• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Riddle DL, Blumenthal T, Meyer BJ, et al., editors. C. elegans II. 2nd edition. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 1997.

Cover of C. elegans II

C. elegans II. 2nd edition.

Show details

Section IVGenes Required for Touch Receptor Function

A. mec-7 and mec-12 Tubulins

The touch receptor cell processes (Fig. 3B) are filled with a distinctive bundle of wide-diameter microtubules that contain 15 protofilaments (Chalfie and Thomson 1979, 1982). (Microtubules present in most C. elegans cells contain 11 protofilaments; those in most organisms contain 13 protofilaments.) Individual microtubules 10−20-μm long do not span the full length of the touch cell processes, which is about 400−500 μm. Rather, overlapping microtubules are positioned along the touch cell process. The microtubule ends appear to be structurally distinct—The end proximal to the cell body appears darkened and is preferentially found on the inside of a microtubule bundle, whereas the distal end is diffusely stained and is always situated outside of the microtubule bundle. Interestingly, the distal end is often juxtaposed to the plasma membrane and thus could potentially form a mechanical link between the microtubule network and mechanosensory receptors in the plasma membrane.

The 15-protofilament microtubules are required for mechanosensory function. In mec-7 null mutants and some mec-12 loss-of-function mutants, the 15-protofilament microtubules are absent from the touch receptor neurons, and, consequently, the animals are touch-insensitive (Chalfie and Sulston 1981). mec-7 encodes a 440-amino-acid β-tubulin and mec-12 encodes a 450-amino-acid α-tubulin (Savage et al. 1989; Hamelin et al. 1992). Both genes are highly expressed in the touch neurons (Mitani et al. 1993; Savage et al. 1994; M. Hamelin et al., pers. comm.). mec-12 is also expressed at high levels in several other cells, including ventral cord DB, VA, and VD neurons and the CAN neuron. Unlike the touch neurons, these cells do not have 15-protofilament microtubules; therefore, mec-12 expression alone is not sufficient for formation of wide-diameter microtubules. Apart from the carboxy-terminal domain, which is highly variable among β-tubulins, only seven amino acid residues distinguish MEC-7 from other isoforms (Savage et al. 1989). These unique amino acids may have an instructive role in determination of protofilament number. Taken together, studies of mec-7 and mec-12 suggest that the 15-protofilament microtubules are composed of MEC-7 and MEC-12 β and α tubulins and are essential for mechanosensory transduction.

What role do the 15-protofilament microtubules have in touch cell function? These microtubules are not essential for axon outgrowth, since touch receptor processes are formed normally in mec-7 mutants (Chalfie and Sulston 1981). Touch cell processes in mec-7 (null) animals are not devoid of microtubules altogether—11-protofilament microtubules, normally all but absent in the touch cells, partially replenish the microtubule content of the axons (Chalfie and Thomson 1982). This suggests that the 11-protofilament microtubules may partially compensate for the lack of MEC-7 tubulin, thereby allowing normal axon outgrowth. The 15-protofilament microtubules do, however, have some capacity to influence process outgrowth. Touch cell processes are extended or produced at ectopic sites as a consequence of two particular mec-7 missense alleles (Savage et al. 1994). Whatever their contribution to touch cell development, microtubule integrity appears to be essential for mechanosensory function of the touch receptor neurons. If touch cell microtubules are disrupted in low concentrations of colchicine, touch sensitivity is lost (Chalfie and Thomson 1982).

B. Genes Affecting the Extracellular Mantle

A second distinguishing feature of the touch receptor neurons is that their processes are surrounded by a specialized extracellular matrix referred to as the mantle (see Fig. 3B) (Chalfie and Sulston 1981). Darkly staining cuticular specializations are positioned periodically along the length of the touch receptor process, in close contact with the mantle. The cuticular specializations look similar to muscle attachment sites and thus they may be sites at which the touch receptor process is fixed to the cuticle.

