Human papillomavirus 16 E7 protein inhibits interferon-c-mediated enhancement of keratinocyte antigen processing and T-cell lysis Fang Zhou, Graham R. Leggatt and Ian H. Frazer The University of Queensland Diamantina Institute, Princess Alexandra Hospital, Brisbane, Qld, Australia Introduction Persistent infection of the cervical epithelium with one of a range of oncogenic human papillomaviruses (HPVs) can initiate cervical cancer. The majority of high-risk papillomavirus infections of immunocompe- tent individuals are cleared, although this can take up to 4 years, and a minority of apparently immunocom- petent individuals will develop persisting infection [1]. These data suggest that HPV may have evolved mech- anisms to enable infected epithelial cells to escape from immune surveillance in vivo [2]. We established a skin graft model to study the role in immune evasion of the HPV 16 nonstructural protein E7, which is overexpres- sed in premalignant lesions associated with HPV 16 infection. In this model, murine skin expressing HPV 16 E7 as the product of a transgene in keratino- cytes (KCs) from a keratin 14 promoter is grafted into naı ¨ ve, otherwise syngeneic, mice. These mice fail to reject such grafts, either spontaneously or after immu- nization with E7 [3]. In contrast, skin grafts similarly expressing ovalbumin (OVA) as a transgene product are spontaneously rejected [4]. In vitro, HPV 16 E7- specific cytotoxic T lymphocytes (CTLs) fail to kill HPV 16 E7 transgenic KCs efficiently, but can kill KCs pulsed with the dominant H-2D b restricted E7 peptide epitope [5] or with OVA [6]. A number of studies have shown that HPV gene expression can interfere with various components of antigen process- ing by uncertain mechanisms [7–9]. Taken together, these results suggest that the HPV 16 E7 may inhibit cellular antigen processing and presentation to enable Keywords antigen processing; CD8 T cells; host–virus interactions; papillomavirus Correspondence I. Frazer, The University of Queensland Diamantina Institute, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Brisbane, Qld 4102, Australia Fax: +61 7 3240 5946 Tel: +61 7 3240 5310 E-mail: i.frazer@uq.edu.au (Received 14 July 2010, revised 2 December 2010, accepted 10 January 2011) doi:10.1111/j.1742-4658.2011.08011.x Infection of epithelium with human papillomavirus (HPV) 16 is generally prolonged, suggesting an ineffective virus-specific immune response, and prolonged infection promotes anogenital cancer. To determine whether poor antigen presentation by HPV-infected keratinocytes (KCs) contributes to prolonged HPV infection, KCs and KCs expressing HPV 16 E7 protein (E7-KCs) were compared for susceptibility to T-cell-mediated lysis directed to ovalbumin (OVA) processed for presentation by the KCs. Interferon (IFN)-c efficiently enhanced susceptibility to lysis of KCs presenting OVA, but not of E7-KCs similarly presenting OVA. E7-KCs also exhibited impaired IFN-c-induced upregulation of transcription of major histocom- patibility complex class I antigen processing and presentation-associated genes, and of membrane SIINFEKL–H-2K b complexes. Thus, expression of HPV 16 E7 protein in KCs may inhibit enhancement by IFN-c of KC sensitivity to T-cell lysis, by impairing antigen presentation. Abbreviations CTL, cytotoxic T lymphocyte; E7-KC, keratinocyte expressing HPV 16 E7; HPV, human papillomavirus; IFN, interferon; IRF, interferon regulatory factor; KC, keratinocyte; MFI, mean fluorescence intensity; MHC, major histocompatibility complex; SD, standard deviation; 2-ME, 2-mercaptoethanol. FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS 955 HPV to evade viral antigen-specific host immune responses. Several viruses have been demonstrated to interfere with antigen processing and presentation by infected cells, through inhibition of peptide recruitment to and processing by the proteasome, or through blocking the production and transport of major histocompatibility complex (MHC) class I complexes [10–13]. Interferon (IFN)-c plays an important role in facilitating CTL- mediated immune effector function, through induction of multiple genes associated with MHC class I antigen presentation [14], and also contributes to the elimina- tion of HPV infection [15]. HPV 16 E7 has been shown to interfere with the transduction of IFN signaling [16–19]. HPV 16 E7 binds to the C-terminal transactivation domain of interferon regulatory factor (IRF)-1 in vitro, and cul- tured fibroblasts transduced with HPV 16 E7 demon- strate reduced transcription of some IFN-c-inducible gene products following overexpression of IRF-1 when compared with untransduced cells [20]. In HPV 18 E6 ⁄ E7 transgenic mice, there is some evidence for reduced transcription of the same gene products in the cervix, as