Nearly 30 years after the first documented cases of AIDS, we still don’t know exactly how HIV destroys the immune system. But inflammation – sustained immune activation – is now seen as a key factor in how it wreaks its damage. Investigation into this is revising the way the virus is understood – and how it might be treated. Derek Thaczuk reports.
The end results of HIV infection have been clear since the epidemic began: left untreated, the virus eventually causes a massive loss of CD4 cells, pivotal players in the body's immune defences (see glossary). If CD4 cell counts fall low enough, the body becomes prey to opportunistic infections and cancers which the previously healthy immune system could defeat.
Perhaps surprisingly, though, we still don't completely understand how HIV depletes CD4 cells. Also, while antiretroviral treatment has allowed HIV-positive people to maintain healthy CD4 counts, all but vanquishing life-threatening opportunistic infections like cytomegalovirus (CMV) and pneumocystis pneumonia (PCP), metabolic problems such as cardiovascular and kidney disease remain widespread. The toxicities of antiretroviral treatment – such as increased cholesterol levels – do not fully explain such complications: HIV infection itself is now understood to significantly raise metabolic risks.
Several emerging concepts may shed light on these questions. Inflammation – the prolonged state of immune activation resulting from the immune system's ongoing battle with the virus – appears to be a key factor in metabolic disorders and cardiovascular disease. Research is also finding that the digestive tract may play a much larger role in HIV disease progression than previously realised, and in fact may be one of the sources of immune activation.
Early infection and the gut
The course of HIV infection follows a largely characteristic pattern in most people. During the first few weeks – acute infection – the immune system has not yet learned to respond to the new intruder. HIV levels are high throughout the body, and the number of CD4 cells in the blood plasma sharply drops.
Evidence now suggests that, by only looking at CD4 cells in the blood, we may have underestimated the overall extent of this early drop. Only a small fraction (2%) of the body's CD4 cells are actually found in circulating blood. Most live in lymph nodes (these include the ‘glands’ you can sometimes feel in the neck and groin when you have an infection), in the gut-associated lymphoid tissue (GALT), where they are present as patches of immune cells lining the length of the gut, and in the mucous membranes lining other organs exposed to foreign substances, such as the lungs and the genitals. Researchers have observed a massive loss of CD4 memory cells in this gut tissue very early after infection.1
Danny Douek, a researcher at the US National Institute of Allergy and Infectious Diseases (NIAID), has studied the process closely: "Once we thought that CD4 cells were lost slowly but surely over the course of the disease. But we are now seeing that most of the memory T-cell pool - which is most of the CD4 cell pool in an adult person - is lost extremely rapidly." Roughly 60% of memory cells may become infected, and the majority of those may disappear within the first two weeks of infection.
Ideally, HIV treatment may need to guard against both immune deficiency and stimulation. This is likely to be a complex goal, and the consensus is that considerable research is still needed.
Besides stripping the tissue of so many CD4 cells, HIV also causes structural damage to the gut immune tissue and to the lymph nodes where many immune cells normally reside. Recent studies have found that these tissues become scarred with collagen tissue during acute infection.2 Researchers speculate that this damage interferes with normal cell growth and interaction, limiting the immune system's ability to fully regenerate the CD4 cells lost in early infection. Gut tissue damage may also contribute to the inflammation that helps drive the later stages of HIV disease – a point we'll return to.3
Chronic infection: why do CD4 cells die?
After the intense few weeks of acute infection, the body begins to produce antibodies and immune cells that specifically target HIV. During this period (known as seroconversion), viral load levels drop and the CD4 cell count returns to near-normal levels. At this point, the disease enters a prolonged phase known as chronic infection.
In the early years of the epidemic, the virus was even thought to lie dormant during the lengthy period of chronic infection. This proved completely wrong: the advent of viral load testing in the mid-1990s proved that the virus continues to actively infect CD4 and other cells from the moment of infection onward, producing millions of new copies every day.
