|Home | About | Journals | Submit | Contact Us | Français|
Foot-and-mouth disease (FMD) remains one of the most devastating livestock diseases around the world. Several serotype-specific vaccine formulations exist, but they require about 5 to 7 days to induce protective immunity. Our previous studies have shown that a constitutively active fusion protein of porcine interferon (IFN) regulatory factors (IRF) 7 and 3 [IRF7/3(5D)] strongly induced type I IFN and antiviral genes in vitro and prevented mortality in an FMD mouse model when delivered with a replication-defective adenoviral vector [Ad5-poIRF7/3(5D)]. Here, we demonstrate that pigs treated with 108, 109, or 1010 PFU of Ad5-poIRF7/3(5D) 24 h before FMDV challenge were fully protected from FMD clinical signs and did not develop viremia, virus shedding or antibodies against FMDV nonstructural proteins. Pigs treated with Ad5-poIRF7/3(5D) had higher levels of IFN and antiviral activity in serum, and upregulated expression of several IFN-stimulated genes in peripheral blood mononuclear cells, compared to pigs treated with Ad5-Blue vector control. Importantly, treatment of porcine cultured cells with Ad5-poIRF7/3(5D) inhibited the replication of all 7 FMDV serotypes. In vitro experiments using cultured embryonic fibroblasts derived from IFN receptor knockout mice suggested that the antiviral response induced by Ad5-poIRF7/3(5D) was dependent on type I and III IFN pathways; however, experiments with mice demonstrated that a functional type I IFN pathway mediates Ad5-poIRF7/3(5D) protection conferred in vivo. Our studies demonstrate that inoculation with Ad5-poIRF7/3(5D) completely protects swine against FMD by inducing a strong type I IFN response and highlights its potential application to rapidly and effectively prevent FMDV replication and dissemination.
IMPORTANCE Foot-and-mouth disease virus (FMDV) causes a fast-spreading disease that affects farm animals, with economically and socially devastating consequences. Our study shows that inoculation with a constitutively active transcription factor, namely, a fusion protein of porcine interferon (IFN) regulatory factors (IRF) 7 and 3 delivered by an adenovirus vector [Ad5-poIRF7/3(5D)], is a new effective treatment to prevent FMD in swine. Animals pretreated with Ad5-poIRF7/3(5D) 1 day before being exposed to FMDV were completely protected from viral replication and clinical disease. It is noteworthy that the doses of Ad5-poIRF7/3(5D) required for protection are lower than those previously reported for similar approaches using Ad5 vectors delivering type I, II, or III IFN, suggesting that this novel strategy would be economically appealing to counteract FMD. Our results also indicate that a dynamic interplay among different components of pigs' innate immune defenses allows potent antiviral effects after Ad5-poIF7/3(5D) administration.
Foot-and-mouth disease virus (FMDV) causes an economically, socially devastating, fast-spreading disease that affects a wide range of farm and wild cloven-hooved animals (1). The existence of seven serotypes (A, Asia1, C, O, and SAT1, -2, and -3) and multiple strains and topotypes makes it difficult to control FMD with a single vaccine (1, 2).
Several vaccine strategies are available or have been described to control FMD, including inactivated vaccines and recombinant vaccines that express partial FMDV sequences in transgenic plants, recombinant baculovirus, vaccinia virus, and adenovirus type 5 (Ad5) vectors (reviewed in reference 3). However, animal movement and trade restrictions (4) in combination with inactivated vaccines formulated with adjuvants are still the most commonly used strategy in FMD enzootic regions (3) or in case of outbreaks in FMD-free countries. Although effective, current inactivated vaccines have several limitations: they provide serotype specific, short-lived immunity with the necessity of booster injection, have a limited shelf life, require expensive high-containment biosafety level 3 (BSL3) manufacturing facilities for production, and do not allow the differentiation of infected from vaccinated animals (DIVA) unless highly purified (2, 3). Recently, an Ad5-FMDV vaccine that overcomes many of these shortcomings has been granted a conditional license for use in cattle in the United States in case of emergency (5). However, similar to the inactivated vaccine, the Ad5-FMD vaccine requires approximately 5 to 7 days to confer complete clinical protection and animals that are exposed to FMDV during that time frame could still develop clinical signs and disseminate the virus (5, 6). Although vaccination of swine is not a common practice, it is occasionally used to control outbreaks. For example, swine vaccination was used in Taiwan and South Korea to contain outbreaks in 1997 and 2010 (7, 8). In addition, avoidance of the disease in pigs, which are great disseminators of the virus, plays a fundamental role in preventing transmission to other species and it is a key component in FMD control strategies. In fact, swine have been associated with the initiation of devastating FMD outbreaks, such as the one that took place in Europe in 2001 (9).
Several studies have demonstrated the value and effectiveness of fast-acting antiviral strategies to control FMD prior to vaccine-induced protection (10). Examples include the use of Ad5 vectors expressing alpha interferon (IFN-α) (11), IFN-β (12), IFN-γ (13), IFN-λ (14), and IFN-α/γ and FMDV-specific small interfering RNA (siRNAs) (15) to protect swine. In addition, combinational approaches such as Ad5-IFN-α and an Ad5 subunit vaccine delivering FMD empty capsids have successfully shortened the window in which vaccinated animals can still become infected (16). However, relatively high doses of Ad5 are required, a fact that restricts large-scale application of these approaches in the field or in emergency situations.
