Martin Chalfie (2008) C.elegans Neuronal Development Meeting "A Protein-Lipid Complex Transduces Touch in C. elegans"

Martin Chalfie

[C.elegans Neuronal Development Meeting, 2008]

Little is known about the molecular basis of mechanical sensing in eukaryotes, in part because of the difficulty of biochemically isolating components needed for mechanosensory transduction. Genetic analysis, however, provides a way of identifying important molecules needed to sense a mechanical stimulus. In the nematode Caenorhabditis elegans, gentle touch to the head or the tail produces a withdrawal response that is sensed by six sensory neurons. The study of the genes identified in screens for touch-insensitive mutants has provided insights into both the differentiation of these cells and the molecular basis of mechanosensation. The genes needed for the function of the cells encode channel components, extracellular matrix proteins, cytoskeletal proteins, and possible enzymes that enable or modulate the touch response. Genetic interactions, molecular analysis of these genes and their products, and electrophysiological studies using heterologous cells as well as in situ recordings indicate that these cells sense touch through a DEG/ENaC channel complex. This complex is localized and, possibly, coupled to extracellular components at attachment sites to the body wall. The channel complex contains at least four proteins (MEC-2, MEC-4, MEC-6, and MEC-10) with MEC-4 and MEC-10, both DEG/ENaC proteins, forming the pore of the channel. Similar amino acid substitutions in MEC-4 and MEC-10 altered the properties of the transduction current, demonstrating that the proteins are part of the transduction machinery. MEC-2 and MEC-6 amplify channel activity 30-40 fold in frog oocytes. MEC-2 appears to act through its ability to bind to the DEG/ENaC proteins, itself, and cholesterol, which is needed for channel function. We have hypothesized that this binding alters the local lipid environment of the channel, allowing it to open when mechanically stimulated. MEC-6 may also affect the lipid microenvironment, since it is similar to the human protein paraoxonase, which has been implicated in the regulation of lipid peroxidation. Our current model for mechanosensory transduction is derived from the elevator model of Ching Kung. In this model the touch stimulus causes the membrane to approach or move away from the attachment sites, causing the tethered channel complexes to move perpendicular to the membrane. This displacement then leads to the opening of the channel. Such a model can explain the rapid response and adaptation seen in the mechanosensory current.

Lawry, S.Terese et al. (2019) International Worm Meeting "Using the Million Mutation Project strains to obtain a range of touch insensitive mutants."

Chalfie, Martin, Walker, Matt, Lawry, S.Terese

[International Worm Meeting, 2019]

The Million Mutation Project (MMP; Thomson et al., Genome Res. 23: 1749-1762, 2013) has become an outstanding resource for mutations in many C. elegans genes. More recently Timbers et al. (PLoS Genet. 12: e1006235, 2016) used a subset of MMP strains to identify genes that could be mutated to give a dye-filling (Dyf) phenotype. Because most of the mutations in the MMP strains are homozygous, we were intrigued about the possibility that the MMP strains could be used to identify mutations having subtle phenotypic effects. We also wanted to assess how complete a screen utilizing the MMP strains would be. To this end we have undertaken a screen of the MMP strains to identify those with extreme to slight defects in the gentle touch response. Classical forward genetic screens (Chalfie and Sulston, 1981; Chalfie and Au, 1989) identified 18 genes necessary for gentle touch sensitivity, although others have also been found (e.g., among genes whose null phenotype is lethality). We have currently screened about 1700 of the 2007 MMP strains. 111 of these strains have an abnormal gentle touch response. 73 of these strains have mutations affecting the protein sequence, UTRs, or splice sites of genes previously known to mutate to touch insensitivity. In particular, we identified strains with mutations in all but one gene (mec-17) found in our initial forward mutageneses. mec-17 is extremely difficult to mutagenize to touch insensitivity; all four mutations in MMP strains failed to make the animals touch insensitive. Thus, the MMP collection, at least for viable phenotypes, appears to be essentially complete. Using MMP defects in the known touch genes that did and did not produce touch sensitivity, we were able to look for important domains within the encoded proteins. We are now identifying causative mutations in the remaining 38 strains, starting with the strains that exhibit a more extreme phenotypic defect. Because the MMP strains are sequenced, identifying the causative mutation should be relatively quick using genetic mapping. We are conducting several other screens using this resource and strongly recommend it to our colleagues.

Topalidou, Irini et al. (2009) International Worm Meeting "Determining the genetic profile of two types of C. elegans mechanosensory neurons: the touch receptor neurons and the FLP neurons."

