Regulators That Control Genes Of The Vertebrate Immune Systems

The adaptive immune system, with its B- and T-lymphocytes, antigen-dependent clonal selection mechanisms, and antigen-specific immunologic memory, is found only in vertebrates within the chordate assemblage of deuterostomes (56-60). Supplementing the innate immune system, this results in a layered system of innate and adaptive responses, interconnected and cemented by many potent regulatory links between the two kinds of response (61). In evolutionary terms, lymphocytes presumably arose by the modification of developmental pathways, leading to more basal hematopoietic cell types. In this section, we consider the relationships between the innovations required for the adaptive immune system and the underlying regulatory structures of deuterostome innate immunity.

4.1. Immune System Innovations of Jawed Vertebrates

Two innovations are conspicuous in this comparison. One, of course, is the elaborate recombinase system, which rearranges antigen receptor gene segments in vertebrates to assemble the transcription units that encode immunoglobulins and T-cell receptors. This first innovation is made possible by the presence of the RAG-1 and RAG-2 genes in the genome and their regulated expression in the differentiation of specialized immune cells. The other innovation is the development of the distinctive lymphocyte cell type. Lymphocytes are not only distinguished by their use of the RAG-mediated DNA rearrangement process, but also by their long lives and antigen-dependent control of proliferation and death. They combine sophisticated, differentiated effector functions with a self-renewal potential that is second only to that of stem cells within the hematopoietic series. The origins of adaptive immunity in a deuterostome ancestor depend on the origins of the RAG genes and the origins of the lymphocyte cell type.

4.2. Potential Evolutionary Bridges Between Innate and Adaptive System Components

RAG gene homologs are not evident in any animals thus far studied except jawed vertebrates (57-62). Sequence data for agnathans and invertebrate deuterostomes are admittedly scarce, and there might be RAG homologs waiting to be discovered in such animals. However, there is also no trace of RAG homologs in protostomes, giving reason to suspect that the discontinuity is real. The RAG-1 and RAG-2 genes could have entered the genomes of vertebrate ancestors only after the split between agnathans and gnathostomes, perhaps by horizontal transfer of a transposable element with its trans-posase gene from a prokaryotic source (63). A true discontinuity of this kind would give little indication of what characteristics to look for in cells of agnathans and invertebrates that might be functional counterparts of the ancestors of vertebrate lymphocytes.

In contrast, the differentiation pathways of lymphocytes do suggest traces of a continuum with other hematopoietic lineages. Evidence for such a continuum emerges from the detailed studies that have been performed in recent years on mammalian lymphocyte precursors. Lymphocytes differentiate continuously from the same hematopoi-etic stem cells that give rise to phagocytic and nonimmune blood cell types. Intermediates in this process retain certain subsets of developmental potentials after having lost others. It has long been assumed that lymphocyte development branches off from the development of all erythroid and myeloid blood cell types at an early stage. There are descriptions of a "common lymphoid progenitor," with the ability to give rise to all classes of lymphocytes but not to other hematopoietic cell types (64,65), and of a complementary "common myeloid progenitor" with the opposite potentials (66). The existence of such partially restricted precursors has been taken to suggest a profound early split. However, it appears that a common lymphoid precursor is not the only intermediate through which lymphocytes can be made.

In fetal life, especially, mouse T- and B-cells arise from precursors that may also give rise to macrophages, although not to other blood cell types (67,68). Many fetal T-cell precursors can still give rise to mixed T/myeloid clones even though they cannot give rise to B-cells, and the reverse is true for B-cell precursors (69). There are numerous cases of cell lines that appear to be immortalized B/macrophage precursors, and recently a naturally occurring cell population in the bone marrow that expresses many B-lineage properties has been shown to be able to generate macrophages or dendritic cells, which play key sentinel roles in the innate immune system (70,71). Mouse and human T-cell precursors after birth similarly retain the ability to differentiate into dendritic cells until they start undergoing antigen receptor gene rearrangement (72-75). Yet another fate as an innate immune system effector is available to T-cell precursors as well: that of becoming a natural killer (NK) cell. These branch points in B- and T-cell development are shown in Fig. 4. Most strikingly, even cells that appear to be committed "common lymphoid progenitors" turn out to retain the ability to differentiate into macrophages and neutrophilic granulocytes, after all, provided they are given an appropriate growth factor receptor signal (76). Thus, the division between the developmental programs leading to adaptive and innate immune function is not so deep. Right up to the first stages of their receptor gene rearrangement events, mammalian lymphoid precursors preserve the developmental alternative of becoming a macrophage or dendritic cell, as shown in Fig. 4.

