Available online at www.sciencedirect.com

ScienceDirect

Journal of the Chinese Medical Association 81 (2018) 699e707www.jcma-online.com

Original Article

Parathyroid hormone gene-activated matrix with DFDBA/collagencomposite matrix enhances bone regeneration in rat calvarial bone defects

Po-Hui Lee a, Tu-Lai Yew b, Yu-Lin Lai c,d, Shyh-Yuan Lee c,d, Hen-Li Chen a,c,d,*a Institute of Oral Biology, National Yang-Ming University, Taipei, Taiwan, ROC

b Department of Biotechnology and Laboratory Science in Medicine, National Yang-Ming University, Taipei, Taiwan, ROCc Department of Stomatology, Taipei Veterans General Hospital, Taipei, Taiwan, ROC

d Department of Dentistry, National Yang-Ming University, Taipei, Taiwan, ROC

Received October 30, 2017; accepted November 10, 2017

Abstract

Background: Gene-activated matrix (GAM) induces sustained local production of growth factors to promote tissue regeneration. GAM contains aplasmid DNA (pDNA) encoding target proteins that is physically entrapped within a biodegradable matrix carrier. GAM with a pDNA encodingthe first 34 amino acids of parathyroid hormone (PTH 1e34) and a collagen matrix enhances bone regeneration in long bone defects. Dem-ineralized freeze-dried bone allograft (DFDBA) is a widely used osteoinductive bone graft. The present study determined the osteogenic effectsof PTH-GAM with a collagen or DFDBA/collagen composite (D/C) matrix for treating craniofacial bone defects.Methods: We constructed a pDNA encoding human PTH 1e34 and performed cyclic AMP ELISA to verify the bioactivity of PTH 1e34. Next,we generated a D/C matrix and PTH-GAMs containing a collagen matrix (PTH-C-GAM) or D/C matrix (PTH-D/C-GAM). Rats with critical-sized calvarial bone defects were divided into four groups, namely, untreated rats (sham group) and rats grafted with D/C matrix, PTH-C-GAM,or PTH-D/C-GAM (D/C, PTH-C-GAM, or PTH-D/C-GAM groups, respectively). PTH expression was determined by performing immuno-histochemical staining after 4 and 8 weeks. New bone formation was evaluated by performing radiography, dual-energy X-ray absorptiometry,microcomputed tomography, and histological examination.Results: PTH pDNA-transfected cells secreted bioactive PTH 1e34. Moreover, PTH was expressed at 4 and 8 weeks after the surgery in rats inthe PTH-C-GAM group but not in rats in the D/C group. New bone formation in the calvarial bone defects, from more to less, was in the order ofPTH-D/C-GAM, PTH-C-GAM, D/C, and sham groups.Conclusion: Our results indicate that PTH-GAM with a collagen matrix promotes local PTH production for at least 8 weeks and boneregeneration in craniofacial bone defect. Moreover, our results indicate that replacement of the collagen matrix with the D/C matrix improves theosteogenic effects of PTH-GAM.Copyright © 2018, the Chinese Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Allografts; Bone morphogenetic proteins; Bone regeneration; Gene therapy; Parathyroid hormone

Conflicts of interest: The authors declare that they have no conflicts of interest

related to the subject matter or materials discussed in this article.

* Corresponding author. Dr. Hen-Li Chen, Department of Dentistry, National

Yang-Ming University, 155, Section 2, Linong Street, Taipei 112, Taiwan,

ROC.

E-mail address: hlchen2@ym.edu.tw (H.-L. Chen).

https://doi.org/10.1016/j.jcma.2017.12.004

1726-4901/Copyright © 2018, the Chinese Medical Association. Published by El

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Gene-activated matrix (GAM) technology, a non-viral genetherapy strategy, promotes sustained local production ofgrowth factors to enhance tissue regeneration.1 GAM containsa plasmid DNA (pDNA) encoding target proteins that isphysically entrapped within a biodegradable matrix carrier.During the healing process, cells migrate into the matrix, takeup the pDNA, and express the encoded proteins. Although the

sevier Taiwan LLC. This is an open access article under the CC BY-NC-ND

Fig. 1. Study design and timeline. SEM ¼ scanning electron microscopy;

DEXA ¼ dual-energy X-ray absorptiometry; IHC staining ¼ immunohisto-

chemical staining.

