Embryonic T-box transcription factor Brachyury as a predictive biomarker and therapeutic target in prostate and lung cancers

Patrícia Fernandes Fontão Embryonic T-box transcription factor Brachyury as a predictive biomarker and therapeutic target in prostate and lung cancers february of 2021 UMinho | 2021 Patrícia Fernandes Fontão Embryonic T-box transcription factor Brachyury as a predictive biomarker and therapeutic target in prostate and lung cancers Universidade do Minho Escola de Ciências Patrícia Fernandes Fontão Embryonic T-box transcription factor Brachyury as a predictive biomarker and therapeutic target in prostate and lung cancers Master’s Dissertation Master in Applied Biochemistry Specialization in Biomedicine Work developed under the supervision of Doctor Olga Catarina Lopes Martinho and Professor Doctor Isabel Correia Neves Universidade do Minho Escola de Ciências february of 2021 ii Copyright and conditions for use of the work by third parties This is an academic work that can be used by third parties as long as the internationally accepted rules and good practices regarding copyright and related rights are respected. Accordingly, this work can be used under the terms set out in the license below. If the user needs permission to use the work under conditions not foreseen in the license indicated, he should contact the author, through RepositóriUM of University of Minho. Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ iii Acknowledgements This chapter of my academic course come to an end, and given its importance I would like to leave a well-earned word of gratitude to the people who made it possible. First of all, I want to thank my supervisor Olga Martinho, my "mother" in science as I like to call her, for everything she taught me during these long months, for all the support and guidance, for always seeing the silver lining no matter what, for always having a word of support and the confidence she have always shown in me, for being unvaluable in this chapter of my life, for being an inspiration as a scientist and as a person, and above all for being the best supervisor I could have. A simple thank you isn’t enough. Secondly, I want to thank my co-supervisor, Professor Isabel, for the opportunity to do this project under her guidance, for having received me in her lab and guided me all step of the way, for her willingness to help me, availability and kindness. Diana and Raquel, thank you for always being present and ready to help me in everything, for the support, the cheerfulness, and kindheartedness that are so characteristic of you. Gabi and Ju, that were right by my side in this stage since day one, thank you for all the support, companionship, and for making this year just a bit easier. To all the lab mates for all the help, company, and joy. I am grateful for my friends, for being there to listen and put up with me, for the encouragement, and above all for the friendship, and for my family, for all the love, unconditional support, care and concern showed, for believing in me and for being a constant presence throughout my life. I want to thank to my parents for giving me strength when it stubbornly ran away and the fear assaulted me, for being light when everything was dark, and for being, always and forever, the great pillars of my life. And finally, and most especially, a huge thank you to my greatest love, my baby sister, the best I could’ve wish for, my other half and best part of me. I hope one day I have just a little bit of the strength and courage you've shown to have and make you proud, as I am of you. I wish you were here. I cannot put into words how thankful I am for having you in my life! Finally, I would also like to mention that the financial support was provided by national funds through the Foundation for Science and Technology (FCT), project PTDC/MED-ONC/31423/2017 (Ref: POCI-01-0145-FEDER-031423). iv STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. Assinado por : Patrícia Fernandes Fontão Num. de Identificação: BI15337390 Data: 2021.02.18 14:05:49+00'00' v Fator de transcrição embrionário T-box Brachyury como biomarcador de resposta e alvo terapêutico em cancros da próstata e pulmão Resumo A Brachyury tem sido reconhecida pelo seu papel em várias neoplasias, tais como cancro da próstata (CPa) e pulmão (CP), entre outros, devido às suas propriedades oncogénicas como por exemplo, a indução da EMT, propriedades estaminais, metastização e mau prognóstico, assim como resistência à terapia em ambos os modelos. Todavia, os mecanismos moleculares através dos quais esta proteína atua no cancro não estão completamente elucidados. Deste modo, os objetivos fundamentais deste trabalho consistiam em elucidar o papel biológico de Brachyury em CP, estudar o seu valor preditivo na resposta a terapias antiandrógenas e a inibidores de EGFR em CPa e CP, respetivamente, bem como explorar estratégias terapêuticas anti-Brachyury. Assim, modulámos a expressão de Brachyury em linhas celulares de CPa e LC, e avaliámos o seu efeito no comportamento biológico das células, na expressão de várias moléculas já associadas com Brachyury, e na modulação da resposta das células a terapias dirigidas a AR e EGFR. Por fim, avaliámos a especificidade dos fármacos afatinib e THZ1 na modulação da Brachyury, assim como desenvolvemos ainda drug delivery systems (DDS) baseados em estruturas zeolíticas e sílica mesoporosa com estes fármacos. Em CPa foi possível observar alguns resultados consistentes com uma indução de EMT parcial e uma expressão aumentada de AR após a sobre-expressão da Brachyury, no entanto, isto não se refletiu na resposta a terapias antiandrógenas. Em relação ao CP, foi possível evidenciar o seu papel oncogénico, considerando o notório aumento de viabilidade celular, migração e formação de colónias, após a sua sobre-expressão. Foram ainda observadas alterações características de uma indução de EMT parcial e um aumento em marcadores de pluripotência. É ainda importante salientar que após a sobre-expressão da Brachyury foi verificada uma menor sensibilidade ao inibidor de EGFR AST1306, na linha celular H292, que poderá ser explicada por uma sobre-activação da via AKT. Por último, demonstrarmos que os fármacos afatinib e THZ1 inibem a expressão de Brachyury em ambos os modelos, e podem ser usados em DDS. Desta forma, concluímos que a Brachyury poderá potencialmente ser uma biomarcadora preditiva em CP, mas não em CPa. Adicionalmente, sugerimos os fármacos afatinib e THZ1 como bons candidatos para reverter o fenótipo maligno induzido pela Brachyury em cancro. Palavras-chave: Biomarcador, Brachyury, Cancro de pulmão, Cancro da próstata, Terapia de cancro. vi Embryonic T-box transcription factor Brachyury as a predictive biomarker and therapeutic target in prostate and lung cancers Abstract Brachyury has been recognized for its role in several cancers, such as prostate cancer (PCa) and lung cancer (LC), among others, due to its oncogenic properties such as EMT induction, stem properties, metastasis and poor prognosis, as well as resistance to therapy in both models. However, the molecular mechanisms through which this protein acts in cancer are not completely elucidated. In this work we intended to dissect the biological role of Brachyury in LC, study the predictive value of this transcription factor in antiandrogen therapies and EGFR-TKIs respectively in PCa and LC, and finally explore therapeutic strategies for Brachyury targeting, including the usage of drug delivery systems (DDS). To accomplish that, we modulated the expression of Brachyury in two PCa and LC cell lines and performed in vitro biological assays, assessed by western blot and qRT-PCR the expression of a number of molecules associated with Brachyury, and evaluated its impact on the modulation of cells response to antiandrogen therapies and EGFR-TKIs. Finally, we assessed the specificity of afatinib and THZ1 drugs at targeting Brachyury expression, and proceeded to develop zeolite and mesoporous silicabased DDS. Overall, in our PCa models we observed some results consistent with an induction of an EMT partial state and upregulated levels of AR upon Brachyury overexpression (OE), however with no effects in antiandrogenic therapies response. Concerning LC, we evidenced the oncogenic role of Brachyury in this malignancy through the increased cell viability, migration and colonies formation. Then again, we were able to note changes characteristics of EMT partial states as well as an increase in stemness markers. Importantly, upon Brachyury OE we found that the H292 cell line became more insensitive to the EGFR inhibitor, AST1306 (allitinib), that might possibly be due to an overactivation of the AKT pathway. Finally, we proved that afatinib and THZ1 are able to inhibit Brachyury expression in PCa and LC, and successfully developed DDS using these drugs. Thus, we conclude that Brachyury might have potential as a predictive biomarker to EGFR targeted therapies in LC, but not to antiandrogens therapies in PCa. Furthermore, our results suggest that afatinib and THZ1 can be used to target Brachyury and potentially revert the oncogenic Brachyuryassociated phenotype. Keywords: Brachyury, Cancer therapy, Lung cancer, Predictive biomarker, Prostate cancer vii Table of contents Copyright and conditions for use of the work by third parties ................................................................ ii Acknowledgements ............................................................................................................................. iii Statement of integrity .......................................................................................................................... iv Resumo............................................................................................................................................... v Abstract.............................................................................................................................................. vi Table of contents ............................................................................................................................... vii List of abbreviations ............................................................................................................................ ix Table of figures .................................................................................................................................. xiii List of Tables ..................................................................................................................................... xiv CHAPTER 1: General Introduction ....................................................................................................... 1 1.1. Cancer ................................................................................................................................ 2 1.2. Prostate cancer ................................................................................................................... 2 1.2.1. Antiandrogen therapies ................................................................................................ 4 1.3. Lung cancer ...................................................................................................................... 10 1.3.1. Molecular targeted therapies: anti-EGFR therapies ...................................................... 12 1.4. Embryonic T-box transcription factor Brachyury .................................................................. 15 1.4.1. Brachyury and cancer ................................................................................................ 16 1.4.1.1. Brachyury in prostate cancer .............................................................................. 19 1.4.1.2. Brachyury in lung cancer .................................................................................... 20 1.4.2. Brachyury as a therapeutic target ............................................................................... 24 1.5. Drug delivery systems ....................................................................................................... 25 1.5.1. Zeolites and mesoporous silicas ................................................................................. 26 CHAPTER 2: Research Objectives ..................................................................................................... 29 CHAPTER 3: Materials and methods ................................................................................................. 30 3.1. Cell lines and cell culture ................................................................................................... 31 xiv List of Tables Table 1. FDA and EMA approved EGFR inhibitors for LC. ....................................................... 15 Table 2. Review of the literature regarding Brachyury expression and its clinical impact in LC patients. ................................................................................................................................. 23 Table 3. Panel of human cancer cell lines used in this work. .................................................. 31 Table 4. Primers sequences used for qRT-PCR. ..................................................................... 32 Table 5. Primary antibodies used for Western Blot. ................................................................ 33 Table 6. Mean IC50 values of antiandrogens in transfected PCa cell lines. ............................... 43 Table 7. Mean IC50 values for afatinib, AST1306 and osimertinib in transfected LC cell lines.. . 49 1 CHAPTER 1: General Introduction 2 1.1. Cancer Cancer is a major health problem and is the second leading cause of death worldwide, expecting to rank as the leading cause of death and the most important barrier to increase life expectancy in the 21st century. Incidence and mortality rates of this disease are rapidly growing, with 18.1 million new cases and 9.6 million cancer deaths worldwide estimated in 20181. Meanwhile, in Portugal were estimated 58199 new cases and 28960 cancer related deaths2. Cancer is a complex disease that involves dynamic changes in the genome, including mutations that activate the so-called oncogenes and inhibit tumor suppressor genes3. To simplify all the complex and intricate alterations that occur during the formation of these malignant neoplasms, Hanahan and Weinberg defined in 2000 the “Hallmarks of Cancer”, in other words, functional capabilities that allow cancer cells to survive, proliferate, and disseminate3. In 2011, the same authors redefined and expanded these Hallmarks of Cancer (Figure 1), adding four more aspects that are characteristic of this disease4. 1.2. Prostate cancer Prostate cancer (PCa) ranks as the 2nd most common diagnosed malignancy in men and is the 5th leading cause of cancer related deaths across the world. Moreover, is the most frequently diagnosed malignancy among men in over half of the countries of the world, including Portugal1,2. Development of PCa has some well-established risk factors, such as ethnicity, age and family history of disease5,6. Age is by far the strongest risk factor for PCa incidence, which rises abruptly with age, being diagnosed in very few people under 50 years (<1 % of all patients) and the mean age at the time of Figure 1. Hallmarks of cancer defined by Hanahan and Weinberg in 20114. 3 diagnosis is 72-74 years old7-9. Overall, when PCa is diagnosed at an early age it is generally associated with a family history of disease6-8, which might be due to a genetic factor or even environmental factors within a family. Some genes and common gene variations have been correlated with an increased risk of PCa, such as mutations in breast cancer genes 1 and 2 ( BRCA1 and BRCA2 , respectively), in homeobox B13 ( HOXB13 ) and in some DNA mismatch repair genes (MMR) namely mutL homolog 1 ( MLH1 ), mutS homologs 2 and 6 ( MSH2 and MSH6 , respectively), postmeiotic segregation increased 2 ( PMS2 ) and other DNA repair genes like, checkpoint kinase 2 ( CHEK2 ), and ataxia telangiectasia mutated ( ATM )6,7,10,11. Finally, regarding ethnicity, there is a great heterogeneity in clinical PCa incidence worldwide, and it is known that black men have a higher risk of PCa incidence and death compared to men from white or Asian backgrounds7. PCa presents a number of challenges for primary care clinicians, mainly because the majority of men with PCa are asymptomatic until the tumor has progressed. Advanced PCa causes symptoms such as lower urinary tract symptoms (LUTS), erectile dysfunction and haematuria7, which also occur in other conditions like benign prostate hyperplasia (BPH) and prostatitis, making it very challenging to distinguish between them based only on symptomatology7. Currently, serum prostate-specific antigen (PSA), the most clinically validated biomarker in PCa, is very used for screening, early diagnosis and for evaluate disease progression as well12,13. PSA, encoded by an androgen-responsive gene, is a serine protease secreted by prostate epithelial cells14, and was first detected in the serum of PCa patients in 198015. In 1986 it was approved by the Food and Drug Administration (FDA) as a biomarker for PCa progression monitoring, and as such, in the 1990’s this test became widely used as a screening method with early detection as a main goal, with the consequent lowering of the mortality rates15. Unfortunately, PSA is not completely specific, since conditions such as BPH and prostatitis increase PSA levels7,13, and some men with PCa also have normal PSA levels7,14, leading to false positive and negative results. These limitations result in overdiagnosis and overtreatment, which constitute the most important adverse effects of PCa screening, being these events vastly more common than in screening for breast, colorectal, or cervical cancer12,16. In spite of these disadvantages, PSA is still a useful biomarker of PCa, but new ones are in need for diagnosis and to predict PCa progression, as well as for distinguish between clinically significant from indolent tumors. In despite of the aforementioned limitations of PSA testing, when a patient has a suspicious high level of PSA it is recommended a digital rectal exam (DRE), and then a transrectal ultrasonography (TRUS) guided biopsy, which remains the standard tool for PCa diagnosis17. 4 Depending on the cell of origin, which is still a controversial topic and an area of active investigation1820, prostatic tumors can be grossly classified as adenocarcinomas (ADC), squamous carcinomas and neuroendocrine (NE) tumors19 (Figure 2), being the vast majority of prostate tumors ADC21. Despite the existence of this classification, prostate tumors are normally multifocal, having multiple independent histologic foci that are often genetically distinct19. Prognostication and treatment stratification at the time of diagnosis are based on serum PSA levels, clinical stage (TNM system), and Gleason score, which is a widely used grading system based on histological patterns of the tumor, originally defined by the pathologist Donald Gleason17,22. Depending on the stage and grade of the tumor, treatment for PCa may involve watchful waiting/active surveillance, surgery, cryosurgery, radiotherapy, brachytherapy, chemotherapy, hormonal therapy, or combinations. In the case of localized disease, the therapeutic approach consists in watchful waiting/active surveillance for low risk tumors16,23,24, and surgery and radiotherapy for intermediate and high risk tumors5,25-27. However, for advanced PCa the first line of treatment is hormonal therapy26,28. Figure 2. Prostate epithelium characterization and origin of the different PCa histological subtypes. Luminal, basal and NE cells, all of which express a specific panel of characteristic proteins, comprise the normal prostate epithelium: luminal cells are defined by expression of cytokeratin (CK) 8, CK18 and androgen receptor (AR); basal cells express high levels of CK5, p63 and very low levels of AR; NE cells, the smallest population, express NE markers such as synaptophysin and chromogranin A and do not express AR. Studies have demonstrated that both luminal cells and basal cells can serve as the cell of origin for PCa; both basal and luminal cells can give rise to ADC, but only basal cells give rise to squamous carcinoma. It remains unknown whether NE cells can be transformed to generate a malignant neoplasm. Adapted from19. 