Ann Hematol DOI 10.1007/s00277-014-2209-7
ORIGINAL ARTICLE
DNA topoisomerase IIβ as a molecular switch in neural differentiation of mesenchymal stem cells Sevim Isik & Merve Zaim & Mehmet Taha Yildiz & Yesim Negis & Tuba Kunduraci & Nihal Karakas & Gulsum Arikan & Guven Cetin
Received: 25 March 2014 / Accepted: 28 August 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Two isoforms of DNA topoisomerase II (topo II) have been identified in mammalian cells, named topo IIα and topo IIβ. Topo IIα plays an essential role in segregation of daughter chromosomes and thus for cell proliferation in mammalian cells. Unlike its isozyme topo IIα, topo IIβ is greatly expressed upon terminal differentiation of neuronal cells. Although there have been accumulating evidence about the crucial role of topo IIβ in neural development through activation or repression of developmentally regulated genes at late stages of neuronal differentiation, there have been no reports that analyzed the roles of topo IIβ in the neural trans differentiation process of multipotent stem cells. Terminal differentiation of neurons and transdifferentiation of Mesenchymal Stem Cells (MSCs) are two distinct processes. Therefore, the functional significance of topo IIβ may also be different in these differentiation systems. MSC transdifferentiation into neuron-like cells represents an useful model to further validate the role of topo IIβ in neuronal differentiation. The aim of this Electronic supplementary material The online version of this article (doi:10.1007/s00277-014-2209-7) contains supplementary material, which is available to authorized users. S. Isik Department of Medical Biology, Faculty of Medicine, Fatih University, Buyukcekmece, Istanbul 34500, Turkey S. Isik (*) : M. Zaim : M. T. Yildiz : Y. Negis : T. Kunduraci : N. Karakas : G. Arikan Department of Biology, Faculty of Arts and Science, Fatih University, Buyukcekmece, Istanbul 34500, Turkey e-mail: [email protected] Y. Negis Department of Medical Biochemistry, Faculty of Medicine, Bahcesehir University, Istanbul, Turkey G. Cetin Division of Hematology, Department of Internal Medicine, Medical Faculty of Bezmialem Vakıf University, Istanbul, Turkey
study is to evaluate the subset of genes that are regulated in neural transdifferentiation of bone marrow-derived human MSCs (BM-hMSCs) in vitro and find genes related with topo IIβ. For this purpose, topo IIβ was silenced by specific small interfering RNAs in hMSCs and cells were induced to differentiate into neuron-like cells. Differentiation and silencing of topo IIβ were monitored by real-time cell analysis and also expressions of topo II isoforms were analyzed. Change in transcription patterns of genes upon topo IIβ silencing was identified by DNA microarray analysis, and apparently genes involved in regulation of several ion channels and transporters, vesicle function, and cell calcium metabolism were particularly affected by topo IIβ silencing suggesting that topoIIβ silencing can significantly alter the gene expression pattern of genes involved in variety of biological processes and signal transduction pathways including transcription, translation, cell trafficking, vesicle function, transport, cell morphology, neuron guidance, growth, polarity, and axonal growth. It appears that the deregulation of these pathways may contribute to clarify the further role of topo IIβ in neural differentiation. Keywords DNA topoisomerase IIβ . siRNA transfection . Human mesenchymal stem cells . Neural differentiation . Microarray analysis
Introduction DNA topoisomerase II (topo II) is a nuclear enzyme which catalyzes the transport of double helix through another, thus alters DNA topology. Several genetic processes such as DNA replication, transcription, recombination, as well as chromosome compaction and segregation depend on its activity [29]. There are two isoforms of topo II in mammalian cells named topoisomerase IIα (topo IIα) and topoisomerase IIβ (topo IIβ). Topo IIα plays a key role in mitotic processes and therefore is
Ann Hematol
only present in proliferating tissues. Unlike topo IIα, topo IIβ is present in all tissues and is abundantly expressed in postmitotic neuronal cells in developing brain [30, 32, 18]. Especially whole body and brain-specific topo IIβ knockout mice studies revealed crucial significance of topo IIβ both in brain development and neural differentiation [32, 18]. Additionally, several studies exist emphasizing a possible link between topo IIβ and neural development/differentiation processes in granule neurons of developing rat cerebellum [27, 28, 9]. Topo IIβ affects neural development through activation or repression of developmentally regulated genes associated with neuronal maturation such as ion channels, receptors, and signal transduction molecules [27, 28]. Transcriptional activation of the genes is dependent on the enzymatic activity of topo IIβ probably through opening of chromatin structure and thereby leading to association of the DNA transcription factors to the promoter regions [6]. As being pluripotent adult stem cells, MSCs are relevant system to study molecular details of neural differentiation because of the multilineage differentiation potential of the cells. Supportingly, studies show that MSCs can transdifferentiate into cardiomyocytes and even into cells of non-mesodermal origin as well as neurons [2]. The capacity of MSCs originated from rat, mouse, and human to differentiate into neurons has been shown for the first time by Sanchez-Ramos et al. and Woodbury et al. using growth factors and chemicals, respectively [24, 31]. Later on, many other synthetic and biological neural inducers have been tested for their abilities of differentiating MSCs into neurons including retinoic acid, cytokines, growth factors, neurotrophins, antioxidants, demethylating agents, forskolin, noggin 3-isobutyl-1methylxanthine (IBMX), dibutyryl cyclic adenosine monophosphate (dbcAMP), and dimethylsulfoxide (DMSO) [15, 16]. On the other hand, in vitro neural differentiation process of MSCs is considered to be substantially different from the terminal differentiation of post-mitotic neurons in developing brain. Therefore, the functional significance of topo IIβ may also be different in these differentiation systems. In the current study, we differentiated bone marrowderived human MSCs into neural-like cells by a cytokine combination. To investigate topo IIβ’s role on gaining neural-like morphology during neural differentiation and interaction of relevant gene sets on this differentiation process, topo IIβ was selectively silenced.
