FEBS Letters 589 (2015) 2058–2065
journal homepage: www.FEBSLetters.org
Suppression of ornithine decarboxylase promotes osteogenic differentiation of human bone marrow-derived mesenchymal stem cells Yo-Hsian Tsai a,1, Kuan-Lian Lin a,1, Yuan-Pin Huang b, Yi-Chiang Hsu c, Chung-Hwan Chen d,e, Yuhsin Chen f, Min-Hua Sie a, Gwo-Jaw Wang d,e,g,h, Mon-Juan Lee a,i,⇑ a
Department of Bioscience Technology, Chang Jung Christian University, Tainan, Taiwan Department of Cosmetics and Fashion Styling, Cheng Shiu University, Kaohsiung, Taiwan Taiwan Graduate Institute of Medical Sciences, Chang Jung Christian University, Tainan, Taiwan d Orthopaedic Research Center, Kaohsiung Medical University, Kaohsiung, Taiwan e Department of Orthopaedics, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan f Taichung District Agricultural Research and Extension Station, Council of Agriculture, Executive Yuan, Taichung, Taiwan g Department of Orthopaedics, National Cheng Kung University Hospital, Tainan, Taiwan h Department of Orthopedic Surgery, University of Virginia, VA, USA i Innovative Research Center of Medicine, Chang Jung Christian University, Tainan, Taiwan b c
a r t i c l e
i n f o
Article history: Received 8 February 2015 Revised 8 June 2015 Accepted 15 June 2015 Available online 30 June 2015 Edited by Quan Chen Keywords: Polyamine a-Difluoromethylornithine (DFMO) Osteogenic differentiation Small interfering RNA (siRNA) Mesenchymal stem cell
a b s t r a c t Ornithine decarboxylase (ODC) is the rate-limiting enzyme for polyamine biosynthesis. Suppression of ODC by its irreversible inhibitor, a-difluoromethylornithine (DFMO), or by RNA interference through siRNA, enhanced osteogenic gene expression and alkaline phosphatase activity, and accelerated matrix mineralization of human bone marrow-derived mesenchymal stem cells (hBMSCs). Besides, adipogenic gene expression and lipid accumulation was attenuated, indicating that the enhanced osteogenesis was accompanied by down-regulation of adipogenesis when ODC was suppressed. A decrease in the intracellular polyamine content of hBMSCs during osteogenic induction was observed, suggesting that the level of endogenous polyamines is regulated during differentiation of hBMSCs. This study elucidates the role of polyamine metabolism in the lineage commitment of stem cells and provides a potential new indication for DFMO as bone-stimulating drug. Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction Ornithine decarboxylase (ODC) is the rate-limiting enzyme for the biosynthetic pathway of polyamines, a series of low molecular-weight polycations involved in various cellular processes such as proliferation, differentiation, and apoptosis [1–4]. Because the level of polyamines and the activities of polyamine biosynthetic enzymes are increased in tumor cells, ODC is one of the potential targets for the development of anticancer drugs [5,6]. Alpha-difluoromethylornithine (DFMO), which is the
Author contributions: M.J.L. conceived and supervised the study; Y.H.T., K.L.L. and M.H.S. carried out the experiments; Y.P.H., Y.C.H. and M.J.L. designed experiments; C.H.C., Y.C. and G.J.W. performed data analysis and manuscript revision; M.J.L. drafted and finalized the manuscript. ⇑ Corresponding author at: Department of Bioscience Technology, Chang Jung Christian University, No. 1, Changda Rd., Gueiren District, Tainan City 71101, Taiwan. Fax: +886 6 2785010. E-mail address: [email protected] (M.-J. Lee). 1 The authors contributed equally to this work.
irreversible inhibitor of ODC, suppresses cancer development in animal models, and is being evaluated in clinical trials for the preventative treatment of colorectal cancer [7,8]. In addition to cancer cells, ODC and polyamine biosynthesis are also important in the proliferation of several types of precursor cells and stem cells. Recent studies indicate that the polyamine metabolic pathway serves as a potential target for eliminating breast cancer stem cells [9]. Earlier studies suggest that ODC is crucial to the proliferation of neural progenitor cells and hematopoietic precursor cells [10–12]. In rat bone-marrow stromal cells, polyamine depletion by DFMO is suggested to reduce TNFa/MG132-induced apoptosis and attenuate cell proliferation [13]. Compared to the established role of ODC in promoting cell proliferation, its function in cell differentiation is diverse and has not been fully explored. Inhibition of ODC by DFMO prevents adipocyte differentiation of 3T3-L1 cells [14,15]. However, DFMO stimulates early cardiac commitment of mesenchymal stem cells [16], as well as the differentiation of embryonal carcinoma cells
http://dx.doi.org/10.1016/j.febslet.2015.06.023 0014-5793/Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
[17–19] and F9 teratocarcinoma stem cells [20–22]. Besides, ODC is active during hair follicle development and hair growth [23]. In our previous report, the mRNA level of ODC was reduced in the presence of exogenous polyamines such as putrescine, spermidine and spermine, which enhance the osteogenic differentiation of human bone marrow-derived stem cells (hBMSCs) [24]. It is therefore hypothesized that suppression of ODC and thus depletion of endogenous polyamines may be correlated with the osteogenic induction of hBMSCs. In this study, hBMSCs were treated with DFMO as well as small interfering RNAs (siRNAs) designed against ODC (siODC) to determine whether suppression of the activity and mRNA expression of ODC alter the differentiation fate of hBMSCs. These results may help to investigate the crosstalk between pathways of osteogenesis and polyamine metabolism, and determine whether the level of intracellular polyamines may be manipulated to promote osteogenic differentiation.
2. Methods 2.1. Cell culture Human BMSCs (CD105+, CD166+, CD29+, CD44+, CD14 , CD34 and CD45 ) were isolated from human bone marrow withdrawn from the posterior iliac crest of the pelvic bone of normal volunteers, and were provided by Lonza, Basel, Switzerland. The cells were cultured in mesenchymal stem cell medium (Lonza, Basel, Switzerland) supplemented with Invitrogen MSC-Qualified Fetal Bovine Serum (Life Technologies, Carlsbad, CA, USA), according to the manufacturer’s instructions. Human BMSCs between passages 3 and 5 were used in this study. Cells were seeded at 3000 cells/well in 96-well plates for cell viability assay and alkaline phosphatase (ALP) activity assay, at 104 cells/well in 24-well plates for alizarin red S and oil red O staining, at 5 104 cells/well in 6-well plates for total RNA and protein extraction, and at 5000 cells/cm2 in 10-cm dishes for HPLC sample preparation. Alizarin red S and oil red O staining were performed as previously described [24].
2059
days after siRNA transfection, cells were cultured in OIM for another 7 to 14 days. For adipogenic induction, cells were maintained in DMEM for 7 days prior to siRNA transfection, and 48 h after siRNA treatment, were incubated in adipogenesis-inducing conditions for another 7 to 14 days. 2.4. Cell viability assay Cell viability was analyzed by WST-1 assay using Quick Cell Proliferation Colorimetric Assay Kit (BioVision, Milpitas, CA, USA) and quantitated spectrophotometrically by measuring the optical density at 450 nm, with a reference wavelength of 650 nm. 2.5. Alkaline phosphatase activity assay The ALP activity of hBMSCs was determined by Alkaline Phosphatase Fluorimetric Assay Kit (BioVision, Milpitas, CA, USA). The kit utilizes the fluorescent signal of 4-methylumbelliferone (4-MU) produced by cleavage of the non-fluorescent substrate, 4-methylumbelliferyl phosphate disodium salt (MUP), by the ALP in cell lysates. Fluorescence intensity was determined at Ex/Em 360/440 nm, and the amount of 4-MU produced was quantified using a standard curve produced by commercial ALP. ALP activity was normalized to the total amount of protein and described as the concentration of 4-MU (pmol/ml) produced per mg lysate per minute. 2.6. Real-time polymerase chain reaction Total RNA was isolated from hBMSCs by the RNeasy Mini Kit (Qiagen, Germntown, MD, USA), and cDNA synthesis was carried out using the SuperScript III First-Strand Synthesis SuperMix for qRT-PCR (Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. The cDNA was diluted to a final concentration of 1 ng/ll and reacted with gene-specific primer pairs and QuantiFast SYBRÒ Green PCR Kit (Qiagen, Germntown, MD, USA). Target genes were detected and amplified by Applied
2.2. DFMO treatment For osteogenic differentiation studies, hBMSCs were treated with 1, 10 or 100 lM DFMO in osteogenic induction medium (OIM, 10 7 M dexamethasone, 10 mM b-glycerolphosphate and 50 mM L-ascorbate 2-phosphate in DMEM) 2 days after seeding, and were cultured for another 7 or 14 days. For adipogenic differentiation analysis, cells were cultured for 7 days after seeding to reach 100% confluence, which favor adipogenesis, before treated with 1, 10 or 100 lM DFMO prepared in adipogenic induction medium (AIM, DMEM supplemented with 1 lM dexamethasone, 0.5 mM isobutylmethylxanthine (IBMX), 10 lg insulin/ml and 100 lM indomethacin) or adipogenesis maintenance medium (AMM, DMEM containing 10 lg/ml insulin). Cells were cultured alternately in DFMO-supplemented AIM or AMM for another 14 days according to the instructions of Chemicon Mesenchymal Stem Cell Adipogenesis kit (Millipore, Billerica, MA, USA). 2.3. siRNA transfection Dharmacon siGENOME siRNA specific for human ODC (siODC), as well as siGENOME GAPD Control siRNA (siMock) were designed and synthesized by Thermo Scientific (Waltham, MA, USA). Transfection of siRNA was carried out 24 h after hBMSCs were plated, using DharmaFECT 1 transfection reagent (Thermo Scientific, Waltham, MA, USA) by following the manufacturer’s instructions. The final siRNA concentration was 3.1, 6.25 and 12.5 nmol/L. Two
Fig. 1. Effect of DFMO and siODC on the viability of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 7 days. Cell viability was determined by WST-1 assay and expressed as optical density at 450 nm (OD450). Error bars represent standard deviations (n P 3). *P < 0.05, compared to the control (DMEM); #P < 0.05, compared to OIM.
