IT was a heavenly setting for a medical conference about such a grim subject. The first international workshop on HIV persistence during drug therapy took place last December on the tropical island of St Martin in the West Indies.
At the end of the meeting, the two American members of the organising committee 鈥� Robert Gallo, the co-discoverer of HIV, and Roger Pomerantz of Thomas Jefferson University in Philadelphia, Pennsylvania 鈥� sat down for lunch at a local cafe. Pomerantz recalls that the air was filled with the little white butterflies for which the island is famous. 鈥淭here were thousands,鈥� he says. 鈥淚 doubt Bob even noticed.鈥�
Sipping his beer, Gallo had his mind on a more serious subject. The men had spent the past three days hearing about HIV鈥檚 ability to hide from even the strongest anti-retroviral medicines available. Pomerantz asked Gallo the question that had been on everyone鈥檚 lips. Did he think it would be possible to eradicate the virus from these hiding places? In other words, would a cure for HIV ever be found?
Advertisement
No, said Gallo, it was impossible. 鈥淚t was a cordial conversation,鈥� says Pomerantz. 鈥淎s cordial as you can be, when you really don鈥檛 agree.鈥� Because Pomerantz thinks the prospects for a cure are only getting better. Is he put off by one of the world鈥檚 most famous HIV researchers dismissing his idea? 鈥淣ot really,鈥� he says with a shrug. 鈥淚 have to admit it, I like to tilt at windmills.鈥�
The current dogma, as espoused by Gallo, is that there is no likelihood of a cure for HIV infection. Of course, we should be thankful that what was once a death sentence has been transformed into a chronic disease that can be kept in check by powerful cocktails of anti-retroviral drugs 鈥� at least for those who can afford them. But HIV is a devious enemy and can hide from those drugs in numerous ways 鈥� in some types of immune cell in the blood, in certain organs, and even cradled in human DNA. Conventional wisdom says it is not possible to completely eradicate the virus from all of its hiding places, so patients will have to keep on taking the tablets鈥or ever.
Pomerantz, and a few other researchers around the world, see things differently. They point out that much progress has been made in the last few years in understanding just how HIV hides from anti-retroviral medicines. New drugs are being designed specifically to hunt down the virus and destroy it, and there are even novel strategies to pry it out of our DNA. Many of those scientists discussed their findings at the St Martin conference, and some came away more confident than ever. 鈥淭he field has matured and we know more about the enemy,鈥� says Pomerantz. 鈥淭his isn鈥檛 trivial, it isn鈥檛 going to happen quickly, but we are making progress on all fronts.鈥�
When HIV was confirmed as the cause of AIDS 20 years ago, it wasn鈥檛 immediately obvious just how devious this virus was, though it would clearly be difficult to defeat. HIV invades and destroys CD4 T-cells, the very immune cells whose job it is to fight viruses. While this onslaught can sometimes be kept at bay for years, in nearly all untreated patients, the virus ultimately gains the upper hand, leading to acquired immune deficiency syndrome or AIDS. With their immune system in tatters, the patient is at the mercy of a host of infections.
The first weapon against HIV emerged in 1986 with the development of anti-retroviral drug zidovudine, also called AZT. Similar medicines followed. Then, in the early 1990s, came a potent new class of anti-retrovirals, the protease inhibitors. By the mid-1990s, the use of powerful combinations of drugs, known as HAART (highly active anti-retroviral therapy) led to a dramatic decline in the number of AIDS cases in some western countries.
Although HAART can make the number of virus particles in patients鈥� blood fall to undetectable levels, scientists knew there was a potential source of new virus. When HIV enters a cell, its genome integrates into the cell DNA. Most of the time it immediately starts replicating 鈥� the integrated genome acting as a template from which further copies scroll off at a blistering pace, spawning legions of new viruses. But sometimes, particularly in a type of immune cell called memory T-cells, the genome goes silent, switches off its genes and, in effect, waits. At a later date, perhaps when the memory T-cell is activated in response to an immune stimulus, the virus wakes from hibernation and starts replicating once more.
