Far from random, evolution follows a predictable genetic pattern, researchers find

Oct 25, 2012
Far from random, evolution follows a predictable genetic pattern, researchers find
The Princeton researchers sequenced the expression of a poison-resistant protein in insect species that feed on plants such as milkweed and dogbane that produce a class of steroid-like cardiotoxins called cardenolides as a natural defense. The insects surveyed spanned three orders: butterflies and moths (Lepidoptera); beetles and weevils (Coleoptera); and aphids, bed bugs, milkweed bugs and other sucking insects (Hemiptera). Above: Dogbane beetle (Photo courtesy of Peter Andolfatto)

Evolution, often perceived as a series of random changes, might in fact be driven by a simple and repeated genetic solution to an environmental pressure that a broad range of specieshappen to share, according to new research.

Princeton University research published in the journal Science suggests that knowledge of a species' genes—and how certain external conditions affect the proteins encoded by those genes—could be used to determine a predictable evolutionary pattern driven by outside factors. Scientists could then pinpoint how the diversity of seen in the natural world developed even in distantly related animals.

"Is evolution predictable? To a surprising extent the answer is yes," said senior researcher Peter Andolfatto, an assistant professor in Princeton's Department of Ecology and and the Lewis-Sigler Institute for Integrative Genomics. He worked with lead author and postdoctoral research associate Ying Zhen, and graduate students Matthew Aardema and Molly Schumer, all from Princeton's ecology and evolutionary biology department, as well as Edgar Medina, a biological sciences graduate student at the University of the Andes in Colombia.

The researchers carried out a survey of from 29 distantly related , the largest sample of organisms yet examined for a single evolutionary trait. Fourteen of these species have evolved a nearly identical characteristic due to one external influence—they feed on plants that produce cardenolides, a class of steroid-like cardiotoxins that are a natural defense for plants such as and dogbane.

Though separated by 300 million years of evolution, these diverse insects—which include beetles, butterflies and aphids—experienced changes to a key protein called sodium-potassium adenosine triphosphatase, or the sodium-potassium pump, which regulates a cell's crucial sodium-to-potassium ratio. The protein in these insects eventually evolved a resistance to cardenolides, which usually cripple the protein's ability to "pump" potassium into cells and excess sodium out.

Andolfatto and his co-authors first sequenced and assembled all the expressed genes in the studied species. They used these sequences to predict how the sodium-potassium pumpwould be encoded in each of the species' genes based on cardenolide exposure.

Scientists using similar techniques could trace protein changes in a species' DNA to understand how many diverse organisms evolved as a result of environmental factors, Andolfatto said. "To apply this approach more generally a scientist would have to know something about the genetic underpinnings of a trait and investigate how that trait evolves in large groups of species facing a common evolutionary problem," Andolfatto said.

"For instance, the sodium-potassium pump also is a candidate gene location related to salinitytolerance," he said. "Looking at changes to this protein in the right organisms could reveal how organisms have or may respond to the increasing salinization of oceans and freshwater habitats."

Milkweed tussock moth (Credit: Peter Andolfatto)

Jianzhi Zhang, a University of Michigan professor of ecology and evolutionary biology, said that the Princeton-based study shows that certain traits have a limited number of molecular mechanisms, and that numerous, distinct species can share the fewmechanisms there are. As a result, it is likely that a cross-section of certain organisms can provide insight into the development of other creatures, he said.

"The finding of parallel evolution in not two, but numerous herbivorous insects increases the significance of the study because such frequent parallelism is extremely unlikely to have happened simply by chance," said Zhang, who is familiar with the study but had no role in it.

"It shows that a common molecular mechanism is used by many different insects to defend themselves against the toxins in their food, suggesting that perhaps the number of potential mechanisms for achieving this goal is very limited," he said. "That many different insects independently evolved the same molecular tricks to defend themselves against the same toxin suggests that studying a small number of well-chosen model organisms can teach us a lot about other species. Yes, evolution is predictable to a certain degree."

Andolfatto and his co-authors examined the sodium-potassium pump protein because of its well-known sensitivity to cardenolides. In order to function properly in a wide variety of physiological contexts, cells must be able to control levels of potassium and sodium. Situated on the cell membrane, the protein generates a desired potassium and sodium ratio by "pumping" three sodium atoms out of the cell for every two potassium atoms it brings in.

