at SciVerse ScienceDirect

Biomaterials 34 (2013) 5776e5784

Contents lists available

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Biologic properties of surgical scaffold materials derived from dermalECM

Katherine M. Kulig a,1, Xiao Luo a,b,1, Eric B. Finkelstein a,c, Xiang-Hong Liu d,Scott M. Goldman d, Cathryn A. Sundback a,c, Joseph P. Vacanti a,c, Craig M. Neville a,c,*

aDepartment of Surgery, Center for Regenerative Medicine, Massachusetts General Hospital, Boston, MA 02114, United Statesb Tongji Hospital, Huazhong University of Science and Technology, Wuhan, People’s Republic of ChinacHarvard Medical School, Boston, MA 02115, United StatesdKensey Nash Corporation, Exton, PA 19341, United States

a r t i c l e i n f o

Article history:Received 31 December 2012Accepted 19 February 2013Available online 2 May 2013

Keywords:CollagenExtracellular matrix (ECM)BioactivityScaffoldChemotaxisDecellularization

* Corresponding author. Center for Regenerative MeCPZN-4809, Massachusetts General Hospital, BostoTel.: þ1 617 643 3379; fax: þ1 815 331 0876.

E-mail address: neville@helix.mgh.harvard.edu (C1 Equal contributions as joint first authors.

0142-9612/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.biomaterials.2013.02.055

a b s t r a c t

Surgical scaffold materials manufactured from donor human or animal tissue are increasingly being usedto promote soft tissue repair and regeneration. The clinical product consists of the residual extracellularmatrix remaining after a rigorous decellularization process. Optimally, the material provides bothstructural support during the repair period and cell guidance cues for effective incorporation into theregenerating tissue. Surgical scaffold materials are available from several companies and are uniqueproducts manufactured by proprietary methodology. A significant need exists for a more thorough un-derstanding of scaffold properties that impact the early steps of host cell recruitment and infiltration. Inthis study, a panel of in vitro assays was used to make direct comparisons of several similar,commercially-available materials: Alloderm, Medeor Matrix, Permacol, and Strattice. Differences in thematerials were detected for both cell signaling and scaffold architecture-dependent cell invasion.Material-conditioned media studies found Medeor Matrix to have the greatest positive effect upon cellproliferation and induction of migration. Strattice provided the greatest chemotaxis signaling and bestsuppressed apoptotic induction. Among assays measuring structure-dependent properties, MedeorMatrix was superior for cell attachment, followed by Permacol. Only Alloderm and Medeor Matrixsupported chemotaxis-driven cell invasion beyond the most superficial zone. Medeor Matrix was theonly material in the chorioallantoic membrane assay to support substantial cell invasion. These resultsindicate that both biologic and structural properties need to be carefully assessed in the considerableongoing efforts to develop new uses and products in this important class of biomaterials.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Biological scaffolds generated from the residual extracellularmatrix (ECM) after tissue decellularization are increasingly beingused to facilitate soft tissue repair in a variety of clinical settings [1e5]. In addition to providing structural support, the matrix can act asan instructive template to induce regenerative processes [6e10]. Amajor motivation in developing this novel class of materials was toprovide a resorbable product that allows for regeneration with lessinflammation and scar formation than current synthetic options,their significant regenerative-inducing properties are just

dicine, 185 Cambridge Street,n, MA 02114, United States.

.M. Neville).

All rights reserved.

beginning to be appreciated and mechanisms explored. Severaltissues are utilized as initial processing sources to generate clinicalproducts, including heart valve, bladder, pericardium, brain, peri-toneum, mesothelium, small intestine, and dermis [11e20]; donorspecies include bovine, porcine, equine, and human. Many pro-cessing steps are required and may include chemical decellulari-zation using detergents, alkali and acidic solutions, proteolyticenzymes, thermal treatment and mechanical agitation, and sub-sequent cross-linking, dehydration, and sterilization [21e24]. Pro-prietary protocols have been developed to generate uniquematerials for improved soft tissue regeneration.

No single surgical scaffold will be optimal for every clinicalapplication. However certain physical and biological properties areconsidered to support scaffold incorporation and tissue repair.Physical properties reported by scaffold manufacturers includeburst and tensile strength, suture retention, and thickness. Inaddition, biologic properties of ECM-derivedmaterials also strongly

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e5784 5777

impact the type and degree of immune response, recruitment andinfiltration of host cells, and eventual incorporation and remodel-ing of the scaffold into regenerated tissue. Scaffold incorporationand remodeling are lengthy processes, requiring several weeks ormonths and results may be suboptimal. Thus, considerable interestexists in developing a better understanding of scaffold parametersthat impact cellular incorporation and quality of outcome. In thepresent study, a panel of in vitro assays was used to investigateseveral similar commercially-available and clinically-used surgicalscaffold products, all derived from the ECM of dermal tissue. Theimpact of these materials on the early steps of host cell recruitmentand infiltration was explored as these properties are essential forsuccessful scaffold integration and tissue regeneration.

