Cytotherapy, 2013; 0: 1e12

Syngeneic adipose-derived stem cells with short-termimmunosuppression induce vascularized composite allotransplantationtolerance in rats

HUI-YUN CHENG1,2, NICOLAE GHETU3, WEI-CHAO HUANG4, YEN-LING WANG1,CHRISTOPHER GLENN WALLACE5, CHIH-JEN WEN6, HUNG-CHANG CHEN7,LING-YI SHIH5, CHIH-FAN LIN6, SHIAW-MIN HWANG8, SHUEN-KUEI LIAO9,10 &FU-CHAN WEI1,5,6

1Center for Vascularized Composite Allotransplantation, Chang Gung Memorial Hospital, Gueishan, Taiwan,2Department of Medical Research and Development Linkou Branch, Chang Gung Medical Foundation, Taoyuan,Gueishan, Taiwan, 3Former Microsurgery Fellow, Chang Gung Memorial Hospital; Regional Oncological Institute,University of Medicine and Pharmacy. “Grigore T. Popa,” Iasi, România, 4Division of Plastic and ReconstructiveSurgery, Tzu Chi General Hospital at Taipei, New Taipei, Taiwan, 5Department of Plastic and Reconstructive Surgery,Chang Gung Memorial Hospital, Gueishan, Taiwan, 6School of Medicine, Chang Gung University, Gueishan,Taiwan, 7Graduate Institute of Biomedical Sciences, Chang Gung University, Gueishan, Taiwan, 8BioresourceCollection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan, 9GraduateInstitute of Cancer Biology and Drug Discovery and Center of Excellence for Cancer Research, Taipei MedicalUniversity, Taipei, Taiwan, and 10R&D Division, Vectorite Biomedica Inc, Taipei, Taiwan

AbstractBackground aims. A clinically applicable tolerance induction regimen that removes the requirement for lifelong immuno-suppression would benefit recipients of vascularized composite allotransplantation (VCA). We characterized the immuno-modulatory properties of syngeneic (derived from the recipient strain) adipocyte-derived stem cells (ADSCs) andinvestigated their potential to induce VCA tolerance in rats. Methods. ADSCs were isolated from Lewis (LEW, RT1Al) rats;their immunomodulatory properties were evaluated by means of mixed lymphocyte reactions in vitro and VCAs in vivo acrossa full major histocompatibility complex mismatch with the use of Brown-Norway (BN, RT1An) donor rats. Two controland four experimental groups were designed to evaluate treatment effects of ADSCs and transient immunosuppressants(anti-lymphocyte globulin, cyclosporine) with or without low-dose (200 cGy) total body irradiation. Flow cytometry wasperformed to quantify levels of circulating CD4þCD25þFoxP3þ regulatory T cells (Tregs). Results. Cultured syngeneicADSCs exhibited CD90.1þCD29þCD73þCD45�CD79a�CD11b/c� phenotype and the plasticity to differentiate to adi-pocytes and osteocytes. ADSCs dramatically suppressed proliferation of LEW splenocytes against BN antigen and mitogen,respectively, in a dose-dependent fashion, culminating in abrogation of allo- and mitogen-stimulated proliferation at thehighest concentration tested. Accordingly, one infusion of syngeneic ADSCs markedly prolonged VCA survival in LEWrecipients treated with transient immunosuppression; of these, 66% developed tolerance. Total body irradiation provided noadditional VCA survival benefit. An important role for Tregs in tolerance induction/maintenance was suggested in vivo andin vitro. Conclusions. Treatment comprising syngeneic ADSCs and transient immunosuppression (i) increased levels ofcirculating Tregs and (ii) induced tolerance in 66% of recipients of major histocompatibility complexemismatched VCAs.

Key Words: adipose-derived stem cell, allotransplantation, regulatory T cell, tolerance, vascularized composite

Introduction

Vascularized composite allotransplantation (VCA)describes the en bloc reconstruction of a recipient’sanatomical unit, such as hand/forearm, abdominal

Correspondence: Shuen-Kuei Liao, PhD, Graduate Institute of Cancer Biology aliaosk@h.tmu.edu.tw; Fu-Chan Wei, MD, Center for Vascularized Composite AlE-mail: fuchanwei@gmail.com

(Received 26 February 2013; accepted 27 June 2013)

ISSN 1465-3249 Copyright � 2013, International Society for Cellular Therapy. Phttp://dx.doi.org/10.1016/j.jcyt.2013.06.020

wall or face, by replacing it with a transplantation ofthe corresponding part from a deceased donor(1e3). The term “allotransplantation” is appliedbecause the donor and recipient are genetically

nd Drug Discovery, Taipei Medical University, Taipei 110, Taiwan. E-mail:lotransplantation, Chang Gung Memorial Hospital, Gueishan 333, Taiwan.

ublished by Elsevier Inc. All rights reserved.

2 H.-Y. Cheng et al.

non-identical but belong to the same species. Unlikesolid-organ transplants, VCAs characteristicallycontain multiple tissue types, such as skin, muscle,nerve and often bone/marrow, which exert differentdegrees of immunogenicity (4). The technique hasthe potential to revolutionize reconstructive surgerybut remains hindered by the requirement for lifelongnon-specific immunosuppressants and their atten-dant toxicities, some of which may be fatal (5,6).Conceivably, these problems could be solved by in-duction of donor-specific transplantation tolerancethat allows complete withdrawal of immunosup-pressants without harming VCA survival (7).

Mesenchymal stromal cells (MSCs) have potentimmunomodulatory properties. They can suppressT- and B-lymphocyte activation and proliferation(8,9), dendritic cell differentiation and maturation(10,11) and natural killer cell activity (12). Donorbone marrowederived MSCs (BM-MSCs) inducedtolerance to semi-allogeneic cardiac (13,14), alloge-neic renal (15), xenogeneic islet cell (16) and allo-geneic skin transplants (17). Kuo et al. (18)successfully induced VCA tolerance in outbred swinewith the use of multiple rounds of donor BM-MSCscombined with non-myeloablative irradiation, donorbone marrow transplantation and transient immu-nosuppression. Similarly, tolerance was successfullyinduced in a swine hemi-face VCA model by use ofmultiple rounds of donor BM-MSCs and transientimmunosuppression but without irradiation or bonemarrow transplantation (BMT) (19). However, analternative, more accessible and plentiful source ofMSCs than marrow is adipose tissue (20). Accord-ingly, Kuo et al. (21) investigated donor-derivedadipocyte-derived stem cells (ADSCs) for hind limbVCA in rats, reporting successful tolerance induc-tion across a full major histocompatibility complex(MHC) mismatch with multiple doses of ADSCsalongside transient immunosuppression. ADSCs ofrecipient origin, however, have not been investi-gated for VCA. Given that human VCA donors aredeceased, recipient-origin ADSCs have a majoradvantage over any form of donor-derived MSCs inthat they can be harvested from the recipient andstored in advance of VCA in optimized form.