Cell Development

Fig. 4. Transcription factor roles in lymphocyte development. Timing of transcription factor action relative to developmental branch points and checkpoints in B- and T-cell development. For details and evidence, see the reviews cited in the text. The relevant stages of lymphocyte development appear to be the same whether the cells are differentiating from a panlymphoid common lymphoid progenitor (CLP) or from a B/myeloid (B, M) or T/myeloid (T, M) precursor, so all three types of potential precursors are shown at the left. The timing of activity of E2A, EBF, and Pax5 in B-cell development and of GATA-3, E2A, HES-1, HEB, and TCF-1 in T-cell development are based on the stages at which homozygous knockouts of these genes cause the most severe phenotype. Arrows leading to alternative fates (NK, or Dendritic/mac) show the latest stages at which individual precursor cells can give rise to both T-or B-lymphocytes and the indicated alternative. Bold arrows, major developmental pathways; thin arrows, confirmed minor pathways; broken arrow, conditional pathway. Stages when RAG-1/2-mediated recombination is active are indicated as R+. Stages at which further developmental progression becomes dependent on successful rearrangement events (checkpoints) are indicated for B- and T-cells by vertical dotted double lines. Important cell surface molecules are shown: for B-cells, immunoglobulin chains ^ and/or 8 and k; for T-cells, CD4, CD8, the T-cell receptor chains a and p, and the alternative class of T-cell receptors TCRy8.

A corollary is that the difference between lymphoid and myeloid developmental programs should be attributable to specific regulatory changes that occur in lym-phoid precursors when these two programs diverge. Cells that do not undergo these events should develop by default into macrophages, dendritic cells, or NK cells. The key regulatory molecules at these points are thus the first candidates we consider as effectors or targets of evolutionary innovations that may have made lymphocyte development possible.

4.3. Critical Regulators of B-Cell Specification

Three transcription factors play key roles in driving B-cells to differentiate from precursors that initially retain the ability to give rise to macrophages (reviewed in refs. 77-80). These factors appear to act in a mutually reinforcing cascade. The first one required for B-lineage specification is a class A basic helix-loop-helix (bHLH) factor encoded by the E2A gene. (differential splicing products E47 and E12 are both active.) The second is the HLH factor early B cell factor (EBF) (Olf-1, COE1), which collaborates closely with E2A to turn on B-cell genes. Even if these two factors have initiated a B-lineage gene expression program, specification is not stabilized into commitment until a third factor, Pax5, is also activated. Pax5 participates in both the positive regulation of B-cell genes and the negative regulation of alternative fates and the genes associated with them (79). The stages at which these factors are required to act are shown in Fig. 4 (upper part).

As probes for the events leading to evolutionary change, these transcription factors are illuminating, although perhaps in an unexpected way. None of them is a vertebrate innovation, or even a deuterostome innovation, in terms of its structure or DNA binding specificity. All three are members of small multigene families of extremely ancient provenance, with close relatives in protostomes as well as deuterostomes. E2A is a homolog of daugh-terless in Drosophila (81) with a close relative in the sea urchin (J.P. Rast, personal communication); EBF has homologs in the Drosophila Collier protein and CeO/E in Caenorhabditis elegans (82); and Pax5 is a member of the pan-bilaterian Pax2/5/8 gene family (83). Furthermore, all these transcription factors have important functions outside the immune system, functions that are conserved from mammals to protostomes. Both EBF and Pax5 are used in the mammalian nervous system, D-Pax2 is used for external sensory organogenesis in Drosophila, and the EBF homolog CeO/E is used in neuronal differentiation in C. elegans (84). The evolutionary innovations that gave these factors roles in B-cell development must have been primarily regulatory, i.e., changes that gave EBF and Pax5 a new domain of expression within the hematopoietic system.

The case of E2A is particularly striking, for the products of this gene are not generally restricted to any particular cell type at all. Mammalian E47 and E12, like Drosophila Daughterless, are expressed ubiquitously. They are used as developmentally neutral dimerization partners of other bHLH transcription factors that enable the heterodimers to carry out various tissue-specific differentiative functions. The only respect in which the B-cell use of E47 is distinct from that of muscle or nerve cells is that B-cells lack any other dimerization partners, and instead employ E47 as a homodimer. It is what the B-lineage cells lack, not what they express, that gives the E2A product(s) their unique roles in B-cells. Thus, in all three cases, the transcription factors that establish the unique B-cell identity in developing hematopoietic cells have been "seconded" to this task from other assignments of greater antiquity.

