Idea Transcript
Stem Cells and Steroid Metabolism In Prostate Cancer Minja Pfeiffer
Päättötyö4-toisinpäin.indd 1
13.3.2012 16.33
Päättötyö4-toisinpäin.indd 2
13.3.2012 16.33
Päättötyö4-toisinpäin.indd 1
13.3.2012 16.33
The research presented in this thesis was performed at the department of Experimental Urology (head: prof. dr. J. A. Schalken), Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands, and was financially supported by the European Commission (MEST-CT-2005-020970). ISBN 978-952-93-0347-2 Thesis layout: Janne Torikka, www.ev.fi Printed by: Ipskamp Drukkers
Päättötyö4-toisinpäin.indd 2
13.3.2012 16.33
Stem Cells and Steroid Metabolism in Prostate Cancer
Minja Pfeiffer
Päättötyö4-toisinpäin.indd 3
13.3.2012 16.33
Promotoren:
Prof. dr. J.A. Schalken Prof. dr. P.F.A. Mulders
Copromotor:
Dr. J.P.M. Sedelaar
Manuscriptcommissie:
Prof. dr. F. Sweep Prof. dr. E.J. van Zoelen Prof. dr. J.T. Isaacs (The Johns Hopkins University School of Medicine)
Paranimfen: R. Cremers N. Rijken
Päättötyö4-toisinpäin.indd 4
13.3.2012 16.33
Stem Cells and Steroid Metabolism in Prostate Cancer
Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. mr. S.C.J.J. Kortmann volgens besluit van het dollege van decanen in het openbaar te verdedigen op woensdag 9 mei 2012 om 10:30 uur precies door
Minja Johanna Pfeiffer geboren op 10 september 1981 te Tampere, Finland
Päättötyö4-toisinpäin.indd 5
13.3.2012 16.33
Doctoral thesis supervisors: Prof. dr. J.A. Schalken Prof. dr. P.F.A. Mulders
Doctoral thesis co-supervisor: Dr. J.P.M. Sedelaar
Doctoral Thesis Committee:
Prof. dr. F. Sweep Prof. dr. E.J. van Zoelen Prof. dr. J.T. Isaacs (The Johns Hopkins University School of Medicine)
Paranimfen: R. Cremers N. Rijken
Päättötyö4-toisinpäin.indd 6
13.3.2012 16.33
Stem Cells and Steroid Metabolism in Prostate Cancer
Doctoral thesis to obtain the degree of doctor from Radboud University Nijmegen on the authority of the Rector Magnificus, prof. dr. S.C.J.J. Kortmann, according to the decision of the Council of Deans to be defended in public on Wednesday May 9th, 2012 at 10:30 hours by
Minja Johanna Pfeiffer born in Tampere, Finland on September 10th, 1981
Päättötyö4-toisinpäin.indd 7
13.3.2012 16.33
Päättötyö4-toisinpäin.indd 8
13.3.2012 16.33
Table of Contents
Chapter 1..........................................................................................................11
Introduction .........................................................................................................................................12 Outline of the thesis ............................................................................................................................20
Chapter 2 .........................................................................................................21 Stem cell characteristics in prostate cancer cell lines Eur Urol 2010;57:246-54.
Chapter 3 .........................................................................................................35 An in vitro model for preclinical testing of endocrine therapy combinations for prostate cancer. Prostate 2010;70:1524-32.
Chapter 4 .........................................................................................................51 Steroidogenic enzymes and stem cell markers are upregulated during androgen deprivation in prostate cancer. Mol Med 2011;17:657-64.
Chapter 5 .........................................................................................................67 AKR1C3 – a potential marker and therapeutic target in castration resistant prostate cancer Manuscript in preparation
Chapter 6 .........................................................................................................79 Discussion
Chapter 7 .........................................................................................................87
Summary in English ...........................................................................................................................88 Summary in Dutch ..............................................................................................................................89 Curriculum vitae ..................................................................................................................................90 List of Publications ..............................................................................................................................91 Acknowledgements ............................................................................................................................92
Päättötyö4-toisinpäin.indd 9
13.3.2012 16.33
10
Päättötyö4-toisinpäin.indd 10
13.3.2012 16.33
Chapter 1
Päättötyö4-toisinpäin.indd 11
13.3.2012 16.33
Chapter
1
Introduction Prostate
The prostate gland is a small, walnut-sized organ in men that surrounds the first part of the urethra under the bladder. Its function is to produce the major components of seminal fluid. The normal function of the prostate and the growth and maintenance of prostate cancer are dependent on androgens (1).
Androgen action
An androgen exerts its biological effect through a ligand-dependent nuclear receptor transcription factor, the androgen receptor (AR). The circulating serum androgen testosterone is taken into prostate cells and converted into dihydrotestosterone (DHT), which has a high affinity for the AR. Upon ligand binding, AR is released from bound chaperones, such as heat-shock proteins, dimerizes with another AR protein, and translocates into the nucleus. In the nucleus it binds to androgen response elements on the DNA and induces gene expression with specific transcription cofactors (reviewed by Heinlein and Chang (2)).
Biosynthesis and metabolism of androgens
Androgens are produced primarily by the testes and the adrenal glands. However, peripheral tissues, such as liver, adipose tissue, skin, brain, and prostate also play roles in converting weak androgens to more potent ones (3-7). De novo synthesis of steroids starts with the conversion of cholesterol by steroidogenic enzymes and passes through several steps to the potent androgens, testosterone and DHT, the two most important androgens in adult men (figure 1).