During a viral infection, the IFN pathway is naturally induced by recognition of viral pathogen-associated molecular patterns (PAMP) that lead to phosphorylation, dimerization, and nuclear translocation of IFN regulatory factor 3 (IRF3) and IRF7. In the nucleus, these IRFs promote type I IFN transcription (17, 18). In turn, IFNs are secreted and, in an endocrine and autocrine manner, bind to specific receptors, triggering the activation of hundreds of IFN-stimulated genes (ISGs) that mediate cellular antiviral pathways (17, 19). Recently, we have demonstrated that a constitutively active fusion protein of porcine IRF3 and IRF7 [poIRF7/3(5D)] strongly induces type I IFN and antiviral genes in swine cells and protects mice from FMD mortality (20). In this study, we demonstrated that pigs inoculated with an Ad5 expressing poIRF7/3(5D) [Ad5-poIRF7/3(5D)] and challenged at 24 h postinoculation (hpi) with FMDV are fully protected against FMD clinical signs, had no viremia or virus shedding, and did not develop antibodies against FMDV nonstructural proteins. Interestingly, the Ad5-poIRF7/3(5D) protective dose was 10- to 100-fold lower than the Ad5-poIFN-α dose previously demonstrated necessary to confer protection against multiple serotypes of FMDV in swine (11, 12). Protection was associated with the presence of systemic antiviral activity and upregulation of ISGs in peripheral blood mononuclear cells (PBMCs). Consistently, treatment with Ad5-poIRF7/3(5D) hindered replication of all 7 FMDV serotypes in porcine cells. By using an FMD mouse model, we demonstrated that Ad5-poIRF7/3(5D) protection relies on the presence of a functional type I IFN receptor, confirming a pivotal role for this IFN against FMD. Our results demonstrate that Ad5-poIRF7/3(5D) is a potent biotherapeutic that can prevent FMD in swine.
Baby hamster kidney cells (BHK-21) were obtained from the American Type Culture Collection (ATCC; CCL-10) and were used to propagate and determine virus titers of stock preparations of FMDV or vesicular stomatitis virus (VSV). BHK-21 cells were maintained in modified minimal essential medium (MEM) containing 10% calf serum and 10% tryptose phosphate broth supplemented with 1% antibiotics and nonessential amino acids. SK-6 and IBRS-2 cells were obtained from the Foreign Animal Disease Diagnostic Laboratory (FADDL) at the Plum Island Animal Disease Center (PIADC). SK-6 cells were used to test the effect of Ad5-poIRF7/3(5D) on growth of multiple FMDV serotypes. IBRS-2 cells were used for FMDV titration of virus yields from infected SK-6 cells. SK-6 and IBRS-2 cells were maintained in Dulbecco modified MEM (DMEM) containing 10% fetal bovine serum (FBS) and supplemented with 1% antibiotics and nonessential amino acids. Human embryonic kidney (HEK) 293 cells were obtained from the ATCC (CRL-1573) and were used to grow Ad5 stocks. These cells were maintained in MEM containing 10% defined FBS and 1% antibiotics and nonessential amino acids. Madin-Darby bovine kidney-t2 (MDBK-t2) cells expressing the chloramphenicol acetyltransferase (CAT) enzyme under the control of the MxA promoter were used for IFN bioassays, as previously described (21). These cells were kindly provided by B. Charleston (Institute for Animal Health, Pirbright, United Kingdom) and were maintained in DMEM containing 10% FBS and supplemented with 1% antibiotics, nonessential amino acids, and blasticidin at 1 mg/ml. Mouse embryonic fibroblasts (MEFs) derived from type I IFN receptor (IFNAR) knockout (KO) mice or from wild-type (WT) mice and cultured in DMEM containing 10% FBS and 1% antibiotics and nonessential amino acids were obtained from D. Levy (New York University). All cell cultures were incubated at 37°C in 5% CO2.
The Janus kinase (JAK) inhibitor CP690550 (catalog number sc207457; Santa Cruz Biotechnologies, Santa Cruz, CA) was used to block JAK-signal transducer and activator of transcription (STAT) signaling (including type I and III IFN signaling). Prior to transfection or infection, cells were incubated for 1 h at 37°C in complete medium containing the inhibitor at concentrations ranging between 1 and 100 μM. A rat anti-mouse IFN-λ2 antibody (catalog number MAB4635; R&D, Minneapolis, MN) was used to neutralize the IFN-λ activity (3 to 5 μg of antibody/ml) of treated cells. The JAK inhibitor or the anti-mouse IFN-λ2 was maintained in the medium during Ad5 transduction and replenished after viral infection.
All work was performed at PIADC under BSL3 agricultural hazard (BSL-3Ag) regulations. FMDV strains A5 Westerwald, A12, Asia-1, C3 Resende, O, SAT1, SAT2, and SAT3 were obtained from E. Rieder (PIADC, USDA) or the FADDL (USDA). Pretreated SK-6 cells were infected with the different FMDV serotypes at a multiplicity of infection (MOI) of 1 and incubated at 37°C in 5% CO2. After a 1-h adsorption, unabsorbed virus was removed by washing the cells with 150 mM NaCl–20 mM morpholineethanesulfonic acid (MES; pH 6.0), followed by replenishment with complete medium. Twenty-four hours postinfection, FMDV was released by one freeze-thaw cycle.
