Disulfide bond disrupting agents activate The unfolded protein response in EGFR- and HER2-positive breast tumor cells

Material Information

Disulfide bond disrupting agents activate The unfolded protein response in EGFR- and HER2-positive breast tumor cells
Series Title:
Oncotarget, 2017, Vol. 8, (No. 17), pp: 28971-28989
Law, Brian
Publication Date:
Physical Description:
Journal Article


Many breast cancer deaths result from tumors acquiring resistance to available therapies. Thus, new therapeutic agents are needed for targeting drug-resistant breast cancers. Drug-refractory breast cancers include HER2+ tumors that have acquired resistance to HER2-targeted antibodies and kinase inhibitors, and “Triple- Negative” Breast Cancers (TNBCs) that lack the therapeutic targets Estrogen Receptor, Progesterone Receptor, and HER2. A significant fraction of TNBCs overexpress the HER2 family member Epidermal Growth Factor Receptor (EGFR). Thus agents that selectively kill EGFR+ and HER2+ tumors would provide new options for breast cancer therapy. We previously identified a class of compounds we termed Disulfide bond Disrupting Agents (DDAs) that selectively kill EGFR+ and HER2+ breast cancer cells in vitro and blocked the growth of HER2+ breast tumors in an animal model. DDA-dependent cytotoxicity was found to correlate with downregulation of HER1-3 and Akt dephosphorylation. Here we demonstrate that DDAs activate the Unfolded Protein Response (UPR) and that this plays a role in their ability to kill EGFR+ and HER2+ cancer cells. The use of breast cancer cell lines ectopically expressing EGFR or HER2 and pharmacological probes of UPR revealed all three DDA responses: HER1-3 downregulation, Akt dephosphorylation, and UPR activation, contribute to DDA-mediated cytotoxicity. Significantly, EGFR overexpression potentiates each of these responses. Combination studies with DDAs suggest that they may be complementary with EGFR/HER2-specific receptor tyrosine kinase inhibitors and mTORC1 inhibitors to overcome drug resistance.
Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Brian Law.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the submitter.


This item is only available as the following downloads:

Full Text


28971 1,* 1 4 5 6 6 7 1 1 Department of Chemistry, University of Florida, Gainesville, FL 32611, USA 2 Department of Pharmacology & Therapeutics, University of Florida, Gainesville, FL 32610, USA 3 UF-Health Cancer Center, University of Florida, Gainesville, FL 32610, USA 4 Center for Cell Biology and Cancer Research, Albany Medical College, Albany, NY 12208, USA 5 Department of Biochemistry, University of Florida, Gainesville, FL, 32610, USA 6 Department of Pathology, Immunology, and Laboratory Medicine, Center for Cellular Reprogramming, University of Florida College of Medicine, Gainesville, FL 32610, USA 7 Department of Medicine, University of Florida, Gainesville, FL, 32610, USA These authors have contributed equally to this work Correspondence to: Brian K. Law, email: email: Keywords: Received: Accepted: Published: Copyright ABSTRACT in vitro


28972 INTRODUCTION The principle of oncogene addiction [1] posits that particular cancers developed and continue to grow and survive only by virtue of their initiating oncogene. This principle has guided much of targeted cancer therapy and has led to treatments for breast tumors that overexpress the Human Epidermal Growth Factor Receptor-2 (HER2) receptor tyrosine kinase. These drugs include the HER2 targeted monoclonal antibodies Trastuzumab [4] and Pertuzumab [8] and the HER2/Epidermal Growth Factor Receptor (EGFR/HER1) tyrosine kinase inhibitor improved the survival of patients with HER2+ breast tumors, but tumor resistance to these agents remains monoclonal antibodies and tyrosine kinase inhibitors, other therapeutic strategies are needed to effectively treat patients whose tumors have acquired resistance to resistance is the functional redundancy among EGFR, HER2, and HER3, whereby inactivation of one family member may be compensated for by another family member [19]. Thus, drugs that act on features common driven by the HER2 or EGFR that have become refractory to available targeted therapies. Cancer is associated with dysregulation of protein synthesis and protein folding, referred to as defective proteostasis [23, 24]. HER2-positive breast cancers are addicted to Endoplasmic Reticulum-Associated protein reliance of these cancers on HER-family proteins and the fact that HER-family members EGFR, HER2, HER3, and HER4 share conserved extracellular cysteine-rich repeats may present a burden to the protein folding machinery. Compounds that disrupt the folding of HERfamily proteins may provide an effective means to target cancers that are addicted to these receptors and could Management of patients with HER2+ breast cancers involves combining these HER2-targeted agents with conventional chemotherapy drugs. Further, primary and occurrence that leaves patients with few options. Hence therapeutics could replace cytotoxic chemotherapy drugs, and might be useful for overcoming resistance by simultaneously downregulating HER1-3. EGFR also contributes to the development and progression of breast cancers [26, 27] and is overexpressed Progesterone Receptor (PR), and HER2 [28]. Subsets of lung cancers and glioblastomas harbor activated point mutants or splice variants of EGFR, but such alterations are rare in breast cancer where EGFR is activated primarily by overexpression of the wild type protein [28]. Thus agents that selectively block the folding of HER13 could have a major impact on the treatment of breast cancer. The molecular chaperone Hsp90 facilitates the folding of a number of oncoproteins including HER2 [29, 30]. The Hsp90 inhibitor Geldanamycin and its analogs are under investigation as anticancer agents [31]. Progress in moving Geldanamycin analogs toward clinical application has been slowed by problems relating to drug toxicity and solubility [32]. We recently described a class of agents when applied to breast cancer cells downregulate HER family members EGFR, HER2, and HER3 in parallel, and inactivate Akt [33]. These agents have the potential to ablate drug resistance by overcoming the functional redundancy among HER1-3 and act downstream at the level of Akt to abrogate drug resistance mediated through upstream activation of Phosphatidylinositol 3-Kinase (PI3K). Here we show that in addition to these useful properties, DDAs also activate the Unfolded Protein Response (UPR). Importantly, DDA-induced UPR is potentiated by overexpression of EGFR or HER2, providing a partial explanation as to how DDAs can effectively kill cancer cells without harming normal tissues. DDAs are chemically and mechanistically distinct from other classes of anticancer agents and selectively exacerbate the ER stress caused by the aberrantly high expression of EGFR and HER2 that occurs in breast cancers. RESULTS DDAs induce ER stress Endoplasmic Reticulum (ER), and HER2+ breast cancers are particularly sensitive to DDAs [33] and ER stress/ activate the Unfolded Protein Response (UPR). In the lines, DDAs activated ER stress as indicated by GRP78 upregulation (Figure 1A). The DDA-resistant MDAexpression, suggesting that they have adapted to persistent ER stress. DDAs upregulated GRP78 at the lowest inhibition of cell division and activation of ER stress


