PMS-601, a New Platelet-Activating Factor Receptor Antagonist That Inhibits Human Immunodeficiency Virus Replication and Potentiates Zidovudine Activity in Macrophages

Abstract

We assessed the anti-human immunodeficiency virus (anti-HIV) activity in vitro of new platelet-activating factor (PAF) receptor antagonists, as PAF and viral replication are thought to be involved in HIV neuropathogenesis. We found that PMS-601 inhibited proinflammatory cytokine synthesis and HIV replication in macrophages and potentiated the antiretroviral activity of zidovudine. These results suggest that PMS-601 is of potential value as an adjuvant treatment for HIV infection.

Human immunodeficiency virus (HIV) type 1 (HIV-1)- associated dementia complex (ADC) is a frequent complication of progressive HIV infection (16, 21). The neuropathogenesis of HIV infection involves (i) the entry of the virus into the central nervous system (CNS), (ii) the production of viral and cellular components that mediate inflammation and neurotoxicity, and (iii) neuronal dysfunction and death. Gendelman et al. (14) recently provided evidence that metabolic encephalopathy is fueled by HIV replication. Infection of the CNS with HIV typically involves selective replication in mononuclear phagocytes and microglial cells, cell activation, and chronic inflammation (15, 32), leading to high levels of proinflammatory compounds (11, 17, 19, 29) such as tumor necrosis factor alpha (TNF-α), β-chemokines, and platelet-activating factor (PAF) (5, 13, 19, 27). The correlation between PAF levels in the cerebrospinal fluid and the severity of clinical dementia suggests a possible role for PAF in immune activation and neuropathological mechanisms (8, 22). Several studies have demonstrated a major slowing of HIV-1 disease progression in patients treated with highly active antiretroviral therapy (HAART) (3, 23). However, this treatment may have less of an impact on ADC than on other AIDS-defining illnesses due to the poor penetration of the drugs into the CNS (7). As PAF and virus replication are thought to be involved in HIV neuropathogenesis, we assessed the anti-HIV activities of piperazine derivatives that function as antagonists of the PAF receptor (28). We describe herein the effects of the most promising molecule, PMS-601, on proinflammatory cytokine synthesis, HIV replication, and the antiretroviral effects of zidovudine (AZT) in monocytes/macrophages.

