Angiogenesis and nerve regeneration in a model of human skin equivalent transplant

www.elsevier.com/locate/lifescie

Life Sciences 73 (2003) 1985–1994

Angiogenesis and nerve regeneration in a model of human skin

equivalent transplant

Agnese Ferrettia, Elena Boschia, Alessandro Stefania, Saturnino Spigab,Marco Romanellic, Monica Lemmia, Anna Giovannettia,

Biancamaria Longonia,d,*, Franco Moscaa

aDivision of General Surgery and Transplants, Department of Oncology, Transplants and Advanced Technologies in Medicine,

University of Pisa, Pisa, ItalybDepartment of Biology, University of Cagliari, Cagliari, Italy

cDepartment of Experimental Pathology, Section of Dermatology, University of Pisa, Pisa, ItalydDepartment of Neuroscience, University of Pisa, Pisa, Italy

Received 29 January 2003; accepted 7 April 2003

Abstract

The angiogenesis and reinnervation were studied in a porcine model of human skin equivalent (SE) graft and

the relationship between the two processes was investigated. Confocal laser scanning microscopy was used to

monitor, during the healing process, the pattern of vascularization and reinnervation at different time points. The

SE was obtained by co-culturing fibroblasts and keratinocytes on a collagen-glycosaminoglycan-chitosan

biopolymer and grafted on dorsal wounds generated by full-thickness resection in 25/30 Kg Large white pigs.

Frozen sections were obtained from biopsies performed in autograft and xenograft, then were immunolabeled by

using the endothelial marker lectin Lactifolia and with the neuronal marker gene product PGP9.5. Cajal staining

was also used to visualize the nerve fibers. The results show that the vascularization precedes the innervation

process. These data are consistent with the view that the development of nervous tissue is driven by nutritional and

trophic factors provided by the vascular system. The arborization of the two systems observed during the third

week from the graft might play a key role in maintaining the healing process and the graft survival.

D 2003 Elsevier Inc. All rights reserved.

Keywords: Confocal microscopy; Immunolabeling; Vascularization; Innervation; Imaging

0024-3205/03/$ - see front matter D 2003 Elsevier Inc. All rights reserved.

doi:10.1016/S0024-3205(03)00541-1

* Corresponding author. Laboratorio di Chirurgia Generale e Trapianti, Via Paradisa 2, Cisanello, Pisa 56100, Italy.

Tel.: +39-050-996820, +39-050-835801; fax: +39-050-543692.

E-mail address: [email protected] (B. Longoni).

A. Ferretti et al. / Life Sciences 73 (2003) 1985–19941986

Introduction

The treatment of severely burned patients is a complex endeavor. In recent years, the reconstruc-

tion of human skin by tissue engineering is considered the best alternative to classic therapy, such as

cadaver skin or xenografts, which are temporary and have various drawback (Berthod and Rouabhia,

1997). There is a considerable interest in the use of cultured keratinocytes for burn therapy

(O’Connor et al., 1981; De Luca et al., 1989; Matouskova et al., 2001). This method allows the

production of human epidermis in large quantities from a small initial skin biopsy (Rheinwald and

Green, 1975). Unfortunately, various obstacles have delayed the use of composite skin substitutes.

Insufficient vascularization has been proposed as the most likely reason for their unreliable survival.

The outcome of tissue transplantation critically depends on the vascularization process. It is known

that once vascularization has taken place, several trophic factors are provided, and cell survival can

be maintained (Boyce et al., 1995). In line with this process, the development of other tissues, such

as nerve regeneration, follows the revascularization process. Various studies showed the important

role of angiogenesis in the reinnervation process (Manek et al., 1993; Gu et al., 1994; Gu et al.,

1995; Hobson et al., 1997; Kangesu et al., 1998). Furthermore studies on regeneration of blood

vessels have been carried out in order to obtain reliable information on graft take. In our

experiments, the choice of the pig model was dictated from the need to investigate whether the

human skin equivalent could be implanted in large quantities with good chance of success. The pig

has been utilized in a number of studies related to wound healing (Winter, 1962; Bustad and

McClellan, 1965) also including burn healing (Winter, 1974; Hoekstra et al., 1993). The remarkable

similarities between porcine and human skin structure have been confirmed by using the confocal

laser scanning microscopy (CLSM) (Vardaxis et al., 1997). For our study, experimental pigs were

severely immunosuppressed, however we should consider that in view of future application to

humans, complications due to rejection are expected to be minimized by autografting. In this regard,

we have reasons to believe that our model of human skin equivalent may be positively used for care

of burned patients.

