Skip Navigation

Integrative and Comparative Biology 2003 43(2):338-346; doi:10.1093/icb/43.2.338
© 2003 by The Society for Integrative and Comparative Biology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrow Request Permissions
Google Scholar
Right arrow Articles by Bartl, S.
Right arrow Articles by Diaz, M.
Right arrow Search for Related Content
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Did the Molecules of Adaptive Immunity Evolve from the Innate Immune System?1

Simona Bartl2,1, Meredith Baish2, Irving L. Weissman3 and Marilyn Diaz4
1 Moss Landing Marine Laboratories, Moss Landing, California 95039
2 Department of Biological Sciences, University of North Carolina, Wilmington, North Carolina 28403
3 Departments of Pathology and Developmental Biology, Stanford University, Stanford, California 94305
4 The Scripps Research Institute, Department of Immunology, La Jolla, California 92037


   SYNOPSIS
TOP
SYNOPSIS
INTRODUCTION
 REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
 CONCLUSIONS
 References
 
The antigen receptors on cells of innate immune systems recognize broadly expressed markers on non-host cells while the receptors on lymphocytes of the adaptive immune system display a higher level of specificity. Adaptive immunity, with its exquisite specificity and immunological memory, has only been found in the jawed vertebrates, which also display innate immunity. Jawless fishes and invertebrates only have innate immunity. In the adaptive immune response, T and B-lymphocytes detect foreign agents or antigens using T cell receptors (TCR) or immunoglobulins (Ig), respectively. While Ig can bind free intact antigens, TCR only binds processed antigenic fragments that are presented on molecules encoded in the major histocompatibility complex (MHC). MHC molecules display variation through allelic polymorphism. A diverse repertoire of Ig and TCR molecules is generated by gene rearrangement and junctional diversity, processes carried out by the recombinase activating gene (RAG) products and terminal deoxynucleotidyl transferase (TdT). Thus, the molecules that define adaptive immunity are TCR, Ig, MHC molecules, RAG products and TdT. No direct predecessors of these molecules have been found in the jawless fishes or invertebrates. In contrast, the complement cascade can be activated by either adaptive or innate immune systems and contains examples of molecules that gradually evolved from non-immune functions to being part of the innate and then adaptive immune system. In this paper we examine the molecules of the adaptive immune system and speculate on the existence of direct predecessors that were part of innate immunity.


   INTRODUCTION
TOP
SYNOPSIS
 INTRODUCTION
REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
 CONCLUSIONS
 References
 
The immune system of higher vertebrates displays exquisite specificity and strong immunological memory. These are the hallmarks of adaptive immunity, which is remarkable for its ability to change and improve over time. The adaptive immune system can be so effective that the host experiences few or no symptoms as many types of pathogens are eradicated from the body. It is these "learned" responses to foreign agents that allow immunizations of humans and their domesticated animals to prevent disease. Immune system molecules that govern specificity and memory are not found in most animals, nor can they be traced through a gradual evolutionary process to their present form. Instead the molecules that orchestrate adaptive immunity are found in the jawed vertebrates and nowhere else. That is, all the molecular components are present in cartilaginous fishes and higher vertebrates while none have been found in jawless fishes. Starting with the isolation of blood proteins in the 1960's and continuing to the present, most of the molecules of adaptive immunity that are pursued are found in sharks and rays and not in earlier fish lineages or invertebrates (Bartl, 1998Go; Clem et al., 1967Go; Clem and Small, 1967Go; Frommel et al., 1971Go; Kurosawa and Hashimoto, 1997Go; Litman et al., 1999Go; Marchalonis et al., 1998Go; Small et al., 1970Go; Suran et al., 1967Go; Voss et al., 1969Go). It is not that these "other" animals do not have effective immune responses, they most certainly do. The immune responses of jawless fish and invertebrates effectively identify and inactivate or remove foreign agents, but without the level of specificity and memory found in jawed vertebrates (Hoffmann and Reichhart, 2002Go; Humphreys and Reinherz, 1994Go; Lackie, 1980Go; Nappi and Ottaviani, 2000Go; Ratcliffe, 1985Go; Salzet, 2001Go; Soderhall and Cerenius, 1998Go). Immune responses such as these that cannot be considered adaptive are called innate. Innate immunity is found in both animals with and without adaptive immunity. The molecular components of the innate immune system do not appear to be closely related to the molecules of the adaptive system.

What would make the molecules of adaptive immunity seem to appear out of thin air? Certainly innate immune functions preceded adaptive functions in evolutionary history. Thus, the adaptive immune system did not replace innate immunity, but rather augmented it. Based on molecular evidence, a system for the effective recognition and removal of foreign agents was in place at the time of the evolution of jawed vertebrates. It appears logical then that adaptive immunity would build on molecules that were present and part of host defense. We will consider two possible explanations for the lack of identified molecular precursors for adaptive immunity. First, the precursor molecules were performing innate immune functions and were structurally similar to the adaptive molecules of today. If so, extant homologs of these precursors still perform innate immune functions but may not be easy to isolate due to inherent diversity. Second, the molecular precursors of the adaptive immune system were performing functions other than immunity and underwent rapid evolution while being recruited to the emerging adaptive immune system. Extant homologs of these precursors are predicted to be quite different from their related immune molecules and serendipitous isolation may only be possible since it is difficult to predict their structure or pattern of expression. In this paper we will consider each of the key molecules of adaptive immunity and assess which explanation best fits our present state of knowledge.

