The curious case of APOBEC3 activation by cancer-associated human papillomaviruses

High-risk human papillomavirus (HR-HPV) infections cause approximately 5% of all human cancers worldwide. These include almost all cervical carcinomas, a leading global cause of cancer deaths, as well as a significant percentage of other anogenital tract cancers and a growing fraction of oropharyngeal carcinomas. To accommodate their life cycles, HR-HPVs need to extensively rewire infected cells. The HR-HPV E6 and E7 proteins are the main drivers of this process, and their expression elicits a barrage of cellular defense responses that restrict this unfriendly takeover of the host cell. Not surprisingly, HR-HPVs have in turn evolved mechanisms to escape or curb antiviral and anti-oncogenic cellular responses. These mechanisms include degradation of the retinoblastoma (RB1) and the p53 (TP53) tumor suppressors by HR-HPV E7 and E6, respectively. If unopposed, RB1 and TP53 would cause cell cycle arrest, senescence, or cell death in response to HR-HPV infection (Fig 1). Other antiviral pathways, including DNA sensing and interferon signaling, are also blunted by HR-HPV E6 and E7 proteins [1]. Surprisingly, HR-HPVs have not evolved strategies to counteract restriction by apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like 3 (APOBEC3, or simply A3). A3s are interferon-regulated DNA cytosine-to-uracil deaminases encoded as a cluster of seven genes (A3A–A3H; there is no A3E) on human chromosome 22, which are all expressed, albeit at vastly different levels, in epithelial cells [2], which are the natural hosts of HPV infection. While their cytidine deaminase activity causes deoxycytidine (C) to deoxythymidine (T) mutations during viral genome synthesis, A3s also restrict viral replication through cytidine deaminase-independent mechanisms [3]. In response, many viruses have evolved mechanisms to evade A3 restriction. The human immunodeficiency virus 1 (HIV1) Viral infectivity factor (Vif) protein targets A3 family members for degradation, and the HIV2 Viral protein X (Vpx) protein targets A3A for degradation [4, 5]. The Hepatitis B Virus X protein impairs this pathway by packaging A3G into exosomes [6]. Human polyomaviruses—including the Merkel Cell Polyomavirus (McPyV)—trigger A3 activity, yet McPyV-positive Merkel cell carcinomas do not show an A3 mutational signature [7, 8]. This strongly suggests that McPyV overrides A3 restriction [8]. While ectopically expressed HR-HPV E7 and E6 have each been reported to increase expression of A3 family members, and A3A can restrict infection with in vitro–generated HPV16 pseudovirions, A3 activity is not blocked by HR-HPVs [3, 9–11]. Despite the fact that HPV genomes contain fewer than predicted A3 recognition sites [12], the mutational drift caused by A3 mutagenesis is extensive; many of the thousands of HPV16 variants that were detected in a recent study exhibit nucleotide changes that are consistent with A3 action [13].