Effects of Medicinal Herb Salvia Miltiorrhiza on OsteoblasticCells In Vitro

Alice Chin, Yanqi Yang, Lei Chai, Ricky W. K. Wong, Abou-Bakr M. Rabie

Orthodontics, Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China

Received 2 March 2010; accepted 6 January 2011

Published online 11 February 2011 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.21376

ABSTRACT: Sufficient osteoinduction is essential for the success and effectiveness of bone grafting. It was previously found that SalviaMiltiorrhiza (SM), a commonly used Chinese herb increased osteogenesis in vivo. The aim of this study is to investigate the effects of SM onbone cells in vitro, in an attempt to get a better understanding on howSMcan promote bone remodeling.MC3T3-E1, an osteoblastic cell line,was cultured with SM for different time intervals (24, 48, and 72 h), whereas the control group consisted of cells cultured without anyintervention. The mRNA expression of alkaline phosphatase (ALP), osteocalcin (OCN), osteoprotegerin (OPG), and the receptor activatorof nuclear factor kappaB ligand (RANKL) were examined by real-time polymerase chain reaction (qPCR). The expression of ALP showed anearly increaseat 24 hby50%(p < 0.001)andat48 hby13%(p < 0.001).OCNwasdecreasedby22%at24 h (p < 0.001)but increasedby50%and 88% at 48 and 72 h, respectively (p < 0.001). RANKL showed an early increase at the first two time points of 24 and 48 h by 45%(p < 0.001) and 36% (p < 0.01), respectively, while OPG was up-regulated at the latter two time points by 10% at 48 h (p < 0.01) and 68%at 72 h (p < 0.001). Thus, OPG/RANKL was down-regulated first, and then up-regulated. SM enhances bone remodeling by regulatingthegeneexpressionofALP,OCN,OPG,andRANKL. It is apotentialmedicinal herb tobeutilized in theapplication that requires stimulationin bone cell activities. � 2011 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 29:1059–1063, 2011

Keywords: Salvia Miltiorrhiza; osteoblasts

Active ingredients from commonly used TraditionalChinese Medicines (TCMs) have been found to beosteoinductive.1,2 Naringin, psoralen and others havebeen found to induce bone formation.1 The abilitiy toinduce osteogenesis and angiogenesis are ideal qualitiesfor bone grafting materials. Therapeutic angiogenesiswith vascular endothelial growth factor (VEGF) has pro-ven to be effective in stimulating localized angiogenesisand improving blood circulation in animal models ofmyocardial and limb ischemia.3–5 However, severehypotension associated with the usage of growth factors,has been attributed to the release of nitric oxide andarterial vasodilation.6 Because of the side effects of thesegrowth factors, such as that of VEGF, the search foran alternative material without these unwanted effectsand at the same time with bone promoting propertiesremains a challenge. These problems thus point us tothe direction of Traditional Chinese Medicine (TCM) asa possible solution to seeking an ideal bone graftingmaterial, the formidable challenge over the years.

Radix Salvia miltiorrhiza (SM), also known asDanshen, is a common TCM that has been used forthousands of years for the treatment of cardiovasculardiseases by improving perfusion of ischemicmyocardiumand enhancing blood circulation.7 Over the years, manystudies have reported the beneficial effects of SMonbonehealing in fractures and in treatment of bone diseasessuch as osteoporosis which is mainly due to its positiveeffects on angiogenesis and enhancement of VEGFexpression. The increased new blood vessels allowmesenchymal cells to be recruited to the area and todifferentiate into cells such as osteoblasts, which willtake part in the bone remodeling process. At the same

time, SM is found to increase bone formation as well. Anin vivo study by Wong et al.8 found that SM extract incollagen matrix had the effect of increasing new boneformation by 478% locally. In addition, at a cellular level,Hu et al.9 found that SM stimulated amore rapid growthof osteoblast-like cells in early stage of culture of chickenembryo calvariae. These studies, both in vivo and in vitrosuggested that SM may increase bone formation. Themechanismmay involve an increase of angiogenesis anda direct effect on bone cells. However, the underlyingmechanism of effect on bone cells at amolecular level hasbeen unclear.

