RNA from a rat liver
tumor (
Morris hepatoma 5123tc) was used to construct cDNAs together comprising the complete coding sequence of rat
oncomodulin mRNA. Information obtained from these cDNAs as well as from primer extension analysis gave a deduced length for the complete
oncomodulin mRNA of approximately 680
nucleotides (excluding the
poly(A) tail) including a 5'-untranslated region of 97 +/- 2
nucleotides, a 324-nucleotide-coding sequence and a 259-nucleotide 3'-noncoding region. Comparison of the
oncomodulin cDNA sequence with those coding for other members of the
calcium-binding protein family shows little homology with the exception of a recently reported
parvalbumin cDNA where the
oncomodulin and
parvalbumin nucleotide sequences are 59% identical in the protein-coding region.
RNA blot analysis of
poly(A+) RNA from normal adult rat liver gave no evidence of
oncomodulin expression in this tissue. A single
RNA species was detected, however, in
RNA extracts from the
hepatoma and from rat and human placentas. A probe prepared from one of the rat
oncomodulin cDNAs hybridized with a single
DNA species in restriction digests of
hepatoma and normal
DNA from rat and sequences in
DNA of humans and other mammals. A 38-nucleotide sequence spanning the 5'-untranslated region and the first seven
codons of the
oncomodulin cDNA, was far less homologous than was the same region of a
parvalbumin cDNA, to a chicken
calmodulin cDNA sequence coding for the first
calcium-binding domain. The
oncomodulin gene appears to have diverged more from that of
calmodulin than has the
parvalbumin gene.