Osteogenic differentiation of Wharton’s jelly-derived mesenchymal stem cellscultured on WJ-scaffold through conventional signalling mechanism

Bahareh Beikia, Bahman Zeynalia, Ehsan Taghiabadib, Ehsan Seyedjafaric and Mousa Kehtaria,d

aDevelopmental Biology Laboratory, School of Biology, College of Science, University of Tehran, Tehran, Iran; bDepartment of RegenerativeBiomedicine, Cell Science Research Center, Royan Institute for Stem Cell Biology and Technology, ACECR, Tehran, Iran; cDepartment ofBiotechnology, College of Science, University of Tehran, Tehran, Iran; dDepartment of Stem Cell Biology, Stem Cell Technology ResearchCenter, Tehran, Iran

ABSTRACTWharton’s jelly-derived extracellular matrix (WJ-ECM) has attracted researcher’s attention for its biomed-ical applications. Previously, we fabricated a biomimetic spongy scaffold from decellularized WJ-ECMand, in this study, we sought to examine the osteogenic inductive potential of this scaffold and itsunderlying mechanism. To address this question, mesenchymal stem cells (MSCs) were isolated fromWJ using a mechanical method and cultured on the scaffold, under dynamic condition, for over21 days in the presence or absence of osteogenic medium. The status of signalling pathways involvedin the osteogenic differentiation and the expression profile of integrins in the WJ-derived MSCs(WJ-MSCs) were examined. WJ-MSCs displayed differentiation capacities and expressed surface anti-gens, characteristics of MSCs. Histologically, WJ-MSCs seeded on the scaffold showed a proper cellularattachment, penetration and migration. They also exhibited a higher degree of alkaline phosphataseactivity, calcium deposition and osteogenic gene expression, than those cultured in 2D condition. Theexpression of Wnt, BMP and TGF-b signalling target genes together with that of a2, av and b1 integ-rins was increased in WJ-MSCs in both presence and absence of osteogenic induction medium. Takentogether, our results demonstrate that WJ-derived scaffold induces osteogenic differentiation of WJ-MSCs, possibly through activating integrins and subsequently conventional intracellular signal-ling pathways.

ARTICLE HISTORYReceived 21 July 2018Revised 15 September 2018Accepted 20 September 2018

KEYWORDSWharton’s jelly; scaffold;tissue engineering;osteogenic differentiation;signalling pathways

Introduction

The development of tissue engineering has made it an attract-ive method with great repairing potential for tissue defects.Bone tissue engineering techniques based on autogenouslycell/tissue transplantation would eliminate problems of donorcompatibility, limitation of autograft implantation, pathogentransfer and immune response of allograft bone transplant-ation [1]. The bone tissue engineering requires cellular compo-nents, scaffolds and growth factors. The scaffold provides theinitial extracellular matrix (ECM) required to support the cellsadhesion, proliferation and differentiation [2]. Currently, thescientific challenges of bone tissue engineering are to developsuitable three-dimensional (3D) biodegradable and biocompat-ible scaffolds that easily are fabricated into a variety of shapesand sizes with desirable mechanical properties and intercon-necting porosity [3,4].

Currently, Wharton’s jelly-derived extracellular matrix (WJ-ECM) has attracted researcher’s attention for its biomedicalapplications [5–7]. Since, similar to bone, the major matrixprotein in WJ-ECM is type I collagen [8], it could have posi-tive effects on bone tissue engineering. Moreover, WJ-ECMalso contains significant amounts of hyaluronic acid, GAGs

and various bioactive molecules like, acidic and basicfibroblast growth factors like (aFGF, bFGF), insulin growth fac-tor-I (IGF-I), platelet-derived growth factor (PDGF), epidermalgrowth factor (EGF), transforming growth factor beta 1(TGF-b1) and other peptide growth factors [9] which couldenhance osteogenic differentiation.

Recently, we fabricated a new biomimetic spongy scaffoldfrom decellularized WJ-ECM using freeze-dried technique anddemonstrated that these biocompatible scaffolds are able toimprove cellular attachment, penetration, and proliferationand accelerate the wound healing process in full thicknesswound healing model [7].

The other important aspect of bone tissue engineering ischoosing an ideal cell source with the high capacity of differ-entiating into osteoblasts. Mesenchymal stem cells (MSCs)have been demonstrated as an available source for tissue-engineering applications due to their multiple differentiationcapabilities to diverse tissues (bone, cartilage, adipose tissue,muscle, etc.) [10]. Among MSCs sources, WJ-MSCs that areperinatal stem cells and show an intermediate state betweenembryonic stem cells (ESCs) and adult MSCs, reflect ideal can-didates for bone tissue engineering [11]. Compared to other

CONTACT Bahman Zeynali zeynalib@ut.ac.ir Developmental Biology Lab., School of Biology, College of Science, University of Tehran, P.O. Box 141556455,Tehran, Iran.

Supplemental data for this article can be accessed here.

� 2018 Informa UK Limited, trading as Taylor & Francis Group

ARTIFICIAL CELLS, NANOMEDICINE, AND BIOTECHNOLOGY2018, VOL. 46, NO. S3, S1032–S1042https://doi.org/10.1080/21691401.2018.1528981

sources of MSCs, WJ-MSCs have some advantages, such aseasy isolation and in vitro expansion, high proliferation ratesand differentiation potential, minimal immune reactivity andimmune modulatory effects [12]. Therefore, the focus of thecurrent study was to examine the osteogenic differentiationof WJ-derived mesenchymal stem cells (WJ-MSCs) onWJ-derived scaffolds under dynamic conditions.