Materials and methods Ethical approval Bone marrow (BM) samples used in this work were collected after informed consent; all procedures were approved by the
Ethics Committee of Fatih University, Medical School, and the study conformed with the 2013 WMA Declaration of Helsinki. Isolation of hMSCs from BM Human BM aspirates were obtained from the (posterior) iliac crest of healthy donors. BM aspirates were provided by Bezmialem University, Medical School, Department of Hematology. Ficoll density gradient centrifugation method was used to isolate hMSCs from BM. Isolated cells were seeded at a density of 1.5–2×105 cells/cm2 into a tissue culture flask (BD Falcon; BD Biosciences, San Jose, CA, USA) including DMEM-LG (Gibco, Grand Island, NY, USA), 20 % MSCqualified Fetal Bovine Serum (MSC-FBS, Gibco), and 0.1 mg/ml primocin (InvivoGen, San Diego, CA, USA), and incubated in 37 °C, 5 % CO2 incubator. Three days after isolation, non-adherent hematopoietic cells were removed via medium refreshment. Culture of hMSCs Isolated cells were seeded into culture flasks, and attached cells started to form colonies after 5–6 days. When cells became 80–90 % confluent, they were detached with 0.25 % Trypsin/EDTA (Gibco) solution, seeded into culture flasks at a density of 1.5×103 cells/cm2 with DMEM-LG containing 10 % MSC-FBS, and incubated at 37 °C, 5 % CO2 incubator. Subculturing of cells was performed at about 5–6 days intervals and medium was refreshed once in between two passages. hMSCs at passage 3 were screened by flow cytometry and used for neural differentiation and siRNA transfection experiments. Immunophenotyping of hMSCs The phenotypic characteristics of hMSCs were confirmed at passage 3, by flow cytometry analysis using antibodies against CD45, anti-HLA-DR, CD90, CD34, CD73, and CD105. Mesodermal differentiation of hMSCs Isolated cells from BM were differentiated into mesodermal (adipogenic, chondrogenic, osteogenic) lineages in order to qualify them as MSCs. Briefly, osteogenic differentiation was assessed by incubating 2 × 10 5 cells/cm 2 in Complete MesenCult Osteogenic Medium including MesenCult MSC Basal Medium, osteogenic stimulatory supplement, β-glycerophosphate, dexamethasone, and ascorbic acid (all from Stemcell) for 5 weeks. Adipogenic differentiation was assessed by incubating 5× 103 cells/cm2 in Complete MesenCult Adipogenic Medium containing MesenCult MSC Basal Medium and 10 % adipogenic stimulatory supplement for 3 weeks.
Ann Hematol
Chondrogenic differentiation was assessed by incubating 1.5× 107 cells/ml for 3 weeks with Stempro chondrocyte differentiation basal medium (Gibco) containing 10 % Stempro chondrogenesis supplement (Gibco). Cell morphologies were observed under light microscopy and histochemical staining was performed. Osteogenic, adipogenic, and chondrogenic differentiation was confirmed by Toluidine Blue, Oil Red O, and Alcian Blue staining, respectively. Transfection of hMSCs with Lipofectamine RNAiMAX reagent The day before transfection, cells were seeded into six-well plates at a density of 1.5×103 cells/cm2. Two different validated siRNAs (TOP2B_5 and TOP2B_6 from Qiagen) were used for silencing topo IIβ. Sequences of TOP2B_5 and TOP2B_6 were TCGGGCTAGGAAAGAAGTAAA and CAGCCGAAAGACCTAAATACA, respectively. siRNAs were diluted with Opti-MEM in 1:100 and Lipofectamine RNAiMAX was diluted with Opti-MEM in 1:50 ratio. Diluted siRNAs (50, 25, and 8 nM topo IIβ-specific siRNAs) and diluted reagent were mixed and incubated for 15 min at room temperature (RT) to allow the siRNA-Lipofectamine RNAiMAX complexes to form. siRNA-Lipofectamine RNAiMAX complexes were added into the wells and incubated at 37 °C in CO2 incubator for 24–48 h. After 24 h of incubation, the medium containing complex was replaced by neural induction medium. Since transfection was performed during neural differentiation, medium was replaced with neural induction medium. Efficiency of transfection was checked by real-time cell analysis (RTCA) and reverse transcription PCR (RT-PCR). Neural differentiation of hMSCs hMSCs were divided into three different experimental groups: Group 1/undifferentiated hMSCs (control) Cells were cultured in expansion medium, DMEM-LG containing 10 % MSC-FBS. Medium refreshment was done every 48 h. Group 2/neural differentiated hMSCs (mN3) hMSCs were precultured for 48 h in expansion medium. Then, culture medium was replaced with neural induction medium called modified N3 (mN3). N3 neural induction medium as previously described [14] includes 0.5 mM IBMX (Sigma), 20 ng/ml human epidermal growth factor (hEGF; Sigma), 40 ng/ml recombinant human basic fibroblast growth factor (bFGF; R&D Systems, Minneapolis, MN, USA), 10 ng/ml fibroblast growth factor-8 (FGF-8; PeproTech, Rocky Hill, NJ, USA), 10 ng/ml recombinant human brain-derived neurotrophic factor (BDNF; R&D Systems), in Neurobasal Medium (Gibco) supplemented with 2 % B27 Supplement
(Gibco). mN3 cytokine combination was obtained by adding 0.25 mg/ml dbcAMP (Sigma) instead of 1 mM dbcAMP, 2 mM L-glutamine (Gibco), and 40 ng/ml nerve growth factor (NGF; Gibco) into N3. Neural induction medium was refreshed every 48 h. Group 3/neural differentiated + topo IIβ silenced hMSCs by siRNA transfection (mN3 + Tr) Topo IIβ silencing (siRNA transfection) was performed 24 h before neural induction. During medium refreshment, transfection was also performed every 48 h. After 12 days, cell morphologies of all experimental groups were observed under light microscopy and brain-specific marker expressions were monitored at mRNA level with RT-PCR. Because of complicated terminology used for experimental groups in the study, we provide an explanatory table to avoid possible confusions (Table 1).
Reverse transcription PCR siRNA silencing efficiency and expression of neural markers, actin, topo IIα, and topo IIβ, were checked at mRNA level using RT-PCR technique. Total RNA was extracted from hMSCs using RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions. Concentration of extracted RNA was quantified using Qubit (Invitrogen, Carlsbad, CA, USA). cDNA was generated by using a Quantitect reverse transcription kit (Qiagen). Actin, topo IIα, and topo IIβ underwent 30, 35, and 35 cycles of amplification at 60, 54, and 60 °C annealing temperatures, respectively (PCR core kit, Qiagen). Forward and reverse primers were CGCACCACTG GCATTGTCAT and GTGGCCATCTCCTGCTCGAA, yielding a 208 bp for actin; ACCATTGCAGCCTGTA and GCTC TTCCCATATTATCC yielding a 596 bp for topo IIα; ATCA AAAGCCACTC-CAGAAAAATC and AGAAGGTGGCTC AGTAGGGAAGTC yielding a 508 bp for topo IIβ. Actin was used as the housekeeping gene. The PCR products were loaded on 2 % agarose gel and were visualized with a gel documentation system.
