Innate Immune Activation by cGMP-AMP Nanoparticles Leads to Potent and Long-Acting Antiretroviral Response against HIV-1

Abstract

HIV-1 evades immune detection by the cGAS-STING cytosolic DNA-sensing pathway during acute infection. STING is a critical mediator of type I IFN production, and STING agonists such as cGMP-AMP (cGAMP) and other cyclic dinucleotides elicit potent immune and antitumor response. In this article, we show that administration of cGAMP, delivered by an ultra-pH-sensitive nanoparticle (NP; PC7A), in human PBMCs induces potent and long-acting antiretroviral response against several laboratory-adapted and clinical HIV-1 isolates. cGAMP-PC7A NP requires endocytosis for intracellular delivery and immune signaling activation. cGAMP-PC7A NP-induced protection is mediated through type I IFN signaling and requires monocytes in PBMCs. cGAMP-PC7A NPs also inhibit HIV-1 replication in HIV⁺ patient PBMCs after ex vivo reactivation. Because pattern recognition receptor agonists continue to show more clinical benefits than the traditional IFN therapy, our data present important evidence for potentially developing cGAMP or other STING agonists as a new class of immune-stimulating long-acting antiretroviral agents.

Figure 1. cGAMP-PC7A nanoparticles elicit potent antiretroviral response against HIV-1 in human PBMCs

(A) Structures of the ultra-pH-sensitive copolymers. (B) Size changes of cGAMP-loaded micelle nanoparticles in buffers at pH above or below pKa (n = 3). (C) Transmission electron micrographs images of cGAMP-loaded PEG-b-PC7A nanoparticles at pH 7.4 and 6.7. Scale bar = 100 nm. (D) A schematic diagram of HIV-1 spreading assay in PBMCs. (E) HIV-BaL replication in PBMCs treated with indicated reagents (right). NP, nanoparticle. Nev, nevirapine (4 µM, same throughout). All NPs were used at 1 µg/mL. (F) HIV-BaL replication in PBMCs treated with increasing amount of cGAMP-NP (PC7A, same below). Two representative donors are shown. (G) HIV-BaL replication in PBMCs treated with indicated reagents (bottom). cGAMP was used at 1 µg/ml. cGAMP-NP was used at 1 µg/ml, and equivalent amount of NP alone was used as control. Each data point represents an individual donor. (H, I) HIV-IIIB (H) or HIV-LAI (I) replication in PBMCs treated with Nev or cGAMP-NP. Assay performed as illustrated in D. Data are representative of at least three independent experiments with independent donors. ***, p<0.005. ns, not significant. Student’s t-test. Error bar, SEM.

Figure 2. cGAMP-induced antiretroviral response in PBMCs is mediated through type I IFN signaling

(A) HIV-BaL replication in PBMCs treated with cGAMP-NP or TLR agonists. (B) A schematic diagram of ‘conditioned media’ assay used in C. (C) HIV-BaL replication in PBMCs. Conditioned media from first collected from PBMCs treated with indicated reagents (right) for two days, then, these media (containing soluble factors) were mixed with HIV-BaL to infect fresh PBMCs. (D) A heat map of secreted cytokines levels after cGAMP or poly(I:C) treatment in C. (E) B18R inhibits cGAMP-mediated antiretroviral response against HIV-BaL. HIV-1 spreading assay was performed as in Figure 1D. Indicated reagents were added at day 0. Data are representative of at least three independent experiments with independent donors.

Figure 3. cGAMP-induced antiretroviral response in PBMCs is dependent on monocytes

(A) Quantitative RT-PCR analysis of IFNB mRNA in indicated cell types. Monocyte, B and CD4+ T cells were isolated from PBMCs by positive selection. Cells were then treated with cGAMP-Lipo (4 µg/ml) and IFNB expression was measured at indicated times. (B) Immunoblots analysis of STING, TBK1, IRF3 protein expression in whole PBMCs and sorted individual cell populations as indicated on top. Equal amount of proteins were loaded for all samples based on the BCA assay. (C,D) HIV-BaL replication in PBMCs depleted of indicated population. Individual data from a representative donor and representative FACS plots confirming depletion efficiency are shown in Supplemental Figure 3. cGAMP-Lipo was used in C and cGAMP-NP was used in D. Data are representative of at least three independent experiments each with 3–4 independent donors. *, p<0.05. ns, not significant. Student’s t-test. Error bar, SEM.

Figure 4. cGAMP is more potent than c-di-GMP at stimulating immune gene expression in PBMCs

(A, B) Immunoblots (A) and quantitative RT-PCR analysis (B) of STING signaling activation in PBMCs by cGAMP (2 µg/ml), c-di-GMP (2 µg/ml) or htDNA (1 µg/ml). Cells were treated with the indicated ligand for 4, 7 or 24 hours as indicated (4 and 24 h for immunoblots, 7 h for qRT-PCR). (C, D) Immunoblots (C) and quantitative RT-PCR analysis (D) of STING signaling activation in THP-1 cells (monocyte cell line) by cGAMP (2 µg/ml), c-di-GMP (2 µg/ml) or htDNA (1 µg/ml). Cells were treated with the indicated ligand for 4, 7 or 24 hours as indicated (4 and 24 h for immunoblots, 7 and 24 h for qRT-PCR). Data are representative of at least two independent experiments.

Figure 5. cGAMP-NP elicits long-acting antiretroviral response against HIV-1 in PBMCs

(A) A schematic diagram of modified HIV-1 spreading assay to include treatment interruption after 1 week (‘long-acting’ experiment). (B, C) HIV-BaL replication in PBMCs treated with Nev (4 uM) or cGAMP-NP (2 µg/mL). Cells were washed with fresh media at day 7, and viral replication was continually monitored till day 27. One representative donor is shown in B. Summary from multiple donors is shown in C. (D) HIV-BaL replication in PBMCs treated with Nev (4 uM) or cGAMP-NP (2 µg/mL) with or without B18R (500 ng/mL). Experimental set up is similar to B. Data are representative of at least three independent experiments with independent donors. **, p<0.01. ns, not significant. Student’s t-test. Error bar, SEM.

Figure 6. cGAMP-NP inhibits HIV-1 replication ex vivo

(A) Quantitative RT-PCR array analysis of indicated immune genes in PBMCs isolated from healthy or HIV-positive donors before and after cGAMP-NP treatment (2 µg/ml) for 24 hours. n=3–4. (B) HIV-1 replication in patients PBMCs ex vivo after re-activation for 7 days and indicated treatment. n=2. (C) HIV-1 replication in patients PBMCs ex vivo after re-activation overnight and immediately followed by indicated treatment. HIV replication was measured in the media at indicated days after reactivation. n=2.