BIOLOGY OF REPRODUCTION 74, 46–57 (2006)Published online before print 14 September 2005.DOI 10.1095/biolreprod.105.045138

Isolation, Characterization, Gene Modification, and Nuclear Reprogrammingof Porcine Mesenchymal Stem Cells1

Pablo Bosch,3 Scott L. Pratt,4 and Steven L. Stice2,3

Department of Animal and Dairy Science,3 University of Georgia, Athens, Georgia 30602ViaGen, Inc.,4 Austin, Texas 78727

ABSTRACT

Bone marrow mesenchymal stem cells (MSCs) are adultpluripotent cells that are considered to be an important resourcefor human cell-based therapies. Understanding the clinicalpotential of MSCs may require their use in preclinical large-animal models, such as pigs. The objectives of the present studywere 1) to establish porcine MSC (pMSC) cultures; 2) tooptimize in vitro pMSC culture conditions, 3) to investigatewhether pMSCs are amenable to genetic manipulation, and 4) todetermine pMSC reprogramming potential using somatic cellnuclear transfer (SCNT). The pMSCs isolated from bone marrowgrew, attached to plastic with a fibroblast-like morphology, andexpressed the mesenchymal surface marker THY1 but not thehematopoietic marker ITGAM. Furthermore, pMSCs underwentlipogenic, chondrogenic, and osteogenic differentiation whenexposed to specific inducing conditions. The pMSCs grew well ina variety of media, and proliferative capacity was enhanced byculture under low oxygen atmosphere. Transient transduction ofpMSCs and isogenic skin fibroblasts (SFs) with a humanadenovirus carrying the gene for green fluorescent protein(GFP; Ad5-F35eGFP) resulted in more pMSCs expressing GFPcompared with SFs. Cell lines with stable genetic modificationsand extended expression of transgene were obtained whenpMSCs were transfected with a plasmid containing the GFPgene. Infection of pMSC and SF cell lines by an adeno-associatedvirus resulted in approximately 12% transgenic cells, whichformed transgenic clonal lines after propagation as single cells.The pMSCs can be expanded in vitro and used as nuclear donorsto produce SCNT embryos. Thus, pMSCs are an attractive celltype for large-animal autologous and allogenic cell therapymodels and for SCNT transgenesis.

developmental biology

INTRODUCTION

Mesenchymal stem cells (MSCs) are pluripotent precursorcells that localize to the stromal compartment of the bonemarrow (BM), where they support hematopoiesis and differ-entiate into mesenchymal lineages. The potential of MSCs toform bone, cartilage, and adipose tissues both in vivo [1–3] andin vitro [4] has been well documented. Their plasticity,however, is not limited to those mesenchymal derivatives.Recent reports have suggested that MSCs can differentiate intoneurons [5, 6], myoblasts [7], and cardiomyocytes [8]. Cells

with features of mesenchymal precursors have been isolatedfrom the BM of many mammals, including laboratory rodents[9, 10], humans [4], cats [11], dogs [12] and pigs [13]. TheMSCs from all species studied to date proliferate in vitro asadherent fibroblastic cells, a feature that has been exploited toenrich MSCs from hematopoietic cells that normally remain insuspension. In humans, pluripotent stem cells derived frommarrow stroma proliferate ex vivo to form a phenotypicallyhomogeneous population of cells that express several surfacemarkers, such as THY1 (also known as CD90), CD44, andTFRC (also known as CD71), but that do not express thehematopoietic markers PTPRC (also known as CD45) andITGAM (also known as CD11b) [4]. Like MSCs from otherspecies, porcine MSCs (pMSCs) were capable of growing andattaching to plastic with a fibroblast-like morphology and thendifferentiating into adipose, bone, and cartilage tissues in vitro[13]. To our knowledge, however, surface marker expressionand culture requirements for ex vivo expansion of MSCs in thisspecies have not been yet defined.

Because of the ability of MSCs to proliferate extensively exvivo while maintaining their pluripotent differentiation capa-bilities (in vivo and in vitro), they are regarded as a particularlyattractive cell type for cell-based therapies in humans. Ofparticular interest is the use of intact or genetically engineeredMSCs for the treatment of skeletal disorders like osteogenesisimperfecta [14, 15]. Moreover, MSCs have attracted muchattention as tools for targeted delivery of anticancer agents intotumors [16, 17]. Before human clinical trials are approved,scaled-up cell production and delivery into a large-animalmodel in which cell doses (number of cells) comparable tothose that will be used in human trials often is required tosatisfy regulatory safety concerns. Beyond safety issues, thereprogramming of pMSCs via somatic cell nuclear transfer(SCNT) lays the foundation for future isogenic comparisonsbetween adult pMSCs and reprogrammed embryonic cellsources (therapeutic cloning) in porcine disease models. Katoet al. [18] recently reported the birth of the first calf originatedfrom a bovine MSC, demonstrating that bovine MSCs can bereprogrammed to drive term development after SCNT.

Development of SCNT has provided a new and faster wayto create transgenic animals. It now is possible to introducegenetic modifications in cultured cells that later can be used asdonor cells to produce cloned animals bearing the genetictransformation (for review, see [19, 20]). Genetic manipu-lations of cultured cells can range from simple, randomintegration of the gene of interest to targeted homologousrecombination to abolish (knock out) or modulate genefunction. In combination with electroporation or a particulartransfection compound, DNA plasmids have been used togenetically modify cultured cells for use in nuclear transfer(NT) [21, 22], and transgenic animals have been generated withthese cells [23, 24]. In addition, transient expression of proteinsin donor cells could open new opportunities, from basic studiesof donor cell physiology and nuclear reprogramming to more

1Supported by ViaGen, Inc.2Correspondence: Steven L. Stice, Edgar L. Rhodes Center for Animaland Dairy Science, University of Georgia, Athens, GA 30602-2771.FAX: 706 542 7925; e-mail: sstice@uga.edu

Received: 30 June 2005.First decision: 23 July 2005.Accepted: 7 September 2005.� 2006 by the Society for the Study of Reproduction, Inc.ISSN: 0006-3363. http://www.biolreprod.org

46

applied studies aimed to improve cloning efficiencies byconditioning donor cells before NT.

The use of viruses as vectors has emerged as a promisingalternative to the classic, mechanical methods of gene delivery.The ability of retroviruses to integrate randomly into the hostgenome has been exploited to stably introduce the greenfluorescent protein (GFP) reporter gene in pig cell lines laterused to produce embryos [25] and transgenic cloned pigs bySCNT [26, 27]. Lentivirus, which is a complex retrovirus, isconsidered to be a promising alternative to the originaloncogenic retroviral vectors because of their ability to infectnondividing mammalian cells and to resist methylation-de-pendent gene silencing. Lentiviral infection of bovine fibroblastsfollowed by SCNT has resulted in the production of transgenicanimals [28].

Adeno-associated virus (AAV) is an integrating, nonpatho-genic human virus that requires coinfection with a helper virus,such as adenovirus or herpesvirus for productive infection. In theabsence of a helper virus, AAV integrates in a site-specificmanner in the host genome, where it remains as latent infection.Vectors derived from AAV are attractive candidates fortransgenesis by virtue of their nonpathogenicity, integrationcapability, infectivity of dividing and nondividing cells, andability to infect a wide variety of cell types. The AAV vectorshave been used to insert small (,20 bp) and large (.1 kb)transgenes by homologous recombination in human cells inculture [29]. More recently, Hirata et al. [30] demonstrated thatAAV vectors can efficiently disrupt one allele of the PRNP genein cultured bovine fibroblasts, expanding the use of AAVvectors to animal transgenesis.

Targeted homologous recombination also has been accom-plished in mammalian cells with adenovirus vectors [31].Because adenovirus very rarely integrates into the host genomeby nonhomologous recombination, replication-defective re-combinant adenoviral vectors are used as efficient expressionvectors. Transient expression of endogenous or even foreignproteins in cultured cells by adenovirus vectors would representa potential tool to manipulate donor cells in culture. Further-more, silencing of endogenous genes by adenovirus-mediatedexpression of small interfering RNA is now a reality [32]. Thesenovel, adenovirus-based approaches could open new possibil-ities for controlling cell processes, such as cycle progression,DNA methylation, or apoptosis in SCNT donor cells.

