ARTICLE IN PRESS

0142-9612/$ - se

doi:10.1016/j.bi

�Correspond��Also to b

Shantou Unive

Tel.: +86754 2

E-mail addr

chengq@stu.ed1Both author

Biomaterials 28 (2007) 3063–3073

www.elsevier.com/locate/biomaterials

The effect of 3-hydroxybutyrate on the in vitro differentiation of murineosteoblast MC3T3-E1 and in vivo bone formation in ovariectomized rats

Yan Zhaoa,1, Bing Zoua,1, Zhenyu Shia, Qiong Wua,�, Guo-Qiang Chena,b,��

aProtein Science Laboratory of Ministry of Science, Department of Biological Science and Biotechnology, Tsinghua University, Beijing 100084, ChinabMultidisciplinary Research Center, Shantou University, Shantou 515063, Guangdong, China

Received 21 November 2006; accepted 8 March 2007

Available online 14 March 2007

Abstract

3-hydroxybutyrate (3HB), one of the degradation products of microbial biopolyesters polyhydroxyalkanoates (PHA), is a high energy

metabolic substrate in animals. This study evaluated the effects of 3HB on growth of osteoblasts in vitro and on anti-osteoporosis in vivo.

Alkaline phosphatase (ALP) assay, Van Kossa assay and Alizarin S red staining were used to study in vitro differentiation of murine

osteoblast MC3T3-E1 cells. The intensity of in vitro cell differentiation measured in ALP was in direct proportion to the concentration of

3HB when it was lower than 0.01 g/L. Calcium deposition, a strong indication of cell differentiation, also showed an obvious increase

with increasing 3HB concentration from 0–0.1 g/L, evidenced by Alizarin red S staining and Van Kossa assay. RT-PCR also showed

significantly higher expression of osteocalcin (OCN) mRNA in MC3T3-E1 cells after 3HB administration. In vivo study using female

Wistar rats (3 months old, n ¼ 80) allocated into normal, sham-operated or ovariectomized (OVX) group that led to decreasing bone

mineral density (BMD), bone histomorphometry and biomechanics compared with normal and sham groups, had demonstrated that

3HB increased serum ALP activity and calcium deposition, decreased serum OCN, prevented BMD reduction resulting from OVX. All

these led to enhanced femur maximal load and bone deformation resistance, as well as improved trabecular bone volume (TBV%). In

conclusion, 3HB monomer containing PHA can be effective bone growth stimulating implant materials.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: 3-hydroxybutyrate; PHA; Osteoblast; Osteoporosis; Polyhydroxyalkanoate

1. Introduction

3-hydroxybutyrate (3HB) containing polyhydroxyalk-anoates (PHA) has shown to be effective as bioimplantmaterials that stimulate tissue regeneration including boneand cartilages tissues [1–4]. 3HB is also one of the mainketone bodies primarily produced in the liver fromdegradation of long-chain fatty acids and transportedthrough plasma to peripheral tissues [5]. The role of 3HB asan energy source and lipogenic precursor has already beenrecognized [6]. Recently, 3HB has been employed to treat

e front matter r 2007 Elsevier Ltd. All rights reserved.

omaterials.2007.03.003

ing author. Tel.: +86 10 6277 1664; fax: +86 10 6278 8784.

e corresponded to. Multidisciplinary Research Center,

rsity, Shantou 515063, Guangdong, China.

901186; fax: +86 754 2901175.

esses: wuqiong@mail.tsinghua.edu.cn (Q. Wu),

u.cn (G.-Q. Chen).

s contributed equally to this research.

traumatic injuries that benefit from elevated levels ofketone bodies such as hemorrhagic shock [7,8], extensiveburns [9], myocardial damage [10], and cerebral hypoxia,anoxia, and ischemia [11]. Furthermore, 3HB has beenfound to be able to reduce death rate of human neuronalcell model culture for Alzheimer’s and Parkinson’s disease[12] and to ameliorate the appearance of corneal epithelialerosion through suppression of apoptosis [13]. 3HB wasalso reported to be able to correct defects in mitochondrialenergy generation [14]. Additional advantages for 3HBinclude good tolerance by humans and a short half-life in

vivo [15]. There have been already several potentialtherapeutic applications reported for 3HB [16].3HB is a degradation product of some PHA, for which

we have demonstrated good biocompatibility in tissueengineering applications [2,17–19]. The foremost abilityamong these properties is to support high levels of cell andtissue growth [4,20–25]. It is speculated that monomers

ARTICLE IN PRESSY. Zhao et al. / Biomaterials 28 (2007) 3063–30733064

released from PHA degradation contribute to improvetissue regeneration [26], which was supported by the factthat 3HB presence activated Ca2+ channel and increasedcalcium influx in the cultured cells [26]. Also, it was foundthat 3HB suppressed the death of cell line L929 whencultured at high density [26]. It was also found that 3HBprevented apoptosis induced by serum withdrawal [27].While the underlying mechanism is still being investigated,it is proposed that 3HB possibly fuels the mitochondriaand thereby promotes cell growth to high density whichrequires extensive energy and nutritional supplies [14].

Previous results showed that PHA copolyester(PHBHHx) consisting of 3HB and 3-hydroxyhexanoatepromoted the growth and differentiation of osteoblasts.Could the degradation product of PHBHHx, namely 3HB,contribute to this phenomenon? If the answer is yes, 3HBcould be a candidate to treat osteoporosis.

Currently, treatments for osteoporosis mainly depend onsuppressing osteoclast activity and bone resorption.Although these treatments can decrease the frequency offracture and increase bone mineral density (BMD), thelong-term suppression of bone resorption seems to beineffective for bone remolding [28]. The uses of growthfactors, hormones or fluoride compounds to stimulate boneformation encountered side effects that limit their applica-tions [29]. It is therefore, significant to develop efficient andsafe treatments for osteoporosis that is attributed toreduced osteoblast activity and bone formation [30–32].

In this study, for the first time, 3HB was used toinvestigate its effect on in vitro growth and differentiationof murine osteoblast MC3T3-E1 cell line, a well-acceptedmodel for osteogenesis study [33]. The in vivo effect of 3HBon osteoporosis was also studied using ovariectomized(OVX) rats, an effective osteoporosis animal model [34].

2. Experiment

2.1. Materials

Murine osteoblastic MC3T3-E1 cells were generously provided by

Professor Rong-qing Zhang, Department of Biological Sciences and

Biotechnology of Tsinghua University, Beijing, China. Fetal bovine serum

(FBS) was purchased from Hyclone (UT, USA), streptomycin from

Amresco (Solon, OH), DL-3-hydroxybutyrate sodium salt (3HB) and

penicillin from Sigma Chemical Co. (St. Louis, MO). All other culture

media were purchased from Gibico-BRL (Gaithersberg, MD). TRIzol

reagent and DEPC were purchased from Invitrogen (CA, USA). RNeasy

Mini Kit with Rnase-Free DNase set was from Qiagen (California, USA).

