|
Author
|
Topic: Evolution caught in the act?
|
Michael J. Behe
Member
Member # 44
|
posted 23. February 2004 20:13
In a recent news story, called “Evolution caught in the act”, featured on a website called “ www.innovations-report.com ”, the work of some University of Michigan and University of Texas researchers is highlighted. The story excitedly remarks that the researchers:
“... used a strain of mutant bacteria ... that had lost their ability to make disulfide bonds. These disulfides are critical for the ability of the bacteria’s propeller-like swimming motor, the flagellum, to work. The researchers then put these non-swimming bugs to the test by placing them on a dish of food where, once they had exhausted the food they could reach, they either had to repair that broken motor or starve to death on the spot.
...
“To the researchers’ surprise, a mutant carrying only two amino acid changes, amounting to less than 2 percent of the total number of amino acids in thioredoxin, restored the ability of the bacteria to move. The altered thioredoxin was able to carry out disulfide bond formation in numerous other bacterial proteins all by itself, without relying on any of the components of the natural disulfide bond pathway. The mutant bug managed to solve the problem in time and swim away from starvation and multiply.”
The breathless writing (typical for press releases) makes it sound as if something akin to Darwinian evolution was rebuilding a shattered bacterial flagellum. But if you go to the trouble to look up the paper (Masip et al, “An *engineered* pathway for the formation of protein disulfide bonds,”Science 303, 1185 (2004)) you find that, as per usual, the “random evolution” part is all hype.
The authors deliberately deleted a cellular system that normally adds a disulfide bond (don't worry about exactly what that is) to a certain protein that is part of the bacterial flagellum. The bacteria then were unable to swim, since they could not make normal flagella. Our intelligent engineers then went to work to find another protein that would do the job of the disulfide-making system they eliminated. They decided that a good candidate would be a protein called thioredoxin, because they knew it had a structure similar to one of the components of the trashed system, and has the basic chemical material needed for the job. However, thioredoxin normally stays inside the cell, whereas the old machinery operates outside the cell membrane. To overcome this problem the workers deliberately added a signal from another protein that is normally targeted to outside the cell, for the express purpose of getting thioredoxin into the right place. Well, nonetheless the exported thioredoxin didn't work. The workers then scratched their heads and thought, "Well, the critical part of thioredoxin is a four amino acid sequence containing the business end of the protein, which actually does the chemistry to make the disulfide bond. Let's use laboratory molecular biological techniques to make a whole bunch of mutants *in just those four residues* and see if any work." They did just that, and then fished out a couple that work, which were able to perform the step to make the flagellar protein and allow the bacteria to swim again.
The mutations were certainly not random. Quite the opposite. The workers added mutations they needed for a proper background in which to look for what they wanted, and then focused on exactly the part of the protein they wanted to substitute which they thought would help it do the job. If this is Darwinian evolution, then there's not a whole lot of difference between Darwinism and deliberate intelligent design.
Someone might ask, "How did the bacteria know which tools to mutate?" The answer is that the bacteria didn't know at all. The researchers knew. The researchers chose a *very* specific region of a *very* specific protein to zero in on.
As a project in deliberate protein engineering, the work is interesting. As an analogy to Darwinian evolution, the work makes a good case for intelligent design.
[ 23. February 2004, 20:15: Message edited by: Michael J. Behe ]
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 23. February 2004 23:56
Hi Mike good to see you contribute to the ISCID forum. You raise some interesting issues and I agree that press releases seldomly represent arguments to their fullest scientific details. Which is why publications in scientific journals need to be looked at. I have gotten a copy of the paper which is quite interesting
quote:
We have engineered a pathway for the formation of disulfide bonds. By imposing evolutionary pressure, we isolated mutations that changed thioredoxin, which is a monomeric disulfide reductase, into a [2Fe-2S] bridged dimer capable
of catalyzing O2-dependent sulfhydryl oxidation in vitro. Expression of the mutant protein in Escherichia coli with oxidizing cytoplasm and secretion via the Tat pathway restored disulfide bond formation in strains that lacked the complete periplasmic oxidative machinery (DsbA and DsbB). The evolution of [2Fe-2S] thioredoxin illustrates how mutations within an existing scaffold can add a cofactor and markedly change protein function.