In the present study, we have isolated and established adultpMSC lines from live animals using a minimally invasive BMaspiration technique. The mesenchymal identity of isolated cellswas determined by expression of surface markers and multi-lineage differentiation potential. We then designed experimentsaimed a) to optimize in vitro culture conditions of pMSCs, b) tocompare transfection/transduction efficiencies of pMSCs andisogenic skin fibroblasts (SFs) exposed to integrating andnonintegrating vectors, and c) to examine the ability of pMSCsto drive development of SCNT embryos to the blastocyst stage.We have shown that adult pMSCs can be genetically modifiedand used to produce SCNT embryos. This is significant in that itprepares us for future large-animal autologous cell/gene therapymodeling comparing the adult cells to embryonic cells derivedthrough SCNT.

MATERIALS AND METHODS

Bone Marrow and Skin Collection

Blood marrow aspirates were obtained from anesthetized, young adult,female pigs (age, ;6 mo). General anesthesia was induced with a combination ofketamine (10 mg/kg body weight i.m.) and xylazine (2 mg/kg body weight i.m.)and maintained with inhalation anesthesia (halothane). Aspirates of BM (;20

ml) were collected from the humeral head with an 11-gauge biopsy-aspirationneedle (Medical Device Technologies, Inc.) attached to a heparinized syringe. Anear-skin sample was obtained from the same animal by punch biopsy. Bonemarrow and skin samples were immediately transported to the laboratory forfurther processing. All animal procedures were approved by the InstitutionalAnimal Care and Use Committee at the University of Georgia.

Isolation of BM pMSCs and SFs

Mononuclear cells were separated by centrifugation of BM aspirates througha solution of polysucrose and sodium diatrizoate (Histopaque; density, 1.077;Sigma) as indicated by the manufacturer. Briefly, 5 ml of Dulbecco phosphate-buffered saline (D-PBS) were added to 3 ml of marrow aspirate and mixed ina 15-ml centrifuge tube. The cell suspension was deposited over 3 ml ofHistopaque and centrifuged at 400 3 g for 30 min at room temperature.Mononuclear cells were recovered with a Pasteur pipette from the opaqueinterface and washed twice with D-PBS. For SF isolation, cartilage tissue wasremoved from the ear-skin sample, followed by scratching the surface of it witha scalpel to partially remove the dermis. Then, specimens were finely choppedwith a scalpel blade and digested by treatment with 10 ml of digestion solutioncontaining 0.2% trypsin (catalog no. T4799; Sigma) and 0.2% collagenase(catalog no. C9263; Sigma) in D-PBS containing 5 mg/ml of BSA at 378C withagitation. At 10-min intervals, the supernatant containing cells was removed andreplaced by fresh digestion solution. Approximately 30 min were required todigest the tissue corresponding to a skin sample. Cells recovered from one skinsample were washed in D-PBS and plated in approximately eight 75 cm2 flasks inMinimum Essential Medium (MEM) Alpha medium (catalog no. 12000-022;Invitrogen Corporation) supplemented with 10% fetal bovine serum (FBS).

Culture of pMSCs

After washing, mononuclear cells were resuspended in MEM Alpha mediumsupplemented with 10% FBS and plated on plastic flasks at a density ofapproximately 500 000 cells/cm2. After 24 h, unattached cells were washed offthe flask during medium exchange. Adherent fibroblast-like cells were allowed togrow for 10–14 days, with media replacement every third day. Cells werepassaged at 80%–90% confluence by trypsinization (0.25% trypsin-EDTAsolution; Sigma) and reseeded at a density of 5000–6000 cells/cm2 in plasticflasks.

Expression of Surface Markers

Expression of surface markers in MSC cultures for phenotypic characteriza-tion was performed by indirect immunofluorescence and flow cytometricanalysis. For immunocytochemistry, cells grown in glass chamber slides werefixed with 2% formaldehyde for 5 min, washed in D-PBS, and blocked with 3%goat serum in D-PBS for 30 min. Then, cultures were incubated with eithera 1:500 dilution of the primary antibody (anti-THY1 or anti-ITGAM) or isotypecontrol (Mouse IgG

1, clone, MOPC-31C; BD Biosciences, Pharmingen) or D-

PBS (negative control) for 1 h and 15 min. Primary antibody used was anti-human THY1 monoclonal antibody that cross-react with pig antigens (Clone5E10; BD Biosciences, Pharmingen) or anti-pig ITGAM monoclonal antibody(Clone 2F4/11; BD Biosciences, Pharmingen). After washing in D-PBS, cellmonolayers were incubated with Alexa Fluor 488 goat anti-mouse IgG (1:100dilution; Molecular Probes) for 1 h. Cells were washed in D-PBS, stained with40,60-diamidino-2-phenylindole (DAPI; 1 lg/ml; Calbiochem), and mountedwith Vectashield mounting medium (Vector Laboratories). Specimens wereexamined under an epifluorescent inverted microscope (Nikon Eclipse TE2000-S; Nikon Corporation) equipped with a digital camera (Qimaging Ratiga 1300;Qimaging).

The same basic staining procedure described for immunocytochemistry wasused to prepare the cells for flow cytometric analysis with minor modifications.Cell cultures were trypsinized, washed in D-PBS, and fixed with 2%formaldehyde solution for 3 min. Nonspecific binding was prevented byincubating the cells in 3% goat serum for 30 min. Cells were incubated in 15 lg/ml of the primary antibody (anti-THY1 or anti-ITGAM) or isotype control for 45min at room temperature. After washing, cells were incubated with a 1:500dilution of Alexa Fluor 488 goat anti-mouse IgG. Fluorescent cell analysis wasperformed with FACSCalibur cytometer (Becton Dickinson ImmunocytometrySystem) and data analyzed by FlowJo software (Tree Star, Inc.).

Lineage Differentiation of MSCs

The pMSC cultures were exposed to chondrogenic, lipogenic, or osteogenicconditions for 14–20 days to determine multipotency. Lipogenic and osteogenicinduction was applied to cells growing in monolayers. Chondrogenic and

APPLICATIONS OF PORCINE MSCS IN TRANSGENESIS 47

osteogenic differentiation was induced on cell masses as described previously inhuman MSCs [33] and in pMSCs [13]. Briefly, aliquots of 200 000 cellswere distributed in 15-ml conical tubes and centrifuged for 5 min at 600 3 g.Sedimented cells were cultured in the tubes with loosened caps to allow gasexchange. Cells formed a spherical mass on the bottom of the tube by 24 hof culture. Composition of differentiation media is shown in Table 1.Differentiation media were replaced every 3–4 days. For lipogenicdifferentiation, cells were first exposed to induction medium for 2–3 daysand then cultured in maintenance medium (Table 1) for another 2–3 days.This alternating treatment was repeated three or four times to achieve fulllipogenic differentiation.

Histochemical stains were used to assess cell differentiation into specificlineages in adherent cell cultures and histological cryosections of cell masses.Cell masses were embedded in a water-soluble embedding medium, frozen inliquid N

2, and sectioned (thickness, 10 lm) with a Leica CM3050 cryostat

(Leica). Accumulation of phosphates and carbonates indicative of osteogenicdifferentiation was demonstrated by the von Kossa silver reduction method [4].Cultures or cryosections were fixed with 4% formaldehyde, exposed to 5%silver nitrate solution, and immediately exposed to direct ultraviolet (UV) lightfor 45–60 min. Specimens were then washed with distilled water and incubatedfor 2–3 min in 5% sodium thiosulfate solution. Expression of alkalinephosphatase (AP) was assessed by a commercial kit (Vector Red AlkalinePhosphatase Substrate Kit I; Vector Laboratories). Intracellular accumulation ofneutral lipids was demonstrated by Oil Red O staining [4]. For this assay,monolayers were fixed and stained with Oil Red O working solution for 1 h.The working solution was made fresh each time by mixing one part distilledwater with 1.5 parts of a saturated Oil Red O solution (0.5% w/v Oil Red O in99% isopropyl alcohol). Acidic mucopolysaccharides in cartilage tissue werestained with alcian blue 8GX (Sigma). Briefly, cryosections were fixed with 3%acetic acid and stained with alcian blue solution (1% w/v alcian blue in 3%acetic acid, pH 2.5) for 30 min. After washing with distilled water, slides weremounted with 90% glycerol and inspected with a transmitted-light microscope.Photographs were taken with a digital camera (Qimaging Ratiga 1300;Qimaging) mounted on the microscope.

Optimization of Culture Conditions for pMSCs

Colony-forming unit fibroblast assay. Passage 2–3 MSCs growing inflasks were trypsinized and plated at density of 10 cells/cm2 in 100- 3 20-mmdishes. Cells were cultured for 14 days under different experimental conditions.Medium was replaced every third to fourth day. Cell colonies were washedwith D-PBS and stained with 3% Crystal Violet (Sigma) in methanol for 15–20min and the number and size of colonies recorded for each experimental group.Number of colonies as well as major and minor axes of each colony weremeasured with the aid of an ocular micrometer. The averages of the major andminor axes are reported as colony diameter. For each experimental condition,the best treatment is the one that induces the highest number of colonies withthe larger diameter.