Cardenolides disrupt the exchange of potassium and sodium, essentially shutting down the protein, Andolfatto said. The human genome contains four copies of the pump protein, and it is a candidate gene for a number of human genetic disorders, including salt-sensitive hypertension and migraines. In addition, humans have long used low doses of cardenolides medicinally for purposes such as controlling heart arrhythmia and congestive heart failure.

The Princeton researchers used the DNA microarray facility in the University's Lewis-Sigler Institute for Integrative Genomics to sequence the expression of the sodium-potassium pump protein in insect species spanning three orders: butterflies and moths (Lepidoptera); and weevils (Coleoptera); and aphids, bed bugs, milkweed bugs and other sucking insects (Hemiptera).

The researchers found that the genes of cardenolide-resistant insects incorporated various mutations that allowed it to resist the toxin. During the evolutionary timeframe examined, the sodium-potassium pump of insects feeding on dogbane and milkweed underwent 33 mutations at sites known to affect sensitivity to cardenolides. These mutations often involved similar or identical amino-acid changes that reduced susceptibility to the toxin. On the other hand, the sodium-potassium pump mutated just once in insects that do not feed on these plants.

Significantly, the researchers found that multiple gene duplications occurred in the ancestors of several of the resistant species. These insects essentially wound up with one conventional sodium-potassium pump protein and one "experimental" version, Andolfatto said. In these insects, the newer, hardier versions of the sodium-potassium pump are mostly expressed in gut tissue where they are likely needed most.

"These gene duplications are an elegant solution to the problem of adapting to environmental changes," Andolfatto said. "In species with these duplicates, the organism is free to experiment with one copy while keeping the other constant, avoiding the risk that the new version of the protein will not perform its primary job as well."

The researchers' findings unify the generally separate ideas of what predominately drives genetic evolution: Protein evolution, the evolution of the elements that control protein expression or gene duplication. This study shows that all three mechanisms can be used to solve the same evolutionary problem, Andolfatto said.

Central to the work is the breadth of species the researchers were able to examine using modern gene sequencing equipment, Andolfatto said.

"Historically, studying genetic evolution at this level has been conducted on just a handful of 'model' organisms such as fruit flies," Andolfatto said. "Modern sequencing methods allowed us to approach evolutionary questions in a different way and come up with more comprehensive answers than had we examined one trait in any one organism.

"The power of what we've done is to survey diverse organisms facing a similar problem and findstriking evidence for a limited number of possible solutions," he said. "The fact that many of these solutions are used over and over again by completely unrelated species suggests that the evolutionary path is repeatable and predictable."

Explore further: Rock-paper-scissors model helps researchers demonstrate benefits of high mutation rates

add to favorites email to friend print save as pdf

Related Stories

Evolution predictable for insects eating toxic plants

Jul 24, 2012

(Phys.org) -- The famous biologist Stephen J. Gould once asked: If we rerun the tape of life, would the outcome of evolution be the same? For years, scientists have questioned whether evolution is predictable, ...

Study: Junk DNA is critically important

Oct 19, 2005

A University of California-San Diego scientist says genetic material derisively called "junk" DNA is important to an organism's evolutionary survival.

Sowing seed on salty ground

Jun 06, 2007

Scientists have discovered a gene that allows plants to grow better in low nutrient conditions and even enhance their growth through sodium uptake, according to a report published online this week in The EMBO Journal.

Recommended for you

Does germ plasm accelerate evolution?

Apr 14, 2014

Scientists at The University of Nottingham have published research in the leading academic journal Science that challenges a long held belief about the way certain species of vertebrates evolved.

User comments : 11

Adjust slider to filter visible comments by rank

Display comments: newest first

rod_russell_9
1.9 / 5 (17) Oct 25, 2012
Isn't evolution amazing? So organized and so structured! Who would have imagined that chance could be so smart?
MrVibrating
3.5 / 5 (8) Oct 25, 2012
Yes, given that there's a finite number of cardenolide-resistance mutations possible, and that individuals not expressing them fare less well than those that do, this kind of inevitable convergent evolution does seem to suggest that Jesus loves us. Or maybe it was the box of chocolate liqueurs i just ate? Either way, i'm left with a nice warm feeling..
Lurker2358
1.7 / 5 (6) Oct 25, 2012
It seems reasonable enough that there are limited number of possibilities at such small scales of structures.

You may be able to make a "larger" structure which may do the same task as well or better, but then you introduce the problem of greater vulnerability to pathogens or other problems.

Therefore, smaller, simpler structures are safer, and then the small, simple structure that works best would be used by the engineer.