28973 Figure 1: DDA responsiveness parallels activation of the unfolded protein response. (A) panel indicates the status of EGFR or HER2 expression and DDA sensitivity. (B) Results are presented as the mean standard deviation of triplicate determinations. (C) DDA sensitive or resistant cell lines were treated (D) The thapsigargin (400 nM) by immunoblot analysis. (E) Luciferase reporter assays measuring the impact of ectopically expressed ATF6 and and are presented as the mean standard deviation of triplicate determinations. (F) Extracts from HEK 293 cells transiently transfected as


28974 ER stress response is mediated by the upstream sensors PERK, IRE1, and ATF6. PERK-dependent activation of an ATF4-CHOP transcriptional axis contributes to cell DDA-sensitive cell lines exhibited upregulation of ATF4 and CHOP in a concentration-dependent manner, while the resistant cell lines expressed high basal ATF4 levels and lacked CHOP expression (Figure 1C). Although the IRE1-Jun kinase axis was previously implicated in ER stress-mediated cell death [36], DDAs did not alter inducers tunicamycin and thapsigargin showed that effective in suppressing Akt phosphorylation (Figure ATF4-CHOP cassettes. Transcriptional reporter assays in HEK293 cells were performed to evaluate whether activity, but potentiated the transcriptional activity of ectopically expressed ATF6 (Figure 1E). Immunoblot analysis of HEK293 cell extracts demonstrated that overexpression of ATF6 increased endogenous GRP78 ATF4, and CHOP levels (Figure 1F). ATF6 activation involves cleavage to release its cytoplasmic domain, which travels to the nucleus to regulate transcription. resulted in higher levels of the cleaved, transcriptionally active form of ATF6. The results in Figure 1 show that DDAs activate all three branches of UPR. Ongoing protein synthesis is required for DDA induction of UPR The sensitivity of HER2+ breast cancer cells to ERAD inhibition depends on continued protein synthesis the translocation step in protein synthesis and by inducing premature chain termination during translation, respectively. DDA activation of the ER stress response was reduced if protein synthesis was inhibited using either PARP cleavage and Akt dephosphorylation, suggesting that induction of ER stress may be partially responsible overcome DDA-mediated downregulation of EGFR or HER3. Similar results were obtained after a 24 hour for their ability to reverse DDA responses. The results indicated that a complicated relationship exists between 2D). A likely explanation for this result is that on the one hand UPR stress is associated with inhibition of protein suppress ER stress induced by misfolding of proteins such as HER1-3, while on the other hand, resolution of ER stress requires the synthesis of proteins including ATF4, CHOP, and GRP78. Overall, the results in Figure 2AD indicate that blockade of protein synthesis with concentration-dependent manner. EGFR or HER2 overexpression sensitizes cancer cells to DDA actions HER3 plays a major role in the survival of HER2+ breast cancers and their resistance to HER2-targeted drugs [20, 21]. It was previously shown that breast cancer cells engineered to overexpress EGFR are sensitized to DDA-induced cell death and Akt dephosphorylation [33], but differential sensitivity to DDA-mediated HER3 downregulation was not examined in that report. T47D cells engineered to overexpress EGFR or HER2 decreased basal HER3 expression (Figure 2E). This decreased baseline, combined with DDA treatment, reduced HER3 expression to very low levels. EGFR overexpression potentiated ER stress as measured by in the EGFR overexpressing cells at this time point, while the levels of these proteins was unchanged in the UPR markers were higher in the EGFR expressing line Akt dephosphorylation was slightly enhanced in the context of EGFR overexpression at this time point. The versus control cells on HER2, HER3, and phospho-Akt EGFR overexpressing line. The ER stress response


