Available online at www.sciencedirect.com

Biomaterials 29 (2008) 1620e1629www.elsevier.com/locate/biomaterials

The clinical application of autologous bioengineered skin based ona hyaluronic acid scaffold

Nicolo Scuderi a, Maria G. Onesti a, Giovanni Bistoni a, Simona Ceccarelli b,Sabrina Rotolo b, Antonio Angeloni b, Cinzia Marchese b,*

a Department of Plastic and Reconstructive Surgery, University Sapienza, Rome, Italyb Department of Experimental Medicine, University Sapienza, Viale Regina Elena 324, 00161 Rome, Italy

Received 17 September 2007; accepted 5 December 2007

Available online 16 January 2008

Abstract

The aim of this work was to generate an in vitro skin substitute harbouring autologous fibroblasts, keratinocytes and melanocytes, to establisha new one-step clinical method in problems associated with skin disorders. Here we present a case of a nine-year-old girl with a congenital giantnevus treated by surgical approach, with primary co-cultures of keratinocytes, melanocytes and fibroblasts obtained from autologous skin biopsy.Generally these lesions need to be removed to avoid the risk of transformation into malignant melanoma. With this purpose we analyzed themelanocytes contained in the new skin substitute for the presence of genetic alterations correlated to increased risk for melanoma. The organo-typical cultures were designed including an engineered scaffold of a non-woven mesh of hyaluronic acid (HYAFF�11). This biomaterial hasbeen previously demonstrated to be the most suitable to maintain polarity and to support the in vitro constructs. Six dermaleepidermal skinsubstitutes were transplanted and 14 days after surgery the re-epithelialized area was about 90%. Our results suggest that this new dermaleepidermal construct not only reduces hospitalization time and ameliorates scar retraction, but might also represent a solution for the highrisk of developing a tumour derived from the original nevus.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Autologous cell; Co-culture; Hyaluronic acid scaffold; Transplantation

1. Introduction

Autologous cell cultures have been used with increasing fre-quency in the field of reconstructive plastic surgery. The estab-lished procedure is based on culturing keratinocytes obtainedfrom a small skin biopsy and represents a valid alternative toskin graft, giving similar results [1,2]. The subsequent introduc-tion of synthetic dermal substitutes to support epithelial cul-tures allowed to overcome technical problems such as poortake efficiency and difficult handling, thus providing their pos-sible applications also in the treatment of pathologies that causeextensive loss or removal of skin substance.

Giant congenital melanocytic nevi, or bathing trunk nevi,are large lesions that can cover more than 50% of the total

* Corresponding author. Tel.: þ39 0649973012.

E-mail address: cinzia.marchese@uniroma1.it (C. Marchese).

0142-9612/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.biomaterials.2007.12.024

body surface. They are considered to be precursors of cancerand prone to transformation into malignant melanomas. Theassociation between giant congenital melanocytic nevi andmalignant melanoma has been established beyond any doubt,and some studies demonstrated a significant increased riskfor melanoma in patients with large congenital nevi measuring20 cm or more in largest diameter, or covering more than 5%of the body surface area [3e5]. Nevertheless, the exact mag-nitude of the risk is still unknown. The lifetime risk of mela-noma for these patients has been reported to range from 5% to40% [6,7], with an actual incidence inversely correlated withpatient’s age. Indeed, malignancy is more likely to set in thefirst decade of life, with an overall incidence of 60%. For thisreason, the options for treatment of congenital giant nevi leantowards staged excision with grafting, dermo-abrasion andcurettage, more than conservative approaches that might im-prove the cosmetic appearance, but often show incomplete

1621N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

clearance and do not eliminate the risk of malignant transfor-mation [8]. In fact, melanocytes in the congenital melanocyticnevi may be found extending into deep dermis and subcutane-ous layer, needing to be excised down to the muscle fascia tocompletely eradicate the possibility of tumour arising [9,10].Split-thickness graft has been often adopted as surgical treat-ment for these pathologies that involves large areas, althoughthis conventional method has been subsequently flanked bythe use of autologous skin equivalents that require less donorskin.

Here we aimed to determine a clinical method for thein vivo application of a new skin equivalent harbouring ker-atinocytes and melanocytes in the epidermal compartmentand fibroblasts in the dermal. This new graft is projected toenhance a proper spatial organization of the different cellcompartments in a well orchestrating network, in order tofacilitate also the deposition of extracellular matrix of autol-ogous origin.

Several genes are known to be implicated in the develop-ment of melanoma malignancies, including the tumour sup-pressor genes p53, PTEN, INK4a and BRAF. In order tominimize the risk of melanoma correlated with melanocyticnevi, here we performed a validation of the melanocytes cul-tures used in the skin substitute, obtained through a molecularanalysis of the presence of known markers of melanoma pro-gression, such as activating BRAF mutations and loss ofwild-type INK4a expression, that both occur at high frequen-cies in melanomas. The BRAF oncogene has demonstrated tobe mutated in a majority of melanoma cell lines [11e13].Remarkably, all BRAF mutations occur within the kinase do-main, with a single amino acid substitution (V599E) account-ing for 95% of BRAF mutations in melanoma cell lines andleading to constitutive kinase activity [14]. With regard toINK4a, this locus encodes two distinct proteins, p16INK4aand p14ARF (CDKN2A), both of which demonstrate tumoursuppressor activity in genetically distinct anticancer pathways[15]. Deletion of INK4a has been detected in 50% of primarytumours and nearly all melanoma cell lines [12,13].