Monocytes were isolated from peripheral blood mononuclear cells by countercurrent centrifugal elutriation (9). Monocytes, macrophages that had been differentiated for 7 days, and promonocytic U1 cells (10) were cultured in medium A, which consisted of RPMI 1640 cell culture medium (Roche Products, Mannheim, Germany) supplemented with 10% heat-inactivated (56°C for 30 min) fetal calf serum (Roche Products), 2 mM l-glutamine (Roche Products), and 1% triantibiotic mixture (penicillin, streptomycin, neomycin; Life Technologies, Grand Island, N.Y.). The cell culture medium was endotoxin-free, as shown by the Limulus amoebocyte lysate test (Sigma Chemical Co, St. Louis, Mo.). The chemically synthesized molecule PMS-601 [1,4-di-(3′,4′,5′-trimethoxybenzoyl)-2-(N,N-diethylaminocarbonyloxymethyl)-piperazine] inhibits platelet aggregation, with a 50% effective dose (ED₅₀) of 8 μM (28). The involvement of PAF in the lipopolysaccharide (LPS)-induced synthesis of proinflammatory cytokines (18, 31) led us to study the effects of PMS-601 on the secretion of these molecules. TNF-α and β-chemokine levels in cell culture supernatants of monocytes stimulated with LPS for 2 h were measured by using an enzyme-linked immunosorbent assay specific for detection of each cytokine (R&D [Oxon, United Kingdom] for chemokines; Immunotech [Luminy, France] for TNF-α). Under these experimental conditions, the increase in the levels of TNF-α, MIP-1α, MIP-1β, and RANTES secreted by induction with 100 ng of LPS per ml was significantly decreased by PMS-601 (27, 29, 17, and 42% inhibition, respectively) (Fig. 1). These anti-inflammatory properties correlated with the anti-PAF activity of PMS-601. The antiretroviral activity of PMS-601 was evaluated with macrophages that had been differentiated for 7 days and that had been infected with reference macrophage-tropic strain HIV-1/Ba-L (12, 28) or with a primary macrophage-tropic isolate, HIV-1-DAS (2) or HIV-1-THI (6). Viral replication was assessed by quantifying the reverse transcriptase (RT) activity in the cell culture supernatant (RetroSys; Innovagen, Lund, Sweden) and the antiretroviral effects of PMS-601 by determining the cumulative RT activity, the percentage of the control RT activity, and the ED₅₀, ED₇₀, and ED₉₀. As shown previously (6), the reference macrophage-tropic strain HIV-1/Ba-L and primary macrophage-tropic isolates HIV-1-DAS and HIV-1-THI replicated efficiently in macrophages, with optimal replication occurring between days 15 and 25 postinfection. Against these HIV-infected macrophages, PMS-601 exhibited similar levels of dose-dependent antiretroviral activity (Fig. 2A; data not shown for strains HIV-1-DAS and HIV-1-THI) for the three viruses tested. The mean ED₅₀, ED₇₀, and ED₉₀ were 10, 30, and 100 μM, respectively. This antiviral activity was not associated with a decrease in cell viability even after 30 days of treatment with 1 mM PMS-601 (Fig. 3A) (selectivity index [SI], >100). The antiretroviral efficiency of PMS-601 in macrophages was confirmed by the inhibition of viral replication (65% inhibition) when cells were exposed to PMS-601 for 24 h (Fig. 2B). HIV replication in macrophages is partly dependent on TNF-α. The effects of PMS-601 on TNF-α-dependent HIV production in chronically HIV-infected promonocytic U1 cells were investigated. HIV production was greatly decreased by 100 μM PMS-601 and by 0.5 μM indinavir (Fig. 2C) (for PMS-601, a 70% decrease; for indinavir, an 81% decrease); this effect was not associated with a decrease in cell growth (Fig. 3B). PMS-601 could be used as adjuvant therapy with antiretroviral molecules, particularly nucleoside RT inhibitors (NRTIs), due to its effects on the late steps in the biological cycle of HIV. We therefore assessed the effects of PMS-601 on the antiretroviral efficiency of AZT by treating macrophages simultaneously with PMS-601 and AZT. In these cells, AZT inhibited HIV-1/Ba-L replication, with an ED₅₀ of 15 nM (Fig. 4), and PMS-601 increased the antiretroviral efficiency of AZT. These effects were additive, as shown by (i) a combination index (CI) of 1, (ii) the decrease in viral replication by 73% by use of an ineffective dose of AZT (1 nM) in association with a dose of PMS-601 that inhibited 46% of viral replication when used alone, and (iii) the complete abolition of viral replication by 10 μM PMS-601 (46% inhibition when used alone) when it was used in association with 10 nM AZT (48% inhibition) (Fig. 4). Another PAF antagonist previously evaluated in our laboratory, RP55778, showed antagonism with AZT (4). RP55778 also decreased cell viability and had no antiviral activity in lymphocytes, whereas PMS-601 decreased the levels of HIV-1/Ba-L and HIV-1-LAI replication (ED₅₀, 5 μM) but had no effect on cell growth when it was used at a concentration of less than 125 μM (Fig. 3C) (50% cytotoxic concentration, 280 ± 90 μM; SI, ≈50).

FIG. 1.

Effect of PMS-601 on the production of TNF-α and chemokines in LPS-stimulated monocytes. Monocytes were simultaneously stimulated for 2 h with the optimal dose (100 ng/ml) of bacterial LPS (Escherichia coli serotype O111:B4; Sigma) and treated with 100 μM PMS-601. PMS-601 was dissolved in dimethyl sulfoxide (Prolabo, Fontenay/Bois, France), and the corresponding dose of dimethyl sulfoxide without PMS-601 was used as a control. This dose of dimethyl sulfoxide had no significant effect on the synthesis of proinflammatory cytokines. These experiments were performed in triplicate at least twice. Data are expressed as a percentage of the control and were analyzed by Student's unpaired t test (Statview, version 4.02, microcomputer software; Abacus Concept Inc., Berkeley, Calif.). □, untreated control; , monocytes treated with 100 μM PMS-601.