Methods

Skin equivalent preparation

Keratinocytes were isolated from human skin specimens removed during reductive surgery of

healthy subjects, grown in Green modified medium and used at the second passage as previously

described (Germain et al., 1993). Fibroblasts were isolated from human dermis and grown in

Dulbecco’s modified Eagle’s medium (DMEM) (Sigma, Milano, Italy) supplemented with 10% fetal

calf serum (Sigma) and antibiotics. The cells were maintained in a humidified 5% CO2 and 95% air

incubator at 37 jC. Skin equivalent (SE) was prepared as previously described (Braye et al., 2001), by

adding a suspension of 3 � 105 fibroblasts/cm2 on top of collagen-glycosaminoglycan-chitosan

biopolymer (Damour et al., 1994) and cultured for 10 days in medium containing 100 Ag/ml ascorbic

acid. Human keratinocytes, at a concentration of 4 � 105 cells/cm2, were plated on this structure and

cultured under submerged conditions for 6 days. The SE was then elevated on the air-liquid interface

for 15 days.

Experimental animals

Large white female pigs about 25/30 Kg of weight were used in the study. They were kept under

standard diet conditions. Animals were immunosuppressed with 10 mg/kg daily cyclosporin A

(Sandimmun, Basel, Switzerland) starting 24 hours before surgery and during the entire experimental

period.

All experiments were performed according to the international rules on animal experimentation, with

the approval of the University of Pisa.

Surgery

The engraftment of an in vitro reconstructed human living SE in a porcine model was performed at the

laboratory for medium and large size animals of the Experimental Surgery Unit in S. Piero a Grado

(University of Pisa) approved by the National Ministry of Health for the validation of systems and

materials as implants. Three immunosuppressed female pigs (Large white) were used in the study. The

surgical operation was performed in aseptic conditions and under general anesthesia using ketamine and

xilazine.

Full-thickness skin resections were performed on the dorsum of each animal. The defects were

covered by pieces of in vitro reconstructed human SE (xenotransplant) and with autologous porcine skin

graft as control (autotransplant).

O-rings for colostomy (55 mm in diameter) were sutured on the circle defects and used as skin graft

chambers to isolate the graft from the surrounding skin.

Wounds were kept moist by dressing with non adherent gauze between the graft and the top of the

chamber. The final dressing was made by compressive packets (Lagrot dressing) and Tensoplast (BSN

medical, Brierfield, England).

Biopsies

Full-thickness skin biopsies of the grafts were taken at different time points during the healing

process: 3, 14, 38 and 64 days after grafting. Normal skin and autologous-graft skin samples, taken from

a similar site as the wound, were used as controls.

Immunohistochemistry

Tissue was fixed in a solution containing paraformaldehyde 4%. Frozen sections 15–60 Am in

thickness were cut, with a cryostat, perpendicular to the cutaneous surface and were collected on gelatin-

coated slides for staining. The sections were washed with phosphate-buffered saline (PBS) (Sigma) at

room temperature (3 times � 10 min). After washing in PBS, the sections were incubated for 15 min at

room temperature with PBS containing 2% (w/v) bovine serum albumin (BSA) (Sigma) and 0.3% (v/v)