1.2.1. Antiandrogen therapies Hormonal therapy arose due to the unraveled role of androgens in PCa, conducted by Charles Huggins29. He demonstrated that the reduction of testosterone levels, by surgical castration, had therapeutic implications in patients with metastatic PCa, awarding him with the Nobel Prize of Medicine 5 in 1966. Androgens and their receptor, androgen receptor (AR), play a pivotal role in PCa development and progression, and the underlying rationale of hormonal therapy is the blockade of their action30,31. AR is a type I hormone receptor that belongs to the nuclear receptors superfamily32,33, which besides being a key factor in PCa it also plays a role in the development and maintenance of the reproductive, musculoskeletal, cardiovascular, immune, neural, and hemopoietic systems34, and is implicated in other cancer types such as breast, ovarian and pancreatic cancers35-38. Structurally, the AR, as the other nuclear receptors, is comprised by several domains, each one with a specific function: a Nterminal domain (NTD), with an activation function, being also responsible for receptor dimerization through interaction with the C-terminal ligand binding domain (LBD); and a hinge region that connects the DNA biding domain (DBD) and the LBD32,33 (Figure 3A). In the basal unbound state, the AR is located in the cytoplasm and bound to heatshock proteins (Hsp90, Hsp70), and other chaperone proteins in a conformation that prevents DNA binding32,34,39 (Figure 3B). The androgens testosterone and dihydrotestosterone (DHT) bind to the LBD of AR, displacing the Hsp and promoting an interaction between the NTD and LBD of the receptor. AR is further translocated into the nucleus, where it dimerizes and bind to androgen response elements (AREs) in the promoter regions of target genes, such as PSA , transmembrane protease serine 2 ( TMPRSS2 ), etc32,40,41. This results in the recruitment of various coregulators to enhance or repress transcription42, activating the transcriptional program of AR and amplifying the signal initiated by hormone binding, that leads to biological responses like growth and survival32,40 (Figure 3B). This constitutes the known genomic functions of AR, however a growing body of evidence suggests that this receptor also has non-genomic functions, that occur in a short time frame being incompatible with the activation of the AR transcriptional program38,43,44. Several studies using cell lines reported that upon ligand binding, AR is able to interact with cytoplasmic proteins, such as the nonreceptor tyrosine-kinase Src and phosphoinositide 3-kinase (PI3K) leading to the activation of the mitogenactivated protein kinase (MAPK) and PI3K/Akt/mTOR signaling pathways45,46, respectively (Figure 3B). These non-genomic functions demonstrate that AR plays a much more complex role than first thought. 6 Figure 3. AR structure domains and respective signaling. A) AR is comprised by a N-terminal domain (NTD), a DNA binding domain (DBD), a C-terminal ligand binding domain (LBD) and a hinge region (H) that connects the DBD with the LBD. Retrived from 34. B) AR has a non-genomic function and is able to interact with cytoplasmic proteins, such as the non-receptor tyrosine-kinase Src and the p85α regulatory subunit of PI3K leading to the activation of the MAPK/ERK and PI3K/Akt/mTOR signaling pathways, respectively, and culminating in increased proliferation and cell survival. Concerning AR genomic function, testosterone enters the cells, the majority being converted into dihydrotestosterone (DHT) by the 5α-reductase enzyme, DHT then binds to AR, displacing Hsp and entering the nucleus. At the nucleus AR dimerizes, interacts with the promoter regions of androgen responsive genes and recruits the basal transcription machinery (BTFs) and coactivators regulating the expression of target genes, such as PSA , TMPRSS2 , etc. Adapted from 43. Inhibition of androgen signaling can be achieved through two main tactics, the inhibition of testosterone production by the testis and adrenal gland, and the direct targeting of the AR. The decreased levels of circulating androgens with the following decline in AR activation and it’s signaling is the rationale underlying androgen deprivation therapy (ADT), the gold standard treatment for PCa47. This can be achieved by castration either surgical (orchiectomy) or chemically, acting in two crucial molecules that control androgen synthesis, luteinizing hormone (LH) and gonadotropin-releasing hormone (GnRH), using for that GnRH (also known as LH releasing hormone (LHRH)) receptor agonists and antagonists32. GnRH agonists desensitize the GnRH receptor by interrupting its physiological intermittent stimulation, whereas 7 the GnRH antagonist blocks directly the stimulation of the receptor32,48, resulting in a decrease of serum testosterone levels, being reported that more than 80-90% of patients show a positive response to androgen ablation9,30,32. The class of drugs that directly target AR, through binding to the LBD and competing with androgens in order to block receptor activation, are denominated antiandrogens or AR antagonists49. Antiandrogens are mostly often used in combination with ADT, being this therapeutic approach known as combined androgen blockade (CAB)32. Flutamide was the first nonsteroidal antiandrogen to be clinically approved by the FDA in 1989 for use in advanced PCa. Few years later, other derivatives of flutamide, namely nilutamide and bicalutamide were also approved, and altogether comprise what is called the 1st generation of antiandrogens32,49. These AR antagonists possess relatively weak affinity for the AR and are not potent enough to completely block AR signaling. The patients often develop resistant mutations, which normally converts them into AR agonists49,50, or other resistance mechanisms, in a time frame of 14-20 months28. When patients no longer respond to ADT and/or CAB, cancer progresses even under castrate levels of testosterone, being this disease now classified as castration resistant prostate cancer (CRPC) with a median survival of 2.8 years28,32,51. With this in mind, a lot of efforts were made, and a 2nd generation of antiandrogens was developed. Enzalutamide was the first AR antagonist of this generation to be approved in 2012 and represented a breakthrough in CRPC treatment due to its properties and effectiveness52,53. Enzalutamide binds to AR with strong affinity, prevents AR nuclear translocation and DNA binding, and led to an improvement on progression of the cancer and in overall survival (OS)52,53. Other 2nd generation antiandrogens, namely apalutamide and darolutamide, were approved for nonand metastatic CRPC (mCRPC) treatment by the FDA in 2018 and 2019, respectively, being subsequently approved by the European Medicines Agency (EMA)49,54-56. Furthermore, the androgen synthesis inhibitor abiraterone was developed and approved just before enzalutamide57,58. This specific drug inhibits the CYP17A1 enzyme that is involved in androgen biosynthesis, preventing this way androgen production in the adrenal glands and in the tumor itself, being reported to result in a survival benefit in CRPC patients57. These two former drugs, abiraterone and enzalutamide, were at the time of their development major breakthroughs for the lethal CRPC, that until then was limited only to the chemotherapeutics docetaxel and cabazitaxel26,27,32,59. Unfortunately, development of resistance to these newer and improved antiandrogens eventually happens leaving CRPC patients with restricted therapeutic options60. More recently, some advances were made with Poly (ADP-ribose) polymerase (PARP) inhibitors, namely rucaparib and olaparib, and antiProgrammed cell death protein 1 (PD1) inhibitor, pembrolizumab, that were approved for a subset of 8 these patients61,62. Several mechanisms that confer resistance were uncovered and comprise two main types, ARdependent and independent mechanisms. Initially it was thought that CRPC was completely independent of AR signaling, however it was demonstrated that in the majority of the cases, CRPC had restored AR signaling which was confirmed by efficacy of the 2nd generation antiandrogens for its treatment63. Mechanisms that restore AR signaling after ADT/CAB treatment are the most common, and comprise AR amplification/overexpression, intratumoral androgen synthesis, AR mutations, ligand independent AR activation, aberrant expression of AR coregulators, expression of AR splice variants (ARvs), and glucocorticoid receptor (GR) upregulation30,32,40,51,64. This plethora of events highlight the importance of AR in PCa carcinogenesis, aggressiveness and metastasis. As mentioned before, CRPC can be totally independent of AR signaling, being the activation of receptor tyrosine kinases (RTKs) and common oncogenic pathways65,66, the epithelial-mesenchymal transition (EMT), acquisition of stem cell properties and neuroendocrine transdifferentiation (NEtD) examples of this kind of mechanisms that overall demonstrate the high plasticity of PCa cells64,67. The concept of EMT was first introduced in 1968 by Elizabeth Hay, that later in 1995 described and defined EMT as a reversible developmental process during which epithelial cells are converted into invasive mesenchymal cells68 (Figure 4), being also involved in processes like inflammation, wound healing and tissue regeneration69,70. More recently, EMT has been identified as a crucial event in cancer invasion and metastasis 4,70, being associated with therapy resistance and cancer stem cells (CSCs) that altogether contribute to a more aggressive state of disease71,72. During this complex process, epithelial tumor cells lose the expression of proteins involved in cell-to-cell adhesion, such as Epithelial-cadherin (E-cadherin) and integrins, and gain expression of proteins typically associated with mesenchymal cells, including Fibronectin, Neural-cadherin (N-cadherin), and Vimentin69,70. The phenotypic switch results in enhanced tumor cell motility and invasiveness and, as a consequence, tumor cells undergoing EMT are thought to be able to detach from the primary tumor and to initiate the cascade of events leading to the establishment of metastasis4,69,70. Regarding the control of this process, it is known that some transcription factors, usually referred as key masters of EMT, like Snail, Slug, Twist1, and Zeb1/2 are activated to promote the molecular changes that occur during the EMT program69,70,73-75 (Figure 4). These transcription factors are in turn controlled by signaling pathways such as nuclear transcription factor κB (NF-κβ), Wnt, Notch, Hedgehog, Activator protein 1 (AP-1), and growth factor signaling, that converge at the level of the aforementioned transcription factors69,70. In the specific case of PCa, it was shown that ADT promotes EMT 9 in animal models as well as in tumor samples of patients after hormonal therapy71,76. Additionally, the previously mentioned ARvs that emerge as an adaptive response to therapy and are increased in CRPC cases, have a regulatory role in EMT as well77. Another interesting aspect of EMT is that it seems to promote the acquisition of stem cell properties by tumor cells72,78. This capacity of the EMT program to generate what some authors affirm to be CSCs72 poses several threats, namely increased migration and invasion, the capacity to form new tumors and the intrinsic resistance to conventional therapies demonstrated by this kind of cells, which ultimately results in cancer metastasis and relapse79,80. Figure 4. Schematic representation of EMT at a molecular level and its role in cancer progression. Induction of EMT leads to the expression of specific transcription factors, such as ZEB, Snail, Slug and Twist, which result in the decrease of epithelial markers and in the increase of mesenchymal ones, leading to the disassembly of epithelial cell–cell junctions and of apical–basal cell ultimately resulting in higher motile and invasion capabilities. Cancer cells undergoing EMT can disseminate from the primary tumor site, migrate into a new location in the body, revert to the epithelial state by undergoing mesenchymal-to-epithelial transition (MET) forming a metastatic lesion. EMT has also been associated with stemness features and therapy resistance, leading to recurrence and a poor prognosis. Adapted from 70,81. Another mechanism associated with therapy resistance in PCa is the NEtD process64,67,82, that occurs in about of 17-25% of patients with mCRPC83,84. Various studies demonstrated that high grade and high stage prostate tumors, specially CRPC, possess a NE phenotype that is associated with poor prognosis64,82,84,85. There is some doubt about the origin of these kind of tumors, but increasing evidence suggests that originate from ADC cells rather than being a de novo secondary tumor, meaning that cancer cells undergo NEtD, acquiring a similar phenotype of the normal NE prostate cells82,85, which is supported by the fact that these NE tumor cells exhibit the same genetic profile as the non-NE cancer cells82,85,86. Tumor cells that had undergone NEtD become more elongated with a neuron-like phenotype, gain cytoplasmic secretory granules, undergo growth arrest, lose AR expression and express higher levels of NE markers such as synaptophysin, chromogranin A, neuron specific enolase (NSE), gastrin and neurotensin82,84-87. In the specific case of PCa it has been shown that patients who had received a longer 16 The ancient T-box family of transcription factors shares a highly conserved DNA binding domain, known as T-box domain, and the members of this family play a key role in embryonic development which is highlighted by the fact that mutations in these genes are associated with anatomical abnormalities and drastic embryonic phenotypes153. Brachyury (or TBXT ) is the founding member of this ancient family and in 1927154 it was found that mutations in this gene in mice resulted in dead in utero in null homozygotes and in a short tail phenotype in heterozygotes155. Only later, after cloning and sequencing it was classified as a DNA-binding protein156-158, the T-domain, was unraveled157 and since these discoveries several Brachyury orthologues in other species were identified, including in humans159-162. The human TBXT gene is located on chromosome 6q27 and consists of eight exons that spans 10 kb, and its open reading frame encodes a protein of 435 amino acids that shares 91% identity with the mouse ortholog162. Brachyury has a pivotal conserved role in embryonic development of vertebrates, specifically in cell movements during gastrulation, mesoderm formation and differentiation, notochord development in chordates and even plays a role in extraembryonic tissues (Figure 7), like the allantois that ultimately gives rise to the umbilical cord162-167, explaining this way why homozygous mutants die in utero . At an early embryonic phase Brachyury is expressed in mesoderm precursors in the primitive streak during gastrulation and later becomes restricted to the notochord, that ultimately develops into the vertebral column in vertebrates163,164,168. These important functions in the proper embryo development were discovered mainly through studies with mutant organisms and lossand gain-of-function experiments and in accordance with the mutant phenotype firstly described in mice, similar phenotypes were observed in other vertebrates as well, demonstrating in fact the conserved role in embryonic development of Brachyury169-172. Later it was found that genetic aberrations in human Brachyury are associated with spinal cord defects such as sacral agenesis173, congenital scoliosis174, neural tube defects and spina bifida175, and also chordoma176-180, which is a rare bone tumor that is thought to arise from notochord remnants176,180. 1.4.1. Brachyury and cancer As mentioned above, genetic alterations in the Brachyury gene, such as single nucleotide polymorphisms (SNPs) and gene duplications, are associated with an increase susceptibility to develop chordoma and are often detected in familial and sporadic tumors of this kind. Moreover, Brachyury seems to be expressed in virtually all chordoma patients, having an oncogenic role in this notochord-derived tumor and is considered part of the molecular identity of chordomas, being used as a specific biomarker for diagnostic purposes176,181-183. 