Bone metastasis is a leading cause of morbidity and mortality in prostate cancer. While cancer stem-like cells have been implicated as a cell of origin for prostate cancer metastasis, the pathways that enable metastatic development at distal sites remain largely unknown. In this study, we illuminate pathways relevant to bone metastasis in this disease. We observed that cyclin A1 (CCNA1) protein expression was relatively higher in prostate cancer metastatic lesions in lymph node, lung, and bone/bone marrow. In both primary and metastatic tissues, cyclin A1 expression was also correlated with aromatase (CYP19A1), a key enzyme that directly regulates the local balance of androgens to estrogens. Cyclin A1 overexpression in the stem-like ALDHhigh subpopulation of PC3M cells, one model of prostate cancer, enabled bone marrow integration and metastatic growth. Further, cells obtained from bone marrow metastatic lesions displayed self-renewal capability in colony-forming assays. In the bone marrow, cyclin A1 and aromatase enhanced local bone marrow-releasing factors, including androgen receptor, estrogen and matrix metalloproteinase MMP9 and promoted the metastatic growth of prostate cancer cells. Moreover, ALDHhigh tumor cells expressing elevated levels of aromatase stimulated tumor/host estrogen production and acquired a growth advantage in the presence of host bone marrow cells. Overall, these findings suggest that local production of steroids and MMPs in the bone marrow may provide a suitable microenvironment for ALDHhigh prostate cancer cells to establish metastatic growths, offering new approaches to therapeutically target bone metastases. Cancer Res; 76(8); 2453–64. ©2016 AACR.

Malignant stem cells may arise within the normal stem cell population or through carcinogenic transformation of a differentiated cell type, enabling the adoption of stem-like features (1–3). It is believed that only a subset of cells, termed cancer stem cells, are capable of initiating metastatic dissemination to distant organs (3–6). Cancer stem cells have been identified in malignancies, including leukemia, brain, breast, colon, and prostate cancers as well (2, 4, 7–12).

In prostate cancer, only rare, phenotypically distinct prostate cancer tumor-initiating cells, also termed stem-like prostate cancer cells, have the capacity to form new tumors (4, 7–9). Metastatic prostate cancer remains a major clinical challenge, as most of prostate cancer patients die from bone metastases (13, 14). Prostate cancer are believed to favor the bone marrow, which are the unique sites enriched with hematopoietic cells, osteoblasts, stromal, extracellular matrixes, and multiple cytokines (5, 13, 15, 16). The stem-like prostate cancer cells are believed to produce high levels of chemokine receptors that enable interaction of prostate cancer with the infiltrated bone marrow cells to facilitate distant metastases (16–18). The growth and metastases of stem-like prostate cancer cells are also influenced by various types of steroid hormones, including androgens and estrogens and their receptors (19–23). Recent reported studies suggest that estrogen and its receptor ERα are present in the hematopoietic stem cells (HSC) and bone cells (24–26). However, it remains largely unknown whether the steriod hormones may promote metastatic growth of the stem-like prostate cancer cells in the bone marrow.

Aromatase is a steroid biosynthesis enzyme that is encoded by the CYP19A1 gene and catalyzes the conversion from androgens to estrogen (27). Aromatase expression in prostate epithelial cells and the infiltrated inflammatory cells is induced by the cytokines and is associated with the progression of prostate cancer (28). CYP19A1 expression is elevated in prostate cancer metastatic tissue with 30-fold higher than that in primary tumors (19, 21, 29). Conversely, deletion of CYP19A1 alleles reduces incidence of prostate cancer in mice after exposure to testosterone and estrogen (19).

The cell-surface markers that recognize cancer stem cells provide possibility to study the unique mechanisms underlying cancer metastases in mouse models (30, 31). Aldehyde dehydrogenase 1 (ALDH1) is a detoxifying enzyme responsible for the oxidation of intracellular aldehydes and has emerged as an ideal marker for isolation of stem-like cells from heterogeneous tumors (32–35). As the ALDH activity is increased in cancer stem cell populations isolated from multiple myeloma, acute myeloid leukemia (15), breast cancer lung, head and neck, gastric and colorectal cancers and recently in prostate cancer (36–45), it is therefore considered as a reliable marker for isolation of cancer stem cells. Importantly, isolated ALDHhigh stem-like prostate cancer cells initiate growth of tumors in orthotopic mouse models (40).

Cyclin A1 is an important cell-cycle regulator and its expression is elevated in prostate cancer (46, 47). It acts as a coregulator of AR on VEGF promoter activity to regulate VEGF expression (46, 47). Induced overexpression of cyclin A1 in PC3 cells initiated metastatic growth in lymph node, lung and liver in subcutaneous and orthotropic xenograft mouse models (47). In metastatic PC3 cells, cyclin A1 is functionally associated with VEGF and MMP2/MMP9 (47). Recently, we reported that cyclin A1 interacts with ERα to promote breast cancer progression in xenograft mouse models (48). Cyclin A1 function is required for hematopoietic stem and progenitor cells (HSPC) to home to their bone marrow niches (49). In this study, we showed that ALDHhigh stem-like cells facilitate metastatic growth by utilizing cyclin A1 and aromatase to increase androgen to estrogen conversion and recruiting extracellular matrix metalloproteinases (MMP9) from the host bone marrow.

Tissue specimens, tissue microarrays, cDNA microarrays, and CGH arrays

Tissue microarrays (TMA) containing primary prostate cancer (n = 17) and metastatic prostate cancer lesions (n = 43) from 14 prostate cancer patients, and paired benign prostate hyperplasia (BPH; n = 48) versus prostate cancer tissues (n = 48) from 48 patients were constructed at the Department of Clinical Pathology and Cytology, Skåne University Hospital, Malmö. The mRNA expression data of cyclin A1 and aromatase (CYP19A1) were extracted from the cBioPortal database with GEO accession number GSE21032 as described (50, 51). The study was approved by the Ethics Committee, Lund University, and the Helsinki Declaration of Human Rights was strictly observed.

Immunohistochemistry analysis

Immunohistochemistry on tumor tissue arrays was performed as previously described (46). The staining procedure was performed using a semiautomatic staining machine (Ventana ES, Ventana Inc.), as previously described (47).

Cell culture and generation of vectors and stable and transient transfection

The androgen-insensitive cell line PC3M was kindly provided by Dr. J Fidler (Department of Urology, MD Anderson Cancer Center, Houston, TX; ref. 52). The cell lines were authenticated by the suppliers. The PC3M cells were received in 2011, and fresh-frozen stocks were used. The total span of years in using PC3M cells in our labs is approximately three years. For transient transfection studies, pMSCV-cyclinA1-EGFP was generated by cloning the full-length (1.8 kb) human cyclin A1 cDNA into the EcoRI site of the pMSCV–EGFP construct (Clonetech Inc.).

Prostate tumor spheroid formation assays

PC3M cells were cultured in polyhema-coated flasks at 5,000/mL in spheroid medium modified from protocols used for mammosphere formation as described (53). To generate second passage spheres, cells derived from the first passage spheres were plated again at a density of 5 × 103 cells/mL and grown in culture medium for 7 days as mentioned above.

