The SARS-coronavirus nsp7+nsp8 complex is a unique multimeric RNA polymerase capable of both de novo initiation and primer extension

te Velthuis, Aartjan J.W.; van den Worm, Sjoerd H. E.; Snijder, Eric J.

Extrapolating to the situation in the viral pp1a and pp1ab precursor polyproteins, in which the nsp8 N-terminus is initially fused to nsp7 ( Figure 1A) , our observations suggested that nsp8 may thus be inactive in the polyprotein context. This would constitute a form of regulation of viral enzyme activity that is not without Figure 6A , presented as the amount of NTP incorporated per mM nsp8 monomer. Error bars represent standard deviations (n = 3). (C) The influence of the pH on nsp(7+8) activity was tested for a pH range of 6-11. A clear optimum was observed around 9.5. (D) Quantification of the results in Figure 6C , presented as the amount of NTP incorporated per mM nsp8 monomer. Error bars indicate standard deviations (n = 3). (E) Schematic presentation of the pulse-chase experiment that was used to test the nsp(7+8) nucleotide incorporation specificity on a primed poly(U) template (see Table 1 ). The reactions were initiated with a limiting concentration of [a-32 P]ATP to allow the formation of a stable polymerase-template complex. Unlabelled nucleotides were used at a final concentration of 50 mM. (F) SARS-CoV nsp(7+8) allowed only limited transversional and transitional mutations. Use of manganese ions as cofactor for polymerase activity resulted in a minor, though noticeable loss of fidelity. Lane 1 represents the input signal to which no unlabelled nucleotides were added. Nucleoside triphosphates are abbreviated to single letters (i.e. A for ATP, G for GTP, U for UTP, C for CTP and R for RTP). precedent, since also the poliovirus 3Dpol is inactive as long as it is fused to the 3C protease in the 3CD precursor (29) . To verify this hypothesis, we expressed nsp7-8-His and tested this protein for RdRp activity. Interestingly, this fusion protein, a potential intermediate of CoV replicase polyprotein processing and a multimer in solution ( Figure 7E ), showed primer extension activities that were comparable to or higher than the activity of nsp(7+8-His) ( Figure 7F ). The de novo initiation activity of nsp7-8-His was, however, $2-fold lower than the activity of nsp8 and nsp(7+8) ( Figure 7D ). In conclusion, this result clearly underlines that the two N-terminal fusion partners other than nsp7 are specifically detrimental to SARS-CoV nsp8 primer-dependent RdRp activity in vitro. It also demonstrates that nsp8 alone may be sufficient to act as a primase.