Mendel wasn't entirely right

Since neither priority of posting nor direct reference to the scientific literature seem to count for anything anymore I might as well post the link to the original research article here.A letter to Nature reports research suggesting that Arabidopsis can, in some instances, revert to an ancestral wild type allele instead of a parental mutant. TTFN,WKThis message has been edited by Wounded King, 03-24-2005 05:28 AMThis message has been edited by Wounded King, 03-24-2005 12:29 PM

In answer to 1.It does affect evolution, but probably not as much as you might think. The proportion of revertants is not so high as to present any real problem for mutation and the spread of mutations, except perhaps in the case where the mutation is particularly detrimental. If anything this would seem to be a helpful accessory mechanism biasing evolution more towards beneficial mutations. It may retard the spread of an allele through the population, but that is about the only negative effect I can see.2: Anyone who is familiar with modern genetics could tell you that Mendel's laws are not the be all and end all of genetics. There are numerous examples of epigenetic markers, segregation affecting mechanisms and cytoplasmic modes of inheritance which are outwith the boundaries of Mendelian genetics.Given that Darwin formulated his theories about evolution with absoloutely no knowledge of Mendelian genetics I fail to see why it should be a nail in evolutions coffin. Certainly the modern synthesis does include a very strong emphasis on genetics, but it is certainly not restricted to Mendelian genetics.I think the problem is more with your view of the importance of Mendelian genetics in evolution than with the impact these findings will actually have.There are already many multi-factorial factors which must be treated as non-mendelian because of the complexity of analysing their inheritance. The only real problem will be for the poor population geneticists.The real key question in terms of evolution as a whole, and what role this phenomenon has played in it, is whether this is actually a wide spread phenomenon, indeed it may not be a wholly replicable one but I have no reason to doubt the research, time will tell. It may be a peculiarity of arabidopsis or a particular sub-domain of the plant kingdom. One other example that is referenced of an ancestral marker recurring is from (Song, 1995) in another plant paper.

I don't know much about Error Control Coding, but I think you may be wrong. There are a number of well understood mechanisms which maintain the genetic code and work to correct errors which arise during DNA synthesis. The new phenomenon may well be a new mechanism by which this can be achieved but it appears to be unique in that the 'repair' is not enacted until the subsequent generation or beyond. It is this delay in the correction of the 'error' that makes it so unusual.It may be that the Hothead gene itself has some peculiar novel role which allows this to happen. If it were a widespread phenomenon in response to de-novo mutation then I would imagine it would have been noticed by now in the many intensive programs of genetic crosses which are performed in plant science, especially on arabidopsis. The paper suggests that the mutation of HTH may greatly increase the frequency of an otherwise very rarely operating mechanism.I have to say that towards the end the authors discussion seems to verge on the teleological but it may just be that their ideas are so vague and numinous as to allow for many interpretations.TTFN,WK

You are making this much more mysterious sounding than what the authors propose. They don't suggest that the template is 'packaged' or 'hidden'. What they propose as a tentative model is that the template consists of a double stranded or otherwise stabilised form of RNA.It isn't 'hidden' simply because no one previously thought to look for it. I imagine the first thing they should do would be to use RT-PCR to show the presence of RNA encoding the ancestral template. This might also give them some idea of how many different templates there are depending on their approach.It is a huge leap to go from the 'cacheing' of previously extant alleles of a gene to the storing of all the future alleles that an organism will ever need. Lets be honest, your zygote would have to consist of almost nothing but dsRNA (for instance) it is almost as patently ridiculous as the preformationist concept of the homonculus.I appreciate that you are just throwing these up as hypothetical objections, but as you have formulated them they seem about as weak as all the current IDist objections to modern evolutionary theory.TTFN,WK

I'm sure they will make these objections, but all it will be doing is once again showing their lack of familiarity with the way science works in practice. Most grant applications require a substantial amount of background research suggesting they will be fruitful and all grant funding is heavily competed for.It is bound to be the case that the evidence for the more outre topics must be more compelling than is required for the everyday. To be frank only having to wait 6 months for the review process to come round again and gain funding seems pretty quick.TTFN,WK

I've corrected the link, it was originally to the advanced on line publication but now Nature has published it so the address moved.Their maximum of 10% revertants was only seen in this one instance of the Hothead mutation. If it was a common feature in arabidopsis it would have been noticed well before hand at that high a frequency. As I suggested before arabidopsis is probably one of the worlds most intesively genetically studied organisms. But the authors suggest that the corrective mechanism is triggered by stress. A beneficial mutation might well relieve stress on the organism or at least not induce it. The whole point about correcting stress inducing mutations is that they return you to a tried and tested allele for your ancestral environment. Your opinion is just that. People probably object to your suggesting that they treat evolutionary biology as the 'holy grail', as they assuredly do to your assuming that it is dogmatic. When people who seem entirely unfamiliar with the theories and practices of evolutionary biology start bemoaning it as dogmatic or a 'religion', those who work in the field get understandably upset. You may think you know it all, most of us aren't so arrogant.TTFN,WKThis message has been edited by Wounded King, 03-24-2005 12:43 PM

You are probably thinking of epigenetic markers such as methylation of DNA, methylation and acetylation of histones and a number of other not strictly genetic factors thought to be important in the development of the early embryo and not neccessarily perfectly reset in SCNT.None of these are transcription factors, but you may be thinking of something completely different.TTFN,WK

