Rate of Evolution Yields Insight into How HIV Causes AIDS

Despite 25 years of intensive research, there is much the experts still don't know about HIV. Among the persistent mysteries is why, exactly, HIV infection causes AIDS. HIV infects and destroys helper T cells, which are needed to mobilize the immune system to fight pathogens. But HIV doesn't seem to kill enough helper T cells—at least not directly—to explain the immune system's eventual collapse. In recent years evidence has been accumulating in favor of a disease model in which the host's own immune response plays a key role in the cause of AIDS. Among this evidence is an analysis of the rate of viral evolution within individual hosts.

Philippe Lemey and colleagues sought to understand how HIV causes AIDS by addressing a related question. Why is that some individuals infected with the virus progress to AIDS in months, whereas others develop immunodeficiency only after decades?

Lemey approached this question by reanalyzing a remarkable data set collected several years ago by Raj Shankarappa and colleagues in the laboratory of James Mullins. Shankarappa followed nine patients from the time the patients first tested positive for HIV until, six to twelve years later, seven of them had developed advanced AIDS. Throughout the study the researchers repeatedly collected samples of HIV from each of the patients, then read the sequence of nucleotides in one of HIV's genes (the one called env, which encodes a precursor to two proteins found on HIV's envelope). Shankarappa's library of samples and sequences is, in essence, a fossil record documenting the history of viral evolution in each of the patients in exquisite detail.

Lemey's first step in reanalyzing Shankarappa's data was to use the nucleotide sequences to estimate the evolutionary tree for the viruses in each of the patients. A small piece of the tree for Patient 1 appears below.

Time progresses from left to right. The dashed lines indicate the times when the researchers collected samples from the patient. The red, blue, and black lines trace the evolutionary diversification of viruses within the patient, from the common ancestor at lower left to the surviving descendants at upper right.

Notice that the tree has many side branches (blue lines) that ultimately end in extinction (all black lines except those at upper right). And it has one branch (red) that persists throughout the infection. This pattern was typical among the nine patients. It is caused by the fact that HIV has a high mutation rate, and thus produces many variants carrying deleterious mutations.

Lemey was interested in the rate of evolution along the persistent red branch. Actually two rates of evolution, both assessed by counting the changes that accumulated in HIV's genome. These changes come in two flavors. One flavor, non-synonymous nucleotide substitutions, change the amino acids in the proteins encoded by the env gene. They are thus potentially subject to natural selection. The other flavor, synonymous substitutions, cause no changes the proteins encoded by env. They are thus likely to be selectively neutral.

Students who complete the HIV Clock lab in EvoBeaker practice using the accumulation of neutral changes to estimate the age of a lineage. Lemey wanted to know if the accumulation of either flavor of change was associated with the rate at which patients progress to AIDS.

We might reasonably have expected that the rate of progression to AIDS would be associated with the rate at which non-synonymous mutations accumulate in a patient's HIV lineage. Immune systems doing a better job of constraining HIV's replication might, for example, impose stronger selection in favor of altered protein structure. Higher rates of non-synonymous HIV evolution might thus be associated with slower progression to AIDS.

In fact, Lemey found no association between the rate of non-synonymous evolution in a patient's HIV lineage and the rate of progression to AIDS.

Instead, Lemey found an association between the rate of synonymous evolution and the rate of progression. Patients whose HIV lineage experienced more rapid neutral evolution advanced more quickly to AIDS.

This result doesn't seem to make much sense, until we consider it in light of a model of AIDS in which the real problem is not the fact that HIV infects and helper T cells, but that HIV infection induces, directly and indirectly, a chronic state of heightened immune activation. This state is characterized by the extensive proliferation of T cells, including helper T cells. The trouble with this is that activated helper T cells are the host cells in which HIV replicates most efficiently. With HIV, a response that ordinarily helps curtail infections instead just feeds the virus. A vicious cycle ensues, producing extensive collateral damage. Eventually the immune system crumbles.

Patients whose immune systems are most highly activated likely experience the highest rates of viral replication. More rounds of replication mean more mutations. And more mutations means a higher rate of neutral evolution.

Lemey's discovery is thus consistent with the model of AIDS in which the trouble is caused not so much by HIV itself as by the way the immune system reacts to it.

Sources:

Illustrations are from Lemey, P., S. L. Kosakovsky Pond et al. 2007. Synonymous substitution rates predict HIV disease progression as a result of underlying replication dynamics. PLoS Computional Biology 3: e29. Published under the Creative Commons Attribution License.

Lemey and colleagues report their discovery in: Lemey, P., S. L. Kosakovsky Pond et al. 2007. Synonymous substitution rates predict HIV disease progression as a result of underlying replication dynamics. PLoS Computional Biology 3: e29.

Shankarappa and colleagues report their study, and their own analysis of the data, in: Shankarappa, R., J. B. Margolick et al. 1999. Consistent viral evolutionary changes associated with the progression of human immunodeficiency virus type 1 infection. Journal of Virology 73: 10489-10502.

For reviews of the current understanding of how HIV causes AIDS, see:

• Douek, D. C., M. Roederer, and R. A. Koup. 2009. Emerging concepts in the immunopathogenesis of AIDS. Annual Review of Medicine 60: 471-484.
• Paiardini, M., I. Pandrea et al. 2009. Lessons learned from the natural hosts of HIV-related viruses. Annual Review of Medicine 60: 485-495.
• Pandrea, I., D. L. Sodora et al. 2008. Into the wild: simian immunodeficiency virus (SIV) infection in natural hosts. Trends In Immunology 29: 419-428.
• Silvestri, G. 2009. Immunity in natural SIV infections. Journal of Internal Medicine 265: 97-109.