28975 frequently peaks and then becomes weaker over time after ER chaperones have been upregulated, protein synthesis has been suppressed, and protein misfolding becomes resolved. Thus, the differences between time points likely results from the peak of the ER stress response occurring earlier in the EGFR overexpressing cells as compared with the control cells. HER4 was not expressed at detectable downregulation at the 24 hour time point, but not at the Figure 2: Ongoing protein synthesis is required for DDA induction of UPR, and elevated DDA sensitivity due to forced EGFR or HER3 overexpression correlates with enhanced HER3 downregulation and increased ER stress. (A) MDAin the indicated combinations and cell extracts were analyzed by immunoblot. (B) (C) (D) cell extracts were analyzed by immunoblot. (E) (F) Vector control or EGFR overexpressing T47D cells (G) Vector control or EGFR overexpressing immunoblot.


28976 DDA effects on HER1-3 and Akt are separable from effects on the ER stress response (2-DOG), thapsigargin, tunicamycin, and dithiothreitol (DTT) to determine whether downregulation of HERfamily receptors, decreased Akt phosphorylation, and induction of cell death is common among all ER stress inducers. 2-DOG strongly activated UPR in both the downregulation of either HER3 expression or Akt phosphorylation, and did not increase cell death as measured by PARP cleavage (Figure 3A). Some ER stress responses result from increased cytoplasmic Ca 2+ Figure 3: DDA activation of UPR is separable from effects on HER1-3 levels and Akt phosphorylation. (A) (2-DOG) in the indicated combinations and cell extracts were analyzed by immunoblot. (B) (C) (D)


28977 mediated through IP 3 receptors. To evaluate the role IP 3 dephosphorylation, or PARP cleavage. Comparison of elevated ATF4 and CHOP expression more effectively weakly activated the ER stress response as measured by GRP78 upregulation. EGFR was downregulated by all of the ER stressors. HER3 levels were particularly sensitive phosphorylation. expression with little upregulation of ATF4 or CHOP expression (Figure 3C). This is in contrast to thapsigargin treatment, which strongly upregulated both ATF4 and CHOP, and more strongly decreased PERK electrophoretic mobility, consistent with its increased phosphorylation downregulated EGFR and HER2 expression, while both decreased Akt phosphorylation. Taken together, the a pattern of ER stress response that is different from that observed with 2-DOG, thapsigargin, and tunicamycin. DTT, although DTT was applied to the cells at a 100 reduces HER-family receptor expression in both cell stressors 2-DOG, thapsigargin, tunicamycin, and DTT with respect to the spectrum of ER stress responses, Akt dephosphorylation, HER-family receptor downregulation, and cell death induction. did not induce an ER stress response in wild type Mouse Embryo Fibroblasts (MEFs) and weakly suppressed Akt phosphorylation and induced PARP cleavage (Figure 3D). However, thapsigargin and tunicamycin induced a robust UPR, markedly suppressed Akt phosphorylation, and strongly upregulated PARP cleavage. In MEFs in which to Ala was heterozygously knocked in [38], CHOP, ATF4, and GRP78 upregulation by thapsigargin and tunicamycin mutation did not affect thapsigarginor tunicamycinexpress very low levels of the HER-family receptors and this may contribute to their relative resistance to thiolCooperation between DDAs and receptor tyrosine kinase inhibitors Since DDAs and receptor tyrosine kinase (RTK) and HER2 through distinct mechanisms, we examined whether these two classes of agents cooperate to inactivate mitogenic signaling pathways and activate to downregulate HER3 levels and Akt phosphorylation (Figure 4A). Under these conditions, the combination treatments did not alter UPR as measured by GRP78, CHOP expression. characterized models for EGFR and HER2 overexpressing breast cancer, but form highly homogenous tumors with questionable relevance to human breast cancer. Patientchoice for studying the effectiveness of anticancer agents against human breast cancers (reviewed in [39]), but the cellular heterogeneity responsible for their higher clinical relevance renders studies of the mechanisms of drug isolated a cell line from the previously described HCI-012 Cell Reprogramming (CCR) [41, 42]. The HCI-012 cell line formed tumors when injected into the mammary morphology of the resulting tumors was similar to that of the parental xenograft line (Supplementary Figure 3A). The HCI-012 cells rapidly initiate cell death if not cultured in the CCR medium (Supplementary Figure approach maintains reversible immortality of epitheliaderived cell lines in vitro associated with upregulation of ER stress markers, on Erk phosphorylation (Figure 4C). Lapatinib partially reduced Akt phosphorylation, and strongly suppressed ERK phosphorylation, but did not alter EGFR, HER2, or HER3 levels, nor did it alter the expression of ER


28978 Figure 4: DDAs may be useful in combination therapies for combating resistance to mTORC1-, EGFR-, or HER2targeted agents. (A) (B) Micrographs of the HCI-012 (C) Immunoblot analysis of HCI-012 cell extracts after the indicated 24 hour treatments. (D) indicated drug combinations. (E) (F) hours of the indicated treatments using the MTT assay. Results are presented as the mean standard deviation of triplicate determinations. (G) drugs combined. (H) by immunoblot. (I) indicate cell death.


