Two rants on endosymbiosis

First off, quick point: Can we please stop using headlines like "Darwin was wrong about [x]"!? Srsly, big deal, some dead dude from the 19th century was wrong about something. Well shit. Evolutionary biology as we know it is now fundamentally flawed. Because Darwin didn't have PCR or fancy sequencers or GFP-tagged whatever. Tragic. What about all the shit Mendel was wrong about? Are we gonna ignore him now? /rant #01

Sometimes when writing up a post on something, I come across 'interesting' sites and papers. I mention my reaction in brackets, and find more crap to rant about. I then move the contents of the brackets to a footnote and unleash an off topic at the bottom of the post. But sometimes, this rant would be -really- off topic, and would be rather distracting if it becomes longer than the post itself. Today, we have come across one such case.

Warm-up - (excessive endosymbiosis)
Before embarking on a little journey of "Did he seriously just write that/get a faculty position/get a degree", a warm up paragraph from this week's Nature:

"In the former, the peptidoglycan layer is sandwiched between the outer and inner membranes, so that it surrounds the inner membrane: in contrast, in the latter there is no inner membrane, and the peptidoglycan layer, located outside the cell, surrounds the outer membrane." (Lake 2009 Nature) (via Catalogue of Organisms, who beat me to it, grrr XP)

That, my friends, is a wonderful example of epic semantics and topology fail. What he's talking about is that double membraned bacteria in question have cytosol-Inner Membrane(IM) - murein wall - Outer Membrane(OM) - outside. Single membraned bacteria have a cytosol-OM-murein wall-outside arrangement [sic]. Ie, somehow OM-M became IM-M-OM, raising the question of how the outer membrane end up on the other side of the murein. Thus, Lake invoked endosymbiosis to explain this 'conundrum' - an OM-M endosymbiont entered another OM-M prokaryote, and the endosymbiont OM became the inner membrane, while the host lost its murein wall. Very complicated stuff.

It may be quite evident at this point what the real 'conundrum' is there. Instead of comparing biochemical properties of the membranes to each other, he compared their relative positions. There is a very interesting topological property where if you have an double membrane and lose the outer layer, the former 'inner membrane' becomes the outer one. I think I may be onto something here...anyone wanna collaborate on a Nature paper? Any mathematicians out there wanna contribute a proof?

So how could someone who's probably a decent scientist fall for something like that? In fact, this seems common as soon as you put the 'hype' into 'hyp[e]othesis' - in this case, the guy seemed desparate for endosymbiosis, to the point of overlooking this very simple point in semantics. The reviewers and editors were no better - they too were getting carried away with the endosymbiosis hype (of course, they've still got ways to go to reach Margulis levels thereof...) For some reason, the fact that there's only one confirmed case of prokaryote-prokaryote endosymbiosis in the literature seems to worry no one...

(Coming from the TC-S camp of eukaryotic evolution, it was probably the double membrane state that was ancestral, with the loss of the outer membrane leading to what TC-S calls 'negibacteria', which eventually gave rise to us Neomurans. Even if you propose that single membraned bacteria came first, there's still no need for endosymbiosis, for they could have perhaps devised a way to form the outer membrane on their own. That would still be more parsimmonious, and more likely, than Lake's hypothesis above...) /rant #02

The big rant - (insufficient endosymbiosis)
We have some major endosymbiosis people in our department, so I never really came across any skeptics of mitochondrial endosymbiosis. The endosymbiotic theory of mitochondrial and plastid origins is pretty much beyond dispute these days, and the evidence is simply overwhelming. However, there always has to be someone to blow against the wind gale. Very rarely, they happen to be right; but even in those cases, their argument tends to be well-reasoned and supported by at least some data from the start. The other 99% of crackpots remain just that - timeless testament to our innate irrationality.

There's some guy who seems mildly annoyed by mitochondrial endosymbiosis. To the point of dedicating an entire website to the topic: http://www.origin-of-mitochondria.net/

So there's a whole page 'critiquing' the endosymbiotic mitochondrial origin theory. I know this is only half a step up from bashing creationists, but it got me a little irritated. Not because I feel threatened, but because it seems to be so easy to get employed as a crackpot, and I'm a little envious of their capabilites. See, if you actually try to abide by proper reason and the scientific method and all that crap, you'll be dirt poor and socially marginalised for the rest of your life. Thus, I shall enjoy one of the few advantages we do get - the feeling of intellectual superiority as you rip into some crackpot's drivel with a barrage of citations and proper data. It's a sport.

