Michael Cox

Associate Professor

Relevant Degree Programs

Affiliations to Research Centres, Institutes & Clusters


Graduate Student Supervision

Doctoral Student Supervision

Dissertations completed in 2010 or later are listed below. Please note that there is a 6-12 month delay to add the latest dissertations.

Cholesterol metabolism as a target in castration-resistant prostate cancer (2018)

Despite clinical benefits of existing prostate cancer treatments, patients continue to develop therapeutic resistance. Persistence of androgen receptor pathway activity is attributed to several mechanisms associated with resistance, including intratumoral androgen receptor agonist synthesis from the precursor cholesterol. Cholesterol has been correlated to poor outcomes in patients and clinically the use of cholesterol synthesis inhibitors, statins, improves prostate cancer survival. The expression of the high-density lipoprotein-cholesterol receptor, scavenger receptor B1 (SR-B1), is elevated in castration-resistant prostate cancer models and has been linked to poor survival of patients. The overarching hypothesis of this thesis is that cholesterol modulation, through either synthesis or uptake inhibition, will impact essential signaling processes impeding the proliferation of prostate cancer. Clinically statin use was found to improve the overall survival of metastatic castration-resistant prostate cancer patients receiving the androgen synthesis inhibitor abiraterone. In vivo experiments demonstrated the ability of statins to impede post-castration biochemical recurrence and reduce tumor growth and androgen receptor agonist synthesis in LNCaP-derived xenograft tumors. SR-B1 was found to be overexpressed in clinical samples from both local and metastatic prostate cancer. Antagonism of SR-B1 in steroid responsive C4-2 cells decreased cholesterol uptake and growth and induced cell cycle arrest. Initially attributed to an observed decrease in de novo steroid synthesis and androgen receptor activity, the inability of exogenous steroid to restore cellular proliferation or androgen receptor activity indicated steroid independent cellular arrest. As such, cellular stress and nutrient deprivation responses were assessed and SR-B1 antagonism was found to induce both autophagy and endoplasmic reticulum stress markers. Given the steroid-independent manner of SR-B1 antagonism mediated cellular arrest, the effects of SR-B1 antagonism on androgen independent PC-3 cells was assessed and found to result in robust cellular death in vitro and decreased growth of xenograft tumors. These findings demonstrate that the reduction of cellular cholesterol availability can impede prostate cancer proliferation through both decreased steroid-synthesis and steroid-independent mechanisms providing a potential therapeutic target for the treatment of prostate cancer.

View record

Inhibition of Gli1 and Gli2 as a targeted therapy for bladder cancer (2018)

The sonic hedgehog (SHH) signaling pathway has been shown to play an integral role in the maintenance and progression of bladder cancer (BCa). Smoothened inhibitors are currently used in the clinic for treatment of some skin cancers, however they have not been evaluated in BCa and SHH inhibition may be an efficacious strategy for BCa treatment. I assessed an in-house human BCa tissue microarray comprising non-invasive, invasive and lymph node metastasized transitional cell carcinoma and found that the transcription factors downstream of SHH, Gli1 and Gli2, were increased in more aggressive tumors. A panel of BCa cell lines show that two invasive lines, UM-UC-3 and 253J-BV, both express these transcription factors but differ in other parameters in the SHH pathway. UM-UC-3 produces greater quantities of SHH ligand, is less responsive in viability to pathway stimulation by recombinant human SHH or SAG, and less responsive to inhibition by a variety of molecules including the Smoothened inhibitors cyclopamine and SANT-1. 253J-BV, on the other hand, was highly responsive to these manipulations and appears more representative of canonical SHH signaling while UM-UC-3 resembles non-canonical autocrine signaling. To overcome this variability I utilized a Gli1 and Gli2 antisense oligonucleotide (ASO) to bypass pathway mechanics and target the transcription factors directly. UM-UC-3 decreased in viability due to both ASOs but 253J-BV was only affected by Gli2 ASO. IC50s were in the nanomolar range. To evaluate in vivo efficacy I developed a murine intravesical orthotopic human bladder cancer (mio-hBC) model for the establishment of non-invasive urothelial cell carcinomas. In this model I pre-treat the bladder with poly-L-lysine for 15 minutes, followed by intravesical instillation of luciferase–transfected human UM-UC-3 cells. Cancer cells are quantified by bioluminescent imaging. Tumors grew to 541.6±0.75 fold (Mean±SE) initial size after 40 days and were confirmed to reflect patient samples by a response to mitomycin C. Treatment of these tumors with Gli2 ASO resulted in decreased tumor size, growth rate and Gli2 mRNA and protein expression. These results validate this model and support the conclusion that Gli2 ASO may be a promising new targeted therapy for BCa.