Syngeneic (from the same animal strain with thesame genetic background as the recipient) andautologous (from the recipient animal itself) ADSCsare equivalent in inbred animals with identical geno-type. We demonstrate for the first time that syngeneicADSCs are profoundly tolerogenic for VCA re-cipients, in that only one round of syngeneic ADSCsis sufficient for tolerance induction across a full MHCmismatch when given in conjunction with transientcyclosporine A (CsA) and anti-lymphocyte globulin(ALG). Additionally, it is demonstrated that VCA

tolerance was accompanied by the persistent elevationof CD4þCD25þFoxP3þ regulatory T cells (Tregs).

Methods

Animals

Male 8- to 12-week-old donor Brown-Norway (BN,RT1An) and recipient Lewis rats (LEW, RT1Al),representing a full MHC mismatch, were purchasedfrom the National Laboratory Animal Center,Taiwan. Rats were housed in pyrogen-free conditionsunder controlled temperature and lighting cycles withwater and commercial rat chow freely available at theChang Gung Memorial Hospital Animal Center. Allexperiments were conducted in accordance with theGuide for the Care and Use of Laboratory Animals of theNational Institutes of Health (Bethesda, MD, USA)and following the Institutional Animal Care and UseCommittee protocol authorized by Chang GungMemorial Hospital, Taiwan.

ADSC preparation

The inguinal fat pad from LEW rats was harvested ina sterile fashion, washed with phosphate-bufferedsaline, minced and digested with type IV collagenase(Life Technologies, Grand Island, NY, USA) andhyaluronidase (Sigma-Aldrich, St Louis, MO, USA)for 1 h at 37�C. After centrifugation, the supernatantwas discarded and the pelleted stromal vascularfraction containing ADSCs was resuspended andplated in stromal medium (low-glucose Dulbecco’smodified Eagle’s medium supplemented with 10%fetal bovine serum and 1% penicillin-streptomycin)and incubated at 37�C in 5% CO2 atmosphere. Themedium was replaced every 2e3 days, and cells with80e90% confluence were detached with the use oftrypsin/ethylenediaminetetraacetic acid (EDTA)(0.5%) and passed. ADSCs at the fourth passagewith 80e90% confluence were detached by lighttrypsinization and used for surface marker charac-terization, differentiation assays and in vitro immunefunction assessments. If these cells met the criteria ofbeing MSCs, they were infused into VCA recipients.

ADSC differentiation

For adipogenic differentiation, ADSCs at 80e90%confluence were cultured with low-glucose Dulbec-co’s Modified Eagle’s medium supplemented with10% fetal bovine serum, 0.5 mmol/L iso-butylmethylxanthine, 100 mmol/L indomethacin,1 mmol/L dexamethasone and 10 mg/mL insulinwith medium change every 3 days. After 21 days,cells were fixed with 10% formalin for 30 min, dried

Stem cells induce composite allotransplant tolerance 3

and stained with 0.5% oil red O. For osteogenicdifferentiation, ADSCs were induced with stromalmedium supplemented with 1 nmol/L dexametha-sone, 2 mmol/L glycerolphosphate and 50 mmol/Lascorbate-2-phosphate, with media change every 2e3days for 14 days. Cells were then fixed in 10%formalin, stained with 40 mmol/L Alizarin red (pH4.1) and inspected by means of light microscopyat 200� magnification.

Mixed lymphocyte reaction

Rat spleens were harvested in sterile conditions. Af-ter erythrocyte lysis, splenocytes were isolated andresuspended in complete RPMI-1640 medium. Inone-way mixed lymphocyte reaction (MLR), sple-nocytes from naive or tolerant LEW (1 � 105 cells/well) were co-cultured with either 2 mg/mL Conca-navalin A (ConA, Sigma-Aldrich) or allogeneic BNstimulator splenocytes that had been irradiated (2000cGy; Gammacell 1000 Elite Nordion International,Ottawa, ON, Canada). Other irradiated cellsincluding ADSCs (added with 1:1, 1:0.5, 1:0.2 and1:0.1 ratios to responder) or ADSC-conditionedsplenocytes (see below) were added according toexperimental design as described.

Cells were cultured in quadruplicate in 96-well,U-shaped plates for 4 days and were then pulsedwith 1 mCi/well 3H-thymidine (3H-TdR, PerkinElmer, Waltham, MA, USA) for 16 h and harvestedover glass fiber filters. Thymidine uptake was quan-tified on a microplate scintillation and luminescencecounter (Packard NXT, Meriden, CT, USA).Thymidine incorporation into spontaneously prolif-erating responders (in medium alone) was the con-trol and set as 1. Ratios of thymidine incorporationunder all other conditions with respect to the controlwere acquired, providing stimulation indices (SI).

ADSC-conditioned splenocytes

LEW or BN splenocytes were cocultured with LEWADSCs with equal numbers in 100-mm culture dishesat 37�C in 5% CO2 atmosphere for 5 days. Cells insuspension were collected carefully as the splenocytefraction. These cells were ADSC-conditioned spleno-cytes and were either directly subjected to flow cyto-metric analysis of Tregs or added in supplement tostimulators and responders at equivalent ratio (stim-ulator:responder:conditioned splenocytes, 1:1:1) inMLR after irradiation.

Flow cytometry

ADSC phenotyping. Cells were collected and ali-quoted to 106 cells per tube in fluorescence-activated

cell sorting (FACS) buffer (2 mmol/L EDTA, 0.5%bovine serum albumin in phosphate-buffered saline)and stained with antibodies for 30 min on ice. Afterwashing and resuspending in FACS buffer, cellswere analyzed by means of FACSCalibur flow cy-tometer (BD Biosciences, San Jose, CA, USA) withthe use of fluorochrome-conjugated antibodies asfollows: PerCP-Cy5.5-conjugated anti-CD90.1(eBioscience, San Diego, CA, USA), PE-conjugatedanti-CD79a (Abcam, Cambridge, MA, USA), APC-conjugated anti-CD11b/c (BioLegend, San Diego,CA, USA), fluorescein isothiocyanateeconjugatedanti-CD45 (BD Pharmingen, San Jose, CA, USA),APC-conjugated anti-CD29 (eBioscience). Cellswere stained with purified anti-CD73 (BD Phar-mingen) followed by biotin-labeled secondary anti-body and APC-conjugated streptavidin.

