169 TSH-induced expression involves regulation of self-renewal and differentiation-related in human bone marrow-derived mesenchymal stem cells

Emin Umit Bagriacik1,2, Melek Yaman2, Rauf Haznedar3, Gulsan Sucak3 and Tuncay Delibasi4 1Immunology Research Center, Gazi University, 06500 Ankara, Turkey 2Department of Immunology, Faculty of Medicine, Gazi University, Besevler, 06500 Ankara, Turkey 3Department of Hematology, Faculty of Medicine, Gazi University, 06560 Ankara, Turkey 4Endocrinology Department, Diskapi Education and Research Hospital, 06030 Ankara, Turkey (Correspondence should be addressed to E U Bagriacik at Department of Immunology, Faculty of Medicine, Gazi University, Besevler; Email: [email protected])

Abstract Bone marrow-derived mesenchymal stem cells are pluripotent stem cells. Expression of 80 genes was studied by real-time cells that are capable of differentiating into a variety of cell types PCR array profiles. Our investigation indicated involvements of including neuronal cells, osteoblasts, chondrocytes, myocytes, interactions between TSH and its in novel regulatory and adipocytes. Despite recent advances in stem cell biology, pathways, which could be the important mediators of neuroendocrine relations, particularly TSH interactions self-renewal, maintenance, development, and differentiation remain elusive. In this study, we investigated expression and in bone marrow-derived mesenchymal stem cells. TSH biological consequence of TSH receptor (TSHR) interactions enhanced differentiation to the chondrogenic cell lineage; in mesenchymal stem cells of cultured human bone marrow. however, further work is required to determine whether To the best of our knowledge, we demonstrated for the first osteoblastic differentiation is also promoted. Our results time that human bone marrow-derived mesenchymal stem presented in this study have opened an era of regulatory cells expressed a functional thyrotropin receptor that was capable events associated with novel neuroendocrine interactions of of transducing signals through cAMP. We extended this study hypothalamic–pituitary axis in mesenchymal stem cell biology to explore possible pathways that could be associated directly and differentiation. or indirectly with the TSHR function in mesenchymal Journal of Endocrinology (2012) 212, 169–178

Introduction Despite these progressions, neuroendocrine interactions in many aspects of mesenchymal stem cell maintenance and The TSH receptor or the thyrotropin receptor (TSHR) is differentiation are still largely unknown. For example, usually expressed by thyroid cells. It is a G protein-coupled whether hBMSCs express the TSHR is totally unknown. receptor and signals through cAMP and inositol triphosphate Certainly, biological consequences of interactions between (IP3) pathways (Kursawe & Paschke 2000, Davies et al. 2002, the TSH and its receptor in maintenance and differentiation Szkudlinski et al. 2002). Stimulation of the TSHR via TSH on of hBMSCs also remain to be elucidated. thyroid cells leads to production of the thyroid hormones In this study, we aimed to investigate the expression and biological function(s) of the TSHR in hBMSCs. (THs), tri-iodothyronine (T3) and thyroxine (T4)(Toccafondi et al. 1982). The TSHR expression by other tissue cells has also TSH-induced was explored particularly. been reported (Haraguchi et al. 1996, Hoermann 1996, Dutton et al. 1997, Bell et al. 2000, Agretti et al. 2002, 2005, Tsai et al. 2004, Ellerhorst et al. 2006). However, exact function Materials and Methods of its expression in extrathyroidal tissues has not been clarified. Human bone marrow-derived mesenchymal stem cells Isolation and maintenance of mesenchymal stem cells (hBMSCs) have been characterized by many properties so far. hBMSCs were isolated by plastic adherence according to They are pluripotent cells that are capable of differentiating a previously published method (Franc¸ois et al. 2006)from into a variety of mesenchymal tissues (Pittenger et al. 1999, iliac crest aspirates of volunteer donors. Volunteer donors Kemp et al. 2005, Kolf et al. 2007, Abdallah & Kassem 2008). of allogeneic bone marrow transplant recipients signed an

Journal of Endocrinology (2012) 212, 169–178 DOI: 10.1530/JOE-11-0404 0022–0795/12/0212–169 q 2012 Society for Endocrinology Printed in Great Britain Online version via http://www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access 170 E U BAGRIACIK and others . TSHR expression by mesenchymal stem cells

