IRF-1 is Essential for Immunity Against West Nile Virus Infection in Mice

The West Nile Virus (WNV) is spread via mosquitoes and has been affecting humans in the United States since 1999 (1). Current research by immunologist further dissects antiviral immunity, in hopes of understanding WNV infectivity and how it can be combated or prevented. Previous research in mice has identified several immune mechanisms of control against WNV including cytokines, chemokines, complement, B CD4+ and CD8+ T cells (reviewed in [1]). These are all various aspects of the innate and adaptive immune system that aid in fighting against pathogens. More specifically, type 1 IFN (IFN-αβ) has been given special attention, as mice deficient in it, rapidly succumb to WNV infection (2). Further, it was determined that IFN induction is dependent on certain transcriptional signals. Past research pointed Brien et al. (2011) in the direction of IRF-1, as it has been reported to contribute to IFN-β induction. For example, it was found that IRF-1 transcription factor activated IFN-β gene transcription and regulated genes that directly impeded replication of several viruses. Further roles of IRF-1 have been found in controlling herpes virus, even though mice lacking IRF-1 did not have defects in their type I IFN response. This would suggest that IRF-1 regulates IFN genes “in a pathogen and cell type-dependent manner” (1). Further research has also identified IRF-1 as a tumor suppressing gene and a regulator of the adaptive immune response (3).
To understand the role of IRF-1, Brien et al. (2011) assessed WNF infectivity in IRF-1 normal and IRF-1 deficient mice (1). As previously mentioned, they chose to manipulate IRF-1 because mice lacking it (IRF-1-/- ) were vulnerable to WNV infection. First, it was confirmed that IRF-1 is required for control of lethal WNV infection. After infection with WNV, Wild-type (WT) mice had a 65% survival rate and a mean time to death of 11 days, whereas IRF-1-/- mice had a 0% survival rate and a mean time to death of 9.5 days.
To better understand how IRF-1 deficiency is a disadvantage for mice with WNV infection, the ‘viral burden’ was measured at various points post infection in serum, several peripheral organs, and the central nervous system (CNS). Indeed, increased levels of viral RNA were found in the serum and lymph nodes in IRF-1-/- mice compared to the WT. Therefore, IRF-1 controls the early stages of WNV infection. Additionally, WNV infected the spleen more rapidly and clearance was delayed in IRF-1-/- mice. WNV was also detected sooner in the kidneys of IRF-1-/- mice, further suggesting that IRF-1 normally functions to control infection in peripheral tissues.

