Thumbs down, what?

Bird fingers confuse me, but the explanations confuse me more, it seems.

I didn’t mean to post today, but I’ve just read a new review/hypothesis paper about the identities of the stunted little things that pass for fingers in the wings of modern birds. The review part is fine, but I’m not sure I get the difference between the hypothesis Čapek et al. (2013) are proposing and the hypothesis they are trying to replace/improve.

To recap: the basic problem with bird fingers is that fossil, genetic and developmental evidence seem to say different things about them.

1. Fossils: birds pretty clearly come from dinosaurs, and the early dinosaurs we have fossils of have five fingers on their hands with the last two being reduced. Somewhat closer to birds, you get four fingers with #4 vestigial. And the most bird-like theropods have only three fingers, which look most like digits 1, 2 and 3 of your ordinary archosaur. (Although Limusaurus messes with this scheme a bit.)

2. Embryology: in developing limb buds, digits start out as little condensations of tissue, which develop into bits of cartilage and then finger bones. Wing buds develop a short-lived condensation in front of the first digit that actually forms, and another one behind the last “surviving” digit. Taking this at face value, then, the fingers are equivalent to digits 2, 3 and 4.

3. Genetics: In five-fingered limbs, each digit has a characteristic identity in terms of the genes expressed during its formation. The first finger of birds is most like an ordinary thumb, both when you focus on individual genes like members of the HoxD cluster and when you take the entire transcriptome. However, the other two digits have ambiguous transcriptomic identities. That is, bird wings have digit 1 and two weirdos.

Add to this the fact that in other cases of digit loss, number one is normally the first to go and number four stubbornly sticks around to the end, and you can see the headache birds have caused.

So those are the basic facts. The “old” hypothesis that causes the first part of my confusion is called the frame shift hypothesis, which suggests that the ancestors of birds did indeed lose digit 1, as in the digit that came from condensation 1 – but the next three digits adopted the identities of 1-2-3 rather than 2-3-4. (This idea, IMO, can easily leave room for mixed identities – just make it a partial frame shift.)

Čapek et al.’s new one, which they call the thumbs down hypothesis, is supposedly different from this. This is how the paper states the difference:

The FSH postulates an evolutionary event in which a dissociation occurs between the developmental formation of repeated elements (digits) and their subsequent individualization.


According to the TDH no change of identity of a homeotic nature occurs, but only the phenotypic realization of the developmental process is altered due to redirected growth induced by altered tissue topology. Digit identity stays the same. Also the TDH assumes that the patterning of the limb bud, by which the digit primordia are laid down, and their developmental realization, are different developmental modules in the first place.

(Before this, they spent quite a lot of words explaining how the loss of the original thumb could trigger developmental changes that make digit 2 more thumb-like.)

I…. struggle to see the difference. If you’ve (1) moved a structure to a different position, (2) subjected it to the influence of different genes, (3) and turned its morphology into that of another structure, how exactly is that not a change in identity?

Maybe you could say that “an evolutionary event” dissociating digit formation and identity is different from formation and identity being kind of independent from the start, but I checked Wagner and Gauthier’s (1999) original frame shift paper, and I think what they propose is closer to the second idea than the first:

Building on Tabin’s (43) insight, we suggest causal independence between the morphogenetic processes that create successive condensations in the limb bud and the ensuing developmental individualization of those repeated elements as they become the functional fingers in the mature hand, thus permitting an opportunity for some degree of independent evolutionary change.

Am I missing something? I feel a little bit stupid now.



Čapek D et al. (2013) Thumbs down: a molecular-morphogenetic approach to avian digit homology. Journal of Experimental Zoology B, published online 29/10/2013, doi: 10.1002/jez.b.22545

Wagner GP and Gauthier JA (1999) 1,2,3 = 2,3,4: A solution to the problem of the homology of the digits in the avian hand. PNAS 96:5111-5116

New genes, new tricks, part 2

In my previous post, I marvelled over the strange and unexpected way duplicated genes behave in fruit flies. The second study I wanted to discuss is also about new fruit fly genes gaining new functions, but unlike the other one, it’s about new genes that didn’t come from pre-existing genes.

