NSG-Pro mouse model for uncovering resistance mechanisms and unique vulnerabilities in human luminal breast cancers.

Most breast cancer deaths are caused by estrogen receptor-α­positive (ER+) disease. Preclinical progress is hampered by a shortage of therapy-naïve ER+ tumor models that recapitulate metastatic progression and clinically relevant therapy resistance. Human prolactin (hPRL) is a risk factor for primary and metastatic ER+ breast cancer. Because mouse prolactin fails to activate hPRL receptors, we developed a prolactin-humanized Nod-SCID-IL2Rγ (NSG) mouse (NSG-Pro) with physiological hPRL levels. Here, we show that NSG-Pro mice facilitate establishment of therapy-naïve, estrogen-dependent PDX tumors that progress to lethal metastatic disease. Preclinical trials provide first-in-mouse efficacy of pharmacological hPRL suppression on residual ER+ human breast cancer metastases and document divergent biology and drug responsiveness of tumors grown in NSG-Pro versus NSG mice. Oncogenomic analyses of PDX lines in NSG-Pro mice revealed clinically relevant therapy-resistance mechanisms and unexpected, potently actionable vulnerabilities such as DNA-repair aberrations. The NSG-Pro mouse unlocks previously inaccessible precision medicine approaches for ER+ breast cancers.

Loss of Nuclear Localized Parathyroid Hormone-Related Protein in Primary Breast Cancer Predicts Poor Clinical Outcome and Correlates with Suppressed Stat5 Signaling.

PURPOSE: Parathyroid hormone-related protein (PTHrP) is required for normal mammary gland development and biology. A PTHLH gene polymorphism is associated with breast cancer risk, and PTHrP promotes growth of osteolytic breast cancer bone metastases. Accordingly, current dogma holds that PTHrP is upregulated in malignant primary breast tumors, but solid evidence for this assumption is missing. EXPERIMENTAL DESIGN: We used quantitative IHC to measure PTHrP in normal and malignant breast epithelia, and correlated PTHrP levels in primary breast cancer with clinical outcome. RESULTS: PTHrP levels were markedly downregulated in malignant compared with normal breast epithelia. Moreover, low levels of nuclear localized PTHrP in cancer cells correlated with unfavorable clinical outcome in a test and a validation cohort of breast cancer treated at different institutions totaling nearly 800 cases. PTHrP mRNA levels in tumors of a third cohort of 737 patients corroborated this association, also after multivariable adjustment for standard clinicopathologic parameters. Breast cancer PTHrP levels correlated strongly with transcription factors Stat5a/b, which are established markers of favorable prognosis and key mediators of prolactin signaling. Prolactin stimulated PTHrP transcript and protein in breast cancer cell lines in vitro and in vivo, effects mediated by Stat5 through the P2 gene promoter, producing transcript AT6 encoding the PTHrP 1-173 isoform. Low levels of AT6, but not two alternative transcripts, correlated with poor clinical outcome. CONCLUSIONS: This study overturns the prevailing view that PTHrP is upregulated in primary breast cancers and identifies a direct prolactin-Stat5-PTHrP axis that is progressively lost in more aggressive tumors.

Prolactin-Stat5 signaling in breast cancer is potently disrupted by acidosis within the tumor microenvironment.

INTRODUCTION: Emerging evidence in estrogen receptor-positive breast cancer supports the notion that prolactin-Stat5 signaling promotes survival and maintenance of differentiated luminal cells, and loss of nuclear tyrosine phosphorylated Stat5 (Nuc-pYStat5) in clinical breast cancer is associated with increased risk of antiestrogen therapy failure. However, the molecular mechanisms underlying loss of Nuc-pYStat5 in breast cancer remain poorly defined. METHODS: We investigated whether moderate extracellular acidosis of pH 6.5 to 6.9 frequently observed in breast cancer inhibits prolactin-Stat5 signaling, using in vitro and in vivo experimental approaches combined with quantitative immunofluorescence protein analyses to interrogate archival breast cancer specimens. RESULTS: Moderate acidosis at pH 6.8 potently disrupted signaling by receptors for prolactin but not epidermal growth factor, oncostatin M, IGF1, FGF or growth hormone. In breast cancer specimens there was mutually exclusive expression of Nuc-pYStat5 and GLUT1, a glucose transporter upregulated in glycolysis-dependent carcinoma cells and an indirect marker of lactacidosis. Mutually exclusive expression of GLUT1 and Nuc-pYStat5 occurred globally or regionally within tumors, consistent with global or regional acidosis. All prolactin-induced signals and transcripts were suppressed by acidosis, and the acidosis effect was rapid and immediately reversible, supporting a mechanism of acidosis disruption of prolactin binding to receptor. T47D breast cancer xenotransplants in mice displayed variable acidosis (pH 6.5 to 6.9) and tumor regions with elevated GLUT1 displayed resistance to exogenous prolactin despite unaltered levels of prolactin receptors and Stat5. CONCLUSIONS: Moderate extracellular acidosis effectively blocks prolactin signaling in breast cancer. We propose that acidosis-induced prolactin resistance represents a previously unrecognized mechanism by which breast cancer cells may escape homeostatic control.

