Butterfly unlocks evolution secret

[C1ay]

From the BBC:

 

Why one species branches into two is a question that has haunted evolutionary biologists since Darwin.

 

Given our planet's rich biodiversity, "speciation" clearly happens regularly, but scientists cannot quite pinpoint the driving forces behind it.

 

Now, researchers studying a family of butterflies think they have witnessed a subtle process, which could be forcing a wedge between newly formed species....

 

For speciation to occur, two branches of the same species must stop breeding with one another for long enough to grow apart genetically.

 

The most obvious way this can happen is through geographical isolation.

 

If a mountain range or river divides a population of animals for hundreds of generations, they might find that if they meet again they are no longer able to breed.

 

But geographical isolation is not enough to explain all speciation. Clearly, organisms do sometimes speciate even if there is no clear river or mountain separating them....

 

More.....

 

A very interesting article illustrating progress via the scientific method.

[Biochemist]

A really interesting post, C1. I like the flavor of the article, mostly because it acknowledges the limits of our current understanding.

 

It does start out with a false premise (we must have speciation because we have species) but the rest is pretty fair. I happen to agree with the conclusion, but this premise is weak.

 

It also brings up more questions than it answers. We have what appears to be a pretty good example of potential sympatric speciation, but there is no identifiable environmental force that would drive this particular selection. We have a tendency for butterflies with different markings to breed with matching butterflies, and weak offspring when the two subspecies interbreed, but no obvious reason why the markings were differentiated in the first place.

 

The superficial view of this information would suggest that a new species is arising in the absence of selective pressure, and may indeed grow into a reproductive isolate without it.

 

This might be a meaningful counterexample to the tenets of natural selection. If a species can actually generate a reproductive isolate in the same geography without selective pressure, what drove the change?

[ThickSet]

I'm having trouble expressing this clearly so I ask you folks to help me here. I'm on a lot of meds.

 

The butterfly observation is nothing new. The reason 'why' is nothing new. 'How' it happens is nothing new. What we are discussing here applies to everything from butterflies to brains: evolution occurs with genetic accidents gaining a foothold due to environmental conditions favouring the particular mutation. Species adapt to their surroundings through survival. Hence tribalism, the vagaries of butterfly (and human) nature.

 

Come back.

[Biochemist]

...The butterfly observation is nothing new. The reason 'why' is nothing new. 'How' it happens is nothing new. What we are discussing here applies to everything from butterflies to brains: evolution occurs with genetic accidents gaining a foothold due to environmental conditions favouring the particular mutation. Species adapt to their surroundings through survival. Hence tribalism, the vagaries of butterfly (and human) nature....

TS-

 

WE have run a number of threads where I have underlined that the mutative mechanism underlying much of the thinking within speciation is open to question. The basic science support for it is mostly conjecture, not evidentiary.

 

It is not supportive to repeat standard dogma (like "we know x" and "we know y") when the topic is actually that we most certainly do NOT know x or y.

 

The article that C1ay posted was a really good example of things to question about our current thinking/presumptions about natural selection. And it says absolutely nothing about a mutative mechanism. This appears to be an example of sympatric speciation in the absence of an environmental pressure. Just take it as the data point that it is. This evidence may well be starkly at odds with the standard thinking on the subject.

[ThickSet]

Hi BioChem - many thanks, you are expanding my mind again.

 

I used the term 'tribalism'. Is the butterfly wing-marking not the same thing as tribalism, only manifested differently? Is it fair to say (generally) all animals tend to band together into smaller units. (I have an engineering background, so biology is out of my experience, and I speak without authority) I always thought evolution was so darned obvious I automatically accepted it. But then again, a lot of people use this same kind of approach to accept 'God' (ouch).

 

I'll try this again. It's always seemed obvious to me, reflecting on human nature, how this whole evolution thingy works - it is a requirement for all 'living' organisms to experiment. The driver is our very deepseated 'ego' = f(survival, domination, accident, environmental, and so on).

[Biochemist]

Is the butterfly wing-marking not the same thing as tribalism, only manifested differently? Is it fair to say (generally) all animals tend to band together into smaller units.

It is an interesting idea, but evolutionary biologists tend to think of tribalism as something that would be driven by a selective process that drove participants toward that sort of behavior.

