Dear Editor,

Interferons (IFNs) were first described for their ability to protect cells from viral infections 1. Since these initial reports, three major types of IFNs have been described. Type I IFNs, including IFNβ and multiple IFNα subtypes, induce antiviral gene programs through the IFNα/β receptor, IFNAR; many of these genes are directly or indirectly involved in curbing viral replication and spread 2. Type I IFNs have been used to treat both chronic viral infections, such as hepatitis B and C, and a variety of neoplastic conditions, such as melanoma, hairy cell leukemia, and non-Hodgkin's lymphoma 3, 4, 5, 6, 7. Recently, PEGylated IFNs, which show decreased clearance as compared to recombinant IFNs, have emerged as the standard of care. Although PEGylation has increased the therapeutic efficacy of IFN therapy, a wide variety of unpleasant and serious side-effects exist. In this study, we set out to determine whether antibody-IFN fusions could be used as an alternative and allow for specific targeting of IFNs, which we believe would be beneficial in reducing undesired side-effects.

We have previously generated chimeric molecules by fusing murine IFNα to the carboxyl-terminus of human IgG3 (IgG3-IFNα) 8. As IFNα is a potent antiviral cytokine, we examined the ability of IgG3-IFNα to activate the antiviral response and inhibit viral replication. Phosphorylated Stat1 was detected in 38C13 cells stimulated with 1 μg of IgG3-IFNα for 60 min (Figure 1A). A recombinant MHV-68 virus (MHV-68-Luc), in which the firefly luciferase gene under the control of the viral M3 promoter is integrated into the viral genome, serves as a convenient readout for viral replication in cultured cells and mice. To determine the ability of IgG3-IFNα to inhibit MHV-68 replication, 38C13 cells were infected with MHV-68-Luc and then treated with either IgG or IgG3-IFNα. IgG3-IFNα inhibited viral replication as measured by luciferase activity 2 days after infection (data not shown). To compare the antiviral efficiency of IgG3-IFNα with that of IFNα, 38C13 cells were infected with MHV-68-Luc virus and then treated with IFNα or IgG3-IFNα at the indicated concentrations. Luciferase activities were measured 2 days after infection. All experiments were performed in triplicate and repeated at least three times. As shown in Figure 1B, IgG3-IFNα was more effective in inhibiting viral protein expression across a wide range of concentrations.

Figure 1
figure 1

(A) 38C13 cells were stimulated with IgG3-IFNα for 30 min, and phosphorylation of STAT1 was assayed by western blot. (B) IgG3-IFNα protects cells from viral infection more effectively than IFNα across a wide range of doses. Anti-HER2-IgG3-IFNα offered better antiviral protection than IgG3-IFNα in 38C13-HER2 cells (see text for details). Cells were infected with MHV-68-Luc and treated with the indicated reagents. Cells were harvested 48 h after infection and assayed for luciferase activity. (C) IgG3-IFNα better inhibited the luciferase expression than IFNα in MHV-68-luc-infected mice. Mice were infected with 5 000 PFU of the virus and treated with the indicated reagents i.p. Bioluminescence was measured on day 5 after infection. (D) Bioluminescence readings of IgG3-IFNα-treated animals showed a statistically significant reduction in luciferase expression compared with the IFNα group in all measured planes (supine and one lateral plane are graphed). (E) On day 7 after infection, mice were euthanized and lung homogenates were assayed for viral load by plaque assays and q-PCR analysis. MHV-68 viral titers are effectively reduced in the lungs of IgG3-IFNα-treated animals compared with those treated with IFNα or IgG alone.

Full size image

In addition to increasing the potency, a potential advantageous feature of antibody-conjugated type I IFN is the possibility of using the antibody specificity to target type I IFN to specific cells. Therefore, we used anti-HER2-IgG3-IFNα, in which IFNα is fused to a HER2/neu-specific antibody, and 38C13 cells stably expressing the HER2/neu receptor (38C13-HER2). 38C13-HER2 cells infected with MHV-68-Luc were treated with IgG3-IFNα or anti-HER2-IgG3-IFNα. Luciferase activity was reduced more effectively after treatment with anti-HER2-IgG3-IFNα across a broad range of therapeutic doses (Figure 1B, right panel). Importantly, the difference between anti-HER2-IgG3-IFNα and IgG3-IFNα was more prominent at low concentrations, suggesting that anti-HER2-IgG3-IFNα increases the effectiveness of IFNα by targeting IFNα to HER2/neu-expressing cells. When parental 38C13 cells that did not express HER2/neu were used, both fusion proteins similarly inhibited viral replication (data not shown).

We used an intranasal model of infection with the MHV-68-Luc virus, followed by bioluminescence imaging, to determine the effectiveness of IgG3-IFNα in inhibiting viral replication in vivo. We first administered 5 000 PFU of MHV-68-Luc through nasal passages and then treated these mice intraperitoneally with 25 000 units of IFNα, 25 000 units of IgG3-IFNα (10 μg), or 10 μg of IgG3 alone. Mice were imaged on day 5 (Figure 1C). Bioluminescence readings of mice imaged in supine and lateral positions were obtained and are presented in the left and right panels of Figure 1D. Whereas mice treated with IFNα exhibited a slight reduction in bioluminescence readings compared with IgG-treated animals, IgG3-IFNα-treated animals exhibited a statistically significant seven-fold reduction in readings obtained in either position (P<0.0001 compared with the IgG-treated group). Indeed, mice treated with IgG-IFN had significantly reduced bioluminescence readings compared with the IFNα-treated mice (P<0.05).

On day 7 after infection with MHV-68, mice in each of the three groups were euthanized, and viral burden in the lung was measured by plaque assays using lung homogenates. qPCR analysis was also used to determine the copy number of the viral genome in the lungs of infected animals. IgG3-IFNα-treated animals exhibited a 100-fold reduction in viral burden as measured by the plaque assay (P<0.05, Student's t-test). Surprisingly, treatment with IFNα provided no protection against viral burden as measured by this assay (Figure 1E, left panel). Similarly, viral genome content in the lungs of IgG3-IFNα-treated animals was 600 times lower than that observed in IgG-treated animals, while IFNα treatment seemed insufficient to suppress viral genome production (Figure 1E, right panel). Thus, IgG3-IFNα proved to be a more potent antiviral agent both in vitro and in vivo.