Herpes simplex virus 2 (HSV-2) infects the genital epithelium before traveling up skin-enervating neurons to establish latency. The virus can later reactivate and travel back through the neurons to release virus at the site of infection, causing recurrent viral shedding and lesions. HSV-2 shedding episodes are generally defined by an initial period of rapid viral replication followed by a protracted phase of immune-mediated clearance, but the duration and titer of viral shedding events demonstrate both individual and episodic variation. The reason for this range of viral control is not understood. CD4+ and CD8+ Tissue-resident memory (Trm) T cells establish residency in the genital tissue after HSV-2 infection and are known to rapidly and potently respond to HSV-2, but viral reactivations are randomly seeded throughout genital tissue, and may occur in regions of low Trm density. However, most HSV-2 shedding episodes are usually quickly curtailed before lesion symptoms develop, but it is not known how the virus is efficiently controlled in the absence of high numbers of Trm. To address this question, Dr. Pavitra Roychoudhury and Dave Swan from the Schiffer group, along with colleagues in the Vaccine and Infectious Diseases Division, created a mathematical model to simulate HSV-2 viral spread and its Trm-mediated containment in genital tissue. They recently published this work in the Journal of Clinical Investigation.
The authors first sought to characterize HSV-2 shedding kinetics within a genital microenvironment. Using HSV-2 titers measured from vaginal swabs collected by HSV-2+ patients every six hours for 60 days, they found that all episodes shared an early peak in viral replication, but that the amount of virus present and the duration of viral shedding varied between samples. Although most shedding episodes were controlled within 20 hours, symptomatic episodes correlated with episodes containing the highest early HSV-2 titers and longer shedding durations. These results confirmed that HSV-2 shedding episodes are defined by initial, robust HSV-2 replication, leading the authors to hypothesize that variance in antiviral immune responses to this shedding may explain why some shedding episodes are rapidly cleared while others persist and cause lesions.
The authors used genital tissue biopsies from 2-8 weeks post-lesion healing to characterize the density of Trm relative to epithelial cells, and found that this effector:target (E:T) ratio varied between samples. Based on biological features of HSV-2 spread within epithelial cells, they next designed a mathematical model that predicts how HSV-2 will spread through a cellular microenvironment given different biologic conditions. Without including any immune response parameter, viral infection of epithelial cells continued unchecked until target cells were either already infected or dead. To link Trm density with the observed heterogeneity in HSV-2 shedding episode kinetics, they incorporated realistic Trm E:T ratios into the model. Based on previously published data, the model assumed Trm randomly patrol the tissue, arrest when they encounter viral antigen, and then proliferate in situ and kill infected cells. However, the model predicted that HSV-2 elimination was less effective than is observed through viral titers in infected tissue, even in simulations with high E:T ratios. This finding prompted the authors to hypothesize that Trm mediate HSV-2 control though not only direct killing but also though other soluble, non-contact-dependent mechanisms.
To determine which Trm-derived cytokines might be responsible for HSV-2 control, the authors returned to the clinical samples and found that Interferon-gamma (IFN-g) and Granzyme B, a pro-inflammatory cytokine and cytotoxic molecule, respectively, were present in high concentrations during viral peaks. The authors then incorporated putative but unproven anti-HSV-2 functions of these cytokines into their model, simulating HSV-2 infection with 500 permutations of IFN-g Granzyme B functions. The models that best fit observed biologic data incorporated at least two putative cytokine-mediated functions, including increasing infected cell death rate and activation of bystander Trm, suggesting that contact-independent mechanisms of infected cell killing are responsible for controlling HSV-2 reactivation.
Further interrogating the antiviral response, the model predicted that, as previously shown, viral reactivation in areas of the tissue with high initial Trm density quickly control viral shedding and prevent lesions through contact-mediated killing, without the need for antiviral cytokine functions. However, shedding events that occur further from Trm, and therefore have more time to replicate before first contact with Trm, are not able to be controlled through direct Trm-mediated killing alone. In these cases, antiviral-cytokines such as IFNg and Granzyme B are necessary to rapidly control the HSV-spread that is already underway.
“This is the first model to reproduce both the viral and molecular immune kinetics of a viral infection, as well as the spatial histopathology of tissue infection,” Dr. Schiffer explained. “We learned that the major mechanism underlying rapid removal of HSV-2 infected cells is likely to be rapid diffusion of antiviral cytokines rather than direct, contact-mediated killing.” This study not only provides a mathematical model that can be altered to study various characteristics of HSV-2 reactivation and the subsequent antiviral immune response but also provides an explanation for how HSV-2 reactivation events are curtailed even when Trm—the first line of defense—density is low. These results provide important immunologic insight for future HSV-2 vaccine research; an effective vaccine must elicit both high numbers of Trm and potent production of antiviral cytokines. Further experiments in HSV-2+ people to identify interferon-stimulated gene signatures and apoptotic cells would strengthen the conclusions generated by this mathematical model.
This work was supported the National Institute of Allergy and Diseases.
UW/Fred Hutch Cancer Consortium members by Larry Corey, Josh Schiffer, Martin Prlic, and Jia Zhu contributed to this work.
Roychoudhury P, Swan DA, Duke ER, Corey L, Zhu J, Davé VA, Richert-Spuhler LE, Lund JM, Prlic M, Schiffer JT. 2020. Tissue-resident T cell derived cytokines eliminate herpes simplex virus-2 infected cells. J Clin Invest. 2020 Mar 3. pii: 132583. doi: 10.1172/JCI132583. [Epub ahead of print]