Peripheral blood immuno-profiling. Rat tail vein bloodwas collected for the analysis of Tregs. Red bloodcells were briefly lysed with ammonium-chloride-potassium buffer at room temperature. After centri-fugation and washing, white cells were stained withAPC-conjugated anti-CD4 (BD Pharmingen) andPE-conjugated anti-CD25 (BD Pharmingen)monoclonal antibodies for 30 min at room temper-ature. Cells were then permeabilized for 18 h at 4�Cwith the use of a permeabilization kit (eBiosecience)followed by staining with PerCP-Cy5.5econjugatedanti-FoxP3 (eBioscience) at room temperature for 1h. Cells were analyzed by FACSCanto II flow cy-tometer (BD Biosciences).

Rat model of VCA

A previously described heterotopic hind limb osteo-myocutaneous VCA model was used (22). Briefly,VCA harvest in the anesthetized donor rat beganwith a longitudinal medial hind limb incision fromankle to groin; this was extended to delineate the skinpaddle (4 � 3 cm). Proximal to the ankle the tendonswere cut, the tibial vessels were cauterized and thetibia was osteotomized. The superficial epigastricvessels were ligated, thigh muscles were sectionedand the femur was osteotomized. The flap was iso-lated on the femoral vessels below the inguinal liga-ment. The flap, including medullary cavities, wasflushed with 10 mL of heparinized saline, wrapped insaline gauze and placed on iced saline. The donor ratwas euthanized.

In recipient rats, 3-cm inguinal and gluteal skinincisions were performed to prepare the recipientfemoral vessels and delineate the recipient defect,respectively.TheVCAwas then inset and its circulationwas restored by 10-0 nylon microanastomoses. Theinguinal incision was closed, and the animal recovered.

4 H.-Y. Cheng et al.

Rats were divided into experimental and controlgroups as follows: group 1 was the syngeneic control(LEW donor and LEW recipient; no treatment).Allogeneic transplantations (BN donor and LEWrecipient) were performed in groups 2e6, and thefollowing treatments were administered: group 2 wasthe allogeneic control without treatment; group 3:ALG (days �1 and 10) and CsA (16 mg/kg, days0e10); group 4: ALG (days �1 and 10) and CsA(16 mg/kg, days 0e10) and ADSC (day 1; 2 � 106

per dose); group 5: ALG (days �1 and 10) and CsA(16 mg/kg, days 0e10) and total body irradiation(TBI) (200 cGy, day �1); group 6: ALG (days �1and 10) and CsA (16 mg/kg, days 0e10) and ADSC(day 1, 2 � 106 per dose) and TBI (200 cGy,day �1).

Follow-up

VCAs were evaluated daily with the use of anestablished semi-quantitative rejection grading sys-tem that ranges in severity from grade 0e4 as follows:grade 0, no rejection; grade 1, pink or slightlyerythematous; grade 2, frank erythema; grade 3, er-ythema or purplish discoloration with blister forma-tion or partial hair loss; and grade 4, dark-purplishdiscoloration with blister formation and major hairloss. Rejection was defined when 80% of the VCAreached grade 4 (23). At 150 days, recipients withsurviving VCAs were BN-skin grafted and monitoredfor another 60 days. At 210 days, recipients wereeuthanized and their splenocytes were assessed foralloreactivity to BN antigen by MLR. Histologicalchanges and lymphocyte infiltration were evaluatedby microscopy after hematoxylin and eosin staining.

Skin grafting

Dorsal cutaneous defects, superficial to the pan-niculus carnosus, were created in recipients withlong-term surviving VCA for insetting BN-originfull-thickness tail skin grafts (2 � 1 cm). These werefixed with tie-overs for 5 days, and successful graftswere evaluated daily for rejection for another 60days. Rejection was suggested if erythema, edema,scaling of the skin, hair loss, epidermolysis anddesquamation occurred; destruction of >80% ofthe graft defined rejection.

Histological examination

Skin biopsy specimens from accepted VCAs wereformalin-fixed, sectioned and stained with hema-toxylin and eosin (H&E) as standard for slidemounting and light microscopy (�200 magnifica-tion) for histological analyses.

Statistics

Data are expressed as mean � standard deviationunless otherwise indicated. Results from MLR andflow cytometry were analyzed by one-way analysisof variance and Tukey honestly significant differ-ences for post hoc pairwise comparison. MedianVCA survival with standard errors were acquiredby the product limit method of Kaplan-Meier andpresented by survival curves. A probability valueof <0.05 was regarded as statistically significant. Allstatistical analyses were conducted with the use ofSPSS software.

Results

ADSCs were immunosuppressive in vitro

Rat ADSCs displayed spindle-shaped and plastic-adherent phenotypes in culture (Figure 1A) anddifferentiated under specific conditions to mesen-chymal cell lineages such as adipocytes and osteo-cytes as shown, respectively, in Figure 1B,C (24).Surface markers of cultured ADSCs at the fourthpassage (P4) were analyzed by means of flow cytom-etry. As expected, ADSCs did not express CD45,CD11b/c and CD79a and positively expressedCD90.1, CD29 and CD73 (Figure 1D,E). Thesephenotypic features are consistent with MSC charac-teristics, as previously reported (25).

The immunosuppressive functions of rat ADSCswere evaluated by means of MLR, in whichLEW splenocytes served as responders and irradiatedBN splenocytes served as allogeneic stimulators.LEW splenocytes proliferated significantly whenco-cultured with stimulators and mitogenic ConAat 24.4-fold and 59.1-fold, respectively, whencompared with spontaneous proliferation (LEWsplenocytes in media only), as shown in Figure 2A,B.Adding syngeneic ADSCs suppressed proliferationof responder LEW splenocytes in a dose-relatedmanner against mitogen (Figure 2A) as well as BNantigen (Figure 2B) stimulations. When added at anequivalent cell ratio (ie, LEW: ADSC ratio of 1:1),mitogen-induced proliferation of responder cellswas almost completely suppressed (Figure 2A). Evenat much smaller ratios (when fewer ADSCs wereadded), responder proliferation against ConAwas still significantly reduced (Figure 2A). Thisdose-related trend was also observed under BNantigen-stimulated conditions (Figure 2B). WhenADSCs, responders and stimulators were co-cultured in equivalent ratios (ADSC:LEW:BN ratioof 1:1:1), responder stimulation was reduced by10.7-fold, again reducing proliferation down to levelsapproaching spontaneous proliferation.

Figure 1. Characterization of LEW ADSCs. (A) ADSCs displayed spindle-shaped, fibroblast-like phenotype and differentiated into (B)adipocytes (note oil red Oestained oil droplets) and (C) osteocytes (note calcium secretion stained by Alizarin red). (D) Expression ofsurface markers by ADSCs was characterized by means of flow cytometry with anti-CD45, CD79a, CD11b/c, CD90.1, CD73 and CD29antibodies (red lines: isotype controls; green lines: antibody-bound ADSCs). (E) Mean fluorescence intensity (MFI) and percentage ofpositive cells from three independent experiments were grouped as shown.

