A new process for domesticating animals and plants, the Genetically Modified Organisms

cn This article has been published in Chinese by the journal Science and Technology in China kjzg, 2004 (no longer available on the WWW)
fr Une discussion en Français a eu lieu à la demande du Sénat de la République Française sur le sujet senat

The scene

The Neolithic revolution, 10,000 years ago, is the revolution of agriculture. Since that time, Man began to domesticate animals and plants. Domestication implies a man-made selection process that directs evolution of life towards organisms thought to be useful for us. Implicitely first, then explicitely, we used in this process our knowledge of heredity to speed up the efficiency of selection. Hence domestication went to be intimately connected to genetics. Genetics, at the core of the science of heredity, is based on an alphabetic metaphor. Genetic heredity is transmitted from generation to generation by a molecule that has the strange property of behaving as a text, written with an alphabet of four letters. This is quite different from Chinese writing, and this needs some explanation.

Alphabetic writing uses strings of symbols that are chosen from a narrow list (usually of the order of 20-30 symbols) and combined sequentially to represent the phonemes that make spoken words (as in the present sentence, if printed out in English). As in most languages, there is no link between these letters and the objects they represent: four (in English), vier (in German), quatre (in French), τεσσερα (in Greek), 四 (in Chinese) represent the same concept of Number Theory, number 4 (we see immediately here an important link between alphabets and numbers: another alphabetic writing, that of numbers, combines ten symbols to represent the abstract meaning of an integer). Indeed different phonemes, and hence words, are represented by the infinite combination of letters. Strings of symbols have many remarkable properties, that will not be discussed here except to emphasize the fact that, when associated into recursive patterns, they can make totally unexpected behaviours emerge (see [1] for a general discussion of this remarkable creative property of strings associated to coding). DNA, the support of genetic heritage, is made of the chaining of four types of chemicals (only four), with an order that has something to do with the general functioning of the cell, the atom of life. In brief, the DNA contained in a cell, its genome, can be represented as a book of recipe, that tells the cell how to behave in a given environment (and in particular how to multiply). The rest of the cell behaves as a kind of computer, that would read (some of) the DNA text (the genetic program) according to cues provided by the environment. This program is decoded through a process that can be considered as recursive (briefly, the routine 'make protein' uses proteins, hence 'make protein' to be put into action). It chains the recipes in an order that allows the organism to maintain itself, reproduce, react to environmental cues, protect itself against aggressions, etc. As we shall see, this precludes total predictability in the case of living organisms, whatever their type, be they "natural" or "artificial".

All living organisms evolve, and they evolve in a way that cannot, because of the very way they are constructed, be predicted. To state this in brief, living organisms are those material systems that have found a way, facing an unforeseeable future, to create an innovative hence unforeseeable progeny, among which some will be able to survive in that always unexpected future [1]. The way selection operates is what separates between the natural and the artificial: natural selection corresponds to all organisms except those that are domesticated. In the latter case, selection is oriented by the human mind, with specific characters chosen to be retained or eliminated. Among artificially selected organisms some have recently undergone a novel selection process, with direct, rather than indirect, action on their genomes. Those are named Genetically Modified Organisms (GMOs for short). It is important at this point to notice that both standard domestication and genetic engineering-based domestication are artificial processes. The practice of agriculture changes the environment in a man-driven way, whether it is the result of selection acting purely on the expressed traits of the organism, the phenotype, or whether it is gene-based.

A first dramatic change between the old recipes meant to domesticate life and the science of genetics happened after the discovery of genes by Mendel, De Vries, Correns, von Tschermak, Morgan, Sturtevant and many others (for important dates see our compilation). Knowing that the phenotype is mostly resulting from the combined action of genes, that were apparently organized as linear sequences and subject to recombination, a first gene-driven approach resulted in a considerable speed up of the creation of new varieties for agriculture. This was at the root of the so-called "green revolution". As an example the weight of an average maize ear doubled between 1970 and 1990. This is still the most widely used way to domesticate plants and animals, for example by using statistical means that give marks to phenotypic traits and connect them to the way genes are organized in the genome (search for quantitative trait loci, QTLs [2, 3]). Genetic engineering is the most recent step in this process of gene-driven domestication of plants and animals. Because of strong socio-political pressure (that cannot be discussed here), it is not (yet) widely used, except in the USA.

