28 March 2010

Introns and design

Mike Gene has posted an interesting series on introns that's worthy of a few comments. His thesis is that "introns facilitated the evolution of multicellular life."

A. The idea is interesting and rational but not novel. Research on introns and evolution is active and lively, and one prominent scientist in the field, Eugene Koonin, has proposed that introns drove many aspects of the evolution of eukaryotes (i.e., non-bacteria).

B. The more general proposal that mobile genetic elements have facilitated evolution of multicellular life is somewhat new (the first major review of the idea appeared in 2004) but also uncontroversial. Thus the notion that non-coding DNA elements are important evolutionary forces is not novel and is clearly not an explanatory problem for evolutionary biology. (And Mike did not claim that it is.)

C. Mike's overall design argument is based on front-loading, i.e., he postulates that "unicellular life was designed to frontload the appearance of metazoan life." The idea is that early unicellular life contained elements which were included mainly to facilitate the development of multicellular life. I'm not a fan of this kind of reasoning, but there's nothing profoundly irrational about it, and Mike can certainly claim to be in distinguished company on this one – Simon Conway Morris once referred to sponges as "animals in waiting," making a similar front-loading allusion without the emphasis on teleology.

Both Mike and Conway Morris are referring to the fact that various genetic elements that seem to be animal-specific or vertebrate-specific are nevertheless found in species thought to represent much more ancient lineages. Conway Morris talks of seemingly brain-specific components found in sponges (which lack a nervous system) and Mike notes that introns are found in single-cell eukaryotes, specifically the kind of organism (a choanoflagellate) that is not an animal but is thought to be related to the single-celled creatures that gave rise to plants and animals. Whether this should be called "front loading," I don't know. But it's pretty clear that the earliest eukaryotic cells, including those which probably gave rise to multicellular life, already had a lot of introns in them.

D. But Mike makes some mistakes that I want to focus on, because one of them illustrates a particularly serious intellectual danger of design-think.

1. Mike asserts that bacteria don't have introns. This is not quite true. The group II introns are present in many bacteria – in some genes – and are thought to be forebears of the group I introns that are so widespread in animal genomes. The implication is that the intron table was already set long before the eukaryotic (much less multicellular) party had started. The "front-loading" would appear to have occurred in bacteria themselves. (Mike alludes to this in the first post of the series.) And that matters because...

2. ...Mike says something really strange early in his series. His first "clue" that introns facilitated multicellular life is the observation that bacteria don't have introns and that bacteria don't make multicellular organisms. He concludes that bacteria are disabled by their lack of introns, and thus are unable to accomplish the feat of generating the wonders of the animal and plant worlds. He asks the bizarre question, "Where are the prokaryotic mice?" He points to the difference in complexity between prokaryotic cells (bacteria, basically) and eukaryotic cells (cells with nuclei, the cells that make up all plants and animals) and summarizes the diversity of cell types that can be generated by eukaryotic cells (brain cells, muscle cells, etc.) in contrast to the simplicity of prokaryotic cell types.  All true. But look carefully at his conclusion:
Note that while the vertebrate expression of the eukaryotic cell has been able to spawn 120 different cell types, both eubacteria and archaebacteria have not moved beyond a meager 2, even though these prokaryotes are both more numerous and older than eukaryotes.
It sounds reasonable, until you think about the tree of life and the reality of common descent, explanations which Mike Gene readily acknowledges. In the context of an unfolding tree of life, his statement reduces to "animals are eukaryotes" and nothing more. Let me explain.

Mike asserts that prokaryotes haven't spawned multicellular life. But that's not true. In fact, all indications are that prokaryotes spawned all multicellular life. The first eukaryotic cells are thought to have been forged by combinations of prokaryotic cells (this is the well-known endosymbiotic hypothesis) and the presence of group II introns in bacteria is just one of numerous observations that point to common ancestry between eukaryotes and prokaryotes. In other words, unless Mike is disavowing common ancestry (the trunk of the tree of life), then his conclusion doesn't make any sense. Prokaryotes really did give rise to multicellular life, by giving rise to ever more complex cell types.

