Edward K. Wakeland

Heterozygosity of h-2 loci in wild mice.

The major histocompatibility complex (MHC) is a cluster of genetic loci coding for cell surface molecules that seem to be the second most important element of the vertebrate immune system after immunoglobulins1. The MHC loci can be divided into three classes. Class I loci code for molecules of about 44,000 molecular weight (MW) which seem to function as markers of self in T-lymphocyte-mediated cytotoxicity. Class II loci code for molecules consisting of two polypeptide chains, of MW 34,000 and 28,000 respectively, which are believed to regulate the immune response by controlling the behaviour of helper and suppressor T lymphocytes. Each molecule may carry several antigens—one ‘private’ (characteristic for the allele controlling this molecule) and several ‘public’ ones2 (shared with molecules controlled by other alleles). Class III loci code for serum proteins. One of the most characteristic and remarkable properties of the MHC loci, in particular class I, is their genetic polymorphism, defined as the existence in a population of two or more alleles at appreciable frequencies3–5. In H–2, the MHC of the mouse, individual alleles at class I K and D loci occur in the two populations that have been studied thus far with an average frequency of about 2% (ref. 5). We estimate, therefore, that these two populations contain some 100 alleles at each of the two loci. The polymorphism of class II locus A seems to be somewhat lower than that of the class I loci (average allelic frequency of 2–5%), while at the second class II locus (E) only five alleles have been identified5. The polymorphism implies that a high degree of heterozygosity should exist at the MHC loci in natural populations, and this has been found at the HLA loci, the human MHC (for review see ref. 6). We report here a similarly high frequency of heterozygosity in the mouse H–2 complex in a population of mice, in spite of a social structure that would tend to reduce heterozygosity.

The population genetics of the H-2 polymorphism in European and North African populations of the house mouse ( Mus musculus L.)

Two hundred and two house mice ( Mus musculus L.) from 29 populations in Europe and North Africa were typed for 16 H-2K and 17 H-2D antigens, each antigen defining a different allele. Among the 13 best characterized populations, 1 to 4 common and 3 to 20 rare antigens were observed. However, an average of 37% of the H-2K and 39% of the H-2D antigens remain to be identified. Ninety-four percent of the 50 mice tested were heterozygous for H-2K antigens and 89% for H-2D antigens. In 4 of the 8 populations tested, the most common H-2K and H-2D antigens occurred in the same individual more often than if randomly associated. Associations between common H-2K and H-2D antigens and excess heterozygosities may be the consequence of the small size and instability of populations composed primarily of related individuals. Estimates of the genetic distances between populations revealed that Danish, Egyptian, and several of the Orkney Island populations were related. These were the only populations in which metacentric chromosomes were not found. In contrast, populations which were antigenically different were also karyotypically different, regardless of taxonomic status of allozymic similarity.

Structural comparisons of serologically identical IA- and IE-encoded antigens from inbred and wild mice

The IA and IE products of B10.S(9R), B10.A, B10.KPB128, and B10.GAA37 were analyzed for primary structural variations by comparative tryptic peptide mapping. The Aα,Aβ, andEβ products of B10.S(9R) and B10.A differed in about 40% of their acid-soluble tryptic peptides, indicating that intra-I-region recombinant strain B10.S(9R) received the genes encoding Aα, Aβ, and Eβ from theH- 2s parental chromosome rather than fromH- 2a. The tryptic peptides of Eα chains from B10.S(9R) and B10.A were indistinguishable, suggesting that B10.S(9R) received the gene encoding the Eα chain from theH- 2a parental chromosome. Consistent with the results of others, these data suggest that the genes encodingAα,Aβ and Eβ chains are centromeric to theIJ subregion, while the gene encoding Eβ chains is telomeric toIJ. The I-region products of two congenic lines carrying wild-derivedH- 2 haplotypes on a C57BL/10 background, designated B10.KPB128 and B10.GAA37, are serologically indistinguishable from those of B10.S(9R). The IA and IE products of B10.S(9R) were compared with those of B10.GAA37 and B10.KPB128 to determine the structural similarity of serologically identical products from allopatric populations of wild mice. The Aα,Aβ, and Eβ products of B10.S(9R) were indistinguishable from those of B10.GAA37 and B10.KPB128 by comparative tryptic peptide mapping. The Eα chains of these three lines differed in one or two of their acid-soluble tryptic peptides. The results indicate that the IA-encoded products of these three lines are structurally very similar and may be identical suggesting that some alleles of the Aα, Aβ, and Eβ chains may be maintained in stable linkage associations in allopatric populations of wild mice. The minor structural variations detected in the Eα chains of these three congenic lines indicate that the Eα chain is encoded by chromosome 17 and suggest that allelic Eα chains exhibit considerably less structural variability than other I-region encoded antigens.

