*3.7. Integration with a Microcephaly-Associated Network*

We generated a new network composed of microcephaly-associated genes/proteins (Table S13) and our merged ZIKV PIN using BisoGenet to impute known interactions between nodes in this network (see Experimental Procedures) (Figure 5A). Four highly cohesive (i.e., highly connected internally, but only sparsely with the rest of the network) clusters emerged (Figure 5B). Of those, only three contained components of the ZIKV PIN: anaphase promoting complex (APC)-dependent proteasomal ubiquitin-dependent protein catabolic process (GO31145), centrosome duplication (GO7099), and COPI coating of Golgi vesicle (GO48205). Finally, we identified four nodes common to both PINs (CEP192, ASXL1, VARS, and EFTUD2). A similarly limited overlap between the microcephaly-associated and merged ZIKV PIN was also obtained with the three other previously determined ZIKV PINs (Figure 5C).

**Figure 5.** Integration of Microcephaly-associated PIN with the merged ZIKV PIN. (**A**) Network of the interactions identified by this study, and proteins related to the Microcephalic phenotype (see Experimental Procedures). (**B**) Clustering of bait sets according to overlapping protein complexes among the integrated network using ClusterONE (Version 1.0) [29]. Gene ontology of these networks were obtained by the Cytoscape plugin, BINGO [30]. Proteins identified in this study are represented as orange nodes. (**C**) Venn diagrams represent the overlap between the Microcephaly-associated PIN and the individual ZIKV PINs.

## *3.8. PIAS1 Modulates NS5 Protein Stability*

The Y2H screening identified PIAS1 (protein inhibitor of activated STAT1) as an interacting partner of NS5. We confirmed this interaction using a GST-pulldown assay in HEK293FT cells co-transfected with GST-tagged PIAS1 and GFP-tagged NS5 (Figure 6A).

PIAS1 is an E3 SUMO-protein ligase implicated in the maintenance of protein stability [41,42]. Curiously, SUMOylation of the DENV NS5 has been linked to its stabilization, thus stimulating viral replication in human cells [43]. This observation prompted us to evaluate the impact of PIAS1 overexpressing on ZIKV NS5 protein stability. HEK293FT cells were transfected with GST-tagged PIAS1 and GFP- or CBP-tagged NS5 constructs and treated with cyclohexamide (CHX) to inhibit de novo protein synthesis. NS5 half-life was evaluated in a time course (Figure 6B). PIAS1 overexpressed is correlated with a decrease in NS5 protein stability, as seen 12 h after CHX treatment (compare lines 4 and 8). PIAS1 modulates steady state levels of NS5 even at lower expression levels of PIAS1 (Figure 6C) and levels of NS5 decrease in a PIAS1 dose-dependent manner (Figure 6D). Taken together, our data sugges<sup>t</sup> that PIAS1 is involved in modulating the stability of ZIKV NS5.

**Figure 6.** PIAS1 interacts with and modulates ZIKV NS5 protein stability. (**A**) GST pulldown was conducted using whole cell lysates of HEK293FT cells transfected with GST-tagged PIAS1 and GFP-tagged NS5 (empty vectors were used as negative controls). Input represents 10% of the lysate used in the GST pulldown assay. (**B**) HEK293FT cells transfected with GST-tagged constructs (empty vector or PIAS1) and pNTAP NS5 and treated with 10 μg/mL cyclohexamide (CHX) for the indicated time points. (**C**) HEK203FT cells transfected with GST and GFP-tagged constructs. Cells were treated, 24 h after transfections, with 10 μg/mL cyclohexamide (CHX) for 12 h. (**D**) HEK293FT cells were transfected with GFP-tagged constructs (empty vector or NS5) and different amounts of GST-tagged PIAS1 cDNA (0.5 μg, 1 μg, 2 μg, or 4 μg). pQCXIH was used to normalize the amount of transfected DNA. At 20 h post-transfection, cells were treated with 10 μg/mL cyclohexamide (CHX) for 12 h. + and – signs indicate presence or absence of the reagen<sup>t</sup> indicated on the left, respectively.