Cell proliferation assay. Passage 2–3 MSCs growing in flasks weretrypsinized and plated at a density of 800 cells/well of 96-well assay plates(black plate with clear bottom; Corning, Inc.). Cells were grown under differentexperimental conditions for 4 days. Media were removed and the plates storedat �808C until the cell proliferation assay was performed following themanufacturer’s instructions (CyQUANT; Molecular Probes). Cells cultured in

medium containing 2% or 20% FBS were included as low- and high-proliferation controls, respectively. Fluorescence in the samples, reported asrelative fluorescence units (RFUs), was measured with a fluorescence micro-plate reader (SPECTRAmax GEMINI; Molecular Devices Corporation) withfilters appropriate for 485 nm (excitation) and 538 nm (emission).

The CyQUANT proliferation assay kit also was used to investigate celladhesion efficiency. Cells were plated in two plates at a density of 13 000 cells/well and incubated under different experimental conditions at 378C in 5% CO

2in

air. After a 5-h incubation, one plate was centrifuged at 600 3 g for 15 min topellet the cells, and the media were carefully removed. This plate was used todetermine total cell numbers. Media from the second plate were removed, andthe wells were washed three times with D-PBS. All plates were stored at�808Cuntil the CyQUANT proliferation assay was performed according tomanufacturer’s instructions.

Experiment 1: Effect of FBS concentration on proliferation of pMSCs.The objective of this experiment was to study the effect of media containingincreasing concentrations of FBS (2.5%, 5%, 10%, 20%, or 30%) on growingcharacteristics of pMSCs in the colony-forming unit fibroblast (CFU-F) assay.

Experiment 2: Effect of media and oxygen tension on proliferation ofMSCs. The ability of different media to support pMSC grow in vitro under anatmosphere of low or high oxygen concentration was investigated using theCFU-F and CyQUANT proliferation assays. The pMSCs were cultured ineither MEM Alpha, low-glucose Dulbecco modified Eagle medium (DMEM)containing 2.2 g/L of sodium bicarbonate (DMEM 2.2; catalog no. 31600-034;Invitrogen Corporation), low-glucose DMEM containing 3.7 g/L of sodiumbicarbonate (DMEM 3.7; catalog no. 31600-034; Invitrogen Corporation), orDMEM/Ham F12 (DMEM/F12; catalog no. D0547; Sigma). All media weresupplemented with 10% FBS. Culture was carried out in low oxygenconcentration (5% O

2, 5% CO

2and 90% N

2) or high oxygen concentration

(5% CO2

in air). In the CyQUANT proliferation experiment, cells were seededin 96-well plates in MEM Alpha media with 10% FBS and allowed to attach for12 h. After this, media were removed and treatments applied. The design of theCFU-F experiment was slightly different, because cells were directly seeded in100-mm dishes in the treatment media.

Experiment 3: Effect of ascorbic acid supplementation and oxygentension on proliferation of pMSCs. The potential stimulatory effect ofincreasing concentrations of ascorbic acid 2-phosphate (0 [control], 5, 50, 500,or 5000 lg/ml; catalog no. A8960; Sigma Chemical) added to the culturemedium (low-glucose DMEM), and proliferation of pMSCs was investigated.The design also included the effect of culture under an atmosphere with low orhigh oxygen concentration on proliferation of pMSCs across all ascorbic acidtreatments.

Transient Genetic Modification

Transduction with a human adenovirus. A chimeric adenovirus type 5that contains the adenovirus type 35 fiber and carries the GFP gene (Ad5F35-eGFP; 5 3 1012 particles/ml, 3.45 3 1010 pfu/ml) was obtained from the VectorDevelopment Laboratory at Baylor College of Medicine, Houston, Texas.

Passage 2–3 pMSC and matching isogenic SF cell lines were seeded ata density of 43 000 cells/cm2 in 12-well plates (150 500 cells/well). Twenty-four hours after plating, cultures were infected with 100 multiplicity ofinfection (defined as pfu/cell) in 500 ll of MEM Alpha with 10% FBS. Thepercentage of GFP-positive cells, relative fluorescence intensity (RFI) of the

TABLE 1. Composition of differentiation media and methods used to assess lineage differentiation.

Lineage Differentiation medium composition Assessment of differentiation

Lipogenic Induction: High glucose MEM Alphasupplemented with ITSþ1 (Sigma),sodium pyruvate (10 mM), methylisobutylxanthine (0.5 mM) anddexamethasone (1lM). Maintenance:High glucose MEM Alpha supplementedwith ITSþ1 (Sigma) and sodium pyruvate(10 mM).

Oil Red O stain: lipid accumulation.

Chondrogenic MEM Alpha supplemented with pTGF-b1(10 ng/ml), dexamethasone (100 nM),ascorbic acid 2-phosphate (50 lg/ml),thyroxine (50 ng/ml) and ITSþ1 (Sigma).

Alcian blue stain: acidic mucopolysaccharides.

Osteogenic MEM Alpha supplemented with 10% FBS,dexamethasone (10�8 M), ascorbic acid(50 lg/ml), and b-glycerophosphate (10 mM).

von Kossa stain: phosphates and carbonates. Alkaline phosphatase (AP) activity.

48 BOSCH ET AL.

GFP-positive cell population and cell viability by exclusion of propidiumiodide (50 lg/ml; Roche Applied Science) was determined 24 h after viralexposure by flow cytometric analysis using a FACSCalibur cytometer andFlowJo software.

Transfection with GFP plasmid. Early passage (2–3) pMSCs and isogenicSFs were plated in 12-well plates (120 000 cells/well) and, 20 h later, weretransfected with a plasmid containing enhanced GFP gene under control ofcytomegalovirus promoter and neomycin-resistant gene under control of anSV40 promoter that allows selection using geneticin (EGFP-N1; ClontechLaboratories). Transfection was carried out in the presence of a polyamine-based transfection reagent (GeneJammer; Stratagene) according to themanufacturer’s recommendations (2 lg plasmid DNA/well). Transfected cellswere sorted based on GFP fluorescence 72 h after transfection using a MoFlofluorescence-activated cell sorting (FACS) set (DakoCytomation) to sort onecell per well of 96-well plates (three plates per cell line). Cells were cultured inMEM Alpha with 15% FBS for 14 days, and development of GFP expressingcolonies was determined at this point by inspection under a microscopeequipped with UV light.

Stable Genetic Modification

Transfection with GFP plasmid. Passage 2–3 MSCs and isogenic SFswere plated in four 100- 3 20-mm plastic dishes per cell line at 1.2 3 106 cells/dish. Cultures were transfected with EGFP-N1 plasmid (12 lg plasmid DNA/dish) using GeneJammer Transfection Reagent according to manufacturer’sspecifications. Selection for transgenic cells was initiated in three dishes percell line 72 h after transfection by culturing the cells in medium containingGeneticin (250 lg/ml; G418; Sigma). Number of GFP-expressing colonies wasdetermined 14 days after transfection. The remaining 100-mm dish waspassaged and propagated in MEM Alpha with 15% FBS. After 8–9 days, cellswere trypsinized and stained with 50 lg/ml of propidium iodide (RocheApplied Science). Viable GFP-expressing cells were sorted with a MoFloFACS cytometer (DakoCytomation) as single cells in 96-well plates (threeplates per cell line) containing culture medium supplemented with 20% FBS,and colonies were allowed to grow for 14 days (with media changed at Day 7).At the end of the culture period, colonies were graded according to theirdevelopment as follows: category 1, colony covering all or almost all thesurface of the well; category 2, colony covering approximately half the well;and category 3, colony covering one-fourth of the well. Colonies also weregraded as GFP positive (high, medium, or low florescence intensity) or as GFPnegative.

AAV transduction. Human AAV vector carrying the GFP gene was kindlyprovided by Vector Development Laboratory at Baylor College of Medicine.The pMSC and SF cultures (passage 2–3) were seeded in four-well plates(40 000 cells/well). Cell cultures were transduced 24 h after plating with 3 3

108 viral particles/well in MEM Alpha with 2% FBS. Serum concentration wasadjusted to 10% by adding FBS 3.5 h after transduction. Transduced cells werepassaged and expanded for 9–10 days in MEM Alpha supplemented with 15%FBS before sorting viable, GFP-positive cells in 96-well plates (one cell perwell, three plates per cell line) using a FACS cytometer (MoFlo;DakoCytomation). Cells were cultured for 14 days in MEM Alpha containing15% FBS (replaced at Day 7 of culture), and colony development wasevaluated as described above.

