Development of the immune system from stem cells

Overview

Stem cells of the adaptive and the innate immune system

All cells of the innate and adaptive immune system can originate from a single pluripotent hematopoietic cell (pHSC). pHSC are characterized by at least three properties. Firstly, upon differentiation in an appropriate environment they can give rise to cells of the erythroid, megakaryocytic, myeloid and lymphoid lineages. This capacity has been experimentally documented by in vivo transplantation into hematopoietically deficient, e.g. lethally irradiated hosts, or by in vitro differentiation under appropriate tissue culture conditions. Secondly, after division, in vivo or in vitro, pHSC are expected to retain their pluripotent stem cell property in at least one of the two daughter cells, i.e. they have the capacity of self renewal. pHSC are expected to have an extended proliferative capacity of this kind. Thirdly, upon transplantation, pHSC are able to home back to heir appropriate sites in the bone marrow, from where they can be re-isolated and replanted into secondary recipients without loss of their three stem cell properties [1-12]. If the rates of repopulations into the different hematopoietic lineages are appropriate and rapid enough, stem cells can protect lethally irradiated recipients from death.

B-Lymphocyte Development

Development of B lymphocytes from progenitor and precursor cells can be followed by the stepwise rearrangements of the V, D and J segments of the immunoglobulin (Ig) heavy (H) and light (L) chain gene loci, by differential expression of specific genes with functions in this developmental pathway, by differential growth properties of cells in this development 'in vitro', and by differential properties to populate B lymphocyte- precursor and -mature cell compartments in recipient mice upon transplantation (for reviews see [13, 14]). B cell development from early progenitors and precursors to mature, surface Ig-expressing (sIg+) B cells can be induced and followed 'in vitro' [15], as well as 'in vivo' after transplantation into severe- combined immunodeficiency (SCID) or RAG-/- mice. 'In vivo', the transplantation of proB or preB-I cells leads to a long-term population of some of the mature B cell compartments [16, 17].

Clonable, transgenetically mutable, transplantable precursor B cells

PreB-I cells from wildtype and a wide variety of mutant mice, which are DH-Jh-rearranged at both alleles of their Ig-heavy(H) chain loci, can proliferate 'in vitro' on stromal cells in the presence of interleukin (IL)-7 for very long periods of time [16]. Some of the cell lines have been carried for over 300 divisions without losing their status of differentiation as preB-I cells Since the efficiency of plating of these preB-I cells is near 100%, it is very easy to derive large numbers of individually DH/DH-rearranged, i.e. genetically marked preB cell clones, from which the differentiated progeny of cells can unambiguously identified as products of the original clone. Since these clones can also be retrovirally transfected with high efficiencies (e.g. with a gene encoding a fluorescent protein) they are suitable recipients of any transgenes with potential activities in B cell development and responsiveness. Neither the thymus nor the bone marrow of the recipient mice are populated from the wildtype-preB cell clones, nor are mature T cells or myeloid cells detectable. Hence wildtype-preB cells do not home back to their original sites in the bone morrow, and they appear committed to the development of one wave of mature, peripheral, long-lived B cells [16]. Transplanted into SCID or RAG-deficient hosts they mainly populate B1 compartments, which can respond to T cell-independent antigens. When helper T cells are co-transplanted conventional B cell compartments are also populated. The hosts become responsive also to T cell-dependent antigens in germinal centers and develop hypermutated, Ig-class-switched antibody responses.

Clonable precursors of myeloid and T lymphoid cell lineages

PAX-5-deficient mice are blocked in B cell development after the preB-I cell stage but develop all other hematopoietic cell lineages normally [18, 19]. PAX-5-deficient preB-I cells, like their wildtype counterparts, are clonable and proliferate for several hundred divisions on stromal cells in the presence of IL-7 [20]. 'In vitro' they differentiate to NK-cells and to different types of myeloid cells, and the direction of differentiation is induced by selective environmental influences of cytokines and cell-cell contacts to which they are induced after the removal of IL-7 [19, 21]. 'In vivo', after transplantation into sublethally irradiated hosts, erythrocytes, all lineages of myeloid cells, NK-cells and T cells develop with varying efficiencies [22, 23]. Thymocytes and peripheral T cells are generated at normal rates in normal numbers at normal sites. Importantly, the PAX-5-deficient preB cells home back to the bone marrow, from where they can be re-isolated and again grown on stromal cells in the presence of IL-7 - and this 'ex vivo/ in vitro' propagation can be repeated for several passages. In summary, with single, genetically DHJH/DHJH-marked and transgenically (e.g. GFP-markable) preBI cell clones from wild-type and from PAX5-/-mice, we are now able to follow homing, self renewal and differentiation into practically all hematopoietic and lymphopoietic cell lineages, in this laboratory specifically to B cells [14]. It should also allow to follow the cooperation of wildtype-derived B cells with PAX5-/--derived T cells and antigen presenting cells in transplanted hosts, and to test immune responses to a variety of experimental and naturally occurring antigens in histocompatible, wildtype, mutant or transgenically altered, reconstituted immune systems of the mouse.