The discovery of the "alphabetic" structure of DNA, coupled to that of restriction enzymes (enzymes that allow one to play "cut and paste" with DNA sequences) started an era where the DNA "text" could be knowingly manipulated at the desk, in silico (for the first definition of that term see [4]), using the familiar cut and paste procedure of standard word processors, and subsequently manipulated in vitro, to create artificial genes. In turn these genes could be introduced in vivo in organisms that were named "Genetically Modified Organisms" to indicate this fact. Used previously, in vivo recombination was also creating artificial genes, but in a generally unpredictable way, since it was only at the level of the phenotype that one could know which gene combination was to be retained. Furthermore, it was not possible to know exactly what kind of modification of the gene text was at work. In contrast, in GMOs the modification of the genome is exactly known, and for this reason it can not only use recombination of genes previously existing in the parent organisms but also introduce genes from elsewhere or even entirely artificial genes, in a way that is much wider than that of standard hybridization, for example.

Perhaps because of that very fact (i.e. introduction of foreign genes into extant organisms) GMOs (in fact plant GMOs, and, curiously, generally not animal GMOs) were considered as frightening, or dangerous by the general public. It would be a research programme in itself to investigate the reasons given by the persons involved (which of course may be different from the real reasons) and to see whether this is based on hard facts. We shall not discuss here the political or economical reasons (which are however, naturally, extremely important) but restrict our discussion to the following features:

•      genomic difference between domestication using GMOs and traditional domestication

•      gene transfer between organisms of the different kingdoms

•      gene transfer between domesticated organisms and undomesticated organisms

•      stability of domesticated organisms in the environment

1. Dragons or Chimeras? Hybridization differs between domestication using GMOs and traditional domestication

The process of selection used by Man was initially to recognize a particular trait and to try to inbreed organisms that possessed the same trait. Because of the laws of genetics, this is a very slow process when using the phenotype alone (one has to wait for many generations to create a really inbred strain, and, for plants, this is usually requiring one year per generation). However, knowledge of the way genes are organized, together with molecular means allowing one to connect a trait with a specific feature of the DNA (restriction enzyme profile, for example) was used to make better choices in crosses, and this was at the root of the green revolution, starting just half a century ago [2, 3]. It must be noticed here that the means employed and still widely accepted as natural by laymen despite their evident artificial construction, are deeply rooted in molecular genetics (a fact that, curiously, mass media usually ignore completely). The result was an enormous increase in yields, both for crops and for animal products, to a point where the general landscape of the world has been dramatically altered, as compared to what it was during the last millenium. Another consequence was that, despite the extraordinary multiplication of Mankind (world-wide, it quadrupled its number in just one century, China reached 600 million people in 1954), famine has considerably regressed, so much that, when it appears, this is not due to lack of agro-food resources, but rather to war or other bad political management.

Human population is still growing and, because this is happening fast, it became obvious that we were facing a difficult situation if we had to use standard genetic techniques to keep on increasing agricultural yields. In fact, genetics aside, the success of the green revolution was mainly caused by a huge consumption in fertilizers and pesticides, yielding extreme pollution (that is dramatically affecting wild-life) and energy consumption, considerably increasing our speed in meeting with global warming. This happened at a time when it became tempting to resort to directed evolution. Instead of selecting, after the fact, the relevant organisms, could we create them with the properties we wished them to have? If one would like to isolate a plant resistant to an insect one could either crossbreed plants (in an intelligent way) until we find one that resists insects (but this is necessarily a very slow process because it can only be ajusted by small increments, that each requires at least two generations to be stabilized) or could we identify genes, the product of which would kill insects? Such genes could be extracted from any source, either from plants, animals or even bacteria. This was the basic idea for the genetic engineering of living organisms.

As can be seen, the main difference with standard genetics are two: first, one can chose the family of genes of interest; second, one can select not only those interesting genes from the same family of plants (or related plants) but even from entirely foreign organisms. This is a considerable extension of an old practice, hybridization. However, hybrids combine whole genomes, possibly and eventually creating new species, while in genetic engineering only one gene or a few genes are introduced in the organism, which certainly does not change its species nature.

This extension of the hybridization process drove a first line of resistance against genetic engineering because of a feeling of lack of "purity" of the resulting modified organisms, a kind of "degeneracy" that is thought to affect the GMOs. This feeling will deeply depend on the cultures: Dragons are positive in China, where Chimeras are usually feared in Europe. Of course, many other features are also important, in particular the fact that, because GMOs are efficient, they are deeply associated to a culture of profit. Hence a moral question posed by the public to the people who, at the same time, say that they work for the benefit of Mankind, but make an enormous profit out of this. To say the least, profit-making hidden behind a moral smoke screen is obviously questionable. Furthermore, naturally, the very fact that GMOs "hybridization" is a deep extension of an old common practice, makes it clear that its consequences have not yet been fully explored. It is therefore quite understandable that this practice rose a heated debate. Rather than enter its details (this would ask for a long book) we shall now dwell on a series of questions raised by the existence of GMOs, questions which can mostly be answered in a fairly clear fashion. However to my view these (misleading) questions, although placed in the limelight, are not the most important questions one should ask about the creation and use of GMOs. We shall say a few words about that in the conclusion of this article.