Reverse transcription reagents were from Tiangen Co., Ltd. (Beijing,

China). Ex Taq DNA polymerase was purchased form TaKaRa (Dalian,

China). The female Wistar rats were purchased from Beijing Vital River

Experimental Animals Co. Ltd. under license no. SCXK (Beijing) 2002-

0003.

2.2. Culture of murine osteoblast MC3T3-E1 cells

The cells of murine osteoblast MC3T3-E1 were grown in Dulbecco’s

modified Eagles medium (DMEM, Gibico) supplemented with 10% (v/v)

FBS (Hyclone, USA), 100U/mL penicillin and 100mg/mL streptomycin.

Incubation was conducted in a CO2 incubator (5% CO2, 95% air) (MCO-

15AC, SANYO Co. Ltd., Japan) at 37 1C. The cells were subcultured

every 2 or 3 days in the presence of 0.25% (w/v) trypsin plus 0.02% (w/v)

ethylenediaminetetraacetic acid tetrasodium salts solution (EDTA)

(Gibico).

2.3. Assay of alkaline phosphatase (ALP) activity

MC3T3-E1 cells were seeded in 48-well plates (104 cells/well) containing

DMEM medium plus 10% FBS. After the cells attached on the bottom of

the wells, the culture medium was changed to DMEM+10%FBS medium

containing 10mM disodium b-glycerophosphate (b-GP) (Sigma, St. Louis,

MO, USA), 0.15mM ascorbic acid (Sigma) and 10�8 M dexamethasone

(Sigma) [35]. Simultaneously, different concentrations of 3HB sodium salt

were added to the culture medium in the wells. After 21-day cultivation,

the cells were washed twice with phosphate buffered saline (PBS) and

harvested in 200mL/well of lysis buffer (pH 8.2, 10mM Tris-HCl, 2mM

MgCl2 and 0.05% Triton X-100). Cells were lysed through 3 cycles of

freezing and thawing. Aliquots were reserved for protein analysis. 300mLof 8mM p-nitrophenyl phosphate (Sigma) in 0.1M sodium carbonate

buffer (pH 10) containing 1mM MgCl2 was added to the reaction mixture,

which was incubated at 37 1C for 30min (total 500mL). The reaction was

stopped by adding 50 mL of 1.0 N NaOH/well [35]. The yellow sample

solutions containing p-nitrophenol as the reaction product were measured

at wavelength of 405 nm using a microplate reader (Versamax, Molecular

Device, USA). A standard curve was prepared using p-nitrophenol

(Sigma). Total protein content of cell lysates was measured according to

Lowry et al. [36], and was expressed as the protein content of the cell

lysate.

2.4. Calcification assay

The calcium deposition of MC3T3-E1 cell culture was studied using

Alizarin red S staining solution. The Alizarin red S solution was freshly

prepared: briefly, 0.1ml of 28% ammonia solution in 100ml distilled water

was added to the solution of Alizarin red S (1 g in 100mL distilled water),

the pH was adjusted to approximately 6.4. The cells were cultured for 21

days under the same conditions as that of the ALP assay [35]. After

incubation, the cell cultures were washed three times with Dulbecco’s PBS

without calcium and magnesium salts (PBS(�)). Subsequently, the cells

were fixed by addition of 10% formalin dissolved in PBS(�) solution for

1 h. After the cell fixing process, the cell cultures were washed three times

with distilled water and stained by Alizarin red S solution for 15min. The

redundant stains were removed by washing the cell culture twice with

distilled water. Digital images (DC 300F, Leica, Germany) of Alizarin red

S stained cultures were obtained and the number of the calcification

nodules was calculated by averaging six values of different sights under the

microscopic counting (DM IRB, Leica, Germany).

2.5. The Van Kossa assay

The cells were cultured for 21 days under the same condition as that of

the experiment of ALP assay. After incubation, the cells ware washed with

150mM NaCl twice and fixed with 10% formalin dissolved in PBS(�)

solution for 1 h. After the fixing process, the cell culture was treated with

100mL of 5% AgNO3. This treatment lasted for 30min under ultraviolet

radiation. Following the removal of the AgNO3 solution, the culture

medium was washed with PBS(�) twice followed by addition of 5%

Na2S2O3 into the plate and sustained for 10min [37]. After washing the

plate with distilled water twice, the cell culture was stained with Neutral

Red for 10min. The redundant stains were removed and the digital images

of the stained cultures were obtained (DM IRB, Leica, Germany).

2.6. RT-PCR study on mRNAs encoding osteocalcin (OCN)

RT-PCR was used to detect the mRNAs encoding OCN [38]. The cells

were cultured for 21 days under the same condition as that of the

ARTICLE IN PRESSY. Zhao et al. / Biomaterials 28 (2007) 3063–3073 3065

experiment of ALP assay. After incubation, total RNA was isolated using

an RNeasy Mini Kit (Qiagen) with Rnase-Free DNase set (Qiagen, USA)

according to the manufacturer’s protocol after cell cultures were washed

twice with PBS. The cDNA was made with M-MuLV reverse transcription

reagents (Tiangen, China) and underwent a process of 5min incubation at

70 1C, 1 h reverse transcription at 44 1C, and 5min inactivation at 75 1C

(XMTB Water-bath Incubator, Changfeng Equipment Co., Ltd., China)

[38]. PCR amplification was performed and specific primer sequences for

OCN were 50-CCGGGAGCAGTGTGAGCTTA-30 and 50-TA-

GATGCGTTTGTAGGCGGTC-30 (Synthesized by Invitrogen, USA).

The PCR amplification was programmed with a 10min Taq Activation at

95 1C, and 35 cycles of denaturation for 15 s at 95 1C followed by an

annealing process for 30 s at 52 1C and an extension for 30 s at 72 1C on an

Eppendoff Mastercycler Gradient PCR System (Eppendoff, Germany)

[38]. The OCN gene was normalized against the mRNAs encoding

glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and the primer

sequences for GAPDH were 50-GTGTACATGGTTCCAGTAT-

GACTCC-30 and 50-AGTGAGTTGTCATATTTCTCGTGGT-30

(Synthesized by Invitrogen, USA). The OCN and GAPDH mRNAs

expression were analyzed by software ImageJ Version 1.33u (National

Institute of Health, USA) [39] and relative OCN mRNA expression was

calculated as follows:

Relative OCN mRNA expression ¼EOCN

EGAPDH,

where EOCN is the OCN mRNA expression analyzed by ImageJ and

EGAPDH the GAPDH mRNA expression analyzed by ImageJ.

