Distinct patterns of HIV-1 evolution within metastatic tissues in patients with non-Hodgkins lymphoma

Abstract

Despite highly active antiretroviral therapy (HAART), AIDS related lymphoma (ARL) occurs at a significantly higher rate in patients infected with the Human Immunodeficiency Virus (HIV) than in the general population. HIV-infected macrophages are a known viral reservoir and have been shown to have lymphomagenic potential in SCID mice; therefore, there is an interest in determining if a viral component to lymphomagenesis also exists. We sequenced HIV-1 envelope gp120 clones obtained post mortem from several tumor and non-tumor tissues of two patients who died with AIDS-related Non-Hodgkin's lymphoma (ARL-NH). Similar results were found in both patients: 1) high-resolution phylogenetic analysis showed a significant degree of compartmentalization between lymphoma and non-lymphoma viral sub-populations while viral sub-populations from lymph nodes appeared to be intermixed within sequences from tumor and non-tumor tissues, 2) a 100-fold increase in the effective HIV population size in tumor versus non-tumor tissues was associated with the emergence of lymphadenopathy and aggressive metastatic ARL, and 3) HIV gene flow among lymph nodes, normal and metastatic tissues was non-random. The different population dynamics between the viruses found in tumors versus the non-tumor associated viruses suggest that there is a significant relationship between HIV evolution and lymphoma pathogenesis. Moreover, the study indicates that HIV could be used as an effective marker to study the origin and dissemination of lymphomas in vivo.

Figure 1. Histopathology of p24 positive macrophages.

Tissues from AM and IV were co-stained to identify infiltrating macrophages (CD68, red) and cells productively infected with HIV (p24, brown). Representative tissues of the left axillary lymph node from patient AM (A) and the spleen from patient IV (B) are shown. Examples of double-stained p24 positive macrophages are highlighted by the black arrows. Images are shown at 400X. The distribution and staining pattern similar in all tumor-containing tissues analyzed (not shown).

Figure 2. Phylogenetic analysis HIV-1 gp120 sequences from different tissues.

Bayesian maximum clade credibility trees assuming a relaxed molecular clock and constant population size coalescent prior generated from the posterior distribution of trees less a 50% burn-in. Branch lengths are shown according to the scale bar at the bottom of each panel, in relative units of time (nucleotide substitutions per site). Posterior probabilities for major nodes are indicated. Internal branches are colored according to the maximum parsimony reconstruction of the ancestral tissue of origin: red = tumor; yellow or green = lymph node; blue = normal tissue. Posterior probability values>80% are shown along the branches. The tissues of origin found within major clades are listed to the right with LN indicating lymph nodes. Number of sequences and V3 loop net charge within the clade for each tissue are given in parentheses. A. Subject AM. Yellow and green lineages represent viral strains from right and left axillary lymph nodes, respectively. B. Subject IV.

Figure 3. HIV-1 population dynamics in tumor and non-tumor tissues.

Bayesian estimates of HIV-1 effective population size (number of infectious genomes effectively contributing to the next generation, y-axis) over time (x-axis) were inferred using the models that fitted best for each data set according to the Bayes Factors test (see Supplemental Table S2). The black line represents the median estimate of the effective population size with the shaded area showing the 95% high posterior density intervals of the estimates. Top panels. Constant population size over time of HIV-1 in normal tissues from patient AM (left panel) and IV (right panel). Bottom panels. Non-parametric estimates (Bayesian skyline plots) of effective population size change over time for subject AM (left panel) and IV (right panel).

Figure 4. Viral gene flow analysis among different.

Light blue arrows represent HIV-1 gene flow from/to normal tissues, tumor tissues and lymph nodes. Viral gene flow among different tissues was inferred with a modified version of the Slatkin and Maddison method^13,28 from the HIV-1 genealogies including strains from different tissues. Numbers on the arrows represent percentage of observed migration in the genealogies. Left panel. Subject AM. Right panel. Subject IV.