Basic Research—Biology

Effects of Platelet-rich Plasma and Cell Cocultureon Angiogenesis in Human Dental Pulp Stem Cellsand Endothelial Progenitor CellsXinzhu Li, MS,*

†Jin Hou, PhD,*

†Buling Wu, PhD,*

†Ting Chen, PhD,*

†and Aoxiang Luo, MS*

†

Abstract

Introduction: Platelet-rich plasma (PRP) has beendescribed as platelet concentrate. Growth factorsreleased by activated platelets can improve wound vas-culogenesis and enhance wound healing. In this study,we used PRP instead of serum to culture human dentalpulp stem cells (hDPSCs) and endothelial progenitorcells (EPCs) and investigated revascularization ability.The effect of hDPSC and EPC coculture on vasculogene-sis was also studied. Methods: PRP was prepared bysecondary centrifugation. Real-time polymerase chainreaction and Western blotting were used to determinethe expression of vasculogenesis-related factorsvascular endothelial growth factor, platelet-derivedgrowth factor, fetal liver kinase 1 (Flk-1), and stromalcell-derived factor 1 (SDF-1) in cultured hDPSCs andEPCs. The cells were divided into 4 groups: EPCs +10% fetal bovine serum (FBS), EPCs + 10% PRP,EPCs + hDPSCs + 10% FBS, and EPCs + hDPSCs +10% PRP. Then, the formation of vessel-like structureswas tested by the tube formation assay. Results: Onday 3, the expression levels of all the markers in thecoculture groups were much higher than in the single-culture groups and were also higher in the PRP groupscompared with the FBS groups (P < .05), except forSDF-1. Expression levels were significantly higher inthe experimental groups (EPCs + 10% PRP,EPCs + hDPSCs + 10% FBS, and EPCs + hDPSCs +10% PRP) than in the control group (EPCs + 10% FBS)and in the PRP groups/coculture groups comparedwith the FBS groups/single-culture groups (P < .01).The tube formation assay showed the area of vessel-like structures formed by the PRP group to be largerthan in the FBS group (P < .05). Conclusions: PRPand coculture can both promote vasculogenesis, andPRP can promote EPCs to form vessel-like structures.(J Endod 2014;-:1–5)

Key WordsCoculture, endothelial progenitor cells, human dentalpulp stem cells, platelet-rich plasma, vasculogenesis

From the *Department of Stomatology, Nanfang Hospital, GuanAddress requests for reprints to Dr Buling Wu, Department of St

0099-2399/$ - see front matterCopyright ª 2014 American Association of Endodontists.

http://dx.doi.org/10.1016/j.joen.2014.07.022

JOE — Volume -, Number -, - 2014

Platelet-rich plasma (PRP) separated from whole blood has been described asplatelet concentrate. Activated PRP includes many growth factors such as platelet-

derived growth factor (PDGF), transforming growth factor beta, insulin-like growth fac-tor, and vascular endothelial growth factor (VEGF) (1–3). When platelets are activated,they release these growth factors almost immediately and continue to synthesizeadditional growth factors for several days (4).

This suggests that PRPmight promote tissue repair. PRP has been reported to alterthe biomechanical and histologic properties of rotator cuff repair during acute injuryresponse (5). Many investigations have shown that growth factors released by activatedplatelets can improve wound angiogenesis and enhance skin wound healing (6–10).The application of PRP in oral clinical research began in the late 1990s. In the fieldof oral medicine, PRP has been used to study bone tissue regeneration. The use ofPRP in oral surgical practice could have beneficial outcomes such as reducedbleeding and enhanced soft tissue healing and bone regeneration (10, 11). It couldalso eliminate concerns of immunogenic reactions and disease transmission. On theother hand, the bioactive factors released by PRP also participate in anabolism,catabolism, and proinflammatory and anti-inflammatory responses, some of whichalso underlie the immune response (12). Therefore, PRP has potential in new applica-tions in tissue engineering.