Reinhardt et al. (2013) wasn’t the best written paper I’ve read, and I had some difficulty figuring out exactly what was going on in places, but there is some interesting stuff in there nonetheless.

The authors investigated six recently evolved new ?protein-coding genes in Drosophila. They wanted to know how they came about and managed to stick. For example, did they first originate as non-coding RNA genes? Did they gain a function through their RNA copies alone before they began to encode a protein? Or did they first awaken from the no man’s land between old genes with protein-coding potential already present?

This harkens back to one of the papers about new genes that I’d previously discussed. Xie et al. (2012) found that the genes for several human-specific proteins began life (and function?) as RNA genes expressed in particular tissues in ancestral primates. What about the six fly genes the new study investigated?

Reinhardt et al.‘s illustration of the two routes to protein-coding geneness is below. Starting with an inactive stretch of DNA (black line), you need two things: (1) an “on” switch or promoter (green box), which causes the transcription of RNA (blue) from the region, and (2) a sequence that can be translated into a decent length protein (an open reading frame or ORF, pink box). These two can theoretically appear in either order.

Before we get into the meat of the paper, let’s borrow the Drosophila family tree from the 12 genomes project page:

D. melanogaster, third from the top, is the species that has been used for every variety of biological investigation for over a hundred years, and also the focus of this study. However, the other species were also used for comparison, to see exactly where and how the genes originated.

Five of the six genes had a relatively long history, with similar sequences being found in D. yakuba and erecta or even further out in D. ananassae. Three of them were not only there in those species, but could also potentially make a nice protein. In two genes, the sequence or part of it was recognisable all the way to ananassae, but it only had long sensible ORFs in melanogaster itself.

In terms of activity… well, first of all I think they screwed up Figure 2. Supposedly, the names of the species in which transcription of these genes was detected are bolded, but actually, all the names are bolded in all the trees, which doesn’t agree with what they say (or with the green dots signifying the origin of transcription in the same figure). Anyway, assuming the bolding was a mistake and the green dots are in the right place, it sounds like four of the six genes were already active in the common ancestor of melanogaster and yakuba or earlier, while another two were only turned on in the melanogaster/sechellia/simulans lineage.

The order of events varies from gene to gene: four genes had good solid ORFs right from the start, while two were transcribed before they were suitable protein templates. The authors note that we can’t actually be sure whether or not the first four developed an ORF before they became active. To be certain of that, we would need more distantly related species with a matching ORF that isn’t transcribed, but in all four cases the species lacking expression of the gene also totally lack any trace of the sequence. So, while the remaining two genes provide positive evidence for the transcription-first scenario, the jury is still out on the ORF-first option.

In D. melanogaster, the presence of the protein product was confirmed for the four genes with the oldest ORFs. The two youngest may still be translated: the protein data came only from embryos, and in fact all six genes contain short signals that are normally associated with the transport of proteins to specific parts of the cell. You might reason that a gene that never makes a protein doesn’t need such signals, but nevertheless, the authors couldn’t positively confirm the existence of these proteins without data from other life stages.

Where these genes are active brings us back to a common theme we encountered in the previous post. In adult D. melanogaster, all six are most strongly expressed in the testicles, and the products of one of them are exclusive to those organs. Likewise, male larvae show more expression of all six genes than females do. The other species show basically the same pattern.

What do these genes do? Actually, do they do anything? Being expressed, even being translated to protein, doesn’t necessarily equate to having a function. Luckily, “function” is not terribly difficult to test for in fruit flies. There are lots of clever tricks that allow you to manipulate their genes and look at the consequences. In this case, Reinhardt et al. bred flies where these genes were turned off. If I understood them correctly, they managed to do this for five genes, four of which resulted in very dead flies. Weirdly, for all four, the affected flies died at the same life stage, just before hatching from the pupa.

With a different strategy that produced only partial knock-down of the genes, they got themselves some grown-up survivors, which allowed them to test the effect of the genes on male fertility (a sensible question given where these genes are most active). Out of three knock-downs with surviving adults of both sexes, only one showed a serious effect, and that was the one that produced generally crappy, short-lived weakling males anyway, so while these genes are active in the testicles and they might disproportionately affect males, they don’t seem to have much to do with fertility per se.