 

The problem is that these "obvious" assessments often are "good from far, but far from good" when you dig into the details. Since you are an engineering sort, it is a little like the initial surprise and discomfort that so many felt with the dual slit experiments in the early days of quantum physics. Some of those results just did not make any sense. But they happenned to be true.

 

In the case at hand, there was no observable pressure for a subset of the animals to act separately. If there was no pressure to do so, and the set of animals did, it is not supporting the notion of natural selection.

 

Really good examples of natural selection are pretty rare. There are some, but most examples are assumed, not demonstrated. Ditto with the notion of a mutative mechanism to expand the gene pool. It is assumed, but not demonstrated.

 

Thanks for the interest. And welcome to Hypography!

[MortenS]

I hope you mean that good examples of speciation by natural selection are rare, not that good examples of natural selection are rare, Bio...

[Tormod]

It does start out with a false premise (we must have speciation because we have species)

 

Bio, this is not a false premise. We would not have species without speciation. When discussing evolution we need to talk the language of the theory. You may not agree that speciation happens but then you should also not use the term "species", since it *does* imply that speciation occurs to separate them.

 

We have the fossil record and also DNA testing as evidence that speciation has occured. What we don't have as of yet is speciation under observation - which is to be expected because these things take time.

 

For those not into the evolution theory, here is a good starting point:

http://evolution.berkeley.edu/evosite/evo101/VC1fEvidenceSpeciation.shtml

[Biochemist]

Bio, this is not a false premise. We would not have species without speciation. ..

Actually, T, I do agree that speciation happens. But I still think this is still a false premise.

 

The specifics of speciation are that species come from other species. I do not, however, think that it is obligatory to assume that species come from pther species just because multiple species exist. It is the evidence of the process (e.g., genetic drift, etc.) that supports speciation, not the existence of many species.

 

Insidious assumptions like these drive lots of false association and false causality as well.

[Biochemist]

I hope you mean that good examples of speciation by natural selection are rare, not that good examples of natural selection are rare, Bio...

MortenS- I did mean that good examples of natural selection are rare. A good example would show several elements:

 

1) a preexisting species

2) a preexisting stress on the species

3) an adaptation that was preferred in response to the stress

4) irreversibility of the adaptation

 

If selection drove speciation, we would add a fifth (for sexually reproductive species): Reproductive isolation.

 

Point #4 above is particularly rare in posited examples of natural selection.

 

We actually have very few examples of the above for sexually reproducing species. We have lots of apparent examples in bacteria, but even some of those examples (like acquired drug resistance via episome transfer) can be interpreted less like natural selection and more like a feature of the species.

 

I am not asserting that natural selection does not occur. I am asserting that the direct support of it is not as pervasive as some presume.

[MortenS]

MortenS- I did mean that good examples of natural selection are rare. A good example would show several elements:

 

1) a preexisting species

2) a preexisting stress on the species

3) an adaptation that was preferred in response to the stress

4) irreversibility of the adaptation

 

If selection drove speciation, we would add a fifth (for sexually reproductive species): Reproductive isolation.

 

Point #4 above is particularly rare in posited examples of natural selection.

 

We actually have very few examples of the above for sexually reproducing species. We have lots of apparent examples in bacteria, but even some of those examples (like acquired drug resistance via episome transfer) can be interpreted less like natural selection and more like a feature of the species.

 

I am not asserting that natural selection does not occur. I am asserting that the direct support of it is not as pervasive as some presume.

 

Ok, first, your list of "requirements".

 

I agree with the three first requirements, and they are not that difficult to demonstrate, provided the selection pressure is strong enough for us to observe during our life-time.

 

In plants, numerous examples of natural selection towards higher tolerance of heavy metals exist. It can be shown that plants not exposed to heavy metals compete very badly with descendants of plants exposed to heavy metals.

 

Same can be demonstrated quite easily with insects and resistance against nearly all types of insecticides:

 

Take insecticide-resistance in malaria carrying mosquitoes such as Culex pipiens or Anopheles gambiae. One single point mutation in the gene ace-1 (acetylcholinesterase), where glycine is replaced with serine, is causing the insecticide resistance in strains of C. pipiens and strains of An. gambiae. (source: http://environmentalrisk.cornell.edu/WNV/WNV-LArchiveOld/5-23-03.html).