But in 1997, a team led by David Ho at the Aaron Diamond AIDS Research Center in New York made the bold claim that giving HAART for long enough could perhaps effect a cure for AIDS. Only months before, Ho had been canonised as Time magazine鈥檚 Man of the Year for leading the development of HAART, so his paper in Nature drew worldwide attention (vol 387, p 188). Although no drugs could attack latent virus in memory T-cells, Ho proposed a different sort of weapon: time. Assuming that the barrage of HAART drugs was stopping all virus replication, they estimated how long it would take for existing latent cells to die off naturally. To everyone鈥檚 delight, they calculated it should take only two to three years.
But the mood of optimism lasted less than a year before the facts began to prove Ho wrong. Some patients had been on HAART for longer than three years by then, and when they stopped taking the drugs the infection roared back to life. That suggested his team had underestimated the lifespan of latent cells. They seemed to last for ever. And, as far as most people were concerned, that was the end of hopes for a cure.
So why do researchers like Pomerantz, think it is time to start talking about the 鈥渃-word鈥� again? To begin with, it seems the initial confidence that HAART was completely squashing viral replication in standard CD4 T-cells was misplaced. When patients started therapy, levels of virus in their blood plummeted from hundreds of thousands of particles per millilitre to below the limit of detection, which at the time was about 50 particles per millilitre. But Pomerantz and his colleagues reported in 1999 that there was no shortage of new virus, if you looked for it with powerful enough tools. His team used more sensitive tests to examine 22 patients receiving HAART. In every one they found that the number of viral particles lurked just below the radar of conventional tests, with an average of 17 copies per millilitre.
The significance was clear. Because of some sort of 鈥渃ryptic replication鈥�, HAART was not laying siege to a dwindling number of enemy troops, it was having to tackle reinforcements all the time. Even with the best treatments available, the virus could still create new progeny, lay down ever greater stocks of latent virus in memory T-cells, evolve resistance to current drugs and continue its destruction of the immune system.
So where was the virus coming from? An important source has turned out to be immune cells in the blood itself. Anti-retroviral drugs were developed with a particular viral breeding ground in mind 鈥� actively dividing cells. Because many of these drugs cannot easily pass through the cell鈥檚 surface membrane, they are given in precursor forms. They depend on cell enzymes to be transformed into their active state, sometimes even requiring more than one activation step.
But CD4 T-cells can enter a resting phase where they don鈥檛 divide. Previously these cells were thought to harbour only latent integrated 搂virus. 鈥淵ou can鈥檛 establish a productive infection in resting T-cells, the theory went,鈥� says Ashley Haase of the University of Minnesota in Minneapolis. Now Haase and his team have blown this idea apart. They tested cells where HIV replication was occurring for the biochemical hallmarks of rapid division. To their surprise, 85 per cent of the cells did not appear to be dividing.
Haase suspects that resting T-cells have low levels of the enzymes needed to activate the anti-retrovirals, creating a drug-free paradise for viral replication. The virus probably reproduces more slowly than with active cells, he says, which explains why they hadn鈥檛 been identified as sources of HIV. 鈥淏ut to reignite an infection you don鈥檛 need a lot of virus,鈥� he says.
Other work by George Pavlakis at the National Institutes of Health in Bethesda, Maryland, suggests that even some actively dividing cells provide hideouts for viral replication. His work focuses on molecules called P-glycoproteins or PGPs, which span the cell鈥檚 surface membrane and act as molecular pumps to eliminate toxins from cells. Unfortunately that includes drugs, particularly protease inhibitors. T-cells and other cells that support HIV were thought to have low numbers of these pumps but Pavlakis鈥檚 work suggests that is dead wrong. His team has found that the number of pumps these cells possess varies widely. Some contain large numbers, making them impervious to normal levels of protease inhibitors.
The fact that standard HAART never achieved the goal of zero replication is, in a way, good news. It suggests that, rather than rethinking the whole approach to HIV, more effective anti-retrovirals may suffice. Pharmaceutical companies are busy developing whole new classes of drugs that penetrate cells more easily, have fewer activation steps or attack new stages of the viral lifecycle. Although these drugs are designed as alternatives for patients whose virus has become resistant to existing anti-retrovirals, they could also become components of future HAART regimens with the more ambitious goal of virus eradication.