informed consent, and the isolation protocol was approved by overnight and then with an anti-mouse-HRP secondary the local ethics committee of Gazi University Medical antibody (Jackson Immunoresearch) for 2 h at room School. Isolated mesenchymal stem cells were maintained temperature. Enhanced chemiluminescence detection system by culturing them in DMEM supplemented with 10% fetal was used for protein detection (Pierce-Thermo Scientific, bovine serum, 4 mM L-glutamine, penicillin–streptomycin, Rockford, IL, USA). non-essential amino acid solution, and pyruvate (all from Gibco-Invitrogen) at 37 8C and 5% CO2. The medium was Assay for cAMP replaced every 3 days. A commercially available colorimetric assay kit (Assay Design, Immunofluorescence and immunohistochemistry Ann Arbor, MI, USA) was used to analyze the cAMP change following TSH stimulation. Briefly, hBMSCs were starved for hBMSCs were grown on chamber slides (Nunc, Roskilde, 24 h in serum-free culture medium and then stimulated by Denmark) for staining to visualize TSHR. For immuno- the recombinant TSH at logarithmically increasing concen- histochemistry, a commercially available staining kit was used trations. Each concentration was tested in triplicate wells. according to the manufacturer’s recommendations (Dako, Via Measurements were performed at 450 nm using a multi-mode Real Carpinteria, CA, USA). Briefly, cells were fixed with microplate reader (Synergy HT, BioTek, Winooski, VT,USA). ice-cold methanol (or in some cases with 4% paraformaldehyde) for 20 min at room temperature and then stained with Genomic DNA-free total RNA isolation antihuman TSHR antibodies (Mouse IgG1 and IgG2a, ABR, Rockford, IL, USA) at room temperature for 2 h in the Total RNA was isolated using a genomic DNA-free total presence of Hoechst 33342 Fluorescent stain (Invitrogen). RNA isolation kit (Qiagen). Cells were lysed by the lysis After washing extensively, slides were treated with anti-mouse solution provided in the kit content and treated with RNase- IgG-FITC (Jackson ImmunoResearch Laboratories, Inc., West free DNAase as instructed before passing through RNA Grove, PA, USA) for 1 h. Confocal laser scanning microscopy isolation columns. Extracted RNA was quantified and stored (Leica, Wetzlar, Germany) was used to visualize slides. at K86 8C until use. Antihuman COL2A1 and antihuman COL9A1 antibodies (Abcam, Cambridge, MA, USA) were used to detect protein Real-time RT-PCR arrays expression for human COL2A1 and COL9A1. Fixed cells were stained with primary antibodies at room temperature for Gene expression was assessed using a RT2 Profiler PCR Array 2 h. Slides were washed extensively and incubated with an kit obtained from SuperArray Bioscience Corporation anti-rabbit secondary antibody conjugated with HRP.Stained (Frederick, MD, USA). The Human Stem Cell RT2 Profiler cells were visualized using a Leica microscope with software PCR Array including the expression of 80 genes of 14 different called ‘The Leica Application Suite’ (Leica, DM4000). pathways and markers related to the identification, growth, and differentiation of stem cells was chosen. The kit also contained FACS analysis the stem cell-specific markers as stem cell differentiation markers and genes in signaling pathways important for stem Adherent cells were trypsinized and washed twice. Cells were cell maintenance. One microgram total RNA per sample was stained with monoclonal antibodies for 30 min at 4 8C, converted into cDNA in reverse transcriptase (RT) reaction. washed twice, and fixed with 2% paraformaldehyde before Real-time PCR was performed using a Light Cycler 480 use. Anti-CD13, CD14, anti-CD29, anti-CD34, anti-CD45, (Roche Applied Science). Amplification continued for 40 anti-CD73, anti-CD105 antibodies, and PE- or FITC- cycles. Results were calculated as fold increase values of gene labeled isotype-matched control antibodies were used expression as instructed by the kit supplier. At least a twofold (eBioscience, San Diego, CA, USA). Analyses were increase or a twofold decrease in a particular gene expression performed using a Cytomics FC500 flow cytometer with over control values was considered as a statistically significant CXP software (Beckman Coulter, Brea, CA, USA). (P!0.05) change. Immunoprecipitation and western blotting Alcian blue staining Immunoprecipitation and western blot experiments were conducted using a previously published protocol (Bagriacik & Alcian blue (Sigma) was prepared using 3% glacial acetic acid. Klein 2000). Briefly, cell lysate was incubated overnight at Fixed cells were stained with 0.1% alcian blue for 30 min. 4 8C with an antihuman TSHR antibody of mouse (ABR) Cells were washed extensively and visualized. and immunoprecipitated with agarose-conjugated protein A (Sigma). Ten percent polyacrylamide ready-made gels Statistical analyses (Thermo Scientific, Rockford, IL, USA) were used for electrophoresis. Immunoblots were treated with another Statistical data were calculated by Student’s t-test or two-way clone of antihuman TSHR primary antibody (ABR) ANOVA as indicated with the level of significance set at

Journal of Endocrinology (2012) 212, 169–178 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access TSHR expression by mesenchymal stem cells . E U BAGRIACIK and others 171

! . ! . P 0 05 or P 0 01. Relative changes as fold change values Early Middle Late K in gene expression were calculated by the formula 2 DDCt using DDCt method (threshold). Isotype

100 101 102 103 100 101 102 103 100 101 102 103 Results 35

Bone marrow-derived cells exhibited the characteristics of CD13 mesenchymal stem cells

We isolated hBMSCs by plate attachment method to study 100 101 102 103 100 101 102 103 100 101 102 103 the TSHR expression and its consequences in mesenchymal stem cells. Confluently grown cells in cultures were typical CD14 mesenchymal stem cells as described by other investigators (Pittenger et al. 1999, Kolf et al. 2007, Abdallah & Kassem 2008). They were adherent monolayer of spindle-shaped cells 100 101 102 103 100 101 102 103 100 101 102 103 having a fibroblastic appearance. We molecularly charac- 32 terized cultured MSCs by flow cytometry (FACS) to make sure that the isolated cells from human bone marrow were CD29 mesenchymal stem cells. For this reason, expression of CD13,

CD14, CD29, CD34, CD73, and CD105 was determined. 100 101 102 103 100 101 102 103 100 101 102 103 There is a growing list of evidence about age-associated decline in MSC-specific marker expression. To find culture stability and purity, we also determined expression of those CD34 MSC markers during early, middle, and late passages of cultured cells. All of the cultured cells at the 5th passage 100 101 102 103 100 101 102 103 100 101 102 103 representing early passages expressed CD13, CD29, CD73, and CD105 at high levels while they lacked the expression of 30 CD14 and CD34 (Fig. 1). They did not express CD45 either CD73 (data not shown). Decline in the frequency of MSC-specific marker expressing cells begun after the 10th passage (data not shown). At the 12th passage, as the mid-passage, only 30–35% 100 101 102 103 100 101 102 103 100 101 102 103 of cultured cells were positive for CD13, CD29, CD73, and 28 CD105 (Fig. 1). None of the cultured cells expressed any of CD105 those markers at the 20th passage as the late passage (Fig. 1). Based on these data, we concluded that our isolates of bone marrow-derived cells were indeed mesenchymal stem cells, 100 101 102 103 100 101 102 103 100 101 102 103 and they maintained surface expression of those mesenchymal Figure 1 FACS analysis of specific markers for hBMSCs. Cells were stem cell markers at high levels until the 10th passage (data not stained with anti-CD13, anti-CD14, anti-CD29, anti-CD34, anti- shown). Therefore, we used the cells from the 5–6th passages CD73, and anti-CD105 antibodies. PE- or FITC-labeled isotype- for TSHR expression and gene expression studies. matched antibodies were used as control. Cells from early (the 5th passage), middle (the 12th passage), and late passages (the 20th passage) were tested. Numbers at the corners represent percent TSHR expression by mesenchymal stem cells values of positive cells for indicated markers. TSHR expression was explored by three powerful methods TSHR (Fig. 2C). These data clearly showed that TSHR for protein detection and visualization. Western blot analysis with a homogeneous pattern was constitutively expressed on was performed using hBMSCs isolated from two different hBMSCs. donors. Two bands of different size appeared on the blot. The larger band was about 100 kDa and the smaller one was about TSHR signaling through cAMP 65 kDa (Fig. 2A). The results were confirmed by repeats of four independent experiments. TSHR was also visualized The next question we asked was whether the TSHR by using immunofluorescence of confocal laser microscopy. expressed on the hBMSCs was a functional receptor. Since Apparently, hBMSCs in culture expressed a homogeneous TSHR is a G protein-coupled receptor, we measured cAMP expression pattern of TSHR at relatively high density levels following stimulation by TSH. To stimulate the (Fig. 2B). We also tested TSHR expression by flow receptor under in vitro conditions in cell cultures, we used a cytometry. MSCs from different passages expressed the commercially available recombinant human TSH (thyrogen), www.endocrinology-journals.org Journal of Endocrinology (2012) 212, 169–178