"The extensive gene transfer that is needed in the endosymbiotic theory would wreak havoc in a complex genome since frequent insertion of random pieces of mitochondrial DNA would disrupt existing functions."

Uhhh...heard of transposons, by any chance? I'm sure those are a few orders of magnitude more plentiful and more violent than the occasional piece of mitochondrial DNA. Yet they still...happen. And genomes have generally been able to deal with it. Random gene insertions do disrupt functions, but then you've got a few million other genomes to take their place! Isn't evolution awesome?

"Most pictures in textbooks of mitochondria resemble bacteria, but in reality, mitochondria form a dynamic network of interconnecting tubules (reticulum)"

May we introduce you to bacteria that don't look like 'bacteria'? Say hi to Streptomyces and Planctomyces, for example.

I'm just making shit up hypothesising here, but it seems like the stereoptypical bacterial morphology may be limited by the fact that it swims. There are certain shapes optimal for a flagellate lifestyle, and netlike/mycelial/branching forms are not among them. Once a bacterium has become commited to living exclusively in the intracellular environment, it no longer needs to be hydrodynamic, and can start taking on whatever other form it likes. I'd imagine that >800my is plenty of time for drastic morphological changes, considering eukaryotes managed it quite well.

"It is said that mitochondria, like bacteria, divide by fission, but the mechanisms are completely different and mitochondria use mainly components of unique eukaryotic origin."

Ever heard of intracellular parasites? An intracellular lifestyle does some weird shit in terms of intense reduction - microsporidia (fungi which shoot their cytoplasm into the host cell, where it takes over and lives off the host's resources, until forming new spores) have highly reduced genomes that have been harsh to introns due to space limitations(eg. 13 introns in an entire genome (E.cuniculi)), as well as a great purge of proteins for nucleotide+amino acid biosynthesis (Keeling & Slamovits 2005 Curr Op Genet Dev)* This makes sense - you don't have to make your own amino acids if you can just steal them from the host! So any degeneration and subsequent loss of previously essential genes is now tolerated, and thereby bound to happen.

*Ok, when you come across a paper with the following introductory paragraph, you just have to read it:
"At the bottom of the rabbit hole, Alice found a bottle labeled, ‘‘Drink Me’’. When she did, Alice shrank to a perfectly functioning, ten-inch miniature of herself. In reality, shrinking can be more difficult than simply drinking a potion, because the component parts of many systems are not themselves shrinkable, and so the system fails to function properly. In the world of eukaryotic nuclear genomes this is probably true, despite the fact that they vary in size by factors of hundreds of thousands (Figure 1), much more than all of Alice’s many transformations combined." (Keeling & Slamovits 2005 Curr Op Genet Dev; free access)
Classical studies geek really shows here...
(I'm quite bothered by the desolate desert around Rhizaria in fig.1 =( )

Similarly, the early mitochondrion no longer desperately needed to maintain its own division machinery, which eventually became transferred over to the host or lost. In a way, it has been able to hijack the host cell to take care of its own division. (so who's 'enslaving' whom here?) Through extreme evolutionary 'laziness', some lineages have been able to lose all genomic DNA entirely, and reduce to tiny membrane bound compartments essentially for specific parts of the host's metabolism. They basically 'disolved' into the host over time! (of course, de Roos' theory would probably claim those lineages as an ancestral state, eventually increasing in complexity. Too bad phylogeny king of stands in the way. Oops.) Just because an organism isn't capable of free life now doesn't mean it ancestrally wasn't either. Again, parasites support that very well. de Roos seems to have fallen for the 'evolution aims to gain complexity' misconception, and had difficluties with it 'going backwards', as it often likes to.

"So, although we see some characteristics that are shared between mitochondria and bacteria, we see many more examples where mitochondria are actually quite different."

Yeah, shit diverges over 850my. Just because they're 'different' doesn't mean they can't share a common origin, even a fairly recent one. Again, microsporidia were considered to be very ancient due to their apparent 'absense' of mitochondria and a highly reduced structure (wiser people were a bit alarmed by the latter detail; parasitism is almost universally a secondary trait (looking back far enough, it always is; first life must have been free living, otherwise we get the chicken-and-egg problem)). Turns out, they're fungi, like the mold in your fridge. We're not well-equiped mentally to deal with such timescales, but a lot can happen in just a few million years.

