bottom-up biology at Birkbeck

Way back in 2001, when I was switching to full-time writing but wanted to keep some kind of connection with the world of academia, I used to trek to London twice a week to reside at the School of Crystallography, Birkbeck College, as a science writer in residence. That was an interesting experience while it lasted, and I got some good articles out of it, but as the railway connection got worse and more expensive over time, it wasn't really sustainable in the long run. (In 1999 I had even applied for a few London-based jobs thinking the railways situation can't get worse but it did!) Moreover, as Birkbeck set up a joint Institute of Structural Molecular Biology (ISMB) with UCL in 2003 and turbo-charged its research in this field, the space I used to have just disappeared.

The ISMB hosts an international symposium every other year, and as I was there when the first one happened, I enjoy the nostalgia trip of attending the latest instalments if and when I'm organised enough to make it happen. This year I was lucky and got there for both days of the symposium. I was rewarded with an amazing meeting that covered both the distinguished history of BBK structural biology (Rosalind Franklin, Aaron Klug, JD Bernal ... ) and its very exciting present. And I got another article out of it which is out now:

Building blocks for bottom-up biology

Current Biology Volume 28, Issue 14, 23 July 2018, Pages R761–R764

FREE access to full text and PDF download



The famously dilapidated pair of Georgian townhouses, 21 and 22 Torrington Square, where Rosalind Franklin worked for the last five years of her life, were later demolished and gave way to this, the Clore Management Centre, which is where the symposium was held (own photo).


PS and I got to test-ride the new rail line Oxford to Marylebone, with Chiltern Rail, which is indeed a bit better than what we had so far.

how microtubules grow

I helped writing up this PR about research done at Birkbeck College, published in the current issue of Cell:


Research sheds light on magic of microtubule growth

Transport within the cell relies on a system of tracks, the microtubules, which continuously grow or decay at one end, while the other end remains tethered and static. Researchers at the ISMB have now elucidated how some of the proteins that coordinate the building work on a growing end recognise the sites where they are needed.

Research into dynamic cellular systems such as the ever-changing microtubule cytoskeleton is complicated by the fact that structural methods typically require static and stable molecular complexes. The group of Carolyn Moores at Birkbeck Crystallography / ISMB, working together with Thomas Surrey at Cancer Research UK, has created a static model of the dynamically growing microtubule end, allowing them to pin down the structural interactions that hold the building site together.

Protein building blocks (tubulin proteins) that are added to the growing end normally carry the energy-rich substance GTP, which is split up into GDP and phosphate within seconds of incorporation. By offering building blocks with a non-cleavable analogue of GTP, the researchers could create a microtubule arrested in its growth but retaining the characteristics of a growing (GTP-carrying) end. Moores’ group at the ISMB could then obtain detailed images of this structure using electron microscopy at very low temperatures (cryo-EM).

Specifically, their work published in the latest issue of the journal “Cell” shows how a crucial group of proteins needed for the extension of the tip and its interaction with cellular features, the End-Binding (EB) proteins, bind to the building blocks that carry the uncleaved (and uncleavable) GTP. The researchers found that each of the EB proteins bridges two strings of tubulin subunits running the length of the tube structure, the so-called protofilaments. Intriguingly, each microtubule has a “seam” where the protofilaments are connected in a different fashion, and the EB proteins appear to recognise that difference, as they don’t bridge the seam.

“The fascinating behaviour of growing microtubule ends has been tantalising biologists for some time,” says Moores. “The biophysical experiments performed in the Surrey lab were essential in optimising sample preparation for our cryo-electron microscopy structural experiments, but we never anticipated the secrets of microtubule biology we would uncover when we started our collaboration.”


With the help of this model system, further research will uncover more details of the wider context in the dynamic behaviour of the cytoskeleton made up of the microtubules.

“Of course, we now have many more questions about the coordination between EB activity and the ebb and flow of microtubule growth and shrinkage,” says Moores, “in particular,: What do microtubule ends look like in the wider context of the multi-protein complexes which the EBs help to assemble? But our futures studies promise to take us in exciting directions in understanding regulation of the microtubule cytoskeleton.”