700 P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

GAM technology is an economic, safe, and simple strategy,the transfection efficiency of GAM is low. Polycations such aspolyethylenimine (PEI) have been used to condense pDNA,facilitate its entry into cells, increase gene expression, andenhance the osteogenic effect of GAMs.2,3 GAMs with pDNAencoding various osteogenic factors and with differentmatrices have been studied for promoting boneregeneration.1,2,4e8

Parathyroid hormone (PTH), a polypeptide containing 84amino acids (PTH 1e84), regulates blood calcium levels byexerting catabolic effect on bones. Intriguingly, PTH and somePTH fragments exert anabolic effects on bones under certainconditions.9 PTH 1e34, which contains 34 amino acids of theN-terminus of PTH, exerts anabolic effects on bones.10,11 PTH1e34 binds to PTH receptors on osteoblasts and activatesseveral signaling pathways. The anabolic effects of PTH onosteogenic cells are mainly mediated by the activation of cy-clic AMP/protein kinase A (cAMP/PKA) signaling pathway.12

PTH 1e34 greatly enhances bone formation. Systemic PTH1e34 administration through a low-dose intermittent subcu-taneous injection increases bone density in osteoporosis,13,14

enhances bone healing,15 and promotes allograft integra-tion16 in rats with calvarial bone defects. Local delivery ofPTH 1e34 along with an RGD peptide by using a hydrogelmatrix exerts beneficial effects on bone defect healing aroundimplants.17 Local application of PTH 1e34 by using a mem-brane effectively enhances the healing of 5 mm calvarial bonedefect in rats.18 The GAM technology has been used to deliverPTH 1e34 for promoting bone regeneration in long bonedefects.1,19,20 However, the effect of PTH-GAM for treatingcraniofacial bone defects is still unclear.

Demineralized freeze-dried bone allograft (DFDBA) is oneof the most widely used bone grafts in dentistry. DFDBAcontains type I collagen, bone morphogenetic proteins (BMP-2, BMP-4, and BMP-7), and other growth factors (includingFGF-1 (fibroblast growth factor-1), IGF-1 (insulin-like growthfactor-1), TGF-b1 (transforming growth factor-beta1), VEGF(vascular endothelial growth factor) and PDGF (platelet-derived growth factor)).21,22 The osteoinductive property ofDFDBA is attributed to the exposure of its inherent BMPsthrough demineralization.23 A DFDBA/collagen composite(D/C) matrix has been used for improving the handlingproperties of powdered DFDBA. Furthermore, this matrix ismore effective than DFDBA24 or collagen25 for promotingbone regeneration.

Several studies indicate that PTH 1e34 and BMPs exertsynergistic effects for promoting bone regeneration. Combineduse of intermittent systemic PTH 1e34 and BMP-2 induceshigher bone formation than the use of PTH or BMP alone in arat model of spinal fusion.26 Similarly, combined use ofintermittent systemic PTH 1e34 and BMP-7 exerted a syn-ergistic effect in a rabbit model of metaphyseal bone heal-ing.27 Furthermore, intermittent systemic PTH 1e34 treatmentimproves BMP-2-induced ectopic and orthotopic bone for-mation.28 Dual delivery of PTH 1e34 and BMP-4 by using atwo-plasmid GAM promotes better bone healing than indi-vidual delivery of PTH 1e34 or BMP-4 by using GAM.19

Local sustained delivery of osteogenic factors throughregional gene therapy may enhance bone defect healing withreduced systemic toxicity and prevent the requirement ofsupraphysiological doses of therapeutic proteins.1,29 PTH-GAMencoding PTH 1e34 and with a collagen matrix promotes thehealing of long bone defects.1,19 Long bones are formedthrough endochondral ossification, whereas craniofacial bonesare formed through intramembranous ossification. Therefore,therapeutic effects of PTH-GAM for treating craniofacial bonedefects should be elucidated. We hypothesized that PTH-GAMwith a collagen matrix promoted bone regeneration in cranio-facial bone defects and that the use of PTH-GAM with a D/Cmatrix exerted better anabolic effects than PTH-GAM with acollagen matrix. Thus, the present study determined the effectsof PTH-GAM with a collagen or D/C matrix in a rat model ofcritical-sized calvarial bone defects.

2. Methods

2.1. Experimental design and timelines

The study design is shown in Fig. 1. A gene vectorencoding human PTH 1e34 was prepared, and cAMP ELISAwas performed to determine the bioactivity of PTH 1e34secreted by cells transfected with this vector. Materials forimplantation into the rat model of calvarial bone defects,including D/C matrix, PTH-GAM with a collagen matrix(PTH-C-GAM), and PTH-GAM with a D/C matrix (PTH-D/C-GAM), were prepared. The structure of PTH-C-GAM wasexamined by performing scanning electron microscopy(SEM). Rats with critical-sized calvarial bone defects weredivided into four groups, namely, untreated rats (Sham group)and rats grafted with D/C, PTH-C-GAM, and PTH-D/C-GAM(D/C, PTH-C-GAM, and PTH-D/C-GAM groups, respec-tively). Endpoint analyses were performed after 4 and 8weeks. PTH expression was determined by performingimmunohistochemical staining, and bone regeneration wasevaluated by performing radiography, dual-energy X-ray ab-sorptiometry (DEXA), microcomputed tomography (mCT),and histological examination (H&E staining).