2060
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
Biosystems 7300 Real-Time PCR System and Applied Biosystems Sequence Detection Software V1.2 (Life Technologies, Carlsbad, CA, USA). Primer sequences and conditions for real-time PCR analysis are the same as those reported previously [24].
proteins were produced with WesternBright ECL HRP substrate (Advansta, Menlo Park, CA, USA) and detected with C-DiGitÒ Blot Scanner (LI-COR Biotechnology, Lincoln, NE, USA). Quantification of Western blot signals was performed with LI-COR Image Studio 4.0 software (LI-COR Biotechnology, Lincoln, NE, USA).
2.7. Western blot analysis Human BMSCs were harvested by Pierce RIPA buffer (Thermo Scientific, Waltham, MA, USA) supplemented with 1 mM phenylmethanesulfonyl fluoride (Sigma–Aldrich, St. Louis, MO, USA). The crude extract was homogenized with a Branson S-450D Sonifier (Emerson Industrial Automation, St. Louis, MO, USA), followed by centrifugation at 14 000g and 4 °C for 15 min to remove cell debris. A total of 50–70 lg of protein from each individual sample was subjected to SDS–PAGE with a 4–12% gradient gel. The resolved proteins were transferred to PVDF membrane using the Trans-Blot Turbo Transfer System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The blots were incubated at room temperature for 1 h in 3% bovine serum albumin, and reacted overnight at 4 °C with primary antibodies diluted in TBST buffer (50 mM Tris, 150 mM NaCl and 0.05% Tween 20, pH 7.5). After washing with TBST buffer, the blots were reacted for 45 min with horseradish peroxidase (HRP)-conjugated secondary antibodies diluted in TBST at a ratio of 1:5000 (v:v). Chemiluminescent signals of target
2.8. Analysis of intracellular polyamine content by high-performance liquid chromatography (HPLC) Human BMSCs were treated with 100 lM putrescine, 1 lM spermidine, 1 lM spermine or 100 lM DFMO in OIM for 7 days. The cells were lysed in 0.2 M perchloric acid, followed by addition of 1,8-diaminooctane (DAO) to a final concentration of 10 lM, which serves as the internal standard. The polyamines in the cell lysate were subjected to benzoylation and analyzed by an Agilent Zorbax 300SB-C18 column (4.6 250 mm) in conjunction with an Agilent 1100 Series HPLC System. Benzoylated polyamines were eluted by a 20-min HPLC gradient program starting with methanol–water (60:40, v/v) as reported in our previous study [25]. 2.9. Statistical analysis Statistical significance relative to the control was analyzed using one-way ANOVA followed by post hoc tests. Data set from
Fig. 2. Effect of DFMO and siODC on the expression of ODC and SSAT in hBMSCs during osteogenic and adipogenic induction. Human BMSCs were treated with osteogenic induction medium (OIM) or adipogenic induction medium (AIM), or with various concentrations of DFMO in OIM or AIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM or AIM for 7 days. Relative gene expression of (A and B) ODC and (C and D) SSAT was determined by real-time PCR. Error bars represent standard deviations (n P 3). *P < 0.05 and **P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
at least three independent experiments was used in the analysis. Levels of significance were expressed as significant, P < 0.05, and highly significant, P < 0.01, respectively. 3. Results
2061
osteogenic and adipogenic induction, and treatment with DFMO or siODC suppressed SSAT mRNA expression as well (Fig. 2C and D). The decrease in SSAT expression of hBMSCs during ODC suppression suggests that the enzyme may be regulated to reduce excretion of intracellular polyamines, so as to maintain polyamine homeostasis.
3.1. Effect of DFMO and siODC treatment on cell viability When treated with DFMO or siODC, the viability of hBMSCs was slightly decreased compared to the control and that incubated in OIM, but was similar to that treated with siMock (Fig. 1). The number of viable cells decreased dramatically when hBMSCs were treated with 1 mM DFMO or 25 nM siRNAs (data not shown). 3.2. Effect of DFMO and siODC treatment on the mRNA level of ODC and SSAT The mRNA expression of ODC was suppressed 2 days after siODC transfection (data not shown). After 7 days of osteogenic or adipogenic induction, the mRNA level of ODC remained lower in cells treated with DFMO and siODC, compared to that of the OIM or AIM controls (Fig. 2A and B). On the other hand, the level of spermidine/spermine N1-acetyl transferase (SSAT), the rate-limiting enzyme of polyamine catabolism, increased upon
3.3. Effect of DFMO and siODC treatment on osteogenic gene expression Treatment of hBMSCs with DFMO resulted in up-regulation of osteogenic genes such as Runt-related transcription factor 2 (Runx2), alkaline phosphatase (ALP), osteocalcin and osteopontin (Fig. 3). Suppression of ODC by siODC also enhanced the expression of early-onset osteogenic genes Runx2 and ALP (Fig. 3A and B), but the increase in the mRNA level of osteocalcin and osteopontin was less significant under siODC treatment compared to that in the presence of DFMO (Fig. 3C and D). 3.4. Effect of DFMO and siODC treatment on ALP activity and matrix mineralization ALP activity and matrix mineralization, both of which are markers for osteoblast maturation, were enhanced under DFMO or
Fig. 3. Effect of DFMO and siODC on the osteogenic gene expression of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 7 days. Relative gene expression of (A) Runx2, (B) ALP, (C) osteocalcin and (D) osteopontin was determined by real-time PCR. Error bars represent standard deviations (n P 3). *P < 0.05 and ** P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
2062
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
cultured in OIM or siMock in the presence of 100 lM DFMO or 12.5 nM siODC (Fig. 5B). 3.5. Effect of DFMO and siODC treatment on adipogenic gene expression During adipogenic differentiation, hBMSCs treated with AIM or siMock had a significantly higher mRNA level of adipogenic genes such as peroxisome proliferator-activated receptor gamma (PPARc), adipocyte protein 2 (aP2) and adipsin, compared to the control (Fig. 6). When treated with DFMO or siODC, the expression of adipogenic genes were significantly suppressed, except for that of aP2 in the presence of 3.1 and 6.25 nM siODC. These results indicate that inhibition of the activity or mRNA expression of ODC resulted in suppression of adipogenic differentiation. 3.6. Effect of DFMO and siODC treatment on lipid accumulation
Fig. 4. Effect of DFMO and siODC on the alkaline phosphatase (ALP) activity of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 7 days. Alkaline phosphatase activity was determined by Alkaline Phosphatase Fluorimetric Assay Kit (BioVision, Milpitas, CA, USA) according to the manufacturer’s instructions. Error bars represent standard deviations (n P 3). ** P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
Lipid accumulation, an indication of adipogenic terminal differentiation, was increased in hBMSCs incubated under adipogenic conditions, or treated with siMock, followed by adipogenic induction (Fig. 7). As expected, the amount of lipid droplets accumulated in hBMSCs was significantly reduced in DFMO- and siODC-treated cells. The amount of lipid droplets decreased to around 0.6-fold or less of AIM or siMock in the presence of DFMO or siODC. It is therefore suggested that the osteogenic activity of DFMO and siODC was correlated with suppression of adipogenic terminal differentiation as well as gene expression. 3.7. Effect of DFMO and siODC treatment on the expression of transcription factors
siODC treatment. As shown in Fig. 4, when treated with 10 or 100 lM DFMO, or with 12.5 nM siODC, the ALP activity of hBMSCs was higher than that cultured in OIM alone or treated with siMock. The enhanced ALP activity coincided with the increase in the mRNA level of ALP (Fig. 3B), indicating that suppression of ODC is correlated with both the activity and mRNA expression of ALP. Results from alizarin red S staining also suggest that ODC suppression accelerated matrix mineralization of hBMSCs (Fig. 5A). The amount of cell-bound alizarin red S (lM alizarin red S/lg DNA) was increased to 1.55- or 1.24-fold, respectively, of those
We analyzed the protein expression of Runx2 and PPARc, the key transcription factors related to osteogenic and adipogenic differentiation, respectively, as well as that of Sp3, a transcription factor that antagonizes ODC and activates the liver/bone/kidney-type ALP promoter [26,27]. As shown in Fig. 8A and B, treatment of hBMSCs with DFMO or siODC resulted in a decrease in the protein expression of ODC under both osteogenic and adipogenic conditions. When hBMSCs were cultured in OIM alone, the expression of Runx2 was increased compared to the control, but was similar
Fig. 5. Effect of DFMO and siODC on matrix mineralization of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 14 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 14 days. (A) The cells were stained with alizarin red S and observed with phase-contrast microscopy. (B) The concentration of cell-bound alizarin red S was quantitated spectrophotometrically at OD570 and normalized to the amount of DNA (lg). Error bars represent standard deviations (n P 3). **P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
2063
Fig. 6. Effect of DFMO and siODC on the adipogenic gene expression of hBMSCs. Human BMSCs were cultured at high cell densities that favor adipogenesis, and treated with AIM or with various concentrations of DFMO in AIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by AIM for 7 days. Relative gene expression of (A) PPARc, (B) aP2 and (C) adipsin was determined by real-time PCR. Error bars represent standard deviations (n P 3). **P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to AIM.
