1. Twenty-three genes and 17 proteins associated with antiviral activity have been identified. Our laboratory identified 25 genes (by microarray), 23 of which were confirmed by qRT-PCR, and 17 proteins (by Western Blotting) associated with the antiviral activity of Type I IFN using conditions we established which allows us to distinguish, in a single MTT assay, between the antiproliferative and antiviral activities of IFN-alpha-treated Daudi (Burkitts Lymphoma) cells. 2. IFIT3 is associated with the antiviral activity of IFN-alpha Our work has focused on the examination of two of these proteins: The first is Interferon-induced protein with tetratricopeptide repeats 3 (IFIT-3) which is also known as Retinoic Acid Induced Gene G (RIG-G), ISG-60 or IFI60. RNA interference of IFIT3 in A549 cells and overexpression of IFIT3 in VERO cells confirmed that IFIT3 is one of the proteins associated with antiviral activity against both Vesicular Stomatitis Virus (VSV) and Endomyocarditis Virus (EMCV). In addition, IFIT3 was induced in both adherent cells (OVCAR-3, 2fTGH and HeLa) and in suspension cells (B-JAB, U937, Jurkat and human monocytes) by IFN-alpha2. We have further elucidated the antiviral role of IFIT3 by showing that overexpression of IFIT3 decreases HCV replication in HUH7 cells and dengue virus replication in HEK293 cells. 3. HIV suppression by IFN- engineered constructs Parental molecules IFN-alpha2c and IFN-alpha21b were used to generate hybrids as well as both site-directed and cassette mutants. From the hybrid designated HY4 (IFN-alpha21b (1-75)/IFN-alpha2c (76-81)/IFN-alpha21b (82-95)/ IFN-alpha2c (96-166), three mutants were generated designated SDM-1 (S86Y), SDM-2 (N90Y), and CM-3(S86K) (12). Mutants SDM-1 and-2 effectively inhibited HIV at concentrations similar to the parental molecules and exceeded the activity of the hybrid from which they were derived (HY4). PKR expression was increased by all IFN-alpha constructs tested compared to untreated macrophages. There was no direct correlation between this increased expression and levels of HIV suppression. Interestingly, the site-directed mutant SDM-1 led to up-regulation of APOBEC3A in macrophages and monocytes at much higher levels than that seen with parental IFNs, HY4 or the CM-3 mutant. These results point to the tyrosine and likely asparagine amino acid residues at positions 86 and 90 respectively as being important not only for regulating AV activity but also for APOBEC induction. IFN-induced phosphorylation of STATs (1,2,3,5 and 6) appears to correlate with the regulation of AV activity and APOBEC3A. The possible link between the antiretroviral activity and toxicity of SDM-1 and 2 was investigated using the IFN-induced indoleamine 2,3-dioxygenase (IDO) as a marker. It was found that both SDM-1 and SDM-2 induced appreciably lower levels of IDO than the parental IFN molecules. We have found that whereas parental IFN-alphas substantially increased the phosphorylation of I-kappaB-alpha which is consistent with the NF kappa B pathway, SDM-1 did not. This may explain the apparent favoring of SDM-1 for AV activity and decreased inflammatory or toxic properties. Since IFN is known to synergize with TNF-alpha for IDO transcription and that TNF-alpha can facilitate HIV infection (13), we investigated the possibility of TNF-alpha being an intermediate link in IFN toxicity. Whereas parental IFNs resulted in rapid and plentiful levels of TNF-alpha both at the gene and protein levels, the same was not true for SDM-1 and SDM-2. In addition, direct addition of exogenous TNF-alpha to macrophages led to enhanced IDO expression even at low (1 ng/ml) concentrations. However, TNF-alpha was not effective for the stimulation of APOBEC3A regardless of concentration which is consistent with dissociation of the two pathways. We would like to further examine the effects of SDM-1 and -2 compared to IFN-alpha2 on other viruses in human monocytes/macrophages and evaluate IDO, TNF-alpha and other biomarkers of toxicity/inflammation. Also we will further examine the mechanism of dissociation of the AV and toxicity pathways of SDM-1 and-2. Finally new IFN constructs will be generated to optimize the therapeutic index by evaluating their bioactivities and markers of toxicity/inflammation. We will examine the potential of these constructs for further product development.