Therefore, this study aims to investigate the effectsof SM on bone cells in vitro by examining the mRNAexpression of alkaline phosphatase (ALP) and osteocal-cin (OCN)—the two important markers for bone for-mation, as well as osteoprotegerin (OPG) and thereceptor activator of nuclear factor kappa B (RANKL)—the twomarkers directly related to bone resorption, in anattempt to get a better understanding on how SM canpromote this physiological process—bone formation andbone resorption, that is, bone remodeling at a molecularlevel.

MATERIALS AND METHODSCell Culture: MC3T3-E1 Cell LineMC3T3-E1 cells, an osteoprogenitor cell line, are cultured inDulbecco’s Modified Eagle Medium (DMEM) containing 10%fetal bovine serum and antibiotics (100 IU/ml of penicillin Gand 100 mg/ml of streptomycin), and incubated at 378C in a 5%CO2humidified atmosphere.Before supplement of SM, the cellswere cultured in differentiation medium (DMEM without ser-um þ 25mg/ml ascorbic acid þ 0.648 mg/ml beta-glycerophos-phate) for 24 h. Corresponding to the control, the conditionedmedium for the test groups was supplemented with SM at afinal concentration of 100 mg/ml. The cells and conditionalmedium were collected at different culture time points of 24,48, and 72 h.

Correspondence to: Yanqi Yang (T: þ852-28590252; F: þ852-25593803; E-mail: yangyanq@hkucc.hku.hk)

� 2011 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.

JOURNAL OF ORTHOPAEDIC RESEARCH JULY 2011 1059

Danshenessencepowderwaspurchased fromChinesemedi-cine pharmacy (PurapharmNong’s, Cat. No. 1007, Hong Kong,China). Before use, the powder was dissolved in distilled waterto make 100� final concentration solution, then autoclaved.

RNA Extraction and cDNA SynthesisTotal RNA was isolated from the cells using TRIzol1 reagentaccording to the manufacturer’s instructions (Invitrogen,Carlsbad, CA). Total RNA (1 mg) was used to generate cDNAin each sample using SuperScriptTM reverse transcriptase(Invitrogen) with 1 mg Oligo(dT)12-18 (500 mg/ml) primers.

Real-Time Polymerase Chain Reaction (qPCR)One microliter of total cDNA was amplified in each PCRreaction mixture containing 10 mM of forward and reverseprimers of selected genes (Table 1). The PCR reaction mixture,SYBR1GreenPCRMasterMix (ABI, FosterCity, CA)wasusedina concentration of 10 ml. PCRamplificationwasperformedona StepOnePlus1 Real-time PCR System (96-well; ABI).

The levels of ALP, OCN, OPG, and RANKL mRNAexpression were examined by qPCR with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a housekeeping gene.The PCR program was carried out as follows: denaturation at958C for 10 min, amplification for 40 cycles (958C for 15 s; 608Cfor 1 min), and final extension at 958C for 15 min.

Statistical AnalysisThe data were analyzed using Graph PadInstat software.Copies of each genewere calculated according toDCt and stand-ard curves, and were reported as mean � SD. The differencebetweenSMgroup and control groupwas compared using t-testwith the level of significance set at p < 0.05.

RESULTSThe mRNA expression of ALP, OCN, OPG, and RANKLwas examined by real-time PCR. The effects of SM onbone formationwere evaluated by ALP andOCN expres-sion, while its effects on bone resorption were evaluatedby the levels of OPG andRANKL expression. ThemRNAexpression of ALP (Fig. 1) showed an early increase at24 h by 50% (p < 0.001) and at 48 h by 13% (p < 0.001).OCN (Fig. 1) was decreased by 22% at 24 h (p < 0.001)but increased by 50% and 88% at 48 and 72 h, respect-ively (p < 0.001).

