Evolution of the immune system

The immune system is one of the most complex and rapidly evolving biological systems in the animal kingdom. Its central task — distinguishing self from non-self and neutralising pathogens — has been a fundamental challenge for all organisms since the origin of life, and the evolutionary solutions to this challenge span a remarkable range of molecular mechanisms, from the antimicrobial peptides of bacteria to the somatic gene rearrangement machinery of jawed vertebrates.^{1, 14} The comparative study of immune systems across the tree of life has revealed that immune defence evolved in a stepwise fashion, with ancient innate mechanisms shared by nearly all animals and more recent adaptive mechanisms restricted to particular vertebrate lineages.¹

Innate immunity: the ancient foundation

Innate immunity is the first line of defence against pathogens and is present in virtually all multicellular organisms, including plants, fungi, and animals. It relies on germline-encoded receptors that recognise conserved molecular patterns common to broad classes of microorganisms, known as pathogen-associated molecular patterns (PAMPs). These receptors, called pattern-recognition receptors (PRRs), include the Toll-like receptors (TLRs), NOD-like receptors, RIG-I-like receptors, and C-type lectins.^{3, 15}

The Toll signalling pathway, first identified in Drosophila as a regulator of dorsal-ventral axis patterning, was subsequently shown by Lemaitre and colleagues to play a central role in the fly's antifungal immune response.⁵ The discovery that mammals possess a family of homologous Toll-like receptors — with TLR4 recognising bacterial lipopolysaccharide, TLR9 recognising unmethylated CpG DNA, and other family members detecting viral RNA, flagellin, and other microbial products — demonstrated that the fundamental logic of innate immune recognition is shared between insects and mammals, indicating a common origin predating the divergence of protostomes and deuterostomes more than 600 million years ago.^{3, 4}

Antimicrobial peptides (AMPs) represent an even more ancient component of innate immunity. These small, typically cationic proteins disrupt microbial membranes and are produced by organisms ranging from bacteria (bacteriocins) to insects (defensins, cecropins) to mammals (defensins, cathelicidins). The widespread phylogenetic distribution of AMPs, and the convergent evolution of similar membrane-disrupting strategies in unrelated lineages, suggests that chemical warfare against microorganisms was among the earliest immune strategies to evolve.¹⁴

The complement system, a cascade of serum proteins that opsonises pathogens, recruits inflammatory cells, and directly lyses microbial membranes, represents an intermediate level of innate immune complexity. Core complement components have been identified in cnidarians, echinoderms, and protochordates, indicating that the lectin and alternative pathways of complement activation predate the vertebrate lineage. The classical pathway, which links complement activation to antibody binding, evolved later in conjunction with the adaptive immune system of jawed vertebrates.^{12, 13}

The origin of adaptive immunity in jawed vertebrates

Adaptive immunity — characterised by antigen-specific receptors, clonal selection, and immunological memory — appears to have arisen in the lineage leading to jawed vertebrates (gnathostomes) approximately 500 million years ago. All jawed vertebrates, from sharks to mammals, possess the hallmarks of the adaptive immune system: immunoglobulin (Ig) and T-cell receptor (TCR) genes that undergo somatic V(D)J recombination to generate an enormous diversity of antigen receptors, MHC molecules that present processed antigen fragments to T cells, and distinct populations of B and T lymphocytes that mediate humoral and cell-mediated immunity, respectively.^{1, 2}

The abrupt appearance of this system in the gnathostome lineage, with no obvious intermediate stages in the jawless vertebrates (lampreys and hagfish) or invertebrate chordates, suggested a sudden evolutionary innovation. The molecular mechanisms underlying V(D)J recombination — specifically, the recombination signal sequences (RSSs) flanking the V, D, and J gene segments and the recombination-activating genes RAG1 and RAG2 that recognise and cut at these sequences — bear a striking resemblance to the terminal inverted repeats and transposases of DNA transposons.⁶ This observation led to the RAG transposon hypothesis.

The RAG transposon hypothesis

The RAG transposon hypothesis proposes that the V(D)J recombination system originated when a DNA transposable element — carrying an ancestral RAG gene and flanked by terminal inverted repeats — inserted into an ancestral immunoglobulin-like receptor gene in an early vertebrate, splitting it into separate gene segments.⁶ The transposase activity of the RAG protein was subsequently co-opted ("domesticated") to catalyse the rearrangement of these segments during lymphocyte development, generating the diversity of antigen receptors that characterises adaptive immunity.

Several lines of evidence support this hypothesis. Agrawal, Eastman, and Schatz demonstrated in 1998 that purified RAG1 and RAG2 proteins can catalyse a transposition reaction in vitro, excising a DNA segment flanked by RSSs and inserting it into a target DNA molecule, precisely as a transposase would.⁶ This finding established that the RAG proteins retain the mechanistic capacity for transposition, even though their normal function in lymphocytes is limited to the cut-and-paste recombination of gene segments. In 2016, Huang and colleagues discovered an active RAG-like transposon, called ProtoRAG, in the genome of the lancelet Branchiostoma belcheri, an invertebrate chordate that lacks adaptive immunity. ProtoRAG consists of RAG1-like and RAG2-like genes flanked by terminal inverted repeats, and it retains the ability to excise itself and insert into new genomic locations — the first demonstration of an intact, active RAG transposon in a living organism.⁷ The discovery of ProtoRAG provided strong evidence that the RAG recombinase of jawed vertebrates is indeed derived from a transposable element that was present in the common ancestor of chordates.