2. Materials and methods

2.1. Overview of experimental design

Several products derived from dermal tissue were evaluated for intrinsic prop-erties that support cellular infiltration and remodeling (Table 1).MedeorMatrix (alsodistributed as XCM Biologic and Arthrex DX) was supplied by the manufacturer.Permacol, Alloderm, and Strattice are commercially-available and were received intheir original packaging. All materials were stored and handled according to anysupplied recommendations, and were used well within the indicated shelf lives.Rehydration and rinsing were performed according to manufacturers’ instructions.Gelfoam Sponge (Pfizer, New York) is a denatured, foamed product made of highlyprocessed, denatured collagen from porcine dermis. Although developed as anabsorbable hemostatic device, it is commonly used as a cell scaffold in tissue engi-neering studies andwas includedas a control. The invitroassays included in thepanelwere chosen to reflect basic biological processes thought tobe important in soft tissueregeneration and defect repair. Specific cell types were chosen for documentedsensitivity for the particular assay and resistance to confounding effects. Preliminaryexperiments were performed, usually comparing serum-containing and serum-freemedium, to validate cell type choices and experimental parameters (data notshown). A minimum of 3 technical replicates were analyzed for each material, witheach experiment performed at least twice. Whenever possible, experiments wereperformed in a blinded manner to minimize experimental bias.

2.2. Preparation of conditioned medium

All materials were processed in parallel in a blinded fashion to minimize po-tential differences in processing. Materials received in a dehydrated state wereweighed and subsequently choppedwith a sterile razor blade in a tissue culture dishto maximize surface area. The resulting fine powder was added to serum-freeDMEM medium with 1% Pen/Strep (both from Invitrogen, Carlsbad, CA) (25 mgpowder/mL medium) in a multiwell dish and placed on a rotator (30 rpm) in anincubator at 37 �C and 5% CO2 for 24 h. Additional control medium was processedsimultaneously so that potential impact upon pH and component stability would beidentical. Conditioning material was removed by processing medium through a lowbinding cellulose acetate spin column (Millipore, Danvers, MA) at 16,000� g for5 min. The conditioning process was modified for materials received in a hydratedstate. These prehydrated materials contained approximately 75% liquid by weightafter removal from packaging. To achieve a consistent ECM solids/medium ratioamong these materials prior to chopping, pieces of these material were centrifugedin a cellulose acetate spin column at 16,000� g for 10 min to remove excess liquid.Each material with some retained liquid was then chopped and mediumwas addedin a mass/volume ratio of 100 mg material/0.75 mL medium. The conditioning stepwas carried out as described for the dehydrated materials. Afterwards, the removedconditioning material was dehydrated and weighed, and the volume of the condi-tioned medium was corrected to a dry mass/volume ratio of 25 mg/mL.

Table 1Dermis-derived surgical scaffold materials.

Material Source Preparation Cross-linking Company

Alloderm human Rehydrated, 40 min No LifeCellGelfoama porcine Rehydrated, 2 min No PfizerMedeor

Matrixporcine Prehydrated,

used as suppliedNo Kensey Nash

Permacol porcine Prehydrated,used as supplied

Yes Covidien

Strattice porcine Rinsed, 2 min No LifeCell

a Control material.

2.3. Cell lines

The cell lines used were chosen based upon the required properties of eachassay. The human breast adenocarcinoma epithelial cell line, MDA-MB-231 (HTB-26, American Type Culture Collection (ATCC), Manassas, VA) was propagated ingrowth medium (DMEM supplemented with 10% fetal bovine serum (SigmaeAldrich, St. Louis MO) and 1% Pen/Strep solution (Invitrogen)). Primary humanforeskin fibroblasts (HFFs) were obtained from Dr. Michael Detmar (MGH) and alsopropagated in growth medium. Human Umbilical Cord Vascular Endothelial Cells(HUVECs) were purchased from Lonza (Hopkinton, MA) and grown in EGM me-dium (Lonza).

2.4. Assays

2.4.1. Proliferation assayA suspension of HFFs was used to seed 2000 cells in 0.05 mL of complete me-

dium/chamber in 96-well dishes. The plates were incubated overnight to allow forcell attachment and recovery from the passaging-induced lag phase. Each well wasrinsed twice with PBS to remove residual serum before being supplied with 0.05 mLof material-conditioned medium. Three wells were used for each batch of condi-tioned medium. Cell starting number was verified by imaging each well at time0 with a Nikon T2000 microscope. After 24 h of incubation at 37 �C and 5% CO2, cellswere fixed with 4% paraformaldehyde, stained with the fluorescent dye DAPI,imaged and recounted.

2.4.2. Apoptosis evaluationIn our preliminary assessments, HUVECs were the cell type with the greatest

apoptotic response and an active caspase 3 signaling pathway (data not shown) andso were chosen for this assay. Opaque 96-well dishes were seeded with20 � 103 cells in 0.05 mL complete EGM. The medium was removed and the cellsrinsed with PBS after 24 h, and 0.05 mL of conditioned medium added. After 24 hfurther culturing, cells were rinsed with PBS, and caspase 3 measured in lysatesusing a chemiluminescent assay (#17-198, Millipore) per manufacturer’s directions.Wells that had continued to be serum-starved were used as positive controls forapoptosis, while serum-containingmediumwas used to demonstrate ability to avertapoptosis within the defined experimental conditions.

2.4.3. Metabolic assayMetabolic activity was measured in the conditioned media-treated HFFs after

completion of the proliferation measurements described in Section 2.4.1. An MTT-based mitochondrial activity kit was used, following the instructions of the manu-facturer (Invitrogen).

2.4.4. Scratch-wound assayFor the scratch-wound assay, MDA-MB-231 cells were harvested by a brief

exposure to trypsin and plated at a density of 25 � 103 cells/chamber of a 24-wellplate in 0.5 mL of growth medium. The following day, after the cells had attachedand grown to form confluent monolayers, single straight 1 mm cell-denuded pathswere made in each chamber with a 200 mL pipette tip and orientation marks markedon the underside. The cells were rinsed twice with PBS to remove cellular debris andgrowth medium. Each well received 0.5 mL conditioned medium and was photo-graphed at 100� magnification. The plates were incubated at 37 �C and 5% CO2 for24 h and rephotographed. Counting of cells migrating into the cleared path wasperformed in a blinded fashion. The experiment was performed twice, each withthree technical replicates; three fields were imaged and counted for each replicate.The mean number and standard deviation of migrating cells/mm2 path weredetermined for each condition.

