Conceived and designed the experiments: RH AB MJ OS. Performed the experiments: RH AD. Analyzed the data: RH AB MJ OS AD. Contributed reagents/materials/analysis tools: MJ OS. Wrote the paper: RH.
The authors have declared that no competing interests exist.
Galiellalactone is a potent and specific inhibitor of STAT3 signaling which has been shown to possess growth inhibitory effects on prostate cancer cells expressing active STAT3. In this study we aimed to investigate the effect of galiellalactone on prostate cancer stem cell-like cells. We explored the expression of aldehyde dehydrogenase (ALDH) as a marker for cancer stem cell-like cells in different human prostate cancer cell lines and the effects of galiellalactone on ALDH expressing (ALDH+) prostate cancer cells. ALDH+ subpopulations were detected and isolated from the human prostate cancer cell lines DU145 and long-term IL-6 stimulated LNCaP cells using ALDEFLUOR® assay and flow cytometry. In contrast to ALDH− cells, ALDH+ prostate cancer cells showed cancer stem cell-like characteristics such as increased self-renewing and colony forming capacity and tumorigenicity. In addition, ALDH+ cells showed an increased expression of putative prostate cancer stem cell markers (CD44 and integrin α2β1). Furthermore, ALDH+ cells expressed phosphorylated STAT3. Galiellalactone treatment decreased the proportion of ALDH+ prostate cancer cells and induced apoptosis of ALDH+ cells. The gene expression of
Prostate cancer is the most commonly diagnosed tumor among men and there is a great need of novel therapies against castration resistant prostate cancer
In the search for specific markers of cancer stem cells, aldehyde dehydrogenase (ALDH) has shown promise as such a marker in different cancers including bladder cancer
Signal transducer and activator of transcription 3 (STAT3) is an important transcription factor in many cancer types and it has been shown to be involved in drug resistance and to have anti-apoptotic effects in prostate cancer cells. Constitutively active STAT3 contributes to oncogenesis through upregulation of genes coding for anti-apoptotic proteins, cell cycle regulators and angiogenesis stimulators, leading to increased survival and uncontrolled growth of cancer cells
In this study we aimed to explore the expression of ALDH as a marker for cancer stem cell-like cells in different human prostate cancer cell lines and the effects of the STAT3 inhibitor galiellalactone on ALDH expressing prostate cancer cells.
The human prostate cancer cell lines DU145, LNCaP (from the American Type culture Collection, [ATCC]) and long-term interleukin-6 (IL-6) stimulated LNCaP cells (LNCaP-IL6 cells)
Prostate cancer cells (DU145, LNCaP and LNCaP-IL6) were subjected to ALDEFLUOR® assay (StemCell Technologies, Aldagen, Inc., Durham, NC) followed by flow cytometry to detect cells with high activity of ALDH (ALDH+). The active reagent BODIPY®-aminoacetaldehyde was added to the cells which was converted by ALDH to the fluorescent BODIPY®-aminoacetate. The ALDH inhibitor diethylaminobenzaldehyde (DEAB) was used as a negative control. DU145 and LNCaP-IL6 cells were treated with 0–50 µM galiellalactone for 24 h and the proportion of ALDH+ was analyzed with ALDEFLUOR® assay. DU145 and LNCaP-IL6 cells were subjected to cell sorting based on the ALDH activity using the ALDEFLUOR® assay. ALDH+ and ALDH− cells were sorted using the BD FACSAria cell sorter (BD Biosciences, San Jose, CA). Non-viable cells were excluded using 7-Amino-actinomycin D staining. ALDH+ and ALDH− subpopulations from DU145 and LNCaP-IL6 cells were re-plated after sorting and analysed with ALDEFLUOR® assay after one week in culture in order to investigate their self-renewing capacity.
