Derek J Rosario1, Gwendolen C Reilly2, Emadaldeen Ali Salah1, Maggie Glover1, Anthony J Bullock2 & Sheila MacNeil2,3

Author for correspondence1Academic Urology Unit, University of Sheffield, Clinical Sciences (South), K Floor, Royal Hallamshire Hospital, Sheffield S10 2JF, UK2Department of Engineering Materials, The Kroto Research Institute, North Campus, University of Sheffield, S3 7HQ,UK3Tel.: +44 114 222 5995;Fax: +44 114 222 5945;E-mail: s.macneil@shef.ac.uk

le to iden- delam-larization, The aims

eristics ofeffects ofpart of

Keywords: bioscaffold, bladder, natural matrix, tissue engineering, urinary bladder matrix

bladder provides the best model for the humanbladder [4]. Porcine bladders are readily availableat a suitable scale for human reconstructiveapplications and decellularized derivatives havebeen used in the clinical setting without induc-ing immune rejection [5]. Initial clinical dataexist to suggest that UBM can provide good sup-port for tissue regeneration [6]. UBM has been

From the literature to date, we were abtify key steps in the processing of UBMination, de-epithelialization, decelludisinfection and terminal sterilization.of the current study were: To assess the mechanical charact

whole porcine bladder and the processing to UBM on these;Decellulariurinary blathe lower u10.2217/17460751.3.2.145 2and bladder stromal cells. Terminal sterilization with ethylene oxide resulted in considerable stiffening of the matrix simultaneous coculture and layered seeding of scaffolds with stromal cells followed by epithelial cells gave similar results with good initial urothelial attachment (followed by loss of cells later) and slow stromal cell penetration. Conclusion: We describe a decellularized sterilized porcine UBM with acceptable mechanical properties that shows promise as a scaffold for producing an in vitro tissue-engineered bladder patch material for lower urinary tract reconstruction. Future work now needs to focus on the conditions for achieving secure epithelial attachment.

Porcine urinary bladder matrix (UBM) is anextracellular matrix (ECM)-derived bioscaffoldthat has been used in bladder reconstruction aswell as other applications [1]. The urinary blad-der is unique, capable of accommodating largechanges in intravesical volume with little changein wall tension and pressure. Functional studiessuggest that bladder compliance comes predomi-nantly from intrinsic properties of the detrusorsmooth muscle cells, with the bladder containingrelatively little elastin [2,3]. Within disease states,changes in collagen content and type are associ-ated with loss of compliance, function and, inthe worst case scenario, renal failure from highbladder pressures [2].

From a functional perspective, the porcine

used both as an acellular scaffold to promoteremodeling [7] and seeded with autologous cellsfor bladder repair [6].

Irrespective of application, the tissue must bemade acellular and sterile for clinical use. Anumber of approaches have been described. Onemethod aims to preserve functional homology toreplace bladder tissue with a full-thicknessmatrix in a like with like fashion [8]. Othersfavor mechanical delamination of the tissueswith removal of the muscularis detrusor to makefurther processing more straightforward. A vari-ety of methods have also been reported for thedisinfection, decellularization and sterilization ofUBM [9]. Any processing may result in damageto the ECM rendering it not fit for purpose.RESEARCH ARTICLE

zation and sterilization of porcine dder matrix for tissue engineering in rinary tractBackground: Several synthetic and natural matrices have been described for tissue engineering of bladder but there is little information on the effects of processing on their subsequent mechanical performance or interaction with human cells. Aim: Our aim was to assess the effects of delamination, decellularization and sterilization on the mechanical properties of porcine urinary bladder matrix (UBM) and to then assess the ability of the UBM to act as a scaffold for reconstruction with human bladder cells. Methods: A total of 20 porcine bladders were assessed before and after mechanical delamination and four porcine bladders were followed at every stage through a comparison of several decellularization and terminal sterilization methodologies examining histological and mechanical characteristics. The sterile UBM was seeded with normal human urothelial and bladder stromal cells either as a simultaneous coculture, or with stromal cells followed by urothelial cells. Results: Mechanical delamination, physical rinsing of the resulting bladder stroma in hypotonic buffer, 0.1% sodium dodecyl sulfate solution and 0.1% peracetic acid resulted in an UBM with acceptable mechanical properties capable of supporting urothelial 008 Future Medicine Ltd ISSN 1746-0751 Regen. Med. (2008) 3(2), 145156 145

k.rowlandText BoxFor reprint orders, please contact:reprints@futuremedicine.com

RESEARCH ARTICLE Rosario, Reilly, Ali Salah, Glover, Bullock & MacNeil

146 To compare the efficacy of different protocolsfor decellularization and terminal sterilizationof porcine bladder tissue to create an acellularsterile matrix suitable for supporting the cultureof human bladder stromal and urothelial cells;

To evaluate the suitability of the sterile UBMgenerated to support in vitro coculture of nor-mal human urothelial (NHU) and bladderstromal cells.