ALDEFLUOR assay

The ALDEFLUOR kit (StemCell Technologies) was used to isolate the population with a high ALDH enzymatic activity (stem-like ALDHhigh cells) according to the manufacturer's protocol. Cells were stained with 7AAD (BD Biosciences) to distinguish viable cells from dead cells and analysis on CyAn ADP flow cytometer or FACS Aria (Beckman Coulter). FACS data were analyzed using FCS Express software (DeNovo Software).

Mouse model of prostate cancer distant metastases

The animal studies were approved by the Swedish Regional Ethical Animal Welfare Committee and the guidelines were strictly followed. Athymic NMRI nude male mice ages 8 to 12 weeks (Taconic Europe) were used. For intravenous injection of tumor cells, mice were sublethally irradiated with two doses of 2 Gy administered 2 hours apart using a 137Cs source at a dose rate of 1 Gy/minutes as described (54). Equal amounts of ALDHhigh cells (1 × 105 cells/mouse) or unsorted PC3M cells (2 × 106 cells/mouse) were injected into mice via tail veins. For intracardiac injection, 1 × 105 cells/mouse were injected into the left ventricle of anaesthetized mice (1% isoflurane through inhalation). In vivo imaging device IVIS imaging system was used (PerkinElmer). For HLA-ABC antibody-based imaging, mice were injected intraperitoneally with 30 μg of HLA-ABC antibody conjugated with 680 DyLight NHS-ester (Life Technologies; ref. 55).

Identification of metastatic prostate cancer cells in the bone marrow of mice by FACS

To identify metastatic prostate cancer cells in the bone marrow of irradiated recipients after transplantation, bone marrow cells from long bones of the recipient mice were isolated and were assessed using FACS methods as described (49).

Evaluation of ability of single cells from metastatic tumors of mice to form colony-forming unit using methylcellulose-based medium

To assess the repopulating ability and differentiation potential of the bone marrow progenitor cells or metastatic prostate cancer cells in the bone marrow, we used a methylcellulose-based colony-forming assay according to the manufacturer's description (MethoCult, Stem Cell Technologies). Colonies containing 30 and more cells were scored after 14 days.

Immunofluorescence analysis

For immunofluorescence analysis, cell suspensions were fixed on slides in 4% paraformaldehyde (Merck KGaA) for 15 minutes at room temperature. The slides were stained with primary antibodies including anti-HLA (Biosite) and anti-cyclin A1 (BD Pharmingen).

Treatment of ALDHhigh or ALDHlow cells with 17-β-estradiol or DTH

For estrogen treatment, ALDHhigh or ALDHlow populations were sorted from PC3M cells and were maintained in 10% charcoal-stripped phenol red free medium (CSS; Gibco, Life Technologies) for 24 hours prior to treatment with 17-β-estradiol at 10 nmol/L and DTH at 5 nmol/L (Sigma-Aldrich Inc.). The effects of 17-β-estradiol and DTH on ALDHhigh or ALDHlow cells were determined using the nonradioactive tetrazolium dye-based proliferation assay (MTS; Promega Biotech) according to the manufacturer's protocol.

Coculture of ALDHhigh or ALDHlow cells with the bone marrow and measurement of estradiol production

For the MTS assay, ALDHhigh or ALDHlow cells were cultured in serum-free phenol red free RPMI medium containing 50% of the bone marrow extracts of NMRI-nude mice. The cells were cultured for 48 hours and were subjected to the MTS assay. For measurement of estradiol production, the Estradiol Assay kit (R&D Systems Ltd.) was used according to the manufacturer's protocol. For treatment with the Aromatase inhibitor, a Type I aromatase inhibitor (Merck Millipore) was used at 100 nmol/L concentration. Concentrations were determined by measuring absorbance on an Infinite M200 multimode microplate reader (Tecan Sunrise). Testosterone and estradiol levels were determined using an immunoassay from Roche Diagnostics (Estradiol III) on a Cobas 6000 Analyzer (Roche Diagnostics).

Statistical analysis

Statistical analysis was performed. Distribution of overall survival (OS) was estimated by the method of Kaplan–Meier, with 95% confidence intervals. Differences between survival curves were calculated using the log-rank test using statistical program (SPSS, 16.0).

Description of experimental procedures and methods are given in Supplementary Materials and Methods.

Expression of cyclin A1 and aromatase in patients with prostate cancer

Immunohistochemical analysis was performed to assess protein expression of cyclin A1 and aromatase on TMAs consisting of BPH (n = 48 patients), primary prostate cancer (n = 65) and metastatic prostate cancer in lymph nodes, bone marrows and lungs (n = 43) from 14 prostate cancer patients who suffered bone metastasis. Cyclin A1 expression was predominantly nuclear, while aromatase expression was cytoplasmic in primary and metastatic prostate cancer tissues (Fig. 1A). Expression of cyclin A1 was higher in primary prostate cancer than in BPH (for cyclin A1, P < 0.001; Fig. 1B). Metastatic lesions displayed increased cyclin A1 expression compared with that of primary prostate cancer (P = 0.019; Fig. 1C). Although aromatase expression did not show significant increase in metastatic prostate cancer compared with primary tumors (Fig. 1C), it was significantly higher in primary prostate cancer than in BPH (P = 0.001; Fig. 1B). Further, expression of cyclin A1 was positively correlated with aromatase expression in both primary and metastatic prostate cancer as determined by the Spearman rank correlation test (R2 = 0.398, P < 0.001).

Figure 1.

Evaluation of the clinical importance of cyclin A1 and aromatase CYP19A1 in prostate cancer (PCa) patients. A–C, immunohistochemical analysis of cyclin A1 and aromatase in primary prostate cancer specimens, and primary prostate cancer versus prostate cancer metastatic lesions in lymph nodes (LN), bone marrow (BM), and lung (Lung). Representative microphotographs are shown. Box plot, quantitative comparison of cyclin A1 and aromatase expression between BPH and primary cancer specimens in B, and between primary and metastatic cancer specimens in C. The paired Wilcoxon rank sum test analyses are shown. ***, P < 0.001; **, P < 0.01; *, P < 0.05. D, box plots present expression of mRNA of cyclin A1 and aromatase in primary and metastatic prostate cancer. E, gene alteration profiles for CYP19A1 and cyclin A1 in primary and metastatic patients. F, Kaplan–Meier survival analysis to show biochemical recurrence (BCR). P values are indicated.

Figure 1.

Evaluation of the clinical importance of cyclin A1 and aromatase CYP19A1 in prostate cancer (PCa) patients. A–C, immunohistochemical analysis of cyclin A1 and aromatase in primary prostate cancer specimens, and primary prostate cancer versus prostate cancer metastatic lesions in lymph nodes (LN), bone marrow (BM), and lung (Lung). Representative microphotographs are shown. Box plot, quantitative comparison of cyclin A1 and aromatase expression between BPH and primary cancer specimens in B, and between primary and metastatic cancer specimens in C. The paired Wilcoxon rank sum test analyses are shown. ***, P < 0.001; **, P < 0.01; *, P < 0.05. D, box plots present expression of mRNA of cyclin A1 and aromatase in primary and metastatic prostate cancer. E, gene alteration profiles for CYP19A1 and cyclin A1 in primary and metastatic patients. F, Kaplan–Meier survival analysis to show biochemical recurrence (BCR). P values are indicated.