Topalidou, Irini, Bounoutas, Alexander, Chalfie, Martin

[International Worm Meeting, 2009]

Cells acquire many of their characteristics during differentiation by expressing specific subsets of genes. The identification of this select gene pool is an important means of understanding cell fate determination. We used gfp cell sorting and cDNA microarrays to identify genes that are upregulated in the touch receptor neurons (TRNs) and FLP neurons, both of which require MEC-3 for their differentiation. We identified channel and receptor subunits, transcription factors, and calcium sensors among the 198 genes expressed at high levels in the TRNs. Using existing mutants and/or feeding RNAi, we tested the role in gentle touch sensation of most of the genes (163/189) not previously identified as needed for touch sensitivity. Eight of the genes were needed for optimal touch sensitivity; genes encoding the transcription factor ALR-1, the PON-1 homolog K11E4.3, and the GABA receptor LGC-37 were among them. We demonstrated that ALR-1 is needed in the TRNs throughout development to regulate the expression of touch neuron-specific genes. Further analysis on the role of K11E4.3 in touch sensation will be given elsewhere in this meeting (Y. Chen and M. Chalfie). Interestingly, most of the mec genes (i.e: mec-18) that were previously identified as TRN-specific were among the 203 genes upregulated in the FLP neurons. QPCR analysis for mec-18 mRNA in animals with genetically-ablated TRNs showed that mec-18 mRNA is indeed expressed in these animals, mainly at the early larvae stages, although no immuno-detectable MEC-18 is produced. Finally, a short-lived gfp driven by the mec-18 promoter (Pmec-18praja::gfp) produces fluorescence in young FLP neurons. We conclude that both post-transcriptional and transcriptional regulation play important roles in differentiating the touch receptor neurons from the FLP neurons.

Chen, Yushu et al. (2009) International Worm Meeting "A second paraoxonase-like gene is expressed in the touch receptor neurons."

Chen, Yushu, Chalfie, Martin

[International Worm Meeting, 2009]

Paraoxonases (PONs) are a family of enzymes with a wide range of activities, including drug metabolism, deoxidization of lipids, and detoxification of organophosphates. The mec-6 gene in Caenorhabditis elegans encodes a paraoxonase-like protein that is part of the MEC-4 channel complex that transduces gentle touch in the touch receptor neurons (TRNs) and is required for neurodegeneration caused by gain-of-function mutations in DEG/ENaC genes (e.g., mec-4 and deg-1). C. elegans has four other PONs-like genes: k11e4.3, e01a2.7, k05f6.11 and e01a2.10 whose products remain uncharacterized. We found that K11E4.3::YFP translational fusion is expressed in the TRNs and a few other neurons. In the processes of the TRNs, K11E4.3::YFP is distributed as puncta that co-localize with MEC-6 puncta. K11E4.3 is required for touch sensitivity in sensitized backgrounds, specifically in mec-6ts animals at the permissive temperature or mec-6 hypomorphic alleles that do not, on their own, produce touch insensitivity. These data suggest that the touch transduction complex may have more than one of these proteins. We are currently examining the expressing of all the C. elegans PON-like genes. Two, k05f6.11 and e01a2.10, are expressed in the hypodermis.

Chen, Xiaoyin et al. (2009) International Worm Meeting "Integrin Signaling is Required for Mechanosensation in the Touch Receptor Neurons."

Chalfie, Martin, Chen, Xiaoyin

[International Worm Meeting, 2009]

The touch receptor neurons are attached to the body wall through components that resemble the dense bodies of the muscles, and many dense body proteins are expressed in both types of cells. These proteins, many of which are associated with integrin signaling may be ultimately linked with extracellular matrix proteins, which are required for touch sensitivity. The role of the integrin signaling proteins in the touch receptor neurons is unknown, however, because elimination of the genes for these proteins usually results in a pat phenotype (paralyzed at two fold stage). Using a strain with enhanced RNAi in neurons and reduced or absent RNAi in non-neuronal tissues, we tested for effects of the loss of these genes on touch sensitivity. RNAi for genes encoding components of the dense bodies in body wall muscles (pat-2 encoding an a-integrin, pat-4 encoding integrin-linked kinase, pat-6 encoding actopaxin, unc-97 encoding PINCH, and unc-112 encoding a Mig-2-like gene) reduced the response of the animals to gentle touch. In accordance with the RNAi result, pat-2 and unc-112 promoter GFP fusions showed expression in the touch receptor neurons, while pat-4, pat-6 and unc-97 have previously been reported to be expressed in these cells. Interestingly, RNAi for unc-97 also altered the number and distribution of MEC-2 puncta in the touch receptor neurons, a phenotype we also observed for the weak allele unc-97(su110). These results suggest that integrin signaling may have roles both in the development and organization of the transduction channels in the touch receptor neurons and in their function. We are currently characterizing the roles that these components and other components of integrin signaling play in touch receptor neuron function.

