Episode 153      29 min 45 sec
Mosquito bytes: Fighting malaria with computational science

Parasitologist Dr Stuart Ralph explains malaria and how computational modelling developed to better understand the disease will aid the global effort to eradicate it. With host Dr Dyani Lewis.

"We’re trying to use bioinformatic means to predict, of all of the proteins that are encoded in a genome, which ones are crucial, which ones can the parasite not do without." -- Dr Stuart Ralph




           



Dr Stuart Ralph
Dr Stuart Ralph

Stuart Ralph is an ARC Future Fellow in the Department of Biochemistry and Molecular Biology at the University of Melbourne. His laboratory researches parasitic diseases, with a primary focus on the causative agent of severe malaria, Plasmodium falciparum. The laboratory is interested in identifying and characterising promising drug targets from P. falciparum and other parasites, as well as studying the modes of action and mechanisms of resistance for existing drugs. Stuart completed his PhD at the University of Melbourne in the School of Botany, and worked as a postdoctoral researcher at the Pasteur Institute in France, then at the Walter and Eliza Hall Institute of Medical Research in Melbourne before starting a research group at the University of Melbourne. His research is supported by grant and fellowship funding from University of Melbourne, the ARC and NHMRC in Australia, as well as international collaborative grants.

Credits

Host: Dr Dyani Lewis
Producers: Kelvin Param, Eric van Bemmel
Audio Engineer: Gavin Nebauer
Voiceover: Dr Nerissa Hannink
Series Creators: Eric van Bemmel and Kelvin Param

View Tags  click a tag to find other episodes associated with it.

 Download mp3 (27.1 MB)

VOICEOVER 
Welcome to Up Close, the research, opinion and analysis podcast from the University of Melbourne, Australia.  

DYANI LEWIS
I’m Dyani Lewis, thanks for joining us.  Malaria has plagued human beings for thousands of years and, while the parasite that causes malaria has been eradicated in many of the wealthier regions of the world, it still infects and kills millions of people each year in poorer countries of Sub-Saharan Africa and the tropics.  Malaria is becoming increasingly resistant to current therapies and there is a constant race to identify new drugs to combat the ever-evolving parasite.  One new tool at our disposal for such drug discovery is bioinformatics.  With an abundance of biological data flowing from genome sequencing projects in recent years, bioinformatics has become and indispensable ally in making sense of the complexities of biological systems.  To tell us more about malaria and the use of bioinformatics in anti-malarial drug discovery we are joined by Dr Stuart Ralph, Senior Lecturer at the Department of Biochemistry and Molecular Biology and Research Fellow at the Bio21 Molecular Science and Biotechnology Institute here at the University of Melbourne, Australia.  Welcome to Up Close, Stuart.

STUART RALPH
Thanks very much, Dyani.

DYANI LEWIS 
Now, as I mentioned, malaria has been with us for millennia, but how big of a problem is it today?

STUART RALPH 
Malaria is still a massive problem.  It’s one of the most important infectious diseases in the world.  It kills still terrible numbers of people and causes enormous economic damage in the countries where malaria is a problem.  So in 2009 there were probably around about 800,000 people who died from complications of malaria and there were probably somewhere around 500 million people who were infected by the parasite that causes malaria, and about half of the people who live in the world live in regions where they’re at danger of contracting malaria.

DYANI LEWIS
So can you explain a bit about what the malaria infection actually looks like, what some of the symptoms are?