Close modal

We next examined mRNA expression of cyclin A1 and aromatase using a prostate cancer cohort consisting of primary (n = 131) and metastatic (n = 19) prostate cancer tissues extracted from the publicly available databases (50). Cyclin A1 and aromatase mRNA expression remained to be similar between primary and metastatic prostate cancer tissues (Fig. 1D). Somatic mutations or amplifications in genes encoding cyclin A1 and aromatase appeared to be rare. Only 4% (8 out of 216) amplifications in gene encoding for cyclin A1 and 2% (4 out of 216) amplifications in the CYP19A1 gene were observed (Fig. 1E). This suggests that overexpression of cyclin A1 and aromatase in prostate cancer may be due to the alterations at protein levels. Nevertheless, we observed that a subset of patients with higher CYP19A1 expression suffered poorer BCR-free survival as compared with those with lower CYP19A1 (P < 0.001; Fig. 1F). Cyclin A1 expression was not associated with disease-free survival of prostate cancer patients in these two cohorts (P = 0.348; Fig. 1F).

Cyclin A1 promoted growth of “tumor spheres” in vitro

The stem-like ALDHhigh and non-stem–like ALDHlow subpopulations of PC3M cells were sorted by FACS using the ALDEFLUOR assay (Fig. 2A and B), and were examined for the expression of cyclin A1 and aromatase. Both ALDHhigh and ALDHlow cells expressed cyclin A1 and aromatase (Fig. 2C). Next, the role of cyclin A1 in tumor-initiating property of stem-like cells was assessed. We used three-dimensional (3-D) culture systems, as this method is well established for assessment of breast cancer stem cells (56, 57). PC3M cells that were stably transfected with pMSCV-EGFP or pMSCV-EGFP-A1 were seeded in 3-D medium and were cultured for a period of 7 days. One of the characteristics of the stem-like cells is the ability to form “tumor spheres.” PC3M cells expressing cyclin A1 formed larger “tumor spheres” with increased cell numbers as compared with the controls (P = 0.03; Fig. 2D and E). We next re-plated the single-cell suspensions prepared from the spheres to test their self-renewal ability. The re-plated cells expressing cyclin A1 again formed larger spheres with higher cell numbers compared with the controls (P = 0.015; Fig. 2D and E). Further, PC3M cells expressing cyclin A1 contained higher frequency of the ALDHhigh subpopulation (5.43% ± 2.69%) with approximately 3-fold increase compared with the controls (1.89% ± 0.85%; P < 0.001; Fig. 2F). These data suggest that PC3M cells contain stem-like cells and that cyclin A1 overexpression increases frequency of the tumor-initiating cells in vitro.

Figure 2.

The role of cyclin A1 in self-renewal and proliferation of prostate cancer stem-like cells. A and B, representative FACS plots of ALDHhigh vs. ALDHlow cells within total PC3M cells. C, immunoblot analysis of ALDHlow vs. ALDHhigh cells using antibodies as indicated. D, quantification of numbers of cells from the “tumor spheres” of the two groups as indicated. E, representative microphotographs show the morphologies of the “tumor spheres.” F, percentages of ALDHhigh cells expressing EGFP or EGFP-A1 are shown. SD ± values indicate means of three independent experiments. *, P = 0.05.

Figure 2.

The role of cyclin A1 in self-renewal and proliferation of prostate cancer stem-like cells. A and B, representative FACS plots of ALDHhigh vs. ALDHlow cells within total PC3M cells. C, immunoblot analysis of ALDHlow vs. ALDHhigh cells using antibodies as indicated. D, quantification of numbers of cells from the “tumor spheres” of the two groups as indicated. E, representative microphotographs show the morphologies of the “tumor spheres.” F, percentages of ALDHhigh cells expressing EGFP or EGFP-A1 are shown. SD ± values indicate means of three independent experiments. *, P = 0.05.

Close modal

Cyclin A1 and aromatase promote metastatic dissemination of ALDHhigh stem-like cells to the bone marrow in mouse xenografts

To investigate whether the ALDHhigh stem-like cells may invade to the bone marrow, we established an in vivo mouse model. Nude mice were sublethally irradiated with low dose γ-irradiation to inhibit host immune function prior to the transplantation of the tumor cells. We observed that expression of aromatase and ERα was higher in ALDHhigh cells overexpressing cyclin A1 compared with the controls (Fig. 3A). Equal amounts of ALDHhigh subpopulations sorted from PC3M cells expressing the vectors were introduced into the irradiated mice via intravenous injection (Fig. 3B). The human HLA-ABC antibody conjugated with fluorescent dye was administrated into the mice for bioluminescent in vivo imaging analysis (Fig. 3C). Tumors that were positive to HLA-ABC fluorescent antibody were detected in both groups of mice using imaging device after 80 days of transplantation (Fig. 3C). We performed FACS analysis and confirmed that the transplanted cancer cells positive to HLA-ABC integrated into the bone marrow (Fig. 3D). There was a clear trend that a higher frequency of ALDHhigh cells overexpressing cyclin A1 homed the host bone marrow compared with the controls (n = 4 mice/per group; P = 0.076; Fig. 3D).

Figure 3.

Cyclin A1 overexpression promotes tumor metastasis by increasing proliferation and metastatic potential of ALDHhigh cells. A, immunoblot analysis of ALDHhigh cells expressing pMSCV-EGFP or pMSCV-EGFP-A1 using antibodies against cyclin A1 and aromatase and ERα. B, a schematic chart depicts the procedure of transplantation of sorted ALDHhigh PC3M cells into irradiated mice through tail-vein injection. C, representative images of bioluminescent in vivo imaging using IVIS imaging device. D, representative FACS plots show the percentage of HLA-ABC–positive metastatic prostate cancer cells in the bone marrow of the two groups of mice. PC3M tumor cells were used as a control as shown in the left (4 mice/group). E and G, representative microphotographs of the bone marrow sections immunostained with antibodies against cytokeratins in E and Ki-67 in G. The enlarged areas are shown in the right. F and H, the expression of cytokeratins and percentage of Ki-67–positive cells are quantified (3 mice/group).

Figure 3.

Cyclin A1 overexpression promotes tumor metastasis by increasing proliferation and metastatic potential of ALDHhigh cells. A, immunoblot analysis of ALDHhigh cells expressing pMSCV-EGFP or pMSCV-EGFP-A1 using antibodies against cyclin A1 and aromatase and ERα. B, a schematic chart depicts the procedure of transplantation of sorted ALDHhigh PC3M cells into irradiated mice through tail-vein injection. C, representative images of bioluminescent in vivo imaging using IVIS imaging device. D, representative FACS plots show the percentage of HLA-ABC–positive metastatic prostate cancer cells in the bone marrow of the two groups of mice. PC3M tumor cells were used as a control as shown in the left (4 mice/group). E and G, representative microphotographs of the bone marrow sections immunostained with antibodies against cytokeratins in E and Ki-67 in G. The enlarged areas are shown in the right. F and H, the expression of cytokeratins and percentage of Ki-67–positive cells are quantified (3 mice/group).