2. Gene transfer between organisms of the different kingdoms

When people discuss the impact of GMOs they usually assume the existence of gene transfer between GMOs domestic plants and wild plants. What evidence do we possess about the ubiquity of gene transfer in the living kingdoms? A few references will tell. We have tried in this article to bring attention to some work that is usually not taken into account in review articles or books dealing with horizontal transfer of genes (which, as a matter of fact, are often stemming from medical work, especially dealing with nosocomial infections [5, 6]).

Although it has been recognized for a long time that horizontal gene transfer could occur in Bacteria, the importance of the phenomenon has only been evaluated recently, with the knowledge of the first complete genomes. In this process, it is important to recognize first the amount of DNA that can be considered as horizontally transferred, as well as the processes permitting the transfer to occur. In particular, it should be understood that, at the present time, we possess only indirect evidence of a widespread process, and, in particular we have almost no evidence for its importance when one considers transfer between the three domains of life, Bacteria and Archaea (usually single cell organisms, without a nucleus) and Eukarya (organisms with cells containing a nucleus, as plants and animals do). It should also be stressed that in general (except for the direct transfer between microbes), the time course of the processes investigated is of the order of one event per million years, or even tens of millions of years. Another important fact has to be taken into account: cells with a nucleus (Eukarya) have a variety of organelles derived from Bacteria that lived inside them (mitochondria, chloloplasts and the like). These symbionts slowly transferred most of their genes to the nucleus of their host. Despite this extraordinarily closed environment, it took probably one billion years to transfer many of the bacterial symbiont genes into the nucleus of the host. This shows that, even when organisms are in close association, stable gene transfer is a very rare event. To understand the processes requires to know something not only about the mechanisms for transfer, but also something about the selective forces that permit its stabilization.

Horizontal transfer between Bacteria is linked to the existence of specific viruses, the bacteriophages (some of which with a very large genome), and to bacterial sexuality. It can also directly involve DNA through the process of transformation. Infection by bacteriophages that became cryptic when integrated in their host's genome (the process called lysogeny) has been recognized by Félix d'Hérelle as early as 1917. Conjugation and sex in bacteria have been recognized by Elie Wollman and François Jacob in the fifties. Both processes, that transfer genes in the wild, were used to transfer genes from homologous organisms in the laboratory since the beginning of the molecular biology era. However the extent of horizontal transfer of genes into chromosomes was really appreciated only in 1991 when Médigue et al., comparing the difference in codon usage in Escherichia coli, proposed that a whole class of genes (now known to comprise 15% of the genes in E. coli K12) bore the signature of horizontal transfer [7]. Nothing, however, could be said about their origin, except that, in most cases, these genes appeared to have been transferred through a conjugation process (and hence came from other bacteria), and, in a minor set of genes through phage lysogeny [8]. It is now widely accepted that there is significant gene transfer between Bacteria (and even between Bacteria and Archaea [9]) but the extent of that transfer is subject to intense debate [10]. In general it appears that horizontal transfer is strongly linked to the existence of persisting unusual environmental conditions (stress conditions). Persistence of the transferred genes requires repeated selection pressure. This can be achieved by co-evolution of genes of the receiver organism, in which case the foreign gene is definitively incorporated into the host genome. In this respect it should be noted that the evidence accumulated so far is highly biased by the choice of genes relevant to one or another interest of the investigator. In reality, what is relevant for selection is only a gene in context, i.e. genomic context and environmental context. This can be perceived in the fact that, when there is horizontal transfer, several genes are in general transferred together (e.g. "pathogenicity islands"). It has been published that organisms such as bacteria belonging to the Bacillus cereus family can have enormous coordinated acquisition of foreign DNA, while they keep constant the core of their genome (with concomitant conserved general phenotype) [11]. This is prominent even in highly transferable vectors, such as plasmids, transposons and the like. It is therefore somewhat irrelevant to argue about the transfer of isolated genes, if one has to make a reasonable risk assessment: the context of the gene is as important as the gene itself. In this respect the widely spread existence of plasmids seems to be a way to keep a genome as intact as possible, while keeping means to obtain foreign genes for some time, when they permit coping with special environmental conditions.

It should be noticed that although Gram negative bacteria are generally better equipped than their Gram positive counterparts (in particular because conjugation is much more frequent in their case), the efficiency of gene expression is good only in Gram-/Gram- exchanges or Gram+/Gram- exchanges, and not in the Gram-/Gram+ direction. This is most probably due to the fact that many Gram+ organism do not possess a ribosomal S1 protein [12]. As a consequence, it is always much easier to express foreign genes in Gram negative bacteria than in Gram positive bacteria. This fact should be taken into account when considering the stability of a gene transfer (and also when considering heterologous gene expression for industrial purposes!).