Methods & materialsSource of tissuesWhole bladders harvested from freshly slaugh-tered market-weight English white pigs weredecontaminated by rinsing in 10% iodine solutionfor 5 min and transported in phosphate-bufferedsaline (PBS) containing penicillin 100 IU/ml,streptomycin 100 g/ml and amphotericin0.625 g/ml (Gibco/Invitrogen, UK). Followingmechanical delamination to separate the detrusormuscle coats from the lamina propria and urothe-lium (Figure 1), the inner layer (UBM) was dividedinto 4 3 cm pieces, noting the bladder of originand the orientation of the long axis of the sampleto the long axis of the bladder. These were refriger-ated overnight at 4C in hypotonic buffer consist-ing of Tris-HCl 10mM (pH 8.0) with aprotonin10 KIU ml-1 and 0.1% EDTA.

Processing protocolsDecellularization was commenced the day fol-lowing harvest. Two protocols for UBM prepara-tion were compared (Figure 2). The first(method 1) involved mechanical delaminationfollowed by the decellularization and steriliza-tion protocol, described for porcine heart valveleaflets [10]. The second (method 2) replaced thedetergent (0.1% sodium dodecyl sulfate [SDS])used in method 1 with a 1 M hyperosmolar solu-tion of NaCl, previously described as an effectivedecellularization agent [11]. UBM was placed ineither 0.1% SDS, made up in hypotonic solu-tion of Tris-HCl ,or in 1 M NaCl. The tissueswere maintained on a rocker at room tempera-ture for 24 h. Nuclease digestion at 37C fol-lowed using deoxyribonuclease I (Sigma, UK)and ribonuclease A (USB) at a final concentra-tion of 50 and 1 Uml-1, respectively, in a filter-sterilized reaction buffer of Tris-HCl (pH 7.5)with added calcium, magnesium and bovineserum albumin. Subsequent sterilization by eth-ylene oxide or by immersion in 0.1% peraceticacid (PAA) solution titrated to pH 7.0 at room

Biomechanical testingAll measurements were performed in air at roomtemperature (21.0 1.5C). Uniaxial biome-chanical testing was carried out using the Elec-troforce (ELF) 3200 materials testing instrument(Bose, USA). Tissue samples were suspendedbetween clamps (Figure 3A) and a single rampedstrain applied from 0 to 6 mm deflection.Stressstrain curves were derived from the load-displacement curves generated and Youngs mod-ulus of elasticity (E) was calculated from the lin-ear (collagen) part of the stressstrain curve(Figure 3B). Samples that slipped in the grips priorto failure were not included in the analysis.

Whole bladderStrips of 3 1 cm of full-thickness bladder tissuewere taken with either a longitudinal or trans-verse orientation from four bladders. Half weresoaked in PBS and the other half in cromakalim(Sigma Aldrich, Dorset, UK), 10-5 M solution,for 30 min at room temperature immediatelyprior to testing [12].

Urinary bladder matrixStrips of 2 1 cm of UBM were taken from eachof the four bladders with either a longitudinal ortransverse orientation and kept moist in PBS.Between five and seven strips were assessed fromeach bladder at each step of processing. The ori-entation of the long axis of the strip with respectto the long axis of the urinary bladder was noted.

Decellularity & sterility verificationThe degree of acellularity was assessed by histolog-ical evaluation (hematoxylin and eosin [H&E]) ateach step. Immunohistochemical staining forbasement membrane proteins was carried out forcollagen IV (cIV22, Dako, UK) and laminin(lam89, Dako). Quantitative evaluation of resid-ual DNA was measured by photospectroscopy. Atotal of three strips of UBM prior to and follow-ing decellularization were air dried to constantweight, minced and any nucleic acid present wasextracted and quantified using a QIAamp DNAmini kit (Qiagen, Germany). The concentrationof DNA was calculated by measuring the absorb-ance at 260 nm (OD260) and 280 nm (OD280).Sterility was tested by incubation in media (Dul-beccos Modified Eagles Medium [DMEM]) sup-plemented with 10% foetal calf serum (FCS) for4 weeks at 37C and on Columbia blood agar for72 h. Any turbidity or discoloration of themedium or growth on blood agar was taken asRegen. Med. (2008) 3(2) future science groupfuture science group

temperature for either 3 or 5 h was carried out. evidence of infection.

Porcine urinary bladder matrix for tissue engineering RESEARCH ARTICLE

future science groupfuture science groupCytotoxicity assayTo determine whether the processed UBM con-tained any leachable toxins, media conditionedwith UBM was produced by adding a 5 5 cmsheet of sterile UBM (whole or diced) to 75 mlDMEM and left overnight on a shaker. The

following day the medium was removed and filter-sterilized through a 0.2 m pore filter. Immediatelyprior to use, the medium was supplemented with10% FCS. As a control, nonconditioned mediumwas prepared in a similar way but without UBM.To compare the viability of bladder stromal cells

Figure 1. Preparation of urinary bladder matrix from whole bladder.

