저작자표시-비영리-변경금지 2.0 대한민국

이용자는 아래의 조건을 따르는 경우에 한하여 자유롭게

l 이 저작물을 복제, 배포, 전송, 전시, 공연 및 방송할 수 있습니다.

다음과 같은 조건을 따라야 합니다:

l 귀하는, 이 저작물의 재이용이나 배포의 경우, 이 저작물에 적용된 이용허락조건을 명확하게 나타내어야 합니다.

l 저작권자로부터 별도의 허가를 받으면 이러한 조건들은 적용되지 않습니다.

저작권법에 따른 이용자의 권리는 위의 내용에 의하여 영향을 받지 않습니다.

이것은 이용허락규약(Legal Code)을 이해하기 쉽게 요약한 것입니다.

Disclaimer

저작자표시. 귀하는 원저작자를 표시하여야 합니다.

비영리. 귀하는 이 저작물을 영리 목적으로 이용할 수 없습니다.

변경금지. 귀하는 이 저작물을 개작, 변형 또는 가공할 수 없습니다.

http://creativecommons.org/licenses/by-nc-nd/2.0/kr/legalcodehttp://creativecommons.org/licenses/by-nc-nd/2.0/kr/

The Role of Glycogen Synthase Kinase 3b in

Apoptosis and Differentiation of Osteoblasts

by

Sun-Il Yun

Major in Molecular Medicine

Department of Medical Sciences

The Graduate School, Ajou University

The Role of Glycogen Synthase Kinase 3b in

Apoptosis and Differentiation of Osteoblasts

by

Sun-Il Yun

A Dissertation Submitted to The Graduate School of

Ajou University

in Partial Fulfillment of the Requirements for the degree

of Ph.D. in Medical Sciences

Supervised by

Yoon-Sok Chung, M.D., Ph.D.

Major in Molecular Medicine

Department of Medical Sciences

The Graduate School, Ajou University

February, 2010

i

- ABSTRACT –

The Role of Glycogen Synthase Kinase 3b in Apoptosis

and Differentiation of Osteoblasts

Differentiation of osteoblasts is critically involved in bone formation and requires several

signal pathways. One of the important pathways is the Wnt signaling where GSK3b plays a

key role in regulation of b-catenin activity, a major effector, through phosphorylation.

Glucocorticoids (GCs) are known to induce osteoporosis via a decrease of differentiation and

apoptosis of osteoblasts and osteocytes in part. However, the detailed mechanism of GC-

induced osteoporosis has not been clearly determined.

This study showed that dexamethasone (Dex) induced apoptosis of MC3T3-E1 osteoblasts

through the GC receptor. Dex activated GSK3b, and inhibition of GSK3b by lithium, a

pharmacological antagonist, or gene knock-down by siRNA prevented the Dex-induced

apoptosis. Unexpectedly, Dex also activated p38 mitogen-activated protein kinase (p38

MAPK); however, the inhibition of p38 MAPK further increased apoptosis in Dex-treated

osteoblasts. These results suggest that p38 MAPK might protect against Dex-induced

apoptosis of osteoblasts.

Etoposide, a genotoxic agent, increased apoptosis of C3H10T1/2 osteoblast progenitor cells

by the activation of both caspase-3 and GSK3b. The pharmacological inhibition (lithium) or

gene knock-down (siRNA) of GSK3b protected the cells from apoptosis. This is quite

similar to the results of Dex-induced apoptosis. In addition, etoposide decreased expression

of Bcl-2, an anti-apoptotic protein, in the C3H10T1/2 cells. LiCl completely recovered the

ii

Bcl-2 expression as shown by both the mRNA and the protein expression levels.

It has recently been reported that Dex decreases differentiation and mineralization of

osteoblasts, but its mechanism is not certainly understood. The present study also showed

that Dex were associated with osteoblast differentiation by employing MC3T3-E1 cells, an

osteoblast cell line. Dex decreased osteocalcin mRNA level compared to that of un-treated

cells at an early differentiation period. Interestingly, Dex continuously accumulated the

expression of heat shock protein 25 (HSP25). Gene knock-down of HSP25 by siRNA

recovered osteocalcin expression up to control level. Furthermore, treatment with lithium

decreased HSP25 expression and also recovered osteocalcin expression. These results

suggest that HSP25 induction was associated with osteoblast differentiation via osteocalcin

modulation and mediated by GSK3β.

In summary, Dex-induced GSK3b activation triggered apoptosis of osteoblasts and was

associated with osteoblast differentiation. Therefore, GSK3b appears to play a key role in the

fate and function of osteoblasts. The results of this study provide additional insights into the

pathogenic mechanism of GC-induced osteoporosis.

Key words: Osteoblasts · Apoptosis · Dexamethasone (Dex) · GSK3b · p38 MAPK ·

Etoposide · Differentiation · HSP25 · Osteocalcin

iii

TABLE OF CONTENTS

ABSTRACT ······························································································ i

TABLE OF CONTENTS ············································································· iii

LIST OF FIGURES ··················································································· vii

I. INTRODUCTION ··················································································· 1

A. Wnt/b-catenin signaling ········································································· 3

B. GSK3b in apoptosis··············································································· 5

C. p38-mitogen-activated protein kinase in apoptosis ·········································· 5

D. Bcl-2 ······························································································· 6

E. Heat Shock Protein 25 ············································································ 7

F. Aims of this study ··············································································· 10 II. MATERIALS AND METHODS ································································ 11

A. MATERIALS ····················································································· 11

1. Reagents ······················································································· 11

2.Materials ························································································ 11

B. METHODS ······················································································· 12

1. Cell culture ·················································································· 12

2. Cell viability assay ··········································································12

3. Trypan blue staining ······································································· 12 4. Mitochondrial membrane potential Analysis ············································12

5. Fluorescence staining ·······································································13

iv

6. Tdt-mediated dUTP-biotin nick End Labeling staining ································13

7. Caspase-3 activity ·········································································14

8. Western blot analysis ······································································ 14 9. Small interference RNA transfection ·····················································15

10. Reverse transcription ······································································ 15

11. Polymerase chain reaction ·································································15

12. Alkaline phosphatase activity ····························································· 16

13. Alizarin red S staining ······································································ 16

14. Luciferase activity ········································································· 17

15. Statistical analysis ········································································· 17

Part I.

Glucocorticoid induces apoptosis of osteoblasts through the activation of glycogen

synthase kinase 3b ·····················································································18

III. RESULTS ···························································································19

1. Dex induces apoptosis of osteoblasts through caspase activation ·················· 19

2. Glucocorticoid receptor mediates Dex-induced apoptosis in osteoblasts ··········21

3. Dex induces GSK3b activation in osteoblasts ········································· 24 4. GR stimulated by Dex reglates the phosphorylation of GSK3b ···················· 25

5. Dex activates p38-MAP in osteoblasts ·················································· 25

6. Gene knock-down of GSK3b and p38-MAPK ········································ 31

IV. DISCUSSION ···················································································· 33

v

Part II.

Glycogen synthase kinase-3beta regulates etoposide-induced apoptosis

via Bcl-2 mediated caspase-3 activation in C3H10T1/2 cells································· 37

III-B. RESULTS ························································································ 38

1. Etoposide induces apoptosis in C3H10T1/2 cells ····································· 38

2. Etoposide induces osteoblast apoptosis via GSK3b activation, which is prevented

by gene silencing ···········································································41

3. Etoposide induces apoptosis via GSK3b activation, which is prevented by gene

silencing······················································································ 44

4. Bcl-2 is mediated by GSK3b in etoposide-induced apoptosis ······················· 47

IV. DISCUSSION ····················································································· 49

Part III.

Dexamethasone regulates osteocalcin via heat shock protein 25 and GSK-3β

in osteoblast Differentiation ······································································· 51

III. RESULTS ·························································································· 54

1. The expression of HSP25 during osteoblasts differentiation ························· 58

2. Effects of Dex on osteocalcin expression in osteoblasts ······························ 58

3. Effects of knock-down of HSP25 on osteocalcin in osteoblasts ····················· 58

4. Lithium controls expression of HSP25 and osteocalcin ······························· 62

IV. DISCUSSION ····················································································· 63

vi

V. CONCLUSION ····················································································· 67

REFERENCES························································································· 69

국문요약 ····························································································· 79

vii

LIST OF FIGURES

Fig. 1. The processes of osteoblast and osteoclast differentiation ································ 1

Fig. 2. Effects of glucocorticoid on bone cells ······················································ 2

Fig. 3 Wnt signaling pathway in the presence of Wnt ············································· 4

Fig. 4. Protein structure and homolog of HSP25 ·················································· 8

Fig. 5. Biochemical activity and cellular properties of HSP25 ································· 9

Fig. 6. Hypothesis of this study ······································································ 10

Part I.

Fig. 7. Dex induced apoptosis in mouse osteoblast cells ······································· 20

Fig. 8. Inhibition of glucocorticoid receptor reduced Dex-induced apoptosis ················ 22

Fig. 9. The activation of GSK3b in MC3T3-El cells by Dex ··································· 26

Fig. 10. The phosphorylation of GSK3b by GR antagonist ····································· 28

Fig. 11. Expression of phospho-p38-MAPK and inhibition of p38-MAPK in Dex-induced

apoptosis ······················································································· 29

Fig. 12. Effects of gene silencing of GSK3b and p38-MAPK on osteoblasts apoptosis ···· 32

Fig. 13. The suggested scheme of Dex-induced apoptosis of osteoblasts ···················· 36

viii

Part II.