Molecules of adaptive immunity
To understand our choice of adaptive molecules we need to provide some background on the components of immune systems. Adaptive immune responses are initiated by a subset of white blood cells, the lymphocytes, which displays highly specific receptors and contains memory cells. Lymphocytes are subdivided into B cells and T cells and they directly recognize and bind antigens. The antigen receptors on B cells are immunoglobulins (Ig) and on T cells are T cell receptors (TCR). Both types of receptors have Ig-fold domains and share other structural features that suggest a shared ancestor (for review see Hood et al., 1985Go). Both require gene rearrangement of germline elements for the expression of the cell surface receptor. Gene rearrangement brings together single gene elements from two or three element pools [encoding the variable (V), joining (J) or diversity (D) regions]. This process is associated with the addition of non-templated nucleotides at the gene element junctions to produce what are called N regions (Gilfillan et al., 1995Go; Komori et al., 1996Go; VanDyk and Meek, 1992Go). The two processes of gene rearrangement and N-region addition have the combined potential to produce a vast pool of unique receptor molecules. Each lymphocyte displays only one receptor from this pool, providing it with a unique specificity. Although antigen receptors on lymphocytes have structural similarity, the nature of the antigen bound by Ig and TCR is different. While Ig binds to free antigen in its native form, TCR binds processed pieces of antigens that are presented on major histocompatibility complex (MHC) molecules (Braciale and Braciale, 1991Go; Engelhard, 1994Go; Hudson and Ploegh, 2002Go; Whitton, 1998Go). These MHC molecules are displayed on the surfaces of functionally named antigen-presenting cells. Thus, the molecular markers for adaptive immunity are Ig, TCR, MHC molecules and the machinery for generating receptor repertoire diversity, namely the recombinase activating gene (RAG) products that rearrange gene elements and terminal deoxynucleotidyl transferase (TdT) that creates the non-templated junctional N-regions. All of these molecules have only been found in the jawed vertebrates. No homologs that could be considered to represent direct predecessors have been isolated from jawless fishes or invertebrates.

Innate immune responses are seen in both invertebrates and vertebrates and arguably are present in single-celled eukaryotes. From available molecular evidence it appears that adaptive immunity evolved in an ancestor of present-day jawed vertebrates. Thus, in jawless fishes and invertebrates, innate immune responses may be the only means of host defense. In the jawed vertebrates, innate immune responses provide an effective first line of defense and assist lymphocytes in the detection and removal of pathogens and other antigens. Non-lymphoid leukocytes contain phagocytes, granulocytes, and natural killer (NK) cells that use a variety of receptors to initiate the innate response (Cooper et al., 2001Go; Figdor et al., 2002Go; Medzhitov, 2001Go; Peiser et al., 2002Go). The cells and molecules of the adaptive immune system work in consort with the innate system and can be considered a more specialized version of the earlier, solely innate, system (Lo et al., 1999Go; Medzhitov and Janeway, 1999Go). In this paper, we will discuss key molecules in adaptive immunity that serve various functions and consider the possible nature of their evolution from precursor molecules. For comparison, we will consider the components of the complement cascade, which exemplify the hypothesis of a gradual evolution from innate immune functions in invertebrates to adaptive responses (Table 1) (for review see Gross et al., 1999Go, Hughes and Yeager, 1997Go, Travis, 1993Go, and Zarkadis et al., 2001Go). In higher vertebrates, the complement cascade can be activated directly by microorganisms (alternative pathway), by lectin binding to microorganisms (lectin pathway) or by immunoglobulins complexed to antigen (classical pathway). Only the latter is considered part of adaptive immunity while the other two are innate responses. The molecules of the classical pathway (C1, C2, C4) are found in jawed vertebrates but molecular evidence for the alternative pathway (factor B, C3) and the lectin pathway (MBL, MASP) is found in more distantly related organisms (Al-Sharif et al., 1998Go; Nonaka et al., 1998Go; Zarkadis et al., 2001Go). Related thioester-containing molecules (C2, factor B, C3, C4) play a central role in all three pathways and have clear relatives in lower vertebrates and invertebrates. These molecules of the {alpha}-2-macroglobulin family contain members that are not linked directly to immunity as well as components of the three complement cascades (Dodds and Law, 1998Go). Thus, complement will be discussed as an example of gradual evolution from predecessors of similar function. Each molecule may provide different clues for the evolution of the molecules of adaptive immunity.


View this table:
[in this window]
[in a new window]
  		TABLE 1. The presence (+) of molecular markers of adaptive immunity and the complement cascade pathways in select taxa.

 

   REARRANGEMENT MACHINERY
TOP
SYNOPSIS
 INTRODUCTION
 REARRANGEMENT MACHINERY
GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
 CONCLUSIONS
 References
 
RAG-1 and RAG-2 encode the molecules that rearrange antigen receptor genes (Oettinger et al., 1990Go; Willerford et al., 1996Go). The mostly intron-less structure and the close linkage of the two genes make them unlike other higher vertebrate genes. It has been suggested that RAGs may have been horizontally transferred from a lower eukaryote or prokaryote into the vertebrate lineage (Bartl et al., 1994Go; Oettinger et al., 1990Go). More recent evidence that RAG recombination is mechanistically similar to retroviral integration and transposition supports the hypothesis that RAGs originated as a retrotransposon (Agrawal, 2000Go; Fugmann, 2001Go; Gellert et al., 1999Go; Schatz, 1999Go). This hypothesis must be considered in the context of recent work on the general principal of horizontal gene transfer from microbes to vertebrates that found greater support for gene loss, sample size effects and variation in evolutionary rate as explanations for the available data (Salzberg et al., 2001Go; Stanhope et al., 2001Go). Horizontal transfer, gene loss or variation of evolutionary rate would all result in no detectable RAG homologs in the close ancestors of the jawed vertebrates.

