Glycoengineering strategies are important and useful tools for the improvement of therapeutic drugs pharmacokinetics. Human IFN-alpha2b (hIFN-alpha2b) is a low molecular weight cytokine that has been used for the treatment of many viral and tumor diseases. However, its low stability and short plasma half-life is responsible
for the administration of high and repeated doses to reach the desired effect. IFN4N constitutes a hyperglycosylated hIFN-alpha2b that was developed in our laboratory by glycoengineering strategies. This new molecule exhibited reduced in vitro antiviral and antiproliferative activity compared to the non-glycosylated IFN.
However, IFN4N showed improved pharmacokinetic properties that lead to a higher in vivo antitumor activity. Since the mutation R23N has been identified as the main responsible for the reduction of IFN4N in vitro antiproliferative ability, here we propose the design, production, purification and characterization of new highly
glycosylated hIFN-alpha2b muteins with better in vitro and in vivo biological activity. Different groups of muteins were designed in order to reach this goal. Group A involved IFN variants with similar glycosylation degree but higher in vitro antiproliferative activity compared to IFN4N. Group B gathered muteins with higher glycosylation degree but lower in vitro activity (R23N mutation was present). Group C combined the best mutations into new muteins that exhibited improved in vitro antiproliferative activity as well as higher glycosylation degree. The new IFN variants were produced in CHO-K1 cells and purified by immunoaffinity chromatography. Physicochemical analyses revealed that the higher the number of potential N-glycosylation sites, the higher the increment in the number of acidic isoforms. As a result, the isoelectric point of the hyperglycosylated variants was reduced. In terms of in vitro biological activity, all hyperglycosylated hIFN-alpha2b variants with restored R23 presented a notably increased in vitro antiproliferative specific biological activity, demonstrating that the groups of muteins were properly designed. Pharmacokinetic studies performed in Wistar rats also showed that the higher the apparent molecular mass, the slower the plasmatic clearance. Finally, in vivo experiments performed in nude mice implanted with prostate cancer derived cells revealed that the new hyperglycosylated variants were able to reduce tumor growth rate. Particularly, muteins showing the same glycosylation degree (IFN4N and group A) slowed the tumor growth rate in a greater extent if the R23N mutation was absent. Nevertheless, muteins from groups B and C demonstrated a lower tumor growth rate in comparison with IFN4N (p<0.001) or muteins from group A  (p<0.001), independently of R23 mutation. Our results showed that properties conferred by hyperglycosylation would be more important than receptor affinity to slow the growth rate of IFN-treated tumors.