701P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

2.2. Preparation of pDNA carrying the human PTH gene(pUMVC1/hPTH)

The human PTH1e34 gene (prepro-PTH 1e34)19 wasinserted into a plasmid (pUMVC1) with a CMV promotor togenerate pUMVC1/hPTH pDNA. pUMVC1/LacZ was used asa negative control for cAMP ELISA. Endotoxin-free plasmidswere prepared using EndoFree plasmid Giga-prep kit (Qiagen,CA, USA).

2.3. Cyclic AMP ELISA

cAMP ELISA was performed to verify the bioactivity ofcell-secreted PTH 1e34 because cAMP is a downstreameffector of PTH.30 HEK293 cells were transfected withpUMVC1/hPTH or pUMVC1/LacZ pDNA by using a trans-fection reagent (FuGENE 6 transfection reagent; Roche, IN,USA). UMR-106 cells express PTH receptors and show theactivation of the cAMP/PKA pathway after PTH stimulation.31

In the present study, UMR-106 cells, which were pretreatedwith isobutylmethylxanthine to block phosphodiesterase ac-tivity, were treated with a conditioned medium obtained byculturing transfected HEK293 cells for 10 min at 37 �C. cAMPin the lysate of UMR-106 cells was measured using cAMPELISA kit (cAMP ELISA Biotrak kit; Amersham Biosciences,NJ, USA).

2.4. Preparation of DFDBA

The hind limb bones of 8-week-old male SpragueeDawley(SD) rats were collected and ground in liquid nitrogen. Boneparticles with a size of 250e500 mm were collected usingsieves and were used in this study. Bone particles was dem-ineralized using 0.5 N sterile HCl, washed by rocking inanhydrous alcohol at 4 �C, and rinsed with sterile distilledwater. Completeness of demineralization was confirmed byperforming radiography (50 kVp and 0.02 s). Next, the boneparticles were freeze-dried and were stored at �80 �C forfurther use.

2.5. Preparation of D/C matrix, PTH-C-GAM, and PTH-D/C-GAM

pUMVC1/hPTH pDNA and branched 25 kDa PEI (Sig-maeAldrich, MO, USA)32 were mixed at a nitrogen/phos-phate molar ratio of 9 and DNA/PEI molecular weight ratio of1:1.25. The mixture was kept at room temperature for 30 minand at �80 �C for 12 h and was lyophilized under vacuum at100 mT and �45 �C for 24 h. Rat type I atelocollagen wasprepared as described previously.33 Next, PEI-pUMVC1/hPTH pDNA (200 mg pDNA) mixture, atelocollagen (2%),and/or DFDBA (collagen/DFDBA ratio, 20:80) were mixed;placed in round Teflon molds (8 mm diameter and 4 mmheight); frozen at �80 �C; and freeze-dried to form sponge-like PTH-D/C-GAM and PTH-C-GAM. D/C matrix was pre-pared in the same manner but without adding the PEI-pUMVC1/hPTH pDNA mixture. Next, the sponge-like

matrices were crosslinked using 25% glutaraldehyde vaporfor 4 h at 37 �C. The remaining aldehyde groups wereremoved using 0.1 M glycine, and the matrices were rinsedthree times with sterile PBS. Finally, PTH-D/C-GAM, PTH-C-GAM, and D/C matrix were sterilized with 70% ethanol andwere freeze-dried before use.

2.6. Scanning electron microscopy

PTH-C-GAM was gold coated in an ion sputter (E101 ionsputter®; Hitachi, Japan) and was observed under a scanningelectron microscope (S-2700; Hitachi).

2.7. Rat model of calvarial bone defect

Protocols for performing animal experiments wereapproved by the Animal Research Committee of the NationalYang-Ming University. The study included 42 male SD rats(age, 4 weeks) that were randomly divided into four groups,namely, Sham, D/C, PTH-C-GAM, and PTH-D/C-GAMgroups, and were examined at two time points, i.e., 4 and 8weeks. At each time point, the Sham group and the otherthree groups used 3 and 6 rats per group, respectively. Therats were anesthetized by intraperitoneally injecting amixture of xylazine and ketamine hydrochloride (mixed in1e4 ratio, 0.17 ml mixture per 100 g body weight). Critical-sized (8 mm diameter) calvarial bone defects were createdusing a trephine bur under saline irrigation.34 The defectswere not treated (Sham) or grafted with materials (D/Ccomposite matrix, PTH-C-GAM, or PTH-D/C-GAM). Ratcalvarias were collected after 4 and 8 weeks for endpointanalyses.