By the way, "Guess and Check" is an intelligent problem solving strategy, albeit not necessarily the most efficient. It is nevertheless useful in some situations that are not easy to put into a simple mathematical equation or shortcut for solving said equation.

Some proposed Quantum Algorithms are actually "Guess and Check". You put in a presumed solution, and the Algorithm spits it back out, and you can tell if it's right or wrong, which might be useful as a search or sort tool in combination with an array of possibilities.
Deathclock
4.5 / 5 (8) Oct 25, 2012
Isn't evolution amazing? So organized and so structured! Who would have imagined that chance could be so smart?


Who would have imagined that a person could be so stupid...
Lurker2358
1.7 / 5 (6) Oct 25, 2012
Once you have the best solution, or at least the solution that works well enough in a real, non-idealistic world, then why would you change it?

Modularity is a sign of intelligence, because the same code or component is being re-used in many different structures or even in entirely different creatures.

What one perceives as "randomized" I perceive as an ordered creation, or perhaps an ordered sort.

If you have two lawn sprinklers, A and B:

A is rated 100% as a sprinkler, but has 5 moving parts.

B is rated 90% as a sprinkler, but has no moving parts.

Even though A is a better sprinkler in the short term, it's too complicated and breaks down a lot, so B ends up being cheaper to manufacture, and lasts longer. If we lived in an rational economy, B would be the best product to manufacture.

Unfortunately for my example, we live in capitalism and it's to the company's benefit to manufacture products that break down so you need to buy them again at mark-up price.

Imperfect example.
Lurker2358
2 / 5 (4) Oct 25, 2012
This line of reasoning can actually explain why there are no "Perfect" life forms.

Gene B may be "Better" than Gene A, but let's say "B" has N extra bases making N/3 extra codons.

Well, B is more complicated to make than A and it is also slightly more vulnerable to viruses, transcription errors, cosmic rays, etc.

And so, while an idealist may say "Gene B is better, let's put that in some GM food animals or food plants!" The real world situation is that Gene A is actually better in the real world, even though it's not the best conceivable gene at it's primary function.

The result here is that while Gene B might prevail in the short term, it is more vulnerable than Gene A, meaning it is more likely to be destroyed or further mutated in the long term anyway.

So life does not favor the genes that an idealist might call "perfect" in any one function or degree of study, but rather the genese that are "good enough" in all respects, not just primary function.
Deathclock
4 / 5 (4) Oct 25, 2012
Some proposed Quantum Algorithms are actually "Guess and Check". You put in a presumed solution, and the Algorithm spits it back out, and you can tell if it's right or wrong, which might be useful as a search or sort tool in combination with an array of possibilities.


Simulated annealing...
Lurker2358
1 / 5 (5) Oct 25, 2012
So even if there was such a thing as a "best possible species" it would, in many cases, be filled with genes which appear "less than perfect" though some idealistic viewpoint.

This also explains why so-called "vestigial organs" are not detrimental in many or even most cases, as the genes involved have more than one function and it simply cannot be helped that they are still needed for functions other than their apparent morphological redundancies or whatever else a complainer might find, i.e. the appendix or male nipples, etc. So it turns out that while an agnostic idealist might complain about male nipples, in fact the male nipples are a good thing, because the genes involved are passed on to the next generation female, and of course every child and ever male is thankful for that indeed!
VendicarD
3.7 / 5 (3) Oct 25, 2012
"Who would have imagined that chance could be so smart?" - Foofie

Only the God that created random chance.

ormondotvos
4.7 / 5 (3) Oct 26, 2012
Stupid headline, designed to attract clicks. "Evolution" is a descriptive term for originally random mutations. Once established, in a particular niche (however broadly defined) parallel adaptations occur. Big whoop.

Still an interesting article, but no great insight, just an interesting discovery.
thingumbobesquire
1 / 5 (4) Oct 26, 2012
Probability theory is a ruse of oligarchic control. Those that buy into it deserve a foolscap along with their degrees. Randomness is in the mind of the beholder. http://thingumbob...pot.com/

More news stories

Ranchers benefit from long-term grazing data

Scientists studying changes in the Earth's surface rely on 40 years of Landsat satellite imaging, but South Dakota ranchers making decisions about grazing their livestock can benefit from 70 years of data ...

Melting during cooling period

(Phys.org) —A University of Maine research team says stratification of the North Atlantic Ocean contributed to summer warming and glacial melting in Scotland during the period recognized for abrupt cooling ...