It appears that the touch cell processes must be closely apposed to the hypodermis for mantle production. In mec-1 mutants, touch cells lack the mantle and associated periodic specializations of the overlying cuticle; the ALM processes run along body wall musculature, rather than within the hypodermis (Chalfie and Sulston 1981). However, where portions of the touch processes are embedded within the hypodermis in mec-1 mutants, the mantle is observed. Two equally plausible models could account for these results: The mantle may be essential for positioning the touch cell processes, or, conversely, failure to correctly position the processes results in the lack of the mantle. Molecular analysis of the mec-1 gene should distinguish between these possibilities. mec-1 action does not appear to be specific to the touch cells since amphidial defects have been noted in some mec-1 mutants (Lewis and Hodgkin 1977; Chalfie and Sulston 1981; Perkins et al. 1986).

The mec-5 and mec-9 genes may encode structural components of the mantle. mec-5 encodes a collagen expressed by the hypodermis (Du et al. 1996). No obvious ultrastructural defects are observed in mec-5 mutants, although a subtle mantle abnormality—inability to bind peanut lectin—has been noted (Chalfie and Sulston 1981; E. Hedgecock and M. Chalfie, unpubl.). Mutations in the mec-9 gene do not detectably alter the mantle, although mec-9 encodes a secreted protein (Chalfie and Sulston 1981; Du et al. 1996). The mec-9 gene encodes two transcripts, the larger of which encodes a 834-amino-acid protein that is expressed only in the touch receptors. The predicted MEC-9 protein contains a glutamic-acid-rich domain and several domains related to the Kunitz-type serine protease inhibitor domain, Ca+-binding EGF repeats and non-Ca+-binding EGF repeats (Du et al. 1996). mec-9 mutations are dominant enhancers of a mec-5 (ts) allele, suggesting that these proteins might interact in the extracellular matrix outside the touch receptor neuron (Du et al. 1996; Gu et al. 1996). Potential roles for MEC-5 and MEC-9 in mechanotransduction are discussed below.

C. Genes That May Encode Subunits of a Mechanosensory Ion Channel

Loss-of-function mutations in mec-4 and mec-10 disrupt touch sensitivity but do not alter touch receptor ultrastructure (Chalfie and Sulston 1981). A few observations suggest that mec-4 and mec-10 could encode subunits of a mechanically gated ion channel. First, these genes encode proteins related to subunits of the vertebrate amiloride-sensitive Na+ channels, which are required for ion transport across epithelia (Driscoll and Chalfie 1991; Huang and Chalfie 1993; Chalfie et al. 1994; Canessa et al. 1993, 1994b; Lai et al. 1996). Although channel activity has not yet been directly demonstrated for either MEC-4 or MEC-10, certain nematode/rat chimeric proteins function in C. elegans and Xenopus oocytes, implying that the nematode and rat proteins are functionally similar (Hong and Driscoll 1994; Waldmann et al. 1995). Second, mec-4 and mec-10 are coexpressed nearly exclusively in the touch receptor neurons (Mitani et al. 1993; Huang and Chalfie 1994). Third, mec-4 and mec-10 are crucial for the mechanosensitivity of the touch neurons. For these reasons, it has been proposed that MEC-4 and MEC-10 could function in a mechanically gated ion channel, i.e., one that opens in response to membrane stretch or to mechanical displacement of a channel domain. Alternatively, the MEC-4/MEC-10-containing channels could be required to maintain the ionic milieu of the touch cells.