What I think is interesting is how they showed how a relatively simple mutation within an existing scaffoled added a cofactor and changed protein function.
The real relevance of these experiments are that they:
".. have, in essence, caught evolution in the act of adding an iron sulfur cluster to thioredoxin, enabling it to act as a thiol oxidant. Our results show that the acquisition of cofactors is evolutionarily a rather simple process, and our ability to generate an artificial pathway for oxidative protein folding highlights the plasticity of redox metabolism."
While undoubtably some intelligent design was needed to conduct these experiments, it is interesting to focus on their evolutionary relevance.
Given evolutionary pressure, thioredoxin mutants have been isolated which restored the motility in these cells.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 24. February 2004 07:36
This experiment appears to illustrate at least three key elements of evolutionary change processes not explained or accounted for by traditional Darwinian theory. First, as Mike points out, the experiment shows that evolutionary change can be produced by intelligent design or engineering processes. Second it demonstrates that external design processes can impact, influence, or control evolutionary change. Finally, it demonstrates a type of evolutionary change that could not have occurred by RM&NS type processes within the time constraints necessary for survival. It might also be noted that this points out the importance of the distinction between experimental results and the interpretation of experimental results.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 24. February 2004 08:21
I have two quick comments I would like to make this morning (time is short):
First, Dr. Behe wrote
quote:
The mutations were certainly not random. Quite the opposite. The workers added mutations they needed for a proper background in which to look for what they wanted, and then focused on exactly the part of the protein they wanted to substitute which they thought would help it do the job. If this is Darwinian evolution, then there's not a whole lot of difference between Darwinism and deliberate intelligent design.
Someone might ask, "How did the bacteria know which tools to mutate?" The answer is that the bacteria didn't know at all. The researchers knew. The researchers chose a *very* specific region of a *very* specific protein to zero in on.
As a project in deliberate protein engineering, the work is interesting. As an analogy to Darwinian evolution, the work makes a good case for intelligent design.
I have heard this argument before - it goes something like this: any lab experiment (or computer situation) can tell us nothing about what natural processes can do because the role of an intelligent agent (a human being) in setting up the experiment adds an element that necessarily turns the situation into an example of intelligent design.
But this can’t be right. One of the things an experimenter must do is simplify the situation and reduce the number of variables present so as to focus on the single variable or small number of variables under consideration. Experiments are “intelligently designed” to help us find out how the natural world works - I can’t see how it could be otherwise. To rule out any results that follow from such as experiment would seem to negate the possibility of doing any research on how natural processes accomplish what they do.
The question I would ask Dr. Behe is this: if experiments such as this are insufficient to investigate the potential and nature of the development of “Darwinian pathways” because of the involvement of an intelligent human being in setting them up, how would you propose that we do study the issue?
Secondly, assuming as I do that this objection about the intelligent design of the experiment invalidated any conclusions about evolution, I have this to say:
It appears to me that once the situation was set up, the existing function of the gene's product got re-worked into something entirely new in the world, and useful to the species. Nobody designed the new form of thioredoxin -- nobody other than the selective pressure. A new and useful biochemical mechanism, consequent to nothing more than mutation and selection imposed by the environment one, arose spontaneously (even though the environment for this was due to an experiment.)
It seems to me that this creation of biological novelty (and hence "new information") is precisely what many ID theorists say can’t be accomplished by natural processes. Sure, it’s very small - but what else would you expect from a short experiment in a laboratory?
The challenge for the ID theorist is to compute some probabilities here. If the probability of this small change is high (which could be investigated both by repeating the experiment and by perhaps catching sub-steps in the development of the new pathway), then we would be moving, perhaps, towards a method for eventually distinguishing design from non-designed things. With that in mind, I would think (or at least hope) that Dr. Behe would take more interest in this type of experiment and its relationship to the possibility (or impossibility) of irreducibly complex things arising via natural processes rather than dismissing the experiment as fatally flawed precisely because it was an experiment.