In this study, we investigated the synergistic effects of 2 kinds of cells: human dentalpulp stem cells (hDPSCs) and endothelial progenitor cells (EPCs). We also used PRPinstead of serum to culture cells so that we could minimize the immune response andinvestigate its vasculogenic potential.

Materials and MethodsPRP Preparation and Activation

The study was approved by the institutional research ethics committee of SouthernMedical University, Guangzhou, China. Human umbilical cord blood was collected fromhealthy volunteers with informed consent in sterile tubes with anticoagulant. Plateletswere separated from umbilical cord blood by secondary centrifugation; 14 mL bloodwas centrifuged at 360g for 20 minutes at room temperature. After the upper layerwas collected, the sample was further centrifuged at 500g for 10minutes. The upper layerwas discarded carefully using a pipette, and the remnants of 1 mL liquid comprised PRP.The platelet number in the PRP was counted by a hematology analyzer (XE-2100; Sysmex,Kobe, Japan). The PRP was then activated by repetitive freeze thawing after the addition ofheparin. Next, centrifugation was performed at 3000g for 20 minutes at room temper-ature to remove platelet membrane fractions. The resulting supernatant was used inour in vitro study.

gzhou, China; and †College of Stomatology, Southern Medical University, Guangzhou, China.omatology, Nanfang Hospital, Guangzhou 510515, China. E-mail address: wubuling623@163.com

Effects of PRP and Cell Coculture 1

Basic Research—Biology

Cell CulturehDPSCs were isolated and cultured using the method

described by Ma et al (13). Dental pulp was separated from healthypremolars or third molars that had been extracted for orthodonticsor impaction from 12- to 25-year-old patients. The hDPSCs werepassaged until 80% confluent. Cell phenotype analysis was per-formed by flow cytometric analysis for CD90/PE, CD105/FITC,CD29/PE, CD44/FITC, and CD14/PE (PharMingen-BD Biosciences,San Diego, CA). After cultured until passage 3, hDPSCs wereused for subsequent experiments.

EPCs were isolated and cultured from human umbilical cordblood obtained from healthy volunteers with informed consent.Mononuclear cells were obtained by density gradient fractionationas described by Reinhold et al and Noriko et al (14, 15). Thecells were resuspended in EBM-2-MV medium (Lonza, Basel,Switzerland) supplemented with 20% fetal bovine serum (FBS)and plated on 6-well plates at a density of 1 � 107 cells/mL thatwere coated with fibronectin. The EPCs were passaged until morethan 80% confluent.

In this study, the cells were grouped as follows: EPCs + 10%FBS (EF), EPCs + 10% PRP (EP), EPCs + hDPSCs + 10% FBS(EDF), and EPCs + hDPSCs + 10% PRP (EDP). In the coculturegroup, the proportion of the 2 kinds of cells and 2 types of mediumwas 1:1 each.

Quantitative Real-time Polymerase Chain ReactionTotal isolated messenger RNA (mRNA) served as the template to

generate complementary DNA through reverse transcription using areagent kit (Invitrogen, Life Technologies, Grand Island, NY). For eval-uation of the gene expression levels of VEGF, PDGF, Flk-1, SDF-1, andglyceraldehyde 3-phosphate dehydrogenase (GAPDH), the primer se-quences were designed as follows:

1. VEGF: sense, 50-CTA CCT CCA CCA TGC CAA GT-3´, and antisense, 50-CAC ACA GGA TGG CTT GAA GA-3´

2. PDGF: sense, 50-ACG TCA GGA AGA AGC CAA AA-3´, and antisense, 50-TCT GGT TGG CTG CTT TAG GT-3´

3. Flk-1: sense, 50-GGT ATT GGC AGT TGG AGG AA-3´, and antisense, 50-ACA TTT GCC GCT TGG ATA AC-3´

4. SDF-1: sense, 50-GCA TTG ACC CGA AGC TAA AG-3´, and antisense,50- ACA CAC ACA CCT GGT CCT CA-3´

5. GAPDH: sense, 50-TCA CCA GGG CTG CTT TTA AC-3´, and antisense,50-GAC AAG CTT CCC GTT CTC AG-3´

The quantitative real-time polymerase chain reactions (qRT-PCRs)were performed using the SYBR Select Master Mix (Life Technologies)on an ABI 7500 real-time polymerase chain reaction system (ABI, Carls-bad, CA). Relative quantization was performed by determining the dif-ference between the threshold cycle (Ct) of GAPDH and the Ct of eachtranscript and computingDDCt. Amplification proceeded as per manu-facturer instructions.

Western Blot AnalysisThe total protein was extracted from the cultured cells and sepa-

rated by sodium dodecyl sulfate polyacrylamide gel electrophoresis(SDS PAGE) and transferred onto a PVDF membrane (Millipore, Biller-ica, MA). The membrane was incubated with rabbit monoclonal anti-VEGF (1:1000; Epitomics, Burlingame, CA) overnight at 4�C. Proteinswere visualized using IRDye 800CW goat antirabbit immunoglobulinG (1:15000; Li-cor, Lincoln, NE) secondary antibody. The membranewas scanned on an Odyssey V3.0 scanner (Li-cor).

2 Li et al.

Tube Formation AssayGrowth factor–reduced basement membrane matrix (356231

Matrigel, BD Biosciences) was placed in 24-well plates per manufac-turer’s instructions. After Matrigel polymerization, 2 � 105 EPCswith corresponding medium were seeded in the wells. In the cocul-ture group, the proportion of the 2 kinds of the cells was 1:1 in themedium (supplemented with 10% FBS). hDPSCs were seeded on Ma-trigel 24 hours after EPCs were seeded. After 48 hours, cells werestained with 2 mmol/L calcein AM (Ebioscience, San Diego, CA). Af-ter 7–48 hours, all the cells were assessed for the presence ofvessel-like structures and imaged using an inverted microscope(Olympus, Tokyo, Japan). The percent tube area was ascertained us-ing ImageJ2x software (NIH, Bethesda, MD) and calculated using thefollowing formula:

Tube area in different groups

Single well area of 24 -- well plates� 100%

Statistical AnalysisStatistical analysis was performed using 1-way analysis of variance

using SPSS 13.0 software (SPSS Inc, Chicago, IL). First, we used the ho-mogeneity test for variance using the Levene test, and if variances wereunequal, we used the Welch test. For multiple comparisons betweengroups, we performed statistical analysis with the Fisher’s Least Signif-icant Difference (LSD) test or the T3 test when variances were equal ornot, respectively. The RQ values (2�DDCT) of the same factor in differentgroups were compared with those for the DF/EF groups (RQ value was1) by using the 1-sample t test. Statistical significance was set at P< .05.

ResultsPRP Characterization

The platelet count range in whole blood was 116–230 � 109/L,whereas PRP averaged 1306–1949 � 109/L, which was significantlyhigher compared with the umbilical cord blood (P < .01).

Effect of PRP on mRNA Levels of Vasculogenesis-relatedFactors

The expression levels of VEGF, PDGF, Flk-1, and SDF-1mRNA weredetected by qRT-PCR on days 3, 7, and 14. The results of qRT-PCR areshown in Figure 1. On day 3, the expression of all the markers in thecoculture groups (EDF and EDP) were higher than in the single-culture groups (EF and EP); similarly, all the markers (except SDF-1) had higher expression levels in the PRP groups (EP and EDP)than in the FBS groups (EF and EDF) (P < .05) (Fig. 1A). The expres-sion levels were significantly higher in the experimental groups (EP,EDF, and EDP) than in the control group (EF) (P < .01) on days7 and 14 (Fig. 1B and C). In addition, the expression levels of 4 markerswere significantly higher in the PRP groups/coculture groups than in theFBS groups/single-culture groups (P < .01) (Fig. 1B and C).