In CNS tissues, WNV spread more rapidly in IRF-1-/- mice compared to the WT. WNV virus was found in the brain of IRF-1-/- mice after four days, while the WT remained uninfected, and the virus remained in higher quantities in subsequent days in IRF-1-/- mice. A very similar pattern of early infection was found in the spinal cords, collectively suggesting that IRF-1 deficiency results in early spread and sustained replicated in CNS tissues.
To follow up on previous findings that IRF-1 is a regulator of IFN-αβ gene transcription, Brien et al. (2011) inquired as to whether IRF-1 might contribute to the induction of type 1 IFN responses in various tissues. They looked specifically to peripheral tissues because of the early viral phenotype found there. IFN-α and β mRNAs were measured in lymph nodes and serum and comparable measures were found in both IRF-1-/- and WT mice. Thus, mice deficient in IRF-1 did not diminish early type 1 IFN levels in lymph tissues or circulation after WNV infection. These results confirm that IRF-1-/- mice do not always show defects in type 1 IFN response (4).
Next Brien et al. (2011) wanted to see how the IRF-1-/- phenotype affects WNV infection in primary mouse embryonic fibroblasts (MEF) and MΦ (macrophages). MEFs were the control (no difference in viral replication in WT vs. IRF-1-/-) cell type to see the effect of WNV on macrophages, the physiologic targets for infection. Increased replication was observed 48 and 72 hours after infection in IRF-1-/- MΦ when compared to WT cells. However, levels of IFN-α and IFN-β were higher after 48 hours in IRF-1-/- MΦ, which was likely due to increased viral replication. Further, type 1 IFN response was also not altered in the MEF control condition. Therefore, IRF-1 does restrict WNV replication, as IRF-1-/- mice experienced enhanced viral growth in the MΦ. However, it was not related to type 1 IFN response.
Previous research has found a role of IRF-1 in IFN-γ signaling, so Brien et al. (2011) compared the contribution of IRF-1 to the antiviral effects of IFN-β and IFN-γ in MΦ after WNV (5). By pre-treating WT and IRF-1-/- MΦ with either IFN-β and IFN-γ and infecting them with WNV, they found that IFN-β inhibited WNV replicated in both mice types, whereas IFN-γ only decreased viral replication in the WT MΦ. The lack of effect of IFN-γ on IRF-1-/- MΦ suggests that IRF-1 is required to mediate the inhibitory effect of IFN-γ, but has no direct influence on the antiviral activity of IFN-β against WNV infection. To confirm the antiviral role of IFN-γ, experiments were performed with anti-mouse IFN-γ antibody and WNV infection was observed. The results imply that WNV-infected MΦ do not produce enough IFN-γ to account for the difference in viral growth between WT and IRF-1-/- cells, suggesting that intrinsic cell difference in viral replication in IRF-1-/- cells occurs independently of IFN-γ response. However, it is important to note that IRF-1 is still essential for mediating the antiviral activity of IFN-γ against WNV in MΦ.
To determine the effect of IRF-1 deficiency on B cell function, Brien et al. (2011) measured levels of WNV-specific IgM and IgG antibodies and found similar levels in both IRF-1-/- and WT mice five and eight days after infection. This confirms that viral infection in IRF-1-/- mice is not due to defective B cell functioning. However, previous research has shown that IRF-1-/- mice do have defective number of CD8+ T cells in peripheral lymphoid organs, as well as defective TH1 responses (6). This raised the possibility that the infectious phenotype in IRF-1-/- mice is due to a failure of the CD8+ T cells. Initial studies confirmed reduced CD8+ T cells in the spleens of IRF-1-/- mice, but also found an unchanged number of CD4+ T cells in IRF-1-/- and WT mice. However, after eights days of WNV infection the total CD8+ population expanded and 60% of them expressed high levels of granzyme B (GrB), creating a large number of potentially cytolytic CD8+ T cells. In the context of WNV infection, IRF-1-/- mice had a stronger antigen-specific T cell response compared to the WT.
The next step was to better understand the mechanism of expansion of CD8+ T cells in IRF-1-/- mice, as this result was not one that the authors had expected to find. To see if WNV-specific IRF-1-/- CD8+ T cells were of higher affinity than the wild type T cells, ex vivo experiments were performed and found no difference in the immune response. They did observe a decreased number of Tregs in the spleens of IRF-1 defective mice, which could cause a skewing of the CD8+ T cells response. However this explanation is ruled out because mice that lack Tregs all together still displayed the T cells response irregularity.
How exactly IRF-1 regulates the expansion of CD8+ T cells after WNV infection is still not understood, so Brien et al. (2011) performed competitive adoptive transfer experiments. They observed a cell-intrinsic effect of IRF-1 on the proliferative potential of CD8+ T cells, as the WT cells initially expanded faster compared to IRF-1-/- cells (consistent with observations above). Therefore, cells lacking IRF-1 are at an initial immunological disadvantage compared to the WT. In additional experiments, it was found that wild type CD8+ T cells generated a larger WNV antigen-specific response in the IRF-1-/- mice compared to the WT mice. Therefore, IRF-1 expression outside of the T cell also regulates the strength of the antigen-specific CD8+ T cell response during WNV infection. It is clear that IRF-1-/- mice experience rapid expansion of antigen-specific CD8+ T cell response. Additionally, a significant higher percentage of WNV-specific CD8+ T cells in IRF-1-/- mice were proliferating (when isolated), confirming that IRF-1-/- CD8+ T cells have an intrinsic capacity to proliferate more rapidly when stimulated through the T cell receptor. Therefore, IRF-1 acts to regulate T cell proliferation.
So what does this accumulation of CD8+ T cells mean for IRF-1-/- mice? First, it was confirmed that in addition to accumulating in the spleen of IRF-1-/- mice, CD8+ T cells also accumulate in the CNS, or brain of WNV-infected, IRF-1 deficient mice. To determine the significance of this, more adoptive transfer studies were performed with WNV-primed CD8+ T cells. WNV-primed CD8+ T cells significantly reduced the virus in the brain, with the greatest inhibitory effect coming with IRF-1-/- mice, compared to the WT. Therefore, WNV-primed IRF-1-/- CD8+ T cells control brain infection, but cannot compete to with the ability of the virus to continue replication in peripheral tissues. Ultimately, viral replication in the peripheral tissues results in early and successful infection in IRF-1-/- mice.
In conclusion, Brien et al. (2011) found that IRF-1 deficient mice allowed enhanced viral replication in peripheral tissues and thus spread to the brain and spinal cord. Further, it was determined that despite a baseline decrease in the number of naïve CD8+ T Cells after WNV infection, IRF-1-/- mice actually rapidly expanded antigen specific B+ CD8+ T cells that were capable of killing and clearing the virus. Unfortunately, even though CD8+ T cells proliferated more rapidly in IRF-1-/- mice, they were not able to reach sufficient numbers quickly enough to fix the damage caused early in the CNS. The early damage was due to a lack of control over WNV replication in peripheral tissues. Therefore, IRF-1 has an essential role in innate and adaptive immune responses against WNV infection. This study has important implications for human immunity as the IRF-1 dependent antiviral phenotype is consistent with recent studies in human cells, “which showed broad antiviral activity of IRF-1 against a range of viruses” (1).

Resources:
(1) Brien et al. (2011). Interferon Regulatory Factor-1 (IRF-1) Shapes Both Innate and CD8+ T Cell Immune Response against West Nile Virus Infection. PLoS Pathog 7, e1002230.
(2) Samuel MA, Diamond MS (2005) Type I IFN protects against lethal West Nile Virus infection by restricting cellular tropism and enhancing neuronal survival. J Virol 79: 13350–13361.
(3) Tanaka N, Ishihara M, Kitagawa M, Harada H, Kimura T, et al. (1994). Cellular commitment to oncogene-induced transformation or apoptosis is dependent on the transcription factor IRF-1. Cell 77: 829–839.
(4) Reis LF, Ruffner H, Stark G, Aguet M, Weissmann C (1994) Mice devoid of interferon regulatory factor 1 (IRF-1) show normal expression of type I interferon genes. Embo J 13: 4798–4806.
(5) Schroder K, Hertzog PJ, Ravasi T, Hume DA (2004) Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol 75: 163–189.
(6) Penninger JM, Sirard C, Mittrucker HW, Chidgey A, Kozieradzki I, et al. (1997) The interferon regulatory transcription factor IRF-1 controls positive and negative selection of CD8+ thymocytes. Immunity 7: 243–254.