The Light In The Black Box

Adela

In the first chapter of his book Darwin’s Black Box, Michael Behe lays out his fundamental purpose: to show that Darwinian evolution does not account for the molecular structure of life (Behe, 25). He goes on to argue that life depends on many molecular systems that are ‘irreducibly complex’, consisting of interlocking parts that must be all in place before they can function. The need for minimal function would preclude the existence of the constituent parts as they would be useless without one another.

Behe cites the human eye as an example: on top of the structure of the eye itself, the biochemical pathways leading up to the recognition of an image in the brain are ‘staggeringly complicated’ (Behe, 18-22). As an analogy, Behe invokes the mousetrap, which requires a minimum number of interacting parts assembled from the outset to catch any mice at all (Behe, 42-44). By the same token, an irreducibly complex organic system such as the eye must be assembled all at once. Its piece by piece evolution would, according to Behe, involve ‘unbridgeable chasms’ from molecular machine to molecular machine (Behe, 15).

Behe contends that a molecular system is irreducibly complex unless the actions of proteins and other molecules involved can be listed in full detail (Behe, 22). He demands an explanation for how the machinery evolved, but at the same time deems natural selection as insufficient to account for their development (Behe, 27-30). Thus, he attributes the fundamental biochemical pathways of life to intelligent design (Behe, 15).

Behe’s argument is fundamentally flawed. First and foremost, it is based on ignorance. Due to a lack of knowledge on the evolutionary history of molecular pathways, he makes the oversimplistic assumption that it does not exist. Also, Behe tends to focus on the current form-function relationship of molecules in cells, without recognizing that they might not have always have had that one same structure or purpose. Finally, from a scientific perspective, intelligent design is unacceptable because it is unfalsifiable. If the same demands for evidence of irreducible complexity were made of Behe, he too, would have to admit to taking logical flights of fancy in Calvin’s box (Behe, 23).

The very purpose of science is to seek answers to the unknown. And indeed, as more and more information about the actions of proteins and other molecules is discovered, we draw ever nearer to elucidating the origins of biochemical pathways. What Behe fails to recognize is that the opening of the black box is precisely the key to the apparent mysteries of the cell, and of our biochemical origins.

Contrary to the purported ‘eerie and complete silence’ meeting Behe’s demand for explanations of how evolution could have occurred (Behe, 5), the international scientific community has produced, and continues to produce, a wealth of literature regarding this topic. While he scoffs at some earlier work attempting to account for the molecular evolution of the immune system (Behe, 136-137), more evidence in support of these accounts has been uncovered since the initial publication of Darwin’s Black Box in 1996. By closely examining Behe’s critique of the immune system in Chapter 6 of his book, and providing recent and well researched counter-examples, we shall demonstrate exactly why his intelligently designed immune system cannot stand.

Simplifying Selection

The first system Behe describes in chapter 6 is the process of clonal selection, which is how the adaptive immune system detects non-self molecules and initiates a response. Antibodies are initially expressed on B cell surfaces (where they are known as B cell receptors or BCRs), and once they bind to a specific antigen, they initiate a signal cascade via a series of modifications of messenger proteins, which ultimately triggers the transcription of genes involved in B cell activation. The B cell then proliferates and differentiates into a plasma cell, which secretes free antibodies that diffuse throughout the tissues of the host organism and bind the same antigen.

Behe maintains that the process of translating the physical binding of the antibody to the B cell for specific antibody secretion is critically dependent on the membrane-bound antibody, a signal transmitter, and the secreted antibody (Behe, 125). In a world of limited resources, it would be extremely wasteful to have a cell with sub-optimal construction for its purpose. Thus, Behe argues that it would make no sense for a lone antibody to be present without the other two components, since it would hardly be able to fight a pathogenic infection by itself (Behe, 126). However, he is overemphasizing the importance of a structure-form relationship. He assumes that an antibody molecule that is incapable of switching between membrane-bound and secreted forms would be useless, but there exists evidence to the contrary. Antibodies are not the only molecules that undergo gene rearrangement to produce a diverse set of antigen receptors. T cell receptors (TCRs) carry out the same function, but are not secreted. So even if antibodies were initially restricted to the membrane-bound form, and were less efficient at binding antigens, they could still have evolved from a TCR-like gene. The ability to rearrange and the ability to switch forms could evolve in separate steps, with each step offering a selectable advantage. Although TCRs that undergo such somatic diversification have not yet been identified in the two extant orders of jawless vertebrates, several homologs α, β, γ and δ T cell antigen receptor genes are present in cartilaginous fish. It appears as if the organization of their TCR loci and the capacity to undergo junctional diversification is reminiscent of that seen in mammals (Rast et al. 1997). This common general mechanism for receptor diversification (to be elaborated on in the proceeding section) provides a clue as to how the BCR might have evolved from the TCR.

