Darwin's Doubt: The Explosive Origin of Animal Life and the Case for Intelligent Design

by Stephen C. Meyer

recent discoveries about the role of epigenetic information in animal development pose a formidable challenge to the standard neo-Darwinian account of the origin of these body plans—perhaps the most formidable of all.

neo-Darwinism lacks an explanation for the origin of organismal form precisely because it cannot explain the origin of epigenetic information.

I asked Wells why developmental biology was so important to evolutionary theory and to assessing neo-Darwinism. I’ll never forget his reply. “Because” he said, “that’s where the whole theory is going to unravel.”

Shannon information

The arrangements of nucleotides in DNA or of amino acids in a protein are highly improbable and thus contain large amounts of Shannon information.

Similarly, animal body plans represent, not only highly improbable, but also highly specific arrangements of matter. Organismal form and function depend upon the precise arrangement of various constituents as they arise during, or contribute to, embryological development.

Perhaps because the information-carrying capacity of the gene can be so easily measured, biologists have often treated DNA, RNA, and proteins as the sole repositories of biological information. Neo-Darwinists have assumed that genes possess all the information necessary to specify the form of an animal. They have also assumed that mutations in genes will suffice to generate the new information necessary to build a new form of animal life.

Many biologists no longer believe that DNA directs virtually everything happening within the cell. Developmental biologists, in particular, are now discovering more and more ways that crucial information for building body plans is imparted by the form and structure of embryonic cells, including information from both the unfertilized and fertilized egg.

Other sources of information must help arrange individual proteins into systems of proteins, systems of proteins into distinctive cell types, cell types into tissues, and different tissues into organs. And different organs and tissues must be arranged to form body plans.

Two analogies may help clarify the point. At a construction site, builders will make use of many materials: lumber, wires, nails, drywall, piping, and windows. Yet building materials do not determine the floor plan of the house or the arrangement of houses in a neighborhood.

Eukaryotic cells have internal skeletons to give them shape and stability. These “cytoskeletons” are made of several different kinds of filaments including those called the “microtubules.” The structure and location of the microtubules in the cytoskeleton influence the patterning and development of embryos. Microtubule “arrays” within embryonic cells help to distribute essential proteins used during development to specific locations in these cells. Once delivered, these proteins perform functions critical to development, but they can only do so if they are delivered to their correct locations with the help of preexisting, precisely structured microtubule or cytoskeletal arrays (see Fig. 14.3). Thus, the precise arrangement of microtubules in the cytoskeleton constitutes a form of critical structural information. These microtubule arrays are made of proteins called tubulin, which are gene products. Nevertheless, like bricks that can be used to assemble many different structures, the tubulin proteins in the cell’s microtubules are identical to one another. Thus, neither the tubulin subunits, nor the genes that produce them, account for the differences in the shape of the microtubule arrays that distinguish different kinds of embryos and developmental pathways. Instead, the structure of the microtubule array itself is, once again, determined by the location and arrangement of its subunits, not the properties of the subunits themselves. Jonathan Wells explains it this way: “Wh...

Another cell structure influences the arrangement of the microtubule arrays and thus the precise structures they form and the functions they perform. In an animal cell, that structure is called the centrosome (literally, “central body”), a microscopic organelle that sits next to the nucleus between cell divisions in an undividing cell. Emanating from the centrosome is the microtubule array that gives a cell its three-dimensional shape and provides internal tracks for the directed transport of organelles and essential molecules to and from the nucleus.16 During cell division the centrosome duplicates itself. The two centrosomes form the poles of the cell-division apparatus, and each daughter cell inherits one of the centrosomes; yet the centrosome contains no DNA.17 Though centrosomes are made of proteins—gene products—the centrosome structure is not determined by genes alone.

When messenger RNAs are transcribed, their protein products must be transported to the proper locations in embryonic cells in order to function properly. Directed transport involves the cytoskeleton, but it also depends on spatially localized targets in the membrane that are in place before transport occurs. Developmental biologists have shown that these membrane patterns play a crucial role in the embryological development of fruit flies.

