Sunday Protist -- Aspidisca: Walking ciliates with scrambled genomes

Complicated cells are a weakness of mine, as you may have noticed. Ciliates satisfy this weakness really nicely, as they are utterly convoluted and insane in terms of both cell and genomic organisation. Ciliates are bizzare in almost every respect possible, from their sex lives to mitoses to nuclei on crack and extremely complicated morphology, and some are even loaded with epi- and endosymbionts, have parasitic bacteria inside their nuclei and there's even a couple anaerobes with reduced mitochondria. Hypotrichs are especially weird morphologically and genomically, and many have one other feature that makes them absolutely adorable: They WALK!

That's right, instead of swimming around like a normal flagellate must do, these creatures have thick bundles of microtubules called cirri which they use as little "feet" to walk on surfaces. One such creature, Aspidisca, apparently loves walking so much it rarely ever swims (Banchetti et al. 2003 Can J Zool)!

If that scalebar were 10cm rather than 10um, Aspidisca would make such a wonderful pet! One can almost imagine what a wonderful feeling it would be to have cirri gently brush against the palm of your hand... (Rosati et al. 1987 Trans Am Microscop Soc)

Aspidisca lives in marine coastal sands, which explains why it doesn't really need to swim much. Like most other ciliates, it's a predator, and probably not one you'd like to encounter if it were of our size. Also, like most, if not all, other ciliates, Aspidisca requires sex to make fresh somatic nuclei (Warning: The following figure contains porn.)

Porn. Mating Aspidisca. Their macronuclei needed a little refreshing. Yes, that's their cytoplasm fusion zone/tube between them. I knew you'd like that. (Rosati et al. 1998 Zoomorphol.)

Now, hypotrichs aren't just weird on the surface -- they are also utterly bizzare at heart. To them, having two differentiated nuclei, one resembling a giant bag of linear eukaryotic 'plasmids' and not even doing mitosis properly, is mundane and boring. They go the extra mile and scramble their genomes. That's right, instead of having your regular eukaryotic gene (interspersed with a few introns, of course), they split up fragments of their genes and mix up their order and direction. (This must really scramble the brains of the unfortunate souls forced to do bioinformatics on these organisms!)

Before we go into gene scrambling, a very brief primer on ciliate genomes is in order. As you may already have heard, ciliates have a transcriptionally-silent micronucleus (MIC) and a transcriptionally-active massively polyploid bag of short chromosomes called a macronucleus (MAC). The MAC may have a ridiculous number of copies of a single chromosome, such as 9000 for some chromsomes in Tetrahymena (Yao & Chao 2005 Annu Rev Genet). The splitting of such nuclei during cell division is a topic worthy of a few whole posts, although I must point out that ciliates have rather insane regulation strategies for DNA replication due to the variable gene copy number after each mitosis.

While the MAC sequences originate from the MIC (right after conjugation), they don't actually match their MIC counterparts exactly; in fact, certain portions are deleted along their way from the MIC! Furthermore, whether a portion is retained or not depends on its presence in the old MAC, which is already destroyed by this point. There's a whole epigenetic process of selected sequence deletion that's also a topic for another time (you can read about it here). Essentially, you have MAC-Destined Sequences (MDS) and Internally Eliminated Sequences (IES), which are retained and excised, respectively. They kind of act like exons and introns on a genomic level. Be very careful though, as it is very easy to begin mixing up IESs and introns -- they're just so alike in sound and concept, and yet so different...

This mechanism of IES excision is very potent at enabling ciliate genomes to quickly fill up with junk where normal organisms wouldn't tolerate it (and die off). An insertion of a random fragment into a coding region of a gene in the MIC would be non-deleterious as it would be quietly removed when the new MAC is formed. This is actually kind of annoying if one needs to insert foreign genes (eg. for GFP-tagged proteins) - adding them to the MIC would do little good, as they wouldn't make it to ever be expressed. One has to either also insert the foreign gene into the MAC as well, or tweak the epigenetic IES excision mechanism with miRNAs and such. It's a mess.

This genome-level "intron" situation gets weirder yet: in some species, the expressed gene fragments (MDSs) are not arranged in a sensible order in the MIC! That is, if you have gene fragments 1-2-3-4 normally, you can sometimes get 2-1-3-4 or even 3-1-4-2, or some other permutation thereof. The MAC sequence remains correct: 1-2-3-4 (otherwise, the organism would die!) Furthermore, those MIC fragments can even be reversed (see above figure)! How do we get from a huge mess in the germline MIC to a regular fully functional sequence in the MAC?

