Free Hit Counter Visitors to these pages since 12-29-97
(Note from the owner of this site: When this page was initially placed on the internet, Richard Deems removed from his internet pages, the paper which the below criticized. Now it has come to my attention that Deems has placed the old, flawed and uncorrected paper back on the internet. I have changed the link below to make sure that the reader can see the original document. It is sad that Deems does not seem open to change.--GRM)
DO NON-STANDARD GENETIC CODES PRESENT A CHALLENGE TO EVOLUTIONARY THEORIES?
R. Joel Duff
Updated 9-16-1998
http://home.entouch.net/dmd/gencode.htm
Free Hit Counter
Visitors to these pages since 12-29-97
It has recently been suggested that the documented cases of deviations from the so-called universal genetic code are "evidence of a weak link in molecular evolutionary theories." It is further suggested that the "problems" created by genomes exhibiting non-universal codes create serious questions about common ancestry which is an assumption at the heart of evolutionary theory. There is always a question of whom the burden of proof lies upon in questions such as these. Is it incumbent upon the anti-evolutionist to show at least strong evidence that his position is at least more likely or is it simply enough to cast some doubt? Alternatively, is the evolutionist required to answer all questions or simply provide a reasonable explanation? If the claims are meant to provide evidence against evolutionary theory and to be used for apologetic purposes, then I would suggest the latter is the case. It is my contention that the specific claims made against evolutionary theory are not insurmountable and neither is their a lack of proposed mechanisms for the observed changes in the genetic code. Here we will examine some of the claimed difficulties as well as outline some of the recent discoveries that are relevant to origin of non-standard codes. Such an examination will show that the burden of proof that non-universal codes are a serious problem for evolutionary theory still lies at the anti-evolutionists feet.
The so-called universal code is the set nucleotide combinations that correspond to a specific set of amino acids. It is the ability of the nucleic acid sequence (DNA sequence) to be "translated" into a corresponding series of amino acids that allows proteins to reliable produced and any living organism alive. Original research into the code suggested that all organisms examined used an identical code. As such it became known as the universal code. As research continued, though, exceptions to this universal code were identified. Both the very existence of exceptions and the presence of altered codes among a diverse group of organisms has prompted many questions about both the mechanism for the evolution of these altered codes. Richard Deem, in particular, has suggested that the existence of non-universal codons presents a serious challenge to the evolutionary hypothesis. The following is an examination of the text of a web page describing the implications of the non-universality of the genetic code. The original text may be found at:
The introduction of this page begins:
"Original studies of the genetic code demonstrated that it was constant throughout the plant, protists, and animal kingdoms. It was taught (even in my college days) that the genetic code should be universal as predicted by the theory of evolution, since alterations of the genetic code would be lethal in those individuals that acquired genetic code mutations. Recently, many examples of variations in the genetic code have been discovered in many species of unrelated organisms. Although first shown in 1979 by Barrell et. al (1), subsequent studies have demonstrated that the genetic codes differs in diverse groups of unrelated animals, plants, and protists (2). There is no pattern of change in the code of related groups of organisms and none of the organisms that possess altered genetic code exhibit any form of evolutionary descent or common ancestry, which might imply that the genetic code had evolved." This appears to be a reasonable introductory statement as I too learned that the genetic code was universal. If you open nearly any textbook from even the last 15 years it will likely still use the term "universal code" and will show the classic table of the amino acids and their corresponding codons, although there will typically be some statement about exceptions to the rule hidden in the text. I would agree that evolutionary theory might predict that the code would be universal for the reasons stated but evolutionary theory would not necessarily prohibit exceptions if a sufficient mechanism were available to overcome the obvious difficulties with acquiring a new code. I would also disagree with the statement that "There is no pattern of change in the code..." and the following statement about not showing any form of evolutionary descent but the reasons for this will hopefully be apparent later.
In a section entitled "The Problem of Mechanism" Deems says:
"Here is the essence of what evolutionists are proposing. They propose that every instance of a specific codon in the DNA is mutated and replaced by another codon. This requires replacement of 1-5% of the entire genetic sequence of an organism. Although this doesn't seem like a large amount of the genome, it is the specificity of replacement that makes this mechanism statistically impossible. In the case of vertebrates, this replacement would involve the specific replacement of millions of nucleotide pairs."
