Alexey V. Bazarov

PIM1 kinase inhibition as a targeted therapy against triple-negative breast tumors with elevated MYC expression

Triple-negative breast cancer (TNBC), in which cells lack expression of the estrogen receptor (ER), the progesterone receptor (PR) and the ERBB2 (also known as HER2) receptor, is the breast cancer subtype with the poorest outcome. No targeted therapy is available against this subtype of cancer owing to a lack of validated molecular targets. We previously reported that signaling involving MYC—an essential, pleiotropic transcription factor that regulates the expression of hundreds of genes—is disproportionally higher in triple-negative (TN) tumors than in receptor-positive (RP) tumors. Direct inhibition of the oncogenic transcriptional activity of MYC has been challenging to achieve. Here, by conducting a shRNA screen targeting the kinome, we identified PIM1, a non-essential serine–threonine kinase, in a synthetic lethal interaction with MYC. PIM1 expression was higher in TN tumors than in RP tumors and was associated with poor prognosis in patients with hormone- and HER2-negative tumors. Small-molecule PIM kinase inhibitors halted the growth of human TN tumors with elevated MYC expression in patient-derived tumor xenograft (PDX) and MYC-driven transgenic mouse models of breast cancer by inhibiting the oncogenic transcriptional activity of MYC and restoring the function of the endogenous cell cycle inhibitor, p27. Our findings warrant clinical evaluation of PIM kinase inhibitors in patients with TN tumors that have elevated MYC expression.

Promotion of variant human mammary epithelial cell outgrowth by ionizing radiation: an agent- based model supported by in vitro studies

IntroductionMost human mammary epithelial cells (HMEC) cultured from histologically normal breast tissues enter a senescent state termed stasis after 5 to 20 population doublings. These senescent cells display increased size, contain senescence associated β-galactosidase activity, and express cyclin-dependent kinase inhibitor, p16INK4A (CDKN2A; p16). However, HMEC grown in a serum-free medium, spontaneously yield, at low frequency, variant (v) HMEC that are capable of long-term growth and are susceptible to genomic instability. We investigated whether ionizing radiation, which increases breast cancer risk in women, affects the rate of vHMEC outgrowth.MethodsPre-stasis HMEC cultures were exposed to 5 to 200 cGy of sparsely (X- or γ-rays) or densely (1 GeV/amu 56Fe) ionizing radiation. Proliferation (bromodeoxyuridine incorporation), senescence (senescence-associated β-galactosidase activity), and p16 expression were assayed in subcultured irradiated or unirradiated populations four to six weeks following radiation exposure, when patches of vHMEC became apparent. Long-term growth potential and p16 promoter methylation in subsequent passages were also monitored. Agent-based modeling, incorporating a simple set of rules and underlying assumptions, was used to simulate vHMEC outgrowth and evaluate mechanistic hypotheses.ResultsCultures derived from irradiated cells contained significantly more vHMEC, lacking senescence associated β-galactosidase or p16 expression, than cultures derived from unirradiated cells. As expected, post-stasis vHMEC cultures derived from both unirradiated and irradiated cells exhibited more extensive methylation of the p16 gene than pre-stasis HMEC cultures. However, the extent of methylation of individual CpG sites in vHMEC samples did not correlate with passage number or treatment. Exposure to sparsely or densely ionizing radiation elicited similar increases in the numbers of vHMEC compared to unirradiated controls. Agent-based modeling indicated that radiation-induced premature senescence of normal HMEC most likely accelerated vHMEC outgrowth through alleviation of spatial constraints. Subsequent experiments using defined co-cultures of vHMEC and senescent cells supported this mechanism.ConclusionsOur studies indicate that ionizing radiation can promote the outgrowth of epigenetically altered cells with pre-malignant potential.

Telomerase activation by c-Myc in human mammary epithelial cells requires additional genomic changes

A central question in breast cancer biology is how cancer cells acquire telomerase activity required for unlimited proliferation. According to one model, proliferation of telomerase(-) pre-malignant cells leads to telomere dysfunction and increased genomic instability. Such instability leads in rare cases to reactivation of telomerase and immortalization. The mechanism of telomerase reactivation remains unknown. We have studied immortalization of cultured human mammary epithelial cells by c-Myc, a positive transcriptional regulator of the hTERT gene encoding the catalytic subunit of telomerase. Retrovirally introduced c-Myc cDNA resulted in immortalization of human mammary epithelial cells in which the cyclin dependent kinase inhibitor, p16INK4A, was inactivated by an shRNA-encoding retrovirus. However, while c-Myc introduction immediately resulted in increased activity of transiently transfected hTERT promoter reporter constructs, endogenous hTERT mRNA levels did not change until about 60 population doublings after c-Myc introduction. Increased endogenous hTERT transcripts and stabilization of telomeric DNA in cells expressing exogenous c-Myc coincided with telomere dysfunction-associated senescence in control cultures. Genome copy number analyses of immortalized cells indicated amplifications of some or all of chromosome 5, where hTERT genes are located. hTERT gene copy number, however, was not increased in one case. The results are consistent with the hypothesis that changes in chromosome 5, while not necessarily increasing hTERT gene copy number, resulted in removal of repressive chromatin structures around hTERT loci, allowing induction of hTERT transcription. These in vitro results model one possible sequence of events leading to immortalization of breast epithelial cells during cancer progression.