This figure shows the mechanical separation of detrusor muscle from lamina propria and the urothelial sheet. (D) shows that the orientation of the strips taken for mechanical testing were maintained with a longitudinal or transverse orientation. (E) shows hematoxylin and eosin staining of urinary bladder matrix (UBM; preprocessing). (F) shows collagen IV staining of UBM.

A B

C D

E F147www.futuremedicine.com

RESEARCH ARTICLE Rosario, Reilly, Ali Salah, Glover, Bullock & MacNeil

148grown in control or UBM-conditioned media3-[4,5-dimethyl(thiazol-2yl)-3,5-diphery] tetrazo-lium bromide (MTT) assays were used. A total of200 l of stromal cell suspension in controlmedium at 2 104 cells/ml, 200 l was added tothe individual wells of 96-well plates and left for2 h to attach. The medium was then replaced witheither conditioned or nonconditioned medium inreplicates of six plates and incubated at 37C in ahumidified atmosphere of 10% CO2 in air for4 days. A single plate was removed daily to assesscell viability with the MTT assay until the end ofthe experiment at day 4. MTT (200 l, 0.5 mg/ml)was added to each well daily and left to incubatefor 4 h at 37C. Formazan crystals were dissolvedusing 2-ethoxyethanol (Cellosolve) and absorbanceread at 570 nm on a plate reader (Biotek EL X800,BioTek Instruments, Inc., VT, USA).

Tissue-engineered bladder assembly Biopsies of normal human urinary tract tissues,obtained with informed consent and Ethics Com-mittee approval, from men undergoing urological

surgery were transported (Hanks balanced saltsolution [HBSS] containing HEPES 10 mmol/lpH 7.6 and aprotinin 20 KIU/ml) and subse-quently dissected and incubated in a strippingmedium (Ca2+- and Mg2+-free HBSS, containingHEPES 10 mmol/l pH 7.6, aprotinin 20 KIU/mland 0.1% (w/v) EDTA) at 37C for 4 h to facili-tate detachment of the urothelium from thestroma [13]. NHU cell cultures were establishedand maintained in keratinocyte serum-freemedium (KSFM), containing recombinant epi-dermal growth factor and bovine pituitary extract(Invitrogen, UK) and supplemented with30 ng/ml cholera toxin [14]. The cultures weremaintained and subcultured as described bySouthgate et al. [15]. Cells were passaged when8590% confluent.

For stromal cell cultures, tissue remaining afterstripping the urothelium was finely minced,resuspended in transport medium, centrifuged at250 g for 5 min, incubated for 2 h in 100 U/mlcollagenase type IV (Sigma Aldrich, Dorset, UK)on a shaker at 37C and then recentrifuged. The

Figure 2. Schema for processing of porcine bladders.

PAA: Peracetic acid; SDS: Sodium dodecyl sulfate.

Urinary bladders from freshly slaughtered pigs (n = 4)

Full thickness strips of bladder (n = 68)

Ethylene oxidesterilization

Ethylene oxidesterilization

3 or 5 h0.1% PAA

3 or 5 h0.1% PAA

Step 3

Step 4

Step 2

Step 1

Sterility check at the end of processing

Method 1 Method 2

Mechanical delamination (n = 68)

Biomechanicaltesting aftereach step(total n = 240)

1 M NaCl andnuclease digestion

Tris-hypotonic buffer, SDS 0.1%and nuclease digestionRegen. Med. (2008) 3(2) future science groupfuture science group

Porcine urinary bladder matrix for tissue engineering RESEARCH ARTICLE

future science groupfuture science groupresulting explants were seeded into six-well Prima-ria (BD Biosciences, UK) plates in a minimumvolume of DMEM augmented with 10% FCSand incubated at 37C in a humidified atmos-phere of 10% CO2 in air. The medium was firstreplaced at day 4 and thereafter every alternateday. Cells were passaged at 8590% confluence.

Pieces of sterile UBM measuring 2 2 cmwere placed in six-well cell culture plates with thesmooth surface uppermost. A 1-cm internaldiameter sterile stainless steel ring was placed inthe center. Two different cell-seeding protocolswere compared. In the first, 3 mls of a 1:1 mix-ture of DMEM and KSFM with 5% FCS wereadded to the wells outside of the stainless steelrings. In total, 3.8 105 P3 bladder stromal cellsand 1.1 106 P4 NHU cells were suspended in0.5 ml of the combination medium above andplaced simultaneously into the center of the ring,giving a seeding density of 6.2 106 per cm2 anda ratio of 2:1. The rings were removed on day 3and half of the composites were raised to anairliquid interface.