Is the virus directly killing off CD4 cells? It's easy to assume that must be the main reason for the eventual drop in CD4 counts. The truth, however, is more complex. Considerably less than 1% of circulating CD4 cells are actually HIV-infected during chronic infection – far too few to explain the overall loss – and millions of new CD4 cells are created every day. In recent years, researchers have uncovered other possible means by which HIV leads to loss of CD4 cells. These include toxic viral proteins, spewed out by infected cells, which can kill off uninfected cells in a so-called ‘bystander effect’. HIV can also trigger cells into ‘committing suicide’ in a process called apoptosis, or programmed cell death.4
Other mechanisms are likely to be at work as well, including – ironically – the immune system's own response to HIV. The virus can only infect activated CD4 cells – those that have been ‘switched on’ to fight against infection. In other words, by the very act of going into action against the virus, CD4 cells make themselves targets for it. This paradox is unavoidable to a certain degree, since immune cell activation is an essential part of immune function. However, there is growing evidence that prolonged and excessive immune activation – inflammation – underlies much of the ongoing damage of HIV disease.5
The idea that inflammation plays a major role in HIV disease was first proposed in the late 1980s,6 but has taken centre stage only recently. One of the first major clues came from the SMART study. This large-scale trial investigated whether people who remained on continuous antiretroviral therapy fared better or worse than those who took structured treatment interruptions – stopping treatment when their CD4 counts climbed above 350 cells/mm3 and resuming it when their counts fell below 250 cells/mm3.
The SMART study was halted early after interim results clearly showed that people interrupting their treatment were over twice as likely to become seriously ill or die. Tellingly, treatment interrupters were not just at risk of ‘traditional’ opportunistic infections. They also had higher rates of heart, liver, and kidney diseases – metabolic problems that are often associated with inflammation. If HIV was driving up levels of immune activation, we would expect to see more inflammation-related diseases in people whose HIV was allowed to replicate – just as was seen in SMART.
Further studies have confirmed that immune activation is actually a very good way to predict how fast HIV disease is progressing. People with higher blood levels of a substance called C-reactive protein (CRP) – known to be a sign of immune activation – progress to worse stages of AIDS more quickly than those with low levels. (CRP is, in fact, a much better predictor of progression than HIV viral load.)7
Why, then, does immune activation persist after treatment, instead of falling to near-normal levels when HIV replication has been controlled by antiretroviral treatment? Thus far, this is one of the most speculative areas of the hypothesis. Yet many researchers are convinced that the answer lies back where we began – in the infected tissues of the digestive tract.
Back to the gut
The lymphoid tissue in the gut keeps watch on microbes in the digestive tract – whether disease-causing organisms from contaminated food or water, or the ‘friendly’ bacteria that colonise the gut and aid digestion, mounting responses that keep the microbes out of the bloodstream. As discussed earlier, the lining of the gut can sustain lasting damage early on during HIV infection, becoming permeable or ‘leaky’.8
Danny Douek explains: "The outer wall of most bacteria in the gut contains what's known as endotoxin, or lipopolysaccharide (LPS). LPS is extremely immunostimulatory. In people with sepsis or toxic shock, you see an overwhelming immune activation due to huge amounts of LPS in their systems. In people with HIV infection, we have found LPS in the bloodstream – not in the same amounts as in sepsis, but enough to activate immune cells. We have also measured elevated levels of other bacterial products, all of which are immune activators, in the bloodstreams of people with HIV infection."
This hypothesis, known as microbial translocation, is currently one of the leading explanations for the persistent immune activation seen in HIV infection.9 However, many researchers suspect that immune activation has many causes. "I'm not convinced that microbial translocation from the gut is the sole answer to HIV-related inflammation," says Robin Weiss, professor of viral oncology at University College London. "We also see sustained immune activation in malaria, and nobody is proposing gut microbes as the source of that." Other candidates for drivers of HIV progression may include the immune stimulation caused by other infections, and the depletion or disabling of regulatory T-cells, which play a key role in cooling down immune activation. Immune cells also produce a variety of ‘messenger chemicals’ known as cytokines, which alert other cells to adjust their immune activity. HIV may confuse this immune communication network by disrupting cytokine production.10
As well as disentangling these complex processes, researchers must also investigate one of the biggest remaining questions: why HIV-positive humans are not able to correct excess immune activation, as they do with other chronic viral infections such as hepatitis C, or as simians (monkeys) are able to do with simian immunodeficiency virus (SIV).
