Painstaking genetic analysis of the lymph nodes of people with HIV suggests that control of inflammation in the lymph nodes in some untreated people might allow treatment to be delayed. In the February 15th issue of the Journal of Infectious Diseases, Quingsheng Li and colleagues from the University of Minnesota suggest that it might be possible to assess disease progression more accurately by taking tiny samples of lymph node tissue. Such biopsies could help in identifying those patients who would benefit from anti-inflammatory agents such as COX-2 inhibitors, which might reduce the stimulus to HIV replication and reduce the damage to lymph nodes that eventually contributes to loss of immune control of the virus.
The authors highlight the slowness of HIV disease progression as a puzzling feature of the disease. Could they find answers to the question of why HIV takes so long to cause disease in the genes and architecture of the lymph nodes, which harbour the bulk of the body’s HIV load during much of a person’s life with HIV?
For background, each cell in the human body (apart from red blood cells) contains a full suite of the approximately 30,000 genes that make up the human genome. Depending on the cellular functions being performed at any given time, particular genes will be switched on or “expressed,” meaning that the DNA that comprises the gene is being translated into messenger RNA, which in turn is converted into a functional protein. The researchers utilised a relatively new technology called microarray analysis that allowed the activity of approximately 12,000 human genes to be evaluated. The microarray tags messenger RNA with a fluorescent probe, causing active genes to literally light up. The degree of illumination can then be measured and analysed using a specialised computer program.
The study evaluated RNA sampled from the inguinal (groin) lymph nodes of nine HIV-infected individuals. Samples were taken at study entry, when all nine were untreated, and then 3 days and one month later. Five individuals started HAART after the first sample was taken, while the remaining four deferred treatment. The comparison of lymph node samples from treated and untreated study participants allowed the researchers to look for changes in gene expression that occurred in response to HAART. The results indicated that the expression of around 200 genes either increased or decreased by a statistically significant amount (between 2- and 10-fold in most cases).
The identification of these “treatment-responsive” genes was just the first step, however. While many human genes have been characterised based on their DNA sequence, knowledge regarding function can be lacking. The researchers painstakingly combed through gene databases and the scientific literature, eventually managing to place over 80% of the treatment-responsive into generic categories including: immune activation and defences, inflammation, cell trafficking, cell repopulation, follicle reformation (follicles are areas of the lymph nodes involved in initiation of immune responses, and are typically damaged by HIV infection) and wound healing.
In the category of immune defences, nearly all genes were decreased in expression after HAART. These included genes such as those for the CD3 and CD8 molecules present on T cells, consistent with the post-HAART decline in HIV-specific CD8 T-cell responses previously documented by other researchers. But Li and colleagues found that “the most striking finding in this category was the change in expression, extending beyond the early into the late stages of infection, of genes predominantly related to innate immunity.” These included natural killer cell genes, complement system genes and multiple genes related to interferons. This suggests to the researchers that innate immunity plays an important role in partially controlling HIV replication in the lymph nodes over the long course of the disease. Genes related to immune activation and inflammation also typically decreased after HAART, mirroring the previously well-documented diminution of the T cell activation markers HLA-DR and CD38.
Conversely, genes associated with tissue repair and remodelling generally increased. This category included genes that make structural components of cells and regulate wound healing; suggesting that the HAART-induced decline in HIV replication allows some restoration of damage the virus has caused in the lymph nodes. There were both increases and decreases in genes related to cell trafficking, reformation of lymph follicles and apoptosis (cell death). Because HAART is known to lead to the release of T cells from the lymph nodes into the circulation, changes related to genes involved in cell trafficking were interpreted as reflecting this process. Similarly, the reduced expression of apoptosis-related genes was considered to be related to the reduction in T-cell death that results from HAART treatment. Increases in the expression of genes involved in follicle reformation were also consistent with improvements in lymph node structure seen in magnified lymph node specimens sampled after HAART.
In conclusion, the researchers propose a model of pathogenesis wherein innate and adaptive immune defences constrain HIV replication to some extent, but these effects are counterbalanced by inflammation and activation, which both maintain these immune defences yet also increase viral activity. They suggest that anti-inflammatory agents such as cyclooxygenase-2 inhibitors may thus have a role as potential HIV therapies.
In an accompanying commentary, Roger Pomerantz highlights some issues that may need to be borne in mind when attempting to interpret the results of this study. For example, while the anti-HIV activity of innate immunity is inferred from changes in gene expression, it is not necessarily proven. Additionally, HAART may have direct effects on immune activation that are not mediated by the reduction in HIV viral load. It can also be tempting, Pomerantz notes, to cherry-pick data generated from such complex analyses that fit previously formulated theories about pathogenesis. In terms of methodology, the evaluation of gene expression may miss changes that occur after messenger RNA transcription yet affect the resultant protein, suggesting that proteins also need to be studied (this field is called proteomics, as opposed to genomics). Small, non-gene RNAs (such as the recently identified small interfering RNAs) may also play key roles in cell function that cannot be captured by looking at genes alone. Nevertheless, Pomerantz concludes that the study “is an excellent initial foray into the field of human functional genomics as it applies to HIV infection.”
References
Li Q et al. Functional genomic analysis of the response of HIV-1-infected lymphatic tissue to antiretroviral therapy. J Infect Dis. 189;4:572-82, 2004
Pomerantz RJ. HIV-1 infection and genomics: sorting out the complexity. J Infect Dis. 189;4:567-571, 2004