28979 suppressed EGFR and HER2 expression and completely abrogated both Akt and Erk phosphorylation. This result suggests that these two agents are complementary in their effects on mitogenic/survival signaling. In the HCI-012 DDA impacts pathways that mediate resistance to HER2and mTORC1-targeted therapeutics resistant, HER2-positive breast cancer, and resistance is thought to be mediated by the activating Phosphatidylinositol 3-kinase (PI3K) mutation H1047R [43]. Observation of cultures revealed that combining Lapatinib cooperated to downregulate EGFR and HER2, to increase fractional PARP cleavage, and to suppress Akt phosphorylation (Figure 4E). The mTORC1 inhibitor dephosphorylation. Lapatinib only weakly potentiated compounds applied individually (Figure 4F). Previous studies demonstrated that in contrast to EGFR or HER2 overexpressing breast cancer lines, the it reduced HER2 expression, but had little effect on the levels or phosphorylation states of the other proteins on HER1-3 expression, or Akt or Erk phosphorylation. HER2, but also strongly downregulated HER3, and suppressed both Akt and Erk phosphorylation. mTORC1 inhibitors such as the rapamycin analogs (rapalogs) inadvertently activate the PI3K/Akt axis by removing negative feedback mediated through S6K1 [44, utility of rapalogs, which are used in immunosuppression, the treatment of human cancers, and the management of Tuberous Sclerosis (TSC) (Reviewed in [46]), the was examined. In TSC, individuals have mutations in the genes coding for the proteins TSC1 or TSC2 and develop benign tumors in multiple tissues in part because the TSC1/TSC2 complex is a GTPase activating protein for the Rheb GTPase responsible for mTORC1 activation (reviewed in [47]). Thus, mTORC1 activation is characteristic of TSC. Rapalogs are FDA-approved for side effect. To address this point, angiosarcoma cells from a TSC2 knockout mouse (TSC2-Ang1; ATCC CRL2620) were used as a model system. Treatment of these which were high under control conditions (Figure 4H). Rapamycin strongly increased Akt phosphorylation and to basal levels. TSC2-Ang1 cell death was only observed arrows), whereas vehicle and rapamycin treated cells continued to proliferate (white arrows) (Figure 4I). The suppressed S6 phosphorylation than rapamycin alone. The results in Figure 4 suggest that DDA combinations with RTK inhibitors might provide improved anticancer actions. Pairing DDAs with rapalogs may both increase mTORC1 inhibition and prevent off-target Akt activation. Preparation and characterization of multivalent DDAs DDA and PEMP-DDA, containing three and four copies of the pharmacophore per molecule, respectively, were synthesized to determine whether they have increased increasing concentration of each compound indicated that PEMP-DDA decreased Akt phosphorylation and This immunoblot analysis was repeated a total of three times and DDA-induced changes in EGFR, phosphoAkt[Thr308], PARP cleavage (cPARP), and GRP78 levels The replicate immunoblot analyses are shown in differences are indicated with P -values obtained using Student's unpaired t -test. All bands were normalized to the corresponding Actin loading control before the ratios between drug treatments were calculated. MTT assays showed that both reduced cell viability in a concentrationDDA pharmacophore [33]. Since the two sulfur atoms of DTDO are involved in DDA chemistry, we examined whether derivatization of the second and third carbon atoms of the four-carbon linker by either hydroxyl or upregulated markers of ER stress, and increased PARP acetylated DTDO derivatives with either cis or trans


28980 through similar mechanisms as the linear forms (e.g., Experiments similar to those carried out with 6. All three DDAs decreased EGFR, HER2, and HER3 expression, increased PARP cleavage, reduced Akt phosphorylation and upregulated the ER stress markers produced similar responses to bi-, tri-, and tetra-functional lines (Supplementary Figure 2A). As expected, the DDAstress in control samples and did not exhibit a response Figure 5: Increasing the number of pharmacophores per DDA molecule improves potency against MDA-MB-468 cells. (A) (B) Immunoblot analyses 308 PARP cleavage (cPARP), and GRP78 expression in panels (C, D, E, and F) P -values. (G) MTT assays performed on cells treated with the indicated concentrations of DDAs for 72 hours. (H) Structural alterations to the parent cyclic, monovalent DDA, DTDO. (I) analyzed by immunoblot.