View record

Hedgehog Pathway in Prostate Cancer Progression: A Novel Therapeutic Approach to Ongoing Adaptive Responses (2016)

Prostate cancer is the most commonly diagnosed cancer in North American men. While ever-improved androgen ablation therapies prolong life in men with advanced disease, remissions are temporary. Identifying targetable pathways underlying castration-resistant progression is essential for improving survival of patients with advanced disease. Hedgehog signalling may be one such pathway. The central hypothesis of this thesis examines the importance of adaptive responses during progression of prostate cancer to castration resistance and subsequent metastasis by reactivating developmental cues. We demonstrate that key Hedgehog pathway regulatory proteins are elevated in advanced prostate cancer compared to benign and untreated. This over-expression of Hedgehog signalling may be an adaptive response, which leads to a transition from a paracrine to autocrine Hedgehog signalling modality. Inhibition of this pathway disrupts progression of androgen sensitive, LNCaP tumours to castration-resistant prostate cancer. This inhibition has no effect on viability or key Hedgehog pathway regulators in vitro and no change in tumour specific Hedgehog genes in vivo, yet by examining host specific murine GLI1, 2 and PTCH1, key regulators of the Hedgehog pathway, we observe significant inhibition. These results suggest that Hedgehog expression promotes castration-resistant prostate cancer progression through reciprocal paracrine signalling within the tumour microenvironment. However, inhibiting the Hh pathway in our castration-resistant prostate cancer models induced down-regulation of the downstream Hh targets, in vitro. This inhibition hinders the migration and invasion of these cell lines in vitro and abrogates their invasive potential in vivo. These results taken with data from our prostate cancer patient cohort that show a significantly higher recurrence rate for patients with elevated SHH and GLI2 levels suggests a higher metastatic capacity driven by ligand-dependant autocrine Hedgehog signalling. Finally, we examine the role of Hedgehog signalling inhibition in sensitizing advanced prostate cancer to palliative therapeutic modalities. Smoothened antagonist in combination with docetaxel demonstrated significant reduction in tumour growth of both LNCaP recurrent and PC3 models, compared to single therapy. These results demonstrate the integral role of Hedgehog signalling as prostate cancer tumours adapt to castration and subsequent therapeutic modalities. Targeting these adaptive responses becomes an integral part of novel cancer therapy in prostate cancer

View record

Regulation of the Physiological Consequences of Neuroendocrine Differentiation in Prostate Cancer by Kinase and Phosphatase Cross Talk (2011)

Prostate cancer is a leading health concern among Canadian males, with one in seven Canadian men developing the disease in their lifetime and one in 27 dying from it. Localized prostatic disease can be treated with surgery, but once metastasis occurs, clinicians rely on the hormone dependent nature of the tumor for treatment. Androgen withdrawal therapy is very effective at limiting tumor growth; however, progression of the tumor to an androgen independent state is inevitable, and at the present time there is no effective therapy for this form of the disease. Neuroendocrine (NE) cells are post-mitotic, secretory cells found distributed throughout both normal and malignant prostate tissue. Increased NE cell content is associated with hormone refractory disease, and it is suspected that these NE cells play a role in the adaptation of surrounding cells to androgen withdrawal conditions through the secretion of paracrine factors. The research in this thesis was aimed towards understanding the biochemical mechanisms by which trans-differentiation of an adenocarcinoma cell to a NE cell occurs. By understanding the nature of this process, rationally designed therapeutic reagents can be developed that either block transdifferentiation of adenocarcinoma cells and inhibit the increase in NE cell content following androgen withdrawal or block the actions of NE cells that promote tumor progression. I used the Kinetworks™ Phospho-Site Screen KPSS 1.1 as well as the Human Operon Version 3.0 microarray to broadly profile changes in protein kinase regulation and mRNA expression levels occurring during NE differentiation of the human prostate cancer cell line model, LNCaP. I found that agents that induce NE differentiation in LNCaP cells cause a perturbation in the phospho-state of two downstream targets of the mammalian target of rapamycin (mTORC1), the ribosomal S6 kinase S6K1 and Rb, as well as increasing vascular endothelial growth factor (VEGF) mRNA expression. Both of these phenomena appear to involve the cAMP-dependent protein kinase PKA and a protein phosphatase PP2A family member. Since mTORC1 is considered to be a critical component in the control of tumourigenicity, and since increased VEGF is associated with advanced tumor progression, these findings appear to address some of the processes by which transdifferentiation of adenocarcinoma cells to NE cells may regulate prostate cancer progression.

View record

Master's Student Supervision

Theses completed in 2010 or later are listed below. Please note that there is a 6-12 month delay to add the latest theses.