Stem cells induce composite allotransplant tolerance 5

ADSCs prolonged allograft survival and inducedtolerance to allotransplants

The in vivo effects of ADSCs on VCA were evalu-ated. Heterotopic hind limb osteomyocutaneousflaps were transplanted in 8- to 12-week-old malerats and divided into six groups. Treatment protocolsand median survival times for each group acquired

Figure 2. LEW ADSCs suppressed responder proliferation against (A)MLR. Cellular proliferations are expressed as SIs with respect to spont*P < 0.05 versus ConA (A), and mitogen (B) stimulation, respectively.

from the Kaplan-Meier test are as tabulated inTable I. Recipient LEW rats received syngeneictransplants (group 1) from LEW rats or allogeneictransplants (groups 2e6) from BN rats on day 0.Groups 1 and 2 did not receive additional treatment.Groups 3e6 were treated with 0.5 mL ALG intra-peritoneally on day �1 and 10 and daily subcutaneous

ConA and (B) alloantigen stimulation, respectively, in one-wayaneous proliferation. #P < 0.05 versus spontaneous proliferation;

Table I. Treatment protocols and median survival times for each group.

GroupTransplant

HOMC (day 0)TBI (cGy)(day �1) Immunosuppressive treatment

LEW ADSC(day 1)

RecipientNo.

No. of recipientswith VCAs that

survived long-term

Median survivaltime � standarderror (days)

1 LEW/LEW 0 No e 4 4 150 � 02 BN/LEW 0 No e 4 0 10 � 13 BN/LEW 0 ALG (0.5 mL, days �1, 10);

CsA (16 mg/kg, days 0e10)e 6 0 32 � 3.7

4 BN/LEW 0 ALG (0.5 mL, days �1, 10);CsA (16 mg/kg, days 0e10)

þ 6 4 150 � 0

5 BN/LEW 200 ALG (0.5 mL, days �1, 10);CsA (16 mg/kg, days 0e10)

e 6 2 49 � 8

6 BN/LEW 200 ALG (0.5 mL, days �1, 10);CsA (16 mg/kg, days 0e10)

þ 6 4 150 � 0

Figure 3. Treatment protocol for VCA recipients (groups 3e6).Group 3: ALG (days �1 and 10) and CsA (16 mg/kg, days 0e10);group 4: ALG (days �1 and 10) and CsA (16 mg/kg, days 0e10)and ADSC (day 1; 2 � 106 per dose); group 5: ALG (days �1 and10) and CsA (16 mg/kg, days 0e10) and TBI (200 cGy, day �1);group 6: ALG (days �1 and 10) and CsA (16 mg/kg, days 0e10)and ADSC (day 1, 2 � 106 per dose) and TBI (200 cGy, day �1).Groups 1 and 2 received no treatment and represented syngeneicand allogeneic controls, respectively.

6 H.-Y. Cheng et al.

CsA (16 mg/kg) from days 0 to 10. Groups 5 and 6received 200 cGy TBI on day �1. Groups 4 and 6received intravenous 2 � 106 LEW-origin ADSCs onday 1 (Figure 3).

All syngeneic transplants survived the wholeobservation period of 150 days without signs ofrejection, whereas group 2 (rejection control) ani-mals rejected VCAs between days 8 and 11. Groups3e6 showed significant prolongation of VCA survivalcompared with group 2. Treatment with immuno-suppressants (ALG and CsA) alone moderatelyprolonged VCA survival time from 10 to 32 days. Ingroup 5, preconditioning with 200 cGy TBI plusimmunosuppressants induced long-term VCA sur-vival in 33.3% of recipients. Infusion of ADSCs(groups 4 and 6) increased VCA acceptance rate to66.6%. The median transplant survival time withstandard errors for each group was acquired bymeans of Kaplan-Meier analysis and is shown inTable I. The presence of ADSCs dramatically pro-longed VCA survival from 32 days to beyond 150days. Interestingly, irradiation did not provide anyadditional benefit to VCA survival over ADSCs(Figure 4A).

To test whether transplantation tolerance haddeveloped, recipient LEW rats with long-term surviv-ing allotransplants received a second allogeneic chal-lenge in the form of skin grafts on postoperative day(POD) 150. BN skin survived completely and grewhair normally at 3 weeks after grafting (Figure 4C).Concurrent third-party skin grafts from Sprague-Dawley (SD) rats were rejected, and the residualwound healed secondarily with wound contraction,demonstrating that tolerance was donor (BN)-specific(Figure 4C). The VCAs exhibited no clinical changesafter skin grafting and developed no signs of rejection(Figure 4B) until POD 210. Microscopic examinationof H&E-stained skin of the original VCAs showed nosigns of inflammation (Figure 4D).

Tolerance was assessed further by means ofMLR in vitro. Splenocytes from three recipients

with long-term surviving VCAs were re-stimulatedwith BN splenocytes at POD 210, followed by eval-uating proliferation with 3H-thymidine incorpora-tion. Splenocytes from all three rats exhibitedsignificantly lower proliferation compared with thosefrom naive Lewis rats when challenged with BNsplenocytes (Figure 4E). This indicates that theserecipients with long-term surviving VCAs lackeddonor-specific allo-response despite repeated allo-geneic stimulation in vivo (skin graft) as well as invitro (MLR), confirming that they had developedimmunological tolerance to BN alloantigen.

ADSCs promoted the generation of CD4þCD25þFoxP3þ

regulatory T cells from splenocytes in vitro

To characterize how LEW ADSCs induced immu-nological tolerance to BN antigen in LEW recipients,an in vitro co-culture system assessing the effects ofLEW ADSC-preconditioned splenocytes on theproliferation of LEW responder splenocytes againstBN stimulator splenocytes, was used. As shown in

Figure 4. ADSCs induced VCA tolerance. (A) Kaplan-Meier VCA survival curves for control and experimental groups. (B) RepresentativeBN VCA of a tolerized LEW recipient in group 4 at post-VCA day 210. (C) Representative accepted skin graft from BN (upper) and rejectedskin graft from SD rat (lower) of a LEW recipient that has already long-term accepted a BN VCA in group 4. (D) No histological evidence ofrejection was observed in the VCA skin of a LEW recipient in group 4 at POD 210 (200� magnification; H&E stained). (E) In response toBN antigen, the responder splenocytes from three VCA-tolerant LEW rats proliferated significantly less than did the splenocytes derivedfrom naive LEW in one-way MLR. Cellular proliferations were expressed as SIs with respect to spontaneous proliferation. #P < 0.05 versusspontaneous proliferation; *P < 0.05 versus BN alloantigenestimulated proliferation, respectively.