The osteogenic differentiation of MSCs has been shownto take place in four steps including osteoprogenitor cells,pre-osteoblasts, osteoblasts and osteocytes that correspondto the processes of proliferation, maturation, matrix synthe-sis and matrix mineralization phases, respectively [13]. Eachstep or phase was shown to depend on the activity ofmultiple signalling pathways, including Wnt, TGF-b and BMPsignalling pathways [14]. There are reports indicating thatWnt pathway promotes the progression of MSCs from pre-osteoblasts into more mature osteoblasts [15]. BMP pathwaywas shown to promote osteogenic commitment and alsoinduce terminal osteogenic differentiation in MSC [16] andTGF-b signalling stimulates proliferation, early differentiation,and commitment of osteoprogenitors into the osteoblasticlineage [14].

In bone formation, the ECM plays an important role inosteoblast differentiation and function. Integrins are involvedin cell-ECM interaction and transmit the signals to cytosol toactivate kinases and signalling pathways including thoseinvolved in osteogenic differentiation [17]. Human MSCs havebeen reported to express a1–6, aV and b1, 3 and 4 integrin[18] which is different between MSCs of various sources [19].It has been demonstrated that several integrins have import-ant role in MSCs osteogenic differentiation. The b1 integrinseems to be important for the differentiation of hMSCs intoosteoblasts and mineralization of matrix [20], a2 integrin isup-regulated during osteogenic differentiation [21] and theexpression of a2b1 integrin is increased during commitmentof hMSCs towards osteoblast fate [22]. Several studies haveshown that MSCs have the ability to change the integrinexpression profile in 3D matrices compared to 2D beds [23],highlighting the importance of choosing appropriate ECM forthe cell culture and differentiation.

Having mentioned the importance of cellular signallingresponse to the 3D scaffolds, there is no much informationavailable. Therefore, in addition to examining osteogenicinduction capacity of our newly fabricated biomimetic scaf-fold, we also investigated the signalling mechanism respon-sible for osteogenic differentiation of WJ-MSCs cultured onthis scaffold.

Materials and methods

Fabrication of WJ-derived scaffold

WJ-ECM was obtained, decellularized and characterized asdescribed [7]. For more details, see supplementary data.

WJ-derived stem cells isolation and characterization

UCs were washed in sterile PBS to remove blood cells con-tamination, and then were cut into 1–2 cm pieces. Umbilical

arteries and vein were removed and WJ explants transferredto a 50ml tube containing PBS. See supplementary data fordetails for WJ-MSCs isolation and osteogenic, chondrogenicand adipogenic differentiation protocols.

WJ-MSCs surface markers expression

1� 105 WJ-MSCs were incubated for 45min with the follow-ing monoclonal antibodies including mouse anti-humanCD105-PE, CD90-FITC, CD44-PE, CD73-PE, CD34/45-PE/FITC,CD133-PE and HLA DR-PE (all from Becton Dickinson, FranklinLakes, NJ). After washing with PBS, cells were analysed inFACSCalibur system (BD Biosciences, San Jose, CA) by flow-ing software.

Cell seeding

A suspension of 2� 105 WJ-MSCs in 100 ml FBS was seededon the WJ-derived scaffold and incubated for 3 h at 37 �C toallow for cell attachment, then 1.5ml of appropriate mediumwas added to each scaffold and left overnight in the incuba-tor before being placed in dynamic condition. For moredetails, see supplementary data.

Cell attachment and migration

Cell migration into the scaffolds was detected by Hoechst33344 (Sigma, St. Louis, MO) staining on 6 mm thick cross sec-tions of scaffolds (see supplementary data for details).

Alkaline phosphatase (ALP) activity and calciumcontent assay

ALP activity and calcium content assay of each WJ-MSCs/scaf-fold construct were measured at day 7, 14 and 21 usingAlkaline Phosphatase Assay kit (Pars Azmoon, Tehran, Iran)and Calcium Content Assay kit (Pars Azmoon, Tehran, Iran),respectively (see supplementary data for details).

Scanning electron microscopy

Matrix mineralization of WJ-MSCs/scaffold construct wasobserved by scanning electron microscope (SEM) analysis(Ais2300, Seron, Uiwang-si, Korea) (see supplementary datafor details).

Mechanical properties of WJ-MSCs/scaffold construct

Mechanical (compressive) property of MSCs/scaffold con-structs at day 7, 14 and 21 of culture in growth and osteo-genic medium was determined using a universal load testmachine (Santam, stm20, Korea). For compressive test, cylin-drical-shaped specimens with approximately 4mm in diam-eter and 8mm in length were conducted at 2mm/mincrosshead speed.

ARTIFICIAL CELLS, NANOMEDICINE, AND BIOTECHNOLOGY S1033

Real-time quantitative PCR analysis (qPCR)

The total RNA of WJ-MSCs seeded on the scaffolds and 2Dculture plates was extracted using RNeasy Mini Kit (Qiagen,Hilden, Germany) at day 7, 14 and 21 of culture in osteogenicand growth medium (see supplementary data for details)(Table 1).

