Xuesen Dong

The use of next-generation anti-androgens in the treatment of prostate cancer has led to an increase in therapy-induced neuroendocrine prostate cancer (t-NEPC). Recent studies have identified t-NEPC as a highly aggressive subtype of prostate cancer with poor survival rates. Our laboratory has recently demonstrated that the Lin28B, a core embryonic protein that suppresses the maturation process of the micro-RNA let-7, plays a crucial role in transforming adenocarcinoma prostate cancer (AdPC) into t-NEPC. Lin28B is an RNA-binding protein with two critical domains: the cold-shock domain (CSD) and the zinc-knuckle domain (ZKD). By utilizing both of CSD and ZKD domains, Lin28B protein can structurally binds to the microRNA let-7, which has been reported as a tumor suppressor and oncogene in NEPC, and inhibits its biosynthesis. Therefore, we hypothesize that blocking the interaction of CSD or ZKD with let-7 using a small molecule chemical may provide therapeutic opportunities to treat t-NEPC.In this study, we used a computer-aided drug design strategy to target Lin28B ZKD with small molecule inhibitors. We identified several inhibitors that effectively block Lin28B from binding to let-7 miRNA and restore let-7 biogenesis. Furthermore, we demonstrated that these inhibitors suppress the expression of stem-like pluripotency gene networks and neuroendocrine biomarkers. Lin28B inhibitors also reduces the expression of cancer stem cell surface markers such as CD44 and CD133. We also showed that these inhibitors suppress t-NEPC cell proliferation and clonogenic abilities. Overall, our study discovered several promising small molecules inhibitors of Lin28B that could be developed into potential drugs against NEPC.

View record

Topoisomerase II (TOP2) is a ubiquitous enzyme in human cells that regulates DNA topology and chromatid separation, which is critical for cells to grow and survive. Popular chemotherapeutics (such as doxorubicin and etoposide) are TOP2 poison inhibitors that induce DNA double-stranded breaks, thereby cytotoxicity to initiate cancer cell death. However, they have significant side effects such as secondary malignancy and cardiotoxicity. TOP2 catalytic inhibitors, on the other hand, cause minimal DNA disruption, have low cytotoxicity, and are efficient at inhibiting cancer cell proliferation. They are sought after as potential anticancer treatments. The development of modern TOP2 catalytic inhibitors is discussed in this thesis. We used a computer-aided drug discovery (CADD) campaign to screen over 6 million molecules from the ZINC15 database through a new druggable pocket of TOP2 protein at its DNA binding domain. T60, the lead drug, is demonstrated to be a catalytic TOP2 enzyme Inhibitor. We defined the mechanism of action of T60 as that it binds TOP2 proteins and blocks TOP2 protein from interacting with DNA thereby preventing TOP2-mediated DNA cleavage. It has low cytotoxicity but effectively inhibits cancer cell proliferation and xenograft growth. T60 also reduces the functions of the androgen receptor and blocks prostate cancer cell growth. Based on these results, we conclude that T60 is a promising candidate compound for the development of new anticancer medications.

View record

Due to the increased utilization of next generation anti-androgens to treat prostate adenocarcinoma (AdPC), therapy-induced neuroendocrine prostate cancer (t-NEPC) has become more prevalent. Although the mechanisms by which t-NEPC is established are not fully understood, emerging evidence suggests that AdPC cells can gain an intermediate pluripotent stem cell (SC)-like phenotype that can promote t-NEPC development. However, it remains unclear whether the core embryonic stem cell genes (ESCs) (LIN28, POU5F1, SOX2, and NANOG) regulate the stem-like state of prostate cancer cells and the switch from luminal epithelial to neuroendocrine lineage during the transition from AdPC to t-NEPC. We hypothesize that LIN28B plays a key role in the transition from AdPC to t-NEPC, and that the overexpression of LIN28B may promote proliferation and trans-differentiation, which may contribute to t-NEPC progression. By comparing the published RNA-seq data on AdPC and t-NEPC, we found that approximately 50% of t-NEPC patient tumors have gained LIN28B and SOX2 expression. Standard molecular and cellular biology techniques were applied to characterize the functions of LIN28B and its relationship with SOX2 using t-NEPC cell and xenograft models. We found that the mRNA levels of LIN28B and SOX2 are positively correlated in patient tumors, patient derived xenografts, transgenic mice, and multiple cell models. LIN28B and SOX2 expression was confirmed to be co-upregulated in a subset of t-NEPC patients by immunohistochemistry. Using our clinically relevant t-NEPC cell/xenograft model, DuNE, we demonstrated that LIN28B is essential for stem cell-like and neuroendocrine marker expression and cell morphology. LIN28B gene depletion by CRISPR inhibited DuNE xenograft initiation and tumor growth. These LIN28B functions are mainly mediated by its inhibitory effects on the microRNA let-7d, which resulted in the upregulation of HMGA2 and HGMA2 mediated SOX2 transcription.Overall, this thesis work adds to the understanding that the LIN28B/let-7/SOX2 axis is an important signaling pathway that regulates a cancer stem-like phenotype to promote t-NEPC development. Ultimately, this knowledge pertains to the clinical implications of LIN28B in informing future therapies that will be effective for managing t-NEPC.

