The gene
egl-5 is the posterior member of the C. elegans Hox gene cluster. Although genetic analysis has allowed the identification of cell lineages affected by
egl-5 (Chisholm, 1991), little was known about the individual cells within these lineages that express the gene. By immunohistochemical staining with affinity-purified anti-EGL-5 polyclonal antibodies we were able to elucidate the
egl-5 expression pattern at the cellular level in most tissues, defining more precisely the time and focus of
egl-5 function. We were also able to establish some cross-regulatory interactions between
egl-5 and the Antennapedia homolog
mab-5. The expression pattern correlates well with the predicted tissues that were shown to require
egl-5 function. EGL-5 is detected in a number of blast cells, neurons, and muscles in the tail region, as well as in the HSN neurons in hermaphrodites, and in the gonad and sex muscles in males. The initial expression pattern, established during embryogenesis, includes expression in rectal epithelial cells, PLM, PVC, and HSN neurons and some posterior body muscle cells. Later, during the L1-L4 larval stages, other cell lineages, derived from P12, M, and V6, and, in the male, some gonad cells start expressing
egl-5. Once
egl-5 expression is established in a cell lineage the gene remains on in that lineage for the rest of larval development and into adulthood. Therefore
egl-5 seems to be typical of Hox genes in being a lasting marker of cell identity. Despite its considerable divergence at the amino acid level,
egl-5 is expressed during development in the anal/cloacal region and male genitalia in a pattern similar to those observed for AbdB in Drosophila and for AbdB homologs in vertebrates. This confirms that
egl-5 is a true AbdB homolog and suggests that patterning the proctodeal region was the ancestral function of this paralog group. In the male seam,
egl-5 is expressed initially in mid-L2 in V6.ppp. Expression persists in the V6.ppp descendants that will form sensory rays 4, 5, and 6. In L3, EGL-5 is also turned on in the ray 3 sublineage. EGL-5 staining in the V6 lineage is absent in
mab-5 males, confirming genetic evidence that
mab-5 function positively regulates
egl-5 in this lineage. We also suggest that
egl-5 function in the V6 lineage is responsible for turning
mab-5 off during L3. We propose that these two genes are in part responsible for the establishment of V ray identities. Cross-regulatory interactions between
egl-5 and
mab-5 were also observed in posterior P lineages in the ventral chord and preanal ganglion. While it was demonstrated that
egl-5 function in P12 is responsible for turning
mab-5 off in that lineage (Salser et al., 1993), we demonstrated that
mab-5 is responsible for turning
egl-5 off in the P10 and P11 lineages. Therefore
egl-5 and
mab-5 exhibit a variety of cross-regulatory interactions in different cellular contexts within the V6 and Pn lineages. Chisholm, A. 1991. Development, 111: 921-932. Salser et al. 1993. Genes & Development, 7: 1714-1724.