 

This seems to be a very effective mutation, one that eventually will occur in most insect populations populations subjected to insecticides targeting acetylcholinesterase.

 

 

As for the fourth requirment, that one is no requirement at all to demonstrate that natural selection occurs or have occurred in the past.

 

Natural selection is currently operating in malaria-infected populations in Africa, where children that are heterozygotes for sicklecell anemia have higher survival rates when infected by the malaria-parasite than children withouth sickle-cell anemia (homozygoty for sickle-cell anemia a fatal condition) This is an example of a trait that will never, ever be fixated, but natural selection will still operate on it. If the malaria-parasite dissappears, there will be selection against heterozygotes of sickle-cell anemia, and the proportion of heterozygotes will eventually go down in the population.

 

This last example do not fulfill your requirements, but it is still natural selection in operation, we are just in the middle of it.

[Biochemist]

...I agree with the three first requirements, and they are not that difficult to demonstrate, provided the selection pressure is strong enough for us to observe during our life-time.......

This last example do not fulfill your requirements, but it is still natural selection in operation, we are just in the middle of it.

Thanks for the thoughtful post, MS-

 

We might be debating terminology, but I am not sure. I am using "natural selection" in the narrow case where a specific feature of a species is preferred due to enviroinmental factors, and as a result becomes the generalized phenotypic expression of the species (or a new species) over time. I think your example of sickle cell anemia as a defense for malaria is a very good example. It would also (over time) tend toward irreversibility since the predominance of the recessive alleles could eventualy extinguish the dominant non-sickle trait. My advocacy for point #4 above, is that if the trait reverses, there was no permanent change in the gene pool, and nothing was actually "selected". Thus, I think it is useful to separate the propensity to select from the evidence of actual selection.

 

Your example of a potential mutative alteration that results in resistance to acetylcholinesterase inhibitors is a less compelling example. The resistance does occur, but the question is whether it is natural selection. You might note that you suggeted it was a "mutation" that substituted the glycine for the serine, and then you suggested that this will spread to other populations of insects exposed to ACHE inhibitors. If this indeed happens, it is highly unlikely that it was a mutative mechanism.

 

The probability of an identical mutative alteration in a gene occurring (exactly one specific amino acid substitution, without other mutative damage to other enzyme systems) in multiple species is vanishingly small. If this were a mutative mechanism, we would expect to see tens of thousands of different genotypic expressions for the same phenotype within a single species (since most surviving mutative geneotypic expressions would not have a phenotypic expression).

 

The fact that the researchers could isolate a single variant residue, suggests that the gene in the insect population is highly homogeneous. This means that there is about a 1 in 20^300 chance for a successful mutation of this gene (assuming 300 residues in the enzyme). And this is before you add the multipliers for a mutative event not causing other lethal damage in some other system.

 

Even if we assume a large breeding population and a short birth cycle, the odds of a specific mutation occurring again in our lifetime in a separate species is pretty small. If it does occur again, it argues that this is an adaptive feature of the species (or the class or order, etc), rather than a mutation. It would be a pretty complex adaptive feature (and a pretty interesting investigative topic) but still a feature, not a mutation. If it is an adaptive feature, the specific phenotypic expression might revert in the absence of ACHE inhibitors. That is, there might not be any permanent chance in the gene pool, so nothing was actually selected.

 

My point is that if an adaptive feature is readily reversible or readily reproducible, it might not be an example of selection. Some pale skinned humans get tan in the sun. The tan is shed over time as the epidermis is shed. We could posit an environment where over time where those that tan easier are more resistant to skin cancer. But it if the environmental stress does not occur to the extent that the feature is essentially permanent, this is only an example of an adaptive feature, not selection, because there was no lasting change in the gene pool.

 

Am I making any sense here?

 

Thanks again for the thoughtful response.

[MortenS]

Thanks for the thoughtful post, MS-

 

You are welcome, I'll try and make another one here ;)

 

We might be debating terminology, but I am not sure. I am using "natural selection" in the narrow case where a specific feature of a species is preferred due to enviroinmental factors, and as a result becomes the generalized phenotypic expression of the species (or a new species) over time.