compared with nontransgenic animals [21]. We therefore investigated, using OVA as a model anti- gen, whether HPV 16 E7, when expressed in KCs at levels similar to those found in HPV infection, could inhibit enhancement by IFN-c of antigen processing and presentation, and CTL-mediated killing of KCs expressing non-self-antigen. Results E7 does not affect the ability of IFN- c to upregulate MHC class I expression and exogenous CTL epitope presentation on KCs We first wished to investigate whether expression of HPV 16 E7 in KCs would alter the expression of mem- brane-associated MHC, the presentation of processed endogenous antigen as a peptide in association with MHC, or CTL-mediated lysis directed at cell mem- brane MHC–peptide complexes. As IFN-c can upregu- late MHC class I expression on keratinocytes [22] and enhances CTL epitope presentation and CTL-mediated lysis of KCs [6], we also wished to investigate whether E7 interfered with IFN-c-induced enhancement of anti- gen processing and presentation. Therefore, KCs and KCs from H-2 b mice transgenic for HPV 16 E7 expressed from a keratin 14 promoter (E7-KCs) were exposed in vitro to IFN-c. We first examined the effect of E7 on IFN-c-mediated induction of MHC expres- sion. Induction by IFN-c of the expression of MHC class I on cultured E7-KCs was similar to induction on nontransgenic KCs (Fig. 1A). We then exposed KCs and E7-KCs to IFN-c and SIINFEKL, a peptide A CD B Fig. 1. Lysis of KCs and E7-KCs treated with IFN-c. Mouse KCs were exposed to SIINFEKL and IFN-c as shown. (A) Expression of MHC class I on KCs and E7-KCs after exposure to IFN-c, as shown, was assessed by flow cytometry with an H-2K b -specific antibody. (B) Expres- sion of SIINFEKL–H-2K b peptide complexes on KCs and E7-KCs after incubation with 10 lM SIINFEKL, and IFN-c, as shown, was assessed with mouse antibody against SIINFEKL–H-2K b . (C) Expression of SIINFEKL–H-2K b complexes on KCs and E7-KCs after incubation with IFN-c at 100 UÆmL )1 and SIINFEKL, as shown, was assessed. (D) Susceptibility of SIINFEKL-exposed KCs to lysis by SIINFEKL-specific CTLs at an effector ⁄ target ratio of 1 : 20 was compared for KCs and E7-KCs pulsed with three concentrations of SIINFEKL. The experiments in (A), (B) and (C) were repeated twice, and those in (D) were repeated three times, with similar results. In (D), mean and SD from triplicate determina- tions of percentage lysis for one experiment are shown. HPV 16 E7 inhibits T-cell KC lysis F. Zhou et al. 956 FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS derived from OVA that is able to associate with mem- brane H-2K b MHC complexes without further process- ing. KCs and E7-KCs showed similar dose-dependent increases in the density of membrane SIINFEKL– H-2K b complexes after IFN-c exposure (Fig. 1B). Furthermore, KCs and E7-KCs exposed to a fixed concentration of IFN-c and to increasing concentra- tions of SIINFEKL displayed similar increased density of SIINFEKL–H-2K b complexes (Fig. 1C). By select- ing appropriate concentrations of peptide, we could then compare the susceptibility of KCs and E7-KCs expressing similar densities of SIINFEKL–H-2K b com- plexes to lysis by a set number of E7-specific T cells, to establish whether E7 expression might impact on the sensitivity of KCs to T-cell lytic mechanisms. For any given level of expression of MHC–peptide com- plex, KCs and E7-KCs were equally susceptible to lysis in vitro by SIINFEKL-specific CTLs (Fig. 1D). Thus, endogenous expression of E7 has no effect on IFN-c- induced enhancement of MHC expression by KCs, or on the susceptibility of KCs pulsed with exogenous SIINFEKL to CTL-mediated lysis. E7 inhibits IFN-c-dependent upregulation of presentation of endogenous antigen by KCs As there was no observed effect of endogenous E7 on MHC expression or on the susceptibility of KCs expressing MHC–peptide complexes to T-cell-mediated lysis, we next wished to test the hypothesis that endog- enous E7 expression in KCs might inhibit the process- ing and presentation of endogenously expressed protein. We therefore compared KCs and E7-KCs, each also expressing OVA endogenously as a transgene product, for susceptibility to lysis by OVA-specific T-cells, both with and without IFN-c pretreatment. E7 and OVA double-transgenic KCs and OVA single- transgenic KCs, if not treated with IFN-c, were simi- larly susceptible to lysis by OVA-specific T cells (Fig. 2A). However, lysis of OVA transgenic KCs by OVA-specific T cells was significantly increased follow- ing IFN-c exposure, whereas lysis of E7 and OVA double-transgenic KCs was not (Fig. 2A). These results, together with those showing that E7 has no effect on the presentation of exogenous peptide or on T-cell-mediated lysis of cells sensitized by exogenous peptide, with or