDB: 3SPI) (Hansen et al., 2011). Basic amino acid residues in the proximity of PI(4,5)P2 are depicted as sticks. (D) PI(4,5)P2 was docked to the Alphafold3 structure of FLWR-1 using AutoDock Vina (Eberhardt et al., 2021). Basic amino acid residues in the proximity of PI(4,5)P2 are depicted as sticks. Arginine 27 and lysine 31 are indicated.