Fig. 1

In silico analyses revealed possible protein kinase C phosphorylation sites on dengue NS5. a Prediction of possible NS5 kinases. A heat map showed possible human kinases that phosphorylate DENV2 NS5 based on in silico predictions by Scansite™, NetPhosK 2.0, and KinasePhos 2.0 (upper panel). Names of possible kinases were listed vertically on the right of the panel. DENV2 proteins were listed horizontally on the top. Color codes (lower panel) indicated numbers of hit from the predicting algorithms; from 0 (very light red) to 3 (dark red). b Sequence conservation of the predicted PKC phosphorylation sites across flaviviruses. Partial amino acid sequences of flaviviruses were listed as indicated by name; Omsk hemorrhagic fever virus (OHFV), Gadgets Gully virus (GGYV), Kama virus (KAMV), and tick-borne encephalitis virus (TBV), St. Louis encephalitis virus (SLEV), dengue 1–4 (DENV1–4), West Nile virus (WNV), yellow fever virus (YFV), Zika virus (ZIKA), Japanese Encephalitis virus (JEV). Predicted PKC phosphorylation sites were showed in RED. In addition, Thr302, and Ser796 that were conserved were underlined. c Three-dimension models of DENV2 NS5 protein. Colored in light gray is the methyltransferase domain, dark gray is the RNA-dependent RNA polymerase (RdRp) domain. Highlighted in yellow, red, and blue are possible PKC phosphorylation sites; Thr244, Ser796, andThr302, respectively. Highlighted in fluorescent green is the GDD RNA polymerase active site. d PKC in vitro kinase assays. Purified DENV2 proteins were used as substrate. Purified GST was used as a non-relevant control for the reaction. Bars represented means of 3 independent experiments, error bars; S.D. e In vitro kinase assays, several human kinases were tested including PKC, PKA, CKI, and CDK4/cyclin D1, as indicated. The signal was normalized by the signal from reactions with GST. Bars represented data from a representative experiment