The tumor suppressor TP53 is the most frequently altered gene in human cancers. The growth-promoting complex, mTORC1 plays a part of the oncogenic profile caused by dysfunctional p53. mTORC1 sits downstream of AMPK and other crucial tumor suppressors/oncogenes, PTEN, LKB1, and Akt. The antifolate pemetrexed was found by this laboratory to activate AMPK via the inhibition of the enzyme AICART in de novo purine synthesis. This work presents a mechanism of mTORC1 activation with p53 loss, as well as of mTORC1 inhibition by pemetrexed-induced AMPK. We have found that mTORC1 activity was substantially upregulated by the loss or mutation of p53. This activation involves the loss of TSC2 from lysosomal membranes, the site of mTORC1 activation by Rheb. We demonstrate that loss of lysosomal TSC2 increased the levels of lysosomal Rheb. Control of mTORC1 was restored by overexpression of TSC2, which correlated with decreased lysosomal Rheb. Surprisingly, pemetrexed-activated AMPK did not phosphorylate TSC2 because of an accumulation of nonfunctional p53, and a subsequent decrease in TSC2 mRNA. Accordingly, lysosomal TSC2 decreased, however, the levels of lysosomal Rheb decreased. Future studies will question whether the robust Raptor phosphorylation by pemetrexed is involved in this decrease in lysosomal Rheb. AMPK activation by pemetrexed also significantly increased the translocation of AMPK to the nucleus, and we will explore the function of this nuclear AMPK. Overall, these findings present a mechanism involved in the oncogenic signaling of mTORC1 with loss of p53 and offer insight into how pemetrexed reinstates control.

The cancers of liver, colon and breast are amongst the top five most prevalent and most fatal worldwide. As the Raf/MEK/ERK pathway is frequently deregulated in hepatocellular carcinoma (HCC), sorafenib, a Raf kinase inhibitor, became the first systemic therapy approved for the treatment of patients with HCC. However, sorafenib only produced modest effects with low response rates in the clinic. Similarly, regorafenib, which was approved for the treatment of metastatic colorectal cancer (CRC), has had a poor response rate in the clinic. Since phosphodiesterase type 5 has been reported to be overexpressed in HCC and CRC, we hypothesized that sildenafil, a phosphodiesterase type 5 inhibitor, could enhance the toxicities of sorafenib and regorafenib in HCC and CRC cells, respectively. Our in vitro data indicated that the drugs interacted strongly to kill cancer cells via induction of ER stress, autophagy and apoptosis. In accordance with these findings, our in vivo data demonstrated a significant reduction in tumor growth. The second study in this manuscript was conducted based on the growing body of evidence about the significant contribution of EGFR and JAK/STAT signaling to the breast tumorigenesis. Our preliminary in vitro data demonstrated that the concurrent inhibition of these two pathways by lapatinib, a dual ERBB1/2 inhibitor, and ruxolitinib, a JAK1/2 inhibitor, synergistically killed breast cancer cells of all types, including the resistant triple negative subtype. Our mechanistic studies showed that the combination of ruxolitinib and lapatinib triggered cytotoxic mitophagy, and autophagy-dependent activation of BAX and BAK leading to the mitochondrial dysfunction.

The paralogous C-terminal binding proteins (CtBP) 1 and 2 are evolutionarily conserved transcriptional coregulators that target and disrupt the expression of several genes essential for multiple cellular processes critical to regulating tumor formation. CtBP’s ability to govern the transcription of genes necessary for apoptosis, tumor suppression, invasion/migration and EMT gives rise to its oncogenic activities. Both isoforms of CtBP are found to be overexpressed in cancers including colorectal, pancreatic, ovarian, and breast, with higher levels correlating to lower overall median survival. Although multiple lines of evidence suggest CtBP plays a role in tumorigenesis, it has never been formally characterized as an oncogene. For this reason, the goal of this dissertation was to design a set of experiments to determine the transforming ability of CtBP2 in vitro using both murine and human fibroblast and in vivo using the Apcmin/+ mouse model of cancer. Specifically, we demonstrate that overexpression of CtBP2 alone can drive transformation of NIH3T3 cells leading to loss of contact inhibition, increased x invasion/migration, and anchorage independent growth. In addition, CtBP2 was found to cooperate with the large T-antigen (LT) component of the simian virus 40 (SV40) to lead to transformation of murine embryonic fibroblasts (MEFs) and with both LT and small T-antigen (ST) to induce migration/invasion and anchorage-independent growth in BJ human foreskin fibroblasts. To confirm the role of Ctbp2 in a mouse tumor model with Ctbp overexpression, we bred Apcmin/+ mice to Ctbp2 heterozygous (Ctbp2+/-) mice, which otherwise live normal lifespans. CtBP is a known target of the APC tumor suppressor and is thus stabilized in APC mutated human colon cancers and is found in high levels in Apcmin/+ polyps. Remarkably, removing an allele of Ctbp2 doubled the median survival of Apcmin/+ mice (P <0.001) and reduced polyp formation to near undetectable levels. These data suggest the importance of CtBP2 in driving cellular transformation and identify it as a potential target for prevention or therapy in APC mutant backgrounds.