Prostate cancer treatment
Prostate cancer is the most frequently diagnosed cancer and the second leading cause of cancerrelated deaths in men in the United States (estimate for 2009) (8) and Europe (estimate for 2006) (9). The incidence of prostate cancer is clearly related to age, and other factors, such as race, family history and diet, have an impact on the risk of developing prostate cancer (10). The role of androgens in prostate cancer was demonstrated by Huggins and Hodges (1), and over 60 years after their pioneer work, hormone ablation still remains the main form of therapy for advanced, metastatic prostate cancer. Endocrine therapy or androgen deprivation therapy aims to lower the levels of circulating testosterone in the patient. This can be achieved surgically (orchiectomy) and pharmacologically (gonadotropin-releasing hormone agonists) either alone or in combination with an anti-androgen as secondary endocrine therapy. The anti-androgen inhibits binding of the ligand to AR. Endocrine therapy reduces the level of a secretory product prostate-specific antigen (PSA) and causes the tumor to regress. Unfortunately, in most of the patients treated with medical or surgical castration, the cancer recurs after an average of 18-24 months and may be considered as castration-resistant prostate cancer (CRPC) (11).
12
Päättötyö4-toisinpäin.indd 12
13.3.2012 16.33
Chapter
1
Figure 1. A schematic overview of the steroidogenic pathway. Potent androgens testosterone and dihydrotestosterone are depicted in boxes highlighted with boundaries. Testosterone is formed mainly by the testes. Dehydroepiandrosterone (DHEA) and androstenedione are adrenal androgen precursors. Abbreviations: AMACR, alpha-methylacylCoA racemase; FASN, fatty acid synthase; SREB, G protein-coupled receptor; StAR, steroidogenic acute regulatory protein; CYP, cytochrome P450; CYB5, cytochrome b5; HSD, hydroxysteroid dehydrogenase; AKR, aldo-keto reductase; SRD, 5α-reductase; UGT, UDP glucuronosyltransferase.
Tissue androgen levels after endocrine therapy
The clinical efficacy of androgen deprivation therapy is evaluated by the reduction in PSA and testosterone levels in the serum. Even though the treatment achieves castrate levels (32 cells were scored.
Single-cell cloning
Colonies with different morphologies were plated again as single cells to test their renewal ability. Single cells from DU145 and PC-3 colonies were seeded onto 96-well plates in a density of 1–2 cells per well with conditioned medium (50%). Morphologies of the established colonies were scored and counted under a light microscope.
Immunohistochemistry
The expression of several markers in the DU145 colonies was examined by immunohistochemistry. Cells were plated at low density on glass slides and left to grow for 8d until colonies formed. Colonies were fixed with ice-cold acetone for 10 min or with 4% paraformaldehyde at room temperature for 20 min. Acetone-fixed colonies were stained with monoclonal antibodies: CK18 (clone DC10, DAKO, Glostrup, Denmark),CK5 (clone RCK103, Euro-Diagnostica, Arnhem, the Netherlands), BCRP (Millipore, Billerica, MA, USA), and CD49b (α2-integrin, clone Gi-14, Dr S Santoso, Giessen, Germany). In addition, 4% paraformaldehyde-fixed colonies were stained with CD133/2 (Miltenyi Biotec) and nestin (Millipore). CK18 (6 ng/ml), CK5 (culture supernatant); CD49b (2.5 ng/ml) antibodies were incubated for 1 h; CD133/2 (0.25 ng/ml) antibodies were incubated for 2 h at room temperature; and antibodies against nestin (5 ng/ml) and BCRP (2.5 ng/ml) were incubated overnight at 4-8°C. 24
Päättötyö4-toisinpäin.indd 24
13.3.2012 16.33
Detection of the antibody binding was performed using Powervision poly-HRP-goat anti-mouse/ rabbit/rat IgG (ImmuniLogic, Duiven, the Netherlands) as the secondary antibody and 303-diaminobenzidine (Power DAB, ImmunoLogic) to observe the peroxidase activity. The nuclei were counterstained with haematoxylin. To evaluate the specificity of the antibodies, known positive tissues or cells were used as controls. Normal human prostate tissue was used as a positive control for CK18, CK5, and α2-integrin; human hippocampus tissue was used to control the nestin staining; and Caco-2 cells were used as a positive control for CD133 and BCRP. In negative controls, only secondary (not primary) antibody was used. The immunohistochemical stainings were studied under a light microscope.
Chapter
2
Results CD133 is expressed only in DU145
From six PCa cell lines, DU145 showed expression for CD133 in a small proportion (0.01 ± 0.01%) of the cells (Fig. 1). Because the amount of positive cells was very low, a reanalysis of the flow cytometry-selected cells could not be performed. In DuCaP, LAPC-4, 22Rv1, LNCaP, and PC-3 cell lines, CD133 expression was not detected in FACS analysis. A small population (0.15 ± 0.20%) of BCRP+ DU145 cells was detected cells. Other cell lines were not tested.
Fig. 1 - Fluorescence-activated cell sorting (FACS) analysis for the CD133 stem cell marker in DU145 prostate cancer cell line; (A) isotype control; (B) DU145 cells showing a small detectable population (0.01%) of CD133+ cells (50 000 cells were analysed).
