Regulation of mechanosensation in C. elegans through ubiquitination of the MEC-4 mechanotransduction channel

Abstract

In Caenorhabditis elegans, gentle touch is sensed by the anterior (ALM and AVM) and posterior (PLM) touch receptor neurons. Anterior, but not posterior, touch is affected by several stress conditions via the action of AKT kinases and the DAF-16/FOXO transcription factor. Here we show that a ubiquitination-dependent mechanism mediates such effects. AKT-1/AKT kinase and DAF-16 alter the transcription of mfb-1, which encodes an E3 ubiquitin ligase needed for the ubiquitination of the mechanosensory channel subunit MEC-4. Ubiquitination of MEC-4 reduces the amount of MEC-4 protein in the processes of ALM neurons and, consequently, the mechanoreceptor current. Even under nonstress conditions, differences in the amount of MFB-1 appear to cause the PLM neurons to be less sensitive to touch than the ALM neurons. These studies demonstrate that modulation of surface mechanoreceptors can regulate the sensitivity to mechanical signals.

Keywords: C. elegans; DEG/ENaC; mechanosensation; neuromodulation; ubiquitination.

Figure 1.

AKT-1 modulates touch sensitivity through MFB-1 and ubiquitination. A, The anterior touch sensitivity of animals with the indicated genotypes or akt-1 animals treated with 1.4 mm PYR-41, 13 μm bortezomib (Bor), 11 μm MG-132, or 0.4 μm concanamycin A (ConcA). Here and in subsequent panels and figures, individual data points, mean, and the 95% confidence interval (CI) are shown. *p < 0.005, comparing akt-1 with wild-type or comparing the indicated mutants with akt-1. B, The anterior touch sensitivity of animals with the indicated genotypes with or without mec-18p::mfb-1 or mec-18p::cav-1::gfp. *p < 0.01. **p < 0.005. C, The anterior touch sensitivity of TU3595 animals treated with RNAi against the indicated genes. *p < 0.01, compared with RNAi against gfp. For Figures 1–5, each data point indicates the value of one biological replicate. Each biological replicate contains at least eight animals. For Figures 6–8, each data point indicates one animal.

Figure 2.

AKT-1 and DAF-16 regulate mfb-1 transcription. A, B, mfb-1p::gfp expression in a wild-type animal (A) and an akt-1 mutant (B) fed with non-RNAi bacteria (top), and in ALM (arrow) and AVM neurons in a animals fed with RNAi against gfp, which reduced gfp expression in non-neuronal tissues (bottom). All pictures of the same magnification were taken at the same exposure. C, Representative pictures of smFISH against mfb-1 (red) and mec-18 (green) in ALM cells in akt-1 and daf-16 mutants. The positions of the mec-18 (green) dots delineate the shape of the ALM cell bodies (dashed circles). An mfb-1 transcript in the cell is indicated with an arrowhead. D–F, Quantification of mfb-1 transcripts in ALM cells (D), AVM cells (E), and PLM cells (F) in akt-1, daf-16, and wild-type animals as measured by single molecule mRNA FISH. Data are mean ± SEM labeled for each genotype. D, n ≥ 70, p < 0.05, comparing akt-1 with daf-16. E, n ≥ 38, p < 0.01, comparing akt-1 with daf-16. F, n ≥ 70, p < 0.0005, comparing akt-1 with daf-16 and p < 0.01, comparing wild-type with daf-16. G, Quantification of akt-1 transcripts in ALM (black) and PLM (white) cells. Data are mean ± SEM labeled for each population of cells. p < 0.0005 comparing ALM and PLM cells.

Figure 3.

MFB-1 ubiquitinates MEC-4 to regulate its expression. A, Western blot against RFP for lysate (right) or after immunoprecipitation against ubiquitin (left) in wild-type animals with (−) or without (+) MEC-4::RFP, and akt-1 and akt-1;mfb-1 animals with MEC-4::RFP. The lysate samples represent 1/20th of the protein of the ubiquitinated samples. No ubiquitinated MEC-4::RFP was immunoprecipitated with a GFP antibody as a negative control (data not shown). B, Quantification of band intensities from three independent biological replicates. All intensities were normalized to wild-type controls. *akt-1 is significantly different (p < 0.05) from wild-type using one-sample t test and from akt-1; mfb-1 (p < 0.01) using Student's t test. N = 3.

Figure 4.

