Starvation triggers secretion of cytoplasmic antioxidant enzymes. (A) Wild-type cells were grown to mid-logarithmic phase, washed twice, and cultured in 2% potassium acetate for 2.5 h. Cell wall proteins (secreted) were extracted and concentrated 139× with exchange to TBS. Separately, a smaller aliquot of the same cells was lysed in nondenaturing conditions in TBS (lysate). Lysate and secreted proteins were separated in nondenaturing 12% polyacrylamide gels and either transferred to a nitrocellulose membrane for Western blot analysis (anti-SOD1) or subjected to the zymography-based in-gel SOD activity assay (SOD1 activity). Samples of lysate fractions were loaded in decreasing amounts at a dilution of 1/10 to compare with the concentrated secreted fraction. The final relative loading is 80× more in the secreted fraction. (B) Mass spectrometry analysis of cell wall–extracted proteins (secreted proteins) from wild-type cells growing in nutrient-rich conditions (“growth”) versus cells cultured in potassium acetate for 2.5 h (“starvation”). Cell wall proteins from cells lacking Vps23 were also analyzed under starvation conditions. All three conditions were analyzed in triplicate, and statistical analyses were performed to classify the secreted proteins as growth versus starvation specific or unchanged (using a cutoff of a Log₂FoldChange of at least ±1 and a P value of <0.06). Within the starvation-specific group, proteins were further classified as dependent on Vps23 or not for their presence in the cell wall, whether they perform an enzymatic activity, and whether they are annotated to affect response to oxidative stress according to the Saccharomyces Genome Database. Complete analyses of proteins identified and the classifications can be found in Table S1. (C) Chart plotting relative protein abundance of the starvation-specific/Vps23-dependent enzymes classified as affecting response to oxidative stress. SOD1, Ahp1, Trx1, and Trx2 directly regulate cellular redox balance, while the remaining are key enzymes in various cellular processes known to affect response to oxidative stress, such as glycolysis, gluconeogenesis, amino acid or nucleotide biosynthesis, and glycerol metabolism (Acb1 is included only as a reference). The average protein areas from the mass spectrometry analysis (Table S1) of the three triplicates for each protein was calculated, with the corresponding SEM (error bars), for each condition. A zoom of the low abundance enzymes is shown in the inset (note the scale difference). Complete descriptions of the enzymes can be found in Table 1. (D) Wild-type and grh1Δ cells were grown to mid-logarithmic phase, washed twice, and incubated in 2% potassium acetate for 2.5 h. Cell wall proteins were extracted from equal numbers of cells followed by precipitation with TCA (secreted). Lysates and secreted proteins were analyzed by Western blot, and the ratio of the secreted–lysate for the indicated cargo protein was determined and compared with that of wild type in each experiment. Statistical analyses were performed for the indicated unconventional cargo proteins and are represented as fold changes in grh1Δ cells compared with wild type (paired Student’s t test). Ahp1 = 0.07 ± 0.03, P = 0.0604; Trx1 = 0.25 ± 0.04, P = 0.0478; Trx2 = 0.23 ± 0.03, P = 0.0065. Error bars indicate SEM, n = 3. Cof1 is a nonsecreted, cytoplasmic protein used to monitor for cell lysis, while Bgl2 is a conventionally secreted cell wall protein. The loading of the secreted fraction is equivalent to 145× that of the lysate. Based on this, the amount of secreted compared with the total pool was calculated to be 0.3% for Ahp1, 0.6% for Trx1, and 0.8% for Trx2.