ALDH+ and ALDH− cells sorted from DU145 and LNCaP-IL6 cells were directly subjected to cytospin or replated and treated with galiellalactone, trypsinized and washed in PBS before subjected to cytospin. The cell suspension was placed in a cytospin funnel clamped to a glass slide and spun at 800 rpm for 2 minutes. The slides were air dried, fixed in 4% paraformaldehyde for 10 minutes and permeabilized with 1% Triton-x in Tris buffer, pH 7.6 for 1 h. The slides were subjected to immunohistochemistry using Dako Autostainer Plus and EnVision™+ kit (Dako). The antibodies used were CD44 (BD Pharmingen), integrin α2β1 (Abcam), CD133, (c-PARP), p-STAT3 and p-NF-κB p65 (Cell Signaling Technology).
Sorted ALDH+ and ALDH− cells from DU145 and LNCaP-IL6 cells were treated with 25 µM galiellalactone for 24 h, subjected to cytospin and stained for the apoptotic marker c-PARP. A minimum of 500 cells were counted from two separate experiments and the number of labeled c-PARP cells were calculated relative the total number of cells. The number of apoptotic cells was expressed as a fraction of the total number of cells.
WST-1 proliferation assay was performed to study the effect of galiellalactone on proliferation and viability of ALDH+ and ALDH− cells sorted from DU145 and LNCaP-IL6 cells. ALDH+ and ALDH− cells were cultured in 96-well plates at the density of 2 000 cells/well in 200 µl of medium. Cells were allowed to set for 24 h. The cells were treated with 0–50 µM galiellalactone for 24 h. Samples were made in triplicate. 20 µl WST-1 solution (Roche Applied Science, Mannheim, Germany) was added per well and incubated at 37°C for 4 h. The absorbance of each well was measured using a scanning multi-well spectrophotometer, ELISA reader at a wavelength of 450 nm and reference wavelength of 690 nm. The results are presented as per cent of untreated control cells.
ALDH+ and ALDH− cells sorted from DU145 and LNCaP-IL6 cells were cultured in 24-well plates at a density of 200 cells per well. The colony formation was measured after one week. Clones were visualized with Crystal Violet staining and counted. The cloning efficiency was calculated as percent of plated cells.
Six to eight-week old male nude NMR1 mice were kept on a 12 h light-dark cycle with access to food and water
Freshly sorted ALDH+ and ALDH− populations of DU145 and LNCaP-IL6 cells were suspended in serum free medium/Matrigel (BD Biosciences) mixture (1∶1 volume). Six to eight-week old male nude NMR1 mice (n = 5) were injected with ALDH+ or ALDH− cells subcutaneously in the right and left flank of the mouse, respectively, in concentrations 10 000 or 100 000 cells per injection site. The tumor growth was monitored during the whole experiment. Tumor size was measured after six weeks using a caliper and the tumor volume (µl) was calculated by the formula length (mm)×width×height×0.5632. The mice were sacrificed with isoflurane and cervical dislocation after 6 week.
The gene expression of CD44, CD133, integrin α2 and ALDH1A1 was investigated with quantitative real-time PCR (q-PCR). Cells were harvested by brief centrifugation directly after cell sorting or after treatment with galiellalactone, and pellets were stored at −80°C until RNA isolation. Total RNA was isolated using QIAshredder spin columns, and further mRNA enrichment and additional washing was performed using the RNeasy kit (Qiagen Sciences). To exclude DNA contamination, samples were DNAse treated for 30 min using RQ1 RNAse-free DNAse (Promega). Complementary DNA (cDNA) was synthesized using random hexamers and reverse transcriptase Superscript II (Invitrogen). Total RNA was isolated from DU145 xenografts using Trizol (Invitrogen). Five micrograms of total RNA was obtained for the cDNA synthesis using a HPLC purified Oligo dTprimer with a T7 sequence.
Quantitative real-time PCR was carried out using 6–20 ng cDNA, 250 nM forward and reverse primer in 2× SYBR Green PCR Master Mix (Applied Biosystems) in a 25 µl reaction. The cycling conditions were: 10 min at 95°C to activate the enzyme, then 40 cycles of 95°C for 15 sec and 60°C for 60 sec. Relative expression levels were quantified by the comparative Ct method
Results are expressed as the mean ± standard error of the mean (SEM). Statistical analysis of data was performed by Dunnett's test or Student's t test using JMP software (SAS Institute Inc, Cary, NC). The results were considered to be statistically significant at p<0.05.
