Despite the frequent expression of N-terminally truncated ErbB2 (DNErbB2/p95HER2) in breast cancer and its association with Herceptin resistance and poor prognosis, it remains poorly understood how DNErbB2 affects chemotherapy-induced cell death. Previously it was shown that DNErbB2 upregulates acid extrusion from MCF-7 breast cancer cells and that inhibition of the Na þ /H þ exchanger (SLC9A1/NHE1) strongly sensitizes DNErbB2-expressing MCF-7 cells to cisplatin chemotherapy. The aim of this study was to identify the mechanism through which DNErbB2 regulates cisplatin-induced breast cancer cell death, and determine how NHE1 regulates this process. Cisplatin treatment elicited apoptosis, ATM phosphorylation, upregulation of p53, Noxa (PMAIP1), and PUMA (BBC3), and cleavage of caspase-9, -7, fodrin, and PARP-1 in MCF-7 cells. Inducible DNErbB2 expression strongly reduced cisplatin-induced ATM-and p53-phosphorylation, augmented Noxa upregulation and caspase-9 and -7 cleavage, doubled p21 WAF1/Cip1 (CDKN1A) expression, and nearly abolished Bcl-2 expression. LC3-GFP analysis demonstrated that autophagic flux was reduced by cisplatin in a manner augmented by DNErbB2, yet did not contribute to cisplatin-induced death. Using knockdown approaches, it was shown that cisplatininduced caspase-7 cleavage in DNErbB2-MCF-7 cells was Noxaand caspase-9 dependent. This pathway was augmented by NHE1 inhibition, while the Na þ /HCO 3 À cotransporter (SLC4A7/ NBCn1) was internalized following cisplatin exposure.Implications: This work reveals that DNErbB2 strongly affects several major pro-and antiapoptotic pathways and provides mechanistic insight into the role of NHE1 in chemotherapy resistance. These findings have relevance for defining therapy regimens in breast cancers with DNErbB2 and/or NHE1 overexpression.

The Na-HCO cotransporter NBCn1 (SLC4A7) is up-regulated in breast cancer, important for tumor growth, and a single nucleotide polymorphism (SNP), rs4973768, in its 3' untranslated region (3'UTR) correlates with increased breast cancer risk. We previously demonstrated that NBCn1 expression and promoter activity are strongly increased in breast cancer cells expressing a constitutively active oncogenic human epidermal growth factor receptor 2 (HER2) (p95HER2). Here, we address the roles of p95HER2 in regulating NBCn1 expression via post-transcriptional mechanisms. p95HER2 expression in MCF-7 cells reduced the rate of NBCn1 mRNA degradation. The NBCn1 3'UTR down-regulated luciferase reporter expression in control cells, and this was reversed by p95HER2, suggesting that p95HER2 counteracts 3'UTR-mediated suppression of NBCn1 expression. Truncation analyses identified three NBCn1 3'UTR regions of regulatory importance. Mutation of putative miRNA-binding sites (miR-374a/b, miR-200b/c, miR-29a/b/c, miR-488) in these regions did not have significant impact on 3'UTR activity. The NBCn1 3'UTR interacted directly with the RNA-binding protein human antigen R (HuR), and HuR knockdown reduced NBCn1 expression. Conversely, ablation of a distal AU-rich element increased 3'UTR-driven reporter activity, suggesting complex regulatory roles of these sites. The cancer-associated SNP variant decreased reporter expression in T-47D breast cancer cells, yet not in MCF-7, MDA-MB-231 and SK-BR-3 cells, arguing against a general role in regulating NBCn1 expression. Finally, p95HER2 expression increased total and plasma membrane NBCn1 protein levels and decreased the rate of NBCn1 protein degradation. Collectively, this is the first work to demonstrate 3'UTR-mediated NBCn1 regulation, shows that p95HER2 regulates NBCn1 expression at multiple levels, and substantiates the central position of p95HER2-NBCn1 signaling in breast cancer.

