SARS-CoV nsp14-inhibitor interactions were further investigated by monitoring the thermal stability of the protein using differential scanning fluorimetry (DSF) (Supporting Information, Fig. S3) [34]. The change in thermal stability of SARS-CoV nsp14 was monitored in response to binding of the natural cofactor substrate SAM, sinefungin and the 9 most active and specific compounds 6 and 9–16 with IC50 < 100 μM. As expected, SARS-CoV nsp14 displayed an increased melting temperature (T m) value with SAM (+6.5 °C) and sinefungin (+4.4 °C) whose structure only differs by a C–NH2 in place of S+-CH3 group in SAM (Fig. 2 ). The binding experiments with the bisubstrate inhibitors showed that all dinucleosides stabilized the SARS-CoV nsp14 protein (T m > 40 °C) with a T m shift from +4.6 °C to +10.8 °C (Supporting Information, Table S1). Eight of the 9 examined compounds increased the stability of SARS-CoV nsp14 more efficiently than the well-known inhibitor sinefungin. More interestingly, T m values for SARS-CoV nsp14 were larger in the presence of five compounds 10 and 12–15 than with the natural enzyme substrate SAM, suggesting strong protein-inhibitor interactions. Remarkably, the highest T m was observed with the most efficient inhibitor 13 (IC50 0.6 ± 0.1 μM) that stabilizes SARS-CoV nsp14 against thermal denaturation with a ΔT m +10.8 °C and exhibits notable binding affinity (apparent KD 1.3 ± 0.87 μM), as deduced from TSA performed with increasing concentration of compound 13. This demonstrates a favorable interaction and highlights the inhibitor potential of compound 13. The T m comparison of compounds 16, 15 and 13 showed a respective increase in SARS-CoV nsp14 stability, settling the importance of the sulfone group, the Cl and NO2 substituents in protein binding with 13. Thus the N-(4-Cl-3-NO2-phenylsulfonamide) moiety is notably preferred for optimal SARS-CoV nsp14 interactions. Another highlight is the thermal shift (+5.4 °C) for SARS-CoV nsp14 observed in the presence of compound 10 compared to 11, both only differ by the position of OCH3 and NO2 substituents in the phenyl ring. A higher stability and inhibition of SARS-CoV nsp14 was observed when the OCH3 group is in “para” position in dinucleoside 10.
Fig. 2 Thermal shifts (ΔTm) of SARS-CoV nsp14 in the absence or presence of SAM, sinefungin and 9 dinucleoside inhibitors 6 and 9–16. Thermal stability of SARS-CoV nsp14 upon ligand binding was monitored by differential scanning fluorimetry. Assays were carried out in reaction mixture [20 mM HEPES (pH 7.5), 150 mM NaCl, 1x SYPRO orange dye] in the presence of 5 μM SARS-CoV nsp14 protein and 1 mM compound previously dissolved in 100% DMSO. The bars and error bars correspond to the mean values from three independent measurements and their s.d.’s, respectively.