STUART RALPH
Malaria is a parasite that lives firstly in our liver and then it breaks out of our liver and infects the cells in our blood that carry oxygen.  So it divides and reproduces within those red blood cells and when it’s finished with them it blows them up. and so there are a number of symptoms that are related to the parasite chewing up those blood cells in our circulatory system and then releasing poisons into our circulatory system.  So some of those symptoms are related to inability to carry oxygen and low levels of iron in your blood, and others are related to the release of those poisons.  So one of the things that happens when those toxins or poisons get released, when the parasite blows up the cell that it’s finished chewing up, is that those poisons then cause quite a rapid and severe fever.  So one of the ways that we can recognise malaria is that people get periodic spikes of fever and those fever spikes are actually diagnostic even of the type of species that causes different sorts of malaria.  So the species of malaria that causes the most disease in the world, which is a parasite called Plasmodium falciparum, it chews up a red blood cell every 48 hours and then busts out to try and find a new one; that happens quite synchronously within our bloodstream so that people get a fever spike every two days.  Something that’s interesting about that fever spike is that because it’s so characteristic we can identify from historical records people who may have had malaria or countries where malaria was endemic based on that description of the two-daily fever spike.  People think that probably Alexander the Great died from Malaria and more recently it’s become quite clear that Tutankhamen had malaria.  It’s not clear whether or not that was what killed him, but people have been able to find evidence of either symptoms in dead bodies that would make us think that people have malaria or descriptions at the time of the symptoms that person had that make us think that that person had malaria and died of it.

DYANI LEWIS
So earlier you said that Plasmodium falciparum is the main cause of malaria in humans, but how many other species of Plasmodium are there?

STUART RALPH
Well, we know of at least five species of parasites that cause malaria in humans.  Four of those are species that we think live only in humans or predominantly in humans; so a human passes the parasite onto a mosquito and that mosquito then goes on to bite another person, so the cycle is within the human and mosquito species.  Then there’s a fifth species which is called Plasmodium knowlesi; that's been identified only in recent years and that's a parasite that crosses over between monkey populations to human populations, so presumably a mosquito bites a monkey and then transmits the disease to humans.  We’re not sure whether that species can continue for a long time within a human population or whether it needs to be transmitted via mosquitoes that have bitten monkeys.

DYANI LEWIS
Now, the idea of eradication has been around for quite a while, so how long have we been trying to really beat malaria and why have we not beaten it so far?

STUART RALPH
People have been aware of malaria for millennia - they haven't necessarily known what causes it, but they’ve had different ways of trying to combat it.  So one of the famous ways was to use preparation of a bark from a tree called the cinchona tree and that then led to the development in the 20th century of a drug based on that active compound in that bark called quinine and then the synthetic derivative of that drug called chloroquine.  So that was a drug that the Jesuits in South America used widely to treat malaria and then they brought those trees back to Europe and those trees were planted wherever the Europeans had colonies throughout the tropics.  So for many years that tree and preparations based on that bark were one of the most important ways of treating malaria.  Then during the Second World War when access to some of those compounds became restricted because of difficulties in shipping, the availability of industrially produced anti‑malarial drugs became more important.  So those are way of killing the parasite once the parasite’s already infected a person.  But we can think of other ways of controlling malaria.  One is to stop people from being bitten in general, and you could do that either by killing mosquitoes in a local area or potentially by stopping the mosquito from biting you.  So you could do that by sleeping in a house that has flyscreens, or sleeping in ideally under a bed net that's been impregnated with insecticides.

DYANI LEWIS
Okay, and what about vaccines?  Do we have a vaccine against malaria?

STUART RALPH
We’ve got vaccines that are in trial against malaria.  So one of the real difficulties with controlling malaria is that unlike some of the other massive infectious diseases, like polio or smallpox, there hasn't been any effective vaccine against malaria or against the parasite that causes malaria.  So the research field in malaria is to some extend split between people who are trying to make vaccines to control malaria and people who are trying to identify compounds they could turn into drugs and then those drugs could either be used to stop people from contracting malaria in the first place or to treat people who have malaria.

DYANI LEWIS
This is Up Close coming to you from the University of Melbourne, Australia.  I’m Dyani Lewis.  Our guest today is molecular biologist, Dr Stuart Ralph, and we’re talking about malaria and bioinformatic approaches to tackling this deadly infection.  So we can't yet prevent malaria with a vaccine but we can treat it with drugs, as you mentioned earlier, Stuart.  But even drugs have their limitations, don’t they?

