HIV and HCV activate the inflammasome in monocytes and macrophages via endosomal Toll-like receptors without induction of type 1 interferon

Abstract

Innate immune sensing of viral infection results in type I interferon (IFN) production and inflammasome activation. Type I IFNs, primarily IFN-α and IFN-β, are produced by all cell types upon virus infection and promote an antiviral state in surrounding cells by inducing the expression of IFN-stimulated genes. Type I IFN production is mediated by Toll-like receptor (TLR) 3 in HCV infected hepatocytes. Type I IFNs are also produced by plasmacytoid dendritic cells (pDC) after sensing of HIV and HCV through TLR7 in the absence of productive pDC infection. Inflammasomes are multi-protein cytosolic complexes that integrate several pathogen-triggered signaling cascades ultimately leading to caspase-1 activation and generation pro-inflammatory cytokines including interleukin (IL)-18 and IL-1β. Here, we demonstrate that HIV and HCV activate the inflammasome, but not Type I IFN production, in monocytes and macrophages in an infection-independent process that requires clathrin-mediated endocytosis and recognition of the virus by distinct endosomal TLRs. Knockdown of each endosomal TLR in primary monocytes by RNA interference reveals that inflammasome activation in these cells results from HIV sensing by TLR8 and HCV recognition by TLR7. Despite its critical role in type I IFN production by pDCs stimulated with HIV, TLR7 is not required for inflammasome activation by HIV. Similarly, HCV activation of the inflammasome in monocytes does not require TLR3 or its downstream signaling adaptor TICAM-1, while this pathway leads to type I IFN in infected hepatocytes. Monocytes and macrophages do not produce type I IFN upon TLR8 or TLR7 sensing of HIV or HCV, respectively. These findings reveal a novel infection-independent mechanism for chronic viral induction of key anti-viral programs and demonstrate distinct TLR utilization by different cell types for activation of the type I IFN vs. inflammasome pathways of inflammation.

Figure 1. Monocytes produce inflammasome cytokines in response to HCV and HIV.

(A) HIV and HCV uninfected PBMC, sorted monocytes, or in vitro differentiated macrophages were cultured with plasma from HCV infected subjects (V, n = 15, plasma vol. creates at the final concentration of 5×10⁵ HCV RNA IU/mL in culture) or plasma from the same subjects prior to infection (Pre, n = 15, matched volume). IL-18 and IL-1β were measured in duplicate after 24 h. Primary human hepatocytes were cultured with V and Pre plasma as well as the culture strain HCV_JFH-1 (n = 5). Monocytes and derived macrophages secrete IL-18 (top panel) and IL-1β (lower panel) in response to V but not Pre plasma. Hepatocytes fail to produce inflammasome cytokines. In (B), IL-18 and IL-1β were measured after HIV and HCV uninfected PBMC, sorted monocytes, T-cells or in vitro differentiated macrophages were cultured with plasma from subjects on HAART (H, n = 6), plasma from elite suppressors (ES, n = 5), or viremic HIV plasma (V, n = 15). Plasma with measurable HIV stimulated significant monocyte IL-18 and IL-1β secretion. IL-18 produced from cells cultured with individual plasma samples (white circles), means (horizontal dash) ± S.D. are shown.

Figure 2. HCV and HIV virions stimulate monocytes to produce IL-18.

HCV viremic plasma (A) or HCV_JFH-1 (B) was separated into fractions by sucrose gradient equilibrium ultracentrifugation. HCV RNA (black circles) for each fraction was determined by RT-PCR. Monocytes are maximally stimulated to secrete IL-18 (grey bars, n = 3) when cultured with fractions in the density range 1.09–1.16 g/mL (*), which are most enriched in HCV RNA. (c) HIV_IIIB, HIV_MN, HIV_RF and HIV_BaL were cultured in activated CD4⁺ T-cells, p24 measured in supernatant, and supernatant transferred to monocytes. Monocytes cultured with HIV_IIIB (black circles), HIV_MN (white triangles), HIV_RF (black squares) and HIV_BaL (white diamonds) culture supernatant secrete IL-18 in a dose dependent manner. Symbols represent the mean ± S.D. of 6 experiments.

Figure 3. Inflammasome activation requires viral entry by clathrin-mediated-endocytosis.

