Fig. 2.

Effects of different stressors on UPR activation at the proteome level. (A) Time course and dose-dependence of different stressors. HeLa cells were treated for 5 and 18 h as indicated with low (DTT 0.2 µM, thapsigargin 0.1 µm, tunicamycin 0.625 µg/ml) and high dose (DTT 2 mM, thapsigargin 1 µm, tunicamycin 5 µg/ml) of stressors. Western blot analysis shows the expression of CHOP (DDIT3) and HSPA5. Tubulin and Ponceau S are shown as loading controls. The experiment was carried out in three replicates. The HSPA5:tubulin ratio (N=3, mean±s.d.), quantified using ImageJ, is shown as a bar graph. Controls at 18 h are shown. (B) Principal component analysis (PCA) comparing the proteome remodeling induced by tunicamycin (5 μg/ml), DTT (2 mM) and thapsigargin (1 µM). Vehicle-treated cells were used as controls. Each treatment was performed in biological triplicates. Data were filtered for 100% valid values in ANOVA-significant proteins (543 proteins). (C) Loadings of B, showing the major separators into components one and two (x- and y-axis, respectively). TIMP1, metalloproteinase inhibitor 1; HSPA5, endoplasmic reticulum chaperone BiP; IFRD1, interferon-related developmental regulator 1; GPRC5A, retinoic acid-induced protein 3; SCD, stearoyl-CoA desaturase 5; PON2, serum paraoxonase/arylesterase 2; KIAA0101, PCNA-associated factor (also known as PCLAF); TK1, thymidine kinase. (D) Unsupervised clustering of ANOVA-significant proteins showing three main clusters, with profiles and corresponding relative enrichments, indicated to the right. Enrichment is based on Fisher's exact test of keyword annotations at 0.04 FDR.