The antiviral efficacy of interferons (IFNs) was evaluated using a vaccinia intranasal infection model in mice in this study. We provide evidence that intranasal administration of IFN-α and IFN-γ (days −1 to +3) resulted in 100 and 90% survival against a lethal respiratory vaccinia infection (8 LD50) in mice, respectively; whereas no animals in the placebo group survived through the study period (21 days). The IFN treatment consisted of a single daily dose of 5×103 U per mouse for 5 consecutive days. The efficacy of IFN-γ was evident even when the IFN-γ treatments started 1–2 days after infection and when a lower dose (2×103 U per mouse) was used. The treatment of IFN-α and IFN-γ reduced the virus titers in the lungs of infected mice by 1000–10,000-fold, when the administration started 1 day after infection. Our data suggest that IFN-α and IFN-γ are effective in protecting vaccinia-infected mice from viral replication in lungs and mortality, and may be beneficial in other human orthopoxvirus infections.
Smallpox and monkeypox are among the most dangerous potential bioterrorism agents . Currently, there is no approved treatment for smallpox; vaccination is the only effective means for pre- and post-exposure prophylaxis for smallpox (if administered within 4 days of exposure) . However, post-vaccination complications have been reported in a small percentage of people, especially those with immunodeficiencies and eczema . Poxviruses may also be genetically engineered to resist current therapies and evade vaccine-induced immunity. In light of the bioweapon potential of smallpox and monkeypox, the search for ways to treat smallpox and monkeypox is especially important.
Poxviruses are sensitive to interferon (IFN)-α/β and IFN-γ in vivo, as mice lacking IFN-α/β or IFN-γ receptor are highly susceptible to poxvirus infections [3,4]. In mice, monkeys, or humans, IFN-α and IFN-γ have been demonstrated to have prophylactic effect against various viruses such as Rift Valley fever (RVF) virus  and hepatitis C virus . Crude IFN-α/β and recombinant IFN-γ effectively reduced lesion formation in a vaccinia tail lesion model . In this report we demonstrate that IFN-α and IFN-γ are effective in protecting vaccinia-infected mice from viral replication in lungs and mortality, and that they may be beneficial in treating other human orthopoxvirus infections.
2 Materials and methods
2.1 Viruses and cells
The WR strain of vaccinia virus (VV) (ATCC #VR-119) was purchased from ATCC (Manassas, VA, USA), propagated in human cervical adenocarcinoma HeLa cells (ATCC #CTL-2.2), and titrated on African green monkey kidney BS-C-1 cells (ATCC #CCL-26) as previously described [8,9]. BS-C-1 and HeLa cells were maintained in Eagle's minimal essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), 100 units (U) ml−1 of penicillin, and 100 µg ml−1 of streptomycin (Gibco-BRL, Rockville, MD, USA). All tissue culture media and phosphorus-buffered solution (PBS) were purchased from Gibco-BRL.
Recombinant murine IFN-αA (4.8×107 U mg−1) and IFN-γ (4.28×106 U mg−1) were purchased from BioSource International, Inc. (Camarillo, CA, USA) and prepared in PBS/0.1% bovine serum albumin (BSA) immediately prior to use.
Female BALB/c mice (5-week-old) were purchased from Charles River Laboratories (Wilmington, MA, USA) and quarantined 1 week prior to use. Female, 15–18 g, BALB/c mice were used in the virus challenge and IFN treatment experiments. Each experimental group was composed of 10 or 20 animals.
2.4 Intranasal challenge with virus and administration with IFN
BALB/c mice were infected intranasally with VV (1×105 plaque-forming units (PFU) per mouse) in a volume of 20 µl per mouse following anesthesia with Avertin (intraperitoneally (i.p.), 0.4–0.6 mg kg−1) . A single stock of VV was used for all experiments. The mice were treated intranasally with IFN-α (2–5×103 U), IFN-γ (2–5×103 U), or PBS/0.1% BSA (placebo) in a 20-µl volume following anesthesia with Avertin. For 21 days, the animals were monitored daily for mortality and weighed every 2–3 days.
2.5 Determination of lung virus titers
The lungs of mice (three mice per group) were taken aseptically on day 6, weighed, frozen at −70°C, homogenized in MEM-2.5, briefly sonicated, and centrifuged at low speed  before plaque assay on BS-C-1 . The virus titers in lungs were expressed as log10 (PFU g−1).