Additional members of this Na+ channel superfamily have been identified. C. elegans family members were called degenerins because dominant alleles of mec-4 and deg-1 induce neurodegeneration. These genes encode proteins similar in size (724−778 amino acids for currently characterized degenerins) which, like all family members, have two putative transmembrane domains. The more amino-terminal of these is hydrophobic, whereas the more carboxy-terminal of these (MSDII) is amphipathic. Genetic evidence (Hong and Driscoll 1994) and electrophysiological characterization of rat and rat/nematode chimeras (Waldmann et al. 1995) support the hypothesis that MSDII constitutes a pore-lining domain and that highly conserved hydrophilic residues in MSDII face into the channel lumen to influence ion flow. Interestingly, both hydrophobic domains are slightly longer than required for a single transmembrane domain. For this reason, it has been proposed that MSDI and MSDII include residues that loop back into the membrane forming a pore, similar to H5 domains of several characterized channel types (Jan and Jan 1994; Renard et al. 1994; García-Añoveros et al. 1995). Analysis of both the vertebrate and C. elegans proteins suggests that the amino and carboxyl termini are cytoplasmic and the central region is extracellular (Canessa et al. 1994a; Renard et al. 1994; Snyder et al. 1994; García-Añoveros et al. 1995; Lai et al. 1996).

Dominant gain-of-function mec-4 alleles induce swelling and death of the touch receptor neurons (Chalfie and Sulston 1981; Chalfie and Au 1989; for discussion of degenerate cell death, see Hengartner, this volume). Toxic mec-4 (d) alleles encode substitutions of large side-chain amino acids for a conserved alanine residue (A713) situated adjacent to MSDII (Driscoll and Chalfie 1991; Lai et al. 1996). Since nontoxic mec-4 alleles encode small side-chain amino acids at this position, steric hindrance is postulated to play a critical part in degeneration. A working model for the initiation of cell death is that the presence of a bulky side chain at position 713 prevents the channel from closing effectively, producing an increased influx of ions that proves toxic. Consistent with such a model, amino acid substitutions in the predicted pore domain (which are likely to disrupt ion influx) block or delay degeneration (Hong and Driscoll 1994). Substitutions at the site corresponding to MEC-4 (A713) in degenerin family members deg-1 and mec-10 ( mec-10 [A673V]) also cause neuronal swelling and death which can be blocked by the presence of a second-site substitution in the predicted pore-lining domain (Huang and Chalfie 1994; García-Añoveros et al. 1995; Shreffler et al. 1995). Thus, the conserved alanine residue and the potential for toxic misregulation of channel activity by large amino acid substitutions at this site thus far appear to be general features of the degenerin gene family.

What is the subunit composition of the degenerin-containing channels? Coexpression of three distinct but homologous subunits of the rat amiloride-sensitive Na+ channel is required to reconstitute in vivo pharmacological properties of the channel in Xenopus oocytes (Canessa et al. 1994b). Furthermore, biochemical characterization of the mammalian epithelial Na+ channels suggests as many as six different subunits (Benos et al. 1987; Ausiello et al. 1992). The mec-6 gene may encode an additional subunit of the C. elegans degenerin ion channels. mec-6 mutations disrupt touch cell function and block mec-4 (d)-and mec-10 (A673V)-induced degeneration (Chalfie and Wolinsky 1990; Huang and Chalfie 1994), consistent with the hypothesis that mec-6 encodes a third protein required for channel assembly or function.

Genetic interactions between mec-4 and mec-10 alleles also suggest that the touch receptor channel is multimeric. Specific mec-4 alleles suppress the toxicity of a mec-4 (d) allele in trans-heterozygotes (Hong and Driscoll 1994). Likewise, one mec-10 allele can act in trans to suppress degeneration induced by the engineered toxic variant of mec-10 , mec-10 (A673V) (Huang and Chalfie 1994). Thus, MEC-4 and MEC-10 homomultimeric interactions are likely to occur in vivo. The fact that mec-4 (lf) mutations suppress cell death induced by mec-10 (A673V) supports the idea that MEC-4 and MEC-10 also interact with each other (Huang and Chalfie 1994). One interpretation of the observed genetic interactions is that the minimal channel complex may include at least two subunits of MEC-4, two of MEC-10, and one of MEC-6. An alternative explanation for the heteroallelic interactions is that individual subunits compete for assembly into a multimeric complex.

