Dissection of the Influenza A Virus Endocytic Routes Reveals Macropinocytosis as an Alternative Entry Pathway

de Vries, Erik; Tscherne, Donna M.; Wienholts, Marleen J.; Cobos-Jiménez, Viviana; Scholte, Florine; García-Sastre, Adolfo; Rottier, Peter J. M.; de Haan, Cornelis A. M.

Inhibitors of growth factor receptor tyrosine kinases and actomyosin network dynamics reduce DYNA-IND entry of IAV The DYNA-IND entry pathway was further characterized by inhibitor profiling using an 80-compound kinase inhibitor library. Serum-induced DYNA-IND entry was examined in 10% FCS using the Gluc-entry assay. 80 mM dynasore was added in order to block CME and any other potential DYNA-DEP entry pathways. This allowed the independent inhibitor profiling of the novel pathway by avoiding the potentially masking effect of the presence of redundant entry pathways. Cells were preincubated with the kinase inhibitors (10 mM) for 1 h at 37uC and then inoculated with virus (MOI 0.5) in the presence of 10% FCS and 80 mM dynasore for 2 h at 37uC (DYNA-IND entry). In parallel, inoculations were also done in PBS to compare the effects of the inhibitors on DYNA-DEP entry. After 2 hr the medium and inhibitor were replaced by full growth medium containing 10% FCS and 10 nM BafA1 to allow the subsequent expression of Gluc activity under identical conditions for the DYNA-IND and -dependent entry assay. Six kinase inhibitors appeared to act non-discriminatively, inhibiting both DYNA-DEP and DYNA-IND entry (Fig. 5A ): the protein kinase C (PKC) inhibitors Ro 31-8220, rottlerin (both displaying moderate cytotoxicity, result not shown) and hypericin, which have all three been previously identified as IAV inhibitors [28, 29] ; the highly cytotoxic pan-specific serine/threonine protease inhibitor staurosporine; the irreversible PI-3 kinase inhibitor wortmannin and the receptor tyrosine kinase inhibitor TYR9. In order to investigate whether some of these inhibitors affect IAV replication during the post-entry phase, we performed the same experiments but now adding the kinase inhibitors after viral entry. Four of the inhibitors thus appeared to induce significant inhibition of post-entry processes (Fig. 5A ). Although unlikely, we cannot formally exclude that post-entry processes specific for only one of the two entry pathways are affected. Interestingly, whereas no specific DYNA-DEP entry inhibitors were identified, 15 inhibitors (none displaying cytotoxic effects, data not shown) caused significant (p,0.05) inhibition (.5-fold) of DYNA-IND entry (Fig. 5B ). This included inhibitors of the calmodulin dependent kinases myosin light chain kinase (MLCK) and CaMKII and seven inhibitors of different growth factor receptor tyrosine kinases. In contrast to the three non-specific PKC inhibitors mentioned above, the PKC inhibitors BIM-1 and HBDDE appeared to have a specific inhibitory effect on DYNA-IND entry. The specific effect of these drugs on DYNA-IND entry is not only shown by the lack of inhibition of DYNA-DEP entry in PBS, but also by the observation that none of the fifteen compounds induced .2-fold inhibition when added post-entry (at t = 2 hr post infection). The kinase library screen was repeated on A549 human epithelial lung carcinoma cells in order to confirm the results in a potentially more natural host cell line. The inhibition profiles obtained were very similar to those found for HeLa cells with the exception of the strong effect of AG879 (99% inhibition) and moderate effects of AG825 (39% inhibition) and Tyr51 (68% inhibition) on DYNA-DEP entry. (Fig. 5C) .