28981 respectively). DDAs are not toxic to cardiomyocytes or MCF10/DCIS cells Cardiotoxicity is a side effect of the HER2 cardiomyocytes differentiated from human induced Pluripotent Stem Cells (iPSCs) as described previously [48, 49]. Microscopic examination of cardiomyocytes appearance (Figure 7A, left panel) and their rate of beating was not altered (Supplementary Videos, Supplementary Video 1Supplementary Video 3). Immunoblot analysis demonstrated that the cardiomyocytes expressed HER2, or suppress Akt phosphorylation (Figure 7A, upper right viability of cardiomyocytes (Figure 7A, lower right panel). The MCF10/DCIS cell line serves as a model of Ductal Carcinoma in situ in which cancer cells aberrantly proliferate, but are unable to invade through the basement membrane to invade locally. MCF10/DCIS cells are considered to express normal levels of EGFR and HER2 DISCUSSION Previous work showed that cancer cell death caused by DDAs correlates with HER1-3 downregulation and Akt dephosphorylation [33]. The results presented here UPR. We previously demonstrated the ability of DDAs to Figure 6: Increasing the number of pharmacophores per DDA molecule improves potency against BT474 cells. (A) (B and C) respectively. P -values. (D) MTT assays performed on cells treated with the indicated concentrations of DDAs for 72 hours.


28982 is a critical component of the folding of both integral membrane and secreted proteins, and interference with this process by treatment with reducing agents such as Figure 1 demonstrate that DDAs activate all three branches Figure 7: DDAs are not toxic to cardiomyocytes or MCF10/DCIS cells. (A) Cardiomyocytes differentiated from iPSCs were treated as indicated for 24 hours and photographed (left panel). Videos demonstrating beating of the cardiomyocytes are presented in Supplementary Videos Supplementary Video 1Supplementary Video 3. Cardiomyocytes treated in parallel were subjected to immunoblot analysis with the indicated antibodies (top right panel), and MTT cell viability assays (bottom right panel). (B) Photomicrographs of


28983 activates UPR at low micromolar concentrations, while millimolar concentrations of DDT are required to induce a similar level of ER stress. It is tempting to speculate that this striking difference in potency relates to the structural the pharmacophore to interconvert between cyclic and linear forms. Alternately, the bifunctional nature of DDAs DTT or similar reducing agents. The observation that DDAs act through mechanisms involving UPR, Akt inactivation, and HER1-3 downregulation raises the question of which of these pathways contributes to DDA anticancer actions, and whether these responses are mechanistically related. provide some insight into these issues. Thapsigargin more effectively suppresses Akt phosphorylation than either tunicamycin or thapsigargin (Figure 1D). Like EGFR and HER3 and suppresses Akt phosphorylation, but under these conditions only weakly induces PARP cleavage and GRP78 expression, and does not upregulate cells as measured by upregulation of GRP78 and ATF4, but does not induce PARP cleavage, or suppress Akt phosphorylation (Figure 3A). Further, 2-DOG does not receptor expression, suppression of Akt phosphorylation, and increased PARP cleavage relate to the thiol reactivity alone. However, results obtained with protein synthesis inhibitors show that while blockade of translation partially overcome PARP cleavage, this treatment did 3 expression (Figure 2AD). Overall, the results of these experiments obtained with the use of 2-DOG cancer cell death results from a combination of UPR activation, HER1-3 downregulation, and decreased Akt phosphorylation. Ideally, cancer therapeutic agents should be toxic to cancer cells with little or no impact on normal of Chronic Myelogenous Leukemia (CML), HER2directed monoclonal antibodies and tyrosine kinase inhibitors for the treatment of HER2+ breast tumors, and ovarian cancers. However, these approaches suffer from cancer escape from therapy through a variety of mechanisms. Thus, in many cases cancer cures may require multiple drugs to overcome both the driver oncogene and potential resistance mechanisms, or the discovery of multifunctional anticancer drugs that target the appropriate mechanisms. Progesterone Receptor (PR-), and HER2 expression (HER2-) at the protein level [28]. The potential for the use of DDAs DDAs are selectively cytotoxic to breast cancer cells that overexpress either HER2 or EGFR and EGFR overexpression potentiates DDA-induced Akt dephosphorylation [33]. In the present study we examined whether EGFR overexpression also potentiates other DDA responses including HER3 downregulation and activation of ER stress. HER3 mediates a number of resistance mechanisms to HER2-targeted therapies through its ability to be phosphorylated by EGFR, IGF-1R, and c-MET T47D ER+ breast cancer cell line ectopic expression of either EGFR or HER2 rendered endogenous HER3 more (Figure 2E), and EGFR overexpression sensitizes EGFR, (Figure 2G). A concern with DDAs relates to their ability to of multiple secreted or membrane proteins. A number of cell types that express normal levels of EGFR and HER2, such as T47D, MCF10/DCIS, MEF lines are unaffected by DDAs. However, T47D cells become responsive to the toxic effects of DDAs upon overexpression of EGFR or HER2 ([33] and herein (Figure 2EG). Since a side effect of Trastuzumab is cardiotoxicity, the possibility that DDAs might also be cardiotoxic is a concern. The results presented in Figure 7 indicate that while the cardiomyocytes expressed high levels of HER2, no effect on the beating of the cardiomyocytes in culture (see videos Supplementary Video 1-Supplementary Video 3 in supplemental material). We speculate that the reason that DDAs do not downregulate HER2 in cardiomyocytes is that in these cells HER2 is expressed at normal levels