D. The Degenerin Gene Family

A relatively large family of degenerins is encoded in the C. elegans genome (at least 13 closely related members reported to date by the Genome Sequencing Project). Multiple subunit genes, differential splicing, and differential expression patterns seem to contribute to the diversity of channel functions mediated by C. elegans family members; this diversity appears to include signal transduction in specialized mechanosensory cells, muscle contraction, and possibly volume control in diverse cell types. Since characterization of other family members provides some insight into the function of the channel in the touch receptors, we briefly digress to review current data on the degenerin family here.

Although the C. elegans degenerins share approximately 25−30% sequence identity with their vertebrate counterparts, clear differences distinguish vertebrate and C. elegans family members. A cysteine-rich domain and a 22-amino-acid region in the ectodomain (García-Añoveros et al. 1995; see below) are unique to the C. elegans proteins. In addition, the carboxyl termini of C. elegans degenerins lack the proline-rich regions important to the function of the vertebrate proteins. In the rat, a proline-rich SH3-binding domain in the rat α-subunit interacts with spectrin to direct the channel to the apical side of the epithelial cell (Rotin et al. 1994). In humans, the proline-rich carboxy-terminal domains of the β and γ subunits of the human epithelial Na+ channel have important roles in regulating channel function since Liddle's syndrome, a hypertensive disorder caused by elevated channel activity, is caused by deletions of these domains (Shimkets et al. 1994; Hansson et al. 1995).

1. deg-1

The first molecularly characterized family member, deg-1 (named for degeneration), was identified by a dominant gain-of-function allele (u38) that induces swelling and death of multiple neurons that are unrelated by cell lineage, position, or function (Chalfie and Wolinsky 1990). The PVC interneuron that mediates the posterior touch response dies late in larval development in the deg-1 (u38) background, and thus the mutants have a Tab (touch abnormal) phenotype: They respond neither to gentle touch nor to strong prods on the posterior. The ASH neurons also die in deg-1 (u38), rendering mutants insensitive to nose touch (A. Hart and J. Kaplan, unpubl.; C. Bargmann, pers. comm.). Interestingly, loss-of-function deg-1 mutations isolated as intragenic suppressors of deg-1 (u38) mediated death have no apparent phenotype, suggesting either that deg-1 activity is not required for PVC and ASH function or that deg-1 is functionally redundant (Chalfie and Wolinsky 1990; García-Añoveros 1995; A. Hart and J. Kaplan, unpubl.). mec-6 mutations suppress deg-1 (u38)-induced cell death, implying that the deg-1 channel complex could include this candidate subunit.

deg-1 encodes a predicted protein of 778 amino acids, although alternative processing of deg-1 transcripts can remove up to 42 amino acids from the protein (Chalfie and Wolinsky 1990; García-Añoveros et al. 1995; Shreffler et al. 1995). As noted above, dominant death-inducing deg-1 mutations affect the conserved alanine residue corresponding to the site affected by dominant degeneration-inducing mutations in the mec-4 (d) mutant (García-Añoveros et al. 1995; Shreffler et al. 1995). A recessive degeneration-inducing deg-1 allele, u506, encodes an A393T substitution in a 22-amino-acid sequence in the predicted extracellular domain that is conserved among the C. elegans family members but is missing from the mammalian proteins (García-Añoveros et al. 1995). Introduction of the same amino acid change into MEC-4 also creates a toxic allele. Ionic influx is implicated in toxicity by this extracellular substitution since mutations that disrupt the predicted channel pore block killing by the deg-1 (A393T) mutation. These results suggest that the 22-amino-acid extracellular domain could regulate ionic flux through these channels. An extracellular gating domain unique to C. elegans family members could confer rapid channel gating properties, distinguishing it from the vertebrate epithelial channel, which has long open times.