Close modal

Because the FACS analysis is dependent on the quality of HLA-ABC antibody and its ability to bind tumor-producing HLA-ABC antigens in vivo, this method presents certain difficulties with large variations. We therefore examined expression of cytokeratins that specifically recognize human prostate cancer cells in the bone marrow (Fig. 3E). A significantly higher number of cytokeratin-positive cells and Ki-67–positive cells was observed in the bone marrow of mice that received ALDHhigh cells expressing cyclin A1 vector (for cytokeratins, P = 0.024; for Ki-67, P < 0.001; Fig. 3E–H). These data showed that ALDHhigh cells integrated into the bone marrow and that cyclin A1 overexpression enhanced metastatic growth of the ALDHhigh cells.

We transplanted unsorted PC3M cells expressing cyclin A1 or EGFP vector into the irradiated mice via intravenous injection. Metastatic tumors in prostate, lung, and liver appeared in 3 of 5 mice that received PC3M cells expressing cyclin A1, while a large subcutaneous tumor was formed near the proximal end of the tail in 1 of 5 control mice (Supplementary Fig. S1). There was a clear trend that elevated levels of cyclin A1 in tumors were positively correlated with poorer overall survivals of the xenograft mice (Supplementary Fig. S1). No metastases to the bone marrow were detected in these xenograft mice. We therefore induced unsorted PC3M cells expressing cyclin A1 or EGFP vectors into mice via intracardiac injection, which allows tumor cells entering directly into systemic circulation, and homing to the bone marrow. Tumor metastases were detected in both groups of mice after 33 days of transplantation (5 mice/group) using bioluminescent in vivo imaging analysis (Supplementary Fig. S2A and S2B). FACS analysis revealed that HLA-ABC–positive PC3M cells integrated into the bone marrow (Supplementary Fig. S2C and S2D). Overexpression of cyclin A1 increased frequency of PC3M homing to the bone marrow relative to controls, as determined by using cytokeratin stainings of tumor cells (P = 0.016 for CK5 and P = 0.0014 for CK20; Supplementary Fig. S3).

Metastatic growth of ALDHhigh PC3M cells in the bone marrow through cyclin A1 and aromatase-associated pathways

We next wanted to assess whether metastatic lesions initiated by ALDHhigh PC3M cells may retain stem-like properties such as self-renewal and differentiation. We isolated bone marrow cells from the mice bearing bone marrow metastases initiated by ALDHhigh cells expressing cyclin A1 or control vector. The isolated bone marrow cells were then subjected to colony-forming unit (CFU) assay in the semisolid MethoCult medium, which is commonly used for the assessment of self-renewal and differentiation ability of the hematopoietic cells (Fig. 4A). Bone marrow cells containing prostate cancer metastatic lesions isolated from the mice (4 mice/group) formed colonies (Fig. 4B). To assess whether the CFUs might be derived from both host hematopoietic stem/progenitor cells, and metastatic ALDHhigh cells, single-cell suspensions were prepared from the colonies and were examined for the coexpression of cyclin A1 and HLA-ABC, a specific human tumor antigen. We observed that the subset of cells collected from the CFUs indeed coexpressed cyclin A1 and HLA-ABC, and the signals were enhanced in cells overexpressing cyclin A1 (Fig. 4C). By using the combination of Giemsa staining and visualization of EGFP positivity in CFUs, we further identified CFUs derived from ALDHhigh cells (Fig. 4D–G). This suggests that the metastatic lesions derived from ALDHhigh cells are able to self-renew and repopulate into new tumor cells in CFU assays, and this characteristic is similar to the host bone marrow hematopoietic stem/progenitor cells. Further, we observed that there was a higher proportion of prostate cancer cells in the CFUs derived from the bone marrow samples of mice bearing metastases with elevated level of cyclin A1 as compared with the controls (P < 0.001; Fig. 4H). These new findings showed that the metastatic lesions derived from ALDHhigh PC3M cells retained self-renewal and repopulating properties.

Figure 4.

Evaluation of bone marrow metastasis initiated by ALDHhigh PC3M cells. A, a schematic chart depicts the experimental procedure. B, CFU counts (3–4 mice/group). C, cell suspensions were collected from the colonies as mentioned in B and were stained with antibodies against cyclin A1 and HLA-ABC. D–G, Giemsa staining and fluorescence analysis of EGFP-expressing CFU cells with respective higher magnification images in F and G. H, percentages of EGFP-positive prostate cancer cells in the total CFUs from each group are shown. I–L, immunoblot analysis of aromatase, ERα, AR, and cleaved caspase-3 in CFUs derived from the bone marrows of mice that received ALDHhigh cells expressing EGFP or EGFP-A1, and from the bone marrow of irradiated tumor-free wild-type mice or cyclin A1-null mice, which served as controls.

Figure 4.

Evaluation of bone marrow metastasis initiated by ALDHhigh PC3M cells. A, a schematic chart depicts the experimental procedure. B, CFU counts (3–4 mice/group). C, cell suspensions were collected from the colonies as mentioned in B and were stained with antibodies against cyclin A1 and HLA-ABC. D–G, Giemsa staining and fluorescence analysis of EGFP-expressing CFU cells with respective higher magnification images in F and G. H, percentages of EGFP-positive prostate cancer cells in the total CFUs from each group are shown. I–L, immunoblot analysis of aromatase, ERα, AR, and cleaved caspase-3 in CFUs derived from the bone marrows of mice that received ALDHhigh cells expressing EGFP or EGFP-A1, and from the bone marrow of irradiated tumor-free wild-type mice or cyclin A1-null mice, which served as controls.

Close modal

We examined the expression of aromatase and ERα in the CFUs derived from the metastatic bone marrow of xenograft mice. Expression of aromatase and ERα retained in the CFUs-derived from the metastatic lesions initiated by the ALDHhigh cells and was enhanced by overexpression of cyclin A1 (Fig. 4I and J). The absence of both aromatase and ERα in the CFUs of tumor-free bone marrow was observed (Fig. 4I and J). In contrast, AR expression was absent in the CFUs derived from the metastatic lesions but was present in the CFUs of tumor-free bone marrow (Fig. 4K), suggesting that AR is produced by the host bone marrow microenvironment. We observed that overexpression of cyclin A1 was positively associated with the decreased apoptosis, as determined by examining the expression of cleaved caspase-3 in the CFUs derived from the metastatic prostate cancer lesions (Fig. 4L).

