REVIEW DIGEST In Vitro Reconstructed Human Skin Equivalents for ...

REVIEW DIGEST

In Vitro Reconstructed Human SkinEquivalents for Autologous Transplantation in Burns,

Chronic Ulcers, and Pigment Defects

Liliana Guerra, Desanka Raskovic,Chiara De Luca, and Liudmila G. Korkina*

Laboratory Tissue Engineering and Skin Pathophysiology, Istituto Dermopatico dell’Immacolata (IDI IRCCS), Rome, Italy.

Background: Cultured epidermal and dermal cells reconstitute normal skinstructure and retain biochemical and molecular characteristics of the originaldonor site. Application of in vitro engineered skin on a well-prepared woundbed allows to permanently regenerate full-thickness wounds because of thepresence of epidermal stem cells in culture. Further, cultured cells producegrowth factors and components of extracellular matrix that facilitate thewound healing process.The Problem: There is a need to develop and optimize both structure andfunctions of the in vitro reconstructed human skin equivalents. Should skinequivalents be different for the treatment of chronic ulcers, burns, or whiteskin spots in vitiligo? How to prepare fully autologous and easy-to-monitorcultures that form ready-to-apply skin sheets?Basic/Clinical Science Advances: Interference between different human skincells in cultures grown on a proper substrate controls cell growth, differentia-tion, cytokine production, and distinct cellular functions. This should be takeninto consideration during preparation of skin equivalents targeting various skinpathologies.Clinical Care Relevance: In acute wounds with an increased risk of scar formationand in chronic nonhealed ulcers as well as in skin pigment disorders whereclinical efficacy of conventional therapies is unsatisfactory, autotransplantationof properly in vitro engineered skin equivalents could substantially improve andaccelerate closure of skin defects and restore esthetical appearance of normalskin.Conclusion: In vitro reconstructed fully autologous and functional skin equiva-lents engineered for different skin defects could be considered as safe and efficientcellular medicinal products for individualized therapy.

BACKGROUND

DURING THE PAST 30 YEARS, severalexperimental and commercial strat-egies have been developed to sup-port wound healing by providingproper substitution for serious skindefects.1,2 The in vitro reconstructedskin equivalents for autotransplan-tation should ideally substitute ma-jor functions of damaged skin, coverlarge damaged=defective areas, and

additionally elicit regeneration re-sponse from the wound bed withoutcausing inflammation or rejection.For safety reasons, in vitro recon-structed skin equivalents should notcontain xenobiotics and cells of non-human origin, and they should becontinuously monitored through anappropriate quality assurance sys-tem as per requirement from regu-latory bodies. From the esthetic point

*Correspondence: Laboratory Tissue En-

gineering and Skin Pathophysiology, Istituto

Dermopatico dell’Immacolata (IDI IRCCS), Via

Monti di Creta 104, Rome 00167, Italy (e-mail:

[email protected]).

Abbreviationsand Acronyms

ECM¼ extracellular matrix

YAG¼ yttzium aluminum garnet

Liudmila G. Korkina

62 j ADVANCES IN WOUND CARE, VOLUME 2 Printed in U.S.A.Copyright ª 2011 by Mary Ann Liebert, Inc. DOI: 10.1089=awc.2010.0278

1____

0 ____

1____

AWC-2010-0278-Guerra_3P.3d 01/27/11 10:07pm Page 62

of view, these autologous skin substitutes shouldbe durable, elastic, and pigmented to resemblenatural skin. Indications for skin substitutes areburns, chronic wounds (leg ulcers), pigment de-fects, plastic, aesthetic, dental surgery, and defectsin oral mucosa. The novel, efficient, and safe cel-lular products manufactured, delivered, and ap-plied in accord with good practices (laboratory,manufacturer, and clinical ones) will start a newera of individualized cell-based therapeutic ap-proach in the management of severe chronic andacute skin defects with high morbidity, risk of in-validity, great social=economical impact,3,4 andheavy psychological burden.5

CLINICAL PROBLEM ADDRESSED

Chronic ulceration of the lower leg is a frequentcondition. Prevalence numbers range from 1% inthe adult population to 3%–5% in the populationover 65 years of age. In western countries, the in-cidence of chronic leg ulceration is rising as a resultof the ageing population and of the increased riskfactors for atherosclerotic occlusion, such assmoking, obesity, and diabetes.3,6 Among thesepatients the amputation rate has been reported tobe 15–70 times that of the general population. Be-cause of the indolent, resistant, or recurrent natureof chronic wounds, a wide variety of treatmentsexist and new wound management techniques arecontinuously being developed.7

