Patient-derived prostate cancer organoids require α6-integrins for self-renewal and recapitulate genetic heterogeneity and histopathology of the original tumors

doi:10.21203/rs.3.rs-3209780/v1

Download PDF

Research Article

Patient-derived prostate cancer organoids require α6-integrins for self-renewal and recapitulate genetic heterogeneity and histopathology of the original tumors

https://doi.org/10.21203/rs.3.rs-3209780/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Modeling the pathophysiology of prostate cancer (PCa) remains a challenge and primary tissue-derived organoid cultures have emerged as promising models representing good tissue mimicry.

Methods: We have established 36 PCa patient-derived organoid cultures that self-renew and can be maintained long-term. Using CRISPR-mediated genomic editing, selected PCa organoids were also genetically engineered to modulate their tumorigenic and self-renewal properties. We show that PCa organoids can also be readily utilized in drug sensitivity tests. In vivo, orthotopic xenograft studies were performed to recapitulate the original tumor phenotype. Finally, we performed scRNA and bulk RNA sequencing to show the multi-cellular involvement in PCa organoid self-renewal and tumorigenesis.

Results: The morphological properties of cancer organoids were found to correlate with the disease grading. Importantly, detailed analyses of selected organoid lines indicated that they maintain the heterogeneity, gene expression profile, and histological architecture of their primary tumor-of-origin and recapitulate these features in vivo as xenografts. Finally, we show that α6-integrin expression is a requisite for the growth and long-term propagation of human PCa organoids while α2-integrin expression is dispensable. High expression levels of both α6- and α2-integrins have been documented in putative prostate epithelial stem cells which likely play an important role in PCa tumorigenesis. Our single-cell analysis identified a distinct cell population with a stem cell profile and high expression of α6- and α2-integrins.

Discussion: Our data suggest that α6-integrin expression promotes PCa stem cell self-renewal and demonstrates the versatile biomedical utilities of PCa-derived organoid models.

Prostate cancer

organoid

integrin

self-renewal

cancer stem cells

Prostate cancer (PCa) is the second most common malignancy in men worldwide[1, 2]. While the majority of PCas are slowly progressing, PCa can also be highly aggressive, and once it has metastasized, typically to the liver, bone, or lungs, there are limited treatment options[3]. Despite progress in targeted molecular therapies, PCa inevitably develops resistance to standard therapeutics posing a major challenge for cancer care[4]. PCas are mostly classified by their histological appearance and graded with Gleason scores and tumor stage at the time of diagnosis[5, 6]. The majority of PCas are classified as acinar prostate adenocarcinoma but rare forms, such as ductal adenocarcinoma, neuroendocrine tumors, and squamous cell carcinoma are also found and tend to be more aggressive[7]. The underlying genetic and transcriptional landscapes of PCa are complex calling for specifically tailored treatment strategies[8]. PCa tumorigenesis is driven by several highly prevalent oncogenic aberrations, including inactivation of PTEN and fusion of androgen-driven TMPRSS2 gene with erythroblast transformation specific (ETS) family transcription factor genes, most often ERG[9–11]. These cases together account for more than 50% of all the PCa cases[12, 13]. These two leading genomic alterations correlate with an increased probability to the development of metastatic castration-resistant PCa (mCRPC) with a poor prognosis[14–16]. PTEN deletion and TMPRSS2-ERG-fusion are accompanied by additional genetic mutations, such as PCa risk-associated single nucleotide variations, and so far incompletely defined microenvironmental factors that contribute to the altered epigenetic regulation of PCa cell genomes[17–19].

Investigations of these critical events and inherent tumor heterogeneity have long been hampered by a lack of relevant human models wherein the early tumorigenesis and initiation of cancer stem cell pool could be characterized in detail. Recently, primary cell-derived organoid technology is gaining prominence as the method of choice as a model for the development of personalized cancer medicine[20, 21]. Given the heterogenous nature of patient-derived organoids, a key challenge will be to generate and characterize PCa organoid biobanks representing specific types of PCa and to develop methods to engineer defined genetic models to address specific scientific questions. Here we report the generation of a panel of PCa organoid cultures derived from 36 PCa patients. For each of the original tumors we have determined the Gleason score, tumor stage and androgen receptor expression (AR) status at the time of prostatectomy. A detailed analysis of selected organoid lines indicated that PCa-derived organoids recapitulate the gene expression profile, genomic landscape and histological architecture of the parental tumors in vitro and in vivo, even after long-term expansion in vitro. In addition, we demonstrate the utility of genetically modified PCa-derived organoids for molecular studies by engineering organoid variants lacking PTEN expression to adjust their tumorigenic potential in vitro. Finally, we show that α6-integrin expression is critical for the maintenance of self-renewing PCa organoid cultures.

Long term PCa organoids recapitulate the original tumor

A functional PCa biobank was established by collecting prostate tumor tissue pieces from prostatectomy samples (∼1 cm³ tissue) with informed written consent from the patients. We received 4-5 biopsies (Table S1) from each prostate and whenever possible, pieces were cut from both cancer and benign lesions, based on pathologist’s evaluation. Each piece was further divided into four parts that were separately processed for organoid derivation, histological analysis, genomic analysis, and patient-derived xenograft (PDX) modeling. Pathologist performed a Gleason scoring analysis for all the 36 histology samples (Table S1). PCa organoid cultures were established as described in experimental procedures and passaged every 14-20 days with a consistent passaging ratio of 1:3 either by gentle pipetting or trypsinization after dissolving the Matrigel^TM blobs. After a couple of passages, aliquots of organoid lines were stored in liquid nitrogen from where they could be thawn for further expansion and assays.

Initially, ~5mm³ of fresh PCa tissue was processed into single-cell suspension (see methods) and seeded into four Matrigel^TM blobs and grown for a minimum of 7 days as organoid cultures and imaged using phase contrast microcopy. The organoids generated from all patients posed heterogeneous morphologies, ranging from irregularly shaped solid cell clusters to spherical structures with a hollow lumen surrounded by columnar epithelial cells. Organoid forming capacity was measured in terms of number of organoids formed per each sample and individual organoids were classified into two main morphological groups, hollow organoids with clearly visible central lumen and solid organoids containing no or only small lumina. The total number of organoids in the different samples varied between 150 and 500 and in most of the samples there was slightly more solid than hollow organoids (Figure 1A). However, in some samples (e.g. PT-5, -6, -16, -20, -21, -23, -24, -29 and -34) more than two thirds of the organoids were solid cell clusters. Interestingly, when patient samples were stratified into two groups based on low (< 7) and high (> 7) post-operative Gleason scoring, it was observed that while hollow organoid numbers were similar in both groups, the number of solid organoids was significantly higher in the high Gleason group (Figure 1B). Upon prolonged culture also the initially hollow organoids started to develop multiple lumina as the clusters grew to form large glandular rosettes (Figure 1C). The size of the organoids continued to increase (Figure 1D, E) while the number of organoids declined during the 2-month culture period (Figure 1D, F). In some cases, highly proliferative prostate fibroblasts appeared to outcompete epithelial organoids in the culture and migrated to the bottom of the plate. In such cases, the rapidly expanding fibroblasts exhausted the nutrients leading to acidification of the culture medium. Therefore, to ensure the epithelial status of organoid cultures, epithelial cells were enriched by FACS-sorting based on the expression of an epithelial cell adhesion molecule (EpCAM, Figure 1G). The EpCAM-negative cells contained stromal cell populations of the prostate[22]. EpCAM-positive epithelial cells grown in Matrigel readily formed organoids and could be propagated in culture on long-term basis (Figure 1H). In contrast, prostate stromal cells did not form organoids but displayed an elongated fibroblastoid morphology and eventually ceased proliferation upon prolonged culture (Figure 1H).