2.8. Radiograph evaluation

Radiographs of the rat calvarias were obtained (50 kVp,8 mA, and 10 s) along with those of an acrylic step wedgecontaining ten 2.7 mm-thick steps. The step wedge was usedas a reference to correct density variation caused by filmexposure and processing. Images on the developed filmswere scanned and digitalized, and average bone densitywithin the calvarial defects was quantified using an imageanalysis software (ImageQuant, Version 5.2; Molecular Dy-namics, CA, USA). Results are presented as acrylic thickness(mm).

2.9. Dual energy X-ray absorptiometry

Bone mineral content (BMC) and new bone area in the ratmodel of calvarial bone defects were measured using a DEXAsystem designed for small animal research (pDEXA Sabre;Norland Medical Systems, WI, USA). Scanning parametersused were width, 1 cm; length, 1 cm; speed, 5 mm/s; andresolution 0.5 � 0.5. DEXA measurements were analyzedusing a software (version 3.9.4) provided in the DEXA system.Original defect area was set as the region of interest for all themeasurements.

702 P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

2.10. Microcomputed tomography

Rat calvarias were scanned using a compact mCT imagingsystem (SkyScan 1174; SkyScan,Belgium). Scanswere acquiredat 8.6 mm image pixel size, 48 kVp, and 120 mA. Next, three-dimensional (3D) reconstruction of the images and bone vol-ume analysis of the defect zone (8 mm) were performed usingSkyScan's volumetric reconstruction software (NRecon) by usinga threshold of 45. This program uses a modified Feldkampalgorithm with an automatic adaptation to scan geometry.

2.11. Immunohistochemical staining and histologicalexamination

Fig. 2. Validation of pUMVC1/hPTH bioactivity. Conditioned media were

collected from untreated HEK293 cells (sham) or HEK293 cells transfected

with pUMVC1/LacZ or pUMVC1/hPTH. The conditioned media were used to

treat UMR-106 cells. The cAMP in the cell lysate of UMR-106 cells was

measured by cAMP ELISA assay. *** p < 0.001.

Rat calvarias were fixed in 4% formaldehyde, decalcified,embedded in paraffin, and sectioned perpendicular to thesagittal suture at the center of the created defect. PTHexpression was determined by performing immunohisto-chemical staining with a monoclonal mouse anti-human PTHantibody (T-9912; dilution, 1/200; BMA Biomedicals,Switzerland) and horseradish peroxidase-conjugated goat anti-mouse IgG (NEF822; dilution, 1/500; PerkinElmer, MA,USA). PTH was visualized using an 3-amino-9-ethylcarbazole(AEC) chromogen. Nuclear counterstaining was performedusing hematoxylin. For histological examination, the tissuessections were stained with hematoxylin and eosin and wereobserved under a light microscope.

2.12. Statistical analysis

Results are presented as mean ± standard deviation. Sta-tistical comparisons were performed using one-way ANOVAwith GraphPad Instat 6.0c (GraphPad Software, CA, USA).Differences between the groups were considered significant atp < 0.05.

3. Results

3.1. Establishment of pUMVC1/hPTH and validation ofhuman PTH 1e34 bioactivity

To verify the bioactivity of PTH 1e34 secreted by cellstransfected with the pUMVC1/hPTH pDNA, conditionedmedium of HEK293 cells transfected with pUMVC1/hPTH orpUMVC1/LacZ was collected and was used for treating UMR-106 cells. Lysates of UMR-106 cells in the pUMVC1/hPTHgroup showed significantly higher cAMP levels than those ofUMR-106 cells in the Shamand pUMVC1/LacZnegative controlgroups (Fig. 2). These results indicate that cells secret bioactivePTH 1e34 after taking up the pUMVC1/hPTH plasmid.

3.2. Morphology of PTH-C-GAM

Morphology of PTH-C-GAM was examined by performingSEM (Fig. 3). SEM images showed that PTH-C-GAM hadhigh porosity, with different-sized pores (ranging from tens tohundreds of microns), and good interpore connectivity.

3.3. Sustained expression of human PTH 1e34 in the ratmodel of calvarial bone defects

In the present study, we used GAMs to promote sustainedlocal administration of PTH. We determined the localexpression of PTH at 4 and 8 weeks after PTH-C-GAM im-plantation by performing immunohistochemical staining(Fig. 4). We observed that rats in the PTH-C-GAM groupshowed stronger PTH expression at both the time points thanrats in the D/C group. These results indicate that local cellstake up the pUMVC1/hPTH pDNA and effectively secret PTHfor at least 8 weeks.