In general, the results sound like new genes that come from random bits of DNA can very quickly become essential to the organism, and it also sounds very much like an overabundance of transcripts in the testicles doesn’t mean that that’s where their function lies – it’s probably more that all kinds of things are expressed in testicles, and these genes are still expressed there because that’s how they started their lives.

Something big missing from the study is actually testing when these genes became functional – we’re told when they became expressed and when they started making a protein, but without manipulating them in relevant non-melanogaster species, it’s impossible to tell whether either of those means function. *disappointed pout*

And what’s up with those four genes that were necessary for the flies’ survival? The knock-downs all did their killing at the same stage. I don’t know what to think about that, and the authors don’t really offer an explanation beyond describing control experiments to make sure the deaths weren’t an unfortunate side-effect of the manipulation itself. Is there something about the development of adults that attracts new genes? Is the process of metamorphosis especially sensitive to even minor mess-ups? (More sensitive than early embryonic development?) Intuitively, I’d find the first possibility more likely, but gods know intuition is a poor guide to reality…



Reinhardt JA et al. (2013) De novo ORFs in Drosophila are important to organismal fitness and evolved rapidly from previously non-coding sequences. PLoS Genetics 9:e1003860

Xie C et al. (2012) Hominoid-specific de novo protein-coding genes originating from long non-coding RNAs. PLoS Genetics 8:e1002942

Superfood shelf. Just-fucking-what.

After feeding every stupid health scare ever with its “free from” labels, Tesco has reached a new low. Or maybe it reached it long ago and my brain just refused to see it.

The damn place has a SUPERFOOD SHELF now. Because salad can’t just be healthy, it has to be SUPER healthy.



I have to go and do some real science now.

The post-eating monster

This is the second time I wrote an entry, pressed publish, and ended up publishing a completely empty post. Why is this thing eating my carefully written blathers?

At least last time I had a saved version I could start again from. This one just disappeared into the aether.

I am sad and pissed and all kinds of grumpy now. Probably not a good time to ask tech support about the problem.

I might be back once I stopped screaming obscenities inside my head.

The return of the giant lichens?

Gosh, can someone tell me if this is bullshit or if he has a point? O.o

It’s rather annoying when a paper comes out that basically threatens to turn what you think you know on its head, and you’re simply not equipped to evaluate its claims. This is the case with Retallack (2012). I’m fascinated by early animals, and endlessly bewildered by the strange fossils of the late Precambrian. While I’m aware that Ediacaran fossils have been interpreted as everything from microbial mats through animals to giant protists, I had the impression that the non-animal interpretations of iconic fossils like Dickinsonia, Spriggina, Parvancorina or Charniodiscus have slowly retreated to the fringe in the decades since their discovery.

And now this guy, whose name I’ve heard enough times to pay attention, gets into Nature arguing that the namesake formation of the Ediacaran period actually originated on dry land, and the iconic fossils are preserved in a manner more like plants, fungi or lichens than animals.

The paltry one semester of introductory geoscience I did years ago is nowhere near enough to comment on all the stuff he says about soils and microbial mats and preservation. I feel completely out of my depth, rocking precariously at the mercy of the waves…

Obviously, this assessment of the original Ediacara site doesn’t affect every fossil site from the period. The latest Precambrian reefs of the Nama Group remain marine reefs containing the remains of unknown animals that grew some of the first mineralised skeletons.

My big question at the moment is how Retallack would interpret the preservation of the White Sea assemblage. This contains similar kinds of fossils to the sites he’s reinterpreted as terrestrial. There’s Dickinsonia and several others like it, there’s Parvancorina, there’s Cyclomedusa*. And this is where hundreds of specimens of my Platonic love Kimberella come from, often associated with crawling and feeding traces. That guy moved around and grazed – plants and lichens seldom do such things! So was Kimberella a land animal? That would be the biggest palaeontological sensation of the decade if not the century. Or did dickinsoniids etc. occur both on land and underwater? Or did the White Sea fossils span a wide variety of environments? (I’m not sure about the distribution of the various White Sea fossils relative to each other…)

Oh my. I wonder what will come out of this. Publication in Nature makes it dead certain that any expert who’d vehemently disagree will find the article. Let’s pull out the pop corn and watch…


*It’s slightly odd that he seemingly treats Cyclomedusa and other “medusoid” fossils as though most people considered them jellyfish. That may have been their original interpretation, but I thought it was widely discredited now.