In this work we realized a new bioengineered skin graft inwhich the melanocytes contained within the epidermal com-partment were investigated for the mutations considered prog-nostic markers of melanoma transformation. The absence ofany of the melanoma related mutation, combined with thisnovel surgical approach, might represent a good alternativeto the previously adopted techniques. The use of this newskin substitute requires smaller donor sites than split-thicknessgraft, although giving similar results, and could significantlyabrogate the risk of melanoma. Moreover it speeds up hospi-talization times and convalescence, providing the possibilityto complete the procedure in a single operation.

2. Materials and methods

2.1. Cell isolation and culture

The patient entered the day-hospital for blood and instrumental preopera-

tive tests. At the same time, a skin biopsy of about 2.5 cm2 was taken from the

right groin, transferred into a Petri-dish and disinfected by submersion in 75%

ethanol for 30 s. Subcutaneous fat and deep dermis were excised from the bi-

opsy sample, and the remaining tissue was cut into small pieces. All fragments

were transferred into a centrifuge tube containing 0.25% trypsin (Gibco, Pais-

ley, UK) and incubated at 37 �C for 10 min in 5% CO2 incubation. Epidermal

fragments were gently pipetted until disintegration into a single cell suspen-

sion. Cells were then counted, seeded into two 35 mm dishes (1� 106 cells/

dish) and cultured in Keratinocyte Serum-Free Medium (K-SFM; Gibco).

Melanocytes cultures were also assessed and cultured in Melanocyte Serum-

Free Medium (Gibco) containing 12-O-tetradecanoylphorbol acetate phorbol

ester (TPA, 10 ng/ml) and 5% FBS. Melanocytes eventually present among

the primary keratinocytes cultures were subsequently isolated by gentle tryp-

sinization, and the purity of the cultures was further assessed by immunoflu-

orescence with specific markers of both keratinocytes and melanocytes.

The dermal fragments obtained from the same biopsy were incubated at

37 �C for 1.5 h in a pre-heated sterile collagenase (625 U/ml; Sigma, St. Louis,

MO, USA) solution. Subsequently, the fragments were gently pipetted until

disintegration into a single cell suspension. Cells were counted, seeded at

1� 105 cells/cm2 into 75 cm2 flasks and maintained in DMEM (Gibco) con-

taining 10% FBS.

2.2. Immunofluorescence analysis

To assess the purity of keratinocytes and melanocytes cultures, the cells

were assayed for the expression of specific markers by immunofluorescence.

Briefly, cells grown on coverslips onto 24-well plates were fixed in 4% para-

formaldehyde in phosphate-buffered saline (PBS) for 30 min at 25 �C, fol-

lowed by treatment with 0.1 M glycine in PBS for 20 min at 25 �C and

with 0.1% Triton X-100 in PBS for additional 5 min at 25 �C to allow per-

meabilization. Cells were then incubated with the following primary anti-

bodies: anti-human cytokeratin monoclonal antibody (1:100 in PBS)

(clone MNF116, Dako, Carpinteria, California) or anti-tyrosinase polyclonal

antibody (1:25 in PBS) (C-19, Santa Cruz Biotechnology, Santa Cruz, Cal-

ifornia). After appropriate washing in PBS, primary antibodies were

visualized using FITC-conjugated goat anti-mouse IgG (1:50 in PBS) or

FITC-conjugated goat anti-rabbit IgG (1:100 in PBS), respectively. Nuclei

were visualized using 40,6-diamido-2-phenylindole dihydrochloride (DAPI)

(1:10,000 in PBS) (Sigma). Fluorescence signals were analyzed by recording

stained images using a CCD camera (Zeiss, Oberkochen, Germany) and

IAS2000/H1 software (Delta Sistemi, Rome, Italy).

2.3. RNA isolation

For RNA extraction, a pellet of 5� 105 melanocytes was immediately

snap-frozen in liquid nitrogen. The human melanoma cell line 624mel, har-

bouring mutations in both BRAF and INK4a, was used as control. RNAs

were extracted using the standard acid guanidinium thiocyanateephenole

chloroform method. Cell pellet was homogenized in 1 ml of extraction reagent

in a separate siliconized DEPC-treated Corex tube using a Polytron homoge-

nizer (Kinematica GmbH., Littau, Switzerland). Between the preparations of

each sample, the homogenizer was washed in Deconex, extensively rinsed

in sterile DEPC-treated distilled water and then sterilized by rinsing in

100% ethanol and flaming. Pellets were resuspended in 10e20 ml of DEPC-

treated water and quantified spectrophotometrically. The quality of the RNA

samples was determined by electrophoresis through denaturing agarose gels

and staining with ethidium bromide; 18S and 28S RNA bands were visualized

under ultraviolet light. cDNA was generated with oligo(dT) from 1 mg of RNA

using the SuperScript III Reverse Transcriptase Kit (Invitrogen Laboratories,

Carlsbad, CA, USA) in a final volume of 20 ml.