STUART RALPH
Yeah, there are several major issues with treating infectious diseases with drugs.  One of those issues is that the parasite will inevitably become resistant to any drug that we use for a long period of time, and that's happened for nearly all of the drugs that we’ve ever used against the malaria parasite.  So if you were to now try and take that drug that I spoke about before - chloroquine, which was developed earlier in the 20th century - most parasites in most places around the world now are no longer killed by chloroquine.  So if you take the dose of chloroquine that would be recommended in the 1970s that would now no longer kill parasites.  If you were trying to take that drug as a prophylactic measure, so say a traveller taking chloroquine, that would probably not protect them and nor would chloroquine be effective in treating a person who had malaria.

DYANI LEWIS
So how quickly does this resistance build up once you introduce a new drug?

STUART RALPH
With chloroquine - a really good drug in many ways - it seemed like it was really hard for the parasite to acquire resistance and it decades of intensive use before resistance initially arose and then spread throughout the world.  But with some of the drugs that have been introduced, subsequent, resistance has arisen and then become quite widespread within only a few years of their use.  So that’s really disheartening that you would bring out a new drug and then quite rapidly that drug becomes useless.  You can imagine that it costs millions of dollars, potentially a billion dollars, to develop that drug and a lot of people’s hard work and then, all of a sudden, now it’s next to useless.

DYANI LEWIS
So is there any way of preventing that resistance form developing?

STUART RALPH
We think we probably can never rule out resistance developing but there are strategies that we could use to slow down resistance and to stop resistance from spreading.  One of the ways that’s commonly used when we treat infectious diseases is to use what’s called combination therapy; so instead of just using one drug at a time, we use two or more drugs at the same time.  Therefore we’re asking the parasites or the microbes to become resistant to multiple drugs at once, and that’s much harder for them to do and hopefully that slows down the introduction and the spread of resistance.  The other thing that that would do is if I take, say, a chloroquine based drug and combine that with a second drug, it means that now I’ve got a better chance of killing more parasites that could infect me.  So I won't necessarily kill the parasites that are chloroquine-resistant but I will be able to kill the parasites that are resistant to that second drug that I’m using.

DYANI LEWIS
Now, Stuart, your work takes a targeted approach to identifying the most promising anti-malarial drugs to develop.  But before we talk about that, I was going to ask you whether you could explain how new drugs have traditionally been discovered.

STUART RALPH
Yeah, so until the last few decades people tried to identify drugs, and in many cases did that very successfully, by screening compounds that they had extracted from a plant that was used by some sort of community to traditionally treat malaria.  I talked before about the cinchona tree and how its bark was used a traditional remedy and then we looked at what the active compound was and then derived a very effective drug.  Another thing that we could do is to just take hundreds of thousands of compounds and then screen them for their ability to kill parasites.  Fortunately, in the case of Plasmodium falciparum we can grow that parasite in culture.  So in my laboratory we get donated blood from the Red Cross and the Red Cross gives us blood that they can't use for transfusions.  We can use that blood to grow parasite cultures of Plasmodium falciparum and we can then tip a bucket of some sort of poison in there and see which of those poisons kill the parasite.  So we can do that with very, very small buckets indeed and we can potentially screen hundreds or thousands of compounds, or in a more industrial lab you could imagine that you could screen millions of compounds at once.  So that’s a physiological based method for identifying drugs.

DYANI LEWIS
But I guess that’s fairly both time and labour intensive, that process?

