Semaphorin‐3F Deletion Prevents Oral Carcinogenesis

Abstract

The Semaphorin‐3F (Sema3F) gene was originally isolated from chromosome 3p21 in a region commonly deleted in small cell lung cancer and was subsequently shown to act as a tumor suppressor. Sema3F is expressed in epithelial cells and is a ligand of the Neuropilin‐2 (Nrp2) receptor. We have previously shown that normal epithelial cells lack Nrp2, but upregulate this receptor during the dysplastic stage of carcinogenesis. Nearly all cancers in the oral cavity are oral squamous cell carcinomas (OSCC) which highly express Nrp2. Preclinical trials performed in our laboratory demonstrate that exogenous SEMA3F protein delivered systemically in mice can regress existing OSCC orthotopic xenografts by inhibiting angiogenesis and tumorigenicity. Here, we seek to understand the role of endogenous Sema3F in oral carcinogenesis and hypothesize that deletion of Sema3F, a putative tumor suppressor, from epithelial cells will increase oral cancer incidence, tumor size, and metastasis. Transgenic mice (K14‐creERT;Sema3Ff/f or K14‐creERT) were treated with tamoxifen and then exposed to carcinogen (4NQO) in their drinking water. Mice were monitored for oral cancer progression and compared between the groups. 100% (10/10) of control mice (K14‐creERT) developed carcinoma in situ or invasive OSCC in the tongue. However, surprisingly 0% (0/8) of mice with inducible K14‐Sema3F‐KO developed OSCC. Since no cancers were found in the mice lacking Sema3F, we predicted that immune cells may be clearing mutated epithelial cells during immune surveillance before they had a chance to grow. To examine the effect of endogenous Sema3F loss on T‐lymphocytes (which express Nrp2), we examined Type IV hypersensitivity assays in control and K14‐Sema3F‐KO mice. Mice lacking Sema3F showed enhanced and prolonged swelling in the ears after antigen challenge, while control mice resolved more quickly. Immunostaining of inflamed ear cryosections showed increased CD4+ T cell infiltration in mice deficient in Sema3F compared to controls. Taken together, our data suggests that Sema3F does not function as a tumor suppressor in all epithelial cells and appears to act indirectly to promote oral carcinogenesis by repressing immune cell activation. Thereby, Sema3F may play duplicitous roles in the tumor microenvironment by inhibiting angiogenesis yet suppressing immune surveillance.