Differentiation of clonable and transplantable precursors of B-, and of myeloid, NK- and T-lymphoid lineages from embryonic stem cells of the mouse

A major limitation of pluripotent hematopoietic stem cells and of clonable, transplantable precursor of B-, and of myeloid-, NK- and T-lymphoid lineages in analyses of the function of genes expressed during hematopoietic development and in immune responses of mature cells of the immune system is our apparent inability to stably transfect and homologously recombine mutated forms of genes at their proper, or at chosen sites of the mouse genome. This is in contrast to mouse embryonic stem cells, in which targeted integration of a gene on both alleles is an established method, which is used for the generation of homologously recombined, targeted mutant mouse strains. Therefore, we intend to develop efficient experimental protocols to generate clonable, transplantable pre B cell lines from wildtype and from PAX5-/- embryonic stem cells by' in vitro' differentiation. On the one side we plan to introduce regulatable forms of transgenes, which might favor autonomous development as well as easy detection of pHSC and lymphoid/myeloid progenitors [14]. On the other side recent publications have indicated that a favorable environments can be generated for hematopoietic development by certain stromal cell lines expressing ligands for the selection and preferred differentiation of hematopoietic cell development [24, 25]. If successful these tissue culture procedures should allow for a rapid screening of a much larger number of candidate genes [26] with potential functions of pHSC, of progenitors and of mature cells of the innate and adaptive immune system, as well as for a functional analysis of genes in positive and negative responses of the immune system.

Retroviral tagging of genes active in the development of hematopoietic cell lineages and immune respnoses

Using retroviral vectors as single insertions into the mouse genome to detect endogenous promotors or polyadenylation sites of actively transcribed genes we plan to identify those genes which are active in DHJH-rearranged preB-I cells from wildtype and from PAX5-/- mice. Puromycin resistance transferred by the vectors is used to select stabily transfected preB cell lines. We follow the expression of tagged, active genes by the green fluorescent protein-encoding reporter gene also transferred by the vector as wildtype preB-I cells are induced to differentiate to B-lineage lymphocytes in vitro and in vivo. We will do the same with PAX5-/- preB-I cells as they are induced to differentiate to natural killer cells and to myeloid cell lineages in vitro and in vivo, and to T-lymphoid lineage cells in vivo. We will also attempt to identify genes, which are not active in preB-I cells but are only turned on when they differentiate along these pathways of differentiation. Special attention will be given to dominant-negative insertional mutations, which might become detectable as a loss-of-function in these hematopoietic differentiation lineages. This mutational analysis is intended to complement the efforts to define the molecular programs of hematopoietic development currently undertaken by subtractive hybridizations of cDNA libraries of different cellular stages of this differentiation, by chip analyses of such cDNA libraries, and by mutational analyses with embryonic stem cells, or of Zebra fish, developing into the hematopoietic cell lineages. So far two retroviral vectors have been constructed which infect mouse preB-I cell lines particularly well. Approximately one in 2x10e4 to 1x 10e5 puromycin-resistant preB cell express GFP. The GFP-expressing cells have been cloned and were shown to have the vector integrated into specific sites of the mouse genome. Several integration sites have already been characterized. The project is being funded at the Biozentrum of the University of Basel, Department of Cell Biology, to F.M. by the Swiss National Founds and is a collaboration with Drs Ulf Grawunder and Ton Rolink at the Department of Molecular Immunology, Pharmazentrum of the University of Basel, Switzerland.

Biochemical and functional characterization of the mouse preB cell receptor

PreB cell receptors (preBcR's) are complexes of mu-heavy chains and surrogate light (SL) chain, the latter being composed of the VpreB-and lambda 5-proteins [13, 27]. Single lambda 5-deficient, double VpreB1-and VpreB2-deficient, and triple VpreB1-, VpreB2-and lambda 5-deficient mice have been generated, and all of them appear deficient in signal transduction from the preBcR, so that preBII do not expand by proliferation. However allelic exclusion of H chains expressed from the second H chain allele is not affected [28]. Recent mutational analysis of the non Ig-like, aminoterminal portion of the lambda 5 gene have shown that this part of the SL-chain is responsible for a preB cell-autonomous, constitutive signalling of the preBcR [29]. Future mutational analyses will attempt to define the role(s) of the non Ig-like, carboxyterminal portion of the VpreB-genes in preBcR signalling, and will analyze the biochemical changes of preBcR leading to the apparently pre B cell autonomous, pre BcR- ligand- independent function of this receptor.