Also, four different genes encoding immunoglobulin (Ig) variable-type domains have been identified in the sea lamprey (Petromyzon marinus), one of which encodes a protein with structural similarities to mammalian VpreB molecules, including the absence of a recognizable transmembrane region, a relatively high proportion of charged amino acids in its C-terminal tail and a distinctive secretion signal peptide (Cannon et al. 2005). This molecule, believed to be the earliest ancestor of the BCR, is proof that animals once survived without the signal transmitting component of the clonal selection system.

It is also possible that the ability to rearrange evolved after the evolution of the alternative splicing pattern, resulting in (rather than as a result of) the ability to switch. This post-splicing switch would require that antigen receptors with a single specificity be effective mediators of immunity. There are several families of innate receptors that do not rearrange, called pattern recognition receptors (PRRs), which can be membrane-bound, secreted, or both, and can mount effective immune responses (Medzhitov and Janeway 1997).

A good example is hemolin, a plasma protein from lepidopteran insects. Its rapid induction to high concentrations in the hemolymph of Manduca sexta larvae and its broad specificity for binding to different types of micro-organisms suggest its role as a PRR that participates in detection and elimination of a variety of lepidopteran pathogens (Yu and Kanost 2002). Significantly, sequence analysis shows that while hemolin is not a precursor to immunoglobulins (the type of Ig-domain it possesses is not found in vertebrates), it is structurally and functionally related to cell-adhesion molecules. This dual function for immune response and cell-adhesion suggests that immune molecules arose from cell-adhesion molecule precursors.

By demonstrating the existence of proteins that do not undergo rearrangement, but are able to switch forms, or that undergo rearrangement without switching forms, Behe’s claim of the need for an ‘irreducible’ clonal selection system for minimal function (Behe, 45) is disproved.

Reconciling Recombination

Next, Behe targets the V(D)J gene recombination system (RS), used by all known jawed vertebrates (gnathosomes) to assemble their antigen receptor genes. T and B lymphocyte progenitors rearrange different sets of prototypic Ig variable (V), diversity (D), and joining (J) gene segments to generate antigen binding regions of their respective T or B cell receptors. These regions are further diversified by the enzymatic addition of nonencoded nucleotides in the joints created during V(D)J assembly. The random nature of this process generates many different receptors for invading pathogens.

Behe holds that three components - genes, signal sequences, and cell - are essential for the the V(D)J RS to generate these millions of antibodies, and that there is no feasible pathway by which the system could evolve (Behe, 131). This, however, is not true. In scoffing at the speculation that “a gene from a bacterium might have luckily been transferred to an animal… the protein coded by the gene could itself rearrange genes… in the animal’s DNA there were signals that were near antibody genes,” (Behe, 137) he grossly underestimates the power of natural selection in stepwise, gradual evolution.

Why, one might ask, would natural selection act on DNA in such a specific manner? What selective forces would be acting on them? The answer lies in a truism from Behe himself: “Threats of aggression can come in all shapes and sizes, so defences have to be versatile.” (Behe, 117) For example, APOBECs, or single-stranded DNA-editing enzymes, are known to cause hypermutation of HIV. In particular, a HIV-encoded virion infectivity factor (Vif) targets APOBEC3G for destruction, causing rapid fixation of mutations that alter amino acids at the protein-protein interface. APOBEC genes have been subject to strong positive selection that appears more ancient than, and is likely only partially caused by, modern lentiviruses (Sawyer et al. 2004). The need for defense against foreign antigens is thus good grounds for RNA and DNA editing in primate genomes.