Table 1 Abbreviations of experimental groups in the study
Group 1 Group 2 Group 3
Experimental groups
Abbreviation used
Undifferentiated hMSCs Neural differentiated hMSCs (mN3 treated hMSCS) Neural differentiated + topo IIβ silenced hMSCs by siRNA transfection
Control mN3 mN3+Tr
Ann Hematol
Immunofluorescent staining
Cell death analysis
hMSCs were seeded into 24-well plates at a density of 1.5×103 cells/cm2 and incubated in neural induction medium for 12 days. Then, cells were permeabilized with TZN buffer (10 mM pH 7.5 Tris–HCl, 0.5 % Nonidet P-40, 0.2 mM ZnCl2) and fixed with 4 % PFA/PBS at RT. Cells were blocked in 0.3 % Triton X/PBS (PBS-Tx) solution containing 10 % goat and 10 % horse serum (Biochrom) for 30 min at RT (Gibco) Then, they were incubated with specific primary antibodies, both diluted in PBS-Tx with 3 % NHS for 2 h at RT. After primary antibody incubation, cells were treated with Alexa Fluorlabeled anti-mouse or anti-rabbit secondary antibodies at RT for 1 h. For nuclear staining, cells were treated with 1:15,000× DAPI (Sigma) solution and incubated at RT for 10 min. Finally, glass coverslips were mounted onto the microscope slides covered with Prolong Antifade Kit (Molecular Probes) or Prolong Gold Antifade Reagent (Invitrogen). Cells were visualized under fluorescent microscope (Nikon Ti, Japan). Primary antibodies against neurofilament (NF; 1:100; Chemicon Temecula, CA), microtubule-associated protein 2 (MAP2; 1:50; Cell Signaling), neuron-specific enolase (NSE; 1:100; Chemicon), neuronal nuclei (NeuN; 1:100; Chemicon Temecula, CA), oligodendrocyte transcription factor 2 (Olig2; 1:100; Chemicon Temecula, CA), glial fibrillary acidic protein (GFAP; 1:100, Chemicon Temecula, CA), and as secondary antibodies GAM-IgG-Alexa Fluor 488 (1:100, Invitrogen) and GAR-IgGAlexa Fluor 594 (1:100, Invitrogen) were used.
hMSCs were seeded into 96-well plates at a density of 1.5× 103 cells/cm2. Twenty-four hours later, topo IIβ was silenced by siRNA treatment as described above. And after 24 h, these cells were treated with mN3 cytokine combination to induce neural differentiation. Induction and transfection media were refreshed every 48 h. Annexin V and Sytox Green staining were performed at 3rd, 7th, and 12th days of neural induction. As positive control of apoptosis, cells were treated with camptothecin (Sigma). Culture medium was replaced with 100 μl of Annexin V-Alexa 568 labeling solution (Roche) and 50 μM Sytox Green dye (Invitrogen) was added to each well to differentiate apoptotic cells from necrotic cells. After 15 min incubation at 25 °C, cells were washed with PBS and treated with 1:15,000 DAPI (Sigma) for 10 min at RT to stain nuclei. Cells were then washed with PBS and dH2O and observed under fluorescence microscopy (Nikon, Eclipse Tí).
Microarray analysis Microarray analysis was done by Febit group (febit GmbH, Heidelberg) for whole set of human transcriptome comprising 22.440 genes. Real-time cell analysis The real-time cell analyzer (RTCA; xCELLigence, Roche) monitors cellular events in real time by the measurement of electrical impedance across interdigitated micro-electrodes integrated on the bottom of tissue culture 96-well E-plates [10]. The effects of siRNA transfection reagent alone, mN3 cytokine combination, and siRNA treatments on hMSCs were monitored using RTCA system. Briefly, cell culture media was added to 96-well E-plates to obtain background intensity. Then, media was discarded and hMSCs were seeded at a density of 1.5×103 cells/cm2. Twentyfour hours later, topo IIβ was silenced by siRNA treatment as described above. And after 24 h, these cells were treated with mN3 cytokine combination to induce neural differentiation. Induction and transfection media were refreshed every 48 h. E-plates containing the cells were placed on the reader in the incubator for continuous recording for 12 days.
Bioinformatics analysis The bioinformatics analysis started with a summary of the measured data. Thereafter, spatial effects on the chip were investigated and corrected. Then, the intensity value distribution of raw data was analyzed and, if required, normalized. For the detection of differentially regulated mRNAs, the following comparison was done as the experimental groups explained in neural induction method are as follows: group 1 vs. group 2 [I-II], group 1 vs. group 3 [IIII], group 2 vs. group 3 [II-III]. Panther pathway analysis Genes not expressed as equal or higher than twofold for each experimental group (I-II, I-III, and II-III) were filtered out before subjecting gene IDs to PANTHER GeneList analysis [4]. Obtained ID numbers for each group were used to run functional classification and statistical overrepresentation tests for detecting related PANTHER Pathways and GO Biological Process comparing with Homo sapiens reference list. Exported results were then used to make the graphical representation by using “obtained value/expected value” ratios for each pathway or biological process.
Results Characterization of hMSCs The immunophenotype characteristics of BM-hMSCs at passage 3 were analyzed by fluorescence activated cell sorting (FACS). Relevant expression levels of cell surface markers are
Ann Hematol
shown in Fig. 1a. hMSCs were positive for CD90 (93.8 %), CD105 (98.2 %), and CD73 (99.9 %) markers and negative for HLA-DR (3.3 %), CD45 (0.8 %), and CD34 (0.3 %). The biological characteristic that most uniquely identifies MSCs is their ability to differentiate into osteocytes, adipocytes, and chondrocytes [3]. Therefore, as indicated in Fig. 1b, MSCs were shown to differentiate into mesodermal lineages under appropriate in vitro differentiating conditions.
Brain-specific marker expressions
Neural differentiation potential of hMSCs
Cell proliferation assay by real-time cell analysis
Morphology
We first analyzed the effect of siRNAs on cell viability; cells were treated with various concentrations (50 nM / pink line, 25 nM /blue line, and 8 nM /green line) of siRNAs and cell index values were obtained, compared with control (black line) hMSCs. Adding siRNAs to hMSCs had no or modest change in the measured microimpedance indicating that the concentrations of siRNAs used in the experiments were in the safe range for hMSCs (Fig. 5a). Cell growth curves of control, mN3, and mN3+Tr were compared in Fig. 5b during 12 days of neural differentiation. In control (black line), after the first 48 h of incubation, the cells grew in an exponential fashion, showing an increasing cell index and then cell proliferation slowed down on day seven. mN3 (purple line) showed a decreased cell index value compared to control;
In order to analyze the role of topo IIβ in neural differentiation process, hMSCs were differentiated into neural-like phenotype by treating with mN3 cytokine combination for 12 days. Control hMSCs showed fibroblast-like morphology and maintained this morphology throughout their period in culture (12 days). mN3 hMSCs started to change their morphology and appeared as neuron-like cells. mN3 + Tr hMSCs also showed neuron-like morphology at day 12. However, topo IIβ silencing decreased by 50 % MSC differentiation to neuron-like cells compared to mN3 (Fig. 2a). Neuron-like cells in mN3 + Tr also had shorter processes (decreased to 25 % of mN3 cellular processes lengths) than cells in mN3 (Fig. 2b).
