To Build New Animals, No New Genetic Information Needed? More in Reply in Charles Marshall
In my previous article responding to Charles Marshall, I argued that his review of Darwin's Doubt in the journal Science ("When Prior Belief Trumps Scholarship") illustrates what has become all too common in the defense of contemporary evolutionary theory: the tendency to affirm as true what evolutionary theory requires, even if that contradicts what we know from experiment and observation about how biological systems actually work. Now I will show that in order to rebut the central argument of Darwin's Doubt, Marshall must also deny (or at least push from view) what we know about what new forms of animal life require as a condition of their existence.
In Darwin's Doubt, I argue that intelligent design provides the best explanation for the origin of the genetic (and epigenetic) information necessary to produce the novel forms of animal life that arose in the Cambrian period. To his credit, and unlike other critics of the book, Marshall addresses this, the main argument of the book, and attempts to refute it. To do so, however, he does not show that any of the main materialistic evolutionary mechanisms can produce the information necessary to build the Cambrian animals. Instead, Marshall disputes my claim that significant amounts of new genetic information (and many new protein folds) would have been necessary to build these animals. Specifically, Marshall claims that "rewiring" of dGRNs would have sufficed to produce new animals from a set of preexisting genes. As he argues:
[Meyer's] case against the current scientific explanations of the relatively rapid appearance of the animal phyla rests on the claim that the origin of new animal body plans requires vast amounts of novel genetic information coupled with the unsubstantiated assertion that this new genetic information must include many new protein folds. In fact, our present understanding of morphogenesis indicates that new phyla were not made by new genes but largely emerged through the rewiring of the gene regulatory networks (GRNs) of already existing genes.1
Yet Marshall's understanding of how animal life originated is problematic for several reasons.
Rewiring Requires Information
First, "rewiring" genetic circuitry would require reconfiguring the temporal and spatial expression of genetic information. Such reconfiguring would entail fixing certain material states and excluding others. Thus, it would constitute an infusion of new information (in the most general theoretical sense) into the biosphere.2 To see why, consider changing a wiring diagram representing a developmental gene regulatory network. Just altering the diagram representing the network to produce a new diagram representing a new network would require changing the arrangement of "nodes" (representing genes) and "edges" (representing interactions between genes and gene products). Changing the arrangement of these elements in order to produce a new network would constitute adding information into the diagram depicting the system. In the same way, changing the arrangements of genetic elements themselves in an actual network would also require informative changes to the arrangement of the network. Thus, Marshall's "rewiring" proposal does not eliminate the need for new information to build the Cambrian animals. Rather, it tacitly invokes additional information of a different, though perhaps partially non-genetic, kind.
In any case, altering the temporal and spatial expression of pre-existing genetic elements assuredly would require the addition of new taxon-specific genes or gene products, and thus, new genetic information. Indeed, experiments on dGRNs in modern representatives of the animal phyla show that different organisms use taxon-specific DNA-binding proteins to regulate the expression of genetic data files. For instance, contrary to theoretical expectations,3 the morphogen Bicoid, essential for normal anterior-posterior body plan specification in Drosophila, is found only in the cyclorrhaphan flies.4 Similarly, the body plan of the freshwater polyp Hydra is specified by taxonomically restricted proteins.5 Nor are these isolated cases. The remarkable disparity of animal morphologies at the macroscopic (i.e., anatomical or body plan) level tends to correspond to differences at other levels (i.e., the microscopic or molecular). Moreover, studies of "evo-devo" model systems have repeatedly revealed that the cell and tissue specification programs that generate distinctive animal morphologies depend upon taxon-specific regulatory factors (proteins and RNAs). As Oliveri and Davidson note:
...the specification apparatus very frequently also includes transcriptional repressors, which, within the specified spatial domain, target key regulatory genes whose expression is required for alternative regulatory states that could have been available to these cells. This is a so-called "exclusion effect," and numerous examples can be found across species...In each developmental case, the identity of the specific transcription factor that executes the repression is distinct, as are the specifically excluded target transcription factors. The design is the same, the biochemical actors diverse.6
Of course, building these species-specific transcription factors necessary to animal development requires genetic information. And the origin of these proteins in the first place would have required the origin of new genetic information.