3.4. PTH-GAMs increase bone mineral density, bonemineral content, new bone area, and bone volume

To clarify whether local delivery of PTH by using PTH-GAMs promoted the healing of calvarial bone defects, weperformed endpoint analysis with radiography, DEXA, andmCT. Radiography data indicated that rats in the Sham groupshowed minimal new bone formation at 4 and 8 weeks. Rats inall the other groups showed higher bone growth than those inthe Sham group (Fig. 5A). Relative bone density was thehighest in rats in the PTH-D/C-GAM group, followed by thatin rats in the PTH-C-GAM, D/C, and Sham groups in thegiven order (Fig. 5B).

DEXA was performed to determine BMC and new bonearea. Rats in the PTH-D/C-GAM group showed higher BMCthan rats in the D/C and PTH-C-GAM groups at both 4 and 8weeks (Fig. 5C). Similar trend was observed for new bone area(Fig. 5D). At 8 weeks after the implantation, rats in the PTH-D/C-GAM group showed 6.15-, 3.34-, and 1.27-fold larger

Fig. 3. Structure of PTH-GAM with collagen matrix. The structure of PTH-C-GAM was examined under a scanning electron microscope at 35X to 500X

magnification.

Fig. 4. In vivo expression of human PTH 1e34. Rat critical-sized calvarial bone defects grafted with D/C (a, c) or PTH-C-GAM (b, d) were examined. The tissues

within the defect sites were immunostained with human-specific PTH (brown color) at 4- (a, b) and 8-week (c, d). Nuclear counterstaining was performed with

hematoxylin (blue color).

703P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

Fig. 5. Bone formation analysis by radiography, dual energy X-ray absorptiometry, and mCT. Rat critical-sized calvarial bone defects were untreated (Sham)

or grafted with D/C, PTH-C-GAM, or PTH-D/C-GAM. The new bone formation at the defect sites was examined by radiography, DEXA, and mCT after 4 and 8

weeks. (A) Representative radiographic images. (B) Average bone density within the defects quantified by radiography with acrylic step wedge as a reference. (C,

D) Bone mineral content (BMC) and new bone area at defect sites determined by DEXA. (E) Representative mCT 3D reconstruction images. (F) Bone volume

within the defects quantified by mCT. * p < 0.05, ** p < 0.01, *** p < 0.001.

704 P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

new bone area than rats in the Sham, D/C, and PTH-C-GAMgroups, respectively (Fig. 5D).

Next, we performed mCT analysis, including 3D recon-struction of images and measurement of bone volume. Resultsobtained from the 3D reconstruction of mCT images indicatedthat rats in the Sham group showed the lowest new boneformation compared with rats in the other groups at 4 and 8weeks. Rats in the PTH-D/C-GAM group showed the highestbone regeneration effect (Fig. 5E). In addition, rats in thePTH-D/C-GAM group showed 7.02-, 4.26-, and 1.47-foldhigher bone volume than rats in the Sham, D/C, and PTH-C-GAM groups, respectively, at 8 week after the implantation(Fig. 5F).

Together, these results indicate that PTH-D/C-GAM fol-lowed by PTH-C-GAM and D/C matrix have the best boneregeneration potential.

3.5. Histological examination showed enhanced newbone formation

H&E staining of tissue sections was performed to examinenew bone formation in sites showing calvarial bone defects.Rats in the Sham group showed thin regenerative tissues, withminimal bone formation (Fig. 6a and e). Apparent new boneformation characterized by thick regenerative tissues andincreased new bone volume was observed in rats in the D/C,PTH-C-GAM, and PTH-D/C-GAM groups at 4 weeks afterthe implantation (Fig. 6bed). In addition, rats in the PTH-D/C-GAM group (Fig. 6d) showed more compact and maturemorphology of the new bone tissue than those in the D/C andPTH-C-GAM groups (Fig. 6b and c). At 8 weeks after theimplantation, rats in the D/C and PTH-C-GAM groups showedloose morphology of the new bone tissue (Fig. 6f and g),

Fig. 6.Histological examination of bone formation. Rat critical-sized calvarial bone defects were untreated (Sham) (a, e) or grafted with D/C (b, f), PTH-C-GAM

(c, g), or PTH-D/C-GAM (d, h). Bone formation at the defect areas was examined at 4- and 8-week by hematoxylin-eosin staining of tissue sections. The triangles

indicate the edges of the calvarial bone defects.