2.4.5. Chemotaxis assayChemoTx culture plates with 8 micron pores (Neuro Probe, Gaithersburg, MD)

were used following supplied instructions to determine the presence of chemo-attractant compounds in conditioned medium. A 106 cells/mL suspension of MDA-MB-231 (labeled prior with 5 mg/mL Hoechst 33342 for 10 min) in unsupple-mented DMEM was used to seed 20 � 103 cells/chamber. Lower chambers werefilled with 0.029 mL conditioned medium and the plate was incubated at 37 �C and5% CO2. After 24 h, migrated cells were counted.

2.4.6. Cell adhesionThe day prior to the experiment, sub-confluent HFFs were removed from the

culture dish with a brief trypsin treatment and replated. This procedure enabledrapid removal of the cells on the following day with a non-enzymatic Cell RemovalBuffer (10% glycerol, 1 mM EDTA, 60 mM sodium acetate) without causing proteolyticdamage to cell adhesion proteins. The HFFs were suspended at a density of2 � 106 cells/mL in growth medium containing 5 mg/mL Hoechst 33342 for 10 min,pelleted, and resuspended at 150� 103 cells/mL in growthmedium. Threemm disksof scaffold material, cut with dermal punches (Miltex, Plainsboro, NJ) were added to1.5 mL of cell suspension in individual 2 mL microfuge tubes. The tubes were rotatedat 1 rpm for 3 h in an incubator at 37 �C and 5% CO2. The disks were removed, gentlysubmerged in PBS to remove nonadherent cells, fixed in 4% paraformaldehyde, and

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e57845778

rinsed in PBS. Fluorescent adherent cells were counted on both sides at 200�magnification.

2.4.7. Invasion assayPunched8mmdisks of each scaffoldmaterialwerefittedon topof themembrane

of cell culture inserts (Millipore, Bedford, MA). Growth medium was used as che-moattractant in the lower chamber of a 24-well plate. MDA-MB-231 cells (5 � 103)that had been serum-starved for 24 h were seeded in the upper chamber in unsup-plemented DMEM. After 24 h incubation at 37 �C, 5% CO2, the disks were removed,embedded in paraffin and sectioned. Infiltrating cells were visualized by stainingnuclei with the fluorescent dye DAPI. Stained sections were imagedwith a Nikon 80imicroscope. Enumeration was done in a blinded fashion by two individuals.

2.4.8. Chorioallantoic membrane assayFertilized white leghorn eggs (Charles River Laboratories, North Franklin, CT)

were incubated in a 38 �C, 65% relative humidity incubator. On incubation day 3, ahole was made on the pointed end of each egg with a syringe fitted with a 18 gneedle and 2e3 mL albumin was removed. The following day, a 1 cm2 window wasmade with a Dremel rotary tool to permit inspection of the embryo for viability;afterwards, the window was sealed with cellophane tape. On day 7, punched 5 mmdisks of each scaffold material were briefly hydrated in PBS. Single disks were placedon each chorioallantoic membrane and the eggs resealed with tape. At least threeeggs were used for eachmaterial and time point. Seven days after implantation, eachwindow was enlarged to permit photographing the material in situ. The materialsamples were surgically removed with surrounding membrane and fixed in 10%formalin, paraffin embedded and sectioned, and stained with hematoxylin and eosin(H&E) or DAPI.

2.5. Statistical analysis

Statistical analysis was performed comparing samples to negative controls usingthe 2-tailed Student t-test with equal variance. Co-variant ANOVA was used forcomparisons between samples. Significance level of p< 0.05 is indicated by * withinthe figures.

3. Results

Four similar extracellular matrix-derived products developedfor soft tissue repair and regeneration and manufactured by sepa-rate companies were evaluated in vitro for basic propertiesconsidered important for material incorporation and remodeling,such as retention of bioactive or matricryptic molecules andintrinsic ability to support host cell invasion.

3.1. Biomaterials

Four dermal and one baseline tissue-derived surgical scaffoldmaterials were tested (Table 1). Three of the experimental materialswere generated from porcine and one from human tissue. All cameas thin, superficially similar, white sheets. Three materials cameprehydrated, and one was desiccated. Permacol was the sole ma-terial to utilize cross-linking. Upon close inspection, most materialsdisplayed sidedness, but only Alloderm indicated appropriateorientation in instructions. All materials appeared to be a singlelayer. Gelfoam was included as a baseline reference material, andcame desiccated.

3.2. Material-conditioned media

A set of media, each conditioned with one of the materials inTable 1, was generated to evaluate the bioactivity of eluted com-ponents such as cytokines or matricryptic peptides that may haveeffects on scaffold cellularization [25,26]. In addition, residualchemicals from the decellularization process could also impactperformance [27].

3.3. Assays

3.3.1. ProliferationBoth initiation and maintenance of cell proliferation re-

quire specific extrinsic mitosis-inducing factors. Ability of eluted

components in conditionedmedia to promote cell proliferationwasevaluated (Fig. 1A). Cells supplied with serum-free medium did notdemonstrate proliferation, and in fact decreased in number due todetachment. Decreases in cell number due to detachment were alsoseen with Gelfoam, Alloderm, and Permacol-conditioned media.Only Medeor Matrix and Strattice-conditioned media supportedcontinued cell growth. Proliferationwas greatest inMedeorMatrix-conditioned medium, at almost 2/3 that of mediumwith serum.