Prostate cancer cell lines show different colony morphologies
When plated in vitro at low density, five (DU145, 22Rv1, LAPC-4, DuCaP, and LNCaP) of six cell lines show all three colony morphologies: holoclones, meroclones, and paraclones. PC-3 cells did not form tightly packed, round holoclones. Fig. 2 shows light microscope pictures of typical colonies from DU145 (Fig. 2A–2C), 22Rv1 (Fig. 2D–2F), LAPC-4 (Fig. 2G–2I), DuCaP (Fig. 2J–2L), LNCaP (Fig. 2M–2O), and PC-3 (Fig. 2P–2Q). Depending on the cell line, growth kinetics (number of colonies) was assessed after 1–3 wk. In some cell lines, the cell morphology differed between the colony types, where holoclones contained generally small and tightly packed cells and paraclones consisted of round or flattened, more irregularly shaped and loosely packed cells. In DuCaP, LAPC-4, and 22Rv1, many of the holoclones and meroclones started to grow three-dimensionally when the cell number in the colonies increased with culturing time, while the paraclones stayed as a monolayer. The colony-forming efficiency of each cell line and the proportion of holo-, mero-, and paraclones in at least three independent experiments are depicted in Fig. 3. 25
Päättötyö4-toisinpäin.indd 25
13.3.2012 16.33
We noticed in DU145 that when there is no selective pressure in terms of plating density on the cell line (ie, with passing the cells in a standard dilution [1:10]), more paraclones start to form when the passage number progresses. Instead, when cells are plated at low density (a high dilution, such as 1:30), the proportion of holoclones increases. In other words, in stable conditions, the stem-like cells that form holoclones are in a more quiescent state but start to proliferate when grown at low density. This observation supports the evidence that DU145 holoclones indeed contain the more stem-like cells because their proportion increases with environmental pressure.
CD133 or breast cancer resistance protein–positive cells are not more clonogenic than the negative cells in DU145
Because Collins et al [10] showed that the CD133+ primary PCa cells were more clonogenic than the negative population, we assayed the colony-forming efficiency of these two populations in the DU145 cell line. If CD133 marked the CSC population, one would expect that CD133+ cells would generate more colonies—primarily holoclones. Yet, FACS isolated CD133+ DU145 cells were not able to generate more colonies or more holoclones than CD133- at low density. The colony-forming efficiency varied strongly between and within experiments. Also, BCRP+ cells did not form significantly more holoclones or any colonies (data not shown). Statistical significance was calculated using a student t test.
Holoclones contain cells that show self-renewal
DU145 holoclones can be maintained in culture for several (at least four) passages, showing renewal ability, whereas paraclones can be kept in culture for one to two passages (Table 1). Cells isolated from holoclones and plated as single cells produced mainly holoclones (80–100%), but meroclones and paraclones were also observed. Cells from paraclones were able to generate only paraclones and occasional meroclones, depending on the morphologic purity of the plated colony. Instead, colonies from PC-3 cell line could be maintained in culture up to four passages (the end point of the study), independent of the morphology. The colonies showed great plasticity, being able to produce different colony morphologies regardless of the original morphology.
26
Päättötyö4-toisinpäin.indd 26
13.3.2012 16.33
Chapter
2
Fig. 2 – Observed colony morphologies in prostate cancer cell lines: DU145 (6 d; A–C), 22Rv1 (13 d; D–F), LAPC-4 (8 d; G–I), DuCaP (35 d; J–L), and LNCaP (12d; M–O) formed three morphologically different colonies: holoclones, meroclones, and paraclones. In PC-3 (11 d; P–O), a typical holoclone phenotype was not observed; only meroclones and paraclones were detected. The holoclones are generally more round and tightly packed, whereas paraclones are irregular in composition and often contain more elongated or flattened cells. The colonies with an intermediate phenotype are meroclones. Colonies with at least 32 cells (five cell doublings) were considered colonies. Bar indicates 200 µm.
27
Päättötyö4-toisinpäin.indd 27
13.3.2012 16.33
Fig. 3 – Colony forming efficiency and the proportion of each colony type in prostate cancer cell lines DU145, 22Rv1, LAPC-4, DuCaP, LNCaP, and PC-3 shown as box plots. The cells were plated at low density as single cells; after 1–4 wk, the colonies that contained >32 cells were counted under a light microscope. The data were analysed using the Statistical Package for the Social Sciences. Median value (horizontal line), outliers (open circle), and extremes (asterisk) are shown. CFE = colony forming efficiency; HOLO = holoclones; MERO =meroclones; PARA = paraclones.
28
Päättötyö4-toisinpäin.indd 28
13.3.2012 16.33
Table 1 – The ability to passage DU145 colonies. After plating single cells at low density (passage 0), some of the formed holoclones and paraclones were isolated and plated again as single cells. The replating of the colonies was continued until the cells were unable to form new colonies. Holoclones could be maintained in culture for at least four passages (the end point of the study), whereas paraclones vanished within one or two passages.
