Appendix The Third Base Donald Forsdyke If I thought that by learning more and more I should ever arrive at the knowledge of absolute truth, I would leave off studying. But I believe I am pretty safe. Samuel Butler, Notebooks Darwin’s mentor, the geologist Charles Lyell, and Darwin himself, both con- sidered the relationship between the evolution of biological species and the evolution of languages [1]. But neither took the subject to the deep informa- tional level of Butler and Hering. In the twentieth century the emergence of a new science – Evolutionary Bioinformatics (EB) – was heralded by two dis- coveries. First, that DNA – a linear polymer of four base units – was the chromosomal component conveying hereditary information. Second, that much of this information was “a phenomenon of arrangement” – determined by the sequence of the four bases. We conclude with a brief sketch of the new work as it relates to William Bateson’s evolutionary ideas. However, imbued with true Batesonian caution (“I will believe when I must”), it is relegated to an Appendix to indicate its provisional nature. Modern languages have similarities that indicate branching evolution from common ancestral languages [2]. We recognize early variations within a language as dialects or accents. When accents are incompatible, communi- cation is impaired. As accents get more disparate, mutual comprehension de- creases and at some point, when comprehension is largely lost, we declare that there are two languages where there was initially one. The origin of lan- guage begins with differences in accent. If we understand how differences in accent arise, then we may come to understand something fundamental about the origin of language (and hence of a text written in that language). It was for this reason that an abstraction was introduced in Chapter 5 – the notion that hereditary information could include an “accent” that might be the key to understanding the origin of species. A distinction was made between a mes- sage itself (primary information) and the accent with which it was spoken (secondary information). In other words, multiple forms of information (e.g. message and accent) can share a common text and we can consider them in hierarchical fashion (primary, secondary, etc.). 668 Appendix There is little new in this. In 1978 in Mind and Nature Gregory Bateson referred to the hierarchical levels of hereditary information postulated by the anti-Darwinian “typologists,” which contrasted with the “synthetic” views of those favoring the “modern synthesis” of Darwinian and Mendelian ideas [3]: It is interesting to note the current controversy between the upholders of ‘synthetic’ theory in evolution (the current Darwinian orthodoxy) and their enemies, the ‘typologists.’ Ernst Mayr, for example, makes fun of the blindness of the typologists: ‘History shows that the typologist does not and cannot have any appreciation of natural selection’ [[4]]. ... Do any of the genetic messages and static signs that determine the phenotype have the sort of syntax ... which would divide the ‘typo- logical’ from ‘synthetic’ thinking? Can we recognize, among the very messages which create and shape the animal form, some messages more typological and some more synthetic? Here Gregory Bateson was thinking of two physical classes of message each conveying a distinct class of information, rather than of one physical class of message capable of conveying two distinct classes of information. He acknowledged possible conflicts between different classes (layers) of information: “Every evolutionary step is an addition of information to an al- ready existing system. Because this is so, the combinations, harmonies, and discords between successive pieces and layers of information will present many problems of survival and determine many directions of change.” One such direction of change would lead to establishment of a new species. Since hereditary information is transmitted in the form of DNA, the task is twofold: First, to determine what is the “accent” of DNA. Second, to deter- mine how changes in that accent can bring about an incompatibility between members of a species (“discord”) such that they do not effectively reproduce with each other. Being reproductively isolated they would be, by definition, independent species even if, in the extreme case, they showed no anatomical or physiological differences. What we see (the conventional phenotype) is largely determined by pro- teins. These are linear polymers of basic units – the amino acids. The se- quence of amino acids distinguishes one protein from another and determines how each folds into a complex unique structure. Thus, its sequence deter- mines its properties, which may be structural (e.g. tendons, muscle) or cata- lytic (e.g. bringing about chemical changes so that muscles contract). What determines the sequence of amino acids, determines the nature of the result- ing protein, and hence