 

That defintion is too narrow for me. Your definition, in my opinion means that we cannot identify natural selection in process, just after the fact. My understanding of natural selection, clearly includes your understanding, but it is bit broader, as I do not view the species, but the population in a particular geographic area as the unit in which natural selection should be studied and observed. Different populations of the same species may be exposed to various environmental stresses in various parts of its range, and develop different adaptations (you can often see this in northern strains and southern strains, and their difference in dealing with cold temperatures)

 

Let me try and clarify my understanding of natural selection:

 

- natural selection acts on the individual by terminating it before reproductive age ( a case of negative selection), reducing its number of offspring compared to other individuals of the same population (another case of negative selection), or by increasing the number of offspring to the next generation (a case of positive selection)

- natural selection can have a diverse effect on the population. It can drive a change in allele composition in the population, but it can also act as a stabilizing mechanism, maintaining status quo, and just weed out deleterious mutations. It is not the only mechanism that can change allele composition in the population: mutations and genetic drift are examples of other mechanisms.

- another important point with natural selection, is that no trait is independent on the environment it is located in. A given trait may be advantageous in one environment, but highly disadvantegous in another environment. Sickle cell anemia is clearly disadvantegous in countries where there is no incidence of malaria, while it will give protection against malaria in countries where the parasite is prevalent.

- natural selection will be easiest to observe when organisms undergo dramatic environmental change

 

 

I can probably write more about this later, if required.

 

I think your example of sickle cell anemia as a defense for malaria is a very good example. It would also (over time) tend toward irreversibility since the predominance of the recessive alleles could eventualy extinguish the dominant non-sickle trait. My advocacy for point #4 above, is that if the trait reverses, there was no permanent change in the gene pool, and nothing was actually "selected". Thus, I think it is useful to separate the propensity to select from the evidence of actual selection.

 

I do not think extinction of the non-sickle cell trait is very likely in this scenario, as this is an example of heterozygote advantage, where the fact that the heterozygote has the highest relative fitness insures the presence of future generations of homozygotes of both normal offspring and offspring with the sickle-cell disease. With mendelian genetics it is impossible to create a breed of pure heterozygotes. Not even genetic drift can do much about this, as long as pairs of heterozygotes continue to produce homozygotes as well as heterozygotes. What natural selection CAN do, and do, is to reduce the amount of homozygotes that are allowed to breed. It cannot however, eliminate the normal gene from the population in this case.

 

From observed adult frequency of sickle-cell anemia, and the expected hardy weinberg frequencies, the relative fitness (and from this you can calculate selection pressure) of the SS genotype is calculated to about 0.14, the relative fitness of SA genotype at 1, and the relative fitness of the AA genotype 0.88. (Based on numbers from a Nigerian population of 12387 individuals)

 

 

Your example of a potential mutative alteration that results in resistance to acetylcholinesterase inhibitors is a less compelling example. The resistance does occur, but the question is whether it is natural selection. You might note that you suggeted it was a "mutation" that substituted the glycine for the serine, and then you suggested that this will spread to other populations of insects exposed to ACHE inhibitors. If this indeed happens, it is highly unlikely that it was a mutative mechanism.

 

I did suggest that it is likely to occur again. And I might have gotten the glycine->serine wrong in my original post, as I discover with more closer reading, that it is the position 119 codon GGC (glycine) that have mutated into AGC (serine) in Anopheles gambiae.

 

They also studied Aedes aegypti and the codon at the same position, as Aedes aegypti is a species of mosquito that has never evolved this resistance. The codon coded for glycine, but had a silent mutation, so that it will require two independent point mutations before it will become serine, instead of one point mutation, as in Culex pipiens and Anopheles gambiae. This makes the mutation to serine a lot less likely to occur in Aedes aegypti, compared to the probability in A. gambiae and C. pipiens.

 

 

The probability of an identical mutative alteration in a gene occurring (exactly one specific amino acid substitution, without other mutative damage to other enzyme systems) in multiple species is vanishingly small. If this were a mutative mechanism, we would expect to see tens of thousands of different genotypic expressions for the same phenotype within a single species (since most surviving mutative geneotypic expressions would not have a phenotypic expression).