Chapter
2
Stem cell and differentiation markers are expressed in DU145 colonies
To support the observation that holoclones contain more stem-like cells, immunohistochemical staining of stem cell and differentiation markers was assayed in the various colonies. DU145 cells express the putative stem cell markers α2-integrin, nestin, and BCRP as well as CK5 and cytokeratin 18 (CK18), but there was no detectable staining for the stem cell marker CD133. The staining of each marker was compared among holoclones and paraclones. Representative pictures of each staining are shown in Fig. 4. The quantity of α2-integrin protein staining was similar in holoclones and paraclones in a lower passage (P91), where three-quarters of all colonies were 95–100% positive for the marker. In a later passage (P107), the α2-integrin staining in the holoclones stayed the same but was reduced in paraclones, where most of the colonies showed no or little staining for the marker. It must be noted that some of the holoclones were negative (0–5% of the cells within a colony stained for the marker) and some paraclones were positive (95–100%). The level of the BCRP protein staining was similar in holoclones and paraclones, with the difference that the staining in holoclones was mostly in the cytoplasm or cell membrane, whereas in paraclones, the localisation of the protein was restricted mainly to the nucleus. In contrast, the staining of the marker for transient amplifying cells, CK5, was more intense in paraclones and less intense in holoclones, although completely negative paraclones were also observed, whereas in most of the holoclones at least some cells were stained. All DU145 cells stained strongly cytoplasmic for the luminal cell marker CK18 and weakly (0–100%) for the stem cell marker nestin regardless of the colony type. A later passage contained generally more nestin-stained colonies than the earlier passage.
Fig. 4 – Immunohistochemical staining of DU145 colonies. Eight-day-old DU145 holoclones and paraclones were stained for the putative stem cell markers CD133, α2-integrin, nestin, and breast cancer resistance protein as well as for the differentiation markers CK5 and CK18. Bar indicates 200 µm.
29
Päättötyö4-toisinpäin.indd 29
13.3.2012 16.33
Discussion The aim of this study was to determine whether in the commonly used PCa cell lines cells with stem cell characteristics could be identified. There are previous reports of stem-like cells in cell lines based on the dye exclusion capacity, revealing a side population in FACS; however, Patrawala et al did not detect a side population in the PCa cell lines DU145, PC-3 and LNCaP [31]. Putative prostate CSCs in cell lines have been found by utilizing the cell surface markers CD44 [26] and CD133 [32]. Based on previous studies with PCa cells, we decided to look at the CD133 expression and the colony formation at low density in PCa cell lines DU145, 22Rv1, LAPC-4, DuCaP, LNCaP, and PC-3. In five out of six PCa cell lines, there was no detectable CD133+ population measured by FACS. Collins et al found a subpopulation of CD44+/α2β1high/CD133+ cells in PCa specimens that show stem cell characteristics in vitro [10]. There is evidence that CD133 is a stem cell marker in the normal prostate [33]. If CD133 marks the CSCs in primary PCa specimens, the lack of CD133 in cell lines may be explained by the long-term in vitro culturing, during which the conditions have altered the protein expression. In the DU145 cell line, we were able to find a small population (0.01%) of CD133+ cells. To examine the properties of the CD133+ and CD133- DU145 cells further, the colony-formation efficiency of FACS-selected positive and negative populations and unselected cells was compared. However, there was no difference in the ability of these three populations to form colonies. If the CD133+ population is enriched with stem-like/ progenitor cells, one would expect these cells to form more colonies. Also, the proportion of holoclones (formed by cells with stem cell properties) was not higher, and the proportion of paraclones (formed by more differentiated cells) was not lower in the CD133+ population than in the CD133- or unselected population. The colony-forming assay was also performed with cells selected with CD133-labelled magnetic beads (MACS, Miltenyi Biotec), but the results were similar (data not shown). Altogether, from these observations we can conclude that the CD133+ population was not enriched with stem-like cells and is not a marker for CSCs in PCa cell lines. However, this does not exclude the possibility that CD133 is a stem cell marker in the cancerous tissue in vivo. Colony-formation assays performed with DU145 BCRP+ (0.15%) and BCRP- cells did not show significant differences between these populations. It has to be noted that the FACS sorted CD133+ and BCRP+ cells formed in fact a shift from the main population, not a clear separate population. Thus, the division into two populations might have been fairly artificial, and positive cells were not truly detectable in FACS. Interestingly, five (DU145, 22Rv1, LAPC-4, DuCaP, and LNCaP) of the examined six PCa cell lines formed holoclones, meroclones, and paraclones at low density. These are thought to derive from stem cells, early progenitors, and late progenitors, respectively [13]. The number of colonies and the proportional number of holo-, mero-, and paraclones may, however, change depending on the culturing conditions. An important factor is the extracellular matrix, which affects cell proliferation and survival. Culturing these cancer cell lines in a friendlier environment, on coated plates, might yield a larger number of colonies. We did not detect morphologically typical holoclones in the PC-3 cell line, because the colonies that were initially considered as possible holoclones rapidly changed their morphology. It was described earlier that PC3 cells have an intrinsic impaired cell–cell adhesion through E-cadherin because of the lack of alpha-catenin expression (homozygous gene deletion) [34]; thus, they are not able to form tightly packed colonies. However, Li et al [35] showed recently that PC-3 cells formed holoclones that contain cells with stem-like properties. We did not observe those round, tightly packed colonies in the PC-3 line. In all of the cell lines, the definition of the three colony morphologies differs somewhat from each other, and there are no strict borderlines between the colony types, which makes the grading fairly subjective. In fact, a continuous gradient of different colony types made the colonies difficult to distinguish from each other. Based on our results and taking the previous studies with different colony forms [13,14] into consideration, we can deduce that at least DU145, 30
Päättötyö4-toisinpäin.indd 30
13.3.2012 16.33
22Rv1, LAPC-4, DuCaP, and LNCaP may contain cells with the hierarchical organization of stem, transient amplifying, and differentiated cells. In addition, the PCa cell lines used in this study are able to grow anchorage independently in soft agar (data not shown), which is a character of stem/ progenitor cells.
