Loss of innexins or increased GJ levels lead to opposing effects on locomotion.

(A) Basal crawling speed (±SEM) of unc-9(e101), inx-11(ok2783) and inx-16(tm1589) mutants compared to WT animals. Number of animals per experiment n, across N=4 independent experiments. (B) Comparison of mean crawling speed of the indicated strains. Each dot represents the mean crawling speed of each measurement. Mean and SEM. One-way ANOVA with Dunnett test (* p < 0.05, *** p < 0.001, **** p < 0.0001). (C) Mean swimming cycles (±SEM) of WT, unc-9, inx-11 and inx-16 mutants. Mean and SEM. Number of animals per experiment n=24-52 (WT), n=34-86 (unc-9), n=28-51 (inx-11) and n=18-72 (inx-16), across N=4 independent experiments. One-way ANOVA with Tukey test (**** p < 0.0001). (D) Basal crawling speed (±SEM) of Cx36::mCherry overexpressing animals compared to WT. Number of animals per experiment n, across N=4 independent experiments. (E) Comparison of mean crawling speed of the indicated strains. Each dot represents the mean crawling speed of each measurement. Mean and SEM. Unpaired t test (* p < 0.05). (F) Mean swimming cycles (±SEM) of WT and Cx36::mCherry overexpressing animals. Number of animals per experiment n=32-68 (WT) and n=26-74 (Cx36), across N=4 independent experiments. Unpaired t-test (*** p < 0.001).

Spontaneous muscle cell activity is visualized by QuasAr2 imaging, revealing differences in unc-9 mutants and Cx36::mCherry overexpressing muscles.

(A) Fluorescence of QuasAr2 expressed in body wall muscles of C. elegans. Measured were ensembles consisting of three muscle cells, yellow outline shows the analyzed ROIs. Numbering of cells is indicated. (B) Representative ΔF/F fluorescence time traces of spontaneous muscle cell activity of WT animals, unc-9 mutants (C) and Cx36::mCherry overexpressing animals (D). Correlation of muscular activity WT, unc-9 mutants (E) and Cx36::mCherry overexpressing animals (I) was calculated using Pearson correlation. Each dot represents one calculated correlation coefficient. Median with interquartile range. Number of animals n=10 (WT, unc-9); n=14 (WT, Cx36). Unpaired t test (**** p < 0.0001), Welch’s t test in I. Single peaks of voltage fluorescence traces were analyzed for amplitude (F, J), peak area (G, K) and full width at half maximum (FWHM) (H, L). Each dot represents the mean result of all peaks of one fluorescence time trace of one animal. Median with interquartile range. Number of animals is the same as in E and I, respectively. Mann-Whitney test in F, K, L; unpaired t test in G, H, J (ns p>0.05).

Cross correlation analysis of single voltage fluorescence peaks reveals changes in maximum correlation.

(A) Cross correlation of single peaks (± 0.25 s from peak) of pairs of the three-cell ensemble of unc-9 mutants and WT. Mean and SEM. Number of animals n=9-10 (WT), n=3-9 (unc-9). Maximum correlation coefficient (B) and lag time at maximum correlation (C) of the cross correlations of ten peaks from each animal. Each dot represents the mean of the first 10 cross correlations of one pair of muscle cells. Median with interquartile range. Unpaired t test (** p<0.01) in B, Welch’s t test (ns p<0.05) in C. (D) Min-max normalization of the cross correlation of single peaks in A. (H) Cross-correlation analysis of single peaks (±0.25 from peaks) of pairs of the three-cell ensemble of Cx36::mCherry overexpressing animals. Number of animals n=14 (WT, Cx36). (I) Maximum correlation coefficient and lag time at maximum correlation (J) of the cross correlation of the ten peaks from each animal. Each dot represents the mean of the first 10 cross correlations of one pair of muscle cells. Median with interquartile range. Mann-Whitney test (ns p>0.05, * p<0.05). (K) Min-max normalization of the mean cross correlations shown in H. (L) Calculated tauoff value of the mean cross correlation of a pair of muscle cells using a one-phase exponential decay fit. Median with interquartile range. Unpaired t test (ns p>0.05).

Voltage imaging of spontaneous muscle activity reveals changes in peak characteristics in inx-16 mutants.

(A) Representative traces of spontaneous muscle cell activity of one WT animal. Representative ΔF/F fluorescence time traces of inx-11 mutants (B) and inx-16 mutants (C). (D) Correlation of muscular activity was calculated using Pearson correlation. Each dot represents one calculated correlation coefficient. Median with interquartile range. Number of animals n=14 (WT), 10 (inx-11), 12 (inx-16). Kruskal-Wallis with Dunn’s test (ns p>0.05). Single peaks of voltage fluorescence traces were analyzed for amplitude (E), peak area (F) and FWHM (G). Each dot represents the mean result of all peaks of one fluorescence time trace of one animal. Median with interquartile range. Number of animals as in D. One-way ANOVA with Tukey test in E, Kruskal-Wallis with Dunn’s test in F, G (ns p>0.05, * p<0.05, ** p<0.01, **** p<0.0001).

