Myofilament lattice assembly begins at the plasma membrane in regions of contact between adjacent muscle cells (Hresko et al., 1994, JCB 124:491-506). UNC-52 and integrin are key early components in the assembly process and are required for organization of cytoskeletal associated proteins such as vinculin and talin (Hresko et al., 1994; Moulder et al., 1996, MBC 7:1181-1193). To further dissect the assembly process, we asked whether integrin and other muscle components affect the localization of UNC-52 in the basement membrane. We stained embryos homozygous for Pat mutations (
pat-2,
pat-3,
pat-4,
pat-5,
pat-11,
pat-12,
unc-112, and
deb-1) with antibodies to UNC-52 (GM1) and MHC A (DM5.6). In wild-type embryos, GM1 stains the basement membranes associated with the pharynx, body wall, and anal muscles. In larvae and adults, staining is especially intense over the dense bodies and M-lines in the body wall muscles. Because the dense body/M-line staining is not observed in embryos, we focused instead on the basement membrane staining. Surprisingly, we found that none of these Pat mutants exhibit defects in the localization of UNC-52. For example, the GM1 staining pattern in
pat-3(
st564) mutant embryos appears to be quite normal, even in arrested embryos. On this basis, we conclude that beta-integrin is not required for the correct localization of UNC-52. Similarly, Pat alleles of
unc-112 have no affect on the localization of UNC-52. We took advantage of a characteristic defect to identify
unc-112 homozygotes prior to the one and a half-fold stage. When stained with GM1, a small break in each ventral muscle quadrant can be seen in wild-type embryos. This break may correspond to the position of the amphid commissure (C. R. Norris, I. A. Bazykina, E. M. Hedgecock, and D. H. Hall, personal communication). In
unc-112 mutants, this break is noticeably larger than in wild-type embryos and can be used to distinguish
unc-112 homozygotes from their wild-type siblings. Consequently, we were able to look at early
unc-112 embryos and establish that there are no defects in the initial localization of UNC-52. Some disorganization of UNC-52 was observed in the milder Pat mutants, including
myo-3 and
pat-12. In these mutants, sections of the dorsal muscle quadrants often pull away from the hypodermis, especially in older embryos (Williams and Waterston, 1994, JCB 124:475-490). The disorganization of UNC-52 in these mutants appears to be a secondary consequence of muscle detachment, rather than a direct effect on UNC-52. On the basis of these results, we conclude that UNC-52 localization is not obviously disrupted by mutations in any of these muscle-affecting genes, including
pat-3 (beta-integrin; Gettner et al., 1995, JCB 129:1127-1141) and
unc-112 (MIG-2; see abstract by Rogalski et al.). Studies on focal adhesion assembly in tissue culture systems indicate that extracellular matrix proteins are required for integrin clustering and cytoskeletal association. However, we suspect that the accumulation of UNC-52 over dense bodies and M-lines during post-embryonic development may be dependent on integrin and/or other membrane associated proteins. The assembly of mature adhesion structures probably involves inductive interactions that remodel and reorganize both the cytoskeleton and associated proteins, and the basement membrane.