Fig. 3: OsPUX8B.2 associates with OsATG8a and OsCDC48-6 via its UBX domain.

a, b Y2H assays determining the interactions between OsPUX8B.2 and rice ATG8 (a) or CDC48 (b). c–e Y2H (c), Co-IP (d), and split-luciferase complementation (SLC) (e) assays determining the binding interface between OsATG8a and OsPUX8B.2. f–h Y2H (e), Co-IP (f), and SLC (g) assays determining the binding interface between OsPUX8B.2 and OsCDC48-6. i, j Co-IP assays showing that OsATG8a (i) and OsCDC48-6 (j) interact with OsPUX8B.2 but not with OsPUX8B.1 or OsPUX8B.3. k, l Competitive Co-IP (k) and SLC (l) assays showing the effects of MoNLE1 presence on the association between OsPUX8B.2 and OsATG8a or OsCDC48-6. m Time-course analysis of OsATG8a and OsCDC48-6 expression patterns upon blast (strain RB22) infection. Data are means ± s.d. from three biological replicates. Data in e, h and l were shown as boxplots of the number of indicated biological replicates displaying the maximum and minimum, first and third quantiles, and the median. Exact p values were determined by two-tailed Student’s t tests in l and m. Protein abundance in SLC assays was determined by immunoblot analysis. Note that all Co-IP experiments were performed in rice protoplasts that were pretreated with MG132 for 10 h before collection. Data in d, e and g–l are representatives of three independent experiments with similar results.