Macroevolution: Its all around us...

I'm not sure if any has already posted this but the Whiting lab themselves are hosting a PDF format version of the full article.One question this paper raised for me was that of how similar the developmental program of the 're-evolved' wings were to the more primitive wingless program.There are certainly enough conserved elements common to both wing development and the development of other systems, such as the nervous system and legs, that most of the actual genes should be maintained but I am interested in how wing specific regulatory elements would be maintained when they should be under no-selective pressure whatever.I don't know how easy it is to observe Phasmid embryonic development or to study gene expression at different instars. But even a purely bioinformatic approach to the genomic DNA of the different species could answer a number of these questions, especially if the development of the wing in its original form is conserved in relation to a well understood model like the drosophila wing, the homology of the wing veins is a promising sign that they might be substantially conserved.Some of these issues are also raised in a review of Whiting's work in Current Biology (Stone and French, 2003)TTFN,WK

quote:

---

Maybe so but it seems there would be a difference between a cow or elephant or whatever becoming a whale and a large feline being bred down to a domestcated housecat.

---

Yes, there is a difference.It's called "selection pressure."

quote:

---

Each different family like felis, canis, equis, and vulpes is a different kind of animal.

---

So, "kind" is actually equivalent to "family"?Well, what happens if I show you an animal which seems to have characteristics of several different "kinds"?Also, where does genetics play out in the definition of "kind"?Do we ignore genetic similarities or use them to determine relatedness between "kinds"?This message has been edited by schrafinator, 05-20-2005 09:51 AM

Hi eclipse, I'd be really grateful if you would consider answering Mick's oink challenge.You keep asserting that there are different types of animals, some micro-level variants of each other, and some completely different "kinds". I am at a loss to understand where this idea comes from.Thanksmick

Isn't a 'neutralist' explanation sufficient ?
I mean wouldn't they be maintained as long they didn't exact any meaningful cost in their non-expressed state?When wings are disadvantageous, the easiest way to 'turn off' their development would be a simple mutation in a key operon that acts a switch. All the other genes subservient to this 'switch' than lie dormant and are under no direct selection to be eliminated.The only possible problem here, is the possibility of some sort of 'genomic cost' for maintaining inactive genes on the chromasome. Is there any evidence for anything like that? It seems conceivable if space on the genome is limiting in any way, but haven't there been many examples of gene duplication where the duplicate was carried along dormant on the geneome for a long period before being reactiviated in a slightly modified form? Wouldn't this point to a (potentially) very low cost for retaining non-active genes ?

They would be maintained as long as there wasn't a mutation that caused their deletion - but the question should be: how long could they be maintained without accumulating such a number of mutations that they become nonfunctional or harmful if expressed? How long could they hang around in a dormant state without becoming pseudogenes?Mick

So your reasoning is that once the controling operon is swtiched off, it would no longer constitute a highly conserved region and deleterious mutations could accumulate without being eliminated by selection because the gene is silent. That makes sense.But do highly conserved regions automatically lose that status as soon as they are no longer transcribed ? Are we safe in assuming that all regions of the genome experience mutation at equivalent rates? (an honest question - I have no idea)Assuning they do, perhaps this would indicate that the 'switch' is thrown somehwere post-transcription. For example, messenger RNA sometimes undergoes splicing or processing prior to translation in higher eukariotic cells, potentially yielding more than one signal protein. Preserving a 'use' for one form of the mRNA would conserve the encoding DNA sequence even though the other form (that triggers wing development) is not produced.Alternatively, if the signal protein itself undergoes processing in the cytoplasm into more than one form, the switch could also be thrown at this level. Retention of utility of one form of the signal protein (other than the one triggering wing development) would then favor conservation of the encoding DNA sequence even in the apterous state.Assuming you are right, and without either of these two types of mechanisms I have proposed to maintain sequence conservation, wouldn't that suggest that the longer a particular lineage remained in the apterous state, the lower the probability of its ever regaining functional wings?Surely that doesn't seem to be the case.
The repeated cycle of 'gain and loss' suggests that the controlling region is well conserved even when it isn't expressed.This message has been edited by EZscience, 05-22-2005 04:03 PM

In terms of being themselves targets of slection? Yes, provided they are indeed wing specific and have no other function. No there are a number of factors which can locally affect mutation rates, transcriptional activation is one such, although not neccessarily such a high concern for germline mutations. I think we are talking across each other to some extent.I am quite happy to accept that all protein coding sequences neccessary for wing development are conserved, because almost all the proteins involved have developmental roles in other systems.What I am less convinced of is the conservation of small regulatory sequences, as small as 5bp even ibp can be crucial to a sites function, which can be located kilobases up or downstream of the gene and serve to drive expression specifically in the wing. Only in the functional regions, look at any multiple alignment of DNA or protein sequences from diverse species and you will often see specific regions of high conservation amid other regions of very low homology, these regions are commonly identified as those most important to the conserved function of the protein. Only those regions which are neccessary to the formation and function of the non-wing specific form. As long as the conformation was maintained enough for proper processing and wing specific functional residues could quite readily be lost. It certainly would suggest that, so one key question is how long ago the various species divergences ocurred. Over tens of millions of years it would be very unlikely that all of the neccessary wing specific regions would be conserved. But not all of the regions would be neccessary for a wing to form, only for one whose development was identical to the ancestral wing. Not neccessarily, which is why a proper developmental genetics/ bioinformatics based approach would be needed to find out if this was true or if other developmental pathways were modified or co-opted to allow the secondary evolution of wings.TTFN,WKThis message has been edited by Wounded King, 05-23-2005 05:09 AM