Somatic Cell Nuclear Transfer

Confluent (passage 2) MSC and SF cultures exposed to roscovitine (15 lM;Sigma) during the last 24 h of culture [34] were used as a source of karyoplaststo produce NT embryos. In vitro-matured oocytes were enucleated, anda single-cell (MSC or SF) was transferred into the perivitelline space. Cell-oocyte couplets were fused in Zimmerman medium with a single electric pulse(250 V/mm for 20 lsec) delivered through a needle-type electrode. The NTunits were electrically activated (two pulses of 75 V/mm for 60 lsec separatedby 5 sec) in a chamber 1 h after fusion and transferred to drops of NCSU-23medium. Embryos were examined for cleavage and blastocyst formation at 2and 7 days, respectively, after NT.

Statistical Analysis

The CFU-F data from experiment 1 were analyzed by one-way ANOVAusing the general linear model (GLM) procedure of the Statistical AnalysisSystem [35] followed by the protected least-significant-difference (LSD) test.The CFU-F and proliferation data from experiments 2 and 3 were analyzed bytwo-way ANOVA using the GLM procedure of SAS under a completelyrandomized factorial design. The model included variation caused by treatment(media in experiment 2 or ascorbic acid in experiment 3), oxygen tension (high

or low), and their interaction. When a significant effect was detected with theANOVAs, treatment means were compared by protected LSD. Student t-testwas used for comparing data from two groups (i.e., pMSC vs. SF). All valuesare presented as the mean 6 SEM from at least three replicates. Differenceswere considered to be significant at P , 0.05.

RESULTS

Isolation of Cell Lines

Mesenchymal stem cell lines were established successfullyfrom BM collected from 10 anesthetized gilts (n ¼ 10). Thenumber of mononuclear cells per BM aspirate (;20 ml)recovered from the density gradient was 2.33 6 0.5 3 108

mononuclear cells, enough to plate approximately six 75-cm2

flasks. Most of the nonadherent cells were removed during thefirst media change at 24–48 h. Discrete colonies of fibroblast-like cells attached to the plastic were evident at Days 4–5 afterinitial seeding. Most cell lines were composed of cells witha characteristic spindle shape, whereas others had cells withpolygonal morphology. The number and size of the coloniesincreased progressively to reach 80% confluency by Days 14–15 after seeding (Fig. 1, A and B).

Expression of Surface Markers

Immunocytochemistry revealed that MSCs from pig BMwere positive for the cell surface marker THY1 (Fig. 2A) andnegative for ITGAM (Fig. 2B). Flow cytometric analysisconfirmed that 99.4% 6 0.20% of the cells expressed THY1antigen (Fig. 2C), and virtually the entire population wasnegative for ITGAM (Fig. 2D).

Lineage Differentiation of MSCs

Results indicated that BM mesenchymal cells acquiremorphological and histochemical characteristics of adipose,cartilage, or bone tissues when exposed to specific differen-tiation-inducing conditions (Fig. 3). Conversely, isogenic SFsexposed to identical induction conditions failed to differentiate(Fig. 3). Cells with discrete, although small, lipid droplets werepresent as early as Days 4–5 of culture. The number of cellswith lipid accumulation and the size of the lipid dropletsincreased until Days 8–9 and then plateaued until the end ofculture period (Days 12–14). The percentage of cells un-dergoing lipogenic differentiation was highly variable amongcell lines, ranging from approxiamtely 1% to 15%. Oil Red Oconfirmed the presence of neutral lipid accumulation indifferentiated pMSCs (Fig. 3A). Lipogenesis was not evidentin pMSCs maintained in culture medium alone (control) (Fig.3B) or isogenic SFs exposed to lipogenic medium (Fig. 3C).Alcian blue staining revealed acidic mucopolysaccharides insections of pMSC masses cultured in chondrogenic medium for14–17 days (Fig. 3D). Cell morphology also was compatiblewith chondrocytes. The pMSC controls and SFs cultured indifferentiation media were negative for alcian blue stain (Fig. 3,E and F). Extensive osteogenic differentiation, as evidenced byblack deposits with von Kossa stain (Fig. 3G) and AP activity(data not shown), was noticeable only in pMSCs exposed toosteogenic conditions. The pMSC controls and SFs cultured inosteogenic differentiation media were both negative for vonKossa stain (Fig. 3, H and I) and AP activity (data not shown).

Optimization of Culture Conditions for pMSCs

Experiment 1: Effect of FBS concentration on pro-liferation of pMSCs. The percentage of FBS in the culturemedium greatly influenced both the number of colonies per

APPLICATIONS OF PORCINE MSCS IN TRANSGENESIS 49

dish and the mean colony diameter. The number of colonies perdish increased with increasing FBS concentrations up to 10%,when a plateau was reached (Fig. 4A). A similar positive effectof FBS on colony diameter also was evident; however,a plateau was not reached with 30% FBS (Fig. 4B).

Experiment 2: Effect of media and oxygen tension onproliferation of pMSCs. The effect of media and oxygentension on CFU-F assay was replicated with four cell lines(obtained from four different animals). In three of four celllines, the number and diameter of colonies were markedlysmaller in DMEM/F12 medium compared with those in theother media studied, whereas in the remaining cell line, theresponse to DMEM/F12 medium was similar to that in theother treatments. Because the inclusion of data from this cellline in the statistical analysis would mask the negative effectobserved for DMEM/F12 medium in three of four cell lines, wedecided to exclude it from the statistical analysis. The numberand diameter of colonies was not different for MSCs grown inMEM Alpha, DMEM 2.2, or DMEM 3.7 (Fig. 4, C and E).Cells cultured in DMEM/F12 responded with fewer andsmaller colonies compared to those cultured with the othermedia studied (Fig. 4, C and E). Despite the fact that oxygen

tension affected neither the number nor the mean diameter ofcolonies in the CFU-F assay (Fig. 4G), we observed darkercolonies in cultures maintained under low oxygen tension(5%), suggesting higher number of cells per colony.

The effect of media type on proliferation of pMSCs in twodifferent oxygen tensions (low or high) was investigated withthe CyQUANT proliferation assay. Significant effect ofmedium type (P , 0.0001) and oxygen tension (P ,0.0001), but not interaction (P¼ 0.49) between these variables,was observed. The number of RFUs, which is correlated to theamount of DNA, was higher in DMEM/F12-treated cellscompared with the other media types studied (Fig. 4D). Lowoxygen tension had a positive effect on cell proliferation, asevidenced by a higher RFU value compared with cellsmaintained in high-oxygen atmosphere (Fig. 4H).

The ability of MEM Alpha and DMEM/F12 to induceadhesion of pMSCs to plastic was compared with theCyQUANT assay. The MEM Alpha had a better ability toinduce pMSC attachment compared with the DMEM/F12medium (MEM Alpha, 189.33 6 10.05 RFU; DMEM/F12,164.66 6 2.96 RFU; P , 0.05).

Experiment 3: Effect of ascorbic acid supplementationand oxygen tension on proliferation of pMSCs. Significanteffects of ascorbic acid concentration (P , 0.0001) and oxygentension (P , 0.0001), but not of the interaction (P ¼ 0.56)between these variables, were observed. Supplementation ofthe culture medium with 5–500 lg ascorbic acid/ml did notaffect cell proliferation compared with the control (Fig. 4F).Addition of 5000 lg ascorbic acid/ml, however, impairedpMSC proliferation (Fig. 4F). Coinciding with results fromexperiment 2, low oxygen tension significantly improved (P ,0.05) pMSC proliferation rate.

Transient Genetic Modification of SFs and pMSCs

Transduction with a human adenovirus. Microscopicinspection of cultures under UV light 24 h after infectionrevealed a superior transduction efficiency in pMSCs comparedto SFs (Fig. 5, E and F), a finding later confirmed by flowcytometry. The percentage of cells expressing GFP wasapproximately 15% higher in pMSCs than that in isogenicSFs (pMSCs, 70.25% 6 5.45%; SFs, 55.31% 6 6.83%; P ¼0.02) (Fig. 5, A–C). Relative fluorescence intensity also washigher in pMSCs compared with that in SFs (pMSCs, 959.666 73.25 RFI; SFs, 585.75 6 19.32 RFI; P ¼ 0.005) (Fig.5A). Percentage of propidium iodide-positive cells was higherin pMSCs (5.35% 6 0.38%) compared with that in SFs (3.45%6 0.24%, P ¼ 0.01) (Fig. 5A).

