I just sent this to my cell physiol prof:
The tree in slide 9 of the first lecture powerpoint is absolutely awful and grotesquely outdated (apparently based on Woese 1990). First of all, it is well-supported that the root is more likely to lie either between Eubacteria and Archaea+Eukarya, or within Eubacteria (I'm partial to the latter). It's rather murky in that particular diagram. Secondly, the eukaryote side of the tree makes one want to cry/punch a giant hole in the wall: Trichomonas and Giardia aren't exceptionaly likely to be the basal eukaryotes, and group reasonably well with Trypanosoma+Euglena, forming what may well be a monophyletic group. In any case, Trichomonas and Giardia seem to form a clade.The diagram comes from the "Big Alberts" text, 2008 edition. They should bloody know better by now! >.<
Dicty is WAAAY more closely related to Humans+Yeast than either Maize or Paramecium -- the later two are on a fundamentally different branch of eukaryotes -- the bikonts (eg. Cavalier-Smith 2009 J. Euk. Microbiol) or the corticates (Cavalier-Smith 2010 Biol. Lett.) Dicty, Humans and Yeast are members of a quite well-supported group (Unikonts), and this tree erroneously suggests that multicellularity evolved in the common ancestor of Plants, Animals and Fungi, which is completely wrong. Multicellularity evolved in each of those lineages independently, and in quite a few other lineages as well, up to ~16 times in the Eukaryotic kingdom (King 2004 Dev. Cell) and arguably a few times among prokaryotes as well (eg. Streptomyces).
I can't believe this is in a 2008 edition of the textbook! This tree has been outdated for at least a decade! How does one go about formally complaining with things like this? I refuse to let any suggestion of "crown/higher eukaryotes" vs. "stem/lower eukaryotes" go by unnoticed. The very notion of higher/lower organisms is thoroughly wrong and must be abandoned. Even in an introductory diagram. ESPECIALLY in an introductory diagram. Especially when they have the nerve to HIGHLIGHT Maize, Yeast and Humans as if they're some special enlightened life forms.
The era of Woese 1990 is over. We can move on anytime now...
Candidate for the Twisted Tree of Life Award?
Also, their 'interesting' spelling of eukaryotes is in itself a rantworthy topic for some other day...
Take a deep breath, there, Psi. As it happens, I think only half of your points are really howlers.
ReplyDeleteFirst off, the root is murky in that diagram, precisely for the reasons that you mention. The location of the root is not actually all that well-supported (you give mutually contradictory options yourself), and while the bacterial branch should diversify closer to its origin, the blobby base of the tree indicates that the matter is not settled yet, and I would argue that this aspect is not unreasonably presented.
But definitely, I am in your corner about the eukaryote side of the tree. SSU-based phylogenies of eukaryote diversity are loaded with FAIL, as anyone who reads any current research on the subject will have hammered into them. While many of these trees (and they get used in way too many places) are labelled as being based on the SSU (so knowledgeable people can disregard them accordingly), the fact that they are presented in the first place, usually without any other trees or contradicting text, is irresponsible.
Incidentally, I think that part of the reason why this tree keeps kicking around is not just that it was for many years the One True Tree, but also that it tells a compelling (if now thoroughly disproven) story: the first eukaryotes were unicellular organisms without mitochondria that became parasites, while the "crown" represents the development of supra-cellular morphological complexity. The fact that the story is completely wrong does not make it any less compelling to many, and it is considerably easier to present than the current tangle of lineages, which is both less intuitive and less unambiguous.
Meanwhile, you make the point that Dicty and opisthokonts form "a quite well-supported group", but this is only the case when considering unrooted trees of the six major eukaryotic groups. There are two problems with this. One is that the location of the root is not known; it could well be between Amoebozoa and Opisthokonta, in which case "unikonts" would be a paraphyletic assemblage. We disapprove of paraphyly, right? Since we do not know whether this is in fact the case, we try to avoid making blanket statements on the topic.
(Incidentally, when I was in my former lab, I was myself in favour of "Unikonta" as a supergroup. This proposition was disputed by our mutual Colombian friend, who made the argument that I just did, but somehow I was unconvinced until I moved away from my previous supervisor's corrupting influence. People at Dal tend to agree with our friend, and in any event I could never come up with a convincing counterargument. Unless, of course, you are convinced that the root lies outside the Opisthokonta+Amoebozoa assemblage, in which case, assert away!)
The other problem with the presentation of "Unikonta" as a clade is that there are eukaryotes that are members of none of the six (or five, if you insist on "Unikonta") supergroups. This further destabilises any assertions of the location of the root. Personally, I think determining the location of the root is premature until we have a better idea of the location of these "orphans" -- but then, I do research on the "orphans", so I would say something like that.