Cross correlation analysis of single peaks reveals higher maximum correlation of inx-16 mutants.

Cross correlation of single peaks (± 0.25 s from peak) of pairs of the three-cell ensemble of inx-11 (A) and inx-16 mutants (B). Mean and SEM. Number of animals n=14 (WT), 9 (inx-11), 12 (inx-16). Maximum correlation coefficient (C) and lag time at maximum correlation (D) of the cross correlations of ten peaks from each animals. Each dot represents the mean of the first 10 cross correlations of one pair of muscle cells. Median with interquartile range. One-way ANOVA with Tukey test (ns p>0.05, * p<0.05) in C, Kruskal-Wallis with Dunn’s test (ns p<0.05) in D. (E) Min-max normalization of the cross correlation of single peaks in A for inx-11, and in B for inx-16 mutants. (G) Calculated tauoff value of the mean cross correlation of a pair of muscle cells using a one-phase exponential decay fit. Median with interquartile range. One-way ANOVA with Tukey test (ns p>0.05, ** p<0.01).

Patch clamp recordings reveal increased excitability of inx-16 mutant muscles.

(A) Resting membrane potential recorded in current clamp of WT, inx-11, inx-16, unc-9 mutants and Cx36::mCherry overexpressing muscles. Median with interquartile range. Number of animals n=8 (WT), 7 (inx-11, inx-16), 9 (unc-9), 8 (Cx36). One-way ANOVA with Tukey test. (B) Capacitance measurements in current clamp mode. Median and interquartile range. One-way ANOVA with Tukey test (* p<0.05, ** p<0.01). Number of animals n=8 (WT, inx-16, Cx36), 9 (inx-11), 11 (unc-9). (C) Input resistance was measured by injection of three consecutive −20 pA pulses (1000ms, 1000ms interval). For normalization, the input resistance was divided by the capacitance of the respective cell. Number of animals n=8 (WT, unc-9, Cx36), 7 (inx-11, inx-16). Median with interquartile range. One-way ANOVA with Tukey test (* p<0.05, ** p<0.01, *** p<0.001). (D) Exemplary traces of induced APs. APs were induced by a 50 ms, +20 pA current pulse. (E) The induced APs were extracted and peak aligned (mean ± SEM). Number of extracted peaks n=40 (WT, inx-11, Cx36), 54 (unc-9), 46 (inx-16). APs were analyzed for amplitude (F), area under the curve (G), FWHM (H) and time to peak (I). Median with interquartile range. Kruskal-Wallis with Dunn’s test (* p<0.05), ** p<0.01, *** p<0.001, **** p<0.0001). (J) Changes of membrane potential in response to current injections of −10, −20 and −30 pA pulses (1000 ms). Number of animals n=8 (WT, unc-9, Cx36), 7 (inx-11, inx-16). Mean and SEM. Two-way ANOVA with Tukey test (* p<0.05).

Cell-specific OVC measurements reveal reduced junctional conductance in unc-9 mutants and increased excitability of inx-11 mutants.

(A) Schematic illustration of a cOVC measurement. Cell 1 is optically clamped to specific membrane potentials (i.e. ΔF/F0 fluorescence levels), while voltage fluorescence changes in cell 2 are observed only. Mean ΔF/F fluorescence (left y-axis) traces of clamped cell 1 and observed cell 2 and mean wavelength changes (right y-axis; represented as the ratio of light intensities of the respective wavelength) in cell 1 of WT animals (B), unc-9 (C), Cx36::mCherry overexpressing muscles (D) and inx-11 mutants (E). Number of animals n=13 (WT), 20 (unc-9), 15 (Cx36), 14 (inx-11). (F) Mean (±SEM) ΔF/F0 fluorescence of clamped cell 1 and unclamped cell 2 (G). (H) Changes of QuasAr2 fluorescence in cell 2 during hyperpolarization and (I) depolarization step. Median with interquartile range. Welch ANOVA with Dunnett’s test in H; one-way ANOVA with Sidak test in I (ns p>0.05, * p<0.05). (J) Calculated rate constant K of the −4% to +4% ΔF/F0 step was calculated using an exponential plateau function. Median with interquartile range. Welch ANOVA with Dunnett’s test (* p<0.05, *** p<0.001, **** p<0.0001). (K) Wavelength changes (ratio of intensities of light of 460 nm over 570 nm, applied to cell 1) during the hyperpolarization and (L) depolarization steps. Median with interquartile range. Welch ANOVA with Dunnett’s test in K; one-way ANOVA with Sidak test in L (ns p>0-05, ** p<0.01). (M) Transition times for 8% and 4% ΔF/F steps (N). Median with interquartile range. Kruskal-Wallis with Dunn’s test (ns p>0.05, ** p<0.01).