Fig. 14. Etoposide induced apoptosis in C3H10T1/2 cells ······································ 39

Fig. 15. Etoposide induced GSK3b activation in C3H10T1/2 cells ··························· 42

Fig. 16. Effects of GSK3b siRNA on etoposide-induced apoptosis ···························· 43

Fig. 17. LiCl inhibited caspase-3 activation and apoptotic cell death in etoposide-induced

apoptosis ····················································································· 45

Fig. 18. Effects of LiCl on etoposide- or staurosporine-induced apoptosis ··················· 46

Fig. 19. Effect of LiCl on Bcl-2 expression in etoposide-induced apoptosis ·················· 48

Part III.

Fig. 20. Effects of Dex on mineralization in MC and MC14 cells ······························ 53

Fig. 21. Effects of Dex on Hsp25 expression during osteoblast differentiation in MC

and MC14 cells ·············································································· 54

Fig. 22. Dex increased HSP25 expression in MC cells ·········································· 55

Fig. 23. Effects of Dex on osteoblast differentiation markers in MC cells ···················· 57

Fig. 24. Osteocalcin expression in MC and MC14 cells ········································· 58

Fig. 25. Effects of HSP25 siRNA on gene expression and ALP activity in MC cells ······· 59

Fig. 26. Lithium inhibited Hsp25 expression induced by Dex in MC cells ··················· 60

Fig. 27. Effects of lithium on the HSP25 and osteocalcin expression in MC cells············ 61

ix

Fig. 28. Dex decreased the osteocalcin activity in MC cells ····································· 62

Fig. 29. Dex signaling was transduced through GR in MC cells ······························· 65

Fig. 30. Schematic signaling in osteoblast differentiation ······································ 66

Fig. 31. Schematic diagram of this study ··························································· 68

- 1 -

I. INTRODUCTION

Bone is a rigid tissue. However, it has dynamic properties. The bone can respond to

mechanical stresses and regulates calcium homeostasis. Bone consists of three types of

cells; osteoblasts, osteocytes, and osteoclasts. Bone cells grow and differentiate, and

apoptosis occurs as part of in this process. Initially, osteoblasts are derived from

mesenchymal stem cells (MSCs). However, the osteoclasts from hematopoietic stem cells

(HSCs) and the osteocytes in bone matrix are differentiated from osteoblasts (Fig. 1).

Osteoblasts play an important role in bone formation, while osteoclasts are involved in

bone resorption (Stein et al, 2004). Both osteoblasts and osteoclasts are involved in bone

remodeling. These cells closely communicate by direct and indirect interactions in a

continuous process maintaining bone homeostasis (Matsuo and Irie, 2008).

A.

B.

Fig. 1. The processes of osteoblast and osteoclast differentiation. Osteoblasts (A) and

osteoclasts (B) (modified from Stein et al, 2004 (A); Matsuo and Irie, 2008 (B)).

- 2 -

Glucocorticoids (GCs), one of the steroid hormones, are secreted from the adrenal

glands in response to stress; they have a variety of effects on the regulation of metabolism

(Molina, 2006). In addition, GCs are widely prescribed for a variety of human diseases

such as autoimmne arthritis, allergic diseases, and asthma. However, patients that are

required to take GCs over a period time develop osteoporosis. Excess GC impairs bone

remodeling that has direct effects on bone cell number and function (Fig. 2; Canalis,

1996; Hurson et al., 2007). Even modest doses of GCs have considerable effects at the

cellular and molecular levels (Canalis and Delany, 2002). However, the molecular

mechanism of GCs in apoptosis and differentiation of osteoblasts is poorly understood.

Fig. 2. Effects of glucocorticoids on bone cells. GCs have a variety of effects on bone cells:

osteoblasts, osteocytes, and osteoclasts. Prolonged exposure to GC leads to a malfunction of

the cells and increases fracture risk (modified from Canalis et al., 2007).

- 3 -

A. Wnt/b-catenin signaling

The Wnt/b-catenin pathway plays a central role in biological processes, including

embryogenesis, morphogenesis, organogenesis, and tumorigenesis. MSCs give rise to

osteoblasts, myocytes, chondrocytes, and adipocytes. However, once Wnt signaling is

activated, it initiates osteoblastogenesis and inhibits differentiation from the MSCs into

myocytes, chondrocytes, or adipocytes (Stein et al., 2004). In bone, the Wnt pathway is

associated with development, bone mass homeostasis, and apoptosis (Pinzone et al.,

2009; Bodine, 2008). However, abnormal Wnt signaling is associated with human disease

such as osteoporosis. Thus, this signaling could be used as a therapeutic target.

There are two Wnt pathways; the canonical and non-canonical pathways. Generally, in

the canonical pathway, Wnt binds to one of the frizzled (FZD) receptor families with low-

density lipoprotein-related protin (LRP)-5/LRP-6. Subsequently, Wnt-FZD-LRP activates

Dishevelled (Dvl) and antagonizes glycogen synthase kinse (GSK) 3b activity. Then, b-

catein is translocated into the nucleus, and binds to the T-cell factor/lymphoid enhancer

(Tcf/Lef) for target gene transcription (Fig. 3). However, in the absence of Wnt, b-catenin

is sequentially phosphorylated by casein kinase 1α (CK1 α) and GSK3b. The

phosphorylated b-catenin forms a complex with scaffolding proteins; adenomatous

polyposis coli (APC) and Axin. The b-transducin repeat containing protein (b-Trcp)

recognizes phosphorylated b-catenin, and the proteasome is consequently degraded. The

Wnt pathway is also activated through b-catenin-independent signaling. This signaling is

part of the called as non-canonical pathway, has not been well described.

The Wnt family consists of 18 members. For bone formation, Wnt3a, Wnt7b, and Wnt

10b stimulates osteoblast differentiation through the canonical pathway. However, Wnt5a

- 4 -

triggers the non-canonical pathway (Caetano-Lopes et al., 2007). In addition, there are

various Wnt antagonists; Wnt inhibitory factors (WIFs), secreted frizzed-related proteins

(sFRPs), dickkopf (DKK)-1, and SOST/sclerostin. In particular, GC treated osteoblasts

suppress Wnt signaling (Ohnaka et al., 2005), and enhance the expression of Dkk-1

(Ohnaka et al., 2004).

Fig. 3. Wnt signaling pathway in the presence of Wnt.

- 5 -

B. GSK3b in apoptosis

GSK3b is known to play a role in osteogenesis through the Wnt/b-catenin signaling

pathway (Krishnan et al., 2006). In addition, GSK3b is a multifunctional regulator of not

only growth but also apoptosis in various cell types (Beurel and Jope, 2006). GSK3b is

distributed in the cytosol, mitochondria, and nuclei (Jope and Johnson, 2004). GSK3b is

phosphorylated on two residues, serine-9 (Ser-9) and tyrosine-216 (Tyr-216). Its

activation and inactivation are regulated by phosphorylation at Ser-9, a critical factor for

many receptor-coupled signaling processes. In addition, this activity is affected by

intracellular localization. The phosphorylation at Tyr-216 of GSK3b is required for

GSK3b activity (Hughes, 1993). The inhibition of GSK3b leads to cell protection from

many apoptotic conditions (Song et al., 2002; Jin et al., 2005; Bijur et al., 2000). Nuclear

GSK3b is involved in the regulation of many transcription factors that influence several

signaling pathway. This results in the control of the expression levels of various genes.

Furthermore, the nuclear accumulation of GSK3b by apoptotic stimuli increases the

interactions between GSK3b and nuclear substrates (Bijur and Jope, 2001).

C. p38-mitogen-activated protein kinase in apoptosis

A general stress response pathway is internalized through the mitogen-activated protein

kinase (MAPK) pathway. p38 is a member of MAPK family . Among various cell types,

p38 MAPK has been linked to the immune response, cellular stress response, apoptosis,

cell growth, survival, and differentiation (Zaurubin and Han, 2005). p38 MAPK is

involved in cell-type specific functions (Wagner and Nebreda, 2009). The p38 MAPK is

- 6 -

activated by phosphorylation. Its activation protects cells from various apoptotic

conditions; acidic preconditioning in endothelial cells (Flacke et al., 2009), DNA damage

in murine tumor cells (Reinhardt et al., 2007), and the progression of myeloma cell in

bone marrow (Gaundar et al., 2009). p38 MAPK also enhances apoptosis under certain

conditions (Lu et al., 2006; Hsu et al., 2007; Osamu et al., 2002). . Dexamethasone (Dex)

activated p38 MAPK in apoptosis of lymphoid cells. Activated p38 MAPK in

osteosarcoma and osteoblasts subsequently activates caspase-3. In cell differentiation,

activation of p38 MAPK is essential for osteoblast differentiation (Hu et al., 2003). These

findings suggest that p38 MAPK may play a regulatory role in a variety of conditions.

D. Bcl-2

Apoptosis is a strongly regulated multi-step process that involves in cell death during

development, diseases, and response to drugs (Bröker et al, 2005). Bcl-2 has been shown

to control apoptosis since it was first identified in B cell lymphomas (Tsujimoto et al.,

1984). The Bcl-2 family is divided into three subfamilies by the Bcl-2 homology (BH)

domain and apoptotic function; anti-apoptotic (Bcl-2, Bcl-xL, Mcl-1), pro-apoptotic (Bax,

Bak), and the BH3-only protein family (Bim, Bid, Puma) (Willis and Adams, 2005). Bcl-

2 antagonizes the pro-apoptotic protein family. Pro-apoptotic Bax is also antagonized by

Bcl-2. Bax released from Bcl-2 has a negative effect on the mitochondria in apoptosis

(Cotter, 2009). The apoptosis of osteoblasts is a tightly regulated physiological process in

terms of the ratio of Bcl-2/Bax proteins (Agas et al., 2008).