A general state of the art in the investigation of gene transfer from plants to bacteria shows that in this particular direction this is an extremely rare event, if it exists at all [13]. For a long time it was assumed that one possessed one example of an enzyme present in bacteria, copper superoxide dismutase, that came from an animal host. However recent evidence argues very strongly to the contrary [14, 15].The idea at that time was that the photoluminescent bacteria that were found in the gut of the swordfish had acquired their copper SOD gene from their host by horizontal transfer, since only manganese and iron SODs were known to be present in bacteria at that time. Recently the subject took a renewed interest however, when it was found that copper SOD could be involved in pathogenesis, and that it was in fact widespread among Gram negative bacteria [16-18]. It becomes therefore difficult at this point to separate between horizontal transfer and divergent evolution. A similar type of study suggested that bacterial catalases may have been acquired by horizontal transfer from a fungus species [19]. Data have accumulated for a long time for other genes, in particular genes involved in the synthesis of antibiotics to suggest their propagation by horizontal transfer. Here the selection pressure seems important: indeed those organisms that make certain antibiotics have to be immune to them, and therefore should either differ in their general metabolism, or synthesise genes of resistance. In the case of penicillins and the related cephalosporins, having been discovered, as everybody knows, in a fungus, it was interesting to study the possibility of their biosynthesis (and resistance) in bacteria. Most rest on non-ribosomal protein synthesis, a process that may be of very ancient origin. Other enzymes, involved in the synthesis and degradation of metabolites specific to eukaryotes, have been found in bacteria [20]. The hypothesis of horizontal transfer stems from the fact that the product not being made in bacteria, it is difficult to see how divergent evolution could have maintained its similarity to an eukaryotic counterpart. It is clear from these studies that, if the transfer occurred, this happened during extremely long periods of time, and often very early during evolution (in particular at a time when oxygen started to be a major pollutant of the Earth [21, 22]). To our knowledge, there is no widespread evidence for recent and efficient direct gene transfer from eukaryotes to prokaryotes (see the review by Nielsen et al, [13]), except in man-made experiments (cloning and heterologous expression of eukaryotic genes in bacteria is of course an obvious case in point). It is interesting to note that implicit perception of this biological fact may have been at the source of much resistance against genetic engineering at its onset, in the mid 1970s.

The main difficulty for this process to occur in a natural environment is not the DNA transfer per se, but the expression of the eukaryotic genes in bacteria (in particular in Gram positive bacteria). Indeed, separation of the compartments for transcription and translation in eukaryotes resulted in a specific way to initiate translation, without the signals that are required in prokaryotes (ribosome binding sites in particular). In addition most eukaryotic genes are interrupted by introns. This means that functional transfer would occur only with genes copied from their mRNA template, through the action of a reverse transcriptase. The fact that it can be performed in the laboratory shows nevertheless that it is in principle a possible, although extremely unlikely, event. The fact that only very rare traces of this event can be found in extant organisms suggests that, unless there is a strong and specific selection pressure, there is no maintenance of the transferred gene in the bacteria where it occurred. Furthermore there is strong accumulating evidence that most proteins do not diffuse freely in the cytoplasm of the cells, but are part of multiprotein complexes. This means that the presence of a foreign gene will be more a burden than a help. Because the organisation of prokaryotes and eukaryotes is so different, this may explain why the gene transfer from these widely different domains of life is exceptional: there is no selective advantage to have developped an efficient means of gene transfer, if the gene products cannot be easily functional after transfer. In brief, transfer from eukaryotes to prokaryotes is an extremely rare event despite the widespread existence of DNA in nature [23].

Transfer from Bacteria to Eukarya is probably much more frequent, in part due to symbiosis of bacteria that became mitochondria and chloroplasts (of course the transformation plasmids of Agrobacterium sp. have also a role, but much smaller than real symbiosis). This natural process was exploited as the first way to produce plant GMOs. Indeed some bacteria have a specific systems meant to transfer genes to plants [24]. Curiously, despite this ubiquitous systems, leading to plant tumors, there is apparently no large invasion of plants by bacterial genes (except, once again, for those that were present when the genes from endosymbiotic bacteria and cyanobacteria led to the formation of mitochondria and chloroplasts). Despite the fact that it is widespread, there is not much indication that bacterial genes (antibiotic synthesis genes, for example) have invaded the plant kingdom. This probably reflects an intrinsic way of the various kingdoms to protect themselves against foreign genes, a phenomenon which has an obvious implication in resistance to viruses, for example.