Thus, as long as we do not have a clear picture of the last common ancestor and its relationship with eukaryotes, it will be difficult to interpret gene similarity as evidence for the endosymbiotic theory.

This is where parsimony helps. Sure phylogeny is fallible (again, see microsporidia), but if an endosymbiont and a free living organism share a significantly large chunk of genes, it takes a lot less explanation and hand waving to invoke endosymbiosis than to craft elaborate hypotheses of weird massive lateral gene transfer stuff. That alone doesn't make it right, but definitely much more probable. And we're really working with probabilities here to begin with.

The mitochondrial genes could be derived from transposable elements, plastids or viruses and could come from either the nuclear genome or a bacterial genome.

'Domestication' of transposons is not as easy or probably as we may like it to be. Also, much of this would have to happen between the proto-eukaryote and the last common ancestor of most eukaryotes alive today, which is an epic question mark at the moment, although it does seem like that time period may not have been that long after all ('short' paper here: Cavalier-Smith 2006 Phil Trans R Soc B; free access). Cell structure, on the other hand, seems to be more malleable than large-scale gene organisation. Also, has there been at least one case of genes randomly congregating into a de novo genome in a random compartment? That would be quite ridiculously unlikely! How did the replication and maintenance machinery get in there then?

And plastid origin of mitochondrial genes? Ok, maybe once or twice that could, in theory, happen (has it?), but we're talking about mitochondrial genes in primarily plastid-less organisms! Does this guy propose a plastid endosymbiosis as ancestral to all eukaryotic lineages with mitochondrial genomes!? He seriously needs to explore something a little outside his metazoa. He needs to take one good look at a proper tree of eukaryotes (one without the 'crown eukaryote' abomination, kthx), and read a TC-S paper or two on eukaryogenesis. Or perhaps we should cross him with Margulis, and the result would have an intermediate phenotype, and perhaps even be a decent scientist!

Intermediates exist in the form of hydrogenosomes and mitosomes from amitochondriate primitive eukaryotes.

Hey, let's pull a little prank! How about we introduce him to Blastocystis and the ciliate Nyctotherus with mitochondria-like organelles (Stechmann et al. 2008 Curr Biol)? Actually, the table in that page, if you can access it, is a powerful demonstration of the dangers of relying on a single morphological for reconstructing evolutionary history. Essentially, if you follow organelle complexity, you'll get something like: Microsporidia, Giardia, Trichomonas, Nyctotherus, Blastocystis, and us. Let's draw that as a tree, mentally (let microsporidia be basal to the rest). So far so good. Ok, let's pull out a certain tree I tend to [ab]use a lot:

(Keeling et al 2005 trends ecol evol)
(I went for quite a while without pulling that out! Did you notice? See, self-restraint works sometimes! Until it doesn't...)

Let's do a little exercise. Grab a mental marker, and let's find Microsporidia. It will be among the opisthokonts, close to chytrids and zygomycetes. Done? Great, now find Giardia. It's a Diplomonad, close to Malawimonas in the Excavates. Trichomonas is a trichomonad, close to hypermastigotes (remember Trichonympha?), again in the Excavates. Now head over to the Chromalveolates, the alveolate neibourhood, for Nyctotherus, a ciliate. And then go down to the Stramenopiles, where you'll find Blastocystis between Actinophryids ('heliozoans') and Bolidophytes. And then point at pretty much everything else. And now look at our single-trait tree, which was built keeping de Roos' hypothesis in mind. So...how'd that go? I think someone needs to read up on basic eukaryote diversity before making shit up about the origins thereof...

See, while both de Roos and Cavalier-Smith like to make up grand hypotheses that tend to contradict the mainstream theories, Cavalier-Smith is actually good at it. He thoroughly reads astounding volumes of literature, formulates rational, testible hypothesis that make sense, and backs off his theories when evidence definitively proves them wrong (as with Archaezoa). de Roos has ways to go to even dream of such level.