Fig. 1 a Representative flow cytometry analyses of cell surface markers in BM-hMSCs at passage 3. hMSCs were positive for CD90, CD105, and CD73 and negative for HLA-DR, CD45, and CD34. b Multilineage mesodermal differentiation. Adipogenic, osteogenic, and chondrogenic
To confirm neural differentiation of hMSCs, the expression of brain-specific markers was detected by immunocytochemistry. Cells in mN3 expressed brain-specific markers such as NF, NeuN, MAP2, Olig 2, NSE, and GFAP (Fig. 3). These marker proteins were not detected in control hMSCs (Supplementary Fig. 1).
differentiation after staining with Oil Red O, Toluidine Blue, and Alcian Blue, respectively. (FITC = fluorescein isothiocyanate, PE = phycoerythrin)
Ann Hematol
Fig. 2 Morphology of control, mN3, and mN3 + Tr hMSCs. a Morphology of hMSCs in control (days 0 and 12), mN3, and mN3 + Tr groups (day 12). hMSCs lost their fibroblastic shape and showed neural-like morphology after 12 days of induction. Transfection decreased neural differentiation potential of hMSCs as most cells are unresponsive to
differentiation. b Quantification of cellular processes of mN3 and mN3 + Tr at day 12. Cellular processes of mN3 were in the range of 263±121 while mN3 + Tr were 64±23 μm. Cellular processes were measured by NIS-Elements BR (Nikon Inc., Japan) software (images ×10; D = day)
the values had not been changed and proceeded in a linear fashion relative to starting date of neural induction. As topo IIβ activity was silenced in a concentrationdependent manner with 50 nM (pink line), 25 nM (blue line), and 8 nM (green line) of topo IIβ-specific siRNAs, cell index values of mN3 + Tr increased exponentially
resembling the growth curve of control hMSCs. Exponential growing was dependent on siRNA concentrations, since higher siRNA concentrations had the faster exponential cell growth. Cell growth continued about 30 h after topo IIβ-specific siRNA treatments reaching to their peak values then declined rapidly to cell index values
Fig. 3 Immunostaining results of neural differentiated hMSCs with mN3 cytokine combination. NF, NeuN, MAP2, NSE, GFAP, and Olig2 marker expressions of differentiated hMSCs (images ×10)
Ann Hematol
close to mN3. However, peak points of cell index values increased along the way of oscillatory wave in exact time points where siRNA treatments reached their maximum time (Fig. 5b). In all experimental groups (control, mN3, and mN3 + Tr), necrotic and apoptotic cell death were evaluated at the 3rd, 7th, and 12th days by Sytox green and Annexin V stainings. In CAM-treated positive control cells, at the 3rd, 7th, and 12th days, 89±10, 91±4, and 92±3 % of necrotic cells were detected, respectively. At the 7th day, 9±2, 4.8±3, and 8± 1 % necrotic cells and at the 12th day, 21±2, 11±3, and 15± 4 % of necrotic cells were detected in control, mN3, and mN3 + Tr groups, respectively. As shown in Fig. 5d, 98.9±2, 98.9±2, and 99.4±1 % of apoptotic cells were detected only for CAM-treated positive control cells at the 3rd, 7th, and 12th days, respectively. In control group, 2±0.9 and 17±2 % apoptotic cells were observed at the 7th and 12th days, respectively. In mN3 and mN3 + Tr groups, apoptotic cells were not detected. RT-PCR In order to confirm selective silencing of topo IIβ with siRNA transfection during neural differentiation of hMSCs, expression patterns of topo II isoforms were validated by RT-PCR analysis (Fig. 4). Before the day of neural induction (day 0), control hMSCs expressed both topo IIα and topo IIβ. Topo IIα expression Fig. 4 RT-PCR results and relative densities of topo IIα and topo IIβ in mN3 and mN3 + Tr hMSCs. Since topo IIα is only found in dividing cells, the expression pattern decreases during neural differentiation in both mN3 and mN3 + Tr, as shown in the graph. However, a slight increase in topo IIα expression is observed at day 12. The increase in topo IIβ expression during neural differentiation confirmed the role of this isoform in late stages of neural differentiation. As topo IIβ was silenced in mN3 + Tr group, no expression was detected. Actin was used as a housekeeping gene
detected only in proliferating cells and it was decreased during neural differentiation as cells enter into differentiation process after final cell division. An increased expression of topo IIβ during neural differentiation process has been observed during 12 days of neural differentiation. RT-PCR results of mN3 + Tr have shown that topo IIβ mRNA expression was suppressed by treatment with topo IIβ-specific siRNAs. This indicated efficient silencing of topo IIβ during neural differentiation. We have also shown that expression level of topo IIα at day 2 and day 6 in mN3 + Tr decreased compared to day 0, was parallel to mN3 at day 2 and day 6, but slightly increased at day 12 (Fig. 4). These findings confirmed the selective silencing of topo IIβ during neural differentiation of hMSCs (Fig. 5). Microarray Gene analysis In order to evaluate the subset of genes that are regulated in neural differentiation of hMSCs and the genes affected by topo IIβ, microarray analysis was performed. A comparison was made between (a) control and mN3 [I-II], (b) control and mN3 + Tr [I-III], and (c) mN3 and mN3 + Tr [II-III]. Many genes were altered in expression in all three groups (Fig. 6). When the gene expression patterns were compared in group [I-II], we detected that 21 genes were differentially modified (absolute value of change equal or greater than 1.5-fold). When topo IIβ was silenced, a
Ann Hematol Fig. 5 Cell viability was compared in control, mN3, and mN3 + Tr hMSCs. a The effect of different concentrations of siRNA treatment was analyzed on cell survival as shown by RTCA analysis. Pink, blue, green, and black lines represent cell index values of 50, 25, and 8 nM of siRNAs-treated hMSCs and control hMSCs, respectively. b RTCA cell index values of 50 nM (pink), 25 nM (blue), and 8 nM (green) topo IIβ-specific siRNAtransfected hMSCs (mN3 + Tr) compared with control (black line) and mN3 (purple) hMSCs during 12 days of neural induction. Days −1, 1, 3, 5, 7, 9, and 11 showing transfection and days 0, 2, 4, 6, 8, and 10 showing neural induction times. Day 0 represents the neural induction time point so day (−1) and day (−2) represent siRNA transfection and cell seeding time points relative to neural induction, respectively. c The determination of the percentage of Sytox Greenpositive cells in CAM-treated positive control, control, mN3, and mN3 + Tr hMSCs at the end of 3rd, 7th, and 12th days of neural induction (p
ORIGINAL ARTICLE