Reference:
P. Maurer, Franck J. Fourniol, Gergő Bohner, Carolyn A. Moores, Thomas Surrey:
EBs Recognize a Nucleotide-Dependent Structural Cap at Growing Microtubule Ends
Cell 2012, 149, 371-382, 13 April 2012




Image source: Cell

protein portal

Gabriel Waksman's group at the Institute for Structural and Molecular Biology (shared between Birkbeck College and University College London) has an exciting article out in tomorrow's issue of Nature, on the molecular machine which builds pili, i.e. the "hairs" that pathogenic bacteria use to attach themselves to their host. I wrote the following summary for the ISMB website:


The ability of bacteria to cause diseases in humans or animals depends, among other things, on their ability to stick to their host in order to be able to establish an infection. Many kinds of bacteria do this with the help of very thin protein hairs, known as pili. Like our hair, these pili grow from the root. With a detailed structural analysis of how this growth happens, ISMB researchers have now laid the foundations for medical applications.

E. coli strains causing urinary tract infections are one of many examples of bacteria depending on pili, as they would otherwise get washed away with the urine. The outermost end of their type I pili is a sticky protein that attaches itself to the surface of the urinary tract, using the host’s carbohydrate receptors.

Following the pilus from the tip inward, we first find two linker units FimG and FimF and then a large number of copies of the main hair builder, FimA, which is anchored in the outer membrane. In 1999 and 2002, the team of Gabriel Waksmann (then at Washington University Medical School in St Louis, USA and now at the ISMB) showed that these subunits fit together like pieces of a linear jigsaw puzzle. Each protein subunit has a well known structural pattern (the same that is also found in antibodies), but one bit of this structure is missing. The following subunit brings along the missing bit that latches into the hole and completes the structure.

This structural incompleteness of each single subunit means that each of them on its own is unstable in the periplasm, the space between the outer and the inner membrane of certain bacteria (only Gram-negative bacteria have this feature), where pili are manufactured. Therefore, they are being looked after by molecular chaperones, i.e. proteins that can protect these contact points from aggregation.

Now Waksman’s group has elucidated the detailed structure of a complex involving the outermost, “sticky” FimH subunit of an E. coli pilus, as it is being exported through the outer membrane by a specialised protein export machine (the FimD usher) and still being guarded by such a chaperone, FimC. The structure, which shows the pilus tip protein lined up inside a hollow cylindrical channel of the usher protein, is unique in that it is the first such structure of a complete protein export machine including the proper protein to be exported.

Previous biochemical studies had implicated the front end of the usher protein, the N-terminal domain, as a binding site for the pilus proteins to be exported. The new crystal structure, however, shows this site to be idle, while a second site, near the other end of the usher protein, the C terminus, binds the FimH-FimC complex. The authors conclude that the usher has two binding sites such that the growing pilus can remain anchored to one, while the next subunit to be added to it docks to the other.

The authors also conducted a crystallographic analysis of the empty FimD usher for comparison (although the structure of a similar, empty usher, PapC, had been solved before). They found that the binding of the substrate proteins induces a major structural rearrangement. The empty usher has an oval pore sealed by a specific part of the protein, the plug domain, while in the presence of FimH-FimC, the protein adopts a perfectly cylindrical shape, and the plug domain moves out of the way to open the channel for the substrate proteins.

Because this highly complex mechanism by which E.coli and similar bacteria assemble their pili has no equivalent in higher organisms, it appears to be a very promising drug target for new antibacterial therapies. Researchers hope that structural details of this mechanism, such as those revealed in this study, will help them find a way of stopping bacterial hair growth. This may offer a fundamentally new way out of the current crisis caused by drug resistant bacteria, which seem to be spreading ever faster.

The crucial difference is that drugs targeting a non-vital virulence factor such as the assembly of pili, will not kill the bacteria, which has two significant advantages. Firstly, there is less evolutionary pressure in favour of resistance genes, and the evolutionary pressure only applies in the location where the pili would normally anchor the bacteria, e.g. the urinary tract. For the survival of bacteria outside this location, the pili aren’t necessary, so any drugs leaked into the environment don’t breed resistance – in marked contrast to leaked antibiotics.