2.7. Animal experiment

The experiments were performed on 80 specific-pathogen-free (SPF) 3-

month old female Wistar rats. The animals were maintained at room

temperature (22–25 1C) with 12:12-h light–dark cycles, they were given

distilled water to drink, and fed with a standard diet ad libitum. Bilateral

ovariectomy or a sham operation was performed under 50mg/kg sodium

pentobarbital anesthesia after 7-day adaptation. A longitudinal incision

was made inferior to the rib cage on the dorsolateral body wall. The

ovaries were exteriorized, ligated, and excised. Rats subjected to the sham

surgical procedure only had the ovaries exteriorized and then replaced.

The success of ovariectomy was confirmed by verifying the absence of

ovarian tissue at the end of each experiment.

The rats were divided into 7 groups (n ¼ 11–12): normal group

(Normal), sham-operated group (Sham), OVX group received no

treatment (Control), OVX groups receiving low 3HB (30mg/kg po daily),

medium 3HB (150mg/kg po daily), high 3HB (750mg/kg po daily), and

nilestriol (1.5mg/kg po weekly) treatments. Nilestriol, which is clinically

approved for application specifically against osteoporosis, acted as a

positive control in these studies. All animals received humane cares in

compliance with the National Institutes of Health Guide for the Care and

Use of Laboratory Animals [40].

Treatment with 3HB and nilestriol started 7 days after the surgical

procedure. 3HB was administered in 0.9% saline via daily oral

administration through instillation of the 3HB solution directly into the

rats’ mouths using burette for 12 weeks at a dose of 30, 150, or 750mg/kg

daily, while nilestriol (Beijing Four Rings Pharmaceutical Co. Ltd., China)

was administered under the above condition with a dose of 1.5mg/kg

weekly as indicated by the user guide. The control groups received the

same volume of 0.9% saline. The animals were weighed weekly and

dosages were adjusted accordingly.

One day following the conclusion of 3HB or saline administration,

animals were sacrificed and the blood samples were collected. The right

and left femoral bones were removed and freed of all connective tissue.

2.8. Biochemical analysis of serum parameters

Blood samples were obtained from the abdominal aorta after the

animals were sacrificed. Serum calcium, inorganic phosphorus and ALP

activity were measured by an autoanalyzer (Hitachi 736, Hitachi Co. Ltd.,

Japan). OCN activity of serum samples was analyzed using a commercially

available radioimmunoassay kit (ZhongSheng BeiKong Biotechnology

and Science Inc., China) based on the manufacturer’s instructions using an

automatic analyzer (ALCYON 300i, Abbott Laboratories Ltd., USA)

[41].

2.9. Bone densitometry

BMD of the right femurs was measured by dual energy X-ray

absorptiometry (DEXA) (XR-236, Norland, USA). DEXA measurements

were performed using the special software for small animals (Version

3.9.4, Norland, USA). The BMD of the distal one third of the femurs,

including the epi-metaphyseal region, were measured.

2.10. Biomechanics

The mechanical properties of intact left femurs were studied using a

bending test with three-point loading. The load was applied perpendicular

to the long axis of the femur in the mid-length of the bone supported on its

epi-physes using a Material Mechanics Testing Machine (Material

Mechanics Testing Machine WD-1, Testing Machine Research Institute

of Changchun, Changchun, China) with a loading velocity of 2mm/min.

The bone load-deformation curves, representing the relationship between

load applied to the bone and deformation in response to the load, were

analyzed.

2.11. Bone histomorphometry

After dissection, the left proximal tibia was fixed with 70% ethanol and

embedded in glycol-methacrylate (Wako Pure Chemical Industries, Japan)

without decalcification. Serial sections (5mm in thickness) were cut

longitudinally using a microtome (Model 2050; Reichert Jung, Buffalo,

NY, USA), and stained with Toluidine Blue to discriminate between

mineralized and un-mineralized bone. All rats were injected subcuta-

neously with 30mg/kg body weight tetracycline hydrochloride (Sigma) 10

and 3 days prior to sacrifice. Five micrometers of unstained sections were

examined under a fluorescent microscope to visualize tetracycline

hydrochloride.

The morphometrical parameters were determined by computer-assisted

image analyzing system (Q550IW, Leica, Germany) with the advanced

software (Leica Qwin Pro, Version 2.2, Germany). Trabecular bone

volume (TBV%) is the percentage of trabecular bone volume to the whole

bone volume. Mean trabecular thickness (MTT, mm) was determined as

the average thickness of the trabecular bones. Mineral apposition rate

(MAR,mm/d) was calculated by dividing the labeling width by the number

of days between the two tetracycline hydrochloride administrations [42].

2.12. Statistical analysis

All data were presented as the mean value7standard deviation (SD) of

each group. Variation between groups was evaluated using the Student’s t-

test, with a confidence level of 95% (po0.05) considered statistically

significance and 99% (po0.01) considered very significant.

3. Results

3.1. Effect of 3HB on differentiation of osteoblast MC3T3-

E1

ALP activity assay, calcium deposition level and assay-ing calcium content were used to evaluate differentiation ofMC3T3-E1 [35–37].

ARTICLE IN PRESSY. Zhao et al. / Biomaterials 28 (2007) 3063–30733066

ALP is a representative enzyme for indication ofosteoblast differentiation [35,36]. In the presence ofdifferent concentrations of 3HB, MC3T3-E1 cells pro-duced increasing ALP activity at concentrations up to0.01 g/L 3HB (Fig. 1) after 21 days. No obvious differencewas observed for 3HB concentration ranging from 0.02 to0.1 g/L. It appeared that low concentration of 3HB couldeffectively enhance the ALP activity of MC3T3-E1 (Fig. 1).

Alizarin red S solution is a traditional approach forevaluating the calcium deposition [35]. After cultivationwith 3HB for 21 days, the calcification parts represented byred color in MC3T3-E1 cell culture showed clearlyincreasing intensity with increasing 3HB concentrations(Fig. 2(a)). Besides the obvious phenomenon of colorchange, the number of calcification nodules from 6 entireviews under microscopic counts showed a similar 3HBproportional dependent trend (Fig. 2(b)). The ALP activityand calcium deposition results strongly suggested that theosteoblast differentiation could be efficiently stimulated inthe presence of 3HB.

Van Kossa assay measuring calcium contents in cellcultures provides further support to the above proposalthat 3HB stimulates osteoblast differentiation (Fig. 3). VanKossa assay shows calcification areas in cell culturesstained as black, whereas the nuclei as red (Fig. 3). It isobviously observable that the calcification area marked asblack increased with increasing 3HB concentration, and thecells cultured with 3HB tended more to stretch along the

**

0

10

20

30

40

50

60

70

80

90

0(-) 0(+) 0.005 0.01 0.02 0.05 0.1

3HB Concentration (g/L)

ALP

Activity (

mU

/mg)

*

Fig. 1. Murine osteoblast MC3T3-E1 differentiation study by alkaline

phosphatase (ALP) activity assay. Cells at an initial concentration of

�104/well were grown in DMEM medium supplemented with 10% FBS.