Unlike most of the other molecules of adaptive immunity, RAGs are quite well conserved within the jawed vertebrates. RAG-1 is especially well conserved containing 80 amino acid blocks with >75% similarity at the protein level (Bernstein et al., 1996Go). If RAG-1 homologs are present in taxa older than jawed vertebrates, then specific PCR primers to highly conserved sites should be a sensitive method of detection. We designed seven sets of primers for RAG-1 and three sets for RAG-2 (Fig. 1). All primer sets except one amplified a band of the correct size using mouse DNA as a template (Table 2). The RAG-2 primer sets did not produce the correct product as consistently as the RAG-1 primers, which is not surprising since this gene is less well conserved. One of the RAG-1 products from nurse shark was isolated and sequenced and shows a degree of overlap with a partial nurse shark clone reported previously and significant similarity with the bull shark and human RAG-1 genes (Fig. 2) (Bernstein et al., 1996Go; Greenhalgh and Steiner, 1995Go; Schatz et al., 1989Go). Multiple PCR products from jawless fishes and protochordates (Botryllus and Styela) that were close to the correct size were sequenced as well. None of these had sequence similarity to RAGs outside of the priming sites (data not shown). We conclude that if RAG-1 existed in ancestors of the jawed vertebrates it is undetectable by this quite sensitive method.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 1. The location of the RAG-1 (A) and RAG-2 (B) primer annealing sites on a schematic of the mouse protein-coding regions (Oettinger et al., 1990Go; Schatz et al., 1989Go). Solid arrows denote forward primers and dashed arrows denote reverse primers. Below the name of each primer is the corresponding 5' nucleotide number. The overall length of each region is shown at the right

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. PCR products* obtained with RAG-1 (11–18) and RAG-2 (21–25) primers on a variety of DNA templates.

 


View larger version (12K):
[in this window]
[in a new window]
  		FIG. 2. Deduced amino acid sequence for human 806–1002 and bull shark 813–1009 RAG-1 fragments aligned with nurse shark partial clones 1 and 2 (Genbank accession numbers M29474 [GenBank] , U62645 [GenBank] , AY102165 [GenBank] , U13982 [GenBank] , from top to bottom). Dots denote identity

 

   GENERATION OF N-REGIONS
TOP
SYNOPSIS
 INTRODUCTION
 REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
 CONCLUSIONS
 References
 
N-region addition at gene element junctions in Ig and TCR genes contributes significantly to the overall diversity of receptor repertoires in the jawed vertebrates (Du Pasquier, 1993Go; Hinds-Frey et al., 1993Go). The addition of non-template-encoded nucleotides during the rearrangement process is mediated by terminal deoxynucleotidyl transferase (TdT). Partial TdT genes were recently isolated from cartilaginous fishes and used as probes to assay lymphoid tissues during shark and skate development (Miracle et al., 2001Go; Rumfelt et al., 2001Go). The expression of TdT in tandem with RAGs is a conserved feature of sites of lymphocyte development in cartilaginous fishes as it is in higher vertebrates. An analysis of the full-length TdT genes from shark and skate finds significant sequence similarity to the pol X gene family (S. B., unpublished data). This family includes other vertebrate TdTs, DNA polymerase beta and yeast DNA polymerase IV (Anderson et al., 1987Go; Holm and Sander, 1995Go; Shimizu et al., 1993Go). Thus, TdT is a member of a larger family of DNA manipulating enzymes but only TdT functions in a template-independent manner.

It is unclear when in the evolution of this gene family a DNA polymerase became template independent. Earlier TdT homologs may be found in invertebrates. The level of similarity between pol X family members makes them detectable by standard PCR methods such that all members could be isolated from jawless fishes and protochordates. These genes may provide clues to TdT evolution by the loss or gain of key DNA binding residues. Though the putative role of TdT prior to gene rearrangement is unclear at this time there is precedent for nucleotidyl transferases in the repair of abasic sites (Masuda et al., 2002Go; Zhang et al., 2002Go).


   ANTIGEN RECEPTORS
TOP
SYNOPSIS
 INTRODUCTION
 REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
COMPLEMENT COMPONENTS
 CONCLUSIONS
 References
 
Both TCRs and Igs have antigen-binding sites composed of protein heterodimers that are encoded by rearranged gene elements. The light chain is composed of V and J elements. The heavy chain is composed of V, D and J elements. Each protein chain of the antigen-binding site creates one half of an Ig-fold. Additional Ig-folds are encoded by the constant (C) regions of each gene and are associated with effector functions of the receptor. If RAGs did not exist prior to the jawed vertebrates then the predecessor of antigen receptors must have been a non-rearranging member of the Ig gene superfamily complete with identifiable regions with similarity to V, J and C regions. If we continue the argument that precursors performed innate immune functions and present-day relatives perform similar functions in the jawed vertebrates, then a recently identified group of molecules are likely candidates. These are the novel immune-type receptors (NITRs) isolated from bony fishes (Hawke et al., 2001Go; Strong et al., 1999Go; Yoder et al., 2001Go). These receptors are part of a large family each having either one or two Ig domains that are structurally similar to the V domain of Ig, TCR and a set of Ig-type receptors (called KIR) expressed on mammalian NK cells. Thus, NITRs have structural similarity to receptors on cells of both the adaptive and innate immune systems. Intriguingly, some of the V domains of the NITRs end with a region encoding GXG characteristic of the J regions of the adaptive antigen receptors (Ig and TCR) and these are linked to constant region domains (Hawke et al., 2001Go). Different NITR members possess either potentially activating or inhibiting cytoplasmic regions. In mixed leukocyte reactions, which measure immune responses to allogeneic cells, some of the NITRs are up-regulated implying an association with immune function (Hawke et al., 2001Go). NITRs display multiple structural similarities with antigen receptors that are best explained by homology rather than convergence. The intriguing possibility is that this large diverse family shares a direct ancestor with the antigen receptors of adaptive immunity. The diversity seen in bony fishes may reflect a lack of specialization by the innate and adaptive immune systems. In contrast, mammals may represent a highly partitioned system where only a remnant of the NITR family remains in the form of the KIR receptors. Further speculation awaits the isolation of NITRs from jawless fish and protochordates and the demonstration of their role in immunity, a difficult feat since assays for immune function are not well established in these organisms.