without IFN-c exposure, allow the conclusion that E7 inhibits the ability of IFN-c to enhance the processing of endogenous antigen for pre- sentation. To confirm these findings in an independent system, we loaded KCs osmotically with OVA, using previously established techniques. KCs and E7-KCs loaded with OVA were treated with IFN-c, or left untreated, and assessed for susceptibility to T-cell-med- iated lysis. CTLs specific for SIINFEKL were equally able to kill untreated KCs or E7-KCs when osmoti- cally loaded with OVA, as expected. As predicted from the findings with OVA transgenic KCs, lysis of KCs osmotically loaded with OVA was similar whether the cells were E7 transgenic or not, but after exposure to IFN-c, substantially increased lysis of OVA-loaded KCs was observed (Fig. 2B), whereas no such enhance- ment was seen for osmotically loaded E7-KCs, con- firming the findings with double-transgenic KCs that endogenous E7 inhibits processing of endogenous anti- gens by KCs for presentation to antigen-specific T cells (Fig. 2B). To further confirm this finding, we used an antibody against SIINFEKL–H-2K b complexes to measure MHC-associated presentation of SIINFEKL, derived from osmotic loading of KCs with OVA. For KCs, expression of SIINFEKL–H-2K b complexes was sub- stantially upregulated in response to IFN-c, whereas for E7-KCs, exposure to IFN-c failed to upregu- late expression of SIINFEKL–H-2K b complexes (Fig. 2C–E). Thus, endogenously expressed E7 inhibits IFN-c-mediated enhancement of processing of endoge- nous antigen by KCs, without inhibiting IFN-c- mediated upregulation of MHC class I expression. HPV 16 E7 blocks the ability of IFN-c to upregulate the transcription of MHC class I antigen processing and presentation-associated genes in keratinocytes As HPV 16 E7 expression in KCs impairs IFN-c-medi- ated processing of endogenous OVA for presentation of SIINFEKL (Fig. 2D), and HPV 16 E7 has been suggested to inhibit IFN-c signal transduction, we hypothesized that endogenously expressed E7 might inhibit transcription of IFN-c-dependent genes that are necessary for MHC class I antigen presentation. To investigate whether HPV 16 E7 could reduce the ability of IFN-c to upregulate transcription in KCs of genes relevant to the processing and presentation of endogenous antigen, we assessed the transcription of three MHC class I antigen processing and presenta- tion-associated genes (pa-28, tap-1 and irf-1) in KCs and E7-KCs both before and after exposure of the cells to IFN-c, using quantitative RT-PCR (Fig. 3). Basal levels of expression of pa-28 and irf-1 were somewhat higher in E7-KCs than in KCs (Fig. 3). However, the increase in level of expression induced by exposure to IFN-c was significantly blunted in E7 KCs, as compared with KCs, for pa28, tap-1 and irf-1 (Fig. 4). The maximal level of expression achieved F. Zhou et al. HPV 16 E7 inhibits T-cell KC lysis FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS 957 after IFN-c exposure was also significantly diminished in E7-KCs, as compared with KCs, for tap-1 and irf-1 (Fig. 3). Discussion In this study, we show that expression of HPV 16 E7 as a transgene product in epithelial cells does not directly impair, but rather slightly increases, MHC class I expression. E7 expression is nevertheless associ- ated with impairment of IFN-c-induced enhancement of presentation of endogenous antigen to CTLs. For E7-KCs, IFN-c treatment is less able to enhance the transcription of genes regulating antigen presentation, including tap-1, irf-1 and pa28. The reduction in gene transcription is from 5- to 10-fold, which is sufficient to reduce antigen presentation about five-fold and impair T-cell-mediated killing in vitro, and may there- fore be sufficient to explain the failure of E7-expressing skin to be rejected in vitro. Viruses use multiple strategies to make infected cells of less interest to virus protein-specific immune effector responses. Papillomavirus nonstructural viral proteins have been shown to interact with several cellular pro- cesses in a manner that could impair MHC class I expression. HPV E5, when overexpressed as a trans- gene product, can trap MHC class I molecules in the Golgi [23]. HPV 16 E7, when overexpressed, can repress the MHC class I heavy chain promoter, as well as the promoters of tap1 and lmp2 [7], and can also P < 0.05 AB C E D Fig. 2. HPV E7 impairs the enhancement by IFN-c of the presentation of endogenous antigen. (A) OVA transgenic KCs (KC-OVA) and KCs expressing E7 and OVA (E7-KC-OVA) were compared for susceptibility to lysis by OVA-primed effector T cells, with or without IFN-c expo- sure. (B) KCs and E7-KCs were loaded osmotically with OVA (OSM ⁄ OVA) or myoglobin (MYO), and, where indicated, exposed to IFN-c (100 UÆmL )1 for 48 h). OVA-loaded