For the second method, the UBM was seededwith 3.8 105 P3 bladder stromal cells inDMEM with 10% FCS for 48 h followed by1.1 106 P4 NHU cells. Bladder compositeswere cultured up to 14 days. Composites wereassayed with Alamar Blue (Serotec Ltd, UK) at

day 7 to test cell viability. Representative samplesof the composites were fixed on day 7 and day 14in 10% formalin, dehydrated, embedded inparaffin wax, sectioned and stained with H&E.

Data analysisData analysis was carried out using SPSS 14.0for windows. The Mann-Whitney U test wasused to detect differences between two unpairedspecimens and the KruskalWallis nonparamet-ric ANOVA was used to compare medians acrossmore than two samples. The Pearsons 2 test wasused to assess significance between groups.

ResultsMechanical testingA typical stressstrain plot of a sample of bladdermatrix loaded to failure is shown in Figure 3B.There were no consistent differences between theresults obtained from the four different bladders,therefore allowing pooling of the data from allfour. Cromakalim resulted in a reduction in elas-ticity as measured by an increase in the Youngsmodulus of elasticity (E) (p < 0.001) in the colla-gen phase (Figure 4A), while mechanical delamina-tion resulted in a more significant reduction inelasticity with an increase in E from a median of0.042.1 MPa (p < 0.0001). It is interesting tonote that whereas whole bladder strips showed

Figure 3. Biomechanical testing of strips of whole bladder and urinary bladder matrix.

(A) Showing close-up of clamps with strip of tissue between jaws. (B) Typical loaddisplacement curve

Strain (%)

Stre

ss (M

Pa)

Break point1.8 MPa140% stain

0 10 20 30 40 50 60 70 80 90 100 110 120 130 1401500.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80149www.futuremedicine.com

showing linear portion of the curve and yield point visible.

RESEARCH ARTICLE Rosario, Reilly, Ali Salah, Glover, Bullock & MacNeil

150

Figure 4. Effect of p

(A) Predelamination showNote change of scale on

Full thickness

EMPa

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00significant anisotropy, this property seemsrelated to the orientation of the muscle layers,removal of which during the processing resultedin similar mechanical properties of the matrix inany orientation (Table 1).

Neither decellularization protocol caused anychange in elastic modulus E (Figure 4B & C).Exposure to 0.1% PAA resulted in a small, butsignificant, reduction in E. However, increasingthe incubation time from 3 to 5 h made no

rocessing on mechanical properties of the bladder with and without cromokalim.

ing effect of cromokalim on smooth muscle of bladder. (B & C) Postdelamination.y-axis from (A).

Method 1 Method 2

Postm

echical delamination

Postdecellularization Tris-SDS

Post-sterilization 3 h Tris

Po

st-sterilization 5 h Tris

EMPa

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00

Full thickness

Postm

echanical delamination

1 M N

aCl postnu

clease

1 M N

aCl post-sterilization 3 h

1 M N

aCl post-sterilization 5 h

EMPa

Full thickness Full thickness with cromokalim

0.10

0.05

0.00Regen. Med. (2008) 3(2) future science groupfuture science group

Porcine urinary bladder matrix for tissue engineering RESEARCH ARTICLE

future science groupfuture science groupappreciable difference. Ethylene oxide steriliza-tion resulted in a log factor increase in E to amean of 50.5 (28.488.4 MPa; p < 0.0001,KruskalWallis ANOVA).

Decellularization protocolImmersion in hypotonic Tris buffer followed by0.1% SDS (method 1) resulted in completedecellularization with little additional histologi-cal change evident following nuclease digestion(Figure 5B & C). Immersion in 1 M NaCl(method 2) resulted in removal of the urothe-lium, however, cells were evident in the laminapropria (Figure 5B & C). Mean DNA content ofthe delaminated UBM prior to decellularizationwas 0.2 0.01 g/mg of tissue. Following 0.1%SDS treatment and PAA sterilization, meanDNA content was 0.01 0.001 g/mg.

SterilityA total of 32 pieces of UBM were placed in fil-ter-sterilized 0.1% (w/v) PAA solution. Thesehad been processed using method 1 (n = 16) ormethod 2 (n = 16) and were further split to besterilized in PAA for either 3 h (n = 16) or 5 h(n = 16). All 16 samples from processingmethod 1 were tested sterile compared with onlyfour of 16 (25%) from processing method 2(p = 0.005, Pearsons 2 test).

Tissue-engineered bladder Figures 6 & 7 show the appearance of the tissue-engineered (TE) bladder prepared on UBM atday 7 (Figures 6A, 6B, 7A & 7B) and 14 (Figures 6C,6D, 7C & 7D) following cell seeding. Penetrationof the matrix by stromal cells was poor by day 7(Figure 6B) but good by day 14 (Figure 6D). Therewas good initial attachment of multilayeredurothelial cells by day 7 (Figure 6A & B) but theattachment was less secure by day 14 (Figure 6C &D). All results indicated good initial attachmentand proliferation of urothelial cells followed bytheir subsequent detachment from the UBM.Prior seeding of stromal cells did not improvethe appearance of the TE bladder.

