Social Life and the Single Nucleotide Foraging Behavior in C. elegans

utes to their tendency to form clumps spontaneously. Since the rise of the field of sociobiology, the study of Though the exact relationship between these behaviors the biological basis of social behavior, scientists have is not known, they all appear to be related to food restriven to assign genetic origins for a variety of social sponse in some manner: in the absence of food, clumpbehaviors. There have been a number of highly publiing and dispersing strains behave very similarly in all cized and often controversial studies of the basis of respects. Various experiments have been performed in human social behaviors such as sexual orientation and which the density, configuration, and composition of the religion. Less trumpeted by the popular press, there bacterial patch were manipulated. These experiments have been a number of more credible advances in the suggest that clumping is partly a result of interaction genetic analysis of complex behavioral traits. Two pabetween the nematodes and the food source, possibly pers in the past year, one in this issue of Cell (de Bono mediated by an odorant response. When cultured toand Bargmann, 1998), have established two interesting gether, animals from dispersing strains fail to join cases of a molecular basis for complex behaviors that clumps formed by clumping strains, indicating that disare arguably relevant to social interactions in natural persing strains have an altered response to theclumping populations. Both have to do with food foraging stratesignal. There is no obvious geographical explanation for gies, one in Drosophila and one in C. elegans. the natural occurrence of the two forms: clumping and The C. elegans case is more clearly identifiable as a dispersing strains each come from many locations, and social behavior, though I will argue later that the Droin one case both types were isolated from the same site sophila case may be similar. Twenty-two natural isolates (Hodgkin and Doniach, 1997). Instead, the two forms of C. elegans have beencollected from various locations may reflect alternative foraging strategies that are faaround the world (Hodgkin and Doniach, 1997). In the vored under different environmental conditions. In Drolaboratory, these strains are cultured on the surface of sophila, laboratory selection experiments suggest that agar-filled petri plates that have been seeded with a active foraging (roving) is selected when population denpatch of bacteria. It was noticed some time ago that sity is high and inactive foraging (sitting) is favored when these natural isolates adopt one of two distinct distribudensity is low (Sokolowski et al., 1997). It is thought tions on the agar surface: about half of the strains disthat more active foraging is advantageous when food perse evenly across the bacterial patch but the other resources are limiting (high density) but is wasteful when strains spontaneously form large dense aggregates, food resources are abundant (low density). Similarly, called clumps. This clumping arises, at least in part, animals foraging as a group may locate scarce or distant from interactions among the nematodes in the clump. food sources more readily, as is thought to be the case de Bono and Bargmann (1998) describe the dispersing with flocks of birds (Wilson, 1975). strains as “solitary” and the clumping strains as “social.” Finding the Gene Before proceeding, let me dispense with the issue of The standard C. elegans laboratory strain N2 is a diswhether “social” is used here in the sense meant by persing strain. In addition to the natural clumping isomost sociobiologists. By the definitions of E. O. Wilson lates, over the years various labs have identified three and others (Wilson, 1975), the case can be argued either mutagen-induced variants of N2 that show strong clumpway. Meanwhile, split your hairs as you prefer, but I will ing behavior. Crosses show that the clumping trait in stick with the reasonable suggestion that the clumping these three mutants and the natural isolates are recesphenomenon is relevant to social behavior. Certainly it sive to the dispersing trait, and that all tested combinais impossible not to be reminded of social interactions tions fail to complement each other, suggesting that when watching a writhing mass of nematodes spontanerecessive alleles of one gene result in the clumping beously clustered together while large expanses of the havior. Genetic mapping, candidate gene identification, plate are nearly unoccupied. and transgenic rescue showed that the gene responsiBehavioral Mechanism of Clumping ble, called npr-1, encodes a transmembrane receptor Two related questions about C. elegans clumping can most similar to the neuropeptide receptor subclass of be only partially answered at this time. First, what is the 7-pass transmembrane receptors. This result, together mechanism by which the clumps form? Second, why with the fact that an npr-1::gfp fusion is expressed in are both clumping and dispersing forms found in the neurons, suggests that the clumping behavior is conwild? Clumping is the most obvious manifestation of a trolled by an unknown neuropeptide acting through the complex set of behavioral differences between the two NPR-1 receptor. Secretion of this neuropeptide is prestrain types (Hodgkin and Doniach; 1997; de Bono and sumably regulated by food and acts as a modulator of Bargmann, 1998). Animals from clumping strains burrow the relevant neuronal circuit, as yet unidentified. The into theagar more readily and accumulate more strongly molecular identity of npr-1 fits nicely with its role in