28984 rather than being overexpressed. Therefore the addition cardiomyocytes. This is consistent with the observation that nearly all breast cancers express HER2, but DDAs are only toxic to the lines that exhibit dramatic HER2 or EGFR overexpression. in the HER-family receptor cysteine-rich extracellular bonds, we hypothesize that overexpression of HERfamily receptors such as EGFR selectively exacerbate the ER stress induced by DDAs. Consistent with this expectation, EGFR overexpression in T47D effect was particularly notable at early time points. We propose the model for DDA action in Figure 8A where DDAs selectively induce the death of EGFR+ and HER2+ cancers through the suppression of Akt phosphorylation, downregulation of HER1-3 expression, and activation of UPR. DDA induction of UPR and Akt dephosphorylation are potentiated by overexpression of EGFR, or to a lesser extent, HER2. Due to their unique Figure 8: Model for the anticancer actions of DDAs. (A) DDAs function to suppress tumor cell division and survival through mechanisms involving suppression of Akt phosphorylation, downregulation of EGFR (HER1), HER2, and HER3, and activation of UPR. Overexpression of EGFR or HER2 potentiates each of these mechanisms. (B) At normal expression levels, EGFR and HER2 are experimentally in Figures 1-6.


28985 and multifunctional mechanisms of action, DDAs may be well suited for targeting the pathways responsible for resistance to HER2and EGFR-targeted agents and prove to be complementary to other therapeutic modalities including monoclonal antibodies and receptor tyrosine kinase inhibitors targeting HERfamily oncogenes. The high sensitivity of EGFR or HER2 overexpressing cancer cells to DDAs may derive proteins combined with the ability of DDAs to prevent In addition to examining DDA effects on breast tumors with overexpression of wild type EGFR or HER2, in future studies it will be important to determine whether mutants or splice variants of these proteins, such as HER2-delta 16 [62] are responsive to DDAs. Cyclic and linear forms of the DDA pharmacophore can interconvert, and elicit similar cellular responses. the DDA pharmacophore result in loss of biological number of pharmacophores per molecule elevates This suggests a modular lead optimization approach in which improvements are made to the activity of the pharmacophore structure, and the optimized pharmacophore is then appended to a polyvalent scaffold to further increase DDA potency. These efforts are currently ongoing in our laboratories. MATERIALS AND METHODS Cell lines, construction of stable cell lines, and recombinant adenoviruses The following cell lines were purchased from American Type Culture Collection (ATCC) (Manassas, The HCI-012 cell line was derived from the HCI-012 Alana Welm [40] using conditional cell reprogramming is shown in Supplementary Figure 3. Wild type and Fibroblasts (MEFs) were described previously [38]. Recombinant retroviruses were prepared and used to produce stable cell lines as described previously [66, 67]. Retroviral vectors encoding EGFR (plasmid 11011, [68]) and HER2 (plasmid 40978 [69]) were from Addgene (Cambridge, MA). Cell culture, preparation of cell extracts, and immunoblot analysis medium (GE Healthcare Life Sciences Logan, UT) 2 3 VO 4 2 H 2 P 2 O 7 as described previously [70]. Immunoblot analysis was carried out using primary (San Jose, CA) [PAI-1, 612024], Millipore (Temecula, quantify immunoblot results, bands were analyzed using ImageJ ( and each band was normalized to the corresponding Actin loading control band. EGF (GF001) was obtained from Chemicon Therapeutics Program. The following reagents were purchased from the indicated sources: tunicamycin, 2-deoxyglucose: Sigma-Aldrich (St. Louis, MO); thapsigargin: AdipoGen (San Diego, CA); Puromycin, Transfection of HEK293 cells and luciferase assays Transfections were performed using Lipofectamine Reagent (Invitrogen, Carlsbad, CA) according to the hours after transfection, and cell extracts analyzed by luciferase assays, with background readings subtracted from the luciferase assay values. Relative Luminescence