PCa organoids recapitulates the morphological and histological features of their parental tissue

To correlate the organoid culture phenotypes with the original tumor tissue, we selected three organoid lines each representing a PCa sample classified into different grade and Gleason score category (O-12: G3a, GS3+4; O-11: G3a, GS4+4; O-9: G3b, GS3+4). The organoids were grown in Matrigel for approximately 21 days, fixed and stained for ERG, PTEN, cytokeratin 8 (CK8) and CK5 and imaged by immunofluorescence (Figure 2A). For AR-staining, organoids were processed for immunohistochemistry as the AR-antibody did not work well for immunofluorescence (Figure 2A). Immunohistochemistry was performed on the matching original tissues using the same marker proteins. The morphological features of the original patient tissue and its matching organoids showed similarity in biomarker expression and in their histological features. For example, PTEN loss is a common scenario in 40% of the PCa patients and was seen in one (PT-11) of the three patient tumors and replicated in O-11 organoid sample. Another common event is overexpression of ERG, often due to fusion with AR-responsive TMPRSS2 gene[23]. Normal prostate epithelium expresses negligible levels of ERG. ERG overexpression was observed in two of the three selected patient tumors and again was replicated in the respective organoid samples (O-9 and O-11). Incidentally, both PT-9 and PT-11 patients carried higher grade tumors with high Gleason score (Table S1). O-12, that derived from a tumor with a lower Gleason score expressed normal PTEN levels and low ERG levels in both in the original tumor lesions and in organoid cultures (Figure 2A). Disappearance of CK5-expressing basal cells is one of the reoccurring features in developing PCa lesions. In line with this, we observed depletion of CK5 positive cells in the PT-11a and corresponding O-11a representing high Gleason PCa (Figure 2A). CK5 positive cells were typically observed at the periphery of cell clusters, but in some cases, CK5 positive cells were also inside the organoids (Figure 2A). Reportedly, basal epithelial cell population is important for the growth and propagation of organoids[24]. Organoids derived from high GS (O-5, GS:4+5) and low GS (O-12, GS:3+4) showed distinct morphological features. Many low GS (O-12) organoids showed normal prostate gland-like polarized epithelial cells surrounding a lumen, whereas higher GS (O-5) organoids appeared to form more solid, tumor-like cysts (Figure 2B). High CK5 expression was observed at the outer surfaces of normal or prostatic intraepithelial neoplasia (PIN) organoids (O-12; Figure 2C). In contrast, O-5 organoids derived from a high Gleason score PCa sample PT-5, showed a more irregular and overall reduced CK5 staining.

To study the morphology and growth of PCa organoids in an in vivo setting, we performed subcutaneous transplantation of long term expanded benign hyperplastic (BH; GS3+4) O-12 and high Gleason (GS4+5) O-5 organoids in immunocompromised SCID mice. OX-12 organoid xenografts formed well-organized glandular structures with distinct layers of basal and luminal epithelium as indicated by CK5 and CK8 staining, respectively (Figure 2E). Curiously, while OX-5 organoid xenografts formed gland-like structures they displayed frequent unpolarized tumor outgrowths (Figure 2E). Overall, OX-5 organoid transplants exhibited faster growth than OX-12 transplants (728 mm³+/-386 mm³ vs. 474 mm³ +/-142 mm³at 12 weeks, respectively) (Figure 2D). Hence, these data demonstrate that the long-term expansion of PCa organoids in vitro and in vivo preserves the growth and structural features of their tumor-of-origin. Collectively, this analysis revealed that PCa organoids express similar biological markers with their corresponding patient-derived tissue of origin, even after long-term expansion in vitro.

In vivo models are much relied in cancer drug discovery. The subcutaneous xenograft models based on established PCa cell lines are widely used in the development of anti-cancer drugs, but currently the success rate of this approach stands at ~5% to find drugs that undergo further clinical approval[25] PCa cell lines are typically adapted to in vitro culture conditions on 2D plastic surfaces over several decades and this might limit their translational relevance. Our data indicates that PDO models established from patient tumors have high resemblance to their tumor-of-origin and could thus serve as accurate models for drug development and precision medicine. To determine whether our organoid models would faithfully recapitulate the PCa phenotype, we placed selected PDOs (O- 5, -11, -12) surgically into the prostate of ICR-SCID mice. The mice were in constant observation until surgery recovery and twice weekly thereafter. Since PCa is a slow progressing cancer, the mice were observed for 90 days after which their prostates were examined in detail. This analysis revealed a remarkable resemblance between the orthotopic organoid xenografts (OOX) and their parental tumor tissue accurately reiterating the morphological traits (Figure 3A). When evaluating PCa biomarkers such as AR, PTEN, CK5 and CK8, we found that the OOX-tumors’ expression profiles resembled the corresponding original tumor tissues (Figure 3B). Importantly, O-11 and OX-11 both displayed PTEN loss seen in the original tumor, but only the OOX-11, transplanted to into mouse prostate recapitulated the histological traits of the parental tumor with cribriform and fused glands along with PTEN loss (Figure 2A, 3A-B). Similarly, OOX-12 transplants derived from low Gleason score adenocarcinoma, showed PTEN and AR expression and formed hyperproliferative glands resembling adenocarcinoma lesions in the matching original tumor. PTEN was not expressed in O-11 organoids or OOX-11 transplants, in agreement with the corresponding original tumor tissue in the respective patients (Table S2, Figures 2A and 3B). Overall, these results demonstrate that PCa organoids recapitulated the histological characteristics and differential marker expression of the original tumor tissues even after long-term culture of the organoids in vitro.

PCa organoids preserves molecular heterogeneity as in its original patient tumors.

Intratumor heterogeneity is a prominent feature of human cancers. Next, we wanted to analyze the genetic properties of the PCa organoids in more detail. The PCa-derived organoids with different Gleason scores were expanded long-term (~6 months) in culture, with a consistent passaging ratio of 1:3-1:4. The growth rate of the different organoid lines was found to vary with passage frequencies between 1-3 weeks (Figure 4A). In some cases, the reduced organoid growth rate was accompanied with an outgrowth of fibroblast-like cells (O-1, -16 and -32). To alleviate this issue and to focus on the PCa cancer cell population, few organoid lines (O-5, -9, -11, -12, -16, -24, -32 and -36) were FACS-sorted to enrich epithelial populations and subsequent genomic and other downstream analysis was performed with sorted epithelial populations of selected organoid lines (Figure 1G, H).

To determine whether the PCa organoid lines retain the original patient tumors’ mutational landscape, we performed whole exome sequencing (WES) analysis of six organoid lines (O-5a, O-5e, O-9, O-11a, O-11c and O-12) derived from four patients (PT-5, PT-9, PT-11 and PT-12). In addition to tumor organoids, matched organoid lines derived from phenotypically normal area of O-5 (5e) and O-11 (11c) were included in the analysis and cultured for more than 2 months. Organoids were sorted for CD49f^hi and EpCAM^low population (basal epithelial cells enriched) and used as an internal control to compare with matching tumor lesions (O5a and O-11a). We analyzed missense variants, stop-gain variants, frameshift and inframe_indels mutations. The global variant profiles of organoid lines and their corresponding original tumor tissue were highly similar with ~95% of the mutant variants retained in almost all the samples (Figure 4B). Though we unfortunately could not get access to the patient’s germline mutational landscape, we filtered for variants present in the Catalogue of Somatic Mutations in Cancer (COSMIC) and found that the majority of all the cancer-related somatic variants present in the patient’s original tissue were retained in the corresponding PCa organoid lines (Figure 4C). We examined the mutational landscape for known driver gene mutations such as PTEN, AR, ERG, TMPRSS2, PIK3CA, SPOP and TP53 that are common in PCa and found several such mutations in our sample set (Figure 4C). Curiously, benign (PT-5e, PT-11c) and tumor (PT-5a, PT-11a) lesions and the corresponding organoids (O-5a,e and O-11a,c) all carried the same tolerated homogeneous or heterogeneous mutations in the PI3KR1 and FOXA1 genes. These types of mutations are generally considered to be associated with low risk whereas deleterious heterogeneous mutations, such as TMPRSS2, is associated with high risk of developing PCa. The observed TMPRSS2 mutations in patients were similarly found in both benign and tumor lesions as well as corresponding organoids (Figure 4C). The PT-9/O-9 pair exhibited deleterious mutation in AR and BRCA1 genes as well as a tolerated heterogeneous mutation in PIK3CA gene. Given that the mutational analysis was consistent in the sorted basal epithelial population suggest that basal stem cell-like population represents the original patient tumor tissue in terms of genetic information. However, we did observe a loss of some key mutations in organoid cultures that were present in the primary tissue. Deleterious heterogeneous mutations of SPOP detected in both PT-11a and PT-11c and PIK3CA detected in PT-11c could not be found in respective organoid cultures suggesting that PCa cells carrying these mutations were eliminated during the long-term in vitro culture.