Unscrambling is still poorly understood, but it seems that in addition to IES removal, some form of recombination happens along the way. Furthermore, errors in the process are somehow corrected before the MAC matures. For Oxytricha ACTIN1, IESs seem to be removed before recombination and translocations (Möllenbeck et al. 2008 PLoS ONE; OA). As scrambling is slightly easier to grasp than unscrambling, a bit more seems to be understood about its evolution. It looks like gene scrambling is a fairly likely result of a system involving intricate sequence excision mechanisms, alternatively known a genome rearrangements.

Real examples of scrambled genes, and models of their evolution: Left - ACTIN1 in Oxytricha. Shows gene structure and then a model for its unscrambling. x's indicate recombination between paired repeats. Right - Model for initial scrambling of Oxytricha ALPHA-TELOMERE BINDING PROTEIN. Fragments of AT-rich DNA can be randomly inserted into the gene as IESs, to be quietly removed afterwards. By IES insertions when the gene is in some awkward topology (eg. loopy), genes can become scrambled. See source for further reading: Prescott 2000 Nature Rev Genet (free access), as well as Chang et al. 2005 PNAS)

So how do we get from genome rearrangements to scrambling? Enter recombination. If recombination scrambles up pieces of genes in most organisms, they die and we don't really see scrambled genes. However, due to their intricate mechanism of eliminating IESs, ciliates are able to unscramble the gene fragments and arrange them in the correct order. Likewise, they are also able to tolerate random insertions within genes, as they get removed in the somatic nucleus anyway. This excess capacity caused by an effective fixing mechanism enabled ciliates to tolerate disordered gene fragments -- a really nice example of a substantial increase in complexity with hardly any adaptive advantage whatsoever; that is, constructive neutral evolution. The evolution of ciliate gene scrambling is discussed in more detail in Stoltzfus 1999 J Mol Evol.

Strictly speaking, it is not known whether Aspidisca itself engages in this weird gene scrambling mess, as its genome is a mystery. It would probably be useful to check Euplotes -- if it scrambles, than most of the stuff in between it and Oxytricha, Stylonichia, Urostyla et al., including Aspidisca. Regardless, Aspidisca is close enough to the known scramblers, so I did have a right to ramble about it anyway.

In case someone else out there cares where Aspidisca branches in Euplotida. (Shen et al. 2010 Eur J Protistol; AOP)

Ciliates are another great model for both cell and molecular biology. Of the awesome Hypotrichs, Oxytricha has its genome sequenced and is ready to be played with. It also has really awesome morphogenesis stuff, but that's also a topic for a later post.

Almost completely off-topic, but apparently there's protein family specific to Alveolates that colocalises with the surface of the characteristic alveolar sacs just at the cortex: the Alveolins:

Blue - nucleo; Green - alveolin (Gould et al. 2008 MBE; free access)

This is really cool! In vivo flurescence imaging is another one of my obsessions, though immunostaining is ok too, provided it actually works...

References
Banchetti, R., Erra, F., Ricci, N., & Dini, F. (2003). Ethogram of Aspidisca sedigita Canadian Journal of Zoology, 81 (1), 14-20 DOI: 10.1139/z02-194

Gould, S., Tham, W., Cowman, A., McFadden, G., & Waller, R. (2008). Alveolins, a New Family of Cortical Proteins that Define the Protist Infrakingdom Alveolata Molecular Biology and Evolution, 25 (6), 1219-1230 DOI: 10.1093/molbev/msn070

Möllenbeck, M., Zhou, Y., Cavalcanti, A., Jönsson, F., Higgins, B., Chang, W., Juranek, S., Doak, T., Rozenberg, G., Lipps, H., & Landweber, L. (2008). The Pathway to Detangle a Scrambled Gene PLoS ONE, 3 (6) DOI: 10.1371/journal.pone.0002330

Prescott DM (2000). Genome gymnastics: unique modes of DNA evolution and processing in ciliates. Nature reviews. Genetics, 1 (3), 191-8 PMID: 11252748

Rosati, G., Verni, F., & Dini, F. (1998). Mating by conjugation in two species of the genus Aspidisca (Ciliata, Hypotrichida): an electron microscopic study Zoomorphology, 118 (1), 1-12 DOI: 10.1007/s004350050051

Rosati, G., Verni, F., Bracchi, P., & Dini, F. (1987). An Ultrastructural Analysis of the Ciliated Protozoon Aspidisca sp. Transactions of the American Microscopical Society, 106 (1) DOI: 10.2307/3226282

Stoltzfus, A. (1999). On the Possibility of Constructive Neutral Evolution Journal of Molecular Evolution, 49 (2), 169-181 DOI: 10.1007/PL00006540

Shen et al. 2010 European Journal of Protistology (Advance online pub, not yet indexed in DOI or ResearchBlogging)