MHC molecules
In higher vertebrates MHC molecules are members of a large gene family (for review see Bartl, 1998Go, Braud et al., 1999Go, and Maenaka and Jones, 1999Go). Most MHC genes are found at one chromosomal location called the major histocompatibility complex. Structural features divide the members into class I and class II types. Within each type are functionally defined classical molecules, which present processed antigens to T lymphocytes, and non-classical molecules. Some examples of the roles of non-classical MHC molecules are iron homeostasis (Hfe), processing of antigens (DM) or presentation to cells other than T lymphocytes (HLA-E) (Braud et al., 1998Go, 1999Go; Lee et al., 1998Go; Maenaka and Jones, 1999Go). In general, the classical MHC molecules are highly polymorphic, encoded in the MHC, and expressed in a temporal and spatial pattern characteristic for each class type. Non-classical MHC molecules usually have few alleles, are not always MHC-encoded, and vary widely in their function and expression pattern.

The evolution of MHC molecules poses a dilemma. There are clear examples of MHC molecules performing innate immune functions in higher vertebrates. For example, NK cells recognize some classical MHC molecules (Lanier, 1998Go). Also, the innate immune system has recently been found to play a role in the recognition of allogeneic tissue grafts, a process previously considered to be T lymphocyte-mediated recognition of non-self MHC molecules (Fox-Marsh and Harrison, 2002Go; Maier et al., 2001Go). On the other hand, some MHC molecules appear to be ancestors in phylogenetic analyses such as the non-classical class I molecules of bony fishes and the DM class II molecules (Bartl, 1998Go; Hughes and Yeager, 1997Go). It has been suggested that the MHC ancestor is a class II rather than a class I molecule (Hughes and Yeager, 1997Go). Thus, DM molecules may represent a direct link to the ancestral molecule. Present-day DM molecules do not have an innate immune function. They serve as chaperones in the pathway that loads antigenic fragments onto classical class II MHC molecules (Brocke et al., 2002Go; Van Kaer, 2001Go).

Thus, the MHC ancestor may have had a role in protein trafficking and not immunity. Recent evidence from a second set of chaperones besides DM may suggest that the MHC predecessor was a heat shock protein (hsp). Several hsp's have been found to function like MHC molecules by binding antigenic peptides, being expressed within cells and on cell surfaces, and mediating T cell activation (Banerjee et al., 2002Go; Li et al., 2002Go; More et al., 2001Go). The immunological properties of one hsp (gp96) appear to be conserved between mammals and amphibians (Robert et al., 2001aGo, bGo). In light of this new evidence it is worth revisiting the similarity noted by Flajnik et al. between the putative peptide-binding site of hsp70 and MHC class I molecules of amphibians (Flajnik et al., 1993Go). Also, hsp70 has been mapped within the MHC locus in frogs as it is in higher tetrapods (Salter-Cid et al., 1994Go). Thus, an MHC predecessor may share the common structural and functional features of MHC molecules and hsp's.

In addition to the lack of ancient genes with clear innate immunological function, two other features make MHC precursors difficult to trace. First, even recent precursors of mammalian MHC genes appear to change from classical to non-classical functions, arguing for multiple evolutionary events towards adaptive functions (Cadavid et al., 1997Go). Even MHC molecules whose functions may be considered solely innate such as HLA-E regulating NK cells or HLA-G involved in maternal-fetal immune responses, appear to have evolved late in the vertebrate lineage (Hughes and Yeager, 1997Go). Second, the high degree of polymorphism in classical MHC genes appears to be a common feature from mammals to cartilaginous fishes (Bartl, 1998Go; Kasahara et al., 1993Go; Okamura et al., 1997Go). The lack of evidence for MHC molecules in jawless fishes and invertebrates may be due to low sequence similarity rather than absence. However, close examination of MHC molecules that branch early in phylogenetic analyses (i.e., DM), and possibly the hsp's, may identify conserved regions that could be successfully targeted by short primers with limited degeneracy to amplify homologs in jawless fishes and protochordates. Alternatively, methods that target functions associated with classical MHC molecules hold the possibility of identifying either the MHC predecessor or unrelated molecules that have converged on a similar function. Either result would contribute to our understanding of the evolution of adaptive immunity. An example of such work is the progress towards the isolation of the Fu/HC locus, which governs histocompatibility in the protochordate Botryllus schlosseri (De Tomaso et al., 1998Go).


   COMPLEMENT COMPONENTS
TOP
SYNOPSIS
 INTRODUCTION
 REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
CONCLUSIONS
 References
 
In contrast to the examples discussed to this point, the adaptive arm of the complement cascade seems to have developed from the components of the innate arm in a direct fashion. There are several identified gene families within the cascade and we will begin with the C3 and C4 components (for review see Gross et al., 1999Go, Nonaka, 2001Go, and Zarkadis et al., 2001Go). These components are members of a protein family in which some members evolved from non-immune to innate to adaptive functions, while other members retained their previous functions in the higher vertebrates. Alpha-2-macroglobulins ({alpha}2M) are related to C3 and C4 and are found in lower invertebrates, protochordates, jawless fishes and higher vertebrates. The progressive evolution from serine protease inhibitors ({alpha}2M) to cell surface tags (C3 and C4) has been discussed elsewhere (Dodds and Law, 1998Go). In lower invertebrates, only {alpha}2M genes have been found, but clear C3 homologs, as well as {alpha}2M, have been isolated from higher invertebrates (echinoderms and protochordates) and jawless fishes.