Cyclin A1 overexpression in ALDHhigh cells increased MMP9 expression in the bone marrow cells

We further examined aromatase expression in a series of bone marrow sections of the xenograft mice bearing metastases using immunohistochemical analysis. Approximately 10% aromatase-positive cells were observed in the metastatic bone marrow of control xenograft mice (Fig. 5A and B). Overexpression of cyclin A1 increased aromatase-positive cells by 60% (P < 0.001) and staining intensity of aromatase (Fig. 5A–C). The integration of ALDHhigh cells overexpressing cyclin A1 into the bone marrow significantly increased MMP9 expression in the host bone marrow (P = 0.016; Fig. 5D, G and H). Similar to AR, MMP9 expression was absent in ALDHhigh cells, but was present in the host bone marrow cells (Fig. 5E and F). These data suggest that ALDHhigh cells with elevated level of cyclin A1 influence the bone marrow–releasing factor such as MMP9.

Figure 5.

Cyclin A1 overexpression increases aromatase and MMP9 expression in mouse bone marrow. Immunostaining and statistical analysis of aromatase (A–C) and MMP9 (D) expression in bone marrow sections from xenograft recipients of ALDHhigh PC3M cells expressing EGPF or EGFP-A1 with tumors cells (arrows; n = 3 mice/group). E–H, immunoblot analysis of MMP9 expression in wild-type mouse bone marrow, ALDHhigh EGFP, and EGFP-A1 subpopulations of PC3M cells (E) and bone marrow and bone marrow–derived CFU-derived cell suspensions from ALDHhigh PC3M cells expressing EGPF or EGFP-A1 recipient xenograft mice (F). Immunoblot (G) and statistical analysis (H) of MMP9 expression in bone marrow ALDHhigh PC3M cells expressing EGPF or EGFP-A1 recipient mice (n = 3/group). *, P < 0.05; ***, P < 0.001.

Figure 5.

Cyclin A1 overexpression increases aromatase and MMP9 expression in mouse bone marrow. Immunostaining and statistical analysis of aromatase (A–C) and MMP9 (D) expression in bone marrow sections from xenograft recipients of ALDHhigh PC3M cells expressing EGPF or EGFP-A1 with tumors cells (arrows; n = 3 mice/group). E–H, immunoblot analysis of MMP9 expression in wild-type mouse bone marrow, ALDHhigh EGFP, and EGFP-A1 subpopulations of PC3M cells (E) and bone marrow and bone marrow–derived CFU-derived cell suspensions from ALDHhigh PC3M cells expressing EGPF or EGFP-A1 recipient xenograft mice (F). Immunoblot (G) and statistical analysis (H) of MMP9 expression in bone marrow ALDHhigh PC3M cells expressing EGPF or EGFP-A1 recipient mice (n = 3/group). *, P < 0.05; ***, P < 0.001.

Close modal

ALDHhigh stem-like subpopulation of PC3M cells are more sensitive to aromatase inhibition than the ALDHlow cells

We further investigated whether the ability of ALDHhigh cells to initiate metastatic growth in the bone marrow is related to their stem-like characteristics with growth advantages. We subjected equal amount of ALDHhigh and ALDHlow cells sorted from PC3M cells expressing EGFP-cyclin A1 or control vectors to the MTS-based proliferation assays. The growth rate of the ALDHhigh subpopulation was significantly higher with 20% increase compared with the ALDHlow cells (P = 0.012; Fig. 6A). Overexpression of cyclin A1 increased growth of ALDHhigh cells by 60% as compared with the ALDHlow cells (P < 0.001; Fig. 6A). Further, overexpression of cyclin A1 increased growth rate only in ALDHhigh cells (P = 0.012) but not in ALDHlow cells (Fig. 6A). This suggests that ALDHhigh cells with cyclin A1 overexpression have growth advantages over that of ALDHlow cells. Next, we inhibited aromatase production in ALDHhigh versus ALDHlow cells using an irreversible type I aromatase inhibitor. Inhibition of aromatase reduced proliferation of control ALDHhigh cells (P = 0.008) and ALDHhigh cells overexpressing cyclin A1 (P < 0.001; Fig. 6B), but had no significant effect on ALDHlow cells. We further examined the effect of bone marrow microenvironment on the growth of ALDHhigh versus ALDHlow cells. We cocultured equal amount of the sorted cells with the bone marrow extracts of mice. ALDHhigh cells displayed higher ability to adapt and grow in the bone marrow–containing medium compared with the ALDHlow cells as determined by MTS proliferation assays (Fig. 6C). Overexpression of cyclin A1 significantly enhanced ability of ALDHhigh cells, but not that of ALDHlow cells to grow in the bone marrow–containing medium (Fig. 6C). Next, we measured the concentration of estradiol, an end product of androgen to estrogen conversion, in ALDHhigh PC3M cells, in bone marrow cells alone and in cocultures of ALDHhigh PC3M cells and bone marrow cells. Estradiol levels in ALDHhigh PC3M cell cultures were similar to those in the medium only (RPMI1640 containing 10% FBS; Fig. 6D), suggesting that ALDHhigh PC3M cells do not produce detectable levels of estrogen in vitro. Bone marrow cells produced detectable levels of estradiol, which were significantly increased in coculture with ALDHhigh (Fig. 6D and E). Overexpression of cyclin A1 further increased estradiol production (for bone marrow alone vs. coculture of bone marrow and ALDHhigh cells expressing EGFP, P < 0.001; for coculture of bone marrow and ALDHhigh cells expressing EGFP-A1 vs coculture of bone marrow and ALDHhigh cells expressing EGFP, P = 0.029; Fig. 6D). Inhibition of aromatase by the aromatase inhibitor in ALDHhigh cells overexpressing cyclin A1 prior to the coculture with the bone marrow cells in serum-free medium led to a significant reduction in the estradiol level in the coculture (P < 0.001; Fig. 6E).

Figure 6.

The effect of inhibition of aromatase on ALDHhigh and ALDHlow cells and their bone marrow microenvironments. A, the proliferation rates of equal amount of ALDHhigh vs. ALDHlow PC3M cells that express EGFP or EGFP-A1. B, proliferation rates were determined in ALDHhigh vs. ALDHlow cells that were treated with aromatase inhibitor or solvent (Ctrl). C, equal amount of ALDHhigh vs. ALDHlow cells were cocultured with the bone marrow extracts and proliferation of the cells were determined using MTS assays. D, estradiol concentrations (pg/mL) were determined in different samples as indicated. The culture medium phenol red–free RPMI-1640 contains 10% serum. E, the effect of an aromatase inhibitor on estradiol concentrations in different samples. The phenol red–free serum-free RPMI-1640 medium was used. SD ± values indicate means of three independent experiments. *, P = 0.05; ***, P < 0.001.

Figure 6.

