Activation of ATF4 mediates unwanted Mcl-1 accumulation by proteasome inhibition by Jinsong Hu, Nana Dang, Eline Menu, Elke De Bryune, Dehui Xu, Ben Van Camp, Els Van Valckenborgh, and Karin Vanderkerken Blood Volume 119(3):826-837 January 19, 2012 ©2012 by American Society of Hematology

Western blot and qRT-PCR analysis of Mcl-1, Bcl-2, and Bcl-xl expression in MM cells during bortezomib treatment. Western blot and qRT-PCR analysis of Mcl-1, Bcl-2, and Bcl-xl expression in MM cells during bortezomib treatment. (A) Top panel: Western blot analysis of bortezomib-treated (for 0, 6, 12, and 24 hours) human LP1 and RPMI-8226 cells and murine 5T33vt cells. The blots were probed with Mcl-1, Bcl-2, Bcl-xl, and β-actin Abs. The concentrations of bortezomib used in human LP1 and RPMI-8226 and murine 5T33vt cells were 50, 10, and 5nM, respectively. The data are representative of 3 independent experiments. Bottom panel: amounts of Mcl-1, Bcl-2, and Bcl-xl were quantified by densitometric analysis of Western blot using ImageJ Version 1.45 software. (B) The pan caspase inhibitor Z-VAD-FMK abrogates the cleavage of Mcl-1. Left panel: Western blot analysis of the Mcl-1 and caspase-3 expression in OPM2 cells. OPM2 cells were untreated (control) or were treated with 5nM bortezomib for 16 hours with or without 2 hours of pretreatment with 50μM Z-VAD-FMK as indicated. Right panel: densitometric analysis of Western blots. (C-E) qRT-PCR analysis of MCL-1, BCL-2, and BCL-XL expression at the mRNA level. The MM cell lines RPMI-8226 (10nM), MMS1 (10nM), U266 (5nM), LP1 (100nM), OPM2 (5nM), and 5T33vt (2.5nM) were treated with bortezomib for 24 hours. RNA was isolated and subjected to qRT-PCR using primers for MCL-1, BCL-2, BCL-XL, and β-actin. Beta-actin was used as an internal control. n = 3 in all experiments. *P < .05 and **P < .01 versus vehicle-treated samples. The columns show the mean results representative of 3 similar experiments; the bars indicate SD. Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology

Bortezomib activates the UPR in MM cells. Bortezomib activates the UPR in MM cells. (A-B) qRT-PCR analysis of the ER stress markers GRP-78 and CHOP expression at the mRNA level. The MM cell lines RPMI-8226 (10nM), MMS1 (10nM), LP1 (100nM), OPM2 (5nM), and 5T33vt (2.5nM) were treated with bortezomib for 24 hours. n = 3 for all experiments. *P < .05 and **P < .01 versus vehicle-treated samples. (C) Left panel: Western blot analysis of GRP-78 and CHOP expression. Whole-cell extracts were prepared from bortezomib-treated (for 0, 6, 12, and 24 hours) human LP1 and MMS1 cells and murine 5T33vt cells; the concentrations of bortezomib used were 100, 10, and 5nM, respectively. The data are representative of 3 independent experiments. Right panel: amounts of GRP-78 and CHOP were quantified by densitometric analysis of Western blot using ImageJ Version 1.45 software. (D) Western blot analysis of the UPR arm of ATF4 activation in bortezomib-treated MM cells. The doses of bortezomib were used at 10nM in both RPMI-8226 and MMS1 cells and at 2.5nM in 5T33vt cells. Left panels show the representative blots of 3 independent experiments; right panels show densitometric analysis. (E-F) qRT-PCR analysis of ATF4 and its target gene ATF3 expression in bortezomib-treated MM cells. The MM cell lines RPMI-8226 (10nM), MMS1 (10nM), LP1 (100nM), OPM2 (5nM), and 5T33vt (2.5nM) were treated with bortezomib for 24 hours. Data represent the means ± SD for 3 separate experiments. *P < .05 and ** P < .01 versus vehicle-treated samples. Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology

ChIP analysis showing the interaction of ATF4 with the Mcl-1 promoter. ChIP analysis showing the interaction of ATF4 with the Mcl-1 promoter. (A) Human Mcl-1 upstream region showing 3 consensus sites for ATF4. There is only 1 consensus site for ATF4 at the promoter of murine Mcl-1. The core consensus motifs are boxed. (B) ChIP analysis of ATF4 binding in bortezomib-treated LP1 cells. LP1 cells were treated with 100nM bortezomib for 16 hours. Immunoprecipitation was performed using control IgG or anti-ATF4 Ab. Samples were analyzed by qRT-PCR using primers for region −1521 to −1297, −1033 to −733, and −423 to −80 nucleotides. *P < .05 versus control IgG-immunoprecipitated samples (n = 3). Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology

Knockdown of ATF4 increases sensitivity to bortezomib. Knockdown of ATF4 increases sensitivity to bortezomib. (A) Western blot analysis of ATF4 and Mcl-1 expression levels in bortezomib-treated U266 cells (2.5nM for 16 hours) and MMS1 cells (5nM for 16 hours). Top panels show the representative blots of 3 independent experiments; bottom panels show densitometric analysis. (B-C) qRT-PCR measurement of ATF4 downstream gene-expression levels after knockdown ATF4 in the presence of bortezomib in U266 cells (2.5nM for 16 hours; B) and MMS1 cells (5nM for 16 hours). n = 3 for all experiments. *P < .05 and **P < .01 versus si-mock samples. Data represent the means ± SD. (D) Knockdown of ATF4 induces apoptosis in U266 cells. ATF4 knockdown or control U266 cells were treated with 2.5nM bortezomib for 16 hours. Bortezomib was added in the medium 4 hours after electroporation-mediated RNAi. Apoptosis of each sample was examined by flow cytometry after annexin V–FITC/7-amino-actinomycin D staining. FACS data shown are representative of 3 independent experiments. Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology

The ER inducer tunicamycin increases Mcl-1 expression and triggers UPR activation in MM cells. The ER inducer tunicamycin increases Mcl-1 expression and triggers UPR activation in MM cells. (A) Tunicamycin treatment induces Mcl-1 accumulation and cleavage in OPM2 cells. OPM2 cells were treated with different doses of tunicamycin for 24 hours. Cell extracts were probed with Mcl-1, ATF4, GRP-78, CHOP, and β-actin pAbs. Left panel shows the representative blots of 3 independent experiments; right panel shows densitometric analysis. (B-D) qRT-PCR analysis of the expression of Mcl-1 and the UPR-related components in response to tunicamycin treatment in RPMI-8226 (B), OPM2 (C), and U266 (D) cells. RPMI-8226, OPM2, and U266 cells were treated with 2.5 μg/mL of tunicamycin for 24 hours. RNA isolation, cDNA synthesis, and qRT-PCR were performed as described in “Methods.” *P < .05 and **P < .01 versus vehicle-treated samples. Data represent the means ± SD of 3 separate experiments, each performed in triplicate. Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology

The role of XBP1 in UPR-mediated Mcl-1 expression. The role of XBP1 in UPR-mediated Mcl-1 expression. (A-F) RT-PCR analysis of XBP1 splicing in different MM cell lines. The MM cell lines OPM2, MMS1, U266, 5T33vt, RPMI-8226, and LP1 were treated with bortezomib and tunicamycin for 24 hours at varying doses. RNA isolation and RT-PCR were performed as described in “Methods.” In the human MM cell lines OPM2, MMS1, U266, RPMI-8226, and LP1, the 210- and 184-bp DNA fragments correspond to unspliced and spliced human XBP1 mRNAs, respectively. In 5T33vt cells, the 343- and 327-bp DNA fragments correspond to unspliced and spliced mouse Xbp1 mRNAs, respectively. (G) Western blot analysis of XBP1 splicing in bortezomib-treated RPMI-8226 and MMS1 cells. Total protein extracts from RPMI-8226 and MMS1 cells treated with 10nM of bortezomib for 0, 6, 12, and 24 hours were electrophoresed and subjected to Western blot analysis using XBP1 Abs. Left panels show the representative blots of 3 independent experiments; right panels show densitometric analysis. (H) qRT-PCR measurement of MCL-1 and UPR-related downstream genes in U266 cells with XBP1 knockdown. n = 3 for all experiments. *P < .05 and **P < .01 versus siRNA mock samples. Data represent the means ± SD. Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology

Schematic diagram depicting the joint roles of the UPR in regulating Mcl-1 expression by proteasome inhibition. Schematic diagram depicting the joint roles of the UPR in regulating Mcl-1 expression by proteasome inhibition. In response to bortezomib, e-IF2α is phosphorylated and the ATF4 branch is activated to trigger protective effects for cell survival by up-regulating the expression of the anti-apoptotic protein Mcl-1. Bortezomib shows completely contrary effects, either promoting or inhibiting XBP1 splicing in different MM cell lines, suggesting that XBP1 splicing is not directly involved in bortezomib-induced Mcl-1 expression. Bortezomib can increase the cleavage of ATF6 to form active ATF6α, which also plays a role in regulating Mcl-1. Jinsong Hu et al. Blood 2012;119:826-837 ©2012 by American Society of Hematology