Transfection with GFP plasmid. No difference wasobserved in the proportions of SFs and pMSCs expressingGFP (SFs, 3.99% 6 0.95%; pMSCs, 8.44% 6 2.33%; P ¼0.22) (Fig. 5D). Viability also was similar between the twoexperimental groups (Fig. 5D). The GFP-positive cells weresorted individually in three 96-well plates per cell line andchecked for GFP-expressing colonies after 14 days of culture.Only 7 of 648 colonies that developed in SF and pMSC plateswere GFP positive.

Stable Genetic Modification

We used two different vectors, namely a GFP plasmid andAAV, to obtain cell populations displaying extended expres-sion of the transgene, which normally is associated withintegration of the transgene into the host DNA. Comparison ofproportions of cells that remained GFP positive after trans-fection/transduction and proliferation in vitro (8–10 days)

FIG. 1. Morphology of adherent fibroblast-like cells, later identified aspMSCs, isolated from pig bone marrow after 14 days from initial plating(A). Same cell line at higher magnification showing detailed fibroblasticmorphology of pMSCs (B). Bar ¼ 200 lm (A) and 100 lm (B).

50 BOSCH ET AL.

revealed a higher proportion of GFP-expressing cells in the

AAV- than in the GFP plasmid-transfected group, irrespective

of cell line (Fig. 6A). No difference was found in transfection/

transduction efficiency between SFs and pMSCs (Fig. 6A).

Transfection with GFP plasmid. The mean number of

colonies per 100-mm dish expressing GFP after 14 days of

G418 selection was 74.32 6 8.34 in SFs and 74.38 6 4.58 in

pMSCs (P . 0.05).

We implemented an approach to reduce the number of cells

with transient expression of GFP from episomes, which

consisted in propagation of transfected cell lines for 10 days

and then sorting individual cells in 96-well plates based on

FIG. 2. Immunofluorescence and flowcytometry for the surface markers THY1 andITGAM in pMSCs. Specific immunoreactiv-ity for THY1 was observed in pMSCsgrowing on chamber slides (A). ITGAMimmunoreactivity was absent in pMSCmonolayers (B; green, immunoreactiveTHY1; blue, DAPI). These results were laterconfirmed by flow cytometry (C and D). ThepMSC suspension was fixed and immuno-stainied for THY1 (C) or ITGAM (D)expression. Histograms show frequencydistribution and fluorescence intensity data.Blue curves represent the distribution ofcells incubated with anti-THY1 or anti-ITGAM primary antibody, whereas redcurves represent the distribution of cellsincubated with the immunoglobulin isotypecontrol. In the histogram (C) from a repre-sentative pMSC line, 99.6% of total cellswere positive for THY1. In D, ITGAM (blue)and isotype (red) curves overlap, indicativeof absence of immunoreactive sites forITGAM on pMSCs. Bar ¼ 100 lm.

FIG. 3. Histochemical stains of SFs andpMSCs exposed to lipogenic, chondrogenic,osteogenic, or control media. The pMSCsunderwent lipogenic (A), chondrogenic (D),and osteogenic differentiation when ex-posed to specific induction media. ThepMSCs cultured in control medium (B, E,and H) and isogenic SFs (C, F, and I)exposed to differentiation conditions failedto differentiate. Bar ¼ 100 lm (A–C) and0.5 mm (D–I).

APPLICATIONS OF PORCINE MSCS IN TRANSGENESIS 51

GFP fluorescence. Two weeks after plating, approximately35% of plated cells formed a colony (Fig. 6B). Irrespective ofcell line (SF or pMSC), the vast majority of these colonies(;92%) did not express GFP. Colony development was similarbetween SFs and pMSCs (Fig. 6C). A higher percentage ofhigh-fluorescence SF colonies was found compared with thesame category in pMSCs (Fig. 6C) (P , 0.05). A pMSCcolony originated from one cell transfected with the GFPplasmid and positive for GFP after 14 days of culture is shownin Figure 6F.

AAV transduction. The proportion of plated cells thatformed a colony was not different between cell lines, rangingfrom 26.39% 6 1.31% in pMSCs to 40.04% 6 7.79% in SFs.Contrasting with results obtained with GFP plasmid-transfectedcells, 90.2% of SF colonies and 96.2% of pMSC coloniesexpressed GFP (Fig. 6, compare B with D). No difference wasobserved in the proportion of colonies in category 1, 2, or 3between cell lines (Fig. 6E). The pMSCs had a largerproportion of high-fluorescence colonies and a lower percent-age of low-fluorescence colonies compared to SFs (P , 0.05)

(Fig. 6E). A pMSC colony positive for GFP after 14 days ofculture is shown in Figure 6G.

Somatic Cell Nuclear Transfer

Both pMSCs and SFs synchronized with roscovitine wereused in two replicates as nuclear donors to produce clonedporcine embryos. Cleavage rates were 44.5% (65/146) forpMSC and 53.1% (60/113) for SF NT embryos. Developmentto blastocyst stage was 4.1% (6/146) in the MSC group and1.77% (2/113) in the SF group. The Yates chi-square testrevealed no statistical difference in cleavage and blastocystdevelopment rates between pMSC- and SF-derived embryos (P¼ 0.21 and P ¼ 0.47, respectively). The number of cells perblastocyst ranged from 7 to 23 cells/blastocyst.

DISCUSSION

Since they were first identified by the pioneering work ofFriedenstein et al. [36, 37] in the early 1970s, MSCs, alsoknown as marrow somatic cells or CFU-F cells, have been the

FIG. 4. Effect of culture conditions onproliferation of pMSCs as determined byCFU-F or CyQUANT assays. Effect ofsupplementation of culture media withdifferent FBS concentrations on the numberof colonies (A) and colony diameter (B).Response of pMSCs to medium type in theCFU-F (C and E) and CyQUANT prolifera-tion assays (D). Effect of increasing con-centrations of ascorbic acid on proliferationof pMSCs in the CyQUANT assay (F).Treatments of 2% and 20% FBS wereincluded as low- and high-proliferationcontrols, respectively (D and F). Effect oflow or high oxygen tension on colonyformation in the CFU-F assay (G) andproliferation in the CyQUANT assay (H).Values are presented as the mean 6 SEM ofat least four independent replicates. Barswith different letters are statistically differ-ent at P , 0.05 (ANOVA followed by LSD).

52 BOSCH ET AL.

subject of numerous studies aimed at deciphering the roles ofthese stem cells in the complex marrow physiology. The MSCsare considered to be nonhematopoietic precursor cells thatsupport hematopoiesis and can differentiate down the lipo-genic, chondrogenic, osteogenic, and tenogenic pathways (forreview, see [38]). The broad differentiation potential (alongwith the extensive ex vivo proliferative capacity) makes thesestromal precursors attractive candidates for autologous andallogeneic cell therapy and, potentially, for SCNT transgenesis.We have isolated and characterized the growing properties ofpMSCs under different culture conditions. Then, we examinedthe ability of pMSCs and isogenic SFs to undergo transient andstable genetic modifications using a combination of GFPplasmid with a transfection reagent and viral vectors. Finally,the present study suggests that pMSCs can undergo nuclearreprogramming to generate cloned blastocysts.

We were able to establish a primary MSC line from each of10 individual animal BM aspirations. Imunocytochemistry andflow cytometric analysis revealed that most cells expressed themesenchymal marker THY1 and did not express ITGAM,a hematopoietic marker in granulocytes, monocytes, naturalkiller cells, subsets of T cells, and subsets of B cells [39].

Expression of cell surface markers in BM-derived cells isolatedin the present study supports the mesenchymal origin of thesecells and agrees with the results of previous work. Thesemarkers are conserved across species, because human and ratMSCs also express THY1 and lack ITGAM expression [4, 6].Furthermore, analysis of forward and scatter light data frompMSCs revealed homogenous cell populations (size andgranularity; data not shown) coinciding with flow cytometricresults from human MSCs [4]. The MSCs, but not thefibroblasts, were capable of differentiating down mesenchymallineages [4, 10, 11, 13], demonstrating that isolated cells frompig BM were truly MSCs. Morphology, surface antigen profile,and pluripotency characteristics provide convincing evidencethat the BM cells isolated in the present study are pMSCs.