DNA topoisomerase IIβ as a molecular switch in neural differentiation of mesenchymal stem cells Sevim Isik & Merve Zaim & Mehmet Taha Yildiz & Yesim Negis & Tuba Kunduraci & Nihal Karakas & Gulsum Arikan & Guven Cetin
Received: 25 March 2014 / Accepted: 28 August 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Two isoforms of DNA topoisomerase II (topo II) have been identified in mammalian cells, named topo IIα and topo IIβ. Topo IIα plays an essential role in segregation of daughter chromosomes and thus for cell proliferation in mammalian cells. Unlike its isozyme topo IIα, topo IIβ is greatly expressed upon terminal differentiation of neuronal cells. Although there have been accumulating evidence about the crucial role of topo IIβ in neural development through activation or repression of developmentally regulated genes at late stages of neuronal differentiation, there have been no reports that analyzed the roles of topo IIβ in the neural trans differentiation process of multipotent stem cells. Terminal differentiation of neurons and transdifferentiation of Mesenchymal Stem Cells (MSCs) are two distinct processes. Therefore, the functional significance of topo IIβ may also be different in these differentiation systems. MSC transdifferentiation into neuron-like cells represents an useful model to further validate the role of topo IIβ in neuronal differentiation. The aim of this Electronic supplementary material The online version of this article (doi:10.1007/s00277-014-2209-7) contains supplementary material, which is available to authorized users. S. Isik Department of Medical Biology, Faculty of Medicine, Fatih University, Buyukcekmece, Istanbul 34500, Turkey S. Isik (*) : M. Zaim : M. T. Yildiz : Y. Negis : T. Kunduraci : N. Karakas : G. Arikan Department of Biology, Faculty of Arts and Science, Fatih University, Buyukcekmece, Istanbul 34500, Turkey e-mail: [email protected] Y. Negis Department of Medical Biochemistry, Faculty of Medicine, Bahcesehir University, Istanbul, Turkey G. Cetin Division of Hematology, Department of Internal Medicine, Medical Faculty of Bezmialem Vakıf University, Istanbul, Turkey
study is to evaluate the subset of genes that are regulated in neural transdifferentiation of bone marrow-derived human MSCs (BM-hMSCs) in vitro and find genes related with topo IIβ. For this purpose, topo IIβ was silenced by specific small interfering RNAs in hMSCs and cells were induced to differentiate into neuron-like cells. Differentiation and silencing of topo IIβ were monitored by real-time cell analysis and also expressions of topo II isoforms were analyzed. Change in transcription patterns of genes upon topo IIβ silencing was identified by DNA microarray analysis, and apparently genes involved in regulation of several ion channels and transporters, vesicle function, and cell calcium metabolism were particularly affected by topo IIβ silencing suggesting that topoIIβ silencing can significantly alter the gene expression pattern of genes involved in variety of biological processes and signal transduction pathways including transcription, translation, cell trafficking, vesicle function, transport, cell morphology, neuron guidance, growth, polarity, and axonal growth. It appears that the deregulation of these pathways may contribute to clarify the further role of topo IIβ in neural differentiation. Keywords DNA topoisomerase IIβ . siRNA transfection . Human mesenchymal stem cells . Neural differentiation . Microarray analysis
Introduction DNA topoisomerase II (topo II) is a nuclear enzyme which catalyzes the transport of double helix through another, thus alters DNA topology. Several genetic processes such as DNA replication, transcription, recombination, as well as chromosome compaction and segregation depend on its activity [29]. There are two isoforms of topo II in mammalian cells named topoisomerase IIα (topo IIα) and topoisomerase IIβ (topo IIβ). Topo IIα plays a key role in mitotic processes and therefore is
Ann Hematol
only present in proliferating tissues. Unlike topo IIα, topo IIβ is present in all tissues and is abundantly expressed in postmitotic neuronal cells in developing brain [30, 32, 18]. Especially whole body and brain-specific topo IIβ knockout mice studies revealed crucial significance of topo IIβ both in brain development and neural differentiation [32, 18]. Additionally, several studies exist emphasizing a possible link between topo IIβ and neural development/differentiation processes in granule neurons of developing rat cerebellum [27, 28, 9]. Topo IIβ affects neural development through activation or repression of developmentally regulated genes associated with neuronal maturation such as ion channels, receptors, and signal transduction molecules [27, 28]. Transcriptional activation of the genes is dependent on the enzymatic activity of topo IIβ probably through opening of chromatin structure and thereby leading to association of the DNA transcription factors to the promoter regions [6]. As being pluripotent adult stem cells, MSCs are relevant system to study molecular details of neural differentiation because of the multilineage differentiation potential of the cells. Supportingly, studies show that MSCs can transdifferentiate into cardiomyocytes and even into cells of non-mesodermal origin as well as neurons [2]. The capacity of MSCs originated from rat, mouse, and human to differentiate into neurons has been shown for the first time by Sanchez-Ramos et al. and Woodbury et al. using growth factors and chemicals, respectively [24, 31]. Later on, many other synthetic and biological neural inducers have been tested for their abilities of differentiating MSCs into neurons including retinoic acid, cytokines, growth factors, neurotrophins, antioxidants, demethylating agents, forskolin, noggin 3-isobutyl-1methylxanthine (IBMX), dibutyryl cyclic adenosine monophosphate (dbcAMP), and dimethylsulfoxide (DMSO) [15, 16]. On the other hand, in vitro neural differentiation process of MSCs is considered to be substantially different from the terminal differentiation of post-mitotic neurons in developing brain. Therefore, the functional significance of topo IIβ may also be different in these differentiation systems. In the current study, we differentiated bone marrowderived human MSCs into neural-like cells by a cytokine combination. To investigate topo IIβ’s role on gaining neural-like morphology during neural differentiation and interaction of relevant gene sets on this differentiation process, topo IIβ was selectively silenced.