2.4. PCR for sequence analysis of BRAF T1796A mutation

Primer sequences were designed according to the database of the National

Center for Biotechnology Information (NCBI, Bethesda, MD, USA, http://

www.ncbi.nlm.nih.gov/BLAST). The primers used for polymerase chain

reaction (PCR) of the BRAF gene (GenBank accession no. AC006344) were

BRAF-F (50-CCTAAACTCTTCATAATGCTTGCTC-30) in intron 15 and

1622 N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

BRAF-R (50-GACTTTCTAGTAACTCAGCAGCATC-30) in intron 14. PCR

products were purified using the QIAquick PCR purification kit (Qiagen,

Valencia, CA) and direct sequencing of PCR products was performed.

2.5. PCR for sequence analysis of INK4a

INK4 exons 1a, 2 and 3 and the alternatively spliced exon 1b of ARF, were

PCR-amplified from genomic DNA. Primer sequences and amplifications were

performed as previously described [16]. PCR products were purified using the

QIAquick PCR purification kit (Qiagen) and direct sequencing was performed

using the ABI PRISM 3700 DNA analyzer (Applied Biosystems, Foster City,

CA).

2.6. Biomaterial

Fig. 1. Preoperative image of the giant melanocytic nevus. Between yellow

lines the area previously treated with laser and dermo-abrasion.

The biomaterial used in the present study was derived from the total ester-

ification of hyaluronan with the benzyl ester, and is referred to as HYAFF�11

(100% nominal degree of substitution, effective range of 92e100%). The

device was obtained from Fidia Advanced Biopolymer S.r.l. (Abano Terme,

Italy). The biomaterial scaffold was configured as a non-woven mesh

(NWM) of 20 mm-thick fibres. In aqueous solution the fibres hydrate with

a weight increase of about 40%. They swell to about double their original

diameter. Degradation occurs by spontaneous hydrolysis of the ester bonds

[17,18].

2.7. Preparation of skin substitute construct

Primary co-cultures of keratinocytes and melanocytes were established by

seeding isolated keratinocytes and melanocytes at a ratio of 20:1. Co-cultures

were then maintained in K-SFM (Gibco) and analyzed by immunofluorescence

with DAPI and with anti-tyrosinase that selectively stains melanocytes, to ver-

ify the percentage of melanocytes present in the co-cultures. The first step of

the skin substitute preparation was to resuspend the primary human fibroblasts,

obtained from the same biopsy, in fibrin gel (Tissucol Baxter) [9], and to seed

them at 2� 105/cm2 into an NWM of HYAFF�11, a benzylic ester of hyalur-

onic acid [17,19]. Fibroblasts were left in place and cultured for about a week

on the 10� 10 cm pieces of NWM of HYAFF�11, in 4 ml of DMEM supple-

mented with 1% L-glutamine, 1% of penicillin/streptomycin and sodium ascor-

bate (50 mg/ml), and enriched with 10% of patient’s own blood serum, until the

cultures reached sub-confluence. Medium was replaced every other day. Sub-

sequently, a suspension of keratinocytes/melanocytes (2� 105 cell/cm2) was

seeded on the surface of the biomaterial. The medium was removed and

changed into chemically defined medium K-SFM (Gibco). The constructs

were kept submerged in the culture medium for about five days. Then, the

cultures were raised to the aireliquid interface for further four days, in order

to promote the in vitro development of a keratinized epidermal surface with

a stratum corneum analogue [2].

2.8. Histology of skin substitute

Both 10 days and one year after surgery, a punch biopsy of the newly re-

epithelialized skin was performed for tissue analysis. Paraffin embedded tissue

was sectioned and 5 mm thick sections were stained with hematoxylin and

eosin.

3. Results

3.1. Case presentation

The patient was a nine-year-old girl with a congenital giantnevus on her back. The caudal portion of the nevus was treatedsince her birth with laser CO2 and dermo-abrasion (Fig. 1, yel-low lines). The removal of the residual lesion would haverequired several operations, creating considerable medical

and psychological problems and long hospitalization times.The patient has been considered eligible for the treatmentand therefore we obtained from her parents the informed con-sent to the treatment with autologous cultured skin substitutesas an alternative approach. This study was approved by theInstitutional Review Committee of the Sapienza MedicalUniversity of Rome.

The patient was admitted to our department for the firsttime to take a skin biopsy of about 2.5 cm2 from the rightgroin. Prophylactic antibiotic therapy was given for subse-quent five days.

