AUSTRALIAN FRONTIERS OF SCIENCE, 2005

Walter and Eliza Hall Institute of Medical Research, Melbourne, 12-13 April

Session 5: Discussion

Question – Have you used conditional mutants – say, temperature sensitive mutants – to try and look at that question of the connection between replication and cell division, et cetera?

Liz Harry – Yes. At the moment we are looking at DNA replication mutants, because we have seen an effect on the positioning. In terms of the whole Z ring positioning area, there are other proteins that affect Z ring positioning. One of them is the Min system, which tends to stop Z rings forming at the poles. That was identified by mutations 30 years ago, and we now know that it stops Z rings forming at the centre, and we have shown that you don’t need it for the precise localisation for midcell but you need it to stop Z rings at other places.

And the chromosome has some inhibitory effect on Z ring formation. What the spirals tell us is that the nucleoid itself or the chromosome doesn’t prevent Z polymerisation per se, but it prevents that transition. One person in Oxford has finally identified a protein that is involved in nucleoid occlusion.

It hasn’t been hugely successful to look for mutations that affect Z rings in other areas, but we do have an FtsZ mutation, the only one known for Bacillus subtilis, which we think stops the spiral going into a ring, because you get a short spiral between two replicated chromosomes. So we are hoping that that will give us some information on how Z ring finds the middle.

Does that answer your question?

Question (continued) – Oh, it is just that you have moved back and forth between a permissive condition and a non-permissive position [inaudible].

Liz Harry – Yes. With this FtsZ mutant it reverses to a ring, so it is very nice to see.

Question – Brett, you talked about not knowing what the function of the toxins was in the cyanobacteria. Presumably, from an ecological perspective, one of the advantages of being toxic is that it stops you from being eaten. Can you comment on that in relation to your work?

Brett Neilan – Yes. That’s the theory that has been around longest, that the reason algae produce toxins is to stop daphnia, water fleas, rotifers, those sort of little things that eat algae. But we have done tests in which, having found that gene, we knocked the gene out and then fed the algae to the rotifers, which still died. So the tests are saying that that toxin is not there to kill ‘grazers’.

The other bit of evidence behind that is that these things evolved, and they evolved their toxin production mechanisms, before those types of eukaryotes came about anyway. It probably is why the dinosaurs became extinct!

Question – I have two questions, one for each of you. On the secret to success of bacteria: there has been a revolution on genome sequencing. I was wondering how much the bacterial genome information, including that of B. subtilis, has helped you along the path of revealing the secrets.

Also, for your type gene cluster for microcystis, is that mirrored by a multi-sub-unit complex as well? Are the proteins all clustered together, or are they just in all different parts? Or is it single polypeptide chain? Cyclosporin synthetase, I think, is a single polypeptide chain.

Liz Harry – Bacillus was the first big genome to be sequenced, and they have now gone through and deleted all genes separately. I think you learn a lot. Sometimes, I think, you don’t know what you are going to learn. Some of the things are obvious, but it is very philosophical. They find that 7 per cent of those genes are essential: if you just delete all of the genes separately and put them onto rich agar, 7 per cent of them are essential. What that means is that you don’t need 93 per cent of them if you grow them on rich agar.

That is informative, it is useful, but it feeds the next question: how many do they need? How many genes do cells need to infect? How many do they need to cause problems? How many do they need to survive in the wild? And that’s a very different question.

In doing things like proteomics, interaction – doing global stuff – is very useful. People come up to me and say, ‘I want to do proteomics but I can’t identify anything when I’ve done it because we don’t know any – or many – of the genes for this organism.’ Yes, global interaction is a big help.

It is also helpful to be able to look at groups of bacteria where you know certain clusters of genes are present. FtsZ is the most highly conserved – it is present in microplasma and microbacteria and all these ones that don’t have cell walls – and that probably tells you immediately that some of the other proteins that are not present in wall-less organisms are there to make the wall, help the cell divide. And the more bacteria you have to say that, the more it becomes a very obvious rule.