Ubiquitous ORFan Genes
Second, recent genomic studies of many animals representing phyla that first arose in the Cambrian show that these animals depend upon many unique genes not present in any other taxa. Moreover, these genes perform many functions besides just specifying body plan development. These sequences, known as taxonomically restricted or "ORFan" genes, are ubiquitous in all animal life and represent 10% or more of the genomes of each species that scientists have investigated.7
The presence of ORFan genes in all sequenced present-day animal genomes -- and, indeed, in all life8 -- suggests that the genomes of Cambrian animals would have likely contained many ORFan genes as well. That, in turn, suggests that a considerable amount of new genetic information not present in simpler Precambrian organisms would have originated before or during the Cambrian radiation in order to build the unique features of the first Cambrian animals. Moreover, even if some universal Precambrian genome originally contained all the genes that later became taxonomically restricted, whatever process distributed these genes to some lineages, but not others, necessarily involved the addition of new information into the biosphere.9
A Telling Admission
Interestingly, in his review and especially when writing elsewhere, Marshall acknowledges the need for new genes and genetic information in order to produce the Cambrian animals. For example, in a 2006 paper entitled "Explaining the Cambrian 'Explosion' of Animals," he noted that: "Animals cannot evolve if the genes for making them are not yet in place. So clearly, developmental/genetic innovation must have played a central role in the radiation."10 Later in the same paper he argues that: "It is also clear that the genetic machinery for making animals must have been in place, at least in a rudimentary way, before they could have evolved."11 Marshall insists that Hox genes, in particular, must have played a necessary causal role in producing the explosion, a point that he also makes in another paper where he explains that developmental considerations "point to the origin of the bilaterian developmental system, including the origin of Hox genes, etc., as the primary cause of the 'explosion.'"12 While in these papers Marshall also emphasizes the importance of rewiring gene regulatory networks to generate new body plans, he clearly acknowledges that new genes would be necessary to produce new animals.
New Animals Require Many New Cell Types and Specialized Proteins
Of course, building the Metazoa (multi-cellular animals) would not have just required new Hox genes, ORFan genes, or genes for building new regulatory (DNA-binding) proteins. Instead, the evolutionary process would need to produce a whole range of different proteins necessary to building and servicing the specific forms of animal life that arose in the Cambrian period. In Darwin's Doubt I note, for example, that the first arthropods would have likely required genes for building the complex protein lysyl oxidase.13 Why? Because what we know from studies of modern arthropods shows that this protein is necessary to support the stout body structure of arthropod exoskeletons.14 Similarly, building Metazoa requires specialized proteins (and metabolic pathways) to produce the kind of extra-cellular matrices that allow developing animals to knit cells into tissues, tissues into organs, and organs and tissues into fully developed animals. Furthermore, different forms of complex animal life exhibit unique cell types and typically each cell type depends upon other specialized or dedicated proteins. As I wrote in Darwin's Doubt:
[new] complex animals [such as arose in the Cambrian period] require more cell types to perform their more diverse functions. Arthropods and mollusks, for example, have dozens of specific tissues and organs, each of which requires "functionally dedicated," or specialized, cell types. These new cell types, in turn, require many new and specialized proteins. An epithelial cell lining a gut or intestine, for example, secretes a specific digestive enzyme. This enzyme requires structural proteins to modify its shape and regulatory enzymes to control the secretion of the digestive enzyme itself. Thus, building novel cell types typically requires building novel proteins, which requires assembly instructions for building proteins -- that is, genetic information.15
Thus, our present observations of animals representing the phyla that first arose in the Cambrian show that these animals would have needed many specialized proteins: proteins for building extracellular matrices or exoskeletons, for facilitating adhesion, for regulating development, for building specialized tissues or structural parts of specialized organs, for servicing gut cells, for producing eggs and sperm as well as many other distinctive functions and structures of individual metazoans. Obviously, these proteins would have had to arise sometime in the history of life. Since most of the Metazoa first arose in the Cambrian explosion, it is reasonable to infer that the proteins necessary to sustain those forms of animal life also arose around that time or just before.