First, it is not completely accurate to suggest that "they" propose that every instance of a specific codon in the DNA is mutated and replaced" but rather it would be more accurate to say that all codons must be lost, though it is true that all codons must eventually be replaced or the genetic translation system will incur lethal errors. Further, by saying this is what "evolutionists" are saying it implies that this is the only hypotheses that evolutionary theory might allow. But, as we will, see it is only one group of evolutionists that suggest that changes in the standard genetic code occur in this fashion. Secondly, the magnitude of this change would be great indeed for most organism the numbers here are deceptive. Although the estimated 1-5% of the entire genetic sequence for eukaryotic nuclear genomes (I believe this must be what is being referenced since the following sentence refers to vertebrate genomes) may be true in generality it is far from the case when the particular codons in questions are examined. Within even most "normal" genomes many of the non-standard codons comprise only a small number of the total amino acids in most protein seqeunces because they represent amino acids with several redundant codons of which the non-standard ones are usually the less frequently. Thus it is not uncommon that among all protein coding sequence the amino acids may be represented by much less than 1%. Add to this that a majority of the genomes with observed non-standard codes have highly skewed base compositions such that they are either high (>75%) in A+T or G+C nucleotides. The effect of this is that some codons are extremely scarce even in a complex eukaryotic genome. Thus the number of mutation required in greatly exaggerated.
Reference to the number of replacements in vertebrate genomes appears very impressive and I would grant they would impressive even after considering the exaggeration of the numbers here, but it is not relevant to the discussion since there are no cases of whole vertebrate nuclear genomes that have been shown to have an altered standard code. There are many cases of vertebrates that exhibit non-standard codes but these alterations are found in the mitochondrial genome of these organisms not the nuclear genome. The only cases of nuclear genomes in vertebrates that have an altered code are mammals and chickens in which the stop codon UGA is changed to SeCys (selenocystine) which is a specific form of cystine found only at certian specific positions in the polypeptide chains of glutathione peroxidase and so effects only a very limited part of the genome.
All eukaryotes have organelles of which some contain their own genomes. Vertebrates have a separate genome in their mitochondria, but this genome, rather than being hundreds of millions of nucleotides in size like the nuclear genome, is typically about 16,000 total nucleotides which code for only a small set of genes. Thus the general claims that millions of codons would have to be changed simultaneously in thousands of genes certainly is overstated for these examples. In fact, these genome have many codons that may be represented fewer than 10 times in the entire genome! One theory for how the code has all copies of a particular codon being lost after which the tRNA might be altered by mutation resulting in a new combination of amino acid-codon (i.e. change in the code). The small size of these genome means that it is not inconceivable that these codons might be lost after which the tRNA might be altered by a single mutation. I think it is very revealing and consistent with the current theories of how code changes might occur that no examples of altered codons have been found in a plant mitochondrial genome. Even the smallest plant mitochondrial genome is much larger than the largest mitochondrial genome in all other organisms (fungi, protists, animals). Thus we would expect plant genomes to exhibit few if any examples of non-standard codons. Currently, the vast majority of all non-standard codons known reside in very small genomes which is consistent with expectations. Continuing from the above quote:
"There is no 'directional mutation pressure' that would cause only one codon sequence to be replaced in an organism, even according to evolutionary theories. Evolution states that selection acts on the protein structure to improve its function. Since we are talking about all the proteins in an organism, there is no one selective pressure that would work to improve the function of all proteins simultaneously (especially by substituting only one specific amino acid for another). It would make much more sense from an evolutionary viewpoint that gene duplication of tRNA would occur, followed by mutation of the duplicated tRNA and gradual conversion of genetic sequences from those that bound to the universal tRNA to those that bound the mutant tRNA. However, such a scenario would be expected to produce intermediate forms of organisms (possessing both forms of tRNA), since the process would obviously be a long one. Even though there are dozens of examples of tRNA mutants, none of them exist as these hypothetical intermediates, indicating that this is not a reasonable mechanism."
Although this is a common understanding of what evolutionary theory states it is in fact only one facet of evolutionary theory. Appealing only to a selectional force to find a way to alter the code ignores a large and complementary school of evolutionary thought that involves neutral or near-neutral mutations and stochastic events that may have just as significant effect on the evolution of an organism as direct selectional pressure on phenotypes. Deem's only modern reference is to Osawa et. al., (1992) who are strong proponents of neutral or near-neutral-theory and so it is no surprise that they do not give much page space to selectionists' theories. In organisms with small genomes such as mitochondria and a few bacteria some codons may be lost completely from the genome. Such is the case in the bacteria Micrococcus (Kano et. al., 1993) in which the AGA and AUA codons are not used at all. Not only are they not found in the nucleic sequence of protein coding genes but further investigation found that Micrococcus could not synthesize proteins that contain in-frame AGA and AUA codons which indicates that they are not only missing the sequences but don't have tRNAs with AGA or AUA anticodons and so cannot place an amino acid into the protein. This means that this code is free to be changed such that it may code for a new amino acid. Should subsequent mutations occur and the codon reappear in a particular protein it would code for a new amino acid. This situation would not be necessarily lethal becsause of the small number of proteins involved. Few genomes have been studied to the detail that Micrococcus has so it may not be as uncommon for genomes to be missing particular genomes as was previously thought.