What are you talking about? As I have stated several times Arabidopsis is widely used as a genetic model system. It is used time and again to investigate genetics in just such controlled experiments over several generations. If there was a 10% reversion in all cases then it would have bee noticed by now. Hell, if it was that frequent Mendel would have noticed it with his peas!! 10% may form a significant proportion of the population, but it will never be the majority of the population, all you seem to be suggesting is that the reversion maintains an effectively ancestrral wild type population, well duh, which is what the paper said in the first place. As pink pointed out you are completely ignoring the action of selection. If the mutation is neutral then it doesn't matter if the allele reverts, if it is deleterious then reversion will increase the chances of extinction of the allele, if it is beneficial then selection should counteract any effect from the revertant, although the revertant may supply a desirable source of potential biodiversity for future environmental changes. In a word, No. The authors only make the most tentative of suggestions. It is really far too early to expect any sort of detailed picture of the mechanism. I could speculate, but that is all it would be.TTFN,WK

Thats rather an exaggeration my dear Holmes.In your rejected post (don't feel too bad, mine got rejected too) you say... This isn't Lamarckian. There are some arguable examples of mechanisms allowing the inheritance of acquired characteristics, but I don't think this could reasonably be classified as one.TTFN,WK

The effects would be enormous, so much so that if that were true we would already know it. The point I have made repeatedly is that this mechanism is clearly not in action at a larger scale for mutation in general. There have been enough genetic cross studies in a wide variety of organisms done for this to be readily apparent.You are making a number of totally unwarranted assumptions about the prevalence, consistency and effects of this mechanism.TTFN,WK

I think Judge is justified in saying it is rare. If this were a frequent occurence in Arabidopsis in any number of mutant backgrounds, or in any plant or organism lacking the Hothead gene, then I can't see how it could have been failed to be noticed. An up to 10% discrepancy from the expected mendelian ratios would be pretty significant in the sorts of numbers commonly used in genetic studies.TTFN,WK

I disagree. For the mechanism to be maintained from that distant ancestral species it surely has to have continued to be selected. If it was no longer conferring an advantage then why would it not have degenerated through processe like drift. Perhaps the many constituent parts are involved in other processes but mechanism specific functions in some would surely have been lost.Given that only one species of Arabidopsis has been studied yet it seems a bit pointless to be making up 'just so' stories out of thin air.The mechanisms operation is rare in the wild type, even a rarely operating mechanism can be maintained provided the advantage it confers is substantial enough. I don't see why it ever needs to have been widespread, and even if it was at some point then Judges question about selection on rarelu utilised traits is still relevant since it has to have been maintained since becoming rare.TTFN,WK

I think it is very interesting that there is such a pronounced parental bias in this phenomenon.While the authors show that reversion re-establishes the paternally inherited, by whatever means, ancestral HTH allele they don't provide quite such strong evidence in the maternal case.In their experiments showing the paternal origin of revertants they find no revertants in similar experiments with a hth/hth mother. In fact the only evidence of a maternal reversion is a couple of HTH/HTH genotype plants. They say... I'm not sure this follows. They may have some sequence data which determines the parental origin of the HTH alleles, but if so it isn't presented. There is no clear evidence of when this mechanism operates is it during germ cell production or in the zygote? The fact that they find more HTH/hth genotypes than wild type phenotypes is interpreted as being due to reversion in the embryonic tissues which do not form part of the adult plant, which may suggest that reversion occurs after fertilisation rather than during germ cell production.If reversion does take place in the zygote or early embryo then it may be that both revertant alleles come from the paternal line. Crosses of hth/hth lines of Columbia and Landsberg would presumably allow this to be determined, provided the rare double revertants turn up.I am fascinated as to the possibility of a store of genetic memory as dsRNA. Is this maintained at specific levels? How many copies of each gene from each generation are there? Are there stores of all of these dsRNAs in all cells? Surely not, if there were wouldn't RT-PCR experiments in Arabidopsis thrown up many bizzare results and false positives?How are these copies established? Are highly transcribed mRNAs better represented in the 'store' than infrequently transcribed ones? Are mRNAs which are highly transcribed specifically in the tissues of the developing germ cells preferentially stored? Do microRNAs play a part in this process?Are many revertants highly mosaic? If they can occur only in non-contributing embryonic tissues can they occur later in particular cell lineages? I am not sure that their genomic blot would neccessarily show up differences due to mosaicism, and I'm sure the PCR would be unlikely to, especially since they themselves note a number of apparently revertant genotypes without a revertant phenotype.Does the reversion occur earlier in the developing germ cells themselves? This might be consistent with the much higher rates of paternal to maternal reversion. I think it would be interesting to syudy the DNA of pollen and ovules and see if they show allelic variation in Hothead.There are so many more possibilites as well. I hope that they are well on the way to a further publication on this topic, it is one of the most fascinating phenomena in years.There is an article referenced in the latest Tnagled Bank which suggests that the phenomenon may simply be due to revertants due to mutation being selected by virtue of improved fertility of HTH pollen. The author suggests that a hth/hth background leads to general genetic instability ather than any targeted mehcanism, and that subsequent selection leads to the reappearance of the ancestral allele.There is no evidence presented which suggests that their other markers of genomic instability actualy represent ancestral alleles rather than simply novel polymorphisms. There is however some evidence along these lines, in that they saw reversion of the Erecta(er) phenotype in hth/hth er/er crosses at a similar frequency.Anyone have any feeling on the comparative merits of the original paper's or the critique's explanations of the phenomenon? I prefer the original papers explanation, but It seems a bit early to tell one way or another, I'm not sure that the selection based model really fully explains the results.TTFN,WK