Materials and methods Ethical approval Bone marrow (BM) samples used in this work were collected after informed consent; all procedures were approved by the
Ethics Committee of Fatih University, Medical School, and the study conformed with the 2013 WMA Declaration of Helsinki. Isolation of hMSCs from BM Human BM aspirates were obtained from the (posterior) iliac crest of healthy donors. BM aspirates were provided by Bezmialem University, Medical School, Department of Hematology. Ficoll density gradient centrifugation method was used to isolate hMSCs from BM. Isolated cells were seeded at a density of 1.5–2×105 cells/cm2 into a tissue culture flask (BD Falcon; BD Biosciences, San Jose, CA, USA) including DMEM-LG (Gibco, Grand Island, NY, USA), 20 % MSCqualified Fetal Bovine Serum (MSC-FBS, Gibco), and 0.1 mg/ml primocin (InvivoGen, San Diego, CA, USA), and incubated in 37 °C, 5 % CO2 incubator. Three days after isolation, non-adherent hematopoietic cells were removed via medium refreshment. Culture of hMSCs Isolated cells were seeded into culture flasks, and attached cells started to form colonies after 5–6 days. When cells became 80–90 % confluent, they were detached with 0.25 % Trypsin/EDTA (Gibco) solution, seeded into culture flasks at a density of 1.5×103 cells/cm2 with DMEM-LG containing 10 % MSC-FBS, and incubated at 37 °C, 5 % CO2 incubator. Subculturing of cells was performed at about 5–6 days intervals and medium was refreshed once in between two passages. hMSCs at passage 3 were screened by flow cytometry and used for neural differentiation and siRNA transfection experiments. Immunophenotyping of hMSCs The phenotypic characteristics of hMSCs were confirmed at passage 3, by flow cytometry analysis using antibodies against CD45, anti-HLA-DR, CD90, CD34, CD73, and CD105. Mesodermal differentiation of hMSCs Isolated cells from BM were differentiated into mesodermal (adipogenic, chondrogenic, osteogenic) lineages in order to qualify them as MSCs. Briefly, osteogenic differentiation was assessed by incubating 2 × 10 5 cells/cm 2 in Complete MesenCult Osteogenic Medium including MesenCult MSC Basal Medium, osteogenic stimulatory supplement, β-glycerophosphate, dexamethasone, and ascorbic acid (all from Stemcell) for 5 weeks. Adipogenic differentiation was assessed by incubating 5× 103 cells/cm2 in Complete MesenCult Adipogenic Medium containing MesenCult MSC Basal Medium and 10 % adipogenic stimulatory supplement for 3 weeks.
Ann Hematol
Chondrogenic differentiation was assessed by incubating 1.5× 107 cells/ml for 3 weeks with Stempro chondrocyte differentiation basal medium (Gibco) containing 10 % Stempro chondrogenesis supplement (Gibco). Cell morphologies were observed under light microscopy and histochemical staining was performed. Osteogenic, adipogenic, and chondrogenic differentiation was confirmed by Toluidine Blue, Oil Red O, and Alcian Blue staining, respectively. Transfection of hMSCs with Lipofectamine RNAiMAX reagent The day before transfection, cells were seeded into six-well plates at a density of 1.5×103 cells/cm2. Two different validated siRNAs (TOP2B_5 and TOP2B_6 from Qiagen) were used for silencing topo IIβ. Sequences of TOP2B_5 and TOP2B_6 were TCGGGCTAGGAAAGAAGTAAA and CAGCCGAAAGACCTAAATACA, respectively. siRNAs were diluted with Opti-MEM in 1:100 and Lipofectamine RNAiMAX was diluted with Opti-MEM in 1:50 ratio. Diluted siRNAs (50, 25, and 8 nM topo IIβ-specific siRNAs) and diluted reagent were mixed and incubated for 15 min at room temperature (RT) to allow the siRNA-Lipofectamine RNAiMAX complexes to form. siRNA-Lipofectamine RNAiMAX complexes were added into the wells and incubated at 37 °C in CO2 incubator for 24–48 h. After 24 h of incubation, the medium containing complex was replaced by neural induction medium. Since transfection was performed during neural differentiation, medium was replaced with neural induction medium. Efficiency of transfection was checked by real-time cell analysis (RTCA) and reverse transcription PCR (RT-PCR). Neural differentiation of hMSCs hMSCs were divided into three different experimental groups: Group 1/undifferentiated hMSCs (control) Cells were cultured in expansion medium, DMEM-LG containing 10 % MSC-FBS. Medium refreshment was done every 48 h. Group 2/neural differentiated hMSCs (mN3) hMSCs were precultured for 48 h in expansion medium. Then, culture medium was replaced with neural induction medium called modified N3 (mN3). N3 neural induction medium as previously described [14] includes 0.5 mM IBMX (Sigma), 20 ng/ml human epidermal growth factor (hEGF; Sigma), 40 ng/ml recombinant human basic fibroblast growth factor (bFGF; R&D Systems, Minneapolis, MN, USA), 10 ng/ml fibroblast growth factor-8 (FGF-8; PeproTech, Rocky Hill, NJ, USA), 10 ng/ml recombinant human brain-derived neurotrophic factor (BDNF; R&D Systems), in Neurobasal Medium (Gibco) supplemented with 2 % B27 Supplement
(Gibco). mN3 cytokine combination was obtained by adding 0.25 mg/ml dbcAMP (Sigma) instead of 1 mM dbcAMP, 2 mM L-glutamine (Gibco), and 40 ng/ml nerve growth factor (NGF; Gibco) into N3. Neural induction medium was refreshed every 48 h. Group 3/neural differentiated + topo IIβ silenced hMSCs by siRNA transfection (mN3 + Tr) Topo IIβ silencing (siRNA transfection) was performed 24 h before neural induction. During medium refreshment, transfection was also performed every 48 h. After 12 days, cell morphologies of all experimental groups were observed under light microscopy and brain-specific marker expressions were monitored at mRNA level with RT-PCR. Because of complicated terminology used for experimental groups in the study, we provide an explanatory table to avoid possible confusions (Table 1).
Reverse transcription PCR siRNA silencing efficiency and expression of neural markers, actin, topo IIα, and topo IIβ, were checked at mRNA level using RT-PCR technique. Total RNA was extracted from hMSCs using RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions. Concentration of extracted RNA was quantified using Qubit (Invitrogen, Carlsbad, CA, USA). cDNA was generated by using a Quantitect reverse transcription kit (Qiagen). Actin, topo IIα, and topo IIβ underwent 30, 35, and 35 cycles of amplification at 60, 54, and 60 °C annealing temperatures, respectively (PCR core kit, Qiagen). Forward and reverse primers were CGCACCACTG GCATTGTCAT and GTGGCCATCTCCTGCTCGAA, yielding a 208 bp for actin; ACCATTGCAGCCTGTA and GCTC TTCCCATATTATCC yielding a 596 bp for topo IIα; ATCA AAAGCCACTC-CAGAAAAATC and AGAAGGTGGCTC AGTAGGGAAGTC yielding a 508 bp for topo IIβ. Actin was used as the housekeeping gene. The PCR products were loaded on 2 % agarose gel and were visualized with a gel documentation system.