2.6 Statistical analysis
Statistical analyses were performed to compare the treated and placebo groups by two-tailed analyses. Survival curves were created by the Kaplan–Meier method using GraphPad Prism® 4 software, and compared using the Mantel–Haeszel logrank test . Fisher's exact test was used to compare the survivors in the placebo and the treated groups. Student's t-test was used to determine if there was a significant difference in the body weight of the mouse and virus titers between the treated and the placebo groups. The P values less than 0.05 were considered statistically significant.
3.1 Intranasal challenge
We successfully tested the respiratory infection model of VV in mice [13,11]. The LD50 of our virus stock was 104.1 PFU per mouse using the Reed–Meunch test . Intranasal inoculation of 8 LD50 (1×105 PFU) of VV resulted in 90–100% mortality of BALB/c mice (6-week-old), depending on the batch of mice. As demonstrated by Smee et al.,  intranasal challenge of BALB/c mice with VV results in pneumonia, remarkable weight loss and death. The infected mice displayed the ruffled furs, arched backs, reduced mobility, and significant weight loss, that are typical symptoms of vaccinia infection .
3.2 Pre-exposure prophylactic administration of IFNs
We first examined whether IFN treatment beginning before virus infection is effective against the lethal respiratory poxvirus infections. We chose to test this treatment using intranasal administration as this route is the most practical and likely to be employed in the treatment of large civilian and military populations in the event of a bioterrorist attack. Our IFN treatment consisted of a single daily dose for 5 consecutive days, starting 1 day before the viral infection. Intranasal administration of IFN-α (5×103 U per mouse) and IFN-γ (5×103 U per mouse) resulted in animal survival rates of 100 and 90%, respectively (Fig. 1). In contrast, no mice in the placebo survived the virus challenge. Doubling the doses of IFN-α and IFN-γ did not result in a significant increase in survival rates (±5%, data not shown). The IFN treatments did not significantly reduce weight loss (10–12%) on day 6 compared to the placebo, although the treated mice regained weight thereafter (Fig. 2).
Effects of IFNs on the survival of mice infected with VV. Groups of 20 BALB/c mice were treated daily (intranasally) with 5×103 U of IFN-α, IFN-γ, or PBS (placebo) for 5 consecutive days. All animals were infected with 8 LD50 of VV on day 0 and were monitored 21 days for mortality. The treatments started 1 day before infection.
Effect of IFN treatment on the weight changes of mice following vaccinia challenge. Groups of 20 BALB/c mice were infected with 8 LD50 of VV on day 0, and treated intranasally with murine IFN-α (5×103 U), IFN-γ (5×103 U), or PBS (placebo) for 5 consecutive days (1, 0, +1, +2, +3). The animals were weighed individually every other day. Values represent the mean of mouse weights of each group. The data for uninfected mice are also shown.
3.3 Effect of start time of IFN-γ treatment
We then studied the effects of delayed treatment of IFN-γ on antiviral efficacy. A 100% survival rate was obtained when the five daily treatments were initiated 1 day before or the day of infection (Table 1). Starting the treatments on days +1, +2, and +3 resulted in 90, 70, and 50% survival rates, respectively. In this study, a 10% survival rate was found in the placebo. We observed a significantly higher survival rate in mice treated 2 days after infection compared to the placebo. There was no significant difference between day 0 and +1; however, the survival rate following day +2 treatment was significantly lower than that following day 0 treatment. Therefore, we chose to start treatment at day +1, although the difference between day +1 and +2 was not significant.
↵Mean day to death of mice that died prior to day 21.
3.4 Effect of IFN treatment on lung virus titers
Next, we studied the effect of IFN treatment on virus replication in lung tissues. The lung virus titers in the treated mice were 1000–10,000-fold lower than that in placebo (P<0.05, Fig. 3), indicating that IFN-α and IFN-γ are effective in suppressing vaccinia replication in the lungs of infected mice when the treatment is started 1 day after infection.
