Friday, August 25, 2006

System 2

Just a few thoughts after having read Matt Inlay’s response to Behe’s immunology chapter section 2.

The mechanism explaining the origin of a rearranging gene segment system to generate antibody diversity.

(CAUTION – I am a self confessed delusional crank with no immunology or evolutionary biology qualifications.)

As I understand it this is a summary of the best guess currently available for the origin of the gene rearranging system to generate variable antibodies.

1. A transposon became inserted either into a sperm cell, and egg cell or a very early embryo cell. Presumably at this stage it is a transposon that only becomes active in transcribed genes rather than having eukaryotic transcription control sequences. It must have initially integrated into an active gene in such a way that it could transpose when that gene was expressed. This is something that is presumably a selective disadvantage.

2. After jumping around the genome for x generations it happened to insert into an innate immune system receptor at the point where its eventual excision would cause variation in the antigen binding site and thus be an advantage to the cell. This is also presumably an exceedingly rare event.


At this point in the description the proposed pathway becomes very sketchy. By means of 4(or 5) genetic duplication events plus an incorrect recombination event we produce a set of variable antibody producing genes with heavy and light chains which fit together beautifully and a set of T-cell receptors using a similar process for a totally different function.

If this is the best that is currently available then I would agree with Behe that it is time someone got their act together or was ready to admit publically that at the moment we really have got lots of major problems explaining the origin of variable antibodies.

(Please note I am not intending to belittle Matt Inlay’s essay. I think he has done an excellent job explaining this to a non-specialist for the web. Thank you Matt!)

(Please note I am not intending to belittle the work of evolutionary immunologists comparing immune systems in different organisms and trying to put together a scheme for its evolution. I am very appreciative indeed of research immunologists. I am merely saying that this account is very sletchy indeed and for it to be convincing you need to already believe that it really did evolve without any intelligent input.)

(BTW are there different RSS sequences at the two join sites and different RAG proteins…what stops them getting muddled?)

22 Comments:

Anonymous Matt Inlay said...

Andrew,

For every sentence in the model you've written, I can point to a dozen research articles that back that sentence up. Each of the steps you described are totally consistent with the evidence we have, and the pathway does not require any miraculous events, which is exactly what Behe claimed was required for the evolution of any IC system. Take your first step for instance. We already know that transposons jump in and out of the genome constantly. Our genomes have more space taken up by transposable elements than they do actual genes! It may be selectively disadvantageous to us (though more often than not neutral), but so are viral infections.

Looking at your second step, that of a transposon inserting into an innate antigen receptor, it's as unlikely as you winning the lottery. And yet every week, someone wins the lottery. If you have a population of millions of jawless vertebrates, and one of them has the correct insertion, then that individual will have a selectable advantage. Again, this is an entirely plausible pathway for the evolution of the system.

Now that you're familiarizing yourself with the evidence, let me ask you a few questions:

a. Why do you think the recombination activating genes (RAG-1 and RAG-2) are so similar to transposases? (so similar that they actuallly contain transposase activity)

b. Why do you think the RAG genes are located right next to each other in the genome of all organisms that possess RAGs, and neither possess introns? (features highly uncommon in eukaryotic genes)

c. Why do the RSSs flanking each V, D, and J segment resemble the ends of transposons?

d. Can you propose an alternative model for the origin of the adaptive immune system that incorporates this evidence? If you appeal to an intelligent designer, how do you specifically test that notion?

8:47 pm  
Blogger Ian Musgrave said...

Matt said:
"We already know that transposons jump in and out of the genome constantly."

An there are a group of SP1 transposons in E. coli which jumps into and out of the flagelin gene, rearranging the flagellin gene so that it escapes the surveillance of the vertebrate immune system.

Matt wrote:
Why do you think the recombination activating genes (RAG-1 and RAG-2) are so similar to transposases?

Heck, for all intents and purposes they are HAT transposases. Finding that out solved the mechanisms of rearrangement of the variable regions of these genes

9:07 am  
Blogger Ian Musgrave said...

Andrew wrote: By means of 4(or 5) genetic duplication events plus an incorrect recombination event we produce a set of variable antibody producing genes with heavy and light chains which fit together beautifully and a set of T-cell receptors using a similar process for a totally different function.

Remember that the Hat trasnposon integrated into a APAR like receptor, which already has a heavy-light structure, and is used for innate immunity. The function is very similar (binding to surface epitopes of invading bacteria/viruses), but now instead of a single epitope, you are able to bind to LOTS of epitopes.