PTP1B suppresses prolactin activation of Stat5 in breast cancer cells.

Basal levels of nuclear localized, tyrosine phosphorylated Stat5 are present in healthy human breast epithelia. In contrast, Stat5 phosphorylation is frequently lost during breast cancer progression, a finding that correlates with loss of histological differentiation and poor patient prognosis. Identifying the mechanisms underlying loss of Stat5 phosphorylation could provide novel targets for breast cancer therapy. Pervanadate, a general tyrosine phosphatase inhibitor, revealed marked phosphatase regulation of Stat5 activity in breast cancer cells. Lentiviral-mediated shRNA allowed specific examination of the regulatory role of five tyrosine phosphatases (PTP1B, TC-PTP, SHP1, SHP2, and VHR), previously implicated in Stat5 regulation in various systems. Enhanced and sustained prolactin-induced Stat5 tyrosine phosphorylation was observed in T47D and MCF7 breast cancer cells selectively in response to PTP1B depletion. Conversely, PTP1B overexpression suppressed prolactin-induced Stat5 tyrosine phosphorylation. Furthermore, PTP1B knockdown increased Stat5 reporter gene activity. Mechanistically, PTP1B suppression of Stat5 phosphorylation was mediated, at least in part, through inhibitory dephosphorylation of the Stat5 tyrosine kinase, Jak2. PTP1B knockdown enhanced sensitivity of T47D cells to prolactin phosphorylation of Stat5 by reducing the EC(50) from 7.2 nmol/L to 2.5 nmol/L. Immunohistochemical analyses of two independent clinical breast cancer materials revealed significant negative correlations between levels of active Stat5 and PTP1B, but not TC-PTP. Collectively, our data implicate PTP1B as an important negative regulator of Stat5 phosphorylation in invasive breast cancer.

Prolactin inhibits BCL6 expression in breast cancer through a Stat5a-dependent mechanism.

BCL6 is a transcriptional repressor that recognizes DNA target sequences similar to those recognized by signal transducer and activator of transcriptions 5 (Stat5). BCL6 disrupts differentiation of breast epithelia, is downregulated during lactation, and is upregulated in poorly differentiated breast cancer. In contrast, Stat5a mediates prolactin-induced differentiation of mammary epithelia, and loss of Stat5 signaling in human breast cancer is associated with undifferentiated histology and poor prognosis. Here, we identify the mammary cell growth factor prolactin as a potent suppressor of BCL6 protein expression in human breast cancer through a mechanism that requires Stat5a, but not prolactin-activated Stat5b, MEK-ERK, or PI3K-AKT pathways. Prolactin rapidly suppressed BCL6 mRNA in T47D, MCF7, ZR75.1, and SKBr3 breast cancer cell lines, followed by prolonged reduction of BCL6 protein levels within 3 hours. Prolactin suppression of BCL6 was enhanced by overexpression of Stat5a but not Stat5b, was mimicked by constitutively active Stat5a, but did not require the transactivation domain of Stat5a. Stat5 chromatin immunoprecipitation demonstrated physical interaction with a BCL6 gene regulatory region, and BCL6 transcript repression required histone deacetylase activity based on sensitivity to trichostatin A. Functionally, BCL6 overexpression disrupted prolactin induction of Stat5 reporter genes. Prolactin suppression of BCL6 was extended to xenotransplant tumors in nude mice in vivo and to freshly isolated human breast cancer explants ex vivo. Quantitative immunohistochemistry revealed elevated BCL6 in high-grade and metastatic breast cancer compared with ductal carcinoma in situ and nonmalignant breast, and cellular BCL6 protein levels correlated negatively with nuclear Stat5a (r = -0.52; P < 0.001) but not with Stat5b. Loss of prolactin-Stat5a signaling and concomitant upregulation of BCL6 may represent a regulatory switch facilitating undifferentiated histology and poor prognosis of breast cancer.