Now, maybe you don't think there's a big mistake here. Mike is just saying that simple cells can't make complex multicellular assemblages, right? Well, no, he's saying something more than that. He's saying that through vast numbers and vast ages, they should have done so. He thinks it's notable that prokaryotes, even after almost 4 billion years and countless individual lives, haven't formed a mouse. I think this is crazy talk.
  • Bacteria actually do form very interesting multicellular assemblages. Organisms? Not really. But let's not ignore the fact that prokaryotes do know how to live in complex and cooperative environments. 
  • Most importantly, Mike's thinking here shows a weakness that I believe arises most commonly in design-oriented analyses of evolution. He seems to think that the presence of prokaryotes today, in all their lame simplicity, is notable because they coexist with those majestic miracles of design and performance, the awe-inspiring flora and fauna of metazoan life.  But that's nonsense. Prokaryotes are successful – wildly successful – in countless niches which have never been colonized by multicellular organisms. Why not ask why they are so successful? Why not write a series in which the presence of introns and mobile elements and organelles is put forth as an explanation for the disastrous failure of multicellular life to dislodge the humble archaea and bacteria from these positions of ecological dominance?
Do you see the mistake? It's in the assumption that a simple species is a failure, and a complex one is a success. Mike didn't ask why the most extreme environments on earth are populated by archaea, those ancient and simple prokaryotes, and not by highly-complex animals. That imbalance, I believe, arises from the emphasis on design, and it's an error we all wrestle with.

Mike's right about introns and their likely role in the origins of multicellular organisms. But he's wrong to associate complexity with success, and I think he's wrong to assert that introns in particular were necessary for the evolution of multicellularity and its associated complexity. I suspect that there are lots of ways to encourage genetic diversity and modularity, but that introns and splicing, dating to the RNA world, made the contribution because they were there when others weren't.

41 comments:

Arthur Hunt said...

It should be noted that Chlamydomonas possesses introns, and that their numbers and sizes are comparable to those seen in "higher" plants. This renders as incorrect (and irrelevant) MG's idea that intron number correlates with organism complexity.

Also, MG's confirmation bias shows thru when he cherry picks old and incomplete data in comparing alternative processing in different groups of eukaryotes.

All in all, I don't find his musings about introns and complexity to be very accurate or informative.

Bill said...

Once again MG starts with a conclusion and fishes for support. Same old creationist, ID nonsense we hear time and time again.

Rather than poof a wombat into existence the Designer poof "front loaded" a sponge because front-loading poofing is obviously easier than wombat poofing.

Yeah, brilliant!

Alexander said...

Slightly OT - it always seemed to me that a successful account of front-loading would have to explain how all the genetic material intended for future organisms (which presumably would not be useful to the earliest organisms) is preserved against deterioration by mutation with no purifying selection. Do you know of any serious attempts by front-loading proponents to address this issue?

Dan said...

I was going to make essentially the same point that Alexander did. Sean Carroll made it in The Making of The Fittest, critiquing Behe's version of front-loading he outlined somewhere. If a basic law of genetics is "use it or lose it" how do genes managed to be preserved until such time as organisms appear that can use them?

NickM said...

AFAI remember Mike doesn't actually accept the tree of life, instead he thinks that the 3 domains of life were specially created, and things were natural after that. So he is always trying to get as much recent design as possible packed into those first cells.

Doug Hayworth said...

Hi, Steve. I just wanted to drop a note to say how much I appreciate your blog. I'm an evolutionary biology PhD am no longer active in that field. This means that don't have the time to do detailed critiques and commentary about those topics myself but still want to read the graduate-school versions instead of the typical layperson ones. I highly value your posts. Thanks for feeding that part of my brain.
Doug

Joe G said...