So I think it is really useful, but again it often gives you more questions.

Brett Neilan – I was asked whether the proteins are clustered together. Yes, they’re all associated. So there are enzymes all together, and we don’t know how they are kept together but probably on internal photosynthetic membranes, in this system anyway, possibly with chaperones – but yes, like a megadalton of protein all sitting together, making a little peptide.

Question – Liz, there are a lot of different organisms, apart from Bacillus. There are ones that have linear chromosomes, there are even some bacteria with nuclear membranes, [inaudible] membranes. Do you think the same process operates here?

Liz Harry – I don’t know. I’m sure that when we solve cell division for bacteria and for humans there will be some analogies or homologies – which they will be, is the question. But I am pretty sure that when we look for antibiotic targets we will find chemotherapeutic agents as well, because there are some proteins in Bacillus that are present in humans, doing cell cycle stuff.

It is a hard question to answer. How do we ever know how different or similar something is, until we actually understand it? When you solve something, you end up saying, ‘Oh yes, all you have to do is tweak this and then everything looks different.’ So it is a really philosophical question. Certainly there will be some similarities, and there will be some differences.

Brett Neilan – Considering we know only half of 1 per cent of the bacteria on the planet, and Bacillus is one of that half per cent, then there are probably a lot more cell division mechanisms to be discovered.

Question – Brett, I liked your really elegant environmental solution to the river bank issue, with growing trees and all the rest to drop the light. What is the solution for open oceans and freshwater dams with all the cyanobacteria?

Brett Neilan – In the dams, the problem cyanobacteria are buoyant; they have gas vacuoles and they come up to the light to make their sugar. And they get away from the light when they have made enough sugar. The wind tends to push the cyanobacteria to one edge of the water system anyway, so it is all impacted onto a bank. So trees around a river or around a lake, first of all – as I said – stop cattle from accessing from any point and defecating wherever they want. (That eutrophication is a big problem.) Trees on that riparian zone filter out superphosphates and nitrates in fertilisers, but also have the remedial effect of not letting the organisms be exposed to levels where they will pump out their toxins. Simple sand filtration will then get rid of the organism – something that engineers in Australia learned from the indigenous people, who used to dig a trench a few metres inland from a river bank so that water would flow through and then the native people could drink that water. They knew that you’d feel sick or even die if you drank blue-green algae.

The only problems with oceans are those red tides that I mentioned. We have never really studied the open ocean to find out whether there are toxic blooms there. You usually find them in coastal zones.

Question (continued) – Yes, but there are no river banks there. And 68 microeinsteins is not a lot of light.

Question – I know that in Western Australia and South Australia they grow algae for beta carotene production. Can you give an idea of just what the technology is for large-scale production of algae and manipulation of the products? They are very complicated organic molecules. You may not be able to touch on the genetics as much as the metabolism. If you wanted to make the variants that you are talking about, and the organic chemists couldn’t do it for you, how could you produce it?

Brett Neilan – That is in the slides I didn’t get onto. That is a system of gene clustering with modular enzymes. The peptide synthetases and the polyketide synthetases are all modular. They either incorporate one amino acid, for peptide, or one acyl group, for polyketide. But then you have tailoring enzymes. So in the rearrangement of those, you could remove an alanine and replace it with a serine module – each module like a tRNA is analogous to selecting which amino acid it is. You can remove the resistance that organisms have to vancomycin, for instance, by very slightly changing its structure. I wouldn’t say that that is simple, because they are huge gene clusters, but it can be done by genetic engineering and then growing up that organism in a new host and purifying the new type of what are being called unnatural natural products.

One of my slides indicates that 300 different amino and hydroxy acids are possible. If you were making an undecapeptide like cyclosporin, for instance, which has a lot of bad side effects on the kidney, you could modify the structure to make one with less side effects but still immunosuppressive, and you have 30011 permutations of what could be made. (They might not all work.)