3.3.2. ApoptosisApoptosis (programmed cell death) results from the activation

of intracellular signaling pathways as a response to stress. A varietyof internal and external molecules closely regulate the process byproviding a balance of inductive and repressive signals. HUVECswere subjected to stress through serum starvation for 24 h. Thematerial-conditioned media were evaluated for potential to alle-viate the apoptotic response bymeasuring enzymatic activity of theapoptosis marker caspase 3 (Fig. 1B). Replacing the medium withserum-containing medium largely alleviated the response as cas-pase levels decreased to approximately 1/3 of that in cells main-tained in serum-free medium. Cells grown in Alloderm-conditioned medium demonstrated the greatest caspase activity,followed by Gelfoam and Medeor Matrix. Permacol- and Strattice-conditioned media generated the lowest apoptotic responses.

3.3.3. MetabolismMedium supporting HFF cells was replaced with conditioned or

control media and mitochondrial activity was measured after 24 h(Fig. 1C) by MTT assay. Serum-free medium was used to establishbaseline activity. Gelfoamwas able to maintain appreciable activityrelative to the serum-free control. However, all four media condi-tioned with scaffold materials did considerably better than Gel-foam. Metabolic activities of cells grown in medium conditionedwith Alloderm, Medeor Matrix, and Permacol was statisticallyindistinguishable from serum-containing medium.

3.3.4. Cell mobilizationThe scratch-wound assay was used to assess the presence of

components in the conditioned media capable of stimulating cellmigration. Although the human adenocarcinoma cell line, MDA-MB-231, is considered to be highly migratory [28], it is relativelyimmobile in the absence of appropriate cytokine signaling. Serum-free medium showed very minimal ability to induce cellularmigration into the cleared path (Fig. 1D). In contrast, all material-conditioned media demonstrated considerable migration-inducing activity. Medium conditioned with Medeor Matrixdemonstrated the greatest degree of cell mobilization, with greaterthan twice the activity of other materials. Strattice, Permacol,Alloderm, and Gelfoam had similar activities.

3.3.5. ChemotaxisSpecific recruitment of mobilized cells is essential for recellu-

larization of the matrix material and initiation of tissue regenera-tion. This process is termed as chemotaxis and is invoked by asignaling molecule concentration gradient. In the absence ofchemotactic-inducing signals, seeded cells remained immobile(Fig. 1E). Addition of serum-containing medium provided chemo-attractant signals that induced cell migration. Gelfoam-conditionedmedium displayed no chemoattractant properties as it functionedsimilarly to serum-free medium. Even the small number of cellsappearing with serum-free medium, presumably due to randommigration, were absent with Alloderm-conditioned medium. BothMedeor Matrix and Strattice-conditioned media demonstratedchemoattractant activities comparable to serum-containingmedium.

0

100

200

300

400

mig

rate

d ce

lls

0.E+00

2.E-05

4.E-05

6.E-05

8.E-05

1.E-04

Abs/

cell

(540

nm)

0

20

40

60

80

100

120

cells

/mm

2

0

250

500

750

1000

Medeo

r

Matrix

Permac

ol

Allode

rm

Strattic

e

Gelfoa

m

+ seru

m

- seru

m

rela

tive

fluor

esce

nce

Chemotaxis

Metabolism Scratch-Wound

Apoptosis

0.4

0.6

0.8

1.0

1.2

1.4

1.6

fold

cha

nge

(cel

l num

ber)

Medeo

r

Matrix

Permac

ol

Allode

rm

Strattic

e

Gelfoa

m

+ seru

m

- seru

m

Medeo

r

Matrix

Permac

ol

Allode

rm

Strattic

e

Gelfoa

m

+ seru

m

- seru

m

Medeo

r

Matrix

Permac

ol

Allode

rm

Strattic

e

Gelfoa

m

+ seru

m

- seru

m

Medeo

r

Matrix

Permac

ol

Allode

rm

Strattic

e

Gelfoa

m

+ seru

m

- seru

m

*

*

*

ProliferationA B

C

E

D

Fig. 1. Material-conditionedmedia studies. Conditionedmediawere used to evaluate the ability of elutable components to induce intrinsic responses in cells. (A) Proliferation. A sparselayer of fluorescently-labeled HFFs was imaged immediately prior and 24 h after replacement of complete medium with material-conditioned medium. Change in cell number wasdetermined and fold-change is plotted. (B) Apoptosis. HUVECs were incubated in serum-free medium for 24 h. Recovery from stress-induced apoptosis after changing to conditionedmediumwas determined bymeasuring the enzymatic activity of the pro-apoptosismarker caspase 3. (C)Metabolism. Relativemitochondrialmetabolic activitywas determined inHFFsby MTT assay after 24 h of growth in conditioned medium. (D) Mobilization. Scratch-wound assay was used to evaluate induction of migration of MDA-MB-231 cells by conditionedmedia. The number of cells moving into the 1 mmwide cleared path was determined 12 h after medium replacement. (E) Chemotaxis. Conditioned media were placed in the bottomchambers of a ChemoTx plate and MDA-MB-231 cells in serum-free medium were seeded on the top. Number of cells migrating to the bottom chamber was determined after 24 h.

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e5784 5779

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e57845780

3.3.6. Cell adhesionAdhesion is required for cells to interact with scaffold material,

and is initiated by a variety of cell surface adhesion proteins andreceptors. Non-enzymatic harvesting of healthy, proliferating HFFswas used to generate a cell suspension. After a 3 h incubation withcontinuous rotation, attached cells were counted (Fig. 2). MedeorMatrix exhibited the greatest adherent property, followed by Per-macol. Strattice was similar to the highly porous Gelfoam. Allodermhad very few adherent cells (<10 cells/mm2). The two materialswith the greatest adhesion values, Permacol and Medeor Matrix,also displayed distinct sidedness with each having approximately1.8-fold difference in cell number between sides.