Begging the Central Question
Although Marshall characterizes my claim that new Cambrian animals would have required new genetic information and new protein folds as "unsubstantiated," he doesn't actually dispute the need for genetic information in order to build the proteins required by each new form of metazoan life. Instead, he only seems to dispute that all that information arose during the Cambrian explosion itself. Indeed, in both his technical publications and his review of Darwin's Doubt, Marshall simply assumes that most of the genetic information necessary to build the Cambrian animals already existed before the Cambrian explosion. In fact, he seems to presuppose the existence of what Susumu Ohno called a "pananimalian genome,"16 a nearly complete set of the genes necessary to build Cambrian animals within some phenotypically simpler, ur-metazoan ancestor. Thus, he states the new animal phyla "emerged through the rewiring of the gene regulatory networks (GRNs) of already existing genes."17 The article "The Causes of the Cambrian Explosion," which accompanies Marshall's review of my book in Science, also presupposes such a universal gene toolkit and suggests that it might have arisen 100 million years or more before the explosion of animal life in the Cambrian period.18
Nevertheless, this question-begging assumption does not solve the central problem posed by Darwin's Doubt -- that of the origin of the genetic (and epigenetic) information necessary to produce the Cambrian animals. It merely pushes the problem back several tens or hundreds of millions of years, assuming that such a universal genetic toolkit ever existed. (Marshall also makes no attempt to rebut my argument about the inability of the mutation/selection mechanism to generate new epigenetic information, a problem that has led other prominent evolutionary biologists to express skepticism about the adequacy of the neo-Darwinian mechanism.19) In any case, Marshall does not explain how the neo-Darwinian mechanism could have overcome the combinatorial search problem described in Darwin's Doubt to produce even the new genetic information necessary to build new proteins and Cambrian animals.
Readers of the book will recall my discussion, in Chapters 9 and 10, of recent mutagenesis experiments. These experiments have established the extreme rarity of functional genes and proteins among the many (combinatorially) possible ways of arranging nucleotide bases or amino acids within their corresponding "sequence spaces." Readers will also recall that the rarity of functional genes and proteins within sequence space makes it overwhelmingly more likely than not that a series of random mutation searches will fail to generate even a single new gene or protein fold within available evolutionary time. This extreme rarity also helps to explain why mathematical biologists, using standard population genetics models, are calculating exceedingly long waiting times (well in excess of available evolutionary time) for the production of new genes and proteins when producing such genes or proteins requires even a few coordinated mutations.20
For these reasons, defining the Cambrian explosion as a 25 million year event, as Marshall does, instead of a 10 million year event, as many other Cambrian experts do (and as I do in Darwin's Doubt), makes no appreciable difference in solving the problem of the origin of genetic information -- such is the extreme rarity of functional bio-macromolecules within their relevant sequence spaces. Nor, for that matter, does positing the origin of a complete set of genes (that is, many more than just one) for building all the Cambrian animals 100 million years before the Cambrian explosion. That merely pushes the problem back and raises other problems such as (a) explaining exactly what selective advantage all these genes for building new animals would have had before they were actually used to build the diverse animals that arose in the Cambrian and (b) how the maintenance of this overly complex genome could have avoided exacting a huge energetic and fitness cost on its host organism, and thus the effects of purifying selection over 100 million years of evolutionary time.
In any case, the experimentally based calculations in Darwin's Doubt show that neither ten million, nor several hundred million years would afford enough opportunities to produce the genetic information necessary to build even a single novel gene or protein, let alone all the new genes and proteins needed to produce new animal forms. Indeed, neither stretch of time is sufficient to allow the mutation/selection process to search more than a tiny fraction of the relevant sequence spaces. Marshall's review does not even allude to a solution to this longstanding mathematical,21 and now experimentally based,22 challenge to the efficacy of the neo-Darwinian mechanism. Instead, his proposal merely presupposes the prior existence of the genetic information necessary to produce the Cambrian animals.
In my next article, I'll address Marshall's claim that the positive argument for intelligent design that I make in the book actually constitutes a wholly negative or critical "God-of-the gaps" argument.
(1) Charles R. Marshall, "When Prior Belief Trumps Scholarship," Science 341 (September 20, 2013): 1344.
(2) See: Claude Shannon, "A Mathematical Theory of Communication," Bell System Technical Journal 27 (1948): 370-423, 623-29.
(3) Nicolas Rasmussen, "A New Model of Developmental Constraints as Applied to the Drosophila System," Journal of Theoretical Biology 127 (1987): 271-99.
