Evolution of genes for allelic and isotypic forms of immunoglobulin kappa chains and of the genes for T-cell receptor beta chains in rabbits

New insights into the evolution of the families of genes encoding immunoglobulins and T-cell receptors of rabbits (Oryctolagus cuniculus) have come from molecular genetic studies. In contrast to human and mouse, rabbits were shown to have two genes for the constant region of immunoglobulin light chains (C kappa 1 and C kappa 2 isotypes) and complex allelic variants of K1 (allotypes). Although K1 allotype protein sequences differed at up to 41% of the amino acid positions, 3' untranslated, 5', and 3' flanking regions were conserved, and in the coding regions 78-80% of the codons with differences had replacement changes. Proportions of silent changes and changes in noncoding regions were comparable. Thus, in spite of their markedly different protein sequences, the K1b4, b5, and b9 allotypes appeared to be products of allelic genes. Molecular genetic analyses suggested that they may have undergone rapid divergence after an ancestral K2-like gene duplicated. Some rabbits were found to have two similar T-cell receptor C beta genes as do humans and many strains of mice, but others appeared to have three different C beta. In addition, we found allotypic forms of C beta. Some of the C beta allotypic differences occurred at positions where analogous C kappa allotypic differences were found. We also found V beta in mouse and human that were more similar to rabbit V beta than closely linked rabbit genes were to each other. This contrasts with rabbit immunoglobulin VH gene sequences that reflect concerted evolution. The data suggested that T-cell receptor V beta genes duplicated prior to mammalian radiation.     

Identification of rabbit genomic Ig-VH pseudogenes that could serve as donor sequences for latent allotype expression

Synthetic DNA oligomers specific for the VHa allotypes of rabbit Ig genes have been used to identify latent allotypic sequences in homozygous a1 and a2 rabbits. Two Ig VH pseudogenes containing latent a3 regions have been cloned from the genome of a homozygous a2 rabbit. Analysis of the regions associated with allotype expression indicates that these two pseudogenes contain VHa- sequences in framework region 1 (FR1) and VHa3 sequences in FR3. One gene has undergone an unusual rearrangement with a third VH gene, deleting their intervening sequences and recombining in FR3 with sequences 5' to the leader exon. Our results demonstrate the presence of latent VH sequences in the genomic DNA of normal rabbits and suggest that a mechanism such as gene conversion is responsible for expression of genetically-unexpected Ig VH genes.     

Complex allotypes of the rabbit immunoglobulin kappa light chains are encoded by structural alleles.

We have isolated the rabbit immunoglobulin b9 Ck light chain gene and compared its nucleotide sequence with the b4, b4var , b5 and bas Ck sequences. In spite of the high number of substitutions found between the different rabbit Ck coding regions, only very few changes are silent. Furthermore, the nucleotide changes are clustered in segments which correlate with the bends and helical regions found in the tertiary structure of the Ck domain of the protein. The flanking regions present a higher degree of conservation than the coding regions. The two genomic EcoRI fragments hybridizing to a b4cDNA probe have been correlated with the two distinct loci, Ck1 and Ck2 : one encodes for the nominal b9 Ck allotype and the other contains the information for the bas Ck region. The b allotypes are true alleles which could have evolved by intergenic conversion.     

Comparison of latent and nominal rabbit Ig VHa1 allotype cDNA sequences

The genetic basis for the expression of a latent VH allotype in the rabbit was investigated. VH region cDNA libraries were produced from spleen mRNA derived from a homozygous a2a2 rabbit expressing an induced latent VHa1 allotype and, for comparison, from a normal homozygus a1a1 rabbit expressing nominal VHa1 allotype. The deduced amino acid sequences of the nominal VHa1 cDNA were concordant with previously published VHa1 protein sequences. A comparison of two complete VH-DH-JH and six partial VHa1 sequences reveals highly conserved sequence within VH framework regions (FR) and considerable diversity in complementarity-determining regions and D region sequences. Two functional JH genes or alleles are evident. Amino acid sequencing of the N-terminal 15 residues of pooled affinity-purified latent VHa1 H chain showed complete sequence identity with the nominal VHa1 sequences. Possible latent VHa1-encoding cDNA clones, derived from the a2a2 rabbit, were selected by hybridization with oligonucleotide probes corresponding to the VHa1 allotype-associated segments of the first and third framework regions (FR1 and FR3). cDNA sequence analysis reveals that the 5' untranslated regions of nominal and latent VHa1 cDNA were virtually identical to each other and to previously reported sequences associated with VHa2 and VHa-negative genes. Moreover, some latent VHa1 genes encode FR1 segments that are essentially homologous to the corresponding segment of a nominal VHa1 allotype. In contrast, other putative latent genes display blocks of VHa1 sequence in either FR1 or FR3 that are flanked by blocks of sequence identical to other rabbit VH genes (i.e., VHa2 or VHa-negative). These composite sequences may be directly encoded by composite germ-line VH genes or may be the products of somatically generated recombination or gene conversion between genes encoding latent and nominal allotypes. The data do not support the hypothesis that latent genes are the result of extensive modification by somatic point mutation.     