705P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

whereas those in the PTH-D/C-GAM group showed the mostmature morphology of the new bone tissue within critical-sized calvarial bone defects (Fig. 6h).

4. Discussion

Local PTH 1e34 administration by using GAM enhancesthe healing of long bones. Results of the present study indicatethat application of PTH-GAM promotes bone regeneration incraniofacial bone defects. In addition, change in the carriermatrix of PTH-GAM from collagen to D/C matrix furtherincreases its osteogenic effect.

The osteoinductive capacity of commercial DFDBA isusually not verified by tissue banks. Many factors maycontribute to the differences in the osteoinductive effect ofDFDBA, such as different processing protocols and donorage.35,36 Variation in the osteoinductive properties of pur-chased DFDBA may significantly affect its regenerative po-tential.37 Therefore, the DFDBA used in this study wasprepared and confirmed for the presence of osteoinductiveeffect via a subcutaneous ectopic bone formation method.

GAM promotes sustained local production of pDNA-encoded growth factors. Results of immunohistochemicalstaining performed at 8 weeks after PTH-C-GAM implanta-tion showed that local cells took up the pDNA present in PTH-C-GAM and secreted PTH for at least 8 weeks (Fig. 4). Theseresults are consistent with those from previous studies, whichshowed that the combination of GAM and PEI promotes theexpression of encoded proteins up to 15 weeks.2,7

Bone regenerative effect was the highest in rats in the PTH-D/C-GAM, followed by that in rats in the PTH-C-GAM, D/C,

and Sham groups in the given order. First, rats in the D/Cgroup showed better new bone formation than rats in the shamgroup, which was consistent with that observed in previousstudies that showed the beneficial effect of the D/C matrix.24,25

Second, PTH-D/C-GAM and PTH-C-GAM showed higherosteogenic potential than the D/C matrix, thus supporting thehigh bone regeneration potential of PTH-GAMs. Third, thedifference between PTH-D/C-GAM and PTH-C-GAM may bedue to the synergistic effect of PTH and BMP reportedpreviously.19,26e28

Effects of PTH on bones depend on the mode of delivery.Intermittent PTH administration through subcutaneous in-jection induces anabolic effects.14e16 Continuous PTHadministration increases osteoclastic bone resorption.38,39

Local PTH production by using GAMs does not seem to beintermittent. Further studies are needed to elucidate mecha-nisms underlying the anabolic effects induced by PTH-GAMson bones.

Use of recombinant proteins such as PDGF-BB40 andBMP-241 in regenerative therapies is simple and feasible.However, these proteins are associated with limitations suchas high cost, short half-life, and difficulties in sustainedlocal delivery.42 The use of GAMs can overcome manylimitations associated with the direct administration ofproteins.43 The results of the present study indicate thatcombined use of the D/C matrix and PEI-condensed pDNAencoding PTH 1e34 has a high potential for treatingdifficult-to-heal large craniofacial bone defects. Themethods and results presented here will be of value for usingPTH-GAMs for promoting craniofacial bone regeneration inthe future.

706 P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

Acknowledgments

This study was funded primarily by the Taipei VeteransGeneral Hospital, Taiwan [grant number VGH94365-6] andthe National Health Research Institutes, Taiwan [grant numberNHRI-EX94-9305SC]. We thank the Taiwan Mouse Clinic[grant number MOST 105-2325-B-001-010] which is fundedby the National Research Program for Biopharmaceuticals(NRPB) at the Ministry of Science and Technology (MOST)of Taiwan for mCT technical support in the rat calvarial defectexperiment.

References

1. Bonadio J, Smiley E, Patil P, Goldstein S. Localized, direct plasmid gene

delivery in vivo: prolonged therapy results in reproducible tissue regen-

eration. Nat Med 1999;5:753e9.

2. Huang YC, Simmons C, Kaigler D, Rice KG, Mooney DJ. Bone regen-

eration in a rat cranial defect with delivery of PEI-condensed plasmid

DNA encoding for bone morphogenetic protein-4 (BMP-4). Gene Ther

2005;12:418e26.

3. Winn SR, Chen JC, Gong X, Bartholomew SV, Shreenivas S, Ozaki W.

Non-viral-mediated gene therapy approaches for bone repair. Orthod

Craniofac Res 2005;8:183e90.

4. Bright C, Park YS, Sieber AN, Kostuik JP, Leong KW. In vivo evaluation

of plasmid DNA encoding OP-1 protein for spine fusion. Spine (Phila Pa

1976) 2006;31:2163e72.

5. Shea LD, Smiley E, Bonadio J, Mooney DJ. DNA delivery from polymer

matrices for tissue engineering. Nat Biotechnol 1999;17:551e4.