For example, early embryo development in the fruit fly Drosophila melanogaster requires the regulatory molecules Bicoid and Nanos (among others). The former is required for anterior (head) development, and the latter is required for posterior (tail) development.19 In the early stages of embryological development, nurse cells pump Bicoid and Nanos RNAs into the egg. (Nurse cells provide the cell that will become the egg—known as the oocyte—and the embryo with maternally encoded messenger RNAs and proteins.) Cytoskeletal arrays then transport these RNAs through the oocyte, where they become attached to specified targets on the inner surface of the egg.20 Once in their proper place—but only then—Bicoid and Nanos play critical roles in organizing the head-to-tail axis of the developing fruit fly. They do this by forming two gradients (or differential concentrations), one with Bicoid protein most concentrated at the anterior end and another with Nanos protein most concentrated at the posterior end. Insofar as both of these molecules are RNAs—that is, gene products—genetic information plays an important role in this process. Even so, the information contained in the bicoid and nanos genes does not by itself ensure the proper function of the RNAs and proteins for which the genes code. Instead, preexisting membrane targets, already positioned on the inside surface of the egg cell, determine where these molecules will attach and how they will function. These membrane targets provid...

Membrane patterns can also provide epigenetic information by the precise arrangement of ion channels—openings in the cell wall through which charged electrical particles pass in both directions. For example, one type of channel uses a pump powered by the energy-rich molecule ATP to transport three sodium ions out of the cell for every two potassium ions that enter the cell. Since both ions have a charge of plus one (Na+, K+), the net difference sets up an electromagnetic field across the cell membrane.21 Experiments have shown that electromagnetic fields have “morphogenetic” effects—in other words, effects that influence the form of a developing organism. In particular, some experiments have shown that the targeted disturbance of these electric fields disrupts normal development in ways that suggest the fields are controlling morphogenesis.22 Artificially applied electric fields can induce and guide cell migration. There is also evidence that direct current can affect gene expression, meaning internally generated electric fields can provide spatial coordinates that guide embryogenesis.

the spatial arrangement and distribution of these ion channels influences the development of the animal.

Biologists know of an additional source of epigenetic information stored in the arrangement of sugar molecules on the exterior surface of the cell membrane. Sugars can be attached to the lipid molecules that make up the membrane itself (in which case they are called “glycolipids”), or they can be attached to the proteins embedded in the membrane (in which case they are called “glycoproteins”). Since simple sugars can be combined in many more ways than amino acids, which make up proteins, the resulting cell surface patterns can be enormously complex. As biologist Ronald Schnaar explains, “Each [sugar] building block can assume several different positions. It is as if an A could serve as four different letters, depending on whether it was standing upright, turned upside down, or laid on either of its sides. In fact, seven simple sugars can be rearranged to form hundreds of thousands of unique words, most of which have no more than five letters.”24 These sequence-specific information-rich structures influence the arrangement of different cell types during embryological development. Thus, some cell biologists now refer to the arrangements of sugar molecules as the “sugar code” and compare these sequences to the digitally encoded information stored in DNA.25 As biochemist Hans-Joachim Gabius notes, sugars provide a system with “high-density coding” that is “essential to allow cells to communicate efficiently and swiftly through complex surface interactions.”26 According to Gabi...

These different sources of epigenetic information in embryonic cells pose an enormous challenge to the sufficiency of the neo-Darwinian mechanism. According to neo-Darwinism, new information, form, and structure arise from natural selection acting on random mutations arising at a very low level within the biological hierarchy—within the genetic text. Yet both body-plan formation during embryological development and major morphological innovation during the history of life depend upon a specificity of arrangement at a much higher level of the organizational hierarchy, a level that DNA alone does not determine. If DNA isn’t wholly responsible for the way an embryo develops—for body-plan morphogenesis—then DNA sequences can mutate indefinitely and still not produce a new body plan, regardless of the amount of time and the number of mutational trials available to the evolutionary process. Genetic mutations are simply the wrong tool for the job at hand. Even in a best-case scenario—one that ignores the immense improbability of generating new genes by mutation and selection—mutations in DNA sequence would merely produce new genetic information. But building a new body plan requires more than just genetic information. It requires both genetic and epigenetic information—information by definition that is not stored in DNA and thus cannot be generated by mutations to the DNA. It follows that the mechanism of natural selection acting on random mutations in DNA cannot by itself genera...

in at least the case of the sugar molecules on the cell surface, gene products play no direct role. Genetic information produces proteins and RNA molecules, not sugars and carbohydrates.