Reference:Retallack GJ (2012) Ediacaran life on land. Nature advance online publication available 12/12/12, doi:10.1038/nature11777

Parasitic sex cells wtf yay?

I’m reading a review about asexual reproduction in colonial sea squirts (Brown and Swalla, 2012). This isn’t really the place I expected to stumble on something that made me giggle and whisper “holy fuck” at the same time, but I was totally wrong. Apparently, some squirt colonies can fuse and create a single “chimaera” colony. And then one member’s germ cells can take over the other’s gonads. Because these squirts keep some germ cells in their blood, so they can go wherever the hell they want in the whole colony. Like that poor sod’s testicles.

It seems everything remotely related to sex in the animal kingdom is at least a little bit kinky.

Also, sea squirts are weird, and this is Captain Obvious speaking.

(Must resist temptation to immediately dive into PubMed for articles on germline parasitism. I have a ton of reading to do that’s actually relevant to my work…)

(P.S.: I wonder if tagging this post “sex” will trigger a giant traffic spike. Consider this an experiment!)



Brown FD, Swalla BJ (2012) Evolution and development of budding by stem cells: ascidian coloniality as a case study. Developmental Biology 369:151-162

Neurulating acorn worms, and the Mammal is confused

To the everyday animal lover, acorn worms have few things going for them. They are pretty hideous as a rule, and they don’t really do anything interesting. (If you count “eating mud all day” as interesting, you are probably a sediment ecologist…)

True to their mud-eating selves, though, acorn worms are great at muddying the water in the evo-devo world. You see, there’s this neat idea that goes back to the early 19th century called dorsoventral inversion. French naturalist Étienne Geoffroy Saint-Hilaire once noted that an arthropod looks rather like an upside down vertebrate in many ways. This diagram from the Wikipedia entry illustrates the concept rather nicely:

The pink and snot-green triangles on the side represent the expression levels of the key genes that determine where the back and belly sides go. dpp (decapentaplegic) is the fly homologue of BMP4 (or BMP2, 4 and 7, to be precise), and sog (short gastrulation) is the fly version of chordin. So not only do major structures like the central nervous system develop on opposite sides of the body, the developmental genetics are also upside down. It seems like a no-brainer. (Well, chordates had to somehow move the mouth back to the belly side, but hey, no big deal ;))

Then those evil acorn worms come in and complicate things.

Acorn worms belong to the phylum Hemichordata. Hemichordates are most closely related to echinoderms like sea urchins. Together with chordates (vertebrates, sea squirts and lancelets), they make up the deuterostomes. Fruit flies and most worms that aren’t “acorn”, on the other hand, are protostomes.

Whether acorn worms and other hemichordates have a central nervous system at all has been the subject of some debate. Recent evidence suggests that they do (Nomaksteinsky et al., 2009). They don’t even just have one, they have two of them, taking the form of nerve cords on the dorsal and ventral sides of the animal. The dorsal nerve cord forms in a way eerily similar to our own spinal cord. Chordates like ourselves undergo a process called neurulation, in which some of the ectoderm (“skin”) on the embryo’s back folds inwards to form a hollow tube that’s the precursor of the central nervous system. Well, look what happens in an acorn worm (figure from Luttrell et al., 2012):

The left side of this image depicts the process in the acorn worm Ptychodera flava (and shows you a young worm – innit cute?), while the right side illustrates what happens in a chick embryo. (I’m not sure why they didn’t use actual photos of chick neurulation, I don’t imagine they’d be very hard to come by…)

In chordates, the notochord (precursor of our spine) sends out chemical signals to direct nerve cord formation. Acorn worms don’t have anything notochord-like at the time the nerve cord develops, but somehow they make it look like neurulation anyway. If it looks like a duck (or chick, as the case may be) and quacks like one, it might just be one.

The problem?