Statistical analysis

One-way analysis of variance (ANOVA) was used for calculat-ing statistically significant differences between groups andt test used for the individual groups. All experimental datawere reported as mean± standard deviation (SD) (n¼ 5) andp values of less than .05 were accepted as significant.

Results

A spongy scaffold was made from decellularized WJ-ECMusing freeze-dried technique. The characterization of the scaf-folds showed heteroporous 3D structure with high degree ofinterconnectivity (data not shown) [7].

WJ-derived MSCs characterization

WJ-derived stem cells after five days culture in DMEM-LGshowed typical MSCs characteristics and displayed spindle-shaped with fibroblastic morphology when adhered toculture plate. WJ-MSCs had ability to differentiate into adipo-genic, chondrogenic and osteogenic lineages, furthermore

expressed mesenchymal CD markers (CD105, CD44, CD90and CD73) and lacked the expression of CD45, CD34, CD133and HLA-DR (Figure 1).

Cellular attachment and migration of WJ-MSCs onthe scaffold

WJ-MSCs attachment and migration into the scaffold weredetected by fluorescence microscopy on serial 6 mm thickcross sections. At day 3, the MSCs were mainly on the surfaceof the scaffold (Figure 2(A)). At day 7, the MSCs hadremained mostly on the top of the scaffold, with only a fewcells penetrating the scaffold (Figure 2(B)). The MSCs hadlargely migrated into the scaffold at day 14 (Figure 2(C)) andby day 21, the MSCs seeded on the scaffolds were distributedthroughout the entire scaffold (Figure 2(D)).

Differentiation and mineralization of WJ-MSCs

The alkaline phosphatase (ALP) activity is an indicator ofcells commitment towards the osteoblastic lineage. WJ-MSCs were cultured in the presence of osteogenic mediumin static or dynamic conditions for 21 days after which theALP activity was examined and compared with TCP as con-trol. As shown in Figure 3, the ALP activity of the cells cul-tured on TCP and scaffold in dynamic condition achievedmaximal levels in day 14 and 21 (there is no significant dif-ference between day 14 and 21). However, the ALP activityon scaffold in static condition did not show significantchanges during time course of osteogenic differentiation.Notably, at each time point in WJ-MSC differentiation, theALP activity of the cells on the scaffold in dynamic condi-tion was significantly higher compared to those in othergroups (p< .05).

The calcium deposition, the indicator of fully maturedosteoblastic cells, was also examined in WJ-MSCs cultured inosteogenic medium for 21 days. Confirming the aboveresults, cells cultured on the scaffolds in dynamic conditionhad the highest calcium deposition on day 21 (Figure 4).The calcium deposition in the scaffold in static conditionhad no significant difference from day 7 to day 21. Theseresults together indicate that the scaffold in static conditionwas not able to induce osteogenic differentiation inWJ-MSCs, therefore, it was excluded from the rest ofexperiments.

SEM analysis of matrix mineralization

The matrix mineralization of WJ-MSCs on the scaffolds ingrowth and osteogenic media was evaluated using SEM. Atday 7 of culture, the cells showed attachment and spreadingbehaviour in both media (Figure 5(A,E)). However, at day 14and 21, the MSCs filled the scaffolds pores with mineral dep-ositions only in osteogenic medium (Figure 5(F,G)). At ahigher magnification, large and rounded aggregations werevisible on the scaffolds in osteogenic medium (Figure 5(H))as compared to the scaffolds in growth medium(Figure 5(D)).

Table 1. Primers used for real-time PCR analysis.

Gene Primer 50–30

ALP F: CAA CAG GGT AGA TTT CTC TTG GR: GGT CAG ATC CAG AAT GTT CC

Osteopontin F: GCC GAG GTG ATA GTG TGG TTR: TGA GGT GAT GTC CTC GTC TG

Actin b F: CTT CCT TCC TGG GCA TGR: GTC TTT GCG GAT GTC CAC

Osteonectin F: GCA AAG GGA AGT AAC AGA CACR: GAA AGG TAA AGG AGG AAA TGG

Osteocalcin F: GTG CAG AGT CCA GCA AAG GTR: TCA GCC AAC TCG TCA CAG TC

Runx2 F: ATGACACTGCCACCTCTGAR: ATG AAA TGC TTG GGA ACT GC

Id1 F: GAC ACA AGA TGC GAT CGT CCR: AGT TGG AGC TGA ACT CGG AA

C-myc F: CTA CCC TCT CAA CGA CAGR: TCT TCT TGT TCC TCC TCA G

CyclinD1 F: GAC CTT CGT TGC CCT CTGR: GGT TCA GGC CTT GCA CTG

Smad7 F: AAA CAG GGG GAA CGA ATT ATCR: ACC ACG CAC CAG TGT GAC

Pai1 F: GAC TCC CTT CCC CGA CTCR: GGG CGT GGT GAA CTC AGT A

ITG-A2 F: TAG CGC TCA GTC AAG GCA TTR: GCA CTG CAT AGC CAA ACT GT

ITG-b1 F: GTG GGT GGT GCA CAA ATT CR: GGT CAA TGG GAT AGT CTT CAG C

ITG-A5 F: TCC TGG CTG GCT GGT ATT AGR: CTT CAA CTT AGA CGC GGA GG

ITG-A6 F: TCA TGG ATC TGC AAA TGG AAR: GCG GGG TTA GCA GTA TAT TCA

ITG-V F: GCA ACA GGC AAT AGA GATR: TGC TGA ATC CTC CTT GAC AA

S1034 B. BEIKI ET AL.