Because little is known about culture conditions that supportpMSC proliferation in vitro, we first sought to characterize thegrowth properties of pMSCs under different culture conditions.Development of MSC colonies depended entirely on thegrowth factors in FBS (Fig. 4, A and B). No colonies werepresent when FBS was omitted, and a clear, positive dose-response relationship was observed between colony numbersand FBS concentrations. A similar response to FBS was

FIG. 5. Transient transduction/transfection of SFs and pMSCs with a human adenovirus (Ad5F35-eGFP) or a plasmid carrying the GFP gene. The SFs andpMSCs were transfected with Ad5F35-eGFP and, after 24 h, were characterized by flow cytometry. The percentage GFP positive, mean fluorescenceintensity, and proportions of viable cells (A) were estimated from the flow cytometric data. Representative dot plots of transduced isogenic SFs (B) andpMSCs (C) showing distribution of cell populations based on GFP intensity (x-axis) and propidium iodide staining (y-axis). Very low proportions of cells arenonviable (top left quadrants in B and C). A larger percentage of cells are viable and GFP positive in pMSCs (74.6%; bottom right quadrant in C) comparedwith that in SFs (44.1%; bottom right quadrant in B). The SFs and pMSCs were transfected with GFP plasmid and characterized by flow cytometry 72 hafter transfection (D). The percentage of GFP-positive and nonviable cells were estimated from the flow cytometric data. Photomicrographs of SFs (E) andpMSCs (F) taken under UV light 24 h after transduction with Ad5F35-eGFP also are shown. Different symbols within each variable denotes significantdifference at P , 0.05 (Student t-test). Bar ¼ 200 lm.

APPLICATIONS OF PORCINE MSCS IN TRANSGENESIS 53

reported for human MSCs [40]. Overall, it is apparent thataddition of 10%–20% of serum to pMSC culture mediumprovides adequate support for pMSC expansion. In three offour pMSC lines, the number and diameter of colonies was notdifferent for pMSCs cultured in MEM Alpha, DMEM 2.2, orDMEM 3.7 but was significantly lower for cells growing inDMEM/F12. The DMEM/F12 medium, however, enhancedproliferation of pMSCs in the CyQUANT assay (Fig. 4D). Thisinconsistent response may have arisen from inherent differ-ences between the assays. The CFU-F assay measured bothplating efficiency and proliferation capacity of media, whereasthe CyQUANT assay only measured proliferation capability.We found that DMEM/F12 medium did reduce rates of cellattachment in the CyQUANT plating assay. Different cellconcentrations between CFU-F (very low cell density) andCyQUANT (high cell density) assays also may havecontributed to the differential outcomes observed.

The effect of oxygen tension during culture on colonyformation was investigated. Despite the fact that oxygentension did not affect the number and size of the colonies (Fig.

4G), pMSCs growing in low oxygen (5%) proliferated fasterthan cells cultured in higher oxygen tension (21%) (Fig. 4H).Culture of cells in reduced oxygen tension has been reported tocause inhibition [41, 42] or stimulation of cell growth in vitro[43–47]. Our results demonstrated an increased proliferationrate of pMSCs in an oxygen concentration (5%) that moreclosely resembles in vivo conditions. Therefore, exogenouscontrol of oxygen tension may have important implications forin vitro propagation of pMSCs and, possibly, differentiation[47, 48].

Ascorbic acid or vitamin C, a primary antioxidant for cells,has been associated with enhancement of cell proliferation [49–51], and it contributes to collagen synthesis in mesenchymalcells [52–54]. One difficulty associated with supplyingascorbate to cultured cells is the instability of this vitaminunder standard culture conditions (neutral pH, 378C, andaerobic environment) resulting from its autoxidation [55]. Toovercome this problem, we used an esterified ascorbate(ascorbic acid 2-phosphate), which is more resistant toautoxidation [50] and, therefore, more stable in aqueous

FIG. 6. Stable genetic modification of SFsand pMSCs with a GFP plasmid or an AAVvector. The SFs and pMSCs were transfectedwith a GFP plasmid or AAV, expanded invitro for 9–10 days, and sorted as singleGFP-positive cells by flow cytometry. Thepercentage of GFP-positive SF- and pMSC-transfected/transduced cells after propaga-tion in vitro is shown (A). The GFP-expressing cells sorted in 96-well plateswere cultured in vitro for 14 days, and thepercentage of GFP-negative and GFP-posi-tive colonies were determined for cellstransfected with GFP plasmid (B) or AAV(D). Colonies were classified according todegree of development (1, 2, or 3) andfluorescence intensity (high, medium, orlow; C and E). Photomicrographs (takenunder UV light) of GFP-positive pMSCcolonies generated from a single cell trans-fected either with GFP plasmid (F) or AAV(G) also are shown. *P , 0.05 (Student t-test). Bar ¼ 200 lm.

54 BOSCH ET AL.

solutions. Addition of ascorbic acid to a vitamin C-freemedium over a wide range of concentrations (5–500 lg/ml) didnot affect the proliferation rate of pMSCs. There may beseveral reasons why we did not observe an effect. The amountof free ascorbate available for the cells is dependent on the rateof conversion of ascorbate 2-phosphate to ascorbate in a givencell type and culture condition. Therefore, the rate ofconversion of ascorbate from its esterified precursor in ourculture system might not have been optimal. Additionally, themethod used to estimate cell numbers may not have beensensitive enough to detect subtle effects of ascorbic acid on cellproliferation. The impaired cell proliferation observed whenascorbate was added at 5000 lg/ml of medium is consistentwith the idea that at high concentrations, ascorbate favors thegeneration of free radicals, promoting in this way a pro-oxidative rather than an antioxidative state [56].

Transient gene transfer into cultured cells with subsequentexpression of the transgene has become a valuable tool forphysiological studies, functional genomics, and gene therapy.Furthermore, ectopic expression of signaling molecules andtranscription factors has been useful in manipulating thedifferentiation fate of stem cells [57, 58]. In the present study,we used a recombinant adenoviral vector carrying the GFPgene to transiently transduce pMSCs and SFs in combinationwith flow cytometric analysis to determine expression of thereporter gene. We have shown that both cell types can beinfected with the nonintegrating human adenovirus vector. Ahigher proportion of pMSCs, however, expressed GFPcompared with isogenic SFs (Fig. 5), and GFP intensity datasuggest that a larger number of adenoviral particles entered intoMSCs than SF cells (Fig. 5A). Because adenoviral entry intothe host cells is mediated through membrane receptors,particularly MCP (also known as CD46) for Ad5F35 [59],the pMSCs likely possess higher density of adenoviralreceptors. We also observed a higher proportion of nonviablepMSCs in these cultures (Fig. 5A), which could be the result ofviral cytopathic effects associated with the higher viralinfection achieved in this group. Nonetheless, overall cellviability was very high (.95%) in both cell lines. WhenpMSCs and SFs were transfected with the GFP plasmid, bothcells types transiently expressed GFP, but far fewer cells(ninefold lower) expressed the reporter gene compared to thatin adenovirus vector transduced cells.

The recent development of SCNT to produce clonedanimals has provided a new method for generating transgeniclivestock [20]. Besides being adaptable to in vitro proliferationconditions, the donor cells used in this process should beamenable to stable genetic manipulation and undergo nuclearreprogramming. We demonstrated stable transgene expressionin pMSCs and SFs using a GFP plasmid and a viral vector(AAV). Plasmid integration was confirmed by selection oftransgenic cells with the antibiotic G418; approximately 1 inevery approximately 16 000 treated cells integrated thetransgene. We then used FACS for clonal propagation ofGFP-positive cells in 96-well plates. At sorting, the percentageof GFP-positive cells transduced with AAV vector doubledthat of cells transfected with the plasmid (Fig. 6A). After 14days in culture, approximately 8% of the clonal colonies wereGFP positive in the plasmid-transfected group versus approx-imately 93% in the AAV vector-transduced groups (pMSCand SF). Both plasmid- and AAV vector-derived GFP-positivecolonies were expanded in vitro up to approximately 1 3 106

cells without losing GFP expression (;21 cell doublings).These data indicate clonal selection for stable transgeneintegration and propagation of transgenic donor cells.Furthermore, the AAV vector used in the present study

clearly was superior to a conventional GFP plasmid. Viralvectors have been used previously to create transgenic celllines later used to generate SCNT transgenic embryos [25] andanimals [26, 27]. Recently, Hofmann et al. [28] obtained hightransduction rates of bovine fibroblasts in culture witha lentivirus vector, and these transgenic cells were able todrive development to term. Therefore, the use of integratingviral vectors like lentivirus and AAV, as we demonstrated inthe present study, is a highly effective alternative method todeliver DNA into cells.