We used PCR primers designed to all known C3/{alpha}2M homologs to amplify cDNA from the protochordate Botryllus schlosseri and isolated 2 partial clones. The deduced amino acid sequences for the Botryllus clones were placed in a neighbor-joining tree with other sequences in the {alpha}2M family (Fig. 3). Although the short length of the sequences analyzed produced some nodes with low bootstrap values, several conclusions can be made. One Botryllus clone (12c) appears to be a homolog of an {alpha}2M gene from another protochordate (Ciona). The other Botryllus clone (22e) is not closely related to either the Ciona {alpha}2M or the protochordate C3 sequences (Ciona and Halocynthia) (Marino et al., 2002Go; Nonaka et al., 1999Go). It is not clear if this clone represents an {alpha}2m or a C3 gene. Thus, in the protochordates there may be two or more {alpha}2M genes that may have been present in earlier ancestors. These early duplication events would have allowed some loci to evolve into C3 genes while retaining other loci for the original {alpha}2M functions. A subsequent diversification of a C3 gene into C4 in the ancestor of jawed vertebrates provided the beginnings of the classical cascade. As appears logical, the complement cascade became more refined with the evolution of the vertebrates, retaining ancestral innate functions and adding new adaptive ones.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 3. A phylogenetic tree of a partial deduced amino acid sequence from select C4, C3 and {alpha}2M genes and two Botryllus clones (Genbank accession numbers AY102166 [GenBank] and AY102167 [GenBank] ) using the neighbor-joining method (Saitou and Nei, 1987Go). The numbers are the percent recovery of that node for 1000 bootstrap replications. The Genbank accession numbers for the aligned sequences are: C4 from human (K02404 [GenBank] ), Medaka (AB025577 [GenBank] ), carp (AB037279 [GenBank] ); C3 from human (K02765 [GenBank] ), trout (L24433 [GenBank] ), lamprey (D10087 [GenBank] ), hagfish (Z11595 [GenBank] ), Amphioxus (AB050668 [GenBank] ), Ciona (AJ320542 [GenBank] , AJ320543 [GenBank] ), Halocynthia (AB006964 [GenBank] ), sea urchin (AF025526 [GenBank] ); {alpha}2M from human (M11313 [GenBank] ), rat (J02635 [GenBank] ), lamprey (D13567 [GenBank] ), Ciona (AJ431688 [GenBank] ), horseshoe crab (D83196 [GenBank] )

 
This progression also is seen when molecules up and downstream of C3/C4 are compared. C4 is a component of both the lectin (innate) and classical (adaptive) cascades (Fig. 4 and Table 1). These cascades are triggered when foreign agents are bound by certain lectins (mannose-binding lectin [MBL] or ficolin) or certain Ig subtypes (Fujita, 2002Go). The alternative (innate) pathway is triggered by the presence of pathogen surfaces that activate C3. For the lectin and classical cascades the first components recruited are MASP (MBL-associated serine protease) or C1, respectively. These then activate C4. Although the lectins and Ig do not share structural similarity, MASP and C1 do. It has been hypothesized that C1 evolved from a MASP-like ancestor that switched its binding capacity from binding a lectin to binding an Ig molecule (Nonaka, 2001Go; Zarkadis et al., 2001Go). In the classical and lectin pathways C4 activation is followed by C2 activation and in the alternative pathway factor B (B) is activated by C3. Factor B and C2 are also structurally similar. The most ancient MASP homologs (as well as a MBL homolog) have been found in protochordates while a factor B homolog is present in echinoderms (Ji et al., 1997Go; Smith et al., 1998Go; Vasta et al., 1999Go). Thus, multiple early components of a lectin/alternative pathway were present during invertebrate deuterosome evolution. Then, during vertebrate evolution three clear pathways emerged, one of which was restricted to activation by the adaptive immune system. Additional downstream steps and the effector functions shown in Figure 4 also appear to have been gradually acquired during complement cascade evolution (Fujita, 2002Go).



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 4. A schematic diagram of the early events in complement cascade activation. All three initiation pathways converge on a C3 activation step that then results in common effector functions. (Adapted from Fujita, 2002Go and Janeway et al., 1999Go)

 

   CONCLUSIONS
TOP
SYNOPSIS
 INTRODUCTION
 REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
 CONCLUSIONS
References
 
In this paper we have made the argument that at least some of the molecules of adaptive immunity arose from structurally similar precursors with related innate immune functions. We began by considering the possibility that predecessors were performing functions unrelated to immunity as less likely. We discussed the initial components of the classical (adaptive), lectin (innate) and alternative (innate) cascades of complement as examples of a gradual evolution of related molecules from non-immune ({alpha}2M) to immune (C3 and C4) function and from innate (MASP) to adaptive (C1) function. In comparison, we addressed the other key molecules in immune evolution and considered which may have had immune functions prior to the evolution of adaptive immunity.

Since the antigen receptors on lymphocytes require rearrangement, we first considered the molecules responsible for this process, RAG-1 and RAG-2. These genes are relatively well conserved and at least RAG-1 is quite easy to isolate from various jawed vertebrates. But there is no evidence that a similar molecule is present in the jawless vertebrates or invertebrates. Although other DNA manipulating enzymes, such as microbial site-specific recombinases, share some structural features with the RAGs, their overall similarity is low (Bernstein et al., 1996Go; Hughes and Yeager, 1997Go). It has been suggested that the RAGs were horizontally transferred in tandem into a jawed vertebrate ancestor and recent evidence for transposition activity by RAGs suggests a viable mechanism. Future analyses of the general process of horizontal gene transfer from microbes to vertebrates may find the RAGs to be a unique exception.

The addition of non-templated nucleotides by the TdT enzyme adds diversity to the receptor repertoire. Part of a well-conserved family of DNA polymerases, the TdTs themselves are also well conserved. It should be quite straightforward to isolate all family members from the jawless fish and key invertebrates and compare structural features. Since structure/function relationships of several family members including TdT have been well characterized, it should be possible to identify intermediary precursors of TdT (Anderson et al., 1987Go; Holm and Sander, 1995Go). However, we find no obvious innate immune role for a TdT precursor. Rather, the recruitment into adaptive immunity of a DNA repair enzyme may have taken very few evolutionary steps to produce the present-day TdTs.