Mechanical properties of WJ-MSCs/scaffold constructs

Compressive modulus of WJ-MSCs/scaffold constructs in growthand osteogenic medium was measured at day 7, 14 and 21 of

culture (Figure 6). The highest compressive modulus was meas-ured on the scaffold in osteogenic medium at day 21(0.71±0.04MPa) that was 4.7 fold increase compared to that of

Figure 1. Characterization of WJ-MSCs. (A) Photomicrograph of MSCs with spindle-shaped morphology after five days of culture, (B) MSCs monolayer at approxi-mately 90% confluence in P2. (C) MSCs colonies (black arrows) in P3. (D) Osteogenic differentiation of P3 MSCs confirmed by alizarin red staining at day 21.(E) Adipogenic differentiation of P3 MSCs confirmed by oil red staining at day 21. (F) Chondrogenic differentiation of P3 MSCs confirmed by Alcian blue at day 21.(G) Flow cytometry analysis of WJ-MSCs surface CD markers expression.

ARTIFICIAL CELLS, NANOMEDICINE, AND BIOTECHNOLOGY S1035

the scaffold in growth medium (�p< .05). This result indicatesthat the mechanical properties of the scaffolds are improved byosteogenic differentiation of MSCs due to the matrixmineralization.

Expression of osteogenic specific genes

The relative expression for the bone associated gene Runx 2,ALP, osteocalcin (OCN) and osteopontin (OPN) was performed by

Figure 2. WJ-MSCs migration within the scaffolds in growth medium. (X) The 3D spongy scaffolds fabricated from WJ-derived ECM (15mm diameter, 3mm thick-ness). (A–D) Histological sections of fluorescent photomicrographs stained with Hoechst 33344 at day 3 (A), 7 (B), 14 (C) and 21 (D). Note that the cells are distrib-uted (arrows) on the surface and inside the scaffolds during three weeks of culture. Scale bars: 25 mm in A and B; 50 mm in C and D.

Figure 3. Alkaline phosphatase (ALP) activity of WJ-MSCs. ALP activity was assayedin all groups (TCP and scaffold in static and dynamic conditions) for over 21 daysin the presence of osteogenic medium. Note that the ALP activity on the scaffoldin dynamic condition was significantly higher than two other groups (�p< .05).

Figure 4. Calcium deposition in WJ-MSCs/scaffold constructs during the21 days culture period in the three groups (TCP, scaffold in static condition, andscaffold in dynamic condition). Note that the calcium deposition on the scaffoldin dynamic condition was significantly higher compared with two other groups(�p< .05), while on the scaffold in static had no significant difference duringthe 21 days.

S1036 B. BEIKI ET AL.

qPCR analysis in WJ-MSCs cultured on the scaffolds in growthand osteogenic medium. When growth medium was used(Figure 7(A)), the expression of all osteogenic markers wasincreased in the cells cultured on the scaffolds up to 28 timesmore than TCP. In the case of differentiation medium (Figure7(B)), as expected, this increased level of gene expression wasup to 70 folds. These results clearly indicate that the scaffoldsignificantly increase the osteogenic differentiation of WJ-MSCseven in the absence of osteogenic induction medium.

Expression of signalling pathways target genes inWJ-MSCs

To investigate the intracellular mechanism by which the scaf-fold enhances osteogenic differentiation of WJ-MSCs, the sta-tus of Wnt, BMP and TGF-b signalling pathways, known toinvolve in osteoblast differentiation, was evaluated in thepresence of osteogenic induction medium. To address thisquestion, the expression of target genes of these signalling

Figure 5. SEM micrographs of WJ-MSCs mineralization on the scaffolds in the presence of growth and osteogenic medium at day 7, 14 and 21. The MSCs areattached and spread on the scaffolds on both groups (A and E), however, only in osteogenic medium they filled the scaffold pores with mineral depositions (F, G).Note that aggregates (most probably calcium salts) are deposited on the scaffold only by MSCs culture in the presence of osteogenic medium (H). Scale bars:500 mm in A–C and E–G; 3 mm in D and E.

ARTIFICIAL CELLS, NANOMEDICINE, AND BIOTECHNOLOGY S1037

pathways was examined in the cells on day 7, 14 and 21 ofculture using qPCR analysis. As shown in Figure 8, theexpression of cyclin D1 and c-Myc (Wnt target genes) was sig-nificantly highest at day 14 compared to TCP (p< .05). Thehighest expression of Id1 (BMP target gene) and pai andsmad 7 (TGF-b target genes) was observed at day 7 (p< .05).These results indicate differential activities of these signallingpathways during osteogenic differentiation of WJ-MSCs.

To investigate the effects of the scaffold alone on theactivities of these signalling pathways, the expression oftheir target genes was examined in WJ-MSCs afterseven days of culture in the absence of osteogenic medium.The effects of the scaffold alone on osteogenic

differentiation of WJ-MSCs showed that the expression of alltarget genes (except of pai) increased significantly comparedto that of TCP (p<.05) (Figure 9). These results indicate thepositive role of the scaffold in activating of the signallingpathways involved in osteogenic differentiation without theinducing factors.

Figure 6. Compressive moduli of WJ-MSCs/scaffold constructs in growth and osteogenic medium at day 7 (dark blue lines), 14 (green lines) and 21 (light blue lines).The constructs in osteogenic medium at day 21 shows a 4.7-fold increase in compressive modulus compared with growth medium (�p< .05).

