Cell-transfer studies presented here distinguish three murine B cell lineages: conventional B cells, which develop late and are continually replenished from progenitors in adult bone marrow; Ly-l B cells (B-la), which develop early and maintain their numbers by self-replenishment; and Ly-1 B "sister" (B-lb) cells, which share many of the properties of Ly-1 B cells, including self-replenishment and feedback regulation of development but can also readily develop from progenitors in adult bone marrow. The sequential emergence of these lineages, the time at which their progenitors function during ontogeny, and the distinctions among their repertoires and functions suggest that evolution has created a layered immune system in which-the immune response potential ofeach successive lineage is adapted to its particular niche. (5) and B-cell neoplasms (6). They tend to use a restricted set of variable-region (V) genes (7-10) and to use N-region insertions less often than conventional B cells (11). These repertoire differences may arise from differences in the diversity-generating mechanisms in individual lineages (12), to the different times that lineages develop (3), and/or to selection by particular antigens (13)(14)(15)(16) (17).In this study, we characterize the progenitor capacity of FL and adult bone marrow (BM) for three kinds of mature B We show here that significant progenitor activity for B-lb cells is present in fetal and adult animals, whereas progenitor activity for B-la cells is readily detectable in FL but is largely missing or nonfunctional in adult BM. We further show (i) that the failure to detect progenitor activity for B-la cells in adult BM is not due to the presence of inhibitors or the absence of inducers that regulate B-la development, (ii) that the progenitors for conventional B cells are already distinct from progenitors for the B-1 subsets in 14-day FL, and (iii) that the distinctive development of the three B-cell populations depends on properties inherent in their progenitors and cannot be explained solely by differential selection.Taken together, these studies indicate that B-ia and B-lb cells belong to separate developmental lineages and that both lineages are distinct from the conventional B lineage. We discuss these findings in the context of our recent hypothesis that these B-cell lineages reflect the existence of an evolutionarily layered immune system in which the immune response potential of each successive lineage is adapted to particular challenges (22). MATERIALS AND METHODS

Long term B lineage chimeras are used here to study the origin of plasma cells in the mouse. Chimeric mice are constructed by reconstituting lethally irradiated mice with peritoneal cells (PerC) and bone marrow cells from congenic pairs of mice differing in Igh-C allotype. All conventional B cells in these mice express the allotype of the bone marrow donor and nearly all Ly-1 B lineage cells express the allotype of the PerC donor. FACS analysis and immunohistology of these mice shows that virtually all (sig+) B cells in peripheral lymphoid organs are derived from the bone marrow donor. However, despite this overwhelming number of bone marrow-derived B cells in these animals, immunohistological staining of lymphoid organs and gut shows that nearly half of the IgM, IgG, and IgA plasma cells derive from the PerC donor. These data demonstrate that the peritoneal cavity contains a major reservoir of self-replenishing cells that play a significant role in the mucosal immune response. The possibility that these are B cells that belong to the Ly-1 B lineage is discussed.

Despite limited data supporting the superiority of dominant over subdominant responses, immunodominant epitopes represent the preferred vaccine candidates. To address the function of subdominant responses in human immunodeficiency virus infection, we analyzed cytotoxic T lymphocyte responses restricted by HLA-B*1503, a rare allele in a cohort infected with clade B, although common in one infected with clade C. HLA-B*1503 was associated with reduced viral loads in the clade B cohort but not the clade C cohort, although both shared the immunodominant response. Clade B viral control was associated with responses to several subdominant cytotoxic T lymphocyte epitopes, whereas their clade C variants were less well recognized. These data suggest that subdominant responses can contribute to in vivo viral control and that high HLA allele frequencies may drive the elimination of subdominant yet effective epitopes from circulating viral populations.

Natural killer (NK) cells are the first lymphocytes to recover after allogeneic stem cell transplantation (SCT) and can exert powerful graft-versus-leukemia (GVL) effects determining transplant outcome. Conditions governing NK cell alloreactivity and the role of NK recovery in sibling SCT are not well defined. NK cells on day 30 post-transplant (NK30) were measured in 54 SCT recipients with leukemia and donor and recipient killer immunoglobulin-like receptor (KIR) genotype determined. In univariate analysis, donor KIR genes 2DL5A, 2DS1, 3DS1 (positive in 46%) and higher numbers of inhibitory donor KIR correlated with higher NK30 counts and were associated with improved transplant outcome. NK30 counts also correlated directly with the transplant CD34 cell dose and inversely with the CD3 þ cell dose. In multivariate analysis, the NK30 emerged as the single independent determinant of transplant outcome. Patients with NK30 4150/ll had less relapse (HR 18.3, P ¼ 0.039), acute graft-versus-host disease (HR 3.2, P ¼ 0.03), non-relapse mortality (HR 10.7, P ¼ 0.028) and improved survival (HR 11.4, P ¼ 0.03). Results suggest that T cell-depleted SCT might be improved and the GVL effect enhanced by selecting donors with favorable KIR genotype, and by optimizing CD34 and CD3 doses.

Studies presented here, conducted with allotype homozygotes, demonstrate the existence of a feedback mechanism that regulates development of Ly-1 B cells from immature progenitors. In the preceding study (P. A. Lalor et al., Eur. J. Immunol. 1989. 19:501), conducted with allotype heterozygotes, we showed that treating neonates with monoclonal antibody to the paternal allotype IgM depletes roughly half of the neonatal B cell population (i.e. those expressing the paternal IgM allotype) and that paternal allotype Ly-1 B cells specificically remain depleted for the life of the animal. Here we show that treating allotype homozygotes with the same antibody depletes all (rather than half) of the B cells and that, under these conditions, relatively normal numbers of Ly-1 B cells reappear shortly after the treatment antibody disappears. The recovery, we also show, is prevented by restoring allotype-congenic Ly-1 B cells to the treated homozygotes, i.e. by reconstituting treated neonates with allotype-congenic peritoneal cells, sorted Ly-1 B cells or a monoclonal population of Ly-1 B "tumor" cells. These findings in essence reveal a feedback mechanism through which mature Ly-1 B cells prevent further Ly-1 B cell development from Ig- precursors. This feedback regulation is independent of Ig secretion by the mature Ly-1 B cells, since the monoclonal Ly-1 B "tumor" population that prevents endogenous Ly-1 B development does not secrete Ig. Furthermore, it appears to be independent of Ly-1 B surface Ig specificity, since a monoclonal population is sufficient to block all Ly-1 B cell development. This mechanism appears to operate normally to fix the composition of the Ly-1 B population, which survives through self-replenishment in adults, in accord with conditions that influence Ly-1 B development during neonatal life.