Chapter
2
In DU145, only holoclones can be maintained in serial culture, thus providing proof of self-renewal— a crucial property of stem cells. In contrast, paraclones contain cells that have only a limited capacity to proliferate, because they died out within two passages. Additionally, cells from paraclones could form only paraclones, never holoclones. In contrast, the replated holoclones produced mainly holoclones, but some occasional meroclones and paraclones were also detected. This gives us proof that some of the cells in holoclones have undergone differentiation in the hierarchical ladder from stem-like cells into more differentiated early and late progenitors. Different colonies of PC-3 cells were all able to stay in culture with morphologically variable daughter colonies, suggesting that the strict morphologic criteria used to identify DU145 colonies cannot be directly applied to PC-3. In fact, many PC-3 colonies of undefined morphology may contain stem-like cells. The differences in the expression of a set of markers were studied in DU145 holoclones and paraclones. There was no expression of the putative stem cell marker CD133. The expression of α2integrin was higher in the holoclones than in paraclones of an older passage, but not in a lower passage. This is in agreement with an earlier study [10] that suggests that α2-integrin is a marker for stem cells and transient amplifying cells. Culturing the cells might push the majority towards differentiation but maintaining a stem-like subset of α2-integrin+ cells that keep forming holoclones. The expression of BCRP did not differ in the intensity between holoclones and paraclones, but the localisation of the protein was restricted in holoclones mostly to the cytoplasm or cell membrane and in paraclones mainly to the nucleus. Because the BCRP actively effluxes cytotoxins from the cell [17], we can assume that the protein has to be present outside the nucleus to contribute to the toxin-recessive phenotype of the stem cells. Therefore, it can be speculated that the BCRP on the cell membrane might be a marker for more stem-like cells. However, there was no significant difference in colony formation between FACS-selected DU145 BCRP+ and BCRP- cells. The expression of the neural stem cell marker nestin in DU145 colonies was mostly very weak, ranging between 0% and 100% positive cells in any colony type. At least in PCa cell lines, nestin does not seem to be a useful marker for defining CSCs. The presence of CK5- cells indicates that there must be cells that are only positive for the luminal marker CK18, because all the cells were CK18+. Also, CK5+/CK18+ cells must exist that present the intermediate (transiently amplifying) basal–luminal cells that are the progenitors of the differentiated CK5-/CK18+ luminal cells [23].
Conclusions Most of the examined cancer cell lines did not express the potential stem cell marker CD133, implicating that it is not a marker for stem cells in PCa cell lines. Instead, the cell lines DU145, 22Rv1, LAPC-4, LNCaP, and DuCaP did contain cells that can form holoclones. Holoclones contain selfrenewing cells and express the putative stem cell markers α2-integrin and BCRP. We suggest that PCa cell lines may contain stem-like cells and should be characterized and studied further. Acknowledgement statement: The authors acknowledge Rob Woestenenk at the central haematology laboratory for performing the flow cytometric sorting. This work is part of the Cancer Cure Early Stage Research Training (CANCURE) project funded by the European Commission (MEST-CT-2005-020970).
31
Päättötyö4-toisinpäin.indd 31
13.3.2012 16.33
References 1. Lapidot T, Sirard C, Vormoor J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 1994;367:645–8. 2. Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 1997;3:730–7. 3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003;100:3983–8. 4. Singh SK, Clarke ID, Terasaki M, et al. Identification of a cancer stem cell in human brain tumors. Cancer Res 2003;63:5821–8. 5. Ricci-Vitiani L, Lombardi DG, Pilozzi E, et al. Identification and expansion of human colon-cancer-initiating cells. Nature 2007;445:111–5. 6. O’Brien CA, Pollett A, Gallinger S, Dick JE. A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 2007;445:106–10. 7. Yin S, Li J, Hu C, et al. CD133 positive hepatocellular carcinoma cells possess high capacity for tumorigenicity. Int J Cancer 2007;120: 1444–50. 8. Li C, Heidt DG, Dalerba P, et al. Identification of pancreatic cancer stem cells. Cancer Res 2007;67:1030–7. 9. Fang D, Nguyen TK, Leishear K, et al. A tumorigenic subpopulation with stem cell properties in melanomas. Cancer Res 2005;65:9328–37. 10. Collins AT, Berry PA, Hyde C, Stower MJ, Maitland NJ. Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res 2005;65:10946–51. 11. Patrawala L, Calhoun-Davis T, Schneider-Broussard R, Tang DG. Hierarchical organization of prostate cancer cells in xenograft tumors: the CD44+{alpha}2{beta}1+ cell population is enriched in tumor-initiating cells. Cancer Res 2007;67:6796–805. 12. Brown MD, Gilmore PE, Hart CA, et al. Characterization of benign and malignant prostate epithelial Hoechst 33342 side populations. Prostate 2007;37:1384–96. 13. Barrandon Y, Green H. Three clonal types of keratinocyte with different capacities for multiplication. Proc Natl Acad Sci U S A 1987;84:2302–6. 14. Tudor D, Locke M, Owen-Jones E, Mackenzie IC. Intrinsic patterns of behavior of epithelial stem cells. J Investig Dermatol Symp Proc 2004;9:208–14. 15. Yin AH, Miraglia S, Zanjani ED, et al. AC133, a novel marker for human hematopoietic stem and progenitor cells. Blood 1997;90:5002–12. 16. Uchida N, Buck DW, He D, et al. Direct isolation of human central nervous system stem cells. Proc Natl Acad Sci U S A 2000; 97:14720–5. 17. Doyle LA, Yang W, Abruzzo LV, et al. A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc Natl Acad Sci U S A 1998;95:15665–70. 18. Hadnagy A, Gaboury L, Beaulieu R, Balicki D. SP analysis may be used to identify cancer stem cell populations. Exp Cell Res 2006;312:3701–10. 19. Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell 1990;60:585–95. 20. Dahlstrand J, Collins VP, Lendahl U. Expression of the class VI intermediate filament nestin in human central nervous system tumors. Cancer Res 1992;52:5334–41. 21. Tohyama T, Lee VM, Rorke LB, Marvin M, McKay RD, Trojanowski JQ. Nestin expression in embryonic human neuroepithelium and in human neuroepithelial tumor cells. Lab Invest 1992;66:303–13. 22. Kleeberger W, Bova GS, Nielsen ME, et al. Roles for the stem cell associated intermediate filament nestin in prostate cancer migration and metastasis. Cancer Res 2007;67:9199–206.