MHC molecules that present antigenic fragments to T lymphocytes are members of an apparently ancient family of diverse molecules that also function in innate immunity. Some of the present-day innate functions may reflect an ancestral role. However, targeting MHC molecules that function in innate immunity presents difficulties. They are a diverse group of molecules with many innate immune functions and in some cases overlapping adaptive immune roles. They also seem to switch easily between innate and adaptive functions on short evolutionary timescales. Alternatively, comparisons of phylogenetically distant genes may predict structural features of the MHC predecessor and have been previously discussed by one of us (Bartl, 1998Go). Following the argument that MHC class II molecules evolved first and that DM genes diverged early in class II evolution, common features between DM and classical class II genes may be ancestral (Bartl, 1998Go; Hughes and Yeager, 1997Go). Recent evidence for immunological roles of some hsp's suggests that similarities between these chaperones and MHC molecules may provide clues to a common predecessor (Banerjee et al., 2002Go; Flajnik et al., 1993Go; Li et al., 2002Go; More et al., 2001Go; Robert et al., 2001aGo, bGo). Thus, it must be considered that the ancestral MHC molecule may have been a chaperone rather then a receptor of the innate immune system. A shift from chaperoning peptides to presenting them to T cells may have been the crucial evolutionary step.

The antigen receptors, Ig and TCR, are related Ig superfamily members that require rearrangement. We would argue that their direct ancestors were similar to the Ig superfamily of NK receptors in mammals and the NITR family found in bony fish. These receptors have immune functions and as the family becomes better characterized we predict that some family members with clear similarity to Ig and TCR will be found in jawless fish and invertebrates as well.

In conclusion, at this time adaptive immunity appears to have evolved "with a bang" or all at once in an ancestor of the jawed vertebrates with all components appearing out of nowhere (Schluter et al., 1999Go). This would be the case if evolution worked rapidly and simultaneously to convert multiple predecessors with non-immune function. Here, we argue for one small bang that occurred when RAGs began their present roles with the other molecules of adaptive immunity gradually evolving from predecessors, some with previous innate functions. We predict that the predecessors of the antigen receptors arose by gene duplication and gradual diversification and that the related genes have persisted with their ancestral functions in invertebrates and cartilaginous fishes. We consider the possibility that some of these related genes may have been lost in higher vertebrates. We predict that the MHC molecule predecessor was recruited from an ancestor involved in protein trafficking. Molecular evidence to support our prediction has been difficult to obtain due to the inherent diversity of the MHC molecules. In the case of TdT relatively few changes would convert a temple-dependent DNA polymerase with no immune function to the template independence that generates diversity in antigen receptors. Starting with four different molecules we end with four different scenarios of the past and forecast an interesting future hunt for evolutionary immunologists.


   FOOTNOTES
 
101 Sequences reported in this paper have been deposited in the Genbank database as accession numbers AY102165 [GenBank] (nurse shark RAG-1), AY102166 [GenBank] (Botryllus 12c), and AY102167 [GenBank] (Botryllus 22e) Back

1 From the Symposium Comparative Immunology presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 2–6 January 2002, at Anaheim, California. Back

2 E-mail: sbartl{at}mlml.calstate.edu Back


   References
TOP
SYNOPSIS
 INTRODUCTION
 REARRANGEMENT MACHINERY
 GENERATION OF N-REGIONS
 ANTIGEN RECEPTORS
 COMPLEMENT COMPONENTS
 CONCLUSIONS
 References
 
Agrawal, A. 2000. Amersham Pharmacia Biotech & Science Prize. Transposition and evolution of antigen-specific immunity. Science, 290:1715-6.[Free Full Text]

Al-Sharif, W. Z., J. O. Sunyer, J. D. Lambris, and L. C. Smith. 1998. Sea urchin coelomocytes specifically express a homologue of the complement component C3. J. Immunol, 160:2983-97.[Abstract/Free Full Text]

Anderson, R. S., C. B. Lawrence, S. H. Wilson, and K. L. Beattie. 1987. Genetic relatedness of human DNA polymerase beta and terminal deoxynucleotidyltransferase. Gene, 60:163-73.[CrossRef][ISI][Medline]

Banerjee, P. P., D. S. Vinay, A. Mathew, M. Raje, V. Parekh, D. V. Prasad, A. Kumar, D. Mitra, and G. C. Mishra. 2002. Evidence that glycoprotein 96 (B2), a stress protein, functions as a Th2-specific costimulatory molecule. J. Immunol, 169:3507-18.[Abstract/Free Full Text]

Bartl, S. 1998. What sharks can tell us about the evolution of MHC genes. Immunol. Rev, 166:317-31.[CrossRef][ISI][Medline]

Bartl, S., D. Baltimore, and I. L. Weissman. 1994. Molecular evolution of the vertebrate immune system. Proc. Natl. Acad. Sci. U.S.A, 91:10769-70.[Free Full Text]

Bernstein, R. M., S. F. Schluter, H. Bernstein, and J. J. Marchalonis. 1996. Primordial emergence of the recombination activating gene 1 (RAG1): Sequence of the complete shark gene indicates homology to microbial integrases. Proc. Natl. Acad. Sci. U.S.A, 93:9454-9.[Abstract/Free Full Text]

Braciale, T. J., and V. L. Braciale. 1991. Antigen presentation: Structural themes and functional variations. Immunol. Today, 12:124-9.[CrossRef][ISI][Medline]

Braud, V. M., D. S. Allan, and A. J. McMichael. 1999. Functions of nonclassical MHC and non-MHC-encoded class I molecules. Curr. Opin. Immunol, 11:100-8.[CrossRef][ISI][Medline]

Braud, V. M., D. S. Allan, C. A. O'Callaghan, K. Soderstrom, A. D'Andrea, G. S. Ogg, S. Lazetic, N. T. Young, J. I. Bell, J. H. Phillips, L. L. Lanier, and A. J. McMichael. 1998. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature, 391:795-9.[CrossRef][Medline]

Brocke, P., N. Garbi, F. Momburg, and G. J. Hammerling. 2002. HLA-DM, HLA-DO and tapasin: Functional similarities and differences. Curr. Opin. Immunol, 14:22-9.[CrossRef][ISI][Medline]

Cadavid, L. F., C. Shufflebotham, F. J. Ruiz, M. Yeager, A. L. Hughes, and D. I. Watkins. 1997. Evolutionary instability of the major histocompatibility complex class I loci in New World primates. Proc. Natl. Acad. Sci. U.S.A, 94:14536-41.[Abstract/Free Full Text]

Clem, I. W., F. De Boutaud, and M. M. Sigel. 1967. Phylogeny of immunoglobulin structure and function. II. Immunoglobulins of the nurse shark. J. Immunol, 99:1226-35.