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access 172 E U BAGRIACIK and others . TSHR expression by mesenchymal stem cells

A C factors, and genes regulating cell–cell communication. As Size (kDa)13 2 some of those genes would have more than one function 25 100 and can be categorized in overlapping groups, we included 20 them in only one family while evaluating our results. For 65 15 example, APC gene is categorized among both cell adhesion 10 molecules and cell cycle regulators. The gene expression 45 5 results for APC gene was demonstrated only under the group

TSHR expression (MFI) TSHR expression 0 of cell adhesion molecules in Table 1. 0 12 20 Treatment of hBMSCs with the recombinant TSH resulted Passage number in upregulation or downregulation of gene expression not B only in a selective way but also in a dose-dependent manner, K i.e. 10 8 M TSH affected gene expression more than K 10 7 M TSH. We observed a moderate increase in TERT, BMP3, and GDF2 genes (Table 1). On the other hand, we found sharp increases in the expression of FGF3, FGF4, and GDF3 (Table 1). ATP-binding cassette sub-family G member 2 (ABCG2), aldehyde dehydrogenase 1 family, member A1 (ALDH1A1), aldehyde dehydrogenase 2 family (mitochondrial) (ALDH2), and fibroblast growth factor receptor 1 (FGFR1 (fms-related 2, Pfeiffer syndrome)) are among the metabolic markers for stem cells. TSH induced both ABCG2 and ALDH1A1 (Fig. 4A). Increase in the expression of ABCG2 was moderate. However, gene expression of ALDH1A1 increased almost K 19 times upon TSH challenge at 10 8 M in comparison with Figure 2 Determination of TSHR expression. TSHR expression the untreated controls. on the cell surface of hMSCs was determined by three different methods. (A) Western blot: lane 1, TSHR in human thyroid cells; Self-renewal is an essential characteristic of embryonic stem lane 2, TSHR in hBMSCs from the first donor; and lane 3, TSHR cells. We investigated whether self-renewal-related markers in hBMSCs from the second donor. (B) In fluorescence confocal were affected by TSH in hBMSCs. Sex-determining region Y microscopy studies, cells were stained with the primary antihuman (SRY)-box 1 (SOX1), sex-determining region Y (SRY)-box TSHR antibody and FITC-conjugated secondary Ab plus Hoechst to 2(), neurogenin 2 (NEUROG2), MYST1, MYST2, stain the nuclei. Inverted Leica DM4000 with 4-laser Leica confocal microscope system (Leica GE) and software, Wetzlar, Germany. and heat-shock 70 kDa protein 9 (HSPA9) genes were tested. (C) TSHR expression was also assessed by flow cytometry. TSH selectively and significantly augmented NEUROG2 and SOX2 expression, whereas TSH decreased or did not which contained both a and b chains of the hormone. cAMP affect the remaining genes (Fig. 4B). The effect of the TSH production increased in a dose-dependent manner (Fig. 3) was again dose dependent. SOX2 expression increased by w K7 K8 upon stimulation and reached the peak level at a concen- 8- and 21-fold at 10 and 10 M TSH application, K tration of 10 8 M recombinant TSH. Forskolin treatment served as a positive control for induction of cAMP 150

production. 125

100 Induction of specific gene expression by TSH To study biological consequences of TSHR expression in the 75

life of hBMSCs, we investigated TSH-induced expression of fmol/ml cAMP 50 genes that were related to the stem cell-specific markers, stem cell differentiation markers, and signaling pathways important 25 for stem cell maintenance. Quantitative gene expression profiles were studied by real-time RT-PCR technology using 0 –11 –10 –9 –8 –7 –6 highly purified DNA-free RNA samples from stimulated cells Ctrl 10 10 10 10 10 10 Forskolin by the recombinant TSH. (TSH) We studied several stem cell-specific markers that were Figure 3 Determination of cAMP in TSH-treated hBMSCs. categorized under various groups of functional gene families An ELISA-based cAMP assay was used to measure cAMP increase followed by TSH treatment. Forskolin served as a positive control. such as cell adhesion molecules, and chromatin Assay was performed in triple wells per sample. The mean value and modulators, cell cycle regulators, cytokines and growth the S.D.s were calculated.

Journal of Endocrinology (2012) 212, 169–178 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access TSHR expression by mesenchymal stem cells . E U BAGRIACIK and others 173