At this point of our reflection, we have alluded to many processes that permit gene transfer, such as conjugation, viral infection, DNA transformation, retrotransposition and similar phenomena. It is interesting to review briefly their existence in nature. It seems remarkable that, instead of metabolic genes, all kinds of retro-elements have repeatedly invaded genomes. This is the case for example of the extremely widespread Mariner element, discovered as a major invador by Hugh Robertson [25-27]. This element has invaded many animal genomes (including primates, at least twice), leaving related organisms free from it. But it should be noticed that, although invasion is frequent, this is of course to be considered at the geological time-scale (for example the invasion of the primate lineage occurred twice, the last time being some 10-30 millions years ago...). The important observation linked to this element is that it may transport with it adjacent genes and therefore greatly contribute to gene transfer. It is also important (but this has been performed only in laboratory experiments) to recognize that this element can work in bacteria [28]. Finally, group I introns (and group II introns [29]) are also major indicators of horizontal transfer. But there is a controversy about their origin and the time of their transfer between genomes. In particular it is still not resolved why, although typical eukaryotic markers, they have been found in bacteriophages such as E. coli T-even bacteriophage T4 and B. subtilis bacteriophages [30-32].

3. Gene transfer between domesticated organisms and undomesticated organisms

Domestication of plants and animals used genetics as soon as this science became available. Before that, domestication was a very slow process [33], with much outbreeding between domesticated organisms and their wild type parents. However, although genes can in principle move freely in either direction, it does not seem to have an important impact. If this were so, domesticated organisms would have invaded the planet in the absence of Man action, and this is not the case. Triggered by profit-making journals, an extremely heated controversy exists in the case of maize, where it has been suggested that genes from GMOs spread into the environment [34]. However this is far from being proven, and, even if it were proven, the impact is extremely low [35]. In any case, the general impact of modern agriculture is enormous. This impact however is not due to gene flow, but to practices that eliminates wild organisms and uses fertilizers, purposedly replacing wild plants and animals by domestic ones, which makes specific assessment of the further possible role of GMOs difficult to perform. There is an immense literature of sayings about gene flow, but not much measurements supporting any important transfer: hunting practices and uprooting plants is by far, the most efficient way to replace wild genes by domesticated genes. In fact, experimental work in the domain is scarce, but  one can find studies indicating that domestication led to hybridization with a sustained flow from domesticated plants to wild plants, in the case of weeds [36, 37]. This is the most obvious situation for such a process to happen, since the only way for weeds to escape human action is to mimick, as much as possible, the phenotype of cultivated plants. However no study tells how far from cultivated fields this type of transfer can be observed. General observation in forests of the lack of similarity between the ancestors of cultivated trees and their present varieties argues against the idea that domesticated trees could easily survive in the wild. Knowledge of genome sequences will, in the near future, help us to resolve this issue in a non controversial way.

4. Stability of domesticated organisms in the environment

Domestication is 10,000 years old. Initially, starting with wild animals and plants, cross-breeding must have been extremely frequent: this is extremely well illustrated in perhaps the case of the oldest domesticated animal, the dog [38]. Hence a significant gene flow between domesticated and wild organisms took place during the whole of the Neolithic. However, as domestication proceeded, the selected traits were less and less adapted to wild life, and, when deprived of specific selection pressure, the gene flow reduced to an extremely low level. This explains why ancestors of present day domesticated grasses, for example, are still extant [39]. As a matter of fact, present day cultivated plants or domestic animals cannot survive in the wild. Most problems caused by biological invasions are due to wild-type animals or plants that conquer new biotopes, either because of present global warming, pollution, or accidental introduction (this is in particular the case in islands, which were until recently remaining isolated, or continents such as Australia, where invasion by European plants and animals has dramatic consequences [40-44]). Domestication of living organisms results in enhancement of specific traits that fit the usage Man wishes to do with them. This is quite visible in races of horse that vary between animals bred for competion, to animals bred for heavy work (this is disappearing with mechanisation) and animals bred for meat. The same apply to cattle with even more specialization, animals being bred either for meat or for milk. They also can be selected to survive in specific ecological niches (such as mountains for example). The main danger facing these animals (or plants) is not mixing up with wild species, nor particular danger towards Man, but, rather to disappear because of lack of financial interest for breeders, while they represent a very important gene pool  selected through centuries or even millenia of hard work. Conservation of domestic animals and plants genetic resources is a major challenge, that can only be met with special germ-line banks and/or association with specific ways to improve revenues of the breeders [45].

Provisional conclusion and  caveat: is life predictable?