It happens on the wrong side. Having read the two papers I’ve cited so far, my impression is that neurulation only happens with the dorsal cord. However, in other respects these animals share the arthropod, not the vertebrate, orientation. Most importantly, their genetic dorsoventral axis is oriented like that of arthropods and other protostomes, with BMP levels in the embryo highest on the dorsal side (Lowe et al., 2006). Chordate embryos can’t even make a nervous system at high BMP levels!

It seems that whatever happened to turn the ancestral bilaterian on its head, it wasn’t as simple as flipping the animal and relocating the mouth. The development of the central nervous system shares serious developmental genetic similarities among deuterostomes like us and protostomes like flies (Arendt and Nübler-Jung, 1999) or ragworms (Denes et al., 2007), indicating that if not a full-blown CNS, then at least the genetic pattern was present in our common ancestor.

The figure above is a schematic comparison of gene expression in the developing nervous systems of fruit flies (Drosophila), ragworms (Platynereis) and vertebrates from Denes et al. (2007). The diagram labelled Enteropneust illustrates the lack of data from acorn worms, which I don’t fully understand given that Lowe et al. (2006) actually studied several of the genes included in this figure. Speaking of that, acorn worms once again prove to be weird and confusing, since the genes in question aren’t expressed in anything like the longitudinal stripes seen in the other animals. In fact, Lowe et al. (2006) found that they aren’t obviously associated with either of the nerve cords.

To be precise, Lowe et al. worked under the assumption that acorn worms had no nerve cords, but if you look at their pictures the lack of resemblance is blindingly obvious. For example, Msx, the light grey gene on the Platynereis and vertebrate diagrams, goes all around acorn worm embryos in a relatively narrow ring. Nk2.2, the red gene, isn’t even expressed in the ectoderm (the embryonic “skin”), whereas central nervous systems invariably come from there. Did Lowe et al. get the wrong genes or what? Don’t think so, Msxes at least are pretty easy to recognise…

To summarise: acorn worm central nervous systems develop much like ours, but on the wrong side of the body, with none of the genetic similarities we share with animals much less closely related to us than acorn worms. To top that, the damn worms have another nerve cord on the opposite side, which doesn’t develop by neurulation unless I’ve misunderstood something.

I… have no idea what’s going on here. Damn you, nature, you need to work on your clarity.



Arendt D & Nübler-Jung K (1999) Comparison of early nerve cord development in insects and vertebrates. Development 126:2309-2325

Denes AS et al. (2007) Molecular architecture of annelid nerve cord supports common origin of nervous system centralization in Bilateria. Cell 129:277-288

Lowe CJ et al. (2006) Dorsoventral patterning in hemichordates: insights into early chordate evolution. PLoS Biology 4:e291

Luttrell S et al. (2012) Ptychoderid hemichordate neurulation without a notochord. Integrative and Comparative Biology 52:829-834

Nomaksteinsky M et al. (2009) Centrarlization of the deuterostome nervous system predates chordates. Current Biology 19:1264-1269

This is when *I* need a good science blogger

Today you get to meet yet another of my random interests: the origin of life. (Is there a person with an interest in living things who isn’t fascinated by the origin of life?) And, since we sciencey types are very anxious about personal biases, I might as well start with a confession.

I love the RNA world hypothesis.

It was just one of those things that you learn at school/uni (I think it was 1st year molecular biology for me), and it’s so neat and elegant and compelling that you immediately fall in love. Sure, later, when you’re out of the inevitable simplicity of class, you learn about the nuances. The difficulties. But the evidence for still seems so convincing that you have no doubt that we’ll eventually solve the problems.

In case you aren’t familiar with it, the RNA world hypothesis is the leading solution to the chicken and egg problem that is the “central dogma” of molecular biology (diagram from Wikipedia):

DNA is great genetic hardware, but it’s nothing without proteins. Proteins are encoded in DNA, but the code is useless without proteins to read it. Making DNA requires proteins. But the proteins come from the DNA code. You see where this is going…

RNA takes the stage

The RNA world is an ingenious idea that elevates RNA from being merely the messenger between DNA and protein to centre stage. While its big brother DNA is a fairly stable and inert molecule, RNA is much more chemically active. It doesn’t like languishing in long, stable double helices – rather, it folds up into all kinds of odd shapes that can, surprisingly, catalyse a variety of chemical reactions. Just like proteins. Yet the “letters” of RNA can form complementary pairs, allowing for faithful copying. Just like DNA.