Healing of third-degree burn wounds is accompa-nied by severe scar formation. These scars are char-acterized by excessive extracellular matrix (ECM)

deposition, altered collagen remodeling, and con-traction. The presence of abundant inflammatorymediators mediates the migration of fibroblast-likecells to the wound area. These peculiar fibroblasts(myofibroblasts) produce high levels of differentcomponents of the ECM, first of all collagen. In nor-mal wound healing the myofibroblasts disappearduring the remodeling phase, whereas in hypertro-phic scars the myofibroblasts remain present andactively produce collagen. Because of the disfiguringand mobility-limiting scars, patients have to undergomany surgeries to release the limitations.4,7,8

There are several options for the treatmentof vitiligo, a skin pathology characterized by theappearance of depigmented spots. Physical (pho-tochemotherapy) and medical therapies (cortico-steroids, calcineurin inhibitors, vitamin D analogs,and antioxidants) are usually utilized in activevitiligo with scarce success. Repigmentation isinitiated by proliferation, migration, and activationof melanocytes still present in the hair follicles,either in the margins or within the depigmentedspots. Advanced cellular therapies, such as auto-transplantation of isolated melanocytes or in vitroengineered skin equivalents, are emerging for sta-ble vitiligo, when medical therapies failed to inducerepigmentation.9,10

RELEVANT BASIC SCIENCE CONTEXT

During the research and development of innova-tive cell-based skin products, several importantbasic science aspects concerning (1) cell-to-cell in-teractions in the cultures,11,12 (2) effects of cell or-igin and culture conditions on the molecularmechanisms of cellular senescence,9 (3) differenti-ation, and (4) prosurvival=death alternatives,13 (5)clonal evolution of skin cells in cultures,14 and (6)stem cell1,13,15 characteristics are essential for thefuture success of clinical application. Distinct ge-netic manipulations with proteins involved in thekeratinocyte senescence and programmed deathallowed to increase proliferation of cultured kera-tinocytes and decrease their differentiation (effectof immortalization) without, however, tumor trans-formation of the cells.7

EXPERIMENTAL MODEL OR MATERIAL—ADVANTAGES AND LIMITATIONS

Skin cell cultures grown on the biologically com-patible material are simultaneously the experi-mental model and the final product in the tissueengineering process. As an experimental model,individual cultures or cocultures of different skin

TARGET ARTICLES

1. Guerra L, Dellambra E, Panacchia L, andPaioni E: Tissue engineering for damaged sur-face and lining epithelia: stem cells, currentclinical applications and available engineeredtissues. Tissue Eng Part B 2009; 15: 91.

2. Panacchia L, Dellambra E, Bondanza S,Paterna P, Maurelli R, Paioni E, and Guerra L:Nonirradiated human fibroblasts and irradiated3T3-J2 murine fibroblasts as a feeder layer forkeratinocyte growth and differentiation in vitroon a fibrin substrate. Cells Tissues Organs 2010;191: 21.

3. Guerra L, Dellambra E, Brescia S, andRaskovic D: Vitiligo: pathogenetic hypothesisand targets for current therapies. Curr DrugMetab 2010; 11: 451.

SKIN EQUIVALENTS FOR BURNS, ULCERS, AND VITILIGO 63

____�____0

____þ


cells (keratinocytes, fibroblasts, and melanocytes)allow to easily manipulate cultural conditions,substrates-scaffolds for cell growth, ratio of cocul-tured cells, addition of biologically active factors,or elimination of foreign molecules and cells.7,9,14

Cellular behavior, the histological and molecularcharacteristics of reconstructed skin, as well as itspossible contamination with microbial and viralmaterial could be effectively monitored at selectedtime points. Serious limitations of the model arelow reproducibility, high cost, and time-consumingprocedures for its maintenance and quality control.