- 7 -

E. Heat Shock Protein 25

Heat shock proteins (HSPs) are divided into high-molecular-weight HSPs (HSP100,

HSP90, HSP70 and HSP60) and small HSPs (HSP25 and α-crystallin). High-molecular-

weight HSPs are ATP-dependent chaperons, while HSP25 (homolog of human Hsp27;

Heat Shock Protein 1; HSBP1), a small HSP, is ATP-independent (Fig. 4; Kostenko and

Moens, 2009; Arrigo et al., 2007). Large HSPs, HSP70 and HSP90, are associated with

hormone signaling pathways. It is well known that GC binds to the glucocorticoid

receptor (GR). The GC-GR complex translocates into the nucleus to modulate

transcription of the target genes. In this pathway, HSP70 and HSP90 have influence on

GR processing. HSP70 participates in the initial GC-GR assembly steps, while HSP90

facilitates the efficient activation of GR. In addition, HSP25 and HSP70 provide

protection against various stimuli.

Osteoblasts induce HSP25 by inflammatory cytokines (Hayashi et al., 2008; Takai et al.,

2006; Shinji et al., 2006). HSP25 forms large oligomers under a variety of cell conditions.

Phosphorylation of HSP25 modifies the oligomer size and activity. It is reversible for the

multiple functions (Fig. 5; Garrido et al., 2006; Hock et al., 2001). HSP25 is also

involved in microfilament organization. Non-phosphorylated HSP25 monomers inhibit

actin polymerization. In addition, HSP25 plays a role in cell differentiation (Mehlen et al,

1997; Favet et al., 2001; Garrido et al., 2006), and is a key controller between

differentiation and apoptosis (Lanneau et al, 2007). In osteoblasts, HSP25 is expressed

during osteoblast differentiation (Shakoori et al., 1992), and stimulated by Dex (Osamu et

al., 2002). However, further study is needed to determine a functional role of HSP25 and

the molecular mechanisms of HSP25 action.

- 8 -

A.

B.

Fig. 4. Protein structure and homolog of HSP25. A: Difference of protein structure

between HSP70 and HSP25. B: The amino acid sequences of human (h) and rat (r) heat

shock protein 27, and the mouse (m) homologue HSP25 are aligned. *Indicates an identical

corresponding residue. Red boxes mark the phosphorylation sites Ser-15, Ser-78, and Ser-82

(resp. Ser-86) (modified from Arrigo et al., 2007; Garrido et al., 2006 (A), and taken from

Kostenko and Moens, 2009).

- 9 -

A.

B.

Fig. 5. Biochemical activity and cellular properties of HSP25. A: Schematic

conformation linked to phosphorylation of HSP25, B: The cellular response related with

HSP27 against stresses (modified from Arrigo et al., 2007 (A); Garrido et al., 2006 (B)).

- 10 -

F. Aims of this study

In this study, the apoptotic effects of Dex, a synthetic GC, on the MC3T3-E1 osteoblast

cell line was investigated. The role of two kinase enzymes, GSK3b and p38 MAPK, was

evaluated in the Dex-induced apoptosis pathway. To test the importance of GSK3b in

apoptosis, a genotoxic agent, etoposide, was studied in the osteoblast precursor cell line,

C3H10T1/2. In addition, whether Dex could induce differentiation of osteoblasts in vitro

and the role of GSK3b in the differentiation of osteoblasts was examined.

Fig. 6. Hypothesis of this study.

- 11 -

II. MATERIALS AND METHODS

A. MATERIALS

1. Reagents

(A) Chemicals: LiCl, Dexamethasone (Dex), Methylthiazolyldiphenyl-tetrazolium

bromide (MTT), 3,3'-dihexyloxacarbocyanine iodide (DiOC6(3)), mifepristone, 3,3'-

Diaminobenzidine tetrahydrochloride (DAB), 2-(4-Amidinophenyl)-6-indolecarbamidine

dihydrochloride (DAPI) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Z-

VAD-fmk was purchased from Biomol (Plymouth Meeting, PA, USA), and SB203580

from Calbiochem (San Diego, CA, USA).

(B) Medium: Minimum Essential Medium Alpha Medium (a-MEM), Dulbecco’s

Modified Eagle’s Medium (DMEM), Phosphate buffered saline (PBS), Antibiotics and

antimycotics, and Fetal bovine serum (FBS) were purchased from GIBCO BRL (Carlsbad,

CA, USA).

(C) Antibodies: GSK3b, cleaved caspase-3, p38 MAPK, b-actin, phospho-GSK3b (Ser-9),

phospho-p38 MAPK (Thr-180/Thr-182) were purchased from Cell Signaling (Boston,

MA, USA). Phospho-GSK3b (Tyr-216) was purchased from Upstate (Chicago, IL, USA),

and HSP25 and HSP70 from Stressgen (Ann Arbor, MI, USA). Horseradish peroxidase

(HRP)-conjugated antibody was purchased from Immunoresearch Laboratories (West

Grove, PA, USA) and Zymed (San Francisco, CA, USA).

- 12 -

2. Materials

Thermo Scientific Forma Series II Water Jacketed CO2 Incubator (Thermo, Waltham,

MA, USA), Light microscope (Olympus, Japan), Axiovert200 inverted fluorescent

microscope (Carl Zeiss, Jena, Germany), Eppendorf Centrifuge 5415R (Eppendorf,

Cambridge, UK), BioTek's PowerWave™ X microplate spectrophotometer (BioTek,

Winooski, VT, USA), and Model 680 microplate reader (Bio-Rad, Cambrige, UK) were

used.

B. METHODS

1. Cell culture

MC3T3-E1 cells were purchased from the RIKEN cell bank (Tsukuba, Japan), and they

were cultured in a-MEM. The osteoblast precursor cell line C3H10T1/2 was purchased

from the Korean Cell Line Bank (Seoul, Korea) and grown in DMEM supplemented with

10% heat-inactivated FBS containing 1% antibiotics and antimycotics. The cells were

incubated in a humid incubator at 37°C (95% O2 and 5% CO2) and maintained in a sub-

confluent state unless otherwise indicated.

2. Cell viability assay

Cells (3x103cells/well) were incubated in 96-well plates overnight and treated with

drugs in 10% serum medium for 48 h. MTT (1 mg/ml, Sigma) was added to the cell

medium for 4 h, and the cells were then solubilized with isopropyl alcohol. The cell

- 13 -

supernatants were measured at 570 nm with the ELISA reader (BioTek). All cells were

treated with 1uM Dex, and the cell viability was assayed after 48 h using the MTT assay.

3. Trypan blue staining

The MC3T3-E1 cells (5x104cells/well) were incubated in 12-well plates overnight and

treated with drugs in 10% serum medium for 24 h. The cells were washed with sterile

phosphate buffered saline, then treated with 0.25% trypsin-EDTA (GIBCO BRL) and

harvested. The cells were diluted in 0.1% trypan blue (GIBCO BRL) and then counted

under a light microscope.

4. Mitochondrial membrane potential analysis

The mitochondrial membrane potential (MMP) was determined by the dye retention

method. The MC3T3-E1 cells (3x105cells/dish) were plated onto 60 mm culture dishes.

After 12 h, Dex was added to cells for 24 h. The cells were loaded with 40 nM DiOC6(3)

during the last 15 min of treatment, and then analyzed with a fluorescence microscope

(Carl Zeiss).

5. Fluorescence staining

The MC3T3-E1 cells were grown in 12-well plates for 24 h with or without a drug. The

cells were washed once with PBS and fixed with MeOH: Aceton (= 2:1) for 10 min at -

- 14 -

20°C. After washing with PBS, the cells were incubated with DAPI (1 mg/ml, Sigma) for

10 min in the dark at room temperature. The cells were washed again and mounted, and

images were captured using the fluorescence microscope (Carl Zeiss).

6. Tdt-mediated dUTP-biotin nick end labeling staining

The cells were grown in 12-well plates and treated with or without a drug for 24 h.

Apoptotic cell death was quantified by ApopTag® (Chemicon, Temecula, CA, USA)

according to the manufacturer’s protocol. Briefly, the cells were washed once with PBS

and fixed with 1% paraformaldehyde (Sigma) for 30 min at room temperature. After

washing with PBS, the cells were post-fixed with acetic acid: EtOH (= 2: 1) for 5 min,

washed twice with PBS, and then quenched with 3% hydrogen peroxide solution for 5

min at room temperature. After equilibrium buffer was applied, the cells were incubated

with working strength TdT enzyme for 1 h at 37°C, and stop buffer was added for 10

min at room temperature. Then, anti-digoxigenin peroxidase conjugate was applied on

cells and incubated for 30 min at room temperature. Peroxidase substrate, DAB, was

added for 3-6 min, and the cell suspension was then washed with distilled water. Nuclei

were counterstained with 0.5% methyl green for 10 min, washed again with distilled

water, and mounted. Images were captured using a light microscope (Olympus, Tokyo,

Japan), and Tdt-mediated dUTP-biotin nick End Labeling (TUNEL) stained cells were

then counted.