RANKL showed an early increase at the first twotime points of 24 and 48 h by 45% (p < 0.001) and 36%(p < 0.01) respectively, while OPG was up-regulatedat the latter two time points by 10% at 48 h (p < 0.01)and 68% at 72 h (p < 0.001). OPG/RANKL was down-

regulated in thefirst two timepoints, that is, 24 and48 h,and then up-regulated at 72 h (Fig. 2). The experimentwas repeated three times and similar results wereobtained.

DISCUSSIONIn this study, SM was found to increase bone for-mation indicated by increase of ALP activity and OCNexpression (Fig. 1). MC3T3-E1 is an osteoprogenitorcell line derived from newborn calvaria that is knownto differentiate along the osteoblast pathway in thepresence of ascorbic acid10 and is able to expressmineral-ization markers. Therefore, this cell line was chosen inthe present study to analyze the stimulatory effects of

Table 1. Primers for Real-Time PCR

Gene Sequence of primers Product size (bp)

RANKL Forward 50-GCTTCTCAGGAGCTCCAGCTAT-30 58Reverse 50-CCTCGCTGGGCCACATC-30

OPG Forward 50-CACCTTGAAGGGCCTGATGT-30 59Reverse 50-TTTTGGGAAAGTGGGATGTTTT-30

ALP Forward 50-TGGCTCTGCCTTTATTCCCTAGT-30 100Reverse 50-AAATAAGGTGCTTTGGGAATCTGT-30

OCN Forward 50-GCATAAGCTCACCGTCCACAA-30 117Reverse 50-GCCCACATCCCCCAAAAT-30

Figure 1. Effects of SM on mRNA expression of (a) ALP inMC3T3-E1 in vitro and (b) OCN in MC3T3-E1 in vitro.

1060 CHIN ET AL.

JOURNAL OF ORTHOPAEDIC RESEARCH JULY 2011

SM on the mRNA expression of ALP and OCN, the twopivotal markers related to bone formation in osteoblasts.Furthermore, SM was found to affect bone resorptionthrough regulation onOPG/RANKLratio (Fig. 2). There-fore, the fact remains that SM promotes bone remodel-ing. Bone is a highly vascularized tissue. In order for itto maintain homeostasis and regeneration, the develop-ment of microvasculature and microcirculation iscrucial.11 The process by which this occurs is known asangiogenesis. SM has a positive effect on promotingangiogenesis. This has been shown by Lay et al.12 inwhich SM crude extract and salvianolic acid B (a

component of SM) enhance angiogenesis in murineSVR endothelial cell line by up-regulating VEGF andthe receptor of VEGF (VEGF-R2) gene expression. At thesame time, regarding on its safety, recent studies whichinvestigated on the side effects of SM showed that therewas no significant cumulative toxicity or the otherside effects found.13,14 Therefore, because of the linkbetween osteogenesis and angiogenesis, and safety ofthe TCM, SM was chosen to be the candidate for bonegrafting material which may show the advantage ofVEGF but avoid the side effects. Thus, the effect of SMon osteogenesis was investigated in the present study.

Bone remodeling is a balance between bone formationand bone resorption. Bone resorption is the essentialstep for bone remodeling before new bone is formed.After the ‘‘old’’ bone is removed, newly formed bone takesthe place. Osteoblasts play an important role not onlyin bone formation but also in regulation of osteoclasts’formation and activity, thus regulate bone remodeling.15

It is believed that there is a direct contact betweenosteoblasts and osteoclasts and that osteoblastsplay an essential role in osteoclast differentiation, whichis confirmed by the discovery of RANKL–RANK inter-action. Thebinding ofRANKLonosteoblasts toRANKonosteoclasts results in the induction of osteoclastic func-tion and activation. At the same time, the secretion ofOPG, the soluble decoy receptor of RANKL, can interferewith RANKL–RANK interactions, thereby suppressingbone resorption. In other words, OPG inhibits osteoclas-togenesis while RANKL supports bone resorption.16,17

