Gallery: A spotter’s guide to human viruses
WILL swine flu virus turn nasty as the northern hemisphere winter gets under way? All previous pandemic flu strains started off mild before becoming deadlier, so health authorities are taking the threat seriously. They know that if 2009 H1N1 flu does become more lethal over the next few months, we will be nearly defenceless: there are already signs of resistance to Tamiflu, and any vaccines will be in very short supply.
H1N1 flu is far from the only threat. A new pathogen could emerge at any time, as the SARS virus did in 2002, or a known virus such as that behind Lassa fever could become much better at passing from person to person and spread beyond Africa. Or a rogue scientist, or just a careless one, could release a deadly virus such as smallpox.
Advertisement
We have been relatively lucky so far. The nature of SARS allowed it to be contained, while H1N1 flu remains mild for now. But our luck could run out tomorrow. “Mother Nature is among the worst terrorists,” says Michael Goldblatt, who once led the biodefence programme for the Pentagon’s research arm, DARPA, and now heads , a biotech company in Gaithersburg, Maryland.
“If you look at the viruses that are the biggest threats of modern times, most of them were unknown through human history: HIV, SARS, Ebola. You don’t know where the next one is coming from. How do you develop therapeutics for the unknown and unknowable, given that you won’t have time to develop a vaccine for a new agent after it appears?” he asks.
Goldblatt and a few other researchers think they have the answer. They are working on an entirely new class of antiviral drugs that should do something seemingly impossible: work against a wide range of existing viruses and also be effective against viruses that have not even evolved yet. What’s more, it should be extremely difficult for any virus to become resistant to these drugs.
This might sound too good to be true, but the first trials of these drugs are already producing encouraging early results. If just a few of them live up to their promise in full-scale human trials – no sure thing – they will be a medical breakthrough on a par with the discovery of penicillin. At last, doctors will be able to treat viral diseases as ably as they do bacterial ones.
“If these drugs live up to their promise, it will be a breakthrough on a par with the development of penicillin”
The conventional strategy for developing antivirals is “one bug, one drug” – finding a drug that blocks viral replication by binding to part of a viral protein. The trouble is, any minor mutation that slightly changes the shape of the protein can render these drugs useless, as is happening with Tamiflu. The of dollars governments worldwide have spent stockpiling this drug could well turn out to be futile.
A few existing antiviral drugs, such as interferons, do work against a wide range of viruses. However, these drugs merely rev up the body’s immune system, which makes them less effective than doctors would like.
Back in the late 1990s, when Goldblatt was at DARPA, he began to wonder whether there was another strategy, one that exploits the key weakness of all viruses: their utter dependence on their hosts. By themselves, viruses are more helpless than newborn babies. They can replicate only by tricking their host cells into making more copies of them, a process that can involve hundreds of host proteins.
What if, Goldblatt wondered, some host proteins are essential for viral replication but not for the survival of the host? If so, disabling these proteins should block viral replication without killing healthy cells.
After moving to Functional Genetics, Goldblatt began putting his idea to the test. He and his colleagues disabled one gene at a time in human cells before exposing them to viruses such as flu. This fishing expedition worked beautifully: they identified more than 100 different human proteins that flu viruses need to replicate but which cells can survive without. Only four were previously known to be involved in viral replication.
One especially promising target is TSG101, a protein involved in the transport of materials within cells that many to break out of cells. Functional Genetics has developed a small-molecule drug that appears to block the interaction between viruses and TSG101. Dubbed , the drug inhibits a wide range of viruses in cell culture, including hepatitis C and HIV, and has also been shown to protect mice against Ebola ().
FGI-104 appears not to be too toxic. Yet it might not even be necessary to inhibit TSG101 to fight infections. When a virus uses TSG101 to escape a cell, some of the normally internal protein ends up exposed on the outside of the cell – a very distinct footprint of viral infection that Goldblatt’s team decided to target. They have designed an antibody, called FGI-101, that will bind to any exposed TSG101. As soon as viruses start to bud off from a cell, binds to the TSG101 and triggers the cell’s destruction. It might not completely prevent viral replication but it should greatly reduce it.
Crucially, because the viral footprint consists of a human rather than a viral protein, it is not specific to one virus. So far, the researchers have identified more than 30 viruses that rely on TSG101, from a wide range of virus families. The antibody binds to cells infected with all those they have tested, including flu, Ebola, HIV and herpes, and also reduces HIV below detectable levels in human cells in culture.
“Lo and behold, every time TSG101 is involved, the same antibody gets it,” says Goldblatt. “All of a sudden, what you have is one therapeutic which identifies any cell infected with a virus that uses this mechanism.” Functional Genetics plans to apply for permission to begin clinical trials of the antibody next year.
Gruesome death
Through random drug screening, the company has also identified another drug candidate, , with broad antiviral effects. How it works isn’t clear, but in experiments with mice its effects are dramatic. “If we give the drug one day post-Ebola-infection, we can save 100 per cent of the animals,” says Goldblatt. “If we wait two days post-infection, we can save 80 per cent. Up to day four, we can still save 40 per cent of the animals from an otherwise gruesome death at day seven. It’s pretty remarkable when you can have that kind of effect with one administration.”
Like the antibody, this drug also works against a number of viruses, including Marburg and parainfluenza, with no apparent serious side effects.
Other research teams are uncovering further viral vulnerabilities. For example, virology textbooks will tell you that the protein coat of a virus assembles spontaneously from the component proteins – which it does if the isolated proteins are placed in a test tube. Yet when Vishwanath Lingappa, then at the University of California, San Francisco, and colleagues added the components to a complex soup more like that found inside cells, other molecules got in the way. In reality, the team has shown, viral assembly depends on the presence of numerous host proteins.