Less differentiated cells may be more amenable thanterminally differentiated cells to nuclear reprogramming onNT (for review, see [60]). For example, postimplantationsurvival of clones originated from mouse embryonic stem cellswas higher than that of adult somatic cells [61], and enhancedin vitro development of preimplantation pig embryos recon-structed with fetal skin-derived stem cells has been reported[62]. Bovine MSCs were able to undergo nuclear reprogram-ming after SCNT and supported development to term [18], butSCNT using other somatic cell types were not included in thatstudy for comparison purposes. In the present study, NT resultssuggest that pMSCs can support blastocyst development afterbeing transferred to enucleated metaphase II oocytes. Althoughboth the pMSCs and SFs produced low cleavage and blastocystrates, this does indicate that pMSCs can undergo nuclearreprogramming at least in support of development to theblastocyst stage. Additional replicates will be necessary toestablish any potential differences in reprogramming abilitybetween cell lines and types. Having shown that pMSC donorcells can develop into initial SCNT blastocyst-stage embryos,future studies might investigate the differentiation potential ordifferences between pMSCs and embryonic cells derived fromthe same donor cells. In addition, this work can lead to studiesthat determine the effects of stable (AAV) or transient (Ad5-F35 vector) expression of certain exogenous genes and theireffect on nuclear reprogramming.

In conclusion, we have been able to establish adult pMSClines from live animals using a minimally invasive BMaspiration technique. These adult stem cells can proliferateextensively in vitro (;30 cell doublings until senescence; datanot shown) and undergo transient and stable genetic modifi-cation with nonviral and viral vectors. Of particular interest isthe highly efficient transduction of MSCs with a nonintegratinghuman adenovirus and AAV vectors. All these characteristicsalong with favorable clonal cell propagation properties makepMSCs an attractive source of cells for large-animal preclinicaltesting. These significant findings will lead to future autolo-gous cell/gene therapy studies comparing easily cultured,genetically modified, adult pMSCs to isogenic embryonic cellsderived via SCNT, thus addressing cell rejection issues innonmurine models for disease and tissue repair.

ACKNOWLEDGMENT

We would like to thank Julie Nelson from the Center for Tropical andEmerging Global Diseases Flow Cytometry Facility at the University ofGeorgia for her assistance with flow cytometric analysis.

REFERENCES

1. Friedenstein AJ, Chailakhyan RK, Latsinik NV, Panasyuk AF, Keiliss-Borok IV. Stromal cells responsible for transferring the microenvironmentof the hemopoietic tissues. Cloning in vitro and retransplantation in vivo.Transplantation 1974; 17:331–340.

2. Beresford JN. Osteogenic stem cells and the stromal system of bone andmarrow. Clin Orthop Relat Res 1989; 240:270–280.

3. Owen M. Marrow stromal stem cells. J Cell Sci Suppl 1988; 10:63–76.4. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD,

APPLICATIONS OF PORCINE MSCS IN TRANSGENESIS 55

Moorman MA, Simonetti DW, Craig S, Marshak DR. Multilineagepotential of adult human mesenchymal stem cells. Science 1999; 284:143–147.

5. Kohyama J, Abe H, Shimazaki T, Koizumi A, Nakashima K, Gojo S, TagaT, Okano H, Hata J, Umezawa A. Brain from bone: efficient ‘‘meta-differentiation’’ of marrow stroma-derived mature osteoblasts to neuronswith Noggin or a demethylating agent. Differentiation 2001; 68:235–244.

6. Woodbury D, Schwarz EJ, Prockop DJ, Black IB. Adult rat and humanbone marrow stromal cells differentiate into neurons. J Neurosci Res 2000;61:364–370.

7. Wakitani S, Saito T, Caplan AI. Myogenic cells derived from rat bonemarrow mesenchymal stem cells exposed to 5-azacytidine. Muscle Nerve1995; 18:1417–1426.

8. Makino S, Fukuda K, Miyoshi S, Konishi F, Kodama H, Pan J, Sano M,Takahashi T, Hori S, Abe H, Hata J, Umezawa A, et al. Cardiomyocytescan be generated from marrow stromal cells in vitro. J Clin Invest 1999;103:697–705.

9. Friedenstein AJ, Gorskaja JF, Kulagina NN. Fibroblast precursors innormal and irradiated mouse hematopoietic organs. Exp Hematol 1976; 4:267–274.

10. Simmons DJ, Seitz P, Kidder L, Klein GL, Waeltz M, Gundberg CM,Tabuchi C, Yang C, Zhang RW. Partial characterization of rat marrowstromal cells. Calcif Tissue Int 1991; 48:326–334.

11. Martin DR, Cox NR, Hathcock TL, Niemeyer GP, Baker HJ. Isolation andcharacterization of multipotential mesenchymal stem cells from felinebone marrow. Exp Hematol 2002; 30:879–886.

12. Huss R, Hoy CA, Deeg HJ. Contact- and growth factor-dependent survivalin a canine marrow-derived stromal cell line. Blood 1995; 85:2414–2421.

13. Ringe J, Kaps C, Schmitt B, Buscher K, Bartel J, Smolian H, Schultz O,Burmester GR, Haupl T, Sittinger M. Porcine mesenchymal stem cells.Induction of distinct mesenchymal cell lineages. Cell Tissue Res 2002;307:321–327.

14. Chamberlain JR, Schwarze U, Wang PR, Hirata RK, Hankenson KD, PaceJM, Underwood RA, Song KM, Sussman M, Byers PH, Russell DW.Gene targeting in stem cells from individuals with osteogenesisimperfecta. Science 2004; 303:1198–1201.

15. Horwitz EM, Prockop DJ, Fitzpatrick LA, Koo WW, Gordon PL, Neel M,Sussman M, Orchard P, Marx JC, Pyeritz RE, Brenner MK. Trans-plantability and therapeutic effects of bone marrow-derived mesenchymalcells in children with osteogenesis imperfecta. Nat Med 1999; 5:309–313.

16. Studeny M, Marini FC, Champlin RE, Zompetta C, Fidler IJ, Andreeff M.Bone marrow-derived mesenchymal stem cells as vehicles for interferon-beta delivery into tumors. Cancer Res 2002; 62:3603–3608.

17. Nakamizo A, Marini F, Amano T, Khan A, Studeny M, Gumin J, Chen J,Hentschel S, Vecil G, Dembinski J, Andreeff M, Lang FF. Human bonemarrow-derived mesenchymal stem cells in the treatment of gliomas.Cancer Res 2005; 65:3307–3318.

18. Kato Y, Imabayashi H, Mori T, Tani T, Taniguchi M, Higashi M,Matsumoto M, Umezawa A, Tsunoda Y. Nuclear transfer of adult bonemarrow mesenchymal stem cells: developmental totipotency of tissue-specific stem cells from an adult mammal. Biol Reprod 2004; 70:415–418.

19. Denning C, Priddle H. New frontiers in gene targeting and cloning:success, application, and challenges in domestic animals and humanembryonic stem cells. Reproduction 2003; 126:1–11.

20. Bosch P, Hodges CA, Stice SL. Generation of transgenic livestock bysomatic cell nuclear transfer. Biotecnologia Aplicada 2004; 21:128–136.

21. Arat S, Gibbons J, Rzucidlo SJ, Respess DS, Tumlin M, Stice SL. In vitrodevelopment of bovine nuclear transfer embryos from transgenic clonallines of adult and fetal fibroblast cells of the same genotype. Biol Reprod2002; 66:1768–1774.

22. Arat S, Rzucidlo SJ, Gibbons J, Miyoshi K, Stice SL. Production oftransgenic bovine embryos by transfer of transfected granulosa cells intoenucleated oocytes. Mol Reprod Dev 2001; 60:20–26.

23. Lee GS, Kim HS, Hyun SH, Lee SH, Jeon HY, Nam DH, Jeong YW, KimS, Kim JH, Han JY, Ahn C, Kang SK, et al. Production of transgeniccloned piglets from genetically transformed fetal fibroblasts selected bygreen fluorescent protein. Theriogenology 2005; 63:973–991.

24. Gong G, Dai Y, Fan B, Zhu H, Zhu S, Wang H, Wang L, Tang B, Li R,Wan R, Liu Y, Huang Y, et al. Birth of calves expressing the enhancedgreen fluorescent protein after transfer of fresh or vitrified/thawedblastocysts produced by somatic cell nuclear transfer. Mol Reprod Dev2004; 69:278–288.

25. Uhm SJ, Kim NH, Kim T, Chung HM, Chung KH, Lee HT, Chung KS.Expression of enhanced green fluorescent protein (EGFP) and neomycinresistant (Neo(R)) genes in porcine embryos following nuclear transferwith porcine fetal fibroblasts transfected by retrovirus vector. Mol ReprodDev 2000; 57:331–337.