DU145, LNCaP and LNCaP-IL6 cells were investigated for their ALDH activity using the ALDEFLUOR® assay and flow cytometry. In DU145 cells 2.62±0.23% of the cells were expressing high ALDH activity (ALDH+), 2.84±0.5% of the LNCaP-IL6 cells were ALDH+ and in LNCaP cells 0.57±0.34% cells showed ALDH activity (n = 3–4) (
DU145, LNCaP and LNCaP-IL6 cells were subjected to ALDEFLUOR® assay in order to identify cells with high ALDH expression (ALDH+). The ALDH inhibitor DEAB was used as a negative control (left panel). The cells without inhibitor shifted to the right were considered ALDH+ cells (right panel).
The characteristics of ALDH+ and ALDH− cell populations isolated from DU145 and LNCaP-IL6 cells were investigated with respect to self-renewing and colony-forming capacity and tumorigenicity.
The sorted ALDH+ and ALDH− subpopulations from DU145 and LNCaP-IL6 cells were re-plated after sorting and analysed with ALDEFLUOR® assay after one week in culture in order to investigate their self-renewing capacity. After one week in culture small populations of ALDH+ cells were detected among the ALDH− DU145 and ALDH− LNCaP-IL6 cells (0.26±0,24% and 0.36±0.05%, respectively, n = 2). However, after one week in culture after sorting, the ALDH+ LNCaP-IL6 and ALDH+ DU145 cell cultures re-established their parental phenotype and constituted a population resembling that of the original parental cell line with 2.24±1.23% and 1.24±0.14% ALDH+ cells, respectively (n = 2).
The colony-forming efficiency was significantly greater in ALDH+ cells compared to ALDH− cells from both LNCaP-IL6 and DU145 cells (
ALDH+ cells show stem cell-like characteristics in terms of increased colony forming capacity and tumorigenicity. A. Clonogenic assay. 200 cells/well were seeded and colonies were counted after 1 week. Clonogenicity was calculated as percent colonies formed in relation to cells seeded (n = 2; p<0.05). B. Tumorigenicity assay. Tumor take after subcutaneous injections of 10 000 or 100 000 cells of freshly sorted ALDH+ and ALDH− cells from DU145 and LNCaP-IL6 (n = 5). C. Tumor size of ALDH+ and ALDH− cell xenografts after 6 weeks (n = 5). ALDH+ cell tumors from 100 000 cells injected subcutaneously were significantly larger than the ALDH− cell tumors for both DU145 cells (p = 0.0002) and LNCaP-IL6 cells (p = 0.0077).
The expression of various biomarkers putatively related to cancer stem cells were investigated in freshly sorted ALDH+ and ALDH− fractions of DU145 and LNCaP-IL6 cells (
A. Freshly sorted ALDH+ and ALDH− cells from DU145 and LNCaP-IL6 cells were subjected to cytospin and stained for CD44, integrin α2β1, p-STAT3 and p-NF-κB p65. B. The relative mRNA expression of CD44 and integrin α2 in ALDH+ and ALDH− cells freshly sorted from DU1145 and LNCaP-IL6 cells.
The gene expression of CD44, integrin α2 and CD133, was investigated with quantitative -PCR. (
DU145 and LNCaP-IL6 cells were treated with galiellalactone (5, 10, 25 or 50 µM) for 24 h and subjected to ALDEFLUOR® assay in order to investigate the effect of galiellalactone on the proportion of ALDH+ cells compared to vehicle. Galiellalactone decreased the proportion of ALDH+ cells of both DU145 and LNCaP-IL6 cells in a dose dependent manner (
A. DU145 and LNCaP-IL6 cells were treated with galiellalactone (5–50 µM) for 24 h and subjected to ALDEFLUOR® assay. The proportion of ALDH+ cells significantly decreased due to galiellalactone treatment in a dose dependent manner. The proportion of ALDH+ cells is expressed as mean ± SEM (n = 2–4). The proportion of ALDH+ DU145 cells were significantly decreased by galiellalactone at the concentrations 10 µM (p = 0.0060), 25 µM (p = 0,0013) and 50 µM galiellalactone (p<0,0001). The proportion of ALDH+ LNCaP-IL6 cells was significantly decreased by galiellalactone at the concentrations 25 µM (p = 0.0441) and 50 µM (p = 0.0445).