17 The association between chordoma and Brachyury was the first link suggesting and supporting an implication of this molecule in cancer, drawing attention of scientists to further study Brachyury in contexts other than embryonic development. One of the first studies that reported an association between Brachyury and cancer was conducted by Palena and collaborators, that using a computer-based differential display (CDD) analysis tool demonstrated that Brachyury is expressed in several carcinomas and is undetectable in most human normal tissues, which was then confirmed at an mRNA level184. In fact, since then a multitude of studies reported this tumor associated pattern of Brachyury, at mRNA and protein levels, being undetectable in most normal tissues, except for testis, thyroid and a few small cell populations detected in some studies185-189, and overexpressed in several malignancies that affect the lung, esophagus, stomach, small intestine, kidney, bladder, uterus, testis, breast, and prostate185-197, and cancer cell lines184-186,189,190,197-199. Through several studies, Brachyury was strongly associated with oncogenic properties, for instance viability, proliferation, migration and invasion in vitro and metastasis formation in vivo 187,189,190,195,196,200-205. Furthermore, Brachyury was correlated with various clinical aspects namely, tumor grade, stage of disease, recurrence and distant metastasis185-187,189,190,193-196,206-208, and shown to be a predictor of poor prognosis in a number of neoplasms such as hepatocellular carcinoma (HCC), gastrointestinal stromal tumors (GISTs), high-risk testicular germ cell tumors, oral squamous cell carcinoma, colorectal, breast, lung and prostate cancers185-187,189-194,207,209,210. Moreover, Brachyury expression was not only detected in primary tumors but also in metastatic tissue200,209, and a limited number of studies reported that Brachyury expression is even higher in metastasis than in the primary tumor186,190,210, which points Brachyury as a key player in metastization. These reports gave insight on Brachyury’s role in cancer and prompted a greater interest of research in this molecule in the oncology field. One of these first studies, conducted by Fernando et. al195, demonstrated in lung and pancreatic cancer cell lines that Brachyury is an EMT inducer given that its overexpression increased the expression of mesenchymal markers (Snail, Slug, Vimentin, N-cadherin, etc.), decreased the expression of epithelial ones (E-cadherin, Plakoglobin, ZO1, etc.) and in accordance the migratory and metastasis formation capacities were also increased. These interesting events were further corroborated by other studies in different models, namely HCC, adenoid cystic carcinoma, colorectal, breast, lung and prostate cancers186,187,189,190,196,197,204,207,208,211 in vitro and more importantly, in a clinical context through tumor tissue analysis189,190,193,205,210,211. Another interest aspect of Brachyury in cancer is its association with stemness: Brachyury overexpression results in an increased self-renewal capacity, measured by tummorsphere formation in vitro and the extreme limiting dilution assay in vivo , and in the expression of stem cell and pluripotency markers Nanog, Oct4, Sox2, Nestin and other CSCs 18 markers186,200,211,212. Both EMT and stemness are intimately related to therapy resistance, and accordingly Brachyury correlates with therapy resistance to a number of drugs, namely cisplatin, docetaxel, cabazitaxel, tamoxifen, vinorelbine, and radiation as well in various cancer models186,187,198,203-205,213,214. In order to understand how Brachyury functions, several researchers investigated the downstream targets of this transcription factor in mesoderm formation215-217, however how and through which downstream targets Brachyury mediates its effects in cancer is less understood (Figure 7). Figure 7. Role of Brachyury in embryonic development and cancer contexts. Brachyury was first described by its vital role in embryonic development, more specifically in cell movements during gastrulation, mesoderm formation and differentiation, notochord development and it also plays a role in extraembryonic tissue formation. These are all typical EMT processes since they involve the massive conversion of epithelial cells into migratory mesenchymal cells. Brachyury mediates its functions through several pathways such as Wnt, TGF-β, NFκβ, FGFR, etc. Cancer cells hijack several pathways normally involved in embryonic development to sustain their characteristic malignant behavior, and this was demonstrated to be the case of Brachyury as well. This transcription factor has been identified in a number of malignancies, with a transcriptional program that involves a number of known oncogenic pathways and it has been strongly correlated with oncogenic properties, EMT, stemness, and therapy resistance that eventually leads to metastasis and patient relapses. Some insights regarding this subject were uncovered mainly by transcriptomic approaches performed both in silico and in vitro using cancer cell lines of different models, for example chordoma, colorectal and breast cancers, and demonstrated that Brachyury is associated with several cancer pathways involved in cell cycle regulation, production of extracellular matrix (ECM) proteins, adhesion proteins and 19 cytokines, calcium signaling pathway, steroid biosynthesis, tumor necrosis factor (TNF) signaling pathway and its gene regulatory network was also associated with the PI3K/AKT signaling200-202,206,218. Some specific molecular players of the Brachyury-mediated effects were unraveled, and it was reported that the Ecadherin promoter possesses a Brachyury binding site, and that this transcription factor functions as a repressor of E-cadherin195,219, being its effects partially mediated by Slug transcription factor195, and moreover some authors reported a positive feedback loop between Brachyury and interleukin-8 (IL-8)/IL8 receptor axis in lung cancer193,220, TGF-β in prostate and lung cancers214, FGF/FGFR in chordoma221, and also a link between Brachyury and YAP axis in chordoma, glioblastoma and lung cancer200, HIF-1α in breast cancer196, MMP12222 and p21 in lung cancer195,204, SOX5 in breast cancer208 and chordoma201, EGF in chordoma201, which all have a well-known role in cancer and may explain Brachyury’s oncogenic properties (Figure 7). Even though these studies represent an advance attained in the research of the molecular players underlying Brachyury’s actions, much more remains to be elucidated. Despite of the already stated evidence of Brachyury as a molecule with an oncogenic function in several models, some contradictory studies reported this transcription factor as a tumor suppressor in gliomas and lung cancer223,224, which suggest that the role of Brachyury might be context-dependent, which needs further addressing. 1.4.1.1. Brachyury in prostate cancer The study conducted by Larocca et al that reported the connection between Brachyury and TGFβ, using a prostate cancer cell line, was the first study to implicate Brachyury in PCa aggressiveness214. Given that a characterization of Brachyury in prostate tumorigenesis was missing, our group evaluated the role of Brachyury in this malignancy190,213. In agreement with the previously tumor associated pattern of Brachyury, it was verified in human samples that Brachyury is aberrantly overexpressed in prostate intraepithelial neoplasia (PIN) lesions, primary and metastatic PCa when compared with normal tissues190. Through in vitro studies, Brachyury showed to have some oncogenic properties: was associated with increased proliferation, viability, migration and invasion; was associated in vitro and in silico with EMTrelated molecular changes, such as a decrease of E-cadherin and concomitant increased expression of mesenchymal genes, N-cadherin, fibronectin, and Snail, and with upregulation of metalloprotease MMP14 which is implicated in ECM matrix degradation and invasion190. Thus, the group confirmed for the first time an association between Brachyury and PCa aggressiveness, as well as showed to be a predictor of poor prognosis in this malignancy190. 20 In a subsequent study, the group demonstrated that Brachyury regulates several biological mechanisms associated with PCa therapy resistance, such as EMT, stem cell properties, NEtD and AR regulation213. Brachyury shown to be a direct regulator of the strongly PCa associated molecules, AR and Alpha-methylacyl-CoA racemase (AMACR), suggesting the possibility to use this transcription factor for clinical diagnosis of PCa213. Furthermore, it was confirmed the role of Brachyury in EMT program, as pointed in the previous study190, being a direct regulator of the mesenchymal markers Snail and Fibronectin, as well as its role in stemness, since it increases prostate-spheres formation capacity and expression of the stem cell markers CD44 and CD15213. Moreover, through in silico analysis Brachyury was associated with the NE markers, chromogranin A and synaptophysin, suggesting the involvement of this transcription factor in the NEtD process, that as previously mentioned gives rise to more aggressive tumors. The T-box transcription Brachyury has been already linked to therapy resistance to chemotherapy in other models and likewise, the group demonstrated that this transcription factor promotes resistance of PCa to 1st line chemotherapeutic drugs used in the treatment of CRPC, docetaxel and cabazitaxel213. Thus, it seems that Brachyury might play a central role in therapy resistance in PCa as well as in the development of CRPC, since it correlates with EMT, stemness and NEtD, and it is a direct regulator of AR. 1.4.1.2. Brachyury in lung cancer Following chordoma, LC is probably one of the cancer models in which Brachyury has been more extensively studied. Several researchers evaluated the mRNA and protein expression of this transcription factor in LC samples (summarized in Table 2) and demonstrated a significant correlation between Brachyury and tumor stage and grade, vascular invasion, lymphatic permeation193,195,205,210, being also a biomarker of poor prognosis193,205,210. The first report in the literature regarding Brachyury in LC223 examined Brachyury expression and promoter methylation in ADC and normal lung samples and cell lines, and the data suggested a possible tumor suppressor role for this transcription factor, however several posterior studies demonstrated quite the opposite. Fernando and his coworkers195 were the first group to point out the oncogenic properties of Brachyury in this model, demonstrating that high mRNA expression of this transcription factor in LC (vs normal lung tissue) is correlated to late stage tumors; moreover, through gainand loss-of-function approaches and in vitro and in vivo assays, these researchers showed Brachyury as a driver of EMT as well as its association with migration, invasion and metastasis formation capacities. Subsequently, 21 numerous studies evaluated not only Brachyury mRNA but also protein expression in tumor samples and established various significant clinical correlations as already mentioned185,205 (Table 2), and reinforced the involvement of this transcription factor in EMT and stemness193,204,205, which altogether suggests that Brachyury plays a key role in LC aggressiveness. As previously mentioned, Brachyury is associated with cell cycle regulation, and in fact some studies performed in LC supported this connection195,204,218. Firstly, the abovementioned study carried out by Fernando et al showed that knockdown of this transcription factor in the H460 cell line increased proliferation in vitro , downregulating CUL1 and p21 and upregulating cyclin D1 levels, all involved in cell cycle, proposing that Brachyury impairs cell cycle, probably at the G1-S transition195. Concordantly, Huang and collaborators also demonstrated that Brachyury negatively regulates the cell cycle, since low levels of this transcription factor resulted in increased proliferation in vitro and tumor growth in vivo . The underlying molecular players of this effect were also studied, however in contrary to the results reported by Fernando et al, the data showed an inverse correlation between Brachyury and cyclin D1, pRb and p21, which was later found to be a direct target of Brachyury204. More recently, Xu et al218 reported dissimilar results to Fernando et al and Huang et al204, more specifically Brachyury knockdown resulted in a decrease of cellular viability and further analyses suggested a positive effect of Brachyury on cell cycle progression and in apoptosis inhibition, which highlights the need to fully elucidate the role of this transcription factor. Another interesting study, performed by Chen et al, proposed a new role for Brachyury in LC, given that it was shown through in vitro experiments that Brachyury suppresses macrophage infiltration, mediated by CCL2 and CCL4 chemokines, an association that was confirmed in lung tumor specimens199. This study suggests a role of Brachyury as a modulator of cancer tumor microenvironment (TME) that until now it was not reported in any cancer model and needs to be further addressed. The T-box transcription factor Brachyury was already associated with therapy resistance in a number of cancer models and the similarly was reported in the specific case of LC. Roselli et al showed for the first time that Brachyury overexpression is associated with resistance to EGFR targeted therapy, namely the AG1478 inhibitor, whereas the knockdown resulted in increased susceptibility to treatment with this drug in NSCLC cell lines185; in the study of Huang and collaborators, the researchers observed that overexpression of this transcription factor conferred a survival advantage in vitro to radiation treatment and to the chemotherapeutic drugs docetaxel, cisplatin, vinorelbine and the cisplatin plus vinorelbine combination, and the opposite after Brachyury knockdown was reported204; these results 22 regarding Brachyury-mediated cisplatin resistance was corroborated by Xu et al205. Additionally, Huang et al also reported an interesting finding regarding therapy resistance, in which the NSCLC A549 cell line was treated with the abovementioned drugs in a chemotherapy regimen and it was demonstrated that the survival cells possessed higher Brachyury protein levels in vitro and in vivo , which strongly points for a role of this transcription factor in recurrence of disease204. 23 Table 2. Review of the literature regarding Brachyury expression and its clinical impact in LC patients. Positive Brachyury Expression (%) Prognostic Value Clinical Correlations Molecule Analyzed ADC SCC Other/Unspecified Nontumour (Technique) Park J et al ., 2008223 Low expression: 80% (12/15) - - 100% (10/10) - - mRNA (RT-PCR) Fernando R et al ., 2010195 - - Stage I: 37.5 % (12/32) 12.5% (2/16) - Stage mRNA (qPCR) Stages II, III, IV: 62.5 % (30/48) Roselli M et al ., 2012185 48% (10/21) 25% (3/12) Undifferentiated carcinoma: 50% (2/4) 43.8% (7/16) - - Protein (IHC1) Bronchioloalveolar carcinoma: 100% (1/1) SCLC: 0 % (0/1) - - 52.5% (42/80) 12.5% (1/8)2 - - mRNA (qPCR) Haro A et al ., 2013193 - - Low expression: 25% (26/104) Yes (5-year DFS and OS rates) Vascular invasion; Lymphatic permeation; Histological grade; TNM stage mRNA (qPCR) Medium expression: 50% (52/104) High expression: 25% (26/104) - - Yes* No expression Protein (IHC)3 Xu K et al ., 2015205 55.56% (25/45) 31.43% (22/70) - 0% (0/115) Yes (OS) TNM stage; LNM Protein (IHC)4 Shimamatsu S et al ., 2016210 Low expression: 58.54% (48/82) Low expression: 45.45% (10/22) Primary tumor: low 63.64 % (7/11), high 36.36 % (4/11) - Primary tumor: Yes (OS after surgery) Histological grade Protein (IHC)3 High expression: 41.46% (34/82) High expression: 54.55% (12/22) LNM: low 48.7 % (56/115), high 51.3 % (59/115) LNM: Yes (OS) Chen S et al ., 2015199 - - Yes* - - - Protein (IHC)* Wan Z et al ., 2016222 - - Primary tumor: low 33.33 % (10/30), medium 23.33 % (7/30), high 13,33% (4/30) - - - Protein (IHC)5 Metastatic tissues: low 22.73% (5/22), medium 27.27 % (6/22), high 18.18% (4/22) Hu Y et al ., 2016225 - - 45 % in different lung tumors - - - Protein (IHC)6 ADC: adenocarcinoma; SCC: Squamous Cell Carcinoma; SCLC: Small Cell Lung Carcinoma; LN: Lymph Node; LNM: Lymph Node Metastasis; IHC: Immunohistochemistry; WB: Western Blot; TMA: Tissue Microarray; RT-PCR: semi-quantitative PCR: qPCR: Real Time PCR; DFS: disease free survival; OS: overall survival. † Comparison of expression between surrounding healthy tissue and tumour tissue out of the 16 Brachyury-positive primary tumours was performed. 1 anti-Brachyury mAb (ab57480; Abcam) (Dilution: 1:100). 2 Using a commercial panel of cDNAs obtained from 40 lung tumour tissues and 8 histologically normal lung tissues obtained from lung cancer patients. 3 anti-Brachyury rabbit polyclonal antibody (Abcam, ab2068) (Dilution: 1:100). 4 goat anti‑human polyclonal anti‑Brachyury (sc‑17745; Santa Cruz Biotechnology) (Dilution: 1:100). 5 rabbit anti‑human polyclonal anti‑Brachyury (sc‑20109; Santa Cruz Biotechnology) (Dilution: 1:400). 6 anti‑Brachyury (Santa Cruz Biotechnology) (Dilution: 1:400). * Values not discriminated on the paper 24 1.4.2. Brachyury as a therapeutic target Due to the pivotal role of Brachyury in therapy resistance and progression in several types of malignancies as well as its tumor associated pattern, therapies targeting this transcription factor are extremely attractive strategies for cancer treatment. In agreement, Robinson et al, recently highlighted the importance of targeting Brachyury in cancer and reviewed the different types of approaches to do so226 (Figure 8). A group of investigators focused on a T-cell mediated approach to target the EMT driver Brachyury. By using an MHC-peptide-binding prediction algorithm, a nonameric Brachyury epitope was selected to generate Brachyury specific T cytotoxic cells from the blood of both normal donors and cancer patients, which was later modified to improve its binding and stimulating properties227. These specific T cells were able to lyse, in vitro and in vivo , tumor cells expressing Brachyury including lung, breast, and colorectal carcinoma cells184,185,187,195,227,228. After these preliminary studies, vaccines expressing the full length human Brachyury, yeast-brachyury (GI-6301) and MVA-brachyury-TRICOM187,228,229, entered clinical trials and were the first vaccines targeting an EMT driver to successfully do that. GI-6301 and MVA-brachyuryTRICOM vaccines showed in Phase I clinical trials to be safe and capable of generate Brachyury-specific T cell immune responses in cancer patients demonstrating evidence of little clinical activity in some of the enrolled patients230,231, being at the moment under evaluation in phase II trials combined with other therapies232-234. Currently, other approaches combining Brachyury vaccines with other therapies are now recruiting patients to initiate clinical trials235,236. Similar to other transcription factors, Brachyury is not easy to directly target, and no smallmolecule inhibitor has been developed yet, however, some small molecule inhibitors have been reported to indirectly target this transcription factor. Recently, Magnaghi et. al237 showed that afatinib is the only EGFR inhibitor active in several chordoma cell lines (EGFR signaling driven). As previously mentioned, Brachyury is overexpressed in virtually all chordomas and silencing of this transcription factor in chordoma cell lines was shown to decrease tumor growth both in vitro and in vivo . Since afatinib also downregulates Brachyury protein, this is thought to be the reason underpinning the great activity of this drug against chordoma237. Furthermore, Sharifnia et. al demonstrated that the transcriptional cyclin-dependent kinase (CDK) inhibitor targeting CDK7/12/13, THZ1, also targets Brachyury, suppressing this way chordoma cell proliferation, both in vitro and in vivo 183. Both studies are an incredible advance for chordoma treatment, given the lack of effective treatment options available in the clinic, and might also be applied in other cancers that overexpress the EMT driver Brachyury, like PCa and LC. 25 Figure 8. Different strategies for Brachyury targeting. The main goal of targeting the embryonic transcription factor Brachyury in cancer is to inhibit its transcriptional program to ultimately eradicate its oncogenic functions. Strategies to target Brachyury can be direct or indirect approaches. Currently, the only available direct strategy is a T-cell mediated approach, being these anti-Brachyury vaccines under evaluation in clinical trials. Small molecules specific to directly target this transcription approach weren’t developed yet but yield tremendous potential. Indirect approaches identified so far comprise the recently identified afatinib237 and THZ1183 drugs. Adapted from 226. 