WK, First let me first thank you for your thoughtful reply.
I have read it several times now.
I am not as familiar as you with the mechanics of genetic expression,
so it is very interesting for me to follow your reasoning on this.
I am just not sure I can contribute much further to what you have said. Seemingly, this would indicate that both lineage divergence events and wing 'gain and loss' events likely occured within short periods of abrupt change, consistent with the 'punctuated equilibria' model. As an ecologist, I am stimulated to imagine the types of abrupt ecological changes that could be responsible for such patterns. As a molecular biologist, you obviously have a good grasp of the genetic constraints limiting these processes. Wouldn't the former be the more parsimonious explanation?
I mean, co-opting new developmental pathways to the same end seems less likely to me than reactivating old ones. If true, it would seem to imply that a tremendous degree of evolutionary flexibility exists w/r/t the assignment of function to regulatory sequences.

Would it? I would have thought that if we assume that the 're-evolved' devlopmental program is identical to the ancestral one then what it would argue is that however long the periods of change were they occurred close together in evolutionary terms, i.e within a million years or less of each other rather than tens of millions. It may be that divergence times could have been calculated from the same data that was used to construct the tree, but I didn't see any in the paper.If the mutation leading to wing loss is only based on a single mutation, at whatever level, then it may well have arisen effectively instantaneously. As to regaining the trait, once again it depends on whether it is effectively a simple backwards mutation to the original form or an co-option of other extant systems. Once again the answer is 'not neccessarily'. It is only parsimonious unless one considers that some additional mechanism must be required to maintain sequences no longer under selective pressure for however long the gap is between the loss and regaining of flight.Without knowing more about the specific regulatory sequences which would actually be involved it is hard to speculate on how much change would be needed. Some core binding sites for transcription factors are very small, only four base pairs, and it is not a big stretch of the imagination to think that a four base sequence may be formed quite frequently in DNA.Regulatory regions are pretty flexible in terms of their downstream effects. Deletions of small regions can radically affect the expression pattern of a gene such as causing it to be globally expressed or lost in a specific tissue. If such a gene is near the top of a regulatory cascade then a lot of othe genes will follow.I realise this may sound to you a lot like the scenario that you were describing, with a single master control gene for wing development regaining its function and the whole system starting up again, but in the Drosophila wing there are a number of wing specific patterns of expression which require specific regulatory elements to pattern the dorsoventral and antero-posterior axes. It is these sort of elements I think would be likely to be lost. The genes might well still be expressed, but not in exactly the patterns seen in the ancestral wing. so you might have a somewhat cruder form of the ancestral wing program, which might then be subsequently further refined. In this scenario I would expect to see quite significant differences in the wing specific upstream regulatory regions compared to normal homologous wing structures, and possibly differences in the details of gene expression during wing development.TTFN,WKThis message has been edited by Wounded King, 05-26-2005 01:14 PM

I've spent the past two hours carefully reading all posts, and have yet to find any credible examples of vertical transformation. I have, however, found numerous references to instances of horizontal variation (particularly EZscience's original thread starting post) - exactly how is this proof of "macroevolution all around us"? There has been no example of boundaries crossed between kinds.

So you are in the dogs giving birth to cats camp then?TTFN,WK

Aye, and it’s a logical choice to be part of that camp. It's simple and irrefutable - what more could you want? The only credible references I keep seeing are those of horizontal variation, hardly proof of "macroevolution all around us"

Dogs giving birth to cats would refute the mainstream theory of evolutiuon.

"Dogs giving birth to cats" is just a simplistic description many people use for the process of macroevolution. I just meant I'm a firm believer in the boundaries between kinds. But that wasn't my point - I was referring to the lack of vertical transformation evidence. Something you'd think you'd find in a thread titled "macroevolution: It's all around us"

Hi MickD, By 'vertical transformation' I can only infer you mean changes that occur over evolutionary or geological time scales. It is these long time scales that are needed for taxa to truly diverge to the point where you might be inclined to consider them different 'kinds' (a terribly unscientific terminology, incidentally).The point is, degrees of relationship between taxa HIGHER THAN SPECIES LEVEL can only be inferred indirectly through measures of morphological similarities or genetic sequence homology. If we can observe speciation in progress, everything else makes sense by simple extrapolation. The (biological) process of separation is the most definitive event in macroevolution because it marks a separation of gene pools. From that point on, the two species exchange no information and follow independent evolutionary trajectories. In time they become more and more different. Extend this process over evolutionary time and you get different Genera, Families, Orders, etc. (and yes, even different 'kinds', if you must use that ambiguous term).So I think what you were expecting to find and didn't, stems from a misconception about what macroevolution really means in 'biological' terms as opposed to how creationists think of it, i.e. an ape giving birth to a human, or a dog to a cat. The key step in macroevolution is speciation. This has always been the sticking point (how it occurs - not 'if' it occurs) and everything beyond speciation is pretty much a logical continuance of diverging traits among (already) separate lineages.This message has been edited by EZscience, 05-27-2005 11:23 AM