Clem, L. W., and P. A. Small Jr. 1967. Phylogeny of immunoglobulin structure and function. I. Immunoglobulins of the lemon shark. J. Exp. Med, 125:893-920.[Abstract]

Cooper, M. A., T. A. Fehniger, and M. A. Caligiuri. 2001. The biology of human natural killer-cell subsets. Trends Immunol, 22:633-40.[CrossRef][ISI][Medline]

De Tomaso, A. W., Y. Saito, K. J. Ishizuka, K. J. Palmeri, and I. L. Weissman. 1998. Mapping the genome of a model protochordate. I. A low resolution genetic map encompassing the fusion/histocompatibility (Fu/HC) locus of Botryllus schlosseri. Genetics, 149:277-87.

Dodds, A. W., and S. K. Law. 1998. The phylogeny and evolution of the thioester bond-containing proteins C3, C4 and alpha 2-macroglobulin. Immunol. Rev, 166:15-26.[CrossRef][ISI][Medline]

Du Pasquier, L. 1993. Phylogeny of B-cell development. Curr. Opin. Immunol, 5:185-93.[CrossRef][ISI][Medline]

Engelhard, V. H. 1994. How cells process antigens. Sci. Am, 271:54-61.[ISI][Medline]

Figdor, C. G., Y. van Kooyk, and G. J. Adema. 2002. C-type lectin receptors on dendritic cells and Langerhans cells. Nature Rev. Immunol, 2:77-84.[CrossRef][ISI][Medline]

Flajnik, M. F., M. Kasahara, B. P. Shum, L. Salter-Cid, E. Taylor, and L. Du Pasquier. 1993. A novel type of class I gene organization in vertebrates: A large family of non-MHC-linked class I genes is expressed at the RNA level in the amphibian Xenopus. Embo. J, 12:4385-96.[ISI][Medline]

Fox-Marsh, A., and L. C. Harrison. 2002. Emerging evidence that molecules expressed by mammalian tissue grafts are recognized by the innate immune system. J. Leukoc. Biol, 71:401-9.[Abstract/Free Full Text]

Frommel, D., G. W. Litman, J. Finstad, and R. A. Good. 1971. The evolution of the immune response. XI. The immunoglobulins of the horned shark, Heterodontus francisci: Purification, characterization and structural requirement for antibody activity. J. Immunol, 106:1234-43.[Abstract/Free Full Text]

Fugmann, S. D. 2001. RAG1 and RAG2 in V(D)J recombination and transposition. Immunol. Res, 23:23-39.[CrossRef][ISI][Medline]

Fujita, T. 2002. Evolution of the lectin-complement pathway and its role in innate immunity. Nat. Rev. Immunol, 2:346-53.[CrossRef][ISI][Medline]

Gellert, M., J. E. Hesse, K. Hiom, M. Melek, M. Modesti, T. T. Paull, D. A. Ramsden, and D. C. van Gent. 1999. V(D)J recombination: Links to transposition and double-strand break repair. Cold Spring Harb. Symp. Quant. Biol, 64:161-7.[CrossRef][ISI][Medline]

Gilfillan, S., C. Benoist, and D. Mathis. 1995. Mice lacking terminal deoxynucleotidyl transferase: Adult mice with a fetal antigen receptor repertoire. Immunol. Rev, 148:201-19.[CrossRef][ISI][Medline]

Greenhalgh, P., and L. A. Steiner. 1995. Recombination activating gene 1 (Rag1) in zebrafish and shark. Immunogenetics, 41:54-54.[CrossRef][ISI][Medline]

Gross, P. S., W. Z. Al-Sharif, L. A. Clow, and L. C. Smith. 1999. Echinoderm immunity and the evolution of the complement system. Dev. Comp. Immunol, 23:429-42.[CrossRef][ISI][Medline]

Hawke, N. A., J. A. Yoder, R. N. Haire, M. G. Mueller, R. T. Litman, A. L. Miracle, T. Stuge, L. Shen, N. Miller, and G. W. Litman. 2001. Extraordinary variation in a diversified family of immune-type receptor genes. Proc. Natl. Acad. Sci. U.S.A, 98:13832-7.[Abstract/Free Full Text]

Hinds-Frey, K. R., H. Nishikata, R. T. Litman, and G. W. Litman. 1993. Somatic variation precedes extensive diversification of germline sequences and combinatorial joining in the evolution of immunoglobulin heavy chain diversity. J. Exp. Med, 178:815-24.[Abstract/Free Full Text]

Hoffmann, J. A., and J. M. Reichhart. 2002. Drosophila innate immunity: An evolutionary perspective. Nat. Immunol, 3:121-6.[CrossRef][ISI][Medline]

Holm, L., and C. Sander. 1995. DNA polymerase beta belongs to an ancient nucleotidyltransferase superfamily. Trends Biochem. Sci, 20:345-7.[CrossRef][ISI][Medline]

Hood, L., M. Kronenberg, and T. Hunkapiller. 1985. T cell antigen receptors and the immunoglobulin supergene family. Cell, 40:225-9.[CrossRef][ISI][Medline]

Hudson, A. W., and H. L. Ploegh. 2002. The cell biology of antigen presentation. Exp. Cell Res, 272:1-7.[CrossRef][ISI][Medline]

Hughes, A. L., and M. Yeager. 1997. Molecular evolution of the vertebrate immune system. Bioessays, 19:777-86.[CrossRef][ISI][Medline]

Humphreys, T., and E. L. Reinherz. 1994. Invertebrate immune recognition, natural immunity and the evolution of positive selection. Immunol. Today, 15:316-20.[CrossRef][ISI][Medline]

Janeway, C. A., Jr., P. Travers, M. Walport, and J. D. Capra. 1999. Immunobiology: The immune system in health and disease. 4th ed. Garland, London.