DISCUSSION OF FINDINGSAND RELEVANT LITERATURE

Normal human skin cells can be isolated by a smallskin biopsy and serially propagated in vitro.1,7,8,16,17

Keratinocytes reconstitute sheets of stratifiedepidermis, the upper layer of the skin, with aphysiological melanocyte-to-keratinocyte ratio. Fi-broblasts may be amplified in a large quantity andmay be cocultured with keratinocytes if an appro-priate biological substrate is used. Interaction be-tween keratinocytes and fibroblasts is crucial forthe full regeneration of a functional epidermis.18

Both keratinocytes and fibroblasts contribute toprotein deposition and to the reconstitution ofthe basal membrane. Keratinocyte- and fibroblast-derived cytokines and growth factors are of utmostimportance in regulating wound healing. Thus,keratinocyte-derived cytokines inhibit synthesisof ECM by fibroblasts, fibroblast-mediated graftcontraction, and scar formation.7,8,18 On the otherhand, fibroblasts exert profound effects on kerati-nocyte proliferation and differentiation.18 Culturedkeratinocytes are essential to maintain melanocytesurvival and functions through production ofgrowth factors targeting melanocytes and cytokinesregulating melanogenesis.9 Of note, keratinocytesisolated from the depigmented spots of vitiligopatients or patients with inherited piebald traitexhibit limited capacity to control a ratio of mela-nocyte survival=death in cocultures.9,11,12

Epidermis needs to constantly replace damagedor dead cells throughout the life. Typically, con-tinuous replacement is maintained because ofthe presence of stem cells. Normally, in humanepidermis at least 10% of basal cells are stemcells.1,7,13,15 If the in vitro engineered skin con-tains stem cells, it enables long survival of thetransplant. Several burn patients treated withthe in vitro reconstructed skin have been followedup for more than 20 years. Although numerousstem cell markers now exist, it is rather difficult

to a priori identify stem cells in culture. Only aposteriori analyses of cell clones, regenerativecapacity, and ‘‘life span’’ of the cultures providesome hints on the presence of functional stemcells.7,11,14

For the in vitro tissue engineering, matrices aredeveloped to support cultured cells and promotetheir differentiation and proliferation toward theformation of a new tissue. Substrates-scaffolds en-abling proper wound coverage should have someessential characteristics that include (1) being easyto handle and apply to the wound site; (2) provideappropriate barrier function; (3) be readily adher-ent; (4) have appropriate physical and mechanicalproperties; (5) undergo controlled degradation; (6)be sterile, nontoxic, and nonantigenic; (7) evokeminimal inflammatory reactivity; (8) incorporateinto the host with minimal pain and low risk offurther scarring; (9) facilitate angiogenesis; and(10) being cost effective.1,7,14 The fibrin matrix isa homogeneous and transparent gel where cellgrowth may be easily monitored. When keratino-cyte-containing fibrin gels are applied onto thewounds, fibrin is quickly degraded, allowing cells totake on the wound bed. Accurate quality controlshave demonstrated that keratinocytes’ ability toform the epidermal stem cell–containing clones,growth rate, and long-term proliferative potentialare not affected by fibrin substrate.14 When appliedto a properly prepared wound bed, the fibrin ad-hesive fixes the cells to the wound and allows betterin-growth.

Accepted practice for third-degree burns is to ac-curately debride the burned tissue, provide a dermalequivalent, and, once this has become well vascu-larized, provide an autologous epidermal layer.8,19

The most commonly used production of graftableepidermis relies on the presence of lethally irradi-ated murine 3T3-J2 fibroblasts.17 The utilization ofnonirradiated autologous human fibroblasts showsexcellent results in terms of unaltered keratinocytebehavior, proliferative potential, cytokine produc-tion, and stem cell presence.14

In chronic ulcers, cultured cells are mainly con-sidered to act as ‘‘biological factories’’ by stimulat-ing the patient’s own wound repair mechanism:(1) production of growth factors and ECM and (2)reepithelialization coming from the edge of thewound.1,2,6,16 The lack of dermis is, however, acritical factor for cell engraftment. A better ap-proach would be to supply the clean and preliminaryrevascularized ulcer with both (1) ready-to-be-vascularized dermis consisting of biocompatible,porous, and oxygen and growth factors penetra-ble three-dimensional matrix plus autologous fi-

64 ADVANCES IN WOUND CARE

1____

0 ____

1____


broblasts and (2) autologous cultured keratinocytes,which could be taken and proliferated.7,14

In the physiological coculture of keratinocytesand melanocytes, (1) keratinocytes regulate mela-nocyte growth and differentiation by a variety ofgrowth and survival factors for melanocytes; (2)keratinocytes regulate the physiological melano-cyte-to-keratinocyte ratio; and (3) melanocytesorganize themselves into the basal layer of thecultured epidermis and develop dendrite ‘‘bridges’’to transfer melanosomes into basal keratinocytes.When autologous cultured epidermal sheets aregrafted onto a well-prepared erbium-YAG (ytt-zium, aluminum, garnet) laser-receiving bed, itfacilitates stable repigmentation of the achromicareas in 90% of vitiligo10 and piebaldism20 pa-tients.