KCs with or without an E7 transgene were compared for susceptibility to lysis with and without IFN-c pretreatment. Nontransgenic KCs treated with IFN-c (KC + IFN-c) are shown as a control. (C, D) Expression of SIINFEKL–MHC class I pep- tide complexes on KCs (C) and E7-KCs (D) osmotically loaded with OVA and exposed to IFN-c (100 UÆmL )1 ) for 48 h was assessed with spe- cific antibody by flow cytometry. (E) Processing of OVA for presentation by MHC as SIINFEKL, according to the protocols for (C) and (D). OVA-loaded KCs exposed to IFN-c (KC + IFN + OVA) had significantly higher expression of SIINFEKL–MHC complexes than OVA-loaded E7 transgenic KCs exposed to IFN-c (E7-KC + IFN + OVA) (P = 0.02, t-test, n = 4). For the positive control (E7-KC + IFN + OVA + SIINFEKL), MHC complexes were saturated with SIINFEKL added to the culture medium. For the negative controls, no IFN-c (KC + OVA; E7-KC + OVA) or no OVA (E7-KC + IFN) was added. Means and SD values for MFI are shown. HPV 16 E7 inhibits T-cell KC lysis F. Zhou et al. 958 FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS bind to TAP1 and inhibit peptide transport [24], and reduce the expression of MHC in cultured murine fi- broblasts [25]. However, these reported effects of E7 do not seem, from our current study, to impact on the ability of KCs expressing E7 at levels more typical of those seen in HPV infection to present exogenous peptide, or on the level of MHC class I expression. Papillomavirus-associated cervical cancers express high levels of E7, and demonstrate impaired membrane expression of MHC class I complexes, which would be expected to impair antigen presentation. However, in cancer cells, reduced MHC class I display is associated with low levels of TAP1 or TAP2 [26], as a result of gene mutations associated with transformation, and the contribution, if any, of overexpression of HPV 16 E7 to reduced MHC class I expression in these cells is unclear. We have recently shown that E7-specific CTLs that are well able to kill E7-expressing transplantable tumors fail to efficiently kill KCs expressing E7 as a transgene product at levels commensurate with those in infected cervical epithelium [5]. As impaired recognition could be overcome by exposure to exogenous E7, it probably reflects either low availability of E7 for pro- cessing for presentation, or impaired antigen processing in E7-expressing cells. To distinguish these possibilities, we studied the processing and presentation of OVA expressed as a transgene product in E7 transgenic and control KCs, using antibody against SIINFEKL–H- 2K b . OVA presentation appeared to be normal in this system, as cells with or without E7 were equally suscep- tible to killing, although the fixed level of OVA expres- sion did not exclude the possibility that high-level OVA expression could overcome any restriction on process- ing. We therefore tested cells loaded osmotically with OVA, where lesser levels of OVA loading were still AB C DE Fig. 3. Transcription of genes associated with antigen processing and presentation in KCs and E7-KCs after IFN-c treatment. mRNA was extracted from KCs and E7-KCs. KCs and E7-KCs were treated or not, as shown, with IFN-c (IFN) at 100 UÆmL )1 for 48 h. Expression levels of mRNA assessed by RT-PCR with specific primers are shown relative to a reference gene, rRNA adenine dimethylase. Transcription after IFN-c expo- sure was higher for KCs than for E7-KCs for irf-1 (P = 0.02, n =7,t-test) and tap-1 (P < 0.01, n = 7). Differences for pa28 and pias1 were nonsignificant by unpaired t-test (n = 3). Error bars represent mean and SD (n = 3). Fig. 4. Upregulation of gene expression in KCs and E7-KCs follow- ing IFN-c exposure. For each tested gene, the ratio of expression level between cells exposed or not exposed to IFN-c is shown for E7-KCs, and also for KCs. Significant differences in magnitude of the IFN-c-induced change in expression between KCs and E7-KCs were seen for irf-1 (P = 0.01, n =3,t-test), tap-1 (P = 0.05, n = 3), and pa-28 (P = 0.05, n = 3). The change for pias1 was nonsignifi- cant by unpaired t-test (n = 3). Error bars represent mean and SD. F. Zhou et al. HPV 16 E7 inhibits T-cell KC lysis FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS 959 equally able to sensitize cells to OVA, whether express- ing E7 or not. However, in both the osmotic loading model and the transgene model, induction by IFN-c of increased expression of MHC–peptide complexes and susceptibility to T-cell-mediated lysis was significantly impaired if KCs expressed E7 as a transgene product. Looking for a mechanism, we assessed levels of tran- scription of genes whose protein products participate in antigen expression. HPV 16 E7 attenuated the ability of IFN-c to induce the transcription of several