Skin colonization with beta papilloma virus drives tissue immunity and resistance to squamous cell cancer

Vast numbers of highly diverse bacterial, fungal and viral microorganisms inhabit our body’s barrier tissues where they constantly interact with the immune system. Critically, exposure to commensal microbes or low-level infection with persisting pathogens can promote tissue homeostasis and enhance resistance to unrelated pathogens. However, whether this also extends to protection against the carcinogenic effects of environmental factors such as UV irradiation or toxic agents has remained unclear. In a recent study published in Nature, Strickley and colleagues provide an intriguing example of how constitutive immune activation by a persisting but controlled skin virus, beta papillomavirus, can drive efficient immunity against the development of squamous cell carcinoma (SCC). As such, their results provide a fresh view on the clinical association between beta human papilloma virus (HPV) infection and SCC in human patients. SCC is a form of epithelial skin cancer linked to UV irradiation. Immunocompromised individuals are at higher risk of developing SCC and insufficient immunity translates to heightened morbidity, with advanced cutaneous SCC posing a significant risk to organ transplant recipients. Of note, SCC lesions commonly harbor beta HPV, which has led to the proposal that beta HPV infection was a cofactor in SCC development. However, in contrast to the well-established role for high-risk alpha strains of HPV in promoting cervical cancer, evidence for a causal link between beta HPV and skin cancer has been missing. To address this question, Strickley et al. started off by colonizing mouse skin with the murine beta papilloma virus strain MmuPV1. While genetically immune-deficient mice developed large numbers of MmuPV1-induced warts, most wildtype mice completely resisted wart development or developed immunity over longer periods which ultimately resulted in wart regression despite virus persistence in the skin epidermis (Figure 1). Virus control was T-cell dependent and immunity could be boosted by transfer of CD8 T cells from immune mice to recipient mice that previously failed to mount sufficient immunity. It further coincided with the generation of epidermal CD8 T cells displaying a CD103 phenotype consistent with tissue-resident memory T (TRM) cells. Intriguingly, immunosuppressive UV irradiation triggered wart development in previously immune mice, emphasizing that suppression of warts relied on ongoing immune control, most likely mediated by TRM cells. The authors went on to show that immune-competent wild-type mice that successfully suppressed wart development were protected from chemically or UV-induced carcinogenesis. This was reflected in a lower tumor burden and incidence, and overall slower kinetics in virusimmune mice than in nonimmune mice. This remarkable finding held true across various genetically distinct strains of MmuPV1-immune mice. Thus, their results paint a picture where TRM cell-driven virus control in the form of a dynamic host– pathogen equilibrium promotes broader tissue immunity, extending to the suppression of skin cancer (Figure 1). Consistent with their preclinical data, Strickley et al. found that normal human skin harbored high numbers of beta HPV-reactive TRM cells. Those cells were significantly reduced in immunosuppressed individuals such as organ transplant recipients, which at the same time presented with increased beta HPV load in skin. Interestingly, beta Correspondence Thomas Gebhardt, Department of Microbiology & Immunology, The University of Melbourne at the Peter Doherty Institute for Infection & Immunity, 792 Elizabeth Street, Melbourne, VIC 3000, Australia. E-mail: gebhardt@unimelb.edu.au