Since the time of the last update of this web page several publications have documented specific cases of intermediates that fit the descriptions of the predictions of the neutralists codon replacement theory and so rather than "not being a reasonable mechanism" it has become a most reasonable mechanism. Castresana et. al., 1998 have identified a group of organisms that fill the predictions of the codon-capture hypothesis amazingly well. Castresana describes the codon capture hypothesis as follows:
"According to the 'codon capture hypothesis,' it would be deleterious for an organism if a codon was assigned to two amino-acids (or an amino acid and polypeptide chain termination) simultaneously. Thus, the first step in the change of the genetic code is assumed to be the complete disappearance of a codon from a genome. Subsequently, the tRNA (or release factor) assigned to this codon loses its capacity to recognize it, so that the codon becomes unassigned, and another tRNA acquires this capacity, allowing the codon to reappear at new positions in protein-coding genes.
The paper reports finding a group of organisms, the hemichordates, that have an intermediate step in codon evolution predicted by the codon-capture theory. To summarize the paper the abstract is reproduced here:
"ABSTRACT: In the mitochondrial genome of the hemichordate Balanoglossus carnosus, the codon AAA, which is assigned to lysine in most metazoans but to asparagine in echinoderms, is absent. Furthermore, the lysine tRNA gene carries an anticodon substitution that renders its gene product unable to decode AAA codons, whereas the asparagine tRNA gene has not changed to encode a tRNA with the ability to recognize AAA codons. Thus, the hemichordate mitochondrial genome can be regarded as an intermediate in the process of reassignment of mitochondrial AAA codons, where most metazoans represent the ancestral situation and the echinoderms the derived situation. We also show that the reassignment of the AAA codon is associated with a reduction in the relative abundance of lysine residues in the mitochondrial proteins."
Nuclear genome changes As we have seen, the examples of non-standard codes that reside in mitochondria and small bacterial genomes do not represent significant hurdles for evolutionary theory. Now we turn to one example of a eukaryotic nuclear genome that exhibits a non-standard code. The fungal genus Candida contains some species what use a non-standard code (note the author's use the term non-standard and standard genomes to avoid continuing the propagation of a misapplied term). Santos et. al., 1997 argue from the results of phylogenetic analyses of Candida species that the "codon change must have occurred in a Candida ancestor with a relatively sophisticated genome encoding thousands of genes." This would appear to qualify as an example of the problems highlighted in Deem's "Problem of Mechanism" quote above. One again though, the case has been somewhat overstated even if it is not wholly inaccurate. Some of the Candida species have low G+C content and thus use UUG/A as the preferred leucine codons and so only a small number of CUG (the altered codon) codons are actually present in the entire genome rather than the 1-5% of the genome or millions of nucleotides. This means that there may be have only been several thousand actual CUG codons in the Candida ancestor that experienced the change. Yet, this does not mean that the neutralists mechanism proposed to explain the changes in some mitochondria and bacteria may be applied here without difficulty. It is unlikely that all of these codons would disappear from the whole genome via A+T pressure or just stochastic events. Hence, thousands of changed codes would still be required and that would certainly have deleterious effects.
The codon-capture hypotheses appears to fail to account for the presence of a non-standard code in Candida. Instead, the genetic code of Candida has been explained by a general selection model called the "ambiguous intermediate theory" first proposed by Schultz and Yarus (1994, 1996; and Yarus and Schultz, 1997). This theory is summarized by Santos et. al., (1997). This theory arose out of the initial observation that;
"10 out of the 18 known codon reassignments involve single nucleotide changes at either the first or the third codon position and that naturally occurring tRNAs are able to misread the reassigned codons. The theory postulates that tRNA species that misread near cognate codons can gradually 'capture' them from their cognate tRNAs or release factors in a gradual process driven by selection. This theory implies that, in any codon reassignment, an intermediary step is required in which a codon codes either for two different amino acids or for an amino acid and translation termination, i.e. such a codon being an ambiguous intermediate codon. This theory assumes that codon ambiguity is not lethal and, consequently, codons do not have to disappear from the genome before their reassignment."
