A Capsid-Encoded PPxY-Motif Facilitates Adenovirus Entry

Wodrich, Harald; Henaff, Daniel; Jammart, Baptist; Segura-Morales, Carolina; Seelmeir, Sigrid; Coux, Olivier; Ruzsics, Zsolt; Wiethoff, Christopher M.; Kremer, Eric J.

To understand the accumulation defect of Ad5-VI-M1 at the MTOC, we expressed wt (VI-wt), mutant (VI-M1) and protein VI deleted in the amphipathic helix (VI-DW) fused to mRFP in cells and analyzed protein VI localization in relation to microtubules ( Figure 4 ). We found VI-wt in a punctuated distribution throughout the cell suggesting association with a vesicular compartment or tubulo-vesicular structures associated with microtubules ( Figure 4 , top row). In contrast, the PPxY motif mutant VI-M1 localized to a more central compartment and was rarely associated with the microtubules (Figure 4 , middle row). Deletion of the amphipathic helix, in contrast, abrogated purified Ad capsids. Electronic microscopy images of purified Ad5-VI-wt capsids (left panel) and purified Ad5-VI-M1 (right panel) are shown. The inset in each figure shows a magnification of individual capsids. The scale bar is 100nm. C) Plaque assay comparison of Ad5-VI-wt vs. Ad5-VI-M1. Quantification of plaques at day 12 for different physical particle per cell ratios for Ad5-VI-wt and Ad5-VI-M1. Shown is the average plaque number per 6 well (+/2 standard deviation) of three individual experiments. D) Focus forming assay. E1 complementing 911 cells were infected with Ad5-VIwt or Ad5-VI-M1 and replication centers were stained with E2A antibodies 24 h post-infection and at different particle per cell ratios. .5 random fields with .200 cells were counted, standard deviation represents field-to-field variations. E) Plaque forming assay comparison of Ad5GFP-VI-wt vs. Ad5GFP-VI-M1. Quantification of GFP positive plaques at day 12 at different physical particle to cell ratios. The values show the average number of GFP positive plaques per 6-well (+/2 standard deviation) of two independent experiments. F) Single round infection assay comparison of Ad5GFP-VIwt vs. Ad5GFP-VI-M1. U2OS cells were infected with increasing amounts of physical particle to cell ratios as indicated. GFP expression was quantified using FACS. Values correspond to the average percentage of GFP positive cells of two experiments done in triplicates (+/2 standard deviation). Note that the ,100% infection of wt infected cells at 100 physical particles per cell is saturated and not comparable to M1. G) Analysis of membrane penetration. A549 cells were infected with Ad5-VI-wt, Ad5-VI-M1, Ad2ts1 grown at 32uC or at 38uC at different physical particle per cell ratios as indicated and in the presence of alpha-Sarcin and translational efficiency was determined by measuring the incorporation of radiolabeled amino acids over time. Values are given in percentage normalized to 100% translation measured in the presence of the toxin but in absence of virus. Conditions are indicated below each bar and are the mean of at least three independent experiments done in triplicates. doi:10.1371/journal.ppat.1000808.g002 Owing to its association with membranes, microtubules and the viral capsid, we next asked whether protein VI displayed intracellular dynamics that could explain virus trafficking. We therefore performed live-cell imaging (LCI) using cells expressing mRFP-VI-wt or mRFP-VI-M1. We found that VI-wt was fast moving with short-and longrange movements whereas VI-M1 was essentially motionless ( Figure 5A and Video S1). The length of the trajectories and the movement of .300 particles were plotted ( Figure 5B ). We found that protein VI-M1 motility was greatly reduced compared to protein VI-wt. We next asked whether VI-wt moti