Table 1 Induction of gene expression for stem cell specific markers

Fold change K K Genbank Function 10 7 Ma 10 8 Ma

Genes Cell adhesion molecules APC NM_000038 Adenomatous polyposis coli C1.02 K1.17 BGLAP NM_199173 Bone gamma-carboxyglutamate (gla) protein (osteocalcin) C1.18 K1.25 CD44 NM_000610 CD44 molecule (Indian blood group) C1.0 K1.33 CDH1 NM_004360 Cadherin 1, type 1, E-cadherin (epithelial) C1.2 C1.71 CDH2 NM_001792 Cadherin 2, type 1, N-cadherin (neuronal) C1.12 C1.04 CTNNA1 NM_001903 Catenin (cadherin-associated protein), alpha 1, 102 kDa C1.03 C1.01 CXCL12 NM_000609 Chemokine (C-X-C motif) ligand 12 (stromal cell-derived factor 1) K1.36 K1.56 NCAM1 NM_000615 Neural cell adhesion molecule 1 K1.19 C1.05 Chromosome and chromatin modulators KAT2A NM_021078 General control of amino-acid synthesis 5-like 2 (yeast) K1.39 K1.14 HDAC2 NM_001527 deacetylase 2 C1.38 C1.67 MYST1 NM_032188 MYST histone acetyltransferase 1 K1.04 C1.02 MYST2 NM_007067 MYST histone acetyltransferase 2 K1.37 K1.38 RB1 NM_000321 Retinoblastoma 1 (including ) K1.09 K1.05 TERT NM_198253 TERT Telomerase reverse transcriptase C2.35* C5.80* Cell cycle regulators AXIN1 NM_003502 Axin 1 K1.54 K3.34* CCNA2 NM_001237 Cyclin A2 C1.11 C1.07 CCND1 NM_053056 Cyclin D1 K1.30 C1.43 CCND2 NM_001759 Cyclin D2 K1.08 C1.13 CCNE1 NM_001238 Cyclin E1 K1.30 K1.33 CDK1 NM_001786 Cell division cycle 2, G1 to S and G2 to M C1.07 C1.10 CDC42 NM_001791 Cell division cycle 42 (GTP binding protein, 25 kDa) K1.24 K1.27 EP300 NM_001429 E1A binding protein p300 K1.39 K2.43* NM_002467 V-myc myelocytomatosis viral oncogene homolog (avian) K1.19 K2.95* PARD6A NM_016948 Par-6 partitioning defective 6 homolog alpha (Caenorhabditis elegans) K1.23 K1.29 Cytokines and growth factors BMP1 NM_006129 Bone morphogenetic protein 1 K1.41 K1.68 BMP2 NM_001200 Bone morphogenetic protein 2 K1.02 C1.45 BMP3 NM_001201 Bone morphogenetic protein 3 (osteogenic) C2.90* C6.31* CXCL12 NM_000609 Chemokine (C-X-C motif) ligand 12 K1.38 K1.56 FGF2 NM_002006 Fibroblast growth factor 2 (basic) K1.43 K1.44 FGF3 NM_005247 Fibroblast growth factor 3 C6.31* C18.35* FGF4 NM_002007 Fibroblast growth factor 4 C11.37* C31.51* GDF2 NM_016204 Growth differentiation factor 2 C3.06* C9.76* GDF3 NM_020634 Growth differentiation factor 3 C4.78* C18.73* IGF1 NM_000618 IGF1 (somatomedin C) K1.08 C1.23 JAG1 NM_000214 Jagged 1 (Alagille syndrome) K1.96 K2.14* Genes regulating cell–cell communication DHH NM_021044 Desert hedgehog homolog (Drosophila) C1.5 C3.06* DLL1 NM_005618 Delta-like 1 (Drosophila) C1.5 C2.54* GJA1 NM_000165 Gap junction protein, alpha 1, 43 kDa K1.19 K1.18 GJB1 NM_000166 Gap junction protein, beta 1, 32 kDa C2.54* C4.78* GJB2 NM_004004 Gap junction protein, beta 2, 26 kDa C1.54 C3.0

*P!0.05. aTSH concentrations used to stimulate hBMSCs. respectively, while the increase in NEUROG2 expression was pathways. For example, transcription factors such as DTX1 relatively lower and moderate. and DLL3 increased by ten times and 16 times, respectively We also found that the gene expressions of the two (Fig. 4D). important signaling pathways related to stem cell maintenance were also affected by TSH. Ten genes from WNT pathway Other genes that were affected by TSH and 12 genes from Notch pathway were studied. Remarkably, only WNT1 within the WNT pathway was upregulated by To find the induction of gene expression that was influenced K almost 11 times at 10 8 M TSH (Fig. 4C). On the other by the presence of the recombinant TSH in hBMSCs, we hand, we observed increases in several members of the Notch tested genes related to stem cell differentiation and to the www.endocrinology-journals.org Journal of Endocrinology (2012) 212, 169–178

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access 174 E U BAGRIACIK and others . TSHR expression by mesenchymal stem cells

AB 25 ABCG2 HSPA9 MYST1 MYST2 NEUROG2 SOX1 SOX2 ALDH1A1 * 20 ALDH2 25 FGFR1 * 20 15

15 10 * 10 * * 5 * * 5 *

Fold change (relative to control) change (relative Fold 0 0 Fold change (relative to control) change (relative Fold –5 –5 –7 –7 10 M TSH 10–8 M TSH 10 M TSH 10–8 M TSH

CDADAR APC AXINI BTRC CCND1 DLL1 DLL3 DTX1 DTX2 DVL1 EP300 15 FRAT1 FZD1 MYC PPARD WNT1 20 KAT2A HDAC2 JAG1 NOTC1 NOTC2 NUMB

* * 15 10 * 10 * 5 * 5 * *

0 0

Fold change (relative to control) change (relative Fold * *

Fold change (relative to control) change (relative Fold * *

* –5 –5 –7 –7 10 M TSH 10–8 M TSH 10 M TSH 10–8 M TSH Figure 4 Regulation of gene expression of hBMSCs. (A) Metabolic markers: ATP-binding cassette sub-family G (white) member 2 (ABCG2), aldehyde dehydrogenase 1 family, member A1 (ALDH1A1), aldehyde dehydrogenase 2 family (mitochondrial) (ALDH2), fibroblast growth factor receptor 1 (FGFR1). (B) Self-renewal markers of hBMSCs: heat-shock 70 kDa protein 9 (HSPA9), MYST histone acetyltransferase 1 (MYST1), MYST histone acetyltransferase 2 (MYST2), neurogenin 2 (NEUROG2), sex-determining region Y (SRY)-box 1 (SOX1), sex-determining region Y (SRY)-box 2 (SOX2). (C) Gene expression for signaling pathways related to stem cell maintenance such as WNT pathway: RNA- specific (ADAR), adenomatous polyposis coli (APC), axin 1 (AXIN1), beta-transducin repeat containing (BTRC), cyclin D1 (CCND1), frequently rearranged in advanced T-cell lymphomas (FRAT1), frizzled homolog 1 (Drosophila) (FZD1), V-myc myelocytomatosis viral oncogene homolog (avian) (MYC), peroxisome proliferator-activated receptor delta (PPARD), wingless-type MMTV integration site family, member 1 (WNT1). (D) Notch pathway: delta-like 1 (Drosophila) (DLL1), delta- like 3 (Drosophila) (DLL3), Deltex homolog 1 (Drosophila) (DTX1), Deltex homolog 2 (Drosophila) (DTX2), Dishevelled, dsh homolog 1 (Drosophila) (DVL1), E1A binding protein p300 (EP300), GCN5 general control of amino-acid synthesis 5-like 2 (yeast) (GCN5L2 or KAT2A), 2 (HDAC2), Jagged 1 (JAG1), Notch homolog 1, translocation-associated (NOTCH1), Notch homolog 2 (NOTCH2), Numb homolog (NUMB) *(P!0.05).