Genetic nomenclature for chicken immunoglobulin allotypes: An extensive survey of inbred lines and antisera

Based on a survey of 36 inbred and 8 partially inbred chicken lines and outbred jungle fowl, and with 29 alloantisera generated in different laboratories, 13 7S Ig and 5 IgM allotypes were designated and a new system of nomenclature for chicken Ig polymorphisms was developed. The survey also revealed considerable genetic polymorphism in the structural gene(s) (G-1) responsible for the production of 7S Ig H chains. IgM H chains, encoded by theM-1 locus were less polymorphic. NineG-1 and fourM-1 gene alleles were delineated in highly inbred lines by the formation of unique combinations ofG-1 orM-1 specificities. Five additionalG-1 alleles were found in chicken lines and jungle fowl segregating for allotypes. Thirty-three percent of theG-1∶M-1 haplotypes theoretically expected, were detected in inbred lines.

An H-2 haplotype possibly derived by crossing-over between the (A α A β ) duplex and the E βlocus

The B10.STA62 strain carries the H-2w27 haplotype derived from a wild mouse captured in the vicinity of Ann Arbor, Michigan. Products of two class II loci composing this haplotype, Aα and Aβ, are serologically, biochemically (by tryptic peptide mapping), and functionally indistinguishable from products controlled by the Aαband Aβ/bgenes of the B10.A(5R) strain. In contrast, the polypeptide chain controlled by the third class II locus, Eβ, is different from that controlled by the Eβ/bgene. This Eβ/w27chain lacks an antigenic determinant present on the Eb molecule and carries determinants lacking on the Eb molecule, the Eβ/band Eβ/w27peptide maps differ in at least six peptides, and cytotoxic T cells specific for the Eβbchains do not react with B10.STA62 target cells. This great difference between the Eβ/band Eβ/w27chains suggests that the corresponding genes have not been derived from one another by a direct mutational conversion; instead, H-2w27 appears to be a recombinant haplotype derived by crossing-over between the AαAβduplex and the Eβlocus. This is the first recombinant discovered separating these class II loci.

The Polymorphism of I-Region-Encoded Antigens among Wild Mice

Genes in the major histocompatibility complex (MHC) have been shown to control the expression of several cell-surface antigens that are involved in the differentiation and activation of cells of the immune system. These antigens can be divided into two classes on the basis of functional and structural criteria (Klein, 1977). The class I antigens are the targets of the transplantation reaction, are involved in the recognition of virus-infected cells, have a molecular weight of about 44,000 daltons, and are nonconva-lently associated with β2-microglobulin. The class II antigens are involved in the regulation of the antibody response to soluble antigens and mediate many of the interactions of immunocompetent cells. Those class II antigens that have been structurally characterized consist of two polypeptide chains and are not associated with β2-microglobulin.

Minor structural variants of H-2K-controlled molecules in wild mice

We extracted H 2 K alleles from wild mouse populations and compared their products, by tryptic peptide mapping, with serologically related molecules found in inbred strains. For these comparisons we cultured splenocytes with mixtures of allor l~C-labeled amino-acids arginine, lysine, leucine and tyrosine, solubilized the cells with Nonidet P-40 detergent, passed the extract over an affinity column of Lens culinaris lectin conjugated to Sepharose 4B, and eluted the glycoproteins from the column with e-methyl-D-mannoside (Hayman and Crumpton 1972). We then precipitated H-2K molecules from the glycoprotein pool by specific alloantisera and isolated them by discontinous NaDodSO4/polyacrylamide gel electrophoresis (Wakeland and Klein 1979). We mixed the samples to be compared at a 3H : 14C cpm ratio of 3:1, digested them with trypsin, loaded the peptides soluble in pyridine acetate onto a cation-exchange column containing Technicon Chromobeads type P, and eluted the peptides by using a linear pH-ionic strength gradient, counted the 3H and a 4C radioactivities of each fraction and corrected them for channel spillover. The lines B10.STC77, B10.KEA5, STU, and Wl2A carry H-2K molecules that are serologically similar to the K d molecule of the B10.D2 strain (ZaleskaRutczynska and Klein 1977, Duncan and Klein 1980). The lines B10.STC77 and B10.KEA5 carry H-2 haplotypes extracted from Michigan wild mice (ZaleskaRutczynska and Klein 1977, Duncan and Klein 1980); the W12A strain carries the H-2 haplotype of a wild mouse captured in Texas (Duncan et al. 1979), and the STU strain carries the H-2 w34 haplotype of a noninbred European albino mouse (Figueroa et al. 1982). To determine the structural similarity of the K d molecules from B10.D2 and the wild lines, we compared these molecules by tryptic peptide mapping (example shown in Fig. 1). The K molecules of B10.STC77, B10.KEA5, Wl2A, and STU differ from the K d molecule of B10.D2 in three, four, five, and six peptides, respectively (Table 1). The wild line B10.CAA2 carrying an H-2 haplotype extracted from Californian wild mice (Zaleska-Rutczynska and Klein 1977) reacts with antisera specific for the K f molecule, but does not absorb out the reactivity of this antiserum against K f of B10.M (E.K. Wakeland and J. Klein, unpublished