Next, we studied how well the overall transcriptomic profiles of the primary tumor/organoid pairs were preserved. To this end, we performed genome-wide transcriptomic RNA-sequencing analysis in three selected pairs, PT/O-9, PT/O-11 and PT/O-12). Principal component analysis (PCA) indicated that the transcriptomic profiles of the organoid cultures were highly similar to their corresponding tumor samples (Figure 4D). Collectively, our genomic and transcriptomic profiling of the PCa organoids authenticate the preservation of the tumor heterogeneity including potential PCa driver mutations.

Prostate organoids show distinct drug sensitivity profiles.

Given the preservation of histological, biochemical, and genetic properties of original tumors in the in vitro organoid cultures we performed a proof-of-concept drug-sensitivity testing in four of the organoid lines (O-5, O-9, O-11 and O-12). This analysis was done to evaluate the capacity of organoid model as a platform to assess potential drug sensitivity and resistance in specific patient tumors. Moreover, when combined with functional drug sensitivity testing, the genomic characterization might give further insight into the potential mechanisms underlying drug resistance. As a primary screening step, the therapeutic response was evaluated by determining the total number and average size of organoids with different doses of enzalutamide or docetaxel, two drugs that are in regular clinical use for PCa treatment. Enzalutamide is a non-steroidal anti-androgenic drug that blocks testosterone binding to androgen receptor, and it is primarily used as androgen deprivation therapy (ADT) in the PCa patients. Docetaxel is a microtubule stabilizing cytotoxic drug typically used to treat metastatic PCa in patients with ADT-resistant tumors.

First, the organoids were grown for 3 weeks without drugs with bi-weekly medium changes followed by a 7-day treatment with a dilution series of either enzalutamide or docetaxel after which organoid number and sizes were determined. Drug dilutions were determined as reported in NCI-DTP portal[26] and by their half-maximal growth inhibitory concentration (GC₅₀) and sensitivity was represented by measuring the area and number of organoids. Except for the O-12 organoid line, the other organoid lines showed significant resistance to enzalutamide as organoid growth was observed even at the highest concentration of enzalutamide (Figure 5A, C and E). O-9 organoids expressed high levels of ERG (Figure 2A) and were shown to contain an AR-mutation that might contribute to enzalutamide resistance (Figure 4C). O-11 organoids were PTEN-negative and expressed high levels of AR both of which could contribute to ADT-resistance[27] (Figure 3B). O-11, as well as O-5 organoids were found to express high levels of ERG that has also been associated with ADT-resistance[28] (Figure 2A). The exome-sequencing analysis revealed several putative candidate mutations in O-5 or O-11 organoid lines that could contribute to their enzalutamide-resistance. Docetaxel-treatment was effective in enzalutamide-resistant organoid samples, except for O-11 that showed some resistance at intermediate concentrations (Figure 5B, D and F). Docetaxel-resistance mechanisms are not well understood but upregulation of multidrug transporters has been suggested as the primary mechanism[29]. Of note, mutations in the DNA-repair machinery, associated with genomic instability in PCa[30], were observed in O-9 organoids (Figure 4C). Taken together, these data demonstrate the potential of the in vitro PCa organoid model as a drug screening platform that can be utilized to study patient specific drug responses.

PTEN depletion induces the growth in PCa organoids

A key property of any model system for cancer cell biological studies is their amenability for robust genetic manipulation. Loss of PTEN is one of the key genomic alterations in PCa that contributes to PCa growth and aggressiveness[9]. We selected three PTEN-positive organoid lines (O-16, O-24 and O-36) and generated isogenic PTEN-deleted variants by using previously described lentiviral constructs expressing Cas9 and PTEN-targeting gRNA[31]. Control population was transduced with the same Cas9-expressing lentiviral vector lacking a gRNA sequence. Positively transduced cells were selected using puromycin selection and 10 days post selection puromycin-resistant organoids were harvested and seeded into Matrigel blobs and grown without puromycin. Loss of PTEN expression was confirmed by immunofluorescence staining of fixed organoids (Figure 6D). To investigate whether loss of PTEN had any effect on organoid growth the different organoid variants were grown for 3 weeks followed by counting the number and overall area of the organoids in culture. While the number of organoids formed by PTEN-KO variants was not significantly increased, PTEN-KO organoids were much bigger than their respective controls (Figure 6B-C). Importantly, while most control vector transduced O-36 organoids had hollow morphology and showed distinct basal and luminal cell populations, O-36 PTEN-KO variants grew as large solid clusters with mixed basal and luminal identity based on CK5/CK8-expression (Figure 6D). Most of the control O-16 and O-24 organoids were already solid clusters but their size increased significantly upon PTEN depletion (Figure 6A).

Integrin α6 regulates PCa organoid growth.

Epithelial cells reside on a laminin-rich basement membrane (BM) that is rich of signals regulating epithelial differentiation and polarity[32]. The integrin family of ECM receptors are the primary cell surface receptors that connect prostate epithelial cells to the BM. In normal prostate epithelium, integrin α6 (CD49f) pairs with β4 to form hemidesmosomes that anchor cells to the BM by binding to laminin. Multiple studies have reported that CD49f^high population in the prostate, as well as many other epithelial organs such as the breast, contains cells with robust organoid formation and tumor initiation capacity[33–35]. High α6-integrin expression is thus considered to mark prostate stem cells and possibly cancer stem cell pool[36]. Counterintuitively, α6β4-integrins are downregulated upon PCa progression[31] However, loss of hemidesmosome organization is due to loss of β4-subunit expression in advanced PCa, while residual α6-subunit expression persists[31, 37]. α6-subunit can also pair with β1 to form integrin α6β1 that has been implicated in the regulation and maintenance of PCa stem cells and attributes to cell survival, resistance to chemotherapy and metastasis[36]. To investigate the role of α6-integrin in PCa organoid cultures, we dissociated organoids into single cells and sorted them by FACS into four different populations using Epithelial Cell Adhesion Molecule (EpCAM, epithelial cell specific marker) and integrin α6 (CD49f) (Figure 7A). CD49f^low/EpCAM^high cells represent luminal epithelial population (LP) and double-positive CD49f^high/EpCAM^high cells consist of basal epithelial cells (BP)[21]. EpCAM^low/CD49f^low cells presumably represent stromal cell populations (SP) whereas the identity of EpCAM^low/CD49f^high population is unclear. It is possible that EpCAM^low/CD49f^high represent a population enriched for cancer stem cell-like cells (PCSC) in the prostate as described for breast cancer organoids[38]. LP formed tiny and partially hollow organoids that could not be efficiently propagated long-term (Figure 7B). In contrast, both BP and PCSC populations efficiently formed organoids that showed mostly the solid phenotype, and which could be cultured on long-term basis (Figure 7B). BP organoid cultures yielded also luminal epithelial cells presumably via basal-to-luminal differentiation, as described previously[21, 39].

In addition to integrin α6 (CD49f), integrin α2 (CD49b) is also expressed in prostate epithelial cells, and high integrin α2 levels have been suggested to depict PCa stem cells[36, 40, 41]. We analyzed the expression of α2- and α6-integrins in selected PCa organoids. Immunofluorescence experiments showed that α6-integrin expression is confined to basal surfaces in both hollow and solid organoids (Figure 7C). In benign O-24 organoids, these basally localized cells also expressed high levels of CK5, a basal epithelial cell marker (Figure 7D). In solid organoids, CK5 was not as strictly restricted to basal side as α6-integrins. α2-integrin expression was also seen in basal cells in benign organoids, but it was not clearly restricted to basal surfaces in solid O-24 organoids and was only gradually decreasing towards the center of cell clusters (Figure 7C).

To test whether high levels of α6-integrins, α2-integrins or both depict prostate stem cells that possess organoid formation capability, we sorted O-24 PCa organoids based on CD49f (α6-integrin) and CD49b (α2-integrin) expression (Figure 7E). The double negative α6^low/α2^low population seeded onto Matrigel blobs hardly grew at all and remained as individual cells and eventually lost viability (Figure 7F). α6^low/α2^high cells exhibited elongated fibroblastoid morphology (Figure 7F). Regardless of the α2-integrin expression levels, cell populations expressing high levels of α6-integrin efficiently formed organoids (Figure 7F). We observed that α6^hi/α2^hi population formed mostly solid organoids while α6^high/α2^low population formed both solid and hollow organoids (Figure 7F).

