Live-cell invasive phenotyping uncovers the ALK2/BMP6 iron homeostasis pathway as a therapeutic vulnerability in LKB1-mutant lung cancer.

The acquisition of invasive properties is a prerequisite for tumor progression and metastasis. Molecular subtypes of KRAS-driven lung cancer exhibit distinct modes of invasion that likely contribute to unique growth properties and therapeutic susceptibilities. Despite this, pre-clinical discovery strategies designed to exploit invasive phenotypes are lacking. To address this, we designed an experimental system to screen for targetable signaling pathways linked to active early invasion phenotypes in the two most prominent molecular subtypes, TP53 and LKB1, of KRAS-driven lung adenocarcinoma (LUAD). By combining live-cell imaging of human bronchial epithelial cells in a 3D invasion matrix with RNA transcriptome profiling, we identified the LKB1-specific upregulation of bone morphogenetic protein 6 (BMP6). Examination of early-stage lung cancer patients confirmed upregulation of BMP6 in LKB1-mutant lung tumors. At the molecular level, we find that the canonical iron regulatory hormone Hepcidin is induced via BMP6 signaling upon LKB1 loss, where intact LKB1 kinase activity is necessary to maintain signaling homeostasis. Furthermore, pre-clinical studies in a novel Kras/Lkb1-mutant syngeneic mouse model show that potent growth suppression was achieved by inhibiting the ALK2/BMP6 signaling axis with single agents that are currently in clinical trials. We show that alterations in the iron homeostasis pathway are accompanied by simultaneous upregulation of ferroptosis protection proteins. Thus, LKB1 is sufficient to regulate both the 'gas' and 'breaks' to finely tune iron-regulated tumor progression.

Loss of the endocytic tumor suppressor HD-PTP phenocopies LKB1 and promotes RAS-driven oncogenesis.

Oncogenic RAS mutations drive aggressive cancers that are difficult to treat in the clinic, and while direct inhibition of the most common KRAS variant in lung adenocarcinoma (G12C) is undergoing clinical evaluation, a wide spectrum of oncogenic RAS variants together make up a large percentage of untargetable lung and GI cancers. Here we report that loss-of-function alterations (mutations and deep deletions) in the gene that encodes HD-PTP (PTPN23) occur in up to 14% of lung cancers in the ORIEN Avatar lung cancer cohort, associate with adenosquamous histology, and occur alongside an altered spectrum of KRAS alleles. Furthermore, we show that in publicly available early-stage NSCLC studies loss of HD-PTP is mutually exclusive with loss of LKB1, which suggests they restrict a common oncogenic pathway in early lung tumorigenesis. In support of this, knockdown of HD-PTP in RAS-transformed lung cancer cells is sufficient to promote FAK-dependent invasion. Lastly, knockdown of the Drosophila homolog of HD-PTP (dHD-PTP/Myopic) synergizes to promote RAS-dependent neoplastic progression. Our findings highlight a novel tumor suppressor that can restrict RAS-driven lung cancer oncogenesis and identify a targetable pathway for personalized therapeutic approaches for adenosquamous lung cancer.

Evaluating antitumor activity of Escherichia coli purine nucleoside phosphorylase against head and neck patient-derived xenografts.

BACKGROUND: Purine nucleoside phosphorylase (PNP) gene transfer represents a promising approach to treatment of head and neck malignancies. We tested recombinant adenovirus already in phase I/II clinical testing and leading-edge patient-derived xenografts (PDX) as a means to optimize this therapeutic strategy. METHODS: Our experiments investigated purine base cytotoxicity, PNP enzyme activity following treatment of malignant tissue, tumor mass regression, viral receptor studies, and transduction by tropism-modified adenovirus. RESULTS: Replication deficient vector efficiently transduced PDX cells and mediated significant anticancer effect following treatment with fludarabine phosphate in vivo. Either 6-methylpurine or 2-fluoroadenine (toxic molecules generated by the PNP approach) ablated head and neck cancer cell proliferation. High levels of adenovirus-3 specific receptors were detected in human tumor models, and vector was evaluated that utilizes this pathway. CONCLUSIONS: Our studies provide the scientific foundation necessary to improve PNP prodrug cleavage and advance a new treatment for head and neck cancer.

The conditional growth suppressor E7.25d.7 is an allele of the sterile-20 kinase misshapen.

The notion of a two-hit or multi-hit model of carcinogenesis dates to at least the 1970's and work done by Alfred Knudson. This concept was considered in the design and execution of a previous FLP/FRT screen in Drosophila melanogaster for conditional growth suppressors. During the course of this work, the lethal allele E7.25D.7 was identified as being of phenotypic interest. Here we report the genetic mapping of E7.25D.7, an allele of the sterile-20 kinase misshapen (msn).

