We have shown that the RGS protein EGL-10 and its partners, the G0/iα GOA-1, the Gqα EGL-30 and the RGS EAT-16 interact genetically in the ASH sensory neuron to modulate avoidance responses of C. elegans to aversive stimuli. Thus the same set of interacting signalling proteins modulates behaviour at the output (neuromuscular junctions) and input (sensory neurons) ends of the neural circuit underlying avoidance. We show that, in ASH, EGL-10 does not affect primary signal transduction but acts downstream of the main signal transducer channel OSM-9 and of the propagation of stimulus-evoked Ca2+ transients to the cell body. The delay in the avoidance response of egl-10 mutants, the frequency of spontaneous, non-stimulus-evoked reversals of locomotion and the genetic interaction with the DAG kinase gene, dgk-1, suggest that EGL-10 contributes to the regulation of neurotransmitter release at the ASH synapses. The results of the genetic interactions and Ca2+ imaging experiments on goa-1 mutants indicate that GOA-1 also acts, in ASH, downstream of OSM-9 and that it interacts with EGL-10 in a fashion similar to that in which these two proteins control acetylcholine release at the neuromuscular junction. The Ca2+ imaging results on egl-30 mutant animals show that, in ASH, this Gqα protein influences Ca2+ transients and depolarisation in response to high osmolarity, indicating a modulatory role for this protein in ASH signal transduction. With regard to EAT-16, our results show that this RGS is required in ASH to modulate avoidance responses with effects opposite to those of EGL-30.
A possible model for the way these proteins function in ASH is depicted in Figure . The model shows the main avoidance signalling pathway in which signals, triggered by different aversive stimuli, converge on the main signal transduction channel of ASH and OSM-9. Gating of OSM-9 generates a signal that is transmitted to the cell body and to downstream neurons to trigger avoidance responses. The model also depicts two opposing modulatory pathways. The negative one, with G0/i
α GOA-1 as the key component, functions downstream of OSM-9 and inhibits neurotransmission by reducing DAG levels. EGL-10 acts in this pathway and affects avoidance by its established function, the inhibition of GOA-1 signalling, thus increasing the concentration of DAG levels at presynaptic sites. Its function is contrasted by the DAG kinase DGK-1 that inactivates DAG. In this model, on the basis of our results, the mechanisms of action of GOA-1 and EGL-10 in ASH appear to be largely the same as those at neuromuscular junctions in motoneurons. The positive modulatory pathway, with Gq
α EGL-30 as the key component, increases primary signalling as Ca2+
transients are reduced in egl-30
mutants in response to high osmolarity (Figure ). The pathway functions upstream of OSM-9 and its effect on behaviour is contrasted by the RGS EAT-16. That EGL-30, in ASH, acts on signal generation and transduction is also supported by previous results showing that serotonin stimulation of the avoidance response to mechanical stimuli is mediated by an increase in ASH Ca2+
]. Since serotonin modulates ASH avoidance responses through the SER-5 receptor and Gq
α signalling [11
], the result is consistent with EGL-30 Gq
α acting upstream of OSM-9 gating. Thus the mechanism of action of EGL-30 in ASH appears to be different from that by which this protein acts at neuromuscular junctions, where it has been shown to act presynaptically in signal transmission by increasing DAG concentration via the phospholipase PLCβ EGL-8 and facilitating transmitter (acetylcholine) release [16
]. Whether, in ASH, EGL-30 also acts presynaptically cannot be established on the basis of our experiments and will require further investigations. However, it is worth mentioning that the avoidance response of egl-8
mutants to the high osmotic strength stimulus was not reduced compared to that of wild-type animals (our unpublished observations) as would be expected on the basis of the neuromuscular junction mechanism. Similarly, our experiments are not sufficient to establish whether the RGS EAT-16, which inhibits the Gq
α positive regulatory pathway, acts in ASH on signal transduction or on transmission or on both.
Figure 6 A model for the ASH modulatory pathways. The model is discussed in the text, in the Discussion section. The molecules in the ovals are those studied in this paper. Only the main players are depicted. The question marks indicate that only the type of molecule (more ...)
We do not know the endogenous and/or environmental cue(s) to which the GOA-1 and EGL-30 modulatory pathways are responding, and they have not been investigated in this paper. The modulatory effects on ASH responses of the presence or absence of food and of serotonin, octopamine, dopamine and other transmitters have been described previously [5
]. Important recent work has identified some of the ligands of the receptors and of the Gα proteins involved in the modulation of the response of ASH to dilute octanol. This response is modulated by the feeding status of the animal through the neurotransmitters serotonin and octopamine, the SER-5 and OCTR-1 receptors, respectively, and Gs
α (GSA-1), Gq
α (EGL-30) and G0/i
α (GOA-1) signalling [9
]. The response to dilute octanol has also been shown to be negatively modulated by dopamine [10
], and DOP-3, a D2-like dopamine receptor, is necessary for this modulation. dop-3
mutants are hypersensitive to dilute octanol, and DOP-3 function is required in ASH as its expression in this neuron is sufficient to rescue the hypersensitivity of dop-3
]. It has also been shown that, at least in cholinergic motoneurons, DOP-3 signals through GOA-1 to inhibit locomotion [36
]. Together these data suggest that dopamine might contribute to the negative modulation of ASH through the DOP-3 receptor and the activation of GOA-1 signalling. The present study is focused on the role of EGL-10, GOA-1, EAT-16 and EGL-30 in the response of ASH neurons, not to dilute octanol, but to various aversive stimuli and in particular to high osmolarity. It is, however, reasonable to hypothesise that the neurotransmitters and/or neurohormones, as well as the receptors and the Gα signalling molecules, involved in the modulation of ASH are largely the same as those involved in the modulation of the response to dilute octanol.
Like the mammalian nociceptive neurons of the dorsal root ganglia, the C. elegans ASH sensory neurons detect stimuli of different nature (polymodality), use a TRPV channel as the main signal transduction channel and glutamate as a neurotransmitter. ASHs have thus proved to be among the most important models to study in live animals, with single-cell resolution, how nociceptive neurons function. The Gα and RGS molecules we have identified are largely conserved in evolution, and mammalian orthologs for each of them can be identified. Our results raise the possibility that also in mammals similar mechanisms might be in place to modulate the activity of nociceptive neurons in the pathway for pain sensation. The discovery of molecules involved in the modulation of signal transduction and signal transmission in nociceptor neurons and the elucidation of their mechanism of action may shed light on how pain is generated and how its control can go astray (that is, chronic pain) and may be useful for designing new pain control therapies.