Ji, X., K. Azumi, M. Sasaki, and M. Nonaka. 1997. Ancient origin of the complement lectin pathway revealed by molecular cloning of mannan binding protein-associated serine protease from a urochordate, the Japanese ascidian, Halocynthia roretzi. Proc. Natl. Acad. Sci. U.S.A, 94:6340-5.[Abstract/Free Full Text]

Kasahara, M., E. C. McKinney, M. F. Flajnik, and T. Ishibashi. 1993. The evolutionary origin of the major histocompatibility complex: Polymorphism of class II alpha chain genes in the cartilaginous fish. Eur. J. Immunol, 23:2160-5.[ISI][Medline]

Komori, T., L. Pricop, A. Hatakeyama, C. A. Bona, and F. W. Alt. 1996. Repertoires of antigen receptors in Tdt congenitally deficient mice. Int. Rev. Immunol, 13:317-25.[Medline]

Kurosawa, Y., and K. Hashimoto. 1997. How did the primordial T cell receptor and MHC molecules function initially? Immunol. Cell. Biol, 75:193-6.[Medline]

Lackie, A. M. 1980. Invertebrate immunity. Parasitology, 80:393-412.[Medline]

Lanier, L. L. 1998. Follow the leader: NK cell receptors for classical and nonclassical MHC class I. Cell, 92:705-7.[CrossRef][ISI][Medline]

Lee, N., M. Llano, M. Carretero, A. Ishitani, F. Navarro, M. Lopez-Botet, and D. E. Geraghty. 1998. HLA-E is a major ligand for the natural killer inhibitory receptor CD94/NKG2A. Proc. Natl. Acad. Sci. U.S.A, 95:5199-204.[Abstract/Free Full Text]

Li, Z., J. Dai, H. Zheng, B. Liu, and M. Caudill. 2002. An integrated view of the roles and mechanisms of heat shock protein gp96-peptide complex in eliciting immune response. Front Biosci, 7:d731-51.[ISI][Medline]

Litman, G. W., M. K. Anderson, and J. P. Rast. 1999. Evolution of antigen binding receptors. Annu. Rev. Immunol, 17:109-47.[CrossRef][ISI][Medline]

Lo, D., L. Feng, L. Li, M. J. Carson, M. Crowley, M. Pauza, A. Nguyen, and C. R. Reilly. 1999. Integrating innate and adaptive immunity in the whole animal. Immunol. Rev, 169:225-39.[CrossRef][ISI][Medline]

Maenaka, K., and E. Y. Jones. 1999. MHC superfamily structure and the immune system. Curr. Opin. Struct. Biol, 9:745-53.[CrossRef][ISI][Medline]

Maier, S., C. Tertilt, N. Chambron, K. Gerauer, N. Huser, C. D. Heidecke, and K. Pfeffer. 2001. Inhibition of natural killer cells results in acceptance of cardiac allografts in CD28-/- mice. Nat. Med, 7:557-62.[CrossRef][ISI][Medline]

Marchalonis, J. J., S. F. Schluter, R. M. Bernstein, and V. S. Hohman. 1998. Antibodies of sharks: Revolution and evolution. Immunol. Rev, 166:103-22.[CrossRef][ISI][Medline]

Marino, R., Y. Kimura, R. De Santis, J. D. Lambris, and M. R. Pinto. 2002. Complement in urochordates: Cloning and characterization of two C3-like genes in the ascidian Ciona intestinalis. Immunogenetics, 53:1055-64.[CrossRef][ISI][Medline]

Masuda, Y., M. Takahashi, S. Fukuda, M. Sumii, and K. Kamiya. 2002. Mechanisms of dCMP transferase reactions catalyzed by mouse Rev1 protein. J. Biol. Chem, 277:3040-6.[Abstract/Free Full Text]

Medzhitov, R. 2001. Toll-like receptors and innate immunity. Nature Rev. Immunol, 1:135-45.[CrossRef][Medline]

Medzhitov, R., and C. A. Janeway Jr. 1999. Innate immune induction of the adaptive immune response. Cold Spring Harb. Symp. Quant. Biol, 64:429-35.[CrossRef][ISI][Medline]

Miracle, A. L., M. K. Anderson, R. T. Litman, C. J. Walsh, C. A. Luer, E. V. Rothenberg, and G. W. Litman. 2001. Complex expression patterns of lymphocyte-specific genes during the development of cartilaginous fish implicate unique lymphoid tissues in generating an immune repertoire. Int. Immunol, 13:567-80.[Abstract/Free Full Text]

More, S. H., M. Breloer, and A. von Bonin. 2001. Eukaryotic heat shock proteins as molecular links in innate and adaptive immune responses: Hsp60-mediated activation of cytotoxic T cells. Int. Immunol, 13:1121-7.[Abstract/Free Full Text]

Nappi, A. J., and E. Ottaviani. 2000. Cytotoxicity and cytotoxic molecules in invertebrates. Bioessays, 22:469-80.[CrossRef][ISI][Medline]

Nonaka, M. 2001. Evolution of the complement system. Curr. Opin. Immunol, 13:69-73.[CrossRef][ISI][Medline]

Nonaka, M., K. Azumi, X. Ji, C. Namikawa-Yamada, M. Sasaki, H. Saiga, A. W. Dodds, H. Sekine, M