Inflammation in HIV-Associated Pulmonary Hypertension: Cause or Effect?

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

In this review, we discuss the possible pathogenetic role of the HIV Nef (negative factor) protein in formation of pulmonary vascular lesions seen in HIV-associated pulmonary hypertension (HAPH). Pulmonary hypertension (PH) affects 0.5% vs 0.0015% in the HIV-infected and general populations, respectively. The lung lesions in HAPH are very similar to those found in idiopathic PH: vascular remodeling, luminal obliteration, and inflammatory cells. In general, individuals with HAPH typically present with progressive subacute dyspnea and occasionally a nonproductive cough. As the disease progresses and as right ventricular involvement ensues, patients may report pedal edema, fatigue, syncope or near-syncope, and chest pain. In the preantiretroviral (ART) era, HIV-infected patients died from immunodeficiency and associated opportunistic infections. However, as these individuals live longer now in the post-ART era, chronic complications associated with HIV infection become more evident.

The conundrum is that not all patients with HIV develop PH, so there must be something about the virus itself or perhaps some environmental or genetic factor that contributes to this disease. Unfortunately, when we first embarked on these studies there was not an appropriate animal model, so we wondered whether similar lesions would be found in simian immunodeficiency virus (SIV)-infected monkeys. Rhesus macaques infected with SIV get a disease very similar to human AIDS. We hoped this monkey model would be a good experimental tool to study the natural history of HAPH. We had access to lungs from several nonhuman primate centers and discovered structures resembling plexiform lesions of PH in the lungs of macaques infected with a chimeric SIV containing the HIV_SF33A nef gene12 or the nef gene isolated from a long-term nonprogressor and subsequently shown to cause disease in a macaque cohort.3 Interestingly, none of the monkeys infected with the parental SIV-_mac239 nef showed lung lesions, suggesting that there are functional differences between the 2 proteins, either alone or in conjunction with other viral or cellular factors.

Nef protein was found in the vascular wall of the pulmonary vessels in the SHIV-nef-infected monkeys (Figure 1),4 which is surprising since the lung vascular cells lack the appropriate HIV receptors. Nevertheless, one of the characteristics of Nef is that it can be secreted by infected cells and taken up by uninfected cells,5 which may account for its presence in the uninfected cells in the lung.

View Figure

• Download as Powerpoint
• Download Figure

Nef is an N-terminus-myristoylated protein with a relative molecular mass of 24–32 kDa, localized to the cytoplasm and membranes of cells. Myristoylation is critical for most of its functions, including membrane localization, and is very highly conserved among nef alleles (Nef structure-function relationships reviewed in reference 6). Nef has no enzymatic activity, but is thought to act as an adaptor molecule, particularly because of its ability to localize to membranes and associate with cellular proteins. Domains of Nef that have been shown to be important for these effects are the proline-rich (PxxP)₃ domain thought to interact with SH3 domains in protein partners and 6 sequence domains that interact with the endocytotic cellular machinery, including a dileucine motif. These domains are important for Nef-mediated downregulation of CD4 and blocking major histocompatibility antigen I (MHC I) trafficking to the membrane, allowing the infected cell to evade immune surveillance. Disruption in Golgi trafficking pathways has been postulated as a pathogenetic mechanism in PH and is an area intensely investigated in Dr Pravin Sehgal's laboratory.7

A new cohort of monkeys was infected with SHIV-nef and the infection allowed to proceed naturally. Monkeys were euthanized at various times post infection and histopathology of lung tissues assessed. At 12 weeks post infection, we found perivascular inflammation, at 24 weeks post infection medial hypertrophy, and by 37 weeks we discovered considerable vascular remodeling. Immunofluorescence studies revealed a linear relationship between Nef, CD34, and smooth muscle actin expression as a function of weeks post infection.

We then examined HAPH lungs and found Nef in the pulmonary vasculature and parenchyma of 2 patients. There was no Nef staining in the uninfected patient with idiopathic PH, while the patient with HAPH showed Factor 8 and Nef staining both in the intimal region and in what appear to be inflammatory cells (Figure 2).4

View Figure

• Download as Powerpoint
• Download Figure

We then hypothesized that certain sequence motifs would be more prevalent in alleles from HIV-infected patients with PH than in HIV-infected pulmonary normotensives. We looked at patients with and without PH from 3 different cohorts: the French cohort, kindly provided by Drs Marc Humbert and Cecile Goujard; a cohort from Dr Nicola Petrosillo from Italy; and a cohort from Dr Priscilla Hsue from the University of California-San Francisco. Table 1 shows the demographic characteristics of these patients. We obtained PBMC DNA, plasma, and lung tissue fragments. The demographics are not remarkable except that HIV patients with PH were more likely to be on ART therapy.

Table 1. General demographic, virological, and immunological characteristics of HIV-infected individuals from initial and validation cohorts analyzed retrospectively. Banked HIV-infected samples were made available for these studies as follows: PBMC DNA from subjects with bona fide diagnosis of HAPH from France and San Francisco, lung tissue from a subject from Italy with echo-based indication of HAPH; controls consisted of plasma from French HIV-infected subjects with no evidence of PH and PBMC DNA for the validation (normotensive) cohort from San Francisco. Duration of HIV infection (in years) was calculated using the HIV diagnosis date and the date the specimen was collected. CD4⁺ T-cell counts are shown in cells/mm³, and viral load is shown in copies/mL plasma. Controls and HAPH groups were compared using the appropriate tests in SAS.P values <0.05 were considered significant. Abbreviations: SD, standard deviation; ART, antiretroviral therapy; p75, 75th percentile; FE, Fisher's exact test; TE/TU, 2-sample t-test assuming equal/unequal variance; LR: Log Rank test; W, Wilcoxon Mann-Whitney with continuity correction

View Fullscreen

We catalogued the mutations and found similar mutations in both human cohorts to those found in the monkeys. We compared the mutations in the French normotensive and HAPH group and in the San Francisco normotensive and hypertensive group and discovered that the mutations found in the hypertensive individuals from each cohort were statistically significantly different from the normotensive individuals in each group. We believe that these mutations alter Nef activity and ability to interact with vascular cells in the lung environment. There have been several in vitro studies of HIV pathogenesis, including studies that show Nef can cross membranes and accumulate in the cytosol of noninfected cells5 and that it induces apoptosis of primary human brain micro-vascular endothelial cells.8

We are developing molecular tools to address several experimental questions by fusing the different nef alleles recovered from patients with HAPH to a reporter protein such as green fluorescent protein. Since Nef has no catalytic activity, a useful assay to measure its activity is by its ability to downregulate CD4. We are validating our newly generated constructs with some Nef-GFP constructs kindly provided by Dr Victor Garcia9 in a HeLa cell line engineered to express CD4 (National Institutes of Health AIDS Reagents Program). We found that GFP alone showed 70% CD4 expression while Nef-GFP showed that only 28% of the cells were CD4-positive, which is in line with previous studies from Dr Garcia's group.

In closing, we believe that Nef, either directly or indirectly, can influence vascular cells to undergo apoptosis, proliferation, and/or release of chemokines, cytokines, and/or growth factors and in the process leading to the remodeling associated with PH.