The level of oncogenic Ras determines the malignant transformation of Lkb1 mutant tissue in vivo.

The genetic and metabolic heterogeneity of RAS-driven cancers has confounded therapeutic strategies in the clinic. To address this, rapid and genetically tractable animal models are needed that recapitulate the heterogeneity of RAS-driven cancers in vivo. Here, we generate a Drosophila melanogaster model of Ras/Lkb1 mutant carcinoma. We show that low-level expression of oncogenic Ras (RasLow) promotes the survival of Lkb1 mutant tissue, but results in autonomous cell cycle arrest and non-autonomous overgrowth of wild-type tissue. In contrast, high-level expression of oncogenic Ras (RasHigh) transforms Lkb1 mutant tissue resulting in lethal malignant tumors. Using simultaneous multiview light-sheet microcopy, we have characterized invasion phenotypes of Ras/Lkb1 tumors in living larvae. Our molecular analysis reveals sustained activation of the AMPK pathway in malignant Ras/Lkb1 tumors, and demonstrate the genetic and pharmacologic dependence of these tumors on CaMK-activated Ampk. We further show that LKB1 mutant human lung adenocarcinoma patients with high levels of oncogenic KRAS exhibit worse overall survival and increased AMPK activation. Our results suggest that high levels of oncogenic KRAS is a driving event in the malignant transformation of LKB1 mutant tissue, and uncovers a vulnerability that may be used to target this aggressive genetic subset of RAS-driven tumors.

YAP1 Expression in SCLC Defines a Distinct Subtype With T-cell-Inflamed Phenotype.

INTRODUCTION: The clinical and biological significance of the newly described SCLC subtypes, SCLC-A, SCLC-N, SCLC-Y, and SCLC-P, defined by the dominant expression of transcription factors ASCL1, NeuroD1, YAP1, and POU2F3, respectively, remain to be established. METHODS: We generated new RNA sequencing expression data from a discovery set of 59 archival tumor samples of neuroendocrine tumors and new protein expression data by immunohistochemistry in 99 SCLC cases. We validated the findings from this discovery set in two independent validation sets consisting of RNA sequencing data generated from 51 SCLC cell lines and 81 primary human SCLC samples. RESULTS: We successfully classified 71.8% of SCLC and 18.5% of carcinoid cases in our discovery set into one of the four SCLC subtypes. Gene set enrichment analysis for differentially expressed genes between the SCLC survival outliers (top and bottom deciles) matched for clinically relevant prognostic factors revealed substantial up-regulation of interferon-γ response genes in long-term survivors. The SCLC-Y subtype was associated with high expression of interferon-γ response genes, highest weighted score on a validated 18-gene T-cell-inflamed gene expression profile score, and high expression of HLA and T-cell receptor genes. YAP1 protein expression was more prevalent and more intensely expressed in limited-stage versus extensive-stage SCLC (30.6% versus 8.5%; p = 0.0058) indicating good prognosis for the SCLC-Y subtype. We replicated the inflamed phenotype of SCLC-Y in the two independent validation data sets from the SCLC cell lines and tumor samples. CONCLUSIONS: SCLC subtyping using transcriptional signaling holds clinical relevance with the inflamed phenotype associated with the SCLC-Y subset.

Leflunomide Suppresses the Growth of LKB1-Inactivated Tumors in the Immune-Competent Host and Attenuates Distant Cancer Metastasis.

Liver kinase B1 (LKB1)-inactivated tumors are vulnerable to the disruption of pyrimidine metabolism, and leflunomide emerges as a therapeutic candidate because its active metabolite, A77-1726, inhibits dihydroorotate dehydrogenase, which is essential for de novo pyrimidine biosynthesis. However, it is unclear whether leflunomide inhibits LKB1-inactivated tumors in vivo, and whether its inhibitory effect on the immune system will promote tumor growth. Here, we carried out a comprehensive analysis of leflunomide treatment in various LKB1-inactivated murine xenografts, patient-derived xenografts, and genetically engineered mouse models. We also generated a mouse tumor-derived cancer cell line, WRJ388, that could metastasize to the lung within a month after subcutaneous implantation in all animals. This model was used to assess the ability of leflunomide to control distant metastasis. Leflunomide treatment shrank a HeLa xenograft and attenuated the growth of an H460 xenograft, a patient-derived xenograft, and lung adenocarcinoma in the immune-competent genetically engineered mouse models. Interestingly, leflunomide suppressed tumor growth through at least three different mechanisms. It caused apoptosis in HeLa cells, induced G1 cell-cycle arrest in H460 cells, and promoted S-phase cell-cycle arrest in WRJ388 cells. Finally, leflunomide treatment prevented lung metastasis in 78% of the animals in our novel lung cancer metastasis model. In combination, these results demonstrated that leflunomide utilizes different pathways to suppress the growth of LKB1-inactivated tumors, and it also prevents cancer metastasis at distant sites. Therefore, leflunomide should be evaluated as a therapeutic agent for tumors with LKB1 inactivation.