Our results indicate that the mRNA expression ofRANKL increased at an early stagewhileOPG increasedat a later stage. The ratio of mRNA expression of OPGto RANKL (OPG/RANKL) was then determined toassess its effect on osteoclastogenesis. As shown inFigure 2C, the OPG/RANKL was down-regulated firstand then up-regulated, which suggests that bone resorp-tion was up-regulated first followed by increased boneformation. This convinces the fact that bone remodelingis a continuous process involvingnot only bone formationbut also bone resorption. In order for new bone to form,bone is first resorbed by osteoclasts and the vacant cavityallow for invasion of new blood vessels together withundifferentiatedmesenchymal cells and osteoprogenitorcells which then engage in osteogenesis. This process iscritical to the formation of healthy bone tissue andallowsthebody to repairmicrofractureswithin theboneexertedby daily stress.

The final up-regulation of OPG/RANKL indicates thefinal effect of SM’s inhibitory effect of osteoclastogenesis.Our results are not completely in agreement with thosereported byKimet al.,18,19 the only twopublished studieson effect of SM on RANKL expression so far. Kimet al.18,19 have found that Tanshinone IIA (isolated fromSM) in a dose-dependent manner has some inhibitoryeffect on RANKL expression in osteoblasts as well asRANK-induced pathways that are involved in osteoclas-tic differentiation, fusion, and actin ring formation, thusleading to decrease in bone resorption. Our results

Figure 2. Effects of SM on mRNA expression of (a) OPG inMC3T3-E1 in vitro, (b) RANKL in MC3T3-E1 in vitro, and (c)OPG/RANKL in MC3T3-E1 in vitro.

SALVIA MILTIORRHIZA ON OSTEOBLASTIC CELLS 1061

JOURNAL OF ORTHOPAEDIC RESEARCH JULY 2011

revealed initial increase of RANKL expression before itdecreased. The different results may be explained by thedifferent experimental design such as time points andconcentrations. Most importantly, the present studyexamines the crude extracts of SM but Kim et al.18,19

only examine Tanshinone IIA, one of the effective com-pounds isolated from the herb. Lee et al.20 investigatedfive compounds isolated from SM: Tanshinone IIA,Tanshinone I, Crypotanshinone, 15,16-dihydrotanshi-none I, and Ferruginol, and found that TanshinoneIIA’s anti-osteoclastogenic effect may lie in its abilityto interfere with differentiation signaling rather thanreduction in osteoclast precursors or deterioration ofthe differentiation-supporting ability of osteoblasts.However, the suppressive effect of the other compoundsmay be partly ascribed to the decrease in osteoclastprecursors and number of osteoblasts due to its inhibi-tory effects on bone marrow cell viability.

Bone remodeling is a process where bone formationand bone resorption occur in a concerned manner.ALP and OCN are two pivotal markers related to boneformation in osteoblasts. Our results revealed anincrease in ALP expression at an early stage (highestincrease at 24 h) andup-regulation ofOCNexpression ata later stage (highest increase at 72 h). The process ofosteogenesis consists of cell proliferation, cell differen-tiation, and mineralization. ALP is an early marker ofosteoblastic differentiation while OCN is a late markerclosely related to osteoblastic maturation.21–23 There-fore, our results coincidewith the chronological functionsof these two bone markers, indicating that SM enhancesbone formation activities of osteoblastic cells through-out the process of osteogenesis. The formation of ECMdownregulates proliferation, and matrix mineralizationdownregulates the expression of genes associated withthe ECM maturation period.24 It explains our resultsthat ALP was up-regulated at the early stage butOCN was downregulated at the same time, while OCNincreased at the late stage but ALP decreased at thesame time. The postproliferative low-level basal expres-sion of OCN increases transcriptionally during theperiod of active ECM mineralization reflecting matu-ration of the bone cell phenotype, and synthesis ofOCN peptide parallels mRNA levels.24