p16(INK4a) -mediated suppression of telomerase in normal and malignant human breast cells.

The cyclin‐dependent kinase inhibitor p16INK4a (CDKN2A) is an important tumor suppressor gene frequently inactivated in human tumors. p16 suppresses the development of cancer by triggering an irreversible arrest of cell proliferation termed cellular senescence. Here, we describe another anti‐oncogenic function of p16 in addition to its ability to halt cell cycle progression. We show that transient expression of p16 stably represses the hTERT gene, encoding the catalytic subunit of telomerase, in both normal and malignant breast epithelial cells. Short‐term p16 expression increases the amount of histone H3 trimethylated on lysine 27 (H3K27) bound to the hTERT promoter, resulting in transcriptional silencing, likely mediated by polycomb complexes. Our results indicate that transient p16 exposure may prevent malignant progression in dividing cells by irreversible repression of genes, such as hTERT, whose activity is necessary for extensive self‐renewal.

BORIS (CTCFL) is not expressed in most human breast cell lines and high grade breast carcinomas

BORIS (CTCFL) is the only known paralog of the versatile regulatory protein CTCF, a multifunctional DNA binding protein that mediates distinct gene regulatory functions involved in cell growth, differentiation, and apoptosis. Unlike CTCF, the expression of BORIS is normally restricted to specific cells in testes (the only cells where CTCF is not expressed), where it may play a role in reprogramming the methylation pattern of male germ line DNA. Frequent amplification of the 20q13.2 region, which contains the BORIS gene, and expression of BORIS transcripts in diverse human tumors and cell lines have led to the hypothesis that aberrant expression of BORIS may play a role in tumorigenesis by interfering with CTCF functions. However, recent studies using more quantitative methods indicate low frequency of BORIS expression in melanoma, ovarian, prostate, and bladder carcinomas. To investigate the relationship between chromosome 20q13 amplification and BORIS mRNA levels within breast cancer cell lines and tissues, we developed a quantitative RT-PCR assay to measure the levels of BORIS mRNA. Endpoint RT-PCR assays were also used to investigate the possible expression of alternatively spliced variants. Using multiple primer sets and controls, we found that neither mature BORIS transcripts nor spliced variants are commonly expressed at detectable levels in malignant breast cells or tissues, although endogenous BORIS transcripts can be induced in MCF-7 cells following 5-aza-2′-deoxycytidine treatment. In conclusion, in most breast cancer cells, endogenous BORIS is unlikely to be expressed at sufficient levels to interfere with CTCF functions. Thus it is improbable that aberrant BORIS expression plays a role in most human breast cancers.

The specific role of pRb in p16INK4A-mediated arrest of normal and malignant human breast cells

RB family proteins pRb, p107 and p130 have similar structures and overlapping functions, enabling cell cycle arrest and cellular senescence. pRb, but not p107 or p130, is frequently mutated in human malignancies. In human fibroblasts acutely exposed to oncogenic ras, pRb has a specific role in suppressing DNA replication, and p107 or p130 cannot compensate for the loss of this function; however, a second p53/p21-dependent checkpoint prevents escape from growth arrest. This model of oncogene-induced senescence requires the additional loss of p53/p21 to explain selection for preferential loss of pRb function in human malignancies. We asked whether similar rules apply to the role of pRb in growth arrest of human epithelial cells, the source of most cancers. In two malignant human breast cancer cell lines, we found that individual RB family proteins were sufficient for the establishment of p16-initiated senescence, and that growth arrest in G1 was not dependent on the presence of functional pRb or p53. However, senescence induction by endogenous p16 was delayed in primary normal human mammary epithelial cells with reduced pRb but not with reduced p107 or p130. Thus, under these circumstances, despite the presence of functional p53, p107 and p130 were unable to completely compensate for pRb in mediating senescence induction. We propose that early inactivation of pRb in pre-malignant breast cells can, by itself, extend proliferative lifespan, allowing acquisition of additional changes necessary for malignant transformation.