3.2. Assessment of keratinocytes, melanocytes andfibroblasts cultures

We first established primary cultures of keratinocytes,melanocytes and fibroblasts from the same full-thicknessskin biopsy. The characteristics of keratinocyte, melanocyteand fibroblast cultures obtained were analyzed by invertedphase-contrast microscopy. Keratinocytes appeared polygo-nal-shaped and possessed a larger nuclearecytoplasmic ratio(Fig. 2A). Melanocytes showed dendritic morphology or slen-der spindle shape (Fig. 2B) whereas fibroblasts exhibitedcharacteristic elongated shape and were clumped together(Fig. 2C).

To further assess the purity of cell cultures and to avoidcross-contamination from different cell types, immunofluores-cence analysis for the expression of specific markers forkeratinocytes or melanocytes was performed. Expression ofcytokeratins was used as a marker for keratinocytes. Since me-lanocytes synthesize and package melanin into melanosomes,anti-tyrosinase antibody, one of the melanogenic enzymesbelonging to the tyrosinase gene family of proteins, wasused as a marker for melanocytes. In our keratinocyte cultures,

Fig. 2. Characteristics of primary cultured keratinocytes, melanocytes and fi-

broblasts. Morphology was observed under an inverted phase-contrast micro-

scope. Epidermal keratinocytes grew into colonies and showed a characteristic

polygonal morphology (A). Melanocytes presented dendritic extensions and

slender spindle shape (B). Dermal fibroblasts displayed bipolar spindle mor-

phology and the typical elongated shape (C). Scale bars 5 mm.

1623N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

all cells resulted positive for the expression of cytokeratins(Fig. 3A), while in melanocyte cultures all cells were positivefor tyrosinase (Fig. 3B).

We then proceeded to establish a co-culture of keratinocytesand melanocytes onto collagen coated culture flasks, usinga seeding ratio of 20:1, on the base of the consensus averageratio. Co-cultures were then analyzed by immunofluorescencewith anti-tyrosinase antibody to verify the percentage of mela-nocytes present in the co-cultures (Fig. 3C).

3.3. Analysis of BRAF mutation and INK4 deletion inmelanocytes cultures used in skin equivalentreconstruction

The BRAF oncogene has been reported to be mutated athigh frequencies in melanomas and among other carcinomas,as well as in a majority of melanoma cell lines [14]. Since thepresence of activating BRAF mutation may be indicative ofa predisposition to melanoma development, we analyzed thesequence of the BRAF mRNA extracted from a representativesample of melanocytes taken from cell suspension before es-tablishing the co-culture with keratinocytes. mRNA extractedfrom the human melanoma cell line 624mel, that is knownto harbour mutations in both BRAF and INK4a, was used asa positive control. mRNA was reverted to cDNA and amplifiedby PCR to be subsequently sequenced. Results shown inFig. 4A did not reveal any mutation concerning the BRAFgene in the primary culture of melanocytes obtained fromthe patient. Furthermore, since homozygous deletions in theINK4 region have been demonstrated in malignant cutaneousmelanoma and are associated with an adverse prognosis, weamplified by PCR (Fig. 4B) and sequenced (data not shown)the INK4 gene. Similarly to the BRAF oncogene, the melano-cytes cultures from the patient analyzed showed an intactINK4 locus, at difference with the tumour cell line that pre-sented a biallelic loss of INK4 locus (Fig. 4B).

3.4. Autologous skin substitute preparation

We thus proceeded to establish the skin substitute, by recon-structing an organotypical culture with a three-dimensional hu-man dermaleepidermal architecture using fibroblast culturesand keratinocyteemelanocyte co-cultures obtained from thepatient’s skin biopsy. Generally, keratinocytes and fibroblastsexpansion on plastic support occurred within two weeks; inthis case, in order to produce a skin substitute containing anupper layer of well-differentiated stratified keratinocytes lininga dermal-like structure of fibroblasts, cells were propagatedfor subsequent passages and then seeded in vitro withina biodegradable fabric made of an NWM of hyaluronic acid(HYAFF�11). This biomaterial served as a scaffold supportfor both fibroblasts and keratinocytesemelanocytes culturesand guaranteed ease of handling in the clinical setting [20].The configuration of the construct was engineered to maintainpolarity and to permit a feasible and reproducible productionof the skin substitute. After 16 days, the three-dimensional

Fig. 3. The purity of cultures was assessed by expression of specific markers. Immunostaining was performed as described in Section 2. Keratinocytes were positive

to staining for human cytokeratin (A). Melanocytes were positive for tyrosinase (B). Co-culture of keratinocytes and melanocytes (seeding ratio 20:1) (C). Me-

lanocytes in the co-culture were identified by immunolabeling with anti-tyrosinase antibody. Nuclei were stained with DAPI. Scale bars 5 mm.

1624 N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

constructs were brought to the operating room under asepticconditions.