2. unc-105

At least one member of the degenerin family can dramatically affect muscle function. Unusual semidominant alleles of the unc-105 gene induce hypercontraction of body wall muscles (Park and Horvitz 1986a). Null alleles of unc-105 do not have an apparent phenotype, suggesting functional redundancy. Molecular analysis established that unc-105 encodes a degenerin and that semidominant unc-105 mutations affect residues near, but not at, the site of the alanine residue affected in mec-4 (d) and deg-1 (d) (B. Shrank et al., pers. comm.). The unc-105 (sd) mutation causes muscle hypercontraction presumably because muscle cells are depolarized by inappropriate ion influx. Specific alleles of let-2 (also known as sup-20 ) restore locomotion to unc-105 (sd) animals (Park and Horvitz 1986b; J. Liu and R. Waterston, pers. comm.). let-2 encodes an essential basement membrane collagen (Sibley et al. 1993; see Kramer, this volume), suggesting that UNC-105 may function in a stretch-responsive channel in body wall muscle that is gated via attachment to collagen in the extracellular matrix. Alternatively, by analogy to the mammalian epithelial sodium channels, UNC-105 channels may primarily mediate vectorial ion transport. In this case, the functional interaction with the LET-2 collagen could reflect the need to localize channels to a particular domain of the plasma membrane. In either case, these results underscore the likely importance of extracellular matrix in the function of degenerin channels.

3. unc-8

Dominant unc-8 alleles cause periodic swelling of subsets of embryonically derived motor neurons (DB3-DA7) and other neurons in the head and tail ganglia (Shreffler et al. 1995). Affected cells do not die—swelling of midbody motor neurons begins after hatching, peaks in severity late in L1 and L2, and then regresses. unc-8 mutants coil and are unable to back up, they lay eggs constitutively, and at least some are resistant to NDG (N-dihydroguairetic acid), a lipoxygenase inhibitor. NDG blocks a signal transduction pathway that opens FMRF-sensitive S-K+ channels in Aplysia (Belardetti et al. 1989). Analysis of mutations that suppress unc-8 (d)-induced defects established that (1) unc-8 null alleles have no apparent phenotype; (2) specific unc-8 alleles can suppress or enhance unc-8 mutations in trans, suggesting that UNC-8:UNC-8 interactions occur; and (3) mec-6 mutations can suppress unc-8 (d)-induced phenotypes (Shreffler et al. 1995). Similar genetic properties are exhibited by deg-1 and some other degenerins, and thus it has been hypothesized that unc-8 could encode another family member. Indeed, molecular analysis has established that unc-8 encodes a degenerin family member (N. Tavernarakis et al., unpubl.).

Further support for UNC-8 association with degenerin-like channels comes from analysis of the dominant extragenic suppressor sup-41 (lb125). In addition to suppressing unc-8 (d) phenotypes, this allele partially suppresses the posterior touch insensitivity (the Tab phenotype) of a deg-1 (d) allele (Shreffler et al. 1995). In contrast, sup-40 (lb130) suppresses all phenotypes associated with unc-8 (d) but does not alter deg-1 (d) induced cell death (Shreffler et al. 1995). Unselected phenotypes of sup-40 (lb130) include slow growth, production of fragile enlarged oocytes, swelling and occasional detachment of adult hypodermal nuclei, and strong NDG resistance. These phenotypes are expressed independently of the unc-8 locus and are not affected by mec-6 mutations.

E. MEC-2, a Potential Protein Link between the Mechanosensory Channel and the Cytoskeletal Network

Mutant mec-2 alleles appear to disrupt touch receptor function specifically. Among the 54 mec-2 alleles are some that are semidominant or fully dominant and exhibit a complex pattern of interallelic complementation (Chalfie and Sulston 1981); these genetic data imply that MEC-2 proteins functionally interact in vivo. Features of the predicted 481-amino-acid MEC-2 protein also imply involvement in protein-protein interactions (Huang et al. 1995). The carboxy-terminal domain of MEC-2 has a proline-rich region that is similar to that of the SH3-binding domains. The central MEC-2 domain (amino acids 114−363) includes a hydrophobic domain (amino acids 114−141) and a cytoplasmic hydrophilic domain that together exhibit 65% identity to the human red blood cell (RBC) protein stomatin. Stomatin is an integral membrane protein that associates with the cytoskeleton and affects ion balance via an unknown mechanism. Interestingly, RBCs that lack stomatin (i.e., hereditary stomatocytosis) swell and exhibit elevated Na+ permeability (Stewart et al. 1993).