A tumor suppressing function in the epithelial adhesion protein Trask

Trask/CDCP1 is a transmembrane glycoprotein widely expressed in epithelial tissues whose functions are just beginning to be understood, but include a role as an anti-adhesive effector of Src kinases. Early studies looking at RNA transcript levels seemed to suggest overexpression in some cancers, but immunostaining studies are now providing more accurate analyses of its expression. In an immuno-histochemical survey of human cancer specimens, we find that Trask expression is retained, reduced or sometimes lost in some tumors compared with their normal epithelial tissue counterparts. A survey of human cancer cell lines also show a similar wide variation in the expression of Trask, including some cell types with the loss of Trask expression, and additional cell types that have lost the physiological detachment-induced phosphorylation of Trask. Three experimental models were established to interrogate the role of Trask in tumor progression, including two gain-of-function models with tet-inducible expression of Trask in tumor cells lacking Trask expression, and one loss-of-function model to suppress Trask expression in tumor cells with abundant Trask expression. The induction of Trask expression and phosphorylation in MCF-7 cells and in 3T3v-src cells was associated with a reduction in tumor metastases while the shRNA-induced knockdown of Trask in L3.6pl cancer cells was associated with increased tumor metastases. The results from these three models are consistent with a tumor-suppressing role for Trask. These data identify Trask as one of several potential candidates for functionally relevant tumor suppressors on the 3p21.3 region of the genome frequently lost in human cancers.

Abstract B34: PIM kinase as a novel therapeutic target for triple-negative breast cancer

The greatest clinical challenge in treating breast cancer occurs in those patients whose tumors lack expression of the estrogen and progesterone receptors and that of the HER2 oncoprotein. No targeted therapeutic strategies currently exist against this aggressive type of “triple negative” breast cancer (TNBC) due to lack of validated targets. We previously found that MYC mRNA, protein, and its signaling were disproportionally elevated in TN compared to receptor positive (RP) breast cancer. We sought to take advantage of the unique molecular feature found in this tumor type to identify potent and effective treatment strategies. Since MYC is an oncogenic transcription factor, rationally designed small molecule inhibitors that can directly inhibit its activity are not available for clinical use. An alternative approach to selectively kill MYC-driven tumors is to inhibit those proteins that are indispensable for the viability of such tumors, but are not essential in non-tumorigenic cells. This form of “indirect” treatment strategy has become known as the “synthetic-lethal” approach. To identify novel targets that are readily druggable for treating MYC-driven TNBC, we conducted a kinome MYC synthetic lethal shRNA screen in non-immortalized human mammary epithelial cells expressing a 4-hydroxytamoxifen (TAM)-activatable MycER transgene (HMEC-MycER). Of 600 human kinases targeted by 2,000 individual shRNA clones, 9 kinases were identified as hits as they were essential specifically for the MYC-activated HMEC cells. Among these hits, we focused on PIM1, a non-essential kinase, the knock-down of which had the greatest efficacy in causing cell death in the MYC-activated cells and had minimum inhibitory effect on the growth of the control cells. We determined that PIM1 expression was elevated in TN tumors and was associated with poor prognosis specifically in patients with hormone receptor-negative tumors. Small molecule PIM kinase inhibitors halted the growth of human TN tumors with elevated MYC expression in a patient-derived tumor xenograft (PDX) mouse model by inhibiting oncogenic transcriptional activity of MYC while simultaneously restoring the function of the endogenous cell cycle inhibitor p27. Thus, our findings warrant clinical evaluation of small molecule PIM kinase inhibitors in patients with TN tumors that exhibit elevated MYC expression. Note: This abstract was not presented at the conference. Citation Format: Dai Horiuchi, Alicia Y. Zhou, Alexandra N. Corella, Christina Yau, Sanjeev Balakrishnan, Kai Kessenbrock, Devon A. Lawson, Roman Camarda, Brittany N. Anderton, Alexey V. Bazarov, Henok Eyob, Julia Rohrberg, Paul Yaswen, Michael T. McManus, Hope S. Rugo, Zena Werb, Andrei Goga. PIM kinase as a novel therapeutic target for triple-negative breast cancer. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Breast Cancer Research; Oct 17-20, 2015; Bellevue, WA. Philadelphia (PA): AACR; Mol Cancer Res 2016;14(2_Suppl):Abstract nr B34.