Figure 7. Expression of osteogenic specific genes in WJ-MSCs cultured on thescaffolds in growth (A) and osteogenic medium (B). The cells were cultured onthe scaffold or TCP for over 21 days in osteogenic medium after which expres-sion of the osteogenic specific marker was examined using real-time PCR. ThemRNA relative expression of osteogenic genes in MSCs cultured in growth andosteogenic medium was clearly higher than that on TCP (�p< .05, ��p< .01).OCN: osteocalcin; OPN: osteopontin; ALP: alkaline phosphatase.

Figure 8. Expression of Wnt, BMP and TGF-b signalling target genes in WJ-MSCs. The cells were cultured on the scaffold or TCP for over 21 days in osteo-genic medium after which expression of the target genes was examined usingreal-time PCR. The expression of C-myc and Cyclind1 (Wnt target genes) wasclearly higher than TCP in the second week. The expression of Id1 (BMP targetgene) and also Smad7 and Pai (TGF-b target genes) was the highest in the firstweek of osteogenic differentiation compared to that of TCP. Dy: dynamic condi-tion (�p< .05, ��p< .01).

Figure 9. Expression of Wnt, BMP and TGF-b signalling target genes inWJ-MSCs cultured on scaffold in the absence of osteogenic induction medium.The cells were cultured on the scaffold or TCP for seven days in growth mediumafter which expression of the target genes were examined using real-time PCR.The expression of all target genes (except of Pai) was clearly higher than that inTCP. D: dynamic condition (�p< .05, ��p< .01).

S1038 B. BEIKI ET AL.

Expression of integrins in WJ-MSCs

Integrins have been shown to mediate cell adhesion and attach-ment and also to regulate multiple intracellular signalling path-ways [24]. Therefore, the expression of several integrin genesincluding a2, av, a5, a6 and b1 involved in osteogenic differenti-ation [25,26], was examined in WJ-MSCs after seven days of cul-ture on the scaffold with or without osteogenic medium. In thepresence of osteogenic medium and dynamic condition, theexpression of all integrins, except the integrin a6, showed a sig-nificant increase compared to that of TCP (Figure 10(B)) (p< .05).In the absence of osteogenic medium (but the presence ofgrowth medium), significant increase in the expression of a2, avand b1 integrins was still observed compared to that of TCP(Figure 10(A)) (p<.05). These results indicate that, regardless ofthe presence or absence of osteogenic induction medium, thescaffold induces expression of the integrins involved in osteo-genic differentiation in WJ-MSCs.

Discussion

The tissue-derived acellular ECM as a natural material hasattracted researcher’s attention because of its valuable sourceof growth factors and bioactive molecules that provide idealsubstrate for tissue engineering and regenerative medicine[27–29]. Many researchers are seeking to use human tissue-derived ECM as a scaffold for clinical purposes to overcomethe risk of immunological reactions and pathogen transmis-sion of animal tissue-derived materials [30].

Previously, we introduced the WJ-ECM as an ideal sourceof natural material for fabrication of biocompatible spongyscaffold and used it as an acceptable skin graft for woundhealing [7]. Since, similar to bone, the major matrix protein inWJ-ECM is type I collagen [8], it could provide ideal substratefor bone tissue engineering. Here, we examine the ability ofour new biomimetic spongy scaffold to induce osteogenicdifferentiation of WJ-MSCs and investigate the signallingpathways involving this differentiation.

In the present study, WJ-MSCs were used to evaluate thecellular response to the scaffold in osteogenic medium anddynamic conditions. Like other researchers [31–34], MSCs wereisolated from the WJ using a mechanical method to avoiddecreased cellular viability, degradation of cellular surfacereceptors, and changed cellular function. Here, we describe,for the first time, a simplified mechanical method to obtain asufficient number of MSCs from all umbilical cord samples,without enzymatic treatment. Moreover, these MSCs exhibitedfunctional properties and expression of surface antigenmarkers similar to MSCs. Our results further revealed that theWJ-MSCs cultured on WJ-scaffold exhibited a high degree ofattachment and migration. The SEM images demonstratedthat all pores of scaffold on day 21, was filled by the mineraldepositions. Analysis of the ALP activity and matrix mineraliza-tion showed that WJ-MSCs cultured on the scaffold weremore differentiated in comparison to TCP. Moreover, ourresults indicated that the increased osteogenic differentiationof WJ-MSCs was observed when these cells were cultured inthe dynamic conditions and not in static condition. Thisincreased osteogenic differentiation in response to dynamic

conditions is consistent with the other reports [35,36], andmight be as a result of the convective transport of nutrientsto the surface of the scaffold, in contrast to the purely diffu-sional transport in static culture, increasing concentration ofoxygen throughout the scaffold [37].