Cell migration is an important component of many physiological and pathological processes such as tissue and organ morphogenesis during development, wound healing, inflammatory immune response, and tumor metastasis. The actin cytoskeleton is the basic engine driving cell migration. In the present study, we elucidate the role of an important actin interacting proteins, Adducins, in motility of normal lung epithelium and lung cancer cells. Adducins are the family of cytoskeleton protein capping the fast growing end and facilitating the bundling of actin filaments. Adducins are encoded by the three closely related genes namely alpha (ADD1), beta (ADD2) and gamma (ADD3) Adducin. ADD1 and ADD3 are ubiquitously expressed, whereas ADD2 is most abundant in brain and erythrocytes. Adducins are also involved in recruiting spectrin to the actin filaments forming spectrin-actin membrane skeletal network. Its role in cell motility remains controversial. In this study, we observed that CRISPR/Cas9 mediated stable knockout of ADD1 and ADD3 in 16HBE normal lung epithelium cells significantly increases transfilter migration of cells. On the other hand, stable overexpression of ADD1 in H1299 Non-Small Cell lung cancer cells significantly decreases wound healing, transfilter migration and Matrigel invasion of the cells. Importantly, the effects of Adducin depletion and overexpression on cell motility were not due to altered cell proliferation. ADD1 overexpressed H1299 cells were characterized by the increased adhesion and spreading on the collagen matrix. Fluorescence microscopy revealed alterations in their cortical actin cytoskeleton that was manifested in the assembly of peripheral F-actin bundles and formation of filopodia-like protrusions. These findings suggest that Adducins are negative regulators of motility of normal lung epithelial and lung cancer cells that act by altering the architecture of submembranous actin cytoskeleton and modulating cell adhesion to the extracellular matrix.

Inhibitor of differentiation 4 (ID4), a member of the helix-loop-helix family of transcriptional regulators has emerged as a tumor suppressor in prostate cancer (PCa). Recent studies have shown that Id4 is highly expressed in the normal prostate and decreases in prostate cancer due to epigenetic silencing. Id4 knockdown in androgen sensitive LNCaP cells has been shown to lead to castration resistant prostate cancer (CRPC) in vitro and in vivo. Id4-/- mice leads to underdeveloped prostate with PIN like lesions without the loss of Androgen Receptor (AR) expression. In this study we demonstrate that the loss of ID4 expression in PCa cell line LNCaP and DU145 may promote tumorigenesis by promoting stemness. LNCaP cells with stably silenced ID4 ((-)ID4) using retroviral based shRNA and LNCaP transfected with non-specific shRNA were used to perform colony forming assay and prostatosphere formation using matrigel. Expression of cancer stem cell markers was determined using western blotting and immunocytochemistry (ICC). FACS analysis was used to sort stem cells and determine the ID4 expression. Xenograft study was performed on SCID mice using CD133 positive LNCaP cells. LNCaP(-)ID4 and DU145 cells lacking ID4 showed increased holoclone as well as decreased paraclone formation, which are believed to be derived from stem cells and differentiated cells respectively, as compared to non-silencing control in the colony forming assay. There was also an increase in prostatosphere development in the LNCaP (-) ID4 cells indicating that the loss of ID4 is responsible for promoting the LNCaP cells towards cancer stem cells. The results were further validated via western blotting and ICC using known cancer stem cell markers on the holoclones and paraclones formed by these cells. Xenograft study showed that 10,000 cells from CD133 positive LNCaP cells developed tumor on SCID mice. This study reports for the first time that loss of ID4 increases holoclone and prostatosphere formation indicating that Id4 may contribute to promoting stemness in prostate cancer cells.