The sequences of rabbit kappa light chains of b4 and b5 allotypes differ more in their constant regions than in their 3'' untranslated regions.

We report the sequence of a cDNA clone encoding the entire variable and constant regions of a rabbit kappa light chain of b5 allotype. The deduced amino acid sequence of the variable region (positions 1-95) is 86% homologous to that of a b4 light chain protein [BS-1) (1) but the b4 and b5 constant regions are only 74% homologous. Comparison of this DNA sequence to that of a cDNA clone encoding a b4 constant region shows that the kappa allotypes b4 and b5 have diverged significantly more in their coding region than in the 3' untranslated regions (86% vs 96% nucleotide sequence homologies). This implies either a function for the 3' untranslated region with evolutionary pressures to conserve or an accelerated divergence of the coding regions.     

Organization and polymorphism of rabbit immunoglobulin heavy chain genes

Germline genes encoding C mu, C gamma, C alpha, and C epsilon heavy chains of rabbit immunoglobulins have been isolated from recombinant phage and cosmid libraries. The JH, C mu, C gamma, and C epsilon are found in a 5'-JH-C mu-C gamma-C epsilon-3' orientation on a 90kb stretch of DNA. Four C alpha genes have been cloned and presumably reside 3' to the other CH genes. Southern blot analysis of rabbit sperm DNA indicates that the rabbit genome contains a single C gamma gene, one C mu gene, and as many as 10 C alpha genes. Restriction site polymorphism is found for C mu, C gamma, and C alpha genes of rabbits of various heavy chain haplotypes. The organization of the rabbit CH genes differs from that of mouse and human CH genes in that the rabbit has multiple C alpha genes, whereas mouse and human have one or two C alpha genes, respectively. In addition, mouse and human have four C gamma genes, whereas rabbit has only one C gamma gene. The presence of a single C gamma gene indicates that at least in the rabbits examined, no germline gene encoding latent or unexpected, C gamma allotypes is present. The genetic control of the expression of latent C gamma allotypes is discussed.     

Partial amino acid sequence of a rabbit immunoglobulin light chain of allotype b5

The amino acid sequence of a rabbit immunoglobulin light chain of allotype b5 has been nearly completed. A comparison of its structure with that of light chains of allotypes b4, b6, and b9 confirms that the constant regions of these various kappa chains differ by 20-35%. The substitutions are clustered in parts of the second half of the chain, and the b5 form bears more resemblance to the b6 chain than to any other, in good agreement with previous serological data. The analysis of the variable region reveals the existence of certain allotype-associated residues which have also been reported in other b5 chains, but not in proteins of the other allotypes. An examination of the rabbit light chain sequences between positions 96 and 107 suggests that this portion of the chain may be encoded separately by a joining "J" DNA segment, as has been described previously for murine and human immunoglobulins. In the rabbit, however, these J kappa regions appear to differ from one allotype to another. Together with the extensive variations of the constant regions, these data suggest that the rabbit kappa gene organization more closely resembles the murine gamma system (four different C gamma genes each flanked by its J segment) than the murine kappa system (a single C kappa gene).     

Multiple sequences related to a constant-region kappa light chain gene in the rabbit genome

Four allelic genes control kappa light chain allotypes in the rabbit. Amino acid sequence studies have revealed an extensive divergence (22%--33%) in the alternative forms of the kappa constant region (b4, b5, b6 and b9). Furthermore, independent studies have shown that a rabbit could express a wrong allotype. To assess the hypothesis that b allotypes are encoded by duplicated genes with a polymorphic control mechanism, we have analyzed the DNAs of four different homozygous rabbits using the Southern blot hybridization technique, with a cloned b4 C kappa probe. DNAs of individual b4, b5, b6 and b9 rabbits were cleaved with Eco RI, Kpn I and Pst I restriction endonucleases. Comparative analysis of restriction patterns shows that all four rabbit DNAs contain multiple DNA fragments hybridizing with the probe under low- and high-stringency washing conditions. In addition, each restriction pattern is distinct, suggesting that the C kappa genes are organized differently in animals expressing different allotypes.     