Effect of PRP on Vasculogenesis-related ProteinsThe expression level of VEGF was the highest among the 4markers.

Therefore, we chose VEGF and detected its protein level by Western blot-ting on days 3, 7, and 14. The results of Western blotting are shown inFigure 2. The expression of VEGF had a rising trend in the EP, EDF, andEDP groups compared with the EF group (Fig. 2A–C). On day 3, therewas no significant difference in the expression of VEGF between the EDFand EDP groups (Fig. 2A). In addition, our data showed higher VEGFexpression in the PRP groups than in the FBS groups (Fig. 2B and

JOE — Volume -, Number -, - 2014

Figure 1. Expression levels of VEGF, PDGF, Flk-1, and SDF-1 by qRT-PCR.A–C represent the expression levels of VEGF, PDGF, Flk-1, and SDF-1 ondays 3, 7, and 14, respectively. #P < .05. *P < .01.

Figure 2. Expression levels of VEGF by Western blotting. A–C represent theresults on days 3, 7, and 14, respectively. The expression of VEGF had a risingtrend in the EP, EDF, and EDP groups compared with the EF group.

Basic Research—Biology

C). VEGF expression was also higher in the coculture groups comparedwith the single-culture groups (Fig. 2B and C).

Tube FormationEPCs seeded on Matrigel began to form vessel-like structures after

7 hours. EPCs in PRP had more complete vessel-like structures thanthose grown in FBS (Fig. 3A1 and A2). And the co-culture group couldnot form vessel-like structures after 7 hours because the number ofEPCs in this group was half of the other 2 groups (Fig. 3A3). EPCsformed an extensive lattice of structures 24–48 hours after seeding(Fig. 3B1 and B2). After EPCs were seeded on Matrigel for 24 hours,hDPSCs were added in the coculture groups (Fig. 3B3). The coculturecells could not form complete vessel-like structures. The vessel-likestructures appeared more clearly with calcein AM staining after48 hours (Fig. 3C). hDPSCs appeared around the vessel-like structures

JOE — Volume -, Number -, - 2014

(Fig. 3C). Percent tube areas in the FBS and PRP groups were63% � 0.17% and 67% � 1.8% (P < .05).

DiscussionPRP contains a large number of autologous growth factors and

high levels of platelets. The platelet count in PRP has been found tobe approximately 5 times that in peripheral human blood (16).Here, platelet counts in PRP averaged 1306–1949 � 109/L, which issignificantly higher compared with whole blood (P < .01). PRP acti-vated by repetitive freeze thawing can release many active growth factors(data not shown). These PDGFs play an important role in tissue remod-eling, including neovascularization (6, 17). Significant positive effects ofPRP on the proliferation rate of osteoblasts or osteoblastlike cells andmesenchymal stem cells have been reported (18, 19). Moreover, thetendency of PRP to promote hDPSC differentiation into dentin hasbeen shown (20), as in our previous experiments (data not shown).PRP can also be used in teeth with apical periodontitis or withoutpulp tissue. Pulplike tissue, mineralized tissue, and some fibrous con-nective tissue can be generated in teeth by using PRP as a scaffold inregenerative endodontic procedures (21–23).

The ability to regenerate lost dental tissue by tissue engineeringtechnology has been speculated. In fact, researchers are increasinglyable to focus on certain stem cells, such as hDPSCs, which are easilyavailable from discarded teeth after extraction. Blood supply is veryimportant for the survival and differentiation of stem cells in tissueregeneration. EPCs are the precursor cells of vascular endothelial cells.Their role in promoting angiogenesis has been verified (24, 25), andstudies have shown that EPCs have 2 types in vitro, early EPCs(eEPCs) and outgrowth endothelial progenitor cells (OECs), alsoknown as late EPCs (26). The cells used in this study were like OECs;they had a stronger ability of angiogenesis and proliferation comparedwith eEPCs.