Fig. 7. Effect of DFMO and siODC on lipid accumulation of hBMSCs. Human BMSCs were cultured at high cell densities that favor adipogenesis, and treated with DFMO under adipogenic induction conditions as described in Section 2 for 14 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by adipogenic induction for 14 days. (A) Cells were stained with oil red O and observed with phase-contrast microscopy. (B) Cell-bound oil red O was extracted with isopropanol and its intensity was determined colorimetrically at OD540. Error bars represent standard deviations (n P 3). **P < 0.01, compared to the control (DMEM); ##P < 0.01, compared to AIM.
to that of hBMSCs co-treated with DFMO or siODC, suggesting that Runx2 may not be directly related to the osteogenic stimulation resulted from ODC suppression (Fig. 8A). Nevertheless, we found that the protein expression of Sp3 was up-regulated in the presence of DFMO or siODC under osteogenic conditions, but not when hBMSCs was cultured in OIM alone or treated with siMock (Fig. 8A), indicating that suppression of ODC may lead to activation of ALP through the up-regulation of Sp3. On the other hand, the expression of PPARc was suppressed in the presence of DFMO or siODC and induced to differentiate adipogenically (Fig. 8B), consistent with the results of PPARc mRNA expression in Fig. 6A. 3.8. Effect of exogenous polyamines and DFMO on intracellular polyamine content Treatment with OIM alone resulted in a decrease in the concentration of all three of the common polyamines, including putrescine, spermidine, and spermine, suggesting that endogenous polyamines may be associated with the regulation of stem cell differentiation (Table 1). Compared to OIM, the total concentration of intracellular polyamines was slightly increased in the presence of 100 lM putrescine. While 1 lM spermidine did not affect total
polyamines, treatment with 1 lM spermine and 100 lM DFMO decreased intracellular polyamines to levels lower than that of OIM alone. It is known that exogenous polyamines may act through pathways distinct from their endogenous counterparts [28], but whether they were sequestered away in other pathways or were simply exported from cells remains to be investigated.
4. Discussion Muscari et al. reported that matrix mineralization of rat BMSCs was unaffected when incubated with 1 mM DFMO for 48 h, followed by induction to differentiate in osteogenic medium for 3 weeks, indicating that polyamine depletion by DFMO does not alter the differentiation potential of rat BMSCs [13]. However, we found that the viability of hBMSCs reduced significantly after treated with 1 mM DFMO in OIM for 7 days. Besides, when maintained for 14 days in OIM supplemented with a lower concentration of DFMO (100 lM), matrix mineralization of hBMSCs was enhanced (Fig. 5). These results suggest that continuous suppression of ODC during osteogenic induction may be critical in promoting osteogenic terminal differentiation.
2064
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
Fig. 8. Effect of DFMO and siODC on protein expression. Human BMSCs were treated with osteogenic induction medium (OIM) or adipogenic induction medium (AIM), or with various concentrations of DFMO in OIM or AIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM or AIM for 7 days. Cells were harvested in RIPA buffer and subjected to Western blot analysis as described in Section 2. (A) Protein expression of ODC, Runx2, and Sp3 in hBMSCs under osteogenic induction. (B) Protein expression of ODC and PPARc in hBMSCs under adipogenic induction. The ratios of Runx 2 and PPARc were determined by normalizing the intensity of each band, determined by LI-COR Image Studio 4.0 software, to that of the control.
Table 1 Intracellular polyamine contents of hBMSCs treated with exogenous polyamines or DFMO in combination with OIM for 7 days. Polyamines (nmol/mg)
Control
OIM
100 lM PUT/OIM
1 lM SPD/OIM
1 lM SPM/OIM
100 lM DFMO/OIM
Putrescine Spermidine Spermine Total polyamines
43.91 ± 2.54 16.55 ± 3.41 16.56 ± 2.35 77.01 ± 4.86
9.08 ± 3.13 3.18 ± 0.36 8.30 ± 2.48 20.56 ± 4.01
16.28 ± 0.38 5.21 ± 1.68 5.13 ± 1.95 26.62 ± 2.60
4.42 ± 0.54 1.13 ± 0.24 15.05 ± 3.80 20.61 ± 3.85
3.21 ± 1.89 4.29 ± 1.51 5.50 ± 0.03 13.01 ± 2.42
3.68 ± 1.01 2.43 ± 0.26 4.26 ± 0.37 10.36 ± 1.11
Although DFMO is known to deplete the intracellular pool of polyamines through the irreversible inhibition of ODC, a strategy to block cancer cell proliferation [29], the effect of SSAT to counteract the effect of ODC suppression was seldom discussed, especially when in this study, hBMSCs were treated with DFMO at concentrations that did not affect cell viability (Fig. 1). By analyzing both the catabolic and anabolic enzyme expression of polyamine biosynthesis, it is suggested that in the presence of exogenous polyamines, ODC was suppressed and SSAT was stimulated to prevent rapid increase in intracellular polyamines [24]. Conversely, SSAT was suppressed when hBMSCs were treated with DFMO or siODC possibly to prevent further depletion of the polyamine pool (Fig. 2C). On the other hand, OIM had no effect on the mRNA expression of ODC (Fig. 2A), but the expression of SSAT was increased (Fig. 2C), which may explain the decrease in total polyamines (Table 1). Similar increase in SSAT mRNA level was also observed in goat-derived adipose stem cells induced to differentiate
osteogenically with spermine and 1,25-dihydroxyvitamin-D3, but their effect on ODC was not reported [30]. Studies on murine embryonal carcinoma cell lines demonstrate that the level of intracellular polyamines must drop below certain threshold to induce differentiation [18]. Indeed, total intracellular polyamines were significantly lowered when hBMSCs were cultured under osteogenic conditions (Table 1). However, co-treatment of hBMSCs with OIM and exogenous polyamines or DFMO had limited effect on total polyamines compared to OIM alone, suggesting that in addition to disrupting polyamine metabolism, exogenous polyamines and DFMO may act through additional pathways to enhance osteogenic gene expression and terminal differentiation. Our results from Western blot analysis indicate that up-regulation of Sp3 by ODC suppression, together with the activation of Runx2 by osteogenic inducing agents in OIM (Fig. 8A), enhanced ALP expression (Fig. 3B) and accelerated the osteogenic differentiation of hBMSCs. However, the mechanism of how ODC
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
