Science Spotlight

Immune surveillance triggers HIV reactivation

From the Hladik lab, Vaccine and Infectious Disease Division

The latent reservoir established by HIV maintains chronic, incurable infection and protects the virus from complete eradication. A potential strategy for curing HIV is to purposely reactivate virus to allow latently infected cells to be killed by anti-viral drugs or immune cells. Although combination antiretroviral therapies (cART) can suppress HIV viral load below the limit of detection, the latent virus naturally rebounds at low levels in periodic “blips” throughout treatment. However, it is not well understood which conditions facilitate viral reactivation. To investigate the triggers of viral blips, Dr. German Gornalusse and colleagues from the Hladik lab in the Vaccine and Infectious Disease Division studied HIV-1 reactivation in the female genital tract (FGT), a site of HIV infection, and found that epithelial cells in the FGT produce immune system signals that reactive latent HIV-1 infection in T cells. These findings, published in the Journal of Virology, also inform strategies aiming to safely reactivate latent virus in a targeted manner for curative therapies.

HIV commonly infects T cells in the female FGT, where HIV activity persists throughout infection. Reasoning that the cells within the FGT may influence latent HIV reactivation, the authors first established primary epithelial cells lines from healthy FGT tissue.  They co-cultured supernatant from these cells with several different cell lines of clonal T cells integrated with HIV-1 constructs expressing GFP under the HIV-1 LTR promoter, providing a reporter for simulated HIV reactivation. They found that incubation with supernatant from unstimulated endocervical, but not vaginal or ectocervical, induced GFP expression in T cells, suggesting that endocervical cells constitutively produce soluble signals that can reactivate HIV-1. Furthermore, co-culture of endocervical cells with HIV-integrated T cells caused HIV-1 reactivation at a higher level than supernatant alone, demonstrating that direct cell contact between FGT endocervical cells and infected T cells enhances reactivation. This effect was lessened when endocervical cells were lysed, showing that intact cells, and not their products or antigens, most potently reverse HIV-1 latency in T cells.

Tumor necrosis factor alpha (TNF-a), a cytokine produced by many immune cells, is known to activate HIV-1. To determine if endocervical-derived TNF-a was responsible for HIV-1 reactivation, the authors next co-cultured endocervical cells and HIV-integrated T cells with increasing concentrations of anti-TNF-a antibody. Predictably, they found that the highest concentrations of blocking antibody almost completely prevented HIV-1 reactivation, demonstrating that TNF-a derived from endocervical cells is partially responsible for reversing latency in T cells. To expand on this mechanism, the authors asked whether herpes simplex virus 2 (HSV-2) infection, which is known to exacerbate HIV-1 infection, could further stimulate TNF-a production by endocervical cells and induce even more HIV-1 reactivation. After demonstrating that HSV-2 infection increases endocervical epithelial cell TNF-a production, they co-cultured supernatant from HSV-2-infected endothelial cells with the HIV-integrated T cells. HIV-1 reactivation, measured by digital droplet PCR, was enhanced with HSV-2 infection. This enhancement was only partially reversed when the epithelial cells were treated with acyclovir (ACL), an antiviral used to treat HSV-2, which could be one explanation why acyclovir treatment of HIV-1/HSV-2 co-infected individuals did not decrease HIV transmission in past studies.

"The human endocervix is a “hot spot” for HIV-1 reactivation. Columnar epithelial cells lining the endocervical canal are targeted by certain stimuli like herpes simplex virus or bacterial flagellin and release inflammatory mediators, among them tumor necrosis factor alpha (TNF-α, in light blue). These soluble molecules can reactivate HIV-1 directly or expand the pool of latently infected T cells by inducing proliferation. The study draws attention to columnar epithelia, such as in the endocervix as demonstrated here or, by inference, possibly throughout the intestinal mucosa, as anatomical sites that require particular attention when attempting to achieve a functional cure of HIV infection," explained Dr. Gornalusse.
"The human endocervix is a “hot spot” for HIV-1 reactivation. Columnar epithelial cells lining the endocervical canal are targeted by certain stimuli like herpes simplex virus or bacterial flagellin and release inflammatory mediators, among them tumor necrosis factor alpha (TNF-α, in light blue). These soluble molecules can reactivate HIV-1 directly or expand the pool of latently infected T cells by inducing proliferation. The study draws attention to columnar epithelia, such as in the endocervix as demonstrated here or, by inference, possibly throughout the intestinal mucosa, as anatomical sites that require particular attention when attempting to achieve a functional cure of HIV infection," explained Dr. Gornalusse. Figure provided by Dr. German Gornalusse

After showing that infection with HSV-2 triggers TNF-a production in FGT cells, the authors investigated the broader innate immune responses within the endocervical epithelium. Epithelial cells are equipped with a wide range of sensors such as Toll-like receptors (TLR) that sense pathogen stimuli known as pathogen-associated molecular patterns (PAMPs) and trigger innate immune responses. To test endocervical cell responses to pathogens, the authors exposed the epithelial cells to various PAMPs known to trigger specific TLRs. They found that the cells responded strongly to polyinosinic-polycytidylic acid (PIC), a synthetic analog of double-stranded viral RNA, and a ligand of TLR3. In addition to stimulating TNF-a, PIC also induced several other inflammatory chemokines from endocervical epithelial cells, providing additional mechanisms by which latent HIV-1 in T cells may be activated. Finally, to test the impact of whole RNA virus on triggering epithelial cytokine production, they infected endocervical cells with hepatitis C virus (HCV) and Sendai virus (SeV), and found that only SeV triggered TNF-a production. Together, these results suggest that endocervical epithelial cells selectively respond to certain innate immune stimuli and produce TNF-a and other cytokines that may, in turn, reactivate latent HIV-1 in T cells within the FGT environment.

This study demonstrates that the endocervix may be a hotspot for reversing HIV-1 latency in adjacent T cells, and that reactivation is at least partially driven by epithelial production of TNF-a. Additionally, the initial triggers for epithelial TNF-a production may be viruses such as HSV-2 or SeV. These results suggest that the endocervical epithelium is a potent site for innate immune surveillance, but that an anti-viral response to one pathogen may concurrently reactivate latent HIV-1. Understanding the conditions for HIV-1 latency reversal could be harnessed to either prevent HIV-1 reactivation to maintain a functional cure or, conversely, induce reactivation for targeted HIV-1 eradication. Further studies are necessary to understand which additional PAMPs or cytokines may lead to HIV-1 reactivation in the endocervix, and to determine if other anatomical sites with similar columnar epithelia, such as the intestinal mucosa, also cause HIV-1 reactivation from adjacent CD4+ T cells or macrophages.

This work was supported by the National Institutes of Health.

UW/Fred Hutch Cancer Consortium members David Koelle, Steve Polyak, David Fredricks, Julie McElrath, Anna Wald, and Florian Hladik contributed to this work.

Gornalusse GG, Valdez R, Fenkart G, Vojtech L, Fleming L, Pandey U, Hughes S, Levy C, Dela Cruz EJ, Calienes F, Kirby AC, Fialkow MF, Lentz GM, Wagoner J, Jing L, Koelle DM, Polyak SJ, Fredricks DN, McElrath MJ, Wald A, Hladik F. 2020. Mechanisms of endogenous HIV-1 reactivation by endocervical epithelial cells. Journal of Virology. 2020 Feb 12. pii: JVI.01904-19. doi: 10.1128/JVI.01904-19. [Epub ahead of print]