The effect of inhibition of aromatase on ALDHhigh and ALDHlow cells and their bone marrow microenvironments. A, the proliferation rates of equal amount of ALDHhigh vs. ALDHlow PC3M cells that express EGFP or EGFP-A1. B, proliferation rates were determined in ALDHhigh vs. ALDHlow cells that were treated with aromatase inhibitor or solvent (Ctrl). C, equal amount of ALDHhigh vs. ALDHlow cells were cocultured with the bone marrow extracts and proliferation of the cells were determined using MTS assays. D, estradiol concentrations (pg/mL) were determined in different samples as indicated. The culture medium phenol red–free RPMI-1640 contains 10% serum. E, the effect of an aromatase inhibitor on estradiol concentrations in different samples. The phenol red–free serum-free RPMI-1640 medium was used. SD ± values indicate means of three independent experiments. *, P = 0.05; ***, P < 0.001.

Close modal

Increasing evidence has suggested that cancer stem cells are the cell origin to initiate distant metastases (3–5). However, the unique cellular pathways that enable cancer stem cells to establish metastatic lesions into the distant organs remain largely unknown. In this study, we observed that cyclin A1 protein expression was significantly higher in prostate cancer metastatic lesions from prostate cancer patients. In agreement with a previously reported study (58), our data also showed that aromatase was abnormally expressed in human prostate cancer. Expression of both cyclin A1 and aromatase proteins was significantly higher in primary prostate cancer than in BPH. A subset of patients with upregulated aromatase mRNA expression suffered poorer BCR-free survival, although this finding was limited to fewer patients. Nevertheless, the present study adds new information on the clinical importance of aromatase expression in prostate cancer progression.

In this study, we found that the ALDHhigh subpopulation of PC3M cells had higher growth ability compared with the ALDHlow subpopulation. It has been shown that γ-radiation at low dose (2–4 Gy) promoted tumor cell infiltration with no pronounced impact on host immune function in nude mice (54). The ALDHhigh subpopulation accounted for 1.89% ± 0.85% of the total PC3M cells that initiated metastatic growth in the bone marrow of the irradiated immune-suppressive host mice, whereas unsorted PC3M cells homed to the bone marrow only when tumor cells were implanted by intracardiac injection. Intracardiac injection delivers tumor cells directly into systemic circulation, while tail-vein injection necessitates that tumor cells pass through lung and liver vasculatures before reaching the bone marrow. Thus, our study establishes a robust mouse model and FACS-based analysis to study metastases of cancer cells and stem-like cancer cells to bone. Our findings are in agreement with the reported studies in breast cancer (36–38) and prostate cancer (39, 40), which suggest that rare subsets of cancer cells are sufficient to initiate metastases to distant organs.

Although ALDHhigh PC3M cells alone are sufficient to initiate metastases into the bone/bone marrow, overexpression of cyclin A1 increased the frequency of integrated ALDHhigh cells by 40% to 50%. These findings are coincident with that overexpression of cyclin A1 promoted proliferation of the ALDHhigh subpopulation, but not ALDHlow subpopulation. Aromatase facilitates the metabolism of testicular testosterone and androstenedione to estrogen and directly regulates the local balance of androgens to estrogens (59). Growing evidence suggests that the microenvironment has profound effects on cancer metastases to preferential tissues/organs. Aromatase is able to utilize exogenous testosterones and extracellular factors from the tumor-associated microenvironment to produce estradiols and other hormones (28).

We found that the expression of bone marrow–releasing factors, including AR, ERα, and MMP9, was induced as a result of metastatic integration of ALDHhigh cells. Our new data show that increased aromatase activity in the ALDHhigh population in the context of the bone marrow microenvironment enhances estradiol production. Moreover, incorporation of a pharmacologic inhibitor establishes the requirement for aromatase in estradiol production in our model. Overexpression of cyclin A1 further increases estradiol production, providing insight into how cyclin A1 overexpression in ALDHhigh cells can influence the tumor microenvironment in the bone marrow niches. We were unable to detect significant changes in the steady-state levels of androgens in monocultures or cocultures of ALDHhigh cells and bone marrow cells (data not included).

Our study suggests that cyclin A1 is a key factor that regulates aromatase-associated pathways. However, the precise cellular mechanisms by cyclin A1 regulates aromatase expression remain to be further explored in future studies. Taken together, our data suggest that the stem-like prostate cancer cells express high levels of several key factors that facilitate metastatic homing to the bone marrow environment. Our study will provide new insights into the prostate cancer bone metastases and help to design novel targeted therapy.

N.P. Mongan is Adjunct Assistant Professor at Weill Cornell Medical College. No potential conflicts of interest were disclosed by the other authors.

Conception and design: R. Miftakhova, J.L. Persson

Development of methodology: R. Miftakhova, C. Allegrucci

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R. Miftakhova, A. Hedblom, J. Semenas, B. Robinson, A. Simoulis, J. Malm, C. Allegrucci, J.L. Persson

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R. Miftakhova, A. Hedblom, J. Semenas, A. Simoulis, N.P. Mongan, N.J. Maitland, C. Allegrucci, J.L. Persson

Writing, review, and/or revision of the manuscript: R. Miftakhova, A. Hedblom, B. Robinson, A. Rizvanov, D.M. Heery, N.P. Mongan, N.J. Maitland, C. Allegrucci, J.L. Persson

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J. Semenas, B. Robinson, A. Rizvanov, N.J. Maitland, C. Allegrucci, J.L. Persson

Study supervision: N.J. Maitland, J.L. Persson

The authors acknowledge support from the University of Nottingham and the Royal Society of London (C. Allegrucci), the Yorkshire Cancer Research (N.J. Maitland), CRUK (D.M. Heery), and Kazan Federal University (A. Rizvanov). They thank Elise Nilsson, Nishtman Dizeyi, Pradeep Kumar, and Richard Karlsson for technical support.