Small molecule antagonists to ERG and ILK increase sensitivity of TMPRSS2-ERG prostate cancer cells to enzalutamide (2019)

While the introduction of second-generation androgen receptor (AR) pathway inhibitors (ARPIs) have dramatically improved survival of patients with advanced prostate cancer (PC), nearly all men treated with these ARPIs go on to develop resistance. Approximately half of PCs harbour the transmembrane protease, serine 2-erythroblast transformation-specific-related gene (TMPRSS2-ERG) fusion. TMPRSS2 allows the transcription factor ERG to be regulated and expressed by the AR, with ERG driving the transcriptional reprogramming that is hypothesized to transformation and enhance the metastatic potential of these tumours. It may be possible to increase sensitivity of ERG-expressing PCs to ARPIs through co-targeting ERG transcriptional function and ERG-associated proteins like integrin-linked kinase (ILK). Integrin-linked kinase (ILK) is an adaptor and scaffold oncoprotein which is involved in upregulating the PI3K-AKT pathway and epithelial-to-mesenchymal transition. ERG has been shown to upregulate ILK in PC models, and a small molecule antagonist of ILK, QLT-0267, phenocopies aspects of ERG antagonism. We have developed small molecule ERG antagonists, including VPC-18005 that can inhibit ERG binding to DNA, transcriptional activity, and metastatic properties of ERG-expressing PC cells. We hypothesize that VPC-18005 (and some of its analogs and derivatives) and QLT-0267 can increase the sensitivity of ERG-expressing, androgen-responsive PC cells like VCaP cells to enzalutamide-mediated cytotoxicity by further suppressing activity of ERG and ILK that is incompletely eradicated by ARPIs. A series of derivatives and analogs to VPC-18005 were screened using a nano-luciferase inhibition assay; VPC-18156 was selected as the most effective inhibitor of ERG transcription and the lead compound for testing. Proliferation and viability assays, performed through live-cell imaging and fluorescent-activated cell sorting, show that QLT-0267 and VPC-18156 increase sensitivity of VCaP cells to enzalutamide, causing increased cell death. Small interfering RNA against ERG and ILK, used as comparisons to the small-molecule antagonists, are also cooperative with enzalutamide at causing cell death in VCaP cells. Neither VPC-18005 nor VPC-18156 affect AR activity in VCaP cells. VPC-18156 decreases protein expression of a downstream target of ERG, SOX9, in VCaP cells. This project could result in a treatment strategy which can improve the effect of ARPIs in ERG-expressing metastatic castration-resistant PCs.

View record

Determining the Role of Scavenger Receptor Class B Type I as a Source of Cholesterol for De Novo Androgen Synthesis in Castration-Resistant Prostate Cancer (2015)

Existing therapies for castration-resistant prostate cancer (CRPC) extend life and provide clinical benefit; however, patients develop therapeutic resistance. Persistent androgen signaling in CRPC is maintained in part by intratumoral steroidogenesis from the precursor cholesterol. The high density lipoprotein-cholesterol (HDL) receptor, scavenger receptor class B type I (SR-BI), is upregulated in CRPC models in vitro and in vivo. This thesis tests the hypothesis that depriving CRPC cells of HDL as a cholesterol source by silencing SR-BI will diminish de novo steroidogenesis and resultant androgen receptor-mediated signaling necessary for CRPC viability. The effects of SR-BI silencing were studied using CRPC C4-2 cells transfected with either Stealth RNAi duplexes targeting SR-BI (SRBI-KD) or Lo GC Negative Control (NC) duplexes. Cells cultured in androgen-depleted conditions for varying times post-transfection were assessed for SR-BI expression, prostate specific antigen (PSA) expression by chemiluminescence, cholesterol levels by fluorometry, and steroid levels by liquid-chromatography-mass spectrometry. HDL-uptake was measured by flow cytometry of the fluorescent cholesterol mimetic 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI-HDL). Cell cycle state was assessed by flow cytometry of propidium iodide DNA staining, while cell cycle markers were assessed by immunoblotting. Adaptive stress responses were assessed by immunoblotting for autophagy markers, as well by flow cytometry for senescence associated beta-galactosidase (SA-β-gal) using a fluorogenic substrate, 5-dodecanoylaminofluorescein di-β-D-galactopyranoside. SRBI-KD treatment reduced SR-BI levels by ~57% by 2 days and ~86% by 4 days post-transfection. This correlated with reduced DiI-HDL-uptake by 22%, reduced cellular testosterone levels two-fold, and reduced PSA secretion by 39% compared to NC cells. These changes were accompanied by reduced proliferation, G₁S cell cycle arrest, and a small, but measurable increase in cell death at 4 to 6 days post-transfection. Cell stress was evidenced by enhanced autophagy activity and induced stress marker expression while senescence was evidenced by increased SA-β-gal activity. These studies indicate that SR-BI silencing reduced DiI-HDL uptake which reduced cellular androgen content and androgen signaling, resulting in induction of an adaptive stress response characterized by cell cycle arrest, autophagy, senescence, and eventually death of these CRPC cells. The data presented herein provide support for SR-BI as a cholesterol source for androgen synthesis in CRPC cells.