Triton X-100 (Sigma) in order to make the tissue permeable to antiserum. Tissue sections were further

treated with 1% (w/v) BSA in PBS for 10 min before incubation with primary antibody applied

overnight at 4 jC. For vascularization studies, the sections were labeled with lectin conjugated to

fluorescein isothiocyanate (FITC) (diluted 1:20; Sigma), a probe which allows to stain the endothelial

cells and their precursors. Antisera against Protein Gene Product 9.5 (PGP9.5) (diluted 1:800;

A. Ferretti et al. / Life Sciences 73 (2003) 1985–1994 1987


Biogenesis, UK) was used for immunostaining by the indirect immunofluorescence method. The general

neuronal marker PGP9.5 is a structural cytoplasmic protein present in all types of mammalian nerves in

normal skin and during nerve regeneration (Dalsgaard et al., 1989; Manek et al., 1993) and also in

cutaneous receptor types such as Merkel cells and Meissner’s corpuscles (Thompson et al., 1983; Wang

et al., 1990). After rinsing in PBS (3 � 15 min), the PGP9.5 labeled sections were incubated with

secondary antisera for 2 h at room temperature in darkness. The antibody used was FITC-conjugated

goat anti-rabbit IgG (diluted 1:100; Vector, USA). All antibody dilutions were made in a PBS solution

containing 0,1% Triton X-100 and 1% BSA. After further PBS washes (3 � 10 min) the sections were

mounted in mounting medium, Immuno-Fluorek (ICN Biomedicals, Milano, Italy) and covered with a

coverslip.

Cajal-De Castro staining

Tissue sections were mounted on electrically charged slides and then fixed for three days using a

mixture of ethanol (50 ml) at 95j (Sigma), twice distilled water (50 ml) (Fresenius Kabi, Verona, Italy),

chloral hydrate (5 g) (Sigma) and nitric acid (3 ml) (Sigma). After 12 hours in a solution of ethanol (100

ml) at 95j and ammonia (6 drops) (Sigma), the tissues were saturated in a 2% solution of silver nitrate

(‘‘Cajal-De Castro’’ method) (Levinson and MacFate, 1961; Mazzi, 1977), in darkness, at a temperature

of 37jC for one week. Thereafter, they were kept in a reduced mixture containing pyrogallic acid (1 g)

(Sigma), twice distilled water (90 ml) and neutralized formaline (Sigma) (10 ml) for 4 hours (Spiga et al.,

1997).

Confocal imaging

Sections were visualized with an inverted microscope (Nikon Eclipse TE300) equipped with a laser

confocal scanning system (Radiance Plus, Bio-Rad, Hercules, CA). Fluorescence images were collected

using the 488 nm excitation wavelength from an argon laser and the 515 nm emission filter. In order to

minimize photo-oxidation of the probe, the laser beam was attenuated to 50% of maximal illumination,

and exposure of tissue sections to light was limited to the image acquisition intervals (about 2 sec every

3 min) via the acquisition software. Analysis of the fluorescent signal was performed with Adobe

Photoshop 5.0 (Adobe, San Jose, CA).

For each area examined in the sections, the fluorescent signal was analyzed in the epidermis and in the

dermis (papillary dermis, dermis and deep dermis). A semi-quantitative assessment of the number of

blood vessels and nerve fibers was performed according to the criteria shown in Table 1.

Table 1

Criteria for the semi-quantitative assessment of immunoreactive blood vessels and nerve fibres

Score Description No of blood vessels/mm2 No of nerve fibers /mm2

0 None 0 0

+/� Occasional 1–50 1–25

+ Sparse 50–200 25–100

+ + Few 200–500 100–200

+ ++ Moderate 500–1200 200–400

+ ++ + Abundant > 1200 > 4000


Confocal analysis of Cajal De-Castro stained sections was performed using a conventional fluorescent

light and Leica 4D Confocal Laser Scanning Microscope (CLSM) with an Argon-Krypton laser. Confocal

images were generated using PL Floutar 40� (n.a. = 1.00� 0.5) and PL Floutar 100� (n.a. = 1.3) oil-

immersion lenses. Each frame was acquired eight times and then averaged to obtain noise-free images.