To get insight into the different cell populations within prostate organoids, we performed single cell RNA sequencing (scRNA-seq) on O-36 (GS: 3+4) organoid sample. Four samples from these parent organoids were prepared: The second passage of the original patient-derived O-36 organoid (sample-1), FACS-sorted EpCAM-positive epithelial cell population derived from O-36 (sample-2), EpCAM-negative stromal population derived from the O-36 organoid (sample-3) and reaggregated epithelial and stromal O-36 populations in a co-culture setup (sample-4). After two weeks in 3D Matrigel blob culture, single cell suspensions were prepared for scRNA-seq and approximately 10,000 cells per sample were loaded into a 10X Genomics single-cell platform and analyzed with Cell Ranger software. Altogether, 40595 cells were included into the analysis that annotated 18 different clusters (Figure 8A). Interestingly, it was observed that the epithelial clusters in single and co-culture setups were relatively similar whereas the stromal populations differed significantly when cultured alone or in the co-culture settings (Figure 8B, D). Moreover, similarity in transcriptomic profiles between the original organoids and the reconstituted coculture setup was noted (Figure 8D). Cell type identification in the different clusters was determined by examining differentially expressed genes (DEGs) indicating stromal (PECAM, FN1, ACTA2) epithelial (CDH1, EPCAM, CD24) and basal (TP63, KRT5, KRT14) epithelial origin (Figure 8B, C).

As discussed above, both high α6- and α2-integrin expressing cells have been reported to carry self-renewal potency. Such stem cell-like cell populations are considered particularly critical for PCa pathogenesis and thus they represent a key target for PCa therapies [36, 42]. To assess if the α6^high and α2^high cells represent distinct cell populations we focused on defining ITGA2- and ITGA6-positive populations in the single cell analysis dataset. ITGA6 was highly expressed in a single epithelial cluster 9 while high ITGA2 expression depicted two clusters (Figure 8C). Comparative analysis of double negative (DN), single positive ITGA6+ and ITGA2+ and double positive (DP) ITGA6+/ITGA2+ cells revealed similarity between ITGA6+ and DP populations (Figure 8E). Curiously, the DN population expressed high levels of prostate stem cell antigen (PSCA) associated with prostate cancer progression[43, 44]. ITGA2+ cells expressed mesenchymal genes such as FN1 and VIM but also some epithelial genes although they were clearly distinct from the ITGA6+ and DP populations (Figure 8E).

Integrin α6 is required for PCa organoid growth whereas integrin α2 is dispensable.

To confirm the functional role of α6- and α2-integrins in the self-renewal capacity of PCa organoid we again implemented lentivirus-mediated CRISPR/Cas9-KO technology (Figure 9A). The knockout experiments were conducted using two different organoid lines (O-16 and O-36). The control virus expressed no specific gRNA whereas α6KO construct expressed a gRNA construct targeting the third exon-3 of ITGA6 and α2KO construct a gRNA targeting exon 5 in the ITGA2 gene[31, 45]. Transduction of the organoids was performed as described above for PTEN-KO and immunofluorescence staining confirmed efficient depletion of α6- and α2-integrins (Figure 9A). It was observed that while the number of α6KO organoids was only modestly reduced, the organoids had a dramatic growth defect when observed after 3 weeks in culture (Figure 9B, C). We could not establish long-term cultures of α6KO organoids as they eventually ceased to grow and became extinct. Any organoid colonies recovered from these cultures were found to express α6-integrins (data not shown). This α6-integrin expression might be because of in-frame editing preserving the open reading frame of ITGA6 mRNA or recovery of α6-expression due to incomplete editing in some cells. In contrast, depletion of α2-integrin expression had no significant effect on the number, growth or survival of PCa organoids. α2KO organoids formed at similar numbers and grew at rates comparable to controls (Figure 9A-C). Finally, the proliferation rates of the different organoid variants were also determined using an independent method based on Ki67 staining. In agreement with the growth analysis, Ki67 staining assay showed that α6KO organoids contained very few proliferative cells (<10%), whereas more than 30% of the cells in α2KO and control organoids showed Ki67 positivity (Figure 9D). Such levels have been considered clinically relevant depicting fast-growing prostate tumors[46]. Collectively, these results indicate that α6-integrin expression is critical for organoid formation and growth and therefore α6-, and not α2-integrins, play a key role in prostate organoid self-renewal and potentially also in the maintenance of PCa stem cell pool.

In this study we generated and characterized a panel of human prostate organoid lines that could be maintained in culture for several months. The long-term cultured organoids maintained the genetic heterogeneity, transcriptomic profiles, histological tumor marker expression and morphological features of the tumor-of-origin. Analysis of our PCa organoid panel corroborates the feasibility of primary PCa organoid culture system for detailed molecular characterization and functional drug sensitivity screening. Drug resistance screening in vitro holds a great value for translational medicine[47] . Given the comparatively slow progression of PCa pathogenesis, in vitro drug resistance screening data using patient derived PCa organoids could be utilized in the clinic to plan the most efficient treatment for the patients. These properties make primary PCa organoids a prominent model for translational precision medicine applications. Recent combined genomic and proteomic analyses of PCa samples have revealed limited correlation between the transcriptome and the corresponding proteome. Curiously, for certain genes the risk for biochemical PCa relapse correlated with low mRNA levels and high protein levels[48, 49]. While the best prognostic value was obtained from combined genomic and proteomic analysis, it was noted that analysis of the proteome was superior in predicting PCa relapse when compared with transcriptomic analysis[49]. The correlation between the proteome and transcriptome or epigenome is complex and further functional studies are needed to assess whether more accurate proteomic correlations can be extracted by comprehensive analysis of genomic data. Proteomic analysis in clinical samples is challenging especially when determining temporal changes. Although we did not perform proteomic experiments in the current study, PCa organoid model could fill this need as most PCa samples could be readily expanded and grown under defined culture conditions in amounts that can be analyzed with current ultrasensitive mass-spectrometry technologies[50].

The observed high success rate of approximately 95% in establishing primary PCa organoids is in line with recent studies that reported efficient establishment of primary PCa organoids. Of which 12 organoid lines were established and passaged for at least 3 passages (Fig. 4A) but low viability of metastatic PCa-derived organoids[51, 52]. However, we did not observe rapid depletion of cancer cells. Also, we observed a significant number of stromal cells in the early passages of unsorted organoid samples. Whether the stromal cell population contributes to maintenance of PCa cells remains to be studied in more detail. Importantly, we demonstrated the feasibility of PCa organoids for genomic engineering thereby circumventing the cancer phenotype maintenance issue. Here, we generated PTEN-defective variants of PCa organoids with intact PTEN expression and showed that loss of PTEN promoted PCa organoid growth. Prostatectomy-derived PCa organoids are an excellent model to study the early events of PCa pathogenesis. Their amenability for genomic editing further strengthens the versatility of the PCa organoid model in detailed functional studies addressing the role of specific genetic factors.

Organoid culture maintenance is based on the ability of prostate epithelial stem cells to self-renew. PCa cells share many characteristics with stem cells, including the ability to self-renew[53]. Cancer stem cells have been identified as a distinct subset of cells within tumors that are responsible for driving tumor growth and recurrence. The presence of stem cells has been associated with poor patient prognosis in several types of cancer and targeting cancer stem cells has emerged as a promising therapeutic strategy[54, 55]. Significant PCa stem cell plasticity has been documented and different prostate epithelial cell types can serve as the PCa cell-of-origin[56, 57] . PCa stem cells pose a serious challenge to successful of PCa therapies and therefore it is important to dissect the molecular machineries promoting stemness and self-renewal of PCa cells. Stem cells reside in specific niches where the microenvironment including cell-ECM interactions help to maintain the stem cell status[58, 59]. High levels of α2- and α6-integrins from the integrin family of ECM receptors has been found in self-renewing prostate epithelial cells[53, 55] . Both integrin receptors can be used to enrich self-renewing stem cells from normal prostate[41, 42, 60–62] . Here, we found that only α6-integrin expression was required for efficient PCa organoid growth and propagation. This finding is in line with several reports where maintenance of α6-integrin expression was noted in PCa despite downregulation of one of its heterodimer partners, β4-integrin [31]. α6-integrins are highly expressed in basal epithelium where prostate stem cells are though to reside[57]. Whether PCa stem cells originating from luminal epithelium also depend on α6-integrins for self-renewal is an interesting question for future studies. In line with the current study, our recent data indicated that downregulation or collagen-binding α1- and α2-integrins correlates with the development of aggressive PCa[45]. However, although α2-integrins do not appear to be required for PCa cell self-renewal or development of early PCa lesions, it is still possible that re-expression of α2-integrins could contribute to metastatic properties of PCa cells[63].

