Accelerated viral dynamics in bat cell lines, with implications for zoonotic emergence

Cara E. Brook; Mike Boots; Kartik Chandran; Andrew P. Dobson; Christian Drosten; Andrea L. Graham; Bryan T. Grenfell; Marcel A. Müller; Melinda Ng; Lin-Fa Wang; Anieke van Leeuwen

In fitting our theoretical model to in vitro data, we estimated the within-host virus 259 transmission rate ( ) and the rate(s) of cellular acquisition to antiviral status ( or + ) (Table 260 1 In contrast to Vero cells, the induced immunity model offered the best fit to all RoNi/7.1 274 data, consistent with reported patterns in the literature and our own validation by qPCR (Table 1; infections on RoNi/7.1 cell lines but here recovered higher estimates for rVSV-MARV than for 278 rVSV-EBOV. This reversal was balanced by a higher estimated rate of acquisition to antiviral 279 status ( ) for rVSV-EBOV versus rVSV-MARV. In general, we observed that more rapid rates 280 of antiviral acquisition (either induced, , constitutive, , or both) correlated with higher 281 transmission rates ( ). When offset by , values estimated for RoNi/7.1 infections maintained 282 the same amplitude as those estimated for immune-absent Vero cell lines but caused gentler 283 epidemics and reduced cellular mortality (Figure 1 ). RoNi/7.1 parameter estimates localized in 284 the region corresponding to endemic equilibrium for the deterministic, theoretical model ( Figure 285 4), yielding less acute epidemics which nonetheless went extinct in stochastic experiments. 286