LTNPs and elite controllers: why does HIV not progress in some people?
For reasons that are not well understood, a minority of HIV-positive individuals – long-term nonprogressors, or LTNPs – maintain high CD4 cell counts much longer than most. One particularly fortunate group, the so-called elite controllers, are able to keep HIV viral load at undetectable levels with no antiretroviral or other treatment.
One reason may lie in the immune system's CD8 cells, which control HIV by destroying infected cells. In most infected individuals, CD8 cells are present in high numbers yet seem unable to properly respond to HIV. LTNPs may be blessed with CD8s that remain able to strongly attack HIV-infected cells. The reasons for this are likely genetic.
In order to reduce cardiovascular risk, comprehensive HIV treatment will need to reduce inflammation, not just control viral replication.
Jean-Pierre Routy, McGill University
In fact, many genetic differences between individuals can affect vulnerability to HIV infection and the speed of disease progression. For instance, to infect a CD4 cell, HIV needs to latch on to two specific pieces of the cell's surface – the CD4 molecule itself, plus one of two ‘co-receptors’ called either CCR5 or CXCR4 – the majority of virus uses CCR5. A small percentage of people lack one or more of the genes needed to make CCR5. In people with a single missing gene, HIV disease develops much more slowly: such people have fewer CCR5 molecules, giving HIV fewer targets. Those who entirely lack the CCR5 genes seem altogether immune to the vast majority of HIV strains and indeed we now have a drug, maraviroc (Celsentri), that mimics this situation by blocking off people’s CCR5 receptors.
Other genes called things like TRIM and APOBEC control other immune defence mechanisms that interfere with various aspects of the life-cycle of viruses (not just HIV). HIV has in turn developed counter-defence genes like nef and vif that neutralise these cellular defences – but we could develop drugs that in turn block these genes and allow the cell to control HIV. Natural variations in these genes may explain why some people control their infection more effectively, and may influence the sensitivity of different populations to infection – the mutation that deletes the CCR5 gene, for instance, occurs in about 1.5% of northern European Caucasians but virtually no black Africans.
What lies ahead?
Regardless of the immune stimulation that seems to help drive HIV disease, in the end we are trying to avoid the opposite – the immune deficiency that leaves people vulnerable to fatal opportunistic infections. Ideally, HIV treatment may need to guard against both immune deficiency and stimulation. This is likely to be a complex goal, and the consensus is that considerable research is still needed.
McGill University's Jean-Pierre Routy believes that, in order to reduce cardiovascular risk, comprehensive HIV treatment "will need to reduce inflammation, not just control viral replication". How we do that will almost certainly include a push toward starting treatment earlier, but "adding anti-inflammatory drugs to antiretroviral treatment may be the best way to prevent long-term immune hyperactivation". Clinical trials of anti-inflammatory agents such as chloroquine are set to begin, but such trials will need to be conducted cautiously so as not to induce the wrong kind of immunosuppression.11
What role, then, for ‘immune-boosting’ treatments such as interleukins? Large and very long-term trials of interleukin-2 (IL-2) recently concluded that, despite raising CD4 counts, IL-2 resulted in no net long-term improvement in the people who took it. Indeed, people who received IL-2 were more likely to develop serious illnesses, especially an array of blood vessel and cardiovascular problems that are likely due to inflammation. However, there are dozens of interleukins and other cytokines governing the immune system, and interacting with each other: ‘immune boosting’ and ‘immune suppressing’ are likely oversimplified ways of viewing such a complex network. Says Routy, "we didn't have this understanding of HIV and inflammation when the IL-2 trials were designed twelve years ago. There may be different benefits with IL-7 or other cytokines."