Cyclin D1 splice variants: polymorphism, risk, and isoform-specific regulation in prostate cancer.

PURPOSE: Alternative CCND1 splicing results in cyclin D1b, which has specialized, protumorigenic functions in prostate not shared by the cyclin D1a (full length) isoform. Here, the frequency, tumor relevance, and mechanisms controlling cyclin D1b were challenged. EXPERIMENTAL DESIGN: First, relative expression of both cyclin D1 isoforms was determined in prostate adenocarcinomas. Second, relevance of the androgen axis was determined. Third, minigenes were created to interrogate the role of the G/A870 polymorphism (within the splice site), and findings were validated in primary tissue. Fourth, the effect of G/A870 on cancer risk was assessed in two large case-control studies. RESULTS: Cyclin D1b is induced in tumors, and a significant subset expressed this isoform in the absence of detectable cyclin D1a. Accordingly, the isoforms showed noncorrelated expression patterns, and hormone status did not alter splicing. Whereas G/A870 was not independently predictive of cancer risk, A870 predisposed for transcript-b production in cells and in normal prostate. The influence of A870 on overall transcript-b levels was relieved in tumors, indicating that aberrations in tumorigenesis likely alter the influence of the polymorphism. CONCLUSIONS: These studies reveal that cyclin D1b is specifically elevated in prostate tumorigenesis. Cyclin D1b expression patterns are distinct from that observed with cyclin D1a. The A870 allele predisposes for transcript-b production in a context-specific manner. Although A870 does not independently predict cancer risk, tumor cells can bypass the influence of the polymorphism. These findings have major implications for the analyses of D-cyclin function in the prostate and provide the foundation for future studies directed at identifying potential modifiers of the G/A870 polymorphism.

Insensitivity of human prolactin receptors to nonhuman prolactins: relevance for experimental modeling of prolactin receptor-expressing human cells.

Prolactin (PRL) receptors are expressed in a broad range of human cell types and in a majority of human breast and prostate cancers. Experimentally, normal and malignant human cells are typically cultured in vitro in media containing bovine PRL (bPRL) from fetal bovine serum or as xenotransplants in vivo in the presence of murine PRL (mPRL). The biological efficacy of bPRL toward hPRL receptors (hPRLR) is controversial, and hPRLR are insensitive to mPRL, but the mechanism is not known. To clarify limitations of current in vitro and in vivo experimental model systems for studies of hPRLR-expressing cells, we tested human and relevant subprimate prolactins in multiple hPRLR bioassays. bPRL and ovine PRL were 10-fold less potent hPRLR agonists than hPRL, although maximal responses at high ligand concentrations (efficacies) equaled that of hPRL. mPRL and rat PRL had greater than 50-fold lower potencies toward hPRLR than hPRL and had 50% reduced efficacies. In fact, mPRL and rat PRL were less effective hPRLR agonists than murine GH. Unexpectedly, mPRL was an effective competitive inhibitor of hPRL binding to hPRLR with an inhibitory constant of 1.3 nm and showed partial antagonist activity, suggesting reduced site-2 binding. Collectively, low bioactivities of bPRL and mPRL toward hPRLR suggest that existing laboratory cancer cell lines grown in 10% bovine serum-supplemented media or in mice are selected for growth under lactogen-depleted conditions. The biology and drug responsiveness of existing human cell lines may therefore not be representative of clinical cancers that are sensitive to circulating PRL.

Cyclin D1b is aberrantly regulated in response to therapeutic challenge and promotes resistance to estrogen antagonists.