Genomic DNA encoding rabbit T cell receptor beta-chains: isotypes and allotypes of C beta

We have discovered sequence differences in DNA encoding the first exon of rabbit T cell receptor beta-chains from unrelated rabbits that probably reflect allelic C beta 1 allotypes. Rabbit I was from a colony bred to maintain the K1-expression mutation Basilea, and rabbit II was from a colony bred to maintain the K1b9 allotype. Genomic DNA from rabbits I and II also exhibit restriction fragment length polymorphism of C beta on Southern blots. In addition, several different restriction enzyme digests of DNA from rabbit I give three bands, whereas DNA from rabbit II gives two when probed with C beta. An approximately 14-kb cloned genomic DNA fragment from rabbit I has two copies of C beta exon 1 and a 6-kb fragment has a third copy, suggesting that rabbit I has three different C beta genes. The DNA sequence of a germ-line genomic DNA fragment encoding the first exon of the beta-chain constant region from rabbit I also has an open reading frame encoding 140 amino acids immediately 5' of the C beta sequence. A corresponding sequence had previously been found in a cDNA clone from the second rabbit (rabbit II).     

Evolution of the rabbit immunoglobulin κ chain genes

We have analyzed the organization and the structure of rabbit chain genes encoding b allotypes in wild rabbits. The 1 gene of the b95 allotype was cloned and its structure determined. The J region is composed of five segments but only J2 appears to be functional and is identical to the J2 segment of the b4 allotype. The J region is highly conserved among the various b allotypes, whereas the constant region exon displays a high level of differences when compared with other allotypes (9%–30% of different amino acids). The b95 J region is closer to that of b4^var and the constant region to b5 allotype constant region. Alignment of nucleotide sequences revealed that the constant region exon displays segmental similarities with b4 and bas constant regions. The mosaic structure of b95 allotype gene indicates that complex allotypes of 1 genes may result from genetic exchanges of gene conversion between the different genes.The nucleotide sequence data reported in this paper have been submitted to the GenBank nucleotide database and have been assigned the accession number M22542. Address correspondence and offprint requests to: P.-A. Cazenave.     

Germline VH genes in an a3 rabbit not typical of any one VHa allotype

We have undertaken investigations in the rabbit of VH genes that may be responsible for the observations of VHa allotypes unexpected from an animal's pedigree (latent allotypes). A short cDNA probe was prepared and shown to be specific for VHa2 mRNA. Southern analyses with short and large probes were unrevealing but screening of a lambda phage library from a VHa3-expressing animal identified a number of unusual genes. These VH genes are remarkable in that they are far closer to one another in the genome (in one case 3085 bps) than VH genes reported in mouse or man, they are highly homologous over long stretches of sequence, and they encode proteins not typical of any one VHa allotype. Proteins similar to the sort encoded by these genomic V-regions may explain some of the observations of latent allotype. Our data suggest that the allelic behavior of VHa allotypes is not due to allelism of a regulatory mechanism that acts upon identical VH genes in rabbits of different VHa phenotypes.     

Enduring antibody responses in "normal" rabbits to maternal immunoglobulin allotypes.

A majority of rabbits born to mothers heterozygous with respect to the b locus allotypic markers "spontaneously" develop antibodies against the noninherited maternal allotypes. Such antibodies generally appear in the 4- to 5-month-old rabbit, after what may be a transient period of tolerance, and persist at constant levels for at least 5 to 6 months. The finding of a high incidence of measurable antimaternal allotype responses in "normal" rabbits, added to similar findings by others in mice, pigs, and man, underscores the generality of this rather unexpected phenomenon and invites further inquiries into biological significance.     

Nucleotide sequences of heat shock activated genes in Drosophila melanogaster. I. Sequences in the regions of the 5' and 3' ends of the hsp 70 gene in the hybrid plasmid 56H8.

We present the sequences at the 5' and 3' ends of one hsp 70 gene variant which is derived from the chromosomal locus 87A7. The 5' end of the hsp 70 mRNA has also been determined. 550 bp upstream from the 5' end of the hsp 70 mRNA, there is a very A+T rich region shown by heteroduplex analysis to be also present at the same position in other hsp 70 genes9. The 5' end of the hsp 70 mRNA was found 26 bp after a characteristic "Hogness box". The first ATG codon was found 250 bp downstream from the 5' end of the hsp 70 mRNA. We also determined the termination codon at the 3' end of the hsp 70 gene. Comparisons with other genes are discussed.     