3.3.7. Cell invasionSupport by the scaffold architecture to facilitate cell infiltration

is critical for efficient remodeling. Serum-containing medium wasused to induce the highly migratory epithelial cell line MDA-MB-231 to invade scaffold materials placed in cell culture insertframes. Thickness of the materials ranged from approximately1 mm for Permacol to 2 mm for Alloderm (Fig. 3A). After 24 h,

120

0

20

40

60

80

100

MeAlloderm Gelfoam

cells

/mm

2

AllodermGelfoam

Medeor MatrixPermacolStrattice

Alloderm Gelfoam Me

*

Fig. 2. Cell adhesion. Tubes containing disks of material in a suspension of HFFs in completenumber of adherent cells on each side was determined. Representative photomicrographs sblack bars of graph). Scale bar ¼ 50 microns.

Permacol was the material having the greatest number of invadingcells, followed by Medeor Matrix and Alloderm (Fig. 3B). Stratticehad the fewest cells. Stained sections were divided into 10 equalhorizontal zones and the number of cells in each was determined.Percent of total infiltrating cells in each zone was determined(Fig. 3C). All cells infiltrating Permacol and Strattice remained su-perficial (in the top zone). While Medeor Matrix and Alloderm hadcells penetrating the first few zones, Medeor Matrix had cells thathad migrated through the material to the bottom surface. Gelfoamhad cells evenly distributed across the scaffold thickness.

3.3.8. Chorioallantoic membraneThe CAM assay is an established model for investigating in-

teractions of a complex developing tissuewith bioactive factors andmaterials [29]. Disks of surgical scaffold materials that had beenplaced on the developing membrane were excised after 7 daystogether with the surrounding tissue (Fig. 4A). Samples weresectioned and stained with H&E (Fig. 4B) or DAPI (Fig. 4C), and cellinfiltration determined. Gelfoam with its highly porous structure,supported the greatest cell infiltration. Cell distribution was

deor Matrix Permacol Strattice

deor Matrix Permacol Strattice

* * **

* **

mediumwere gently rotated during incubation. After 3 h, the disks were rinsed and thehowing attached cells on each side of the tested materials (top row corresponds to the

perc

ent c

ells

Alloderm StratticePermacolMedeor MatrixGelfoam

B

A

0

1000

2000

3000

4000

MedeorMatrix

PermacolAlloderm StratticeGelfoam

thic

knes

s (µ

m)

0

25

50

75

100

0

25

50

75

100

0

25

50

75

100

0

25

50

75

100

0

25

50

75

100

Top Bottom Top Bottom Top Bottom Top BottomTop Bottom

Fig. 3. Cellular invasion of material. (A) Average thickness of the materials. (B) The filters of cell culture inserts were replaced with disks of material. Cells in the serum-free mediumwere seeded in the insert; serum-containing mediumwas placed in the outside chamber as the chemoattractant. After 24 h, the number of invading cells was determined. Enumerated sections were divided into 10 equal horizontal zones and the fraction of invading cells in each wasdetermined and plotted. Representative photomicrographs are shown. Scale bar ¼ 50 microns.

K.M

.Kuliget

al./Biom

aterials34

(2013)5776

e5784

5781

Start Incubation

Day 0

Remove Albumen

Create Window

Implant Material

Harvest Material

3 4 7 14

A

B

Material Infiltration (cells/mm2) +/- SD

Medeor Matrix 344 5Permacol 41 71

Alloderm 0 0

Strattice 18 0

Gelfoam 1952 296

C

Alloderm Gelfoam Medeor Matrix Permacol Strattice

Fig. 4. CAM. (A) Time line of CAM assay. Inset: surgical scaffold material on CAM at 7 days post-implantation. At this time, materials were excised and processed for histology.Sections were stained with H&E (B) or DAPI (C) to allow for visualization and counting of infiltrating cells. Cells invading at least 100 mm were counted. Scale bar ¼ 50 microns.

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e57845782

relatively homogenous throughout all Gelfoam samples. Permacoland Strattice, with their dense architecture, had few infiltratingcells. Alloderm and Medeor Matrix had similar, generally open, lessdense structures; however, Alloderm was largely devoid of infil-trating cells while Medeor Matrix supported the greatest cellinfiltration of the tested surgical scaffold materials.

4. Discussion

In this study, four similar extracellular matrix-derived materialsdeveloped for soft tissue repair and regenerationwere evaluated ina panel of in vitro assays; their abilities to both support andmediatecell infiltration were assessed. The materials are all availablecommercially and the porcine-derived materials have the requiredFDA clearance. All are derived by decellularization of dermal tissueresulting in grossly similar materials, but the tissue source andproprietary processes of each manufacturer results in productswith unique properties and characteristics. Products based upondecellularized ECMs continue to be developed as a replacement forsynthetic surgical mesh because these materials have the potentialto facilitate cell invasion and attachment while decreasinginflammation.

