(A) Coomassie stained gels showing the input amounts of the Atg12~Atg5-Atg16-mCherry complex (upper gel) and of the GST-prApe1(1–41) + Atg19 or GST proteins on the beads (lower gel) used for the experiment shown in (B). (B) GST-prApe1(1–41) + Atg19 or GST coated glutathione beads imaged in the presence of Atg12~Atg5-Atg16-mCherry complex (wild type, K57E, N84E or K57E,N84E). (C) Quantification of three independent experiments of the relative mCherry signal intensity measured at the beads. One experiment used for the quantification is shown in (B). The signal measured for the wild type Atg5~Atg12-Atg16-mCherry complex was set to 100%. Due to optical reasons very low signals at the beads resulted in values lower than the background and thus negative values. Error bars represent standard deviations. (D) Coomassie-stained gel showing the result of a liposome co-sedimentation assay using wild type Atg12~Atg5-Atg16-mCherry and the indicated point mutants thereof. Liposome binding allows the protein to be pelleted (P). The unbound protein remains in the supernatant (S). (E) Quantification of in vitro Atg8 conjugation assays using the indicated mutants of Atg12~Atg5. The amount of conjugated Atg8 and un-conjugated Atg8 was measured as the band intensity signal on a Coomassie stained gel and set as 100%. Amounts of conjugated Atg8 were determined relative to this. Averages of these values were calculated from three independent experiments and the final values are plotted together with standard deviations. See also Figure 5—figure supplement 1. (F) Co-immunoprecipitation using Atg5-TAP or Atg5 K57E,N84E-TAP as bait in the presence of 6xmyc-Atg19 and either Atg16 wild type or Atg16 E102A. 6xmyc-Atg19 pulled down protein in wild type Atg5 and Atg16 expressing samples was set to 100. All the other conditions were quantified in relation to this (Atg5 wild type and Atg16 E102A, 85% (±59) p-value = 0.63, n.s.; Atg5 K57E,N84E and Atg16 wild type, 13% (±12.9) p-value < 0.0001; Atg5 K57E,N84E and Atg16 E102A, 2.9% (±4.6) p-value < 0.0001). The p-values were calculated using a two-tailed Student t-test. Proteins were detected using anti-TAP and anti-Myc antibodies, anti-Pgk1 was used as loading control. Shown is a representative blot of three experiments. (G) prApe1 processing assay using an atg5Δ strain transformed with the indicated expression constructs. The lower Ape1 band indicates prApe1 processing and thus its delivery into the vacuole. The prApe1 and Ape1 bands were detected with an anti-Ape1 antibody. The expression levels of Atg5 were visualised with an anti-Myc antibody. The Pgk1 signal served as a loading control. The bar graph to the right shows a quantification of six independent experiments. The p-values were calculated using a two-tailed Student t-test. (H) prApe1 processing assay using yeast atg16Δ strain with Atg5 wild type-TAP or Atg5 K57E,N84E-TAP stably integrated in the genome and transformed with the indicated Atg16 constructs. The blot shows the prApe1 processing in the Atg5 mutants in combination with Atg16 wild type in rich conditions (the full set of tested Atg16 can be found in Figure 5—figure supplement 2). A blot showing the full set of Atg16 mutants after 6 hr nitrogen-starvation is shown in Figure 5—figure supplement 3. The Ape1 bands were detected using an anti-Ape1 antibody. The bar graph shows quantification of the prApe1 processing of four independent experiments. The p-value was calculated using a two-tailed Student t-test. (I) A representative blot of a GFP-Atg8 cleavage assay is shown. (J) Pho8∆60-activity assay under rich (black bars) and 5 hr N-starvation (white bars) growing conditions using a pho13∆, pho8∆60, atg5Δ strain transformed with the indicated Atg5 expression constructs or an empty vector. At least three independent experiments were conducted and the mean values for each conditions were plotted. The error bars represent the standard deviation.