Scanware 4.2a Leica, Corel Photo Paint version 9 and Epson Stylus photo 790 for image analysis and

printing. For 3D reconstruction, the ‘‘maximum intensity’’ algorithm was used on 10 images scanned in

Fig. 1. Confocal images of blood vessels immunostained with lectin at different time points after transplantation. A) 3 days after

grafting: moderate number of linear vessels observed; B) 14 days after grafting: higher density of blood vessels observed; C) 5

weeks after grafting: anasthomotic branches dividing from linear vessels. Images are representative of similar experiments

performed on three different pigs.


a 10 Am Z-range. Resulting images were white-labeled on a black background, in a gray scale ranging

from 0 (black) to 255 (white).

Results

The revascularization and reinnervation processes took place about one week after grafting, with

innervation following the vascularization process. Blood vessels, one week after grafting were clearly

more numerous than in control. The illustrations are the results of the sum of optical sections scanned

throughout the sample, by confocal microscopy, in order to obtain a two-dimensional representation of

the structure. Positive staining of the skin graft with the endothelial marker lectin, was observed in the

sample tested at different time points from the implantation day (Fig. 1, panels A, B and C).

Fig. 2. Confocal images of nerves immunostained for PGP9.5 at different time points after transplantation. A) 38 days after

grafting: few sparsely distributed nerve fibers observed in dermis; B) 63 days after grafting: nerves more densely distributed,

showing sprouting of collateral branches. Images are representative of similar experiments performed on three different pigs.

Fig. 3. Confocal image of nerve fibers obtained by Cajal/De Castro staining. At 38 days after transplantation nerve fibers were

observed in dermis (white asterisks) and in epidermis (white arrows). Image is representative of similar experiments performed

on three different pigs.


A moderate number of blood vessels were visualized in the area of the graft, 3 days after grafting (Fig.

1, panel A). Beginning two weeks after grafting, a greater density of blood vessels was observed (Fig. 1,

panel B). At this time, blood vessels were oriented mainly perpendicularly to the surface of the

Table 2

Confocal microscopy analysis of blood vessels and nerve fibres density using FITC-conjugated lectin and antisera to protein

gene product 9.5 (PGP9.5) respectively

Control Days after grafting

3 14 38 63

Vascularization

Papillary dermis + + + +/� + + + +/�Dermis + + + + + + + + + ND

Deep dermis + + + + ND + ND

Innervation

Epidermis + 0 0 0 0

Papillary dermis + + + 0 0 0 + ++

Dermis + + 0 +/� + + + + +

Deep dermis ND 0 0 + + + +

The semi-quantitative assessment was carried out on several sections. Since each group contained three animals, an average

score was calculated for the final analysis.

0, +/�, +, ++, +++: see Table 1.

ND: not detected.


epidermis. Five weeks after grafting, anasthomotic branches and new pattern of sprouting could be

observed (Fig. 1, panel C). Positive staining of the skin graft with the pan-neuronal marker, PGP 9.5,

was observed in the sample tested at different time points from the implantation day (Fig. 2, panels A,

B). Nerve fibers were distributed throughout all the layers of the epidermis and dermis. A time-

dependent sprouting and network were observed with multiple branching fibers extending to the dermis

at 63 days after transplantation (Fig. 2, panel B). The distribution of reinnervation followed that of

neovascularization. Nerves developed towards the surface of the epidermis with a slower temporal

pattern than that of vascularization. These unmyelinated fibers increase in thickness as they extend into

the dermis. A further confirm of the presence of these fibers was obtained by Cajal/De Castro staining of

the sections. Confocal image depicted in Fig. 3, shows the presence of nerve fibers at 38 days from the

implantation. Anasthomotic branches were localized at epidermic and dermis level. A semi-quantitative

analysis of both the vascularization and reinnervation process is reported in Table 2.