INNOVATION

The most innovative aspect of the target articles byGuerra’s group7,9,14 is that the classical procedureof human skin equivalents preparation in the invitro conditions described by Rheinwald and Green35 years ago,17 with the obligatory use of lethallyirradiated murine fibroblasts as a feeder layer V,is substituted by another one, where autologoushuman nonirradiated fibroblasts perfectly main-tained normal growth, differentiation, and sta-minality (the presence of stem cells) of cultivatedkeratinocytes.14 Skin equivalents manufacturedwithout cells and molecules of animal origin wouldbetter meet international safety requirements forcell-based medicinal products.

SUMMARY ILLUSTRATION

The figure opposite shows the standardoperations for the in vitro reconstructionand autotransplantation of skin equiv-alents to (1) third-degree burns (leftcolumn)—subconfluent primary kerati-nocyte cultures are expanded on fibringel, and layers of confluent keratinocytes(secondary cultures) grown on fibrin areapplied on the top of dermal reconstructsubstituting a soft tissue defect. Biopsyas small as 1 cm2 gives skin equivalentsto cover large burn areas (more than1 m2), which usually heal without exces-sive scar formation; (2) chronic ulcers(middle column)—subconfluent primarykeratinocytes are layered on the autolo-gous fibroblasts expanded within a fibringel. Completely autologous equivalents

having architecture of normal skin are transferredto properly prepared ulcer bed; (3) vitiligoor piebaldism stable pigment defect (right col-umn)—confluent primary keratinocyteþ melano-cyte cultures (primary cultures) obtained from thebiopsy of pigmented parts of the skin are seriallyexpanded (secondary cultures) andkeratinocyte=melanocyte ratio as well as melaninproduction by cultivated melanocytes are thor-oughly controlled. The in vitro reconstructed pig-mented skin equivalents cover large surfaces ofdepigmented skin preliminary slightly ablated(deepithelialized) by erbium-YAG laser (for details,see refs.10,20).

CAUTION, CRITICAL REMARKS,AND RECOMMENDATIONS

The technology of in vitro tissue engineering hasbeen shown to be feasible and several productshave been successfully marketed. There is still agreat need for the development of new advancedproducts with higher clinical impact.

Translation of research from the bench to bed-side requires long-term rather than short-termquality controls of cell-based products. Biological‘‘drugs’’ for the somatic cell therapy must be thor-oughly evaluated before proposing them for humanuse either in clinical trials or as commercial prod-ucts. In accord with current regulations, productsbased on manipulated human cells for cell therapyare considered as medicinal preparations to bemanufactured in compliance with good manu-facturing practice, which considerably increasescost of the production.

TAKE HOME MESSAGE

Basic science advances

� Interference between different human skin cells in cultures grown on asubstrate mimicking ECM influences cell growth, differentiation, cytokineproduction, and several other cellular functions.

Clinical science advances

� Distinct procedures of the in vitro reconstruction of skin equivalentscontaining different cells and an appropriate scaffold lead to differenti-ated production of cell-based medicinal preparations targeting eitheracute or chronic wounds or pigment skin defects.

Relevance to clinical care

� In acute wounds with an increased risk of scar formation and in chronicnonhealed ulcers as well as in skin pigment disorders with unmettherapeutic needs, autotransplantation of purposefully in vitro engineeredskin equivalents substantially improves and accelerates wound healingand restores esthetical appearance of normal skin.

____�____0

____þ



The speed with which regulatory mechanismswill be developed and implemented is a criticalfactor for advanced cell-based drugs and therapies.

Financing the high costs of engineered tissueproduction is always a challenge. Moreover, thecost of autologous cells=skin equivalents is veryhigh when compared with allogenic products.However, there are many advantages in terms ofsafety and efficacy in the usage of completely au-

tologous cellular medicinal products; therefore, thor-ough cost=efficacy analysis should be carried out toprove their market viability.