inhibition led to suppression of PPARc remains to be investigated. PPARc not only is the master regulator of adipocyte differentiation [31], but also plays a role in the control of cell proliferation and apoptosis, and is known to activate SSAT transcription [6,32]. The mRNA levels of PPARc, ODC and SSAT, as well as polyamine levels were significantly higher in human colorectal carcinoma than in normal tissue [33]. In hBMSCs, we demonstrate that a decrease in the expression of PPARc, ODC and SSAT, accompanied by reduced intracellular polyamines, was responsible for enhancing osteogenesis. The fact that both ODC and PPARc are associated with cell proliferation implicates that they may share common effectors in the differentiation and proliferation of stem cells. Conflict of interest statement The authors confirm that there is no conflict of interest. Acknowledgments This study is supported by the Ministry of Science and Technology, Taiwan (NSC101-2314-B-309-001-MY3), the Academic Research Funds of Chang Jung Christian University, Tainan, Taiwan, and Taichung District Agricultural Research and Extension Station, Council of Agriculture, Executive Yuan, Taichung, Taiwan. We thank the Toxicology Research Center, Chang Jung Christian University, for the support on high-performance liquid chromatography. References [1] Pendeville, H., Carpino, N., Marine, J.C., Takahashi, Y., Muller, M., Martial, J.A. and Cleveland, J.L. (2001) The ornithine decarboxylase gene is essential for cell survival during early murine development. Mol. Cell. Biol. 21, 6549–6558. [2] Lux, G.D., Marton, L.J. and Baylin, S.B. (1980) Ornithine decarboxylase is important in intestinal mucosal maturation and recovery from injury in rats. Science 210, 195–198. [3] Heby, O. (1981) Role of polyamines in the control of cell proliferation and differentiation. Differentiation 19, 1–20. [4] Pignatti, C., Tantini, B., Stefanelli, C. and Flamigni, F. (2004) Signal transduction pathways linking polyamines to apoptosis. Amino Acids 27, 359–365. [5] Casero Jr., R.A. and Marton, L.J. (2007) Targeting polyamine metabolism and function in cancer and other hyperproliferative diseases. Nat. Rev. Drug Discov. 6, 373–390. [6] Gerner, E.W. and Meyskens Jr., F.L. (2004) Polyamines and cancer: old molecules, new understanding. Nat. Rev. Cancer 4, 781–792. [7] Raj, K.P., Zell, J.A., Rock, C.L., McLaren, C.E., Zoumas-Morse, C., Gerner, E.W. and Meyskens, F.L. (2013) Role of dietary polyamines in a phase III clinical trial of difluoromethylornithine (DFMO) and sulindac for prevention of sporadic colorectal adenomas. Br. J. Cancer 108, 512–518. [8] Zhou, P., Cheng, S.W., Yang, R., Wang, B. and Liu, J. (2012) Combination chemoprevention: future direction of colorectal cancer prevention. Eur. J. Cancer Prev. 21, 231–240. [9] Cirenajwis, H., Smiljanic, S., Honeth, G., Hegardt, C., Marton, L.J. and Oredsson, S.M. (2010) Reduction of the putative CD44+CD24 breast cancer stem cell population by targeting the polyamine metabolic pathway with PG11047. Anticancer Drugs 21, 897–906. [10] Malaterre, J., Strambi, C., Aouane, A., Strambi, A., Rougon, G. and Cayre, M. (2004) A novel role for polyamines in adult neurogenesis in rodent brain. Eur. J. Neurosci. 20, 317–330. [11] Zangheri, E.O., Labanca, A.M. and Santana, H. (1996) Role of endogenous polyamines in the proliferation of normal hematopoietic progenitor cells: high-proliferative potential colony-forming cells (HPP-CFC), and lowproliferative potential colony-forming cells (LPP-CFC). Studies ‘‘in vitro’’. Biocell 20, 97–103.
2065
[12] Niskanen, E., Kallio, A., McCann, P.P. and Baker, D.G. (1983) The role of polyamine biosynthesis in hematopoietic precursor cell proliferation in mice. Blood 61, 740–745. [13] Muscari, C., Bonafe, F., Stanic, I., Flamigni, F., Stefanelli, C., Farruggia, G., Guarnieri, C. and Caldarera, C.M. (2005) Polyamine depletion reduces TNFalpha/MG132-induced apoptosis in bone marrow stromal cells. Stem Cells 23, 983–991. [14] Ishii, I., Ikeguchi, Y., Mano, H., Wada, M., Pegg, A.E. and Shirahata, A. (2012) Polyamine metabolism is involved in adipogenesis of 3T3-L1 cells. Amino Acids 42, 619–626. [15] Vuohelainen, S. et al. (2010) Spermidine is indispensable in differentiation of 3T3-L1 fibroblasts to adipocytes. J. Cell Mol. Med. 14, 1683–1692. [16] Muscari, C. et al. (2008) Difluoromethylornithine stimulates early cardiac commitment of mesenchymal stem cells in a model of mixed culture with cardiomyocytes. J. Cell. Biochem. 103, 1046–1052. [17] Uhl, L., Kelly, M. and Schindler, J. (1986) alpha-Difluoromethylornithine induces differentiation of a human embryonal carcinoma cell line in vitro. Biochem. Biophys. Res. Commun. 140, 66–73. [18] Schindler, J., Kelly, M. and McCann, P.P. (1985) The response of several murine embryonal carcinoma cell lines to stimulation of differentiation by alphadifluoromethylornithine. J. Cell. Physiol. 122, 1–6. [19] Oredsson, S.M., Billgren, M. and Heby, O. (1985) Induction of F9 embryonal carcinoma cell differentiation by inhibition of polyamine synthesis. Eur. J. Cell Biol. 38, 335–343. [20] Frostesjo, L. and Heby, O. (1999) Polyamine depletion up-regulates c-Myc expression, yet induces G(1) arrest and terminal differentiation of F9 teratocarcinoma stem cells. J. Cell. Biochem. 76, 143–152. [21] Frostesjo, L., Holm, I., Grahn, B., Page, A.W., Bestor, T.H. and Heby, O. (1997) Interference with DNA methyltransferase activity and genome methylation during F9 teratocarcinoma stem cell differentiation induced by polyamine depletion. J. Biol. Chem. 272, 4359–4366. [22] Bjersing, J.L., Brorsson, A. and Heby, O. (1997) Increased expression of c-jun, but not retinoic acid receptor beta, is associated with F9 teratocarcinoma stem cell differentiation induced by polyamine depletion. J. Cell. Biochem. 67, 378– 385. [23] Nancarrow, M.J., Nesci, A., Hynd, P.I. and Powell, B.C. (1999) Dynamic expression of ornithine decarboxylase in hair growth. Mech. Dev. 84, 161–164. [24] Lee, M.J., Chen, Y., Huang, Y.P., Hsu, Y.C., Chiang, L.H., Chen, T.Y. and Wang, G.J. (2013) Exogenous polyamines promote osteogenic differentiation by reciprocally regulating osteogenic and adipogenic gene expression. J. Cell. Biochem. 114, 2718–2728. [25] Lee, M.J., Huang, C.Y., Sun, Y.J. and Huang, H. (2005) Cloning and characterization of spermidine synthase and its implication in polyamine biosynthesis in Helicobacter pylori strain 26695. Protein Expr. Purif. 43, 140– 148. [26] Yusa, N., Watanabe, K., Yoshida, S., Shirafuji, N., Shimomura, S., Tani, K., Asano, S. and Sato, N. (2000) Transcription factor Sp3 activates the liver/bone/kidneytype alkaline phosphatase promoter in hematopoietic cells. J. Leukoc. Biol. 68, 772–777. [27] Kumar, A.P. and Butler, A.P. (1997) Transcription factor Sp3 antagonizes activation of the ornithine decarboxylase promoter by Sp1. Nucleic Acids Res. 25, 2012–2019. [28] Poulin, R., Soulet, D., Gagnon, B., Rivest, S. and Audette, M. (2006) Evidence for a Multistep Model for Eukaryotic Polyamine Transport in: Polyamine Cell Signaling: Physiology, Pharmacology, and Cancer Research (Wang, J.Y. and Casero, R.A., Eds.), pp. 415–432, Humana Press Inc., Totowa, New Jersey. [29] Oredsson, S.M., Anehus, S. and Heby, O. (1984) Reversal of the growth inhibitory effect of alpha-difluoromethylornithine by putrescine but not by other divalent cations. Mol. Cell. Biochem. 64, 163–172. [30] Tjabringa, G.S., Zandieh-Doulabi, B., Helder, M.N., Knippenberg, M., Wuisman, P.I. and Klein-Nulend, J. (2008) The polymine spermine regulates osteogenic differentiation in adipose stem cells. J. Cell Mol. Med. 12, 1710–1717. [31] Lee, M.J. et al. (2013) PPARgamma silencing enhances osteogenic differentiation of human adipose-derived mesenchymal stem cells. J. Cell Mol. Med. 17, 1188–1193. [32] Fajas, L., Egler, V., Reiter, R., Miard, S., Lefebvre, A.M. and Auwerx, J. (2003) PPARgamma controls cell proliferation and apoptosis in an RB-dependent manner. Oncogene 22, 4186–4193. [33] Linsalata, M., Giannini, R., Notarnicola, M. and Cavallini, A. (2006) Peroxisome proliferator-activated receptor gamma and spermidine/spermine N1acetyltransferase gene expressions are significantly correlated in human colorectal cancer. BMC Cancer 6, 191.