Many of these proteins have no obvious link to viral assembly. For instance, one normally inhibits an enzyme that cuts up RNA (). “What does that have to do with assembling viral capsids?” says Lingappa. “But the virus comes in and says, ‘You – stop doing whatever you’re doing for the host, and work for me now.'”
These hijacked proteins could be a target for antiviral therapy, Lingappa realised, and he founded of San Francisco to commercialise the idea. His team has found that each family of viruses seems to rely on a different set of between 20 and 40 hijacked proteins for coat assembly, but that all members of a family exploit much the same set of proteins.
This means each of Prosetta’s drug candidates targets a smaller number of viruses than those of Functional Genetics, but they still cover a far wider range than conventional antivirals. One compound, for example, has been shown to be effective against six different flaviviruses in tests in cell cultures, including dengue fever, yellow fever, West Nile fever and hepatitis C. In all, the company is developing drugs targeting 14 of the 21 virus families known to cause human disease, and hopes to begin negotiating partnerships with big pharmaceutical companies later this year.
Many host proteins can be disabled without serious side effects, Lingappa is finding. “The virus, however – at least this is our current working hypothesis – is absolutely dependent on a particular enzyme,” he says. “If you stop it, the virus is out of luck. But the host can turn to other [proteins] and say, ‘Can you get by? Can you do 50 per cent of the work?'”
Eventually, Lingappa hopes to design combinations of drugs that slightly inhibit several proteins at once, instead of completely disabling a single protein, which might reduce side effects still further.
Perhaps the most promising broad-spectrum antiviral is being developed by , a pharmacologist at the University of Texas Southwestern Medical Center in Dallas. Like Goldblatt, Thorpe has discovered a “footprint” of viral infections.
In healthy cells, certain molecules are found only on the inner surface of the cell membrane. These molecules include a fatty substance called phosphatidylserine, one of the most common constituents of cell membranes. When cells are stressed – when infected by a virus, for instance – some phosphatidylserine ends up on the outer face of the membrane. “It’s clearly present on all the virally infected cells we’ve dealt with,” says Thorpe. “It stands out like a beacon.”
So, just as the FGI-101 antibody targets exposed TSG101, Thorpe and his colleagues have designed an antibody, now known as bavituximab, that binds to exposed phosphatidylserine. Once bavituximab binds to an infected cell, the immune system should quickly destroy it, limiting viral replication.
In addition, many common viruses wrap themselves in the cell membrane of their host cells as they bud off. This membrane “envelope” may act as an invisibility cloak, allowing a virus to hide from the immune system. Bavituximab, however, will bind to this cloak, revealing the viruses to the immune system. So, at least in theory, bavituximab should also trigger the destruction of enveloped viruses before they infect cells.
Thorpe’s team recently tested bavituximab on guinea pigs infected with Pichinde virus, the guinea-pig equivalent of Lassa fever virus. Half the animals treated with bavituximab survived, while all those injected with a control antibody died, the team reported last year (). “We were doing something that was quite extraordinary,” says Thorpe. “We had let the animals progress in their disease until they were heavily symptomatic, and then started treatment. It was a major accomplishment to knock that disease back.”
The company developing bavituximab, of Tustin, California, is now testing it against a variety of other viruses, including HIV, hepatitis C, influenza, measles and members of the smallpox and rabies virus families. So far, every virus they have checked has left a phosphatidylserine footprint. “We feel like there’s a good potential that they’ll all have it exposed,” says the head of Peregrine, Steven King. If he is right, drugs like bavituximab could help fight every known human virus.
Peregrine has already begun testing bavituximab against HIV and hepatitis C. Early results look promising, says Thorpe. The drug appears to reduce viral numbers without major side effects, but it is too early to tell whether extended use can clear the viral infections completely. More advanced clinical trials are also testing bavituximab as an anticancer agent, since phosphatidylserine also ends up on the surface of tumour cells. These trials, too, suggest that the drug appears safe. And similar treatments could target other components of the cell membrane that are exposed during infection.
It remains early days for host-targeted antiviral therapy, as this approach is known. Indeed, many experts are sceptical about the entire notion. “People in infectious disease are comfortable with targeting the pathogen. They’re not comfortable with targeting the host,” says Goldblatt. One reason is the higher risk of side effects, especially with infections that require lengthy treatment.
“Until we can get these things into humans and test them, it’s a little bit of a crapshoot as to whether they will work,” says Michael Kurilla, who coordinates biodefence research at the US National Institute of Allergy and Infectious Diseases in Bethesda, Maryland.
On the flip side, host-targeted antivirals offer some big advantages over conventional antivirals. First and foremost, they provide a way to cope with new viruses. They will not be ready in time to help us fight H1N1 flu, but the next time a new virus starts spreading among humans, we might not be as defenceless as we are now. In a decade or two, there could be a whole range of approved antivirals to pull off the shelf and try.
The other huge advantage of targeting the host is an evolutionary one. When an antiviral drug targets a viral protein, it produces intense selection pressure favouring any mutations that allow the virus to dodge the onslaught – and just tiny changes in a viral protein can be enough to achieve this.
When a drug targets the host, though, it is much harder for viruses to dodge the attack. It is not clear, for instance, how viruses could change to avoid leaving a phosphatidylserine footprint. And to evolve resistance to TSG101-targeting drugs, viruses would have to find another way to exit cells, a drastic change that would involve far more than one or two mutations. “I am sure that somehow, somewhere, pathogens will find a way to circumvent this,” says Goldblatt. “But it’s not obvious how easy that will be or how it will occur.”
Gallery: A spotter’s guide to human viruses