Table 1 Abbreviations of experimental groups in the study
Group 1 Group 2 Group 3
Experimental groups
Abbreviation used
Undifferentiated hMSCs Neural differentiated hMSCs (mN3 treated hMSCS) Neural differentiated + topo IIβ silenced hMSCs by siRNA transfection
Control mN3 mN3+Tr
Ann Hematol
Immunofluorescent staining
Cell death analysis
hMSCs were seeded into 24-well plates at a density of 1.5×103 cells/cm2 and incubated in neural induction medium for 12 days. Then, cells were permeabilized with TZN buffer (10 mM pH 7.5 Tris–HCl, 0.5 % Nonidet P-40, 0.2 mM ZnCl2) and fixed with 4 % PFA/PBS at RT. Cells were blocked in 0.3 % Triton X/PBS (PBS-Tx) solution containing 10 % goat and 10 % horse serum (Biochrom) for 30 min at RT (Gibco) Then, they were incubated with specific primary antibodies, both diluted in PBS-Tx with 3 % NHS for 2 h at RT. After primary antibody incubation, cells were treated with Alexa Fluorlabeled anti-mouse or anti-rabbit secondary antibodies at RT for 1 h. For nuclear staining, cells were treated with 1:15,000× DAPI (Sigma) solution and incubated at RT for 10 min. Finally, glass coverslips were mounted onto the microscope slides covered with Prolong Antifade Kit (Molecular Probes) or Prolong Gold Antifade Reagent (Invitrogen). Cells were visualized under fluorescent microscope (Nikon Ti, Japan). Primary antibodies against neurofilament (NF; 1:100; Chemicon Temecula, CA), microtubule-associated protein 2 (MAP2; 1:50; Cell Signaling), neuron-specific enolase (NSE; 1:100; Chemicon), neuronal nuclei (NeuN; 1:100; Chemicon Temecula, CA), oligodendrocyte transcription factor 2 (Olig2; 1:100; Chemicon Temecula, CA), glial fibrillary acidic protein (GFAP; 1:100, Chemicon Temecula, CA), and as secondary antibodies GAM-IgG-Alexa Fluor 488 (1:100, Invitrogen) and GAR-IgGAlexa Fluor 594 (1:100, Invitrogen) were used.
hMSCs were seeded into 96-well plates at a density of 1.5× 103 cells/cm2. Twenty-four hours later, topo IIβ was silenced by siRNA treatment as described above. And after 24 h, these cells were treated with mN3 cytokine combination to induce neural differentiation. Induction and transfection media were refreshed every 48 h. Annexin V and Sytox Green staining were performed at 3rd, 7th, and 12th days of neural induction. As positive control of apoptosis, cells were treated with camptothecin (Sigma). Culture medium was replaced with 100 μl of Annexin V-Alexa 568 labeling solution (Roche) and 50 μM Sytox Green dye (Invitrogen) was added to each well to differentiate apoptotic cells from necrotic cells. After 15 min incubation at 25 °C, cells were washed with PBS and treated with 1:15,000 DAPI (Sigma) for 10 min at RT to stain nuclei. Cells were then washed with PBS and dH2O and observed under fluorescence microscopy (Nikon, Eclipse Tí).
Microarray analysis Microarray analysis was done by Febit group (febit GmbH, Heidelberg) for whole set of human transcriptome comprising 22.440 genes. Real-time cell analysis The real-time cell analyzer (RTCA; xCELLigence, Roche) monitors cellular events in real time by the measurement of electrical impedance across interdigitated micro-electrodes integrated on the bottom of tissue culture 96-well E-plates [10]. The effects of siRNA transfection reagent alone, mN3 cytokine combination, and siRNA treatments on hMSCs were monitored using RTCA system. Briefly, cell culture media was added to 96-well E-plates to obtain background intensity. Then, media was discarded and hMSCs were seeded at a density of 1.5×103 cells/cm2. Twentyfour hours later, topo IIβ was silenced by siRNA treatment as described above. And after 24 h, these cells were treated with mN3 cytokine combination to induce neural differentiation. Induction and transfection media were refreshed every 48 h. E-plates containing the cells were placed on the reader in the incubator for continuous recording for 12 days.
Bioinformatics analysis The bioinformatics analysis started with a summary of the measured data. Thereafter, spatial effects on the chip were investigated and corrected. Then, the intensity value distribution of raw data was analyzed and, if required, normalized. For the detection of differentially regulated mRNAs, the following comparison was done as the experimental groups explained in neural induction method are as follows: group 1 vs. group 2 [I-II], group 1 vs. group 3 [IIII], group 2 vs. group 3 [II-III]. Panther pathway analysis Genes not expressed as equal or higher than twofold for each experimental group (I-II, I-III, and II-III) were filtered out before subjecting gene IDs to PANTHER GeneList analysis [4]. Obtained ID numbers for each group were used to run functional classification and statistical overrepresentation tests for detecting related PANTHER Pathways and GO Biological Process comparing with Homo sapiens reference list. Exported results were then used to make the graphical representation by using “obtained value/expected value” ratios for each pathway or biological process.
Results Characterization of hMSCs The immunophenotype characteristics of BM-hMSCs at passage 3 were analyzed by fluorescence activated cell sorting (FACS). Relevant expression levels of cell surface markers are
Ann Hematol
shown in Fig. 1a. hMSCs were positive for CD90 (93.8 %), CD105 (98.2 %), and CD73 (99.9 %) markers and negative for HLA-DR (3.3 %), CD45 (0.8 %), and CD34 (0.3 %). The biological characteristic that most uniquely identifies MSCs is their ability to differentiate into osteocytes, adipocytes, and chondrocytes [3]. Therefore, as indicated in Fig. 1b, MSCs were shown to differentiate into mesodermal lineages under appropriate in vitro differentiating conditions.
Brain-specific marker expressions
Neural differentiation potential of hMSCs
Cell proliferation assay by real-time cell analysis
Morphology
We first analyzed the effect of siRNAs on cell viability; cells were treated with various concentrations (50 nM / pink line, 25 nM /blue line, and 8 nM /green line) of siRNAs and cell index values were obtained, compared with control (black line) hMSCs. Adding siRNAs to hMSCs had no or modest change in the measured microimpedance indicating that the concentrations of siRNAs used in the experiments were in the safe range for hMSCs (Fig. 5a). Cell growth curves of control, mN3, and mN3+Tr were compared in Fig. 5b during 12 days of neural differentiation. In control (black line), after the first 48 h of incubation, the cells grew in an exponential fashion, showing an increasing cell index and then cell proliferation slowed down on day seven. mN3 (purple line) showed a decreased cell index value compared to control;
In order to analyze the role of topo IIβ in neural differentiation process, hMSCs were differentiated into neural-like phenotype by treating with mN3 cytokine combination for 12 days. Control hMSCs showed fibroblast-like morphology and maintained this morphology throughout their period in culture (12 days). mN3 hMSCs started to change their morphology and appeared as neuron-like cells. mN3 + Tr hMSCs also showed neuron-like morphology at day 12. However, topo IIβ silencing decreased by 50 % MSC differentiation to neuron-like cells compared to mN3 (Fig. 2a). Neuron-like cells in mN3 + Tr also had shorter processes (decreased to 25 % of mN3 cellular processes lengths) than cells in mN3 (Fig. 2b).