This work was supported by the Swedish Cancer Foundation, the National Research council, the Government Health Grant, the Malmö Cancer Foundation, the Skåne University Hospital Foundation to J.L. Persson.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Mimeault
M
,
Batra
SK
. 
Concise review: recent advances on the significance of stem cells in tissue regeneration and cancer therapies.
Stem Cells
2006
;
24
:
2319
45
.
2.
Leong
KG
,
Wang
BE
,
Johnson
L
,
Gao
WQ
. 
Generation of a prostate from a single adult stem cell.
Nature
2008
;
456
:
804
8
.
3.
Dalerba
P
,
Cho
RW
,
Clarke
MF
. 
Cancer stem cells: models and concepts.
Annu Rev Med
2007
;
58
:
267
84
.
4.
Patrawala
L
,
Calhoun
T
,
Schneider-Broussard
R
,
Li
H
,
Bhatia
B
,
Tang
S
, 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
.
5.
Wang
X
,
Kruithof-de Julio
M
,
Economides
KD
,
Walker
D
,
Yu
H
,
Halili
MV
, et al
A luminal epithelial stem cell that is a cell of origin for prostate cancer.
Nature
2009
;
461
:
495
500
.
6.
Pang
R
,
Law
WL
,
Chu
AC
,
Poon
JT
,
Lam
CS
,
Chow
AK
, et al
A subpopulation of CD26+ cancer stem cells with metastatic capacity in human colorectal cancer.
Cell Stem Cell
2010
;
6
:
603
15
.
7.
Collins
AT
,
Berry
PA
,
Hyde
C
,
Stower
MJ
,
Maitland
NJ
. 
Prospective identification of tumorigenic prostate cancer stem cells.
Cancer Res
2005
;
65
:
10946
51
.
8.
Marcinkiewicz
K
,
Scotland
KB
,
Boorjian
SA
,
Nilson
EM
,
Person
JL
,
Abrahamsson
PA
, et al
The androgen receptor and stem cell pathways in prostate and bladder cancers (Review).
Int J Oncol
2012
;
40
:
5
12
.
9.
Rosen
JM
,
Jordan
CT
. 
The increasing complexity of the cancer stem cell paradigm.
Science
2009
;
324
:
1670
3
.
10.
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
(
Pt 16
):
3539
45
.
11.
Collins
AT
,
Habib
FK
,
Maitland
NJ
,
Neal
DE
. 
Identification and isolation of human prostate epithelial stem cells based on alpha(2)beta(1)-integrin expression.
J Cell Sci
2001
;
114
(
Pt 21
):
3865
72
.
12.
Goldstein
AS
,
Lawson
DA
,
Cheng
D
,
Sun
W
,
Garraway
IP
,
Witte
ON
. 
Trop2 identifies a subpopulation of murine and human prostate basal cells with stem cell characteristics.
Proc Natl Acad Sci U S A
2008
;
105
:
20882
7
.
13.
Semenas
J
,
Allegrucci
C
,
Boorjian
SA
,
Mongan
NP
,
Persson
JL
. 
Overcoming drug resistance and treating advanced prostate cancer.
Curr Drug Targets
2012
;
13
:
1308
23
.
14.
Coleman
R
,
Body
JJ
,
Aapro
M
,
Hadji
P
,
Herrstedt
J
. 
Bone health in cancer patients: ESMO Clinical Practice Guidelines.
Ann Oncol
2014
;
25
Suppl 3
:
iii124
37
.
15.
Kaplan
RN
,
Riba
RD
,
Zacharoulis
S
,
Bramley
AH
,
Vincent
L
,
Costa
C
, et al
VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche.
Nature
2005
;
438
:
820
7
.
16.
Shiozawa
Y
,
Havens
AM
,
Jung
Y
,
Ziegler
AM
,
Pedersen
EA
,
Wang
J
, et al
Annexin II/annexin II receptor axis regulates adhesion, migration, homing, and growth of prostate cancer.
J Cell Biochem
2008
;
105
:
370
80
.
17.
Olechnowicz
SW
,
Edwards
CM
. 
Contributions of the host microenvironment to cancer-induced bone disease.
Cancer Res
2014
;
74
:
1625
31
.
18.
Luo
J
,
Ok Lee
S
,
Liang
L
,
Huang
CK
,
Li
L
,
Wen
S
, et al
Infiltrating bone marrow mesenchymal stem cells increase prostate cancer stem cell population and metastatic ability via secreting cytokines to suppress androgen receptor signaling.
Oncogene
2014
;
33
:
2768
78
.
19.
Nelles
JL
,
Hu
WY
,
Prins
GS
. 
Estrogen action and prostate cancer.
Expert Rev Endocrinol Metab
2011
;
6
:
437
51
.
20.
Hu
WY
,
Shi
GB
,
Lam
HM
,
Hu
DP
,
Ho
SM
,
Madueke
IC
, et al
Estrogen-initiated transformation of prostate epithelium derived from normal human prostate stem-progenitor cells.
Endocrinology
2011
;
152
:
2150
63
.
21.
Montgomery
RB
,
Mostaghel
EA
,
Vessella
R
,
Hess
DL
,
Kalhorn
TF
,
Higano
CS
, et al
Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for castration-resistant tumor growth.
Cancer Res
2008
;
68
:
4447
54
.
22.
Prins
GS
,
Birch
L
,
Tang
WY
,
Ho
SM
. 
Developmental estrogen exposures predispose to prostate carcinogenesis with aging.
Reprod Toxicol
2007
;
23
:
374
82
.
23.
Ricke
WA
,
McPherson
SJ
,
Bianco
JJ
,
Cunha
GR
,
Wang
Y
,
Risbridger
GP
. 
Prostatic hormonal carcinogenesis is mediated by in situ estrogen production and estrogen receptor alpha signaling.
FASEB J
2008
;
22
:
1512
20
.
24.
Burnett-Bowie
SA
,
McKay
EA
,
Lee
H
,
Leder
BZ
. 
Effects of aromatase inhibition on bone mineral density and bone turnover in older men with low testosterone levels.
J Clin Endocrinol Metab
2009
;
94
:
4785
92
.
25.
Khosla
S
,
Melton
LJ
 3rd
,
Riggs
BL
. 
Clinical review 144: Estrogen and the male skeleton.
J Clin Endocrinol Metab
2002
;
87
:
1443
50
.
26.
Nakada
D
,
Oguro
H
,
Levi
BP
,
Ryan
N
,
Kitano
A
,
Saitoh
Y
, et al
Oestrogen increases haematopoietic stem-cell self-renewal in females and during pregnancy.
Nature
2014
;
505
:
555
8
.
27.
Grumbach
MM
,
Auchus
RJ
. 
Estrogen: consequences and implications of human mutations in synthesis and action.
J Clin Endocrinol Metab
1999
;
84
:
4677
94
.
28.
Ellem
SJ
,
Risbridger
GP
. 
Treating prostate cancer: a rationale for targeting local oestrogens.
Nat Rev Cancer
2007
;
7
:
621
7
.
29.
Stanbrough
M
,
Bubley
GJ
,
Ross
K
,
Golub
TR
,
Rubin
MA
,
Penning
TM
, et al
Increased expression of genes converting adrenal androgens to testosterone in androgen-independent prostate cancer.
Cancer Res
2006
;
66
:
2815
25
.
30.
Singh
SK
,
Hawkins
C
,
Clarke
ID
,
Squire
JA
,
Bayani
J
,
Hide
T
, et al
Identification of human brain tumour initiating cells.
Nature
2004
;
432
:
396
401
.
31.
Cheng
L
,
Wu
Q
,
Guryanova
OA
,
Huang
Z
,
Huang
Q
,
Rich
JN