OK so how does the theory of evolution explain introns, exons, editing and splicing?

IOW what is the evidence that demonstrates blind, undirected processes can produce editing and splicing machinery?

Joe G said...

Dan:
If a basic law of genetics is "use it or lose it" how do genes managed to be preserved until such time as organisms appear that can use them?

That may be a "law" of blind watchmaker genetics.

Not pre-programmed front-loaded genetics.

Anonymous said...

http://designmatrix.wordpress.com/2010/03/29/crazy-rabbit/

Stephen Matheson said...

Art, I agree that the correlation is imperfect and that Mike's analysis is incomplete. Can we agree that introns and mobile elements are very likely to have played a decisive role in eukaryotic and/or metazoan evolution? Mike gets props for thinking, IMO. So Bill, I share your skepticism but I insist that there's a vast gulf of thoughtful integrity between Mike Gene and the crew of the Discovery Institute or Reasons To Believe. That matters to me.

Dan & Alexander, I agree, and that's one big reason why I don't buy "front-loading."

Nick, I'm looking forward to Mike's response where he might correct me on that assumption.

Doug, thanks, and welcome! I'll get a link up to your blog.

JoeG, having seen your comment at Mike's blog and your own fulminations on yours, I offer you no response and bid you a polite farewell.

Joe G said...

IOW Steve, you can't support your position.

Understood.

Anonymous said...

Steve,

You might be interested in Mike's response to that issue here:

http://designmatrix.wordpress.com/2009/02/10/misunderstandings-about-front-loading-3/

Alan Fox said...

Re Alexander and Dan and the point that front-loaded genes would need to be preserved against mutation, would there then not also need to be a trigger for them to be "unpreserved" at the appropriate moment when the right organism is in the right environment? Parsimony?

Anonymous said...

Alan, you might want to look into "moonlighting" proteins.

James said...

Also, how do proponents of front-loading explain the distribution of genes in organisms today? If all future required genetic information was built into an original organism from which all eukaryotes are descended, why do plants not have the genes that animals use? Why do animals not have genes for producing chlorophyll? Are there answers to this or am I misunderstanding their proposal?

Anonymous said...

I'm kind of rusty I've been away from this debate due to work, and I need to catch up.

But to stick with the subject of the OP, introns in Chlamy have some characteristics that are shared with other plant lineages that appeared later in the green tree of life, and some other characteristics that are shared with animals. I'm sure there are many more examples of this when it comes to other factors. Divergence of form then occurred later, with different forms utilizing pre-existing genes for different purposes. Some of the pre-existing genes were subsequently lost.

Arthur Hunt said...

Hi Steve,
You asked:

Can we agree that introns and mobile elements are very likely to have played a decisive role in eukaryotic and/or metazoan evolution?

Just speaking of introns for now (to keep with the theme of the OP), I would say, in a word, no.

The way I look at this, there are three options: introns played a decisive role in multicellularity, some property of multicellularity played a decisive role in the dynamic of intron evolution in multicellular organisms, or there is no connection between the two (any perceived connection is actually just coincidence). If we are to presume some sort of cause-and-effect, then I think a better case can be made for the second of these options than the first. Specifically, I think that regulatory changes were more important for the evolution of multicellularity than were the sizes of proteomes (which, if I read things correctly, is the outcome of the expansion of number, size, and functioning of introns). And it is quite possible that regulatory changes, especially those that affect RNA metabolism, promoted subsequent intron evolution.

As an aside, here is another study that contradicts MG's assertions, starting with the very base of the eukaryotic tree. From the abstract:

"We find: (i) intron density in the plant-animal ancestor was high, perhaps two-thirds that of humans and three times that of Drosophila; and (ii) intron density in the ancestral bilateran was also high, equaling that of humans and four times that of Drosophila."

Bilbo said...