32
Päättötyö4-toisinpäin.indd 32
13.3.2012 16.33
23. van Leenders G, Dijkman H, Hulsbergen-van de Kaa C, Ruiter D, Schalken J. Demonstration of intermediate cells during human prostate epithelial differentiation in situ and in vitro using triple- staining confocal scanning microscopy. Lab Invest 2000;80:1251–8.
Chapter
2
24. van Leenders G, van Balken B, Aalders T, Hulsbergen-van de Kaa C, Ruiter D, Schalken J. Intermediate cells in normal and malignant prostate epithelium express c-MET: implications for prostate cancer invasion. Prostate 2002;51:98–107. 25. Zheng H, Wasylyk C, Ayadi A, et al. The transcription factor Net regulates the angiogenic switch. Genes Dev 2003;17:2283–97. 26. Patrawala L, Calhoun T, Schneider-Broussard R, et al. Highly purified CD44+ prostate cancer cells from xenograft human tumors are enriched in tumorigenic and metastatic progenitor cells. Oncogene 2006;25:1696–708. 27. Locke M, Heywood M, Fawell S, Mackenzie IC. Retention of intrinsic stem cell hierarchies in carcinoma-derived cell lines. Cancer Res 2005;65:8944–50. 28. Resnicoff M, Medrano EE, Podhajcer OL, Bravo AI, Bover L, Mordoh J. Subpopulations of MCF7 cells separated by Percoll gradient centrifugation: a model to analyze the heterogeneity of human breast cancer. Proc Natl Acad Sci U S A 1987;84:7295–9. 29. Hirschmann-Jax C, Foster AE, Wulf GG, et al. A distinct ‘‘side population’’ of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci U S A 2004;101:14228–33. 30. Setoguchi T, Taga T, Kondo T. Cancer stem cells persist in many cancer cell lines. Cell Cycle 2004;3:414–5. 31. Patrawala L, Calhoun T, Schneider-Broussard R, Zhou J, Claypool K, Tang DG. Side population is enriched in tumorigenic, stem-like cancer cells, whereas ABCG2+ and ABCG2- cancer cells are similarly tumorigenic. Cancer Res 2005;65:6207–19. 32. Wei C, Guomin W, Yujun L, Ruizhe Q. Cancer stem-like cells in human prostate carcinoma cells DU145: the seeds of the cell line? Cancer Biol Ther 2007;6:763–8. 33. Richardson GD, Robson CN, Lang SH, Neal DE, Maitland NJ, Collins AT. CD133, a novel marker for human prostatic epithelial stem cells. J Cell Sci 2004;117:3539–45. 34. Morton RA, Ewing CM, Nagafuchi A, Tsukita S, Isaacs WB. Reduction of E-cadherin levels and deletion of the alphacatenin gene in human prostate cancer cells. Cancer Res 1993;53:3585–90. 35. Li H, Chen X, Calhoun-Davis T, Claypool K, Tang DG. PC3 human prostate carcinoma cell holoclones contain selfrenewing tumor initiating cells. Cancer Res 2008;68:1820–5.
33
Päättötyö4-toisinpäin.indd 33
13.3.2012 16.33
34
Päättötyö4-toisinpäin.indd 34
13.3.2012 16.33
Chapter 3
Päättötyö4-toisinpäin.indd 35
13.3.2012 16.33
Chapter 3 Prostate 2010;70:1524-32.
An In Vitro Model for Preclinical Testing of Endocrine Therapy Combinations for Prostate Cancer Minja J. Pfeiffer, Peter F. Mulders, and Jack A. Schalken Department of Urology, Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands
Abstract Background: Even though patients with prostate cancer commonly respond to endocrine treat-
ment, in most cases the disease progresses to castration resistant prostate cancer (CRPC). Our objective was to generate a novel cell line model representing the endocrine treatment naïve prostate cancer for testing treatments that target the androgen receptor (AR) and androgen metabolism.