- 15 -

7. Caspase-3 activity

The C3H10T1/2 cells were plated onto 100 mm culture dishes in the presence or

absence of a drug for 24 h. The cells were harvested, resuspended in cell lysis buffer, and

subjected to 3 freeze-thaw cycles at -70°C. Cell lysates were incubated on ice for 15 min,

the supernatant fractions were collected, and the protein concentrations were estimated

using the Bradford method (Bio-Rad, Hercules, CA, USA). Caspase-3 activities were

analyzed with equal protein concentrations of cell lysates (300-500 mg per sample) using

CaspACETM assay system kit (Promega, Madison, WI, USA) according to the

manufacturer’s protocol. The activity was measured at 405 nm with the ELISA reader

(BioTek).

8. Western blot analysis

The cells were grown overnight in 100-mm dishes to a density of 5-7x105 viable cells

per dish, and then pretreated with each drug for 1 h before the addition of Dex. At various

time points, the cells were harvested and analyzed by immunoblot analysis. The cells

were harvested with ice-cold PBS and solubilized in RIPA lysis buffer (Upstates,

Temecula, CA, USA) supplemented with a mixture of protease inhibitors (1 mM PMSF, 1

mg/ml aprotinin, 1 mg/ml leupeptin, 1 mg/ml pepstatin, 1 mM DTT) and protease

inhibitors (50 mM NaF). Cell debris was removed by centrifugation at 13,000 rpm for 10

min. Clarified samples were collected and the protein concentration was then estimated

using the Bradford method (Bio-Rad, Hercules, CA, USA). Whole cell lysate (50-80 mg)

was mixed with 5x sample buffer and boiled for 10 min. The samples were separated by

- 16 -

10% to 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)

and transferred onto polyvinylidene fluoride (PVDF) membranes (Amersham,

Burkinghamshire, UK). The membranes were washed with 0.1% Tween 20 in TBS [140

mM NaCl, 20 mM Tris (pH 7.4)] and blocked with 5% skim milk. Immunoblots were

performed with specific antibodies according to the manufacturer’s instructions (Cell

Signaling), and the membranes were probed with HRP-conjugated antibody. The

membrane was developed using enhanced chemiluminescence (Amersham) and exposed

to X-ray film.

9. Small interference RNA transfection

The cells were grown overnight in 12-well plates to a density of 5x104 viable cells per

well, and the siRNA of GSK3b (Cell signaling) or the siRNA of HSP25 (designed by

IDT, Coralville, IA, USA) was then transfected with the TransIT-siQUESTÒ transfection

reagent (Mirus Bio Co., Madison, WI, USA) according to the manufacturer’s protocol.

After 24 hours, the cells were harvested and analyzed by immunoblot or RT-PCR

(Reverse transcriptase-PCR).

10. Reverse transcription

The cells were grown on culture dishes and the total RNA was isolated from the cultured

cells using TrisolÒReagent (Invitrogen, Carlsbad, CA, USA) according to the

manufacturer’s protocol. Reverse transcription to synthesize cDNA was carried out in 20

- 17 -

ml reactions using 0.5mg of RNA for one hour at 42°C, primed with an oligo(dT) primer

(GeneDEPOT, San Jose, CA, USA).

11. PCR

All other PCR conditions and reagents in the Maxim PCR premix kit were used for the

experiments (i-StarTaq) (iNtRON, Seongnam, South Korea). PCR was performed to

generated the genes with 10 pM each primer as follows: Bcl-2 sense, 5’-

GGGAGAACAGGGTATGATAA-3’; antisense, 3’-GTAGCGACGAGAGAAG

TCAT-5’; HSP25 sense, 5’-TGTCCCTGGACGTCAACCACTT-3’; antisense, 3’-

ACCTGGAGGGAGCGTGTATTTC-5’(Ho et al., 2006); osteocalcin sense, 5’-TTGAAG

ACCGCCTACAAACG-3’ ; antisense, 3’-ACATCCATACTTGCAGGGCA-5’ ;

osteopontin antisense, 5’-CTTTACAGCCTGCACCCAGA -3’; sense, 3’-TCCATGTGG

TCATGGCTTTC-5’; Collagen type I sense, 5’-GCAATCGGGATCAGTACGAA-3’,

antisense, 3’-CTTTCACGCCTTTGAAGCCA-5’ (Marzia et al., 2000); Runx2 antisense,

5’-ATGAGAGTAGGTGTCCCGCC-3’; sense, 3’-GAAGGCCAGAGGCAGAAGTC-5’;

GAPDH antisense, 5’-CTCACTCAAGATTGTCAGCA-3’, sense, 3’-GTCATCATACTT

GGCAG GTT-5’. PCR amplification was performed for 23-28 cycles at 94°C for 20 s,

55-60°C for 10 s, and 72°C for 30 s, followed by a 5 min at 72°C. The PCR product were

separated on a 1.5% agarose gels and visualized by staining with ethidium bromide.

Densitometric values for each band were quantified using the Image Gel-Doc (Bio-Rad).

- 18 -

12. Alkaline phosphatase activity

To measure the alkaline phosphatase (ALP) activity, the MC3T3-E1 cells were seeded

on 12-well plates. After the cells were grown for 2 days, they were replaced with

osteogenic induction medium that was supplemented with ascorbic acid (50 μg/ml,

Sigma) and glycerol 2-phosphate (10 mM, Sigma) with and without Dex in α–MEM

media. After 3 days of culture, the cells were washed with PBS, harvested, and then lysed

with NaHCO3 buffer (0.25 M, pH 7.5) containing 0.01% Triton X. The cells were

sonicated for 5 seconds three times on ice and the cell lysates were then centrifuged at

13,000 rpm for 10 min. An equal volume of supernatant was added to the substrate buffer

[0.1 M NaHCO3 (pH 10.3) and 1 mM MgCl2 (Sigma)]. The ALP activity was determined

using the supernatant reacted with substrates, p-nitorphenyl phosphate (Biochemica, UK).

After incubation of the mixture at 37°C for 20-60 min, absorbance was measured at 405

nm. The activity was standardized by the protein content using the Bradford method and

expressed as fold increase compared to the control.

13. Alizarin red S staining

To evaluate the effect of Dex on the matrix mineralization of the MC3T3-E1 cells, the

cells were seeded on 6-well plates. After the cells were grown for 2 days, they were

replaced with and without Dex in the osteogenic induction medium. Every 2 or 3 days the

osteogenic medium was exchanged. After 30 days of culture, the cells were fixed with

100% EtOH at room temperature, washed with PBS and stained with 40 mM alizarin red

S (Sigma, pH 4.2) for 10 min at room temperature. The stained plates were washed with

- 19 -

distilled water, dried, and alizarin red S stained wells of each group were captured.

14. Luciferase activity

The MC3T3-E1 cells were grown overnight in 12-well plates to a density of 5x104

viable cells per well. The pGL2-basic was used as an internal control (Promega), or

pGL2-OSE (that was kindly provided by professor Choi JY), the osteocalcin gene

promoter fused to luciferase, was transfected into the cells by ExGen500Ò transfection

reagent (Fermentas, Madison, WI, USA) according to the manufacturer’s protocol. After

overnight culture, Dex or lithium (Sigma) was added to the media for 24h and harvested,

lysed, and then analyzed using the Luciferase assay system kit (Promega, WI, USA). The

activity was measured by TD-20 (Turner BioSystems). As an evaluation of transfection

efficiency, a β-gal expression vector was used, and the values of luciferase activity were

normalized by values acquired from the β-gal activity.

15. Statistical analysis

Data were expressed as the mean ± standard deviation (SD). All experiments were

repeated at least three times unless otherwise mentioned. Statistical significance among

three groups was evaluated by the analysis of variance (ANOVA) or statistical

significance between two groups was evaluated by the Student’s t-test.

- 20 -

Part I.

Glucocorticoid induces apoptosis of osteoblasts through the

activation of glycogen synthase kinase 3b

(J Bone Miner Metab 27:140-148, 2009)

- 21 -

III. RESULTS

1. Dex induces apoptosis of osteoblasts through caspase activation

To examine the effect of Dex on osteoblasts, MC3T3-E1 cells were treated with Dex.

As shown in Fig. 7A, Dex significantly reduced the cell proliferation in a dose-dependent

manner, measured by MTT assay. To assess whether the decreased viability was

associated with apoptosis, the activation of caspase-3 and apoptotic nuclei fragmentation

were respectively determined by Western blot analysis and TUNEL staining. Treatment

of the cells with various doses of Dex increased caspase-3 activity (Fig. 7B). In addition,

TUNEL assay showed that Dex treatment significantly increased apoptosis dose-

dependently (Fig. 7C). A relatively high-dose of Dex was required to induce apoptosis of

the cells. The concentration of 1 mM Dex was selected to achieve the maximum effect of

apoptosis.

- 22 -

A.

B.

C.

- 23 -

Fig. 7. Dex induced apoptosis in osteoblast cells. A: MC3T3-E1 cells were exposed to 10,

1 mM, and 100, 10, 1 nM Dex for 48h. The cell proliferation was measured by MTT assay.

B: The cells were exposed to 1 mM, and 100, 10, 1 nM Dex for 24 h. The apoptotic cells with

condensed nuclei were counted by TUNEL staining. A percentage of the total cell number is

represented. The values are expressed as mean ± SD. C: Western blot analysis revealed

cleaved caspase-3 after incubation with 10, 1 mM, and 100, 10, 1 nM Dex for 24h. The

cleaved caspase-3 was analyzed by immunoblots using ECL. Equal amounts of whole cell

lysates were analyzed by 12% SDS-PAGE, and subsequently transferred to PVDF

membranes. b-actin was used for an internal control.