Hi Prof. Matheson,

I appreciate your responding to Mike's ideas. For those of us who follow his blog, it's more interesting when others challenge his views. And it's had the added benefit of flushing Art out of the bushes. I'll be curious to see if Mike responds to him as well.

One question: I can't help but notice your strong antipathy to a design perspective. I'm curious if part of that is rooted in your Reformed theology, which I believe takes a very dim view of Natural theology.

Bilbo said...

Hi Bill,

Mike doesn't start with a conclusion, he starts with a hypothesis, where fishing for supporting evidence is allowed.

I've never tried poofing a bacterium, but I imagine it would be easier than poofing a wombat. ;)

Bilbo said...

Alexander and Doug,

Mike Gene is the only guy I know working on the front-loading hypothesis. He would say that the way you preserve elements for multicellularity in prokaryotes is by making them useful in the prokaryotes.

Bilbo said...

Oops, that was Dan, not Doug.

Anonymous said...

IMO, the paper Arthur references doesn't contradict Mike's hypothesis, it complements it. It just studies genomic complexity from the point of view of intron loss.

That the early ancestors of eukaryotes had such genomic complexity, which may have lead to the origin of spliceosomal introns and the spliceosomal machinery, fits well with an evolutionary path that was "front-loaded". The early ancestors were therefore phenotypically plastic, followed by the evolution of modern organisms whose forms were stabilized by selection. Since those were the initial conditions (intron-rich ancestral genome), the ratio between intron losses and gains increased in many lineages, resulting in the observed trends towards losses.

A protein functions "arsenal" of a higher organism contains more
functions than the arsenal of a simpler one. And the functions are produced via a network of alternative splicing events with higher complexity.

Arthur Hunt said...

Hi Ashe,

My comments about introns in "lower" eukaryotes speaks to this remark by MG:

"Let me now provide a couple of clues to support the hypothesis that introns facilitated the evolution of multicellular life.

First, as a general rule, introns are far more common in multicellular genomes than single-celled genomes."


It's pretty clear that, "as a general rule", introns are not necessarily "far more common in multicellular organisms than single-celled genomes".

I guess I could have made more clear the connection between my comments and MG's claims. Sorry about that.

Anonymous said...

Still doesn't contradict MG's hypothesis, since he regards ancestors as exceptions that prove the rule, see his discussion on chaonoflagellates and chlamy.

You might be interested in this quotation:

"Protein-coding genes have, on average, 8.3 exons per gene and are intron-rich relative to other unicellular eukaryotes and land plants (15) (fig. S21); only 8% lack introns (Table 1) (1). The average Chlamydomonas intron is longer (373 bp) than that of many eukaryotes (16), and the average intron number and size are more similar to those of multicellular organisms than those of protists (fig. S21) (1, 17)."

http://www.sciencemag.org/cgi/content/abstract/318/5848/245

Alan Fox said...

Mike Gene points to a quote in a 1999 Scientific American Article where Sandro J. de Souza is quoted:

In general, nuclear introns are widespread in complex eukaryotes, or higher organisms. Simple prokaryotes and eukaryotes (such as fungi and protozoa) lack them.

Does Introns in protein-coding genes in Archaea indicate ongoing developments?

Anonymous said...

MG's wording is more accurate.

"First, as a general rule, introns are far more common in multicellular genomes than single-celled genomes,”

Arthur Hunt said...

Hi Ashe,

You said recently:

"MG's wording is more accurate.

"First, as a general rule, introns are far more common in multicellular genomes than single-celled genomes,”"


MG's wording does not accurately represent what Drs. Scott Roy and Walter Gilbert said in the paper I linked to above. To repeat:

""We find: (i) intron density in the plant-animal ancestor was high, perhaps two-thirds that of humans and three times that of Drosophila; and (ii) intron density in the ancestral bilateran was also high, equaling that of humans and four times that of Drosophila.""

If you don't mind my prying, I wonder why you accept MG's opinion and not that of Roy and Gilbert. Is there something about the approach the latter use that calls their conclusions into question?