Methods: After culturing DuCaP cells 20 passages with additional 1 nM R1881, DuCaP-N(naive) cell line was developed and validated for testing endocrine therapy combinations. Cell viability, apoptosis and cell cycle distribution were assessed in DuCaP and DuCaP-N when interfering with the hormonal content.
Results: Addition of 1 nM R1881 to DuCaP reduces cell viability and induces cell cycle inhibition
and apoptosis. Eventually, an androgen accustomed DuCaP-N cell line developed. An antiandrogen (bicalutamide), a histone deacetylase (HDAC) inhibitor (trichostatin A) and a 5α-reductase (SRD5A) inhibitor (finasteride) reduce cell viability, and their combinations give a synergistic response in inducing apoptosis.
Conclusions: The TMPRSS2-ERG expressing DuCaP-N cell line represents human prostate cancer prior to endocrine treatment, and its parental DuCaP cell line is a model for CRPC. These cell lines can be used for preclinical evaluation of compounds that target the androgenic pathway.
36
Päättötyö4-toisinpäin.indd 36
13.3.2012 16.33
Introduction Androgens play a critical role in the development and growth of prostate cancer [1]. Hence, current endocrine therapy treatments for metastatic prostate cancer aim to reduce the amount of circulating androgens by castration. This can be combined with AR antagonists, such as flutamide and bicalutamide [2]. However, although serum androgen levels reach castration levels (36). We were able to detect StAR, FASN, HSD17B2, HSD17B4, HSD17B10, AKR1C2, AKR1C3, SRD5A1, SRD5A3, RDH5, and AR mRNA in all of the samples, but with no statistically significant differences. The most prevalent difference between these two cell lines seems to be the up-regulation of AR in low androgen environment (Fig. 4).
40
Päättötyö4-toisinpäin.indd 40
13.3.2012 16.33
Chapter
3
Fig. 2. Induction of apoptosis and cell cycle inhibition in DuCaP with the addition/depletion of androgens. (A) ApoONE caspase-3/7 assay shows that addition of 1nM and 10 nM R1881 to RPMI-1640 supplemented with 10% FCS induces apoptosis. (B) The addition of 0.1 nM, 1 nM, and 10 nM R1881, and hormone depletion (CSS) induces cell cycle arrest. RFU, relative fluorescence units. Error bars, SEM. *P≤0.05; **P≤0.01.
Fig. 3. Expression of AR and TMPRSS2-ERG in DuCaP and DuCaP-N cells. (A) A representative image of a Western blot for AR protein expression. (B) A representative agarose gel image of the expression of TMPRSS2-ERG fusion transcripts. β-2M, β-2-microglobulin. RT, negative control.
41
Päättötyö4-toisinpäin.indd 41
13.3.2012 16.33
Fig. 4. Expression of genes involved in steroid metabolism in DuCaP and DuCaP-N. The expression value foldchanges in DuCaP (gray bars) compared to DuCaP-N (black bars) measured by real time PCR. Genes CYP11A1, CYP17A1, HSB3B2, HSB17B3 and SRD5A2 were excluded from the figure based on high Cp-values. Error bars, SEM of three samples.
Validation of DuCaP-N as a Preclinical Model for Endocrine Therapy Combinations
In order to validate the DuCaP-N model, we tested compounds that interfere with AR signaling at different levels. Cells were grown with the physiological circulating androgen, testosterone (1 nM). We used an AR antagonist (bicalutamide) and a 5α-reductase inhibitor (finasteride). Since HDAC inhibitors have a strong effect on gene fusion positive prostate cancer cell lines [18], we included a prototype HDAC inhibitor, trichostatin A, in our assays. To study whether a change in chromatin structure could enhance the effect of AR signaling inhibition also treatment combinations were tested. The cell viability of DuCaP-N was impaired (P≤0.000) when treated with agents targeting AR (bicalutamide 5 mM and TSA 5 nM) or testosterone metabolism (finasteride, 5 mM), and when depleted of hormones. Treatment with bicalutamide, TSA, or finasteride resulted in a mean relative cell viability of 50%, 51%, and 72%, respectively. Interestingly, when DuCaP-N is grown in medium containing CSS, some cells do survive despite the hormone depletion (mean cell viability 41%). Furthermore, combinations of agents decreased the number of viable cells compared to single agent treatments. Bicalutamide with TSA (mean cell viability 33%), and finasteride with TSA (mean cell viability 39%) had an additive effect (P=0.001 and P=0.005, respectively). Bicalutamide together with finasteride did not result as additive (mean cell viability 51%). When bicalutamide is administrated in combination with TSA, or with both TSA and finasteride (mean cell viability 20%, P≤0.000 when compared to combinations with two agents), the cell viability is inferior to hormonal depletion suggesting combi42
Päättötyö4-toisinpäin.indd 42
13.3.2012 16.33
natorial drug treatment with different mechanism of action being a better approach than hormone depletion only (Fig. 5). Also other concentrations were tested for bicalutamide (1–10 mM), TSA (1–10 nM), and finasteride (1–10 mM). All concentrations of bicalutamide gave a corresponding 50% cell growth inhibition, but TSA and finasteride acted in a dose dependent manner (results not shown). The primary interest of our research was to create a cell line model that resembles hormone treatment naïve prostate cancer, and therefore less focus was given to the effect of therapeutic agents on the parental DuCaP cells. Since no positive growth response was observed in DuCaP cells when cultured in RPMI-1640 supplemented with 10% FCS and 0.001–1 nM R1881, this cell line did not offer us additional interest within the study objective. We did observe, however, agonistic behavior of bicalutamide (10 mM) when added to RPMI-1640 supplemented with 10% CSS and 1 nM testosterone or R1881 in cell proliferation assays (results not shown).