After the cells attached on the bottom of the wells, the culture medium was

changed with fresh DMEM+10% FBS medium containing three bone

inducing compounds, including 10mM disodium b-glycerophosphate (b-GP), 0.15mM ascorbic acid and 10�8 M dexamethasone. Simultaneously,

DL-3-hydroxybutyrate sodium salt (3HB) with concentrations of 0, 0.005,

0.01, 0.02, 0.05 or 0.1 g/L was added to the above cultures with each

containing 6 parallel studies. The incubation lasted 21 days. The cell lysate

containing p-nitrophenol as the reaction product was measured at a

wavelength of 405 nm. The total protein content of cell lysates was

measured, and was expressed as the protein content of the cell lysate (mU/

mg). 0(�): cells were cultured on DMEM+10% FBS; 0(+) and all 3HB

containing studies: cells were grown on DMEM+10% FBS containing

the above three bone inducing compounds. *: po0.05, compared with the

0(+) group. **: po0.01, compared with the 0(+) group.

longitudinal axis compared with the control groups. Suchtendency became more obvious at 3HB concentrations of0–0.02 g/L. In addition, the intact nuclei areas stained asred spots indicated that the cells were in a healthy growthstate.Elevated osteoblast differentiation under 3HB adminis-

tration was also evidenced by RT-PCR. The expression ofmRNAs encoding OCN, which is an important marker forosteoblast and bone late-stage differentiation, was exam-ined [38] (Fig. 4). Based on the calcification level observedin Alizarin red S staining assay and Van Kossa assay,positive control group as well as cells administrated with0.01, 0.05 and 0.1 g/L 3HB, were selected to be evaluated.RT-PCR and image analysis showed a significant increas-ing tendency in relative OCN expression level when the3HB concentration increased (Figs. 4(a) and (b)). Thisphenomenon strongly supports the elevated calcificationlevel observed in Alizarin red S staining assay and VanKossa assay. All these phenomena confirmed the positiveeffect of 3HB on osteoblast differentiation, that is,MC3T3-E1 cells differentiated more quickly under the3HB administration than under the normal condition.

3.2. Effect of 3HB on in vivo bone tissue growth evaluated

by biochemical analysis of serum parameters

In vitro results showed the stimulating effect of 3HB onosteoblast differentiation. To verify potential applicationof 3HB as anti-osteoporosis agent, in vivo experiment wascarried out using 3 month old female Wistar rats withbilateral ovariectomy performed to induced osteoporosis;3HB was orally administered on control, sham or OVXrats. The bone growth in the experimental rats wasevaluated based on biochemical analysis of serum para-meters, bone densitometry, biomechanics, and bonehistomorphometry.Animal blood samples were collected from the abdominal

aorta for the biochemical analysis of serum parameters.There was no significant difference on serum phosphateamong various animal groups with or without OVX, andthe decrease on ALP activity following ovariectomy wasnot significant either (Table 1). However, ALP activity wassignificantly increased in cells cultivated in mediumconcentration of 3HB and in nilestriol compared with thatof the normal and sham groups. Bilateral ovariectomy alsoled to a decrease in serum calcium (by 26.2%) and astatistically significant increase in serum OCN (by 27.9%).This phenomenon resulted in elevated bone resorption,which would induce the OCN in the extracellular matrix tobe released into blood [43]. Noticeably, all animalsadministered with 3HB and nilestriol showed enhancedserum calcium and maintained those calcium levels close tothe normal and sham values compared with the OVXcontrol group. In addition, administrations of medium andhigh concentrations of 3HB and nilestriol, respectively,inhibited the increase of serum OCN that was observed inOVX animals due to the lack of estrogen, whereas the ALP

ARTICLE IN PRESS

0(-) 0(+) 0.005 g/L

0.01 g/L 0.02 g/L 0.05 g/L 0.1 g/L

0

10

20

30

40

50

60

70

80

90

100

0(-) 0(+) 0.005 0.01 0.02 0.05 0.1

3HB Concentration (g/L)

Am

ount of C

alc

ific

ation N

odule

s

* *

* *

* * * *

* *

Fig. 2. Murine osteoblast MC3T3-E1 differentiation study by Alizarin red S staining. (a) Calcium deposition study of MC3T3-E1 stained by Alizarin red

S, the calcium deposition nodules were stained as dark red areas. (b) Average numbers of calcium deposition nodules obtained by counting 6 microscopic

views (400� ) for each group. The Alizarin red S solution was freshly prepared. The cells were cultured for 21 days under the same conditions as that of the

experiment for ALP assay described in Fig. 1. After the incubation, the cells were washed three times with Dulbecco’s phosphate-buffered saline without

calcium and magnesium salts (PBS(�)), followed by 10% formalin fixation in PBS(�) solution for 1 h. Subsequently, the cell cultures were washed three

times with distilled water and stained using Alizarin red S solution for 15min. After staining, the cell cultures were washed twice with distilled water to

completely remove the redundant stains. 0(�): cells were cultured on DMEM+10% FBS; 0(+) and all 3HB containing studies: cells were grown on

DMEM+10% FBS containing the three bone inducing compounds described in Fig. 1; **: po0.01, compared with the 0(+) group.

Y. Zhao et al. / Biomaterials 28 (2007) 3063–3073 3067

activity in the blood was not significantly affected comparedwith the OVX control animals (Table 1). In fact, ALPactivity and calcium content in the blood samples of OVXrats were higher compared with OVX control animals.These in vivo phenomena are in agreement with thedifferentiation experiment done in vitro. It appeared that3HB helped maintain the conditions for normal bone tissuegrowth under disordered physiological conditions likeovariectomy.