, et al
Elevated invasive potential of glioblastoma stem cells.
Biochem Biophys Res Commun
2011
;
406
:
643
8
.
32.
Duester
G
. 
Families of retinoid dehydrogenases regulating vitamin A function. Production of visual pigment and retinoic acid.
Eur J Biochem
2000
;
267
:
4315
24
.
33.
Magni
M
,
Shammah
S
,
Schiró
R
,
Mellado
W
,
Dalla-Favera
R
,
Gianni
AM
. 
Induction of cyclophosphamide-resistance by aldehyde-dehydrogenase gene transfer.
Blood
1996
;
87
:
1097
103
.
34.
Sophos
NA
,
Vasiliou
V
. 
Aldehyde dehydrogenase gene superfamily: The 2002 update.
Chem Biol Interact
2003
;
143–144
:
5
22
.
35.
Yoshida
A
,
Rzhetsky
A
,
Hsu
LC
,
Chang
C
. 
Human aldehyde dehydrogenase gene family.
Eur J Biochem
1998
;
251
:
549
57
.
36.
Ginestier
C
,
Hur
MH
,
Charafe-Jauffret
E
,
Monville
F
,
Dutcher
J
,
Brown
M
, et al
ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome.
Cell Stem Cell
2007
;
1
:
555
67
.
37.
Matsui
W
,
Huff
CA
,
Wang
Q
,
Malehorn
MT
,
Barber
J
,
Tanhehco
Y
, et al
Characterization of clonogenic multiple myeloma cells.
Blood
2004
;
103
:
2332
6
.
38.
Pearce
DJ
,
Taussig
D
,
Simpson
C
,
Allen
K
,
Rohatiner
AZ
,
Lister
TA
, et al
Characterization of cells with a high aldehyde dehydrogenase activity from cord blood and acute myeloid leukemia samples.
Stem Cells
2005
;
23
:
752
60
.
39.
Doherty
RE
,
Haywood-Small
SL
,
Sisley
K
,
Cross
NA
. 
Aldehyde dehydrogenase activity selects for the holoclone phenotype in prostate cancer cells.
Biochem Biophys Res Commun
2011
;
414
:
801
7
.
40.
van den Hoogen
C
,
van der Horst
G
,
Cheung
H
,
Buijs
JT
,
Pelger
RC
,
van der Pluijm
G
. 
The aldehyde dehydrogenase enzyme 7A1 is functionally involved in prostate cancer bone metastasis.
Clin Exp Metastasis
2011
;
28
:
615
25
.
41.
Goossens-Beumer
IJ
,
Zeestraten
EC
,
Benard
A
,
Christen
T
,
Reimers
MS
,
Keijzer
R
, et al
Clinical prognostic value of combined analysis of Aldh1, Survivin, and EpCAM expression in colorectal cancer.
Br J Cancer
2014
;
110
:
2935
44
.
42.
Li
XS
,
Xu
Q
,
Fu
XY
,
Luo
WS
. 
ALDH1A1 overexpression is associated with the progression and prognosis in gastric cancer.
BMC Cancer
2014
;
14
:
705
.
43.
Liu
Y
,
Lv
DL
,
Duan
JJ
,
Xu
SL
,
Zhang
JF
,
Yang
XJ
, et al
ALDH1A1 expression correlates with clinicopathologic features and poor prognosis of breast cancer patients: a systematic review and meta-analysis.
BMC Cancer
2014
;
14
:
444
.
44.
Wei
D
,
Peng
JJ
,
Gao
H
,
Zhang
T
,
Tan
Y
,
Hu
YH
. 
ALDH1 expression and the prognosis of lung cancer: a systematic review and meta-analysis.
Heart Lung Circ
2015
;
24
:
780
8
.
45.
Zhou
C
,
Sun
B
. 
The prognostic role of the cancer stem cell marker aldehyde dehydrogenase 1 in head and neck squamous cell carcinomas: a meta-analysis.
Oral Oncol
2014
;
50
:
1144
8
.
46.
Wegiel
B
,
Bjartell
A
,
Ekberg
J
,
Gadaleanu
V
,
Brunhoff
C
,
Persson
JL
. 
A role for cyclin A1 in mediating the autocrine expression of vascular endothelial growth factor in prostate cancer.
Oncogene
2005
;
24
:
6385
93
.
47.
Wegiel
B
,
Bjartell
A
,
Tuomela
J
,
Dizeyi
N
,
Tinzl
M
,
Helczynski
L
, et al
Multiple cellular mechanisms related to cyclin A1 in prostate cancer invasion and metastasis.
J Natl Cancer Inst
2008
;
100
:
1022
36
.
48.
Syed Khaja
AS
,
Dizeyi
N
,
Kopparapu
PK
,
Anagnostaki
L
,
Harkonen
P
,
Persson
JL
. 
Cyclin A1 modulates the expression of vascular endothelial growth factor and promotes hormone-dependent growth and angiogenesis of breast cancer.
PLoS One
2013
;
8
:
e72210
.
49.
Miftakhova
R
,
Hedblom
A
,
Batkiewicz
L
,
Anagnosaki
L
,
Zhang
Y
,
Sjolander
A
, et al
Cyclin A1 regulates the interactions between mouse haematopoietic stem and progenitor cells and their niches.
Cell Cycle
2015
;
14
:
1948
60
.
50.
Taylor
BS
,
Schultz
N
,
Hieronymus
H
,
Gopalan
A
,
Xiao
Y
,
Carver
BS
, et al
Integrative genomic profiling of human prostate cancer.
Cancer Cell
2010
;
18
:
11
22
.
51.
Robinson
MD
,
McCarthy
DJ
,
Smyth
GK
. 
edgeR: a Bioconductor package for differential expression analysis of digital gene expression data.
Bioinformatics
2010
;
26
:
139
40
.
52.
Pettaway
CA
,
Pathak
S
,
Greene
G
,
Ramirez
E
,
Wilson
MR
,
Killion
JJ
, et al
Selection of highly metastatic variants of different human prostatic carcinomas using orthotopic implantation in nude mice.
Clin Cancer Res
1996
;
2
:
1627
36
.
53.
Dontu
G
,
Abdallah
WM
,
Foley
JM
,
Jackson
KW
,
Clarke
MF
,
Kawamura
MJ
, et al
In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells.
Genes Dev
2003
;
17
:
1253
70
.
54.
Kobayashi
S
,
Otsu
H
,
Noda
Y
,
Ogiu
T
. 
Comparison of dose-dependent enhancing effects of gamma-ray irradiation on urethan-induced lung tumorigenesis in athymic nude (nu/nu) mice and euthymic (nu/+)littermates.
J Cancer Res Clin Oncol
1996
;
122
:
231
6
.
55.
Suemizu
H
,
Kawai
K
,
Higuchi
Y
,
Hashimoto
H
,
Ogura
T
,
Itoh
T
, et al
A versatile technique for the in vivo imaging of human near-infrared fluorochrome-conjugated macromolecule probes.
PLoS ONE
2013
;
8
:
e82708
.
56.
Ginestier
C
,
Wicinski
J
,
Cervera
N
,
Monville
F
,
Finetti
P
,
Bertucci
F
, et al
Retinoid signaling regulates breast cancer stem cell differentiation.
Cell Cycle
2009
;
8
:
3297
302
.
57.
Ponti
D
,
Costa
A
,
Zaffaroni
N
,
Pratesi
G
,
Petrangolini
G
,
Coradini
D
, et al
Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties.
Cancer Res
2005
;
65
:
5506
11
.
58.
Ellem
SJ
,
Schmitt
JF
,
Pedersen
JS
,
Frydenberg
M
,
Risbridger
GP
. 
Local aromatase expression in human prostate is altered in malignancy.
J Clin Endocrinol Metab
2004
;
89
:
2434
41
.
59.
Smith
IE
,
Dowsett
M
. 
Aromatase inhibitors in breast cancer.
N Engl J Med
2003
;
348
:
2431
42
.