DiscussionWe have systematically investigated the effects ofessential steps of tissue processing, namely dela-mination, decellularization and sterilization, onthe properties of bladder tissue.

The mechanical characteristics of any tissueare dictated by the 3D arrangement and compo-sition of its ECM and the cellular compositionof the tissue. The results of this study confirmprevious functional data indicating that smoothmuscle cells contribute most to the mechanicalproperties of the bladder [12]. Cromakalim, apotassium channel activator, causes smooth

Table 1. Effects of the stages of processing on mechanical properties.

Stage of processing n Mean elastic modulus E standard error (MPa)

Full-thickness bladder: longitudinal orientation of strips 20 0.04 0.005

Full-thickness bladder: longitudinal orientation with cromokalim added

15 0.11 0.02*

Full-thickness bladder: transverse orientation of strips* 21 0.03 0.005

Full-thickness bladder: transverse orientation of strips with cromokalim added

15 0.06 0.01*

Following mechanical delamination: longitudinal 20 2.33 0.28

Following mechanical delamination: transverse 20 2.41 0.23

Following Tris and nuclease treatment 21 2.24 0.36

Following Tris and sterilization in 0.1% PAA for 3 h 21 1.56 0.21

Following Tris and sterilization in 0.1% PAA for 5 h 20 1.49 0.21

Following ethylene oxide terminal sterilization 43 50.45 2.2

The elasticity modulus (E) was calculated from the collagen phase of the matrix. The addition of cromokalim resulted in a small yet significant decrease, particularly on those strips taken from a longitudinal orientation, indicating anisotropy of the detrusor muscle bundle arrangement of the bladder. Mechanical delamination results in a log factor (~2030-fold) increase in E and loss of anisotropy. Terminal sterilization with ethylene oxide results in a further log factor (2030-fold) increase in E (reduction in elasticity). *p < 0.01(Mann-Whitney) for difference between E for longitudinal and transverse strips after addition of cromokalim. PAA: Peracetic acid.151www.futuremedicine.com

RESEARCH ARTICLE Rosario, Reilly, Ali Salah, Glover, Bullock & MacNeil

152muscle relaxation that resulted in a significantchange in biomechanical behavior of the blad-der tissue. This was more evident followingcomplete mechanical removal of the smoothmuscle coats.

Bolland et al. have described an alternativemethod to mechanical delamination usingchemical rather than mechanical means toremove smooth muscle cells [8]. The investigatorsexamined uniaxial tensile loading of fresh and

Figure 5. Histology of porcine bladder matrix.

(A) Cellularized delaminated tissue. (B) Effects of SDS visible extensively decellularized matrix. 1M NaCl has resulted in de-epithelialization but stromal cells are still visible within the matrix. (CE) Adequate decellularization with Tris/SDS/nuclease but continuing cellularity of method 2. (D & E) show the retention of collagen IV and laminin following processing.SDS: Sodium dodecyl sulfate.

b. Post 0.1% SDS / hypoosmolar treatment

b. Post 0.1% SDS / treatment

Method 1: Tris-HCl + 0.1% SDS Method 2: 1 M NaCl

Postdelamination

Post 0.1% SDS/hypo-osmolar treatment

Postnuclease

Postperacetic acid0.1% collagen IV

Postperacetic acid0.1% lamininRegen. Med. (2008) 3(2) future science groupfuture science group

Porcine urinary bladder matrix for tissue engineering RESEARCH ARTICLE

future science groupfuture science group

Figure 6. Ability of epithelial growth aafter coculture.

Urinary bladder matrix wdetailed in the methods interface for 7 and 14 dastaining for hematoxylin but poor penetration of reasonable stromal cell pHere, it can be seen that of sectioning.decellularized full-thickness bladder matrix andfound similar effects on bladder distensibilitywithout change in ultimate tensile strength. Fur-thermore, they found a degree of anisotropy asso-ciated with the tissues obtained, mechanicalresults varying according to the orientation of thetissue. We found that with whole bladder, therewas significant anisotropy (Table 1); however, fol-lowing delamination, there was no significantdifference identified. These findings are in keep-ing with those described by Freytes et al. [16],indicating that any anisotropy seen is related tothe orientation of the detrusor muscle bundlesand not to the ECM orientation in UBM. Themain advantages of using delaminated tissue aremore effective sterilization and potentially morerapid revascularization following engraftment.

The uniaxial mechanical properties of a blad-der acellular matrix graft (BAMG) derived fromvarious species have previously been described[17]. The final moduli of elasticity for BAMG of0.4, 0.6 and 1.43 MPa from pig, human and ratbladders, respectively, are of a similar order ofmagnitude to those described in the currentstudy, although the method of preparation wassomewhat different.