26. Park KW, Cheong HT, Lai L, Im GS, Kuhholzer B, Bonk A, Samuel M,Rieke A, Day BN, Murphy CN, Carter DB, Prather RS. Production ofnuclear transfer-derived swine that express the enhanced green fluorescentprotein. Anim Biotechnol 2001; 12:173–181.

27. Lai L, Park KW, Cheong HT, Kuhholzer B, Samuel M, Bonk A, Im GS,Rieke A, Day BN, Murphy CN, Carter DB, Prather RS. Transgenic pigexpressing the enhanced green fluorescent protein produced by nucleartransfer using colchicine-treated fibroblasts as donor cells. Mol ReprodDev 2002; 62:300–306.

28. Hofmann A, Zakhartchenko V, Weppert M, Sebald H, Wenigerkind H,Brem G, Wolf E, Pfeifer A. Generation of transgenic cattle by lentiviralgene transfer into oocytes. Biol Reprod 2004; 71:405–409.

29. Hirata R, Chamberlain J, Dong R, Russell DW. Targeted transgeneinsertion into human chromosomes by adeno-associated virus vectors. NatBiotechnol 2002; 20:735–738.

30. Hirata RK, Xu C, Dong R, Miller DG, Ferguson S, Russell DW. EfficientPRNP gene targeting in bovine fibroblasts by adeno-associated virusvectors. Cloning Stem Cells 2004; 6:31–36.

31. Mitani K, Wakamiya M, Hasty P, Graham FL, Bradley A, Caskey CT.Gene targeting in mouse embryonic stem cells with an adenoviral vector.Somat Cell Mol Genet 1995; 21:221–231.

32. Xia H, Mao Q, Paulson HL, Davidson BL. siRNA-mediated genesilencing in vitro and in vivo. Nat Biotechnol 2002; 20:1006–1010.

33. Mackay AM, Beck SC, Murphy JM, Barry FP, Chichester CO, PittengerMF. Chondrogenic differentiation of cultured human mesenchymal stemcells from marrow. Tissue Eng 1998; 4:415–428.

34. Gibbons J, Arat S, Rzucidlo J, Miyoshi K, Waltenburg R, Respess D,Venable A, Stice S. Enhanced survivability of cloned calves derived fromroscovitine-treated adult somatic cells. Biol Reprod 2002; 66:895–900.

35. SAS. SAS User’s Guide: Statistics. Cary, NC: SAS Institute, Inc.; 1985.36. Friedenstein AJ, Deriglasova UF, Kulagina NN, Panasuk AF, Rudakowa

SF, Luria EA, Ruadkow IA. Precursors for fibroblasts in differentpopulations of hematopoietic cells as detected by the in vitro colony assaymethod. Exp Hematol 1974; 2:83–92.

37. Friedenstein AJ, Chailakhjan RK, Lalykina KS. The development offibroblast colonies in monolayer cultures of guinea-pig bone marrow andspleen cells. Cell Tissue Kinet 1970; 3:393–403.

38. Kassem M. Mesenchymal stem cells: biological characteristics andpotential clinical applications. Cloning Stem Cells 2004; 6:369–374.

39. Mazzone A, Ricevuti G. Leukocyte CD11/CD18 integrins: biological andclinical relevance. Haematologica 1995; 80:161–175.

40. Castro-Malaspina H, Gay RE, Resnick G, Kapoor N, Meyers P, ChiarieriD, McKenzie S, Broxmeyer HE, Moore MA. Characterization of humanbone marrow fibroblast colony-forming cells (CFU-F) and their progeny.Blood 1980; 56:289–301.

41. Rempel D, Abrahamsson SO. The effects of reduced oxygen tension oncell proliferation and matrix synthesis in synovium and tendon explantsfrom the rabbit carpal tunnel: an experimental study in vitro. J Orthop Res2001; 19:143–148.

42. Wang DW, Fermor B, Gimble JM, Awad HA, Guilak F. Influence ofoxygen on the proliferation and metabolism of adipose derived adult stemcells. J Cell Physiol 2005; 204:184–191.

43. Studer L, Csete M, Lee SH, Kabbani N, Walikonis J, Wold B, McKay R.Enhanced proliferation, survival, and dopaminergic differentiation of CNSprecursors in lowered oxygen. J Neurosci 2000; 20:7377–7383.

44. Bradley TR, Hodgson GS, Rosendaal M. The effect of oxygen tension onhemopoietic and fibroblast cell proliferation in vitro. J Cell Physiol 1978;97:517–522.

45. Taylor WG, Richter A, Evans VJ, Sanford KK. Influence of oxygen andpH on plating efficiency and colony development of WI-38 and Vero cells.Exp Cell Res 1974; 86:152–156.

46. Nevo Z, Beit-Or A, Eilam Y. Slowing down aging of cultured embryonalchick chondrocytes by maintenance under lowered oxygen tension. MechAgeing Dev 1988; 45:157–165.

47. Lennon DP, Edmison JM, Caplan AI. Cultivation of rat marrow-derivedmesenchymal stem cells in reduced oxygen tension: effects on in vitro andin vivo osteochondrogenesis. J Cell Physiol 2001; 187:345–355.

48. Fink T, Abildtrup L, Fogd K, Abdallah BM, Kassem M, Ebbesen P,Zachar V. Induction of adipocyte-like phenotype in human mesenchymalstem cells by hypoxia. Stem Cells 2004; 22:1346–1355.

49. Ivanov VO, Ivanova SV, Niedzwiecki A. Ascorbate affects proliferation ofguinea-pig vascular smooth muscle cells by direct and extracellularmatrix-mediated effects. J Mol Cell Cardiol 1997; 29:3293–3303.

50. Hata R, Senoo H. L-Ascorbic acid 2-phosphate stimulates collagenaccumulation, cell proliferation, and formation of a three-dimensionaltissue-like substance by skin fibroblasts. J Cell Physiol 1989; 138:8–16.

51. Chepda T, Cadau M, Girin P, Frey J, Chamson A. Monitoring of ascorbate

56 BOSCH ET AL.

at a constant rate in cell culture: effect on cell growth. In Vitro Cell DevBiol Anim 2001; 37:26–30.

52. Kurata S, Senoo H, Hata R. Transcriptional activation of type I collagengenes by ascorbic acid 2-phosphate in human skin fibroblasts and itsfailure in cells from a patient with a2(I)-chain-defective Ehlers-Danlossyndrome. Exp Cell Res 1993; 206:63–71.

53. Phillips CL, Tajima S, Pinnell SR. Ascorbic acid and transforming growthfactor-b1 increase collagen biosynthesis via different mechanisms:coordinate regulation of pro-a1(I) and pro-a1(III) collagens. ArchBiochem Biophys 1992; 295:397–403.

54. Ono M, Aratani Y, Kitagawa I, Kitagawa Y. Ascorbic acid phosphatestimulates type IV collagen synthesis and accelerates adipose conversionof 3T3-L1 cells. Exp Cell Res 1990; 187:309–314.

55. Buettner GR. Ascorbate autoxidation in the presence of iron and copperchelates. Free Radic Res Commun 1986; 1:349–353.

56. Buettner GR, Jurkiewicz BA. Catalytic metals, ascorbate, and freeradicals: combinations to avoid. Radiat Res 1996; 145:532–541.

57. Ishizaka S, Shiroi A, Kanda S, Yoshikawa M, Tsujinoue H, Kuriyama S,

Hasuma T, Nakatani K, Takahashi K. Development of hepatocytes from

ES cells after transfection with the HNF-3b gene. FASEB J 2002; 16:

1444–1446.

58. Gratsch TE, O’Shea KS. Noggin and chordin have distinct activities in

promoting lineage commitment of mouse embryonic stem (ES) cells. Dev

Biol 2002; 245:83–94.

59. Gaggar A, Shayakhmetov DM, Lieber A. CD46 is a cellular receptor for

group B adenoviruses. Nat Med 2003; 9:1408–1412.

60. Rideout WM III, Eggan K, Jaenisch R. Nuclear cloning and epigenetic

reprogramming of the genome. Science 2001; 293:1093–1098.

61. Rideout WM III, Wakayama T, Wutz A, Eggan K, Jackson-Grusby L,

Dausman J, Yanagimachi R, Jaenisch R. Generation of mice from wild-

type and targeted ES cells by nuclear cloning. Nat Genet 2000; 24:109–

110.

62. Zhu H, Craig JA, Dyce PW, Sunnen N, Li J. Embryos derived from

porcine skin-derived stem cells exhibit enhanced preimplantation de-

velopment. Biol Reprod 2004; 71:1890–1897.

APPLICATIONS OF PORCINE MSCS IN TRANSGENESIS 57