An attenuated mouse-adapted FMDV O1-A12 chimeric virus previously described (22) was used for MEF infections. Briefly, KO or WT MEFs were pretreated with Ad5-poIRF7/3(5D) or control vector Ad5-Blue at an MOI of 20. After 24 h, cells were washed with MEM, followed by infection with chimeric FMDV O1-A12 at an MOI of 1. After 1 h of adsorption, unabsorbed virus was removed, cells were washed, and complete medium was replenished. At 24 h, virus was released by one freeze-thaw cycle.
Ad5-poIRF7/3(5D) or Ad5-Blue was isolated, propagated, and titrated in HEK 293 cells. Ad5s were purified by two rounds of CsCl gradient centrifugation as previously described (23).
Vesicular stomatitis virus (VSV) serotype New Jersey field strain 95COB was obtained from L. Rodriguez (PIADC, USDA) and was used for IFN bioassays on MDBK-t2 cells at an MOI of 2 to determine overall antiviral activity by endpoint dilution titration (24).
Viral titers of infected MEFs or SK6 cells were determined by endpoint dilution titrations on IBRS-2 or BHK-21 cells (23). Results were expressed as log10 50% tissue culture infective doses (TCID50) per milliliter. Viral titers of FMDV-infected swine sera or nasal swabs were determined by plaque assay on BHK-21 cells (25) and expressed as the number of PFU per milliliter.
Animal experiments were performed in the BSL-3Ag high-containment facilities of the PIADC. All experiments were conducted in compliance with the Animal Welfare Act (AWA), the Guide for Care and Use of Laboratory Animals (26), the Public Health Service Policy for the Humane Care and Use of Laboratory Animals (27). Specific animal protocols (number 151-13R for swine and number 204-11-R for mice experiments) were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the Plum Island Animal Disease Center (USDA/APHIS/AC certificate number 21-F-0001).
Nine Yorkshire gilts (4 to 6 weeks old; approximately 20 kg each) free of FMDV and FMDV antibodies were randomly allocated into 3 groups of 3 animals each. After 1 week of acclimation, pigs were inoculated subcutaneously (s.c.) at two opposite sites in the neck with 1 ml of the following Ad5 combinations: 1010 PFU/animal of Ad5-Blue, 109 PFU of Ad5-poIRF7/3(5D) and 109.95 PFU of Ad5-Blue/animal, or 1010 PFU of Ad5-poIRF7/3(5D)/animal (Fig. 1A). Twenty-four hours after Ad5 inoculation (hpi), all animals were challenged by intradermal inoculation in the heel bulb (IDHB), using 4 sites of inoculation, 100 μl per site, with FMDV A24 obtained from vesicular fluid of previously infected pigs. A challenge dose of 20 TCID50/pig of FMDV A24 Cruzeiro was selected based on previous experiments in which such a dose consistently caused clinical disease in all pigs by 2 to 4 days postchallenge (dpc) (12). Rectal temperatures and clinical signs, including alertness, lameness, and development of vesicles, were monitored daily for 9 dpc and at 14 dpc, along with collection of anticoagulated blood and nasal swabs for detection of viremia, viral RNA, and virus shedding, respectively. Heat-inactivated serum samples from 0, 4, 7, and 14 dpc were used to measure the FMDV antibody response. At 14 dpc, all pigs were euthanized. Clinical scores were determined by recording the presence of vesicular lesions in each toe and in the snout and/or mouth (maximum score = 17); lesions restricted to the site of inoculation were not counted (29).
Twelve gilts (4 to 6 weeks old; approximately 20 kg each) were randomly distributed in 3 groups of 4 animals each and were inoculated s.c. at two opposite sites in the neck with 1 ml of Ad5 combinations as follows: phosphate-buffered saline (PBS; control), 108 PFU of Ad5-poIRF7/3(5D) and 108.95 PFU of Ad5-Blue, or 109 PFU of Ad5-poIRF7/3(5D), as shown in Fig. 2A. Twenty-four hpi, all animals were challenged by IDHB inoculation with the same dose of FMDV A24 as described above. Evaluation of disease and sampling were performed as indicated above and illustrated in Fig. 2A.
Ten 6-week-old female C57BL/6 WT mice or 10 IFN-α/β receptor (IFNAR) KO (stock number 032045-JAX) mice were purchased from the Mutant Mouse Resource and Research Center (MMRRC; Jackson Laboratories, Farmington, CT). Mice were acclimated for 1 week before the experiment. Two groups of five WT mice each were inoculated with either Ad5-poIRF7/3(5D) (1 × 108 PFU/mouse) or Ad5-Blue (1× 108 PFU/mouse) s.c. in the dorsal flank. Analogously, 2 groups of five IFNAR KO mice were inoculated with either Ad5-poIRF7/3(5D) (1 × 108 PFU/mouse) or Ad5-Blue (1× 108 PFU/mouse) s.c. in the dorsal flank (see Fig. 7A). The dose of Ad5-poIRF7/3(5D) selected (108 PFU/mouse) was based on previous data (20) and accounted for the fact that Ad5-poIRF7/3(5D) does not induce as high a level of antiviral activity in murine cells as it does in porcine cells. Forty-eight hours after Ad5 treatment, all mouse groups were challenged with FMDV A24 Cruzeiro at the lethal dose of 5 × 104 PFU/mouse. Challenge with FMDV A24 was conducted at 48 hpi as previously reported, to minimize the effect of the antiviral activity induced by the adenoviral vector alone (20, 30). Serum samples were collected at specified days (see Fig. 7A) for viremia detection. Mouse survival was recorded for 7 days after challenge.