Prior to any clinical use, complete decellulari-zation is essential to prevent immunological reac-tion to the grafted tissue. The protocol describedin method 1 produced a completely decellular-ized matrix that was evident on light microscopyand confirmed on DNA quantification. Hyper-osmolar saline has been described as an effectivede-epithelialization agent, but was ineffective indecellularizing stromal tissues in this study. Bycontrast, 0.1% SDS proved a highly effectivedecellularization agent with minimal effects onstiffness, as indicated by Youngs modulus.

Maintenance of the basal lamina proteins waspossible using this protocol and may be of advan-tage for successful reseeding of epithelial cells.This has found to be the case for skin [11,18,19]and for buccal mucosa [20] allowing progressionof such TE skin [21] and buccal mucosa [20] intoclinical use.

A number of techniques have been investi-gated for terminal sterilization, including gammairradiation, ethylene oxide, glycerol and PAA.The 0.1% PAA solution is an effective sterilizingagent for adequately decellularized UBM, with-out having a deleterious effect on the mechanicalproperties. Our experiences in this and otherstudies lead us to suggest that tissue thickness,density and incomplete decellularization proba-bly impede ingress of PAA into the interstices ofthe matrix, resulting in an inadequate steriliza-tion [22]. Therefore, this method is only appro-priate for relatively thin matrices in which thecollagen arrangement is not overly compact.

The UBM scaffold obtained followingprocessing with 0.1% SDS followed by PAAsolution sterilization was able to support humanbladder cell growth. We found no evidence ofcytotoxicity on testing indeed the stromal cellsshowed improved proliferation when exposed inculture to the extract of the prepared UBM, sug-gesting persistence of growth factors. At 14 days,stromal cells penetrated into the matrix andurothelial cells proliferated, although thestrength of their attachment to the UBM waspoor. This may have reflected the low-calciumenvironment culture conditions that wereadopted for the initial expansion of urothelialcells. Under low-calcium conditions epithelialcells will migrate well horizontally but do notform secure attachments to the underlyingstroma, requiring the latter addition of physio-logical levels of calcium [23]. Future studies willfocus on using this UBM to produce a TEurothelium of sufficient robustness to progress to

urinary bladder matrix to support nd stromal cell penetration 7 and 14 days

as combined with urothelial cells and stromal cells as & materials section and cultured at an airliquid ys before examination by conventional histology and and eosin. (A & B) Day 7: show good epithelial growth stromal cells into the matrix. (C & D) Day 14: show enetration but relatively poor epithelial adherence. epithelial layers are detaching owing to the shear force 153www.futuremedicine.com

clinical studies.

RESEARCH ARTICLE Rosario, Reilly, Ali Salah, Glover, Bullock & MacNeil

154

Figure 7. Epithelial 14 days after initialfollowed by norma

Urinary bladder matrix wdetailed in the methods At this point, human uroto an airliquid interface histology and staining foepithelial growth but poo(C & D) Day 14: show readherence. There is no siFigures 6 & 7.

A

COther xenogeneic ECM have been describedfor bladder reconstruction, the most commonamongst these being small intestinal submucosal(SIS), which is commercially available (e.g., Sur-gisis, Cook Biotech, IN, USA). As the ECM ofan organ is secreted by the cells that populatethat organ and in turn influences the behavior ofthe cells of that organ [1], it seems intuitive thatECM of bladder origin, rather than small bowel,may be preferable for bladder repair. The colla-gen fiber arrangement of SIS confers significantanisotropy on the scaffold, with the scaffoldshowing greater stiffness and strength in the pre-ferred direction of the fibers [16]. The collagenfiber orientation of the urinary bladder shows amuch more isotropic orientation [24]. This isborne out in the current study (Table 1) by thesimilarity of modulus of elasticity of strips ofUBM taken with a longitudinal or transverse ori-entation. It would seem mechanically preferableto use ECM of bladder origin rather than smallbowel for bladder repair.

There was delay in stromal cell penetration ofthe matrix, regardless of whether the stromal

concurrent coculture was carried out. The physi-cal properties of the scaffold, such as pore size,influence cell coverage and penetration, as dem-onstrated in a study comparing different naturaland synthetic matrices [25]. In the current experi-ment, the ingress of stromal cells was slow (butdid occur) and early epithelial attachment to thematrix was good but deteriorated over time.Other groups have investigated stromal cell infil-tration of UBM and found coculture to signifi-cantly enhance stromal penetration [26]. Otherfactors, such as inter alia, calcium concentration,serum concentration and seeding density, willneed to be studied further to optimize epithelialcoverage and achieve a secure epithelial attach-ment before this could move to the clinic. In par-ticular, the effects of mechanical stimulation ofthe scaffold on cell ingress and epithelial attach-ment need assessment and this is currentlyoccuring in several laboratories, as discussed byKorossis et al. [27].

In conclusion, mechanical delamination,physical rinsing in hypotonic buffer, 0.1% SDSsolution and 0.1% PAA results in porcine UBMwith acceptable mechanical properties capable ofsupporting bladder cell culture in vitro. Thismatrix merits further investigation for lower uri-nary tract reconstruction focusing now on theconditions for adequate recellularization andformation of a secure epithelium.