genes, and also of IRF-1, a master regulator of IFN-inducible genes. IFN-c is a potent inducer of antigen processing and of MHC class I expression for many cell types [14]. Signal transduction occurs via the JAK–STAT path- way, and upregulation of the expression of the down- stream genes relevant to antigen processing and presentation is mediated by members of the IRF family [27], particularly IRF-1[28]. E7 blocked the ability of IFN-c to efficiently induce irf-1 transcription to the level observed in nontransgenic KCs. Thus, E7 may block the ability of IFN-c to induce IRF-1 expression, in turn inhibiting the expression of downstream genes related to MHC class I antigen processing and presentation. E7 could potentially also interfere with IFN-c-mediated upregulation of IRF-1 expression by inhibiting upstream transduction of IFN signaling [16,17,29,30]. Furthermore, HPV 16 E7 can also inter- fere with IRF-1 function without affecting IRF-1 tran- scription and translation. This occurs through alteration of the DNA-binding capacity and promoter transactivation of IRF-1 without alteration of IRF-1 level [20,21,31], probably by direct binding to E7 [20]. Differences between the effects of E7 on induction by IFN-c of IRF-1 mRNA in our study and on the steady-state levels of IRF-1 measured by others may reflect the different experimental systems, or effects of E7 on post-transcriptional regulation of IRF-1 produc- tion or destruction. HPV 16 is a member of the mucosotropic a-papillo- mavirus clade. Papillomaviruses from the genetically and functionally distinct skin tropic b-clade use alter- native means to impair antigen presentation. HPV 38 E6 inhibits STAT1 expression and phosphorylation induced by IFN-b and IFN-c in human KCs, and inhibits IRF-1, TAP1 and MHC class I expression in host cells [29]. An impaired response of E7-KCs to IFN-c has significant implications for immunotherapy of HPV-associated skin lesions, which have proven refractory to induced antigen-specific immunotherapy [32,33]. IFN-c, secreted by CD8 T cells, by activated NK and NKT cells and by dendritic cells [34,35], is a key intermediate mediator of CD8 T-effector cell function, enhancing antigen presentation as well as polarizing the immune response to the Th1 type. Impaired antigen presentation may thus be one of the reasons why HPV infection is slow to clear in the face of adequate cellular immunity, and why immunother- apy has proven ineffective for persisting HPV infec- tion. It may also explain why local administration of supraphysiological concentrations of IFNs can contrib- ute to the clearance of HPV-associated genital warts [36]. Administration of proinflammatory mediators that can enhance antigen presentation by an IFN- independent pathway, perhaps through toll-like recep- tor signaling [34], may therefore facilitate the immune clearance of HPV-associated disease. Experimental procedures Immunogen, peptide and mice An 8-mer peptide (SIINFEKL) corresponding to the major CTL epitope of OVA was synthesized by AusPep (Park- ville, Vic., Australia) to bind to MHC class I H-2K b (amino acids 258–266 of OVA). C57BL ⁄ 6J mice, C57BL ⁄ 6J mice expressing SIINFEKL [37] or HPV 16 E7 [38] from the keratin 14 promoter, and C57BL ⁄ 6J mice expressing OVA from the keratin 5 promoter [4], were bred under conven- tional conditions in specific pathogen-free holding rooms in the Princess Alexandra Hospital biological resources facility (Brisbane, Qld, Australia). The protocols of these experi- ments were approved by the institutional Animal Ethics Committee. Generation of effector cells C57BL ⁄ 6J female mice (6–8 weeks of ages) were immunized once with 100 lg of SIINFEKL ⁄ 30 lg of keyhole limpet hemocyanin (Sigma Pharmaceuticals, Melbourne, Victoria, Australia) and 30 lg of QuilA (Spikoside; ISCOTEC AB, Lulea, Sweden). Lymph node cells were collected from immunized mice 4 days after immunization. Lymphocytes were cultured in filtered Click’s medium [50% Eagle’s ⁄ Ham’s amino acids (Sigma), 50% RPMI-1640 (Gibco; Invitrogen, Carlsbad, CA, USA), 10% heat-inactivated fetal bovine serum, containing 1 ngÆmL )1 mouse inter- leukin-2 (Pharmingen, San Diego, CA, USA) and 0.05 lm SIINFEKL, 2 · 10 5 UÆmL )1 penicillin ⁄ 2 · 10 5 UÆmL )1 streptomycin, 200 mml-glutamine and 5 · 10 )5 m 2-mercap- toethanol (2-ME)] for 4 days. Generation of target cells Isolation and culture of KCs from mouse skin has been described previously [5]. In brief, KCs were cultured in epidermal cell culture 3 : 1 medium [for 500 mL of 3 : 1 medium: 365 mL of DMEM (Gibco), 5 mL of HPV 16 E7 inhibits T-cell KC lysis F. Zhou et al. 960 FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS l-glutamine ⁄ penicillin ⁄ streptomycin, 200 mm ⁄ 2 · 10 5 U ⁄ 2 · 10 5 UÆmL )1 , 125 mL of Ham’s F12, 50 mL of fetal