journal homepage: www.FEBSLetters.org
Suppression of ornithine decarboxylase promotes osteogenic differentiation of human bone marrow-derived mesenchymal stem cells Yo-Hsian Tsai a,1, Kuan-Lian Lin a,1, Yuan-Pin Huang b, Yi-Chiang Hsu c, Chung-Hwan Chen d,e, Yuhsin Chen f, Min-Hua Sie a, Gwo-Jaw Wang d,e,g,h, Mon-Juan Lee a,i,⇑ a
Department of Bioscience Technology, Chang Jung Christian University, Tainan, Taiwan Department of Cosmetics and Fashion Styling, Cheng Shiu University, Kaohsiung, Taiwan Taiwan Graduate Institute of Medical Sciences, Chang Jung Christian University, Tainan, Taiwan d Orthopaedic Research Center, Kaohsiung Medical University, Kaohsiung, Taiwan e Department of Orthopaedics, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan f Taichung District Agricultural Research and Extension Station, Council of Agriculture, Executive Yuan, Taichung, Taiwan g Department of Orthopaedics, National Cheng Kung University Hospital, Tainan, Taiwan h Department of Orthopedic Surgery, University of Virginia, VA, USA i Innovative Research Center of Medicine, Chang Jung Christian University, Tainan, Taiwan b c
a r t i c l e
i n f o
Article history: Received 8 February 2015 Revised 8 June 2015 Accepted 15 June 2015 Available online 30 June 2015 Edited by Quan Chen Keywords: Polyamine a-Difluoromethylornithine (DFMO) Osteogenic differentiation Small interfering RNA (siRNA) Mesenchymal stem cell
a b s t r a c t Ornithine decarboxylase (ODC) is the rate-limiting enzyme for polyamine biosynthesis. Suppression of ODC by its irreversible inhibitor, a-difluoromethylornithine (DFMO), or by RNA interference through siRNA, enhanced osteogenic gene expression and alkaline phosphatase activity, and accelerated matrix mineralization of human bone marrow-derived mesenchymal stem cells (hBMSCs). Besides, adipogenic gene expression and lipid accumulation was attenuated, indicating that the enhanced osteogenesis was accompanied by down-regulation of adipogenesis when ODC was suppressed. A decrease in the intracellular polyamine content of hBMSCs during osteogenic induction was observed, suggesting that the level of endogenous polyamines is regulated during differentiation of hBMSCs. This study elucidates the role of polyamine metabolism in the lineage commitment of stem cells and provides a potential new indication for DFMO as bone-stimulating drug. Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction Ornithine decarboxylase (ODC) is the rate-limiting enzyme for the biosynthetic pathway of polyamines, a series of low molecular-weight polycations involved in various cellular processes such as proliferation, differentiation, and apoptosis [1–4]. Because the level of polyamines and the activities of polyamine biosynthetic enzymes are increased in tumor cells, ODC is one of the potential targets for the development of anticancer drugs [5,6]. Alpha-difluoromethylornithine (DFMO), which is the
Author contributions: M.J.L. conceived and supervised the study; Y.H.T., K.L.L. and M.H.S. carried out the experiments; Y.P.H., Y.C.H. and M.J.L. designed experiments; C.H.C., Y.C. and G.J.W. performed data analysis and manuscript revision; M.J.L. drafted and finalized the manuscript. ⇑ Corresponding author at: Department of Bioscience Technology, Chang Jung Christian University, No. 1, Changda Rd., Gueiren District, Tainan City 71101, Taiwan. Fax: +886 6 2785010. E-mail address: [email protected] (M.-J. Lee). 1 The authors contributed equally to this work.
irreversible inhibitor of ODC, suppresses cancer development in animal models, and is being evaluated in clinical trials for the preventative treatment of colorectal cancer [7,8]. In addition to cancer cells, ODC and polyamine biosynthesis are also important in the proliferation of several types of precursor cells and stem cells. Recent studies indicate that the polyamine metabolic pathway serves as a potential target for eliminating breast cancer stem cells [9]. Earlier studies suggest that ODC is crucial to the proliferation of neural progenitor cells and hematopoietic precursor cells [10–12]. In rat bone-marrow stromal cells, polyamine depletion by DFMO is suggested to reduce TNFa/MG132-induced apoptosis and attenuate cell proliferation [13]. Compared to the established role of ODC in promoting cell proliferation, its function in cell differentiation is diverse and has not been fully explored. Inhibition of ODC by DFMO prevents adipocyte differentiation of 3T3-L1 cells [14,15]. However, DFMO stimulates early cardiac commitment of mesenchymal stem cells [16], as well as the differentiation of embryonal carcinoma cells
http://dx.doi.org/10.1016/j.febslet.2015.06.023 0014-5793/Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
[17–19] and F9 teratocarcinoma stem cells [20–22]. Besides, ODC is active during hair follicle development and hair growth [23]. In our previous report, the mRNA level of ODC was reduced in the presence of exogenous polyamines such as putrescine, spermidine and spermine, which enhance the osteogenic differentiation of human bone marrow-derived stem cells (hBMSCs) [24]. It is therefore hypothesized that suppression of ODC and thus depletion of endogenous polyamines may be correlated with the osteogenic induction of hBMSCs. In this study, hBMSCs were treated with DFMO as well as small interfering RNAs (siRNAs) designed against ODC (siODC) to determine whether suppression of the activity and mRNA expression of ODC alter the differentiation fate of hBMSCs. These results may help to investigate the crosstalk between pathways of osteogenesis and polyamine metabolism, and determine whether the level of intracellular polyamines may be manipulated to promote osteogenic differentiation.
2. Methods 2.1. Cell culture Human BMSCs (CD105+, CD166+, CD29+, CD44+, CD14 , CD34 and CD45 ) were isolated from human bone marrow withdrawn from the posterior iliac crest of the pelvic bone of normal volunteers, and were provided by Lonza, Basel, Switzerland. The cells were cultured in mesenchymal stem cell medium (Lonza, Basel, Switzerland) supplemented with Invitrogen MSC-Qualified Fetal Bovine Serum (Life Technologies, Carlsbad, CA, USA), according to the manufacturer’s instructions. Human BMSCs between passages 3 and 5 were used in this study. Cells were seeded at 3000 cells/well in 96-well plates for cell viability assay and alkaline phosphatase (ALP) activity assay, at 104 cells/well in 24-well plates for alizarin red S and oil red O staining, at 5 104 cells/well in 6-well plates for total RNA and protein extraction, and at 5000 cells/cm2 in 10-cm dishes for HPLC sample preparation. Alizarin red S and oil red O staining were performed as previously described [24].
2059
days after siRNA transfection, cells were cultured in OIM for another 7 to 14 days. For adipogenic induction, cells were maintained in DMEM for 7 days prior to siRNA transfection, and 48 h after siRNA treatment, were incubated in adipogenesis-inducing conditions for another 7 to 14 days. 2.4. Cell viability assay Cell viability was analyzed by WST-1 assay using Quick Cell Proliferation Colorimetric Assay Kit (BioVision, Milpitas, CA, USA) and quantitated spectrophotometrically by measuring the optical density at 450 nm, with a reference wavelength of 650 nm. 2.5. Alkaline phosphatase activity assay The ALP activity of hBMSCs was determined by Alkaline Phosphatase Fluorimetric Assay Kit (BioVision, Milpitas, CA, USA). The kit utilizes the fluorescent signal of 4-methylumbelliferone (4-MU) produced by cleavage of the non-fluorescent substrate, 4-methylumbelliferyl phosphate disodium salt (MUP), by the ALP in cell lysates. Fluorescence intensity was determined at Ex/Em 360/440 nm, and the amount of 4-MU produced was quantified using a standard curve produced by commercial ALP. ALP activity was normalized to the total amount of protein and described as the concentration of 4-MU (pmol/ml) produced per mg lysate per minute. 2.6. Real-time polymerase chain reaction Total RNA was isolated from hBMSCs by the RNeasy Mini Kit (Qiagen, Germntown, MD, USA), and cDNA synthesis was carried out using the SuperScript III First-Strand Synthesis SuperMix for qRT-PCR (Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. The cDNA was diluted to a final concentration of 1 ng/ll and reacted with gene-specific primer pairs and QuantiFast SYBRÒ Green PCR Kit (Qiagen, Germntown, MD, USA). Target genes were detected and amplified by Applied
2.2. DFMO treatment For osteogenic differentiation studies, hBMSCs were treated with 1, 10 or 100 lM DFMO in osteogenic induction medium (OIM, 10 7 M dexamethasone, 10 mM b-glycerolphosphate and 50 mM L-ascorbate 2-phosphate in DMEM) 2 days after seeding, and were cultured for another 7 or 14 days. For adipogenic differentiation analysis, cells were cultured for 7 days after seeding to reach 100% confluence, which favor adipogenesis, before treated with 1, 10 or 100 lM DFMO prepared in adipogenic induction medium (AIM, DMEM supplemented with 1 lM dexamethasone, 0.5 mM isobutylmethylxanthine (IBMX), 10 lg insulin/ml and 100 lM indomethacin) or adipogenesis maintenance medium (AMM, DMEM containing 10 lg/ml insulin). Cells were cultured alternately in DFMO-supplemented AIM or AMM for another 14 days according to the instructions of Chemicon Mesenchymal Stem Cell Adipogenesis kit (Millipore, Billerica, MA, USA). 2.3. siRNA transfection Dharmacon siGENOME siRNA specific for human ODC (siODC), as well as siGENOME GAPD Control siRNA (siMock) were designed and synthesized by Thermo Scientific (Waltham, MA, USA). Transfection of siRNA was carried out 24 h after hBMSCs were plated, using DharmaFECT 1 transfection reagent (Thermo Scientific, Waltham, MA, USA) by following the manufacturer’s instructions. The final siRNA concentration was 3.1, 6.25 and 12.5 nmol/L. Two
Fig. 1. Effect of DFMO and siODC on the viability of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 7 days. Cell viability was determined by WST-1 assay and expressed as optical density at 450 nm (OD450). Error bars represent standard deviations (n P 3). *P < 0.05, compared to the control (DMEM); #P < 0.05, compared to OIM.