Locked out
One promising new class of drugs, for example, are known as the entry inhibitors. Fuzeon, marketed by Roche, was approved last year, and others are on the way. These drugs act at the cell surface to stop HIV invading new cells, so they should avoid the problems of poor enzymatic activation or PGP clearing. David Ho and other researchers have told New 杏吧原创 they are planning to conduct clinical trials combining entry inhibitors with standard HAART to see if the combination can squash cryptic replication.
George Pavlakis is investigating a different strategy to knock out PGPs involving an existing anti-retroviral, the protease inhibitor ritonavir. This drug is already used to boost the potency of other protease inhibitors, apparently by slowing the rate at which the liver breaks down the other drugs. But it has recently been shown that ritonavir also binds to PGPs so strongly that it reduces their ability to pump out other drugs from cells. This suggests that using ritonavir alongside a second protease inhibitor is a way to target immune cells with many PGPs. Pavlakis is planning to carry out trials to assess whether this is a second reason ritonavir boosts the potency of other protease inhibitors. 鈥淪trategies to improve therapy must take this under serious consideration,鈥� he says.
Promising as these approaches sound, there are other sources of HIV persistence that lie outside the blood鈥檚 immune cells. Immune and microglial cells of the brain are one hideout of particular concern because of the relative impermeability of the blood vessels that perfuse the brain, the so-called blood-brain barrier. Because of this, some anti-retroviral drugs fail to reach effective levels in brain tissue, and some patients develop neurological problems known as HIV-associated dementia. The fact that this condition occurs even in patients with low blood levels of the virus has led to the brain being called a 鈥渟anctuary site鈥� for HIV.
But neuroscientists have been working on ways to get drugs past the blood-brain barrier for some time and in recent years have made progress on several fronts, raising the possibility that these strategies could be co-opted for HIV therapies. One promising technique is to put drugs inside microscopic spheres called liposomes. These are bound to antibodies that latch on to the endothelial cells that make up the brain鈥檚 blood vessels, enabling the liposomes鈥� contents to enter the brain (New 杏吧原创, 22 March 2003, p 16).
Another strategy may already be within our grasp. One reason for the impermeability of the brain鈥檚 blood vessels is that their endothelial cells have P-glycoprotein pumps. So a regimen of ritonavir plus a second protease inhibitor George Pavlakis plans to investigate may knock out HIV replication in the brain, as well as in the blood鈥檚 immune cells. 鈥淲e can try to fine-tune the cocktail to improve penetration to the brain,鈥� Pavlakis says.
The male genital tract is another sanctuary site for HIV. It is known that some patients on HAART have HIV in their semen despite having undetectable blood levels of virus. The exact location of the infected cells responsible is unclear, but for some years a group led by Angela Kashuba of the University of North Carolina at Chapel Hill has been studying which anti-retrovirals are best at reducing virus levels in semen. She has shown that some drugs are actually more concentrated in semen than in the blood, and surmises that the immune cells in the genitals must be activating the drugs less efficiently, perhaps because they lack the necessary enzymes.
No sanctuary
Kashuba鈥檚 team is using this data to design the best combination of anti-retrovirals for eliminating HIV from semen. Although their primary aim is to develop a HAART regimen that reduces sexual transmission of HIV, the researchers believe it could also play a role in future eradication therapy. One of Kashuba鈥檚 colleagues at North Carolina, Myron Cohen sees it like this: 鈥淐uring HIV was a big deal in the 1980s and disappeared as people became daunted by the challenge. Now no one likes to talk about it, but I don鈥檛 think we can avoid it.鈥�
The brain and male genitals may not be the body鈥檚 only sanctuary sites 鈥� the female genitals and the lymph nodes are among other possible ones. Given the progress being made in drugs for the genital tract, it is not a great stretch to hope that treatments can be developed to maximise their impact in different tissues, says Cohen. But few scientists are trying these strategies right now, because of the assumption a cure is impossible. 鈥淥ther people need to come to believe this is not impossible, and study how to eliminate the virus in different compartments,鈥� he says.