Future perspectiveSignificant progress has been made in tissue engi-neering of the bladder and several approaches toproduce synthetic scaffolds, as well as naturalscaffolds, are reported. However, there is still aneed for an off-the-shelf scaffold suitable forautologous tissue repair. We suggest that porcinebladder is the most likely candidate as this isreadily available and is structurally similar tohuman bladder.

However, there are a number of factors thatmust be considered. It is not sufficient to makethis acellular and suitable for reconstitution withhuman cells, it must also be sterilized. To the bestof our knowledge the issue of sterilizing humanbladder for clinical use has not been routinelyaddressed and sterilization can impact adverselyon the mechanical properties of the tissue. Thisstudy undertakes a comparison of a number ofsuch methodologies. From these it is possible toidentify a protocol of mechanical delamination,decellularization with detergent and sterilizationwith PAA which produces a scaffold that could

growth and stromal cell penetration 7 and seeding with human bladder stromal cells, l human urothelial cells 48 h later.

as combined with human bladder stromal cells as & materials section and cultured submerged for 48 h. thelial cells were introduced and the composites raised for 7 and 14 days before examination by conventional r hematoxylin and eosin. (A & B) Day 7: show good r penetration of stromal cells into the matrix.

asonable stromal cell penetration but poor epithelial gnificant difference in the overall appearance between

B

DRegen. Med. (2008) 3(2) future science groupfuture science group

cells were seeded initially in isolation or whether be used clinically.

Porcine urinary bladder matrix for tissue engineering RESEARCH ARTICLE

future science groupfuture science group

Executive summary

Why use pig bladder f

If one wishes to repaihuman bladder and y

What are the challeng

For clinical use the pomechanical properties

It is well accepted thaand biological proper

How were bladders p

A total of 20 intact pomechanical delaminat

Four porcine bladders The histology and me

What was achieved?

Results showed that idecellularization in a h

Such stromas had acc These stromas suppor

What next?

The next stage is to nbladder patches that

Bibliography1. Badylak SF: Xenogeneic

matrix as a scaffold for treconstruction. Transpl. 12, 367377 (2004).

2. Charlton RG, Morley AGillespie JI: Focal changand connective tissue inunstable human bladder953960 (1999).

3. Murakumo M, Ushiki TMatsumura K, Shinno YThree-dimensional arranlarger defects.

Financial & competing interests disclosure The authors have no relevant affiliations or financialinvolvement with any organization or entity with a financial

tutional review board approval or have followed the princi-ples outlined in the Declaration of Helsinki for all human oranimal experimental investigations. In addition, for investi-gations involving human subjects, informed consent has beenobtained from the participants involved.

or human bladder repair?

r human bladder then there is a need for a suitable scaffold that has the physical properties close to the et is readily available pig bladder is such a candidate.

es in producing this?

rcine bladder must be rendered free of all porcine cells and be sterilized and yet retain appropriate and be fit for purpose, in other words, be capable of supporting urothelial and stromal cells. t processing of tissues to remove cells and to sterilize them may adversely affect their physical, mechanical ties.

repared?

rcine bladders were used for this study and their mechanical properties were examined before and after ion. were then followed through several protocols for removing cells and sterilization. chanical properties of the resulting bladder stroma were examined at each stage.

t was possible to produce a bladder stroma by mechanical delamination from the intact bladder, ypertonic buffer with 0.1% detergent and sterilization using 0.1% peracetic acid.

eptable mechanical properties. ted the ingrowth of stromal cells and the initial attachment of urothelial cells.

ow use these sterilized acellular pig bladder stromas to develop a reliable protocol for the production of can be investigated thoroughly in vitro before future evaluation in the clinic.

extracellular issue Immunol.

R, Chambers P, es in nerve, muscle normal and . BJU Int. 84,

, Abe K, , Koyanagi T:

and elastin fibers in the human urinary bladder: a scanning electron microscopic study. J. Urol. 154, 251256 (1995).

4. Hashitani H, Brading AF: Electrical properties of detrusor smooth muscles from the pig and human urinary bladder. Br. J. Pharmacol. 140, 146158 (2003).

5. Badylak SF: Regenerative medicine and developmental biology: the role of the extracellular matrix. Anat. Rec. B New Anat.

6. Atala A, Bauer SB, Soker S, Yoo JJ, Retik AB: Tissue-engineered autologous bladders for patients needing cystoplasty. Lancet 367, 12411246 (2006).

7. Badylak S, Meurling S, Chen M, Spievack A, Simmons-Byrd A: Resorbable bioscaffold for esophageal repair in a dog model. J. Pediatr. Surg. 35, 10971103 (2000).

8. Bolland F, Korossis S, Wilshaw SP et al.: Development and characterisation of a full-thickness acellular porcine bladder matrix for tissue engineering. Biomaterials Arguably, such natural scaffolds that are acellu-lar and sterile could also be used cell free for repairof small defects; however, the weight of evidenceappears to suggest that for repair of major defectsit is necessary to introduce urothelial cells.