Serum and nasal swabs samples were examined for the presence of FMDV RNA by real-time reverse transcription-quantitative PCR (RT-qPCR) as previously described (31). Genome copies of FMDV were calculated using a standard curve with 10-fold dilutions of a known concentration of pure in vitro-synthesized viral RNA (molecular weight [MW] = 2.50 × 106). Threshold cycle (CT) values of ≥40 were considered negative.
Total RNA was isolated from approximately 107 peripheral blood mononuclear cells (PBMCs) using an RNAeasy extraction kit (Qiagen, Valencia, CA), followed by cDNA synthesis using random hexamers with qScript kit mix (Quanta Biosciences, Gaithersburg, MD) according to the manufacturer's instructions. cDNAs were diluted 10 times and used as the template for RT-qPCR with PerfeCTa SYBR green FastMix (Quanta Biosciences). Samples were run in an Applied Biosystems 7500 apparatus (Applied Biosystems, Carlsbad, CA). Relative quantification was performed for several cytokines, including interleukin 1β (IL-1β), IL-10, IL-15, IL-6, IL-12, IFN-γ-induced protein 10 (IP-10), IFN-γ, IFN-β, and a panel of ISGs, including those for 2′,5′-oligoadenylate synthetase (OAS1), myxovirus resistance 1 (Mx1), ubiquitin-specific peptidase 18 (USP18), regulated on activation, normal T cell expressed and secreted (RANTES), and IRF7, as previously described (32, 33). The expression of each gene of interest was normalized using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and β-actin. Data were analyzed using the comparative threshold cycle (ΔΔCT) method (33, 52). IFN concentrations were determined by an Mx-CAT enzyme-linked immunosorbent assay (ELISA) as described below.
Serotype-specific neutralizing antibody titers were determined by endpoint dilution serum neutralization tests (SNTs) using samples from 0, 4, 7, and 14 dpc (34). Neutralizing titers were reported as the log10 of the serum dilution yielding a 50% reduction in the TCID50 induced by FMDV A24 in BHK-21 cells.
The presence of antibodies against FMDV nonstructural proteins was determined by using a commercially available ELISA (PrioCHECK FMDV NS; Thermo Fisher, Waltham, MA) by following the instructions from the manufacturer. The optical density (OD) was determined at 450 nm in an ELISA reader (VersaMax; Molecular Devices). The percent inhibition ± standard deviation for each group average is shown in graphs (35).
IFN bioactivity was determined by a VSV infection inhibition assay on MDBK-t2 cells as previously described (36). Briefly, 2-fold serial dilutions of serum or cell supernatants were prepared in 96-well plates and incubated in cells for 24 h, followed by challenge with VSV at an MOI of 2. Antiviral activity was expressed as the reciprocal of the highest dilution of serum/cell supernatant able to suppress the VSV-induced cytopathic effect on MDBK-t2 cells in 50% of the wells assayed.
IFN bioactivity test was also measured by a CAT ELISA as described by Fray et al. (21). In brief, dilutions of sera were applied to MDBK-t2 cells seeded into 24-well tissue culture plates. In parallel, known amounts (from 1.95 to 1,000 U/ml) of recombinant human IFN-α (PBL Interferon Source, Piscataway, NJ) were used to determine a standard curve. Twenty-four hours after incubation at 37°C with 5% CO2, cells were lysed and CAT expression was determined using a commercially available ELISA kit (Roche Applied Sciences, Indianapolis, IN) in accordance with the manufacturer's protocol. Units of antiviral activity per milliliter of the samples were calculated from the standard curve.
Treatment differences were determined by Student's t test using Microsoft Excel. Values are expressed as means ± standard errors of the means (SEM), and statistical significance is indicated in the figures.
In our previous studies, we showed that administration of Ad5-poIRF7/3(5D) prevents mortality in an FMD mouse model (20). Therefore, we investigated whether Ad5-poIRF7/3(5D) administration could prevent FMD in swine, an FMD natural host. Groups of three pigs each were treated with different doses of Ad5-poIRF7/3(5D) (109 PFU/animal or 1010 PFU/animal) or with Ad5-Blue (1010 PFU/animal) as a control. All doses were adjusted with Ad5-Blue in order to administer a total of 1010 PFU of Ad5 per animal and account for the possible baseline antiviral effect induced by the vector alone (20, 30). Twenty-four hpi, animals were challenged with FMDV A24 (Fig. 1A). As previously described, control pigs showed clinical signs of infection by 2 dpc (Fig. 1B), with an average peak of fever by 4 dpc and reaching an average lesion score of ≥13 by 5 dpc. After 3 consecutive days, temperature declined to normal (<39.5°C) in all control animals, except for one pig, which died of FMD-associated myocarditis by 6 dpc. In contrast, pigs treated with either 109 or 1010 PFU/animal of Ad5-poIRF7/3(5D) did not show FMD clinical signs or fever (temperature > 39.5°C is indicated by a gray shaded area) during the entire course of the experiment (Fig. 1C). These results indicated that Ad5-poIRF7/3(5D) effectively protects swine against clinical manifestation of FMDV infection.