The present study demonstrates that commercial ECM-derivedsurgical scaffold materials have distinct structural properties thatimpact cell infiltration and retain varying amounts of biologic activity.Although this class of materials was originally developed as an alter-native to polymer-based products that often incite significant in-flammatory responses, decellularized ECM also has the additional

potential to play an instructive role in tissue regeneration [30e35].Materials derived from many different tissues, including bladder[36,37], dermis [3e5,15],mesothelium [14], pericardium [38e40] andsmall intestine [1,37,41,42], have all demonstrated an ability to pro-mote healing and tissue regeneration. In vivo preclinical studies haveshown that implanted ECM materials promote angiogenesis andconstructive remodeling, in part by modulating macrophage-mediated inflammation [32,33,35,43]. In addition to providingstructural support, the large fibrous components of the extracellularmatrix also reversibly bind growth factors and cytokines that caninfluencemacrophage activity. These factors alsoplay important rolesin signaling nearby tissue during tissue regeneration, to inducemitosis and cell migration and recruitment. Because the factors areboth inherently labile and not immobilized by covalent linkage in thenative tissue, they are susceptible to inactivation or loss from theextracellular matrix during material processing. Furthermore, matri-cryptic peptides with significant beneficial or detrimental bioactivity[44e48]maybe generated through thebiomaterial synthesis process.Potential residual processing chemicals from the manufacturingprocess could also impact surgical scaffold material biocompatibility.To assess retention of biologic activity, we generated conditionedmedium fromeach of thematerials to evaluate on specific human celllines that are sensitive indicators for the assayed properties.

The dermal tissue-derived materials displayed a range of re-sponses in our assays. The assays were collectively designed toassess two general properties, presence of soluble cell mobilizingsubstances that could potentially signal neighboring tissue, andmaterial structure that facilitates invasion of the mobilized cells. An

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e5784 5783

additional clinical material, Gelfoam, was included in the study.Although Gelfoam is derived from porcine dermis, it is highly pu-rified and denatured collagen, and as such was not expected toretain significant biological activity. This expectationwas validated,as Gelfoam-conditioned medium supported neither proliferationnor chemotaxis, did not mitigate apoptosis, and had the lowestvalues for cell mobilization and metabolism. Its structure of inter-connected pores created by the foaming manufacturing processwas highly supportive of infiltration, with values far exceeding allother materials. On the other hand, cell attachment to Gelfoamwasmiddling, perhaps the result of being constituted of denaturedcollagen. Among the four surgical scaffold materials, mediumconditioned with Medeor Matrix best supported cell proliferation,cell migration signaling, and mitochondrial metabolism, and wassecond to Strattice in the chemotaxis assay. Strattice not onlyprovided chemotaxis signals greater than serum, but also was mosteffective in suppressing apoptotic signaling. Medeor Matrix pro-vided the best surface for promoting cell attachment. Both it andthe next best performing material, Permacol, displayed distinctsidedness. This was not necessarily disadvantageous, as even theless adhesive side of Medeor Matrix still outperformed other ma-terials. Differences in structural properties inherent in thematerialsalso greatly impacted cell infiltration. Permacol had the mostinfiltrated epithelial cells in the invasion assay, however the densestructure of both Permacol and Strattice appears to have beenhighly restrictive to more than superficial cell penetration. Cellspenetrated several hundredmicrons for both Alloderm andMedeorMatrix. Additionally, a small fraction of cells was consistently seento have completely penetrated Medeor Matrix to the bottom layerof thematerial. Because few cells were detected in the intermediateregion, transit may have been facilitated by structural aspects of thematerial such as channels formed from remnants of blood vessels,hair follicles or other physiological structures. This is an interestingfinding and future monitoring of the assay by real time fluorescentmicroscopy could provide insight. Implantation on the CAM wasalso designed to examine cell invasion, and proved to be an evenmore demanding assay. The open structure of Gelfoam, asmentioned above, proved highly supportive of cell infiltration.Among the four scaffold materials, Medeor Matrix was the mosthighly cellularized, with clusters of cells clearly visible in certainopen structural features. The othermaterials performedmarginally,with few or no infiltrating cells. It is possible that cell signalingcomponents combined with an open architecture facilitatedincorporation of cells into Medeor Matrix.

Certain caveats exist in interpreting this study. The in vitro as-says used in the present study are unlikely to fullymimic conditionsduring clinical use of the materials, but they do reflect many of thegeneral processes thought to be important for tissue regeneration.As a panel, they represent a quantitative measure of biologicalproperties that would be difficult to assess in vivo. However, thecell lines used to represent a limited number of cell types involvedin tissue regeneration. Both the absolute concentration and relativeratio with other trans-acting factors can be important for modu-lating a particular cell response and this could be different in vivo oreven in specific applications. The cell-based assays presented heredo not specifically detect the presence or amount of specific cyto-kines or factors eluted into conditioned medium. However, they dohave an advantage over ELISAs or similar biochemical assays asthey measure only the effects of factors that have maintainedbioactivity after the decellularization process. Furthermore, theidentity and activity of matricryptic peptides are incomplete, butinclude molecules such as endostatin and restin that have beenshown to be involved in normal tissue repair and regeneration[49,50]. Further validation of these findings with follow up in vivostudies is planned.

5. Conclusions

Both xenogeneic and allogeneic biologic scaffolds derived fromthe ECM of decellularized dermal tissue are increasingly used topromote the regenerative process in many reconstructive surgicalprocedures. These scaffolds are often grouped together and assumedto perform similarly. However due to the variations inmanufacturingmethods, they have altered properties that can lead to outcomedifferences. In our investigations, most pronounced were differencesin the ability to support cell infiltration; this essential requirementfosters integration of the scaffold in the regenerating tissue ratherthan transforming it into a barrier. In addition, we detected signifi-cant differences in the ability of the generated eluents to promote cellmigration, metabolism, proliferation, and chemotaxis. These resultsindicate that both biologic and structural properties need to becarefully assessed in ongoing efforts to develop new uses and prod-ucts in this important class of biomaterials.