Stem cells induce composite allotransplant tolerance 7

Figure 5, ADSC-preconditioned splenocytes, eitherfrom BN or LEW, had suppressive effects on BNantigen-stimulated proliferation of LEW splenocytes.It suggested that ADSCs induced a suppressivephenotypic change in splenocytes during co-culture.It was previously shown that ADSCs in vitro inducedthe generation of CD4þCD25þFoxP3þ Tregs, whichare known to be critical in self-tolerance as well astransplantation tolerance (26). Therefore, flowcytometry was used to assess the effects of ADSCs oncellular composition, specifically Tregs, in the sple-nocyte culture. As demonstrated in Figure 6, LEWADSCs considerably increased the percentage ofTregs among CD4þ lymphocytes when incubatedwith LEW splenocytes, BN splenocytes or cocul-tured LEW and BN splenocytes (Figure 6). Synge-neic co-culture of LEW splenocytes with LEWADSCs increased the percentage of Tregs from 1.2to 6.2% (Figure 6) among CD4þ lymphocytes.Additionally, the potential of LEW ADSCs to induceTregs was preserved across histocompatibility bar-riers for BN splenocytes as well. Specifically, LEWADSCs induced the increase of Tregs in CD4þ cellsfrom 4.3 to 7.4% (Figure 6). These Tregs, at leastpartially, are responsible for the suppressive function

of ADSC-conditioned splenocytes (either from BNor LEW) on BN alloantigen-elicited proliferation ofLEW splenocytes, as shown in Figure 5. Interest-ingly, although co-culture of BN and LEW spleno-cytes induced Tregs from 4.3% (BN splenocytealone) and 1.2% (LEW splenocytes alone) to 10.9%,the presence of LEW ADSCs enhanced these effectseven further to 24.4% (Figure 6).

The levels of Tregs in VCA recipients were alsoevaluated. As shown in Figure 7A, group 4 recipientswith long-term surviving VCAs were assessed forTreg levels after their allografts had survived for>200 days; they had significantly higher levels ofcirculating Tregs compared with naive rats. In therejection group (group 3, ALG/CsA), Tregs weremeasured prior to rejection onset; they were at asignificantly lower level than that of group 4. In long-term surviving VCA recipients, peripheral Tregs at1 month after VCA were at levels similar to those innaive rats but then gradually rose and were main-tained at a higher steady level (Figure 7B).

Thus, ADSCs from the recipient LEW straindisplayed potent in vitro immunosuppressive effectspotentially through the induction of suppressiveTregs in the co-culture. Furthermore, ADSCs

Figure 5. ADSC-conditioned splenocytes inhibit alloantigen-stimulated proliferation. ADSC-conditioned cells were preparedby co-culturing splenocytes (LEW or BN as indicated) with LEWADSCs followed by isolation and irradiation. The addition ofADSC-conditioned splenocytes halved one-way MLR prolifera-tion of naive LEW splenocytes against BN antigen. Cellular pro-liferations were expressed as SIs with respect to spontaneousproliferation. *P < 0.05 compared with LEW spontaneousproliferation.

8 H.-Y. Cheng et al.

induced donor-specific VCA tolerance whenadministered alongside ALG and a short course ofCsA. Elevated levels of circulating Tregs weredemonstrated and may be critical for maintaining thetolerant state.

Discussion

Although more than 100 patients have benefitedfrom various VCAs since the late 1990s, the risks oflife-long immunosuppression hamper the wider

Figure 6. Increased proportion of CD4þCD25þFoxP3þ Tregs in splenoccultured alone or with ADSCs as labeled above each graph for 5 days anCD4, followed by CD25 and FoxP3. The percentage of CD25þFoxP3þ cupper right corner.

application of VCAs because, unlike most solid or-gan transplants, they are not life-saving. Conse-quently, transplantation tolerance is a major focus inVCA research because its safe induction and main-tenance eliminate the requirement for harmful non-specific immunosuppression (27). Among thenumerous experimental strategies that have beeninvestigated (7), donor non-vascularized BMT plusimmunosuppressants has proven to be a reliablemethod to induce VCA tolerance (28,29). Vascu-larized BMT (VBMT), however, provides additionaladvantages such as instant engraftment and a naturalstromal-supporting microenvironment (30). We pre-viously reported that the combination of VBMT,non-myeloablative TBI and a short course of immu-nosuppression (ALG and FK-506 or CsA) inducedrobust donor-specific tolerance in a rat hind limbVCAmodel (31). In the current study, we show that asingle dose of syngeneic ADSCs together with ALGand transient CsA induced donor-specific tolerancein 66% of VCA recipients across a full MHC barrier(group 4). Interestingly, the addition of 200 cGy ofirradiation did not provide any survival benefit(group 6). However, the absence of ADSCs and theprovision of 200 cGy irradiation in the inductionregimen halved the reliability of induction and/ormaintenance of long-term allograft survival (group 5).

We also demonstrate that the induction ofCD4þCD25þFoxP3þ Tregs may play an importantrole in ADSC-mediated tolerance induction. In vitro,

ytes co-cultured with ADSCs. Splenocytes from LEW or BN wered then were analyzed by flow cytometry. Cells were first gated withells among CD4þ cells from the lymphoid gate is indicated at each

Figure 7. Increased level of Treg in VCA-tolerant LEW recipients. (A) Percentage of Tregs among peripheral lymphocytes in VCA-tolerantLEW recipients (survival), VCA-rejecting LEW recipients (rejection) and untreated LEW (naive). *P < 0.05; NS indicates no significantdifference. (B) The proportion of Tregs among peripheral lymphocytes in VCA-tolerant LEW gradually increased between 27 and 41 daysafter VCA to a level that was then maintained beyond 200 days. *P < 0.05 compared with the level at POD 27. The red dotted line indicatesthe level of Tregs in naive LEW. The blue bar represents the level of Tregs in VCA-rejecting recipients at the average rejection time point of33 days. Error bars denote standard deviation.

Stem cells induce composite allotransplant tolerance 9

ADSCs dramatically suppressed mitogen-inducedas well as alloantigen-induced proliferation of sple-nocytes. Hence, such effects may be explained bythe preferential differentiation of splenocytes toTregs mediated by ADSCs. Interestingly, ADSCsdisplayed similar Treg-inducing effects towardsyngeneic as well as allogeneic splenocytes, whichsuggests that some constitutively expressed factorsby ADSCs may participate in the process. This issupported by an earlier report that constitutivesecretion of transforming growth factor-b andprostaglandin-2 by ADSCs plays a key role in theinduction of Tregs (32). It will be of interest tofurther investigate if such factors also take part inthe induction of Treg by ADSCs in our current co-culture model.