- 24 -

2. Glucocorticoid receptor mediates Dex-induced apoptosis in osteoblasts

An earlier study showed that Dex-induced apoptosis was dependent on glucocorticoid

receptor (GR) binding and it was inhibited by GR antagonist (Cifone et al, 1999). In order

to confirm whether the effect of Dex in osteoblasts was mediated via GR, MC3T3-E1

cells were pretreated with 20 mM mifepristone, a GR antagonist, for 1 h prior to Dex

treatment. Dex reduced the cell proliferation of osteoblasts. However, this was recovered

by mifepristone treatment (Fig. 8A). In addition, the result of apoptotic nuclear

fragmentation was significantly reduced by mifepristone similar to caspase inhibitor, Z-

VAD-fmk (Fig 8B). Mitochondrial membrane potential (MMP) was measured in Dex-

treated cells in terms of mitochondrial membrane function. In the MMP assay, Dex

induced a loss of MMP, however, mifepristone and Z-VAD-fmk significantly prevented

the Dex-induced loss of MMP (Fig. 8C). All these results indicated that Dex induced

apoptosis in the cells through GR.

- 25 -

A.

B.

C.

- 26 -

Fig. 8. Inhibition of glucocorticoid receptor reduced Dex-induced apoptosis. A: MC3T3-

E1 cells were pretreated for 1 h with 20 mM mifepristone in culture media before an addition

of 1 mM Dex. Cell viability was measured by MTT assay after 48 h. Three independent

experiments were performed. B: The cells were pretreated respectively with 40 mM Z-VAD-

fmk and 10 mM mifepristone prior to an addition of 1 mM Dex for 24 h. Dead cells were

counted by DAPI staining. Treatment with Z-VAD-fmk or mifepristone for 24 h reduced the

dead cells induced by Dex (slant: dead cells; black: live cells). C: Mitochondrial membrane

potential (MMP) was reduced by Dex treatment. The 10 mM mifepristone or 40 mM Z-VAD-

fmk prevented the loss of MMP. The values are expressed as mean ± SD.

- 27 -

3. Dex induces GSK3b activation in osteoblasts

GSK3b is one of the best characterized intracellular signaling factor in the canonical

Wnt/b-catenin pathway in osteoblasts (Hurson et al, 2007; Krishnan et al, 2006), and

previous several studies showed that GSK3b induced apoptosis (Song et al., 2002; Jin et al.,

2005; Al-Assar et al., 2005). In this study, the cellular mechanism of Dex-induced

apoptosis was investigated. Furthermore, it was reported that the activation and regulation

of GSK3b are dependent on phosphorylation (Grimes and Jope, 2001). Therefore, the

activation of GSK3b was examined in Dex-treated MC3T3-E1 cells and determined by

Western blotting at a various time points. As a result of this, Dex decreased the

phosphorylation of GSK3b at Ser-9 compared to Dex-untreated cells, and started to reduce

its phosphorylation from early time at 4 h (Fig. 9A). This result indicates that Dex induces

early GSK3b activation in MC3T3-E1 cells in a sense that the phosphorylation of GSK3b

at Ser-9 signifies the inactivation, whereas that of GSK3b at tyrosine 216 (Tyr-216)

indicates the activation. If GSK3b plays a key role in Dex-induced apoptosis, the inhibition

of the GSK3b activation attenuates Dex-induced apoptosis. As expected, LiCl treatment, a

GSK3b specific inhibitor (Klein and Melton, 1996; Stambolic et al, 1996; Davies et al,

2000; Zhang, 2003), inhibited Dex-induced apoptosis (Fig. 9B).

Furthermore, the treatment of LiCl significantly increased the phosphorylated Ser-9

GSK3b level in MC3T3-E1 cells (Fig. 9C, 9D). However, this did not changed Tyr-216

level. This result suggests that GSK3b activation by Dex stimulates the induction of

apoptosis through the phosphorylation of Ser-9. In Dex-treated cells, the phosphorylation

level of GSK3b at Ser-9 was decreased. On the contrary, that of GSK3b at Tyr-216 was

increased compared to Dex-untreated cells (Fig. 9C, 9D).

- 28 -

To confirm the activation of GSK3b by Dex, intrinsic substrates of GSK3b, c-Myc and b-

catenin, were examined. Phosphorylated c-Myc and b-catenin were detected by an addition

of a proteasome inhibitor, MG132 (Fig. 9E). These results indicate that Dex induces

osteoblast apoptosis through the activation of GSK3b.

4. GR stimulated by Dex regulates the phosphorylation of GSK3β

As shown in Fig. 8, Dex signaling was passing via GR and also induced GSK3b

activation in osteoblasts. Therefore, to observe that the Dex signaling stimulated through

GR was related with GSK3b activation, MC3T3-E1 cells were pretreated with 5 and 10 μM

mifepriston and was examined the phospo-GSK3b level. Mifepristone increased the

GSK3b phosphorylation compared to that of Dex (Fig. 10).

5. Dex activates p38 MAPK in osteoblasts

Previous reports showed that p38 MAPK activation is essential for GC-induced

apoptosis (Lu et al., 2006; Cai et al., 2006). Dex increased the phosphorylation level of p38

MAPK (Fig. 11A). To confirm whether p38 MAPK is involved in Dex-induced apoptosis,

cells were treated with SB203580, a p38 MAPK inhibitor. To our surprise, SB203580

significantly decreased osteoblast viability in Dex-induced apoptosis measured by MTT

and TUNEL staining (Fig. 11B, 11C). The Ser-9 phosphorylation of GSK3b was

significantly decreased during the inhibition of p38 MAPK activity by SB203580 (Fig.

11D).

- 29 -

A.

B.

C.

D.

- 30 -

E.

Fig. 9. The activation of GSK3b in MC3T3-El cells by Dex. A: MC3T3-E1 cells were

treated with 1 mM Dex for 2, 4, 8, and 24 h. The protein expressions of phospho-GSK3b

(Ser-9 and Tyr-216), GSK3b, and b-actin were analyzed by Western blotting. The

representative result of two independent experiments was shown. B: The cells were

pretreated with 25 mM LiCl for 1 h. The 1 mM Dex was added for an additional 24 h. The

dead cells were counted by trypan blue staining. The values are expressed as mean ± SD. C,

D: The cells were treated with 1 mM Dex for 24 h. The phosphorylation of GSK3b (Ser-9

and Tyr-216) was analyzed by Western blotting. The immunoblots were probed with

phospho-GSK3b specific antibodies, and b-actin was used for an internal control. The results

showed that phosphorylation of Ser-9 decreased after Dex treatment. LiCl with 10 mM

concentration increased Ser-9 phosphorylation. There was no significant change in Tyr-216

phosphorylation after treatment with Dex or LiCl. The immunoblots were quantified by

densitometer. The results shown are representative data from two independent experiments.

E: The cells were treated with and without 1 mM MG132 prior to an addition of 1 mM Dex

for 0, 1, 3, 5, and 7h. The expression of phospho-c-Myc, phospho-b-catenin, and b-catenin

was analyzed by immunoblots.

- 31 -

A.

B.

Fig. 10. The phosphorylation of GSK3b by GR antagonist. MC3T3-E1 cells were

pretreated with 5, 10 mM mifepristone for 1 h, and then added 1 mM Dex in media for 24 h.

The expression of phospho-GSK3b (Ser-9) was analyzed by immunoblotting (A) and

quantified by densitometer (B). b-actin was used for an internal control. The results were

obtained from two independent experiments.

- 32 -

A.

B.

C.

- 33 -

D.

Fig. 11. Expression of phospho-p38 MAPK and inhibition of p38 MAPK in Dex-

induced apoptosis. A: After MC3T3-E1 cells were treated with 1 mM Dex, phosphorylation

of p38 MAPK was time-dependently detected by immunoblots. Whole cell lysates were

analyzed by 10% SDS-PAGE, and b-actin was then measured with the same immunoblot. B,

C: The cells were treated with 10 mM SB203580 for 1 h, and 1 mM Dex was then added.

After 48 h, cell proliferation was analyzed by MTT (B), and apoptotic cell death was

determined by TUNEL assay after 24 h (C). Data are expressed as mean ± SD. (D) The cells

were pretreated with 10 mM SB203580 for 1 h, and then exposed to 1 mM Dex for an

additional 4, 8, 16, and 24 h. Phosphorylation of GSK3b (Ser-9 and Tyr-216) and p38

MAPK were analyzed by Western blotting. The immunoblots were probed with specific

antibodies, and an antibody for b-actin served as control.

- 34 -

6. Gene knock-down of GSK3b and p38 MAPK

To determine whether the gene level of GSK3b altered or whether p38 MAPK affected

Dex-induced apoptosis, siRNA experiments were performed. GSK3b gene knock-down by

the siRNA significantly decreased Dex-induced apoptosis, which is assessed by TUNEL

assay and GSK3b activity test (Fig. 12A). In addition, GSK3b siRNA clearly decreased

both phosphorylated (Ser-9) and total GSK3b in compared to that of the siRNA control.

The siRNA of GSK3b also reduced p38 MAPK expression (Fig. 12B). These results

indicated that inhibition of GSK3b protected the cells from Dex-induced apoptosis.

On the contrary, transfection of siRNA for p38 MAPK increased TUNEL positive

apoptotic cells. Knock-down of p38 MAPK showed a strong effect on cell death compared

to Dex (Fig. 12C, 12D). As the expression level of p38 MAPK was decreased by siRNA,

total GSK3b level was also decreased. According to these results, p38 MAPK is thought to

be related to a survival signaling for the compensation of apoptosis induced by Dex.