An earlier start to HIV treatment
While many details remain to be investigated, consensus is growing around one key point: the need for earlier antiretroviral treatment. If ongoing HIV infection poses greater future health risks than HIV treatment does – as SMART and other studies suggest – then earlier treatment would be warranted. Large comparisons of cohort studies are finding that, as treatment is started at higher CD4 counts, the risk of AIDS-defining illnesses or death steadily decreases. The trend holds true up to beginning treatment at CD4 counts of 350 cells/mm3, although the benefits of starting treatment at even higher counts are less clear.
Is the case for earlier treatment persuasive enough to change treatment decisions for people with HIV? As an example, Richard Carson, diagnosed in 2005, is uncertain. By current treatment guidelines, Richard's robust CD4 counts (635 cells/mm3 at last count) and low viral load (1550 copies/ml) have allowed him to look at antiretrovirals as a distant prospect. He has heard the earlier-is-better arguments, but is not quite prepared to jump into treatment as a result. "In the end, I'll do whatever is best for me," he says. "I've heard a lot of reasons why I should not start treatment yet – the side-effects, the risk of resistance. If there's more solid evidence that I shouldn't wait, then I may change my mind."
One final challenge may be simply trying to accommodate new evidence and new insights into a pre-existing model that no longer fits. CD4 cell depletion has often been understood by a simple ‘tap and plughole’ analogy: picture the CD4 cell count as the level of water in a sink, with the drain open and the tap running. CD4 cells are destroyed as they are infected by HIV (the drain), but replenished as the body produces more (the tap). When the tap can no longer keep pace with the drain, CD4 counts fall.
As we realise the many factors that affect disease progression, will it be hard to abandon this simple picture for a more complex, if more accurate one? Danny Douek thinks not: "I don't necessarily think that more accurate means messier and more complicated. The original model of a tap and drain is actually a pretty good model. I think we've simply realised that there may be more taps and more drains. I think the model still stands pretty well, but it's becoming more complete and sophisticated. Ultimately it will be simpler because it will make more sense and leave less unanswered."
Types of immune cells
The immune system is a complex array of different cells that do different jobs.
Some mount fast, non-specific reactions such as allergies to get rid of foreign substances.
Some, the monocytes, engulf and sometimes digest invaders.
B-cells secrete antibodies, proteins that surround specific invaders and either physically block them from infecting cells or flag them for destruction.
T-cells divide into CD8 cells, which destroy already infected cells, or CD4 cells, which regulate and amplify other parts of the immune response,
Both B- and T-cells can be memory cells, sensitised to specific invaders for a quick response to them in the future. Vaccines work by priming this memory.
B- and T-cells can also be resting or activated. Activated cells work at infection sites, and they are short-lived.
One theory of how HIV slowly destroys the immune system is that it causes too many T-cells to stay in a permanently activated state, and thus ‘exhausts’ this branch of the immune system.
1. Douek D Making sense of HIV pathogenesis. The PRN Notebook 10(8):9-12, 2005.
2. Estes J et al. Collagen deposition limits immune reconstitution in the gut. J Infect Dis 198:456-464, 2008.
3. Mehandru S The gastrointestinal tract in HIV-1 infection: questions, answers, and more questions. The PRN Notebook 2007.
4. Levy JA HIV and the pathogenesis of AIDS. 3rd ed. ASM Press, 2007.
5. El-Sadr W HIV and inflammation: a paradigm shift. 5th International AIDS Society Conference on HIV Pathogenesis, Treatment, and Prevention. Cape Town, South Africa, abstract MOPL104, 2009.
6. Ascher MS and Sheppard HW AIDS as immune system activation: a model for pathogenesis. Clin Exp Immunol 73:165-167, 1988.
7. Lau B et al. C-reactive protein is a marker for human immunodeficiency virus disease progression. Arch Intern Med 166:64-70, 2006.
8. Douek D HIV disease progression: immune activation, microbes, and a leaky gut. Top HIV Med 15(4):114-117, 2007.
9. Brenchley JM et al. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med 12:1365-1371, 2006.
10. Sodora DL, Silvestri G Immune activation and AIDS pathogenesis. AIDS 22(4):439-446, 2008.
11. Forsman A, Weiss RA Why is HIV a pathogen? Trends in Microbiology 16(12):555-560, 2008.