Alternative adaptive immunity in jawless vertebrates

The discovery that lampreys and hagfish possess a form of adaptive immunity based on an entirely different molecular system was one of the most surprising findings in comparative immunology. Pancer and colleagues showed in 2004 that jawless vertebrates generate diverse antigen receptors not through V(D)J recombination of immunoglobulin-domain genes but through somatic rearrangement of leucine-rich repeat (LRR) modules to assemble variable lymphocyte receptors (VLRs).⁸ Lampreys produce three types of VLR (VLRA, VLRB, and VLRC) expressed on distinct lymphocyte populations that are functionally analogous to the T and B cells of jawed vertebrates. VLRB, for example, is secreted as a decameric protein that functions analogously to an antibody, binding antigens with high specificity and affinity.¹⁷

The existence of two independent adaptive immune systems — one based on immunoglobulin-domain receptors and V(D)J recombination in gnathostomes, the other based on LRR receptors and gene conversion-like assembly in lampreys and hagfish — represents a remarkable case of convergent evolution at the functional level. Both systems achieve the same fundamental outcome (enormous receptor diversity generated somatically during lymphocyte development) but through completely different molecular mechanisms, indicating that the selective advantage of adaptive immunity was so strong that it drove the independent evolution of receptor diversification strategies in two separate vertebrate lineages.^{1, 8, 17}

MHC evolution and balancing selection

The major histocompatibility complex (MHC) is a multigene family encoding cell-surface glycoproteins that bind peptide fragments derived from intracellular (MHC class I) and extracellular (MHC class II) antigens and present them to T cells, enabling the adaptive immune system to detect infected or abnormal cells. MHC genes are the most polymorphic loci in vertebrate genomes: in humans, some HLA (human leukocyte antigen) loci possess thousands of alleles, far exceeding the level of variation found at any other gene.⁹

This extraordinary polymorphism is maintained by balancing selection, driven primarily by the advantage of rare alleles in recognising novel pathogen-derived peptides. Several forms of balancing selection have been implicated, including heterozygote advantage (individuals heterozygous at MHC loci can present a wider range of pathogen peptides and therefore resist a broader spectrum of infections), negative frequency-dependent selection (rare alleles confer resistance to pathogens adapted to exploit common alleles), and fluctuating selection driven by temporal and spatial variation in pathogen communities.¹⁰

One of the most striking signatures of balancing selection at MHC loci is trans-species polymorphism, in which allelic lineages at MHC genes predate the speciation events that separated the species carrying them. Klein and colleagues demonstrated that some MHC alleles in humans are more closely related to alleles in chimpanzees or gorillas than they are to other human alleles, indicating that these allelic lineages have been maintained by balancing selection for millions of years, persisting through multiple speciation events.¹¹ Trans-species polymorphism of this depth has been documented in primates, rodents, birds, and fish, confirming that pathogen-mediated balancing selection is a pervasive and ancient force shaping MHC evolution across vertebrates.^{9, 11}

Immunoglobulin gene evolution

The immunoglobulin (Ig) superfamily is one of the largest and most diverse protein families in vertebrate genomes. The Ig domain — a characteristic beta-sandwich fold of approximately 100 amino acids — predates the adaptive immune system: Ig-domain proteins are found in invertebrates and even bacteria, where they function in cell adhesion, receptor signalling, and other non-immune contexts.¹ The adaptive immune system co-opted the Ig domain for antigen recognition, and successive gene duplications produced the multi-segment Ig and TCR loci of modern gnathostomes.

The organisation of Ig loci varies considerably across vertebrate lineages. Mammals typically possess translocon-type Ig loci, in which many V, D, and J gene segments are arrayed in tandem and a single rearrangement event selects one segment from each class. Cartilaginous fishes (sharks and rays), by contrast, possess a "cluster" organisation in which many small, independently rearranging VDJ units are arrayed across the locus.¹ The number and types of antibody classes (isotypes) also vary: mammals produce IgM, IgD, IgG, IgA, and IgE, while sharks produce IgM and IgNAR (a unique heavy-chain-only antibody), and bony fishes produce IgM, IgD, and IgT/IgZ.¹ The diversification of antibody isotypes reflects lineage-specific adaptations to different pathogen environments and mucosal architectures, with each new isotype evolving through gene duplication and divergence of the constant-region genes.^{1, 16}

Immune evolution in a coevolutionary context

The evolution of the immune system cannot be understood in isolation from the pathogens that drive it. Host-pathogen coevolution imposes perpetual selective pressure on immune genes, leading to some of the fastest rates of molecular evolution observed in any gene family. The Red Queen dynamics operating between hosts and their parasites maintain immune gene polymorphism through negative frequency-dependent selection, while arms-race dynamics drive the diversification of immune receptor repertoires and the invention of new defence strategies.^{10, 11}

The stepwise elaboration of immune complexity across the animal tree of life — from antimicrobial peptides and phagocytosis in invertebrates, through complement and TLR signalling in basal deuterostomes, to the full adaptive immune system of gnathostomes — reflects a series of evolutionary innovations, each driven by the relentless selective pressure exerted by coevolving pathogens. The co-option of a transposable element to create the V(D)J recombination system, the independent evolution of VLR-based adaptive immunity in jawless vertebrates, and the extreme polymorphism maintained at MHC loci across all jawed vertebrates all testify to the extraordinary creative power of host-pathogen coevolution in shaping one of the most intricate biological systems known.^{1, 7, 9}