[ 24. February 2004, 09:25: Message edited by: Evan ]
IP: Logged
|
|
Paul A. Nelson
Member
Member # 26
|
posted 24. February 2004 09:52
Evan wrote:
quote:
I would think (or at least hope) that Dr. Behe would take more interest in this type of experiment and its relationship to the possibility (or impossibility) of irreducibly complex things arising via natural processes rather than dismissing the experiment as fatally flawed precisely because it was an experiment.
Is that really what Mike is saying above -- i.e., "I don't buy it because it's an experiment"?
No. Mike writes:
quote:
They decided that a good candidate would be a protein called thioredoxin, because they knew it had a structure similar to one of the components of the trashed system, and has the basic chemical material needed for the job. However, thioredoxin normally stays inside the cell, whereas the old machinery operates outside the cell membrane. To overcome this problem the workers deliberately added a signal from another protein that is normally targeted to outside the cell, for the express purpose of getting thioredoxin into the right place. Well, nonetheless the exported thioredoxin didn't work. The workers then scratched their heads and thought, "Well, the critical part of thioredoxin is a four amino acid sequence containing the business end of the protein, which actually does the chemistry to make the disulfide bond. Let's use laboratory molecular biological techniques to make a whole bunch of mutants *in just those four residues* and see if any work." They did just that, and then fished out a couple that work, which were able to perform the step to make the flagellar protein and allow the bacteria to swim again.
Mike is saying that this is a biologically, or evolutionarily, unrealistic experiment -- a line of critique common in the origin-of-life literature (see, e.g., many of Leslie Orgel's papers).
[ 24. February 2004, 10:01: Message edited by: Paul A. Nelson ]
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 24. February 2004 10:09
Thank you for bringing up this more substantive issue.
However, it should be clear from my post that I didn’t mean to say that this all Mike is saying is that "I don't buy it because it's an experiment." However, this is one of the things he did say, and I think that issue bears discussion.
The other issue, which is related, is to what extent the experiment is biologically relevant. Yes, the experimenters zeroed in on a particular site (thus “intelligently designing” a part of the solution), but also the bacteria reached a solution that was not pre-planned, was unexpected, and taught the researchers something about a possible biological structure. Thinking about what this means, how to fit it into design theory vis a vis probabilities, and thinking about how to refine this type of experiment to learn more about what is biologically relevant outside of the laboratory ought to be the goal of our analysis of this paper.
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 24. February 2004 10:34
Posts that do not contribute to the discussion but merely advocate or act as a "cheerleading" service, will be deleted.
[ 24. February 2004, 10:41: Message edited by: Moderator ]
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 24. February 2004 11:49
I see the main importance of this paper in that it showed how just _two_ mutations can generate a novel solution to an existing problem. This indicates to me that evolutionary processes may be able to generate novelty.
It may be interesting from an ID perspective to evaluate the amount of information and complexity generated by just two mutations.
Warren seems to suggest that these findings 'not explained or accounted for' by Darwinian theory. I would surely be interested in Warren showing support for his thesis but I do not want to distract from this exciting paper/experiment and its relevance to ID and evolution.
As far as Evan's comments are concerned, he raises an interesting point, if experimental results can be used as evidence of 'intelligent design' then how can experiments be conducted in a manner which would not 'invalidate' them? The basic components of this experiment are random variation and strong selection. What more would be needed?
There seems to be an exciting challenge here from an information theoretical perspective worth pursuing.
These findings may tie together the claims made in EMBO Journal (2001) 20, 1555–1562, "Turning a disulfide isomerase into an oxidase: DsbC mutants that imitate DsbA " by Martin Bader et al.
quote:
It has been reported that thioredoxin or protein disulfide isomerase (PDI), when exported to the periplasm, is able to rescue a dsbA null phenotype (Jonda et al., 1999; Debarbieux and Beckwith, 2000). The observed rescue depends on the presence of the dsbB gene, suggesting that thioredoxin and PDI are re-oxidized by DsbB. According to our results, DsbB also re-oxidizes the thioredoxin fold of monomeric DsbC. Thus, DsbB seems to be rather promiscuous in recognizing the thioredoxin folds of DsbA, monomeric DsbC, thioredoxin and PDI. However, this promiscuity does not interfere with disulfide bond isomerization in wild-type cells because the thioredoxin fold of wild-type DsbC is not oxidized by DsbB. It is very likely that DsbC evolved from thioredoxin and that evolutionary pressure forced DsbC to find a mechanism whereby its active site is protected from DsbB. Apparently, this was achieved by dimerization of the thioredoxin domains of DsbC, thus allowing DsbC to function as an isomerase despite the presence of DsbB.