- 35 -

A. B.

C. D.

Fig. 12. Effects of gene silencing of GSK3b and p38 MAPK on osteoblast apoptosis.

MC3T3-E1 cells were transfected with 25 nM siRNA of GSK3b (A, B) and p38 MAPK (C,

D) for 24 h, and then exposed to 1 mM Dex for an additional 4 h. TUNEL assay was repeated

twice (A, C). Equal amounts of whole cell lysates were analyzed by 10% SDS-PAGE, and

the equal protein contents were loaded for measurement (B, D). The representative result of

three independent experiments is shown. *Con: control.

- 36 -

IV. DISCUSSION

GCs are known to have an effect on bone metabolism (Canalis E, 1996) and to induce

apoptosis of osteoblasts (Gohel et al., 1999; Spreafico et al., 2006), osteocytes (Kogianni

et al., 2004), and numerous lymphoid and myeloid tissues (Lu et al., 2006; Cai et al.,

2006). Although the effects of GCs in bone remodeling have been reported, the

mechanism of GC-induced apoptosis is poorly understood. Thus, to clarify the apoptotic

mechanism of GC in osteoblasts, a relatively high-dose of Dex was used as an apoptotic

inducer in MC3T3-E1 cells. The results of caspase-3 analysis and TUNEL staining

showed that Dex induced apoptotic cell death.

Several studies have shown that Dex-induced apoptosis is dependent on GR binding

(Cifone et al., 1999; Schmidt et al., 2004; Ni Chonghaile et al., 2006). This study

examined the effects of GR on Dex-induced osteoblast apoptosis. A GR antagonist,

mifepristone, attenuated the effect of Dex. Moreover, the loss of MMP on the

mitochondria in Dex-treated cells was prevented by not only mifepristone, but also Z-

VAD-fmk, a caspase inhibitor. The present results indicate that Dex induces osteoblast

apoptosis via GR and the caspase activation.

To clarify the mechanism of Dex-induced cell death in osteoblasts, two kinase enzymes,

GSK3b and p38 MAPK, were examined, because they are known to be involved in

apoptosis (Beurel and Jope, 2006; Lu et al., 2006). First, this study demonstrated that

GSK3b played a crucial role in apoptosis of osteoblasts caused by GC. In osteoblasts,

GSK3b is not only a main actor in osteoblastogenesis (Krishnan et al., 2006), but also a

multi-tasking regulator in many biological events, including structure, gene expression,

- 37 -

mobility, and apoptosis (Doble and Woodgett, 2003; Kim et al., 2002). The activation of

GSK3b is regulated by serine and tyrosine phosphorylation, by protein complex

formation, and by its intracellular localization (Grimes and Jope, 2001).

GSK3b is related to apoptosis as shown by aforementioned several studies (Jin et al.,

2005; Al-Assar et al., 2005). Thapsigargin induced apoptosis of neuronal cells through the

activation of GSK3b, and this was prevented by LiCl (Song et al., 2002). A recent study

showed that GSK3b overexpression was sufficient to facilitate staurosporine- and heat

shock-induced apoptosis, whereas LiCl attenuated caspase-3 activation (Bijur et al., 2000).

However, it has not yet to be known whether GSK3b participates in Dex-induced

osteoblast apoptosis. The present study showed that Dex induced GSK3b activation at

early time and this proceeded to apoptosis. There are recent reports that GSK3b can be

activated in early time ( Bijur and Jope, 2001; Kim et al., 2002). LiCl, a GSK3b inhibitor,

reduced Dex-induced apoptosis. In addition, the effect of GSK3b siRNA was similar to

this result. In summary, the activation of GSK3b was considerably associated with Dex-

induced apoptosis.

GSK3b is a potentially very effective and putative substrate in cellular processes (Kim

et al., 2003). Further, Dex has recently been known to alter the gene expression in

osteoblasts during specific developmental pathways (Hurson et al., 2007), and to increase

the expression of Dkk-1, a member of the dickkopf family of secreted potent antagonists

of Wnt signaling (Ohnaka et al., 2004). In addition, GCs in osteoblasts decrease the c-

myc expression, a cell cycle regulatory factor that is modulated by GSK3b (Smith et al.,

2002). It is, therefore, possible that the GSK3b activity modulates the Wnt signal

pathway and alters the cell cycle in Dex-treated osteoblasts.

- 38 -

Second, this study demonstrated that p38 MAPK has a crucial role in osteoblast

apoptosis caused by GC. Previous reports have shown that p38 MAPK was involved in

apoptosis (Hsu et al., 2007; Cai et al., 2006) and activated during the GC-induced

apoptosis (Osamu et al., 2002). In addition, it was shown that p38 MAPK plays many

critical roles in apoptosis and survival of osteoblasts (Alikhani et al., 2007; Chae et al.,

2001; Saffar et al., 2008). Furthermore, the core region of GSK3b conformation has a

topology similar to the equivalent region in the MAPK such as ERK2 and p38 (Dajani et

al., 2001).

In this study, the correlation between p38 MAPK and osteoblast apoptosis was

determined. The present study showed that Dex induced p38 MAPK activation.

Unexpectedly, the inhibition of p38 MAPK by inhibitor or siRNA for gene knock-down

did not decrease osteoblast apoptosis. Thornton et al. (2008) showed that p38 MAPK

inactivated GSK3b by phosphorylation in the brain and thymocytes and resulted in an

accumulation of b-catenin. As a result of this, p38 MAPK is related with survival

signaling for the compensation of apoptosis induced by Dex. According to the results of

this study, the suggested mechanism of osteoblast apoptosis caused by Dex was illustrated

as summarized below. Dex induced osteoblast apoptosis via GR in the way of GSK3b

activation, while Dex activated p38 MAPK as a compensatory mechanism to protect the

cells from apoptosis (Fig. 13, Yun et al., 2009).

- 39 -

Fig. 13. The suggested scheme in Dex-induced apoptosis of osteoblasts. Dex induced

apoptosis and decreased phosphorylation of GSK3β at Ser-9. However, inhibition of GSK3β

by lithium (a GSK3β inhibitor) and GSK3β siRNA protected the cells from apoptosis. In

addition, Dex increased the phosphorylation of p38 MAPK during osteoblast apoptosis.

Inhibition of p38 MAPK by SB203580 (a p38 MAPK inhibitor) and p38 MAPK siRNA

increased apoptotic cell death. There could be a compensatory mechanism between GSK3β

and p38 MAPK in Dex-induced apoptosis of osteoblasts.

- 40 -

Part II.

Glycogen synthase kinase-3beta regulates etoposide-induced

apoptosis in C3H10T1/2 cells.

(Apoptosis 14:771-777, 2009)

- 41 -

III. RESULTS

1. Etoposide induces apoptosis C3H10T1/2 cells

Most chemotherapeutic agents induce cell death. In particular, genotoxic stress is a

suitable stimulus that effectively induces apoptosis (Bladwin and Osheroff, 2005).

Therefore, the effect of etoposide, a DNA-damaging agent, was investigated in C3H10T1/2

osteoblast progenitor cells.

The effects of etoposide on the cells were examined using MTT assay, TUNEL staining and

Western blotting. To determine a suitable concentration, the cells were treated with varying

doses of etoposide for 24 h (Fig. 14A). Etoposide at 50 mM concentration decreased the

cell viability by one-half after 24 h and this was sustained up to 48 h (Fig. 14B). The effect

of etoposide on the osteoblasts was identical to that shown in a previous study (Davis et al.,

2002). Therefore, this concentration (50 mM) was used for inducing apoptosis. As shown in

Fig. 14C and 14D, etoposide significantly increased TUNEL-positive cells and triggered

caspase-3 activation, which resulted in the cleavage of poly ADP-ribose polymerase

(PARP) (Fig. 14E).

- 42 -

A. B. C.

- 43 -

D.

E.

Fig. 14. Etoposide induced apoptosis in C3H10T1/2 cells. A: C3H10T1/2 cells were

treated with varying doses (10, 25 and 50 mM) of etoposide for 24 h. B: Cells were incubated

with 50 mM etoposide for 4, 8, 16, 24 and 48 h. Cell viability was measured by MTT assay.

Results were means ± SD value, * p ≤0.01 compared with untreated controls. (C, D) Cells

were treated with 50 mM etoposide for 24 h after which nuclei were stained using TUNEL

assay. E: Equal amounts of total cellular lysates were analyzed by 12% SDS-PAGE, and

subsequently transferred to PVDF membranes. Cleavage of PARP and caspase-3 was

analyzed by immunoblots using ECL detection to assess activation of an apoptotic signal. b-

actin was used as a protein internal control.

- 44 -

2. Etoposide induces osteoblast apoptosis via GSK3β activation, which is prevented

by gene silencing

To determine if GSK3β was involved in etoposide-induced osteoblast apoptosis, the

GSK3β activity was measured by Western blot analysis. Etoposide induced the

dephosphorylation of GSK3β at Ser-9 in a time-dependent manner, while total GSK3β used

as a control protein was constantly expressed (Fig. 15). To determine if GSK3β was

involved in osteoblast apoptosis, siRNA for GSK3β was transfected into the cells. This

resulted in a knock-down expression of GSK3β. Caspase-3 activation was then analyzed by

Western blotting. This showed that GSK3β siRNA significantly inhibited caspase-3

activation and decreased the TUNEL-positive cells (Fig. 16).

- 45 -

A.

B.