In conclusion, a future perspective on the use ofporcine bladder tissues is that they can be pre-pre-pared and stored for use in bladder repair either ascell-free scaffolds for small defects or reconstitutedwith the patients laboratory expanded cells for

interest in or financial conflict with the subject matter ormaterials discussed in the manuscript. This includesemployment, consultancies, honoraria, stock ownership oroptions, expert testimony, grants or patents received orpending, or royalties.

No writing assistance was utilized in the production ofthis manuscript.

Ethical conduct of research The authors state that they have obtained appropriate insti-155www.futuremedicine.com

gement of collagen 287(1), 3641 (2005). 28, 10611070 (2007).

RESEARCH ARTICLE Rosario, Reilly, Ali Salah, Glover, Bullock & MacNeil 9. Gilbert TW, Sellaro TL, Badylak SF: Decellularization of tissues and organs. Biomaterials 27, 36753683 (2006).

10. Mirsadraee S, Wilcox HE, Korossis SA et al.: Development and characterization of an acellular human pericardial matrix for tissue engineering. Tissue Eng. 12, 763773 (2006).

11. Ghosh MM, Boyce S, Layton C, Freedlander E, MacNeil S: A comparison of methodologies for the preparation of human epidermal-dermal composites. Ann. Plast. Surg. 39, 390404 (1997).

12. Foster CD, Speakman MJ, Fujii K, Brading AF: The effects of cromakalim on the detrusor muscle of human and pig urinary bladder. Br. J. Urol. 63, 284294 (1989).

13. Kimuli M, Eardley I, Southgate J: In vitro assessment of decellularized porcine dermis as a matrix for urinary tract reconstruction. BJU Int. 94, 859866 (2004).

14. Cross WR, Eardley I, Leese HJ, Southgate J: A biomimetic tissue from cultured normal human urothelial cells: analysis of physiological function. Am. J. Physiol. Renal Physiol. 289, F459F468 (2005).

15. Southgate J, Masters JR, Trejdosiewicz LK: Culture of human urothelium. In: Culture of Epithelial Cells. Freshney RI, Freshney MG (Eds). J Wiley & Sons, Inc., NY, USA 381400 (2002).

16. Freytes DO, Badylak SF, Webster TJ, Geddes LA, Rundell AE: Biaxial strength of multilaminated extracellular matrix scaffolds. Biomaterials 25, 23532361 (2004).

17. Dahms SE, Piechota HJ, Dahiya R, Lue TF, Tanagho EA: Composition and biomechanical properties of the bladder acellular matrix graft: comparative analysis in rat, pig and human. Br. J. Urol. 82, 411419 (1998).

18. Chakrabarty KH, Dawson RA, Harris P et al.: Development of autologous human dermalepidermal composites based on sterilized human allodermis for clinical use. Br. J. Dermatol. 141, 811823 (1999).

19. Ralston DR, Layton C, Dalley AJ, Boyce SG, Freedlander E, MacNeil S: The requirement for basement membrane antigens in the production of human epidermal/dermal composites in vitro. Br. J. Dermatol. 140, 605615 (1999).

20. Bhargava S, Chapple CR, Bullock AJ, Layton C, MacNeil S: Tissue-engineered buccal mucosa for substitution urethroplasty. BJU Int. 93, 807811 (2004).

21. Sahota PS, Burn JL, Brown NJ, MacNeil S: Approaches to improve angiogenesis in tissue-engineered skin. Wound Repair Regen. 12, 635642 (2004).

22. Huang Q, Dawson RA, Pegg DE, Kearney JN, MacNeil S: Use of peracetic acid to sterilize human donor skin for production of acellular dermal matrices for clinical use. Wound Repair Regen. 12, 276287 (2004).

23. Hernon CA, Harrison CA, Thornton DJ, MacNeil S: Enhancement of keratinocyte performance in the production of tissue-engineered skin using a low-calcium medium. Wound Repair Regen. 15, 718726 (2007).

24. Badylak SF: The extracellular matrix as a biologic scaffold material. Biomaterials 28, 35873593 (2007).

25. Brehmer B, Rohrmann D, Becker C, Rau G, Jakse G: Different types of scaffolds for reconstruction of the urinary tract by tissue engineering. Urol. Int. 78, 2329 (2007).

26. Brown AL, Brook-Allred TT, Waddell JE et al.: Bladder acellular matrix as a substrate for studying in vitrobladder smooth muscle-urothelial cell interactions. Biomaterials 26, 529543 (2005).

27. Korossis S, Bolland F, Ingham E, Fisher J, Kearney J, Southgate J: Review: tissue engineering of the urinary bladder: considering structurefunction relationships and the role of mechanotransduction. Tissue Eng. 12, 635644 (2006).156 Regen. Med. (2008) 3(2) future science groupfuture science group