Worm Breeder's Gazette 15(2): 33 (February 1, 1998)

These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.

The bli-1 and bli-2 genes encode collagens involved in strut assembly.

Jennifer R. Crew, James M. Kramer

Department of Cell and Molecular Biology, Northwestern University Medical School, Chicago, IL 60611.

The adult cuticle of Caenorhabditis elegans consists of cortical and
basal layers connected through a medial layer by struts.  The medial
layer and struts are unique to the adult cuticle.   The struts are
periodically placed within the cuticle, lying roughly in rows along
either side of the annulae, oriented perpendicular to the plane in which
the rest of the cuticle is arranged. 
        The bli-1 and bli-2 genes have been physically and genetically
mapped.  Transformation rescue has been used to show that the bli-1 gene
corresponds to sequence CO9G5.6, which encodes an unusual cuticle
collagen.  The amino- and carboxyl-terminal domains of this protein are
roughly 400 amino acids, approximately 10 times larger than the
corresponding domains in most cuticle collagens.  These terminal domains
contain large proportions of proline and charged residues. Although no
known motifs or domains implicated in protein interactions are evident
in these terminal domains, the large size and number of charges found in
these domains imply that BLI-1 may connect the layers flanking the
medial layer by interacting with other cuticle components. The mutant
phenotype of bli-2 is rescued by cosmid C32E11, which contains an open
reading frame encoding a more standard cuticle collagen.
        The mutant phenotype of these genes is cuticles covered by
fluid-filled swellings.  We have made double mutant combinations of
bli-1 and bli-2 with several heterochronic genes and observed that such
combinations yield blisters only when an adult cuticle is present. Using
a temperature-sensitive allele of bli-2, we showed that the function of
its product is necessary in the late L4 stage, the time during which the
adult cuticle is synthesized.  In addition, mRNAs for the bli-1 and
bli-2 genes are expressed highly only during the L4 stage of development
(B. D. Ackley and J. M. K., unpub.).  These data together imply a defect
specific to the adult cuticle. 
        Transmission electron microscopy of bli-1 and bli-2 mutants
shows that the struts are absent from the medial layer, which is filled
by granular electron dense material or morphologically abnormal struts,
implying that these proteins are involved in strut construction.  The
hypodermis of these mutants can vary in width and can also contain large
membrane bound structures.  These may be due to secretory defects caused
by incorrectly processed collagens, as have been noted previously in
vertebrates and for collagen type IV in the worm.
        Some bli-1 and bli-2 mutant animals have abnormalities in the
three-pronged adult alae, such as four-pronged alae or gaps and small
discontinuities in the treads.  Because tread assembly is controlled by
the lateral seam, we have examined its integrity.  Animals mutant in
bli-1 and bli-2 have normal numbers of seam cells that appear to fuse
correctly.  Some mutant animals also exhibit irregularities in the
spacing of the annulae.  It is thought that the circumferential actin
bundles that arise in the hypodermis prior to each molt may aid in
spacing these indentations.  We have examined these actin bundles in
animals undergoing the L4-adult molt; there do not appear to be any
abnormalitites in bundle orientation or distribution in mutant animals. 
        Blistered animals can have greatly shortened lifespans (3-4
days) and reproductive capabilities due to the severity of the
blistering.  We have observed that blisters can "heal" in a small
percentage of animals, allowing them to live essentially a normal
lifespan.  Although the severity of the blistering phenotype differs
between alleles, the frequency of healing events observed is about 10%
in every case.  We are unsure how this healing occurs, but it may
involve the fusion of the cortical and basal layers.
        Mutations in bli-1 and bli-2 exhibit allele-specific, partially
penetrant intergenic non-complementation, suggesting that their gene
products may interact.  We have also found similar interactions with the
other Bli genes.  In addition, we have TEM evidence that the medial
layer is the layer affected in animals mutant in all of the Bli genes. 
These data lead us to believe that these genes may constitute the
proteins necessary for constructing struts.  Therefore, this system
could be a valuable one for understanding the assembly and maintenance
of  a discrete extracellular structure.