<p>Supplementary figure 2: Inhibition of ATM and knockdown of noxa, caspase-9, μ-calpain and cathepsin B. MCF-7 vector (V) and ΔNErbB2 (ΔN) cells were washed free of tetracycline 72 h before the experiment. Cells were exposed to KU-55933 (10 μM) for 19 h, cisplatin (25 μM) or control conditions for 18 h and subsequently processed for immunoblotting (A-B). Top panels show representative immunoblots, lower panels show protein level relative to that in vector cells under control conditions. A. p-ATM B. p-p53. ΔNErbB2 cells were washed and transfected with siRNA (200nM) 72 h before the experiment. Cells were exposed to cisplatin (25 μM), and subsequently processed for immunoblotting (C-F). Top panels show representative immunoblots, lower panels show protein level relative to that in cells transfected with mock control. C. Noxa D. Caspase-9. E. μ-calpain. F. Cathepsin B. Data are representative of 3-4 independent experiments for each condition. Western blot data are shown as means with SEM error bars. * indicates comparison of the value in cisplatin treated to that in the respective cell line control, # indicate comparison of values in vector- and ΔNErbB2 cell lines under same treatment. */#, **/##, ***/### and ****/#### indicate p < 0.05, p < 0.01, p < 0.001 and p < 0.0001, respectively (C-F), one-way ANOVA with Bonferroni post-test) (A-B, two-way ANOVA with Bonferroni post-test).</p>

<p>Supplementary figure 1: Effects of cisplatin treatment on expression of ΔNErbB2 and NHE1, and subcellular localization of cytochrome c. MCF-7 vector (V) and ΔNErbB2 (ΔN) cells were washed free of tetracycline 48 h before the experiment, and exposed to cisplatin (25 μM) or control conditions for 18 h. Cells were subsequently processed for immunoblotting (A), immunofluorescence analysis (B) or cell surface biotinylation (C-D). A. ΔNErbB2, top panel shows representative immunoblots, lower panel shows ΔNErbB2 protein level relative to that in vector cells under control conditions. B. Subcellular localization of cytochrome c. Nuclei are stained with DAPI. C. Representative immunoblots of NHE1 in total lysate fractions (TLF), intracellular fractions (ITF) and biotinylated fractions (BF) in MCF-7 vector (V) and ΔNErbB2 (ΔN) cells. F. Top panel shows representative immunoblots of NHE1 in TLF and BF in native MCF-7 cells, lower panel shows NHE1 protein level relative to that in native MCF-7 cells under control conditions. Data are representative of 3-5 independent experiments for each condition. Western blot data are shown as means with SEM error bars. * indicates comparison of the value in cisplatin treated to that in the respective cell line control, # indicate comparison of values in vector- and ΔNErbB2 cell lines under same treatment. */#, **/##, ***/### and ****/#### indicate p < 0.05, p < 0.01, p < 0.001 and p < 0.0001, respectively (C, one-way ANOVA with Bonferroni post-test) (A, two-way ANOVA with Bonferroni post-test).</p>

<p>Supplementary figure 1: Effects of cisplatin treatment on expression of ΔNErbB2 and NHE1, and subcellular localization of cytochrome c. MCF-7 vector (V) and ΔNErbB2 (ΔN) cells were washed free of tetracycline 48 h before the experiment, and exposed to cisplatin (25 μM) or control conditions for 18 h. Cells were subsequently processed for immunoblotting (A), immunofluorescence analysis (B) or cell surface biotinylation (C-D). A. ΔNErbB2, top panel shows representative immunoblots, lower panel shows ΔNErbB2 protein level relative to that in vector cells under control conditions. B. Subcellular localization of cytochrome c. Nuclei are stained with DAPI. C. Representative immunoblots of NHE1 in total lysate fractions (TLF), intracellular fractions (ITF) and biotinylated fractions (BF) in MCF-7 vector (V) and ΔNErbB2 (ΔN) cells. F. Top panel shows representative immunoblots of NHE1 in TLF and BF in native MCF-7 cells, lower panel shows NHE1 protein level relative to that in native MCF-7 cells under control conditions. Data are representative of 3-5 independent experiments for each condition. Western blot data are shown as means with SEM error bars. * indicates comparison of the value in cisplatin treated to that in the respective cell line control, # indicate comparison of values in vector- and ΔNErbB2 cell lines under same treatment. */#, **/##, ***/### and ****/#### indicate p < 0.05, p < 0.01, p < 0.001 and p < 0.0001, respectively (C, one-way ANOVA with Bonferroni post-test) (A, two-way ANOVA with Bonferroni post-test).</p>