The use of Trichomonas vaginalis purine nucleoside phosphorylase to activate fludarabine in the treatment of solid tumors.

Treatment with fludarabine phosphate (9-ß-D-arabinofuranosyl-2-F-adenine 5'-phosphate, F-araAMP) leads to regressions and cures of human tumor xenografts that express Escherichia coli purine nucleoside phosphorylase (EcPNP). This occurs despite the fact that fludarabine (F-araA) is a relatively poor substrate for EcPNP, and is cleaved to liberate 2-fluoroadenine at a rate only 0.3% that of the natural E. coli PNP substrate, adenosine. In this study, we investigated a panel of naturally occurring PNPs to identify more efficient enzymes that may be suitable for metabolizing F-araA as part of experimental cancer therapy. We show that Trichomonas vaginalis PNP (TvPNP) cleaves F-araA with a catalytic efficiency 25-fold greater than the prototypic E. coli enzyme. Cellular extracts from human glioma cells (D54) transduced with lentivirus stably expressing TvPNP (D54/TvPNP) were found to cleave F-araA at a rate similar to extracts from D54 cells expressing EcPNP, although much less enzyme was expressed per cell in the TvPNP transduced condition. As a test of safety and efficacy using TvPNP, human head and neck squamous cell carcinoma (FaDu) xenografts expressing TvPNP were studied in nude mice and shown to exhibit robust tumor regressions, albeit with partial weight loss that resolved post-therapy. F-araAMP was also a very effective treatment for mice bearing D54/TvPNP xenografts in which approximately 10% of tumor cells expressed the enzyme, indicating pronounced ability to kill non-transduced tumor cells (high bystander activity). Moreover, F-araAMP demonstrated activity against D54 tumors injected with an E1, E3 deleted adenoviral vector encoding TvPNP. In that setting, despite higher F-araA cleavage activity using TvPNP, tumor responses were similar to those obtained with EcPNP, indicating factors other than F-Ade production may limit regressions of the D54 murine xenograft model. Our results establish that TvPNP is a favorable enzyme for activating F-araA, and support further studies in combination with F-araAMP for difficult-to-treat human cancers.

Vimentin Is Required for Lung Adenocarcinoma Metastasis via Heterotypic Tumor Cell-Cancer-Associated Fibroblast Interactions during Collective Invasion.

Purpose: Vimentin is an epithelial-to-mesenchymal transition (EMT) biomarker and intermediate filament protein that functions during cell migration to maintain structure and motility. Despite the abundance of clinical data linking vimentin to poor patient outcome, it is unclear if vimentin is required for metastasis or is a correlative biomarker. We developed a novel genetically engineered mouse model (GEMM) to probe vimentin in lung adenocarcinoma metastasis.Experimental Design: We used the LSL-KrasG12D/Lkb1fl/fl/Vim-/- model (KLV-/-), which incorporates a whole-body knockout of vimentin and is derived from the Cre-dependent LSL-KrasG12D/Lkb1fl/fl model (KLV+/+). We compared the metastatic phenotypes of the GEMMs and analyzed primary tumors from the KLV models and lung adenocarcinoma patients to assess vimentin expression and function.Results: Characterization of KLV+/+ and KLV-/- mice shows that although vimentin is not required for primary lung tumor growth, vimentin is required for metastasis, and vimentin loss generates lower grade primary tumors. Interestingly, in the KLV+/+ mice, vimentin was not expressed in tumor cells but in cancer-associated fibroblasts (CAFs) surrounding collective invasion packs (CIPs) of epithelial tumor cells, with significantly less CIPs in KLV-/- mice. CIPs correlate with tumor grade and are vimentin-negative and E-cadherin-positive, indicating a lack of cancer cell EMT. A similar heterotypic staining pattern was observed in human lung adenocarcinoma samples. In vitro studies show that vimentin is required for CAF motility to lead tumor cell invasion, supporting a vimentin-dependent model of collective invasion.Conclusions: These data show that vimentin is required for lung adenocarcinoma metastasis by maintaining heterotypic tumor cell-CAF interactions during collective invasion. Clin Cancer Res; 24(2); 420-32. ©2017 AACR.