Acknowledgments

We thank Chris Bowley for help in reviewing the protocols. XiaoLuo, M.D. was the recipient of a national scholarship from the ChinaScholarship Council.

References

[1] Smith MJ, Paran TS, Quinn F, Corbally MT. The SIS extracellular matrixscaffold-preliminary results of use in congenital diaphragmatic hernia (CDH)repair. Pediatr Surg Int 2004;20:859e62.

[2] Saray A. Porcine dermal collagen (Permacol) for facial contour augmentation:preliminary report. Aesthetic Plast Surg 2003;27:368e75.

[3] Jess P, Bulut O. Small bowel obstruction after reconstruction of the pelvic floorwith porcine dermal collagen (Permacol) after extended abdominoperinealextirpation for rectal cancer: report of two cases. Colorectal Dis 2010;12:e178e179.

[4] Glasberg SB, Light D. AlloDerm and Strattice in breast reconstruction: acomparison and techniques for optimizing outcomes. Plast Reconstr Surg2012;129:1223e33.

[5] Huston TL, Taback B, Rohde CH. Chest wall reconstruction with porcineacellular dermal matrix (Strattice) and a latissimus myocutaneous flap. AmSurg 2011;77:e115e116.

[6] Badylak SF, Brown BN, Gilbert TW, Daly KA, Huber A, Turner NJ. Biologicscaffolds for constructive tissue remodeling. Biomaterials 2011;32:316e9.

[7] Badylak SF, Freytes DO, Gilbert TW. Extracellular matrix as a biological scaffoldmaterial: structure and function. Acta Biomater 2009;5:1e13.

[8] Badylak SF. The extracellular matrix as a biologic scaffold material. Bio-materials 2007;28:3587e93.

[9] Flynn LE, Prestwich GD, Semple JL, Woodhouse KA. Proliferation and differ-entiation of adipose-derived stem cells on naturally derived scaffolds. Bio-materials 2008;29:1862e71.

[10] Flynn L, Prestwich GD, Semple JL, Woodhouse KA. Adipose tissue engineeringwith naturally derived scaffolds and adipose-derived stem cells. Biomaterials2007;28:3834e42.

[11] Hodde JP, Badylak SF, Brightman AO, Voytik-Harbin SL. Glycosaminoglycancontent of small intestinal submucosa: a bioscaffold for tissue replacement.Tissue Eng 1996;2:209e17.

[12] Ota T, Gilbert TW, Schwartzman D, McTiernan CF, Kitajima T, Ito Y, et al.A fusion protein of hepatocyte growth factor enhances reconstruction ofmyocardium in a cardiac patch derived from porcine urinary bladder matrix.J Thorac Cardiovasc Surg 2008;136:1309e17.

[13] Hoganson DM, O’Doherty EM, Owens GE, Harilal DO, Goldman SM,Bowley CM, et al. The retention of extracellular matrix proteins and angio-genic and mitogenic cytokines in a decellularized porcine dermis. Bio-materials 2010;31:6730e7.

[14] Hoganson DM, Owens GE, O’Doherty EM, Bowley CM, Goldman SM,Harilal DO, et al. Preserved extracellular matrix components and retainedbiological activity in decellularized porcine mesothelium. Biomaterials2010;31:6934e40.

[15] Teicher EJ, Madbak FG, Dangleben DA, Pasquale MD. Human acellular dermalmatrix as a prosthesis for repair of a traumatic diaphragm rupture. Am Surg2010;76:231e2.

[16] Wolf MT, Daly KA, Brennan-Pierce EP, Johnson SA, Carruthers CA, D’Amore A,et al. A hydrogel derived from decellularized dermal extracellular matrix.Biomaterials 2012;33:7028e38.

[17] Wolf MT, Daly KA, Reing JE, Badylak SF. Biologic scaffold composed of skeletalmuscle extracellular matrix. Biomaterials 2012;33:2916e25.

K.M. Kulig et al. / Biomaterials 34 (2013) 5776e57845784

[18] Brown AL, Brook-Allred TT, Waddell JE, White J, Werkmeister JA, Ramshaw JA,et al. Bladder acellular matrix as a substrate for studying in vitro bladdersmooth muscle-urothelial cell interactions. Biomaterials 2005;26:529e43.

[19] Mase Jr VJ, Hsu JR, Wolf SE, Wenke JC, Baer DG, Owens J, et al. Clinicalapplication of an acellular biologic scaffold for surgical repair of a large,traumatic quadriceps femoris muscle defect. Orthopedics 2010;33:511.

[20] DeQuach JA, Yuan SH, Goldstein LS, Christman KL. Decellularized porcinebrain matrix for cell culture and tissue engineering scaffolds. Tissue Eng Part A2011;17:2583e92.

[21] Crapo PM, Gilbert TW, Badylak SF. An overview of tissue and whole organdecellularization processes. Biomaterials 2011;32:3233e43.

[22] Gilbert TW, Sellaro TL, Badylak SF. Decellularization of tissues and organs.Biomaterials 2006;27:3675e83.

[23] Srokowski EM, Woodhouse KA. Decellularized scaffolds. Compr Biomater2011;2:369e86.

[24] Gilbert TW. Strategies for tissue and organ decellularization. J Cell Biochem2012;113:2217e22.

[25] Agrawal V, Tottey S, Johnson SA, Freund JM, Siu BF, Badylak SF. Recruitment ofprogenitor cells by an extracellular matrix cryptic peptide in a mouse modelof digit amputation. Tissue Eng Part A 2011;17:2435e43.

[26] Marcal H, Ahmed T, Badylak SF, Tottey S, Foster LJ. A comprehensive proteinexpression profile of extracellular matrix biomaterial derived from porcineurinary bladder. Regen Med 2012;7:159e66.