Assuming that the last remnants of viral replication can be laid to rest, another source of HIV has to be taken care of, which represents perhaps the biggest challenge of all: latent virus integrated into the genome of dormant memory T-cells. This time no one is satisfied with just waiting for the cells to die. A number of labs are now working on ways to destroy sleeping memory cells.
Many of the strategies under investigation use chemicals to activate the cells and induce them to divide, and then kill them. Two recent experiments along these lines have shown some promise. Using an animal model of HIV that involves transplanting human immune cells into mice, Dean Hamer and colleagues at the NIH found both the immune chemical messenger IL-7 and a plant-derived growth factor called prostratin can wake up memory cells. They also used a specially engineered toxin to locate and destroy cells that were stimulated by the treatment. Last September they reported that the combination destroyed 80 per cent of the memory cells (Immunity, vol 19, p 413). Hamer sees the results as cause for optimism. 鈥淚t鈥檚 worthwhile trying to get rid of these viral reservoirs and see what happens,鈥� he says. 鈥淚t takes a lot of steps, but that was also true for retroviral therapy.鈥�
The group led by Pomerantz has conducted a pilot study in three patients, combining highly potent HAART with antibodies designed to stimulate sleeping T-cells. In two of the patients there was significant change in the genetic profile of the latent HIV. 鈥淭hat says we can change that reservoir, perturb it,鈥� Pomerantz says. It also suggests that different compounds may activate viruses of different genetic backgrounds. So as with HAART, a combination of drugs might be necessary to destroy all latent virus. 鈥淲e are just starting to have a reasonable number of these compounds,鈥� says Pomerantz. 鈥淲e are where antiviral therapy was almost ten years ago.鈥�
Rousing HIV with the right mixture of drugs is clearly important. T-cells play a crucial role in the immune system, and activating too many at once 鈥� a process that usually kills T-cells 鈥� would leave the patient unable to respond to future infections. One way around this problem would be to wake up the virus directly, rather than its host T-cells.
The HIV genome stays sleeping because inhibitory proteins act as a blanket covering its promoter region, and so prevent RNA copies from being made, says David Margolis, of the University of Texas Southwestern. He has devised a way to yank off those proteins using hairpin-shaped molecules called polyamides (see 鈥淗ow to wake a sleeping virus鈥�). These bind to the promoter region in preference to the inhibitory proteins 鈥� but unlike the inhibitory proteins they do not prevent the genome from replicating (Journal of Virology, vol 76, p 12349).
Polyamides are probably years away from being turned into safe medications, but Margolis says the experiment proves in principle that you can activate latent HIV directly. More broadly, he says it shows that new ways to attack the virus are still emerging. 鈥淚t won鈥檛 be simple to develop these new classes of drugs but in theory they could be amazingly potent and safe,鈥� he says. 鈥淪ome day we might look back on AZT and think of it the way we think of arsenic now.鈥�
Despite these promising avenues, however, many HIV researchers remain sceptical about prospects for a cure. Robert Gallo, who struck the note of pessimism at the recent St Martin conference, thinks there are just too many ways for the virus to evade attack. 鈥淚 don鈥檛 see this happening in my lifetime,鈥� he says.
Even David Ho, who originally raised the prospect of a cure in 1997, talks more cautiously these days. Eradication is no longer a major thrust of his research, although he maintains it would be unwise to give up all hope. 鈥淚t鈥檚 wrong to be falsely optimistic and raise hopes,鈥� Ho says. 鈥淏ut it鈥檚 also wrong to be unduly pessimistic.鈥�
Pomerantz describes his own view as realistic. He envisages giving patients a supercharged version of HAART to eradicate all replicating virus in the blood, plus drugs that can reach sanctuary sites, as well as chemicals to activate and destroy HIV integrated into the genome. With all the recent work in this field, he thinks there is now a critical mass of dedicated labs.
鈥淭reatments are always easier than cures,鈥� he says. 鈥淢edical science has been working on a cure for heart disease for a hundred years and nobody says we should give up. Why should we give up on HIV?鈥�
