Yasuyoshi Nishida

Rearrangements of Immunoglobulin Genes during Differentiation and Evolution

Immunoglobulin genes are shown to undergo dynamic rearrangements during differentiation as well as evolution. We have demonstrated that a complete immunoglobulin heavy chain gene is formed by at least two types of DNA rearrangement during B cell differentiation. The first type of rearrangement is V-D-J recombination to complete a variable region sequence and the second type is S-S recombination to switch a constant region sequence. Both types of recombination are accompanied by deletion of the intervening DNA segment. Structure and organization of CH genes are elucidated by molecular cloning and nucleotide sequence determination. Organization of H chain genes is summarized as VH-(unknown distance)-JH-(6.5 kb)-C mu-(4.5 kb)-C delta-(unknown distance)-C gamma 3-(34 kb)-C gamma 1-(21 kb)-C gamma 2b-(15 kb)-C gamma 2a-(14.5 kb)-C epsilon-(12.5 kb)-C alpha. The S-S recombination takes place at the S region which is located at the 5 side of each CH gene. Nucleotide sequence of the S region comprises tandem repetition of closely related sequences. The S-S recombination seems to be mediated by short common sequences shared among S regions. A sister chromatid exchange model was proposed as a mechanism for S-S recombination. Comparison of nucleotide sequences of CH genes indicates that immunoglobulin genes have scrambled by intervening sequence-mediated domain transfer during their evolution.

Long terminal repeat-like elements flank a human immunoglobulin epsilon pseudogene that lacks introns.

There are at least three immunoglobulin epsilon genes (C epsilon 1, C epsilon 2, and C epsilon 3) in the human genome. The nucleotide sequences of the expressed epsilon gene (C epsilon 1) and one (C epsilon 3) of the two epsilon pseudogenes were compared. The results show that the C epsilon 3 gene lacks the three intervening sequences entirely and has a 31‐base A‐rich sequence 16 bases 3′ to the putative poly(A) addition signal, indicating that the C epsilon 3 gene is a processed gene. The C epsilon 3 gene sequence is homologous to the five separate DNA segments of the C epsilon 1 gene; namely, a segment in the 5′‐flanking region (100 bases) and four exons, which are interrupted by a spacer region or intervening sequences. Long terminal repeat (LTR)‐like sequences which contain TATAAA and AATAAA sequences as well as terminal inverted repeats are present in both 5′‐ and 3′‐flanking regions. The 5′ and 3′ LTR‐like sequences do not, however, constitute a direct repeat, unlike transposable elements of eukaryotes and retroviruses. The 3′ LTR‐like sequence is repetitive in the human genome, but is not homologous to the Alu family DNA. Models for the evolutionary origin of the processed gene flanked by the LTR‐like sequences are discussed. The C epsilon 3 gene has a new open frame which codes potentially for an unknown protein of 292 amino acid residues.

Organization and Reorganization of Constant Region Genes of Immunoglobulin Heavy Chains: Genetic Basis for Class Switching

We have determined the complete organization of the mouse CH gene family, which is comprised of the 8 CH genes in the order 5’-JH-6.5kb-Cμ-4.5kb-Cδ-55kb-Cγ3-34kb-Cγ1-21kb-Cγ2b-15kb-Cγ2a-14kb-Ce-12kb-Cα-3’. The S regions, which contain characteristic tandemly repeated unit sequences, are located 5’ to each CH gene except for the Cδ gene. There are at least two types of repetitive sequences dispersed in this 200 kb region. No pseudogenes are present. The arrangements of the CH genes in BALB/c and C57BL mice are similar, but the lengths of the S regions vary. The basic structures of all the CH genes are similar in that coding sequences are interrupted at the junctions of the domains and the hinge regions. Comparison of the nucleotide sequences of the CH genes revealed that sequence segments have been exchanged among members of the CH gene family. Cloning and characterization of human Cγ genes, i.e. Cγ1, Cγ2, Cγ3, Cγ4 and φCγ, indicate that the human Cγ gene family evolved by dynamic DNA rearrangements, including gene duplication, exon duplication, and exon reassortment by unequal crossing-over. A human pseudo-epsilon gene (Ce3) is a processed gene that has completely spliced out introns. The presence of movable genetic elements surrounding the Ce3 gene suggests that the Ce gene evolved by a translocation mechanism. Although S-S recombination has been shown to take place in myelomas and hybridomas secreting a large amount of immunoglobulin, analyses of the CH gene organization in normal spleen B cells bearing immunoglobulin on their surface suggest that RNA splicing may be responsible for the first step in class switching, followed by S-S recombination. The nucleotide sequences of S regions contain short common sequences, TGGG(G) and (G)AGCT. Comparison of nucleotide sequences surrounding recombination sites revealed common sequences TGAG and TGGG. A sister chromatid exchange model was proposed to explain deletion of CH genes accompanying S-S recombination. We have found that the S region serves as a preferred recombination site in E.coli extracts.