The
lrp-1 (I) gene of C. elegans encodes a product that is similar to megalin (also called
gp330), one of two > 500 kD members of the LDL receptor family of proteins in mammals. Megalin is expressed on the apical surface of certain epithelial cells where it endocytoses a variety of extracellular ligands, including proteases, complexes of proteases and their inhibitors, and lipoprotein particles that have been enriched for apoE. A reverse-genetic analysis in mice has revealed that megalin is essential for viability: homozygous pups die soon after birth and display many developmental abnormalities, most of which are consistent with a failure to endocytose cholesterol (1). This hypothesis, however, has not been rigorously proved, and pleiotropy has rendered difficult an analysis of the phenotype. A reverse genetic approach has therefore been undertaken with C. elegans. Null mutations in
lrp-1 confer a general failure to thrive; homozygotes are Dpy, frequently paralysed, and rarely develop past the L3 stage. A striking effect of the mutations is a defect in molting: the homozygotes are often either completely stuck in old cuticle or have old cuticle attached to their tails. Problems in molting occur during each molt, and the final arrest of growth usually occurs at the L3 to L4 molt. Although arrested, they can live in this state for many days. The persistence of old cuticle suggests a problem with its degradation, a problem consistent with either a failure to regulate extracellular proteases or to endocytose sterols, an essential nutrient for nematodes and a possible source of steroid derivatives that could regulate molting (2). Genetic mosaic analysis is consistent with a function for LRP-1/megalin in
hyp7, the major component of the epidermis and the major source of cuticle. Monoclonal antibodies support this idea and reveal a further resemblance to mammalian megalin: LRP-1 is primarily expressed on the apical surface of the ridges of
hyp7. The extracellular part of LRP-1 must therefore be between the plasma membrane and the cuticle. Starvation of worms for cholesterol confers to some degree a similar defect in molting, suggesting that LRP-1 may be a receptor for the uptake of sterols. As one means of addressing this hypothesis, mutations in other genes are being isolated that confer a phenotype similar to that of the
lrp-1 mutations and to the effect seen with cholesterol starvation. These mutations may be in genes encoding proteins that help mediate the uptake or transport of sterols or that are more directly involved in hormonal signalling that controls molting. That C. elegans may use steroid-like signalling molecules for molting is supported by recent studies (3) of CHR3, a possible hormone receptor that resembles an ecdysone-inducible gene product from Drosophila. Finally, the genome sequencing consortium has revealed a second gene that can encode a large member of the LDL receptor family. The predicted gene product is distinct from that of
lrp-1 and has a greater resemblance to the mammalian LDL receptor-related protein. The presence of this second gene in C. elegans suggests that both large proteins appeared early in the evolution of animals. The function of this protein is also being investigated. (1) T. E. Willnow et al. PNAS 93, 8460-8464 (1996). (2) D. J. Chitwood, Nematode sterol biochemistry, in Physiology and Biochemistry of Sterols. 1992. (3) M. Kostrouchova et al. Development 125, 1617-1626 (1998).