In this study, we cocultured hDPSCs and EPCs to investigate thepotential effects on vasculogenic differentiation. qRT-PCR showed asignificant increase in the expression of vasculogenesis-related factorsin the experimental groups compared with the control group(P < .01), and the expression levels of 4 markers were significantly

Effects of PRP and Cell Coculture 3

Figure 3. EPCs form vascularlike structures. (A) EPCs seeded on Matrigel formed vessel-like structures after 7 hours. (A1) EPCs with 10% FBS. (A2) EPCs with10% PRP. (A3) Coculture of EPCs and hDPSCs (1:1) (magnification, 10�). (B) EPCs seeded on Matrigel formed vessel-like structures after 24 hours (magni-fication, 4�). (C) The cells were stained with 2 mmol/L calcein AM after 48 hours (magnification, 10�). The lower panels are the zoomed-in boxed area of theimages shown in the upper panels.

Basic Research—Biology

higher in the PRP groups/coculture groups than in the FBS groups/single-culture groups (P < .01) on days 7 and 14. Because of its high-est expression among the 4 tested markers, we chose VEGF and testedits expression at the protein level. The result of Western blot analysiswas consistent with that of qRT-PCR. We confirmed that PRP not onlymaintained the growth and proliferation of cells but also promotedvasculogenesis. Tube formation assay confirmed that PRP promotedEPCs to form vessel-like structures. However, the coculture of hDPSCsand EPCs could not form a complete vessel-like structure. In addition,vessel-like structures formed in cocultures persisted longer withoutany degradation (20), and this phenomenon was also discovered dur-ing the experiment.

In conclusion, the present study showed that the coculture ofdifferent kinds of cells (hDPSCs and EPCs) could promote vasculogen-esis, as opposed to a culture with only EPCs. In addition, PRP promotedvasculogenesis better than FBS. Moreover, autologous PRP was found tobe a safe, feasible, and reliable newmedium for culture cells. Thus, PRPcan be used in tissue engineering and might even be applied in toothregeneration in the future.

4 Li et al.

AcknowledgmentsXinzhu Li and Jin Hou contributed equally to this work.The authors thank Ms Jiao Hu and Dr Jun Wen for their assis-

tance with this study.Supported by a grant from National Natural Science Founda-

tion of China (grant no. 81371137).The authors deny any conflicts of interest related to this study.

References1. Eppley BL, Woodell JE, Higgins J. Platelet quantification and growth factor analysis

from platelet-rich plasma: implications for wound healing. Plast Reconstr Surg2004;114:1502–8.

2. Pietramaggiori G, Kaipainen A, Czeczuga JM, et al. Freeze-dried platelet-rich plasmashows beneficial healing properties in chronic wounds. Wound Repair Regen 2006;14:573–80.

3. Robson MC, Phillips LG, Thomason A, et al. Recombinant human platelet-derivedgrowth factor-BB for the treatment of chronic pressure ulcers. Ann Plast Surg1992;29:193–201.

4. Heldin C-H, Westermark B. Mechanism of action and in vivo role of platelet-derivedgrowth factor. Physiol Rev 1999;79:1283–316.

JOE — Volume -, Number -, - 2014

Basic Research—Biology

5. Beck J, Evans D, Tonino PM, et al. The biomechanical and histologic effects of platelet-

rich plasma on rat rotator cuff repairs. Am J Sports Med 2012;40:2037–44.6. Roy S, Driggs J, Elgharably H, et al. Platelet-rich fibrin matrix improves wound

angiogenesis via inducing endothelial cell proliferation. Wound Repair Regen2011;19:753–66.

7. Yang HS, Shin J, Bhang SH, et al. Enhanced skin wound healing by a sustained releaseof growth factors contained in platelet-rich plasma. Exp Mol Med 2011;43:622–9.

8. Ravari H, Hamidi-Almadari D, Salimifar M, et al. Treatment of non-healing woundswith autologous bone marrow cells, platelets, fibrin glue and collagen matrix.Cytotherapy 2011;13:705–11.