 

Actually, I think you are quite wrong here. Point mutations are quite common in insects. On average 0.14 mutations pr gamete pr generation. Considering the millions of eggs laid pr generation, and the number of generations pr year of African mosquitoes, the number of point mutations in the mosquito population totally are quite enormous. But as you say, most of the mutations are silent. Since this particular mutation is not a deleterious one, and it has the same chance to happen as a silent mutation, I really do not see the problem with it occuring in the population at the same rate as silent mutations.

 

The fact that the researchers could isolate a single variant residue, suggests that the gene in the insect population is highly homogeneous. This means that there is about a 1 in 20^300 chance for a successful mutation of this gene (assuming 300 residues in the enzyme). And this is before you add the multipliers for a mutative event not causing other lethal damage in some other system.

 

I do not agree with your calculations. We do know the genotype before the mutation, and we do know the genotype after the mutation. We do not need to add any multipliers in this case, as we do know that the particular mutation is advantegous.

 

The only thing we need to know to calculate how often this particular mutation are likely to occur, is the rate of mutations pr bp pr generation, the number of individuals pr generation, and the number of generations.

 

 

Even if we assume a large breeding population and a short birth cycle, the odds of a specific mutation occurring again in our lifetime in a separate species is pretty small. If it does occur again, it argues that this is an adaptive feature of the species, rather than a mutation. It would be a pretty complex adaptive feature (and a pretty interesting investigative topic) but still a feature, not a mutation. If it is an adaptive feature, the specific phenotypic expression might revert in the absence of ACHE inhibitors. That is, there might not be any permanent chance in the gene pool, so nothing was actually selected.

 

If it occurred in the same species I would be inclined to agree with you, problem is that this particular one-point mutation has occurred independently at least 4 times in 3 different species in 2 different subfamilies: Anopheles gambiae, Anopheles albimanus, and Culex pipiens.

 

The same gene is of course subject to a lot of other mutations that also enhance resistance in a lot of other insect species. In total, I think there are nearly 20 different mutations in this gene in various species of insects, some of which is presented in this table: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=362867&rendertype=table&id=T1

 

 

Am I making any sense here?

 

Some of it yes, but I disagree on some of the numbers, especially since mutation rates have actually been measured.

 

 

Thanks again for the thoughtful response.

 

Hmm, I wonder if this response is equally thoughtful....it is so late that my brain hurts...

 

I might go back and edit this post for clarity when i wake up. so long.

[Biochemist]

This is another great post, MS.

...That defintion is too narrow for me. Your definition, in my opinion means that we cannot identify natural selection in process, just after the fact....

I think I concede that your defintiion of natural selection is more commonly used than mine. It does leave me with an inability to differentate between adaptation and selection when we are in discussion. Could you make a suggestion on this?

I do not think extinction of the non-sickle cell trait is very likely in this scenario, as this is an example of heterozygote advantage

After additional thought, I agree.

Actually, I think you are quite wrong here. Point mutations are quite common in insects. On average 0.14 mutations pr gamete pr generation.

This is an amazing number, if I understand it. Are you saying every parent has a 0.14 chance of a mutation for a single generation of offspring? Every 7th parent will have a mutation? I would love to talk about what that number means. This might take an entire sepearate thread.

I do not agree with your calculations....

I have more to say about this, but I want to hear back from you on the question above first.

 

Another great response, MS. This is a great thread.

[MortenS]

This is another great post, MS.I think I concede that your defintiion of natural selection is more commonly used than mine. It does leave me with an inability to differentate between adaptation and selection when we are in discussion. Could you make a suggestion on this?After additional thought, I agree.This is an amazing number, if I understand it. Are you saying every parent has a 0.14 chance of a mutation for a single generation of offspring? Every 7th parent will have a mutation? I would love to talk about what that number means. This might take an entire sepearate thread. I have more to say about this, but I want to hear back from you on the question above first.

 

Another great response, MS. This is a great thread.

 

Let me try and separate adaption and natural selection, as I understand it.

 

Adaption, in my usage, is used about the actual phenotypic trait that offer an advantage to the indivual. Natural selection is the differential survival and reproduction of the individuals carrying the adaption. Natural selection is propagating the adaption over the organism when the selection pressure is there. The greater the selection pressure, the faster the adaption will spread in the population.