3.3. Effect of 3HB on in vivo bone tissue growth evaluated

by bone mechanics

Right femurs of the OVX rats were used to evaluatethe BMD. The BMD of the femurs was significantly

reduced by 8.9% in the OVX group (Table 2). Oraladministrations using medium, high concentrationsof 3HB and nilestriol efficiently prevented significantBMD decrease. Instead, the BMD remained close to thesame level as that of the normal and sham groups.This phenomenon strongly suggests that the 3HB couldenhance osteoblast differentiation. Moreover, it efficientlymaintained the bone tissue growth under abnormalcondition.Intact left femurs of the OVX rats were collected for

testing bone mechanics. Results indicated that differenceon maximal deformation between OVX and sham groupswas statistically significant (Table 3). Maximal loaddecreased in the OVX control group with less statisticalsignificance. Besides, there was no significant difference on

ARTICLE IN PRESS

Fig. 3. Murine osteoblast MC3T3-E1 differentiation study using Van Kossa assay. The cells were cultured for 21 days under the same conditions as that

of the experiment for ALP assay described in Fig. 1. The calcium containing area was stained as black in color and the nuclei area as dark red spot. The

differentiated cells appeared to stretch along the longitudinal axis compared with that of the control ones. 0(�): cells were cultured on DMEM+10%

FBS; 0(+) and all 3HB containing studies: cells were grown on DMEM+10% FBS containing the three bone inducing compounds described in Fig. 1;

arrows: (A) The nuclei areas stained as dark red spot; (B) the calcium containing area stained as black in color.

Y. Zhao et al. / Biomaterials 28 (2007) 3063–30733068

ARTICLE IN PRESS

3HB Concentration (g/L) 0 0.01 0.05 0.1

GAPDH�

OCN�

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0 0.01 0.05 0.1

3HB Concentration (g/L)

Re

lative

OC

N E

xp

ressio

n

Fig. 4. Relative osteocalcin (OCN) expression level study by RT-PCR.

The cells were cultured for 21 days under the same conditions as that of

the experiment for ALP assay described in Fig. 1. Cells were grown on

DMEM+10% FBS containing the three bone inducing compounds

described in Fig. 1. (a) Osteocalcin (OCN) and glyceraldehyde-3-

phosphate dehydrogenase (GAPDH) mRNA expression. (b) Relative

OCN mRNA expression normalized against GAPDH (data analyzed

using software ImageJ Version 1.33u). The OCN and GAPDH mRNAs

expression were analyzed by software ImageJ Version 1.33u (National

Institute of Health, USA) and relative OCN mRNA expression was

calculated as follows: Relative mRNA expression ¼ EOCN=EGAPDH;

EOCN: OCN mRNA expression analyzed by ImageJ and EGAPDH:

GAPDH mRNA expression analyzed by ImageJ.

Table 1

Effect of 3HB on the serum parameters of the experimental animals

Group n ALP (IU/L) Ca

Normal 11 37.773.0 0.

Sham 11 31.075.2 0.

OVX

Control 11 29.072.6 0.

Low 3HB 12 30.776.4 0.

Medium 3HB 12 42.374.2b 0.

High 3HB 12 34.075.3 0.

Nilestriol 11 41.570.7b 0.

The experiments were performed on 80 specific-pathogen-free (SPF) 3-month

light–dark cycles. Bilateral ovariectomy or a sham operation was performed u

longitudinal incision was made inferior to the rib cage on the dorsolateral body

the sham surgical procedure only had the ovaries exteriorized and then replace

acted as a positive control in the current study.

Treatment with 3HB and nilestriol started 7 days after the surgical procedure.

weeks at a dose of 30, 150, or 750mg/kg daily. Nilestriol was administered in 0

kg. Control groups received same volume of 0.9% saline.apo0.01 vs. Sham.bpo0.01 vs. Control.cpo0.05 vs. Control.

Y. Zhao et al. / Biomaterials 28 (2007) 3063–3073 3069

femur rigidness among groups (Table 3). Administration of3HB with different concentrations showed significantincrease on maximal deformation of the femurs. Inaddition, administration of 3HB with medium concentra-tion and nilestriol significantly increased the maximal loadof the femurs. In summary, both the maximal load and themaximal deformation of the OVX rat femurs wereimproved by 3HB compared with the OVX control.Although the rigidness of the bone did not fluctuateamong groups, it could still be concluded that 3HBadministration improved the bone mechanics of the OVXrats compared with the OVX controls.Bone histomorphometry of the rats was evaluated in

terms of TBV%, MTT and MAR (Table 4). Clearly, alltrabecular microstructure parameters in the tibia decreasedfollowing OVX. However, only the TBV% data haddifferences with statistical significance, the changes ofMTT and MAR were less statistically significant. In allcases, administration of 3HB and nilestriol showedsignificant improvement on TBV%, while differences inMTT and MAR following these administrations werenot statistically significant. The data again indicatedthat 3HB effectively reduced bone defects induced byovariectomy.

4. Discussion

Good biocompatibility is one of the most importantrequirements for the development of any medical implantmaterials. The biopolyester family PHA, especially thecopolyesters of 3HB and 3-hydroxyhexanoate (PHBHHx),had shown strong and controllable mechanical properties[18] combined with a good biocompatibility to various cells[17–24,44]. The foremost ability among these properties is

(mmol/L) P (mmol/L) Osteocalcin (ng/mL)

9270.22 1.0470.53 1.8670.19

8470.02 0.9570.04 1.8370.14

6270.07a 0.9470.17 2.3470.26a

8870.17b 0.8370.34 2.1470.19

7370.11c 0.9370.12 2.0670.21c

8870.16b 0.9670.25 2.0470.19c

8570.23c 0.9270.21 1.9570.21b

old female Wistar rats kept at room temperature (25 1C) with 12:12-h

nder 50mg/kg sodium pentobarbital anesthesia after 7-day adaptation. A

wall. The ovaries were exteriorized, ligated, and excised. Rats subjected to

d. Nilestriol, which is a clinically certificated specific against osteoporosis,

3HB was administered in 0.9% saline by daily oral administration for 12

.9% saline by weekly oral administration for 12 weeks at a dose of 1.5mg/

ARTICLE IN PRESS

Table 2

Effect of 3HB on the bone mineral density (BMD) of rat femurs

Group n BMD

Normal 11 0.26270.015

Sham 11 0.25970.017

OVX

Control 11 0.23670.018a

Low 3HB 12 0.24670.017

Medium 3HB 12 0.25970.017b

High 3HB 12 0.25870.013b

Nilestriol (positive control) 11 0.26370.018c

Experimental conditions were the same as described in Table 1.apo0.01 vs. Sham.bpo0.05 vs. Control.cpo0.01 vs. Control.

Table 3

Effect of 3HB on the bone biomechanics of rat femurs

Group n Maximal load (mm) Maximal deformation (kg) Rigidness (kg/mm)

Normal 11 8.5470.96 0.64670.124 17.6174.41

Sham 11 8.3570.54 0.65770.044 16.5370.94

OVX

Control 11 8.1270.39 0.60370.054a 17.0870.86

Low 3HB 12 8.1070.58 0.66570.065b 16.1471.58

Medium 3HB 12 8.6570.49b 0.65170.053b 18.7271.80

High 3HB 12 8.3570.65 0.66870.063b 17.2271.80

Nilestriol 11 8.7370.64b 0.63870.049 17.5571.57

Experimental conditions were the same as described in Table 1.apo0.01 vs. Sham.bpo0.05 vs. Control.

