Field of Science

Guest Post - Don't judge a cell by it's coat

Today's post features a Guest Post from Lucas Brouwers of Thoughtomics. He's written some wonderful science posts (including some great ones about evolution), so I was quite excited when he agreed to write a post for me. Give his blog a look, it's well worth it :)


This post was chosen as an Editor's Selection for ResearchBlogging.orgAs a reader of this blog, you have become closely familiar with a wide variety of aspects of the bacterial world. In this guest-post, I would like to take the opportunity to entertain you with a story on the evolution of eukaryotes (animals and plants), with a big prokaryotic twist!

One of the first things you'll notice when you compare an eukaryotic cell with a bacterial cell, is that eukaryotic cells look incredibly messy. There are all sorts of subcompartments and membranes laying on top of and besides each other. Bacteria look like slick and minimalist living machines in comparison!

Eukaryotes shuttle a lot of proteins and other compounds between these compartments. But since all these comportments are isolated from eachother via membranes, this shuttling requires some special tricks. All eukaryotic membranes contain 'membrane coat' proteins that can fuse together to create vesicles that bud from the mother membrane, and fuse with a target membrane. Inside such vesicle, proteins and compounds can be transported to their new destination. As a picture is worth a 1000 words, and a video even more, you can see the process of vesicle formation for yourselves in the video below.

There are three classes of these membrane coat proteins in eukaryotes, which all share the same three-dimensional structure. A structural biologist would tell you that the features they have in common are a 'beta propeller' and a 'SPAH domain'. Because they all look so similar, scientists believe that these proteins have a common origin. Since they're present in every single eukaryote that we know of, it's reasonable to assume that such a membrane coat protein was already present in the latest common ancestor of all eukaryotes. This last common ancestor would still be very prokaryote-like, so we could expect to find at least some similar membrane coat proteins in prokaryotes. The problem is that up till now, nobody was able to detect these membrane coat proteins in prokaryotes!

The usual way scientists search for similar proteins across species is via sequence similarity: if two proteins in different species roughly have the same amino acids, it's likely that they do the same things in both species. But as time progresses, the differences between such proteins of different species can build up beyond recognition. In this particular case this seems to be the problem, since we're looking for proteins that diverged billions of years ago, before the eukaryotic/prokaryotic split! That is why Rachel Santarella-Mellwig and colleagues decided to take a different approach. Instead of searching for proteins with a similar sequence, they searched for proteins with a similar three-dimensional structure, using structure prediction algorithms.

Surprisingly, they managed to find membrane-coat-like proteins in several bacterial species. All the bacteria where they identified these proteins, belong to the superphylum of Planctomycetes, Verrucomicrobia and Chlamydiae (PVC). Many of these bacteria have the ability to turn their internal membrane inwards, surrounding the DNA with a double membrane (that certainly sounds familiar!). To investage whether the predicted membrane coat proteins also associate with membranes in these bacteria, the team did some further experiments. They choose one of the PVC bacteria analyze in greater detail: the lucky candidate was the freshwater bacterium G. obscuriglobus. The proteins were targeted with gold-coated antibodies, so that they could be visualized with electron microscopy. They found that 95% of the protein localized to the space between the inner and outer membrane, of which more than one third were in close contact with a membrane of a vesicle. If you look carefully in figure five from the paper, you can see small black dots that are the gold particles binding to the proteins (for the people who don't look carefully: there are big black arrows pointing at them). They're located at the edges (membranes) of these dark blobs, which are the vesicles within G. obscuriglobus.
So although the sequences of these proteins have changed beyond recognition, the three-dimensional structure and the overall function of these proteins seem to have been retained through evolution! The authors did not find evidence for an alternative scenario, in which the PVC-ancestor 'gobbled up' the gene from an eukaryote. Thus the likeliest scenario seems to be that a simple membrane folding mechanism evolved in the common ancestor of PVC's and the protoeukaryote, spurred by or accompanied with the evolution of these membrane coat proteins. The authors are a bit more careful in their conclusion, at the end of their paper they write: "... this suggest that the PVC bacterial superphylum contributed significantly to eukaryogenesis". All in all, I think this is a great example of bioinformatics and experimental work coming together in a fascinating story about eukaryotic origins. It makes me wonder what else lies hidden in all databases, waiting for the right questions to be asked...


Santarella-Mellwig R, Franke J, Jaedicke A, Gorjanacz M, Bauer U, Budd A, Mattaj IW, & Devos DP (2010). The compartmentalized bacteria of the planctomycetes-verrucomicrobia-chlamydiae superphylum have membrane coat-like proteins. PLoS biology, 8 (1) PMID: 20087413


Guest Posts

I'll be continuing with my own posts, for those of you that miss the enthusiastic monologues about plants and bacteria :p but I am still accepting guest posts to help keep a regular schedule during what's turning out to be an insanely busy term. So if you've ever felt like writing a lab-rat blog post send me an email!


takluyver said...

Not having read the paper: without the sequence similarity, what makes them believe that these proteins are related to those of eukaryotes, rather than being an example of parallel evolution?

Charles Kennett said...

Didn't know that PVCs can enfold their genetic material in a double layer membrane. How cool! Wonder how Lynn Margulis would work this fact into endosymbiotic origin of eukaryotes.

Lucas Brouwers said...

@Thomas The authors devote a paragraph in their discussion to your question. They argue that the absence of sequence similarity between eukaryotes and prokaryotes is uninformative, because the sequence similarity between eukaryotic MCs themselves has also been lost.
Since structure is more conserved than sequence during evolution, the retention of the same architecture + similar function can be seen as an argument in favor of divergent evolution (this is comparable to FtsZ / tubulin and MreB /actin)
Hope to have answered your question :).