 

If we use the insecticide resistance example (and let us forget for the moment whether the resistance mutation arose once, or several independent times, as it is a minor problem for discussing natural selection, but clearly an interesting topic on its own), the resistance itself is the adaption, but the differential survival and reproduction of the resistance-carrying mosquito larvae and adults is the natural selection. Those that die or fail to reproduce due to the lack of the resistance gene (or allele) are negatively selected, while those that survive due to the presence of the resistance gene is positively selected.

 

When the insecticides were first applied in Africa, the eradication of mosquitoes were highly successful for a while. Today, around 80-90% of the mosquito population (Anopheles gambiae) in certain African locations displays resistance.

 

Insecticide resistance evolve amazingly fast. If you look at all the sorts of insecticides we have thrown at insects during history, there is recorded instances of evolved resistance to nearly all chemicals after just 2-20 years of exposure. It is the reason why chemical warfare against insects usually is a lost battle, even before it starts.

 

As for the number that is given: 0.14 mutations pr gamete pr generation, means just that (and mutations in this context means base substitution, insertions and deletetions, but not larger types of mutations, like duplication, reversion etc): A gamete is a sexual cell (ovum or sperm). Each ovum or sperm on average contains 0.14 mutations.

 

One female mosquito lay between 50 and 200 eggs during each oviposition, and may oviposit around 4-6 times during its life (=generation)

 

So using the mutation rate pr gamete pr generation, the total number of mutations in the offspring of one female mosquito during one generation should be between 30 and 170 mutations. I am not saying that these mutations will be expressed as phenotypic traits.

 

Now, the number of 0.14 was of interest to you and, it might be of interest to look at how it is arrived at. It is an average for insects investigated.

I can try and calculate the number for the mosquito Anopheles gambiae.

 

We need the substitution rate pr nucleotide pr generation and the haploid genome size of an organism, and multiply these. The haploid genome size of A. gambiae is 278 x 10^6 nucleotides. The substitution rate in A. gambiae I do not know, but in for insects the substitution rate seems to be between 1 x 10^(-8) and 1x10^(-9) substitutions pr nucleotide pr generation. These numbers will likely change as more species are investigated.

 

Multiplying these numbers we get a number between 0.278 and 2,78 substitutions pr gamete pr generation. Different genome sizes obviously will get us different numbers.

 

I guess the crucial number to estimate is the mutation rate pr nucleotide pr generation, and these numbers are obviously not easy to estimat, and various methods have been used. A summary of mutation rates in various viruses, bacteria and eukaryotes can be found here: http://www.genetics.org/cgi/content/full/148/4/1667

[Biochemist]

...As for the number that is given: 0.14 mutations pr gamete pr generation, means just that...So using the mutation rate pr gamete pr generation, the total number of mutations in the offspring of one female mosquito during one generation should be between 30 and 170 mutations. I am not saying that these mutations will be expressed as phenotypic traits.

Wow.

 

These little beggars are mutation machines. If the mutation rate is this high, and some fraction of genomic mutations produce transcribable proteins (even if there was no phenotypic expression) there would tend to be a lot of trash protein fabricated and degraded intracellularly. Is this true?

[MortenS]

I have been reading a bit more about the 0.14 number I mentioned. It is actually the number of mutations pr effective genome pr sexual generation. The effective genome is the part of the genome where deleterious mutations are likely to have an effect (i.e, the coding parts of the DNA). Mutations in non-coding parts of the DNA is left out in that number.

 

The comparable number in humans is 1.6, so humans actually have a higher mutation rate than Drosophila pr sexual generation. A part of this explanation, is that humans have way more cell divisions between our egg-stage and the finished ovum or sperm than Drosophila. While it is about 25 cell divisions in Drosophila, it is about 400 in humans. At each cell division, it is a chance for a DNA-copying error to occur.

 

The reason insects evolve and accumulate mutations more rapidly than humans are solely because of the difference in generation time, and the enormous difference in population size in many insect species compared to that of man. The larger the population size, the more varied mutations will surface, and with short generation times, not only will there quickly be new opportunities for mutations, but any advantageous mutations, however rare they may be, will quickly get a chance to spread through the population.

 

During one human generation time, Drosophila can undergo over 4-500 generations in the lab (a bit less in nature, due to lower temperatures). Mosquitoes in Africa can probably undergo 100-300 generations during one human generation.