Table 4

Effect of 3HB on the bone histomorphology of rat tibia

Group n TBV (%) MTT (mm) MAR (mm/d)

Normal 11 17.970.74 60.25711.15 1.0570.21

Sham 11 18.070.63 58.4379.03 0.9670.20

OVX

Control 11 15.070.44a 54.3978.50 0.7170.13

Low 3HB 12 17.871.50b 59.7279.42 0.8270.19

Medium 3HB 12 17.672.43b 59.5178.77 0.8970.24

High 3HB 12 18.571.83c 62.5179.79 0.8970.17

Nilestriol 11 17.571.46b 63.02710.01 0.9170.24

Experimental conditions were the same as described in Table 1.

TBV%: trabecular bone volume.

MTT: mean trabecular thickness.

MAR: mineral appositional rate.apo0.01 vs. Sham.bpo0.05 vs. Control.cpo0.01 vs. Control.

Y. Zhao et al. / Biomaterials 28 (2007) 3063–30733070

to support cell and tissue growth. However, the detailedmechanism for the good biocompatibility of PHA is stillnot clear. If PHA is to be used for medical implantapplication, it is highly desirable to clearly explain themechanism of its good biocompatibility.

3HB is one of the important degradation products ofPHA. 3HB is also produced in the liver by the degradationof long-chain fatty acids; it is transported through plasmato peripheral tissues as an important energy source [6,14].3HB has been speculated as one of the key factors for 3HB

ARTICLE IN PRESSY. Zhao et al. / Biomaterials 28 (2007) 3063–3073 3071

containing PHA such as PHBHHx to support growth ofdifferent types of cells. Noticeable, 3HB was reported todelay neurocytes apoptosis [12].

Based on the previous study of osteoblast cell growth onPHBHHx and tissue reactions induced by implanting PHA[1–3,20,24], the effect of 3HB that is also a monomerreleased by degradation of 3HB containing PHA, on bonecell growth, should be explained to support the clinicalapplication of this type of PHA.

In this study, the in vitro stimulative effect of 3HB ondifferentiation of osteoblast MC3T3-E1 was confirmed(Figs. 1–4). 3HB was found to increase ALP activity ofMC3T3-E1 in a dose-dependent way, especially at 3HBconcentrations lower than 0.01 g/L (Fig. 1). Meanwhile,3HB also led to increasing calcium deposition in MC3T3-E1, especially at concentrations lower than 0.02 g/L(Fig. 2(a)), the number of calcium deposition nodules isin direct proportion to 3HB concentration (Fig. 2(b)).Similar observation was obtained in results of Van Kossastaining to MC3T3-E1, which indicated that the extra-cellular and intracellular calcification levels were enhancedby 3HB (Fig. 3). The elevated OCN mRNA expressionlevel detected by RT-PCR strengthened the observation inAlizarin red S staining and Van Kossa assay (Fig. 4), andthis phenomenon could be a result of a quicker progressionof cellular differentiation under the 3HB administration.All the results demonstrated enhanced differentiation ofMC3T3-E1 cultured in the presence of different 3HBconcentrations.

Besides, it is noticeable that the in vitro and in vivo

studies on OCN seemed to have contrary results (Fig. 4,Table 1). However, these two results are totally constant,and they all strongly support the favorable effect of 3HBon osteoblast differentiation and bone formation. In the in

vitro experiment, OCN is a late-stage differentiation andbone formation marker, and the increased OCN mRNAexpression indicated the elevated osteoblast differentiationlevel. On the other hand, in the in vivo experiment,ovariectomy would induce quick bone absorption andsubsequently induce OCN to release from extracellularmatrix into blood. Thus the serum OCN would beincreased after ovariectomy [43]. The decreased serumOCN by 3HB administration indicated that 3HB couldreduce bone absorption and maintain normal bonefunction (Table 1). Thus, both the in vitro and in vivo

study on OCN strongly support that 3HB could effectivelystimulate osteoblast differentiation and maintain bonenormal function in the OVX rats.

The in vivo experiment also confirmed the positive effectof 3HB on bone tissue growth. The serum ALP activity, thecalcium content, BMD, bone mechanisms and TBV% inall 3HB treated bilateral ovariectomy (OVX) rats weremaintained, or even more, exceeded the normal level,compared with the OVX control group without adminis-trations of 3HB and nilestrol (Tables 1–4). The efficacy of3HB is comparable to that of the positive control drugnilestriol, a type of estrogen that promotes growth of

osteoblasts and osteocytes [45], nilestrol has already beenapproved as a drug to treat the osteoporosis resulting frommenopause and depressed estrogen level.Estrogen deficiency resulting from OVX leads to an

increased rate of bone remodeling (both resorption andformation) and an imbalance between bone resorption andformation [46]. In the OVX rats, administration of 3HB withmedium and high concentration helped maintain the femurBMD at a normal value (Table 2). 3HB administration alsomaintained the maximal deformation and maximal load ofthe OVX femurs compared to the control value (Table 3).3HB and nilestriol administration significantly improvedTBV% compared with OVX control group (Table 4), yet thechanges in MTT and MAR were not statistically significant.Although these results did not directly support the fact that3HB counteracts the accelerated bone loss resulting fromovariectomy, it did confirm that 3HB administration helpednormalize the BMD in femurs and improved the mechanicalproperties of the femurs (Table 4), which could be attributedto the improved growth of osteoblasts as evidenced by thein vitro experiment in this study.Administration of 3HB prevented the decrease of

calcium content and ALP activity resulting from ovar-iectomy as effectively as nilestriol compared with that ofthe OVX control (Table 1). Meawhile, ovariectomyincreased serum OCN, which is released from extracellularmatrix and resulted in quick bone absorption [43].Administration of 3HB with medium or high concentrationalso effectively decreased OCN as nilestriol did. The in vivo

results agreed well with that of the in vitro data (Figs. 1–4and Tables 1–4).It was reported that both the in vitro hydrolytic

degradation and in vivo biodegradation released 3HB wasused in this study with comparable concentrations [19,47].Moreover, previous study indicated the orally admini-strated 3HB would elevate the ketone body in blood to alevel comparable to that in the in vitro experiment in thisstudy (0.005–0.1 g/L) [43], as demonstrated by 3HBconcentration and its effects are shown both in the in vivo

and in vitro experiment here (Figs. 1–4, Tables 1–4).Admittedly, osteoblasts undergo an orderly develop-

mental progression in the bone multicellular units thatultimately ends in apoptosis [48]. The equilibrium ofosteoblast proliferation, differentiation, and apoptosisdetermine the size of the osteoblast population at anygiven time. However, mineralization takes place late in thematuration and aging phase [49]. Prolonging cell life at thispoint is critical for bone formation. Conversely, early celldeath (apoptosis) will diminish bone formation. Obviously,the correlation between the effect of 3HB administrationon osteoblast differentiation and the maintenance of bonenormal function in the OVX rats is reasonable.