determines the organism. We can equate phenotype with protein and genotype with DNA. The lin- ear sequence of the bases in an organism’s DNA codes for the linear se- quences of amino acids in its proteins. If we know its DNA sequence we The Third Base 669 know which proteins the organism can make. Hence, if we fully understand the properties of those proteins, we can predict the organism’s anatomical and physiological characteristics. In short, DNA (the genotype) determines protein (the phenotype). The code which relates a particular base sequence to a particular amino acid is a triplet code. Each amino acid corresponds to a three base sequence (“codon”) in DNA. For example, the amino acid glycine is encoded by GGT, and the amino acid threonine is encoded by ACT. So a protein sequence that happened to consist just of these two amino acids – say glycine, alanine, gly- cine, glycine, alanine, glycine – could be encoded by the sequence: GGTACTGGTGGTACTGGT ………… (A-1) Where is the accent of DNA? It turns out that the code is degenerate so that each amino acid has four possible codons. Glycine can be encoded by GGT, GGC, GGA or GGG. Threonine can be encoded by ACT, ACC, ACA or ACG. As far as an amino acid is concerned, what matters is the first two bases of its codon. The third base indicates the accent of DNA. Thus, two or- ganisms may make the above amino acid sequence, but their corresponding DNA sequences may have different accents – for example, instead of the above sequence, one might have the sequence: GGCACCGGCGGCACCGGC ………… (A-2) This also encodes the sequence glycine, alanine, glycine, glycine, alanine, glycine. In this particular case in every codon the third base is C (under- lined). But any of the four bases in the third codon position will suffice as far as the “primary information” – i.e. for the amino acid sequence – is con- cerned. So organisms can vary in their accent (“secondary information”) while maintaining the same primary information (and hence the same pheno- type). Although somewhat simplified, the above account is uncontentious and widely agreed upon. More contentious is whether, in the general case, it is this difference in accent that initiates reproductive isolation between two groups living in the same territory. Furthermore, even if this accent differ- ence did bring about reproductive isolation, what would be the mechanism? William Bateson, under some prompting from Crowther (Chapter 17), agreed that the most fundamental form of reproductive isolation, manifest at an early stage as the hybrid sterility seen when recently diverged (allied) species are crossed, was due to an incompatibility that could be characterized cytologi- cally as an incompatibility between paternal and maternal chromosomes when attempting to pair during meiosis (see Fig. 9-5). Thus, if we can understand what makes chromosomes incompatible, we can understand hybrid sterility. And if we understand hybrid sterility, we can understand an origin of species. 670 Appendix We seek to understand how chromosomes that are homologous (i.e. are alike) can pair with each other. Do they pair by virtue of this likeness (like- with-like), or by virtue of some key-in-lock (sword-in-scabbard) comple- mentarity, which implies that they are not really alike? One must be the sword, the other the scabbard. This paradox was resolved when it was appre- ciated that chromosomal DNA is a duplex consisting of two complementary strands – a “Watson” strand and a “Crick” strand – that pair with each other on the basis of their complementarity. So, in Crowther’s terminology, potentially the sword of one strand can pair with the scabbard of the other (and vice versa). For this swords have to be unsheathed from their own scab- bards and then each inserted into the scabbard of the other. Thus, the Watson strand of one chromosome would pair with the Crick strand of its homolo- gous chromosome (and vice versa). This would require that the Watson strand be displaced from pairing with the Crick strand of its own chromo- some. Likewise, the Crick strand of the homologous chromosome would be displaced from pairing with the Watson stand of its own chromosome. Then the cross-pairing could occur. It was Francis Crick, a codiscoverer of the structure of DNA, who worked it out. Most pictures of DNA molecules show them as double-helices with two strands of DNA containing inward-looking bases, which pair with each other, thus joining the two strands to form a duplex. Crick proposed that under certain circumstances parts of the strands would separate (“unpair”) from each other and become outward-looking. In this circumstance, the out- ward-looking bases in the DNA of a maternal chromosome could pair with complementary outward-looking bases of a paternal chromosome [5]. What about the accent of DNA? It has been suggested that for chromo- somes to pair (as when a sword “pairs” with its scabbard) their DNA’s must have similar accents [6–8].