Chapter
3
Fig. 5. DuCaP-N relative cell viability after 10 days of culture with therapeutic agents. Cells were grown for 10 days in RPMI-1640 supplemented with 10% CSS, 1 nM testosterone and therapeutics (B= bicalutamide 5 μM, T= TSA 5 nM, F= finasteride 5μM) or in hormone depleted medium only (RPMI-1640with10%CSS). Statistical significant differences were calculated between non-treated and single agent treated cells or hormone stripped sample, single agent and twoagent combinational treatments, and between two- and triple-agent treatments. Errorbars, SEM. *P≤0.05; **P≤0.01.
Reduced DuCaP-N Cell Viability Is Mediated Partly Through Apoptosis
When DuCaP-N cells were treated for 48 hr with bicalutamide 10 mM, TSA 10nM or finasteride 10 mM, a slight, although statistically non-significant, increase in apoptotic cells was observed (1.2-, 1.1-, and 1.2-fold, respectively). Instead, when bicalutamide was combined with TSA, the apoptotic rate increased 2.3-fold (P≤0.000) compared to treatment with only bicalutamide or TSA, showing a strong synergy with these two compounds. Similarly, finasteride together with TSA resulted in a synergistic 1.4-fold (P=0.039) response compared to treatment with only finasteride or TSA. A 1.6-fold response (P=0.006) was observed with CSS (Fig. 6). Used therapeutic agents did not significantly inhibit the cell cycle (results not shown).
43
Päättötyö4-toisinpäin.indd 43
13.3.2012 16.33
Fig. 6. Induction of apoptosis after 48 hr of incubation with therapeutic agents. DuCaP-N cells were grown in RPMI1640 supplemented with 10% CSS, 1 nM testosterone and therapeutics (bicalutamide 10 μM, TSA 10 nM, finasteride 10 μM) or in hormone depleted medium only (CSS). Error bars, SEM. **P≤0.01.
DuCaP and DuCaP-N Cells Contain a Subset of Cells Of Mouse Origin
Staining of the DNA with Hoechst 33342 revealed the presence of cells of mouse origin in DuCaP and DuCaP-N. Mouse cells can easily be distinguished from human cells by the speckled staining pattern of the nuclei [19]. Less than 1% of the cells in DuCaP (Fig. 7A) and approximately 10% of DuCaP-N cells showed speckled staining of the nucleus (Fig. 7B). Estimation of the amount of the mouse cells is difficult due to the clustering of DuCaP and DuCaP-N cells. By growing DuCaP-N cells in high calcium (1.8 mM) containing MEM-alpha medium supplemented with 10% CSS, we were able to select for cells of mouse origin (Fig. 7C). The cells were confirmed to contain mouse chromosomes by FISH (Fig. 7D). No human chromosomes 1, 2, or 3 were observed. The chromosomal count was 53.4±3.9 chromosomes per cell, whereas the normal chromosomal count in a metaphase cell in mice is 40. Many chromosomal aberrations, such as chromosome gaps, breaks, and fragments were observed.
44
Päättötyö4-toisinpäin.indd 44
13.3.2012 16.33
Chapter
3
Fig. 7. Cells of mouse origin in DuCaP and DuCaP-N cells. (A) Mouse cells can be distinguished by a simple Hoechst 33342 staining from human cells. Speckled nuclei reveal mouse cells in DuCaP and in (B) DuCaP-N. White arrows indicate the speckled mouse nuclei. (C) Cells of mouse origin were selected by culturing DuCaP-N cells in MEM-alpha supplemented with 10% CSS. All cells have speckled nuclei. (D) FISH analysis with Cy3 labeled mouse whole chromosome 8 probe (arrow heads) confirms the cells with speckled nuclei to be of mouse origin. Figures show representative results. Scale, 0.2mm.
Discussion In this study, we describe the development and validation of a novel cell line, DuCaP-N, that can serve as a good model for preclinical testing of drug combinations for the treatment of hormone therapy naive metastatic prostate cancer. When the hormone sensitive parental DuCaP cell line is cultured with additional 1 nM R1881, apoptosis and cell cycle inhibition is triggered. However, some cells survive and proliferate in the excess androgen environment. After a culture period of 20 passages DuCaP-N is generated. The most striking difference between the DuCaP and DuCaP-N cell lines is the over-expression of AR in low androgen environment (DuCaP). AR is over-expressed in most CRPCs, among which 10–20% exhibit amplification of the AR gene [20]. In regard to AR expression, DuCaP-N resembles the primary prostate cancer, and DuCaP the castration resistant form. Androgen concentrations in medium with 10% FCS appear to be extremely low [21]. We also determined the total testosterone and dihydrotestosterone levels in this culture medium after ether extraction followed by chromatography and detection with radioimmunoassay. The amount of tested androgens was below detectable levels (testosterone