STUART RALPH
It is, although with robotics that's probably not a person who’s testing all of those things, it’s a robot that’s dispensing all of the small amounts and robots that are then reading the results so that they can see which drugs are actually working.  But there are some drawbacks to that approach.  One of the drawbacks is that we don’t necessarily know how the compound works just because we’ve identified that it kills.  So we can find a compound and we can note that it kills parasites very effectively but we’d really like to know how it’s killing the parasite.  One of the reasons that we’d like to know why it’s killing the parasite is that reason I talked about before, that we would like to be able to understand resistance mechanisms.  The other reason is a slightly more pragmatic reason of getting drugs approved.  So in many countries if we can define how the drug works then the regulatory bodies are more likely to approve that drug for use.

DYANI LEWIS
So, Stuart, let’s talk about bioinformatics then.  You use bioinformatics to determine the best possible targets for drugs that might target malaria, so how does this approach work?

STUART RALPH
The process that we’re using is to try and take advantage of the biological information that's been revealed by the genome projects for these organisms.  As most people will be aware, since the mid 1990s an explosion of genome sequencing projects. Many people will be familiar with the human genome project but there have also been genome projects performed on the microbes that cause many human diseases.  In 2002 or 2003, I can't remember, the full genome of the parasite that causes human malaria was sequenced and so now we know all of the genes in that genome and we think we know pretty much all of the proteins that that genome encodes.  So genes encode for proteins - and there are about 5000 genes that encode proteins in the malaria parasite - so we would consider all of those proteins to be starting points for target identification now.  So we would now go through all of those potential starting points and use computational methods to try and identify which of those protein targets are most likely to be the sorts of proteins that might interact with drugs in a way that would bring about death of the parasite. 

DYANI LEWIS
So what sort of factors are you looking for that might determine whether a particular enzyme made by the malaria parasite will be a good drug target?

STUART RALPH
Well, one of the things that we would be interested in is whether or not that protein is actually a protein that's present in a disease-causing stage.  So when we think about malaria, the parasite that causes malaria, as we alluded to before, has stages in the mosquito and stages in the human.  So if you're thinking about a drug that you would give to the human rather than to the mosquito then you would want to be looking for proteins that are important and active in the human stages of the parasite’s lifecycle.  A second thing that we are interested is in identifying genes that are essential; so there are some genes in every genome that are dispensable and that means that you can get rid of that gene, either by deleting the gene from the genome or by somehow chemically inhibiting that gene or the product of that gene, and in many cases that has absolutely no effect on the organism.  So we’re trying to use bioinformatic means to predict, of all of the proteins that are encoded in a genome, which ones are crucial, which ones can the parasite not do without, and we think that only those proteins would make very good targets.

DYANI LEWIS
You're essentially whittling down all of this information to have a priority list of drug targets, is that right?

STUART RALPH
Exactly, and then one thing that we’re quite interested in is trying to find targets that are different in the parasite to their versions in the human.  So what we wouldn't want to happen is that we would try and inhibit a protein that’s exactly the same as a protein that’s found in the human genome.  We would hope that we wouldn't be using a drug that killed the parasite but also kills the person by inhibiting that same protein.  So we can use bioinformatics to identify the proteins in the genome of the parasite that are very different from the proteins in the genome of the host, which is us; so comparative genomics - comparing the genomes of humans and of the malaria causing parasite - to find the targets that are very different between those organisms.

DYANI LEWIS
What would exclude a gene from your list of potential targets?  Would it be a percentage of similarity to the Plasmodium gene equivalent? 