In vivo activation of quiescent B cells by anti-immunoglobulin. I. Induction of latent VH allotype production in adult rabbits by treatment with heterologous antibodies or antibody fragments

We demonstrated previously that injection of adult rabbits with homologous anti-VHa1 allotype antibody induces the appearance in serum of genetically unexpected (latent) a1 immunoglobulin (Ig) and a1-like internal images. We have now investigated the mechanism for this induction effect and have found that treatment of animals with polyclonal goat or monoclonal mouse anti-a1 antibody similarly results in production of unexpected a1 determinants. The N-terminal amino acid sequences of deblocked heavy chains showed that latent a1 Ig induced by monoclonal antibody treatment was identical to nominal a1 Ig at 18 of 19 residues, with the one amino acid difference at position 13 probably due to a single nucleotide change. Like nominal a1, these molecules expressed multiple VHa1 framework epitopes, as demonstrated by direct immunoelectron microscopic visualization of immune complexes. Whereas F(ab)'2 fragments of rabbit anti-a1 antibody retained inducing activity, F(ab) fragments were effective only when administered in an aggregated form; this result suggests that cross-linking of surface Ig receptors plays a critical role in the induction process. We conclude that all rabbits contain dormant B cells expressing germ-line-encoded latent allotypes, and that in vivo administration of anti-Ig antibody causes activation of these cells directly rather than through perturbation of an allotype regulatory network.     

Multiplicity of constant kappa light chain genes in the rabbit genome: a b4b4 homozygous rabbit contains a kappa-bas gene

We have constructed a genomic library of homozygous b4b4 rabbit DNA in the pJB8 cosmid vector. Clones containing Ckappa-like sequences were screened with a b4 cDNA probe and were characterized by restriction mapping. One of the clones contained a Ckappa sequence different from the b4 allotype normally expressed by the animal. We report here the nucleotide sequence of this gene and show that it probably corresponds to a kappa-bas form of the Basilea allotype. It appears to be a structurally complete gene without any stop codons within the coding region and containing the dinucleotide AG as a splice site acceptor for the J-C junction, just 5' of the coding block. Comparison with the b4 cDNA nucleotide sequence shows a separate evolution of the Ckappa-coding and 3'-untranslated sequences, since the 3'-untranslated regions are more conserved than the coding regions. Genomic blot analysis would suggest that the kappa-bas gene is isotypic in the domestic rabbit population, since it lies within a genomic EcoRI or PstI restriction fragment, which was shown to be common to all homozygous b4, b5, b6 and b9 rabbit DNAs.     

Structural analysis of the three vitellogenin genes in Drosophila melanogaster.

Genomic fragments coding for sequences expressed as abundant mRNA in female Drosophila melanogaster were isolated from a lambda library. Hybridization of these clones to polytene chromosomes. in situ, identified four which mapped to X chromosomal region 9A to 9B, the locus for yolk proteins 1 and 2 (Ypl,2) and two which mapped to 12A6-7 to 12D3, the locus for Yp3. These clones were mapped with restriction enzymes, and the coding regions and regions of homology determined by Southern blots probed with cDNA, 5'-end-labelled RNA and nick-translated DNA. Heteroduplex and R-loop mapping confirmed that three of the clones carried two genes separated by about 1.4 kb and oriented in opposite directions. Southern blots probed with cDNA made from alkali-hydrolyzed RNA showed that these genes had their 5' ends next to each other. All 3 genes show homology to each other and have a main coding region of about 1.3 kb, the approximate size for the mRNAs.     

Human growth hormone DNA sequence and mRNA structure: possible alternative splicing.

We have determined the complete sequence of the human growth hormone (hGH) gene and the position of the mature 5' end of the hGH mRNA within the sequence. Comparison of this sequence with that of a cloned hGH cDNA shows that the gene is interrupted by four intervening sequences. S1 mapping shows that one of these intervening sequences has two different 3' splice sites. These alternate splicing pathways generate hGH peptides of different sizes which are found in normal pituitaries. Comparison of sequences near the 5' end of the hGH mRNA with a similar region of the alpha subunit of the human glycoprotein hormones reveals an unexpected region of homology between these otherwise unrelated peptide hormones.     

Structure of a cluster of mouse histone genes.

The four mouse histone genes (2 H3 genes, an H2b gene and an H2a gene) present in a cloned 12.9 kilobase fragment of DNA have been completely sequenced including both 5' and 3' flanking regions. These genes are expressed in cultured mouse cells and the 3' and 5' ends of the mRNA have been determined by S1 nuclease mapping. These genes code for a minor fraction of the histone mRNAs expressed in cultured mouse cells. They comprise at most 5-8% of the total histone mRNA of each type. The two H3 genes code for H3.2 and H3.1 histone proteins, while the H2b gene codes for an H2b.1 protein with a single amino acid change (val-leu) at position 18. Only the 3' portion of the H2a gene is contained in the clone and there is an amino acid change (alanine-proline) at position 126. Comparison of the 5' and 3' flanking sequences reveals a conserved sequence at the 3' end of the mRNA which forms a hairpin loop structure. The codon usage in the genes is non-random and there has been no discrimination against CG doublets in the coding region of the genes.