ADSCs exerted similar immunosuppressive ef-fects in vivo. At 4 weeks after VCA, circulatingTregs in VCA recipients resembled those in naiverats. In VCA-tolerant recipients, circulating Tregsgradually doubled and maintained this higher levellong-term. In contrast, recipients that rejected VCAdid not demonstrate this evolution of Treg levels,and, indeed, Treg levels were not statisticallydifferent when compared with naive rats. Accord-ingly, higher levels of Tregs potentially are criticalfor inducing and maintaining tolerance in thismodel. They may patrol in the circulation orlater be recruited to the allograft to sustain hypo-alloreactivity locally. The latter behavior wouldreflect the recruitment of VCA-protective Tregs tothe VCA/recipient interface and/or throughout theallograft (33e35). Ge et al. (15) reported a signifi-cant elevation in intragraft FoxP3þ cells after MSCtreatment, which also suggests intragraft recruit-ment of Tregs for solid organ transplants. Further-more, FoxP3þ Tregs were enriched in the skin ofbilateral hand VCAs 6 years after transplantation ina human case report (36). Potential targets for

Tregs include both residual alloresponsive T-cellsand new thymic emigrants (37). Interestingly, theevolution of Treg levels in the current study differsfrom those reported by Kuo et al.(21), who reportedcomparatively higher increases of peripheral Tregsat 4 weeks after VCA, followed by a persistent andeventual decrease in Tregs to naive levels (21).Because the intra-VCA Treg recruitment was crit-ical to tolerance induction and/or maintenance(35,38), it would appear that the kinetic differencesin circulating Tregs between our findings and thoseof Kuo et al. (21) might be explained, first, by adifference in the rapidity of intragraft recruitmentfrom the periphery, and second, by the duration ofthe phase of Treg induction by the treatment. Thespecific mechanisms are as yet unclear and furtherinvestigation is underway to characterize thesephenomena more rigorously. It is also noteworthythat in human studies, ADSCs were reported toexpress HLA-G molecules, which were suggested toparticipate in tolerance induction for organ trans-plant recipients (39e41). These findings furtherconsolidate the clinical potential to apply ADSC forinducing allotransplant tolerance.

Stem cells have been shown also to exert theirimmunomodulatory effects on other cell pop-ulations, such as B cells, dendritic cells and naturalkiller cells, in addition to Tregs (42,43). Such me-chanisms may have influenced induction or main-tenance of tolerance to VCA in our experimentalsystem, as implicated by the different behaviorsamong splenocytes isolated from the three tolerantrecipients in MLR (Figure 4E). Although the allo-grafts had no signs of rejection, splenocytes from rat3 showed higher proliferation potential against BNantigens, suggesting that T lymphocytes (the prin-ciple population assessed by MLR) may not be theonly cells responsible for tolerance. Studies on theroles of different cell populations in induction and/or

10 H.-Y. Cheng et al.

maintenance of VCA tolerance are worthwhile andcurrently underway.

MSCs used for transplantation tolerance induc-tion are almost exclusively of donor origin thusfar. Although earlier reports suggested that MSCsdo not express co-stimulators such as CD80, CD86and CD40 that are important for alloreactivity(13,18,19,21,44), recent studies have provided evi-dence indicating that MSCs also retain some degreeof immunogenicity in certain circumstances that maylimit their longevity and attenuate their beneficialeffects (45). For example, Huang et al. found thatallogeneic MSCs were eliminated from the hostearlier than were syngeneic MSCs (46). Our currentresult indicates that, together with transient immu-nosuppressants, robust induction of tolerance can beobtained for MHC-mismatched VCA from just onedose of syngeneic stem cells, contrasting with earlierstudies that relied on multiple doses of allogeneicstem cells (18,19,21,42). We hypothesize that thiscan be explained, at least in part, by the longer lifespans of syngeneic versus allogeneic MSCs in thehost. This hypothesis, if proven correct, would alsolend explanation to the extended duration ofCD4þCD25þFoxP3þ Treg elevations found in thecirculation. We plan to track the fate of infused stemcells in vivo to characterize their destination(s), ki-netic behaviors and longevity as tolerance is inducedand maintained.

It has also been demonstrated in vivo that allo-geneic MSCs may transit from an immunoprivilegedto an immunogenic state after differentiation (46).Beggs et al. (47) and Poncelet et al. (48) reported thatallo-antibodies reacting to donor cells were inducedby MSC infusions in baboons and pigs, respectively,even though MSCs showed low immunogenicity invitro. In vivo immunogenicity of donor-derivedMSCs raises uncertainty over their future clinicalapplicability. Nevertheless, recipient-derived ADSCsprovide the distinct advantage that they are easilyaccessible by liposuction (49,50) and amenable toisolation, storage and optimization to intensify theirtherapeutic benefits in advance of transplantation.Additionally, it may be possible to pre-treat a VCArecipient with autologous ADSCs because they arenot donor-dependent. These factors, as well as thestrong effects that recipient strain ADSC infusionexerted in this study, suggest that this system oftolerization may have an important position in futureclinical VCA practice. In fact, a recent clinical studydemonstrated the safety and feasibility of adminis-tering autologous MSC after kidney transplantation.A gradual increase of peripheral Tregs was alsodocumented for these patients (51).

Although only one infusion of autologous ADSCswith transient immunosuppression produced 66%

VCA tolerance, there remains room for improvementof the protocol to reliably obtain tolerance (ie,100%). Dosing for both MSCs and transient im-munosuppressants, in particular, probably willbenefit from further optimization (such as timings,concentrations, frequency and number of adminis-trations), as will the choice of immunosuppressant(s)(52,53). We have, for example, previously demon-strated the superiority of CsA over tacrolimus whentolerizing recipients of concurrent VCA/VBMTthat had been preconditioned with low-dose irra-diation, supporting the notion that the choice ofimmunosuppressant may facilitate, or indeedhinder, the benefits of stem cells or the cells thatthey invoke (31). Similarly, rapamycin appears tobe superior to calcineurin inhibitors at inducingCD4þCD25þFoxP3þ Tregs in solid organ trans-plant recipients (54,55). The final outcome ofVCA, be it rejection or tolerance, is determined byan intricate balance of a multitude of events be-tween numerous cell types derived from donor and/or recipient. It will be worthwhile to finely tunethese potential factors to maximize VCA survivaland the reliability of tolerization protocols.

Acknowledgments

We acknowledge the financial support of the Na-tional Science Council, Taiwan, ROC (NSC97-2314-B-182-022-MY3), the Department of Health,Taiwan, ROC (DOH99-TD-C-111-008), andChang Gung Memorial Hospital, Taiwan, ROC(CMRPG3A0421 and CMRPG3A0431).