Flavonoids are polyphenolic secondary metabolites found in plants that have bioactive properties including antiviral, antioxidant, and anticancer. Two isomeric flavone were extracted from Gnaphalium elegans and Achyrocline bogotensis, plants used by the people from the Andean region of South America as remedies for cancer. 5,7-dihydroxy-3,6,8-trimethoxy-2-phenyl-4H-chromen-4-one (5, 7–dihydroxy- 3, 6, 8 trimethoxy flavone/ flavone A) and 3,5-dihydroxy-6,7,8-trimethoxy-2-phenyl-4H-chromen-4-one (3, 5–dihydroxy-6, 7, 8–trimethoxy flavone/ flavone B) have shown antineoplastic activity against colon cancer cell lines dependent upon their differentiation status. Pharmacokinetic studies reported herein were used to determine dosing for antitumor assays, as well as determine target tissue concentration. These included the development of methods to extract the flavones from plasma or colon tissue and reverse phase high performance liquid chromatography methods for quantification. Quantification methods were linear (r2 ≥ 0.99) with plasma calibration curves ranging from 250 - 2,500 ng/mL and 2,500 - 100,000 ng/mL for both flavones and colon calibration curves ranging from 250 – 100,000 ng/g (flavone A) and 1,000-25,000 ng/g (flavone B). Intravenous administration of a 20 mg/kg dose in rats yielded half-lives of 83.68 ± 56.61 and 107.45 ± 53.31 minutes with clearance values of 12.99 ± 13.78 and 80.79 ± 35.06 mL/min/kg for flavones A and B, respectively. Analysis of colon tissue yielded concentrations of 1639 ± 601 ng/g (flavone A) and 5975 ± 2480 ng/g (flavone B), suggesting both may be good candidate for individual or adjunct therapy for colon cancer due to distribution to the target tissue. Preliminary studies in colon cancer cells CaCo 2 and HCT 116 using either flavone in combination with 5-fluorouracil (5-FU) suggested synergistic activity of these compounds. The combination treatment increased induction of apoptosis by enhancing the DNA damaging mechanism of 5-FU. In vivo, preliminary xenograft experiments using HCT 116 cells showed smaller tumors in mice dosed with flavone B as compared to the 5-FU or combination treatment. Further experiments are warranted to confirm these observations.

Phylloquinone is a natural analog of vitamin K that has been shown to both inhibit cancer cell growth and induce apoptosis in several cancer cell lines. This study examined these effects in a non-Hodgkin lymphoma cell line, known as U937. Cell growth inhibition and apoptosis were assessed through the quantification of cell density and area, following treatment with several concentrations of phylloquinone. In addition, apoptosis was detected and quantified using immunofluorescent markers of apoptosis (i.e. annexin V, APO-BrdU). Treatment with phylloquinone resulted in reduced overall cell density, increased overall cell area, and an increased frequency of apoptosis in U937 cells. Increasing both phylloquinone concentration and treatment time enhanced these effects. These results are significant because they document the anti-cancer effects of this analog of vitamin K, as well as provide insight into the morphological changes that occur during apoptosis in U937 cells.

The FDA-approved platinum compound, cisplatin, is commonly used as a chemotherapy drug to treat many forms of cancer. However, this compound also has several associated side-effects, including ototoxicity. This has made the development of novel platinum compounds that reduce cancer cell viability, while causing fewer and milder side-effects, an area of significant research interest. In the present study, we examined the apoptotic effects that four monofunctional platinum compounds, pyriplatin, phenanthriplatin, Pt(diethylenetriamine)Cl, and Pt(N,Ndiethyldiethylenetriamine) Cl, had on zebrafish inner ear auditory epithelial cells. We then compared the apoptotic effects of these compounds to those of cisplatin, which is a bifunctional platinum compound. Our hypothesis was that the four monofunctional platinum compounds would cause reduced inner ear apoptosis compared to cisplatin. Zebrafish were injected with either vehicle solution, cisplatin or with one of the monofunctional compounds. Later, at 24-hour and 48-hour time points, the zebrafish were euthanized, and two of their auditory inner ear endorgans, the utricle and saccule, were dissected out. A terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labeling (TUNEL) assay was then used to label apoptotic cells, and the inner ear organs were viewed under a microscope. The number of apoptotic cells on each sample was quantified and the data were analyzed for significant differences between treatments. We found that with the exception of pyriplatin in the saccules, and with the exception of pyriplatin, Pt(N,N-diethyldiethylenetriamine)Cl, and phenanthriplatin in the utricles, the monofunctional compounds and cisplatin did not induce apoptosis in the inner ear of zebrafish at either time point. Based on these results we conclude that the monofunctional platinum compounds largely do not induce zebrafish inner ear apoptosis and if they were to produce ototoxicity, it would not be through an apoptotic mechanism.