1.5. Drug delivery systems Major advances are being accomplished regarding anticancer therapy, such as the discovery of novel therapeutic targets, predictive biomarkers of response and so on, however this is not sufficient to greatly improve cancer treatment and overcome the underlying problems of conventional systemic therapies, namely poor specificity, high toxicity and the development of drug resistance238. This mirrors the current need for improved therapeutic approaches and in that sense, drug delivery is an area of active and attractive research with the aim of develop more efficient and less toxic strategies. Drug delivery systems (DDS) can be defined as technologies that are designed to improve the specificity of therapeutics by stabilizing them in vivo , controlling their release, and localizing their effect238-240. These systems alter pharmacokinetics and biodistribution of the associated drugs, and are able to reduce side effects, improve bioavailability and reduce degradation in the human body, protecting them from harsh environments238-240. Due to the promising results obtained with this kind of approaches, several DDS were already approved by FDA and EMA for clinical use, and a lot more are currently being evaluated in clinical trials241. In the biomedical field, it is worth to mention that these DDS besides having therapeutic applications can also be used in diagnostic settings241,242. Currently, there is a plethora of materials that can be used for drug 32 full cDNA of TBXT is inserted together with Geneticin (G418) resistance gene, that further is exploited for selection of successfully transfected cells. LNCaP, 22RV1, H292 and HCC827 cells were plated on 6-well plates at a density of 5x105 cells per well in RPMI/DMEM 10% FBS and allowed to adhere overnight. In the next day, transfection was done using the FUGENE HD reagent (Roche) according to the manufacturer’s protocols, with 2 µg of the plasmid at a ratio of 6:2 (reagents:plasmid), in serum free Opti-MEM media. After 48 hours, stable LNCaP, 22RV1, H292 and HCC827 cell pools with TBXT expression were maintained with 500-800 µg/mL G418 treatment. The Empty vector was used as control and as such in this study the resultant cell clones from transfection with this vector will be further designated as Empty and the ones resulting from transfection with the vector containing the TBXT full cDNA will be called TBXT. 3.4. Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) analysis The cells were plated on a 6-well plate at a density of 1x106 cells per well and allowed to adhere overnight. In the next day, the cells were subjected to more 24 hours in 10% FBS culture media. The total RNA was isolated using TRIZOL reagent (GRiSP Research Solutions) according to the manufacturer’s instructions, through a chloroform-based phase separation and isopropanol RNA precipitation. One microgram of RNA was reverse transcribed into cDNA with the Xpert cDNA Synthesis Mastermix (GRiSP Research Solutions). The expression levels of several genes were then assessed by qRT-PCR and the primers used are presented in Table 4. Real-time PCR was conducted by using 200 ng of cDNA as template and the reagent SsoFast™ EvaGreen® Supermix (Bio-Rad), using the Thermal cycler CFX96 (BioRad). The thermocycler program used was as follows: 95°C for 10 minutes for enzyme activation, and 40 cycles at 94°C for 15 seconds for denaturation, 58°C for 30 seconds for annealing and 72°C also for 30 seconds for extension. PCR mixture without the cDNA template was used as negative control and β-actin was used as an internal control to normalize gene expression. Data was analyzed using the formula: Ratio=2∆Ct. Table 4. Primers sequences used for qRT-PCR. Transcript target Forward primer Reverse primer Snail 5’-CTCTAGGCCCTGGCTGCTAC-3’ 5’-TGACATCTGAGTGGGTCTGG-3’ Slug 5’-CTTTTTCTTGCCCTCACTGC-3’ 5’-ACAGCAGCCAGATTCCTCAT-3’ E-cadherin 5’-TGCCCAGAAAATGAAAAAGG-3’ 5’-GTGTATGTGGCAATGCGTTC-3’ β-catenin 5’-GAAACGGCTTTCAGTTGAGC-3’ 5’-CTGGCCATATCCACCAGAGT-3’ Vimentin 5’-GGGACCTCTACGAGGAGGAG-3’ 5’-AAGATTGCAGGGTGTTTTCG-3’ 33 Nanog 5’-ATACCTCAGCCTCCAGCAGA-3’ 5’-CTGGGGTAGGTAGGTGCTGA-3’ Oct4 5’-GCTCCTGAAGCAGAAGAGGA-3’ 5’-CTCCAGGTTGCCTCTCACTC-3’ Sox2 5’-GAGAACATGCTCTTGGCACA-3’ 5’-GCACATCTCTGCCAGTTGAA-3’ β-actin 5’-GGACTTCGAGCAAGAGATGG-3’ 5’-AGCACTGTGTTGGCGTACAG-3’ 3.5. Western Blot analysis The cells were plated on a 6-well plate at a density of 1x106 cells per well and allowed to adhere overnight. In the next day, the cells were serum starved for two hours, and when necessary, two hours followed by a 24-hour treatment with drugs. In some experiences when necessary, PCa cells were also stimulated with 10 nM of DHT for 24 hours and LC cells were stimulated with 10 ng/ml of EGF for 15 minutes in 0.5% FBS medium. To obtain the protein extracts the cells were washed with Phosphate-Buffered Saline (PBS) and then scrapped in lysis buffer containing 50 mM of Tris (pH 7.6–8), 150 mM of NaCl, 5 mM of EDTA, 1 mM of Na3VO4, 10 mM of NaF, 10 mM of sodium pyrophosphate, 1% of NP-40 and 1/7 of protease inhibitors (Roche, Amadora, Portugal). After a centrifugation of 13000 rpm for 15 minutes, total protein was quantified using the Bradford method (Sigma-Aldrich). Aliquots of 40 μg of total protein from each sample were separated on 8/10% polyacrylamide gel by sodium dodecyl sulfate polyacrylamide gel electrophoresis (100V) and transferred onto a nitrocellulose membrane (Amersham Biosciences) in 25mM Tris-base/glycine buffer using the Trans-Blot Turbo Transfer System (25V, 1A for 30 minutes). The membranes were blocked with milk 5% Tris-Buffered Saline/0.1% Tween (TBS-Tween) for 1 hour at room temperature (RT) and incubated overnight with the primary antibodies at 4°C (Table 5). Next, after washing in TBS-Tween, the membranes were incubated with the respective secondary antibody coupled to horseradish peroxidase (1:2500, Cell Signaling). Tubulin was used as loading control. Blots detection was done by chemiluminescence (Supersignal West Femto kit, Pierce, Thermo Scientific) using the Sapphire Biomolecular Imager (Azure Biosystems). Table 5. Primary antibodies used for Western Blot. Protein target Reference Dilution (Secondary antibodies) Brachyury D2Z3J (CS) 1:500 (Rabbit) Snail C15D3 (CS) 1:1000 (Rabbit) E-cadherin 24E10 (CS) 1:1000 (Rabbit) N-cadherin D4R1H (CS) 1:1000 (Rabbit) Vimentin D21H3 (CS) 1:1000 (Rabbit) β-catenin D10A8 (CS) 1:1000 (Rabbit) 34 FAK 3285T (CS) 1:1000 (Rabbit) Paxilin D9G12 (CS) 1:1000 (Rabbit) Talin C45F1 (CS) 1:1000 (Rabbit) Vinculin 4650T (CS) 1:1000 (Rabbit) YY1 SC-7341 1:1000 (Mouse) AR ab108341 1:1000 (Rabbit) p-EGFR (Tyr1068) D7A5 (CS) 1:1000 (Rabbit) EGFR D38B1 (CS) 1:1000 (Rabbit) p-AKT (Ser473) D9E (CS) 1:1000 (Rabbit) AKT C67E7 (CS) 1:1000 (Rabbit) p-MEK1/2 (Ser217/221) 41G9 (CS) 1:1000 (Rabbit) MEK1/2 L38C12 (CS) 1:1000 (Mouse) p-ERK1/2 (Trh202/tyr204) D13.14.4E (CS) 1:2000 (Rabbit) ERK1/2 137F5 (CS) 1:2000 (Rabbit) p21 2947 (CS) 1:1000 (Rabbit) α-Tubulin SC-73242 1:2500 (Mouse) CS: Cell Signaling Technology; SC: Santa Cruz Biotechnology; ab: Abcam 3.6. Immunofluorescence analysis The cells were seeded on glass cover slips placed on 12-well plates until ~60% of confluence and were allowed to adhere overnight. The cells were then fixed and permeabilized in cold methanol for 5 minutes. After blocking with Ultra V Block solution (Thermo Scientific) for 10 minutes, the cells were incubated overnight at 4°C with the primary antibody for Brachyury (1:200, #81694, Cell Signaling). After washing in PBS, the TRITC Alexa Fluor-conjugated secondary antibody (Molecular Probes, Invitrogen) was used at a dilution of 1:500 for 1 hour at RT protected from light. Finally, after washing in PBS, cells were mounted in Vectashield Mounting Media with 4’,6-diamino-2-phenylindone (Sigma) and images were obtained with a fluorescence microscope (Olympus BX61), using Cell P software. 3.7. Cellular viability assay To assess cellular viability overtime, H292 cells were plated on 48-well plates in triplicate at a density of 3x104 cells per well and allowed to adhere overnight in DMEM 10% FBS. In the following day, the cells were submitted to DMEM 0.5% FBS and let incubate for 24, 48 and 72 hours. The total biomass was quantified in time zero after the cells were fixed with cold 10% trichloroacetic acid (TCA) for at least 1 hour at 4ºC and stained with Sulforhodamine B (Sigma-Aldrich) for 30 minutes. To remove the excess of dye, cells were repeatedly washed with 1% acetic acid and protein-bound dye was dissolved in 10mM of Tris-Base solution (pH=10.5) for absorbance measurement at 490nm using the Thermo-Scientific Varioskan Flash SkanIt software (Thermo-Scientific). The same was done to quantify the remaining time 35 points. The results were calibrated to the starting value (time 0 hours, considered as 100% of proliferation) and expressed as the mean ± SD. The assay was done in triplicate at least three times. To perform all cytotoxicity assays, cells were plated on 96-well plates at a density of 5-8x103cells per well, depending on the cell line, and allowed to adhere overnight in DMEM/RPMI 10% FBS. On the next day, the cells were treated with increasing concentrations of the drugs, zeolite/mesoporous silica and DDS or with DMSO alone (in the case of the drugs), both diluted in 0.5% FBS culture medium. After 72 hours, cell viability was quantified using the Cell Titer96 Aqueous cell proliferation assay (Promega). The results were expressed as the mean percentage ± SD of viable cells relative to the DMSO or medium alone (considered as 100% viability). The IC50 was calculated by nonlinear regression analysis using GraphPad Prims software version 8. 3.8. Wound Healing migration assay H292 cells were seeded on 6-well plates and cultured to at least 95% confluence. Monolayer cells were scrapped with a plastic 1000μl pipette tip, washed with PBS and then incubated with fresh DMEM 0.5% FBS medium. The “wounded” areas were photographed after 12, 24, 48 and 72 hours by phase contrast microscopy using the Olympus IX53 microscope. The migration distance was measured using the beWound software (version 1.7, BeSurg) and the relative migration distance was calculated by the following formula: 𝑊 (%)=𝑊0−𝑊𝑡 𝑊0×100, where W0 is the width of the cell wounds before incubation, and Wt is the width of cell wounds after incubation. The results are expressed as the mean ± SD. The assay was done in triplicate at least three times. 3.9. Clonogenicity assay H292 cells (750 cells/well) were seeded on 12-well plates and incubated overnight to adhere. Medium was replaced for fresh DMEM 0.5% FBS and was let incubate for 8-12 days, with medium renewal after 3 days. The colonies were fixed with cold methanol for at least 10 minutes at -20ºC and stained with 5% Crystal violet for 30 minutes and manually counted. Results were expressed as the mean colonies ± SD. The assay was done in triplicate. 3.10. Drug Delivery Systems preparation For the DDS development three porous materials, NaY, MCM-41 and SBA-15 were used, and only two were selected to encapsulate each drug (afatinib or THZ1) taking into consideration their molecular structure (Figure 11). The preparation of DDS based in porous materials was carried out based on a previously established method259,265,266, that is simply a encapsulation method. Firstly, NaY, MCM-41 and 36 SBA-15 materials were dehydrated at 120°C for 2 hours in order to remove water from the pores, which is important given the poor water solubility of afatinib and THZ1. So, in order to prepare afatinib DDS, 200 mg of NaY and SBA-15 were each one added to a solution of 10 mg (20.58 µmol) of afatinib in 10 mL of ethanol; THZ1 DDS, 200 mg of MCM-41 and SBA-15 materials were added to a solution of 5 mg of THZ1 (7.825 µmol) in 10 mL of acetone. The resulting suspensions were stirred (100 rpm) for 48 hours at RT. In this case the suspensions were not filtered nor heated in order to minimize drug loss, instead they were allowed to sediment and the major part of the solvent was evaporated at RT during 48120 hours. After this all the DDS were dried in an oven at 60 °C for 48 hours in order to fully evaporate the solvent, and further stored in a desiccator. Figure 11. Molecular structure of Afatinib and THZ1. 3.11. Statistical analysis Statistical analysis was done using GraphPad Prism 8 version. The level of significance in all statistical analysis was set at p<0.05. Student’s t-test was used to do single comparisons between two different conditions. Afatinib THZ1 37 CHAPTER 4: Results 38 4.1. Biological and predictive role of Brachyury in Prostate Cancer 4.1.1. Role of Brachyury in modulation of EMT proteins in prostate cancer cell lines The first main aim of this work was to study the impact of Brachyury expression in antiandrogen therapy response in PCa. For that we chose two androgen-dependent cell lines, namely LNCaP and 22RV1, which were described by our group190 and others195 as positive and negative, respectively, for Brachyury expression. Thus, we genetically modulated Brachyury, in order to achieve an OE of our molecule of interest, using an expression vector that was different from the one used in the first report of our group190. By immunofluorescence analysis (Figure 12), it was possible to observe a low cytoplasmic expression of Brachyury in LNCaP cells transfected with the empty vector (Empty cells), which is undetectable by western blot (Figure 13). The cells transfected to overexpress Brachyury (TBXT cells), presented high levels of this protein in the nucleus, which was indicative of a successful and functional transfection, and of a specific antibody, since Brachyury OE in the two transfected cell lines, was also confirmed by western blot (Figure 13A and 13B). Furthermore, it is important to notice a heterogeneous population of TBXT cells, in an expression level context, which is due to the fact that these cells were stably transfected but selected with antibiotic treatment instead of single cell cloning or cell sorting. Figure 12. Immunofluorescence analysis of Brachyury expression in transfected LNCaP cells. Brachyury expression was analyzed in the control (Empty cells) and Brachyury OE (TBXT) cells by immunofluorescence analysis (20X and 40X magnification), in which the red and blue signals represent Brachyury and DAPI stained nucleus, respectively. 39 Further, as in this work we used either a different expression vector and antibody for Brachyury detection, we intended to validate the success of our transfection by recapitulating some of the previous results obtained by the group in the 22RV1 cell line (Figure 13A), for which was found a direct correlation between Brachyury mRNA expression and EMT related genes190,213. Herein we exposed the cells to different culture conditions, 0.5% FBS and 10% FBS, as well as treated them with 10 nM of DHT to stimulate AR, and determined the expression of both Brachyury and EMT related proteins (Figure 13). Firstly, it is interesting to note that Brachyury expression increased over the stimulating conditions in 22RV1 cell line, but not in LNCaP cells (Figure 13A and 13B). Secondly, as it can be observed in Figure 13A, the 22RV1 epithelial cell line does not express several EMT related proteins and for the ones that they do express, E-cadherin and β-catenin (epithelial markers), a slight decrease was found between Empty and TBXT cells (Figure 13A). Regarding LNCaP cell line, for which there is no data concerning Brachyury’s effects in EMT genes, we analyzed both protein (Figure 13B) and mRNA expression levels (Figure 13C). In this cell line, Brachyury OE resulted in upregulation of YY1 and β-catenin proteins, in this last case at mRNA level as well (Figure 13C). Vimentin protein level was also increased, but only in the control (CTR) condition (0.5% FBS), not being confirmed at mRNA level. In contrast, E-cadherin was found slightly downregulated at mRNA level (Figure 13C), but slightly overexpressed at protein level (Figure 13B). Regarding the two transcription factors that regulate EMT, Snail and Slug, LNCaP cells do not express them at a protein level, but they can be detected at mRNA level, however no differences were noted between Empty and TBXT cells (Figure 13C). Complementary, we also evaluated some focal adhesion (FA) associated proteins, namely FAK, paxillin, talin and α-actinin and a significant downregulation of talin protein was found upon Brachyury upregulation, indicating a potential loss of adhesion properties. 40 Figure 13. Characterization of Brachyury, AR, FA-, and EMT-related molecules in 22RV1 and LNCaP transfected cell lines. A) Western blot analysis of AR and of the epithelial markers E-cadherin and β-catenin in transfected 22RV1 cells. B) Analysis of AR, epithelial (E-cadherin) and mesenchymal (vimentin) markers, YY1 transcription factor and FA associated proteins (FAK, paxillin, talin and α-actinin) in our genetically modulated LNCaP model. All these analyses were conducted in 0.5% FBS, 10% FBS RPMI media and in DHT (10 nM, 24 hours) stimulating conditions. Tubulin was used as a loading control. The western blots presented here are representative assays of two independent experiments. C) Real time analysis (N=1) of EMT related genes in Empty and TBXT LNCaP cells (10% FBS growth condition). The experiment was done in triplicate and presented as relative expression in relation to β-actin. Finally, given the reported positive association and identification of AR as a target of Brachyury in PCa213, we also analyzed the AR expression levels in our models in basal and stimulating conditions (Figure 13A and 13B). Taking a closer look to the results and in agreement to the previously reported, Brachyury OE upregulated AR (full length) in LNCaP cells for all tested conditions, especially under stimulating conditions (10% FBS and DHT). However, the same was not observed in the 22RV1 cell line, in which TBXT cells had an increased full length AR expression in the conditions 0.5% and 10% FBS, but not in the stimulated one (DHT), where AR is slightly downexpressed in comparison to Empty cells. Altogether, even though some apparent contradictory results were found in comparison to the group’s published results, we herein show that Brachyury OE can induce EMT and AR expression changes 41 in the chosen cell lines, however, further analysis are in need, mainly regarding the biological behavior of TBXT cells that we established here (as invasion and migration assays). Importantly, by all the expression assays we present, and many other validation tests we conducted (data not shown), we have complete confidence in the overexpressing cell lines we generated as well as in the new commercial Brachyury antibody we chose, which allowed us to move to the next aim of the work with certitude. 4.1.2. Brachyury’s effect in the modulation of prostate cancer cells response to AR inhibitors As mentioned in the beginning and given the evidence of AR as a target of this transcription factor and the positive association between these two molecules in PCa213, which we confirmed mainly in the more androgen-dependent cell line, LNCaP (Figure 13B), we hypothesized that Brachyury could be a potential predictive biomarker to antiandrogen therapy response in PCa. To do so we performed cytotoxic assays with LNCaP and 22RV1 cells using several drugs widely used in the clinic, namely the 1st generation antiandrogens flutamide and bicalutamide, the 2nd generation antiandrogen enzalutamide and the CYP17A1 inhibitor abiraterone (Figure 14A). Analyzing Figure 14A, the sensitivity of these cell lines to the selected drugs is quite dissimilar, especially for bicalutamide, enzalutamide and abiraterone. Given that 22RV1 cell line, besides expressing the full length AR, also expresses a AR truncated version (ARv)282 (Figure 14B) which is constitutively active and generally associated with resistance to antiandrogens32,51,64, explaining in this way the sensitivity differences among the two cell lines. Unexpectedly, 22RV1 cell line was more sensitive to abiraterone than LNCaP cells and both were unresponsive to flutamide (Figure 14A). Through these assays we were able to determine the IC50 concentrations of each drug, when possible, being these values summarized in Table 6. Looking at these results it is extremely clear that Brachyury OE did not impact the response to these androgen targeted therapies, since no statistically significant differences on IC50 values between Empty and TBXT cells were found (Figure 14A and Table 6). 