Disclosure of interests: The authors have nocommercial, proprietary, or financial interest in theproducts or companies described in this article.

References

1. Siemionow M, Ozturk C. An update on facial transplantationcases performed between 2005 and 2010. Plast Reconstr Surg.2011;128:707ee20e.

2. Shores JT, Imbriglia JE, Lee WP. The current state of handtransplantation. J Hand Surg Am. 2011;36:1862e7.

3. KaufmanCL, BreidenbachW.World experience after more thana decade of clinical hand transplantation: update from the Louis-ville hand transplant program. Hand Clin. 2011;27:417e21.

4. Lee WP, Yaremchuk MJ, Pan YC, Randolph MA, Tan CM,Weiland AJ. Relative antigenicity of components of a vascu-larized limb allograft. Plast Reconstr Surg. 1991;87:401e11.

5. Schneeberger S, Landin L, Jableki J, Butler P, Hoehnke C,Brandacher G, et al. Achievements and challenges in com-posite tissue allotransplantation. Transpl Int. 2011;24:760e9.

6. SiemionowM. Impact of reconstructive transplantation on thefuture of plastic and reconstructive surgery. Clin Plast Surg.2012;39:425e34.

Stem cells induce composite allotransplant tolerance 11

7. Page EK, Dar WA, Knechtle SJ. Tolerogenic therapies intransplantation. Front Immunol. 2012;3:198.

8. Franquesa M, Hoogduijn MJ, Baan CC. The impact ofmesenchymal stem cell therapy in transplant rejection andtolerance. Curr Opin Organ Transplant. 2012;17:355e61.

9. Cui L, Yin S, Liu W, Li N, Zhang W, Cao Y. Expandedadipose-derived stem cells suppress mixed lymphocyte reac-tion by secretion of prostaglandin E2. Tissue Eng. 2007;13:1185e95.

10. Djouad F, Charbonnier LM, Bouffi C, Louis-Plence P,Bony C, Apparailly F, et al. Mesenchymal stem cells inhibitthe differentiation of dendritic cells through an interleukin-6-dependent mechanism. Stem Cells. 2007;25:2025e32.

11. Choi YS, Jeong JA, Lim DS. Mesenchymal stem cell-mediated immature dendritic cells induce regulatory T cell-based immunosuppressive effect. Immunol Invest. 2012;41:214e29.

12. Spaggiari GM, Capobianco A, Becchetti S, Mingari MC,Moretta L. Mesenchymal stem cell-natural killer cell in-teractions: evidence that activated NK cells are capable ofkilling MSCs, whereas MSCs can inhibit IL-2-induced NK-cell proliferation. Blood. 2006;107:1484e90.

13. Casiraghi F, Azzollini N, Cassis P, Imberti B, Morigi M,Cugini D, et al. Pretransplant infusion of mesenchymal stemcells prolongs the survival of a semiallogeneic heart transplantthrough the generation of regulatory T cells. J Immunol.2008;181:3933e46.

14. Ge W, Jiang J, Baroja ML, Arp J, Zassoko R, Liu W, et al.Infusion of mesenchymal stem cells and rapamycin synergizeto attenuate alloimmune responses and promote cardiacallograft tolerance. Am J Transplant. 2009;9:1760e72.

15. Ge W, Jiang J, Arp J, Liu W, Garcia B, Wang H. RegulatoryT-cell generation and kidney allograft tolerance induced bymesenchymal stem cells associated with indoleamine 2,3-dioxygenase expression. Transplantation. 2010;90:1312e20.

16. Ito T, Itakura S, Todorov I, Rawson J, Asari S, Shintaku J,et al. Mesenchymal stem cell and islet co-transplantationpromotes graft revascularization and function. Trans-plantation. 2010;89:1438e45.

17. Sbano P, Cuccia A, Mazzanti B, Urbani S, Giusti B, Lapini I,et al. Use of donor bone marrow mesenchymal stem cells fortreatment of skin allograft rejection in a preclinical rat model.Arch Dermatol Res. 2008;300:115e24.

18. Kuo YR, Goto S, Shih HS, Wang FS, Lin CC, Wang CT,et al. Mesenchymal stem cells prolong composite tissue allo-transplant survival in a swine model. Transplantation. 2009;87:1769e77.

19. Kuo YR, Chen CC, Goto S, Huang YT, Wang CT, Tsai CC,et al. Immunomodulatory effects of bone marrow-derivedmesenchymal stem cells in a swine hemi-facial allo-transplantation model. PLoS One. 2012;7:e35459.

20. Izadpanah R, Trygg C, Patel B, Kriedt C, Dufour J,Gimble JM, et al. Biologic properties of mesenchymal stemcells derived from bone marrow and adipose tissue. J CellBiochem. 2006;99:1285e97.

21. Kuo YR, Chen CC, Goto S, Lee IT, Huang CW, Tsai CC,et al. Modulation of immune response and T-cell regulationby donor adipose-derived stem cells in a rodent hind-limballotransplant model. Plast Reconstr Surg. 2011;128:661ee72e.

22. Adamson LA, Huang WC, Breidenbach WC, Rahhal D,Xu H, Huang Y, et al. A modified model of hindlimb osteo-myocutaneous flap for the study of tolerance to compositetissue allografts. Microsurgery. 2007;27:630e6.

23. Zdichavsky M, Jones JW, Ustuner ET, Ren X, Edelstein J,Maldonado C, et al. Scoring of skin rejection in a swinecomposite tissue allograft model. J Surg Res. 1999;85:1e8.

24. Bunnell BA, Flaat M, Gagliardi C, Patel B, Ripoll C.Adipose-derived stem cells: isolation, expansion and differ-entiation. Methods. 2008;45:115e20.

25. Schaffler A, Buchler C. Concise review: adipose tissue-derived stromal cells: basic and clinical implications for novelcell-based therapies. Stem Cells. 2007;25:818e27.

26. Sakaguchi S. Naturally arising Foxp3-expressing CD25þCD4þ

regulatory T cells in immunological tolerance to self and non-self. Nat Immunol. 2005;6:345e52.

27. Petruzzo P, Testelin S, Kanitakis J, Badet L, Lengele B,Girbon JP, et al. First human face transplantation: 5 yearsoutcomes. Transplantation. 2012;93:236e40.

28. Prabhune KA, Gorantla VS, Maldonado C, Perez-Abadia G,Barker JH, Ildstad ST. Mixed allogeneic chimerism andtolerance to composite tissue allografts. Microsurgery. 2000;20:441e7.

29. Wachtman GS, Wimmers EG, Gorantla VS, Lin CH,Schneeberger S, Unadkat JV, et al. Biologics and donor bonemarrow cells for targeted immunomodulation in vascularizedcomposite allotransplantation: a translational trial in swine.Transplant Proc. 2011;43:3541e4.