3.5. Clinical case

Sixteen days after skin biopsy, the patient was admitted toour department and the next day she underwent surgery for theremoval of about 400 cm2 of the giant congenital nevus downto the muscular fascia. After careful haemostasis, six dermaleepidermal skin substitutes were placed over the iatrogenic sub-stance. Paraffin gauzes were applied and compressive bandagewas performed. Prophylactic antibiotic therapy was adminis-tered to prevent infection. Seven days after surgery, whenthe dressing was removed, the re-epithelialized area was ap-proximately 70% (Fig. 5A, detail in A0), reaching 90% after14 days from surgery (Fig. 5B, detail in B0). Within 10 daysfrom surgery, the newly re-epithelialized skin exhibited an ini-tial cornified layer starting from the wound edge, as assessedby histological examination of a punch biopsy of the newlyformed skin (Fig. 6A). The patient was discharged from thehospital on the postoperative day 18 and followed-up forone year by ambulatory, with the first control after two months(Fig. 7A). After six months, the cellular organization achieved

by the transplanted skin was acceptable, the scar displayedsome hypertrophic areas and appeared moderately erythema-tous, particularly at the level of the flexioneextension areasat the upper limb joint (Fig. 7B). At the one year follow-upthe modulation of the healing was complete (Fig. 7C) andthe skin sample displayed a regular multilayered epitheliumstructure with normal development and regeneration process.The skin showed also an arranged connective tissue in the der-mis and a well-defined cornified layer (Fig. 6B, C). Further-more, there were no major signs of retraction or limitationto the mobility of the limbs or neck: the scar was not painfuleither spontaneously or on palpation. The grafted tissue pro-vided good cover for the deep structures with good mechanicalproperties. No corneal desquamation was appreciable and thenew tissue appeared well-hydrated (Fig. 7C).

4. Discussion

Three-dimensional artificial skin constructs have been de-veloped to meet the growing demand in tissue-engineeredskin equivalents and resulted to be very promising for thetreatment of a variety of pathologies, including burns, ulcersand giant nevi. The surgical treatment of choice for such

Fig. 5. Aspect of skin substitutes seven days (A, A0) and 14 days (B, B0) postgraft

(detail in A0). After 14 days, the re-epithelialization degree was significantly high

Fig. 4. PCR and direct fluorescent sequencing, as described in Section 2, of

a representative fraction of primary melanocytes from patient’s specimen in

comparison with melanoma 624mel cell line samples, as positive control. A

standard PCR reaction using specifying primers that amplify B-actin as house-

keeping was performed in parallel (lower panels). (A) Detection of BRAF

V600E mutations of exon 15. Ethidium bromide-stained PCR gel and follow-

ing DNA sequence showed the presence of BRAF V600E mutation in mela-

noma 624mel, as expected, and absence in primary melanocytes (upper

panel). (B) Detection of loss of the INK4a. Ethidium bromide-stained RT-

PCR gel demonstrated the loss of the INK4a in melanoma 624mel, as positive

control, and wild-type cDNA in primary melanocytes.

1625N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

pathologies frequently involves the use of split-thickness au-tografts. The major problems correlated with the use of thisconventional approach for the treatment of congenital giantnevi still remain the amount of donor skin required to com-plete the wound closure and the possible presence of satellitenevi or potentially transforming melanocytes in the donor site[21]. For that reasons, in such clinical setting, this solution isnot adequate and skin substitutes offer a viable alternativesolution that presents some advantages, including the elimina-tion of donor site pain, morbidity, development of scar, pruritusand chronic wounds [2]. Furthermore, it has been previouslydemonstrated that cultured skin substitutes constructed ontocollageneglycosaminoglycan substrates have a qualitative out-come after grafting similar to treatment with meshed, split-thickness autografts, and offered the essential advantage ofminimizing the requirement for harvesting of donor skin [2].

In the skin, epidermis undergoes a process of continuousand complex differentiation that is not strictly autonomousbut is linked to the underlying connective tissue. In healthyskin, the balance between proliferation and cell loss is pre-cisely maintained [20,22,23]. Therefore, the skin substitutemust be able to recreate this delicate balance in vitro, interact-ing with a complex in vivo system. Hyaluronic acid, a naturallyoccurring molecule present in all soft tissues, plays an essen-tial role in the maintenance of the normal extracellular matrixstructure. The first attempt to reconstruct, in vitro, autologous

ing. After seven days, the grafted area presented some re-epithelialized zones

er (detail in B0).

Fig. 6. Hematoxylin and eosin staining of skin biopsies, showing both the epidermal and dermal components. Histological appearance of the graft 10 days after

surgery (A) exhibited well-developed, stratified epidermis (e), with an initial cornified layer (c), and dermis (d). At the one year follow-up (B, C) the biopsy showed

a clearly arranged connective tissue in the dermis (d) and a regular multilayered epithelium structure (e) with a well-defined cornified layer (c). Original magni-

fications: A, B, �5; and C, �10.