All this might tell us that GMOs do not pose any problem. In fact, I wish to end this article by a caveat (that applies to any type of man-driven selection, standard domestication included), to stress that, because life is in essence unpredictable, we should never say that we know what will be the outcome of any of our actions in the domain. The central concept often named the Central Dogma of Molecular Biology, is that "information transfer", always associated with two processes essential for life, metabolism and compartimentalization, rests on two laws. The first one is complementarity, which to a sequence of primordial motifs associates a complementary sequence of motifs (chain of letters) that results in exact and complete specification of the former. The second law is coding, which uses a cipher allowing the rewriting of the DNA text symbolized by a four letter alphabet into a second text symbolized by a twenty letter alphabet. Molecular biology used to rely on the alphabetic description of the genes. It is now further exploring the metaphor, in which the genome is understood as providing the complete text of the recipe allowing the construction, the development, the survival and the evolution of any living organism. Let us stress again that it would be a deep mistake to mix up the book of recipe with the meal, as journalists do when they try to make the public believe that the knowledge of the Human Genome sequence will allow us to cure all diseases! What this metaphor, which comprizes an alphabetic text and a coding process, accounts for is the possibility of a true creation (i.e. sudden appearance of an entity which cannot be predicted from what was existing before, but only accounted for a posteriori in the chain of evolution), as repeatedly witnessed when analyzing the evolution of species. Comparing genomes allows one to tell those places where creation operates: the Darwinian trio Variation / Selection / Amplification makes material systems evolve, evolution creates new functions, which, to come to being, recruit preexisting structures (hence the "tinkering" features stressed by François Jacob). And hence the impossibility of predicting a function from the structure alone [46]. Naturally, because any creation is unforeseeable, one will only be able to explain a posteriori how this or that recruitment allowed the development of that particular function. It is there that we begin to construct the means to explore, from what we know of the past, what might be conditions for new creation to come, as well as the places where it will be possible (this is a central question, for example, when one wishes to understand how and where new diseases can emerge). The alphabetic metaphor of the genetic program is so appropriate that one can represent the cell and its program as a computer and its program. That particular computer would have the particularity, in the program it deciphers and modifies, to provide the appropriate instructions not only for its own duplication, but for the duplication of the machine itself. The most elementary machine of this type, as proposed by Alan Turing, is made of a read/write head, of a mobile "tape" which passes through the head and which carries only a linear sequence of symbols, and of a mechanical device which allows the tape to go forward, go backward, stop, or read the symbols in the band, as a function of the previous readings by the head. This machine is essentially defined by the fact that it allows formal separation between the machine proper (the read/write head and the mechanics needed to make the tape move), the data which set the conditions under which the program is executed, and the program itself. The major conceptual question asked by this metaphor is whether one is allowed to separate the program from the machine. As I argued elsewhere [1], the experiments which are at the root of the molecular techniques of biotechnology provide us with a first indication that this separation is effective. It is demonstrated in an illuminating way in these experiments of cell cloning that have been so fashionable since 1997. For a living organism the genetic program leads to the real construction of the machine. The logic needed for an effective complementarity between matter and its symbolic representation has long been discussed by von Neumann in the mid-sixties [47]. It requires some "jump" from matter to abstractness. If the program is a model for a universal construction rule that plans the duplication of both its own description and that of the machine, then the duplication of the program together with that of the machine becomes logically possible. But this requires that the machine and its symbolic representation (program in the program) be somehow separated. Thus there is a need, in addition to the genetic program, for the existence, somewhere, of an image of the machine. This is still an open question, that must be answered if we wish to pursue the consequences of the alphabetic metaphor.

If this metaphor is appropriate, then we have to accept its deepest consequences. And they are that, in principle, it will never be possible to predict exactly what will be the concrete outcome of a given genetic program. Of course when things are kept very similar to one another they must more or less behave in the same way, but very minute differences, such as those that separate dogs from wolves, can have enormous consequences [48]. We should always bear this in mind when considering our practices involving living organisms.


1. Danchin A: The Delphic boat. What genomes tell us. Cambridge, USA: Harvard University Press; 2003. [Comment]

2. Ordon F, Friedt W, Scheurer K, Pellio B, Werner K, Neuhaus G, Huth W, Habekuss A, Graner A: Molecular markers in breeding for virus resistance in barley. J Appl Genet 2004, 45:145-159. pubmed

3. Rothschild MF: Porcine genomics delivers new tools and results: this little piggy did more than just go to market. Genet Res 2004, 83:1-6. pubmed

4. Danchin A, Médigue C, Gascuel O, Soldano H, Hénaut A: From data banks to data bases. Res Microbiol 1991, 142:913-916. pubmed

5. Yokoyama K, Doi Y, Yamane K, Kurokawa H, Shibata N, Shibayama K, Yagi T, Kato H, Arakawa Y: Acquisition of 16S rRNA methylase gene in Pseudomonas aeruginosa. Lancet 2003, 362:1888-1893. pubmed

6. Dzidic S, Bedekovic V: Horizontal gene transfer-emerging multidrug resistance in hospital bacteria. Acta Pharmacol Sin 2003, 24:519-526. pubmed

7. Médigue C, Rouxel T, Vigier P, Hénaut A, Danchin A: Evidence for horizontal gene transfer in Escherichia coli speciation. J Mol Biol 1991, 222:851-856. pubmed