4.4. Regulators of T-Cell Specification

The initial events leading to specification of T-cells from multilineage or bipotent precursors are not yet as well understood as those for B-cells (reviewed in refs. 75,85, and 86). Still, mutational and expression analysis in mice suggests that at least four kinds of transcription factors are needed specifically in T-cell differentiation (Fig. 4, lower part). The zinc finger factor GATA-3 is required cell autonomously from a very early stage (87,88). The bHLH repressor HES-1 (a target of activation by Notch signal ing) is also required in immature T-cells, particularly for proliferative expansion (89). Class A bHLH proteins are essential for T-cell development in general: inhibition of these transcription factors can be sufficient to drive the earliest precursor cells into an NK developmental pathway (90). The class A bHLH activator Hela E-box binding protein (HEB) and the high mobility group (HMG box) factor T-cell factor-1 (TCF-1; a target of Wnt pathway signaling) are also needed later, for transition through the first T-cell receptor-dependent checkpoint (91,92). Precursor cells in older adult mice require TCF-1 even earlier in the T-cell differentiation pathway, perhaps as early as they require GATA-3 (93). By analysis of the cis-regulatory elements of T-cell-specific genes, it is evident that additional classes of factors, e.g., Ets, Runx, and Myb, are also essential (75,86,94), but individually these do not reveal T-cell-specific effects either because of the presence of multiple, partially redundant family members (for Ets: ref. 95), or because they are needed in hematopoietic precursors generally.

Like the transcription factors that orchestrate B-lineage specification, the T-cell differentiation factors are members of pan-bilaterian families that are used in diverse developmental programs. HEB is another member of the same small bHLH factor family as the E2A gene products, and there is extensive overlap between the functions of HEB and E2A even in mammalian lymphocyte development (96). Like the E2A products, HEB is a homolog of Daughterless in Drosophila. The bHLH repressor, HES-1, is closely related to Drosophila Hairy and the products of the various Enhancer of split complex genes. TCF-1 and its close vertebrate relative Lef-1 have a sea urchin homolog (97) and are members of the same family as Drosophila Pangolin. GATA-3 and its close vertebrate relative GATA-2 are similarly closely related to Drosophila dGATAc, as well as to sea urchin SpGATAc.

There are broad similarities in the ways that these transcription factors are utilized throughout the bilaterian radiation. As for the B-cell factors, all the T-cell transcription factors are used in other tissues as well as in lymphocytes, and it is these nonlymphoid sites of action that are conserved. GATA-3 (and dGATAc) and the bHLH activators and repressors all play prominent roles in neurogenesis, in both vertebrates and flies (98,99). TCF/Lef factors in vertebrates and in sea urchins mediate signals from the Wnt pathway (97,100-102); similarly, Pangolin mediates Wg signals throughout the fly embryo (103). Among the T-cell factors, only GATA-3 appears to have an additional, specific role in hematopoiesis that may be shared with animals that do not have lymphocytes. GATA-3 and its close relative GATA-2 are used in vertebrate hematopoietic stem cells (104,105), and we have already seen how SpGATAc, a GATA-2/3 homolog, is used by sea urchin coelomocytes. Even Drosophila hemocytes express a kind of GATA factor, Serpent, although this is not an ortholog of deuterostome hematopoietic GATA factors (106,107). Such instances of hematopoietic use cannot be simply interpreted as evidence that the invertebrates have T-lymphocyte-like cells per se, however, because the nonlymphoid sites of function of each of these factors and their close relatives are so diverse. The conclusion is evident that the T-cell developmental program, as far as it is understood, is again a new application for old regulators.

4.5. Regulators of Lymphomyeloid Precursor Generation

To find evidence for regulatory molecules that might contribute novel functions for lymphocyte development, perhaps surprisingly, we must look at the factors that control

Lymphoid Development And Ikaros

Fig. 5. Timing of transcription factor activities in hematopoiesis. The figure emphasizes the roles of Ikaros and PU.1 in definitive fetal hematopoiesis and in the differentiation of various hematopoietic cell types. Primitive, extraembryonic yolk sac hematopoiesis does not depend on either of these transcription factors, nor on Runx1 or c-Myb. Definitive intraembryonic stem cell generation, in contrast, depends acutely on Runx1 and c-Myb. Two kinds of derivatives of definitive stem cells show further dependence on PU.1 and Ikaros: these are the special long-term repopulating set of definitive stem cells (lower left) and the lymphocytes (NK, T, B). Myeloid cells also depend for their development on PU.1 to greater (macrophages) or lesser (granulocytes) extents, but they do not depend on Ikaros. In the figure, cell types that are Ikaros-depen-dent are shown in gray. Cell types that are at least partially PU.1-dependent are shown vertically striped. Heavy vertical stripes in B-cells and macrophages indicate the complete absence of cells of these lineages in PU.1-/- mice. Megakaryo, megakaryocyte.