To investigate the role of the scaffold alone, on the osteo-genic differentiation of WJ-MSCs and its underlying mecha-nisms, these cells were cultured on the scaffolds in thedynamic conditions and in the presence or absence of osteo-genic inductive medium. In the presence of the medium, theexpression of both early (Runx2, highest at day 7 and ALP,highest at day 14) and late specific osteogenic gene markers(OCN highest at day 14 and OPN, highest at day 21) was sig-nificantly increased in the cells cultured on the scaffold com-pared to that of TCP. The status of osteogenic signallingpathways (by examining the expression of their target genes)showed that while highest activity of TGFb and BMP signallingwas observed at day 7, Wnt/b-catenin signalling displayedhighest activity at day 14. This activity pattern of Wnt, TGFband BMP signalling pathways during osteogenic differentiationhas already been reported [38–43]. The expression of integ-rins, as the signal transducer from the ECM to the intracellularsignalling pathways, was also examined in the WJ-MSCs. Ourresults demonstrated that the expression of a2, av, a5 and b1integrins was significantly increased in the cells cultured onthe scaffold compared to that of TCP. These results indicatethat, in the presence of osteogenic medium, differentiation ofWJ-MSCs cultured on the scaffold is improved compared to

Figure 10. Expression of integrins in the WJ-MSCs cultured on the scaffold inthe growth (A) or osteogenic medium (B) under dynamic conditions (Dy). Thecells were cultured on the scaffold or TCP for seven days after which expressionof integrins was examined using real-time PCR. The expression of a2, av and b1integrins was clearly higher than that in TCP in both osteogenic and growthmedia after seven days of culture. Dy: dynamic condition (�p< .05, ��p< .01).

ARTIFICIAL CELLS, NANOMEDICINE, AND BIOTECHNOLOGY S1039

that of TCP, most probably through conventional signallingmechanisms. Consistent with this conclusion, it was reportedthat the interaction of cell surface integrins with collagenouscompounds of scaffolds activated signalling pathways induc-ing osteogenic specific genes expression [44,45].

In the absence of osteogenic inductive medium, interest-ingly, the expression of all osteogenic gene markers wasincreased, in all time point, in WJ-MSCs cultured on the scaf-fold compared to that of TCP. This result indicates that ourfabricated scaffold induces osteogenic differentiation of WJ-MSCs even in the absence of osteogenic induction medium.The roles of collagenous scaffolds to induce osteogenic differ-entiation in the absence of inducing factor was also reportedby Byrne et al. who examined the expression of osteogenicgenes in MSCs cultured on a cow collagen-glycosaminogly-can-derived scaffold, and indicated that collagen-based scaf-folds are capable to induce bone gene expression in theabsence of bone induction factors [46]. As we expected, theexpression of the integrins (a2, av and b1) was also increasedin the MSCs cultured on the WJ-scaffold in the absence ofosteogenic inductive medium. Generally, it is well acceptedthat 3D scaffolds increase expression or activation (dimeriza-tion) of integrins [47,48]. More specifically, it has beenreported that the binding of MSCs through their a2b1 integ-rins to collagen type I based substrates is sufficient to inducebone differentiation, even in the absence of bone inductionfactors [44,49]. Therefore, our results are consistent with theother reports [20,21] that collagen based matrix increases thedifferentiation of osteoblasts by interacting with integrins.

The signalling mechanism by which the 3D scaffolds affectstem cell differentiation is not well understood. It is wellknown that Wnt, BMP and TGFb signalling pathways aremajor player in osteogenic differentiation in vitro and in vivo[14]. Interestingly, we observed a remarkable increase in theactivity of these signalling pathways (in a different pattern,summarized in Figure 11) in the MSCs, in the absence ofosteogenic inductive medium, indicating that our fabricatedscaffold improves the osteogenic differentiation of WJ-MSCsthrough the conventional signalling mechanisms.

Conclusions

In the present study, we described for the first time a simpli-fied mechanical method to obtain a sufficient number of

MSCs from the WJ samples with minimum time consuming.These MSCs were then cultured on the WJ-derived scaffoldto examine whether the scaffold enhances their osteogenicdifferentiation. Our results showed that WJ-derived spongyscaffolds increased the osteoblastic differentiation of theMSCs compared to TCP, in the presence or absence of boneinduction factors, confirming the positive effect of the scaf-fold on bone differentiation possibly through providing moresurface area and adhesion sites for the cells. The increasedexpression of integrins, and Wnt, TGF-b and BMP signallingtarget genes even in the absence of osteogenic inductionmedium suggest that activation of integrins and subse-quently these signalling pathways, by the scaffold, could besufficient to induce osteogenic differentiation WJ-MSCs.Further experiments are needed to clarify more precisely therole of integrins on osteogenic differentiation of MSCs and toinvestigate how scaffold-activated integrins induce signallingpathways involved in osteogenic differentiation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Funding

This study was supported by University of Tehran and Iranian Council forStem Cell Sciences and Technologies.

References

[1] Liu X, Ma PX. Polymeric scaffolds for bone tissue engineering.Ann Biomed Eng. 2004;32:477–486.

[2] Porter JR, Ruckh TT, Popat KC. Bone tissue engineering: a reviewin bone biomimetics and drug delivery strategies. BiotechnolProg. 2009;25:1539–1560.

[3] Xu C, Su P, Chen X, et al. Biocompatibility and osteogenesis ofbiomimetic bioglass–collagen–phosphatidylserine compositescaffolds for bone tissue engineering. Biomaterials. 2011;32:1051–1058.

[4] Le Pape F, Richard G, Porchet E, et al. Adhesion, proliferation andosteogenic differentiation of human MSCs cultured under perfu-sion with a marine oxygen carrier on an allogenic bone substi-tute. Artif Cells Nanomed Biotechnol. 2018;46:95–107.

[5] Stocco E, Barbon S, Radossi P, et al. Autologous chondrocytes asa novel source for neo-chondrogenesis in haemophiliacs. CellTissue Res. 2016;366:51–61.