Cyclin D1 is a key mediator of cell cycle progression that is aberrantly regulated in multiple cancers, especially in breast cancers. A number of studies have indicated that a polymorphism in a splice donor site in the cyclin D1 gene is associated with alternative splicing and the production of the alternative cyclin D1b transcript. Furthermore, this polymorphism is selectively associated with disease outcomes. However, relatively little is known regarding the protein product of the alternatively spliced message, cyclin D1b. Using antibodies specific for cyclin D1b, it was found that this protein is readily detectable in a number of cancer cell lines and primary breast cancers. Whereas cyclin D1b interacts with cyclin-dependent kinase 4 (CDK4), it is relatively inefficient at mediating RB phosphorylation and cell cycle progression in model systems due to the lack of exon 5 of cyclin D1-encoded sequences. However, cyclin D1b protein levels are not significantly attenuated by DNA damage or antiestrogen treatment, indicating that the protein may have significant effect on the response to such therapeutic modalities. Whereas enforced expression of cyclin D1b was not sufficient to abrogate DNA damage checkpoint responses, it did efficiently overcome cell cycle arrest mediated by antiestrogen therapeutics. This action of cyclin D1b was not associated with effects on estrogen receptor activity, but was rather dependent on functional association with CDK4. Combined, these studies indicate that the cyclin D1b protein is aberrantly regulated and could contribute to therapeutic failure in the context of ER-positive breast cancer.

Insulin receptor substrate-1 regulates the transformed phenotype of BT-20 human mammary cancer cells.

Although originating from a human breast cancer, BT-20 cells do not form colonies in soft agar. BT-20 cells do not express insulin receptor substrate-1 (IRS-1), which is known to promote both normal and abnormal growth and to inhibit differentiation. Stable expression of IRS-1 confers to BT-20 cells the ability to form colonies in soft agar. BT-20 cells form tumors in xenografts in mice, but the size of tumors is twice as large when the cells express IRS-1. The increased transformed phenotype is characterized by occupancy of the rDNA and cyclin D1 promoters by IRS-1 and the activation of the cyclin D1, c-myc, and rDNA promoters. In addition, the retinoblastoma protein, which is detectable in the rDNA promoter of quiescent BT-20/IRS-1 cells, is replaced by IRS-1 after insulin-like growth factor-I stimulation. Our results indicate that in BT-20 human mammary cancer cells, expression of IRS-1 activates promoters involved in cell growth and cell proliferation, resulting in a more transformed phenotype. Targeting of IRS-1 could be effective in inhibiting the proliferation of mammary cancer cells.

Ultrahigh density microarrays of solid samples.

We present a sectioning and bonding technology to make ultrahigh density microarrays of solid samples, cutting edge matrix assembly (CEMA). Maximized array density is achieved by a scaffold-free, self-supporting construction with rectangular array features that are incrementally scalable. This platform technology facilitates arrays of >10,000 tissue features on a standard glass slide, inclusion of unique sample identifiers for improved manual or automated tracking, and oriented arraying of stratified or polarized samples.

Phosphorylation of BATF regulates DNA binding: a novel mechanism for AP-1 (activator protein-1) regulation.

BATF is a member of the AP-1 (activator protein-1) family of bZIP (basic leucine zipper) transcription factors that form transcriptionally inhibitory, DNA binding heterodimers with Jun proteins. In the present study, we demonstrate that BATF is phosphorylated in vivo on multiple serine and threonine residues and at least one tyrosine residue. Reverse-polarity PAGE revealed that serine-43 and threonine-48 within the DNA binding domain of BATF are phosphorylated. To model phosphorylation of the BATF DNA binding domain, serine-43 was replaced by an aspartate residue. BATF(S43D) retains the ability to dimerize with Jun proteins in vitro and in vivo, and the BATF(S43D):Jun heterodimer localizes properly to the nucleus of cells. Interestingly, BATF(S43D) functions like wild-type BATF to reduce AP-1-mediated gene transcription, despite the observed inability of the BATF(S43D):Jun heterodimer to bind DNA. These data demonstrate that phosphorylation of serine-43 converts BATF from a DNA binding into a non-DNA binding inhibitor of AP-1 activity. Given that 40% of mammalian bZIP transcription factors contain a residue analogous to serine-43 of BATF in their DNA binding domains, the phosphorylation event described here represents a mechanism that is potentially applicable to the regulation of many bZIP proteins.