In the same paper the authors comment
quote:
Overproduction of wild-type DsbC, in strains grown in minimal media, resulted in no measurable motility. To investigate the level of DsbC overexpression from our promoter constructs, we prepared periplasmic extracts from the mutants and compared them to the starting strain. The level of expression of DsbC was unchanged in our mutants, showing that their ability to rescue dsbA null mutants was not due to enhanced DsbC expression (data not shown). Interestingly, SDS–PAGE revealed a 45 kDa protein that was present in all motile mutants and absent in both disulfide-negative strains (JCB817, JCB817 pMB69). The first 10 N-terminal amino acid residues of this band were sequenced and match exactly with the first 10 residues of E.coli alkaline phosphatase (RTPEMPVLEN). Alkaline phosphatase contains two disulfide bonds necessary for correct folding. If the disulfide bonds are not formed, the protein does not fold correctly and is degraded (Bardwell et al., 1991). This result provides additional evidence that the DsbC variants G49R and G49E are efficiently forming disulfide bonds in the absence of DsbA, the native donor of disulfide bonds in the periplasm.
The authors predict "We reasoned that a rational design of mutants in the dimerization interface should reveal more DsbC variants that are able to rescue DsbA"
So in other words, these experiments seem to be supported in various ways. In this case the authors used "a plasmid library of random mutations in the dsbC gene (in which the expression of DsbC was under arabinose control) was created and transformed into a dsbA null strain. After incubation, 28 motile colonies were observed in which motility was dependent on the presence of arabinose"
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 24. February 2004 14:13
I would agree that the question of biological or evolutionary realism is important in evaluating the significance of this experiment. I am not, however, sure that the answer to the question is entirely clear. It seems reasonably clear that the experimenters used relatively complex manipulation or engineering aimed at achieving a specific goal. We have reasonable evidence that natural evolutionary processes have changes far more complex than the changes achieved in the experiment. At least IMO, it is reasonable to argue that this experiment suggests that bacteria may have the capacity to perform complex engineering or intelligent design.
If you argue that the experiment represents an unrealistic evolutionary process, then you are left without a realistic option for explaining evolutionary changes which clearly appear to have occurred. If you argue that the experiment may represent a realistic evolutionary process, then you are left with the technical question of ‘How are bacteria capable of performing complex engineering or intelligent design processes?’.
IP: Logged
|
|
rafe gutman
Member
Member # 134
|
posted 24. February 2004 23:13
dr. behe, you wrote:
quote:
As a project in deliberate protein engineering, the work is interesting. As an analogy to Darwinian evolution, the work makes a good case for intelligent design.
when i first read this statement, my initial impression was that you were just trying to spin the results of the experiment. however, since this is your first ISCID post, i really should give you the benefit of the doubt. would you care to elaborate on this point? how, exactly, does this experiment make a good case for intelligent design?
IP: Logged
|
|
Neal
Member
Member # 1137
|
posted 26. February 2004 14:44
Well I am not a molecular biologist so I have questions rather than answers. Regarding the experiment, the way I understand it is that there is this thing called a flagella used to move a bacteria around. The flagella is comprised of several? dozen? several dozen? proteins (parts/functions) all of which or most of which are essential for the flagella to even exist and beyond that to perform its function of moving a bacterium. Now a protein ("system") I guess that normally resides in or around the bacteria cell and is essential for the process of producing one of the proteins in the flagella is eliminated or compromised such that it can no longer perform at least one vital task which is creating the disulfide bonds that presumably hold this one protein together? or allow this protein to bond with other proteins in the flagella? Experimenters, then used a protein similar to the protein system that creates these disulfide bonds and that is? or is not? normally part of this bacteria. This other protein is the thioredoxin. The thioredoxin is then intentionally modified in four key places such that it is then identical to or at least similar to the missing native protein system that normally performs the task of creating these disulfide bonds in one of the proteins of the flagella. The intentionally altered thioredoxin proteins are then introduced at just the correct place? and at just the correct time? in order to rebuild (fix)? or manufactured anew? a fully functioning flagella. What are the temporal and spacial constraints in this experiment beyond the protein tools required to make this experiment work?