FUTURE DEVELOPMENTS

Tissue-engineered skin should include all the skinappendages (hair follicles, sweat glands, and sen-sory organs) and layers (epidermis and dermis),

1____

0 ____

1____



with rapid development of functional vascular andnerve network and scar-free integration with thesurrounding host tissue. Such a construct shouldallow the skin to fulfill its many normal functions:barrier formation, pigmentary defense against ul-traviolet irradiation, thermoregulation, and me-chanical and aesthetic functions. Considerablefocus should be placed on multipotent adult stemcells, such as bone marrow– or adipose tissue–derived mesenchymal stem cells or hair bulge stemcells, which could accelerate wound repair or evenreconstitute the wound bed. Moreover, next-gen-eration skin equivalents will be constructed onsmart scaffolds carefully engineered to releasesignaling molecules, growth and differentiationfactors, and special proteins to facilitate cell mi-gration and adhesion. Properly designed clinicaltrials on safety and efficacy of stem cell–bearing

skin substitutes engineered on advanced biomate-rials should be carried out.

ACKNOWLEDGMENTS

Tissue engineering basic and clinical research waspartly funded by the Italian Ministry for Health(grant IDI IRCCS-RC-2009). Financial and ad-ministrative support to Cell Factory, a spin off ofthe corresponding author’s laboratory, producingskin equivalents for clinical application, providedby the Congregation ‘‘Figli dell’Immacolata’’ ishighly appreciated.

AUTHOR DISCLOSURE STATEMENT

The authors have no competing interests. This ar-ticle was not written by any writer other than theauthors.

REFERENCES

1. Metcalfe AD and Ferguson MWJ: Tissue engineeringof replacement skin: the crossroads of biomaterials,wound healing, embryonic development, stem cellsand regeneration. J R Soc Interface 2007; 4: 413.

2. MacNeil S: Progress and opportunities for tissue-engineered skin. Nature 2007; 445: 874.

3. Sen CK, Gordillo GM, Roy S, Kirsner R, Lambert L,Hunt TK, Gottrup F, Gurtner GC, and Longaker MT:Human skin wounds: a major and snowballingthreat to public health and the economy. WoundRepair Regen 2009; 17: 763.

4. Hickerson WL, Compton CC, Fletchall S, and SmithLR: Cultured epidermal autografts and allodermiscombination for permanent burn wound coverage.Burns 1994; 20 suppl 1: S52.

5. Sampogna F, Raskovic D, Guerra L, Pedicelli C,Tabolli S, Leoni L, Alessandroni L, and Abeni D:Identification of categories at risk for high qualityof life impairment in patients with vitiligo. Br JDermatol 2008; 159: 351.

6. Mekkes JR, Loots MAM, Van Der Waal AC, andBos JD: Causes, investigation and treatment of legulceration. Br J Dermatol 2003; 148: 388.

7. Guerra L, Dellambra E, Panacchia L, and Paioni E:Tissue engineering for damaged surface and liningepithelia: stem cells, current clinical applicationsand available engineered tissues. Tissue Eng Part B2009; 15: 91.

8. Ronfard V, Rives JM, Neveux Y, Carsin H, andBarrandon Y: Long-term regeneration of human

epidermis on third degree burns transplanted withautologous cultured epithelium grown on a fibrinmatrix. Transplantation 2000; 70: 1588.

9. Guerra L, Dellambra E, Brescia S, and Raskovic D:Vitiligo: pathogenetic hypothesis and targets forcurrent therapies. Curr Drug Metab 2010; 11: 451.

10. Guerra L, Primavera G, Raskovic D, Pellegrini G,Gollisano O, Bondanza S, Luci A, and De Luca M:Erbium: YAG laser and cultured epidermis in thesurgical therapy of stable vitiligo. Arch Dermatol2003; 139: 1303.

11. Bondanza S, Maurelli R, Paterna P, Migliore E, DiGiacomo F, Primavera G, Paionni E, Dellambra E,and Guerra L: Keratinocyte cultures from involvedskin in vitiligo patients show an impaired in vitrobehaviour. Pigment Cell Res 2007; 20: 288.

12. Kostyuk V, Potapovich A, Cesareo E, Brescia S,Guerra L, Valacchi G, Pecorelli A, Deeva IB, Ras-kovic D, De Luca C, Pastore S, and Korkina LG:Dysfunction of glutathione S-transferase leads toexcess 4-hydroxy-2-nonenal and H2O2 and im-paired cytokine pattern in cultured keratinocytesand blood of vitiligo patients. Antioxid RedoxSignal 2010; 13: 607.

13. Pellegrini G, Dellambra E, Golisano O, Martinelli E,Fantozzi I, Bondanza S, Ponzin D, McKeon F, andDe Luca M: p63 identifies keratinocyte stem cells.Proc Natl Acad Sci USA 2001; 98: 3156.