2060
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
Biosystems 7300 Real-Time PCR System and Applied Biosystems Sequence Detection Software V1.2 (Life Technologies, Carlsbad, CA, USA). Primer sequences and conditions for real-time PCR analysis are the same as those reported previously [24].
proteins were produced with WesternBright ECL HRP substrate (Advansta, Menlo Park, CA, USA) and detected with C-DiGitÒ Blot Scanner (LI-COR Biotechnology, Lincoln, NE, USA). Quantification of Western blot signals was performed with LI-COR Image Studio 4.0 software (LI-COR Biotechnology, Lincoln, NE, USA).
2.7. Western blot analysis Human BMSCs were harvested by Pierce RIPA buffer (Thermo Scientific, Waltham, MA, USA) supplemented with 1 mM phenylmethanesulfonyl fluoride (Sigma–Aldrich, St. Louis, MO, USA). The crude extract was homogenized with a Branson S-450D Sonifier (Emerson Industrial Automation, St. Louis, MO, USA), followed by centrifugation at 14 000g and 4 °C for 15 min to remove cell debris. A total of 50–70 lg of protein from each individual sample was subjected to SDS–PAGE with a 4–12% gradient gel. The resolved proteins were transferred to PVDF membrane using the Trans-Blot Turbo Transfer System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The blots were incubated at room temperature for 1 h in 3% bovine serum albumin, and reacted overnight at 4 °C with primary antibodies diluted in TBST buffer (50 mM Tris, 150 mM NaCl and 0.05% Tween 20, pH 7.5). After washing with TBST buffer, the blots were reacted for 45 min with horseradish peroxidase (HRP)-conjugated secondary antibodies diluted in TBST at a ratio of 1:5000 (v:v). Chemiluminescent signals of target
2.8. Analysis of intracellular polyamine content by high-performance liquid chromatography (HPLC) Human BMSCs were treated with 100 lM putrescine, 1 lM spermidine, 1 lM spermine or 100 lM DFMO in OIM for 7 days. The cells were lysed in 0.2 M perchloric acid, followed by addition of 1,8-diaminooctane (DAO) to a final concentration of 10 lM, which serves as the internal standard. The polyamines in the cell lysate were subjected to benzoylation and analyzed by an Agilent Zorbax 300SB-C18 column (4.6 250 mm) in conjunction with an Agilent 1100 Series HPLC System. Benzoylated polyamines were eluted by a 20-min HPLC gradient program starting with methanol–water (60:40, v/v) as reported in our previous study [25]. 2.9. Statistical analysis Statistical significance relative to the control was analyzed using one-way ANOVA followed by post hoc tests. Data set from
Fig. 2. Effect of DFMO and siODC on the expression of ODC and SSAT in hBMSCs during osteogenic and adipogenic induction. Human BMSCs were treated with osteogenic induction medium (OIM) or adipogenic induction medium (AIM), or with various concentrations of DFMO in OIM or AIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM or AIM for 7 days. Relative gene expression of (A and B) ODC and (C and D) SSAT was determined by real-time PCR. Error bars represent standard deviations (n P 3). *P < 0.05 and **P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
at least three independent experiments was used in the analysis. Levels of significance were expressed as significant, P < 0.05, and highly significant, P < 0.01, respectively. 3. Results
2061
osteogenic and adipogenic induction, and treatment with DFMO or siODC suppressed SSAT mRNA expression as well (Fig. 2C and D). The decrease in SSAT expression of hBMSCs during ODC suppression suggests that the enzyme may be regulated to reduce excretion of intracellular polyamines, so as to maintain polyamine homeostasis.
3.1. Effect of DFMO and siODC treatment on cell viability When treated with DFMO or siODC, the viability of hBMSCs was slightly decreased compared to the control and that incubated in OIM, but was similar to that treated with siMock (Fig. 1). The number of viable cells decreased dramatically when hBMSCs were treated with 1 mM DFMO or 25 nM siRNAs (data not shown). 3.2. Effect of DFMO and siODC treatment on the mRNA level of ODC and SSAT The mRNA expression of ODC was suppressed 2 days after siODC transfection (data not shown). After 7 days of osteogenic or adipogenic induction, the mRNA level of ODC remained lower in cells treated with DFMO and siODC, compared to that of the OIM or AIM controls (Fig. 2A and B). On the other hand, the level of spermidine/spermine N1-acetyl transferase (SSAT), the rate-limiting enzyme of polyamine catabolism, increased upon
3.3. Effect of DFMO and siODC treatment on osteogenic gene expression Treatment of hBMSCs with DFMO resulted in up-regulation of osteogenic genes such as Runt-related transcription factor 2 (Runx2), alkaline phosphatase (ALP), osteocalcin and osteopontin (Fig. 3). Suppression of ODC by siODC also enhanced the expression of early-onset osteogenic genes Runx2 and ALP (Fig. 3A and B), but the increase in the mRNA level of osteocalcin and osteopontin was less significant under siODC treatment compared to that in the presence of DFMO (Fig. 3C and D). 3.4. Effect of DFMO and siODC treatment on ALP activity and matrix mineralization ALP activity and matrix mineralization, both of which are markers for osteoblast maturation, were enhanced under DFMO or
Fig. 3. Effect of DFMO and siODC on the osteogenic gene expression of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 7 days. Relative gene expression of (A) Runx2, (B) ALP, (C) osteocalcin and (D) osteopontin was determined by real-time PCR. Error bars represent standard deviations (n P 3). *P < 0.05 and ** P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
2062
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
cultured in OIM or siMock in the presence of 100 lM DFMO or 12.5 nM siODC (Fig. 5B). 3.5. Effect of DFMO and siODC treatment on adipogenic gene expression During adipogenic differentiation, hBMSCs treated with AIM or siMock had a significantly higher mRNA level of adipogenic genes such as peroxisome proliferator-activated receptor gamma (PPARc), adipocyte protein 2 (aP2) and adipsin, compared to the control (Fig. 6). When treated with DFMO or siODC, the expression of adipogenic genes were significantly suppressed, except for that of aP2 in the presence of 3.1 and 6.25 nM siODC. These results indicate that inhibition of the activity or mRNA expression of ODC resulted in suppression of adipogenic differentiation. 3.6. Effect of DFMO and siODC treatment on lipid accumulation
Fig. 4. Effect of DFMO and siODC on the alkaline phosphatase (ALP) activity of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 7 days. Alkaline phosphatase activity was determined by Alkaline Phosphatase Fluorimetric Assay Kit (BioVision, Milpitas, CA, USA) according to the manufacturer’s instructions. Error bars represent standard deviations (n P 3). ** P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
Lipid accumulation, an indication of adipogenic terminal differentiation, was increased in hBMSCs incubated under adipogenic conditions, or treated with siMock, followed by adipogenic induction (Fig. 7). As expected, the amount of lipid droplets accumulated in hBMSCs was significantly reduced in DFMO- and siODC-treated cells. The amount of lipid droplets decreased to around 0.6-fold or less of AIM or siMock in the presence of DFMO or siODC. It is therefore suggested that the osteogenic activity of DFMO and siODC was correlated with suppression of adipogenic terminal differentiation as well as gene expression. 3.7. Effect of DFMO and siODC treatment on the expression of transcription factors
siODC treatment. As shown in Fig. 4, when treated with 10 or 100 lM DFMO, or with 12.5 nM siODC, the ALP activity of hBMSCs was higher than that cultured in OIM alone or treated with siMock. The enhanced ALP activity coincided with the increase in the mRNA level of ALP (Fig. 3B), indicating that suppression of ODC is correlated with both the activity and mRNA expression of ALP. Results from alizarin red S staining also suggest that ODC suppression accelerated matrix mineralization of hBMSCs (Fig. 5A). The amount of cell-bound alizarin red S (lM alizarin red S/lg DNA) was increased to 1.55- or 1.24-fold, respectively, of those
We analyzed the protein expression of Runx2 and PPARc, the key transcription factors related to osteogenic and adipogenic differentiation, respectively, as well as that of Sp3, a transcription factor that antagonizes ODC and activates the liver/bone/kidney-type ALP promoter [26,27]. As shown in Fig. 8A and B, treatment of hBMSCs with DFMO or siODC resulted in a decrease in the protein expression of ODC under both osteogenic and adipogenic conditions. When hBMSCs were cultured in OIM alone, the expression of Runx2 was increased compared to the control, but was similar
Fig. 5. Effect of DFMO and siODC on matrix mineralization of hBMSCs. Human BMSCs were treated with osteogenic induction medium (OIM) or with various concentrations of DFMO in OIM for 14 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM for 14 days. (A) The cells were stained with alizarin red S and observed with phase-contrast microscopy. (B) The concentration of cell-bound alizarin red S was quantitated spectrophotometrically at OD570 and normalized to the amount of DNA (lg). Error bars represent standard deviations (n P 3). **P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to OIM.