To test if, despite the absence of clinical signs, FMDV actively replicated or disseminated in pigs treated with Ad5-poIRF7/3(5D), the presence of virus was determined in sera (viremia) or in nasal secretions (virus shedding) by virus isolation. As shown in Fig. 1D, pigs in the control group had detectable viremia by 2 dpc and virus shedding by 3 dpc, lasting for approximately 3 days. In contrast, no virus was isolated from sera or nasal secretions of animals inoculated with Ad5-poIRF7/3(5D) (Fig. 1E). Furthermore, no viral RNA was detected in sera of animals inoculated with either dose of Ad5-poIRF7/3(5D) at any time postchallenge, whereas viral RNA was detected in the control animals starting at day 4 and remained detectable until day 7 (Table 1).
Thus far, our lab has shown that the minimum dose of an Ad5-poIFN-α required to fully protect swine from challenge with FMDV A24 Cruzeiro is 109 PFU/animal (10). Even higher Ad5 doses (1011 focus-forming units [FFU]/animal) were required when a recombinant virus produced by an industrial partner (Gen Vec Inc., Gaithersburg, MD) was tested in swine against FMDV A24, O1 Manisa, and Asia-1 (12). We therefore tested if treatment with Ad5-poIRF7/3(5D) could be effective at lower doses, making this strategy more attractive for use in veterinary medicine. Groups of four pigs were treated with 108 or 109 PFU/animal of Ad5-poIRF7/3(5D) or with PBS as a control (Fig. 2A). As shown in Fig. 2B, control pigs started to show signs of disease by 3 dpc, with average elevated body temperatures by 5 dpc. In contrast, pigs treated with 108 or 109 PFU of Ad5-poIRF7/3(5D) did not show any clinical signs or fever (Fig. 2C). Consistently, viremia and virus shedding were detected only in the control group. All animals treated with Ad5-poIRF7/3(5D), independently of the dose, did not show viremia, and only a signal below the dynamic range for the RT-qPCR detection method (40 > Ct > 45) was detected for viral RNA in nasal secretion (Fig. 2D and andE).E). These results confirmed that Ad5-poIRF7/3(5D) effectively protects swine from FMD even at a dose lower than the dose required for similar protection with Ad5-poIFN-α.
FMDV replication promptly induces a humoral response against structural (capsid) and nonstructural proteins. To test for the presence of FMDV neutralizing antibodies in pigs from experiments 1 and 2, neutralization assays were performed using strain A24 with sera collected at 0, 4, 7, and 14 dpc (Fig. 3A and andB).B). All groups were negative for neutralizing antibodies before FMDV challenge (Fig. 3A and andB).B). In experiment 1, the control group displayed neutralizing antibody titers greater than 2 log10 at 7 and 14 dpc and animals treated with Ad5-poIRF7/3(5D) showed a nonsignificant increase at 7 dpc (Fig. 3A). Similarly, in experiment 2, the control group displayed an increase in the neutralizing antibody titers at 7 and 14 dpc but the overall levels were lower than in experiment 1; no neutralizing antibody titers were detected in the groups treated with either 108 or 109 PFU/animal of Ad5-poIRF7/3(5D) at any analyzed time point (Fig. 3B). It is noteworthy that the appearance of the clinical signs, viremia, and virus shedding in experiment 2 were slightly delayed compared to that in experiment 1. Although the FMDV A24 Cruzeiro challenge dose was the same and the environmental conditions were similar in both experiments, it is possible that higher variability across experimental subjects (age, size, and overall physical condition of animals that are not inbred) may have affected the course of disease.
Inoculation with FMDV induces a humoral response even when the virus does not spread from the initial site of inoculation; however, only the absence of antibodies against nonstructural proteins suggests that viral replication has not taken place (12). In order to determine if treatment with Ad5-poIRF7/3(5D) could provide protection against FMDV replication, the sera of treated animals were evaluated for the presence of antibodies against viral nonstructural proteins. As seen in Fig. 3C and andD,D, antibodies against FMDV 3ABC were below the detection limit in the animals treated with Ad5-poIRF7/3(5D) even at the lowest dose tested (108 PFU/animal). Altogether, the absence of virus or viral RNA in serum as well as in nasal secretions, in addition to the absence of antibodies against FMDV nonstructural protein 3ABC, indicates that treatment with Ad5-poIRF7/3(5D) fully protects swine against FMDV replication.
IRF7 has been defined as the master regulator for IFN-α expression (18). As such, expression of the fusion protein IRF7/3(5D) induces the production of IFN in cells that are directly infected by Ad5-poIRF7/3(5D) vector (20). The levels of systemic antiviral activity and porcine type I IFN were measured at 1 dpi (0 dpc) and at 2 dpi (1 dpc) with Ad5-poIRF7/3(5D), by using a VSV bioassay and MxA-CAT ELISA, respectively (Fig. 4A and andB).B). Sera from all animals showed no or low basal levels of antiviral activity the day before Ad5 administration (data not shown), and little or no induction of antiviral activity was detected in the control group treated with Ad5-Blue at 1 and 2 dpi. However, antiviral activity increased in groups of animals treated with Ad5-poIRF7/3(5D) by 1 dpi and decreased considerably by 2 dpi. Average type I IFN activity and average MxA-CAT ELISA activity at 1 dpi ranged between ~250 and 450 IU/ml of serum in the group treated with of Ad5-poIRF7/3(5D) (1010 PFU/animal). Consistently, a lower induction of antiviral activity (~100 to 200 IU/ml) was detected when a lower dose of Ad5-poIRF7/3(5D) (109 PFU/animal) was used (Fig. 4A and andB).B). Increased antiviral activity in all groups treated with Ad5-poIRF7/3(5D) was associated with upregulation in expression of known ISGs such as OAS1 (Fig. 4C) and Mx1 (Fig. 4D), RANTES (Fig. 4E), IRF7 (Fig. 4F), and USP18 (Fig. 4G) as well as the cytokine IFN-β (Fig. 4F) at 1 dpi. In experiment 2, we observed a tendency of ISG and IL-10 upregulation in pigs treated with 108 or 109 PFU/pig of Ad5-poIRF7/3(5D) compared to the control group, but the differences were not statistically significant (data not shown). It is noteworthy that in this experiment, for unknown reasons, the levels of ISG basal expression in one of the control animals were unexpectedly high.