14. Panacchia L, Dellambra E, Bondanza S, Paterna P,Maurelli R, Paioni E, and Guerra L: Nonirradiated

human fibroblasts and irradiated 3T3-J2 murinefibroblasts as a feeder layer for keratinocytegrowth and differentiation in vitro on a fibrinsubstrate. Cells Tissues Organs 2010; 191: 21.

15. Pellegrini G, Golisano O, Paterna P, Lambiase A,Bonini S, Rama P, and De Luca M: Location andclonal analysis of stem cells and their differenti-ated progeny in the human ocular surface. J CellBiol 1999; 145: 769.

16. Eming SA, Smola H, and Kreig T: Treatment ofchronic wounds: state of the art and future con-cepts. Cells Tissues Organs 2002; 172: 105.

17. Rheinwald JG and Green H: Serial cultivation ofstrains of human epidermal keratinocytes: theformation of keratinizing colonies from singlecells. Cell 1975; 6: 331.

18. Werner S, Kreig T, and Smola H: Keratinocyte-fibroblast interactions in wound healing. J InvestDermatol 2007; 127: 998.

19. Tonello C, Vindigni V, Zavan B, Abatangelo S,Abatangelo G, Brun P, and Cortivo R: In vitroreconstruction of an endothelialized skin substi-tute provided with a microcapillary networkusing biopolymer scaffols. FASEB J 2005; 19:1546.

20. Guerra L, Primavera G, Raskovic D, Pellegrini G,Gollisano O, Bondanza S, Luci A, and De Luca M:Permanent repigmentation of piebaldism by erbi-um: YAG laser and autologous cultured epidermis.Br J Dermatol 2004; 150: 715.

____�____0

____þ


AWC-2010-0278-Guerra_3P.3d 01/27/11 10:07pm 7

REVIEW DIGEST

Macrophage Heterogeneity and Wound Healing

Grace Pinhal Enfield and Samuel Joseph Leibovich*

Department of Cell Biology and Molecular Medicine, The Cardiovascular Research Institute, New Jersey Medical School,

University of Medicine and Dentistry of New Jersey, Newark, New Jersey.

Background: Macrophages (Mfs) participate in host defense by orchestratinginflammation, immune responses, and wound healing. In response to micro-environmental stimuli, Mfs adopt either a classically activated (M1) pro-inflammatory phenotype or an alternatively activated (M2) wound healingphenotype.The Problem: M1 Mfs are induced by products of pathogenic agents andinterferon-g. M2 Mfs are induced in response to interleukin-4 (IL-4) andIL-13. However, recent studies show that there are IL-4=IL-13-independentpathways that induce phenotypes resembling M2 Mfs. Further character-ization is necessary to elucidate the different activating agents and functionalcharacteristics of these M2-like Mfs.Basic/Clinical Science Advances: Unstimulated Mfs express low levels ofinflammatory cytokines and growth factors such as tumor necrosis factor-a(TNF-a), IL-12, and vascular endothelial growth factor (VEGF). Stimulation ofMfs with Toll-like receptor agonists, such as lipopolysaccharide, induces M1activation, characterized by production of TNF-a and IL-12. In the inflamma-tory environment, the retaliatory metabolite adenosine rapidly accumulatesextracellularly as an adenosine triphosphate breakdown product. Costimula-tion of Mfs with toll-like receptor (TLR) agonists and adenosine switches Mfsfrom an M1 to a novel M2-like phenotype by upregulating IL-10 and (VEGF)and downregulating TNF-a and IL-12 in an adenosine A2A receptor-dependentand IL-4=IL-13-independent manner, as depicted in Fig. 1.Clinical Care Relevance: M2-like Mfs and their induction pathways presentattractive potential targets for pharmacological regulation of wound healingand inflammatory diseases.Conclusion: Switching from an M1 into an M2-like Mf phenotype can occur inan adenosine A2A receptor-dependent and IL-4=IL-13-independent manner andmay be critical to wound healing.

BACKGROUND

MACROPHAGES (MfS) PLAY A pivotalrole in host defense through orches-tration of inflammation and woundhealing. While resting Mfs are rela-tively quiescent, activation yieldspolarized phenotypes based on loca-tion and microenvironmental influ-ences. These include the presence ofactivating agents, cytokines, hyp-oxia, and ischemia, which in turnregulate expression of numerous in-

ducible genes that may be involvedin the development of Mf subtypes.Although categorization can beblurred by assorted combinations ofactivating agents, Mfs are typicallydescribed as classically activated(M1) or alternatively activated (M2).Based on the nature of activatingstimuli, Mfs may switch activationstate to other phenotypes.1–4

M1 Mfs promote inflammatoryprocesses during injury and infection

Samuel Joseph Leibovich

*Correspondence: Department of Cell Biology

and Molecular Medicine, The Cardiovascular

Research Institute, New Jersey Medical School,

University of Medicine and Dentistry of New

Jersey, 185 South Orange Ave., Medical Science

Building G606, Newark, New Jersey 07103

(e-mail: [email protected]).