48 Figure 17. Characterization of FA-, EMTand stemness-associated molecules in transfected LC cells. A, B) Western blot analysis of several EMT related proteins such as epithelial (ZO1, β-catenin and E-cadherin) and mesenchymal markers (N-cadherin, Vimentin) and one of the key players in EMT regulation (Snail) and also FA associated proteins (FAK, vinculin, paxillin) in H292 (A) and HCC827 (B) transfected cells, at basal and stimulating conditions (10% FBS and EGF 10ng/ml). Tubulin was used as loading control. C) Real time analysis (N=1) of EMTand stemness-related genes in H292 Empty and TBXT cells (10% FBS growth condition). The experiment was done in triplicate and presented as relative expression in relation to β-actin. 4.2.2. Brachyury’s effect in the modulation of lung cancer cells response to EGFR inhibitors To conclude this section of the work concerning the LC model, we aimed to dissect whether Brachyury can modulate LC cells response to EGFR targeted therapies. For that we chose to use some clinically relevant EGFR inhibitors, such as 2nd and 3rd generation EGFR inhibitors, namely afatinib, AST1306 (allitinib) and osimertinib, respectively. Cytotoxic assays were performed using those drugs in 49 Empty and TBXT clones of both H292 and HCC827 cell lines (Figure 18). As evidenced in Figure 18, the two cell lines are differentially sensitive to EGFR inhibitors, which is in accordance to their EGFR mutational status, with HCC827 cell line being more responsive. Concerning the drugs actions, AST1306 was the most potent drug to H292 cell line, with a lower IC50 value (Table 7), while for HCC827, afatinib was the one that demonstrated to be the most potent, not excluding the possibility of osimertinib being as potent as the latter mentioned 2nd generation EGFR inhibitor since it was not possible to determine the osimertinib IC50 value (Figure 18 and Table 7). Comparing the response between Empty and TBXT cells, no differences were detected, except in the case of AST1306 inhibitor, for which H292 TBXT cells showed a significantly higher IC50 value than Empty cells (Figure 18B and Table 7). It should be noted that only one assay was performed with the HCC827 cell line and that the dose scale for afatinib and osimertinib drugs are not fully optimized, but even so these preliminary data (Figure 18A) showed that Brachyury had no effect on HCC827 cells response to EGFR targeted inhibitors. Table 7. Mean IC50 values for afatinib, AST1306 and osimertinib in transfected LC cell lines. For H292 cells, IC50 values (µM) are presented as the mean ± SD of, at least, three independent assays, whilst the IC50 values (nM) for HCC827 cells were determined from only one assay. All assays were performed in triplicate. Mean IC50 H292 (µM) HCC827 (nM) Empty TBXT Empty TBXT Afatinib 2.599±0.219 2.606±0.558 10.59 10.36 AST1306 1.422±0.222 2.324±0.272 64.95 37.72 Osimertinib 2.840±0.403 2.436±0.276 <100 <100 50 Figure 18. Evaluation of our genetically modulated H292 and HCC827 LC models response to EGFR targeted therapies. A) To assess the cytotoxicity of afatinib, AST1306 and Osimertinib, Empty and TBXT cells were treated with increasing concentrations of these drugs for 72 hours, and the cell viability was measured by MTS assay. The graphs are represented as the mean±SD (relative to DMSO alone) and are representative assays of three independent assays for H292 cells, done in triplicate. B) Comparative analysis of EGFR inhibitors IC50 values for Empty and TBXT H292 cells. 51 Given the differences of response between Empty and TBXT cells, in particular with AST1306 in H292 cells, we next aimed at giving some insight regarding the mechanisms underlying this divergent response, by determining the basal activation levels of EGFR signaling, namely AKT and MAPK pathways, in our genetically modulated H292 and HCC827 models and under different culture conditions (Figure 19). As expected, in FBS and EGF stimulating conditions the levels of p-AKT, p-MEK1/2 and p-ERK1/2 were upregulated compared to basal conditions (0.5% FBS medium). Furthermore, as previously mentioned, HCC827 cell is EGFR mutant, having a constitutive activation of both EGFR (Figure 15B) and its signaling compared to H292 cells, EGFR WT (Figure 19). Focusing on H292 cell line results, it can be noted that TBXT had higher levels of AKT, MEK1/2 and ERK1/2 activation, even at basal conditions. When looking at HCC827 cells results, they also present the same tendency concerning p-ERK1/2 and p-MEK1/2 , but herein TBXT cells did not present increased levels of p-AKT when compared to the control Empty cells (Figure 19). Thus, the overactivation of AKT found in H292 TBXT, but not in HCC827 TBXT cells, suggests us that AKT pathway overactivation could be one of the major players underlying the lower responsiveness of H292 TBXT cells to AST1306 inhibitor. Figure 19. Characterization of EGFR signaling in H292 and HCC827 transfected cell lines. The activation of some AKT and MAPK pathways proteins, namely AKT, MEK1/2 and ERK1/2, was analyzed by western blot at 0.5% FBS, 10% FBS and EGF (10 ng/ml) stimulating conditions (15 minutes). Representative assays of two independent experiments. Tubulin was used as loading control. Taken altogether, LC cells with Brachyury overexpression have an overactivation of important signaling pathways in this pathology, which can result in a lower sensitivity of these cells to EGFR inhibitors. However, these are preliminary studies that deserve to be dissected in the near future. 52 4.3. Targeting Brachyury expression in cancer 4.3.1. In vitro evaluation of the impact of small molecule inhibitors in Brachyury expression Given the recent findings of two small molecule inhibitors, afatinib and THZ1, that can downregulate Brachyury expression in chordoma183,237, we aimed to study whether this transcription factor could also be pharmacological inhibited in PCa and LC models, in order to revert its oncogenic potential in both tumor models. To do so, we first used two colon cancer cell lines, SW620 and SW480, that endogenously express high levels of our molecule of interest195,212,284, to validate the results obtained in chordoma183,237. These cell lines were treated with both drugs in three different concentrations, that were chosen accordingly to the literature, and Brachyury protein levels were evaluated by Western Blot. As reflected in Figure 20A, both THZ1 and afatinib significantly decreased Brachyury protein levels in SW620 cell line, while for SW480 THZ1 was far more effective comparing to afatinib, suggesting that the modulation of Brachyury expression by these drugs could be cell line specific (Figure 20A). Next, we move on to our genetically modulated PCa and LC models to assess the specificity of these drugs in downregulating the ectopically-induced Brachyury expression (Figure 20B and 20C). Since our major aim from the beginning was to exploit the potential predictive role of Brachyury in the therapeutic response to antiandrogens and anti-EGFR drugs, being its role in the regulation of AR and EGFR expression always assessed, we herein decided to firstly proceed only with the cell lines that are strongly positive for those targets, LNCaP and HCC827, respectively. Focusing on our PCa model (Figure 20B), and as it can be noted, afatinib was more effective than THZ1 at targeting Brachyury, since that only the highest concentration of THZ1 used was able to decrease Brachyury expression levels. Looking at AR, it is possible to observe that these drugs can also affect its expression, both in Empty and TBXT cells, but more strongly in TBXT cells in which Brachyury was concomitantly inhibited. These findings can somehow validate that in fact Brachyury modulation has an impact in AR expression (Figure 20B). Concerning our LC model (Figure 20C) and in contrary to the results obtained with LNCaP cells, Brachyury was markedly more efficiently targeted by THZ1 than afatinib, for which the effect was similar to the one obtained with LNCaP cells (Figure 20B). However, it is important to note that both drugs were more effective at downregulating the shorter isoform of Brachyury than its full-length form. Additionally, afatinib completely abrogated the activation of EGFR in both Empty and TBXT cells, as expected. Likewise, it seems that THZ1 downactivated EGFR, but only in TBXT cells when Brachyury expression was 53 completely abolished (Figure 20C). We have previously shown that TBXT cells had an overactivation of ERK1/2 and upregulation of Snail, compared to the Empty cells (Figures 17B and 19) and herein we validate those findings, and further verified that TBXT cells retained a basal p-ERK1/2 expression, even in depletion conditions upon afatinib treatment. Similarly, both drugs were able to downregulate Snail expression, with lower efficacy in Brachyury-overexpressing cells. Intriguingly, we can see that THZ1 treatment stimulated ERK1/2 activation in both Empty and TBXT cells, a phenomenon that was already evidenced by other authors285. Thus, with this analysis we were able to conclude that these small molecule inhibitors are capable to target Brachyury, expressed endogenously and ectopically, in other models apart from chordoma. Furthermore, the results obtained here demonstrated that the Brachyury targeting of both drugs is probably dependent on the cell line genetic background. Finally, we were able to validate that Brachyury upregulated EGFR signaling, Snail and AR expression, and demonstrate that even though not being Brachyury specific, THZ1 and afatinib can somehow be used to revert the Brachyury-induced phenotype in both PCa and LC. 54 Figure 20. Analysis of Brachyury targeting by THZ1 and afatinib small molecule inhibitors in colon, prostate and lung cancer cells. A) Western blot analysis of endogenously Brachyury expression in the colon cancer cell lines SW620 (left) and SW480 (right) upon a 24-hour treatment with fixed concentrations of THZ1 and afatinib (AFA). B, C) Evaluation of ectopic Brachyury expression, and other relevant proteins associated with the Brachyury-induced phenotype (AR, EGFR, p-EGFR, Snail, p-ERK1/2), then again upon treatment with THZ1 and AFA in LNCaP (B) and HCC827 (C) cells. These are representative western blots of two independent experiments. Tubulin was used as loading control. 4.3.2. Development of zeolite and mesoporous silica based drug delivery systems Due to the promising results obtained before, in this particular part of the work we intended to encapsulate both afatinib and THZ1 in zeolite and mesoporous silica, to improve the delivery of these drugs to cancer cells. For that, we chose three different silica porous materials, NaY, MCM-41 and SBA15 with different size pores to evaluate which material would be capable to host higher quantities of the drug, in order to increase the efficacy of the respective DDS. First, we went ahead to evaluate the toxicity of the chosen parental materials in HCC827 cell line in a wide range of concentrations (0.025-0.75 55 mg/ml) and proved the biocompatibility of NaY, MCM-41 and SBA-15 materials, with no cytotoxicity observed (Figure 21A). It is noteworthy to mention that we started with limited available quantities of each drug and as such only two materials were further selected to function as matrixes for the development of THZ1 and afatinib DDS. Having in consideration the chemical structure of both drugs (Figure 11) and the porosity of the materials, we chose NaY and SBA-15 to entrap afatinib, which will allow to compare between zeolite and mesoporous silica structures; and the mesoporous silica materials MCM-41 and SBA15 as hosts for THZ1, which in terms of chemical structure is larger than afatinib. Following the development of those DDS we evaluated the cellular viability of HCC827 cells after 72 hours of treatment with increasing concentrations of THZ1 and afatinib DDS. Altogether, the results obtained demonstrated the successful development of these DDS, in other words, these host structures were able to entrap the selected drugs and were able to deliver them to the cells, ultimately resulting in a decrease of cellular viability (Figure 21B and 21C). The decrease of viability was higher with the afatinib DDS (Figure 21C) which can be due to the higher starter quantities of afatinib used in the development of the respective DDS or due to the higher sensitivity of HCC827 cell line to afatinib compared to THZ1. Given that we hadn’t yet determined the real doses of afatinib and THZ1 entrapped in these DDS we can’t conclude about which was the best host material and the efficacy of the DDS compared to the free drugs. So, further characterization work of the samples is needed in order to understand the host-guest interactions of the drugs and the silica porous materials and their viability as DDS for cancer therapy. 56 Figure 21. Biocompatibility of NaY, MCM-41 and SBA-15 materials and development of THZ1 and afatinib DDS. A) The toxicity of each parental material, NaY, MCM-41 and SBA-15, was assessed in HCC827 cell line by MTS assay upon a 72-hour incubation with increasing doses of the respective nanomaterial. B, C) THZ1 (THZ@MCM and THZ@SBA) and afatinib (Afa@NaY and Afa@SBA) DDS were developed. Their impact on cell viability was evaluated in HCC827 cells by MTS assay after a 72-hour treatment with both THZ1 (B) and afatinib (C) DDS. These are representative assays of at least four independent assays done in triplicate, being the graphs represented as the mean±SD, relative to medium alone (100% viability). 57 CHAPTER 5: General Discussion 64 functionality of the transfection in our LC cell lines, we found the reported negative association between Brachyury and p21 in H292 TBXT cells, while in HCC827 TBXT cells no association was evidenced. Cyclin kinases are proteins required for cell cycle progression, and the CKI p21, a master p53 target is well known for its role as a negative cell cycle regulator303. The decreased levels of p21 of H292 TBXT cells, could suggest a positive effect in cell cycle progression and possibly explain the advantage in cell viability of these cells, however a cell cycle and apoptosis analysis would give us more insight regarding this matter especially since dissimilar results of Brachyury role in cell cycle has also been described195,204,218. Following this, as for PCa, we analyzed the expression of EMTand FA-associated molecules at mRNA and protein levels upon transfection and while only an upregulation of Snail protein was found for HCC827 cell line, some expected and unexpected results were obtained for H292 cells. Given the significant increase of migration of H292 TBXT compared to Empty cells, it would be expected an induction of an EMT state, in other words, a decrease of epithelial markers and an increase of mesenchymal ones as well as EMT related transcription factors. In agreement to this, in H292 TBXT cells, in comparison to Empty cells, it was observed an increase in Snail and Slug mRNA and in Vimentin protein in the 10% FBS culture condition, while vimentin mRNA levels remained unchanged. Indeed, a positive association between Brachyury, Snail and Slug mRNA levels in NSCLC cell lines195,204 as well as in tumor specimens193 was already evidenced. Additionally, we also evaluated the expression of Nanog, Oct4 and Sox2 at mRNA level, which are stem-related and pluripotent transcription factors highly associated with CSCs as well as with tumorigenesis, metastasis and resistance to therapy, including to EGFR-TKIs in ADC tumors304,305. More interestingly, Brachyury has been associated with these markers in several malignancies211,212, including LC200. In accordance with these findings, in H292 TBXT cells there was an increase of mRNA levels of these CSCs markers, specially Nanog and Oct4, compared to Empty cells. These results imply that Brachyury might have a role in an induction of stemness in LC tumors, that could be a consequence of the induction of the EMT program itself or these pluripotency transcription factors could possibly be direct targets of Brachyury or only targets of its transcriptional program. Moreover, Chiou et al. reported a role of Nanog and Oct4 in EMT induction, besides their well-known role in stemness, which tightens even more the connection between stemness, EMT and the pluripotency transcription factors306. The fact that Brachyury has been associated with all these latter processes, makes it possible to realize that altogether our results are suggestive of an EMT induction, an increase in stemness and a more aggressive phenotype in H292 cells upon Brachyury OE. However, we had some unexpected results as well, such as no 65 alterations regarding N-cadherin protein levels, and also an increase in E-cadherin mRNA expression, which given that this cell line does not express this marker at a protein level and that the mRNA analysis was done only once, those results should be carefully analyzed and further validated. Unfortunately, we were not able to conduct the biological assays in HCC827 cell line, to confirm whether the aggressive Brachyury-mediated phenotype is dependent on p21 downregulation, or even EMT and stemness induction. Finally, considering our interesting preliminary results and in order to study our hypothesis of Brachyury playing a role in EGFR-TKIs resistance, we explored the impact of Brachyury in the modulation of LC cells response to EGFR inhibitors. First, we evaluated EGFR expression and EGFR activation (pEGFR) levels after transfection in three different culture conditions, for both H292 and HCC827, but no differences in total and phosphorylated EGFR levels were noted. As previously mentioned, a group of researchers already demonstrated the role of Brachyury in the modulation of response to EGFR targeted therapies185, however it should be noted that in the mentioned study it was used two KRAS mutant NSCLC cell lines (A549 and H460), which are not the more adequate models to study EGFR targeted therapy given the KRAS predictive value of poor response to EGFR inhibitors307,308. In this work we intended to use more adequate cell lines, such as the EGFR WT H292 and the EGFR mutant HCC827, that constitute better models to study EGFR targeted therapy, specially HCC827 given that nowadays only EGFR mutant patients are clinically directed to this kind of treatment. Moreover, in the work conducted by Roselli and collaborators185 it was used the AG1478 EGFR inhibitor, that possesses a similar structure and mechanism of action of the 1st generation EGFR inhibitors gefitinib and erlotinib, and then again to achieve more translational results, in this work we used more clinically relevant EGFR inhibitors, such as 2nd and 3rd generation EGFR inhibitors, namely afatinib, AST1306 (allitinib) and osimertinib. Even though allitinib is not widely used in LC at a clinical context, this inhibitor belongs to the 2nd generation of irreversible EGFR inhibitors and therefore is more potent than AG1478307. Focusing first in the H292 cell line, we were able to demonstrate that despite not being EGFR mutant, it is still very sensitive to these drugs with IC50 values ranging from 1.4-3.2 µM. Comparing the response between Empty and