And, so the theory goes, there was a time when RNA was both the genome and the enzymes (enzymes made of RNA are called ribozymes). The right sort of RNA molecule could have copied itself without proteins [1], and performed whatever chemistry a primitive life form needed – also without proteins. Crucially, the right sort of RNA molecule could have invented proteins [2].

One of the key revelations to lend support to the RNA world hypothesis is that proteins in cells today are still made by RNA. Proteins are manufactured in ribosomes. A modern ribosome is a very complicated structure made of several folded-up RNA molecules and dozens of proteins. However, investigations of its structure (see Cech [2000] for a quick review) revealed that the place where amino acids are joined into a protein chain is all RNA – the proteins may support the RNA, but it seems to be the RNA that actually does the job.

Beautiful hypothesis vs. ugly facts?

So, everything is shiny and awesome and exciting. Ribozymes capable of all sorts of interesting chemistry [3] abound, and we have some very neat ideas regarding how RNA paved the way towards the modern protein-and-DNA world [2].

And then Harish and Caetano-Anollés (2012) come along, and I don’t know what to think.

A large part of the problem is that their methods go way over my head. I get the gist of their message. They figured out the relative ages of the RNA and protein components of the ribosome. The protein-synthesis parts – RNA and protein alike – turned out relatively new. They also found that the oldest protein parts interact with the oldest RNA parts – and seem to have coevolved. That, they say, would suggest that RNA and fairly large pieces of protein had a common history together before the future ribosome became capable of making proteins.

Yes, that means either that RNA didn’t invent proteins, or at the very least, that the “inventor” was not a precursor of the ribosome.

I really really don’t want to believe the former, and the latter possibility is a butchery of Occam’s razor without further evidence. But what else is left, if the study is correct?

One part of their results that I found intriguing is the structural similarity of the most ancient parts of ribosomal RNA to – you’d never guess – lab-evolved RNA-copying ribozymes. That is… oh, I don’t really know what it is, aside from “fascinating”. Did the ribosome start out as replication machinery, and turn into a protein factory only later? Or are the structures similar because reading the primitive genetic code required the same sort of molecular machine as copying RNA? Or is it even just coincidence?

And this is why I need a good science blogger. I need someone who deeply understands the paper and can translate it into something I can digest. Because at the moment, I can’t make heads or tails of this. I’m rather attached to the RNA world; it makes sense to me, and as far as scientific hypotheses go, it’s simply beautiful. Yet I can’t point to any obviously bullshit reasoning in the new study, other than where they seem to imply that because modern ribosomes need proteins to work, proteins must have been present in the ribosome from the start. (Which is a bit like every damn irreducible complexity argument advanced by creationists.) I just don’t have a good enough grasp on the methodology to tell whether it’s all solid or whether any of it is dodgy. Words fail to express how much that bugs me.



[1] Lincoln and Joyce (2009) and Wochner et al. (2011) came tantalisingly close to making/evolving the right sort of RNA molecule in the lab. The former’s pair of ribozymes can only copy each other by stitching together two half-ribozymes, but they can keep going at it forever and ever. Wochner et al.’s molecule can copy RNA using single letters as ingredients, but it runs out of steam after 90 or so of them. That’s several times better than the previous record, but still not long enough for the ribozyme to replicate its twice-as-long self.

[2] This excellent video describes one way it could have happened. When it comes to science education, cdk007 never fails to deliver!

[3] Including attaching amino acids to other RNA molecules (Turk et al., 2010) – look up tRNA if you don’t see why this is exciting 😉



Cech TR (2000) The ribosome is a ribozyme. Science 289:878-879

Harish A & Caetano-Anollés G (2012) Ribosomal history reveals origins of modern protein synthesis. PLoS ONE 7:e32776

Lincoln TA & Joyce GF (2009) Self-sustained replication of an RNA enzyme. Science 323:1229-1232

Turk RM et al. (2010) Multiple translational products from a five-nucleotide ribozyme. PNAS 107:4585-4589

Wochner A et al. (2011) Ribozyme-catalyzed transcription of an active ribozyme. Science 332:209-212