STUART RALPH
One way would be to make an arbitrary cut-off, a percentage similarity.  So proteins have a number of residues; an average protein might have, for example, 500 building blocks, so we could set an arbitrary score and we could say, well, no more than 200 of those building blocks - which are called amino acids - can be identical between the human version and the parasite version.  Those sorts of arbitrary cut-offs can be useful in reducing a larger list to a smaller list, but there are potentially more sophisticated ways of doing that.  When it comes to drug interactions, the thing that's probably most important is not just the similarity of those text strings - so we can imagine those amino acid sequences, proteins, as text strings - that’s not so important as the three dimensional shape that those text strings fold into.  So proteins are three dimensional objects. And if we could blow them up then we would see little mountains and valleys and rivers and clefts and canyons.  When it comes to drug discovery, one of the things that we often try and do is to find a little canyon or a cleft or a pocket and we try and find a drug that would fit into that pocket.  So when we try and find drugs that inhibit a parasite enzyme but not the human enzyme, we could make a three dimensional map of how that pocket looks in the human version and how that pocket looks in the parasite version and say, yeah, I can see a way of getting a drug that’s going to fit into the parasite version but wouldn't fit into the human version.  So that sort of three dimensional mapping is actually quite computationally intensive to perform and up until the late 1990s, anyway we didn't have the computational power to be able to do high throughput analysis of three dimensional structures of proteins.  That's now becoming more and more available.  In 2010 Melbourne University established a new collaboration with IBM to build a new supercomputer and one of the things that we are now using that supercomputer to do is to look at the three dimensional structure of human enzymes and compare those to parasite enzymes so that we can identify which of the proteins that are encoded by the genome are proteins that would interact with a drug-like molecule, and which ones are proteins that are dissimilar enough from their human counterparts that we could inhibit them in a selective way without making the human sick.  Of course, these things are just computational predictions, and one thing that is a very valid criticism of this sort of computational work is that it’s all meaningless until you actually test it and show that it works.  We try to straddle the lab work, where we come in with our white coats and beakers and bad haircuts and testings in the lab, and combine that with sitting behind a laptop or a supercomputer.  So what we’re trying to do is to make predictions about which proteins would make good targets and then we actually try and verify those in the lab, so that we show that those proteins are indeed inhibited by such and such a compound and that indeed those compounds don’t kill human cells.

DYANI LEWIS
I’m Dyani Lewis and my guest today is molecular biologist, Dr Stuart Ralph.  We’re talking about malaria and bioinformatics on Up Close, coming to you from the University of Melbourne, Australia.  So how about from the drug perspective, Stuart, can you also shortlist possible drugs and what sort of characteristics would you be looking for?

STUART RALPH
So that’s a field that’s called chemoinformatics and there are people in that field who certainly are trying to go through all of the possible chemical compounds that you could make and then derive rules about which ones of those are plausible as drugs and which ones could you never imagine making into drugs.  One very simple rule would be that if a compound was extremely large it would be maybe expensive to make that compound and also maybe easy to break that compound down.  So one of the things that we look at when we try and find compounds that are drug-like compounds is finding compounds that are under a certain size.  Then there are also things like how well that compound is likely to be able to cross the membranes in the human body and how likely it is to be excreted in urine by the person.  So some of the characters that lead to those properties we can identify, bioinformatically or chemoinformatically, the properties that would make those compounds undesirable.  But that’s still a field that’s in its infancy.

DYANI LEWIS
Then how do the pharmaceutical companies fit into this picture, because they’re traditionally been fairly reluctant to spend a lot of money on research into malaria and they’re also fairly reluctant to share data?

STUART RALPH
It’s true that there has historically been a lack of interest in anti-malarial drug discovery.  When we think about the kind of people who get malaria, unfortunately people living in disease-endemic countries, those countries that are often - nearly invariably - much poorer than the developed countries which have managed to eradicate or control malaria.  So if you think about your target market for a particular drug being people who have very little disposable income then that’s quite an unattractive market to pursue.  So that factor has meant that diseases like malaria - and malaria’s by no means the only disease that fits into that category - have been neglected by big pharmaceutical companies who are trying to make a profit for their shareholders.  In fact, malaria is reasonably lucky that it does at least have a high profile and it has enjoyed research support from some pharmaceutical companies and also from some military research organisations.  The other thing about malaria is that people are very aware of malaria when they travel and they’re eager not to contract malaria, which is of course a good idea, and so there’s a market for malaria in prophylactic drugs.  So when you or I want to travel to a part of the world which has malaria parasites circulating then we’d be wise to take drugs that protect us from those parasites and we would be willing, of course, to pay some money for that and so there’s some commercial market there in the prophylactic market against malaria. 