Effects of IFN treatment on the virus titer in the lungs of VV-infected mice. Groups of five BALB/c mice were treated daily (intranasally) with 5×103 U of IFN-α, IFN-γ, or PBS (placebo) for 5 consecutive days. All animals were infected with 8 LD50 of VV on day 0, killed on day 6 for collection of lungs. The virus titers in the lung were measured by plaque assay. The treatments started 1 day after infection.
3.5 Efficacy of lower doses of IFN-α and IFN-γ
We further examined the efficacy of lower doses of IFNs. We observed an 80% survival rate when 2000 U of IFN-γ was used, compared to the 90% survival that resulted from the 5000 U of IFN-γ-treated group; however, the difference was not statistically significant. In contrast, only 10% of the animals in placebo survived the vaccinia infection. When a lower dose of IFN-α (2000 U) was used, only 40% survival rate was seen compared to 90–100% survival when 5000 U of IFN-α was used.
We demonstrate that intranasal administration of IFN-α and IFN-γ can reduce lung virus titers and mortality due to respiratory vaccinia infection in mice. VV is sensitive to IFNs in vivo, as mice lacking IFN-α/β or IFN-γ receptor are highly susceptible to poxvirus infections . This is consistent with previous reports that vaccinia tail lesions were reduced greatly by five i.p. administrations of crude IFN within 24 h prior to infection . Monkeys treated with daily intramuscularly (i.m.) or intravenous (i.v.) injections of leukocyte IFN (5×105 U kg−1) from day −1 to day +7 after vaccination were completely protected and the severity of the skin lesions was decreased. Lesion scores correlated inversely with the dose of IFN . IFN-α is also effective against other viruses in mice and monkeys, such as St. Louis encephalitis virus , Venezuelan equine encephalitis (VEE) virus , and RVF virus .
The dose of recombinant murine IFN used (5×103 U) approximates that used in previous studies [20–22]. No apparent toxic effects were observed in mice that received a single daily intranasal dose of IFN-α (1×104 U) or IFN-γ (1×104 U) for 5 consecutive days as judged by insignificant weight loss and mobility reduction (data not shown). We chose an IFN regime consisting of a single daily intranasal dose (5×103 U) for 5 days because it provided the best protection without apparent toxic effects according to our preliminary and published work. The virus challenge dose (8 LD50 or 105 PFU) used in this study was moderate. Further studies are necessary to clarify the efficacy of IFNs against severe orthopoxvirus infections using 100 LD50 of VV.
The mechanisms by which IFNs prevent mortality resulting from vaccinia infection are not completely understood. One possible explanation for the reduced mortality in IFN-treated mice may be that VV replication is inhibited as a consequence of the direct antiviral effect of IFNs. VV titers in lungs of IFN-α- and IFN-γ-treated mice were reduced relative to titers in the placebo mice by approximately 3–4 log, which may account for the observed reduction in mortality. IFN-α/β and IFN-γ can induce an antiviral state by triggering the Janus kinase-signal transducer and activator of transcription (JAK-STAT) pathway and eventually by inducing the 2′,5′-oligoadenylate synthetase (OAS)/RNase L and dsRNA-dependent protein kinase (PKR) [23,24]. In addition, IFNs regulate immune responses mediated by effector cells such as natural killer (NK) cells, macrophages, and cytolytic T lymphocytes (CTL) .
The efficacy of IFN-γ has been evaluated in a variety of experimental infections, such as cytomegalovirus (CMV) infections in mice and rats [26,27]. Pretreatment of mice with anti-IFN-γ delayed LCMV elimination , increased mortality in mouse hepatitis virus (MHV)3-infected mice [29,30], and enhanced the spread of murine ectromelia virus to and replication in the spleen, lungs, ovaries and livers . Morrill and his colleagues demonstrated that prophylactic treatment of rhesus macaques with recombinant IFN-γ (104–106 U, i.m., days −1 to +3) reduced viremia and diminished the signs of RVF . IFN-γ can enhance the overall development of cell-mediated immune responses, macrophage activation, antigen presentation, and chemokine gene expression, direct antiviral effect, and induction of NO-mediated antiviral activity by macrophages [33,34], and regulation of IFN-α/β production . This supports the fact that the in vivo efficacy of IFNs tends to be higher when the IFN administration starts before the infection, allowing the effects of IFNs to have a ‘head start’ on the host cells during the race between virus replication and IFN activity. Taken together, both innate and adaptive immune responses are likely to play important roles in this IFN-induced antiviral effect.