In addition, there exists in the form of the huge Ig superfamily a breeding ground of potential for genetic rearrangement. Ig domains are extremely versatile and used by TCR, BCR, and major histocompatibility complex (MHC) proteins, invariant activating and inhibiting receptors on natural killer (NK) cells, F_C receptors on phagocytes, and other cell surface molecules that regulate T and B cell interactions with each other and with antigen presenting cells. There exist several well-studied invertebrate immune systems involving Ig domain manipulation that bear remarkable similarity to the V(D)J RS that might shed some light on the mechanisms of how a monomorphic germline receptor gives rise to a somatically variable antigen recognition system.

Amphioxus (Branchiostoma floridae, a cephalochordate representing the most phylogenetically proximal sister group to the vertebrates) expresses variable region-containing chitin binding proteins (VCBPs), which have been implicated in innate immunity against intestinal pathogens. VCBPs possess two tandem V region domains fused to a single, C-terminal chitin-binding domain (CBD), and are structurally closely related to the rearranging antigen-binding receptors of jawed vertebrates. Not only are the constituent VCBP domains prototypic of Igs, VCBP3 has particularly been shown to share structural similarities with the more recently derived antigen receptors in folding and three-layer packing, leading to characteristic dimerization, and creation of advantageous binding properties. In addition, it is suggested that VCBPs or their forerunners are adapted to effecting immune recognition through the clustering of hypervariable residues at the solvent-accessible CBD comprising both V domains. As chitin is a very abundant polymer of N-acetylglucosamine in sea water, it is possible that the CBD of a VCBP can interact with chitin degradation products in ingested seawater and multimerize VCBPs as V region-based immune receptors. Natural selection for hypervariable clusters might then cause stabilization of diversified genes encoding V regions into the germline before the innovation of a basic mechanism of somatic reorganization in vertebrates (Prada 2006).

Somatic variation with individualization of the response can also be seen in the modulation of gene expression via alternative splicing in the fibrinogen related proteins (FREPs) of the freshwater snail (Biomphalaria glabrata). They are made of one or two amino-terminal Ig domains and a carboxyl-terminal fibrinogen domain. Recent studies have shown that these genes are diversified extensively at the genome level in individual animals through both point mutations and recombinatorial processes in somatic tissues. In addition, individual variation of another kind, whether due to gene conversion or some other mechanism, is observed in FREPs where a considerably larger number of sequences can be seen at the somatic level than at the genomic level (Zhang 2004). While it is unlikely that the current V(D)J gene structure arose from a common ancestor shared with FREP-employing protostomes, a completely plausible mechanism of genetic rearrangement occurring separately from somatic variation (that could also apply to the V(D)J RS) has been demonstrated, and may become clearer when other deuterostome genomes are annotated.

Moreover, the recent discovery of variable lymphocyte receptors (VLRs) containing leucine-rich repeat (LRR) modules with highly variable amino acid sequences explains the failure to find molecules of adaptive immunity including Ig-like structures, V(D)J gene segments or MHC genes in surviving jawless vertebrates (agnathans, sister group and nearest living phylogenetic relatives of gnathosomes), despite their displaying adaptive immune responses, such as accelerated rejection of secondary skin allografts, elevated agglutination in response to antigens. The sole VLR gene encodes the signal peptide, parts of the N- and C-terminals, and the invariant stalk region of the receptor. It is flanked by many cassettes encoding one, two or three LRR modules that are randomly incorporated into the germline VLR gene via a multistep assembly, thus creating a repertoire comparable in size to that of antibodies in gnathosomes. LRR modules lack recombinatorial signal sequences; instead, short homology sequences may serve as anchorage sites for assembly, suggesting that gene conversion is a probable mechanism for VLR diversification (Alder 2005).

It is hypothesized that the V(D)J RS arose in ostracoderms (agnathans with dermal skeletons, a gnathosome ancestor (Cooper and Alder 2006). There are two possible pathways by which agnathans and gnathosomes diverged: firstly, VLR RS might have evolved in an ancestor shared only by lampreys and hagfishs, and other agnathans fell prey to pathogen-mediated extinction; secondly, VLR RS might have evolved in an ancestor common to both agnathans and gnathosomes, and V(D)J RS was subsequently acquired, leading to existence of receptors of both self and non-self receptors and mixed lymphocyte activation, thus favoring the loss of either system.

