We are interested in studying the in vivo functions of collagen IV, a major component of basement membranes (BMs). Type IV collagen is a heterotrimeric molecule mostly existing as [
a1]2[
a2]. In C.elegans,
emb-9 and
let-2 have been found to encode these two chains respectively. We have identified and characterized 11
emb-9 and 17
let-2 mutants. The majority of these mutants are temperature sensitive (ts), semi-dominant embryonic lethals that arrest at about the 2X stage. Most of these mutations are substitutions for the glycines that occur at every third position in the Gly-X-Y triple-helical domain of the molecule. By contrast, the putative null mutations of
emb-9 that we have generated are less severe since they develop to a later stage (>3X) in embryogenesis and are also recessive. This suggests that the semidominance of the glycine substitution alleles is caused because the mutant protein interferes not only with its wild type copy but also with other molecules. We have generated several specific antibodies against both the
a1 and
a2 chains. By immunofluorescence, we find that both chains colocalize and are found in most, but not all, regions where basement membranes are present. Antibody binding is observed around the pharynx, intestine, gonad and under the four body wall muscle quadrants, but not between the muscle quadrants. In glycine substitution mutants, little or none of the type IV collagen is secreted and it accumulates inside the body wall muscle cells. In ts mutants, at high temperature all type IV accumulates intracellularly, while at lower temperatures some type IV collagen is secreted and some accumulates intracellularly. The putative
emb-9 null mutants do not stain with anti-EMB-9 antibodies and show intracellular accumulation of LET-2, suggesting that LET-2 cannot assemble and be secreted on its own. To study where collagen IV is expressed, we generated transgenics containing lac-Z- promoter fusion constructs. In embryos, intense LacZ stain is seen in the body wall muscle cells, but no stain is seen in pharyngeal cells. Also, by immunofluorescence, none of the
emb-9 or
let-2 mutants show intracellular accumulation in pharyngeal cells. The pharynx is surrounded by a thick basement membrane that stains strongly with antibodies against both collagen IV chains. This suggested that the pharynx may not make its own collagen IV, but rather utilizes what is made elsewhere. To confirm this, we have generated transgenics carrying an HA epitope-tagged
emb-9 gene that is driven by the
unc-54 body wall muscle myosin promoter. Preliminary results show that the pharynx BM does stain with anti-HA antibody, suggesting that type IV collagen expressed in muscle cells can assemble into the pharyngeal BM. These results are inconsistent with the prevalent view that formation of the extracellular collagen network occurs by self-assembly. We have also isolated several suppressors of
emb-9 mutations, in order to find other BM components that might interact with collagen IV (intergenic suppressor), or get a better understanding of the collagen IV structure itself (intragenic suppressor). We have isolated 9 suppressors of the lethality of the
emb-9(
cg56), a glycine substitution mutation. These map to at least 3 different chromosomes. By immunofluorescence, we find that in suppressed animals there is relatively less intracellular accumulation and more secreted collagen IV than in
emb-9(
cg56) alone. We are currently characterizing these suppressors in greater detail.