Thanks.

Anonymous said...
This comment has been removed by the author.
Anonymous said...

I have no idea what you're talking about. I don't disagree with the Roy's conclusion at all, and neither does MG, and neither does Roy's conclusion contradict MG's hypothesis.

It looks like all you did here was completely ignore my response. See my above response and let me know if you have any questions.

Alan Fox said...

It seems Mike Gene has decided:

(quote)

No, the issue of introns is not central to my position. I raised the issue after reading Steve’s review of Meyer’s book, as I thought some people might be interested in a different approach than the usual evolution vs. design template. The intron issue is just one facet of my front-loading thesis and I once again demonstrated that a teleological perspective can a) generate testable hypotheses that are b) supported by some evidence.

(unquote)

and has switched moderation on (at least for me) at his blog. Not sure where the demonstrations took place that Gene refers to.

Arthur Hunt said...

More data that contradicts MG's "clue" can be found in Table 2 of this paper. Focus on the first two columns, and notice that there is considerable overlap between the extents intron contents of many unicellular organisms and animals (even without normalizing for genome sizes or gene numbers).

"First, as a general rule, introns are far more common in multicellular genomes than single-celled genomes,"

I don't think so.

Bilbo said...

Hi Art,

I'm enjoying watching the fight. Better than watching my Pistons lose. I was wondering if you're going to respond to Mike's latest thread.

Anonymous said...

Alan, try clicking on the big "Intron" word in his cloud on the sidebar (try reading the whole thing this time :))

Bilbo, I'm more interested where Arthur takes the "more data" from, as he has previously failed to provide any data that contradict the MG hypothesis.

Arthur, just curious, which of those organisms in that table do you think has "overlaps", in terms of intron number, density, size, etc, with multicellular critters?

Anonymous said...

Notice also how he is carefully avoiding answering my specific questions. It's bizarre.

Arthur Hunt said...

Arthur, just curious, which of those organisms in that table do you think has "overlaps", in terms of intron number, density, size, etc, with multicellular critters?

Fully half of the 14 organisms in the table that have intron numbers greater than Drosophila are not multicellular plants or animals. I haven't bothered to check into genome sizes, but it's a fair bet that most or all have gene numbers that are lower than Drosophila.

"...as a general rule, introns are far more common in multicellular genomes than single-celled genomes..."

That's not what the data say.

Bilbo, seeing that this is Steve's blog, I'm going to limit my further participation in this thread to back-and-forth with Steve (if there is any). So I guess I won't be having much to say about MG's efforts to rescue his hypothesis.

Bilbo said...

Art,

I'm not sure Steve would mind you taking on Mike here. But even if he did, there's always your own blog. We sports enthusiasts are willing to travel vast distances to see a good fight.

Anonymous said...

Looks like Mike responded:

http://designmatrix.wordpress.com/2010/04/04/the-general-rule/

Arthur Hunt said...

Steve,

I notice in your discussion of multicellularity in bacteria that you did not mention bacteria like filamentous cyanobacteria, myxobacteria, and some magnetic bacteria. These all have multicellular lifestyles (even obligatorily multicellular) that are quite akin to what we see in filamentous fungi, oomycetes, and the like. That these bacteria have such clear multicellular developmental programs really makes it hard to argue that introns in protein-coding genes are important for multicellularity in a general sense.

Anonymous said...

Arthur, It appears as if you once again misunderstood. MG's hypothesis is that introns is one of the factors that facilitated mammalian-like multicellular complexity. He acknowledged that bacteria can form multicellular assemblages in his response to Steve.

See here:
http://designmatrix.wordpress.com/2010/03/31/cell-plans-and-evolution/#more-2255

Brad said...

Steve, have you ever followed up on Mike's replies to this post?

Stephen Matheson said...

Hi Brad. No, I haven't. He got the last word, and I'm fine with that. I thought the discussion here was more enlightening, anyway.