GFP expression in the body-cavity neurons of C. elegans. Courtesy of S. Reichelt, MRC-LMB, Cambridge, UK.

The nematode worm Caenorhabditis elegans can be shy or gregarious when feeding time arrives. New work uncovers some of the neurons and genes that are involved in regulating social feeding behaviour in the worm, and points to multiple systems of antagonistic signalling that control whether, and when, the worms aggregate into feeding groups.

The standard laboratory strain of C. elegans is a loner, preferring solitary feeding. But worms with a valine-to-phenylalanine mutation at residue 215 of NPR-1 — a putative G-protein-coupled receptor — come together to form aggregates when they encounter bacteria (their food source). And npr-1-knockout worms also aggregate, indicating that the form of the receptor that contains valine (NPR-1 215V) normally suppresses aggregation. In two studies, de Bono and colleagues take advantage of the small nervous system of the worm and the powerful genetic tools available to delve deeper into the control of social feeding.

The first study investigated how and where NPR-1 acts. They constructed a transgene that expressed NPR-1 215V that was tagged with green fluorescent protein (GFP) and driven by the npr-1 promoter. When this transgene was expressed in npr-1-knockout worms, it suppressed aggregation, and the worms showed GFP expression in a number of neurons. By using different promoters to drive expression of the transgene in subsets of neurons, the authors showed that expression in just three sensory neurons — AQR, PQR and URX — was sufficient to suppress aggregation.

These three neurons are exposed to the fluid in the body cavity. Their firing can be inhibited by the selective expression of a gain-of-function mutant of a potassium channel, EGL-2, and this also suppresses aggregation. The ability of these neurons to mediate social feeding seems to depend on signalling through a cyclic-GMP-gated ion channel, as neuron-specific mutations in tax-2 or tax-4 , which encode the subunits of this channel, also inhibited aggregation. So, it seems that NPR-1 suppresses aggregation by antagonizing signalling through TAX-2 and TAX-4 in these sensory neurons.

The second study investigated how external stimuli might elicit aggregation. A screen for mutations that interfered with aggregation in npr-1-null animals identified four genes. Two of these, osm-9 and ocr-2 , are thought to encode subunits of a TRP (transient receptor potential)-related cation channel in C. elegans chemosensory neurons, and are required for avoidance of various noxious stimuli. The other two genes, odr-4 and odr-8 , are required to localize some chemosensory receptors to sensory cilia.

Localized expression of minigenes in specific neurons of double-knockout animals showed that the expression of ocr-2 and odr-4 is required in specific nociceptive neurons to rescue social feeding. Laser ablation of these neurons (the ASH and ADL pairs) abolished social feeding behaviour. Another piece of the puzzle came from studies of worms in which the osm-3 kinesin was knocked out. This gene product is required for the correct formation of sensory cilia on ASH and ADL, and on other sensory neurons. Although knocking out osm-3 interferes with the development of ASH and ADL, it doesn't suppress social feeding. The authors propose that, as well as blocking the ability of ASH and ADL to promote social feeding, lack of OSM-3 blocks antagonistic signals that normally inhibit this behaviour. In support of this idea, removing OSM-3 restored social feeding to animals defective in ODR-4 or OCR-2. So, as with the body-cavity neurons, nociceptive sensory neurons might be involved in a system of antagonism between signals that promote aggregation and those that suppress it.

As these neurons are required for responses to stressful or aversive stimuli, de Bono et al. propose that aggregation is a response to an aversive stimulus that is produced by bacteria. The next step will be to find out what the aversive stimulus that promotes aggregation is — and how these different control systems interact to determine when social feeding occurs.