Figure 11. Schematic image of signalling pathways that regulates the key stages of bone differentiation. The BMP pathway plays a role in all stages of MSCs differ-entiation into osteocytes. TGF-b plays an inductive role in the proliferation of osteoprogenitors and their early differentiation into immature osteoblasts; however, ithas also an inhibitory role in the further differentiation of immature osteoblasts through inhibiting bone matrix mineralization. Wnt signalling pathway inducesimmature osteoblasts into mature osteoblasts.

S1040 B. BEIKI ET AL.

[6] Herrero-Mendez A, Palomares T, Castro B, et al. HR007: a family ofbiomaterials based on glycosaminoglycans for tissue repair.J Tissue Eng Regen Med. 2015;11:989–1001.

[7] Beiki B, Zeynali B, Seyedjafari E. Fabrication of a three dimen-sional spongy scaffold using human Wharton’s jelly derived extracellular matrix for wound healing. Mater Sci Eng C Mater BiolAppl. 2017;78:627–638.

[8] Nanaev AK, Kohnen G, Milovanov AP, et al. Stromal differentiationand architecture of the human umbilical cord. Placenta. 1997;18:53–64.

[9] Sobolewski K, Małkowski A, Ba�nkowski E, et al. Wharton’s jelly asa reservoir of peptide growth factors. Placenta. 2005;26:747–752.

[10] Mauney JR, Volloch V, Kaplan DL. Role of adult mesenchymalstem cells in bone tissue engineering applications: current statusand future prospects. Tissue Eng. 2005;11:787–802.

[11] Kim DW, Staples M, Shinozuka K, et al. Wharton’s jelly-derivedmesenchymal stem cells: phenotypic characterization and opti-mizing their therapeutic potential for clinical applications. IJMS.2013;14:11692–11712.

[12] La Rocca G, Lo Iacono M, Corsello T, et al. Human Wharton’s jellymesenchymal stem cells maintain the expression of key immuno-modulatory molecules when subjected to osteogenic, adipogenicand chondrogenic differentiation in vitro: new perspectives forcellular therapy. CSCR. 2013;8:100–113.

[13] Neve A, Corrado A, Cantatore FP. Osteoblast physiology in normaland pathological conditions. Cell Tissue Res. 2011;343:289–302.

[14] Chen G, Deng C, Li YP. TGF-b and BMP signaling in osteoblastdifferentiation and bone formation. Int J Biol Sci. 2012;8:272–288.

[15] Case N, Rubin J. Beta-catenin—a supporting role in the skeleton.J Cell Biochem. 2010;110:545–553.

[16] Hu W, Ye Y, Wang J, et al. Bone morphogenetic proteins inducerabbit bone marrow-derived mesenchyme stem cells to differenti-ate into osteoblasts via BMP signals pathway. Artif Cells NanomedBiotechnol. 2013;41:249–254.

[17] Saidak Z, Le Henaff C, Azzi S, et al. Wnt/beta-catenin signalingmediates osteoblast differentiation triggered by peptide-inducedalpha5beta1 integrin priming in mesenchymal skeletal cells. J BiolChem. 2015;290:6903–6912.

[18] Majumdar MK, Keane-Moore M, Buyaner D, et al. Characterizationand functionality of cell surface molecules on human mesenchy-mal stem cells. J Biomed Sci. 2003;10:228–241.

[19] Brooke G, Tong H, Levesque JP, et al. Molecular trafficking mecha-nisms of multipotent mesenchymal stem cells derived fromhuman bone marrow and placenta. Stem Cells Dev. 2008;17:929–940.

[20] Gronthos S, Simmons PJ, Graves SE, et al. Integrin-mediated inter-actions between human bone marrow stromal precursor cells andthe extracellular matrix. Bone. 2001;28:174–181.

[21] Foster LJ, Zeemann PA, Li C, et al. Differential expression profilingof membrane proteins by quantitative proteomics in a humanmesenchymal stem cell line undergoing osteoblast differentiation.Stem Cells. 2005;23:1367–1377.

[22] Walsh S, Jordan GR, Jefferiss C, et al. High concentrations of dexa-methasone suppress the proliferation but not the differentiationor further maturation of human osteoblast precursors in vitro:relevance to glucocorticoid-induced osteoporosis. Rheumatology(Oxford). 2001;40:74–83.

[23] Martino MM, Mochizuki M, Rothenfluh DA, et al. Controlling integ-rin specificity and stem cell differentiation in 2D and 3D environ-ments through regulation of fibronectin domain stability.Biomaterials. 2009;30:1089–1097.

[24] Walling BL, Kim M. LFA-1 in T cell migration and differentiation.Front Immunol. 2018;9:952.

[25] Becerra-Bayona SM, Guiza-Arguello VR, Russell B, et al. Influenceof collagen-based integrin alpha1 and alpha2 mediated signalingon human mesenchymal stem cell osteogenesis in three dimen-sional contexts. J Biomed Mater Res A. 2018.

[26] Wang H, Li J, Zhang X, et al. Priming integrin alpha 5 promotesthe osteogenic differentiation of human periodontal ligament

stem cells due to cytoskeleton and cell cycle changes.J Proteomics. 2018;179:122–130.

[27] Swinehart IT, Badylak SF. Extracellular matrix bioscaffolds in tissueremodeling and morphogenesis. Dev Dyn. 2016;245:351–360.