signaling pathways being important for stem cell mainten- expression at mRNA level may not reflect protein expression. ance. Mesenchymal cell lineage markers, embryonic cell Therefore, we first studied protein expression for COL2A1 lineage markers, and neuronal cell lineage markers were and COL9A1. Assessment of protein expression was included among the stem cell differentiation markers. TSH performed using immunohistochemistry. Our findings treatment changed expression of markers such as COL2A1 showed that COL2A1 protein expression elevated in TSH- and COL9A1. In particular, COL2A1 expression increased treated cells (Fig. 5B). However, COL9A1 protein expression almost 26-fold whereas COL9A1 increased about eightfold was less than COL2A1 expression (Fig. 5A). In supporting the (Table 2). Interestingly, TSH treatment abrogated alkaline protein expression studies, we treated cells with TSH for phosphatase (ALPI) expression by more than 200-fold decline 21 days in order to follow differentiation process. Treated while it did not affect PPARG expression significantly. MSCs stained alcian blue positive (Fig. 5C). These data Decrease in aggrecan gene expression was important as well suggested that MSCs had capacity to differentiate into (Table 2). These data suggested that TSH might contribute to chondrocyte-like cells in the presence of TSH. the regulation of differentiation processes by which mesench- Among the embryonic stem cell lineage markers tested, ymal stem cells differentiate into chondrocyte-like cells in significant upregulations in ACTC1, ASCL2, PDX1, and cultures. However, one may argue that increase in gene T expression occurred. ASCL2 elevated by ninefold while

Journal of Endocrinology (2012) 212, 169–178 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access TSHR expression by mesenchymal stem cells . E U BAGRIACIK and others 175

Table 2 Induction of gene expression for stem cell differentiation markers

Fold change K K Genbank Function 10 7 Ma 10 8 Ma

Genes Mesenchymal cell lineage markers ACAN NM_001135 Aggrecan K1.45 K3.53* ALPI NM_001631 Alkaline phosphatase, intestinal K226.28* K103.39* BGLAP NM_199173 Bone gamma-carboxyglutamate (gla) protein (osteocalcin) C1.18 K1.25 COL1A1 NM_000088 Collagen, type I, alpha 1 K1.61 K1.59 COL2A1 NM_001844 Collagen, type II, alpha 1 C12.71* C25.59* COL9A1 NM_001851 Collagen, type IX, alpha 1 C4.22* C7.45* PPARG NM_015869 Peroxisome proliferator-activated receptor gamma C1.32 C1.96 Neural cell lineage markers NCAM1 NM_000615 Neural cell adhesion molecule 1 K1.19 C1.05 OPRS1 NM_005866 Opioid receptor, sigma 1 C1.37 K1.04 S100B NM_006272 S100 calcium binding protein B C2.90* C3.45* TUBB3 NM_006086 Tubulin, beta 3 K1.11 K1.45 Embryonic cell lineage markers ACTC1 NM_005159 Actin, alpha, cardiac muscle 1 C7.66* C13.81* ASCL2 NM_005170 Achaete–scute complex homolog 2 (Drosophila) C4.88* C9.11* FOXA2 NM_021784 Forkhead box A2 C1.12 C2.21* PDX1 NM_000209 Pancreatic and duodenal 1 C4.22* C7.40* ISL1 NM_002202 ISL LIM homeobox 1 C1.67 C2.82* KRT15 NM_002275 Keratin 15 C2.08* C3.47* MSX1 NM_002448 Msh homeobox 1 C1.10 K1.04 MYOD1 NM_002478 Myogenic differentiation 1 C2.05* C7.98* T NM_003181 T, homolog (mouse) C2.59* C5.49*

*P!0.05. aTSH concentrations used to stimulate hBMSCs. ACTC1 increased by 13-fold in comparison with the control 2000). Additionally, dendritic cells and a few subsets of T cells group, which was not treated with the recombinant TSH express functional TSHR that is capable of signaling via both (Table 2). Elevations in PDX1 and T were also significant cAMP and JAK2 (Whetsell et al. 1999, Bagriacik & Klein but relatively lower (Table 2). 2000). These previously accumulated data indicate that TSHR expressed in extrathyroidal tissues probably mediates differential functions. Increased phagocytosis of fluorescent- labeled particles and elevated cytokine secretions are a Discussion A B In this study, we showed that hBMSCs expressed a functional TSHR that was able to signal through cAMP upon stimulation by a recombinant TSH. More importantly, we explored TSH-induced gene expression in hBMSCs. Our results indicated that TSH affected expression patterns of many genes involved in various developmental and differen- tiation pathways in mesenchymal stem cell life and provided 10×10 10×10 further support for the existence of regulatory roles of C hypothalamic–pituitary–thyroid axis in the developmental stages of hBMSCs. It is very well known that stimulation of TSHR by TSH results eventually in production of THs, T3 and T4 in thyroid cells (Toccafondi et al. 1982, Szkudlinski et al. 2002). However, we also know that TSHR can be expressed by extrathyroidal tissues or cells (Haraguchi et al. 1996, Hoermann 1996, Dutton et al. 1997, Bell et al. 2000, Agretti Figure 5 Chondrogenic-like cell differentiation and protein et al. 2002, 2005, Tsai et al. 2004, Ellerhorst et al. 2006). For expression of COL2A1 and COL9A1. (A) Immunohistochemistry for COL9A1 detection. (B) Immunohistochemistry for COL2A1 example, kidney cells were shown to express a functional detection. (C) Alcian blue staining of differentiated cells by TSH TSHR that transmitted signals through cAMP (Sellitti et al. stimulation. www.endocrinology-journals.org Journal of Endocrinology (2012) 212, 169–178