TBXT clones, our experiments demonstrated that Brachyury upregulation increased the resistance to AST1306 treatment with a statistically significant higher IC50 value in the TBXT clone, but this was not the case of afatinib and osimertinib. Regarding HCC827 cell line, as expected, these cells were highly sensitive to EGFR-TKIs compared to H292 cell line, with IC50 values in the nanomolar range (not determined in the osimertinib case). Bearing in mind that even though our results 66 came from only one assay and the drug scales of afatinib and osimertinib should be further optimized, this allowed us to speculate that Brachyury does not modulate the response to EGFR inhibitors, in EGFR mutant cell lines. These dissimilarities between the effect of Brachyury in our two models led us to examine some possible underlying reasons, and to do so, we evaluated the expression of proteins involved in important oncogenic pathways such as, PI3K/AKT and MAPK/MEK/ERK, that constitute part of the downstream EGFR signaling. In H292 cell line it was clear that TBXT cells had elevated levels of p-AKT, p-MEK1/2 and p-ERK1/2, which suggests that Brachyury leads to an overactivation of PI3K/AKT and MAPK/MEK/ERK pathways. Furthermore, since these pathways are known for their role in growth, proliferation, migration and so on, these last results are in agreement with the previous observed increase in cell viability, migration and colony formation abilities upon Brachyury OE. Regarding HCC827 model and differently from what it was seen with the H292 cell line, in this EGFR mutant model, Brachyury OE resulted only in a slight increase in MAPK activation at basal and stimulating conditions. Curiously, when in depleting conditions, upon afatinib treatment, HCC827 TBXT cells still expressed basal levels of p-ERK1/2 in contrary to Empty cells despite the total abrogation of EGFR activation, corroborating the finding that Brachyury sustains MAPK activation. Furthermore, comparing H292 and HCC827 cells, in the last case, Brachyury OE did not increased the levels of p-AKT when compared to the control Empty cells, as it happened in H292 cell line. Thus, the overactivation of AKT found in H292 TBXT, but not in HCC827 TBXT cells, suggest us that AKT pathway overactivation could be the one of major players underlying the lower responsiveness of H292 TBXT cells to AST1306 inhibitor, as was already described before309 Altogether, the results showed that Brachyury OE induces aggressiveness, modulation of p21, EMT, stemness and EGFR signaling activation, which could be behind the lower responsiveness of the cells to AST1306 inhibitor in the EGFR WT cell line used here, while the same phenotype was not observed in HCC827 cell line. Regarding this subject, it is important to integrate our results of the relative expression of Brachyury isoforms in HCC827 TBXT cells, as well as the genetic and inherent characteristics of this specific cell line. In other words, the fact that these cells express the short isoform over the longer one, which have been proved to potentially have different roles in cell cycle and prognosis at least302, could have had an impact on the different responses regarding EGFR inhibition we observed here, a finding that was the opposite of H292 TBXT cell line. In addition, we must keep in mind that we were limited in the number and type of assays done in HCC827 cell line, and as such we cannot totally exclude the possibility of a similar phenotype in EGFR mutant cell lines. 67 Transcription factors are often dysregulated in cancer and as such targeting these molecules that govern gene expression is a fundamental anticancer treatment. Nevertheless, given the subcellular location and lack of ligand binding domain/pocket of these molecules, strategies to specifically and effectively target most of transcription factors remained elusive until few years ago310. Even though no small molecule to direct target Brachyury has been developed yet, two studies reported two drugs that indirectly target this transcription factor in chordoma cell lines183,237. The first one to be reported was afatinib, a pan-ErbB inhibitor that besides targeting EGFR, as abovementioned it is also capable to target all ErbB receptors101. Magnaghi and his coworkers performed a screening of several TKIs in a panel of chordoma cell lines and showed that afatinib was the only inhibitor that showed activity in 6 out of 7 cell lines. After this screening the authors demonstrated that this EGFR inhibitor downregulates Brachyury protein possibly being the reason behind the great activity of afatinib in chordoma cell lines237. The second drug to be described as able to target Brachyury was THZ1, a covalent CDK inhibitor that targets CDK7, and CDK12/13 though only at higher doses311. The last main aim of this work was to study the potential of Brachyury as a therapeutic target in PCa and LC, mainly in those with high AR and EGFR expression, respectively. First, we intended to expand the abovementioned findings regarding afatinib and THZ1 effects on Brachyury expression in other models besides chordoma, using for that cell lines that endogenously express high levels of this transcription factor. Specifically, we chose two colon cancer cell lines, SW480 and SW620, that were reported in the literature as positive for Brachyury expression195,212,284. THZ1 treatment demonstrated a great ability to downregulate the expression of our molecule of interest in both cell lines, while afatinib was effective in SW620 and SW480 cell lines, but in this latter only at high concentrations. Regarding PCa, in LNCaP cells afatinib was much more effective on decreasing Brachyury protein levels compared to THZ1, in which only at a concentration of 1 µM was able to downregulate this transcription factor. Looking at AR, it is possible to observe that these drugs can also affect its expression, both in Empty and TBXT cells, but more strongly in TBXT cells in which Brachyury was concomitantly inhibited. Rasool and his colleagues shown that CDK7, the main target of THZ1, acts downstream of AR312. CDK7 inhibition by small interfering RNA (siRNA) or pharmacologically through THZ1 results in the inhibition of AR transcriptional program but not in the levels of AR itself in LNCaP and VCaP cell lines312. Even though the degree of downregulation of both Brachyury and AR was only modest, it occurs both with THZ1 and afatinib, which somehow validate that in fact Brachyury modulation has an impact in AR expression213. 68 Concerning our results in the LC model, even though afatinib being able to abrogate EGFR activation, was only able to slightly decrease the short isoform of Brachyury whilst no effect on the long isoform expression was observed in HCC827 cell line. The results suggest that Brachyury inhibition by afatinib doesn’t occur in concomitance with EGFR activation inhibition. In contrary, THZ1 was very effective at targeting both Brachyury isoforms in this cell line. Overall, our results strongly suggest a cell line dependent mechanism of the tested drugs to downregulate endogenously and ectopically expression of Brachyury, as it was observed for chordomas183,237. Magnaghi and collaborators conducted some assays with proteasome and autophagy inhibitors (MG-132 and bafilomycin) demonstrating that in these conditions afatinib was not capable to downregulate Brachyury protein in chordomas, which led them to hypothesize that the underlying mechanisms of afatinib to target Brachyury involve these pathways of protein degradation, more than EGFR inhibition237. In fact, our results are concordant with their findings given that LNCaP313 and SW620314 cells are p-EGFR and EGFR negative, respectively, and afatanib was very effective at targeting Brachyury expression in these models. Concerning our THZ1 results, in the study conducted by Sharifnia and her coworkers it was shown that the underlying mechanism of the THZ1-mediated Brachyury downregulation in chordoma was the existence of a super enhancer region associated to this transcription factor locus 183, and indeed this CDK7 inhibitor has been described for its effects in targeting transcriptional addictions in cancer and genes associated with super enhancers315-320. Moreover, through an experiment in which Brachyury was ectopically expressed under the regulation of an exogenous promoter rather than Brachyury’s endogenous regulatory elements, it was shown that THZ1 treatment barely had an impact on the ectopic protein levels183. Intriguingly, even though our PCa and LC models lack a TBXT associated super enhancer, THZ1 was still able to downregulate Brachyury expression in particular in the HCC827 cell line. The reason behind this downregulation of ectopically expressed Brachyury could be a post-translational mechanism or we can also hypothesize that the CDK7/12/13 inhibitor downregulates Brachyury through the inhibition of an upstream regulator of this transcription factor, which deserves further exploitation. Given these interesting results, we move further to our objective of using silica porous materials as nanocarriers to improve the delivery of these drugs to cancer cells. Having in mind the chemical structures of afatinib and THZ1, we chose two different zeolite and/or mesoporous silica materials for each drug, which would allow us to compare and conclude about the potentiality of these hosts. So, we used the zeolite NaY and the mesoporous silica SBA-15 to entrap afatinib and the mesoporous silica 69 materials MCM-41 and SBA-15 to encapsulate THZ1, which when used alone all showed to be biocompatible with no cytotoxic effects in HCC827 cell line. After the development of these DDS we performed cytotoxic assays again and demonstrated the successful entrapment of afatinib and THZ1 in the respective materials, and to the best of our knowledge there are no studies reporting the development of DDS using these porous materials as matrix to host any of these drugs. To conclude about the potential of the developed DDS, additional experiments are needed, such as the determination of afatinib and THZ1 load in the porous materials, that would allow us to compare the efficacy of these DDS with the free drugs and further assess if these DDS maintained the drug’s specificity to target Brachyury expression. Overall, our results demonstrated that indeed the small molecule inhibitors, afatinib and THZ1, are able to downregulate Brachyury protein levels not only in chordoma, as first reported, but in several models and curiously in both endogenously and ectopically expressed Brachyury. Furthermore, as far as we are aware our study was the first to demonstrate the inhibition of more than one of Brachyury isoforms by afatinib and THZ1. Interestingly, our results concerning the inhibition of Snail protein levels upon afatinib and THZ1 treatment point out to a possible way to revert the Brachyury-associated phenotype, albeit these drugs not being Brachyury specific. In this sense, we can in the future treat our genetically modulated models with these drugs, free or in DDS, and assess their effects on the reversion of Brachyuryinduced EMT, stemness, viability, migration, invasion, colony formation as well as use it in combination with other drugs used for PCa and LC treatment. 70 In this work the role of Brachyury in the biological behavior, therapy resistance as well as its targeting in PCa and LC models was assessed. First, we were able to expand the previous findings of the group to an androgen-dependent cell line never used before, LNCaP, and explore our hypothesis of Brachyury mediating resistance of the cells to antiandrogen drugs, concluding that this was not the case, having demonstrated that the overexpression of this transcription factor did not impact the cells response to these kind of therapies. In regard to our LC model, using two cell lines with different genetic backgrounds, never applied by others groups, the oncogenic role of Brachyury in one of these models was evidenced by the advantages in cell survival, migration and colonies formation upon Brachyury OE. In addition, since Brachyury was already reported to confer resistance to an EGFR inhibitor, we intended to further explore this using more clinically relevant models and EGFR-TKIs. We proved indeed, that in one of our LC models Brachyury conferred resistance to one out of three of the chosen EGFR inhibitors that we further hypothesized to be mediated by an hyperactivation of the AKT pathway, that we observed in these cells. Finally, we were able to demonstrate that both afatinib and THZ1 are capable to target Brachyury in other models besides chordoma and gave some insight on the potential of these drugs to reverse the Brachyury-associated phenotype. Finally, given these interesting results we move on to develop novel DDS to entrap the above cited small molecule inhibitors, but unfortunately, we were only able to prove their successful development and not their efficacy. In conclusion, even though Brachyury has shown to not confer resistance to antiandrogen therapies in PCa it was also proved to be a potential predictive biomarker to EGFR targeted therapies in LC, which needs further addressing. Furthermore, since the demonstrated efficacy of afatinib and THZ1 small molecule inhibitors at targeting Brachyury we hypothesized that the usage of these drugs could be a strategic tactic to sensitize CRPC to docetaxel and also LC to EGFR targeted therapy, as well as to attenuate the Brachyury-associated aggressive phenotype. Despite the interesting results obtained here, some limitations of this specific work should be acknowledged. First, to further complement these results several assays that we intended to perform from the beginning of this work, such as the biological in vitro studies in LNCaP and HCC827 cells, as well as additional cytotoxic assays for the EGFR-TKIs in HCC827 cell line, but that we were not able to do CHAPTER 6: Conclusions and future perspectives 71 it, due to the limited time and difficulty to work with these cells, should be performed. Second, it is of vital importance to have in mind that our mRNA analyses were done only once, being necessary to do more and different assays, as to study other EMT and FA proteins expression. Also, both EMT and FA processes are reversible, highly dynamic with a complex and tight regulation, which make them tremendously difficult to study and indeed the standard analyses of expression used here function almost as a snapshot at a single timepoint, and as such do not represent the overall process of these mechanisms. In other words, the fact that in a specific timepoint the cells do not express an epithelial and/or mesenchymal marker do not exclude them from being under an EMT or FA process. Third, it is worth mentioning that in vitro models, despite being of extreme significance in cancer research, they do not recapitulate the general complexity of a tumor, which results in observations that might not be translatable to either in vivo models and cancer patients, and such results should be further validated with more complex models. Given our promising results that left several aspects open for discussion as well as the abovementioned limitations, we intend to further explore our initial hypotheses and new ones that arose from this work and as future perspectives we seek to: ‒ Deepen the role of Brachyury (both isoforms when applicable) in PCa and LC models, performing a comprehensive study of its impact in several signaling pathways mentioned here, namely TGFβ, Wnt/β-catenin, PI3K/AKT, MAPK, etc, and also cell cycle and apoptosis analyses; ‒ Conduct in vivo studies to fully validate the lack of Brachyury predictive value to antiandrogenic therapies in PCa; ‒ Recapitulate the results obtained here in regard to the biological behavior and in the response to EGFR-TKIs of LC cells, through more in vitro studies and moving further for in vivo models; ‒ Explore the possible underlying mechanisms of afatinib and THZ1 actions in Brachyury expression as well as their implications on the Brachyury-associated biological behavior of PCa and LC cells, in vitro and in vivo ; ‒ Perform a thorough characterization of the DDS developed here to determine its efficacy in comparison to the free drugs, and being this the case we hope to validate the potential, safety and efficacy of these DDS on Brachyury-targeting and the potential phenotype reversion, using in vitro and in vivo models. 72 1 Bray, F. et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians 68, 394-424, doi:10.3322/caac.21492 (2018). 2 Portugal, Globocan 2018 , <https://gco.iarc.fr/today/data/factsheets/populations/620portugal-fact-sheets.pdf> 3 Hanahan, D. & Weinberg, R. A. The hallmarks of cancer. Cell 100, 57-70, doi:10.1016/s00928674(00)81683-9 (2000). 4 Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646-674, doi:10.1016/j.cell.2011.02.013 (2011). 5 Heidenreich, A. et al. EAU guidelines on prostate cancer. Part 1: screening, diagnosis, and treatment of clinically localised disease. European urology 59, 61-71, doi:10.1016/j.eururo.2010.10.039 (2011). 6 Brawley, O. W. Prostate cancer epidemiology in the United States. World journal of urology 30, 195-200, doi:10.1007/s00345-012-0824-2 (2012). 7 Merriel, S. W. D., Funston, G. & Hamilton, W. Prostate Cancer in Primary Care. Advances in therapy 35, 1285-1294, doi:10.1007/s12325-018-0766-1 (2018). 8 Chan, J. M., Stampfer, M. J. & Giovannucci, E. L. What causes prostate cancer? A brief summary of the epidemiology. Seminars in cancer biology 8, 263-273, doi:10.1006/scbi.1998.0075 (1998). 9 Grönberg, H. Prostate cancer epidemiology. Lancet 361, 859-864, doi:10.1016/s01406736(03)12713-4 (2003). 10 Zhen, J. T. et al. Genetic testing for hereditary prostate cancer: Current status and limitations. Cancer 124, 3105-3117, doi:10.1002/cncr.31316 (2018). 11 Heidegger, I. et al. Hereditary prostate cancer - Primetime for genetic testing? Cancer treatment reviews 81, 101927, doi:10.1016/j.ctrv.2019.101927 (2019). 12 Schröder, F. H. et al. Screening and prostate-cancer mortality in a randomized European study. The New England journal of medicine 360, 1320-1328, doi:10.1056/NEJMoa0810084 (2009). 13 Eskra, J. N., Rabizadeh, D., Pavlovich, C. P., Catalona, W. J. & Luo, J. Approaches to urinary detection of prostate cancer. Prostate cancer and prostatic diseases 22, 362-381, doi:10.1038/s41391-019-0127-4 (2019). 14 Tricoli, J. V., Schoenfeldt, M. & Conley, B. A. Detection of prostate cancer and predicting progression: current and future diagnostic markers. Clinical cancer research : an official journal of the American Association for Cancer Research 10, 3943-3953, doi:10.1158/1078-0432.Ccr03-0200 (2004). 15 Schaid, D. J. The complex genetic epidemiology of prostate cancer. Human molecular genetics 13, R103-121, doi:10.1093/hmg/ddh072 (2004). 16 Bangma, C. H., Roemeling, S. & Schröder, F. H. Overdiagnosis and overtreatment of early detected prostate cancer. World journal of urology 25, 3-9, doi:10.1007/s00345-007-0145-z (2007). 17 Lapointe, J. et al. Gene expression profiling identifies clinically relevant subtypes of prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 101, 811-816, doi:10.1073/pnas.0304146101 (2004). CHAPTER 7: References 73 18 Wang, X. et al. A luminal epithelial stem cell that is a cell of origin for prostate cancer. Nature 461, 495-500, doi:10.1038/nature08361 (2009). 19 Wang, G., Zhao, D., Spring, D. J. & DePinho, R. A. Genetics and biology of prostate cancer. Genes & development 32, 1105-1140, doi:10.1101/gad.315739.118 (2018). 