Taken together, the patient derived PCa organoid model fulfills the criteria of a reliable in vitro PCa model, recapitulating the tumor heterogeneity and molecular properties ranging from genetic and transcriptomic traits to histological features. PCa organoids serve as a clinically relevant cellular models that can be studied in detail in vitro, thereby potentially reducing the use of animal models in PCa research. The organoid model is also amenable as a platform to drug testing. Propagation of PCa organoids from the α6-integrin positive stem cells and their culture with or without stromal cells of the prostate restores patient-specific material that can be stored in biobanks, thereby providing an important resource for precision medicine approaches.

Human specimens

Prostate tumour specimen (~1-2 cm³) were obtained from radical proctectomy performed at University of Oulu Hospital on patients who had PCa with approval of the Regional Ethics Committee of the Northern Ostrobothnia Hospital District (EETTMK:4/2015) and in compliance with the Declaration of Helsinki. All patients provided informed consent. All clinical data used in this project are anonymized at the source by the personnel of Oulu University Hospital or Biobank Borealis and researchers have no access to any sensible information nor any means of personal identification. Lymph nodes were also collected from the patients suspected to have nodal metastasis. Handling and processing of samples was performed according to qualified pathologists and according to guidelines. Samples were confirmed to be tumour or normal based on histopathological assessment. Tumor grades and Gleason scores diagnosis of each case was confirmed on routine hematoxylin and eosin-stained slides by an independent histopathologist (Table-S1). For each tumour specimen, samples were split into 4 parts and processed for histology, RNA and DNA isolation, PDX modeling or dissociated and processed for organoid culture.

Isolation and propagation of human prostate organoids

Part of the patient-derived ~5 mm³biopsies was minced and incubated at 37°C with 5ml of digestion solution containing 7.5 mg of Collagenase II and 100µg of Hyaluranidase). Tissue digestion was done overnight to get a good yield of cells, and the resulting cell suspension was filtered through a 70µm nylon cell strainer and spun for 5 minutes at 300-400 x g. The pellet was washed in cold advanced DMEM/F12 (GIBCO) and mixed with ice-cold basement membrane extract (BME, Matrigel^TM). Approximately 5-10 x 10³ cells were mixed with 50µl of BME and seeded as a blob per well in a 24-multi-well plate. After BME had solidified, the samples were cultured in prostate organoid isolation medium Advanced DMEM/F12 supplemented with 1% Primocin, 1% Glutamax, 1:50 B27 supplement, 1.25mM n-Acetyl-L-cysteine, 10% (vol/vol) Rspondin-1 and Noggin conditioned medium, 30% (vol/vol), 10mM nicotinamide, 50ng/ml recombinant human EGF, 100ng/ml recombinant human FGF10, 25ng/ml recombinant human HGF, 5μM A8301, 1nM of DHT and 10μM Y27632 as described in[64]. After organoid culture establishment, they were maintained for 2-3 weeks in culture (depending on the sample) and were visually inspected in regular intervals. Upon attainment of dense culture, the growing organoids were passaged by mechanical dissociation and transferred to fresh matrix as previously described. Prostate organoid medium (without Y27632) was changed twice a week and cultures were split upon attainment of dense culture. To prepare frozen stocks, organoid cultures were dissociated and mixed with freezing medium (90% FBS and 10% DMSO) and frozen following standard procedures. When required, the cultures were thawed using standard thawing procedures and cultured as described above. For the first 3-4 days (freshly formed organoids from PT) or first 2 weeks after thawing, the culture medium was supplemented with 10 µM Y-27632.

Histology and staining

Prostate tissues were fixed for 24 hours in 4% paraformaldehyde at room temperature, and then transferred to 70% ethanol followed by paraffin embedding as follows: briefly, tissues were processed through a graded ethanol series followed by xylene, and then embedded in paraffin, cut at 5μm, and stained (HE and immunohistological staining). For immunohistological staining, paraffin slides were deparaffinized and subjected to antigen retrieval using citrate sodium solution pH=6. Organoids were processed for cryosections, and immunofluorescence experiments were conducted on the organoids. Briefly, organoids were cryofixed with OCT compound (TissueTek) and cryosections were prepared using cryotome. Stainings were done as follows. The slides were washed briefly with PBS following formalin fixation. To reduce background nonspecific staining, slides were incubated with a 5% BSA in TBS solution for 1 hour. Endogenous peroxidase activity was blocked for 15 min in a 3% hydrogen peroxide/methanol buffer. Primary antibodies (listed in the Supplementary Dataset – Table S4) were then applied at appropriate dilutions for overnight at 4°C. Slides were then washed in TBS and incubated with a secondary antibody-HRP conjugate for 1 hour at room temperature and finally developed with 3,3′-diaminobenzidine (DAB) for 5 min, counterstained with haematoxylin, and mounted with DPX (Sigma). Slides were also stained in the absence of primary antibodies to evaluate nonspecific secondary antibody reactions. For immunofluorescence staining, organoids were processed as described in 21. Briefly, after primary antibody incubation, the sections were incubated with a secondary antibody conjugated to a fluorophore. Nuclei were stained with DAPI. Confocal images were captured on a Zeiss LSM inverted confocal microscope.

Organoid formation Assay

To assess the organoid formation efficiency in hollow vs solid organoids, pictures of entire blob of BME obtained per patient were photographed using a Leica microscope 2-3 weeks after isolation (depending on the sample) and all viable organoid structures were counted.

Drug sensitivity assay

For the drug sensitivity assays, organoids were dissociated into 2-5 cell clumps by enzymatic dissociation with TrypLE (Life Technologies). Then, cell viability assays were conducted by plating 500 clumps per well of a 96-well cell culture plate in 200μl of expansion medium. The following day a concentration dilution series of enzalutamide, docetaxel or vehicle (DMSO) control was added to the organoids. The concentration selected was based on the cell viability data from NCI-DTP portal, the results from the screening or the literature. Medium was changed 2 times a week for 3 weeks. Viable cells were assessed by their ability to generate organoid de novo. Representative pictures of the viability result were taken 2-3 weeks after starting the treatment. All cell viability experiments were conducted in triplicate in at least two independent experiments with different passages.

Mouse xenograft studies

Experiments involving mice were performed under permission from the National Animal Experiment Board following the regulations of the EU and national legislation (ESAVI/3706/2021). All mouse experiments are conducted in Oulu laboratory animal center, University of Oulu, strictly following the 3R principles. For Subcutaneous grafts, suspensions containing 1 million cells were prepared (normal and tumor derived organoid lines) in Advanced DMEM/F12 (GIBCO) containing BME and were injected into both flanks of male ICR-SCID mice (Taconic life sciences). Visible tumours developed in approximately 2–4 weeks and were followed until 3 months. Mice were culled when the tumours reached limit end-point (size or ulceration). For orthotopic graft, organoid suspensions were prepared in Advanced DMEM/F12 (GIBCO). These were grafted into the anterior prostates of 7 weeks old male, testosterone-supplemented ICR-SCID mice. Briefly, a transverse incision was made in the lower abdomen and the bladder, seminal vesicles and prostate were partially pulled out of the abdominal cavity to expose the anterior prostate. A needle puncture was made through the capsule of the anterior prostate between the two main ducts. The organoid suspension was injected into the pocket created under the prostatic capsule. The exteriorized organs were then returned into the body cavity, and the incisions in the body wall and skin closed using a running suture of 4-0 silk. After 12 weeks of growth, the host mice were euthanized for gross examination of prostate and lymph nodes. Both prostate and lymph nodes were examined for possible metastasis and bisected, and one-half subjected to histological examination and the other as frozen sections for further engraftment.