5. Conclusion

3HB, one of the key degradation products of PHA,supported in vitro differentiation of murine osteoblast

ARTICLE IN PRESSY. Zhao et al. / Biomaterials 28 (2007) 3063–30733072

MC3T3-E1 in direct proportion to its concentration.Administration of 3HB to ovariectomized (OVX) rats alsoimproved the quality of bone tissues compared with that ofthe OVX rats without 3HB administration. Combined withits nontoxicity and rapid metabolizable ability, 3HB canbecome an effective agent against osteoporosis. Moreover,3HB-containing PHA could be used for bone tissue repairdue to the favorable properties of its degradation product3HB.

Acknowledgments

This research was supported by National NaturalScience Foundation for Outstanding Young InvestigatorAward (Grant no. 30225001) and Natural SciencesFoundation of China (Grant nos. 30570024 and20334020). The Foundation for Basic Research in Tsin-ghua University (Grant no. JCjc2005070), 973 BasicResearch Fund and 863 High Tech project (Grant no.2006AA02Z242) awarded to Chen GQ (Grant no.2007CB707804) also contributed to this study. The authorsthank Professor Rong-qing Zhang for his donation ofosteoblast MC3T3-E1 cells. Assistance from First Hospitalaffiliated to Peking University on analysis of serumparameters, Third Hospital affiliated to Peking Universityon the BMD analysis, China Academy of Chinese MedicalSciences on the biomechanics analysis, and GeneralHospital of People’s Liberation Army (PLA) on the bonehistomorphology analysis are gratefully acknowledged. Wealso thank Dr. Mike Leski for his help in improving themanuscript both in the language and discussion.

References

[1] Wang YW, Wu Q, Chen GQ. Poly(3-hydroxybutyrate-co-3-hydro-

xyhexanoate) scaffolds with good biocompatibility for osteoblast

proliferation and differentiation. Biomaterials 2004;25:669–75.

[2] Wang YW, Wu Q, Chen JC, Chen GQ. Evaluation of hydroxyapatite

(HAP) incorporation into 3-dimensional scaffold made of poly(3-

hydroxybutyrate-co-3- hydroxyhexanoate) (PHBHHx) for bone

reconstruction. Biomaterials 2004;26:899–904.

[3] Wang YW, Yang F, Cheng YC, Yu PHF, Chen JC, Wu Q, et al.

Effect of composition of poly (3-hydroxybutyrate-co-3-hydroxylhex-

anoate) on growth of fibroblast and osteoblast. Biomaterials

2005;26:755–61.

[4] Deng Y, Zhao K, Zhang XF, Hu P, Chen GQ. Study on the three-

dimensional proliferation of rabbit articular cartilage derived

chondrocytes on polyhydroxyalkanoate scaffolds. Biomaterials

2002;23:4049–56.

[5] Robinson A, Williamson D. Physiological roles of ketone bodies as

substrates and signals in mammalian tissues. Physiol Rev

1980;60:143–87.

[6] Curi R, Lagranha CJ, Doi SQ, et al. Molecular mechanisms of

glutamine action. J Cell Physiol 2005;204(2):392–401.

[7] Katayama M, Hiraide A, Sugimoto H, Yoshioka T, Sugimoto T.

Effect of ketone bodies on hyperglycemia and lactic acidemia in

hemorrhagic stress. J Parenter Enteral Nutr 1994;18:444–6.

[8] Lin T, Koustova E, Chen HZ, Rhee PM, Kirkpatrick J, Alam HB.

Energy substrate-supplemented resuscitation affects brain monocar-

boxylate transporter levels and gliosis in a rat model of hemorrhagic

shock. J Trauma 2005;59:1191–202.

[9] Mizobata Y, Hiraide A, Katayama M, Sugimoto H, Yoshioka T,

Sugimoto T. Oxidation of D(-)3-hydroxybutyrate administered to rats

with extensive burns. Surg Today 1996;26:173–8.

[10] Zou Z, Sasaguri S, Rajesh KG, Suzuki R. DL-3-hydroxybutyrate

administration prevents myocardial damage after coronary

occlusion in rat hearts. Am J Physiol-Heart C Physiol 2002;283(5):

H1968–74.

[11] Suzuki M, Sato K, Dohi S, Sato T, Matsuura A, Hiraide A. Effect of

beta-hydroxybutyrate, a cerebral function improving agent, on

cerebral hypoxia, anoxia and ischemia in mice and rats. Jpn J

Pharmacol 2001;87:143–50.

[12] Kashiwaya Y, Takeshima T, Mori N, Nakashima K, Clarke K,

Veech RL. D-beta-hydroxybutyrate protects neurons in models of

Alzheimer’s and Parkinson’s disease. Proc Nat Acad Sci USA

2000;97:5440–4.

[13] Nakamura S, Shibuya M, Saito Y, Nakashima H, Saito F, Higuchi

A, et al. Protective effect of D-beta-hydroxybutyrate on corneal

epithelia in dry eye conditions through suppression of apoptosis.

Invest Ophthalmol Vis Sci 2003;44:4682–8.

[14] Tieu K, Perier C, Caspersen C, Teismann P, Wu DC, Yan SD,

et al. D-beta-hydroxybutyrate rescues mitochondrial respiration and

mitigates features of Parkinson disease. J Clin Invest 2003;112:

892–901.

[15] Plecko B, Stoeckler-Ipsiroglu S, Schober E, Harrer G, Mlynarik V,

Gruber S, et al. Oral beta-hydroxybutyrate supplementation in two

patients with hyperinsulinemic hypoglycemia: monitoring of beta-

hydroxybutyrate levels in blood and cerebrospinal fluid, and in the

brain by in vivo magnetic resonance spectroscopy. Pediatr Res

2002;52:301–6.

[16] Veech RL, Chance B, Kashiwaya Y, Lardy HA, Cahill Jr GF.

Ketone bodies, potential therapeutic uses. IUBMB Life 2001;51:

241–7.

[17] Zhao K, Deng Y, Chen JC, Chen GQ. Polyhydroxyalkanoate (PHA)

scaffolds with good mechanical properties and biocompatibility.

Biomaterials 2003;24:1041–5.

[18] Chen GQ, Wu Q. The application of polyhydroxyalkanoates as tissue

engineering materials. Biomaterials 2005;26:6565–78.

[19] Qu XH, Wu Q, Zhang KY, Chen GQ. In vivo studies of poly(3-

hydroxybutyrate-co-3-hydroxyhexanoate) based polymers: biodegra-

dation and tissue reactions. Biomaterials 2006;27:3540–8.