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access 176 E U BAGRIACIK and others . TSHR expression by mesenchymal stem cells

few examples for variable consequences of TSH-induced COL2A1 and COL9A1 are among the lineage markers for gene expression in dendritic cells (Bagriacik & Klein 2000). mesenchymal stem cells. COL9A1 gene encodes one of the In fact, our results in this study showed that stimulation via four a chains found in type IX collagen. COL2A1 is the gene TSHR induced a variety of gene expressions, at least at encoding for the a chain of type II collagen. Type IX collagen mRNA level. Particularly, genes involved in self-renewal and is closely associated with type II collagen in cartilage maintenance were affected by the presence of TSH. components (Wu et al. 1992, Eyre et al. 2004). Members of Several studies showed that SOX2 and OCT4 gene the SOX family have been found to be responsible for the expressions were required for self-renewal process in transcriptional regulation of type II and IX collagen genes embryonic stem cells (Niwa et al. 2000, Ivanova et al. 2006). including COL2A1 and COL9A1 (Bell et al. 1997, Stokes Additionally, forced expression of SOX2 was related to et al. 2001). Particularly, SOX9 played an important role in maintenance of expansion and differentiation capabilities of COL9A1 gene expression (Zhang et al. 2003). Therefore, one hBMSCs (Go et al. 2008) implying that maintenance of can argue that via members of SOX family, TSH may be expansion governed by SOX2 would be regulated by TSH– involved in the regulation of collagen synthesis during TSHR interactions. Also, we know that SOX2 is in the cartilage generation. In fact, Miura et al. (2002) showed that regulation of FGF4 expression in embryonic stem cells THs enhanced cartilage differentiation in the growth plate in (Johnson et al. 1998, Kamachi et al. 2000, Gao et al. 2009). organ-cultured mouse tibias. By doing so, longitudinal bone Our results showed that TSH upregulated expressions of both growth was stimulated. They also found that T3 exposure FGF4 and FGF3 significantly. Therefore, we speculated that increased expression of type II collagen gene in ATDC5 cells, TSH would affect FGF4 expression probably by increasing a mouse chondrogenic cell line. Their findings about THs SOX2 expression through an indirect way. However, it having functional effects in chondrocyte differentiation would remains to be investigated. be strong support for our study. However, our argument

TSH Adenylate cyclase TSHR Gs Gq ATP cAMP Phospholipase C

JAK2 PKA PIP2 ???

IP3 DAG Ca2+ PKC Up- or down-regulation of transcription factors and others SOX2, DTX1, WNT1, ALPI FGF3, FGF4, GDF3, TERT, ACTC1, GDF4, COL2A1, COL9A1, etc.

Self-renewal differentiation maintenance Figure 6 A model for associated functions with TSHR expression in hBMSCs. The TSHR is a G protein- coupled receptor. When TSH binds to its receptor, subunits of the G protein activate some biochemical pathways as shown in the figure. As a consequence of these activations, expressions of some transcription factors or some growth factors are regulated in processes of self-renewal, differentiation, and maintenance. TSH receptor (TSHR), G protein s (Gs), G protein q (Gq), protein kinase A (PKA), phosphatidyl inositol 4,5-bisphosphate (PIP2), inositol 1,4,5-trisphosphate (IP3), diacylglycerol (DAG), protein kinase C (PKC), and Janus kinase 2 (JAK2).