To better understand the mechanism of protection induced by Ad5-poIRF7/3(5D) treatment, the expression of other cytokines and chemokines was measured. Relative transcript levels were determined in total RNA isolated from PBMCs of Ad5-poIRF7/3(5D)- or Ad5-Blue-treated pigs (experiment 1). Transcript levels of IP-10 and IL-10 (Fig. 5) were significantly increased in groups treated with Ad5-poIRF7/3(5D) compared to those in the control group. While transcript levels for IP-10 were the most upregulated (>9-fold), only a mildly significant increase was detected for IL-6 or IL-10 after Ad5-poIRF7/3(5D) treatment. Significant downregulation of IL-1B protein was observed after Ad5-poIRF7/3(5D) treatment in experiment 2. Although the same tendency was observed in experiment 1, differences were not statistically significant (data not shown). Similarly, other cytokines, such as IL-15, IL-12, and tumor necrosis factor alpha (TNF-α), were not differentially expressed after Ad5-poIRF7/3(5D) treatment compared to the control (data not shown).
Previously, we reported the antiviral activity of poIRF7/3(5D) against FMDV serotype A12 in vitro in tissue culture or A24 in vivo in mice (20). We therefore investigated whether poIRF7/3(5D) has antiviral properties against all 7 FMDV serotypes. Porcine (SK-6) cells were transduced for 24 h with Ad5-poIRF7/3(5D) or Ad5-Blue (control), followed by challenge with FMDV serotype A, Asia, C, O, or SAT1/2/3. FMDV yields were determined at 24 hpi. As shown in Fig. 6, FMDV yield was substantially reduced in all cases when cells were transduced with Ad5-poIRF7/3(5D) compared to Ad5-Blue. The extent of inhibition varied according to the tested serotype or subtype. A reduction of approximately 5 to 7 log10 was detected for A5, A12, Asia 1, O1 Caseros, and SAT1, and a reduction of at least 4 log10 was detected for SAT2 and SAT3 viruses; however, a difference of only 3 to 5 log10 was detected for O1 Campos and C3 Resende. Thus, expression of poIRF7/3(5D) effectively reduces viral growth of all 7 FMDV serotypes in porcine cells.
In the past, we have shown that treatment with Ad5-poIRF7/3(5D) protects mice from lethal FMDV challenge and that the antiviral activity induced by Ad5-poIRF7/3(5D) is only partially reversed by the addition of the inhibitor B18R, a product of vaccinia virus that competes with IFN protein for binding to the type I IFN receptor (20). To elucidate whether Ad5-poIRF7/3(5D) induces a type I IFN-independent antiviral activity, we took advantage of our previous success with Ad5-poIRF7/3(5D) in the FMD mouse model (20) and the availability of IFNAR KO C57BL/6 mice.
WT or IFNAR KO mice were treated with Ad5-poIRF7/3(5D), or Ad5-Blue and challenged with FMDV A24 48 h later (Fig. 7A). While all mice treated with Ad5-Blue died by day 4 (WT) and 3 (IFNAR KO), an 80% survival rate was detected in WT mice treated with Ad5-poIRF7/3(5D). Ad5-poIRF7/3(5D) treatment did not protect IFNAR KO mice from viral infection, as all mice died by day 2 (Fig. 7B). Consistently, WT and IFNAR KO mice treated with Ad5-Blue and IFNAR KO mice treated with Ad5-poIRF7/3(5D) had average viremia levels of >106 PFU/ml at 1 and 3 dpc, respectively. In contrast, viremia was not detected at 3, 5, or 7 dpi in WT mice treated with Ad5-poIRF7/3(5D), and lower viremia (3 × 104 PFU/ml) was detected in 2 out of 5 animals at 1 dpi (Fig. 7C). These data suggest that Ad5-poIRF7/3(5D) inoculation does not protect mice from FMDV challenge in the absence of functional type I IFN signaling.
Interestingly, when MEFs derived from WT or IFNAR KO mice were mock, Ad5-poIRF7/3(5D), or Ad5-Blue transduced for 24 h followed by infection with an FMDV adapted to infect murine cells (22), an approximately 5-log10 reduction in viral titer was observed in IFNAR KO MEFs previously treated with Ad5-poIRF7/3(5D) compared to the same cells mock or Ad5-Blue transduced (Fig. 7D). As expected, FMDV grew to higher titers in MEFs lacking a functional type I IFN response than in WT MEFs. Antiviral activity induced by Ad5-poIRF7/3(5D) transduction in IFNAR KO MEFs was only partially reversed when cells were independently pretreated with a JAK inhibitor which blocks type I and III IFN pathways or with a mouse IFN-λ2 neutralizing antibody before Ad5-poIRF7/3(5D) transduction (Fig. 7E). However, the antiviral activity induced by Ad5-poIRF7/3(5D) was fully reversed when cells were pretreated with a combination of mouse IFN-λ2 neutralizing antibody and the JAK inhibitor, suggesting that type I and III IFN signaling was responsible for the antiviral activity observed in IFNAR KO MEFs.