Genetic evidence suggests that the MEC-2 protein functionally interacts with both the degenerin channels and the microtubule network. Certain mec-2 alleles partially suppress mec-10 (d)-induced death (Huang and Chalfie 1994). In addition, some recessive mec-2 alleles act as dominant enhancers of a weak mec-4 (ts) allele (Huang et al. 1995; Gu et al. 1996). In a wild-type background, a MEC-2LacZ fusion protein is distributed along the touch receptor axon as well as in the cell body; this distribution depends on the integrity of the 118-amino-acid amino-terminal MEC-2 domain, which is situated in the cytoplasm (Huang et al. 1995). The axonal distribution of a MEC-2LacZ fusion protein is mildly disrupted in a mec-7 (null) or mec-12 strong loss-of-function background. More dramatically, two specific mec-12 missense alleles interfere with localization of MEC-2 fusion proteins, restricting the fusion proteins to the cell body (Huang et al. 1995). If the implied interactions are direct, a simple hypothesis is that MEC-2 may tether the 15-protofilament microtubules to the degenerin channel, an association that might enable mechanical deflection of microtubules to open the channel (Huang et al. 1995). These models must be considered highly speculative until the predicted biochemical interactions have been documented.

F. Molecular Biology of Other mec Genes

Mutations in mec-14 disrupt touch receptor function without altering ultrastructure (Chalfie and Au 1989). mec-14 encodes a member of the superfamily that includes the β subunits that associate with, and modify the activity of, Shaker-type K+ channels (N. Hom and M. Chalfie, pers. comm.). Since mec-14 alleles can partially suppress mec-10 (A673V)-induced death (Huang and Chalfie 1994), it appears that MEC-14 influences mechanosensory channel function. However, whether MEC-14 interacts with the touch receptor channel directly to modify activity via mechanisms similar to that of K+ channel β subunits (see, e.g., Rettig et al. 1994) is not known.

mec-8 most likely affects touch cell function indirectly by influencing expression of other mec genes. mec-8 alleles disrupt touch sensitivity (Chalfie and Sulston 1981), but they also affect amphid and phasmid development, attachment of body wall muscle to the hypodermis and cuticle, and embryonic and larval development (Perkins et al. 1986; Lundquist and Herman 1994). The MEC-8 protein, which includes two RNA-binding motifs, is required for proper splicing of several messages including its own, that of the UNC-52 protein and that of the MEC-2 protein (M. Huang 1995; Lundquist et al. 1996).

The molecular identities of mec-1 , mec-6 , mec-15 , mec-17 , and mec-18 remain to be elucidated. A mec-15 allele partially suppresses mec-10 (A673V), and mec-18 alleles enhance mec-10 (A673V)-induced deaths (Huang and Chalfie 1994), suggesting that the products of these genes could directly influence channel function.

G. A Model for Mechanotransduction in Touch Receptor Neurons

The molecular identities of genes required for touch cell function suggest a model for how a mechanical stimulus such as gentle touch is transduced into a locomotory response (for a discussion of this model, see Huang et al. 1995; Du et al. 1996; G. Gu et al., pers. comm.). This model shares features of the proposed gating mechanism of mechanosensory channels that respond to auditory stimuli in the vertebrate inner ear (Fig. 4) (for review, see Hudspeth 1989; Pickles and Corey 1992). In the inner ear, hair cells that have bundles of a few hundred stereocilia on their apical surface mediate sensory transduction. The stereocilia are connected at their distal ends to neighboring stereocilia by filaments called tip links. Directional deflection of the stereocilia relative to each other creates tension on the tip links; this tension is proposed to open the channels directly. Consistent with this model, the integrity of the tip links is essential for channel opening, and the mechanosensitive channels appear to be situated at the ends of the stereocilia, near the connecting tip links (Hudspeth 1982; Assad et al. 1991; Lumpkin and Hudspeth 1995; Denk et al. 1996).