CRISPR-Cas9 based genetic manipulation of organoids experiments

To perform genetic modification in organoids, sgRNAs targeting loci of interest were cloned into plentiCRISPRv2 vector. A list of gRNAs and primers are presented in Supplementary Table S3. To generate frameshift mutations in genes of interest, organoids were transfected with plentiCRISPRv2 vector containing the locus-specific sgRNA. All the plasmids were verified by sequencing. Lentiviral particles were produced by co-transfecting plentiCRISPRv2 containing specific gRNA sequence, psPax2, and pVSV-G at the ratio of 4:3:1 (second-generation lentivirus packaging system) into human embryonic kidney 293T cells (ATCC, LGC Standards GmbH, Wesel, Germany; CRL-11268) using Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific). The transduced organoids were either selected with 4 μg/ml puromycin or 100 μg/ml hygromycin B for 10 days or 3–7 days, respectively. Subsequently, these organoids were grown for at least 7–10 days without selection antibiotics and further analyses were conducted by measuring the organoids or immunofluorescence staining.

RNA-sequencing (RNA-seq) and differential gene expression analysis

RNA library was sequenced as paired-end reads of 150 bp. Raw sequence reads were submitted for quality check by FastQC. Trimmomatic[65] was employed for quality control with settings “TruSeq3-PE-2.fa:2:30:10, SLIDINGWINDOW:5:20, MINLEN:15”. FastQC was applied again on cleaned reads to ensure quality. STAR[66] (v.2.7.2a) was used to map cleaned reads to the human genome assembly hg38 with default settings. HTSeq[67] was applied to quantitate mapped reads with parameters “-s no, -I gene_name”. DESeq2[68] was utilized to detect differentially expressed genes. Genes with low expressions were filtered out. Differentially expressed genes were identified by cutoff FDR <0.1.

Single-Cell Isolation and Sequencing

For the scRNA-Seq, four samples were derived from the O-36 (GS: 3+4) organoid: original patient-derived O-36 organoid, EpCAM-positive epithelial cell population, EpCAM-negative stromal population, and reaggregated epithelial and stromal O-36 populations in a coculture setup. The organoid samples were collected from the Matrigel blobs after two weeks of growth, incubated with cold media and washed three times with cold DPBS to remove the remaining Matrigel. Single-cell suspensions were achieved by incubating the organoids in trypsin-ETDA solution at 37°C. The cells were pelleted and re-suspended in PBS with 1% BSA (Sigma-Aldrich, St. Louis, MO).

Encapsulation of the cells and scRNA-seq library preparations were performed using a Chromium Next GEM Single Cell 5' Kit v2 with 10X Chromium system (10X Genomics), following the manufacturer’s protocol and starting with 6000–12 000 cells per sample. The ready gene expression libraries were paired-end sequenced as one pool (1.5 pM) by using NextSeq550 system (Illumina) with Mid Output kit v2.5 (150 Cycles). The following sequencing parameters were used as recommended in the protocol of Single Cell 5' kit: R1: 26, R2: 90, i7 index: 10, and i5 index: 10 cycles, respectively.

The sequencing data was processed using Cell Ranger (version 7.0.0, 10X Genomics) with cellranger count/aggr pipelines to map the fastq files to the GRCh38 reference genome, to build feature-barcoding matrices, and to generate cell clusters. The data was further filtered using Loupe Browser (10X Genomics) with the following parameters: only cells with 1) UMIs 250–7000, 2) expressed genes per cell 150–3000 and 3) mitochondrial UMIs: 0–20% were included in the analysis. The resulting 38 921 cells were re-clustered and analyzed with Loupe Browser to investigate the cluster/sample-specific gene expression levels and particular cell types, and to create gene expression heatmaps and violin plots of the relevant genes in clusters.

Statistical Analyses

All summary data are presented as mean ± SD or representative images of at least 2 independent experiments. All statistical analyses were performed in R and GraphPad Prism software (GraphPad 7.0). Sample size (n) values used for statistical analyses are provided in the relevant figures and supplementary figures. Tests for differences between two groups were performed using Mann-Whitney’s two-tailed test, Student's two-tailed unpaired t-test or log-rank test as specified in the figure legends. When using t-test we assumed normality and equal distribution of variance between the different groups. No data points were excluded from the statistical analyses. Significance was set at p-value ≤ 0.05.

ADT Androgen deprivation therapy

AR Androgen Receptors

BP Basal Population

CD Cluster of Differentiation

CK Cytokeratin

COSMIC Catalogue of Somatic Mutations in Cancer

CRISPR Clustered Regularly Interspaced Short Palindromic Repeats

DEGs Differentially expressed genes

DHT Dihydro Testosterone

DP Double positive population

ECM Extra Cellular Matrix

EpCAM Epithelial Cell adhesion molecules

ERG ETS‑related gene

FACS Fluorescence Activated Cell Sorting

G Grade

GS Gleason Score

LP Luminal Population

mCRPC Metastatic castration resistant PCa

O Organoid

OOX Orthotopic organoid xenografts

OX Organoid Xenograft

PCa Prostate Cancer

PCA Principal component analysis

PCSC Prostate cancer stem cell-like cells

PDO Patient derived organoid

PDX Patient-derived xenograft

PT Primary tissue

PTEN Phosphatase and tensin homolog

RNA Ribonucleic acid

SCID Severe combined immuno deficient

SP Single positive population

SN Single negative population

TMPRSS Transmembrane protease, serine 2

t-SNE t-distributed stochastic neighbor embedding

UMAP Uniform Manifold Approximation and Projection

WES Whole exome sequencing

Ethics approval and informed consent to participate - The Oulu PCa cohort samples derived from radical prostatectomy patients were obtained under written informed consent with approval of the Regional Ethics Committee of the Northern Ostrobothnia Hospital District (EETTMK 4/2015) in compliance with the Declaration of Helsinki. Mouse cancer models and protocols are approved by the National Animal Experiment Board (ESAVI/3901/2021). In all animal work, the principle of 3R (reduction, refinement, replacement) is respected.

Consent for publication - We, hereby give our consent for the publication of identifiable details, which can include photograph(s) and/or videos and/or case history and/or details within the text (“Material”) to be published in the Journal of Experimental and Clinical Cancer Research.

Data Availability - The datasets generated during and analyzed during the current study will be made available in the European Nucleotide Archive (ENA) upon publication of the study.

Competing interests - The authors have no relevant financial or non-financial interests to disclose.

Funding - The research was funded by the Academy of Finland (#311934 to AM), the Jane and Aatos Erkko Foundation (#190046 to AM) and Cancer Foundation Finland (#61-6158 to AM).

Author contributions - RD conceived, designed and performed all the experiments, analyzed and interpreted results. AH and SK performed experiments, AT and TM performed single cell RNA seq analysis, QZ performed bulk RNAseq analysis, AA, JN and MV provided patient materials and performed histopathology diagnosis. AM, VI, KP and GHW provided insights to the project, AM interpreted the results, conceptualized, and funded the project. AM and RD wrote the manuscript. All authors commented on the manuscript.

Acknowledgments - We thank Riitta Jokela for her excellent technical assistance; Dr. Virpi Glumoff for support in FACS analyses; Jaana Träskelin and Mari Sujala at Biocenter Oulu Virus Core Laboratory and Dr. Veli-Pekka Ronkainen for support in microscopy at Biocenter Oulu - Tissue Imaging Center and Biocenter Finland; and Oulu Laboratory Animal Centre and Biocenter Transgenic and Tissue Phenotyping Core Facility for the facilities, services and assistance in mouse work; and Biocenter Oulu Sequencing Center and Leena Keskitalo, and Mari Sujala for the assistance in scRNA-Seq and WES.