Fig. 1 a Representative flow cytometry analyses of cell surface markers in BM-hMSCs at passage 3. hMSCs were positive for CD90, CD105, and CD73 and negative for HLA-DR, CD45, and CD34. b Multilineage mesodermal differentiation. Adipogenic, osteogenic, and chondrogenic
To confirm neural differentiation of hMSCs, the expression of brain-specific markers was detected by immunocytochemistry. Cells in mN3 expressed brain-specific markers such as NF, NeuN, MAP2, Olig 2, NSE, and GFAP (Fig. 3). These marker proteins were not detected in control hMSCs (Supplementary Fig. 1).
differentiation after staining with Oil Red O, Toluidine Blue, and Alcian Blue, respectively. (FITC = fluorescein isothiocyanate, PE = phycoerythrin)
Ann Hematol
Fig. 2 Morphology of control, mN3, and mN3 + Tr hMSCs. a Morphology of hMSCs in control (days 0 and 12), mN3, and mN3 + Tr groups (day 12). hMSCs lost their fibroblastic shape and showed neural-like morphology after 12 days of induction. Transfection decreased neural differentiation potential of hMSCs as most cells are unresponsive to
differentiation. b Quantification of cellular processes of mN3 and mN3 + Tr at day 12. Cellular processes of mN3 were in the range of 263±121 while mN3 + Tr were 64±23 μm. Cellular processes were measured by NIS-Elements BR (Nikon Inc., Japan) software (images ×10; D = day)
the values had not been changed and proceeded in a linear fashion relative to starting date of neural induction. As topo IIβ activity was silenced in a concentrationdependent manner with 50 nM (pink line), 25 nM (blue line), and 8 nM (green line) of topo IIβ-specific siRNAs, cell index values of mN3 + Tr increased exponentially
resembling the growth curve of control hMSCs. Exponential growing was dependent on siRNA concentrations, since higher siRNA concentrations had the faster exponential cell growth. Cell growth continued about 30 h after topo IIβ-specific siRNA treatments reaching to their peak values then declined rapidly to cell index values
Fig. 3 Immunostaining results of neural differentiated hMSCs with mN3 cytokine combination. NF, NeuN, MAP2, NSE, GFAP, and Olig2 marker expressions of differentiated hMSCs (images ×10)
Ann Hematol
close to mN3. However, peak points of cell index values increased along the way of oscillatory wave in exact time points where siRNA treatments reached their maximum time (Fig. 5b). In all experimental groups (control, mN3, and mN3 + Tr), necrotic and apoptotic cell death were evaluated at the 3rd, 7th, and 12th days by Sytox green and Annexin V stainings. In CAM-treated positive control cells, at the 3rd, 7th, and 12th days, 89±10, 91±4, and 92±3 % of necrotic cells were detected, respectively. At the 7th day, 9±2, 4.8±3, and 8± 1 % necrotic cells and at the 12th day, 21±2, 11±3, and 15± 4 % of necrotic cells were detected in control, mN3, and mN3 + Tr groups, respectively. As shown in Fig. 5d, 98.9±2, 98.9±2, and 99.4±1 % of apoptotic cells were detected only for CAM-treated positive control cells at the 3rd, 7th, and 12th days, respectively. In control group, 2±0.9 and 17±2 % apoptotic cells were observed at the 7th and 12th days, respectively. In mN3 and mN3 + Tr groups, apoptotic cells were not detected. RT-PCR In order to confirm selective silencing of topo IIβ with siRNA transfection during neural differentiation of hMSCs, expression patterns of topo II isoforms were validated by RT-PCR analysis (Fig. 4). Before the day of neural induction (day 0), control hMSCs expressed both topo IIα and topo IIβ. Topo IIα expression Fig. 4 RT-PCR results and relative densities of topo IIα and topo IIβ in mN3 and mN3 + Tr hMSCs. Since topo IIα is only found in dividing cells, the expression pattern decreases during neural differentiation in both mN3 and mN3 + Tr, as shown in the graph. However, a slight increase in topo IIα expression is observed at day 12. The increase in topo IIβ expression during neural differentiation confirmed the role of this isoform in late stages of neural differentiation. As topo IIβ was silenced in mN3 + Tr group, no expression was detected. Actin was used as a housekeeping gene
detected only in proliferating cells and it was decreased during neural differentiation as cells enter into differentiation process after final cell division. An increased expression of topo IIβ during neural differentiation process has been observed during 12 days of neural differentiation. RT-PCR results of mN3 + Tr have shown that topo IIβ mRNA expression was suppressed by treatment with topo IIβ-specific siRNAs. This indicated efficient silencing of topo IIβ during neural differentiation. We have also shown that expression level of topo IIα at day 2 and day 6 in mN3 + Tr decreased compared to day 0, was parallel to mN3 at day 2 and day 6, but slightly increased at day 12 (Fig. 4). These findings confirmed the selective silencing of topo IIβ during neural differentiation of hMSCs (Fig. 5). Microarray Gene analysis In order to evaluate the subset of genes that are regulated in neural differentiation of hMSCs and the genes affected by topo IIβ, microarray analysis was performed. A comparison was made between (a) control and mN3 [I-II], (b) control and mN3 + Tr [I-III], and (c) mN3 and mN3 + Tr [II-III]. Many genes were altered in expression in all three groups (Fig. 6). When the gene expression patterns were compared in group [I-II], we detected that 21 genes were differentially modified (absolute value of change equal or greater than 1.5-fold). When topo IIβ was silenced, a
Ann Hematol Fig. 5 Cell viability was compared in control, mN3, and mN3 + Tr hMSCs. a The effect of different concentrations of siRNA treatment was analyzed on cell survival as shown by RTCA analysis. Pink, blue, green, and black lines represent cell index values of 50, 25, and 8 nM of siRNAs-treated hMSCs and control hMSCs, respectively. b RTCA cell index values of 50 nM (pink), 25 nM (blue), and 8 nM (green) topo IIβ-specific siRNAtransfected hMSCs (mN3 + Tr) compared with control (black line) and mN3 (purple) hMSCs during 12 days of neural induction. Days −1, 1, 3, 5, 7, 9, and 11 showing transfection and days 0, 2, 4, 6, 8, and 10 showing neural induction times. Day 0 represents the neural induction time point so day (−1) and day (−2) represent siRNA transfection and cell seeding time points relative to neural induction, respectively. c The determination of the percentage of Sytox Greenpositive cells in CAM-treated positive control, control, mN3, and mN3 + Tr hMSCs at the end of 3rd, 7th, and 12th days of neural induction (p