20 Goldstein, A. S., Huang, J., Guo, C., Garraway, I. P. & Witte, O. N. Identification of a cell of origin for human prostate cancer. Science 329, 568-571, doi:10.1126/science.1189992 (2010). 21 Grignon, D. J. Unusual subtypes of prostate cancer. Modern pathology 17, 316-327, doi:10.1038/modpathol.3800052 (2004). 22 Epstein, J. I. et al. The 2014 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma: Definition of Grading Patterns and Proposal for a New Grading System. The American journal of surgical pathology 40, 244-252, doi:10.1097/pas.0000000000000530 (2016). 23 Loeb, S. et al. Active surveillance for prostate cancer: a systematic review of clinicopathologic variables and biomarkers for risk stratification. European urology 67, 619-626, doi:10.1016/j.eururo.2014.10.010 (2015). 24 Tseng, K. S., Landis, P., Epstein, J. I., Trock, B. J. & Carter, H. B. Risk stratification of men choosing surveillance for low risk prostate cancer. The Journal of urology 183, 1779-1785, doi:10.1016/j.juro.2010.01.001 (2010). 25 Babaian, R. J. et al. Best practice statement on cryosurgery for the treatment of localized prostate cancer. The Journal of urology 180, 1993-2004, doi:10.1016/j.juro.2008.07.108 (2008). 26 Litwin, M. S. & Tan, H. J. The Diagnosis and Treatment of Prostate Cancer: A Review. Jama 317, 2532-2542, doi:10.1001/jama.2017.7248 (2017). 27 Boulos, S. & Mazhar, D. The evolving role of chemotherapy in prostate cancer. Future oncology (London, England) 13, 1091-1095, doi:10.2217/fon-2016-0464 (2017). 28 Sharifi, N., Gulley, J. L. & Dahut, W. L. Androgen deprivation therapy for prostate cancer. Jama 294, 238-244, doi:10.1001/jama.294.2.238 (2005). 29 Huggins, C. & Hodges, C. Studies on Prostatic Cancer. I. The Effect of Castration, of Estrogen and of Androgen Injection on Serum Phosphatases in Metastatic Carcinoma of the Prostate. Cancer Res 1, 283-297, doi:10.1097/01.ten.0000196289.33630.45 (1941). 30 Heinlein, C. A. & Chang, C. Androgen receptor in prostate cancer. Endocrine reviews 25, 276308, doi:10.1210/er.2002-0032 (2004). 31 Basu, S. & Tindall, D. J. Androgen action in prostate cancer. Hormones & cancer 1, 223-228, doi:10.1007/s12672-010-0044-4 (2010). 32 Tan, M. H., Li, J., Xu, H. E., Melcher, K. & Yong, E. L. Androgen receptor: structure, role in prostate cancer and drug discovery. Acta pharmacologica Sinica 36, 3-23, doi:10.1038/aps.2014.18 (2015). 33 Weikum, E. R., Liu, X. & Ortlund, E. A. The nuclear receptor superfamily: A structural perspective. Protein science 27, 1876-1892, doi:10.1002/pro.3496 (2018). 34 Davey, R. A. & Grossmann, M. Androgen Receptor Structure, Function and Biology: From Bench to Bedside. The Clinical biochemist. Reviews 37, 3-15 (2016). 35 Christopoulos, P. F., Vlachogiannis, N. I., Vogkou, C. T. & Koutsilieris, M. The Role of the Androgen Receptor Signaling in Breast Malignancies. Anticancer research 37, 6533-6540, doi:10.21873/anticanres.12109 (2017). 36 Okitsu, K. et al. Involvement of interleukin-6 and androgen receptor signaling in pancreatic cancer. Genes & cancer 1, 859-867, doi:10.1177/1947601910383417 (2010). 37 Mizushima, T. & Miyamoto, H. The Role of Androgen Receptor Signaling in Ovarian Cancer. Cells 8, doi:10.3390/cells8020176 (2019). 80 147 Kim, W. J. et al. Histological transformation from non-small cell to small cell lung carcinoma after treatment with epidermal growth factor receptor-tyrosine kinase inhibitor. Thoracic cancer 6, 800804, doi:10.1111/1759-7714.12217 (2015). 148 Levin, P. A. et al. Histologic Transformation from Adenocarcinoma to Squamous Cell Carcinoma as a Mechanism of Resistance to EGFR Inhibition. Journal of Thoracic Oncology 10, E86-E88, doi:10.1097/JTO.0000000000000571 (2015). 149 Jukna, A. et al. Squamous Cell Carcinoma "Transformation" Concurrent with Secondary T790M Mutation in Resistant EGFR-Mutated Adenocarcinomas. Journal of thoracic oncology 11, e4951, doi:10.1016/j.jtho.2015.12.096 (2016). 150 Park, S. et al. Paired genomic analysis of squamous cell carcinoma transformed from EGFRmutated lung adenocarcinoma. Lung cancer 134, 7-15, doi:10.1016/j.lungcan.2019.05.024 (2019). 151 Shaurova, T., Zhang, L., Goodrich, D. W. & Hershberger, P. A. Understanding Lineage Plasticity as a Path to Targeted Therapy Failure in EGFR-Mutant Non-small Cell Lung Cancer. Frontiers in genetics 11, 281, doi:10.3389/fgene.2020.00281 (2020). 152 Vaquerizas, J. M., Kummerfeld, S. K., Teichmann, S. A. & Luscombe, N. M. A census of human transcription factors: function, expression and evolution. Nature reviews. Genetics 10, 252-263, doi:10.1038/nrg2538 (2009). 153 Naiche, L. A., Harrelson, Z., Kelly, R. G. & Papaioannou, V. E. T-box genes in vertebrate development. Annual review of genetics 39, 219-239, doi:10.1146/annurev.genet.39.073003.105925 (2005). 154 Dobrovolskaia-Zavadskaia. Sur la mortification spontanee de la queuw chez la spuris nouveau et sur l'existence d'un caractere (facteur) hereditaire non viable. Compr Soc Bio 97, 114-119 (1927). 155 Chesley, P. Development of the short‐tailed mutant in the house mouse. Journal of Experimental Zoology 70, 429-459, doi:10.1002/jez.1400700306 (1935). 156 Herrmann, B. G., Labeit, S., Poustka, A., King, T. R. & Lehrach, H. Cloning of the T gene required in mesoderm formation in the mouse. Nature 343, 617-622, doi:10.1038/343617a0 (1990). 157 Kispert, A. & Herrmann, B. G. The Brachyury gene encodes a novel DNA binding protein. The EMBO journal 12, 3211-3220, doi:10.1002/j.1460-2075.1993.tb05990.x (1993). 158 Kispert, A., Koschorz, B. & Herrmann, B. G. The T protein encoded by Brachyury is a tissuespecific transcription factor. The EMBO journal 14, 4763-4772, doi:10.1002/j.14602075.1995.tb00158.x (1995). 159 Smith, J. C., Price, B. M., Green, J. B., Weigel, D. & Herrmann, B. G. Expression of a Xenopus homolog of Brachyury (T) is an immediate-early response to mesoderm induction. Cell 67, 7987, doi:10.1016/0092-8674(91)90573-h (1991). 160 Schulte-Merker, S., Ho, R. K., Herrmann, B. G. & Nüsslein-Volhard, C. The protein product of the zebrafish homologue of the mouse T gene is expressed in nuclei of the germ ring and the notochord of the early embryo. Development 116, 1021-1032 (1992). 161 Kispert, A., Herrmann, B. G., Leptin, M. & Reuter, R. Homologs of the mouse Brachyury gene are involved in the specification of posterior terminal structures in Drosophila, Tribolium, and Locusta. Genes & development 8, 2137-2150, doi:10.1101/gad.8.18.2137 (1994). 162 Edwards, Y. H. et al. The human homolog T of the mouse T(Brachyury) gene; gene structure, cDNA sequence, and assignment to chromosome 6q27. Genome research 6, 226-233, doi:10.1101/gr.6.3.226 (1996). 163 Beddington, R. S., Rashbass, P. & Wilson, V. Brachyury--a gene affecting mouse gastrulation and early organogenesis. Dev 116, 157-165 (1992). 81 164 Wilkinson, D. G., Bhatt, S. & Herrmann, B. G. Expression pattern of the mouse T gene and its role in mesoderm formation. Nature 343, 657-659, doi:10.1038/343657a0 (1990). 165 Willison, K. The mouse Brachyury gene and mesoderm formation. Trends in genetics 6, 104105, doi:10.1016/0168-9525(90)90106-g (1990). 166 Wilson, V. & Beddington, R. Expression of T protein in the primitive streak is necessary and sufficient for posterior mesoderm movement and somite differentiation. Developmental biology 192, 45-58, doi:10.1006/dbio.1997.8701 (1997). 167 Inman, K. E. & Downs, K. M. Brachyury is required for elongation and vasculogenesis in the murine allantois. Development 133, 2947-2959, doi:10.1242/dev.02454 (2006). 168 Technau, U. Brachyury, the blastopore and the evolution of the mesoderm. BioEssays 23, 788794, doi:10.1002/bies.1114 (2001). 169 Cunliffe, V. & Smith, J. C. Ectopic mesoderm formation in Xenopus embryos caused by widespread expression of a Brachyury homologue. Nature 358, 427-430, doi:10.1038/358427a0 (1992). 170 Schulte-Merker, S., van Eeden, F. J., Halpern, M. E., Kimmel, C. B. & Nüsslein-Volhard, C. no tail (ntl) is the zebrafish homologue of the mouse T (Brachyury) gene. Development 120, 10091015 (1994). 171 Haworth, K. et al. Canine homolog of the T-box transcription factor T; failure of the protein to bind to its DNA target leads to a short-tail phenotype. Mammalian genome 12, 212-218, doi:10.1007/s003350010253 (2001). 172 Buckingham, K. J. et al. Multiple mutant T alleles cause haploinsufficiency of Brachyury and short tails in Manx cats. Mammalian genome 24, 400-408, doi:10.1007/s00335-013-9471-1 (2013). 173 Postma, A. et al. Mutations in the T (brachyury) gene cause a novel syndrome consisting of sacral agenesis, abnormal ossification of the vertebral bodies and a persistent notochordal canal. Journal of Medical Genetics 51, 90-97 (2014). 174 Feng, X. et al. Genetic variants of TBX6 and TBXT identified in patients with congenital scoliosis in Southern China. Journal of orthopaedic research , doi:10.1002/jor.24805 (2020). 175 Jensen, L. E. et al. The human T locus and spina bifida risk. Human genetics 115, 475-482, doi:10.1007/s00439-004-1185-8 (2004). 176 Vujovic, S. et al. Brachyury, a crucial regulator of notochordal development, is a novel biomarker for chordomas. The Journal of pathology 209, 157-165, doi:10.1002/path.1969 (2006). 177 Yang, X. R. et al. T (brachyury) gene duplication confers major susceptibility to familial chordoma. Nature genetics 41, 1176-1178, doi:10.1038/ng.454 (2009). 178 Kelley, M. J. et al. Characterization of T gene sequence variants and germline duplications in familial and sporadic chordoma. Human genetics 133, 1289-1297, doi:10.1007/s00439-0141463-z (2014). 179 Pillay, N. et al. A common single-nucleotide variant in T is strongly associated with chordoma. Nature genetics 44, 1185-1187, doi:10.1038/ng.2419 (2012). 180 Nibu, Y., José-Edwards, D. S. & Di Gregorio, A. From notochord formation to hereditary chordoma: the many roles of Brachyury. BioMed research international 2013, 826435, doi:10.1155/2013/826435 (2013). 181 Tirabosco, R. et al. Brachyury expression in extra-axial skeletal and soft tissue chordomas: a marker that distinguishes chordoma from mixed tumor/myoepithelioma/parachordoma in soft tissue. The American journal of surgical pathology 32, 572-580, doi:10.1097/PAS.0b013e31815b693a (2008). 182 Jambhekar, N. A. et al. Revisiting chordoma with brachyury, a "new age" marker: analysis of a validation study on 51 cases. Archives of pathology & laboratory medicine 134, 1181-1187, doi:10.1043/2009-0476-oa.1 (2010). 82 183 Sharifnia, T. et al. Small-molecule targeting of brachyury transcription factor addiction in chordoma. Nature medicine 25, 292-300, doi:10.1038/s41591-018-0312-3 (2019). 184 Palena, C. et al. The human T-box mesodermal transcription factor Brachyury is a candidate target for T-cell-mediated cancer immunotherapy. Clinical cancer research 13, 2471-2478, doi:10.1158/1078-0432.Ccr-06-2353 (2007). 185 Roselli, M. et al. Brachyury, a driver of the epithelial-mesenchymal transition, is overexpressed in human lung tumors: an opportunity for novel interventions against lung cancer. Clinical cancer research 18, 3868-3879, doi:10.1158/1078-0432.Ccr-11-3211 (2012). 186 Hamilton, D. H. et al. Brachyury, a vaccine target, is overexpressed in triple-negative breast cancer. Endocrine-related cancer 23, 783-796, doi:10.1530/erc-16-0037 (2016). 187 Palena, C. et al. Overexpression of the EMT driver brachyury in breast carcinomas: association with poor prognosis. Journal of the National Cancer Institute 106, doi:10.1093/jnci/dju054 (2014). 188 Miettinen, M. et al. Nuclear Brachyury Expression Is Consistent in Chordoma, Common in Germ Cell Tumors and Small Cell Carcinomas, and Rare in Other Carcinomas and Sarcomas: An Immunohistochemical Study of 5229 Cases. The American journal of surgical pathology 39, 1305-1312, doi:10.1097/pas.0000000000000462 (2015). 189 Du, R. et al. Overexpression of brachyury contributes to tumor metastasis by inducing epithelialmesenchymal transition in hepatocellular carcinoma. Journal of experimental & clinical cancer research 33, 105, doi:10.1186/s13046-014-0105-6 (2014). 190 Pinto, F. et al. T-box transcription factor brachyury is associated with prostate cancer progression and aggressiveness. Clinical cancer research 20, 4949-4961, doi:10.1158/1078-0432.Ccr-140421 (2014). 191 Pinto, F. et al. The embryonic Brachyury transcription factor is a novel biomarker of GIST aggressiveness and poor survival. Gastric cancer 19, 651-659, doi:10.1007/s10120-015-05050 (2016). 192 Pinto, F. et al. Brachyury oncogene is a prognostic factor in high-risk testicular germ cell tumors. Andrology 6, 597-604, doi:10.1111/andr.12495 (2018). 193 Haro, A. et al. Expression of Brachyury gene is a significant prognostic factor for primary lung carcinoma. Annals of surgical oncology 20 Suppl 3, S509-516, doi:10.1245/s10434-0132914-9 (2013). 194 Kilic, N. et al. Brachyury expression predicts poor prognosis at early stages of colorectal cancer. European journal of cancer 47, 1080-1085, doi:10.1016/j.ejca.2010.11.015 (2011). 195 Fernando, R. I. et al. The T-box transcription factor Brachyury promotes epithelial-mesenchymal transition in human tumor cells. The Journal of clinical investigation 120, 533-544, doi:10.1172/JCI38379 (2010). 196 Shao, C., Zhang, J., Fu, J. & Ling, F. The potential role of Brachyury in inducing epithelial-tomesenchymal transition (EMT) and HIF-1α expression in breast cancer cells. Biochemical and biophysical research communications 467, 1083-1089, doi:10.1016/j.bbrc.2015.09.076 (2015). 197 Li, K. et al. SMC1 promotes epithelial-mesenchymal transition in triple-negative breast cancer through upregulating Brachyury. Oncology reports 35, 2405-2412, doi:10.3892/or.2016.4564 (2016). 198 Li, K. et al. Brachyury promotes tamoxifen resistance in breast cancer by targeting SIRT1. Biomedicine & pharmacotherapy 84, 28-33, doi:10.1016/j.biopha.2016.09.011 (2016). 199 Chen, S. et al. T-box transcription factor Brachyury in lung cancer cells inhibits macrophage infiltration by suppressing CCL2 and CCL4 chemokines. Tumour biology 36, 5881-5890, doi:10.1007/s13277-015-3260-2 (2015). 83 200 Shah, S. R. et al. Brachyury-YAP Regulatory Axis Drives Stemness and Growth in Cancer. Cell reports 21, 495-507, doi:10.1016/j.celrep.2017.09.057 (2017). 201 Nelson, A. C. et al. An integrated functional genomics approach identifies the regulatory network directed by brachyury (T) in chordoma. The Journal of pathology 228, 274-285, doi:10.1002/path.4082 (2012). 202 Jezkova, J. et al. Brachyury regulates proliferation of cancer cells via a p27Kip1-dependent pathway. Oncotarget 5, 3813-3822, doi:10.18632/oncotarget.1999 (2014). 203 Kobayashi, Y. et al. Knockdown of the T-box transcription factor Brachyury increases sensitivity of adenoid cystic carcinoma cells to chemotherapy and radiation in vitro: implications for a new therapeutic principle. International journal of oncology 44, 1107-1117, doi:10.3892/ijo.2014.2292 (2014). 204 Huang, B. et al. The embryonic transcription factor Brachyury blocks cell cycle progression and mediates tumor resistance to conventional antitumor therapies. Cell death & disease 4, e682, doi:10.1038/cddis.2013.208 (2013). 205 Xu, K., Liu, B. & Liu, Y. Impact of Brachyury on epithelial-mesenchymal transitions and chemosensitivity in non-small cell lung cancer. Molecular medicine reports 12, 995-1001, doi:10.3892/mmr.2015.3348 (2015). 206 Otani, R. et al. Brachyury gene copy number gain and activation of the PI3K/Akt pathway: association with upregulation of oncogenic Brachyury expression in skull base chordoma. Journal of neurosurgery 128, 1428-1437, doi:10.3171/2016.12.Jns161444 (2018). 207 Imajyo, I. et al. T-box transcription factor Brachyury expression is correlated with epithelialmesenchymal transition and lymph node metastasis in oral squamous cell carcinoma. International journal of oncology 41, 1985-1995, doi:10.3892/ijo.2012.1673 (2012). 208 Chen, M. et al. Transactivation of SOX5 by Brachyury promotes breast cancer bone metastasis. Carcinogenesis 41, 551-560, doi:10.1093/carcin/bgz142 (2020). 209 Lee, K. H. et al. Prognostic significance of expression of epithelial-mesenchymal transition driver brachyury in breast cancer and its association with subtype and characteristics. Oncology letters 15, 1037-1045, doi:10.3892/ol.2017.7402 (2018). 210 Shimamatsu, S. et al. Prognostic Significance of Expression of the Epithelial-Mesenchymal Transition-Related Factor Brachyury in Intrathoracic Lymphatic Spread of Non-Small Cell Lung Cancer. Annals of surgical oncology 23, 1012-1020, doi:10.1245/s10434-016-5530-7 (2016). 211 Shimoda, M. et al. The T-box transcription factor Brachyury regulates epithelial-mesenchymal transition in association with cancer stem-like cells in adenoid cystic carcinoma cells. BMC cancer 12, 377, doi:10.1186/1471-2407-12-377 (2012). 212 Sarkar, D., Shields, B., Davies, M. L., Müller, J. & Wakeman, J. A. BRACHYURY confers cancer stem cell characteristics on colorectal cancer cells. International journal of cancer 130, 328337, doi:10.1002/ijc.26029 (2012). 213 Pinto, F. et al. Brachyury as a potential modulator of androgen receptor activity and a key player in therapy resistance in prostate cancer. Oncotarget 7, 28891-28902, doi:10.18632/oncotarget.8499 (2016). 214 Larocca, C. et al. An autocrine loop between TGF-β1 and the transcription factor brachyury controls the transition of human carcinoma cells into a mesenchymal phenotype. Molecular cancer therapeutics 12, 1805-1815, doi:10.1158/1535-7163.Mct-12-1007 (2013). 215 Morley, R. H. et al. A gene regulatory network directed by zebrafish No tail accounts for its roles in mesoderm formation. Proceedings of the National Academy of Sciences of the United States of America 106, 3829-3834, doi:10.1073/pnas.0808382106 (2009). 84 216 Hotta, K., Takahashi, H., Satoh, N. & Gojobori, T. Brachyury-downstream gene sets in a chordate, Ciona intestinalis: integrating notochord specification, morphogenesis and chordate evolution. Evolution & development 10, 37-51, doi:10.1111/j.1525-142X.2007.00212.x (2008). 217 Martin, B. L. & Kimelman, D. Regulation of canonical Wnt signaling by Brachyury is essential for posterior mesoderm formation. Developmental cell 15, 121-133, doi:10.1016/j.devcel.2008.04.013 (2008). 218 Xu, J. et al. The Role of Transcriptional Factor Brachyury on Cell Cycle Regulation in Non-small Cell Lung Cancer. Frontiers in oncology 10, 1078, doi:10.3389/fonc.2020.01078 (2020). 219 Sun, S. et al. The T-box transcription factor Brachyury promotes renal interstitial fibrosis by repressing E-cadherin expression. Cell communication and signaling 12, 76, doi:10.1186/s12964-014-0076-4 (2014). 220 Fernando, R. I., Castillo, M. D., Litzinger, M., Hamilton, D. H. & Palena, C. IL-8 signaling plays a critical role in the epithelial-mesenchymal transition of human carcinoma cells. Cancer Res 71, 5296-5306, doi:10.1158/0008-5472.Can-11-0156 (2011). 221 Hu, Y., Mintz, A., Shah, S. R., Quinones-Hinojosa, A. & Hsu, W. The FGFR/MEK/ERK/brachyury pathway is critical for chordoma cell growth and survival. Carcinogenesis 35, 1491-1499, doi:10.1093/carcin/bgu014 (2014). 222 Wan, Z. et al. T-box transcription factor brachyury promotes tumor cell invasion and metastasis in non-small cell lung cancer via upregulation of matrix metalloproteinase 12. Oncology reports 36, 306-314, doi:10.3892/or.2016.4792 (2016). 223 Park, J. C. et al. Epigenetic silencing of human T (brachyury homologue) gene in non-small-cell lung cancer. Biochemical and biophysical research communications 365, 221-226, doi:10.1016/j.bbrc.2007.10.144 (2008). 