Fig. 15. Etoposide induced GSK3β activation in C3H10T1/2 cells. C3H10T1/2 cells were

incubated with 50 μM etoposide for 24 h. A: Activation of GSK3β was analyzed by Western

blotting, which assessed the phosphorylation of GSK3β at Ser-9. Total GSK3β was used as a

protein control. B: GSK3β (Ser-9) phosphorylation level was quantified using ImageGauge

program (Ver 3.12). Data were corrected using total GSK3β. The experiments were

independently repeated three times.

- 46 -

A.

B.

Fig. 16. Effects of GSK3b siRNA on etoposide-induced apoptosis. GSK3β gene

expression levels. Gene expression of GSK3β in C3H10T1/2 cells was modified using

siRNA of GSK3β. After cells had been transfected with 25 nM siRNA for overnight, 50 (M

etoposide was added for 24h. The cells were harvested for Western blot analysis. A: The

gene knock-down of GSK3β was confirmed by immunoblotting. A representative

immunoblot was shown. B: The cells were fixed for TUNEL assay and results were

presented. The experiments were repeated three times. *Con: control.

- 47 -

3. Etoposide induces apoptosis via GSK3β activation, which is prevented by gene

silencing

To confirm the role of GSK3β during etoposide-induced apoptosis, cells were treated with

LiCl, a GSK3β inhibitor. Before the addition of etoposide, the cells were pre-treated with

LiCl at varying doses for 1 h. Caspase-3 activation and PARP cleavage were inhibited in a

dose-dependent manner (Fig. 17A). To prevent caspase activation by etoposide, cells were

treated with 40 μM Z-VAD-fmk, a general caspase inhibitor. This treatment decreased the

proportion of TUNEL-positive cells induced by etoposide. In addition, the effect of LiCl

coincided with the prevention of DNA degradation, as in the case for Z-VAD-fmk (Fig.

17B). When cells were treated with LiCl, caspase-3 activity and PARP cleavage were

significantly decreased in response to etoposide (Fig. 18A). Staurosporine, which is another

agent that causes cellular apoptosis, also induced osteoblast apoptosis. The effective dose

for staurosporine was checked in C3H10T1/2 cells. The effect of a decreased amount of

0.1μM staurosporine was similar to that of etoposide. Staurosporine significantly increased

caspase-3 activation. However, LiCl had no effects on preventing the cells from

staurosporine-induced caspase-3 activation and cleavage of PARP (Fig. 18B).

- 48 -

A.

B.

Fig. 17. LiCl inhibited caspase-3 activation and apoptotic cell death in etoposide-

induced apoptosis. A: C3H10T1/2 cells were pre-treated with 5, 10 or 20 mM LiCl for 30

min before adding 50 mM etoposide and then incubated for 24 h. PARP, cleaved caspase-3

and b-actin were analyzed by Western blotting. B: After treatment with 10 mM LiCl or 40

mM Z-VAD-fmk for 30 min, 50 mM etoposide was added and the cells were incubated for an

additional 24h.

- 49 -

A.

B.

Fig. 18. Effects of LiCl on etoposide- or staurosporine-induced apoptosis C3H10T1/2

cells increased caspase-3 activity after treatment by both 50 μM etoposide (A) and 0.1 μM

staurosporine (B). Cells were pre-treated with 10 mM LiCl for 30 min before the addition of

etoposide (A) or staurosporine (B), then incubated for an additional 24 h. The caspase-3

activity was estimated by EIA method using equal amounts of total cellular lysates. The

cleavage of caspase-3 and PARP was examined by Western blotting. Three independent

experiments were performed and representative results were shown.

- 50 -

4. Bcl-2 is mediated by GSK3b in etoposide-induced apoptosis

To determine mediation through the mitochondrial pathway, Bcl-2 and Bax expressions

were observed over time during etoposide-induced osteoblast apoptosis. Etoposide

decreased the expression of Bcl-2, an anti-apoptotic protein. However, the level of the pro-

apoptotic protein, Bax, did not change during the 24 h period (Fig. 19A, B). To ascertain if

GSK3b influenced these results, the cells were pre-treated with 10 mM LiCl for 1 h before

the addition of etoposide for 24 h. LiCl clearly improved the expression level of Bcl-2

protein. In contrast, lithium did not rescue the staurosporine-induced (mediated) Bcl-2

protein decrease (Fig. 19C). In addition, the decreased transcription level of Bcl-2 by

etoposide was recovered by LiCl (Fig. 19D).

- 51 -

A. B.

C. D.

Fig. 19. Effects of LiCl on Bcl-2 expression in etoposide-induced apoptosis. C3H10T1/2

cells were treated with 50 μM etoposide for 4, 8, 16 and 24h. A: Equal amounts of total

cellular lysates were loaded, and both Bcl-2 and Bax were analyzed by Western blotting. A

representative blot from three independent experiments was shown. B: The expression levels

of Bcl-2 and Bax were quantified by ImageGauge at 24h compared to zero time, and were

corrected using β-actin. C: Cells were pre-treated with 10 mM LiCl for 30 min before the

addition of 50 µM etoposide or 0.1 µM staurosporine and then incubated for an additional

24h. The protein expression of Bcl-2 was analyzed by Western blotting. D: The

transcriptional level was examined by quantitative RT-PCR. Results are representative of

three separate experiments. β-actin (A, B) and GAPDH (C) were used as internal controls.

- 52 -

IV. DISCUSSION

Increased osteoblast apoptosis can lead to osteoporosis, which is also induced by

glucocorticoids. C3H10T1/2 cells, which give rise to osteoblasts, showed apoptosis after

exposure to etoposide, a well-known anti-cancer agent. First, to determine the apoptotic

effects, cell numbers were examined. Etoposide decreased cell proliferation in both a dose-

and time-dependent manner. Etoposide showed apoptotic characteristics, including

procaspase-3 processing, proteolysis of PARP and TUNEL-positive staining. Therefore,

etoposide was a suitable agent for inducing apoptosis, as shown in a previous report using a

human osteoblast-like cell line (MG-63) (Davis et al., 2002). The proportion of apoptotic

cells under etoposide condition was decreased by the co-treatment with Z-VAD-fmk. This

indicates that etoposide-induced osteoblast apoptosis proceeds via caspase-3 activation.

GSK3β, one of serine/threonine kinases, plays central roles in regulating cellular functions,

structure and survival (Jope et al., 2007). In a recent report, GSK3b showed differential roles

in the regulation of transcriptional activation (Doble and Woodgett, 2003). Although GSK3

has been recently shown to be a key regulator of cell fate (Beurel and Jope, 2006), the roles

in genotoxic stress-induced osteoblast apoptosis have yet to be identified. Therefore, the role

of GSK3b in osteoblast apoptosis signaling was examined.

Etoposide decreased the phosphorylation of GSK3b at Ser-9 in a time-dependent manner.

This result is similar to an earlier study using dexamethasone-induced apoptosis (Yun et al.,

2009). LiCl, a GSK3b inhibitor, decreased apoptosis by the inhibition of caspase-3 activation.

These results indicate that GSK3b is associated with etoposide-induced apoptosis. The

effects of another stress-inducing drug on osteoblast were studied. Staurosporine, which is a

- 53 -

potent inducer of apoptosis and a highly potent protein kinase C inhibitor (Caponigro et al.,

1997), induces osteoblast apoptosis by the activation of caspase-3-like proteases (Chae et al.,

2000). Both etoposide and staurosporine significantly increased caspase-3 activation.

However, LiCl had no effects on caspase-3 activity in staurosporine-induced apoptosis.

Therefore, although both etoposide and staurosporine induced apoptosis, their effects were

mediated via different signaling pathways. This implies that GSK3b activation does not

induce all types of apoptosis. A recent study showed that GSK3b was involved in apoptosis

of colorectal cancer cells through a direct mitochondrial pathway (Tan et al., 2005).

In a recent report, parathyroid hormone (PTH) rapidly stimulated the expression of Bcl-2

for osteoblast survival (Bellido et al., 2003). In addition, etoposide-induced apoptosis

decreased Bcl-2 level and increased Bax expression via a p53-dependent pathway in human

osteosarcoma cells (Yuan et al., 2008). Therefore, members of the Bcl-2 family of proteins,

Bcl-2 and Bax, were examined whether they are associated with osteoblast apoptosis by

etoposide. In this study, etoposide decreased Bcl-2 expression throughout the 24h test period,

but did not change the levels of Bax. However, LiCl recovered the decreased expression of

Bcl-2 in terms of both the protein and the mRNA levels. This suggests that GSK3b regulates

the expression of Bcl-2 in etoposide-induced osteoblast apoptosis. The protective effect of

LiCl was confirmed in osteoblast apoptosis, as seen with neuronal mechanisms (Chen and

Chuang, 1999). It is important to note that LiCl protected osteoblasts from apoptosis by the

phosphorylation-induced inactivation of GSK3b, which resulted in an increased Bcl-2

expression level due to the blockade of caspase-3 activity.

The results of this study demonstrated that etoposide induced osteoblast apoptosis by an

increase of GSK3b activation through Bcl-2 mediated signaling, which led to caspase-3

- 54 -

activation. It was shown that LiCl treatment recovered the Bcl-2 expression level and

inhibited caspase-3 activity. In addition, the modulation of GSK3b expression by siRNA

resulted in an increase of cell survival. These results imply that GSK3b is an essential

component of etoposide-induced osteoblast apoptosis and that Bcl-2 is one of the key

regulators of GSK3b for cell survival. Therefore, GSK3b inhibitors can block GSK3-

facilitated apoptosis and may be effective agents for osteoporosis treatment.