bovine serum, 500 lL of transferrin (Sigma), 5 mgÆmL )1 in 0.1% BSA in NaCl ⁄ P i , 500 lL of insulin (Sigma), 5mgÆmL )1 in 1 mm HCl, 500 lL of cholera toxin (Sigma), 8.4 mgÆmL )1 in NaCl ⁄ P i , 100 lL of hydrocortisone (Sigma), 1.2 mgÆmL )1 in 90% ethanol in water, 1000 lLof 184 mm adenine (Sigma), 17 mgÆmL )1 in 0.1% BSA in NaCl ⁄ P i , 500 lL of gentamicin, 20 mgÆmL )1 , 500 lLof 2-ME stock (35 lL of 2-ME in 10 mL of RPMI-1640 med- ium for 2-ME stock)] for 48 h, and then transferred into serum-free KC medium (Gibco) for 2 days. KCs were seeded in 96-well plates at 2 · 10 4 cells per well for CTL assays, or suspended at 5 · 10 5 cells per tube for FACS experiments. Osmotic loading of KCs The techniques have been described previously [39]. In brief, 5 · 10 6 KCs were suspended in 1 mL of RPMI-1640 osmotic loading buffer [25 mm Hepes, 0.5 m sucrose (w ⁄ v), 10% polyethylene glycol, pH 7.2] containing 10 mgÆmL )1 OVA (Sigma) or myoglobin (Sigma), and incubated at 37 °C for 10 min. The cells were then diluted into 14 mL of a mixture of 60% RPMI-1640 medium and 40% water, and held at 37 °C for 2 min. The loaded KCs, pelleted at 300 g for 7 min, were resuspended in RPMI-1640 medium and pelleted at 300 g for 5 min. Finally, the cells were resuspended in culture medium and incubated with or with- out IFN-c at 100 UÆ mL )1 for 24 h [40]. CTL assay ( 51 Cr release) A standard 5-h 51 Cr release assay was conducted as described previously [5]. CTL assay data were expressed as percentage specific lysis according to the following formula: The data were analyzed by t-test, and results were regarded as significantly different when P < 0.05. Flow cytometry KCs (5 · 10 5 ) were incubated with the first antibody (mouse anti-SIINFEKL–H-2K b clone D25.1.1.16 [41], 50 lL per sample, provided by D. Purcell, University of Melbourne), mouse anti-H-2K b (clone AF6-88.5; Pharmingen), or mouse anti-(rat IgM) (Pharmingen), an isotype control, for 1 h at 4 °C. Cells were then washed twice and incubated with the secondary antibody [fluorescein isothiocyanate-conjugated rabbit anti-mouse IgG (Dako Cytomation, Copenhagen, Denmark)] for 1 h at 4 °C, washed, and fixed with 5% formaldehyde. Data collected on a FACSCalibur (Becton Dickinson, San Diego, CA, USA) were analyzed with winmdi 2.8 (Joseph Trotter, Scripps Research Institute, La Jolla, CA, USA). Viable KCs were selected preferentially by excluding small particles. The change in mean fluorescence intensity (MFI) was calculated as the difference in MFI between test and isotype control samples. Real-time PCR mRNA was extracted from KCs and converted to cDNA with the use of random primers and PowerScript RT (Gene- Works, Hindmarsh, SA, Australia), according to the manu- facturer’s protocol. cDNA samples dissolved in the PCR mix buffer (FastStart SYBR Green Master; Roche Applied Science, Mannheim, Germany) were used to conduct quanti- tative PCR under the following conditions: 50 °C for 2 min; 95 °C for 10 min; and 40 cycles of 94 °C for 1 min, 55 °C for 1 min and 72 °C for 1 min. The following primers were used: TAP1, forward, 5¢-ACC TGG CTA CGG TAC ACC TG-3¢; TAP1, reverse, 5¢-CCT CTG AGC TCC CAC TTG AC-3¢; IRF-1, forward, 5¢-CCT GGG TCA GGA CTT G- GA TA-3¢; IRF-1, reverse, 5¢-TTC GGC TAT CTT CCC T- TC CT-3¢; PA28, forward, 5¢-CCG CTC CTC CTT CTC TTT CT-3¢; PA28, reverse, 5¢-AAG CCA AGG TGG ATG TGT TC-3¢; JAK1, forward, 5¢-TCA ACC TTC CCA AAG TGA CC-3¢; JAK1, reverse, 5¢-CAT GAC TCG CTG CAT GAA CT-3¢; PIAS1, forward, 5¢-AAG TGC TCA - CAG CCT TGG AT-3¢; PIAS1, reverse, 5¢-TCC CTA GGT GCA TGT TCT CC-3¢; rRNA adenine dimethylase, for- ward, 5¢-GGA GGG CCC ATC AGT TTA AT-3¢; rRNA adenine dimethylase, reverse, 5¢-AAA CAA TTG CAT TGC ATA GTGC-3¢. The data were analyzed with rotor- gene 6000. Statistical analysis All experimental data, including DMFI of FACS data, were analyzed with unpaired t-tests. Error bars represent mean and standard deviation (SD). Results were regarded as showing significant differences if P-values were < 0.05. Acknowledgements The authors are grateful to the staff of the biological research facility at the Princess Alexandra Hospital for their assistance. This work was funded by program grant No. 352439 from the National Health and % specific lysis ¼ mean sample release (c.p.m.) À mean spontaneous release mean maximum release (c.p.m.) À mean spontaneous release (c.p.m.) Â 100 F. Zhou et al. HPV 16 E7 inhibits T-cell KC lysis FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS 961 Medical Research Council of Australia, and grants from the Lions Medical Research Foundation, the Australian Cancer Research Foundation, the Cancer Council Queensland, and the Princess Alexandra hos- pital Foundation. Scholarship funding to F. Zhou was from the Cancer Collaborative Group, Princess Alex- andra Hospital and from ANZ Trustees. I. Frazer was recipient of a Queensland Government Premier’s fel- lowship. The authors declare that they have no conflict of interest or financial interests regarding the research findings described in this article. References 1 Koshiol JE, Schroeder JC, Jamieson DJ, Marshall SW, Duerr A, Heilig CM, Shah KV, Klein RS, Cuuvin S & Schuman P et al. (2006) Time to clearance of human papillomavirus infection by type and human immunodeficiency virus serostatus. Int J Cancer 119, 1623–1629. 