9. Tashnizi MA, Alamdari DH, Khayami ME, et al. Treatment of non-healing sternumwound after open-heart surgery with allogenic platelet-rich plasma and fibringlue-preliminary outcomes. Indian J Plast Surg 2013;46:538–42.

10. Albanese A, Licata ME, Polizzi B, et al. Platelet-rich plasma (PRP) in dental and oralsurgery: from the wound healing to bone regeneration. Immun Ageing 2013;10:23.

11. Kaul RP, Godhi SS, Singh A. Autologous platelet rich plasma after third molar sur-gery: a comparative study. J Maxillofac Oral Surg 2012;11:200–5.

12. Boswell SG, Cole BJ, Sundman EA, et al. Platelet-rich plasma: a milieu of bioactivefactors. Arthroscopy 2012;28:429–39.

13. Ma D, Gao J, Wu B, et al. Changes in proliferation and osteogenic differentiation ofstem cells from deep caries in vitro. J Endod 2012;38:796–802.

14. Reinhold J, Christina L, Mervyn W, et al. Outgrowth endothelial cells: characteriza-tion and their potential for reversing ischemic retinopathy. Invest Ophthalmol Vis Sci2010;51:5906–13.

15. Noriko O, Qin L, Yuko H, et al. Optimized method for culturing outgrowth endothe-lial progenitor cells. Inflamm Regen 2011;31:219–27.

16. Mooren RE, Hendriks EJ, van den Beucken JJ, et al. The effect of platelet-rich plasmain vitro on primary cells: rat osteoblast-like cells and human endothelial cells. Tis-sue Eng Part A 2010;16:3159–72.

JOE — Volume -, Number -, - 2014

17. Dohan Ehrenfest DM, Rasmusson L, Albrektsson T. Classification of platelet concen-trates: from pure platelet-rich plasma (P-PRP) to leucocyte- and platelet-rich fibrin(L-PRF). Trends Biotechnol 2009;27:158–67.

18. Ogino Y, Ayukawa Y, Kukita T, et al. Platelet-rich plasma suppresses osteoclastogen-esis by promoting the secretion of osteoprotegerin. J Periodontal Res 2009;44:217–24.

19. Zaky SH, Ottonello A, Strada P, et al. Platelet lysate favours in vitro expansion ofhuman bone marrow stromal cells for bone and cartilage engineering. J TissueEng Regen Med 2008;2:472–81.

20. Dissanayaka WL, Zhan X, Zhang C, et al. Coculture of dental pulp stem cells withendothelial cells enhances osteo-/odontogenic and angiogenic potential in vitro.J Endod 2012;38:454–63.

21. Martin G, Ricucci D, Gibbs JL, et al. Histological findings of revascularized/revital-ized immature permanent molar with apical periodontitis using platelet-rich plasma.J Endod 2013;39:138–44.

22. Torabinejad M, Faras H. A clinical and histological report of a tooth with an openapex treated with regenerative endodontics using platelet-rich plasma. J Endod2012;38:864–8.

23. Jadhav G, Shah N, Logani A. Revascularization with and without platelet-rich plasmain nonvital, immature, anterior teeth: a pilot clinical study. J Endod 2012;38:1581–7.

24. Foster TE, Puskas BL, Mandelbaum BR, et al. Platelet-rich plasma: from basic sci-ence to clinical applications. Am J Sports Med 2009;37:2259–72.

25. Dong Z, Li B, Liu B, et al. Platelet-rich plasma promotes angiogenesis ofprefabricated vascularized bone graft. J Oral Maxillofac Surg 2012;70:2191–7.

26. Medina RJ, O’Neill CL, Humphreys MW, et al. Outgrowth endothelial cells: charac-terization and their potential for reversing ischemic retinopathy. Invest OphthalmolVis Sci 2010;51:5906–13.

Effects of PRP and Cell Coculture 5