(4) David Rudel and Ralf Sommer, "The evolution of developmental mechanisms," Developmental Biology 264 (2003): 15-37; p. 25: "Phylogenetic evidence suggests that bcd may be a new innovation in the anterior positional information gene network during the evolution of Dipterans. Despite repeated attempts, it has not been possible to clone bcd homologues outside of the Cyclorraphan flies (Stauber et al., 1999). Additionally, bcd is not present in the Antennapedia complex of the flour beetle Tribolium castaneum (Brown et al., 2002). This has caused speculation that bcd may have evolved late in the evolution of the Dipterans."
(5) Konstantin Khalturin, Friederike Anton-Erxleben, Sylvia Sassmann, Jörg Wittlieb, Georg Hemmrich, Thomas C. G. Bosch, "A Novel Gene Family Controls Species-Specific Morphological Traits in Hydra," PLoS Biology 6 (2008): e278.
(6) Paolo Oliveri and Eric H. Davidson, "Built to Run, Not to Fail," Science 315 (2007): 1510-11 (emphasis added).
(7) Amanda K Gibson, Zach Smith, Clay Fuqua, Keith Clay and John K. Colbourne, "Why so many unknown genes? Partitioning orphans from a representative transcriptome of the lone star tick Amblyomma americanum," BMC Genomics 14 (2013): 135.
(8) Konstantin Khalturin, Georg Hemmrich, Sebastian Fraune, René Augustin and Thomas C.G. Bosch, "More than just orphans: are taxonomically-restricted genes important in evolution?" Trends in Genetics 25 (2009): 404-13.
(9) Marshall seems to have an idiosyncratic view of animal evolution, depicting the evolution of animals as a reductive process in which pre-existing genetic information from a universal Precambrian gene set is selectively lost to some lineages but not to others. This contrasts markedly with a more standard neo-Darwinian view in which form and information gradually accumulate over time.
(10) Charles R. Marshall, "Explaining the Cambrian 'Explosion' of Animals," Annual Reviews of Earth and Planetary Sciences 34 (2006): 355-84.
(12) Charles R. Marshall, "Nomothetism and Understanding the Cambrian 'Explosion,'" Palaois 18: 195-96 (June, 2003).
(13) See Darwin's Doubt, p. 191.
(14) Susumu Ohno, "The notion of the Cambrian pananimalia genome," Proceedings of the National Academy of Sciences, U.S.A. 93 (1996): 8475-78.
(15) Darwin's Doubt, p. 162.
(16) Susumu Ohno, "The notion of the Cambrian pananimalia genome," Proceedings of the National Academy of Sciences, U.S.A. 93 (1996): 8475-78.
(17) Charles R. Marshall, "When Prior Belief Trumps Scholarship," Science 341 (September 20, 2013): 1344 (emphasis added).
(18) M. Paul Smith and David A. T. Harper, "Causes of the Cambrian Explosion," Science 341 (September 20, 2013): 1355-56. (Smith and Harper propose "an apparent >100-million-year gap between the evolutionary innovation and its consequences.")
(19) Gerd B. Müller and Stuart A. Newman, "Origination of Organismal Form: The Forgotten Cause in Evolutionary Theory."In: Origination of Organismal Form: Beyond the Gene in Developmental and Evolutionary Biology, Edited by Gerd B. Müller and Stuart A. Newman. Cambridge, MA: MIT Press, 2003, 7-8.
(20) See Darwin's Doubt, Chapters 9-12.
(21) Murray Eden, "Inadequacies of Neo-Darwinian Evolution as a Scientific Theory." In: Mathematical Challenges to the Neo-Darwinian Interpretation of Evolution, edited by P. S. Moorhead and M. M. Kaplan, 9-11. Wistar Institute Symposium Monograph. New York: Liss, 1967. Marcel Schützenberger, "Algorithms and the Neo-Darwinian Theory of Evolution." Also in: Mathematical Challenges to Neo-Darwinian Theory of Evolution. 73-80.
(22) John Reidhaar-Olson, and Robert Sauer. "Functionally Acceptable Solutions in Two Alpha-Helical Regions of Lambda Repressor." Proteins: Structure, Function, and Genetics 7 (1990): 306-16. Douglas D. Axe, "Estimating the Prevalence of Protein Sequences Adopting Functional Enzyme Folds," Journal of Molecular Biology 341 (2004): 1295-1315.