The Lymphopoiesis Unit studied major events implicated in the establishment of a functional immune system:


Generation of hematopoietic stem cells, in the mouse embryo.


Hematopoietic stem cells (HSC) guaranty the production of blood cells throughout life, first in the fetal liver and in the bone marrow, after birth. HSC are generated during a short period on embryonic life, in the dorsal aorta. Lympho-myeloid hematopoietic progenitors are detected in the dorsal aorta starting at E9 and we showed that these progenitors are endowed with long-term reconstitution capacity, but only engraft natural killer (NK)-deficient Rag2gc–/– mice. This novel population, called immature HSCs, evolves in culture into HSCs, defined by acquisition of CD45 and MHC-1 expression and by the capacity to reconstitute NK-competent mice. This evolution occurs during ontogeny, as early colonization of fetal liver by immature HSCs precedes that of HSCs. Moreover, organ culture experiments show that immature HSCs acquire, in this environment, the features of adult HSCs (Kieusseian et al Development 2012).


B cell development.


Lineage commitment is regulated during hematopoiesis, with stepwise loss of differentiation potential ultimately resulting in lineage commitment. We described a novel population of B/NK bipotent precursors among common lymphoid progenitors in the fetal liver and the bone marrow. The absence of T cell precursor potential is due to low Notch1 expression and secondary to inhibition of E2A activity by members of the inhibitor of DNAbinding (Id) protein family. Our results demonstrated a new, Id protein-dependent, molecular mechanism of Notch1 repression, operative in both fetal and adult common lymphoid progenitors, where T cell potential is selectively inhibited without affecting either the B or NK programs. We thus identified Id proteins as negative regulators of T cell specification, before B and NK commitment (Pereira de Sousa et al J. Immunol. 2012).

TLR9 is expressed in cells of the innate immune system, as well as in B lymphocytes and their progenitors. We investigated the effect of the TLR9 ligand CpG DNA on the proliferation of pro-B cells. CpG DNA inhibits the proliferation of pro-B, but not pre-B, cells by inducing caspase-independent cell death through a pathway that requires the expression of cathepsin B. This pathway is operative in Rag-deficient mice carrying an immunoglobulin transgene, in which B lymphopoiesis is compromised, to reduce the size of the B lymphocyte precursor compartments in the bone marrow. Thus, TLR9 signals can regulate B lymphopoiesis in vivo (Lalanne et al J. Immunol. 2010).


T cell development.


The generation of T cells depends on the migration of hematopoietic progenitor cells to the thymus throughout life. The identity of the thymus-settling progenitor cells has been a matter of considerable debate. We found that thymopoiesis is initiated by a first wave of T cell lineage–restricted progenitor cells with limited capacity for expansion but accelerated differentiation into mature T cells. They gave rise to ab and gd T cells that constituted Vg3+ dendritic epithelial T cells. Less-differentiated progenitor cells that retained the potential to develop into B cells and myeloid cells subsequently maintained thymopoiesis. In that second wave, which started before birth, progenitor cells had high proliferative capacity but delayed differentiation capacity and no longer gave rise to embryonic gd T cells. Our work reconciles conflicting hypotheses on the nature of thymus-settling progenitor cells (Ramond et al Nat. Immunol. 2014).

To define how T cells engage into the gd or ab T cell lineages we analyzed the differentiation potential of single thymocytes from wild-type and TCRgd-transgenic mice at two sequential early developmental stages. Double-negative (DN) 3 progenitors from both wild-type and transgenic mice retain the capacity to engage into both pathways, indicating that full commitment is only completed after this stage. DN2 and DN3 progenitors from TCRgd transgenic mice have strong biases for opposite fates, indicating that developmentally regulated changes, other than the production of a functional TCR, altered their likelihood to become a gd or an ab T cell. Thus, unlike the differentiation in other hematopoietic lineages, T cell progenitors do not restrict, but rather switch their differentiation potential as they develop (Pereira et al J. Immunol. 2013)


Innate lymphoid cell development


The transcription factor RORgt is required for the development of several innate lymphoid populations, such as lymphoid tissue–inducer cells (LTi cells) and cells that secrete interleukin 17 (IL-17) or IL-22. The progenitor cells as well as the developmental stages that lead to the emergence of RORgt+ innate lymphoid cells (ILCs) were not well defined. We identify the chemokine receptor CXCR6 as an additional marker of the development of ILCs and show that common lymphoid progenitors lost B cell and T cell potential as they successively acquired expression of the integrin a4b7 and CXCR6. Whereas fetal RORgt+ cells matured in the fetal liver environment, adult bone marrow–derived RORgt+ ILCs matured outside the bone marrow, in a Notch2-dependent manner. Therefore, fetal and adult environments influence the differentiation of RORgt+ cells differently (Possot et al Nat. Immunol. 2011)

The transcriptional regulator promyelocytic leukemia zinc finger (PLZF) is highly expressed during the differentiation of natural killer T (NKT) cells and is essential for the acquisition of their effector/memory innate-like phenotype. We defined two subsets that differ phenotypically and functionally: a PLZF+NK1.1 subset composed of mostly quiescent cells that secrete more IL-4 than IFN-γ upon activation and a PLZF+/−NK1.1+ subset that expresses CD127, NK1.1, and other NK-cell markers, secrete more IFN-γ than IL-4 upon activation and contains a sizable fraction of dividing cells. The size of the NK1.1+population is very tightly regulated and NK1.1+αβ and γδ thymocytes compete for a thymic niche (Pereira and Boucontet, Eur. J. Immunol. 2012)  

A large fraction of innate NKTgd T cells uses TCRs composed of a semi-invariant Vd6.3/6.4-Dd2-Jd1 chain together with more diverse Vg1-Jg4 chains. To address the role of gdTCR specificity in their generation, we analyzed their development in mice transgenic (T) for a Vg1-Jg4 chain frequently expressed by NKTgd cells (Tg-g) and in mice Tg for the same Vg1-Jg4 chain together with a Vd6BDd2Jd1 chain not usually found among NKTgd cells (Tg-gd). Surprisingly, both promyelocytic leukemia zinc finger (PLZF)+ and NK1.1+ NKTgd cells were found in the thymus of Tg-gd albeit at lower numbers than in Tg-g mice, and virtually all of them expressed the Tg TCR. However, the PLZF+ subset, but not the NK1.1+ subset, also expressed an endogenous Vd6.3/6.4 chain, and its size was severely reduced in TCRd2/2 Tg-gd mice. These results suggest that the PLZF+ and the NK1.1+ subsets are developmentally unrelated. However, PLZF+ and NK1.1+ NKTgd cells express identical Vd6.3/6.4 chains, and NK1.1+ cells can be obtained upon intrathymic injection of sorted PLZF+ cells, thus indicating their developmental relationship.

The NK1.1+ gd thymocytes present in Tg-gd mice correspond to a small subset of NK1.1+ gd thymocytes in wild-type animals, which express a more diverse repertoire of TCRs and can be recognized by the expression of the CD62L. Collectively, the data demonstrates that TCR specificity is essential for the development of most NKTgd T cells and revealed a developmental heterogeneity in gd T cells expressing the NK1.1 marker (Pereira et al. J. Immunol. 2013).




Updated on 11/03/2014


Lymphopoiesis Unit
Institut Pasteur
25 Rue du Docteur Roux
75724 Paris Cedex 15 FRANCE

Tel 33 1 45 68 82 55
Secretary 33 1 45 68 85 93
Fax 33 1 45 68 89 21



Access to all events