View record

The androgen receptor (AR) gene is important for prostate cancer development and tumor progression. The protein encoded by the AR gene is the mainstay therapeutic target for metastatic prostate cancers. Alterations in mRNAs transcribed by the AR gene are also biomarkers of therapy-resistant prostate cancers. These alterations are induced by alternative RNA splicing of the AR gene, AR gene amplification, or gain-of-function mutations of the AR gene. Non-coding RNAs derived from the AR gene may also serve as biomarkers of disease progression of prostate cancers. Circular RNA (circRNA) is one subtype of non-coding RNAs that have been demonstrated to be abundantly expressed in human cells and is highly resistant to exonuclease digestion. It is generated by RNA splicing machinery through back-splicing processes. In this thesis, I have demonstrated that there are two circRNAs derived from the AR gene, namely CirAR2 and CirAR3. These circRNAs are widely expressed in AR-positive prostate cancer cells. The expressions of these circRNAs are elevated by androgen deprivation and anti-androgens. I have constructed expression vectors encoding CirAR3 and showed that CirAR3 does not alter AR protein expression as well as ligand-dependent AR transcriptional activities. However, I demonstrated that CirAR3 is resistant to Ribonuclease R digestion, and has a significantly longer half-life than linear AR mRNAs. CirAR3 can be detected by real-time PCR in less than 10 AR-positive 22Rv1 prostate cancer cells. In summary, these studies suggest that circRNAs derived from the AR gene may be potential biomarkers of prostate cancers.

View record

Treatment-induced neuroendocrine (NE) prostate cancer (t-NEPC) is an aggressive subtype of prostate cancer (PCa) that can arise as a consequence of rigorous androgen receptor pathway inhibition (ARPI) therapies now used to treat castration resistant disease (CRPC). While the PI3K/AKT pathway has been investigated as a co-therapeutic target with ARPI for advanced prostate adenocarcinoma, whether this strategy has implications on t-NEPC progressionremains unknown. Findings from this work indicate that PI3K/AKT inhibition alone reduces protein expression of the RE-1 silencing transcription factor (REST) and induces multiple NE markers in PCa cells. The loss of REST by PI3K/AKT inhibition is through protein degradation mediated by the E3-ubiquitin ligase β-TRCP and REST phosphorylations at the S1024, S1027, and S1030 sites. Since AR inhibition was previously reported to deplete REST, results from this project reveal that the combined inhibition of PI3K/AKT and AR further aggravates REST protein reduction. Upon profiling the transcriptomes of AKT inhibition, AR inhibition, and AKT/AR co-inhibition in the LNCaP cell model, Gene Set Enrichment Analysis (GSEA) shows that these transcriptomes are highly correlated with the REST-regulated gene signature. Co-targeting AKT and AR resulted in an even higher correlation comparing to those of single treatment. Comparing these transcriptomes to the RNA-seq gene signature of t-NEPC patients by GSEA, it was observed that adding AKT inhibition to AR blockade enhanced the expression of neurogenesis-related genes and resulted in a stronger and broader upregulation of REST-regulated genes specific to t-NEPC. Collectively, these results indicate that AKT pathway inhibition can induce NE transdifferentiation in PCa cells via REST protein degradation. It delineates a potential risk for the AR and PI3K/AKT co-targeting strategy as it may further facilitate t-NEPC development.

View record

The prostate composes of epithelium and stroma, both of which are kept in balance to maintain normal prostate function. The balance between epithelium and stroma can be disrupted by the abnormal growth of stromal cells which results in prostate diseases such as benign prostatic hyperplasia. The epithelial-stromal interaction plays important roles not only in normal prostate homeostasis maintaining but also in prostate cancer development and progression. In prostate tumor, cancer associated fibroblasts enhance the secretions of cytokines and growth factors to favor cancer cells growth and metastasis. Androgen receptors are reported to regulate the development and maintenance the function of prostate. Progesterone receptor (PR) which belongs to the same steroid hormone receptor family as androgen receptors are little known in prostate. PR was reported to express in prostate, but there is no clear conclusion about the localization and function of PR in human prostate. The objective of this thesis is to investigate the expression and function of PR in human prostate. Two PR isoforms, PRA and PRB, are detected in subsets of the human prostate stromal cells by applying immunohistochemistry assays. Both PR isoforms express specifically in human prostate stromal fibroblasts and smooth muscle cells. Both PRA and PRB are demonstrated to play an inhibitory role in prostate stromal cell proliferation. PR suppresses the expression of cyclin A, cyclinB and cdc25c to delay cell cycle. PRA and PRB are demonstrated to regulate different transcriptomes by gene microarray assay. Immunohistochemistry assays were applied to human prostate cancer tissue biopsies, and PR levels are detected to decrease in the cancer associated stroma compared to the paired normal stroma. The conditioned media from PR positive stromal cell inhibit PC-3 and C4-2B cell motility through down-regulating the secretion of stromal cell derived factor 1 and interleukin 6. We conclude that PRA and PRB express in prostate stromal cells and inhibit the stromal cells proliferation. Decreased expression of PR in cancer associated stroma contributes to prostate tumor progression.

View record