FIG. 2.

Effects of long-term treatment (A) and 24-h treatment (B) with PMS-601 on HIV-1/Ba-L replication in macrophages. A total of 10⁵ cells were infected with 1,000 50% tissue culture infective doses on day 0, and viral replication was determined throughout the culture period by quantifying the RT activity in the cell culture supernatant. Data are expressed as the mean ± standard deviation RT activity for three independent culture wells. (C) Effects of PMS-601 on HIV-1 production in chronically HIV-infected promonocytic cells. U1 cells were simultaneously stimulated with TNF-α (10 ng/ml) and treated with 100 μM PMS-601, which was noncytostatic. HIV particle production was measured after 48 h of incubation by quantifying the RT activity in the cell culture supernatant. Data are expressed as the mean ± standard deviation percent inhibition for four independent culture wells. These experiments were reproduced with cells from at least three independent healthy donors. DMSO, dimethyl sulfoxide.

FIG. 3.

Effects of PMS-601 on primary macrophage (A), promonocytic U1 cell (B), and peripheral blood lymphocyte (C) viability. Cells were treated throughout the culture with a range of PMS-601 doses and were counted either by trypan blue staining (for healthy lymphocytes and promonocytic U1 cells) or by neutral red staining, using a standard curve (for healthy macrophages). Data are representative of experiments performed with cells from three independent healthy blood donors.

FIG. 4.

Effects of PMS-601 on the antiretroviral efficacy of AZT in macrophages. Cells were treated with various associations of PMS-601 and AZT and were infected with 1,000 50% tissue culture infective doses/10⁵ cells infected with reference strain HIV-1/Ba-L. Results are expressed as the mean ± standard deviation percent inhibition for three independent culture wells on the basis of the cumulative RT activity. Data were analyzed by Student's unpaired t test (Statview, version 4.02, microcomputer software; Abacus Concept Inc.). These experiments were reproduced with cells from three independent healthy donors.

Our results show that PMS-601, a PAF receptor antagonist, combines significant anti-inflammatory and broad antiretroviral properties. Nevertheless, the lack of correlation between the anti-PAF and antiretroviral activities of the members of this family of piperazine derivatives (28) suggests that the antiretroviral activity of PMS-601 was not solely due to its anti-PAF and anti-inflammatory properties. PMS-601 probably also has other biological effects. The inhibition of viral replication observed for the 24-h treatment of macrophages and comparison of the properties of PMS-601 and RP55778 support this idea. First, the results of the 24-h treatment experiment strongly suggest that PMS-601 interacts with an early stage of the HIV biological cycle. However, its precise mode of action is not yet clear: no significant modulation of RT activity was detected, and specific interaction with HIV-1 coreceptors is unlikely because PMS-601 inhibited similarly the replication of viral strains that differed in their binding to HIV coreceptors (1). Second, RP55778, another PAF receptor antagonist, has been shown to have anti-HIV-1 and anti-inflammatory effects in vitro against cells of the macrophage lineage (20, 30). Like PMS-601, RP55778 decreased the levels of TNF-α production and the release of HIV particles from monocytes/macrophages. However, RP55778 had no antiviral activity in lymphocytes and reduced the antiretroviral efficiency of NRTIs, particularly that of AZT (CI, ≈10) (4). The potentiation of the effects of AZT is a key property in favor of the use of PMS-601 as an adjuvant therapy with NRTIs. In some tissues, such as those of the CNS, HIV replication is not completely abolished by HAART. Instead, it is maintained by the inflammatory environment that exists in parallel with monocyte recruitment (24, 25). Due to its various biological properties and its lipophilicity, PMS-601 is an excellent candidate molecule for decreasing HIV replication, leukocyte recruitment, and invasive tissue inflammation.

Our results, obtained in vitro, suggest that the piperazine derivative PMS-601 may effectively decrease the neuropathological symptoms associated with HIV. Recent data obtained from studies with HIV-infected patients treated with lexipafant, another PAF receptor antagonist, provide further evidence that this type of molecule may be valuable for the future treatment of certain symptoms associated with HIV infection (26).