C. elegans strains used in this study.

(A) Fluorescence micrograph showing expression pattern of Cx36::mCherry in BWMs. Scale bar: 50 µm. (B) Representative ΔF/F fluorescence time traces of strain ZX2476, the WT control for unc-9 mutants. (C) Correlation coefficients of all three different WT strains expressing different QuasAr2 arrays were calculated using Pearson correlation. Median with interquartile range. Kruskal-Wallis with Dunn’s test (ns p>0.05). (D) Correlation coefficients for the three pairs of Cx36 overexpressing muscle cells analyzed in an ensemble were calculated using Pearson correlation. Median with interquartile range. Kruskal-Wallis with Dunn’s test (** p<0.01).

(A) Mean (±SEM) ΔF/F traces of extracted peaks (±0.25 s) used for cross correlation analysis of single peaks of unc-9 mutants, (F) Cx36::mCherry, (K) inx-11 and (L) inx-16 animals, as well as their respective WT controls. (B) Mean (±SEM) ΔF/F traces of extracted peaks (±0.65 s) used for cross correlation analysis of longer time windows of unc-9 mutants, (G) Cx36::mCherry, (M) inx-11 and (N) inx-16, as well as their respective WT controls. Number of animals n=7-10 (WT), 3-10 (unc-9) in B, n=12 (WT), 14 (Cx36) in G, n=14 (WT), 9 (inx-11), 12 (inx-16) in K and L. (C) Cross-correlation analysis of longer time windows (±0.65 from peaks) of pairs of the three-cell ensemble of unc-9 mutants, (H) Cx36::mCherry animals, (O) inx-11 and (P) inx-16 mutants. (D) Maximum correlation coefficient of unc-9, (I) Cx36 animals, and (Q) inx-11 and inx-16 mutants of the cross correlations of five extracted time windows from each animal. (E) Lag time at maximum correlation of unc-9, (J) Cx36 animals, and (R) inx-11 and inx-16 mutants of the cross correlations of five extracted time windows from each animal. Each dot represents the mean of the first 5 cross correlations of one pair of muscle cells. Median with interquartile range. Unpaired t test in D, Welch t test in E, Mann-Whitney test in I and J, Kruskal-Wallis with Dunn’s test in Q and R (ns p>0.05, ** p<0.01, *** p<0.001).

(A) Hardware setup for cOVC measurements: QuasAr2 is excited with a 637 nm laser [1], equipped with a Galilean beam expander [2] at 1.8 W/mm2 and imaged at 700 nm, using a 700/75 ET bandpass filter [10]. BiPOLES components were activated using light from a video projector with digital light processing unit [4]. Projector light was narrowed by additional filters [6]. In front of the accessory tube lens [8] a 1% transmission ND filter [7] was added. Between the beam combination tube equipped with a laser beam splitter [5], and the Galilean beam expander [2], an adapter with lenses of the epifluorescence path of the microscope [3] was added. Imaging was performed on a Zeiss Axiovert 200 inverted microscope [9] equipped with a 40x oil immersion objective [11] and a Kinetix22 sCMOS camera [12]. (B) Spectral characteristics of beamer output were shaped with excitation filters and ND filters. (C) Relative intensity as a function of color pixel value (0 (full off) – 255 (full on)) for blue and green RGB colors. (D) Measured width of different projected spot sizes using a 40x objective. (E) Measured spot sizes using a 40x objective. Width was measured from traces shown in (D) at the point where intensity drops to 25% of maximum value.. (F) cOVC measurement, where the ROI for ‘clamped cell 1’ was moved outside of the worm, adjacent to a BWM cell (ROI 2). Illumination protocol of ROI1, going from equal blue and green illumination to 100% blue and then to 100% green, is indicated (black curve, right y-axis; N=13 experiments). (G) Mean±SEM ΔF/F0 traces of strain only expressing QuasAr2 in BWMs (red and blue traces for cells 1 and 2), i.e., without BiPOLES, while the cOVC stimulation program attempts to run a 0, −4, +4% ΔF/F protocol (black traces, mean±SEM). N=15 experiments were averaged. (H) Independent clamping of individual BWMs located at opposite muscle strands in the head; mean±SEM ΔF/F0 traces of QuasAr2 fluorescence of the two muscles, indicated in blue and red color. The light stimulation protocol is indicated in black and grey for the opposite cells, as indicated. Average of N=5 experiments. (I) Long term −3% ΔF/F0 clamping of one individual cell. Shown are the achieved fluorescence values (black trace, bottom) and the ratio of the required wavelengths illuminated on the cell (upper half, referring to right y-axis).