HIV co-receptor use in the blood and central nervous system differs in 20% of patients

HIV from the blood and cerebrospinal fluid of most patients with detectable viral loads uses the same co-receptors to enter cells, according to a study published in the electronic edition of The Journal of Infectious Diseases on 9^th February.

However, the study found that a minority of patients have different types of HIV in the two compartments. This may influence the risk of developing AIDS-related dementia and the effectiveness of entry inhibitor drugs that target HIV’s co-receptors.

In addition to the CD4 receptor, HIV also requires the presence of other co-receptors on the cell surface to infect its target cells. Most HIV strains use the CCR5 receptor to infect macrophage cells. It is thought that macrophages infected with HIV are a primary source of HIV crossing into the central nervous system.

Other HIV strains, called X4 viruses, preferentially use the CXCR4 receptor to infect lymphocytes as well as macrophages and microglial cells in the central nervous system.

“In general, R5 viruses predominate during early infection,” explain the study’s authors. “X4 viruses are most often detected later in the course of infection and may be associated with more-rapid CD4 T-cell loss.”

To assess the relationship between co-receptor use by HIV in the blood plasma and in the cerebrospinal fluid that surrounds the brain and spinal cord, investigators from San Francisco analysed co-receptor use in paired samples from a cohort of 46 HIV-positive subjects with viral loads above 1000 copies/ml. The patients had a median CD4 cell count of 245 cells/mm³ (range 5 to 612 cells/mm³), and 9 (20%) were receiving antiretroviral therapy.

“Most cerebrospinal fluid populations utilised CCR5 as the principal co-receptor, and the majority of subjects had concordant tropism in the two compartments,” they conclude. “However, approximately one fifth of subjects displayed the R5+X4 phenotype in one or both fluids, and one tenth of subjects had discordant plasma and cerebrospinal fluid tropism.”

Each patient underwent lumbar puncture to extract a sample of cerebrospinal fluid at the same time as a blood sample was taken. The Pheno-Sense HIV entry assay was used to determine the co-receptor favoured by the HIV particles extracted from each sample, and this was compared to the patients’ CD4 cell count, viral load and mental function.

In 38 plasma samples and 39 cerebrospinal fluid samples, R5 virus predominated. In contrast, none of the samples contained pure X4 virus, although six plasma and seven cerebrospinal fluid samples contained a combination of R5 and X4 virus.

R5 virus was more common in patients with higher CD4 cell counts, as has been observed in previous studies. However, the four patients with AIDS-related dementia all had R5 virus in the cerebrospinal fluid. This is consistent with the known role of infected macrophages in transporting HIV into the central nervous system and R5 viruses damaging brain cells.

Forty-one (89%) of the pairs of samples were concordant (36 for R5 and five for R5+X4). However, two patients had R5+X4 in the cerebrospinal fluid and R5 in the blood plasma, and three had the opposite pattern.

The investigators selected three patients with R5+X4 virus in blood, cerebrospinal fluid or both for detailed genetic analysis of the relationship between HIV species in the two compartments. In all three cases, they found evidence of multiple exchanges of HIV species between the blood and the cerebrospinal fluid. However, they also concluded that some independent evolution HIV occurs within each compartment, which may explain the low prevalence of discordance seen in their full analysis.

This, they argue, raises the possibility that measurements of co-receptor use by HIV in the blood may not reflect the make-up of the HIV throughout the body, increasing the risk of failure of new entry inhibitor drugs that only target one co-receptor.

“Our detailed analysis…provides evidence of extensive, subtle discordance between and variation within compartments,” they write. “Our results have implications not only for better understanding of viral compartmentalisation but also for optimising future therapy with chemokine-receptor entry inhibitors.

“Screening of viral tropism in plasma alone may be inadequate for determination of optimal therapy with entry inhibitors… Assessment of co-receptor utilisation in cerebrospinal fluid and other tissue compartments will contribute to understanding the uses and short-comings of this mode of therapy during clinical trials.”