Discussion and conclusion

The application of immunohistochemistry in conjunction with a lectin labeling of the sections

obtained from the biopsies of the SE graft, allowed us to study both the neuronal and the

vascularization process. Furthermore, the possibility to perform virtual sections of the sample during

observation with confocal microscopy and subsequent reconstruction of the image, allowed us to

obtain a three dimensional picture with high resolution ready to examine innervation and

vascularization. In the present study, by using confocal laser scanning microscopy, we have shown

the pattern of neovascularization and reinnervation in a model of human SE grafted in pig. The

observations made by confocal microscopy provide the advantage to eliminate the out-of-focus

images. The imaging obtained by summing the serial virtual sections resulted in a significant

amelioration of resolution and contrast. The results show a clear improvement in neovascularization

during the three weeks of observation of the healing process, followed with a slower time course by

the nerve growth. In fact, it is known that the process of healing cannot take place in the absence

of angiogenesis (Arnold and West, 1991). A well-organized network of fibers could be observed

around the third week from the graft. According with other reports, we observed that the

developing of nerves followed that of revascularization (Gu et al., 1995; Hobson et al., 1997;

Kangesu et al., 1998), probably due to the fact that the blood supply is necessary for developing of

other tissue with high metabolic request. This, suggesting an interaction between regenerating nerves

and blood vessels. Taken together, these results show that the model described above can be

promising for future applications to humans. Despite severe immunosuppression was induced in

pigs in order to achieve the success of the skin graft, amelioration of the technique is expected

when considering the facilitation involved by autotransplantation versus xenotransplantation.

Acknowledgements

ARPA Foundation is greatly acknowledged for financial support. We thank Dr. O. Damour for kindly

providing the biopolymer for skin equivalent preparation, Dr. C. Gargini for helpful discussion and Mr.

A. Tacchi for excellent technical assistance.


References

Arnold, F., West, D.C., 1991. Angiogenesis in wound healing. Pharmacology and Therapeutics 52, 407–422.

Berthod, F., Rouabhia, M., 1997. Exhaustive review of clinical alternatives for damaged skin replacement. In: Rouabhia, M.

(Ed.), Skin Substitute Production by Tissue-Engineering: Clinical and Fundamental Application. Landes Bioscience, Austin,

pp. 23–45.

Boyce, S.T., Supp, A.P., Harriger, M.D., Greenhalgh, D.G., Warden, G.D., 1995. Topical nutrients promote engraftment and

inhibit wound contraction of cultured skin substitutes in athymic mice. Journal of Investigative Dermatology 104, 345–349.

Braye, F.M., Stefani, A., Venet, E., Pieptu, D., Tissot, E., Damour, O., 2001. Grafting of large pieces of human reconstructed

skin in a porcine model. British Journal of Plastic Surgery 54, 532–538.

Bustad, L.K., McClellan, R.O., 1965. Use of pigs in biomedical research. Nature 208, 531–535.

Dalsgaard, C.J., Rydh, M., Haegerstrand, A., 1989. Cutaneous innervation in man visualized with protein gene product 9.5

(PGP9.5) antibodies. Histochemistry 92, 385–390.

Damour, O., Gueugniaud, P.Y., Berthin-Maghit, M., Rousselle, P., Berthod, F., Sahuc, F., Collombel, C., 1994. A

dermal substrate made of collagen-GAG-chitosan for deep burn coverage: first clinical uses. Clinical Materials 15,

273–276.

De Luca, M., Albanese, E., Bondanza, S., Megna, M., Ugozzoli, L., Molina, F., Cancedda, R., Santi, P.L., Bormioli, M., Stella,

M., 1989. Multicentre experience in the treatment of burns with autologous and allogenic cultured epithelium, fresh or

preserved in a frozen state. Burns 15, 303–309.

Germain, L., Rouabhia, M., Guignard, R., Carrier, L., Bouvard, V., Auger, F.A., 1993. Improvement of human keratinocyte

isolation and culture using thermolysin. Burns 19, 99–104.