Modulation of Bax and mTOR for Cancer Therapeutics.

A rationale exists for pharmacologic manipulation of the serine (S)184 phosphorylation site of the proapoptotic Bcl2 family member Bax as an anticancer strategy. Here, we report the refinement of the Bax agonist SMBA1 to generate CYD-2-11, which has characteristics of a suitable clinical lead compound. CYD-2-11 targeted the structural pocket proximal to S184 in the C-terminal region of Bax, directly activating its proapoptotic activity by inducing a conformational change enabling formation of Bax homooligomers in mitochondrial membranes. In murine models of small-cell and non-small cell lung cancers, including patient-derived xenograft and the genetically engineered mutant KRAS-driven lung cancer models, CYD-2-11 suppressed malignant growth without evident significant toxicity to normal tissues. In lung cancer patients treated with mTOR inhibitor RAD001, we observed enhanced S184 Bax phosphorylation in lung cancer cells and tissues that inactivates the propaoptotic function of Bax, contributing to rapalog resistance. Combined treatment of CYD-2-11 and RAD001 in murine lung cancer models displayed strong synergistic activity and overcame rapalog resistance in vitro and in vivo Taken together, our findings provide preclinical evidence for a pharmacologic combination of Bax activation and mTOR inhibition as a rational strategy to improve lung cancer treatment. Cancer Res; 77(11); 3001-12. ©2017 AACR.

Targeting adhesion signaling in KRAS, LKB1 mutant lung adenocarcinoma.

Loss of LKB1 activity is prevalent in KRAS mutant lung adenocarcinoma and promotes aggressive and treatment-resistant tumors. Previous studies have shown that LKB1 is a negative regulator of the focal adhesion kinase (FAK), but in vivo studies testing the efficacy of FAK inhibition in LKB1 mutant cancers are lacking. Here, we took a pharmacologic approach to show that FAK inhibition is an effective early-treatment strategy for this high-risk molecular subtype. We established a lenti-Cre-induced Kras and Lkb1 mutant genetically engineered mouse model (KLLenti) that develops 100% lung adenocarcinoma and showed that high spatiotemporal FAK activation occurs in collective invasive cells that are surrounded by high levels of collagen. Modeling invasion in 3D, loss of Lkb1, but not p53, was sufficient to drive collective invasion and collagen alignment that was highly sensitive to FAK inhibition. Treatment of early, stage-matched KLLenti tumors with FAK inhibitor monotherapy resulted in a striking effect on tumor progression, invasion, and tumor-associated collagen. Chronic treatment extended survival and impeded local lymph node spread. Lastly, we identified focally upregulated FAK and collagen-associated collective invasion in KRAS and LKB1 comutated human lung adenocarcinoma patients. Our results suggest that patients with LKB1 mutant tumors should be stratified for early treatment with FAK inhibitors.

LKB1 represses focal adhesion kinase (FAK) signaling via a FAK-LKB1 complex to regulate FAK site maturation and directional persistence.

Liver kinase ß1 (LKB1, also known as STK11) is a serine/threonine kinase that has multiple cellular functions including the regulation of cell polarity and motility. Murine proteomic studies show that LKB1 loss causes aberrant adhesion signaling; however, the mechanistic underpinnings of this relationship are unknown. We show that cells stably depleted of LKB1 or its co-activator STRADα have increased phosphorylation of focal adhesion kinase (FAK) at Tyr(397)/Tyr(861) and enhanced adhesion to fibronectin. LKB1 associates in a complex with FAK and LKB1 accumulation at the cellular leading edge is mutually excluded from regions of activated Tyr(397)-FAK. LKB1-compromised cells lack directional persistence compared with wild-type cells, but this is restored through subsequent pharmacological FAK inhibition or depletion, showing that cell directionality is mediated through LKB1-FAK signaling. Live cell confocal imaging reveals that LKB1-compromised cells lack normal FAK site maturation and turnover, suggesting that defects in adhesion and directional persistence are caused by aberrant adhesion dynamics. Furthermore, re-expression of full-length wild-type or the LKB1 N-terminal domain repressed FAK activity, whereas the kinase domain or C-terminal domain alone did not, indicating that FAK suppression is potentially regulated through the LKB1 N-terminal domain. Based upon these results, we conclude that LKB1 serves as a FAK repressor to stabilize focal adhesion sites, and when LKB1 function is compromised, aberrant FAK signaling ensues, resulting in rapid FAK site maturation and poor directional persistence.