1626 N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

bioengineered skin based on a scaffold of esterified hyaluronicacid is described in 1998 [18], and nowadays the most suitabledevices for cell culture are found to be those made with the to-tal esterified derivative of hyaluronan, commercially availableas HYAFF�11, in the three-dimensional configuration of non-woven mesh [17,19,24,25]. This biomaterial is completelybiodegradable, immunologically inert, and do not elicit com-plement activation [26,27]. The development of human skinequivalents on a hyaluronic acid scaffold is of particular sig-nificance, since previous keratinocytes cultures containing col-lagen hydrogels demonstrated to have some defects, such aslimited survival, poor resistance to contraction, faulty anchor-age of the epidermis to the collagen matrix, and lack of dermalsupport [20,28]. HYAFF�11 avoids many of the drawbackspreviously encountered with collagen scaffolds, as it showsa good cell adhesiveness even in the absence of any coating,ensures a three-dimensional structure that provide mechanicalstability and easy handling, therefore allowing a wider clinicalapplication, and has a better resistance to contraction than thecollagen-based materials [17]. Moreover, three-dimensionalscaffolds made of HYAFF�11 have been demonstrated tosupport growth and differentiation of fibroblasts, keratinocytesand chondrocytes [17,29]. Epithelial sheets prepared onHYAFF�11 scaffolds have been applied in experimental

models of cutaneous grafting [18,30], and they have alsobeen successfully used in humans for the treatment of cutane-ous lesions [31,32]. In rabbits, such biomaterial has alsodemonstrated to provide successful cell scaffolds for tissue-engineered repair of bone and cartilage [29,33].

It is known that the use of cultured skin equivalents derivedfrom only fibroblasts and keratinocytes cultures give raise toa skin that lacks natural pigmentation [2,34]. This inconve-nient becomes more serious when the treated area is wide,as in giant congenital nevi, originating both aesthetical andpsychological problems. For this reason, in recent yearssome protocols have been established for manufacturing skinequivalents using fibroblasts, keratinocytes and melanocytes,in order to ameliorate skin substitutes pigmentation. A recentwork by Liu et al. [35] showed successful grafting of pig-mented skin equivalents in animal models, offering the basisfor application in humans. Therefore, in this study we seededa mixed suspension composed of keratinocytes and melano-cytes, with a ratio of 1:20, on a scaffold of esterified hyalur-onic acid containing autologous fibroblasts to create a skinsubstitute harbouring melanocytes.

In the treatment of congenital nevi, the major risk is rep-resented by the possibility of melanoma transformation. Sev-eral genes have been demonstrated to be involved in the

Fig. 7. Aspect of the grafted area at the two months (A), six months (B) and

one year (C) follow-up, when complete re-epithelialization can be detected.

1627N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

development of melanomas. Among them, BRAF oncogenehas been reported to be mutated in a majority of melanomacell lines [11,14], while deletion of INK4a is known to bea common genetic trait detected in 50% of primary tumours

and nearly all melanoma cell lines [12,13]. Nevertheless,none of the previous attempts to develop an in vitro model ofco-culture system that resumed the human dermaleepidermalarchitecture took in account the possibility that melanocytesobtained from nevomelanocytic lesions and then propagatedin vitro could possess mutations leading to melanoma transfor-mation. In this study we generated a skin equivalent containingkeratinocytes, melanocytes and fibroblasts, at the same timemanaging the risk of melanoma correlated with the treated pa-thology through a validation of the melanocytes cultures usedin the skin substitute, obtained by performing a molecular anal-ysis of the presence of known markers of melanoma progres-sion. From a clinical point of view, the skin substitute wedeveloped behaves like very thin autologous skin mimickingthe split-thickness grafts taken with a dermatome, but signifi-cantly reducing the risk of melanoma transformation.

Finally, it should be pointed out that bioengineered skingrafts lack the important adnexa and accessory epithelial struc-tures, such as sweat glands, sebaceous glands and hairs, withnegative repercussions on skin lubrication and heat exchange.Moreover, experimental in vitro models seem to suggest thatthe passive transport routes through skin from bioengineeredepidermal cultures might be different from that through naturalskin. In fact, while natural skin routes are limited to the extra-cellular space, epidermal cultures seem to follow both extra-cellular and intracellular routes [36].

Other aspects to be evaluated in patients treated with bioen-gineered skin are scar retraction and hypertrophy. Clinicalevidence on autologous skin grafts have pointed out thatdeep dermis components are able to suppress the activity ofmiofibroblasts involved in cell retraction [37], with the conse-quence that grafts carrying thicker dermis components resultin lesser scar retraction.

Recently, skin equivalents have been further improved by thegeneration of a dermaleepidermal architecture including amicrocapillary network in a three-dimensional biomaterial[38,39], somehow addressing an essential topic for all kinds ofgrafts that is skin substitutes vascularization. Our results high-light the possibility to realize a skin substitute that would notonly reduce the risk to propagate genetic alterations but wouldalso provide a useful tool for the treatment of skin diseases re-quiring transplantations of large areas. However, it still mightbe improved by the insertion of autologous endothelial cells.

