Fig. 4: Co-translational protein delivery regulates protein composition.

A The noise of protein distribution with variable copy numbers (1x, 2x, 3x) of GFP-tagged Tim50 integration plasmid in fzo1Δ strain (n > 46). The results represent the mean ± standard deviation of three independent experiments (P = 0.006 (1x – 3x)). B–E, G Relationship between the noise of protein distribution and mitochondrial size in A–D, and F. The noise value was calculated by dividing the population of individual mitochondrial fragments into eight groups. Error bars represent the SEM of three independent experiments. R squared values and p-values for the F statistic hypothesis test are shown on the side. C Schematic representation of plasma membrane tethering models. In plasma membrane tethering, MCP is bound to CAAX (a ras protein family member). D The noise of protein distribution in plasma membrane tethering (CAAX) of GFP-tagged Tim50 integration plasmid in fzo1Δ strain (n > 47). The results represent the mean ± standard deviation of three independent experiments (P = 0.002). F The noise of protein distribution of GFP-tagged Tim50 upon CCCP treatment in WT (No drug), 10 min, and 30 min (n > 64). The results represent the mean ± standard deviation of three independent experiments. H RNA expression from previously published studies normalized to TIM50 mRNA expression (# not available). I The noise of protein distribution of GFP-tagged proteins. Green bar represents co-translationally localized mitochondrial proteins. Orange bar represents post-translationally localized mitochondrial proteins (n > 40). Each co-translationally localized protein was compared to both post-translationally localized controls (Pdx1 and Rsm7) (Tim50 (P = 0.02 and 0.009), Mia40 (P = 0.01 and 0.03), Dld1 (P = 0.01 and 0.0007), Cox15 (P = 0.009 and 0.02). Statistical significance was assessed using the two-tailed Student’s t test for D and two-tailed Anova followed by t-test for pairwise comparison for A–F and I. Source data are provided as a Source Data file.