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
2063
Fig. 6. Effect of DFMO and siODC on the adipogenic gene expression of hBMSCs. Human BMSCs were cultured at high cell densities that favor adipogenesis, and treated with AIM or with various concentrations of DFMO in AIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by AIM for 7 days. Relative gene expression of (A) PPARc, (B) aP2 and (C) adipsin was determined by real-time PCR. Error bars represent standard deviations (n P 3). **P < 0.01, compared to the control (DMEM); #P < 0.05 and ##P < 0.01, compared to AIM.
Fig. 7. Effect of DFMO and siODC on lipid accumulation of hBMSCs. Human BMSCs were cultured at high cell densities that favor adipogenesis, and treated with DFMO under adipogenic induction conditions as described in Section 2 for 14 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by adipogenic induction for 14 days. (A) Cells were stained with oil red O and observed with phase-contrast microscopy. (B) Cell-bound oil red O was extracted with isopropanol and its intensity was determined colorimetrically at OD540. Error bars represent standard deviations (n P 3). **P < 0.01, compared to the control (DMEM); ##P < 0.01, compared to AIM.
to that of hBMSCs co-treated with DFMO or siODC, suggesting that Runx2 may not be directly related to the osteogenic stimulation resulted from ODC suppression (Fig. 8A). Nevertheless, we found that the protein expression of Sp3 was up-regulated in the presence of DFMO or siODC under osteogenic conditions, but not when hBMSCs was cultured in OIM alone or treated with siMock (Fig. 8A), indicating that suppression of ODC may lead to activation of ALP through the up-regulation of Sp3. On the other hand, the expression of PPARc was suppressed in the presence of DFMO or siODC and induced to differentiate adipogenically (Fig. 8B), consistent with the results of PPARc mRNA expression in Fig. 6A. 3.8. Effect of exogenous polyamines and DFMO on intracellular polyamine content Treatment with OIM alone resulted in a decrease in the concentration of all three of the common polyamines, including putrescine, spermidine, and spermine, suggesting that endogenous polyamines may be associated with the regulation of stem cell differentiation (Table 1). Compared to OIM, the total concentration of intracellular polyamines was slightly increased in the presence of 100 lM putrescine. While 1 lM spermidine did not affect total
polyamines, treatment with 1 lM spermine and 100 lM DFMO decreased intracellular polyamines to levels lower than that of OIM alone. It is known that exogenous polyamines may act through pathways distinct from their endogenous counterparts [28], but whether they were sequestered away in other pathways or were simply exported from cells remains to be investigated.
4. Discussion Muscari et al. reported that matrix mineralization of rat BMSCs was unaffected when incubated with 1 mM DFMO for 48 h, followed by induction to differentiate in osteogenic medium for 3 weeks, indicating that polyamine depletion by DFMO does not alter the differentiation potential of rat BMSCs [13]. However, we found that the viability of hBMSCs reduced significantly after treated with 1 mM DFMO in OIM for 7 days. Besides, when maintained for 14 days in OIM supplemented with a lower concentration of DFMO (100 lM), matrix mineralization of hBMSCs was enhanced (Fig. 5). These results suggest that continuous suppression of ODC during osteogenic induction may be critical in promoting osteogenic terminal differentiation.
2064
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
Fig. 8. Effect of DFMO and siODC on protein expression. Human BMSCs were treated with osteogenic induction medium (OIM) or adipogenic induction medium (AIM), or with various concentrations of DFMO in OIM or AIM for 7 days. For siODC treatment, hBMSCs were cultured in various concentrations of siODC for 2 days, followed by OIM or AIM for 7 days. Cells were harvested in RIPA buffer and subjected to Western blot analysis as described in Section 2. (A) Protein expression of ODC, Runx2, and Sp3 in hBMSCs under osteogenic induction. (B) Protein expression of ODC and PPARc in hBMSCs under adipogenic induction. The ratios of Runx 2 and PPARc were determined by normalizing the intensity of each band, determined by LI-COR Image Studio 4.0 software, to that of the control.
Table 1 Intracellular polyamine contents of hBMSCs treated with exogenous polyamines or DFMO in combination with OIM for 7 days. Polyamines (nmol/mg)
Control
OIM
100 lM PUT/OIM
1 lM SPD/OIM
1 lM SPM/OIM
100 lM DFMO/OIM
Putrescine Spermidine Spermine Total polyamines
43.91 ± 2.54 16.55 ± 3.41 16.56 ± 2.35 77.01 ± 4.86
9.08 ± 3.13 3.18 ± 0.36 8.30 ± 2.48 20.56 ± 4.01
16.28 ± 0.38 5.21 ± 1.68 5.13 ± 1.95 26.62 ± 2.60
4.42 ± 0.54 1.13 ± 0.24 15.05 ± 3.80 20.61 ± 3.85
3.21 ± 1.89 4.29 ± 1.51 5.50 ± 0.03 13.01 ± 2.42
3.68 ± 1.01 2.43 ± 0.26 4.26 ± 0.37 10.36 ± 1.11
Although DFMO is known to deplete the intracellular pool of polyamines through the irreversible inhibition of ODC, a strategy to block cancer cell proliferation [29], the effect of SSAT to counteract the effect of ODC suppression was seldom discussed, especially when in this study, hBMSCs were treated with DFMO at concentrations that did not affect cell viability (Fig. 1). By analyzing both the catabolic and anabolic enzyme expression of polyamine biosynthesis, it is suggested that in the presence of exogenous polyamines, ODC was suppressed and SSAT was stimulated to prevent rapid increase in intracellular polyamines [24]. Conversely, SSAT was suppressed when hBMSCs were treated with DFMO or siODC possibly to prevent further depletion of the polyamine pool (Fig. 2C). On the other hand, OIM had no effect on the mRNA expression of ODC (Fig. 2A), but the expression of SSAT was increased (Fig. 2C), which may explain the decrease in total polyamines (Table 1). Similar increase in SSAT mRNA level was also observed in goat-derived adipose stem cells induced to differentiate
osteogenically with spermine and 1,25-dihydroxyvitamin-D3, but their effect on ODC was not reported [30]. Studies on murine embryonal carcinoma cell lines demonstrate that the level of intracellular polyamines must drop below certain threshold to induce differentiation [18]. Indeed, total intracellular polyamines were significantly lowered when hBMSCs were cultured under osteogenic conditions (Table 1). However, co-treatment of hBMSCs with OIM and exogenous polyamines or DFMO had limited effect on total polyamines compared to OIM alone, suggesting that in addition to disrupting polyamine metabolism, exogenous polyamines and DFMO may act through additional pathways to enhance osteogenic gene expression and terminal differentiation. Our results from Western blot analysis indicate that up-regulation of Sp3 by ODC suppression, together with the activation of Runx2 by osteogenic inducing agents in OIM (Fig. 8A), enhanced ALP expression (Fig. 3B) and accelerated the osteogenic differentiation of hBMSCs. However, the mechanism of how ODC
Y.-H. Tsai et al. / FEBS Letters 589 (2015) 2058–2065