Schultz and Yarus (1996) outlined some of the observations that have been made that support an ambiguous intermediate. First, they point out that the nonambiguity assumption of codon-capture models contradicts chemical principle and experimental evidence. Apparently missense translation in normal E. coli has been estimated at 4 x 10-4 per codon (Ellis and Gallant 1982). This means that an appreciable basal ambiguity is tolerated within cells without evident harm. Also, mutational studies have even pushed the error rate to greater than 1 amino acid in every 250 without apparent change in phenotype. Secondly, it is observed that the known examples of reassignments are very nonrandom and that the types of reassignments frequently fall within examples of demonstrated equivocal reading of those tRNAs. All changes could be accomplished with either a G-U, C-A or G-A intermediate with 8 of 15 being G-U. The base position of the changes in the codon are also non-random. Thirdly, "phylogenetic distribution of reassignment is consistent with ambiguous intermediates." For example the authors note:
"Tourancheau et a. (1995) have made the initially surprising observation that UAA/UAG in ciliates have been reassigned to glutamine at least three times independently (on the bases of the rRNA tree), instead of depending on a common ancestral reassignment. This striking phylogenetic cluster of identical but independent reassignments has no apparent explanation in the codon-disappearance scheme. However, such a cluster is easily explained within the ambiguous intermediate mechanism by a tendency to equivocal reading of these codons inherited from an ancestor. Such an ambiguity might be conserved within a group of species if used for an important regulatory event like stop codon readthrough."
Fourthly, "Molecular fossil and functional evidence of translational ambiguity accompanies known cases of reassignment." Evidence of this is that sequenced tRNAs of captured codons from the ciliate Tetrahymena contains unusual sequences that have been identified as enhancers of equivocal coding in E. coli (Shultz and Yarus, 1994).
Further evidence of ambiguous intermediates is seen in the case of Candida. Some species of Candida have been found to have an ambiguous codon such that a single codon (CUG) may code for the standard leucine or as serine. These species appear to be intermediate between those that use the standard and non-standard code more exclusively. This selectional driven hypothesis combined in some cases with low G+C content and stochastic forces yields a very viable mechanism for the origination of non-standard codes. Schultz and Yarus argue that the codon-caputure and ambigous intermediate hypotheses are not incompatible but that they may in many cases work in conjunction to create the circumstances that would allow for changes in the code.
A further interesting aspect of the Candida example is that we see that within this genus there are some species that have an altered code while other do not. At the same time each species exhibits differing amounts of ambiguity in reading the CUG codon. Thus it appears we are viewing several stages of evolution in this genus. I would also point out that these species are so similar that before any knowledge of different genetic codes was known creation scientists would almost certainly suggest that all these species would have been derived from one Candida "kind" or even more broadly from the family in which Candida is placed. If such 'microevolution' is allowed then one would have to accept that the change in the standard code has come about within a normal providential, i.e. natural, context and thus must not be as impossible as it is being alleged. The other example of eukaryotic nuclear genomes with non-standard codes is seen in multiple lineages of ciliated protozoa. These protozoa have a combination of both small nuclear genomes and high A+T content which suggests that there would be few representatives of the changed codons. More likely the A+T pressure has allowed a situation where ambiguous reassignments might occur, especially in the case of stop codons for which many natural ambiguous codons are know. Currently the Glycine tRNA genes in a number of these ciliate groups are being looked at to test this hypothesis. The ciliated protists are yet another example in which many closely related species and genera show combinations of standard and non-standard codes. Thus it would appear that, in this case, the standard code has been altered in a recent evolutionary history that most creation scientists would put in the context of microevolution.
What could be the evolutionary advantage of changes in the genetic code if, in fact, such changes are driven by selectional vs. neutral processes? For the ciliates it has been proposed that since they are phagotrophic (take in their food through the oral apparatus) a non-standard code would constitute a very effect mechanism in preventing the expression of foreign DNA. In the case of Candida an even more interesting hypothesis has been forwarded that the presence of ambiguous decoding has actually evolved to "generate genetic diversity and allow for rapid adaption to environmental challenges." The reason for this is that under different environmental stimuli it has been shown that the percentage of mischarged (non-standard amino acid applied to the tRNA) can change and this would obviously change the composition of many many enzymes. Thus it is possible for this organism to make many different proteins from the same DNA transcript. Although this may lower the fitness of some enzymes there would be mixture of normal and mutant enzymes at all times thus allowing the organism to survive but also allowing for many new proteins to be tried out under new conditions.