5. Conclusions

This study accomplished the in vitro reconstruction of skinequivalents that contained an autologous well-differentiatedepidermal layer covering an underlying living dermal structureand provided a molecular genetic profile of melanocytes thatrepresented a good prognostic value in order to avoid therisk for further melanocyte transformation. Our findingsfrom the in vivo experiment with autologous cultures of fibro-blasts, keratinocytes and melanocytes on scaffolds of NWMHYAFF�11 showed how interaction between the two mainskin components in vitro is essential for the development of

1628 N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

a stable structure and confirmed that the skin equivalentsmethod is effective in repairing human skin defects, providingshorter hospitalization times and the possibility of completingthe procedure in a single operation. Although encouraging andwell-supported by the clinical evidence, our results need fur-ther confirmation by the treatment of additional patients anda longer follow-up observation.

Acknowledgement

This work was supported in part by funding from Ministerodella Salute, Progetti Regionali 2005.

References

[1] Boyce ST, Kagan RJ, Meyer NA, Yakuboff KP, Warden GD. The 1999

clinical research award. Cultured skin substitutes combined with integra

artificial skin to replace native skin autograft and allograft for the closure

of excised full-thickness burns. J Burn Care Rehabil 1999;20:453e61.

[2] Boyce ST, Kagan RJ, Yakuboff KP, Meyer NA, Rieman MT,

Greenhalgh DG, et al. Cultured skin substitutes reduce donor skin har-

vesting for closure of excised, full-thickness burns. Ann Surg 2002;235:

269e79.

[3] Quaba AA, Wallace AF. The incidence of malignant melanoma (0 to 15

years of age) arising in ‘‘large’’ congenital nevocellular nevi. Plast

Reconstr Surg 1986;78:174e81.

[4] Ruiz-Maldonado R, Tamayo L, Laterza AM, Duran C. Giant pigmented

nevi: clinical, histopathologic, and therapeutic considerations. J Pediatr

1992;120:906e11.

[5] Marghoob AA, Schoenbach SP, Kopf AW, Orlow SJ, Nossa R, Bart RS.

Large congenital melanocytic nevi and the risk for the development of

malignant melanoma. A prospective study. Arch Dermatol 1996;132:

170e5.

[6] Kaplan EN. The risk of malignancy in large congenital nevi. Plast

Reconstr Surg 1974;53:421e8.

[7] Sober AJ, Burstein JM. Precursors to skin cancer. Cancer 1995;75:

645e50.

[8] Tannous ZS, Mihm Jr MC, Sober AJ, Duncan LM. Congenital melano-

cytic nevi: clinical and histopathologic features, risk of melanoma, and

clinical management. J Am Acad Dermatol 2005;52:197e203.

[9] Rhodes AR, Wood WC, Sober AJ, Mihm Jr MC. Nonepidermal origin of

malignant melanoma associated with giant congenital nevocellular

nevus. Plast Reconstr Surg 1981;67:782e90.

[10] Arons MS, Hurwitz S. Congenital nevocellular nevus: a review of the

treatment controversy and a report of 46 cases. Plast Reconstr Surg

1983;72:355e65.

[11] Davies H, Bignell GR, Cox C, Shephens P, Edkins S, Clegg S, et al.

Mutations of the BRAF gene in human cancer. Nature 2002;417:

949e54.

[12] Flores JF, Walker GJ, Glendening JM, Haluska FG, Castresana JS,

Rubio MP, et al. Loss of the p16INK4a and p15INK4b genes, as well as

neighboring 9p21 markers, in sporadic melanoma. Cancer Res 1996;56:

5023e32.

[13] Walker GJ, Flores JF, Glendening JM, Lin AH, Markl ID, Fountain JW.

Virtually 100% of melanoma cell lines harbor alterations at the DNA

level within CDKN2A, CDKN2B, or one of their downstream targets.

Genes Chromosomes Cancer 1998;22:157e63.

[14] Pollock PM, Harper UL, Hansen KS, Yudt LM, Stark M, Robbins CM,

et al. High frequency of BRAF mutations in nevi. Nat Genet 2003;

33:19e20.

[15] Sharpless NE, Ramsey MR, Balasubramanian P, Castrillon DH,

DePinho RA. The differential impact of p16(INK4a) or p19(ARF) defi-

ciency on cell growth and tumorigenesis. Oncogene 2004;23:379e85.

[16] Fargnoli MC, Chimenti S, Keller G, Soyer HP, Dal Pozzo V, Hofler H,

et al. CDKN2a/p16INK4a mutations and lack of p19ARF involvement in

familial melanoma kindreds. J Invest Dermatol 1998;111:1202e6.

[17] Campoccia D, Doherty P, Radice M, Brun P, Abatangelo G, Williams DF.

Semisynthetic resorbable materials from hyaluronan esterification.

Biomaterials 1998;19:2101e27.

[18] Zacchi V, Soranzo C, Cortivo R, Radice M, Brun P, Abatangelo G. In

vitro engineering of human skin-like tissue. J Biomed Mater Res

1998;40:187e94.

[19] Galassi G, Brun P, Radice M, Cortivo R, Zanon GF, Genovese P, et al. In

vitro reconstructed dermis implanted in human wounds: degradation

studies of the HA-based supporting scaffold. Biomaterials 2000;21:

2183e91.