One of the most serious concerns of current FMDV vaccines is the limited capacity to provide rapid cross-protection against all FMDV serotypes. Vaccines are formulated with specific strains depending on the geographical area of interest, and they require 5 to 7 days to provide effective protection (2, 37, 38). Here, we show that in swine, administration of Ad5-poIRF7/3(5D) has potential as an effective biotherapeutic strategy against FMD as early as 24 h posttreatment. Two independent pig experiments, using Ad5-poIRF7/3(5D) doses ranging from 108 to 1010 PFU/animal, demonstrated the effectiveness of this agent in rapidly inducing protection against FMD in a highly susceptible natural host. Sterile immunity has been described as the absence of productive viral replication in animals challenged with FMDV (12). Detection of antibodies against the FMDV nonstructural protein 3ABC is considered a sensitive indicator of present or past infection with FMDV (39). To assess whether Ad5-poIRF7/3(5D) treatment effectively blocked productive FMDV replication in animals that did not show clinical signs, we assayed for viremia, virus shedding, and viral RNA in serum or nasal samples and antibodies against viral nonstructural protein 3ABC. Except for an almost negligible increase in FMDV neutralizing antibody titers measurable only at 7 dpc, virus replication could not be observed in animals treated with Ad5-poIRF7/3(5D) (experiment 1). It is possible that this low antibody response was elicited by immune detection of the challenge virus at the inoculation site. Indeed, the lack of antibodies against nonstructural proteins in groups treated with Ad5-poIRF7/3(5D) indicated that viral replication was absent or minimal after challenge. Thus, our results indicated that pretreatment of pigs with Ad5-poIRF7/3(5D) prevented FMDV replication and induced sterile protection even when the virus was directly inoculated in the epithelium of the heel bulb.
In our study, systemic antiviral activity and IFN levels increased in groups of animals treated with Ad5-poIRF7/3(5D) by 1 dpi and waned by 2 dpi. Previous reports have demonstrated that antiviral activity or IFN levels in serum are not good predictors of the extent or duration of the induced protection against FMD (16). For instance, pigs treated with Ad5-pIFNα had no detectable IFNα in their plasma prior to challenge at 7 dpi, while they showed delayed and lower levels of viremia and vesicular lesions after challenge (16). In the same studies, pigs inoculated with a combination of Ad5-A24 and Ad5-pIFNα and challenged 5 days later had no detectable antiviral activity or IFN levels by the day of challenge but were fully protected from FMD (16). Furthermore, studies with swine with Ad5-IFNλ have also shown that animals can be fully protected against FMD when challenged at 24 hpi, in the absence of detectable systemic antiviral activity (14). It is known that IFN-α/β has immunomodulatory roles and enhances the humoral immune response to soluble antigens by stimulating dendritic cells (40); in fact, it has been demonstrated that type I IFN functions as an adjuvant of an Ad5-FMD vaccine in swine (41). Therefore, it is possible that IFNs induced by Ad5-poIRF7/3(5D) may also function as an adjuvant when combined with FMDV vaccines.
The livestock industry demands minimal allocation of expenses for therapeutic agents to maintain lower prices for the consumer. For this reason, although genetically modified adenoviruses are attractive vectors for the delivery of exogenous DNA to mammalian cells for therapeutic applications (42), large-scale production is costly and generates concern on the applicability of such an approach in the field. Here, we show that a dose of 108 PFU of Ad5-poIRF7/3(5D) protects swine against FMD when administered 24 h before challenge. This dose is at least 10 times lower than the minimum protective dose (MPD) reported for an identical vector expressing IFN-α, Ad5-poIFN-α (11), and 1,000 times lower than the MPD reported for an industrially produced Ad5-poIFN-α (1011 FFU/animal), which are necessary to completely protect swine against A24, O1 Manisa, and Asia-1 (12). These results highlight the potential biotherapeutic application of Ad5-poIRF7/3(5D) in livestock. However, future animal experiments are needed to evaluate the MPD, duration of protection, and the effectiveness of Ad5-poIRF7/3(5D) against other FMDV serotypes.
A dose-dependent significant induction of ISGs such as the genes for OAS, Mx1, and IFN-β was observed after treatment with Ad5-poIRF7/3(5D) at 109 or 1010 PFU/ml in experiment 1, and a similar tendency of ISG upregulation was detected in pigs treated with 108 or 109 PFU/pig of Ad5-poIRF7/3(5D) in experiment 2, but the difference was not statistically significant. It is noteworthy that one of the control pigs had increased levels of ISGs before FMDV challenge and a consistent delay in the appearance of FMD clinical signs. This result suggested that the IFN pathway in this pig might have been primed for reasons unrelated to the current experiment, therefore affecting the group mean and the statistical significance on ISG expression analysis.