Please help me understand this better.
Thanks
IP: Logged
|
|
Argon
Member
Member # 276
|
posted 27. February 2004 09:45
Making mountains out of molehills. That is what happens with poorly written press releases (as well as with critiques and discussions of poorly written press releases).
The title of the paper is: "An engineered pathway for the formation of protein disulfide bonds". The quick description of the work in Science magazine also says that the authors "engineered a new pathway".
Behe writes: "The researchers chose a *very* specific region of a *very* specific protein to zero in on."
That is true. As they hoped, they got a protein which generated disulphide bonds. What they also got, quite unexpectedly and unintentionally, was a 2-iron, 2-sulfur, bridged dimer from two amino acid subsitutions. Interestingly, the 2Fe-2S form was selected from a pool of mutagenized genes on the basis of selectable, biological activity. If we read the last paragraph of the paper, we specifically see that it was the unexpected acquistion of the cofactor (To quote: "adding an iron sulfur cluster to thioredoxin, enabling it to act as a thiol oxidant", pg 1189), that the authors mention as catching "evolution in the act": Not the addition of leader sequences nor the choice of sites to mutate.
Considering that this discussion is of a press release and not an actual paper per se, I am not sure that this discussion really belongs in the "Literature review" section.
[ 27. February 2004, 09:50: Message edited by: Argon ]
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 27. February 2004 11:53
Neil writes,
quote:
The thioredoxin is then intentionally modified in four key places such that it is then identical to or at least similar to the missing native protein system that normally performs the task of creating these disulfide bonds in one of the proteins of the flagella. The intentionally altered thioredoxin proteins are then introduced at just the correct place? and at just the correct time? in order to rebuild (fix)? or manufactured anew? a fully functioning flagella.
This isn't what happened, as argon explained.
I also agree with argon that this doesn't belong in the literature forum because. among other things, we don't have the actual paper. Is this online someplace? Could the full paper be posted for discussion?
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 27. February 2004 12:40
In the opening post, Dr. Behe wrote,
quote:
"Well, the critical part of thioredoxin is a four amino acid sequence containing the business end of the protein, which actually does the chemistry to make the disulfide bond. Let's use laboratory molecular biological techniques to make a whole bunch of mutants *in just those four residues* and see if any work." They did just that, and then fished out a couple that work, which were able to perform the step to make the flagellar protein and allow the bacteria to swim again.
It seems to me that maybe Dr. Behe didn't quite see the whole picture of what happened. After they "made the mutants" and "fished out a couple that worked," they discovered that the ones that worked had evolved, through further mutation, a novel solution to the problem - a solution that did not restore the pathway that had been inactivated, but rather was a different way of getting the proper substances so the bacteria could move again.
It was not the original setup of the experiment that was important, it was the result. Dr. Behe doesn't seem to address that.
IP: Logged
|
|
Argon
Member
Member # 276
|
posted 27. February 2004 13:48
Rafe Gutman provided a link to the paper at the ARN board. Here is the link. I do not know if everyone will have access to the online version and so here is the print reference: L. Masip, JL Pan, S Haldar, JE Penner-Hahn, MP DeLisa, G Georgiou, JCA Bardwell, J-F Collet, "An Engineered Pathway for the Formation of Protein Disulphide Bonds" Science, 303(5661), pp 1185-1189 20-Feb-2004. Science magazine is available in many public libraries and often at bookstores like Borders and Barnes & Noble.
This topic is being discussed in a little more detail on the ARN board in this thread.
[ 27. February 2004, 16:17: Message edited by: Argon ]
IP: Logged
|
|
|
Printer-friendly view of this topic