Allelic exclusion of surface IgM allotypes on spleen and bursal B cells in the chicken

Influence of antibody diversification on the mechanism of haplotype exclusion of immunoglobulin gene expression

Allelic, or haplotype, exclusion of immunoglobulin gene expression ensures that the products of a single allele or light chain isotype are expressed on the B cell surface. Evidence has accumulated in rodent and primate models to indicate that the products of successful rearrangement regulate this process. In contrast, haplotype exclusion of chicken immunoglobulin gene expression is regulated at the level of variable region gene rearrangement. We discuss here alternative models for ensuring haplotype exclusion that may operate in the chicken and extend the discussion to address the issue as to how two apparently distinct mechanisms may have evolved to yield the same outcome.

Selective integration of avian leukosis virus in different hematopoietic tissues

Hematopoietic tissues obtained from avian leukosis virus (ALV)-infected Hyline SC chickens were analyzed for the presence of integrated viral DNA sequences. Cells were prepared from bone marrow, bursa, spleen, thymus, and peripheral blood. Following the removal of erythrocytes, cellular DNAs from each of these tissues were examined by Southern analysis. During the first few weeks of infection, DNA from the bone marrow contained as many as 0.5 copies of viral DNA per haploid genome. Cells from the bursa and peripheral blood contained between 0.05 and 0.15 copies per haploid genome. In contrast, neither splenic nor thymic DNA contained significant levels of viral DNA sequences even though infected birds developed titers of circulating virus between 10(5) and 10(6) IU/ml of plasma. DNA prepared from erythrocytes isolated from the peripheral blood of these birds contained approximately 0.4 copies of integrated viral sequences per haploid genome at 2 weeks after infection. Despite greater levels of integrated sequences in other tissues, by 9 weeks after infection viral sequences were detected only in DNA from bursal lymphocytes. Cells prepared from the spleen and thymus of infected birds were also examined for their size distribution, their internal complexity and their surface expression of immunoglobulin. None of the populations examined differed from normal, uninfected control preparations. These results suggest that ALV infection occurs primarily in the bone marrow and that the different tissues of the hematopoietic system are selectively infected. Further, these results indicate that ALV infection persists longer in bursal lymphocytes than in other hematopoietic tissues. Previous studies have demonstrated that the lymphoid tumors that develop in white leghorn chickens following ALV infection are bursal-dependent B-cell lymphomas that express immunoglobulin M. The observations presented in this communication offer, in part, an explanation for the restricted phenotype of the lymphoid tumor that develops in the SC chicken. Further, the data suggest an explanation for the bursal-dependent nature of the ALV-induced lymphoma.

Avian B cell precursors: surface immunoglobulin expression is an early, possibly bursa-independent event

The avian bursa of Fabricius contains about 1 X 10(4) discrete follicles, each of which is colonized by a small number of lymphoid progenitor cells during embryonic life. We have previously shown (J.R.L. Pink et al., Eur. J. Immunol. 1985. 15:617) that all, or almost all B cell progenitors in the bursae of 4-day-old chicks express cell surface IgM. In this report, we have analyzed the distribution of cell surface (s)IgM-1 allotypes within individual follicles of (M-1a/M-1b) allotype heterozygous birds. Although the majority of follicles contained a mixture of sIgM-1a+ and sIgM-1b+ cells, a significant proportion of isolated follicles contained exclusively sIgM-1a+ or sIgM-1b+ cells. Statistical analysis of the frequency of such "M-1a" and "M-1b" follicles demonstrated that the sIg+ B cells in the bursae of 4-8-week-old birds are derived from 2-4 allotypically committed precursor cells per follicle. Since we have previously shown that each bursal follicle is colonized by 2-5 pre-bursal stem cells, these cells must be committed to the eventual expression of one or other allotypic haplotype before they have undergone extensive proliferation within the bursa. In addition, we show that almost all B progenitor cells from the bursae of chicks which had been allotype suppressed as embryos were committed to synthesis of the nonsuppressed allotype, showing that this commitment was essentially complete at the time of suppression (i.e. before 19 days of incubation). Finally the bone marrow of 16-day embryos was used to reconstitute the bursal lymphocytes of cyclophosphamide-treated host embryos. Reconstitution was inhibited by anti-Ig antiserum indicating that most 16-day embryonic BM-derived bursal cell precursors also express sIgM. These results raise the possibility that expression of sIgM may be controlled by a "biological clock" rather than by any inductive capacity of the bursal microenvironment. Furthermore, these results provide further evidence that in normal birds a self-renewing sIg+ B cell population in the hatched chicken is the sole source of B cells in the adult.

Immunoglobulin-bearing stem cells for clones of B (bursa-derived) lymphocytes

Treatment of neonatal chickens with cyclophosphamide depletes bursal lymphocytes while maintaining the bursal epithelium intact. The bursae of normal young chickens contain "bursal stem cells" which can reconstitute the lymphoid compartment in the bursa of the cyclophosphamide-treated recipient. Using bursal stem cells from IgM allotype-heterozygous donors we show that most bursal follicles in the reconstituted host are colonized by single stem cells which are committed to the expression of one or other IgM allotype. In addition we show that the reconstituting bursal stem cells express allelically excluded surface IgM at the time of transfer. Our results suggest that B lymphocyte numbers in hatched chickens are maintained by self-renewal of committed precursors rather than by de novo production from multipotential stem cells.