28986 Units (RLUs) were normalized to the number of micrograms of protein assayed. The results are presented as the mean of triplicate determinations standard deviation. The ATF6 Reporter (plasmid 11976 [71]) and obtained from Addgene. MTT cell viability assays Cell viability was evaluated using MTT instructions (kit CGD1, Sigma-Aldrich, St. Louis, MO). Thymidine incorporation assays Tritiated thymidine incorporation assays were performed as described previously [73]. Results are presented as the mean standard deviation of triplicate or quadruplicate determinations. RT-PCR follows: Forward: CCTGGTTGCTGAAGAGGAGG and Reverse: CCATGGGGAGATGTTCTGGAG. Forward: GGATGCAGAAGGAGATCAC and Reverse: AAGGTGGACAGCGAGGCCAG. Reactions were gels. Synthesis of DDAs DDA synthesis and characterization is presented in Supplemental Material (Supplementary Information and Supplementary Figure 4). Cardiomyocyte differentiation To induce cardiomyocyte differentiation we utilized the PSC Cardiomyocyte Differentiation Kit iPSCs were grown in feeder-free conditions using hES mm dish using Gentle Cell Dissociation Reagent (Stem single cell suspension. The cells were divided equally among the wells of a 12-well plate coated with Matrigel using mTeSR1 medium and ROCK inhibitor (10 uM changed daily with mTeSR1 until the iPSCs formed a mesoderm differentiation, Cardiomyocyte Differentiation Medium A was added for 48 hours (Days 0-2). For cardiac For cardiomyocyte maturation, cells were maintained in Cardiomyocyte Maintenance Medium for the duration of culture (Day 4+), replacing medium every other day. Spontaneous cell contraction began on day 10. ACKNOWLEDGMENTS Assistant Secretary of Defense for Health Affairs through (RC). These studies were supported in part by Florida University of Florida for a Graduate School Fellowship. this article was funded in part by the University of Florida Open Access Publishing Fund. CONFLICTS OF INTEREST REFERENCES 1. 2. addiction--a rationale for molecular targeting in cancer 3. Pagliarini R, Shao W, Sellers WR. Oncogene addiction: pathways of therapeutic response, resistance, and road maps 4. J, Rosen PP, Twaddell T, Henderson IC, et al. Phase II study of weekly intravenous trastuzumab (Herceptin) in patients with HER2/neu-overexpressing metastatic breast cancer. Semin Oncol. 1999; 26:78-83. Molina MA, Codony-Servat J, Albanell J, Rojo F, Arribas J,


28987 receptor monoclonal antibody, inhibits basal and activated Her2 ectodomain cleavage in breast cancer cells. Cancer Res. 2001; 61:4744-4749. 6. Pegram MD, O'Callaghan C. Combining the anti-HER2 antibody trastuzumab with taxanes in breast cancer: 7. McKeage K, Perry CM. Trastuzumab: a review of its use in the treatment of metastatic breast cancer overexpressing HER2. Drugs. 2002; 62:209-243. 8. S, Tang S, Song P, Liu Q, Ringgold K, Pilaro AM, Tilley approval of dual anti-HER2 regimen: pertuzumab in combination with trastuzumab and docetaxel for HER2positive metastatic breast cancer. Clin Cancer Res. 2013; 19:4911-4916. 9. Heinrich G, Untch M, Jackisch C. HER2 Dimerization Inhibitor Pertuzumab Mode of Action and Clinical Data 10. O'Sullivan CC, Swain SM. Pertuzumab : evolving therapeutic strategies in the management of HER213:779-790. 11. Johnston SR, Leary A. Lapatinib: a novel EGFR/HER2 12. Montemurro F, Valabrega G, Aglietta M. Lapatinib: a dual inhibitor of EGFR and HER2 tyrosine kinase activity. 13. 2007; 99:348-349. 14. LS, Arbushites MC, Maltzman JD, Forster JK, Rubin SD, overexpressing advanced or metastatic breast cancer that regimens. Ann Oncol. 2009; 20:1026-1031. PI-3K/Akt activation leads to a multidrug resistance in human breast adenocarcinoma cells. Oncogene. 2003; 16. 17. Valabrega G, Montemurro F, Aglietta M. Trastuzumab: mechanism of action, resistance and future perspectives in HER2-overexpressing breast cancer. Ann Oncol. 2007; 18:977-984. 18. of lapatinib resistance in HER2-positive breast tumor cells: 19. cancer inhibitory activity against EGFR, HER2, HER3, and VEGF through disruption of HER/MET crosstalk. Cancer 20. pseudokinase HER3 in the acquired resistance against Trans. 2014; 42:831-836. 21. autocrine signaling axis can mediate acquired lapatinib 22. LM. Downregulation of HER3 by a novel antisense of EGFR and HER2 tyrosine kinase inhibitors in animal models. Mol Cancer Ther. 2013; 12:427-437. 23. Dufey E, Urra H, Hetz C. ER proteostasis addiction in 33:40-47. 24. reticulum: a new perturbation site for targeted cancer therapy. Cell Res. 2011; 21:867-883. driven breast cancer cells require ER-associated degradation 26. Milanezi F, Carvalho S, Schmitt FC. EGFR/HER2 in breast cancer: a biological approach for molecular diagnosis and therapy. Expert Rev Mol Diagn. 2008; 8:417-434. 27. HER2) and the epidermal growth factor receptor (EGFR) in breast cancer. Pathobiology. 1993; 61:268-282. 28. Mazzucchelli L, Frattini M. Molecular characterization of EGFR and EGFR-downstream pathways in triple negative breast carcinomas with basal like features. Histol 29. W. Geldanamycin destabilizes HER2 tyrosine kinase and suppresses Wnt/beta-catenin signaling in HER2 overexpressing human breast cancer cells. Oncol Rep. 2007; 17:89-96. 30. geldanamycin dimer that induces the selective degradation of HER-family tyrosine kinases. Cancer Res. 2000; 60:2090-2094.