Mice with DU145 xenografts were treated with vehicle or 1 mg/kg galiellalactone daily for three weeks, the tumors were harvested and the gene expression was analyzed. The mRNA expression of
Sorted ALDH+ and ALDH− cells from DU145 and LNCaP-IL6 cells were treated with 25 µM galiellalactone for 24 h and immunostained for the apoptotic marker c-PARP (
A. ALDH+ and ALDH− cells sorted from DU145 and LNCaP-IL6 cells were treated with 25 µM galiellalactone for 24 h. Apoptotic cells were detected by c-PARP staining. B. Quantification of c-PARP stained apoptotic cells in ALDH+ and ALDH− cells sorted from DU145 and LNCaP-IL6 cells and treated with 25 µM galiellalactone for 24 h. Treatment with galiellalactone significantly increased the amount of c-PARP expressing cells in ALDH+ DU145, ALDH+ LNCaP-IL6 and ALDH− LNCaP-IL6 cells compared to untreated controls (p = 0.019, 0.035 and 0.009 respectively; n = 2). The difference in apoptotic response between galiellalactone treated ALDH+ and ALDH− cells from DU145 and LNCaP-IL6 cells was not significant (p = 0.059 and p = 0.119, respectively; n = 2). C. Galiellalactone decreased the viability of ALDH+ and ALDH− cells sorted from DU145 and LNCaP-IL6 cells. The ALDH+ DU145 cells were significantly more sensitive to galiellalactone compared to the corresponding ALDH− cells at 5 µM (p = 0.0017), 10 µM (p = 0.0010), 25 µM (p = 0.0019) and 50 µM (p = 0.0025). Results are presented as mean per cent of untreated control (n = 2).
Prostate cancer stem cells are shown to be androgen independent and lack expression of a functional androgen receptor
In the present study we identified cancer stem cell-like subpopulations in cultured prostate cancer cell lines which responded to treatment with the STAT3 inhibitor galiellalactone. A small fraction (2–4%) of the cell line populations showed high ALDH activity. These ALDH+ cells revealed stem cell-like characteristics such as increased colony forming and self-renewing capacity and high tumorigenicity as well as expression of putative prostate cancer stem cell markers. These cells responded to treatment with galiellalactone.
The capability of reconstituting a heterogenous mass is a definition of cancer stem cells, and we found that isolated ALDH+ prostate cancer cells were capable of self-renewal and re-establishment of the parental cell line and possessed high tumorigenicity
The cancer stem cell-like ALDH+ population was greater in long term IL-6 stimulated LNCaP cells compared to LNCaP cells supporting the view that prostate cancer stem cells show a pro-inflammatory phenotype
Other natural products besides galiellalactone have been shown to inhibit cancer stem cells. Sesquiterpene lactone parthenolide and curcumin are cytotoxic to cancer stem cells and target the cells by inhibiting the activity of NF-κB and STAT3
In conclusion, prostate cancer cell lines contained ALDH+ subpopulations with stem cell-like characteristics which expressed phosphorylated STAT3. These subpopulations were clearly inhibited by the STAT3 inhibitor galiellalactone. These findings emphasize that targeting the STAT3 pathway in prostate cancer cells, including prostate cancer stem cell-like cells, may be a novel potent treatment strategy in patients with advanced prostate cancer resistant to ADT and cytotoxic therapy and that galiellalactone is an important compound for studying STAT3 signaling in prostate cancer and a potential starting point for the development of future prostate cancer drugs.
At the Department of Clinical Sciences, Malmö, Lund University, Skåne University Hospital, Malmö, Sweden we wish to thank Susan Evans Axelsson for excellent help with the animal studies, Elise Nilsson for performing immunohistochemistry and Per-Anders Bertilsson for excellent help with cell sorting. We also wish to thank professor Zoran Culig at the Department of Urology, Innsbruck Medical University, Innsbruck, Austria for the generous gift of LNCaP-IL6 cells.