(A) The HIV entry receptor CD4 was blocked on monocytes with mAb prior to addition of HIV_BaL culture supernatant (HIV_BaL,). IL-18 production at multiple anti-CD4 mAb concentrations is shown relative to isotype control-treated monocytes. Symbols represent the mean ± S.D. of n = 4 experiments, (black triangles represents the mAb concentration previously determined to prevent HIV infection of CD4+ T-cells. CD4 blockade did not alter IL-18 production from monocytes at concentrations >100 fold that required to block infection. (B) The HCV entry receptor CD81 was similarly neutralized using anti-CD81 mAb. CD81 blockade did not alter IL-18 production from monocytes cultured with plasma from HCV_{Subject 180} (HCV_{Subject 180}). Symbols represent the mean ± S.D. of n = 4 experiments. (C) Receptor-mediated HIV entry or fusion was inhibited by adding Maraviroc (MVC) or T20, respectively to cultures of HIV_BaL and monocytes. IL-18 production was measured after 24 h with no inhibition seen. Bars represent the mean ± S.D. of n = 8 replicates. Clathrin-mediated or clathrin-independent endocytosis was inhibited by pre-incubating monocytes with MβCD, Genistein or Dynasore then adding (D) HIV_BaL or (E) HCV_{Subject 180}. IL-18 production was significantly reduced by Dynasore and MβCD, suggesting the clathrin-mediated endocytosis of the virus is required for inflammasome activation by HCV and HIV. (F) LPS was added to treated cells and IL-18 or TNF-α production measured after 24 h to assess cellular function independent of inflammasome activation. Bars represent the mean ± S.D. of n = 8 replicates, (*) denotes comparisons with p≤0.05 compared to the untreated cells.

Figure 4. Differential importance of endosomal TLRs in inflammasome sensing of HIV and HCV.

Monocytes were cultured without stimulation or with HIV_BaL or HCV_{Subject 180} and IL-1β, IFNα and measured at multiple timepoints. Fold change in expression of IL-1β (black bar) and IFNα (grey bar) following HIV_BaL or HCV_{Subject 180} relative to stimulation with media alone is shown at 6 h (A). Functional knockdowns of the endosomal located TLRs in monocytes were generated by RNA interference. Monocytes were mock transfected or transfected with either siRNA targeting TLR7 or a non-targeting sequence (Scramble). After 24 h, cell lysates were prepared and efficacy of knockdown determined by (B) western blot and (C) qRT-PCR. (D) Specificity of the TLR7 siRNA was confirmed by qRT-PCR using primers for TLR3, TLR7, TLR8, and TLR9. (*) denotes comparisons with p≤0.05 compared to the mock transfected cells. Monocytes in which endosomal TLR knockdown were generated were cultured with HIV_BaL (solid bars) or HCV_{Subject 180} (hatched bars) and pro-IL-1β mRNA transcription measured at 6 h (E–H) and IL-18 secretion measured at 24 h (I–L). Shown are the relative production of pro-IL-1β mRNA and IL-18 in TLR8 (E, I), TLR7 (F, J), TLR3 (G, K) and TLR9 (H, L) knockdown monocytes normalized to mock transfected monocytes (no siRNA) stimulated with the same viruses. Bars represent the mean ± S.D. for n = 6–9 independent transfection experiments, (**) denotes comparisons with p≤0.05 and (***) denoted p≤0.001 compared to the scramble siRNA transfected cells.

Figure 5. MyD88 is required for inflammasome sensing of HIV and HCV.

Knockdowns of the TLR adaptors MyD88 (A, B) and TICAM-1 (TRIF, E, F) were generated and confirmed by previously described RNA interference techniques. (*) denotes comparisons with p≤0.05 compared to the mock transfected cells. Monocytes in which MyD88 was knocked down were cultured with HIV_BaL (solid bars) or HCV_{Subject 180} (hatched bars) and pro-IL-1β mRNA transcription measured at 6 h (C, G) and IL-18 secretion measured at 24 h (D, H). Shown are the relative production of pro-IL-1β mRNA and IL-18 in MyD88 (C, D) or TICAM-1 (G, H) knockdown monocytes normalized to mock transfected monocytes (no siRNA) stimulated with the same viruses. Bars represent the mean ± S.D. for n = 6–9 independent transfection experiments, (**) denotes comparisons with p≤0.001 compared to the scramble siRNA transfected cells.

Figure 6. Differential importance of cytoplasmic sensors in inflammasome response to HIV and HCV.

Functional knockdowns of the cytoplasmic sensors NLRP3, RIG-I and AIM2 in monocytes were created as previously described for TLRs. (A) Knockdown monocytes were cultured with HIV_BaL, HIV_RF, HIV_IIIB, or HIV_RF and IL-18 measured after 24 h. IL-18 produced relative to that of mock transfected cells is shown. (B) Similarly prepared monocytes were cultured with a panel of plasma from HCV infected individuals (HCV_{Subject 117}, HCV_{Subject 54}, HCV_{Subject 16}, HCV_{Subject 180}) and IL-18 measured at 24 h. Bars represent the mean ± S.D. for n = 6–9 independent transfection experiments.