Our results of increased expression of bone markersare coincident with those reported by Ding et al.25 andCui et al.26 who found that SM caused a significantincrease in ALP activity in a dose-dependent manner,which supports SM’s ability to increase osteoblasticcell activity. However, Liu et al.27 reported a diverseregulation of SM on bone marrow cell activity.Liu et al.27 found that salvianolic acid B or SalB(a hydrophilic component of SM) at a low concentrationincreased the total metabolic activity of osteoblastic cellsand expression of ALP significantly, but SalB at a higherconcentration significantly inhibited growth of bonemarrow cells. These findings may imply that differentconcentrations of SM may produce different effects onbone cell activities. At the same time, Liu et al.27 also

examined the effect of protocatechuic aldehyde or Pca,another component found in the aqueous extract of SM.They found that a high concentration (above 0.006 mg/ml) of Pca reduced the effect of SalB on osteoblasticactivity. Therefore, the higher concentration (5 mg/ml)used in Ding et al.’s study25 may not have reached thethreshold concentration that inhibits ALPactivity due toa possible opposing effect of Pca within the SM extract.We investigated the crude extract of SM in this studysince this is the common form of the herb to be used andmay be the candidate for bone grafting material or boneactivity enhancer rather than individual componentsbecause it is difficult to extract a single compound fromthe herb. The above discussion indicates that in order forSM to be used as a potential bone grafting material, thedosage and the type of extract chosenare important sinceTCMs do not only consist of pure components.

In summary, the present study revealed SM’s abilityto inhibit bone resorption and promote bone formation(Fig. 3). Even though different compounds can be iso-lated from SM and each may have different mechanismson how they affect bone remodeling, SM’s ability toinhibit bone resorption and promote bone formation can-not be denied. These results have shed some light on themechanism on how SM can increase osteogenesis andmay become a potential bone grafting material in thefuture.

CONCLUSIONSM is a safe, cheap, and strong bone anabolic agent thatmay exert its osteogenic effects by regulating the geneexpression ofALP,OCN,OPG, andRANKL.The low costand safety of this TCMmake it appealing for applicationthat requires stimulation in bone cell activities such asbone grafting.

ACKNOWLEDGMENTSWe thank Mr. Shadow Yeung and Mr. Raymond Tong fortheir technical assistance. This work is supported by SmallProject Funding (2010071761494), HKU (201007176194).

Figure 3. Flow chart of the effect of SM on bone remodeling.

1062 CHIN ET AL.

JOURNAL OF ORTHOPAEDIC RESEARCH JULY 2011

REFERENCES1. WongRWK,Rabie ABM. 2006. Effect of naringin on bone cells.

J Orthop Res 24:2045–2050.2. Wong RWK, Rabie ABM. 2011. Effect of psoralen on bone

formation. J Orthop Res 29:158–164.3. Takeshita S, Pu LQ, Stein LA, et al. 1994. Intramuscular

administration of vascular endothelial growth factor inducesdose-dependent collateral artery augmentation in a rabbitmodel of chronic limb ischemia. Circulation 90:1128.

4. Harada K, Friedman M, Lopez JJ, et al. 1996. Vascular endo-thelial growth factor administration in chronic myocardialischemia. Am J Physiol 270:1791.

5. Baffour R, Berman J, Garb JL, et al. 2004. Enhanced angio-genesis and growth of collaterals by in vivo administration ofrecombinant basic fibroblast growth factor in a rabbit model ofacute lower limb ischemia: Dose response effect of basic fibro-blast growth factor. J Vasc Surg 16:181.

6. Simons M, Bonow RO, Chronos NA, et al. 2000. Clinical trialsin coronary angiogenesis, issues, problems, consensus: Anexpert panel summary. Circulation 102:73.

7. Zhu YP, Dan S. 1998. Chinese materia medica, chemistry,pharmacology and applications, 1st ed. Netherlands: HarwoodAcademic Publishers; 459–463 p.

8. Wong RWK, Rabie ABM. 2008. Effect of Salvia miltiorrhizaextract on bone formation. J Biomed Mater Res 85A:506–512.

9. HuMZ. 1993. Effect of radix Salvia miltiorrhiza on the mitoticactivity of osteoblast-like cells isolated from chicken embryocalvariae cultured in vitro. ZhonghuaWai Ke Za Shi 3:251–253.