Abstract LB-122: PIM1 kinase inhibition halts the growth of MYC-overexpressing triple-negative breast tumors

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA The proto-oncogene MYC affects diverse cellular processes ranging from cell proliferation, metabolism, to pluripotency of embryonic stem cells. While de-regulated MYC signaling has been commonly identified in a variety of human malignancies, no therapeutic strategies have been clinically established to specifically target tumors that present elevated MYC expression. We have previously reported that triple-negative (TN) breast cancers, which lack the expression of predictive biomarkers of response (i.e., estrogen/progesterone receptors, human epidermal growth factor receptor 2), exhibit significantly elevated MYC expression as well as MYC pathway activation (Horiuchi, D., et al. 2012 JEM). To understand the vulnerabilities of TN tumors, we performed a kinome synthetic-lethal shRNA screen in human mammary epithelial cells in which MYC activity can be controlled (i.e., HMEC-MycER cells). The screen yielded 10+ potential synthetic lethal partners of MYC, including recently described AMPK-related kinase 5 (Liu, L., et al. 2012. Nature). One of our top hits was the proto-oncogene PIM1, a non-essential serine/threonine kinase previously shown to be a genetic enhancer of MYC in transgenic mouse models of lymphomas and prostate cancer. Our bioinformatics studies performed on four independent clinical cohorts (n: 146 ∼ 683 patients each) showed that in all cohorts both MYC and PIM1 were significantly enriched in the TN populations. PIM1 expression was correlated with that of MYC in three out of the four cohorts. Furthermore, we found that PIM1 overexpression alone was a poor prognostic factor associated with diminished recurrence-free survival. Our in vivo efficacy studies using conventional as well as novel “human-in-mouse,” patient derived orthotopic xenograft (PDX) models demonstrated that PIM was essential for the growth of MYC-overexpressing human TN tumors. In the PDX tumors, PIM inhibition nearly completely arrested tumor growth, which was associated with significant up-regulation of an endogenous cell cycle inhibitor p27, a known PIM1 substrate, down-regulation of protein synthesis machinery, and inhibition of MYC activity itself. These observations held true in a panel of breast cancer cell lines, and elevating the p27 expression alone was sufficient to inhibit cancer cell proliferation. Thus, our findings warrant the use of PIM kinase inhibitors in treating TN tumors that exhibit the elevated expression of MYC and/or PIM. Citation Format: Dai Horiuchi, Alicia Y. Zhou, Alexandra N. Corella, Christina Yau, Devon A. Lawson, Alexey V. Bazarov, Paul Yaswen, Michael T. McManus, Zena Werb, Alana L. Welm, Andrei Goga. PIM1 kinase inhibition halts the growth of MYC-overexpressing triple-negative breast tumors. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-122. doi:10.1158/1538-7445.AM2014-LB-122

Abstract B232: PIM1 kinase is essential for the growth of MYC-overexpressing triple-negative breast tumors and is an efficacious therapeutic target.

While de-regulated MYC signaling has been identified in a wide variety of human malignancies, no targeted therapeutic strategies have been established to target tumors with elevated MYC expression. We have previously reported that receptor “triple negative” (TN) breast cancers, which lack predictive biomarkers of response (i.e., estrogen/progesterone receptors, human epidermal growth factor receptor 2), exhibit significantly elevated MYC expression as well as MYC pathway activation (Horiuchi, D., et al. 2012 JEM). In an effort to uncover the potential Achilles’ heels of TN tumors, we performed a kinome, synthetic-lethal shRNA screen in human mammary epithelial cells in which MYC activity can be controlled (i.e., HMEC-MycER cells). The screen yielded 13 potential synthetic lethal partners of MYC, including recently described AMPK-related kinase 5 (Liu, L., et al. 2012. Nature). One of our top hits is the proto-oncogene PIM1, previously shown by others to be a genetic enhancer of MYC in transgenic mouse models of lymphomas and prostate cancer. Our bioinformatics studies performed on four independent clinical cohorts (n: 146 ∼ 683 patients each) show that in all the cohorts both MYC and PIM1 are significantly enriched in the TN populations. MYC and PIM1 expression are correlated in three out of the four cohorts. Furthermore, we find that PIM1 overexpression is a poor prognostic factor associated with diminished recurrence-free survival. Our in vivo efficacy studies using conventional as well as novel “human-in-mouse” orthotopic xenograft models demonstrate that PIM1 is required for the growth of MYC-overexpressing human TN tumors. We are currently investigating the cellular mechanisms by which PIM kinase inhibition causes such growth defects. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B232. Citation Format: Dai Horiuchi, Alicia Y. Zhou, Alexandra N. Corella, Christina Yau, Devon A. Lawson, Alexey V. Bazarov, Paul Yaswen, Michael T. McManus, Alana L. Welm, Zena Werb, Andrei Goga. PIM1 kinase is essential for the growth of MYC-overexpressing triple-negative breast tumors and is an efficacious therapeutic target. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B232.