224 Pinto, F. et al. The T-box transcription factor brachyury behaves as a tumor suppressor in gliomas. The Journal of pathology 251, 87-99, doi:10.1002/path.5419 (2020). 225 Hu, Y., Feng, X., Mintz, A., Petty, W. J. & Hsu, W. Regulation of brachyury by fibroblast growth factor receptor 1 in lung cancer. Oncotarget 7, 87124-87135, doi:10.18632/oncotarget.13547 (2016). 226 Robinson, H., McFarlane, R. J. & Wakeman, J. A. Brachyury: Strategies for Drugging an Intractable Cancer Therapeutic Target. Trends in cancer 6, 271-273, doi:10.1016/j.trecan.2020.01.014 (2020). 227 Tucker, J. A. et al. Identification and characterization of a cytotoxic T-lymphocyte agonist epitope of brachyury, a transcription factor involved in epithelial to mesenchymal transition and metastasis. Cancer immunology, immunotherapy 63, 1307-1317, doi:10.1007/s00262-0141603-2 (2014). 228 Hamilton, D. H. et al. Immunological targeting of tumor cells undergoing an epithelialmesenchymal transition via a recombinant brachyury-yeast vaccine. Oncotarget 4, 1777-1790, doi:10.18632/oncotarget.1295 (2013). 229 Heery, C. R. et al. Phase I Study of a Poxviral TRICOM-Based Vaccine Directed Against the Transcription Factor Brachyury. Clinical cancer research 23, 6833-6845, doi:10.1158/10780432.Ccr-17-1087 (2017). 230 Heery, C. R. et al. Phase I Trial of a Yeast-Based Therapeutic Cancer Vaccine (GI-6301) Targeting the Transcription Factor Brachyury. Cancer immunology research 3, 1248-1256, doi:10.1158/2326-6066.Cir-15-0119 (2015). 231 Collins, J. M. et al. Phase I Trial of a Modified Vaccinia Ankara Priming Vaccine Followed by a Fowlpox Virus Boosting Vaccine Modified to Express Brachyury and Costimulatory Molecules in Advanced Solid Tumors. The oncologist 25, 560-e1006, doi:10.1634/theoncologist.2019-0932 (2020). 85 232 ClinicalTrials.gov. QUILT-3.011 Phase 2 Yeast-Brachyury Vaccine Chordoma , <https://www.clinicaltrials.gov/ct2/show/NCT02383498> 233 ClinicalTrials.gov. BN Brachyury and Radiation in Chordoma , <https://www.clinicaltrials.gov/ct2/show/NCT03595228> 234 ClinicalTrials.gov. Perioperative Atezolizumab With MVA-BN-Brachyury and PROSTVAC For Intermediate-Risk And High-Risk Localized Prostate Cancer (AtezoVax) , <https://www.clinicaltrials.gov/ct2/show/NCT04020094> 235 ClinicalTrials.gov. Phase I/II Study of Immunotherapy Combination BN-Brachyury Vaccine, M7824, ALT-803 and Epacadostat (QuEST1) , <https://www.clinicaltrials.gov/ct2/show/NCT03493945> 236 ClinicalTrials.gov. BN-Brachyury, Entinostat, Adotrastuzumab Emtansine and M7824 in Advanced Stage Breast Cancer (BrEAsT) , <https://www.clinicaltrials.gov/ct2/show/NCT04296942> 237 Magnaghi, P. et al. Afatinib Is a New Therapeutic Approach in Chordoma with a Unique Ability to Target EGFR and Brachyury. Molecular cancer therapeutics 17, 603-613, doi:10.1158/15357163.Mct-17-0324 (2018). 238 Din, F. U. et al. Effective use of nanocarriers as drug delivery systems for the treatment of selected tumors. International journal of nanomedicine 12, 7291-7309, doi:10.2147/ijn.S146315 (2017). 239 Sun, T. et al. Engineered nanoparticles for drug delivery in cancer therapy. Angewandte Chemie 53, 12320-12364, doi:10.1002/anie.201403036 (2014). 240 Tibbitt, M. W., Dahlman, J. E. & Langer, R. Emerging Frontiers in Drug Delivery. Journal of the American Chemical Society 138, 704-717, doi:10.1021/jacs.5b09974 (2016). 241 Anselmo, A. C. & Mitragotri, S. Nanoparticles in the clinic: An update. Bioengineering & translational medicine 4, e10143, doi:10.1002/btm2.10143 (2019). 242 Mitchell, M. J. et al. Engineering precision nanoparticles for drug delivery. Nature reviews. Drug discovery , 1-24, doi:10.1038/s41573-020-0090-8 (2020). 243 Allen, T. M. & Cullis, P. R. Drug delivery systems: entering the mainstream. Science 303, 18181822, doi:10.1126/science.1095833 (2004). 244 ClinicalTrials.gov. A First in Human Study Using 89Zr-cRGDY Ultrasmall Silica Particle Tracers for Malignant Brain Tumors , <https://clinicaltrials.gov/ct2/show/NCT03465618> 245 ClinicalTrials.gov. Targeted Silica Nanoparticles for Real-Time Image-Guided Intraoperative Mapping of Nodal Metastases , <https://clinicaltrials.gov/ct2/show/NCT02106598> 246 Arruebo, M. Drug delivery from structured porous inorganic materials. Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology 4, 16-30, doi:10.1002/wnan.132 (2012). 247 Sing, K. et al. IUPAC Manual of Symbols and Terminology Appendix 2, Pt. 1. Colloid Surface Chemistry, Pure Appl. Chem 31, 578 (1972). 248 Servatan, M. et al. Zeolites in drug delivery: Progress, challenges and opportunities. Drug discovery today 25, 642-656, doi:10.1016/j.drudis.2020.02.005 (2020). 249 Alothman, Z. A. A review: fundamental aspects of silicate mesoporous materials. Materials 5, 2874-2902, doi:10.3390/ma5122874 (2012). 250 Rimoli, M. G. et al. Synthetic zeolites as a new tool for drug delivery. Journal of biomedical materials research. Part A 87, 156-164, doi:10.1002/jbm.a.31763 (2008). 251 Helliwell, M., Jones, R. H., Kaucic, V. & Logar, N. Z. The use of softer X-rays in the structure elucidation of microporous materials. Journal of synchrotron radiation 12, 420-430, doi:10.1107/s0909049504032327 (2005). 86 252 Mumpton, F. A. La roca magica: uses of natural zeolites in agriculture and industry. Proceedings of the National Academy of Sciences of the United States of America 96, 3463-3470, doi:10.1073/pnas.96.7.3463 (1999). 253 Perez, E. et al. Encapsulation of α-tocopheryl acetate into zeolite Y for textile application. Industrial engineering chemistry research 49, 8495-8500, doi:10.1021/ie100483v (2010). 254 Baeza, A., Manzano, M., Colilla, M. & Vallet-Regí, M. Recent advances in mesoporous silica nanoparticles for antitumor therapy: our contribution. Biomaterials science 4, 803-813, doi:10.1039/c6bm00039h (2016). 255 Platas-Iglesias, C. et al. Zeolite GdNaY nanoparticles with very high relaxivity for application as contrast agents in magnetic resonance imaging. Chemistry 8, 5121-5131, doi:10.1002/15213765(20021115)8:22<5121::Aid-chem5121>3.0.Co;2-w (2002). 256 Tsotsalas, M. M. et al. Encapsulating (111)In in nanocontainers for scintigraphic imaging: synthesis, characterization, and in vivo biodistribution. ACS nano 4, 342-348, doi:10.1021/nn901166u (2010). 257 Ndiege, N., Raidoo, R., Schultz, M. K. & Larsen, S. Preparation of a versatile bifunctional zeolite for targeted imaging applications. Langmuir 27, 2904-2909, doi:10.1021/la2000409 (2011). 258 Faraji Dizaji, B. et al. Synthesis of PLGA/chitosan/zeolites and PLGA/chitosan/metal organic frameworks nanofibers for targeted delivery of Paclitaxel toward prostate cancer cells death. International journal of biological macromolecules 164, 1461-1474, doi:10.1016/j.ijbiomac.2020.07.228 (2020). 259 Martinho, O. et al. In vitro and in vivo studies of temozolomide loading in zeolite structures as drug delivery systems for glioblastoma. RSC Advances 5, 28219-28227, doi:10.1039/C5RA03871E (2015). 260 Uglea, C. V. et al. Drug delivery systems based on inorganic materials: I. Synthesis and characterization of a zeolite-cyclophosphamide system. Journal of biomaterials science. Polymer edition 6, 633-637, doi:10.1163/156856294x00572 (1994). 261 Chen, D. et al. Cancer Cell Membrane-Decorated Zeolitic-Imidazolate Frameworks Codelivering Cisplatin and Oleanolic Acid Induce Apoptosis and Reversed Multidrug Resistance on Bladder Carcinoma Cells. ACS omega 5, 995-1002, doi:10.1021/acsomega.9b02261 (2020). 262 Arruebo, M. et al. Sustained release of doxorubicin from zeolite-magnetite nanocomposites prepared by mechanical activation. Nanotechnology 17, 4057-4064, doi:10.1088/09574484/17/16/011 (2006). 263 Baerlocher, C., McCusker, L. B. & Olson, D. H. Atlas of zeolite framework types . (Elsevier, 2007). 264 Vilaça, N. et al. Encapsulation of α-cyano-4-hydroxycinnamic acid into a NaY zeolite. Journal of Materials Science 46, 7511, doi:10.1007/s10853-011-5722-2 (2011). 265 Amorim, R. et al. Zeolite structures loading with an anticancer compound as drug delivery systems. J. Phys. Chem. C 116, 25642-25650, doi:10.1021/jp3093868 (2012). 266 Vilaça, N. et al. Potentiation of 5-fluorouracil encapsulated in zeolites as drug delivery systems for in vitro models of colorectal carcinoma. Colloids and surfaces. B, Biointerfaces 112, 237244, doi:10.1016/j.colsurfb.2013.07.042 (2013). 267 Vilaca, N. l. et al. Micro-and mesoporous structures as drug delivery carriers for salicylic acid. J. Phys. Chem. C 119, 3589-3595, doi:10.1021/jp5117849 (2015). 268 Slowing, II, Vivero-Escoto, J. L., Wu, C. W. & Lin, V. S. Mesoporous silica nanoparticles as controlled release drug delivery and gene transfection carriers. Advanced drug delivery reviews 60, 1278-1288, doi:10.1016/j.addr.2008.03.012 (2008). 269 Vallet-Regí, M., Balas, F. & Arcos, D. Mesoporous materials for drug delivery. Angewandte Chemie 46, 7548-7558, doi:10.1002/anie.200604488 (2007). 87 270 Wang, Q. et al. Multiple Drug Delivery from Mesoporous Coating Realizing Combination Therapy for Bare Metal Stents. Langmuir 35, 3126-3133, doi:10.1021/acs.langmuir.8b04080 (2019). 271 Jeong, H. J. et al. Macrophage cell tracking PET imaging using mesoporous silica nanoparticles via in vivo bioorthogonal F-18 labeling. Biomaterials 199, 32-39, doi:10.1016/j.biomaterials.2019.01.043 (2019). 272 Zhang, Y. et al. Core-Shell Magnetic Mesoporous Silica Microspheres with Large Mesopores for Enzyme Immobilization in Biocatalysis. ACS applied materials & interfaces 11, 10356-10363, doi:10.1021/acsami.8b18721 (2019). 273 Rosenholm, J. M. et al. Cancer-cell-specific induction of apoptosis using mesoporous silica nanoparticles as drug-delivery vectors. Small 6, 1234-1241, doi:10.1002/smll.200902355 (2010). 274 Popova, M. et al. Tamoxifen Delivery System Based on PEGylated Magnetic MCM-41 Silica. Molecules 25, doi:10.3390/molecules25215129 (2020). 275 Tao, Z., Toms, B., Goodisman, J. & Asefa, T. Mesoporous silica microparticles enhance the cytotoxicity of anticancer platinum drugs. ACS nano 4, 789-794, doi:10.1021/nn9015345 (2010). 276 Saroj, S. & Rajput, S. J. Tailor-made pH-sensitive polyacrylic acid functionalized mesoporous silica nanoparticles for efficient and controlled delivery of anti-cancer drug Etoposide. Drug development and industrial pharmacy 44, 1198-1211, doi:10.1080/03639045.2018.1438467 (2018). 277 Peyvand, P. et al. Imidazolium-based ionic liquid functionalized mesoporous silica nanoparticles as a promising nano-carrier: response surface strategy to investigate and optimize loading and release process for Lapatinib delivery. Pharmaceutical development and technology 25, 11501161, doi:10.1080/10837450.2020.1803909 (2020). 278 Beck, J. S. et al. A new family of mesoporous molecular sieves prepared with liquid crystal templates. Journal of the American Chemical Society 114, 10834-10843, doi:10.1021/ja00053a020 (1992). 279 Zhao, D. et al. Triblock copolymer syntheses of mesoporous silica with periodic 50 to 300 angstrom pores. Science 279, 548-552, doi:10.1126/science.279.5350.548 (1998). 280 Narayan, R., Nayak, U. Y., Raichur, A. M. & Garg, S. Mesoporous Silica Nanoparticles: A Comprehensive Review on Synthesis and Recent Advances. Pharmaceutics 10, doi:10.3390/pharmaceutics10030118 (2018). 281 Heikkilä, T. et al. Evaluation of mesoporous TCPSi, MCM-41, SBA-15, and TUD-1 materials as API carriers for oral drug delivery. Drug delivery 14, 337-347, doi:10.1080/10717540601098823 (2007). 282 Hille, C. et al. Detection of Androgen Receptor Variant 7 (ARV7) mRNA Levels in EpCAM-Enriched CTC Fractions for Monitoring Response to Androgen Targeting Therapies in Prostate Cancer. Cells 8, doi:10.3390/cells8091067 (2019). 283 Hamilton, D. H., McCampbell, K. K. & Palena, C. Loss of the Cyclin-Dependent Kinase Inhibitor 1 in the Context of Brachyury-Mediated Phenotypic Plasticity Drives Tumor Resistance to Immune Attack. Frontiers in oncology 8, 143, doi:10.3389/fonc.2018.00143 (2018). 284 Hamilton, D. H., Litzinger, M. T., Fernando, R. I., Huang, B. & Palena, C. Cancer vaccines targeting the epithelial-mesenchymal transition: tissue distribution of brachyury and other drivers of the mesenchymal-like phenotype of carcinomas. Seminars in oncology 39, 358-366, doi:10.1053/j.seminoncol.2012.02.005 (2012). 285 Ji, W. et al. Efficacy of the CDK7 Inhibitor on EMT-Associated Resistance to 3rd Generation EGFRTKIs in Non-Small Cell Lung Cancer Cell Lines. Cells 9, doi:10.3390/cells9122596 (2020). 88 286 Valenta, T., Hausmann, G. & Basler, K. The many faces and functions of β-catenin. The EMBO journal 31, 2714-2736, doi:10.1038/emboj.2012.150 (2012). 287 Guo, X. & Wang, X. F. Signaling cross-talk between TGF-beta/BMP and other pathways. Cell research 19, 71-88, doi:10.1038/cr.2008.302 (2009). 288 Cheruku, H. R. et al. Transforming growth factor-β, MAPK and Wnt signaling interactions in colorectal cancer. EuPA Open Proteomics 8, 104-115, doi:10.1016/j.euprot.2015.06.004 (2015). 289 Kashyap, V. & Bonavida, B. Role of YY1 in the pathogenesis of prostate cancer and correlation with bioinformatic data sets of gene expression. Genes & cancer 5, 71-83, doi:10.18632/genesandcancer.12 (2014). 290 Bonavida, B. & Baritaki, S. The novel role of Yin Yang 1 in the regulation of epithelial to mesenchymal transition in cancer via the dysregulated NF-κB/Snail/YY1/RKIP/PTEN Circuitry. Critical reviews in oncogenesis 16, 211-226, doi:10.1615/critrevoncog.v16.i3-4.50 (2011). 291 Yang, T. et al. hnRNPM, a potential mediator of YY1 in promoting the epithelial-mesenchymal transition of prostate cancer cells. The Prostate 79, 1199-1210, doi:10.1002/pros.23790 (2019). 292 Xia, W. et al. Transcription factor YY1 mediates epithelial-mesenchymal transition through the TGFβ signaling pathway in bladder cancer. Medical oncology 37, 93, doi:10.1007/s12032-02001414-5 (2020). 293 Zhang, C. et al. YY1 mediates TGF-β1-induced EMT and pro-fibrogenesis in alveolar epithelial cells. Respiratory research 20, 249, doi:10.1186/s12931-019-1223-7 (2019). 294 Desiniotis, A. & Kyprianou, N. Significance of talin in cancer progression and metastasis. International review of cell and molecular biology 289, 117-147, doi:10.1016/b978-0-12386039-2.00004-3 (2011). 295 Takahashi, H. et al. Regulation of notochord-specific expression of Ci-Bra downstream genes in Ciona intestinalis embryos. Zoological science 27, 110-118, doi:10.2108/zsj.27.110 (2010). 296 Sakamoto, S., McCann, R. O., Dhir, R. & Kyprianou, N. Talin1 promotes tumor invasion and metastasis via focal adhesion signaling and anoikis resistance. Cancer Res 70, 1885-1895, doi:10.1158/0008-5472.Can-09-2833 (2010). 297 Fang, K. P., Zhang, J. L., Ren, Y. H. & Qian, Y. B. Talin-1 correlates with reduced invasion and migration in human hepatocellular carcinoma cells. Asian Pacific journal of cancer prevention 15, 2655-2661, doi:10.7314/apjcp.2014.15.6.2655 (2014). 298 Huang, Z., Barker, D., Gibbins, J. M. & Dash, P. R. Talin is a substrate for SUMOylation in migrating cancer cells. Experimental cell research 370, 417-425, doi:10.1016/j.yexcr.2018.07.005 (2018). 299 McCarty, D. J. et al. Novel galeterone analogs act independently of AR and AR-V7 for the activation of the unfolded protein response and induction of apoptosis in the CWR22Rv1 prostate cancer cell model. Oncotarget 8, 88501-88516, doi:10.18632/oncotarget.19762 (2017). 300 Thress, K. S. et al. Acquired EGFR C797S mutation mediates resistance to AZD9291 in nonsmall cell lung cancer harboring EGFR T790M. Nature medicine 21, 560-562, doi:10.1038/nm.3854 (2015). 301 Niederst, M. J. et al. The Allelic Context of the C797S Mutation Acquired upon Treatment with Third-Generation EGFR Inhibitors Impacts Sensitivity to Subsequent Treatment Strategies. Clinical cancer research 21, 3924-3933, doi:10.1158/1078-0432.Ccr-15-0560 (2015). 302 Ma, J. et al. Identification of the Different Roles and Potential Mechanisms of T Isoforms in the Tumor Recurrence and Cell Cycle of Chordomas. OncoTargets and therapy 12, 11777-11791, doi:10.2147/ott.S232526 (2019). 89 303 Georgakilas, A. G., Martin, O. A. & Bonner, W. M. p21: A Two-Faced Genome Guardian. Trends in molecular medicine 23, 310-319, doi:10.1016/j.molmed.2017.02.001 (2017). 304 Hu, F. et al. Lung adenocarcinoma resistance to therapy with EGFR‑tyrosine kinase inhibitors is related to increased expression of cancer stem cell markers SOX2, OCT4 and NANOG. Oncology reports 43, 727-735, doi:10.3892/or.2019.7454 (2020). 305 Sławek, S. et al. Pluripotency transcription factors in lung cancer-a review. Tumour biology 37, 4241-4249, doi:10.1007/s13277-015-4407-x (2016). 306 Chiou, S. H. et al. Coexpression of Oct4 and Nanog enhances malignancy in lung adenocarcinoma by inducing cancer stem cell-like properties and epithelial-mesenchymal transdifferentiation. Cancer Res 70, 10433-10444, doi:10.1158/0008-5472.Can-10-2638 (2010). 307 Silva-Oliveira, R. J. et al. Cytotoxicity of allitinib, an irreversible anti-EGFR agent, in a large panel of human cancer-derived cell lines: KRAS mutation status as a predictive biomarker. Cellular oncology 39, 253-263, doi:10.1007/s13402-016-0270-z (2016). 308 Pao, W. et al. KRAS mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. PLoS medicine 2, e17, doi:10.1371/journal.pmed.0020017 (2005). 309 Silva-Oliveira, R. J. et al. AKT can modulate the in vitro response of HNSCC cells to irreversible EGFR inhibitors. Oncotarget 8, 53288-53301, doi:10.18632/oncotarget.18395 (2017). 310 Bushweller, J. H. Targeting transcription factors in cancer - from undruggable to reality. Nature reviews. Cancer 19, 611-624, doi:10.1038/s41568-019-0196-7 (2019). 311 Kwiatkowski, N. et al. Targeting transcription regulation in cancer with a covalent CDK7 inhibitor. Nature 511, 616-620, doi:10.1038/nature13393 (2014). 312 Rasool, R. U. et al. CDK7 Inhibition Suppresses Castration-Resistant Prostate Cancer through MED1 Inactivation. Cancer discovery 9, 1538-1555, doi:10.1158/2159-8290.Cd-19-0189 (2019). 313 Martín-Orozco, R. M. et al. EGF prevents the neuroendocrine differentiation of LNCaP cells induced by serum deprivation: the modulator role of PI3K/Akt. Neoplasia 9, 614-624, doi:10.1593/neo.07337 (2007). 314 Fournier, C. et al. EGF signalling in epithelial carcinoma cells utilizes preformed receptor homoclusters, with larger heteroclusters post activation. bioRvix , 305292, doi:10.1101/305292 (2018). 315 Chipumuro, E. et al. CDK7 inhibition suppresses super-enhancer-linked oncogenic transcription in MYCN-driven cancer. Cell 159, 1126-1139, doi:10.1016/j.cell.2014.10.024 (2014). 316 Eliades, P. et al. High MITF Expression Is Associated with Super-Enhancers and Suppressed by CDK7 Inhibition in Melanoma. The Journal of investigative dermatology 138, 1582-1590, doi:10.1016/j.jid.2017.09.056 (2018). 317 Nagaraja, S. et al. Transcriptional Dependencies in Diffuse Intrinsic Pontine Glioma. Cancer cell 31, 635-652.e636, doi:10.1016/j.ccell.2017.03.011 (2017). 318 Christensen, C. L. et al. Targeting transcriptional addictions in small cell lung cancer with a covalent CDK7 inhibitor. Cancer cell 26, 909-922, doi:10.1016/j.ccell.2014.10.019 (2014). 319 Zhang, Z. et al. Preclinical Efficacy and Molecular Mechanism of Targeting CDK7-Dependent Transcriptional Addiction in Ovarian Cancer. Molecular cancer therapeutics 16, 1739-1750, doi:10.1158/1535-7163.Mct-17-0078 (2017). 320 Wang, Y. et al. CDK7-dependent transcriptional addiction in triple-negative breast cancer. Cell 163, 174-186, doi:10.1016/j.cell.2015.08.063 (2015).