The introduction of the Recombination Activating Gene (RAG) enzyme as a transposase into a VJ gene target appears to be a hallmark of generating Ig-type diversity in jawed vertebrates. Homologs of vertebrate Rag1 and Rag2 have been found in the purple sea urchin (Strongylocentrotus purpuratus, an echinoderm invertebrate). In combination with the apparent absence of V(D)J recombination in echinoderms, this finding strongly suggests that linked Rag1- and Rag2-like genes were already present and functioning in a different capacity in the common ancestor of living deuterostomes, and that their specific role in the adaptive immune system was acquired later in an early jawed vertebrate (Fugmann et al. 2006). The identification of RAG-like genes outside of the context of an adaptive immune system thus supports the hypothesis of transposase-mediated junctional rearrangement, or in other words, a mechanism for incorporation of junctional diversification into the V(D)J RS.

Ultimately, there are sufficient grounds for natural selection, as well as a variety of mechanisms by which it may occur to account for the incipient stages of development of complicated gene recombination systems, and to refute Behe’s claim of irreducable complexity.

Complimenting Complement

Finally, Behe addresses the complement system, which is responsible for opsonization of antigens, initiating phagocytosis, and assembly of the membrane attack complex. It is composed of very intricate pathways, requiring precise sequences of activating molecules and regulation by factor molecules. The absence or malfunction of just one protein would disrupt the cascade and render the system ineffective. One of Behe’s main concerns is the centrality of the C3 molecule in the complement system: without it or the molecules that activate it, the lectin, alternative, and classical pathways would all cease to function (Behe, 135). Yet again Behe’s motif is repeated: the myriad requirements of the structure-function relationship block gradual, Darwinian-style evolution (Behe, 138).

But Behe fails to realize that the current components of the complement cascade might not always have been present, nor have played the same roles that they do today. It turns out that certain parts and functions of the vertebrate complement system were absent in their ancestors, but the presence of ancestral homologous proteins suggests their gradual, stepwise accumulation through evolution.

Urochordates do not have a C2 gene that encodes C3 convertase. Instead, they employ activating enzymes knowns as MASPs, which appear to be related by sequence homology to the C1 enzymes of the classical pathway. Analysis of the human MASP-1/3 gene, which encodes two proteases of the lectin pathway, has revealed alternatively used serine-protease-encoding regions for the gene's two protein products. Phylogenetic studies indicate that one arose by retrotransposition early in vertebrate evolution, and after gene duplication, one of the duplicates lost the downstream serine-protease-encoding region, resulting in a gene of the MASP-2 type. Further gene duplications might have generated the C1r and C1s genes, resulting in the four genes found in humans (Dahl et al. 2001).

The proposed pathway of gene evolution provides a feasible account of how the complement system came to straddle innate and adaptive immunity. By adding components such as C2 and C4 onto the invertebrates' cascade, they retained the lectin pathway, but also linked it to adaptive immunity via employment of immunoglobulins in the classical pathway. It is then possible that the C3 convertase in the vertebrate lectin pathway, C2b, gradually took over the function of the invertebrate MASP, such that now the vertebrate complement system is critically dependent on C2b.

Evidence for ancestral homologs of complement component genes with characteristic domains abounds. They have been found not only in jawed vertebrates but also in jawless fish and non-vertebrate deuterostomes (Zarkadis et al. 2001). The identification of homologues of complement components C3 and factor B in sea urchins is another indication that invertebrates possess a simple complement system homologous to the alternative pathway in higher vertebrates (Smith et al. 1998). Also, studies of the crystal structures of native C3 and its final major proteolytic fragment C3c have revealed thirteen domains, nine of which were unpredicted, suggesting that the proteins of the α2-macroglobulin family evolved from a core of eight homologous domains in vertebrate ancestors (Janssen et al. 2005).

Behe defines an irreducibly complex system as one that “cannot be produced gradually by slight, successive modifications of a precursor system, since any precursor to an irreducibly complex system is by definition nonfunctional”. The complement cascade has been shown to be composed of several parts that came to be integrated together in jawed vertebrates, belying Behe's claim and proving that the complement pathways do not present a challenge to step-by-step evolution, but rather, a fitting tribute.