- 55 -

Part III.

Dexamethasone Regulates Osteocalcin via Heat

Shock Protein 25 and GSK-3β in Osteoblast

Differentiation

(Preparing to submission)

- 56 -

III. RESULTS

1. The expression of HSP25 during osteoblast differentiation

Prolonged passage of MC3T3-E1 (MC) cells could result in un-mineralization. The

MC3T3-E1 sub-clone 14 (MC14) cells have been shown to form mineral nodules (Wang

et al., 1999). In this study, both MC and MC14 cells were cultured with and without Dex

for 30 Days. In the presence of Dex, MC14 cells formed mineral nodules, but MC cells

did not (Fig. 20).

I investigated whether HSP25 expression was related to osteoblast differentiation. In the

presence of Dex, MC cells continued to accumulate HSP25 protein during differentiation

(from day 3 to day 20), but gradually decreased in MC14 cells (Fig. 21A, 21B). The

HSP25 level was significantly increased in the MC cells compared to the MC14 cells (Fig.

21C). The increased protein level of HSP25 showed a dose- and time-dependent increase.

However, the HSP70 expression that plays a role in cytoprotection induced by stress did

not (Fig. 22).

- 57 -

A.

B.

Fig. 20. Effects of Dex on mineralization in MC and MC14 cells. The MC (A) and MC14

cells (B) were cultured with osteogenic induction medium supplemented with 1 μM Dex.

Both the cells were induced under the same conditions. Alizarin red S staining was

performed on day 30. Data shown are representative of three independent experiments.

- 58 -

A.

B.

C.

Fig. 21. Effects of Dex on HSP25 expression during osteoblast differentiation in MC

and MC14 cells. A, B: The MC and MC 14 cells were cultured with osteogenic induction

medium supplemented with 1 μM Dex. On the indicated days (3, 5, 10, and 20 days), the

cells were harvested and equal amounts of whole cell lysates were immunoblotted with anti-

HSP25 antibodies. b-actin was used for an internal control. C: HSP25 mRNA expression

was examined in the MC and MC 14 cells. GAPDH was used as the internal control. The

relative fold change of Hsp25 compared to the control was analyzed. Bars represent the

standard deviation. Data shown are average of three repeat experiments.

- 59 -

A.

B.

Fig. 22. Dex increased HSP25 expression in MC cells. A: The MC cells were treated with

1 μM Dex for 1, 2, and 3 days. B: The cells were exposed to 10, 1, 0.1, 0.01 and 0.001 μM

Dex for 24h. The protein expression of HSP25, HSP70, and β-actin was analyzed by

Western blot. The fold increase of HSP25 was quantified and normalized by β-actin.

- 60 -

2. Effects of Dex on osteocalcin gene expression in osteoblasts

To investigate whether the HSP25 induction by Dex was related to osteoblast

differentiation, osteoblast differentiation markers were measured by RT-PCR. The

expression level of mRNA was evaluated. Dex increased HSP25 expression and then

decreased osteocalcin expression in the MC cells. There were no effects on osteopontin,

Runx2, and collagen type I (Fig. 23). In addition, the osteocalcin level also showed a

difference between the MC and MC14 cells. The MC14 cells showed a high expression of

osteocalcin compared to the MC cells (Fig. 24).

3. Effects of knock-down of HSP25 on osteocalcin in osteoblasts

This study examined whether the HSP25 modulated by Dex could influence osteocalcin

expression. To knock-down HSP25, the siRNA of HSP25 was transfected into the MC cells.

The inhibition of HSP25 was determined by PCR and immunoblot analysis (Fig. 25A,

25C). HSP25 siRNA considerably recovered osteocalcin expression up to control level (Fig.

25A, 25B). Dex significantly decreased ALP activity, and HSP25 siRNA treatment

recovered the ALP activity (Fig. 25D).

- 61 -

Fig. 23. Effects of Dex on osteoblast differentiation markers in MC cells. The MC cells

were cultured in osteogenic conditioned media for 3 days. The effects of Dex on osteocalcin,

osteopontin, HSP25, Runx2, and collagen type I (Col I) were analyzed by quantitative RT-

PCR. The results of mRNA level were normalized by GAPDH. The fold increase compared

to the untreated control is shown. The results were repeated independently over three times.

- 62 -

A.

B.

Fig. 24. Osteocalcin expression in MC and MC14 cells. After the MC and MC14 cells

were cultured in osteogenic conditioned media with or without 1 μM Dex for 3 days. The

effect of Dex on the expression level is shown (A) as analyzed by RT-PCR (B). Both the

cells were normalized by GAPDH. The results shown are representative data from at least

three separate experiments.

- 63 -

A.

B.

C.

D.

- 64 -

Fig. 25. Effects of HSP25 siRNA on gene expression and ALP activity in MC cells. A-D:

After the MC cells were transfected with 30 nM HSP25 siRNA for 24h, the cells were

cultured in osteogenic conditioned media with or without 1 μM Dex for an additional 3 days.

The knock-down effect of HSP25 on the expression level was analyzed by RT-PCR (A) and

western blotting (C), and the representative data are shown. β-actin was used as an internal

protein control. B: The osteocalcin mRNA level was shown by RT-PCR and normalized by

GAPDH. D: After that, both MC cells and the transfected cells with HSP25 siRNA were

harvested, lysed, and evaluated for the ALP activity. The results of the ALP activity were

normalized by reacted protein. The fold increases compared to the control are shown. These

results are from triplicate experiments.

- 65 -

4. Lithium controls expression of HSP25 and osteocalcin

This study examined whether the induction of HSP25 was associated with Wnt signaling.

Lithium is known to inhibit GSK3β during Wnt signaling (Hedgepeth, 1997; Klein and

Melton, 1996). Lithium significantly reduced HSP25 expression; however, it had no

influence on HSP70 (Fig. 26). To confirm at the transcription level, expressions of HSP25

and osteocalcin were analyzed by RT-PCR. Similar to the HSP25 siRNA, lithium reduced

HSP25 and increased osteocalcin transcription (Fig. 27).

To determine the role of GSK3β, transcription activity of an osteocalcin reporter

containing the osteocalcin element sequence (pGL2-OSE), was measured in the MC cells.

Dex significantly decreased the osteocalcin transcriptional activity. Lithium recovered

osteocalcin activity compared to the vector control (Fig. 28 top). The effects of lithium

were confirmed at both the mRNA and protein levels (Fig. 28 bottom).

Fig. 26. Lithium inhibited HSP25 expression induced by Dex in MC cells. The MC cells

were exposed to 1 μM Dex or 1 μM Dex with 10mM LiCl for 4, 8, 16, and 24h. The cells

were harvested and the cell lysates were then immublotted with anti-HSP25, anti-HSP70,

anti-phospho Ser9 GSK3 β, and anti-β-actin for an internal control.

- 66 -

A.

B.

C.

Fig. 27. Effects of lithium on the HSP25 and osteocalcin expression in MC cells. A-C:

The MC cells were exposed to 1 μM Dex, Dex with 10 mM LiCl, or 10 mM LiCl for 3

days in osteogenic medium. RT-PCR analysis was then done to estimate the transcription

level of osteocalcin (B) and HSP25 (C). A: The representative data are shown. GAPDH

was used as a positive control and the results were normalized by GAPDH. The results

were repeated independently three times.

- 67 -

Fig. 28. Dex decreased the osteocalcin activity in MC cells. The MC cells were transfected

with an empty vector (pGL2-basic) or the pGL2-OSE vector. The cells were cultured for 24h,

and then exposed to 1 μM Dex, Dex with 10 mM LiCl, 10 mM LiCl for an additional 24 h.

The empty vector was set as 100% and the relative fold increases were calculated for the

others. The results were normalized by β-gal activity (top). The expression of HSP25 and

osteocalcin was also analyzed by RT-PCR or immunoblot (bottom). The results are from

triplicate experiments.

- 68 -

IV. DISCUSSION

GCs, such as Dex, have deleterious effects on osteoblast physiology (Hong et al., 2009;

Kanazawa et al., 2009). A previous study showed that Dex increased HSP25 expression in

osteoblasts (Osamu et al., 2002). HSP25 was induced by TGF-β (Hayashi et al., 2008),

prostaglandin (Takai et al., 2006), and sphingosine 1-phosphate (Shinji et al, 2006) in

osteoblasts. HSP25 controls protein misfoldings in cellular compartments. However, several

studies have shown that HSP25 influences the differentiation of keratinocytes (Davidson

and Morange, 2000), cardiomyocytes (Winger et al., 2007), and chondrocytes (Favet et al.,

2001). However, the functional role and underlying molecular mechanisms associated with

HSP25 have not yet been reported.

In this study, the role of HSP25 was examined in osteoblasts. First, to determine the effects

of Dex, differentiation was examined in MC and MC14 cells. Treatment with Dex in the

MC14 cells induced differentiation and the formation of mineral nodules. However, Dex in

the MC cells didn’t have the same effect. Second, the protein expression of HSP25 was

examined in MC and MC14 cells during differentiation. The HSP25 protein was increased in

the MC cells, while HSP25 protein was decreased in the MC14 cells after Dex treatment.

Dex increased the HSP25 mRNA in both the MC and MC14 cells. This implies that HSP25

plays an important role during the differentiation of osteoblasts.

A previous study reported that HSP25 controlled actin polymerization, and its function was

related to dynamic structural organization (Benndorf et al., 1994). Dynamic cytoskeletal

changes resulted in an increase of ALP activity, osteocalcin secretion and mineralization

(H