Journal of Endocrinology (2012) 212, 169–178 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access TSHR expression by mesenchymal stem cells . E U BAGRIACIK and others 177 about induction of chondrogenic lineage activities in that TSH and TSHR interaction was related to induction of hBMSCs is a consequence of a direct TSH–TSHR several genes regulating self-renewal, cellular maintenance, interaction rather than a TH–TH receptor interaction. and differentiation processes such as cartilage differentiation MSCs that were incubated in the presence of TSH for from hBMSCs. However, currently we do not know how the 21 days stained alcian blue positive. Alcian blue, a TSH is involved in orchestrating a hierarchy of regulated phthalocyanine, reacts with proteoglycans such as chondroitin transcription factors that are critical for self-renewal and residue, hyaluronic residue, and sulfated residue. Therefore, it differentiation pathways. The precise sequential mechanism as can be used for detection of chondrogenic cell differentiation the consequences of interactions between the TSHR and (Quintarelli et al. 1964, Scott et al. 1964, Cowman et al. 1984). TSH is under investigation in our laboratories. To support the Significant increases in alcian blue positivity and COL2A1 results of our in vitro experimental system, we plan to design expression in TSH-treated cells were indications of chon- in vivo studies in the future. drogenic-like cell differentiation. Taken all together, these In conclusion, this study has provided novel information data suggested that TSH–TSHR interaction enhanced regarding the regulatory role of neuroendocrine interactions commitment of hMSCs to differentiate toward chondro- of the hypothalamic–pituitary axis in human mesenchymal genic-like cells. However, further studies are also necessary to stem cell biology, self-renewal, and differentiation. TSH– understand whether TSH–TSHR interaction may promote TSHR interaction would be one particular mechanism osteogenic differentiation as well. of triggering those regulatory events. Future studies should It is believed that WNT, FGF, Notch, Hedgehog, and clarify the remaining questions in establishing the biological TGFh/BMP signaling network is involved in the mainten- importance of the TSHR expression by hBMSCs. ance of tissue homeostasis not only by regulating self-renewal of normal stem cells but also by regulating proliferation or differentiation of progenitor cells as well (Katoh & Katoh Declaration of interest 2007). We found that WNT1 gene of the WNT pathway was The authors declare that there is no conflict of interest that could be perceived upregulated significantly by TSH. Also TSH induced gene as prejudicing the impartiality of the research reported. expression of important members of the Notch signaling pathway. In particular, inductions in expression of DLL3 and DTX1 were significantly high. Current literature has no Funding satisfactory information concerning a distinct correlation between those genes and any selective function. In fact, we This study was supported by a governmental grant from the State Planning observed selective and important upregulations among Organization in Turkey (DPT 2003K120470-37 and 2009K120670). members of WNT, FGF,Notch, and BMP signaling network. Many other genes, which were known to be important in Acknowledgements developmental events in stem cell’s life, were affected by the recombinant TSH treatment. Therefore, one can argue that The authors thank Prof. Dr Sevim Ercan for her kind help in the confocal TSH indeed may contribute to maintenance and the microscopy experiments. They also thank Prof. Dr John R Klein from regulations of self-renewal and tissue homeostasis of hBMSCs The Dental Branch of The University of Texas at Houston for his critical during their proliferation and differentiation process. As a review of the manuscript and for his kind advice. support for our speculation, Kim et al. (2007) found that TSH increased expression of WNT1 gene in FRTL5 cells, a rat thyroid cell line. The constitutive expression pattern of the TSHR References suggested that the TSHR might be required to play an Abdallah BM & Kassem M 2008 Human mesenchymal stem cells: from basic essential role in self-renewal, maintenance, and differentiation biology to clinical applications. Gene Therapy 15 109–116. (doi:10.1038/sj. processes as a mediator of activating a variety of pathways. In gt.3303067) establishing functions associated with the expression of the Agretti P, Chiovato L, De Marco G, Marcocci C, Mazzi B, Sellari- TSHR, we proposed a primitive model of signaling events Franceschini S, Vitti P, Pinchera A & Tonacchera M 2002 Real-time PCR (Fig. 6) based on the data we obtained in the study. This new provides evidence for thyrotropin receptor mRNA expression in orbital as model would serve as a base to establish further molecular well as in extraorbital tissues. European Journal of Endocrinology 147 733–739. (doi:10.1530/eje.0.1470733) pathways as a part of neuroendocrine interactions of the Agretti P, De Marco G, De Servi M, Marcocci C, Vitti P, Pinchera A & hypothalamic–pituitary axis. Tonacchera M 2005 Evidence for protein and mRNA TSHr expression in Taken all together, experimental system in this study has fibroblasts from patients with thyroid-associated ophthalmopathy (TAO) proven that hBMSCs expressed a functional TSHR. This is after adipocytic differentiation. European Journal of Endocrinology 152 the first original report about the expression of TSHR by 777–784. (doi:10.1530/eje.1.01900) Bagriacik EU & Klein JR 2000 The thyrotropin (thyroid-stimulating hBMSCs and also indicates the induction of various genes hormone) receptor is expressed on murine dendritic cells and on a subset of that are involved in regulations of critical pathways in the CD45RB high lymph node T cells: functional role for thyroid-stimulating mesenchymal stem cell life. Our findings strongly suggested hormone during immune activation. Journal of Immunology 164 6158–6165. www.endocrinology-journals.org Journal of Endocrinology (2012) 212, 169–178

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access 178 E U BAGRIACIK and others . TSHR expression by mesenchymal stem cells