6. Jang JH, Rives CB, Shea LD. Plasmid delivery in vivo from porous tissue-

engineering scaffolds: transgene expression and cellular transfection. Mol

Ther 2005;12:475e83.

7. Elangovan S, D'Mello SR, Hong L, Ross RD, Allamargot C, Dawson DV,

et al. The enhancement of bone regeneration by gene activated matrix

encoding for platelet derived growth factor. Biomaterials 2014;35:737e47.

8. Lindsay R, Nieves J, Formica C, Henneman E, Woelfert L, Shen V, et al.

Randomised controlled study of effect of parathyroid hormone on

vertebral-bone mass and fracture incidence among postmenopausal

women on oestrogen with osteoporosis. Lancet 1997;350:550e5.

9. Mohan S, Kutilek S, Zhang C, Shen HG, Kodama Y, Srivastava AK, et al.

Comparison of bone formation responses to parathyroid hormone(1-34),

(1-31), and (2-34) in mice. Bone 2000;27:471e8.

10. Mosekilde L, Sogaard CH, Danielsen CC, Torring O. The anabolic effects

of human parathyroid hormone (hPTH) on rat vertebral body mass are

also reflected in the quality of bone, assessed by biomechanical testing: a

comparison study between hPTH-(1-34) and hPTH-(1-84). Endocrinology

1991;129:421e8.

11. Oxlund H, Ejersted C, Andreassen TT, Torring O, Nilsson MH. Para-

thyroid hormone (1-34) and (1-84) stimulate cortical bone formation both

from periosteum and endosteum. Calcif Tissue Int 1993;53:394e9.

12. Yang D, Singh R, Divieti P, Guo J, Bouxsein ML, Bringhurst FR. Con-

tributions of parathyroid hormone (PTH)/PTH-related peptide receptor

signaling pathways to the anabolic effect of PTH on bone. Bone 2007;40:

1453e61.

13. Dobnig H. A review of teriparatide and its clinical efficacy in the treat-

ment of osteoporosis. Expet Opin Pharmacother 2004;5:1153e62.14. Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster JY,

et al. Effect of parathyroid hormone (1-34) on fractures and bone mineral

density in postmenopausal women with osteoporosis. N Engl J Med 2001;

344:1434e41.

15. Silva ED, Vasconcelos DF, Marques MR, Silva MA, Manzi FR, Barros SP.

Intermittent administration of parathyroid hormone improves the repairing

process of rat calvaria defects: a histomorphometric and radio-

densitometric study. Med Oral Patol Oral Cir Bucal 2015;20:e489e93.

16. Sheyn D, Cohn Yakubovich D, Kallai I, Su S, Da X, Pelled G, et al. PTH

promotes allograft integration in a calvarial bone defect.Mol Pharm 2013;

10:4462e71.

17. Valderrama P, Jung RE, Thoma DS, Jones AA, Cochran DL. Evaluation of

parathyroid hormone bound to a synthetic matrix for guided bone

regeneration around dental implants: a histomorphometric study in dogs.

J Periodontol 2010;81:737e47.

18. Missana LR, Jammal MV. Critical size defect regeneration by rhPTH-

collagen membrane as a new tissue engineering tool. J Biomed Mater

Res A 2014;102:4358e64.

19. Fang J, Zhu YY, Smiley E, Bonadio J, Rouleau JP, Goldstein SA, et al.

Stimulation of new bone formation by direct transfer of osteogenic

plasmid genes. Proc Natl Acad Sci U S A 1996;93:5753e8.

20. Chen H, Frankenburg EP, Goldstein SA, McCauley LK. Combination of

local and systemic parathyroid hormone enhances bone regeneration. Clin

Orthop Relat Res 2003:291e302.21. Shigeyama Y, D'Errico JA, Stone R, Somerman MJ. Commercially-pre-

pared allograft material has biological activity in vitro. J Periodontol

1995;66:478e87.22. Wildemann B, Kadow-Romacker A, Pruss A, Haas NP, Schmidmaier G.

Quantification of growth factors in allogenic bone grafts extracted with

three different methods. Cell Tissue Bank 2007;8:107e14.

23. Lohmann CH, Andreacchio D, Koster G, Carnes Jr DL, Cochran DL,

Dean DD, et al. Tissue response and osteoinduction of human bone grafts

in vivo. Arch Orthop Trauma Surg 2001;121:583e90.

24. Tsai CH, Chou MY, Jonas M, Tien YT, Chi EY. A composite graft ma-

terial containing bone particles and collagen in osteoinduction in mouse.

J Biomed Mater Res 2002;63:65e70.