Figure 4. A model for mechanotransduction in the ear.

Figure 4

A model for mechanotransduction in the ear. Adjacent stereocilia in a hair bundle are connected at their apical surface by tip links (shown as springs). The short end of each tip link is thought (more...)

Central to the model of mechanotransduction in the touch cells (Fig. 5) is the heteromeric degenerin channel, composed of MEC-4, MEC-10, and MEC-6 subunits. These subunits assemble to form a pore in the membrane that is lined by hydrophilic residues in MSDII. Subunits are oriented such that their amino and carboxyl termini project into the cytoplasm and their cysteine-rich regions extend outside the cell. Localized tension, which is expected to be required for regulated opening and closing, is likely to be administered by tethering the extracellular channel domains to the specialized extracellular matrix and anchoring intracellular domains to the microtubule cytoskeleton. On the extracellular side, channel subunits may interact with MEC-1, MEC-5, and/or MEC-9 in the touch receptor mantle. Inside the cell, channel subunits are likely to interact with the 15-protofilament microtubules, which may contact the channel at their distal ends. Microtubules might be connected to the channel via MEC-2, a linker protein that may interact both with the MEC-12 α-tubulin and with intracellular channel domains.

Figure 5. Speculative model for mechanotransduction in the touch receptor neurons.

Figure 5

Speculative model for mechanotransduction in the touch receptor neurons. (A,B) The mechanosensitive channel in the touch receptors is hypothesized to include MEC-4 (more...)

How is the touch signal transduced? A touch stimulus could deform the microtubule network, which could tug the channel open from the intracellular side. Alternatively, a touch stimulus could perturb the mantle connections and pull the channel open from the extracellular side. In either case, Na+ influx would activate the touch receptor to signal via gap junction connections to interneurons in the touch relay circuit, eliciting locomotion in the appropriate direction. This model provides a number of specific predictions that can be easily tested with available tools, although experiments such as reconstituting channel triggering in heterologous expression systems will likely be challenging.

Is mechanotransduction in C. elegans similar to that in other organisms? One mechanosensitive ion channel gene, mscL, has been cloned from Escherichia coli (Sukharev et al. 1994). The MscL channel forms a large conductance nonselective ion channel. MscL encodes a 136-amino-acid protein that appears to have two transmembrane domains with intracellular amino and carboxyl termini (P. Blount and C. Kung, pers. comm.) but does not share striking primary sequence similarity to MEC-4 or MEC-10. Purified MscL reconstituted in liposomes is mechanosensitive, so it appears that membrane stretch is sufficient to generate the gating force for this channel. Overall, the primary sequence of the MscL channel and the means of delivering gating tension appear to be different from that proposed to mediate mechanotransduction in the C. elegans touch receptor neurons.

As discussed above, the mechanically gated ion channel that mediates auditory transduction in the vertebrate inner ear has been studied using elegant biophysical approaches (for review, see Hudspeth 1989; Pickles and Corey 1992), but the genes encoding the auditory channel subunits have not been cloned. Might the hair cell channel be a degenerin family member? Such a possibility remains a plausible hypothesis, but it should be noted that the hair cell channel is a relatively nonselective cation channel that is clearly permeable to K+, whereas indirect evidence (Waldmann et al., 1995) suggests that the touch cell channel, like the rENaC channel, may be Na+-selective. In addition, if a degenerin family member is a component of the hair cell channel, it appears that the gene encoding that channel may not be highly conserved with MEC-4 and MEC-10, since PCR and low-stringency screening approaches have not quickly yielded a degenerin homolog expressed in the hair cells.

Copyright © 1997, Cold Spring Harbor Laboratory Press.
Bookshelf ID: NBK20187
PubReader format: click here to try

Views

  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...