8. Weinbauer MG, Rassoulzadegan F: Are viruses driving microbial diversification and diversity? Environ Microbiol 2004, 6:1-11. pubmed

9. Futterer O, Angelov A, Liesegang H, Gottschalk G, Schleper C, Schepers B, Dock C, Antranikian G, Liebl W: Genome sequence of Picrophilus torridus and its implications for life around pH 0. Proc Natl Acad Sci U S A 2004, 101:9091-9096. pubmed

10. Kurland CG, Canback B, Berg OG: Horizontal gene transfer: a critical view. Proc Natl Acad Sci U S A 2003, 100:9658-9662. pubmed

11. Carlson CR, Kolsto AB: A small (2.4 Mb) Bacillus cereus chromosome corresponds to a conserved region of a larger (5.3 Mb) Bacillus cereus chromosome. Mol Microbiol 1994, 13:161-169. pubmed

12. Nitschké P, Guerdoux-Jamet P, Chiapello H, Faroux G, Hénaut C, Hénaut A, Danchin A: Indigo: a World-Wide-Web review of genomes and gene functions. FEMS Microbiol Rev 1998, 22:207-227. pubmed

13. Nielsen KM, Bones AM, Smalla K, van Elsas JD: Horizontal gene transfer from transgenic plants to terrestrial bacteria--a rare event? FEMS Microbiol Rev 1998, 22:79-103. pubmed

14. Cornish-Bowden A: The amino-acid sequences of the copper/zinc superoxide dismutases from swordfish and Photobacter leiognathi confirm the predictions made from the compositions. Eur J Biochem 1985, 151:333-335. pubmed

15. Bannister WH, Bannister JV, Barra D, Bond J, Bossa F: Evolutionary aspects of superoxide dismutase: the copper/zinc enzyme. Free Radic Res Commun 1991, 12-13 Pt 1:349-361. pubmed

16. Wilks KE, Dunn KL, Farrant JL, Reddin KM, Gorringe AR, Langford PR, Kroll JS: Periplasmic superoxide dismutase in meningococcal pathogenicity. Infect Immun 1998, 66:213-217. pubmed

17. Dunn KL, Farrant JL, Langford PR, Kroll JS: Bacterial [Cu,Zn]-cofactored superoxide dismutase protects opsonized, encapsulated Neisseria meningitidis from phagocytosis by human monocytes/macrophages. Infect Immun 2003, 71:1604-1607. pubmed

18. Farrant JL, Sansone A, Canvin JR, Pallen MJ, Langford PR, Wallis TS, Dougan G, Kroll JS: Bacterial copper- and zinc-cofactored superoxide dismutase contributes to the pathogenesis of systemic salmonellosis. Mol Microbiol 1997, 25:785-796. pubmed

19. Klotz MG, Klassen GR, Loewen PC: Phylogenetic relationships among prokaryotic and eukaryotic catalases. Mol Biol Evol 1997, 14:951-958. pubmed

20. Baker ME: Evolution of mammalian 11beta- and 17beta-hydroxysteroid dehydrogenases-type 2 and retinol dehydrogenases from ancestors in Caenorhabditis elegans and evidence for horizontal transfer of a eukaryote dehydrogenase to E. coli. J Steroid Biochem Mol Biol 1998, 66:355-363. pubmed

21. Watts RA, Hunt PW, Hvitved AN, Hargrove MS, Peacock WJ, Dennis ES: A hemoglobin from plants homologous to truncated hemoglobins of microorganisms. Proc Natl Acad Sci U S A 2001, 98:10119-10124. pubmed

22. Moens L, Vanfleteren J, Van de Peer Y, Peeters K, Kapp O, Czeluzniak J, Goodman M, Blaxter M, Vinogradov S: Globins in nonvertebrate species: dispersal by horizontal gene transfer and evolution of the structure-function relationships. Mol Biol Evol 1996, 13:324-333. pubmed

23. Breitbart M, Felts B, Kelley S, Mahaffy JM, Nulton J, Salamon P, Rohwer F: Diversity and population structure of a near-shore marine-sediment viral community. Proc R Soc Lond B Biol Sci 2004, 271:565-574. pubmed

24. Gelvin SB: Agrobacterium-mediated plant transformation: the biology behind the "gene-jockeying" tool. Microbiol Mol Biol Rev 2003, 67:16-37, table of contents. pubmed

25. Robertson HM, Lampe DJ: Recent horizontal transfer of a mariner transposable element among and between Diptera and Neuroptera. Mol Biol Evol 1995, 12:850-862. pubmed

26. Robertson HM, Martos R: Molecular evolution of the second ancient human mariner transposon, Hsmar2, illustrates patterns of neutral evolution in the human genome lineage. Gene 1997, 205:219-228. pubmed