DYANI LEWIS
Do these pharmaceutical companies share their information readily with the research community? 

STUART RALPH
In recent years they have been participating more in collaborative research aimed at finding anti‑malarial compounds as well as compounds to treat other neglected tropical diseases.  So they’ve been doing that in a number of ways; one is to provide access to their proprietary compounds.  So a lot of large pharmaceutical companies will have compound libraries - big catalogues of these chemicals that I talked about before - and in some cases they have tested those against a very large number of diseases.  So for all of their million compounds, for example, they’ve tested whether or not it kills the bacteria that causes tuberculosis and the parasite that causes malaria and whether or not it kills tumour cells and whether or not it makes you feel better if you've got the flu.  So some of that data in recent years has now been shared with the public academic research community and that’s been really helpful at identifying compounds that could serve as leads for drug development.  Some of those organisations like Novartis and GSK have made concerted efforts to specifically test all of their useful compounds against the parasites that cause malaria and to then make that data available to researchers so that we could identify leads for drug discovery.

DYANI LEWIS
So with all of these more recent changes then, how optimistic can we be about our prospects of overcoming malaria some day in the future?

STUART RALPH
I don't know about that, Dyani.  There’s been optimism in the world before about our ability to control or eradicate malaria and that's proved to be unfounded, unfortunately.  In the late 1940s when we had some good drugs against malaria and we’d recently discovered insecticides like DDT there was then optimism that, using those tools, people would be able to eradicate malaria from the entire world within a number of decades.  For a little while that actually looked very promising, but for a number of reasons that then didn't lead to eradication and so a lot of people were really very disillusioned by the promise of governments and scientists and research bodies and that led then to several decades of pessimism about our ability to control or eradicate malaria.  In recent years there’s been some renewed optimism based on a number of factors.  One is that the number of people being infected, or at least the number of people dying from malaria has started to go down over the last decade.  It’s clear that malaria is on the decline in many parts of the world, so that leads to some hope for our ability to control or eradicate malaria.  The other reasons that we might be hopeful is that there are finally appearing - not on the market yet but perhaps on the market within the next few years, so in 2012 it’s possible that there will be the first vaccine registered for use against malaria.  Also there’s been now a recognition that there is a large amount of money needed to solve this problem and that money has been committed by national governments and also by philanthropists like Bill and Melinda Gates, who have put in the money that’s necessary both to perform the research to develop vaccines and drugs, and also that if those vaccines and drugs were to be developed that those philanthropists will pay for those vaccines and drugs, or governments will pay for those vaccines and drugs to be distributed to the people who need them but who can't afford to pay market prices for them.

DYANI LEWIS
Stuart, thank you for being our guest on Up Close today.

STUART RALPH
It’s a pleasure, Dyani.  Thank you.

DYANI LEWIS 
That was Dr Stuart Ralph, Senior Lecturer at the Department of Biochemistry and Molecular Biology and Research Fellow at the Bio21 Molecular Science and Biotechnology Institute here at the University of Melbourne, Australia, giving us an understanding of research into bioinformatic approach to overcoming malaria.  Relevant links, a full transcript and more info on this episode can be found at our website at upclose.unimelb.edu.au.  Up Close is a production of the University of Melbourne Australia.  This episode was recorded on Thursday 21 July 2011.  Our producers for this episode were Kelvin Param and Eric van Bemmel.  Audio engineering by Gavin Nebauer.  Up Close is created by Eric van Bemmel and Kelvin Param.  I’m Dyani Lewis, until next time, goodbye.

VOICEOVER
You've been listening to Up Close.  For more information visit upclose.unimelb.edu.au.  Copyright 2011, the University of Melbourne.


show transcript | print transcript | download pdf