“Rome was not built in a day”; neither were the highly specialized proteins involved in immunity. Having examined the systems of clonal selection, Ig gene recombination and complement in some detail, we have hopefully arrived at a better understanding of how the various components of the immune system might have evolved to work in tandem. Evidently, in contrast to Behe’s claims, the complexity of life’s foundation has hardly ‘paralyzed science’s attempt to account for it’ (Behe, 5), but rather, has spurred on much research, which every day is providing more knowledge about the intricate, but definitely not irreducible, molecular intricacies of life; and will, ultimately, uncover the light at the bottom of the big black box.

REFERENCES

M. J. Behe. 2006.Darwin's Black Box. 10th anniv. ed. New York: Free Press.

J. P. Rast, M. K. Anderson, S. J. Strong, et al. 1997. α, β, γ and δ T cell antigen receptor genes arose early in vertebrate phylogeny. Immunity 6:1–11.

John P. Cannon, Robert N. Haire, Zeev Pancer, M. Gail Mueller, Diana Skapura, Max D. Cooper and Gary W. Litman. Variable domains and a VpreB-like molecule are present in a jawless vertebrate. 2005. Immunogenetics Volume 56, Number 12.

R. Medzhitov, C. A. Jr. Janeway. 1997. Innate immunity: the virtues of a nonclonal system of recognition. Cell 3:295-8.

Xiao-Qiang Yu and Michael R. Kanost. Binding of hemolin to bacterial lipopolysaccharide and lipoteichoic acid. 2002. Eur. J. Biochem. 269:1827-1834.

S. L. Sawyer, M. Emerman, H. S. Malik. Ancient adaptive evolution of the primate antiviral DNA-editing enzyme APOBEC3G. 2004. PLoS Biol. 2(9):E275.

José A Hernández Prada, Robert N Haire, Marc Allaire, Jean Jakoncic, Vivian Stojanoff, John P Cannon, Gary W Litman & David A Ostrov. 2006. Ancient evolutionary origin of diversified variable regions demonstrated by crystal structures of an immune-type receptor in amphioxus. Nature Immunology 7:875-882.

Si-Ming Zhang, Coen M. Adema, Thomas B. Kepler, Eric S. Loker. 2004. Diversification of Ig Superfamily Genes in an Invertebrate. Science 305:251-254.

Matthew N. Alder, Igor B. Rogozin, Lakshminarayan M. Iyer, Galina V. Glazko, Max D. Cooper, Zeev Pancer. 2005. Diversity and Function of Adaptive Immune Receptors in a Jawless Vertebrate. Science 310:1970–1973.

Sebastian D. Fugmann, Cynthia Messier, Laura A. Novack, R. Andrew Cameron, and Jonathan P. Rast. 2006. An ancient evolutionary origin of the Rag1/2 gene locus. PNAS Vol. 103 10:3728-3733.

I. K. Zarkadis, D. Mastellos, J. D. Lambris. Phylogenetic aspects of the complement system. 2001. Dev Comp Immunol. 25(8-9):745-62.

L. C.Smith, C. S. Shih, S. G. Dachenhausen. 1998. Coelomocytes express SpBf, a homologue of factor B, the second component in the sea urchin complement system. J Immunol. 161(12):6784-93.

Bert J. C. Janssen, Eric G. Huizinga, Hans C. A. Raaijmakers, Anja Roos, Mohamed R. Daha, Kristina Nilsson-Ekdahl , Bo Nilsson and Piet Gros. 2005. Structures of complement component C3 provide insights into the function and evolution of immunity. Nature 437:505-511.

Masaru Nonaka, Fumiko Yoshizaki. 2004. Primitive complement system of invertebrates. Immunological Reviews Vol. 198: 203.

M. R. Dahl, S. Thiel, M. Matsushita, T. Fujita, A. C.Willis, T. Christensen, T. Vorup-Jensen, J. C. 2001. MASP-3 and its association with distinct complexes of the mannan-binding lectin complement activation pathway. Jensenius. Immunity 15(1):127-35.