Sung H, Ferlay J, Siegel RL, et al (2021) Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71:209–249. https://doi.org/10.3322/caac.21660
Ferlay J, Colombet M, Soerjomataram I, et al (2021) Cancer statistics for the year 2020: An overview. Int J Cancer. https://doi.org/10.1002/ijc.33588
Sartor O, de Bono JS (2018) Metastatic Prostate Cancer. N Engl J Med 378:645–657. https://doi.org/10.1056/NEJMra1701695
Watson PA, Arora VK, Sawyers CL (2015) Emerging mechanisms of resistance to androgen receptor inhibitors in prostate cancer. Nat Rev Cancer 15:701–711. https://doi.org/10.1038/nrc4016
Mottet N, van den Bergh RCN, Briers E, et al (2021) EAU-EANM-ESTRO-ESUR-SIOG Guidelines on Prostate Cancer-2020 Update. Part 1: Screening, Diagnosis, and Local Treatment with Curative Intent. Eur Urol 79:243–262. https://doi.org/10.1016/j.eururo.2020.09.042
Epstein JI, Egevad L, Amin MB, et al (2016) 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. Am J Surg Pathol 40:244. https://doi.org/10.1097/PAS.0000000000000530
Schrecengost R, Knudsen KE (2013) Molecular pathogenesis and progression of prostate cancer. Semin Oncol 40:244–258. https://doi.org/10.1053/j.seminoncol.2013.04.001
Grasso CS, Cani AK, Hovelson DH, et al (2015) Integrative molecular profiling of routine clinical prostate cancer specimens. Ann Oncol Off J Eur Soc Med Oncol 26:1110–1118. https://doi.org/10.1093/annonc/mdv134
Jamaspishvili T, Berman DM, Ross AE, et al (2018) Clinical implications of PTEN loss in prostate cancer. Nat Rev Urol 15:222–234. https://doi.org/10.1038/nrurol.2018.9
Tomlins SA, Rhodes DR, Perner S, et al (2005) Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 310:644–648. https://doi.org/10.1126/science.1117679
Mitelman F, Johansson B, Mertens F (2007) The impact of translocations and gene fusions on cancer causation. Nat Rev Cancer 7:233–245. https://doi.org/10.1038/nrc2091
Kumar A, Kasikci Y, Badredine A, et al (2021) Patient-matched analysis identifies deregulated networks in prostate cancer to guide personalized therapeutic intervention. Am J Cancer Res 11:5299–5318
Boyd LK, Mao X, Lu Y-J (2012) The complexity of prostate cancer: genomic alterations and heterogeneity. Nat Rev Urol 9:652–664. https://doi.org/10.1038/nrurol.2012.185
Carver BS (2014) Strategies for targeting the androgen receptor axis in prostate cancer. Drug Discov Today 19:1493–1497. https://doi.org/10.1016/j.drudis.2014.07.008
Carver BS, Tran J, Gopalan A, et al (2009) Aberrant ERG expression cooperates with loss of PTEN to promote cancer progression in the prostate. Nat Genet 41:619–624. https://doi.org/10.1038/ng.370
Tan W, Zheng T, Wang A, et al (2022) Dynamic changes in gene alterations during chemotherapy in metastatic castrate resistant prostate cancer. Sci Rep 12:4672. https://doi.org/10.1038/s41598-022-08520-6
Krohn A, Diedler T, Burkhardt L, et al (2012) Genomic Deletion of PTEN Is Associated with Tumor Progression and Early PSA Recurrence in ERG Fusion-Positive and Fusion-Negative Prostate Cancer. Am J Pathol 181:401–412. https://doi.org/10.1016/j.ajpath.2012.04.026
Yoshimoto M, Joshua AM, Cunha IW, et al (2008) Absence of TMPRSS2:ERG fusions and PTEN losses in prostate cancer is associated with a favorable outcome. Mod Pathol 21:1451–1460. https://doi.org/10.1038/modpathol.2008.96
Giannareas N, Zhang Q, Yang X, et al (2022) Extensive germline-somatic interplay contributes to prostate cancer progression through HNF1B co-option of TMPRSS2-ERG. Nat Commun 13:7320. https://doi.org/10.1038/s41467-022-34994-z
Gao D, Vela I, Sboner A, et al (2014) Organoid cultures derived from patients with advanced prostate cancer. Cell 159:176–187. https://doi.org/10.1016/j.cell.2014.08.016
Karthaus WR, Iaquinta PJ, Drost J, et al (2014) Identification of multipotent luminal progenitor cells in human prostate organoid cultures. Cell 159:163–175. https://doi.org/10.1016/j.cell.2014.08.017
Henry GH, Malewska A, Joseph DB, et al (2018) A Cellular Anatomy of the Normal Adult Human Prostate and Prostatic Urethra. Cell Rep 25:3530-3542.e5. https://doi.org/10.1016/j.celrep.2018.11.086
Cerveira N, Ribeiro FR, Peixoto A, et al (2006) TMPRSS2-ERG gene fusion causing ERG overexpression precedes chromosome copy number changes in prostate carcinomas and paired HGPIN lesions. Neoplasia N Y N 8:826–832. https://doi.org/10.1593/neo.06427
Höfner T, Eisen C, Klein C, et al (2015) Defined Conditions for the Isolation and Expansion of Basal Prostate Progenitor Cells of Mouse and Human Origin. Stem Cell Rep 4:503–518. https://doi.org/10.1016/j.stemcr.2015.01.015
Day C-P, Merlino G, Van Dyke T (2015) Preclinical mouse cancer models: a maze of opportunities and challenges. Cell 163:39–53. https://doi.org/10.1016/j.cell.2015.08.068
| Databases & Tools | Developmental Therapeutics Program (DTP). https://dtp.cancer.gov/dtpstandard/dwindex/index.jsp. Accessed 27 Apr 2023
Yang Y, Bai Y, He Y, et al (2018) PTEN loss promotes intratumoral androgen synthesis and tumor microenvironment remodeling via aberrant activation of RUNX2 in castration-resistant prostate cancer. Clin Cancer Res Off J Am Assoc Cancer Res 24:834–846. https://doi.org/10.1158/1078-0432.CCR-17-2006
Bernasocchi T, El Tekle G, Bolis M, et al (2021) Dual functions of SPOP and ERG dictate androgen therapy responses in prostate cancer. Nat Commun 12:734. https://doi.org/10.1038/s41467-020-20820-x
Hwang C (2012) Overcoming docetaxel resistance in prostate cancer: a perspective review. Ther Adv Med Oncol 4:329–340. https://doi.org/10.1177/1758834012449685
Zhang W, van Gent DC, Incrocci L, et al (2020) Role of the DNA damage response in prostate cancer formation, progression and treatment. Prostate Cancer Prostatic Dis 23:24–37. https://doi.org/10.1038/s41391-019-0153-2
Wenta T, Schmidt A, Zhang Q, et al (2022) Disassembly of α6β4-mediated hemidesmosomal adhesions promotes tumorigenesis in PTEN-negative prostate cancer by targeting plectin to focal adhesions. Oncogene 41:3804–3820. https://doi.org/10.1038/s41388-022-02389-5
Matlin KS, Myllymäki S-M, Manninen A (2017) Laminins in Epithelial Cell Polarization: Old Questions in Search of New Answers. Cold Spring Harb Perspect Biol 9:a027920. https://doi.org/10.1101/cshperspect.a027920
Guo C, Liu H, Zhang B-H, et al (2012) Epcam, CD44, and CD49f Distinguish Sphere-Forming Human Prostate Basal Cells from a Subpopulation with Predominant Tubule Initiation Capability. PLoS ONE 7:e34219. https://doi.org/10.1371/journal.pone.0034219
Mei W, Lin X, Kapoor A, et al (2019) The Contributions of Prostate Cancer Stem Cells in Prostate Cancer Initiation and Metastasis. Cancers 11:434. https://doi.org/10.3390/cancers11040434
Dekkers JF, Whittle JR, Vaillant F, et al (2019) Modeling Breast Cancer Using CRISPR-Cas9–Mediated Engineering of Human Breast Organoids. JNCI J Natl Cancer Inst 112:540–544. https://doi.org/10.1093/jnci/djz196
Harris KS, Kerr BA (2017) Prostate Cancer Stem Cell Markers Drive Progression, Therapeutic Resistance, and Bone Metastasis. Stem Cells Int 2017:8629234. https://doi.org/10.1155/2017/8629234
Davis TL, Cress AE, Dalkin BL, Nagle RB (2001) Unique Expression Pattern of the α6β4 Integrin and Laminin-5 in Human Prostate Carcinoma. The Prostate 46:240–248