Our data show that post-exposure IFN-γ treatment is effective against lethal vaccinia infection in mice. Previous reports showed that IFNs are effective against viral infections in animals usually when they are given prior to the infection. Pegynated IFN-α (0.4–5 ×105 U per mouse, i.p., days −2 to +5) was shown to be effective against aerosolized VEE virus . Treatment of mice with IFN-γ prior to infection (−24 h and −4 h, i.p. or i.m.) with murine CMV (MCMV) significantly reduced mortality in a dose-dependent manner; whereas IFN-γ administered after MCMV infection had no apparent effect . Prophylactic IFN treatment may be clinically useful in some viral infections where exposure to the virus can be predicted; however, post-exposure IFN-γ treatment would be more useful in antiviral therapy.
Our initial experiments demonstrate that IFN-α is slightly, but not significantly, more effective than IFN-γ in suppressing virus replication in lungs (Fig. 3) and virus-induced weight loss (Fig. 2); however, IFN-γ appears to be more effective when the treatment was delayed and a lower dose was used. IFN-γ has been suggested to replace IFN-α/β as it appears to be less toxic than IFN-α. In addition, we found IFN-γ, but not IFN-α, enhanced the in vivo anti-poxviral efficacy of other antiviral drugs (data not shown). As smallpox vaccination can cause serious complications in immunocompromised individuals and as poxvirus infection can cause an immunosuppressive effect , prophylactic treatment using IFN-γ may have greater efficacy than IFN-α. For these reasons, our further studies thus will focus on the IFN-γ application in control of poxvirus infection in mice.
Based on the above results, we concluded that intranasal administration of IFN-γ could be effective against poxvirus infection even when treatment is delayed and when a lower dose (2000 U) is used. Recombinant human IFN-γ (Actimmune®, Genentech, Inc.) has been approved to treat chronic granulomatous disease, a rare genetic disorder characterized by deficient phagocyte oxidative metabolism [36,37]. The Actimmune (IFN-γ1β) is recommended to be administered subcutaneously three times weekly (1.5×106 U m−2). To our knowledge, beneficial effects of post-exposure IFN-γ treatment have not been reported. Here, we provide the first evidence that IFN-γ can be used as an anti-poxviral post-exposure prophylaxis. Our data thus extend the application of this biological response modifier to control of poxvirus infection. IFNs and IFN inducers may also be beneficial in reduction of vaccine-induced inflammatory effects. For instance, simultaneous immunization and treatment of monkeys with IFN-α/β or IFN inducers induced immunity, prevented viremia, and lessened local inflammatory reactions caused by the vaccination . Intranasal administration of IFN-α and IFN-γ provides nearly full protection against respiratory VV infection in mice; therefore, our results have implications in the utilization of IFN as a prophylactic tool on a scenario of aerosol smallpox and monkeypox.
This work was supported by Defense Advanced Research Project Agency (MDA972–01-C-0084). We thank Drs. D.F. Smee, J.R. Bennink, and Edith Grene for many helpful discussions.
(1998) Preparation of cell cultures and vaccinia virus stocks. In: Current Protocols in Molecular Biology (AusubelF.M., BrentR., KingstonR.E., MooreD.D., SeidmanJ.G., SmithJ.A., StruhlK., Eds.), Vol. 3, pp. 16.16.1–16.16.5. Wiley, New York.
(1998) Titration of vaccinia virus stocks by plaque assay. In: Current Protocols in Molecular Biology (AusubelF.M., BrentR., KingstonR.E., MooreD.D., SeidmanJ.G., SmithJ.A., StruhlK., Eds.), Vol. 3, pp. 16.16.5–16.16.6. Wiley, New York.
(2000) Pegylated alpha interferon is an effective treatment for virulent venezuelan equine encephalitis virus and has profound effects on the host immune response to infection. J. Virol. 74, 5006–5015.
(1989) Mechanism of recovery from acute virus infection. VIII. Treatment of lymphocytic choriomeningitis virus-infected mice with anti-interferon-gamma monoclonal antibody blocks generation of virus-specific cytotoxic T lymphocytes and virus elimination. Eur. J. Immunol. 19, 1283–1288.