30. Gordon CR, Tai CY, Suzuki H, Strande LF, Ramsamooj R,Matthews MS, et al. Review of vascularized bone marrowtransplantation: current status and future clinical applications.Microsurgery. 2007;27:348e53.

31. Huang WC, Liao SK, Wallace CG, Chang NJ, Lin JY,Wei FC. Greater efficacy of tolerance induction with cyclo-sporine versus tacrolimus in composite tissue allotransplantswith less myeloablative conditioning. Plast Reconstr Surg.2011;127:1141e8.

32. Engela AU, Baan CC, Dor FJ, Weimar W, Hoogduijn MJ. Onthe interactions between mesenchymal stem cells and regu-latory T cells for immunomodulation in transplantation.Front Immunol. 2012;3:126.

33. Bozulic LD, Wen Y, Xu H, Ildstad ST. Evidence that FoxP3þ

regulatory T cells may play a role in promoting long-termacceptance of composite tissue allotransplants. Trans-plantation. 2011;91:908e15.

34. Pilat N, Baranyi U, Klaus C, Jaeckel E, Mpofu N, Wrba F,et al. Treg-therapy allows mixed chimerism and trans-plantation tolerance without cytoreductive conditioning. Am JTransplant. 2010;10:751e62.

35. Lin JY, Tsai FC, Wallace CG, Huang WC, Wei FC, Liao SK.Combined treatment with regulatory T cells and vascularizedbone marrow transplantation creates mixed chimerism andinduces donor-specific tolerance to vascularized compositeallografts without cytoreductive conditioning. J Surg Res.2012;178:974e81.

36. Eljaafari A, Badet L, Kanitakis J, Ferrand C, Farre A,Petruzzo P, et al. Isolation of regulatory T cells in the skin of ahuman hand-allograft, up to six years posttransplantation.Transplantation. 2006;82:1764e8.

37. Graca L, Chen TC, Le Moine A, Cobbold SP, Howie D,Waldmann H. Dominant tolerance: activation thresholds forperipheral generation of regulatory T cells. Trends Immunol.2005;26:130e5.

38. Lin JY, Tsai FC, Wallace CG, Huang WC, Wie FC, Liao SK.Combined treatment with regulatory T cells and vascularizedbone marrow transplantation creates mixed chimerism andinduces donor-specific tolerance to vascularized compositeallografts without cytoreductive conditioning. Chimerism.2012;4:20e2.

39. Yang HM, Sung JH, Choi YS, Lee HJ, Roh CR, Kim J,et al. Enhancement of the immunosuppressive effectof human adipose tissue-derived mesenchymal stromalcells through HLA-G1 expression. Cytotherapy. 2012;14:70e9.

12 H.-Y. Cheng et al.

40. Selmani Z, Naji A, Zidi I, Favier B, Gaiffe E, Obert L, et al.Human leukocyte antigen-G5 secretion by human mesen-chymal stem cells is required to suppress T lymphocyte andnatural killer function and to induceCD4þCD25highFOXP3þ

regulatory T cells. Stem Cells. 2008;26:212e22.41. Zarkhin V, Talisetti A, Li L, Wozniak LJ, McDiarmid SV,

Cox K, et al. Expression of soluble HLA-G identifies favor-able outcomes in liver transplant recipients. Transplantation.2010;90:1000e5.

42. Nauta AJ, Fibbe WE. Immunomodulatory properties ofmesenchymal stromal cells. Blood. 2007;110:3499e506.

43. Machado Cde V, Telles PD, Nascimento IL. Immunologicalcharacteristics of mesenchymal stem cells. Rev Bras HematolHemoter. 2013;35:62e7.

44. Kuo YR, Chen CC, Shih HS, Goto S, Huang CW, Wang CT,et al. Prolongation of composite tissue allotransplant survivalby treatment with bone marrow mesenchymal stem cells iscorrelated with T-cell regulation in a swine hind-limb model.Plast Reconstr Surg. 2011;127:569e79.

45. Griffin MD, Ritter T, Mahon BP. Immunological aspects ofallogeneic mesenchymal stem cell therapies. Hum Gene Ther.2010;21:1641e55.

46. Huang XP, Sun Z, Miyagi Y, McDonald Kinkaid H,Zhang L, Weisel RD, et al. Differentiation of allogeneicmesenchymal stem cells induces immunogenicity and limitstheir long-term benefits for myocardial repair. Circulation.2010;122:2419e29.

47. Beggs KJ, Lyubimov A, Borneman JN, Bartholomew A,Moseley A, Dodds R, et al. Immunologic consequences ofmultiple, high-dose administration of allogeneic mesenchymalstem cells to baboons. Cell Transplant. 2006;15:711e21.

48. Poncelet AJ, Vercruysse J, Saliez A, Gianello P. Although pigallogeneic mesenchymal stem cells are not immunogenic invitro, intracardiac injection elicits an immune response invivo. Transplantation. 2007;83:783e90.

49. Gir P, Oni G, Brown SA, Mojallal A, Rohrich RJ. Humanadipose stem cells: current clinical applications. PlastReconstr Surg. 2012;129:1277e90.

50. Siemionow M, Madajka M, Cwykiel J. Application of cell-based therapies in facial transplantation. Ann Plast Surg.2012;69:575e9.

51. Perico N, Casiraghi F, Introna M, Gotti E, Todeschini M,Cavinato RA, et al. Autologous mesenchymal stromal cellsand kidney transplantation: a pilot study of safety and clinicalfeasibility. Clin J Am Soc Nephrol. 2011;6:412e22.

52. Eggenhofer E, Renner P, Soeder Y, Popp FC, Hoogduijn MJ,Geissler EK, et al. Features of synergism between mesen-chymal stem cells and immunosuppressive drugs in a murineheart transplantation model. Transpl Immunol. 2011;25:141e7.

53. Popp FC, Eggenhofer E, Renner P, Slowik P, Lang SA,Kaspar H, et al. Mesenchymal stem cells can induce long-term acceptance of solid organ allografts in synergy with low-dose mycophenolate. Transpl Immunol. 2008;20:55e60.

54. Chu Z, Zhang J, Zhao Y, Ji Q, Zhong J, Zhang C, et al. In-fluence of immunosuppressive drugs on the development ofCD4(þ)CD25(high) Foxp3(þ) T cells in liver transplant re-cipients. Transplant Proc. 2010;42:2599e601.

55. Lu L, Qian XF, Rao JH, Wang XH, Zheng SG, Zhang F.Rapamycin promotes the expansion of CD4(þ) Foxp3(þ)regulatory T cells after liver transplantation. Transplant Proc.2010;42:1755e7.