Finally, yes, "eucaryote" is an odd spelling, but since its stems are transliterated from Greek, often by way of Latin, the letters "C" and "K" are technically interchangeable. (I prefer to stick to "K" myself, as it is less ambiguous, but I cannot speak for everybody.) The initial definition of the three domains used the names "Bacteria", "Archaea", and "Eucarya", so officially the "C" is enshrined in legal immunity as a proper name. I might not agree with that, but nomenclature is an aggravatingly formal thing, and so that is unlikely to change.
Ok the murky root is excuseable, although most people would probably agree Archaea+Eukarya are 'holophyletic' (for our mutual Colombian friend =P); in fact, it's hard to imagine otherwise without invoking some rather fucked up HGT and prokaryotic endosymbiosis and utter madness (à la JA Lake). So the real murky part is the debate over the paraphyly vs monophyly (now to annoy our mutual Colombian friend) of Eubacteria. Which you could indicate by a polytomy, but they don't expect cell biologists to be even marginally literate in phylogeny. (rightly so =( )
ReplyDeleteIn non-evol textbooks they usually don't even bother indicating what the tree is based on, as if it's irrelevant. As if we have THE tree, and all sequences reveal THE tree unanimously. So now whenever people look at A tree (as they all are), they assume once again it's THE tree, and madness follows. So yes, thoroughly irresponsible.
Yeah, we need to have a chat about the Eukaryotic root =P I think the 'hloboseans' could come in handy there, as they have a nice messy distribution of amoeboid vs flagellate lineages. Ie, the big question is, which change is more likely: amoeboid -> flagellate or vice versa? Or equally likely? I'm still pondering over that.
Also, MyoII comes into the picture -- if the root is in Unikonta, then myoII becomes ancestral, and lost just before the divergence of bikonts or corticates. If the root is in the Eozoa (à la some crazy guy*, 2010), the excavate amoebae might make a little more sense. It seems Naegleria has some MyoII-like myosins (), which said crazy guy (2009 JEM) suggests happened through LGT from amoebozoa (!?); if more MyoII-esque genes are found in Excavate (amoebae, and who knows, flagellates too?), but NOT in the corticate amoebae (eg. the ochrophyte amoebae, chlorarachniophytes and other cercozoa), this may suggest a loss of MyoII in the corticates, contrary to its gain in unikonts proposed by Richards & TCS 2005 Nature. This would also suggest that amoeboid --> flagellate is more common/likely than flagellate --> amoeboid. Btw, by amoeboid I mean the serious, actin-based-and-tubulin-poor stuff, not the occasional floppy flagellate.
How do your bugs fit into all that? In the likely event of none of the above making any sense, I'm just sort of thinking outloud...
Of course, Myo II could also have arisen in paraphyletic unikonts, been mostly degraded in possibly-monophyletic excavates, and completely lost before the divergence of corticates. We'd have to look at way more genes though. Can I play?
For now I'll go with unikont holophyly. And maybe even bikont holophyly. I need to take this up with our mutual Colombian ex-student of Tom's =P I'll even use 'holophyly' to be diplomatic.
Btw, speaking of holophyly, DO we disapprove of paraphyly? ^_~
Your previous supervisor's corrupting influence... ahhh, yeah, I have some tingly feelings about the Chromalveolate Hypothesis. I know what you mean. I keep my trap shut. But...weird feeling. Seriously. It's as if all those non-photosynthetic and seemingly plastidless paraphyletic basal groups are trying to communicate some message of deep cosmic meaning to me. Hmmm.
So yeah, hurry up with those damn orphans. Btw, what do you think of Tom's recent shoving of Apusozoa into unikonts?
*for the confused: I mean this in an endearing way...
My link kinda failed; for the source of the Naegleria MyoII-oid things, see:
ReplyDeletehttp://genomebiology.com/2007/8/9/R196
Hmm... the location of the eukaryotic root is worthy of a blog post in and of itself, and I have neglected my own blog for ages now... but then I would feel compelled at least to explain what a eukaryote is in the first place (for those nonexistent-to-few people who read my blog but are not already informed on such matters), and that would amount to another blog post... hmm....
ReplyDeleteWell, I will get to that soon, I hope (it is good to have a mission for these things!). In the meantime, I can say only that I entertain no certainty about the location of the root. I expect that it might split up the excavates (but probably not in the way that our affectionately regarded crazy person suggests!), and it is possible that some of the "orphans" might turn out to branch before the unikont-bikont split, but I do my best to keep my mind open on this matter.
As for monophyly-versus-holophyly, I use the former term pretty much exclusively, but then I am still naive enough to think that holophyly is possible in a ranked (Linnaean) classification system. Thus, if paraphyly is to be avoided, the need for a term for "non-polyphyly" is lessened. But if more than a few people objected, I would happily adopt "holophyly" in place of the stricter meaning of "monophyly".
Hmm... I should get to that afore-allusioned blog post, but I should get to bed first....