Tumor metastasis currently presents the greatest obstacle for effective cancer remediation. Metastatic growth necessitates both degradation of a specialized form of extracellular matrix (ECM) known as the basement membrane (BM) and the invasion of surrounding tissues thereafter. The thoracic air sacs of fruit flies (Drosophila melanogaster), which develop and operate in a fashion comparable to the human lung, provide a unique model for identifying and characterizing factors that contribute to its own development as well as tumoral invasion. We investigated the involvement of both Janus kinase (JAK)/Signal transducer and activator of transcription (STAT) signaling and a BMassociated protein during the development of air sac primordia (ASPs), the precursors to Drosophila air sacs. We find that JAK/STAT signaling occurs in ASP tip cells and that misexpression of core pathway components via the GAL4/UAS system negatively impacts ASP development. Further, we identify Unpaired 2 (Upd2) as the primary activating ligand for JAK/STAT activity in the ASP. Knockdown of the BM-associated protein using GAL4 drivers associated with a fibroblast growth factor (FGF) receptor gene, breathless (btl), and segment polarity gene, patched (ptc), prevented larval development beyond the second larval instar (L2). Knockdown of the BM-associated protein in the wing also produced bristle defects, but its overexpression did not have an effect anywhere other than in the ASP, where the proportion of mutant phenotypes increased significantly (p < .0001) in response. Finally, we find that collagen IV localization was unaffected by knockdown of the BM-associated protein. Together, our data constitute a significant step forward in understanding the role of both this BM-associated protein and JAK/STAT signaling in the ASP and similar mammalian structures.

Chronic Lymphocytic Leukemia (CLL) is defined by the accumulation of clonally expanded CD5+ and CD19+ B lymphocytes in blood and secondary lymphoid organs with impaired apoptotic mechanisms. CLL represents one third of all leukemia cases with an average age of 72 years at diagnosis making it the most common adult leukemia. The Eµ-Tcl1 mouse serves as an excellent model to study the development of CLL as they progress to a CLL like disease by 9-14 months of age, due to overexpression of an oncogene, T cell Leukemia 1(Tcl1), specifically in B cells through the Ig VH promoter and Eµ enhancer (Bichi et al. PNAS. 2002). In an adoptive transfer model, intravenous or intraperitoneal injection of primary CD5+CD19+ CLL cells from the Eµ-Tcl1 CLL mouse into recipient syngeneic mice leads to the development of a CLL like disease within 3-8 weeks of transfer. We have characterized the growth of CLL cells in these mice by periodic submandibular bleeding, spleen ultrasonography and flow cytometry. We find that Eµ-Tcl1 CLL cells express more Prostate apoptosis response-4 protein (Par-4), a known pro-apoptotic tumor suppressor protein, than normal B-1 or B-2 cells in mice. Par-4 is silenced by promoter methylation in more than 30% of all cancers and has been shown to be secreted and to induce apoptosis selectively in various types of cancer cells but not in normal cells. We found that CLL cells have constitutively active B-cell receptor signaling (BCR) and that inhibition of BCR signaling with FDA approved drugs causes a decrease in Par-4 protein, mRNA levels, and an increase in apoptosis. In particular, activities of Src family kinases, spleen tyrosine kinase and Bruton’s tyrosine kinase are required for Par-4 expression in CLL cells, suggesting a novel regulation of Par-4 through BCR signaling in both Eµ-Tcl1 CLL cells and primary human CLL samples. Consistent with this, lenti-viral shRNA mediated knockdown of Lyn kinase leads to a decrease in Par-4 expression in MEC-1 cells, a human CLL derived cell line. Igα (CD79a) silencing in primary human CLL cells also results in down regulation of Par-4 expression. Additionally, we knocked down expression of Par-4 in MEC-1 cells which resulted in a decrease in cell growth that could be attributed to an increase in p21 expression and a reduction in the G1/S cell cycle transition. We have also observed this phenomenon by crossing mice deficient in Par-4 with the Eµ-Tcl1 mouse where lack of Par-4 delays CLL growth in the mouse significantly (time to euthanization due to poor body condition - Eµ-Tcl1: 8.9mo vs Par4-/-EµTcl1: 11.97 mo, p = 0.0472) and splenic B-CLL cells from these mice also have increased expression of p21. Since mice in this cohort are whole body knockout for Par-4, the difference in survival times between the Par-4 +ve and Par-4 –ve EµTcl1 mice could be due to the influence of Par-4 on CLL cells as well as the effect of Par-4 secreted by the CLL cells on the microenvironment. There could be other potential roles for Par-4 in the context of CLL which are under further investigation. We have also investigated the site of CLL growth in mouse models to determine that the spleen is the primary organ to accumulate the CLL tumor burden. We have found that splenectomy significantly delays the development of CLL in the primary Eμ-Tcl1 mouse model and prevents growth and development in the adoptive transfer model. Interestingly, splenectomy did not delay CLL development as significantly in animals deficient for Par-4 compared to C57BL/6 wild type mice. Par-4 appears to regulate a specific microenvironment required for CLL growth. Current studies are investigating the role of Par-4 in the microenvironment and the cell types that are critical for CLL growth within the splenic niche.