Viable Lin(-) CD27(+) c-kit(Hi) Sca-1(Hi) GFP(+) cells recovered from heterozygous RAG1/GFP knockin mice progressed through previously defined stages of B, T, and NK cell lineage differentiation. In contrast to the GFP(-) cohort, there was minimal myeloid or erythroid potential in cells with an active RAG1 locus. Partial overlap with TdT(+) cells suggested that distinctive early lymphocyte characteristics are not synchronously acquired. Rearrangement of Ig genes initiates before typical lymphoid lineage patterns of gene expression are established, and activation of the RAG1 locus transiently occurs in a large fraction of cells destined to become NK cells. These early lymphocyte progenitors (ELP) are distinct from stem cells, previously described prolymphocytes, or progenitors corresponding to other blood cell lineages.

Sex steroids negatively regulate B lymphopoiesis in adult mice. Paradoxically, lymphocytes arise during fetal life, when estrogen levels are high and maternal lymphopoiesis is suppressed. Here we demonstrate that embryonic B lymphopoiesis was unaffected by estrogen, but sensitive to glucocorticoids. Both fetal and adult precursors contained glucocorticoid receptor transcripts, but only adult precursors expressed estrogen receptor ␣ and ␤ together with the androgen receptor. Fetal hematopoietic cells did not efficiently acquire functional estrogen receptors after transplantation to irradiated adult mice. Sex steroid receptors were also expressed in a stage-and developmental age-dependent fashion in human precursors. A developmental switch in responsiveness of hematopoietic cells to sex steroids may be essential for formation of the immune system.

Adiponectin, an adipocyte-derived hormone, is attracting considerable interest as a potential drug for diabetes and obesity. Originally cloned from human s.c. fat, the protein is also found in bone marrow fat cells and has an inhibitory effect on adipocyte differentiation. The aim of the present study is to explore possible influences on lymphohematopoiesis. Recombinant adiponectin strongly inhibited B lymphopoiesis in long-term bone marrow cultures, but only when stromal cells were present and only when cultures were initiated with the earliest category of lymphocyte precursors. Cyclooxygenase inhibitors abrogated the response of early lymphoid progenitors to adiponectin in stromal cell-containing cultures. Furthermore, PGE2, a major product of cyclooxygenase-2 activity, had a direct inhibitory influence on purified hematopoietic cells, suggesting a possible mechanism of adiponectin action in culture. In contrast to lymphopoiesis, myelopoiesis was slightly enhanced in adiponectin-treated bone marrow cultures, and even when cultures were initiated with single lymphomyeloid progenitors. Finally, human B lymphopoiesis was also sensitive to adiponectin in stromal cell cocultures. These results suggest that adiponectin can negatively and selectively influence lymphopoiesis through induction of PG synthesis. They also indicate ways that adipocytes in bone marrow can contribute to regulation of blood cell formation.

RAG1/GFP knockin mice were exploited to isolate and characterize fetal lymphoid progenitors. CD11b and IL-7Ralpha are expressed in a developmental stage-dependent fashion, revealing how substantial numbers of early lymphoid progenitors were discarded or neglected in previous studies. The myeloerythroid potential of fetal progenitors in clonal assays declined in synchrony with activation of the RAG1 locus but was not completely extinguished. Lymphoid differentiation corresponded to patterns of gene expression previously found for adult marrow, but no fraction of fetal liver was enriched with respect to B + T progenitors. Also, unlike adults, fetal lymphoid progenitors transiently expressed endothelial cell markers. These findings help to reconcile discrepancies in previous reports and suggest that the fetal immune system arises via unique mechanisms.

No abstract

Substantial progress has been made in determining developmental relationships between lymphocyte precursors and those corresponding to other blood cell lineages. Indeed, exploitation of RAG1/GFP knock-in mice has recently made it possible to chart the entire sequence of lymphocyte differentiation events in adult bone marrow and thymus. However, the differentiation pathways proposed for fetal life are very different from this model. We review many examples where the results of gene targeting experiments are substantially dependent on developmental age. In mice, adult patterns of gene expression and corresponding properties of lymphocyte precursors are not fully established until several weeks after birth, and the same might be true for humans. Furthermore, examples are cited where fetal hematopoietic cells did not efficiently acquire those properties when transplanted to an adult environment. There are several important implications of these findings. Cognizance of developmental age-related changes might resolve apparent conflicts in the literature. Hematopoietic stem cells and their lymphoid lineage progeny appear in waves, and a direct connection is yet to be established between fetal stem cells and ones that sustain adult blood cell formation. There is the possibility that adult stem cells derive from founders with an unknown origin.

Glucocorticoids are extensively used in anti-inflammatory therapy and may contribute to the normal regulation of lymphopoiesis. This study utilized new information about the early stages of lymphopoiesis in mouse and man to determine precisely which cell types are hormone sensitive. Cycling B lineage precursors were depleted in dexamethasone-treated mice, while mature, non-dividing CD45R(Hi) CD19(Hi) lymphocytes, myeloid progenitors and stem cells with the potential for lymphocyte generation on transplantation were spared. Lineage marker-negative (Lin(-)) IL-7R(+) Flk-2(+) pro-lymphocytes also declined, but not as rapidly as the terminal deoxynucleotidyl transferase-positive cells within an early Lin(-) c-kit(Hi) Sca-1(Hi) fraction of bone marrow. Hormone-sensitive cells with additional properties of early lymphoid progenitors (ELP) were identified within the same Lin(-) c-kit(Hi) Sca-1(Hi) subset using human mu transgenic mice and recombination-activating gene 1 (RAG1)/green fluorescent protein knock-in animals. Furthermore, cells with a recent history of RAG1 expression were more glucocorticoid sensitive than mature lymphocytes in marrow and spleen. Lymphocyte progenitors in mice bearing a human bcl-2 transgene were protected from dexamethasone treatment. However, isolated progenitors from either wild-type or bcl-2 transgenic mice were directly sensitive to the hormone in stromal cell-free cultures, suggesting that additional factors must determine vulnerability to glucocorticoids. B lineage lymphocyte precursors were found to be abnormally elevated in the bone marrow of adrenalectomized or RU486-treated mice. This suggests that glucocorticoids may normally contribute to steady-state regulation of lymphopoiesis. Finally, parallel studies revealed that the earliest events in human lymphopoiesis are susceptible to injury during glucocorticoid therapy.

No abstract