inhibition led to suppression of PPARc remains to be investigated. PPARc not only is the master regulator of adipocyte differentiation [31], but also plays a role in the control of cell proliferation and apoptosis, and is known to activate SSAT transcription [6,32]. The mRNA levels of PPARc, ODC and SSAT, as well as polyamine levels were significantly higher in human colorectal carcinoma than in normal tissue [33]. In hBMSCs, we demonstrate that a decrease in the expression of PPARc, ODC and SSAT, accompanied by reduced intracellular polyamines, was responsible for enhancing osteogenesis. The fact that both ODC and PPARc are associated with cell proliferation implicates that they may share common effectors in the differentiation and proliferation of stem cells. Conflict of interest statement The authors confirm that there is no conflict of interest. Acknowledgments This study is supported by the Ministry of Science and Technology, Taiwan (NSC101-2314-B-309-001-MY3), the Academic Research Funds of Chang Jung Christian University, Tainan, Taiwan, and Taichung District Agricultural Research and Extension Station, Council of Agriculture, Executive Yuan, Taichung, Taiwan. We thank the Toxicology Research Center, Chang Jung Christian University, for the support on high-performance liquid chromatography. References [1] Pendeville, H., Carpino, N., Marine, J.C., Takahashi, Y., Muller, M., Martial, J.A. and Cleveland, J.L. (2001) The ornithine decarboxylase gene is essential for cell survival during early murine development. Mol. Cell. Biol. 21, 6549–6558. [2] Lux, G.D., Marton, L.J. and Baylin, S.B. (1980) Ornithine decarboxylase is important in intestinal mucosal maturation and recovery from injury in rats. Science 210, 195–198. [3] Heby, O. (1981) Role of polyamines in the control of cell proliferation and differentiation. Differentiation 19, 1–20. [4] Pignatti, C., Tantini, B., Stefanelli, C. and Flamigni, F. (2004) Signal transduction pathways linking polyamines to apoptosis. Amino Acids 27, 359–365. [5] Casero Jr., R.A. and Marton, L.J. (2007) Targeting polyamine metabolism and function in cancer and other hyperproliferative diseases. Nat. Rev. Drug Discov. 6, 373–390. [6] Gerner, E.W. and Meyskens Jr., F.L. (2004) Polyamines and cancer: old molecules, new understanding. Nat. Rev. Cancer 4, 781–792. [7] Raj, K.P., Zell, J.A., Rock, C.L., McLaren, C.E., Zoumas-Morse, C., Gerner, E.W. and Meyskens, F.L. (2013) Role of dietary polyamines in a phase III clinical trial of difluoromethylornithine (DFMO) and sulindac for prevention of sporadic colorectal adenomas. Br. J. Cancer 108, 512–518. [8] Zhou, P., Cheng, S.W., Yang, R., Wang, B. and Liu, J. (2012) Combination chemoprevention: future direction of colorectal cancer prevention. Eur. J. Cancer Prev. 21, 231–240. [9] Cirenajwis, H., Smiljanic, S., Honeth, G., Hegardt, C., Marton, L.J. and Oredsson, S.M. (2010) Reduction of the putative CD44+CD24 breast cancer stem cell population by targeting the polyamine metabolic pathway with PG11047. Anticancer Drugs 21, 897–906. [10] Malaterre, J., Strambi, C., Aouane, A., Strambi, A., Rougon, G. and Cayre, M. (2004) A novel role for polyamines in adult neurogenesis in rodent brain. Eur. J. Neurosci. 20, 317–330. [11] Zangheri, E.O., Labanca, A.M. and Santana, H. (1996) Role of endogenous polyamines in the proliferation of normal hematopoietic progenitor cells: high-proliferative potential colony-forming cells (HPP-CFC), and lowproliferative potential colony-forming cells (LPP-CFC). Studies ‘‘in vitro’’. Biocell 20, 97–103.
2065
[12] Niskanen, E., Kallio, A., McCann, P.P. and Baker, D.G. (1983) The role of polyamine biosynthesis in hematopoietic precursor cell proliferation in mice. Blood 61, 740–745. [13] Muscari, C., Bonafe, F., Stanic, I., Flamigni, F., Stefanelli, C., Farruggia, G., Guarnieri, C. and Caldarera, C.M. (2005) Polyamine depletion reduces TNFalpha/MG132-induced apoptosis in bone marrow stromal cells. Stem Cells 23, 983–991. [14] Ishii, I., Ikeguchi, Y., Mano, H., Wada, M., Pegg, A.E. and Shirahata, A. (2012) Polyamine metabolism is involved in adipogenesis of 3T3-L1 cells. Amino Acids 42, 619–626. [15] Vuohelainen, S. et al. (2010) Spermidine is indispensable in differentiation of 3T3-L1 fibroblasts to adipocytes. J. Cell Mol. Med. 14, 1683–1692. [16] Muscari, C. et al. (2008) Difluoromethylornithine stimulates early cardiac commitment of mesenchymal stem cells in a model of mixed culture with cardiomyocytes. J. Cell. Biochem. 103, 1046–1052. [17] Uhl, L., Kelly, M. and Schindler, J. (1986) alpha-Difluoromethylornithine induces differentiation of a human embryonal carcinoma cell line in vitro. Biochem. Biophys. Res. Commun. 140, 66–73. [18] Schindler, J., Kelly, M. and McCann, P.P. (1985) The response of several murine embryonal carcinoma cell lines to stimulation of differentiation by alphadifluoromethylornithine. J. Cell. Physiol. 122, 1–6. [19] Oredsson, S.M., Billgren, M. and Heby, O. (1985) Induction of F9 embryonal carcinoma cell differentiation by inhibition of polyamine synthesis. Eur. J. Cell Biol. 38, 335–343. [20] Frostesjo, L. and Heby, O. (1999) Polyamine depletion up-regulates c-Myc expression, yet induces G(1) arrest and terminal differentiation of F9 teratocarcinoma stem cells. J. Cell. Biochem. 76, 143–152. [21] Frostesjo, L., Holm, I., Grahn, B., Page, A.W., Bestor, T.H. and Heby, O. (1997) Interference with DNA methyltransferase activity and genome methylation during F9 teratocarcinoma stem cell differentiation induced by polyamine depletion. J. Biol. Chem. 272, 4359–4366. [22] Bjersing, J.L., Brorsson, A. and Heby, O. (1997) Increased expression of c-jun, but not retinoic acid receptor beta, is associated with F9 teratocarcinoma stem cell differentiation induced by polyamine depletion. J. Cell. Biochem. 67, 378– 385. [23] Nancarrow, M.J., Nesci, A., Hynd, P.I. and Powell, B.C. (1999) Dynamic expression of ornithine decarboxylase in hair growth. Mech. Dev. 84, 161–164. [24] Lee, M.J., Chen, Y., Huang, Y.P., Hsu, Y.C., Chiang, L.H., Chen, T.Y. and Wang, G.J. (2013) Exogenous polyamines promote osteogenic differentiation by reciprocally regulating osteogenic and adipogenic gene expression. J. Cell. Biochem. 114, 2718–2728. [25] Lee, M.J., Huang, C.Y., Sun, Y.J. and Huang, H. (2005) Cloning and characterization of spermidine synthase and its implication in polyamine biosynthesis in Helicobacter pylori strain 26695. Protein Expr. Purif. 43, 140– 148. [26] Yusa, N., Watanabe, K., Yoshida, S., Shirafuji, N., Shimomura, S., Tani, K., Asano, S. and Sato, N. (2000) Transcription factor Sp3 activates the liver/bone/kidneytype alkaline phosphatase promoter in hematopoietic cells. J. Leukoc. Biol. 68, 772–777. [27] Kumar, A.P. and Butler, A.P. (1997) Transcription factor Sp3 antagonizes activation of the ornithine decarboxylase promoter by Sp1. Nucleic Acids Res. 25, 2012–2019. [28] Poulin, R., Soulet, D., Gagnon, B., Rivest, S. and Audette, M. (2006) Evidence for a Multistep Model for Eukaryotic Polyamine Transport in: Polyamine Cell Signaling: Physiology, Pharmacology, and Cancer Research (Wang, J.Y. and Casero, R.A., Eds.), pp. 415–432, Humana Press Inc., Totowa, New Jersey. [29] Oredsson, S.M., Anehus, S. and Heby, O. (1984) Reversal of the growth inhibitory effect of alpha-difluoromethylornithine by putrescine but not by other divalent cations. Mol. Cell. Biochem. 64, 163–172. [30] Tjabringa, G.S., Zandieh-Doulabi, B., Helder, M.N., Knippenberg, M., Wuisman, P.I. and Klein-Nulend, J. (2008) The polymine spermine regulates osteogenic differentiation in adipose stem cells. J. Cell Mol. Med. 12, 1710–1717. [31] Lee, M.J. et al. (2013) PPARgamma silencing enhances osteogenic differentiation of human adipose-derived mesenchymal stem cells. J. Cell Mol. Med. 17, 1188–1193. [32] Fajas, L., Egler, V., Reiter, R., Miard, S., Lefebvre, A.M. and Auwerx, J. (2003) PPARgamma controls cell proliferation and apoptosis in an RB-dependent manner. Oncogene 22, 4186–4193. [33] Linsalata, M., Giannini, R., Notarnicola, M. and Cavallini, A. (2006) Peroxisome proliferator-activated receptor gamma and spermidine/spermine N1acetyltransferase gene expressions are significantly correlated in human colorectal cancer. BMC Cancer 6, 191.