[27] Reing JE, Brown BN, Daly KA, Freund JM, Gilbert TW, Hsiong SX, et al. Theeffects of processing methods upon mechanical and biologic properties ofporcine dermal extracellular matrix scaffolds. Biomaterials 2010;31:8626e33.

[28] Mathew AC, Rajah TT, Hurt GM, Abbas Abidi SM, Dmytryk JJ, Pento JT. Influ-ence of antiestrogens on the migration of breast cancer cells using an in vitrowound model. Clin Exp Metastasis 1997;15:393e9.

[29] Ribatti D, Nico B, Vacca A, Presta M. The gelatin sponge-chorioallantoicmembrane assay. Nat Protoc 2006;1:85e91.

[30] Badylak SF, Gilbert TW. Immune response to biologic scaffold materials.Semin Immunol 2008;20:109e16.

[31] Badylak SF, Valentin JE, Ravindra AK, McCabe GP, Stewart-Akers AM. Macro-phage phenotype as a determinant of biologic scaffold remodeling. Tissue EngPart A 2008;14:1835e42.

[32] Brown BN, Valentin JE, Stewart-Akers AM, McCabe GP, Badylak SF. Macro-phage phenotype and remodeling outcomes in response to biologic scaffoldswith and without a cellular component. Biomaterials 2009;30:1482e91.

[33] Valentin JE, Stewart-Akers AM, Gilbert TW, Badylak SF. Macrophage partici-pation in the degradation and remodeling of extracellular matrix scaffolds.Tissue Eng Part A 2009;15:1687e94.

[34] Orenstein SB, Qiao Y, Klueh U, Kreutzer DL, Novitsky YW. Activation of humanmononuclear cells by porcine biologic meshes in vitro. Hernia 2010;14:401e7.

[35] Brown BN, Ratner BD, Goodman SB, Amar S, Badylak SF. Macrophage polari-zation: an opportunity for improved outcomes in biomaterials and regener-ative medicine. Biomaterials 2012;33:3792e802.

[36] Badylak S, Meurling S, Chen M, Spievack A, Simmons-Byrd A. Resorbablebioscaffold for esophageal repair in a dog model. J Pediatr Surg 2000;35:1097e103.

[37] Piechota HJ, Gleason CA, Dahms SE, Dahiya R, Nunes LS, Lue TF, et al. Bladderacellular matrix graft: in vivo functional properties of the regenerated ratbladder. Urol Res 1999;27:206e13.

[38] Deeken CR, Melman L, Jenkins ED, Greco SC, Frisella MM, Matthews BD.Histologic and biomechanical evaluation of crosslinked and non-crosslinkedbiologic meshes in a porcine model of ventral incisional hernia repair. J AmColl Surg 2011;212:880e8.

[39] James NL, Poole-Warren LA, Schindhelm K, Milthorpe BK, Mitchell RM,Mitchell RE, et al. Comparative evaluation of treated bovine pericardium as axenograft for hernia repair. Biomaterials 1991;12:801e9.

[40] Rocco G, Serra L, Fazioli F, Mori S, Mehrabi-Kermani F, Capasso A, et al. The useof veritas collagen matrix to reconstruct the posterior chest wall after cost-overtebrectomy. Ann Thorac Surg 2011;92:e17e18.

[41] Ueno T, Pickett LC, de la Fuente SG, Lawson DC, Pappas TN. Clinical appli-cation of porcine small intestinal submucosa in the management of infectedor potentially contaminated abdominal defects. J Gastrointest Surg 2004;8:109e12.

[42] Hodde JP, Record RD, Tullius RS, Badylak SF. Retention of endothelial celladherence to porcine-derived extracellular matrix after disinfection andsterilization. Tissue Eng 2002;8:225e34.

[43] Allman AJ, McPherson TB, Merrill LC, Badylak SF, Metzger DW. The Th2-restricted immune response to xenogeneic small intestinal submucosa doesnot influence systemic protective immunity to viral and bacterial pathogens.Tissue Eng 2002;8:53e62.

[44] Ricard-Blum S, Ballut L. Matricryptins derived from collagens and pro-teoglycans. Front Biosci 2011;16:674e97.

[45] Saito Y, Owaki T, Matsunaga T, Saze M, Miura S, Maeda M, et al. Apoptoticdeath of hematopoietic tumor cells through potentiated and sustainedadhesion to fibronectin via VLA-4. J Biol Chem 2010;285:7006e15.

[46] Nguyen NM, Kelley DG, Schlueter JA, Meyer MJ, Senior RM, Miner JH.Epithelial laminin alpha5 is necessary for distal epithelial cell maturation,VEGF production, and alveolization in the developing murine lung. Dev Biol2005;282:111e25.

[47] Hocking DC, Kowalski K. A cryptic fragment from fibronectin’s III1 modulelocalizes to lipid rafts and stimulates cell growth and contractility. J Cell Biol2002;158:175e84.

[48] Davis GE, Bayless KJ, Davis MJ, Meininger GA. Regulation of tissue injury re-sponses by the exposure of matricryptic sites within extracellular matrixmolecules. Am J Pathol 2000;156:1489e98.

[49] FelborU,Dreier L, BryantRA, PloeghHL,OlsenBR,MothesW.Secreted cathepsinL generates endostatin from collagen XVIII. EMBO J 2000;19:1187e94.

[50] Ramchandran R, Dhanabal M, Volk R, Waterman MJ, Segal M, Lu H, et al.Antiangiogenic activity of restin, NC10 domain of human collagen XV: com-parison to endostatin. Biochem Biophys Res Commun 1999;255:735e9.