[20] Stark HJ, Willhauck MJ, Mirancea N, Boehnke K, Nord I, Breitkreutz D,

et al. Authentic fibroblast matrix in dermal equivalents normalises

epidermal histogenesis and dermoepidermal junction in organotypic

co-culture. Eur J Cell Biol 2004;83:631e45.

[21] Earle SA, Marshall DM. Management of giant congenital nevi with arti-

ficial skin substitutes in children. J Craniofac Surg 2005;16:904e7.

[22] Pellegrini G, Ranno R, Stracuzzi G, Bondanza S, Guerra L, Zambruno G.

The control of epidermal stem cells (holoclones) in the treatment of

massive full-thickness burns with autologous keratinocytes cultured on

fibrin. Transplantation 1999;68:868e79.

[23] Ronfard V, Rives JM, Neveux Y, Carsin H, Barrandon Y. Long-term

regeneration of human epidermis on third degree burns transplanted

with autologous cultured epithelium grown on a fibrin matrix. Transplan-

tation 2000;70:1588e98.

[24] Aigner J, Tegeler J, Hutzler P, Campoccia D, Pavesio A, Hammer C,

et al. Cartilage tissue engineering with novel nonwoven structured

biomaterial based on hyaluronic acid benzyl ester. J Biomed Mater Res

1998;42:172e81.

[25] Brun P, Abatangelo G, Radice M, Zacchi V, Guidolin D, Daga Gordini D,

et al. Chondrocyte aggregation and reorganization into three-dimensional

scaffolds. J Biomed Mater Res 1999;46:337e46.

[26] Benedetti L, Cortivo R, Berti T, Berti A, Pea F, Mazzo M, et al. Biocom-

patibility and biodegradation of different hyaluronan derivatives (Hyaff)

implanted in rats. Biomaterials 1993;14:1154e60.

[27] Cortivo R, Brun P, Rastrelli A, Abatangelo G. In vitro studies on biocom-

patibility of hyaluronic acid esters. Biomaterials 1991;12:727e30.

[28] Stark HJ, Szabowski A, Fusenig NE, Maas-Szabowski N. Organotypic

cocultures as skin equivalents: a complex and sophisticated in vitro

system. Biol Proced Online 2004;6:55e60.

[29] Radice M, Brun P, Cortivo R, Scapinelli R, Battaliard C, Abatangelo G.

Hyaluronan-based biopolymers as delivery vehicles for bone-marrow-

derived mesenchymal progenitors. J Biomed Mater Res 2000;50:

101e9.

[30] Myers SR, Grady J, Soranzo C, Sanders R, Green C, Leigh IM, et al. A

hyaluronic acid membrane delivery system for cultured keratinocytes:

clinical ‘take’ rates in the porcine kerato-dermal model. J Burn Care

Rehabil 1997;18:214e22.

[31] Donati L, Veronesi AM, Garbin S, Ordanini MN, Ferrone MP, Falcone L,

et al. In vitro development and clinical application in burn patients of

keratinocytes cultured on a hyaluronic acid ester membrane. In:

Williams DF, editor. Novel biomaterials based on hyaluronic acid and

its derivatives. Proceedings of a workshop held at the annual meeting

of the European society for biomaterials; 1994. p. 52e61. Pisa, Italy.

[32] Donati L, Marazzi M, Veronesi AM, Ordanini MN, Falcone L,

Ferrone M, et al. Treatment of cutaneous wound with cultured human

keratinocytes on hyaluronic acid membrane. Wound Repair Regen

1995;3:363.

[33] Solchaga LA, Dennis JE, Goldberg VM, Caplan AI. Hyaluronic acid-

based polymers as cell carriers for tissue-engineered repair of bone and

cartilage. J Orthop Res 1999;17:205e13.

[34] Bello YM, Falabella AF, Eaglstein WH. Tissue-engineered skin. Current

status in wound healing. Am J Clin Dermatol 2001;2:305e13.

[35] Liu Y, Suwa F, Wang X, Takemura A, Fang YR, Li Y, et al. Reconstruc-

tion of a tissue-engineered skin containing melanocytes. Cell Biol Int

2007;31:985e90.

1629N. Scuderi et al. / Biomaterials 29 (2008) 1620e1629

[36] Simonetti O, Hoogstraate AJ, Bialik W, Kempenaar JA, Schrijvers AH,

Bodde HE, et al. Visualization of diffusion pathways across the

stratum corneum of native and in-vitro-reconstructed epidermis by

confocal laser scanning microscopy. Arch Dermatol Res 1995;

287:465e73.

[37] Rudolph R. The effect of skin graft preparation on wound contraction.

Surg Gynecol Obstet 1976;142:49e56.

[38] Supp DM, Wilson-Landy K, Boyce ST. Human dermal microvascular

endothelial cells form vascular analogs in cultured skin substitutes after

grafting to athymic mice. FASEB J 2002;16:797e804.

[39] Tonello C, Vindigni V, Zavan B, Abatangelo S, Abatangelo G, Brun P,

et al. In vitro reconstruction of an endothelialized skin substitute

provided with a microcapillary network using biopolymer scaffolds.

FASEB J 2005;19:1546e8.