And Andrew, don’t think of yourself as a delusional crank, think of yourself as an astrologer (one who now has to cope with Pluto not being a planet).

(off to a party now. Yea!)

9:20 am  
Blogger Ian Musgrave said...

Party was great, but then I got up early to see the Saturn-Venus conjunction, and got pipped by clouds.

Anyway, thinking more about it, Astrology is an appropriate comparison with ID supporters. Today, something like ¾ of all UK citizens will read a horoscope (the figures are probably similar for Australian and US citizens). These are ordinary folks, not the sad anoraks who hang out on hillsides waiting for aliens to make crop circles. You wouldn’t call the horoscope readers delusional, but we know astrology doesn’t work. So why does belief in astrology persist? There are many reasons. Firstly, the idea that there is some force that guides our destiny is quite attractive to most people, no matter how fuzzily it is defined. Like astrology (some force connects us to the stars, or maybe its just psychology or something), ID is so fuzzily defined (something, somewhere, was designed, somehow) that the designer can be Greys or Vishnu, and this keeps almost everyone happy. Astrology has never come up with positive evidence for anything, (which planet or asteroid was discovered by astrologers), they are purely reactive to the discoveries of astronomers, with astounding post hock rationalizations. ID has never uncovered anything about biology, with Dembski snorting that ID doesn’t have to produce “that pathetic level of detail”, that they demand of biologists. Astrologers use the maths developed by astronomers to locate planets precisely, and then add layers of mumbo jumbo to it. IDers abscond with mathematics, and bury it in meaningless mumbo jumbo as well. Finally, the evidence against astrology is substantial, but it tends to be technical, and buried in technical journals. Whereas astrological predictions are Rorschach blots that people can predict their wishes and desires onto, and this make them more liable to “go with their gut” than try and wade through technical literature. Similarly, the appeal of there being a designer of some sort appeals to people very strongly.

So, you don’t have to be delusional to be seduced by ID, but you do have to be aware, that like astrology, ID’s appeal is non-rational.

12:30 am  
Anonymous Hrafn said...

This article in Dispatches from the Culture Wars sheds some light on the precedents applying to expert testimony such as Behe's:

Malone's advice includes the following:
"Beware the expert who claims that he or she doesn’t need to read the literature on the subject because no amount of literature could change his or her mind."


It would seem that Judge Jones had precedent solidly on his side in being highly skeptical of Behe's claims.

7:34 am  
Blogger Andrew Rowell said...

Matt & Ian,

Thanks for your help here.

I have a couple of queries.

1. Can you clarify the point in the process where the light and heavy chains evolve? Ian, you imply that they were present in an APAR like receptor prior to the insertion of the transposon. Matt, In your article you say that the heavy and light chains are the result of a duplication of the entire locus after the origin of the variable V and J segments.

2. Can you clarify for me the original nature of the RAG transposase. Was it prokaryotic in origin and reliant upon the eukaryotic transcription signals?

3. Do either of you know the answer to my last question at the bottom of the post... or maybe it is a silly question???
I assume that there are two of these transposon like segments that excise. The VD excision and the DJ excision. Do the two excisions use the same machinery? ie the same RSS and excision proteins?

8:25 am  
Blogger Andrew Rowell said...

Ian & Matt,

Another question.... Do these transposons insert on an entirely random basis into the host DNA or do they tend to target actively transcribed regions when they insert?

8:46 am  
Anonymous Matt Inlay said...

From the OP:
BTW are there different RSS sequences at the two join sites and different RAG proteins…what stops them getting muddled?

RSSs have a general structure of a conserved heptamer, a spacer sequence, and a conserved nonamer. There are two types of RSSs, ones with a 12 nucleotide spacer, and ones with a 23 nucleotide spacer. Rearrangement can only occur between an RSS with a 12 bp spacer and an RSS with a 23 bp spacer. This is commonly known as the "12/23 rule". It should be noted that the Transib transposason recently discovered in sea urchin is structurally identical (heptamer-space-nonamer, 12/23 rule), something that was predicted by transposon model years before it was discovered.

To answer a few of your other questions:

1. Can you clarify the point in the process where the light and heavy chains evolve?

Great question. Based on the structures of the known rearranging antigen receptors (e.g. IgH, Igk, TCRa, TCRb, to name a few), the ancestral antigen receptor gene must have looked like this:

----V---C-----

Note that this gene formed a homodimer. After the insertion of the RAG transposon into the V region, it split the V into a V and a J:

----V---J-----C-----

This general structure is similar to the TCRalpha locus, and the Igkappa locus. A gene duplication would have created two light chains, which could form a heterodimer. Now here's the catch, IgH (heavy chain) and TCRbeta both V, D, and J gene segments, so a 2nd event created the D segment. Initially it was presumed that a 2nd RAG transposon insertion must have created it. However, given the improbable nature of the 1st insertion, it was considered highly improbable that a 2nd insertion would occur right next to the first. Definitely a challenge for the transposon model. However, it was later discovered that an improper rearrangement could duplicate the region in between the V and J segments, which could create another gene segment between them.

