Mono-ADP-ribosylation by ARTD10 restricts Chikungunya virus replication by interfering with the proteolytic activity of nsP2

Sarah Krieg; Fabian Pott; Laura Eckei; Maud Verheirstraeten; Mareike Bütepage; Barbara Lippok; Christine Goffinet; Bernhard Lüscher; Patricia Verheugd

interfere with viral replication, we determined the abundance of auto-proteolytically 188 . CC-BY-NC-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.01.07.896977 doi: bioRxiv preprint 9 processed nsP3. Therefore, we made use of EGFP-encoding variants of the replicon ( 2 EGFP 189 and 3 EGFP, in which EGFP is integrated after amino acids 466 or 383 in nsP2 or nsP3, 190 respectively Fig. 1b and [48] ), enabling us to visualize processed nsP2 or nsP3 proteins using a 191 GFP-specific antibody (Fig. 2) . HEK293 cells stably expressing ARTD10 or ARTD10-GW were 192 employed. These cells were then transfected with or without ARTD12 expressing constructs 193 and induced for ARTD10 expression prior to transfection with 3 EGFP replicon RNA. The 194 analysis of whole cell lysates showed a reduction in processed nsP3 in presence of either the 195 enzymatically active ARTD12 or ARTD10. NsP3 was further reduced when both enzymes were 196 expressed (Fig. 2a) . These findings led us hypothesize that MARylation hampers polyprotein 197 processing. Proper polyprotein processing is a prerequisite for RNA replication [43], consistent 198 with impaired replication of a mutant replicon with an inactive protease (nsP2-C478A/S482A, 199 referred to as CASA) (Fig. 2b,c) [44]. Similarly, a functionally active macrodomain is needed for 200 replication as substitution of key amino acids in the macrodomain (D10A, V33E, for details 201 concerning the replicon construct see Fig. 1b ) interfered with replication ( Fig. 2b ,c) [6, 9] . To 202 analyze how the lack of a functional macrodomain compromised replication, we determined 203 the abundance of proteolytically processed nsP2 (Fig. 2d , for the specificity of the antibody 204 see Supplementary Fig. 3a ,b). As expected, nsP2 was detectable after transfection of the 205 wildtype (wt) but not the CASA mutant replicon (Fig. 2d) . Similarly, nsP2 was not detectable, 206 when expressed from hydrolase deficient replicons (Fig. 2d) , implying a defect in nsP2-207 mediated polyprotein processing in the absence of MAR hydrolase activity. This observation 208 was corroborated with the 3 EGFP and 2 EGFP replicons and mutants thereof ( Fig. 2e-g) . 209