More important, the location of specific sugar molecules on the exterior surface of embryonic cells plays a critical role in the function that these sugar molecules play in intercellular communication and arrangement. Yet their location is not determined by the genes that code for the proteins to which these sugar molecules might be attached. Instead, research suggests that protein patterns in the cell membrane are transmitted directly from parent membrane to daughter membrane during cell division rather than as a result of gene expression in each new generation of cells.29 Since the sugar molecules on the exterior of the cell membrane are attached to proteins and lipids, it follows that their position and arrangement probably result from membrane-to-membrane transmission as well.

Finally, it is not only the molecular structure of these membrane targets, but also their specific location and distribution that determines their function. Yet the location of these targets on the inner surface of the cell is not determined by the gene products out of which they are made any more than, for example, the locations of the bridges across the River Seine in Paris are determined by the properties of the stones out of which they are made. Similarly, the sodium-potassium ion pumps in cell membranes are indeed made of proteins. Nevertheless, it is, again, the location and distribution of those channels and pumps in the cell membrane that establish the contours of the electromagnetic field that, in turn, influence embryological development. The protein constituents of these channels do not determine where the ion channels are located.

The locations of specified target sites on the interior of the cell membrane also help to determine the shape of the cytoskeleton. And, as noted, the gene products out of which these targets are made do not determine the location of these targets. Similarly, the position and structure of the centrosome—the microtubule-organizing center—also influences the structure of the cytoskeleton. Although centrosomes are made of proteins, the proteins that form these structures do not entirely determine their location and form.

Biologists have shown that microsurgery on the cell membranes of ciliates can produce heritable changes in membrane patterns without altering the DNA.34 This suggests that membrane patterns (as opposed to membrane constituents) are impressed directly on daughter cells. In both cases—in membrane patterns and centrosomes—form is transmitted from parent three-dimensional structures to daughter three-dimensional structures directly. It is not entirely contained in DNA sequences or the proteins for which these sequences code.35 Instead, in each new generation, the form and structure of the cell arises as the result of both gene products and the preexisting three-dimensional structure and organization inherent in cells, cell membranes, and cyto-skeletons. Many cellular structures are built from proteins, but proteins find their way to correct locations in part because of preexisting three-dimensional patterns and organization inherent in cellular structures. Neither structural proteins nor the genes that code for them can alone determine the three-dimensional shape and structure of the entities they build.

36 If this is so, then natural selection acting on genetic variation and mutations alone cannot produce the new forms that arise in the history of life.

When I explain this in public talks, I can count on getting the same question. Someone in the audience will ask whether mutations could alter the structures in which epigenetic information resides. The questioner wonders if changes in epigenetic information could supply the variation and innovation that natural selection needs to generate new form, in much the same way that neo-Darwinists envision genetic mutations doing so. It’s a reasonable thing to ask, but it turns out that mutating epigenetic information doesn’t offer a realistic way of generating new forms of life. First, the structures in which epigenetic information inheres—cytoskeletal arrays and membrane patterns, for example—are much larger than individual nucleotide bases or even stretches of DNA. For this reason, these structures are not vulnerable to alteration by many of the typical sources of mutation that affect genes such as radiation and chemical agent. Second, to the extent that cell structures can be altered, these alterations are overwhelmingly likely to have harmful or catastrophic consequences. The original Spemann and Mangold experiment did, of course, involve forcibly altering an important repository of epigenetic information in a developing embryo. Yet the resulting embryo, though interesting and illustrative of the importance of epigenetic information, did not stand a chance of surviving in the wild, let alone reproducing. Altering the cell structures in which epigenetic information inheres will...