Changes in the inflammatory response prior to viral challenge were observed at the transcript level in pigs treated with 109 or 1010 PFU/animal of Ad5-poIRF7/3(5D) (experiment 1). In addition, pigs inoculated with Ad5-poIRF7/3(5D) had higher transcript levels of type I IFNs, IL-6, IL-10, and IP-10 than control pigs. IP-10 has been shown to mediate the IFN-α-induced mechanisms of protection against FMD (30, 33). Interestingly, major anti-inflammatory cytokines, such as IL-6 and IL-10 (43), were also upregulated after Ad5-poIRF7/3(5D) treatment. Consistent with this anti-inflammatory response, transcript levels of IL-1β, a proinflammatory cytokine (43), were downregulated after Ad5-poIRF7/3(5D) treatment. At the protein level, a trend for upregulation of IL-10 and downregulation of IL-1B was observed in both pig experiments after Ad5-poIRF7/3(5D) administration, but in some cases, the differences were not statistically significant, probably due to the high variability across experimental subjects (data not shown). Induction of a negative-feedback pathway downstream of IFN has been shown to subdue the inflammatory response and reduce IL-1β production (44). Type I IFN inhibits IL-1 production by suppressing caspase-1-dependent IL-1β maturation and by inducing IL-10, which reduces the abundance of pro-IL-1α and pro-IL-1β (45), and many ISGs are induced during an inflammatory response (44). We detected upregulation of USP18, an ubiquitin-specific protease induced by type I and type III IFNs that specifically inactivates the IFN-α response (46). A dynamic balance seems to exist between ISGs and proinflammatory and anti-inflammatory components after Ad5-poIF7/3(5D) administration that allows potent antiviral effects regulated by complex mechanisms.
Previous reports have demonstrated the role of IRF7 as a major regulator of both type I and type III IFN responses in epithelial cells (47). Similarly, we demonstrated the involvement of type III IFN in the antiviral response induced by Ad5-poIRF7/3(5D) in IFNAR KO MEFs (Fig. 7D and andE).E). It is noteworthy that the antiviral activity induced by Ad5-poIRF7/3(5D) in IFNAR KO MEFs was only partly reverted after addition of a neutralizing anti-IFN-λ2 antibody. Perhaps the dose of added antibody was not sufficient to fully neutralize all IFN-λ produced. Alternatively, the anti-IFN-λ2 antibody might have not neutralized other subtypes of IFN-λ, such as IFN-λ3 (IL28B), despite the high amino acid sequence homology (48). In support of these hypotheses, we observed that the antiviral activity induced by Ad5-poIRF7/3(5D) in IFNAR KO MEFs was neutralized when the anti-IFN-λ2 antibody was combined with a JAK inhibitor, a molecule that abrogates the JAK-STAT signaling induced by type I and III IFNs (17). Additional experiments are needed to elucidate if additional molecules activated in a JAK-dependent manner could have also played a role in the antiviral activity observed.
IFNAR KO mice were not protected from FMD lethal infection after Ad5-poIRF7/3(5D) treatment, suggesting that the type III IFN contribution is not sufficient to protect mice against FMDV challenge. In agreement with this hypothesis, the antiviral response observed when IFNAR KO MEFs were treated with Ad5-poIRF7/3(5D) could be the result of the use of a mouse-adapted FMDV strain that was attenuated in this system (22), while experiments in vivo were conducted using a lethal dose of virulent WT FMDV. It is also possible that the pathway induced by Ad5-poIRF7/3(5D) to protect against FMDV in vivo is exclusively type I IFN dependent. Interestingly, the contribution of type III IFN (poIFN-λ3) to prevent FMDV viremia and delay clinical signs in swine has been previously reported (14). Perhaps the induction of type III IFN also contributes to the robust antiviral response induced after Ad5-poIRF7/3(5D) administration in swine. Further experiments are needed to address this hypothesis.
We also show that treatment with Ad5-poIRF7/3(5D) can limit viral replication of all 7 FMDV serotypes in porcine cells. Although the degree of inhibition varied by the serotype, a reduction of 4 to 7 log10 in viral titer was detected for most serotypes, except O1 Campos and C3 Resende, which have been recognized as fast replicating and highly virulent (49, 50). For SAT2 and SAT3 FMDV, treatment with Ad5-poIRF7/3(5D) reduced viral titer approximately 4 log10, but overall, these viruses reached lower endpoint titers, probably due to differences in adaptation to propagate in SK-6 cells (51).
In summary, our work identifies Ad5-poIRF7/3(5D) as a new effective biotherapeutic candidate able to induce full protection against FMD in swine and sheds light on the mechanism involved in inhibition of FMDV replication. In addition, in vitro studies demonstrated the effectiveness of Ad5-poIRF7/3(5D) against all 7 FMDV serotypes. Future studies will determine the effectiveness of the IRF7/3(5D) constructs against other relevant FMDV serotypes, also including evaluation in other susceptible species, evaluation of duration of poIRF7/3(5D)-mediated protection, and the effect of therapeutic combinations to induce rapid and long-lasting protection against FMD.
This work was funded by ARS-CRIS project 1940-32000-057-00D and interagency agreements with the Science and Technology Directorate of the U.S. Department of Homeland Security (award numbers HSHQDC-11-X-00189 and HSHQPM-13-X-00113) and Specific Cooperative Agreement number 58-1940-4-003 between the USDA and University of Connecticut. L.R.-C. and E.R.-M. are recipients of a Plum Island Animal Disease Center Research Participation Program fellowship, administered by the Oak Ridge Institute for Science and Education (ORISE) through an interagency agreement with the U.S. Department of Energy.
We acknowledge James Zhu for helpful discussions in the conception of the hypotheses tested in this work. We thank Gisselle Medina, Marla Koster, Ignacio Fernandez, and Danielle Hickman for technical and scientific support and Marvin J. Grubman for critical reading of the manuscript. We are grateful to the PIADC animal research branch for professional assistance with animal experiments.