2 Kanodia S, Fahey LM & Kast WM (2007) Mecha- nisms used by human papillomaviruses to escape the host immune response. Curr Cancer Drug Targets 7, 79–89. 3 Dunn LA, Evander M, Tindle RW, Bulloch AL, De Kluyver RL, Fernando GJ, Lambert PF & Frazer IH (1997) Presentation of the HPV16E7 protein by skin grafts is insufficient to allow graft rejection in an E7- primed animal. Virology 235, 94–103. 4 Azukizawa H, Kosaka H, Sano S, Heath WR, Takah- ashi I, Gao XH, Sumikawa Y, Okabe M, Yoshikawa K & Itami S (2003) Induction of T-cell-mediated skin dis- ease specific for antigen transgenically expressed in keratinocytes. Eur J Immunol 33, 1879–1888. 5 Leggatt GR, Dunn LA, De Kluyver R, Stewart T & Frazer IH (2002) Interferon-gamma enhances cytotoxic T lymphocyte recognition of endogenous peptide in keratinocytes without lowering the requirement for sur- face peptide. Immunol Cell Biol 80, 415–424. 6 Zhou F, Frazer IH & Leggatt GR (2009) Keratinocytes efficiently process endogenous antigens for cytotoxic T-cell mediated lysis. Exp Dermatol 18, 1053–1059. 7 Georgopoulos NT, Proffitt JL & Blair GE (2000) Tran- scriptional regulation of the major histocompatibility complex (MHC) class I heavy chain, TAP1 and LMP2 genes by the human papillomavirus (HPV) type 6b, 16 and 18 E7 oncoproteins. Oncogene 19, 4930–4935. 8 Evans M, Borysiewicz LK, Evans AS, Rowe M, Jones M, Gileadi U, Cerundolo V & Man S (2001) Antigen processing defects in cervical carcinomas limit the pre- sentation of a CTL epitope from human papillomavi- rus 16 E6. J Immunol 167, 5420–5428. 9 Thomas KJ, Smith KL, Youde SJ, Evans M, Fiander AN, Borysiewicz LK & Man S (2008) HPV16 E6 29-38- specific T cells kill cervical carcinoma cells despite partial evasion of T-cell effector function. Int J Cancer 122, 2791–2799. 10 Hill A, Jugovic P, York I, Russ G, Bennink J, Yewdell J, Ploegh H & Johnson D (1995) Herpes simplex virus turns off the TAP to evade host immunity. Nature 375, 411–415. 11 Levitskaya J, Coram M, Levitsky V, Imreh S, Steiger- wald-Mullen PM, Klein G, Kurilla MG & Masucci MG (1995) Inhibition of antigen processing by the internal repeat region of the Epstein–Barr virus nuclear antigen-1. Nature 375, 685–688. 12 Ahn K, Meyer TH, Uebel S, Sempe ´ P, Djaballah H, Yang Y, Peterson PA, Fruh K & Tamp R (1996) Molecular mechanism and species specificity of TAP inhibition by herpes simplex virus protein ICP47. EMBO J 15, 3247–3255. 13 Ahn K, Gruhler A, Galocha B, Jones TR, Wiertz EJ, Ploegh HL, Peterson PA, Yang Y & Fruh K (1997) The ER-luminal domain of the HCMV glycoprotein US6 inhibits peptide translocation by TAP. Immunity 6, 613–621. 14 Boehm U, Klamp T, Groot M & Howard JC (1997) Cellular responses to interferon-gamma. Annu Rev Immunol 15, 749–795. 15 Seresini S, Origoni M, Lillo F, Caputo L, Paganoni AM, Vantini S, Longhi R, Taccagni G, Ferrari A, Doglioni C, Secchi P & Protti MP (2007) IFN-gamma produced by human papilloma virus-18 E6-specific CD4 + T cells predicts the clinical outcome after surgery in patients with high-grade cervical lesions. J Immunol 179, 7176–7183. 16 Barnard P & McMillan NAJ (1999) The human papillo- mavirus E7 oncoprotein abrogates signaling mediated by interferon-a. Virology 259, 305–313. 17 Chang YE & Laimins LA (2000) Microarray analysis identifies interferon-inducible genes and Stat-1 as major transcriptional targets of human papillomavirus type 31. J Virol 74 , 4174–4182. 18 Koromilas AE, Li S & Matlashewski G (2001) Control of interferon signaling in human papillomavirus infection. Cytokine Growth Factor Rev 12, 157–170. 19 Nees M, Geoghegan JM, Hyman T, Frank S, Miller L & Woodworth CD (2001) Papillomavirus type 16 onco- genes downregulate expression of interferon-responsive genes and upregulate proliferation-associated and NF-kappaB-responsive genes in cervical keratinocytes. J Virol 75, 4283–4296. 20 Park JS, Kim EJ, Kwon HJ, Hwang ES, Namkoong SE & Um SJ (2000) Inactivation of interferon regulatory factor-1 tumor suppressor protein by HPV E7 oncopro- tein – implication for the E7-mediated immune evasion mechanism in cervical carcinogenesis. J Biol Chem 275 , 6764–6769. HPV 16 E7 inhibits T-cell KC lysis F. Zhou et al. 962 FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS 21 Um SJ, Rhyu JW, Kim EJ, Jeon KC, Hwang ES & Park JS (2002) Abrogation of IRF-1 response by high- risk HPV E7 protein in vivo. Cancer Lett 179, 205–212. 