Gu, X.H., Terenghi, G., Kangesu, T., Navsaria, H.A., Springall, D.R., Leigh, I.M., Green, C.J., Polak, J.M., 1995. Regeneration

pattern of blood vessels and nerves in cultured keratinocyte grafts assessed by confocal laser scanning microscopy. British

Journal of Dermatology 132, 376–383.

Gu, X.H., Terenghi, G., Purkis, P.E., Price, D.A., Leigh, I.M., Polak, J.M., 1994. Morphological changes of neural and vascular

peptides in human skin suction blister injury. Journal of Pathology 172, 61–72.

Hobson, M.I., Brown, R., Green, C.J., Terenghi, G., 1997. Inter-relationships between angiogenesis and nerve regeneration: a

histochemical study. British Journal of Plastic Surgery 50, 125–131.

Hoekstra, M.J., Hupkens, P., Dutrieux, R.P., Bosch, M.M., Brans, T.A., Kreis, R.W., 1993. A comparative burn wound model in

the New Yorkshire pig for the histopathological evaluation of local therapeutic regimens: silver sulfadiazine cream as a

standard. British Journal of Plastic Surgery 46, 585–589.

Kangesu, T., Manek, S., Terenghi, G., Gu, X.H., Navsaria, H.A., Polak, J.M., Green, C.J., Leigh, I.M., 1998. Nerve and blood

vessel growth in response to grafted dermis and cultured keratinocytes. Plastic and Reconstructive Surgery 101, 1029–1038.

Levinson, S.A., MacFate, R.P., 1961. Trattato di diagnostica e di tecnica di laboratorio. Piccin, Padova.

Manek, S., Terenghi, G., Shurey, C., Nishikawa, H., Green, C.J., Polak, J.M., 1993. Neovascularisation precedes neural

changes in the rat groin skin flap following denervation: an immunohistochemical study. British Journal of Plastic Surgery

46, 48–55.

Matouskova, E., Broz, L., Pokorna, E., Konigova, R., 2001. Necrobiotic process causing burn wound conversion may be

prevented by allogenic keratinocytes delivered by the recombined human/pig skin. Folia Biologica 47, 135–142.

Mazzi, V., 1977. Manuale di tecniche istologiche e istochimiche. Piccin, Padova.

O’Connor, N.E., Mulliken, J.B., Banks-Schlegel, S., 1981. Grafting of burns with cultured epithelium prepared from autolo-

gous epidermal cells. Lancet 1, 75–78.

Rheinwald, J.G., Green, H., 1975. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing

colonies from single cells. Cell 6, 331–343.

Spiga, S., Lallai, L., Mulargia, C., Loy, F., Orru, F., Siddi, A., Argiolas, M., Serra, G.P., 1997. Application of the Cajal-De

Castro method to the laser scanning confocal microscope in studying the peripheral nervous system. Italian Journal of

Anatomy and Embryology 102, 173–193.

Thompson, R.J., Doran, J.F., Jackson, P., Dhillon, A.P., Rode, J., 1983. PGP9.5-a new marker for vertebrate neurons and

neuroendocrine cells. Brain Research 278, 224–228.

Vardaxis, N.J., Brans, T.A., Boon, M.E., Kreis, R.W., Marres, L.M., 1997. Confocal laser scanning microscopy of porcine skin:

implications for human wound healing studies. Journal of Anatomy 190, 601–611.


Wang, L., Hilliges, M., Jernberg, T., Wiegleb-Edstrom, D., Johansson, O., 1990. Protein gene product 9.5-immunoreactive

nerve fibres and cells in human skin. Cell Tissue Research 261, 25–33.

Winter, G.D., 1962. Formation of the scab and the rate of epithelisation of superficial wounds in the skin of the young domestic

pig. Nature 193, 293–294.

Winter, G.D., 1974. In: Maibach, H.L., Rovee, D.T. (Eds.), Healing. Year Book Medical Publishers, Chicago, pp. 71–112.

Angiogenesis and nerve regeneration in a model of human skin equivalent transplant

Documents

Transcript of Angiogenesis and nerve regeneration in a model of human skin equivalent transplant