Bell DM, Leung KK, Wheatley SC, Ng LJ, Zhou S, Ling KW, Sham MH, Kolf CM, Cho E & Tuan RS 2007 Mesenchymal stromal cells. Biology of Koopman P, Tam PP & Cheah KS 1997 Sox9 directly regulates the type-II adult mesenchymal stem cells: regulation of niche, self-renewal and collagen gene. Nature Genetics 16 174–178. (doi:10.1038/ng0697-174) differentiation. Arthritis Research and Therapy 9 1–10. (doi:10.1186/ar2116) Bell A, Gagnon A, Grunder L, Parikh SJ, Smith TJ & Sorisky A 2000 Kursawe R & Paschke R 2000 Modulation of TSHR signaling by Functional TSH receptor in human abdominal preadipocytes and orbital posttranslational modifications. Trends in Endocrinology and Metabolism fibroblasts. American Journal of Physiology. Cell Physiology 279 C335–C340. 18 199–206. (doi:10.1016/j.tem.2007.05.002) Cowman MK, Slahetka MF, Hittner DM, Kim J, Forino M & Gadelrab G Miura M, Tanaka K, Komatsu Y, Suda M, Yasoda A, Sakuma Y, Ozasa A & 1984 Polyacrylamide-gel electrophoresis and Alcian Blue staining of Nakao K 2002 Thyroid hormones promote chondrocyte differentiation in sulphated glycosaminoglycan oligosaccharides. Biochemical Journal 221 mouse ATDC5 cells and stimulate endochondral ossification in fetal mouse 707–716. tibias through iodothyronine deiodinases in the growth plate. Journal of Bone Davies T, Marians R & Latif R 2002 The TSH receptor reveals itself. Journal of and Mineral Research 17 443–454. (doi:10.1359/jbmr.2002.17.3.443) Clinical Investigation 110 161–164. Niwa H, Miyazaki J & Smith AG 2000 Quantitative expression of Oct-3/4 Dutton CM, Joba W, Spitzweg C, Heufelder AE & Bahn RS 1997 defines differentiation, dedifferentiation or self-renewal of ES cells. Nature Thyrotropin receptor expression in adrenal, kidney, and thymus. Thyroid Genetics 24 372–376. (doi:10.1038/74199) 7 879–884. (doi:10.1089/thy.1997.7.879) Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, Ellerhorst JA, Sendi-Naderi A, Johnson MK, Cooke CP, Dang SM & Moorman MA, Simonetti DW, Craig S & Marshak DR 1999 Multilineage Diwan AH 2006 Human melanoma cells express functional receptors for potential of adult human mesenchymal stem cells. Science 284 143–147. thyroid-stimulating hormone. Endocrine-Related Cancer 13 1269–1277. (doi:10.1126/science.284.5411.143) (doi:10.1677/erc.1.01239) Quintarelli G, Scott JE & Dellovo MC 1964 The chemical and histochemical Eyre DR, Pietka T, Weis MA & Wu JJ 2004 Covalent cross-linking of the properties of Alcian Blue. II. Dye binding of tissue polyanions. Histochemie NC1 domain of collagen type IX to collagen type II in cartilage. Journal of 4 86–98. (doi:10.1007/BF00306150) Biological Chemistry 279 2568–2573. (doi:10.1074/jbc.M311653200) Scott JE, Quintarelli G & Dellovo MC 1964 The chemical and histochemical Franc¸ois S, Bensidhoum M, Mouiseddine M, Mazurier C, Allenet B, properties of alcian blue I. the mechanism of alcian blue staining. Semont A, Frick J, Sache´ A, Bouchet S, Thierry D et al. 2006 Local Histochemie 4 73–85. (doi:10.1007/BF00306149) irradiation not only induces homing of human mesenchymal stem cells Sellitti DF,Akamizu T,Doi SQ, Kim GH, Kariyil JT,Kopchik JJ & Koshiyama H at exposed sites but promotes their widespread engraftment to multiple 2000 Renal expression of two ‘thyroid-specific’ genes: thyrotropin receptor organs: a study of their quantitative distribution after irradiation damage. and thyroglobulin. Experimental Nephrology 8 235–243. (doi:10.1159/ Stem Cells 24 1020–1029. (doi:10.1634/stemcells.2005-0260) 000020674) Gao F, Kwon SW, Zhao Y & Jin Y 2009 PARP1 poly (ADP-ribosylates Sox2 Stokes DG, Liu G, Dharmavaram R, Hawkins D, Piera-Velazquez S & to control Sox2 protein levels and FGF4 expression during embryonic stem Jimenez SA 2001 Regulation of type-II collagen gene expression during cell differentiation. Journal of Biological Chemistry 284 22263–22273. (doi:10. human chondrocyte de-differentiation and recovery of chondrocyte- 1074/jbc.M109.033118) specific phenotype in culture involves Sry-type high-mobility-group box Go MJ, Takenaka C & Ohgushi H 2008 Forced expression of Sox2 or Nanog (SOX) transcription factors. Biochemical Journal 360 461–470. (doi:10.1042/ in human bone marrow derived mesenchymal stem cells maintains their 0264-6021:3600461) expansion and differentiation capabilities. Experimental Cell Research 314 Szkudlinski MW, Fremont V, Ronin C & Weintraub BD 2002 Thyroid- 1147–1154. (doi:10.1016/j.yexcr.2007.11.021) stimulating hormone and thyroid-stimulating structure– Haraguchi K, Shimura H, Lin L, Saito T, Endo T & Onaya T 1996 Functional function relationship. Physiological Reviews 82 473–502. expression of thyrotropin receptor in differentiated 3T3-L1 cells: a possible Toccafondi R, Rotella CM, Tanini A, Aterini S, Borrelli D, Loddi L & model cell line of extrathyroidal expression of thyrotropin receptor. Arcangeli P 1982 Effects of TSH on cAMP levels and thyroid hormone Biochemical and Biophysical Research Communications 223 193–198. (doi:10. release in human thyroid ‘autonomous’ nodules: relationship with 1006/bbrc.1996.0868) iodothyronine and iodine content in thyroglobulin. Clinical Endocrinology Hoermann R 1996 Stimulation of thyroidal and extrathyroidal thyrotropin 17 537–546. (doi:10.1111/j.1365-2265.1982.tb01626.x) receptors. Experimental and Clinical Endocrinology and Diabetes 104 (Suppl 4) Tsai JA, Janson A, Bucht E, Kindmark H, Marcus C, Stark A, Zemack HR & 88–91. (doi:10.1055/s-0029-1211710) Torring O 2004 Weakevidence of thyrotropin receptors in primary cultures Ivanova N, Dobrin R, Lu R, Kotenko I, Levorse J, DeCoste C, Schafer X, of human osteoblast-like cells. Calcified Tissue International 74 486–491. Lun Y & Lemischka IR 2006 Dissecting self-renewal in stem cells with (doi:10.1007/s00223-003-0108-3) RNA interference. Nature 442 533–538. (doi:10.1038/nature04915) Whetsell M, Bagriacik EU, Seetharamaiah GS, Prabhakar BS & Klein JR 1999 Johnson LR, Lamb KA, Gao Q, Nowling TK & Rizzino A 1998 Role of the Neuroendocrine-induced synthesis of bone marrow-derived cytokines Sox-2 in the expression of the FGF-4 gene in with inflammatory immunomodulating properties. Cell Immunology 192 embryonal carcinoma cells. Molecular Reproduction and Development 50 159–166. (doi:10.1006/cimm.1998.1444) 377–386. (doi:10.1002/(SICI)1098-2795(199808)50:4!377::AID- Wu JJ, Woods PE & Eyre DR 1992 Identification of cross-linking sites in MRD1O3.0.CO;2-F) bovine cartilage type IX collagen reveals an antiparallel type II–type IX Kamachi Y, Uchikawa M & Kondoh H 2000 Pairing sox off: with partners in molecular relationship and type IX to type IX bonding. Journal of Biological the regulation of embryonic development. Trends in Genetics 16 182–187. Chemistry 267 23007–23014. (doi:10.1016/S0168-9525(99)01955-1) Zhang P, Jimenez SA & Stokes DG 2003 Regulation of human Col9A1 Katoh M & Katoh M 2007 WNT signaling pathway and stem cell signaling gene expression: activation of the proximal promoter region by Sox9. network. Clinical Cancer Research 13 4042–4045. (doi:10.1158/1078-0432. Journal of Biological Chemistry 278 117–123. (doi:10.1074/jbc. CCR-06-2316) M208049200) Kemp KC, Hows J & Donaldson C 2005 Bone marrow-derived mesenchymal stem cells. Leukemia and Lymphoma 46 1531–1544. (doi:10.1080/ 10428190500215076) Kim WB, Lewis CJ, McCall KD, Malgor R, Kohn AD, Moon RT & Received in final form 22 November 2011 Kohn LD 2007 Overexpression of Wnt-1 in thyrocytes enhances cellular growth but suppresses transcription of the thyroperoxidase gene via Accepted 29 November 2011 different signaling mechanisms. Endocrinology 193 93–106. (doi:10.1677/ Made available online as an Accepted Preprint JOE-06-0025) 29 November 2011

Journal of Endocrinology (2012) 212, 169–178 www.endocrinology-journals.org

Downloaded from Bioscientifica.com at 09/29/2021 10:03:51PM via free access