22 Niederwieser D, Aubock J, Troppmair J, Herold M, Schuler G, Boeck G, Lotz J, Fritsch P & Huber C (1988) IFN-mediated induction of MHC antigen expres- sion on human keratinocytes and its influence on in vi- tro alloimmune responses. J Immunol 140, 2556–2564. 23 Ashrafi GH, Haghshenas MR, Marchetti B, O’Brien PM & Campo MS (2005) E5 protein of human papillo- mavirus type 16 selectively downregulates surface HLA class I. Int J Cancer 113, 276–283. 24 Vambutas A, DeVoti J, Pinn W, Steinberg BM & Bona- gura VR (2001) Interaction of human papillomavirus type 11 E7 protein with TAP-1 results in the reduction of ATP-dependent peptide transport. Clin Immunol 101, 94–99. 25 Park JS, Boyer S, Mitchell K, Gilfor D, Birrer M, Darlington G, El Deiry W, Firestone GL, Munger K, Band V, Fisher PB & Dent P (2000) Expression of human papilloma virus E7 protein causes apoptosis and inhibits DNA synthesis in primary hepatocytes via increased expression of p21Cip-1 ⁄ WAF1 ⁄ MDA6. J Biol Chem 275, 18–28. 26 Ritz U, Momburg F, Pilch H, Huber C, Maeurer MJ & Seliger B (2001) Deficient expression of components of the MHC class I antigen processing machinery in human cervical carcinoma. Int J Oncol 19, 1211–1220. 27 Tennant LM, Renard C, Chardon P & Powell PP (2007) Regulation of porcine classical and nonclassical MHC class I expression. Immunogenetics 59, 377–389. 28 Briscoe J, Guschin D, Rogers NC, Watling D, Muller M, Horn F, Heinrich P, Stark GR & Kerr IM (1996) JAKs, STATs and signal transduction in response to the interferons and other cytokines. Phil Trans R Soc Lond B Biol Sci 351, 167–171. 29 Cordano P, Gillan V, Bratlie S, Bouvard V, Banks L, Tommasino M & Campo MS (2008) The E6E7 onco- proteins of cutaneous human papillomavirus type 38 interfere with the interferon pathway. Virology 377, 408–418. 30 Li SY, Labrecque S, Gauzzi MC, Cuddihy AR, Wong AHT, Pellegrini S, Matlashewski GJ & Koromilas AE (1999) The human papilloma virus (HPV)-18 E6 oncoprotein physically associates with Tyk2 and impairs Jak-STAT activation by interferon-a. Oncogene 18, 5727–5737. 31 Perea SE, Massimi P & Banks L (2000) Human papillo- mavirus type 16 E7 impairs the activation of the inter- feron regulatory factor-1. Int J Mol Med 5, 661–666. 32 Frazer IH (2004) Prevention of cervical cancer through papillomavirus vaccination. Nat Rev Immunol 4, 46–55. 33 Frazer IH & Appleton S (2006) Vaccines to prevent and treat human papillomavirus associated anogenital dis- ease. In The Cervix, 2nd edn (Jordan JA & Singer A eds), pp. 609–620. Blackwell Publishing, Oxford. 34 Wang RF, Miyahara Y & Wang HY (2008) Toll-like receptors and immune regulation: implications for can- cer therapy. Oncogene 27, 181–189. 35 Frucht DM, Fukao T, Bogdan C, Schindler H, O’Shea JJ & Koyasu S (2001) IFN-gamma-production by anti- gen-presenting cells: mechanisms emerge. Immunol Today 22, 556–560. 36 Frazer IH & McMillan NA (1997) Papillomatosis and condylomata acuminata. In Clinical Applications of the Interferons (Stuart-Harris R & Penny RW eds), pp. 79–91. Chapman and Hall Medical, London. 37 Stefanski HE, Mayerova D, Jameson SC & Hogquist KA (2001) A low affinity TCR ligand restores positive selection of CD8 + T cells in vivo. J Immunol 166, 6602–6607. 38 Herber R, Liem A, Pitot H & Lambert PF (1996) Squa- mous epithelial hyperplasia and carcinoma in mice transgenic for the human papillomavirus type 16 E7 oncogene. J Virol 70, 1873–1881. 39 Moore MW, Carbone FR & Bevan MJ (1988) Intro- duction of soluble protein into the class I pathway of antigen processing and presentation. Cell 54, 777–785. 40 Carbone FR & Bevan MJ (1990) Class I-restricted pro- cessing and presentation of exogenous cell-associated antigen in vivo. J Exp Med 171, 377–387. 41 Porgador A, Yewdell JW, Deng Y, Bennink JR & Germain RN (1997) Localization, quantitation, and in situ detection of specific peptide–MHC class I complexes using a monoclonal antibody. Immunity 6, 715–726. F. Zhou et al. HPV 16 E7 inhibits T-cell KC lysis FEBS Journal 278 (2011) 955–963 ª 2011 The Authors Journal compilation ª 2011 FEBS 963 . Human papillomavirus 16 E7 protein inhibits interferon-c-mediated enhancement of keratinocyte antigen processing and T-cell lysis Fang Zhou, Graham R. Leggatt and Ian H. Frazer The. expression of HPV 16 E7 protein in KCs may inhibit enhancement by IFN-c of KC sensitivity to T-cell lysis, by impairing antigen presentation. Abbreviations CTL, cytotoxic T lymphocyte; E7- KC, keratinocyte. infection, could inhibit enhancement by IFN-c of antigen processing and presentation, and CTL-mediated killing of KCs expressing non-self -antigen. Results E7 does not affect the ability of IFN- c to upregulate