Vast numbers of highly diverse bacterial, fungal and viral microorganisms inhabit our body's barrier tissues where they constantly interact with the immune system. Critically, exposure to commensal microbes or low-level infection with persisting pathogens can promote tissue homeostasis and enhance resistance to unrelated pathogens. 1,2 However, whether this also extends to protection against the carcinogenic effects of environmental factors such as UV irradiation or toxic agents has remained unclear. In a recent study published in Nature, Strickley and colleagues provide an intriguing example of how constitutive immune activation by a persisting but controlled skin virus, beta papillomavirus, can drive efficient immunity against the development of squamous cell carcinoma (SCC). 3 As such, their results provide a fresh view on the clinical association between beta human papilloma virus (HPV) infection and SCC in human patients.
SCC is a form of epithelial skin cancer linked to UV irradiation.
Immunocompromised individuals are at higher risk of developing SCC and insufficient immunity translates to heightened morbidity, with advanced cutaneous SCC posing a significant risk to organ transplant recipients. 4 Of note, SCC lesions commonly harbor beta HPV, 5 which has led to the proposal that beta HPV infection was a cofactor in SCC development. However, in contrast to the well-established role for high-risk alpha strains of HPV in promoting cervical cancer, evidence for a causal link between beta HPV and skin cancer has been missing. To address this question, Strickley et al. started off by colonizing mouse skin with the murine beta papilloma virus strain MmuPV1. While genetically immune-deficient mice developed large numbers of MmuPV1-induced warts, most wildtype mice completely resisted wart development or developed immunity over longer periods which ultimately resulted in wart regression despite virus persistence in the skin epidermis 3 (Figure 1).
Virus control was T-cell dependent and immunity could be boosted by transfer of CD8 + T cells from immune mice to recipient mice that previously failed to mount sufficient immunity. It further coincided with the generation of epidermal CD8 + T cells displaying a CD103 + phenotype consistent with tissue-resident memory T (T RM ) cells. Intriguingly, immunosuppressive UV irradiation triggered wart development in previously immune mice, emphasizing that suppression of warts relied on ongoing immune control, most likely mediated by T RM cells. The authors went on to show that immune-competent wild-type mice that successfully suppressed wart development were protected from chemically or UV-induced carcinogenesis. This was reflected in a lower tumor burden and incidence, and overall slower kinetics in virusimmune mice than in nonimmune mice. This remarkable finding held true across various genetically distinct strains of MmuPV1-immune mice. 3 Thus, their results paint a picture where T RM cell-driven virus control in the form of a dynamic hostpathogen equilibrium promotes broader tissue immunity, extending to the suppression of skin cancer ( Figure 1).
Consistent with their preclinical data, Strickley et al. found that normal human skin harbored high numbers of beta HPV-reactive T RM cells. Those cells were significantly reduced in immunosuppressed individuals such as organ transplant recipients, which at the same time presented with increased beta HPV load in skin. Interestingly, beta HPV-DNA and transcripts were significantly reduced in SCC tumors compared with surrounding skin and similar observations were made in a proportion of MmuPV1immune mice that eventually developed warts. 3 Together, these results imply selection against virusinfected keratinocytes and thus provide a novel link between antiviral immune surveillance by T RM cells and localized protection from cancer development (Figure 1).
The mode of T RM control in driving protection from both persisting virus infection and skin carcinogenesis remains to be defined. Malignant transformation and increased division of cancerous keratinocytes may result in stronger or altered presentation of papilloma virus antigens and thus T-cell receptor-dependent T RM activation. At the same time, T RM cells may recognize a broad array of stress molecules which are likely to be expressed at elevated levels by virusinfected SCC cells, including the NKG2D ligands MICA (human) and Rae-1 (mouse). Therefore, T RM cellmediated protection may at least in part be a result of direct recognition of virus-infected cancer cells, explaining the apparent selection of virus-negative escape variants during eventual SCC development. In addition, intermittent or continuous low-level T RM stimulation by virus persisting in normal keratinocytes may promote innate immune mechanisms that suppress SCC development in an antigen nonspecific manner. Consistent with this idea, strong T RM cell activation by cognate antigens can elicit potent nonspecific antiviral immunity. 6,7 Likewise, T RM cells induced by colonization of mouse skin with Staphylococcus epidermidis have been shown to enhance immunity toward infection with Candida albicans. 1 Finally, while T RM cellmediated immunity may result in rapid elimination of malignant keratinocytes during transformation, T RM cells may also act to control microscopic and clinically occult SCC lesions in a cancer-immune equilibrium, as recently shown in a mouse model of melanoma. 8 Overall, the study by Strickley et al.
provides an intriguing alternative explanation for the association of beta HPV with SCC by demonstrating that virus colonization of skin facilitates immune control of cancer. This type of cross-protective tissue immunity is orchestrated by a localized pool of T RM cells, which echoes the welldocumented functions of T RM cells in other types of infections and cancers, as well as overall tissue homeostasis. 9 It is therefore tempting to speculate that skin colonization with other commensal microorganisms may similarly promote T RM -mediated immunity to SCC and potentially other cancers. Future studies will have to address these questions and will thereby open new avenues for cancer immunotherapies by harnessing tissue immunity to commensal microflora. Importantly in this respect, Masopust and colleagues recently demonstrated that viral

Cross-protection
Immune escape peptide stimulation of tumorassociated T RM cells can elicit highly effective cancer immunity. 10 Finally, although mechanistically somewhat different, these strategies are related to recent attempts at exploiting the intriguing link between gut microbiota and improved responses to checkpoint blockade cancer immunotherapy.
These are fascinating examples of how we derive a major benefit from our intimate coexistence with the myriad of microorganisms that inhabit our body, and of how future therapies may successfully harness these interactions to fight cancer.