Topalidou, Irini et al. (2011) International Worm Meeting "The Messiness of Combinatorial Control."

Topalidou, Irini, Chalfie, Martin

[International Worm Meeting, 2011]

Because of combinatorial expression, transcription factors can determine the development of more than one type of cell. For example, the transcription factor complex of MEC-3 and UNC-86 is required in embryos for the development of both the touch receptor neurons (TRNs) and the FLP neurons. Despite sharing these transcription factors, the TRNs, but not the FLP neurons, express several proteins needed for gentle touch, e.g., MEC-2, MEC-4, and MEC-18. Both the embryonic TRNs and the FLP neurons upregulate mRNAs for approximately 300 genes. Twenty three percent of the genes are upregulated in both cells. We were surprised to find that some of the commonly upregulated genes are MEC-3-regulated genes that had previously been identified as being specific to the TRNs. Although the mRNAs for these genes are detected in the FLP cells in low amounts, these mRNAs do not appear to be translated to detectable levels. These results suggest that transcriptional control is relatively inexact, but these apparent transcription errors are well tolerated and do not alter cell fate. Either a threshold amount of transcription is required for translation or these mRNAs are subjected to some sort of translational control. The FLP cells are, nonetheless, poised to make TRN-specific proteins. Expression of the transcription factor ALR-1, a downstream target of MEC-3 that ensures, but does not direct, TRN differentiation caused FLP neurons to make higher levels of the mRNAs for these genes and their proteins. This ectopic expression also prevented the FLP neurons from forming the fine branches that characterize the adult cells. Thus, alr-1 appears to change the fate of these cells. Our experiments suggest that combinatorial control, although potentially able to produce many different types of cells, can be messy. Such messiness may need to be controlled, but it also could be advantageous, allowing cells to take on additional characteristics under different conditions.

Chen, Xiaoyin et al. (2013) International Worm Meeting "Neuromodulation of C. elegans mechanosensation."

Chen, Xiaoyin, Chalfie, Martin

[International Worm Meeting, 2013]

Sensory perception is modified by cues detected by the sensory cells themselves and from signals initiated by other sensory cells. We show that mechanosensation in the touch receptor neurons (TRNs), which receive no synaptic input, is modulated by both mechanical and non-mechanical cues. These signals alter touch sensitivity through two signaling pathways that modulate touch sensitivity by converging on a common mechanism. Sustained mechanical signals, such as continuous vibration, increase TRN sensitivity to force. This long-term sensitization requires the activation of a secondary mechanosensory system involving the integrins and is independent of the MEC-4 mechanotransduction channels. Non-mechanical stress cues, such as dauer formation, hypoxia, and high salt, reduce touch sensitivity by reducing the expression of two insulin-like peptides, INS-10 and INS-22, which act as long-range neuromodulators to activate insulin signaling in the TRNs. Both insulin and integrin signaling compensate for each other and converge by using AKT-1 and DAF-16. Loss of AKT-1 activity increases the transcription of mfb-1, which reduces the amount of MEC-4 mechanotransduction channel available on the plasma membrane, thus changing the TRN sensitivity to force. These alterations in TRN sensitivity adjust the acuity and priority of the touch response to benefit the animal under diverse conditions. During sustained background stimulation, habituation reduces the response to background stimuli, but long-term sensitization counters habituation to maintain TRN response to strong stimuli, thus distinguishing background stimulation from stronger signals. The reduction of touch sensitivity under non-mechanical stress conditions increases the efficiency of the animal's ability to do non-mechanosensory tasks. For example chemotaxis is more efficient in the presence of mechanical distractions. Thus, modulation of touch can alter the balance between senses to prioritize sensory responses that can facilitate an escape from the stress conditions. Our findings demonstrate a non-synaptic neuromodulatory network that integrates multi-modal signals and alters behavior to optimize the animal's response to multiple conditions.

Chen, Xiaoyin et al. (2013) International Worm Meeting "Pleiotropic genes affecting touch sensitivity in C. elegans."