10. Simons M, Bonow RO, Chronos NA, et al. 2000. Clinical trialsin coronary angiogenesis issues, problems, consensus: Anexpert panel summary. Circulation 102:E73–E86.

11. Schmid J, Wallkamm B, Hammerle CH, et al. 1997. The sig-nificance of angiogenesis in guided bone regeneration. A casereport of a rabbit experiment. Clin Oral Implant Res 8:244–248.

12. Lay IS, Chiu JH, ShiaoMS, et al. 2003. Crude extract of Salviamiltiorrhiza and salvianolic acid B enhance in vitro angeio-genesis in murine SVR endothelial cell line. Planta Med 69:26–32.

13. Li G, Gao Y, Li S, et al. 2009. Study on toxicity of danshensu inbeagle dogs after 3-month continuous intravenous infusion.Toxicol Mech Methods 19:441–446.

14. Gao Y, Liu Z, Li G, et al. 2009. Acute and subchronic toxicityof danshensu in mice and rats. Toxicol Mech Methods 19:363–368.

15. Suda T, Udagawa N, Nakamura I, et al. 1995. Modulation ofosteoclast differentiation by local factors. Bone 17:87S–91S.

16. Akatsu T,Murakami T, NishikawaM, et al. 1998. Osteoclasto-genesis inhibitory factor suppresses osteoclast survival byinterfering in the interaction of stromal cells with osteoclast.Biochem Biophys ResCommun 250:229–234.

17. Yasuda H, Shima N, Nakagawa N, et al. 1999. A novel mol-ecular mechanism modulating osteoclast differentiation andfunction. Bone 25:109–113.

18. KimHK,WooER,LeeHW, et al. 2008.The correlation of Salviamiltiorrhiza extract-induced regulation of osteoclastogenesiswith the amount of components tanshinone I, tanshinone IIA,cryptotanshinone, and dihydrotanshinone. Immunopharma-col Immunotoxicol 30:347–364.

19. KimHH,KimJH,KwakHB, et al. 2004. Inhibitionof osteoclastdifferentiation and bone resorption by tanshinone IIA isolatedfrom Salvia miltiorrhiza Bunge. Biochem Pharmacol 67:1647–1656.

20. Lee SY, Choi DY,Woo ER. 2005. Inhibition of osteoclast differ-entiation by Tanshinones from the root of Salvia miltiorrhizaBunge. Arch Pharm Res 28:909–913.

21. Franceschi RT, Iyer BS. 1992. Relationship between collagensynthesis and expression of the osteoblast phenotype inMC3T3-E1 cells. J Bone Miner Res 7:235–246.

22. ASubin JE, Liu F, Malaval L, Gupta AK. 1995. Osteoblast andchondroblast differentiation. Bone 17:77S–83S.

23. Arnett TR,HendersonB. 1998.Methods in bone biology, 1st ed.London: Chapman and Hall; 10–12 p.

24. Bilezikian JP, Raisz LG, Rodan GA. 1996. Principles of bonebiology. San Diego: Academic Press.

25. Ding Y, Soma S, Takano Yamamoto T, et al. 1995. Effects ofsalvia miltiorrhiza bunge (SMB) on MC3T3-E1 cells. J OsakaUniv Dent Sch 35:21–27.

26. CuiL, ZhouLY,LiuYY, et al. 1994.Preventing cancellous boneloss in steroid-treated rats and stimulating bone formation bywater extract of Salvia Miltiorrhiza and Danshensu. ChinPharmacol Bull 20:286–291.

27. Liu YR, Qu SX, Maitz MF, et al. 2007. The effect of the majorcomponents of SalviaMiltiorrhiza bunge on bonemarrow cells.J Ethnopharmacol 111:573–583.

SALVIA MILTIORRHIZA ON OSTEOBLASTIC CELLS 1063

JOURNAL OF ORTHOPAEDIC RESEARCH JULY 2011