28988 31. Synthesis and biological evaluation of geldanamycin analogs against human cancer cells. Cancer Chemother 32. prodrug in mPEG-b-PCL micelles greatly enhances tolerability and pharmacokinetics in rats. J Control Release. 2008; 129:33-40. 33. that downregulate EGFR, HER2, and HER3 in parallel. oncotarget.3398. 34. death by promoting protein synthesis and oxidation in the stressed endoplasmic reticulum. Genes Dev. 2004; 18:3066-3077. 36. Free cholesterol-induced macrophage apoptosis is mediated by inositol-requiring enzyme 1 alpha-regulated activation (Shanghai). 2008; 40:226-234. 37. Kumar R, Azam S, Sullivan JM, Owen C, Cavener DR, activates the eukaryotic initiation factor 2alpha kinase, 38. Translational control is required for the unfolded protein response and in vivo glucose homeostasis. Mol Cell. 2001; 39. Whittle JR, Lewis MT, Lindeman GJ, Visvader JE. Patientderived xenograft models of breast cancer and their 40. L, Randall RL, Stijleman IJ, et al. Tumor grafts derived 41. A, Minas T, Simic V, Johnson M, Albanese C, Schlegel R, cooperate with ROCK inhibitor to conditionally reprogram and immortalize epithelial cells. Am J Pathol. 2013; 183:1862-1870. 42. feeder cells induce the conditional reprogramming of 43. JA, Arteaga CL. H1047R phosphatidylinositol 3-kinase mutant enhances HER2-mediated transformation by heregulin production and activation of HER3. Oncogene. 44. smooth muscle cell differentiation through insulin receptor substrate-1/phosphatidylinositol 3-kinase/Akt2 feedback tyrosine kinase signaling and activates Akt. Cancer Res. 46. Sadowski K, Kotulska K, Jozwiak S. Management of side effects of mTOR inhibitors in tuberous sclerosis patients. 47. Sampson JR. Therapeutic targeting of mTOR in tuberous 48. 49. Singh AM, Li FQ, Hamazaki T, Kasahara H, Takemaru K, pathway, facilitates cardiomyocyte differentiation of murine targets breast ductal carcinoma in situ (DCIS) stem/ Kaji EH, Lodish HF. In vitro unfolding of retinol-binding protein by dithiothreitol. Endoplasmic reticulum-associated bond formation and protein folding in the endoplasmic Workman P. Cancer genome targets: RAF-ing up tumor cells to overcome oncogene addiction. Expert Rev Anticancer Ther. 2002; 2:611-614. Kaelin WG, Jr. The concept of synthetic lethality in the Garber K. Synthetic lethality: killing cancer with cancer. J


28989 OI, Moore DT, Perou CM. Phenotypic evaluation of the basal-like subtype of invasive breast carcinoma. Mod Pathol. 2006; 19:264-271. O'Hare MJ. Epidermal growth factor receptor expression on human breast luminal and basal cells in vitro Epithelial pathways as determinants of lapatinib unresponsiveness in HER2-positive gastric cancer. Clin Cancer Res. 2014; 60. contribute to enhanced proliferation and carcinogenesis 61. Insulin-like growth factor-1 receptor inhibition induces a resistance mechanism via the epidermal growth factor receptor/HER3/AKT signaling pathway: rational basis for cotargeting insulin-like growth factor-1 receptor and epidermal growth factor receptor in hepatocellular 62. delta-HER2 and its clinicopathological correlation in HER2-overexpressing breast cancer. Mol Med Rep. 2016; 63. formation and metastasis evoked by a HER2 splice variant. 64. Jones FE. Oncogenic HER2{Delta}16 suppresses miRresistance of breast tumors. Carcinogenesis. 2010; Shad A, Dritschilo A, et al. Use of reprogrammed cells to Med. 2012; 367:1220-1227. 66. P, Kim JS, et al. Glucocorticoids and histone deacetylase inhibitors cooperate to block the invasiveness of basal-like breast cancer cells through novel mechanisms. Oncogene. 2013; 32:1316-1329. 67. containing protein 1 and the epidermal growth factor Res. 2016; 18:80. 68. Greulich H, Chen TH, Feng W, Janne PA, Alvarez JV, WR, Meyerson M. Oncogenic transformation by inhibitor2:e313. 69. 70. Law ME, Aakre ME, Covington C, Moses HL. Rapamycin potentiates transforming growth factor beta-induced growth arrest in nontransformed, oncogene-transformed, and 71. 72. senses endoplasmic reticulum (ER) stress and causes 73. farnesyltransferase inhibitor involves inhibition of the