Allotype suppression in the chicken. V. Abnormal isotype ratios of chronically suppressed IgM and IgG allotypes

Chronic allotype suppression was generated in M-1 (C mu), G-1 (C gamma) heterozygous B14 line chickens either by embryonal injection or by maternal transfer of anti-M-1b allotype antibodies. Using a sensitive radioimmunoassay to detect the suppressed IgM-1b, a striking disparity was observed between the degree of suppression of the IgM-1b allotype and that of the genetically linked IgG-1i allotype. The amount of suppressed IgM-1b ranged between 0.2 and 3% of total serum IgM in chronically suppressed birds. The levels of the genetically linked IgG-1i, however, comprised 5 to 25% of total serum IgG in such birds. The allotypes measured in suppressed chickens were qualitatively identical to those in normal, nonsuppressed birds by serological criteria. These results are discussed within the context of isotype regulation in the chicken.

Immunoglobulin expression in diploid and polyploid interspecies hybrid of Xenopus: evidence for allelic exclusion

The expression of species-specific surface IgM was studied in various Xenopus species hybrids. Whether the hybrids are diploid or polyploid, a lymphocyte does not produce all of the immunoglobulins (Ig) encloded by the Ig genes present in its genome. This suggests that the selection pressure to make a lymphocyte synthesize only one antibody is high and that allelic exclusion already existed in the ancestors of amphibians which appeared 300 million years ago. The apparent lack of cells producing multiple Ig in polyploid hybrids having more than one pair of functional Ig genes suggests that if a stochastic model for allelic exclusion is correct, the frequency of multiple successful rearrangements has to be very low, resulting in a huge waste of lymphocyte precursors. Given the particularity of frog development it is argued that this is not likely to be the case.

Allotype suppression in the chicken. II. Suppression in homozygous chickens with antiallotype antibody and allotype-disparate B cells

Injection of heterozygous (M-1a/M-1b, G1g/G-1i) B 14-line chickens with antisera directed against either IgM-1a or IgM-1b induced suppression of the relevant IgM-1 and genetically linked IgG-1 allotypes, whereas a mixture of anti-M-1a and anti-M-1b antibodies failed to produce allotype suppression. Injection of anti-M-1 antiserum into M-1 homozygous chickens induced only a transient delay of a few days in the appearance and rise of serum IgM-1 levels. However, suppression of host allotypes was induced by injecting M-1, G-1 homozygous neonatal or embryonal recipients with anti-M-1 antisera together with B locus- histocompatible allotype-disparate spleen, bone marrow or bursal cells. The active cell type were donor B cells, which established chimerism in the injected host, whereas peripheral blood T lymphocytes from agammaglobulinemic donors were ineffective. Allotype suppression was attributed to a homeostatic control mechanism which is exerted by normal B cells (but not T cells) over B cell recruitment in anti-M-1 antibody-treated, immature hosts.

Allotype suppression in the chicken. IV. Deletion of B cells and lack of suppressor cells during chronic suppression

Embryonally induced allotype suppression in M-1,G-1 heterozygous chickens was stable for at least 18 months after hatching. Suppression was established rapidly since injection of antigen only 4 days after anti-IgM-1 antiserum failed to abrogate its effect. Injected chickens had undetectable serum levels (i.e. less than 40 micrograms/ml) of the suppressed IgM-1 and low (0.3-0.6 mg/ml) levels of the linked IgG-1 allotype. This correlated with a complete depletion of cells bearing the relevant IgM-1 allotypes and a compensatory increase in the alternative nonsuppressed IgM-1 allotype-bearing cells in the spleen, peripheral blood and bursa. Cell transfer studies suggested that suppression could not be attributed to allotype-specific suppressor cells.

Allotype suppression in the chicken. III. Analysis of the recovery form suppression by neonatally injected or maternal antibodies

Injection of M-1 (Cmu), G-1 (Cgamma) heterozygous chickens on the day of hatch with anti-IgM-1 antiserum induced allotype suppression from which chickens recovered over a period of approximately 4 months. The suppression of the serum IgM-1 levels was matched by a decrease in the number of splenic and peripheral blood B cells bearing the relevant IgM-1 allotype, and a compensatory increase in the number of cells bearing the alternative nonsuppressed IgM-1 allotype. However, the proportion of IgM-1 bearing bursal cells was only marginally altered. The recovery from suppression was due to B cell recruitment and could be abrogated by bursectomy. Allotype suppression induced in ovo or maintained by repeated injection of anti-IgM-1 anti-serum resulted in chronic suppression and depletion of the relevant peripheral as well as bursal IgM-1-bearing cells. Antibody titers of the relevant allotype in partially suppressed chickens generally correlated with serum allotype levels without clonal restriction in antibody response of the suppressed allotype.