MEC-4 expression regulates touch sensitivity. Representative pictures (A) and quantifications (B) of permeabilized antibody staining of MEC-4 (normalized to wild-type) in the processes of ALM neurons of animals with the indicated phenotype, or treated with RNAi [indicated by (i)] against unc-112, pat-2, or ins-10, or treated with high salt (NaCl) or sustained vibration (vib). *p < 0.05, compared with wild-type. **p < 0.005, compared with wild-type. ***p < 0.005, comparing akt-1; mfb-1 with akt-1. N ≥ 3. Representative pictures (C) and quantifications (D) of permeabilized antibody staining of MEC-4 (normalized to wild-type) in the processes of ALM neurons of akt-1 animals with or without bortezomib (Bor) or concanamycin A (ConcA). *p < 0.005, compared with no drug condition. N = 3 for all conditions. Representative pictures (E) and quantifications (F) of permeabilized antibody staining of MEC-4 (normalized to wild-type ALM neurons) in the processes of PLM neurons of animals with the indicated phenotype, or treated with RNAi against ins-10, or treated with high salt (NaCl). *p < 0.05, comparing wild-type PLM to wild-type ALM (B). **p < 0.005, comparing akt-1; mfb-1 to akt-1or to wild-type. N ≥ 3. G, The anterior touch sensitivity of animals with the indicated genotypes with (circles) or without (dots) MEC-4 overexpression in the TRNs. *p < 0.05, comparing each strain with or without MEC-4 overexpression. **p < 0.005, comparing each strain with or without MEC-4 overexpression.

Figure 5.

Environmental conditions alter MEC-4 expression. A, Quantification of antibody staining of MEC-4 (normalized to wild-type) in the processes of ALM neurons of animals with the indicated phenotype with or without sustained vibration (vib), and wild-type animals recovered from sustained vibration (recovery). *p < 0.005, compared with wild-type; and p < 0.05, compared with recovery. N ≥ 3. B, The anterior touch sensitivity of wild-type and mfb-1 animals grown under normal condition, hypoxia, or high salt. p < 0.005, comparing mfb-1 with wild-type under hypoxia and high salt conditions. N ≥ 4.

Figure 6.

MRCs in ALM neurons. A–C, Representative MRCs to saturated response in the ALM neurons in wild-type (A), akt-1 (B), and akt-1; mfb-1(C) animals. Gray lines indicate overlays of 20 repeats of stimuli to a single animal. Black lines indicate the average. Red lines below indicate the voltage driving the stimulus probe. D, MRCs of ALM neurons in wild-type, akt-1, and akt-1; mfb-1 animals. *p < 0.0005, compared with wild-type; and p < 0.005, compared with akt-1; mfb-1.

Figure 7.

MFB-1 alters surface MEC-4 in cultured TRNs. A, Representative pictures of permeabilized (Total) and nonpermeabilized (Surface) antibody staining of MEC-4 (green) and MEC-18 (red) in the processes of cultured ALM neurons. B, Quantifications of permeabilized (Total) and nonpermeabilized (Surface) antibody staining of MEC-4 in the processes of cultured ALM neurons. *p < 0.05, comparing surface MEC-4 (Welch's test). **p < 0.001, comparing total MEC-4 (Student's t test).

Figure 8.

MRCs in PLM neurons. A–C, Representative MRCs to saturated response in the PLM neurons in wild-type (A), akt-1 (B), and akt-1; mfb-1(C) animals. Gray lines indicate overlays of 20 repeats of stimuli to a single animal. Black lines indicate the average. Red lines below indicate the voltage driving the stimulus probe. D, MRCs of PLM neurons in wild-type, akt-1, and akt-1; mfb-1 animals. *p < 0.005, compared with wild-type; and p < 0.0005, compared with akt-1; mfb-1. **p < 0.0001, compared with wild-type ALM MRCs in Figure 6D. E, PLM MRCs elicited by stimulation at different positions in one animal. The vulva is labeled with a white triangle. The approximate positions of the stimulation sites are labeled with black triangles, and the distances to the PLM cell body are labeled beneath with the corresponding MRCs. For the MRCs, gray lines indicate overlays of 20 repeats of stimuli at a specific site, and the black lines indicate the average.

Figure 9.

Regulation of mechanotransduction in the TRNs. MFB-1-dependent ubiquitination and removal of MEC-4 mediates the change in touch sensitivity by both integrin (green pathway) and insulin signaling (red pathway). Integrin signaling may increase MEC-4 activity through additional pathways (green dotted pathway and “?”). The same pathway also partially contributes to less sensitivity in the posterior TRNs (blue pathway). Dashed lines indicate that the signaling strength is reduced in the specific conditions/cells. Dotted lines indicate the hypothesized pathway. In the posterior TRNs, the size of the labels of proteins indicates the relative abundance (or activity) of that protein compared with the anterior TRNs under normal conditions.