A2AR¼ adenosine A2A receptor

Arg1¼ arginase-1

CD204¼ SR-A=scavengerreceptor, class A type I=II=IMfscavenger receptor-I (MSRI)

CD206¼MRCI=mannosereceptor, class C type I

DAMP¼ damage-associatedmolecular patterns

dectin-1¼ b-glucan receptor

Fizz1¼ found in inflammatoryzone-1

GC¼ glucocorticoid

IC¼ immune complex

IFN-g¼ interferon-g

IL-4¼ interleukin-4

(continued)

j 89ADVANCES IN WOUND CARE, VOLUME 2 Printed in U.S.A.Copyright ª 2011 by Mary Ann Liebert, Inc. DOI: 10.1089=awc.2010.0252

____�____0

____þ

AWC-2010-0252-Leibovich_3P.3d 01/27/11 10:04pm Page 89

that are vital for intracellular patho-gen removal. These Mfs are inducedby pathogen-associated molecularpatterns (PAMPs), such as lipopoly-saccharide (LPS), damage-associatedmolecular patterns (DAMPs), suchas those derived from mitochon-dria of apoptotic neutrophils, andinterferon-g (IFN-g) (produced by ac-tivated CD4þT helper 1 [Th1], CD8þT cytotoxic 1, and natural killer cells)to enhance pro-inflammatory re-sponses through production of cyto-kines (e.g., interleukin-1 [IL-1], IL-6,IL-12, and tumor necrosis factor-a[TNF-a]) and inflammatory media-tors (e.g., nitric oxide [NO] throughinducible NO synthase [iNOS]),and to increase phagocytic and anti-gen presenting activity. In contrast,M2 Mfs are critical for induction ofangiogenesis, tissue remodeling,and repair. M2 Mfs participate inresolving inflammation through pro-duction of anti-inflammatory cyto-kines, and growth and angiogenicfactors, and by phagocytosis andelimination of debris.1,2,4–7


Because of their role in regulatinginflammation and repair, modulationof M1 and M2 Mfs and their down-stream effects may provide thera-peutic benefits. Characterization ofM2 Mf subtypes and mechanismsby which they are induced may pro-vide targets for therapy in variousdiseases. Because of their woundhealing and anti-inflammatory prop-erties, induction of M2 phenotypesmay enhance angiogenesis and gran-ulation tissue formation in woundhealing. On the other hand, mainte-nance of an M1 phenotype may pro-mote inflammation and host defense,while reducing excessive granulationtissue formation. Angiogenesis inheart disease and tumor growthalso involves Mfs, and may be regu-lated by manipulation of M1=M2polarization.7–12

RELEVANT BASICSCIENCE CONTEXT

Mfs participate in the regulationof inflammation and wound repairby secretion of cytokines, chemokines,and growth factors and by phago-cytosis. Two main categories ofactivated Mfs have been described:pro-inflammatory (M1) Mfs and anti-inflammatory=wound healing (M2)Mfs. M2 Mfs are defined as Mfsactivated by IL-4 and IL-13 throughthe IL-4 receptor-a, and are charac-terized by upregulated expression ofIL-10, transforming growth factor-b(TGF-b), and vascular endothelialgrowth factor (VEGF ) and low ex-pression of TNF-a and IL-12.1,4 M2Mfs also express cell surface and in-tracellular markers, which are listedin Table 1. Recent studies, however,have shown that IL-4 and IL-13are not essential for induction ofthe M2-like phenotype and thatmouse wounds lacking IL-4=IL-13still contain Mfs with M1 and M2-like characteristics.2 While numerouspathways, including recognition andphagocytosis of apoptotic cells (seechapter by Roy in this volume), maymediate Mf activation, we set outto define one subtype of M2-like Mfsthat is induced by costimulation ofMfs with Toll-like receptor (TLR)and adenosine A2A receptor (A2AR)agonists.13,14