[20] Wang YW, Wu Q, Chen GQ. Increased hydrophilicity of microbial

polyhydroxyalkanoates via hyaluronan coating reduced growth of

mouse fibroblast L929 cells. Biomaterials 2003;25:4621–9.

[21] Deng Y, Lin XS, Zheng Z, Deng JG, Chen JC, Ma H, et al.

Poly(hydroxybutyrate-co-hydroxyhexanoate) promoted production

of extracellular matrix of articular cartilage chondrocytes in vitro.

Biomaterials 2003;24:4473–81.

[22] Yang XS, Zhao K, Chen GQ. Effect of surface treatment on the

biocompatibility of microbial polyhydroxyalkanoates. Biomaterials

2002;23:1391–7.

[23] Zhao K, Deng Y, Chen JC, Chen GQ. Polyhydroxyalkanoate (PHA)

scaffolds with good mechanical properties and biocompatibility.

Biomaterials 2003;24:1041–54.

[24] Yang M, Zhu SS, Chen Y, Chang ZJ, Chen GQ, Gong YD, et al.

Studies on bone marrow stromal cells affinity of poly(3-hydroxybu-

tyrate-co-3-hydroxyhexanoate). Biomaterials 2004;25:1365–73.

[25] Kang IK, Choi SH, Shin DS, Yoon SC. Surface modification of

polyhydroxyalkanoate films and their interaction with human

fibroblasts. Int J Biol Macromol 2001;28(3):205–12.

[26] Cheng S, Wu Q, Yang F, Xu M, Leski M, Chen GQ. Influence of DL-

beta-hydroxybutyric acid on cell proliferation and calcium influx.

Biomacromolecules 2005;6:593–7.

[27] Cheng S, Chen GQ, Leski M, Zou B, Wang Y, Wu Q. The effect of D,

L-X-hydroxybutyric acid on cell death and proliferation in L929 cells.

Biomaterials 2006;27:3758–65.

[28] Shaarawy M, Hasan M. Serum bone sialoprotein: a marker of bone

resorption in postmenopausal osteoporosis. Scand J Clin Lab Inv

2001;61(7):513–21.

ARTICLE IN PRESSY. Zhao et al. / Biomaterials 28 (2007) 3063–3073 3073

[29] Morley P, Whitfield JF, Willick GE. Parathyroid hormone: an

anabolic treatment for osteoporosis. Curr Pharm Design

2001;7(8):671–87.

[30] Lee TC, McHugh PE, O’Brien FJ, O’Mahoney D, Taylor D, Bruzzi

M, et al. Bone for life: osteoporosis, bone remodeling and computer

simulation. In: Prendergast PJ, McHugh PE, editors. Topics in bio-

mechanical engineering. Galway, Ireland, 2004. p. 58–93.

[31] Canalis E. Novel treatments for osteoporosis. J Clin Invest

2000;106:177–9.

[32] Delmas PD. Treatment of postmenopausal osteoporosis. Lancet

2002;359:2018–26.

[33] Schiller PC, D’Ippolito G, Roos BA, Howard GA. Anabolic or

catabolic responses of MC3T3-E1 osteoblastic cells to parathyroid

hormone depend on time and duration of treatment. J Bone Miner

Res 1999;14:1504–12.

[34] Corral DA, Amling M, Priemel M, Loyer E, Fuchs S, Ducy P, et al.

Dissociation between bone resorption and bone formation in

osteopenic transgenic mice. Proc Nat Acad Sci USA 1998;95:

13835–40.

[35] Kazuo Isama, Toshie Tsuchiya. Enhancing effect of poly(l-lactide) on

the differentiation of mouse osteoblast-like MC3T3-E1 cells. Bioma-

terials 2003;24:3303–9.

[36] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein

measurement with the Folin phenol reagent. J Biol Chem

1951;193:265–75.

[37] Harnano T, Chiba D, Nakatsuka K, Nagahata M, Teramoto A,

Kondo Y, et al. Evaluation of a polyelectrolyte complex (PEC)

composed of chitin derivatives and chitosan, which promotes the rat

calvarial osteoblast differentiation. Polym Adv Technol

2002;13:46–53.

[38] Chen VJ, Smith LA, Ma PX. Bone regeneration on com-

puter-designed nano-fibrous scaffolds. Biomaterials 2006;27:

3973–9.

[39] Feige JN, Sage D, Wahli W, et al. PixFRET, an ImageJ plug-in for

FRET calculation that can accommodate variations in spectral bleed-

throughs. Microsc Res Techn 2005;68(1):51–8.

[40] National Research Council. Guide for the care and the use of

laboratory animals. Washington DC: National Academy Press; 1996.

[41] Iwasaki-Ishizuka Y, Yamato H, Murayama H, Ezawa I, Kurokawa

K, Fukagawa M. Menatetrenone rescues bone loss by improving

osteoblast dysfunction in rats immobilized by sciatic neurectomy.

Life Sci 2005;76:1721–34.

[42] Xie F, Wu CF, Lai WP, Yang XJ, Cheung PY, Yao XS, et al. The

osteoprotective effect of Herba epimedii (HEP) extract in vivo and

in vitro. Evid Based Complement Alternat Med 2005;2(3):353–61.

[43] Van Soesbergen RM, Lips P, et al. Bone metabolism in rheumatoid

arthritis compared with postmenopausal osteoporosis. Ann Rheum

Dis 1986;45:149.

[44] Engelmayr GC, Hildebrand DK, Sutherland FWH, Mayer JE, Sacks

MS. A novel bioreactor for the dynamic flexural stimulation of tissue

engineered heart valve biomaterials. Biomaterials 2003;24:2523–32.

[45] Tomkinson A, Gevers EF, Wit JM, Reeve J, Noble BS. The role of

estrogen in the control of rat osteocyte apoptosis. J Bone Miner Res

1998;13:1243–50.

[46] Raisz LG. Pathogenesis of osteoporosis: concepts, conflicts, and

prospects. J Clin Invest 2005;115:3318–25.

[47] Shangguan YY, Wang YW, Wu Q, Chen GQ. The mechanical

properties and in vitro biodegradation and biocompatibility of UV-

treated poly (3-hydroxybutyrate-co-3-hydroxyhexanoate). Biomater-

ials 2006;27:2349–57.

[48] Lynch MP, Capparelli C, Stein JL, Stein GS, Lian JB. Apoptosis

during bone-like tissue development in vitro. J Cell Biochem

1998;68:31–49.

[49] Hock JM, Krishnan V, Onyia JE, Bidwell JP, Milas J, Stanislaus D.

Osteoblast apoptosis and bone turnover. J Bone Miner Res

2001;16:975–84.