25. Kohles SS, Vernino AR, Clagett JA, Yang JC, Severson S, Holt RA.

A morphometric evaluation of allograft matrix combinations in the

treatment of osseous defects in a baboon model. Calcif Tissue Int 2000;67:

156e62.

26. Kaito T, Morimoto T, Kanayama S, Otsuru S, Kashii M, Makino T, et al.

Modeling and remodeling effects of intermittent administration of

teriparatide (parathyroid hormone 1-34) on bone morphogenetic

protein-induced bone in a rat spinal fusion model. Bone Rep 2016;5:

173e80.

27. Morgan EF, Mason ZD, Bishop G, Davis AD, Wigner NA,

Gerstenfeld LC, et al. Combined effects of recombinant human BMP-7

(rhBMP-7) and parathyroid hormone (1-34) in metaphyseal bone heal-

ing. Bone 2008;43:1031e8.

28. Kempen DH, Lu L, Hefferan TE, Creemers LB, Heijink A, Maran A, et al.

Enhanced bone morphogenetic protein-2-induced ectopic and orthotopic

bone formation by intermittent parathyroid hormone (1-34) administra-

tion. Tissue Eng Part A 2010;16:3769e77.29. D'Mello S, Atluri K, Geary SM, Hong L, Elangovan S, Salem AK. Bone

regeneration using gene-activated matrices. AAPS J 2017;19:43e53.

30. Kano J, Sugimoto T, Fukase M, Chihara K. Direct involvement of cAMP-

dependent protein kinase in the regulation of alkaline phosphatase activity

by parathyroid hormone (PTH) and PTH-related peptide in osteoblastic

UMR-106 cells. Biochem Biophys Res Commun 1994;199:271e6.

31. Erclik MS, Mitchell J. The role of protein kinase C-delta in PTH stimu-

lation of IGF-binding protein-5 mRNA in UMR-106-01 cells. Am J

Physiol Endocrinol Metab 2002;282:E534e41.

32. Boussif O, Lezoualc'h F, Zanta MA, Mergny MD, Scherman D,

Demeneix B, et al. A versatile vector for gene and oligonucleotide transfer

into cells in culture and in vivo: polyethylenimine. Proc Natl Acad Sci U S

A 1995;92:7297e301.

33. Chen DC, Lai YL, Lee SY, Hung SL, Chen HL. Osteoblastic response to

collagen scaffolds varied in freezing temperature and glutaraldehyde

crosslinking. J Biomed Mater Res A 2007;80:399e409.

34. Takagi K, Urist MR. The reaction of the dura to bone morphogenetic

protein (BMP) in repair of skull defects. Ann Surg 1982;196:100e9.

35. Schwartz Z, Somers A, Mellonig JT, Carnes Jr DL, Dean DD,

Cochran DL, et al. Ability of commercial demineralized freeze-dried bone

allograft to induce new bone formation is dependent on donor age but not

gender. J Periodontol 1998;69:470e8.

707P.-H. Lee et al. / Journal of the Chinese Medical Association 81 (2018) 699e707

36. Schwartz Z, Mellonig JT, Carnes Jr DL, de la Fontaine J, Cochran DL,

Dean DD, et al. Ability of commercial demineralized freeze-dried bone

allograft to induce new bone formation. J Periodontol 1996;67:918e26.

37. Polson AM, Heijl LC. Osseous repair in infrabony periodontal defects.

J Clin Periodontol 1978;5:13e23.38. Shen V, Birchman R, Wu DD, Lindsay R. Skeletal effects of parathyroid

hormone infusion in ovariectomized rats with or without estrogen reple-

tion. J Bone Miner Res 2000;15:740e6.39. Uzawa T, Hori M, Ejiri S, Ozawa H. Comparison of the effects of inter-

mittent and continuous administration of human parathyroid hormone(1-

34) on rat bone. Bone 1995;16:477e84.

40. Nevins M, Giannobile WV, McGuire MK, Kao RT, Mellonig JT,

Hinrichs JE, et al. Platelet-derived growth factor stimulates bone fill and

rate of attachment level gain: results of a large multicenter randomized

controlled trial. J Periodontol 2005;76:2205e15.

41. Khan SN, Lane JM. The use of recombinant human bone morphogenetic

protein-2 (rhBMP-2) in orthopaedic applications. Expet Opin Biol Ther

2004;4:741e8.

42. Evans CH. Gene delivery to bone. Adv Drug Deliv Rev 2012;64:1331e40.43. Lu CH, Chang YH, Lin SY, Li KC, Hu YC. Recent progresses in gene

delivery-based bone tissue engineering. Biotechnol Adv 2013;31:

1695e706.