And finally,

"In order for an evolutionary theory to be considered a scientific fact or a valid scientific theory, there are some basic requirements. First, it is necessary to have a reasonably detailed mechanism that explains the basic steps in the endosymbiotic scenario. [done] Second, this mechanism should be placed in the context of current Darwinian evolutionary theory and should contain no fundamental problems or falsifications[huh...?]. Third, a substantial body of empirical evidence that directly supports this scenario should be present.[nope, no evidence whatsoever... I know of a lab where people just sit around on their asses all day because there's simply no data in that field. Also, they don't publish any ridiculous number of papers, thereby making us cell biologists very jealous.] Fourth, no credible or logically sound alternatives should exist[huh? Since when is that a requirement for a valid theory?]. If these criteria are not met, the endosymbiotic theory cannot be considered to be a scientific fact that has been proven beyond reasonable doubt. Remarkably, the endosymbiotic theory fails all points." [bolded edits mine]

Yeah, to all my friends working on endosymbiosis: IT IS A LIE! OH NOES!!1!

Seriously, how can people argue that bullshit, and SOMEHOW be employed in biology?! This guy is apparently an actual biologist (although more of a biochem/bioinformatics background; As a devoted cell biologist, I have an obligation to hate them a little...you see, the academic community has ascended far beyond the primordial practices of stone age tribalism.) After a brief search, I found another interesting abstract, although we don't have access to this paper (and I can't be bothered to ILL it):

"Current theories about the origin of the eukaryotic cell all assume that during evolution a prokaryotic cell acquired a nucleus. Here, it is shown that a scenario in which the nucleus acquired a plasma membrane is inherently less complex because existing interfaces remain intact during evolution. Using this scenario, the evolution to the first eukaryotic cell can be modeled in three steps, based on the self-assembly of cellular membranes by lipid-protein interactions. First, the inclusion of chromosomes in a nuclear membrane is mediated by interactions between laminar proteins and lipid vesicles. Second, the formation of a primitive endoplasmic reticulum, or exomembrane, is induced by the expression of intrinsic membrane proteins. Third, a plasma membrane is formed by fusion of exomembrane vesicles on the cytoskeletal protein scaffold. All three self-assembly processes occur both in vivo and in vitro. This new model provides a gradual Darwinistic evolutionary model of the origins of the eukaryotic cell and suggests an inherent ability of an ancestral, primitive genome to induce its own inclusion in a membrane." (de Roos 2006 Artificial Life; emphasis mine)

Huh? Umm...this...just...like...errr...no! NO! Does not compute! AAAAAH! My eyes! I can feel my brain liquifying and oozing out of all sorts of orifices! See why the computery bioinformatics folk must be kept away from any mention of an actual organism? (ok, admittedly, some can manage it well, but that doesn't mean I shouldn't stereotype for personal fun =P )

I'll fix this sometime within the coming week, and 'translate' for you a real hypothesis on eukaryotic origins.

>Psi Wavefunction casts lvl10 TC-S Attack on lvl8 Crackpot for 500 damage.
Lvl5 Crackpot sustains 500 damage; HP 130/630
>Crackpot uses Copy Attack to cast Psi's lvl10 TC-S attack.
Psi sustains 0 damage due to TC-S Immunity.
>Psi casts lvl40 HAHAPWNEDLULz! on Crackpot for 1000 damage.
>Crackpot defeated!
*cue Final Fantasy victory music
>YOU gain 2000XP

Now back to working on this week's Sunday Protist ^.^

References
Cavalier-Smith, T. (2006). Cell evolution and Earth history: stasis and revolution Philosophical Transactions of the Royal Society B: Biological Sciences, 361 (1470), 969-1006 DOI: 10.1098/rstb.2006.1842

KEELING, P., & SLAMOVITS, C. (2005). Causes and effects of nuclear genome reduction Current Opinion in Genetics & Development, 15 (6), 601-608 DOI: 10.1016/j.gde.2005.09.003

Lake, J. (2009). Evidence for an early prokaryotic endosymbiosis Nature, 460 (7258), 967-971 DOI: 10.1038/nature08183

STECHMANN, A., HAMBLIN, K., PEREZBROCAL, V., GASTON, D., RICHMOND, G., VANDERGIEZEN, M., CLARK, C., & ROGER, A. (2008). Organelles in Blastocystis that Blur the Distinction between Mitochondria and Hydrogenosomes Current Biology, 18 (8), 580-585 DOI: 10.1016/j.cub.2008.03.037