---V---D---J----C-----

So that is the proposed pathway for the evolution of the heavy chain gene from the light chain gene.

Again, do you have an alternate pathway for the origin of these genes that is both detailed and testable?

6:42 pm  
Anonymous Matt Inlay said...

A point of clarification:
I assume that there are two of these transposon like segments that excise. The VD excision and the DJ excision. Do the two excisions use the same machinery? ie the same RSS and excision proteins?

The RSSs flanking the V regions of the IgH locus have 23 bp spacers, as well as the RSSs of the J regions. Therefore, they are incapable of rearranging with one another (12/23 rule). The D region has 12 bp spacers on both sides of it (one to rearrange with the J segments and another for the V segments). The 12/23 rule is also what prevents one J from rearranging with another J, Ds with Ds, and Vs with Vs.

2. Can you clarify for me the original nature of the RAG transposase. Was it prokaryotic in origin and reliant upon the eukaryotic transcription signals?

I'm guessing that the original RAG transposase was prokaryotic in origin. To the best of my knowledge, transposons have their own promoters, but are still somewhat reliant on the transcriptional activity of the region they insert into. In other words, if they insert into a transcriptionally dead region (i.e. heterochromatin), they won't be expressed.

Because of the abundance of transposons, organisms have developed their own mechanisms to repress transposition. It is generally thought that DNA methylation evolved as a mechanism to repress transposons. I'm not sure if I have the story right though.

12:33 am  
Blogger Andrew Rowell said...

Matt,

Looking at the adaptive immune system as a whole and especially at some of the difficult problems of its origin do you ever have doubts about the ability of unintelligent activity to make it?

9:03 am  
Blogger Andrew Rowell said...

Matt,

If your pathway is correct is this a reasonable hypothesis?

Take an organism that relies on only on innate immunity. Insert a transposon like contruct with the RAG genes and the correct border sequences into a suitable innate system receptor. This should result in a selective advantage for that organism.

9:18 am  
Blogger Andrew Rowell said...

Matt,
(I hope I am not trying your patience!)

Is there a diagram that shows how your sequence of gene duplications, genome duplication and incorrect recombination events produces the correct RSS sequences flanking the V, D and J regions?

9:26 am  
Blogger Andrew Rowell said...

Matt,

Are you saying that you disagree with Ian about the the RAG insertion into an APAR like receptor which already had a heavy/light chain structure?

9:33 am  
Blogger Richard H said...

Andrew,

I felt a need to ask a few related, more general questions at this point, but please do not let this interrupt your knowledge acquisition on the theory of evolution.

1. If you do not fully understand the science behind evolutionary theory, as you clearly do not like most people, why do you so easily dismiss it in favour of a supernatural cause?

2. Why can you not at least say "Both evolution and a supernatural cause are possibilities for the origin of life"?

3. On what scientific basis can you completely exclude evolution as the origin of life?

2:58 pm  
Blogger Andrew Rowell said...

Hello Richard,

1. I would say that fully understanding all the science of evolutionary theory ie understanding all of the littel that we do know about all the areas of biology is impossible for a single person. Even if you read all your life you would not be able even to read all the reviews let alone all the original papers. If someone claims to know the whole story... or gives the impression that they know the whole story then you can be sure that they are bluffing.

Why do I dismiss it? I do not. I believe that micro-evolution is a fact and that it was a very important discovery.

The issue over which I am sceptical is whether it can be extrapolated to explain the origin of all of biology.

I would say that my upbringing, personal experience and the study of molecular biology are the three things which are behind my resistance to swallowing the entire macro-evolution story.

2. (NB Evolutionists do not include abiogenesis as part of evolution) My answer would be if chance could do it alone why add in another unecessary cause?