Lim E, Vaillant F, Wu D, et al (2009) Aberrant luminal progenitors as the candidate target population for basal tumor development in BRCA1 mutation carriers. Nat Med 15:907–913. https://doi.org/10.1038/nm.2000
Strand DW, Goldstein AS (2015) The many ways to make a luminal cell and a prostate cancer cell. Endocr Relat Cancer 22:T187–T197. https://doi.org/10.1530/ERC-15-0195
Ojalill M, Parikainen M, Rappu P, et al (2018) Integrin α2β1 decelerates proliferation, but promotes survival and invasion of prostate cancer cells. Oncotarget 9:32435–32447. https://doi.org/10.18632/oncotarget.25945
Sottnik JL, Daignault-Newton S, Zhang X, et al (2013) Integrin alpha2beta1 (α2β1) Promotes Prostate Cancer Skeletal Metastasis. Clin Exp Metastasis 30:569–578. https://doi.org/10.1007/s10585-012-9561-6
Collins AT, Berry PA, Hyde C, et al (2005) Prospective Identification of Tumorigenic Prostate Cancer Stem Cells. Cancer Res 65:10946–10951. https://doi.org/10.1158/0008-5472.CAN-05-2018
Raff AB, Gray A, Kast WM (2009) Prostate Stem Cell Antigen: A Prospective Therapeutic and Diagnostic Target. Cancer Lett 277:126–132. https://doi.org/10.1016/j.canlet.2008.08.034
Saeki N, Gu J, Yoshida T, Wu X (2010) Prostate stem cell antigen (PSCA): a Jekyll and Hyde molecule? Clin Cancer Res Off J Am Assoc Cancer Res 16:3533–3538. https://doi.org/10.1158/1078-0432.CCR-09-3169
Cruz SP, Zhang Q, Devarajan R, et al (2023) TEAD1 regulates ITGA1 and ITGA2 to control prostate cancer progression. 2023.04.12.536554
Kristiansen G (2012) Diagnostic and prognostic molecular biomarkers for prostate cancer. Histopathology 60:125–141. https://doi.org/10.1111/j.1365-2559.2011.04083.x
Pampaloni F, Reynaud EG, Stelzer EHK (2007) The third dimension bridges the gap between cell culture and live tissue. Nat Rev Mol Cell Biol 8:839–845. https://doi.org/10.1038/nrm2236
Latonen L, Afyounian E, Jylhä A, et al (2018) Integrative proteomics in prostate cancer uncovers robustness against genomic and transcriptomic aberrations during disease progression. Nat Commun 9:1176. https://doi.org/10.1038/s41467-018-03573-6
Sinha A, Huang V, Livingstone J, et al (2019) The Proteogenomic Landscape of Curable Prostate Cancer. Cancer Cell 35:414-427.e6. https://doi.org/10.1016/j.ccell.2019.02.005
Brunner A-D, Thielert M, Vasilopoulou C, et al (2022) Ultra-high sensitivity mass spectrometry quantifies single-cell proteome changes upon perturbation. Mol Syst Biol 18:e10798. https://doi.org/10.15252/msb.202110798
Servant R, Garioni M, Vlajnic T, et al (2021) Prostate cancer patient-derived organoids: detailed outcome from a prospective cohort of 81 clinical specimens. J Pathol 254:543–555. https://doi.org/10.1002/path.5698
Establishment and characterization of prostate organoids from treatment‑naïve patients with prostate cancer. https://www.spandidos-publications.com/10.3892/ol.2021.13124. Accessed 12 Jun 2023
Moltzahn F, Thalmann GN (2013) Cancer stem cells in prostate cancer. Transl Androl Urol 2:242–253. https://doi.org/10.3978/j.issn.2223-4683.2013.09.06
Yun E-J, Zhou J, Lin C-J, et al (2016) Targeting cancer stem cell in castration resistant prostate cancer. Clin Cancer Res Off J Am Assoc Cancer Res 22:670–679. https://doi.org/10.1158/1078-0432.CCR-15-0190
Wolf I, Gratzke C, Wolf P (2022) Prostate Cancer Stem Cells: Clinical Aspects and Targeted Therapies. Front Oncol 12:935715. https://doi.org/10.3389/fonc.2022.935715
Skvortsov S, Skvortsova I-I, Tang DG, Dubrovska A (2018) Concise Review: Prostate Cancer Stem Cells: Current Understanding. Stem Cells Dayt Ohio 36:1457. https://doi.org/10.1002/stem.2859
Li JJ, Shen MM (2019) Prostate Stem Cells and Cancer Stem Cells. Cold Spring Harb Perspect Med 9:a030395. https://doi.org/10.1101/cshperspect.a030395
Kise K, Kinugasa-Katayama Y, Takakura N (2016) Tumor microenvironment for cancer stem cells. Adv Drug Deliv Rev 99:197–205. https://doi.org/10.1016/j.addr.2015.08.005
Bahmad HF, Jalloul M, Azar J, et al (2021) Tumor Microenvironment in Prostate Cancer: Toward Identification of Novel Molecular Biomarkers for Diagnosis, Prognosis, and Therapy Development. Front Genet 12:652747. https://doi.org/10.3389/fgene.2021.652747
Rabinovitz I, Nagle RB, Cress AE (1995) Integrin alpha 6 expression in human prostate carcinoma cells is associated with a migratory and invasive phenotype in vitro and in vivo. Clin Exp Metastasis 13:481–491. https://doi.org/10.1007/BF00118187
Lawson DA, Xin L, Lukacs RU, et al (2007) Isolation and functional characterization of murine prostate stem cells. Proc Natl Acad Sci U S A 104:181–186. https://doi.org/10.1073/pnas.0609684104
Hall CL, Dubyk CW, Riesenberger TA, et al (2008) Type I collagen receptor (alpha2beta1) signaling promotes prostate cancer invasion through RhoC GTPase. Neoplasia N Y N 10:797–803. https://doi.org/10.1593/neo.08380
Ziaee S, Chung LW (2014) Induction of integrin α2 in a highly bone metastatic human prostate cancer cell line: roles of RANKL and AR under three-dimensional suspension culture. Mol Cancer 13:208. https://doi.org/10.1186/1476-4598-13-208
Drost J, Karthaus WR, Gao D, et al (2016) Organoid culture systems for prostate epithelial and cancer tissue. Nat Protoc 11:347–358. https://doi.org/10.1038/nprot.2016.006
Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30:2114–2120. https://doi.org/10.1093/bioinformatics/btu170
Dobin A, Davis CA, Schlesinger F, et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinforma Oxf Engl 29:15–21. https://doi.org/10.1093/bioinformatics/bts635
Anders S, Pyl PT, Huber W (2015) HTSeq—a Python framework to work with high-throughput sequencing data. Bioinformatics 31:166–169. https://doi.org/10.1093/bioinformatics/btu638
Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550. https://doi.org/10.1186/s13059-014-0550-8
Shoemaker RH (2006) The NCI60 human tumour cell line anticancer drug screen. Nat Rev Cancer 6:813–823. https://doi.org/10.1038/nrc1951

Supplementary Tables 1 to 4 are not available with this version.

Download PDF

Version 1

posted

You are reading this latest preprint version

Patient-derived prostate cancer organoids require α6-integrins for self-renewal and recapitulate genetic heterogeneity and histopathology of the original tumors

Status:

Version 1

Abstract

Figures

Introduction

Results

Long term PCa organoids recapitulate the original tumor

PCa organoids recapitulates the morphological and histological features of their parental tissue

PCa organoids preserves molecular heterogeneity as in its original patient tumors.

Prostate organoids show distinct drug sensitivity profiles.

PTEN depletion induces the growth in PCa organoids

Integrin α6 regulates PCa organoid growth.

Integrin α6 is required for PCa organoid growth whereas integrin α2 is dispensable.

Discussion

Experimental procedures

Human specimens

Isolation and propagation of human prostate organoids

Histology and staining

Organoid formation Assay

Drug sensitivity assay

Mouse xenograft studies

CRISPR-Cas9 based genetic manipulation of organoids experiments

RNA-sequencing (RNA-seq) and differential gene expression analysis

Statistical Analyses

Abbreviations

Declarations

References

Supplementary Tables

Status:

Version 1