View record

Establishing methods to screen novel small molecules targeting insulin-like growth factor / insulin-like growth factor binding protein interaction (2013)

Insulin-like growth factors (IGFs) are important systemic mediators of growth and survival that suppress apoptosis and promote cell cycle progression, angiogenesis and metastatic activities in various cancers by activating IGF-IR tyrosine kinase-mediated signaling. These effects depend on the bioavailability of IGFs, which is regulated by IGF binding proteins (IGFBPs). Increased IGFBP-2 and IGFBP-5 expression observed in castration-resistant prostate cancer is thought to promote tumor progression by enhancing IGF-mediated signaling. IGFBPs have cooperative carboxy-terminal and amino-terminal low and a high affinity IGF binding sites. I hypothesize that blocking the high affinity IGF binding site can affect the bioavailability of IGFs to target tissues and thus be used for treatment of various IGF-responsive diseases including prostate cancer. I initially characterized immunologic reagents capable of being used in sandwich ELISA formats to detect IGF-I and IGFBP-5 and attempted several configurations to establish an IGF-I/IGFBP-5 “bridged” sandwich ELISA platform to measure association and dissociation of IGF-I/IGFBP-5 complex formation. The inability of all bridged ELISA formats tested to measure IGF-I/IGFBP-5 binding, lead me to developed a Bio-Layer Interferometry-based assay that measures IGF-I/ IGFBP-5 binding kinetics that will allow for screening of factors that can affect this intermolecular interaction.I demonstrated that biotinylated IGF-I bound to streptavidin-coated biosensors can be used to measure binding of recombinant IGFBP-5 [2.24 nm shift in optical density (Response)]. I also demonstrated that IGF-I could efficiently disrupt this interaction (0.21 nm shift), while the amino-terminal IGF-I mutant, E3R, exhibits an intermediate competitive activity (1.47 nm shift) and insulin exhibits a low competitive activity (1.83 nm shift). In addition, I demonstrated that IGF-I can competitively disrupted this interaction, resulting in a dissociation rate constant (Kdis 1.5-³ 1/s), In contrast, the amino terminal IGF-I mutant, E3R binds with an intermediate affinity (Kdis 5.6-⁴ 1/s), and buffer free sample results in a (Kdis) of 1.5-⁴ (1/s). These results demonstrate the capacity of this BLI-based assay to differentiate relative competitive activity of compounds that target the high affinity IGF-I binding site of IGFBPs and establish a platform to screen for factors that might be developed as rationale therapeutics to disrupt sequestration of IGF-I by IGFBPs.

View record

Studying the effect of OGX-225, a novel antisense therapy, on prostate cancer cells (2010)

Prostate cancer (PCa) is the most common malignancy diagnosed in Canadian men. If the disease is not detected early the patients would present with metastatic disease. Metastatic PCa represents a challenge for treatment. The primary treatment at this stage is androgen ablation (castration). However, metastatic PCa invariably progress to a castration-resistant form, which does not respond to androgen ablation. This form is a lethal form of the disease, and there are several strategies that have been described to treat this form of cancer. The Insulin-like Growth Factor (IGF) system plays a pivotal role in prostate development and castration-resistant progression. It is tightly regulated by seven high affinity Insulin-like Growth Factor Binding Proteins (IGFBP 1-7). IGFBP-2 and IGFBP-5 are over-expressed in PCa and play a role in castration-resistant progression, while IGFBP-3 is downregulated and inhibits growth of PCa cells. OGX-225 is a novel second- generation antisense drug (ASO) that targets both IGFBP-2 and IGFBP-5. M. Muramaki et al, at the prostate center in Vancouver, showed in their unpublished data that it inhibits the growth PCa cells both in vitro and in vivo, and that it downregulates IGFBP-2 and IGFBP-5. In my thesis, I showed the inability to rescue PCa cells exposed to OGX-225 by adding back IGFBP-5, indicating that the action of OGX-225 is irreversible and targets several targets in addition to IGFBP-2 and IGFBP-5. I also showed the growth inhibitory effect of IGFBP-3 through a mechanism other than apoptosis, while not having an additive effect on PCa cells when combined with OGX-225. This thesis reports the multi-targeting ability of OGX-225 resulting in the inability to rescue PCa cells exposed to it, and gives some insight into the mechanism of action of IGFBP-3 and its effect when combined with OGX-225.

View record


If this is your researcher profile you can log in to the Faculty & Staff portal to update your details and provide recruitment preferences.


Get key application advice, hear about the latest research opportunities and keep up with the latest news from UBC's graduate programs.