27. Lampe DJ, Grant TE, Robertson HM: Factors affecting transposition of the Himar1 mariner transposon in vitro. Genetics 1998, 149:179-187. pubmed

28. Akerley BJ, Rubin EJ, Camilli A, Lampe DJ, Robertson HM, Mekalanos JJ: Systematic identification of essential genes by in vitro mariner mutagenesis. Proc Natl Acad Sci U S A 1998, 95:8927-8932. pubmed

29. Toro N: Bacteria and Archaea Group II introns: additional mobile genetic elements in the environment. Environ Microbiol 2003, 5:143-151. pubmed

30. Shub DA, Gott JM, Xu MQ, Lang BF, Michel F, Tomaschewski J, Pedersen-Lane J, Belfort M: Structural conservation among three homologous introns of bacteriophage T4 and the group I introns of eukaryotes. Proc Natl Acad Sci U S A 1988, 85:1151-1155. pubmed

31. Goodrich-Blair H, Scarlato V, Gott JM, Xu MQ, Shub DA: A self-splicing group I intron in the DNA polymerase gene of Bacillus subtilis bacteriophage SPO1. Cell 1990, 63:417-424. pubmed

32. Sandegren L, Sjoberg BM: Distribution, sequence homology, and homing of group I introns among T-even-like bacteriophages: evidence for recent transfer of old introns. J Biol Chem 2004, 279:22218-22227. pubmed

33. Khush GS: Green revolution: the way forward. Nat Rev Genet 2001, 2:815-822. pubmed

34. Quist D, Chapela IH: Transgenic DNA introgressed into traditional maize landraces in Oaxaca, Mexico. Nature 2001, 414:541-543. pubmed

35. Wisniewski JP, Frangne N, Massonneau A, Dumas C: Between myth and reality: genetically modified maize, an example of a sizeable scientific controversy. Biochimie 2002, 84:1095-1103. pubmed

36. Papa R, Gepts P: Asymmetry of gene flow and differential geographical structure of molecular diversity in wild and domesticated common bean (Phaseolus vulgaris L.) from Mesoamerica. Theor Appl Genet 2003, 106:239-250. pubmed

37. Coulibaly S, Pasquet RS, Papa R, Gepts P: AFLP analysis of the phenetic organization and genetic diversity of Vigna unguiculataL. Walp. reveals extensive gene flow between wild and domesticated types. Theor Appl Genet 2002, 104:358-366. pubmed

38. Leonard JA, Wayne RK, Wheeler J, Valadez R, Guillen S, Vila C: Ancient DNA evidence for Old World origin of New World dogs. Science 2002, 298:1613-1616. pubmed

39. Salamini F, Ozkan H, Brandolini A, Schafer-Pregl R, Martin W: Genetics and geography of wild cereal domestication in the near east. Nat Rev Genet 2002, 3:429-441. pubmed

40. Drake JM, Lodge DM: Global hot spots of biological invasions: evaluating options for ballast-water management. Proc R Soc Lond B Biol Sci 2004, 271:575-580. pubmed

41. Courchamp F, Chapuis JL, Pascal M: Mammal invaders on islands: impact, control and control impact. Biol Rev Camb Philos Soc 2003, 78:347-383. pubmed

42. Stachowicz JJ, Terwin JR, Whitlatch RB, Osman RW: Linking climate change and biological invasions: Ocean warming facilitates nonindigenous species invasions. Proc Natl Acad Sci U S A 2002, 99:15497-15500. pubmed

43. Kennedy TA, Naeem S, Howe KM, Knops JM, Tilman D, Reich P: Biodiversity as a barrier to ecological invasion. Nature 2002, 417:636-638. pubmed

44. Garcia-Ramos G, Rodriguez D: Evolutionary speed of species invasions. Evolution Int J Org Evolution 2002, 56:661-668. pubmed

45. Huby M, Griffon L, Moureaux S, De Rochambeau H, Danchin-Burge C, Verrier E: Genetic variability of six French meat sheep breeds in relation to their genetic management. Genet Sel Evol 2003, 35:637-655. pubmed

46. Danchin A: From protein sequence to function. Curr Opin Struct Biol 1999, 9:363-367. pubmed

47. von Neumann J: The Computer and the Brain. New Haven: Yale University Press; 1958 (reed 1979 ).

48. Saetre P, Lindberg J, Leonard JA, Olsson K, Pettersson U, Ellegren H, Bergstrom TF, Vila C, Jazin E: From wild wolf to domestic dog: gene expression changes in the brain. Brain Res Mol Brain Res 2004, 126:198-206. pubmed

uk-flag The site in English

fr-flagTout le site en Français

<< A Presocratic view of BiologyUne vision Présocratique de la Biologie>>