Secondly, the “disarmed” bacteria may remain present in the organism for long enough to allow our immune system to learn how to fight them most efficiently, such that they act as an immunization against the disease.

Some substances targeted at pilus formation are already under investigation as potential drugs, but the detailed molecular understanding of the process will certainly help to address this target more systematically.

Reference:

G. Phan et al, Nature 2011, 474, 49–53 doi:10.1038/nature10109


Background:

Gross M:
Education in Chemistry 2008, 45 No 4, 141-143
Better than antibiotics?

Groß M:
Nachrichten aus der Chemie 2008, 56, Nr 2, 148
Blickpunkt Biowissenschaften: Bakterien am Schopf gepackt

perforin pore in German

Following up on my Chemistry World news story on the perforin pore, I have done a longer article about this for Spektrum der Wissenschaft, which appears in this month's issue:

Zweischneidige Killerwaffe
Spektrum der Wissenschaft Nr. 1, S. 18

Beginning of the article and restricted access to pdf file

This is the only German piece published this month, so that's the January round-up done with.




(image courtesy of Helen Saibil, Birkbeck College London)

Reference:
R.H.P. Law et al. Nature 468, 447–451 (18 November 2010) doi:10.1038/nature09518

perforin pore unveiled

The group of Helen Saibil at Birkbeck College determined the shape of the perforin pore, which killer cells of our immune system use to eliminate cancerous or virus-infected cells, which suggested how the protein subunits could be arranged. In collaboration with researchers at Monash University (Melbourne, Australia) who solved the crystal structure of the individual protein blocks that assemble to form this pore, the group has now come to a conclusive model, which is “the wrong way round” compared to what they were expecting based on similar bacterial pores.
Read my news item in Chemistry World here.




(image courtesy of Helen Saibil, Birkbeck College London)

Reference:
R.H.P. Law et al. Nature online DOI: 10.1038/nature09518

BBK magazine

A while ago, I helped my friends at the school of crystallography at Birkbeck College prepare a press release on their latest research in assisted protein folding:

Protein folding in the womb

This piece has now been published (anonymously) in Birkbeck's magazine, BBK, which is accessible to all in a PDF file. My piece is on the bottom half of page 5 of the magazine (page 7 of the pdf file). The first piece is also about structural biology work from the Birkbeck crystallography department, by Gabriel Waksman.

protein in the womb

Scientists at Birkbeck College, London, have obtained the first images of a new-born protein in the “womb” of an enclosure that helps it find the proper structure.

Most of our genes serve the production of different proteins, which first emerge as a long string of building blocks, the amino acids. This string has to arrange itself into a well-defined pattern of coils, twists, and turns, so the protein can proceed to fulfill its biological function in the cell. For this process, known as folding, many proteins need helper proteins known as molecular chaperones. An important group of these helpers comes in the shape of a molecular barrel, which accommodates the new-born protein and releases it when it has attained its properly folded structure.

As the newborn moves and wriggles around in this barrel, researchers have so far been unable to get a clear picture of it. The group of Helen Saibil at the department of crystallography, at London’s Birkbeck College, in collaboration with the group of Saskia van der Vies at the Free University, Amsterdam, has now succeeded in imaging such a “newborn protein in the womb” by using a particularly large protein baby. The researchers studied the case of a virus that infects bacteria (a bacteriophage) and uses the bacterial barrel to help fold its own shell protein. This protein is too large for the barrel, such that the virus brings its own lid along so its protein can be securely locked up in the folding chamber.

This hijacking by a virus is rather unfortunate and fatal for the bacterium concerned, but it is a lucky break for structural biologists, as the bulky virus protein can’t move around as much as other proteins do. In fact, its position inside the barrel was sufficiently well defined such that Saibil’s team succeeded in taking its picture using low temperature electron microscopy to image the proteins in their natural state, and computer image processing to distinguish between complexes captured at different stages of binding and assembly.

Comparing their images to the existing pictures of the empty barrel, the researchers could gain valuable insights into how a protein is bound and released during the assisted folding process. They could also observe that the enclosure, like a womb, has to stretch to accommodate this relatively large protein baby.

------------------

This is a draft press release I wrote, regarding a paper that appeared in the current issue of Nature (1.1.), vol. 457, p. 107