3. There is no strictly scientific basis at the moment by which I can completely exclude chance & nat sel. as the basis for the origin of life. (I think that some of the elements are there with IC ideas and probability type arguments but if you define scientific as what has convinced most scientists clearly ID is nowhere near atm)Likewise I do not believe that there is any scientific basis for excluding intelligence as the basis for the origin of life... other than artificial boundary definition. That is what the ID people are seeking to do --- historically and as a current fact of society many people have some kind of conviction as they look at nature that chance and selection cannot account for it. ID asks -Has that conviction got any real rational scientific basis or is it just evolution hoodwinking us? The IC argument is an attempt to put into scientific form or even mathematical form an argument for design of some kind. I think it is a fascinating area of intellectual endeavour.

6:50 pm  
Blogger Ian Musgrave said...

Illness, deadlines, marking. Will be back later.

10:47 pm  
Anonymous Matt Inlay said...

"Looking at the adaptive immune system as a whole and especially at some of the difficult problems of its origin do you ever have doubts about the ability of unintelligent activity to make it?"

Not really. I say "not really" instead of "no" because it's not something I think about. As scientists, all we can do is seek explanations for phenomenons. If an explanation cannot be tested, we usually disregard it. That doesn't mean that the explanation isn't correct, but if we have no way to study it or test it, then scientifically, there isn't much we can do with it. This is essentially why no one takes ID seriously.

I do have some doubts that we'll ever be able to fully explain all aspects of the origin of the immune system, however, I'm really not worried about it. I feel that we're trying to find the origin of the immune system in the best way we can, through the scientific method. I don't really care what the final answer is.

10:49 pm  
Anonymous Matt Inlay said...

"If your pathway is correct is this a reasonable hypothesis?

Take an organism that relies on only on innate immunity. Insert a transposon like contruct with the RAG genes and the correct border sequences into a suitable innate system receptor. This should result in a selective advantage for that organism."


This is something I've pondered for awhile. I think if you set up the experiment correctly, it could work. For example, instead of an innate immune receptor, we used an antibiotic resistance gene. We insert the RAG transposon into the gene within the binding pocket. We then treat the bacteria with a variety of antibiotics, selecting for a strain that has the transposon leave the gene, and create some kind of sequence difference within the gene that allows resistance to the antibiotic. There are some holes in this setup, but I think the basic idea is sound. I bet Ian could come up with a better test.

10:57 pm  
Anonymous Matt Inlay said...

"Is there a diagram that shows how your sequence of gene duplications, genome duplication and incorrect recombination events produces the correct RSS sequences flanking the V, D and J regions?"

There's several papers in the 58 articles that address this issue. One of them has a great figure that I think contains everything you asked for, but I don't remember which paper it is. Here's a paper that comes close:

http://www.jem.org/cgi/content/full/191/10/1631

Check out figure 2. Email me if you don't have access.

It's important to note that there's a lot more complexities to the system than what is written in DBB. For example, one item that's often forgotten is that there are multiple Vs, Ds, and Js for each receptor, and some are organized in unusual ways, depending on which organism they're in. However, the differences between each organism help to establish a putative pathway for how the system evolved. That is, similarities between organisms suggest events that occurred before the organisms split, and differences suggest events that occurred afterwards.

11:10 pm  
Anonymous Matt Inlay said...

Are you saying that you disagree with Ian about the the RAG insertion into an APAR like receptor which already had a heavy/light chain structure?

I think this is the paper that Ian is referring to:
http://www.jimmunol.org/cgi/content/full/174/5/2885

The APAR receptors here don't have constant regions, which is where the difference between heavy and light chains lie.

11:25 pm  
Blogger Andrew Rowell said...

Matt,
"This is something I've pondered for awhile. I think if you set up the experiment correctly, it could work."

I was more interested in the idea of the innate immune system receptors being significantly immproved in their efficiency by what is essentially random mutation in their binding sites. Is there any evidence for this idea?

8:23 pm  
Anonymous Matt Inlay said...

"I was more interested in the idea of the innate immune system receptors being significantly immproved in their efficiency by what is essentially random mutation in their binding sites. Is there any evidence for this idea?"

Yes and no. The notion that innate immune receptors would be improved by diversifying is completely non-controversial. Think about the flu vaccines we get every year. The surface molecules on the flu viruses mutate, and so the antibodies we possess for last years flu can no longer bind. Basically, all organisms are in an arms race with viruses, only viruses can evolve much more rapidly than other organisms.

There are other mechanisms by which antibody genes can diversify. One of which is called somatic hypermutation. In this process, point mutations are introduced (randomly) into the antibody genes. The cells that express the antibody genes are then selected for those that can produce antibodies with a higher affinity (than the other mutating cells). This is one of the reasons why our immune systems improve with each infection. So the evidence you're looking for already exists in the form of another subsystem of the adaptive immune system.

3:06 am  

Post a Comment

<< Home