Robustness and changeability of the code
One other significant development in this field has come from Maeshiro and Kimura (1998). They have examined why the standard genetic code has such a highly regular structure and derive models that "quantitatively evaluates genetic codes and accurately predicts known deviant codes even without the appropriate anticodon list." They find that all the examples of non-standard codes fall within a short list of codons which if changed actually improve the s-robustness (robustness against nonsense mutations) and changeability (alterability of phenotypes by a single base mutation of codons) without decreasing the u-robustness (the unalterability of phenotypes caused by a single base mutation of codons, where the phenoptyes denote any of 20 amino acids and the stop codon). Essentially a selectionist model, Maeshiro and Kimura provide a means of looking at the code in terms of a dynamic code with contrasting selectional forces acting upon it. I would direct the reader to the article for the details of the models.
Conclusions
More cases of non-standard codes could be examined and more detail could be provided but the data and references provided here should suffice to demonstrate that the presence of non-standard genetic codes is not a "huge problem" for evolutionary theory. Rather we find that the cases of non-standard codes are found in rather non-standard genomes that are unusual in their own right and thus it is not unexpected that they have undergone unusual patterns of evolution. I would also predict that many more cases of non-standard genetic codes will be found in the near future. Only a small fraction of the organisms on earth have been examined in the amount of detail that is required to determine how the code is used. The fact that several significant discoveries have been found in the past year regarding potential mechanism for such changes suggests that until our knowledge of genomes is more complete the use of alternative genetic codes as a "proof" against evolution is premature and therefore its use as an apologetic tool unwise. I would suggest there is amply reason to believe that the presence of non-standard codes do not present an insurmountable challenge to evolutionary theories.
References
Bessho, Y., T. Ohama, and S. Osawa. 1991. Planarian mitochondria II: The unique genetic code as deduced from COI gene sequences. Journal Molecular Evolution 34:324-330.
Castresana, J., G. Feldmaier-Fuchs, and S. Paabo. 1998. Codon reassignment and amino acid composition in hemichordate mitochondria. Proceedings National Academy Sciences USA 95:3703-3707.
Edelman, L. and M. Culbertson. 1991. Exceptional codon recognition by the glutamine tRNAs in Saccharomyces cerevisiae. EMBO Journal 10:1481-1491.
Ehara, M., H-I, Yasuko, I. Yuji, and T. Ohama. 1997. Use of a deviant mitochondrial genetic code in yellow-green algae as a landmark for segregating members within the phylum. Journal Molecular Evolution 45:119-124.
Giulio, M. D. 1997. On the origin of the genetic code. Journal Theoretical Biology 187:573-581.
Kano, A., T. Ohama, R. Abe, and S. Osawa. 1993. Unassigned or nonsense codons in Micrococcus luteus. Journal Molecular Biology 230:51-56.
Hayashi-Ishimaru, Y., T. Ohama, Y. Kawatsu, K. Nakamura,and S. Osawa. 1996. UAG is a sense codon in several chlorophycean mitochondria.
Himeno, H., H. Masaki, T. Ohta, I. Kumagai, K-I. Miura, and K. Watanbe. 1987. Unusual genetic codes and a novel genome structure for tRNA SerAGY in starfish mitochondrial DNA. Gene 56:219-230.
Karlovsky, P. and B. Fartmann. 1992. Genetic code and phylogenetic origin of oomycetous mitochondria. Journal Molecular Evolution 34:254-258.
Maeshiro, T. and M. Kimura. 1998. The role of robustness and changeability on the origin and evolution of genetic codes. Proceedings National Academy Sciences USA 95: 5088-5093.
Osawa, S., T. Jukes, K. Watanabe, and A. Muto. 1992. Recent evidence for evolution of the genetic code. Microbiological Reviews 56(1):229-264.
Osawa, S. and T. Jukes. 1995. On codon reassignment. Journal Molecular Evolution 41:247-249.
Sanchez, J-S. 1995. On the origin and evolution of the genetic code. Journal Molecular Evolution 41:712-716.
Santos, M. A. S., T. Ueda, K. Watanabe, and M. Tuite. 1997. The non-standard genetic code of Candida spp.: an evolving genetic code or a novel mechanism for adaption? Molecular Microbiology 26(3):423-431.
Schultz, D. W. and M. Yarus. 1994. Transfer RNA mutation and the malleability of the genetic code. Journal Molecular Biology 235:1377-1380.
Schultz, D. and M. Yarus. 1996. On malleability in the genetic code. Journal Molecular Evolution 42:597-601.
Tourancheau, A. B., N. Tsao, L. Klobutcher, R. Pearlman, and A. Adoutte. 1995. Genetic code deviations in the ciliates: evidence for multiple and independent events. The EMBO Journal 14(13):3262-3267.
Yasuko, H-I., M. Ehara, Y. Inagaki, and T. Ohama. 1997. A deviant mitochondrial genetic code in prymnesiopytes (yellow-aglae): UGA codon for tryptophan. Current Genetics 32:296-299.