[28] Wolf K, Friedl P. Extracellular matrix determinants of proteolyticand non-proteolytic cell migration. Trends Cell Biol. 2011;21:736–744.

[29] Taylor-Weiner H, Schwarzbauer JE, Engler AJ. Defined extracellularmatrix components are necessary for definitive endoderm induc-tion. Stem Cells. 2013;31:2084–2094.

[30] Yang C, Hillas PJ, B�aez JA, et al. The application of recombinanthuman collagen in tissue engineering. BioDrugs. 2004;18:103–119.

[31] De Bruyn C, Najar M, Raicevic G, et al. A rapid, simple, and repro-ducible method for the isolation of mesenchymal stromal cellsfrom Wharton’s jelly without enzymatic treatment. Stem CellsDev. 2011;20:547–557.

[32] La Rocca G, Anzalone R, Corrao S, et al. Isolation and characteriza-tion of Oct-4þ/HLA-Gþmesenchymal stem cells from humanumbilical cord matrix: differentiation potential and detection ofnew markers. Histochem Cell Biol. 2009;131:267–282.

[33] Wu LF, Wang NN, Liu YS, et al. Differentiation of Wharton’s jellyprimitive stromal cells into insulin-producing cells in comparisonwith bone marrow mesenchymal stem cells. Tissue Eng A. 2009;15:2865–2873.

[34] Petsa A, Gargani S, Felesakis A, et al. Effectiveness of protocol forthe isolation of Wharton’s Jelly stem cells in large-scale applica-tions. In Vitro Cell Dev Biol Anim. 2009;45:573–576.

[35] Meinel L, Karageorgiou V, Fajardo R, et al. Bone tissue engineer-ing using human mesenchymal stem cells: effects of scaffoldmaterial and medium flow. Ann Biomed Eng. 2004;32:112–122.

[36] Wang TW, Wu HC, Wang HY, et al. Regulation of adult humanmesenchymal stem cells into osteogenic and chondrogenic line-ages by different bioreactor systems. J Biomed Mater Res A. 2009;88:935–946.

[37] Sikavitsas VI, Bancroft GN, Mikos AG. Formation of three-dimen-sional cell/polymer constructs for bone tissue engineering in aspinner flask and a rotating wall vessel bioreactor. J BiomedMater Res. 2002;62:136–148.

[38] Komori T. Regulation of bone development and maintenance byRunx2. Front Biosci. 2008;13:898–903.

[39] Tang Y, Wu X, Lei W, et al. TGF-beta1-induced migration of bonemesenchymal stem cells couples bone resorption with formation.Nat Med. 2009;15:757–765.

[40] Yan X, Liao H, Cheng M, et al. Smad7 protein interacts withreceptor-regulated Smads (R-Smads) to inhibit transforminggrowth factor-b (TGF-b)/Smad signaling. J Biol Chem. 2016;291:382–392.

[41] Javed A, Afzal F, Bae JS, et al. Specific residues of RUNX2 areobligatory for formation of BMP2-induced RUNX2-SMAD complexto promote osteoblast differentiation. Cells Tissues Org. 2009;189:133–137.

[42] Phimphilai M, Zhao Z, Boules H, et al. BMP signaling is requiredfor RUNX2-dependent induction of the osteoblast phenotype.J Bone Miner Res. 2006;21:637–646.

[43] Tachi K, Takami M, Sato H, et al. Enhancement of bone morpho-genetic protein-2-induced ectopic bone formation by transform-ing growth factor-b1. Tissue Eng A. 2011;17:597–606.

[44] Mizuno M, Kuboki Y. Osteoblast-related gene expression of bonemarrow cells during the osteoblastic differentiation induced bytype I collagen. J Biochem. 2001;129:133–138.

[45] Li N, Zhang M, Drummen GP, et al. Sika deer antler collagen typeI-accelerated osteogenesis in bone marrow mesenchymal stemcells via the Smad pathway. Evid Based Complement AlternatMed. 2016;2016:1.

[46] Byrne EM, Farrell E, McMahon LA, et al. Gene expression by mar-row stromal cells in a porous collagen-glycosaminoglycan scaffoldis affected by pore size and mechanical stimulation. J Mater Sci:Mater Med. 2008;19:3455–3463.

ARTIFICIAL CELLS, NANOMEDICINE, AND BIOTECHNOLOGY S1041

[47] Heckmann L, Fiedler J, Mattes T, et al. Mesenchymal progenitorcells communicate via alpha and beta integrins with a three-dimensional collagen type I matrix. Cells Tissues Org. 2006;182:143–154.

[48] Jakubikova J, Cholujova D, Hideshima T, et al. A novel 3D mesen-chymal stem cell model of the multiple myeloma bone marrow

niche: biologic and clinical applications. Oncotarget. 2016;7:77326–77341.

[49] Salasznyk RM, Williams WA, Boskey A, et al. Adhesion to vitro-nectin and collagen I promotes osteogenic differentiation ofhuman mesenchymal stem cells. J Biomed Biotechnol. 2004;2004:24–34.

S1042 B. BEIKI ET AL.

Copyright of Artificial Cells, Nanomedicine & Biotechnology is the property of Taylor &Francis Ltd and its content may not be copied or emailed to multiple sites or posted to alistserv without the copyright holder's express written permission. However, users may print,download, or email articles for individual use.