Chen, Xiaoyin, Chalfie, Martin

[International Worm Meeting, 2013]

Six touch receptor neurons (TRNs) sense gentle touch along the body in C. elegans. Although saturated mutageneses have identified many genes needed for mechanosensation in the TRNs, these screens were ineffective in identifying several groups of genes, including pleiotropic genes. To circumvent this problem, we used neuronally-enhanced feeding RNAi to screen 1005 pleiotropic genes for effects on mechanosensation and identified 61 genes affecting touch sensitivity. In addition to 11 genes involved in general transcription and translation, which likely play housekeeping functions, the remaining 50 genes are involved in protein degradation, calcium signaling, cell adhesion and cytoskeleton, mitochondrial function, endocytosis and exocytosis, and classical signaling pathways such as wnt, hedgehog, small GTPase and MAP kinase. We further confirmed six genes (cdk-1, tag-170, wmr-1, ifb-1, tom-1, and mca-3) that affect anterior touch sensitivity using available viable alleles. tag-170 encodes a thioredoxin domain-containing protein required for cell division and microtubule growth. Loss of tag-170 eliminated acetylated tubulin and disrupted microtubule organization in the ALM and PLM neurons, but less so in the AVM and PVM neurons. Consistent with previous findings of Bounoutas et al. (2011), the disruption of microtubules further led to reduced protein production from genes needed for mechanosensation, including mec-2 and mec-18. mca-3 encodes a plasma membrane calcium ATPase (PMCA) that should pump cytosolic calcium out of the cell. Mutations in mammalian orthologs of mca-3 cause deafness and balance defects in rats and exacerbate hearing loss in human. A partial loss-of-function allele of mca-3 reduced the touch-induced calcium response in the TRNs by four fold, but not the response to potassium depolarization, suggesting that mechanotransduction may be regulated by cytosolic calcium level. In conclusion, reverse genetic screens using neuronally-enhanced feeding RNAi complement forward mutagenesis screens to reveal additional genes needed for TRN development and function.

Chen, Yushu et al. (2011) International Worm Meeting "The paraoxonase-like protein K11E4.3 is needed for gentle touch and mec-4(d)-mediated neurodegeneration."

Chen, Yushu, Chalfie, Martin

[International Worm Meeting, 2011]

Paraoxonases (PONs) are a family of enzymes with various activities, including deoxidization of lipids, drug metabolism, and detoxification of organophosphates. Paraoxonases (PONs) have also been implicated in humans to protect against atherosclerosis, but the molecular role of the proteins is uncertain. In Caenorhabditis elegans, however, a PON-like protein MEC-6 is a necessary component of the MEC-4 mechanosensory ion channel complex in the six touch receptor neurons (TRNs), which transduces gentle touch. C. elegans has four other PONs-like genes (k11e4.3, e01a2.7, k05f6.11 and e01a2.10) whose products remain uncharacterized. We found that K11E4.3 is also expressed in the TRNs and a few other neurons. K11E4.3 co-localizes with MEC-6 in puncta along the TRN processes. K11E4.3 is required for touch sensitivity in sensitized backgrounds, specifically in mec-4 and mec-6 temperature sensitive alleles at the permissive temperature or mec-6 hypomorphic alleles that do not, on their own, produce touch insensitivity. Like MEC-6, K11E4.3 is required for neuronal degeneration caused by gain-of-function mutations in mec-4 [mec-4(d)]. Loss of either K11E4.3 or MEC-6 lowers the amount of MEC-4::YFP in the TRN process. K11E4.3 physically interacts with MEC-4(d) and increases MEC-4(d) channel activity in Xenopus oocytes. These data suggest that, as with the DEG/ENaC and stomatin-like proteins, the touch transduction complex has multiple PON-like proteins. These proteins may also be important in stabilizing the channel complex or in transporting it to the plasma membrane.

Coblitz, Brian et al. (2011) International Worm Meeting "A screen for mutants with defective hypodermal attachment of mechanosensory neurons."

Chalfie, Martin, Coblitz, Brian

[International Worm Meeting, 2011]

The touch receptor neurons (TRNs) project processes anteriorly along the body adjacent to the hypodermis. Hemidesmosomal-like structures on the hypodermis link it to TRN extracellular proteins. In newly hatched larvae the TRN processes lie next to the body wall muscle, but as the animals mature intervention by the hypodermis separates the TRN processes from the muscle. Mutations affecting the extracellular proteins MEC-1, MEC-5 (a collagen), and HIM-4 (hemicentin) disrupt this separation. To identify other components needed for the separation, we mutagenized animals in which the body wall muscles were labeled with Pmyo-3mCherry and the TRNs were labeled with mec-18::GFP. Attachment mutants had ALM processes adjacent to the muscle. In addition to identifying new alleles for mec-1, mec-5, and him-4, we also found alleles in at least 7 other genes. Two of these new genes cause the mechanosensory abnormal (Mec) phenotype, suggesting novel roles of mec genes in attachment. Efforts are underway to map and sequence the new mutations.