(A) The average speed of mutants identified in a genetic screen for suppressors of slo-1(gf). dyb-1, a dystrobrevin homolog; dys-1, a dystrophin homolog; stn-1, a syntrophin homolog. Error bars represent s. e. m. (n>10). Asterisks represent significant difference (P<0.05). (B) The number of eggs retained in uteri of slo-1(gf) suppressor mutants. Error bars represent s. e. m. Data points between slo-1(gf) and all of other strains are significantly different (P<0.001). (C) Quantitative analysis for the first angle of head bending. Each data set for the first angle is significantly different from that of wild-type animals (P<0.01). Yellow dots indicate the five most anterior of the 13 midline points for a wild-type animal (See also Materials and Methods). Error bars represent s. e. m. (n = 10).

(A) The gene structure of ctn-1. Our genome analysis indicates that unlike WormBase annotation (WS210) ctn-1 consists of 13 exons and is predicted to encode a 784 amino-acid protein (Figure S1). The homology regions with α-catenin/vinculin I (homologous to the N-terminal talin/α-actinin-binding region of vinculin, 213 amino acids), α-catenin/vinculin II (homologous to the C-terminal F-actin/inositol phospholipids-binding region of vinculin, 145 amino acids) and the coiled-coil domain (35 amino acids) are depicted at the bottom of the gene structure. The mutation sites for two different alleles (eg1167 and cim6) are shown on the top of the gene structure. The predicted amino acid sequence is available in Figure S1. (B) Rescue of the head bending phenotype with a variety of ctn-1 constructs. Ex[ctn-1] represents the transgene carrying the genomic ctn-1 DNA extrachromosomal array. Ex[P_myo-3ctn-1] represents the muscle-specific myo-3 promoter-driven ctn-1 transgene, whereas Ex[P_H20ctn-1] represents the neuron-specific H20 promoter-driven ctn-1 transgene. Error bars represent s. e. m. (n = 10). Single asterisks indicate significant difference between two groups (P<0.001, unpaired two-tailed t-test) whereas ns indicates no significant difference. (C) Rescue of the defect in egg-laying muscle with a variety of ctn-1 constructs. Error bars represent s. e. m. (n = 15). Single asterisks indicate significant difference between two groups (P<0.001, unpaired two-tailed t-test) whereas ns indicates no significant difference. (D) Rescue of the average locomotory speed with a variety of ctn-1 constructs. Error bars represent s. e. m. (n>10). Single asterisks indicate significant difference between two groups (P<0.001, unpaired two-tailed t-test) whereas ns indicates no significant difference. (E–H) The expression pattern of a ctn-1 promoter-tagged GFP reporter. Expression in (E) body wall muscles and the ventral cord neurons (arrowheads), (F) nerve ring (arrowheads) and pharyngeal muscle (arrow), (G) egg-laying muscle, and (H) enteric (arrow) and sphincter (arrowhead) muscle. Scale bar, 10 µm.

(A) An integrated transgenic line expressing the lowest level of GFP-tagged CTN-1 was used for staining with anti-GFP (CTN-1, green) and anti-vinculin/DEB-1 (DEB-1, red) antibodies. Dashed box area is enlarged in the bottom left of the panel (Merged) to show detail. Arrowheads indicate the attachment plaques that adhere tightly adjacent muscle cells. Scale bar, 10 µm. (B) Transgenic animals expressing integrated GFP-tagged SGCA-1 were used to analyze the localization of SGCA-1 in wild-type (WT), dys-1 ctn-1 and slo-1 animals. Scale bar, 10 µm. The graph shows quantified puncta intensities. Error bars, s.e.m. Wild-type vs. dys-1 (P<0.01), Wild-type vs. ctn-1 (P<0.01), Wild-type vs. slo-1 (P>0.05). (C) Transgenic animals expressing integrated mCherry-tagged ISLO-1 were used for analyzing the localization of ISLO-1 in wild-type (WT) and ctn-1 animals. Scale bar, 10 µm. The graph shows quantified puncta intensities. Error bars, s.e.m. Wild-type vs. ctn-1 (P<0.0001).

The same integrated array, SLO-1::GFP, was used for this analysis in different genetic backgrounds. (A–B) Muscular localization of SLO-1::GFP in wild-type, dys-1 and ctn-1 animals. Arrowheads represent the ventral (or dorsal) nerve cords. (B) The graph showing quantification of puncta intensities. Error bars, s.e.m. Wild-type vs. dys-1 or ctn-1 (P<0.0001). Scale bar, 10 µm. (C–D) Neuronal localization of SLO-1::GFP in wild-type, dys-1 and ctn-1 animals (See also Figure S2). Regions of the ventral nerve cord (dashed boxes) are enlarged in the bottom left of each panel to show detail. (D) The graph showing quantification of puncta intensities. Error bars, s.e.m. Wild-type vs. dys-1 (P>0.05), Wild-type vs. ctn-1 (P<0.0001). Scale bar, 10 µm.

(A) Representative evoked current responses from wild-type, slo-1(gf), ctn-1;slo-1(gf), ctn-1 and slo-1(lf) animals. (B) Normalized evoked current responses from (A). (C) Evoked amplitude response. Wild-type (n = 31), slo-1(gf) (n = 15), ctn-1;slo-1(gf) (n = 12), ctn-1 (n = 16), slo-1(lf) (n = 21). There is no significant difference between wild-type and each genotype used (P>0.05). (D) Half-time decay. Wild-type vs. slo-1(gf), P<0.05; wild-type vs. ctn-1;slo-1(gf), P>0.05; wild-type vs. ctn-1, P<0.05; wild-type vs. slo-1, P<0.01. (E) Evoked charge integral. Wild-type vs. slo-1(gf), P<0.01.

In muscles, CTN-1 interacts with the dystrophin complex. ISLO-1 links the dystrophin complex to SLO-1. The N-terminal region or the coiled-coiled domain of CTN-1 is likely to interact with the dystrophin complex. The C-terminal region may interact with other cytoskeletal proteins. In neurons, CTN-1 interacts with SLO-1 through possible unknown intermediates other than the dystrophin complex.