Posts Tagged ‘LGT’
Tangled Trees?
David Quammen book, The Tangled Tree: A Radical New History of Life was reviewed in Nature Aug. 2 by John Archibald (not incidentally the author of a book on symbiosis). It was also reviewed in the New York Times Book Review on Aug 14 by Erika Check Hayden under the title of “Uprooted”. On Aug 19, Quammen himself published an article in the New York Times Magazine titled “The Scientist Who Scrambled Darwin’s Tree of Life”.
Quammen’s book is largely about Carl Woese, the great pioneer of molecular phylogenetics, who in 1977 with coauthor Fox, made clear from their study of a component of ribosomes (protein factories in cells) that there are three not two domains of life – the Archaea in addition to Bacteria and organisms with nucleated cells like plants, animals, fungi and some unicellular organisms (Eukarya). Woese was apparently not a fan of Darwin – he commented to a prospective co-author of a book to be titled Beyond God and Darwin, “Jan, you accord Darwin so much more substance than the bastard deserves”.
Anyway, the theme of the reviews and article (and I assume of the book which I have not, or at least not yet read), is that life is better described as a “network”, a “tangled web”, or a “topiary” rather than Darwin’s tree because of the existence of horizontal i.e. lateral gene transfer (LGT ). LGT can take place by transformation (taking up DNA from the environment), transduction (a piece of host DNA carried to another organism by a virus), and conjugation (bacterial “sex” which is unidirectional). Unfortunately Hayden casually equates LGT with “swapping genes” which is exactly what it is not i.e. bidirectional rather than unidirectional. Not does its existence invalidate Darwin’s tree metaphor for evolution as these authors generally seem to think. Quammen’s article is illustrated with outrageous illustrations of what looks like an early hominid with a chicken’s head, one with a fish’s head, a human with what looks like a frog’s body between head and shoulders on top and lower legs and feet on the bottom. Perhaps we should not blame Quammen for these illustrations but the point is that because laterally transferred DNA is normally just a tiny part of the genomes of eukaryotes, the best metaphor I have seen is “trees with some cobwebs”. (In 2005 Liza Gross in PloS attributes this metaphor to Fan Ge and coauthors.)
Yet, once anisogamy (micro and macro gametes) have evolved, then sex again becomes unidirectional, i.e. from males or male functions to females or female functions. But that is unidirectional within a species – indeed the ability to interbreed after Mayr is what is most commonly said to define species boundaries. Of course Woese was right that we know very little of what went on in the early history of life. However, only if hybridization (sex between members of different species) were rampant in the Eukarya could Darwin’s trees truly be said to be entangled and despite the occasional case there is no evidence of that.
What is the difference between LGT in prokaryotes and sex in eukaryotes?
Biologists are typically adamant that lateral or horizontal (used interchangeably) gene transfer in prokaryotes is not “sex” and should not be called that. Yet I keep running into people in print and in person, who, with little if any explanation, want to call it that. Are they just naive or could there be something more to it? So I decided to give the question some thought and here is what I came up with. Actually, it is quite difficult to pin down a principled difference between LGT in prokaryotes and sex in eukaryotes.
Consider a stereotypical case of LGT – conjugation in E-coli. Superficially it seems easy to characterize the difference. LGT is lateral (genes move between peers rather than being transmitted from parents to offspring), partial (only some of them do), and unidirectional (they move in one direction only).
Further consideration however suggests that things are not so simple. Eukaryotic recombination is, at least in part, just as “lateral” as is the prokaryote. Imagine an A1B1 haploid eukaryote engaging in sex and recombination with an A2B2. Whether the two loci are on different or on the same chromosome, together, they give rise to four offspring – two parental types an A1B1 and an A2B2 and two recombinant types an A1B2 and an A2B1. One way of looking at the recombinants is that a B2 has been transferred into an A1B1 displacing B1 while a B1 has been transferred into an A2B2 displacing B2 (or alternatively, an A2 has been transferred into a A1B1 displacing A2 and A1 has been transferred into a A2B2 displacing A2). Eukaryotic recombination then, is no less “lateral” than is the prokaryotic. This suggests we should stop talking about prokaryotic sexual processes as “lateral gene transfer” as if “lateral” were unique to them and talk about them simply as “gene transfer” (as some indeed already do).
Moreover, both prokaryotic and eukaryotic transfers are only partial. In eukaryotes, the non-recombinants A1B1 and A2B2 are, in effect, transmitted vertically rather than laterally, and with the recombinants, in the first version above A1 and A2 are as well, while only B1 and B2 are being transmitted through peers. This is not in principle unlike prokaryotes in which part only of the donor genome is transferred.
Is the bi-directional transfer in eukaryotes rather than uni-directional transfer in prokaryotes then the defining difference? With isogamy in eukaryotes, that seems to be the case. However, once anisogamy/oogamy has evolved from isogamy (as is thought to have been the case), genes are transferred there unidirectionally as well i.e. from the microgamete/sperm or its producer to the macrogamete/egg or its producer. Eukaryotic sex is associated with cellular reproduction (and involves syngamy, gene duplication and a pair of meiotic divisions) while prokaryote sexual processes are not and do not. In the eukaryotic case, cells “go”, at least initially, while in the prokaryotic case only genes do. So to some degree bidirectionality seems to be a fundamental difference with isogamy, but ultimately with anisogamy and oogamy in eukaryotes, that distinction too loses its force.
Another possibility is that prokaryotes are not organized into “good biological species” with sexual processes possible within species boundaries but not outside them as eukaryotes tend to be (occasional hybridizations notwithstanding). At first blush, it seems obvious that prokaryotes could not be. If F+ were driving through a closed population, sooner or later F- would go extinct and that would be that. However, negative frequency dependence could obtain between transmission of F+ through peers and F- through offspring with each being favoured when the other is rare as has been suggested. But that is not unlike the eukaryotic case in which Fisher’s ‘one mother, one father principle’ means that one mating type/gender is favoured when the other is common, and vice-versa.
Perhaps, related to uni and bi-directionality, the difference is based on the different natures of the social relationships among the parties involved. It has often been suggested that gene donors in prokaryotes are parasites, F+ and their like commonly being a plasmid, transposon or other mobile genetic element with a few host genes being dragged along with it (parasites of the parasite?). In some cases however, gene transfer seems to be in the interest of the recipient – antibiotic resistant factors can be transferred for example. However, gene transfer in prokaryotes may be in the interests of only one, or the other, but not both partners in various cases, while sex understood as genetic recombination in eukaryotes has historically universally been thought of as in the two parties mutual interest. Ultimately of course like bi-directionality, the cooperative theory of eukaryotic sex fails with the coming of anisogamy and oogamy if (and it is becoming a bigger if) the traditional explanation of gender differences and relations is accepted. There proto-males in anisogamy and males in oogamy are viewed as reproductive parasites of proto-females and females respectively.
If principled distinctions based on laterally, partiality, unidirectionality, good species and antagonistic versus cooperative social relationships all fail, at least ultimately, where does that leave us? In a muddle actually.
Blute Blog 