Complete sequence of a cloned cDNA encoding rabbit secreted μ-chain of V(H)a2 allotype: Comparisons with V(H)a1 and membrane μ sequences

The complete sequence of a cDNA clone (pμ3) encoding a secreted IgM heavy chain from a rabbit of V(H)a2 allotype has been determined. Comparison of the nucleic acid and amino acid sequences with a comparable mouse μ-chain shows unusually high homologies of the sequences of the fourth domains and secretory portions of the molecules (80 to 90% vs 56 to 73% for the other domains). This high degree of homology is also seen with the sequences of the fourth domain and secreted terminus of human μ-chain. The DNA sequence of a second cDNA clone encoding the membrane form of rabbit IgM is ~ 90% homologous with the mouse and human coding sequences for the membrane terminus, and there is 98 and 97% amino acid sequence homology. We found an unusually long (156 base pair) 5' untranslated region in clone pμ3 and discovered that 105 bases at the 5' end were complementary to the terminal portion of the C(H)4 domain of our μ sequence. Southern blotting analyses and the finding of a 13 base pair segment in rabbit genomic V region DNA that is homologous to the complementary strand of a segment of the C(H)4 of rabbit μ strongly suggests that this extended sequence resulted from a reverse transcriptase 'hooking error'. We have also obtained and sequenced another cDNA clone from a V(H)a1 cDNA library. Comparison of the V region of clone pμ3 from V(H)a2 with the 117 available positions of a1 DNA sequence shows 25 differences, nine of which are in allotype-associated codons. There is also a two-codon deletion in the V(H)a1 compared to V(H)a2 in the third framework region that may contribute to allotypic differences. Development of V(H)a allotype-specific probes, now in progress, is complicated by the fact that most of the allotype-associated amino acids of a1, a2, and a3 can be derived from each other by single base changes.