Fig. 5. Timing of transcription factor activities in hematopoiesis. The figure emphasizes the roles of Ikaros and PU.1 in definitive fetal hematopoiesis and in the differentiation of various hematopoietic cell types. Primitive, extraembryonic yolk sac hematopoiesis does not depend on either of these transcription factors, nor on Runx1 or c-Myb. Definitive intraembryonic stem cell generation, in contrast, depends acutely on Runx1 and c-Myb. Two kinds of derivatives of definitive stem cells show further dependence on PU.1 and Ikaros: these are the special long-term repopulating set of definitive stem cells (lower left) and the lymphocytes (NK, T, B). Myeloid cells also depend for their development on PU.1 to greater (macrophages) or lesser (granulocytes) extents, but they do not depend on Ikaros. In the figure, cell types that are Ikaros-depen-dent are shown in gray. Cell types that are at least partially PU.1-dependent are shown vertically striped. Heavy vertical stripes in B-cells and macrophages indicate the complete absence of cells of these lineages in PU.1-/- mice. Megakaryo, megakaryocyte.

the development of uncommitted lymphomyeloid precursors. It has become increasingly clear in the past decade that the long-term self-renewing hematopoietic stem cells of adult vertebrates are not a "primitive" cell type, but rather the product of a regulated developmental program that maintains their pluripotentiality (108). Several additional transcription factors are involved in the generation of these definitive (post-yolk sac) hematopoietic precursors that can later continue to play essential roles in gene expression throughout lymphocyte differentiation, linking the stem cell and lymphocyte regulatory states. Differentiation programs for which these factors are required are shown in Fig. 5. In mice, these factors include the Runt class transcription factor Runxl (AML1, CBFa2, PEBP2aB), the Myb class transcription factor c-Myb, the zinc finger transcription factor Ikaros, and the divergent Ets subfamily member PU.1 (Spi-1). There is nothing lymphocyte-specific about these transcription factors, and yet, as we shall see, it is PU.1 that provides our best current evidence for a factor that is a vertebrate lymphohematopoietic innovation.

These transcription factors regulate lymphocyte and stem cell development via sets of target genes that may overlap but are certainly not the same. In mammals, c-Myb and Runxl are essential for the establishment of definitive hematopoietic stem cells (109-111) (Fig. 5). Runx is also essential for the expression of T-cell receptor a, P, y, and 5 genes, and c-Myb is used by T-cells not only to drive their cell cycle progression in response to stimulation by antigen (112), but also for expression of the T-cell receptor y and 5 genes (data not shown). Ikaros (113) and PU.l (114) are important for establishing the pool of definitive long-term repopulating stem cells in the bone marrow prior to birth (Fig. 5). Even in the fetal stages at which other blood cell types can develop without them, though, these two transcription factors still turn out to be essential for any lymphocyte development whatsoever (Fig. 5).

Both the Myb and the Runx/Runt family are extremely ancient, at least predating the protostome/deuterostome split. The cases of Ikaros and PU.l (or the PU.l subfamily of Ets factors) are a little different, as described in the next section.

Both Ikaros and PU.l have particularly complex functions that include inhibitory as well as stimulatory effects on particular target genes. Recent studies imply that one of the major roles of Ikaros may be as a repressor of subset-specific lymphoid genes that are inappropriate for the current differentiation state (115); it also appears to keep lymphocytes from responding too easily to subthreshold activation signals (116). PU.l is intimately involved in positive regulation of B-cell cytokine receptor and antigen receptor genes (117,118). However, in its most general role, PU.l is a potent positive regulator of the differentiation and growth of macrophages and granulocytes, cells of the innate immune system (119) (Fig. 5). Whereas at moderate levels of expression it is needed for early events in both T- and B-lymphocyte lineages (120,121), at higher levels, PU.l can actually block lymphocyte development (122,123). The fact that lymphocytes depend on the activity of a myeloid transcription factor for their initial development is an additional feature linking the lymphocyte lineages with the phago-cytic lineages. The compatibility between the regulators needed to initiate lymphoid and phagocytic differentiation programs in jawed vertebrates is consistent with the possibility that the lymphocyte developmental program evolved as a modification of a phagocytic cell program.

Was this article helpful?

0 0
How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book


Responses

Post a comment