Both TLRs and A2ARs may be ex-pressed on Mfs and are importantregulators of inflammation and repair.Elucidation of signaling from thesereceptors may lead to approaches toenhance or block downstream effectsthrough regulation of Mf mediators(intracellular or secreted) or throughthe use of receptor agonists or antag-onists. For example, we have shownthat LPS activation of TLR4 inducesTNF-a expression, which is down-regulated by adenosine receptor ago-nists, whereas costimulation of TLR4and A2AR upregulates IL-10 andVEGF expression.13 The switch to an

iNOS¼ inducible nitric oxidesynthase

LPS¼ lipopolysaccharide

M1¼ classically activated

M2¼ alternatively activated

Mf¼macrophage

NO¼ nitric oxide

PAMP¼ pathogen-associatedmolecular pattern

TGF-b¼ transforming growthfactor-b

Th1¼ CD4þ T helper 1

TLR¼ Toll-like receptor

TNF-a¼ tumor necrosisfactor-a

VEGF¼ vascular endothelialgrowth factor

Ym1=Ym2¼ chitinase 3-like 3(CHI3l3)=chitinase 3-like 4(CHI3l4)

Abbreviations andAcronyms (continued)


1____

0 ____

1____


M2 phenotype favors angiogenesis and granulationtissue formation in wound healing. Knowledge of thesignaling pathways involved in this process shouldprovide an understanding of the phenotypic switchesoccurring in Mfs and yield potential therapeutictargets for enhancing wound healing or for dimin-ishing abnormal wound healing. Although Mfshave varying roles that are dependent on activa-tion by agents in the surrounding milieu, a clearunderstanding of how to manipulate and switchthese phenotypic profiles may provide potentialbenefits.


There are limitations to consider in the character-ization of M2 Mf gene expression. These includevarying expression patterns when comparing mod-els in different species and variable effects of theinflammatory milieu surrounding Mfs. AlthoughM1=M2 Mf polarization occurs in humans, studiesof M2 Mf marker expression have been predomi-nantly restricted to mice. Human homologs havenot always been found, giving rise to questions ofidentification, and indicating the need for furtherelucidation of M1=M2 Mf polarization in the hu-man system.1 It is critical to recognize that changesin the Mf background influence Mf activation.While in vitro models enable identification of acti-vation factors, exclusive focus on Mfs in culturemay overlook interactions with other cell types.

A limited focus on cytokines (such as IL-4=IL-13)in regulating Mf activation may also result in ef-fects of nonimmune-dependent metabolites, such asadenosine, being overlooked. Recent studies indi-cate IL-4=IL-13-independent M2-like Mf activa-tion, and our studies show that adenosine, throughA2AR induction by TLR agonists, mediates an M1 toM2-like switch.2,13–15 The characteristics of acti-vated Mfs also vary with time to result in earlyand late markers.16 Early wound Mfs show higherlevels of TNF-a expression due to M1 activation,which diminishes with time, whereas later woundMfs express elevated VEGF as a result of switch-ing from an M1 to an M2-like phenotype.2 Thisswitch is driven in a temporally orchestrated man-ner that depends upon changes in the microenvi-ronment of the wound (see Discussion of Findingsand Relevant Literature section).


Mfs orchestrate inflammation, tissue repair, andimmune responses by phagocytosis and destructionof foreign organisms through recognition of PAMPsand DAMPs, by production of cytokines and growthfactors, and by antigen presentation.5,6 Tissues con-tain heterogeneous populations of Mf, which havethe capacity to dramatically change their pheno-type as a result of differentiated plasticity as well asmicroenvironmental tissue- and immune-specificinfluences.2

Upon exposure to stimuli, Mfs develop phago-cytic and secretory phenotypes oriented for spe-cific functional activities. Two general categoriesof activated Mfs have been defined: M1 and M2.The M1 phenotype is driven by Th1 cytokines,such as IFN-g, and TLR agonists to induce a pro-inflammatory and cytocidal response against in-tracellular pathogens and certain transformedand cancer cells. These M1 Mfs predominate inthe early wound environment and secrete highlevels of pro-inflammatory cytokines, low levels ofanti-inflammatory cytokines, and high levels ofNO (through iNOS induction) (see Table 1).1,4,7

An M1 phenotype may be induced by PAMPs as-sociated with exogenous or commensal organisms,by endogenous TLR agonists, or by DAMPs.Failure of the innate immune response to elimi-nate these stimuli might result in the persistentinduction of M1 Mfs. M2 Mfs can be induced in aTh2-dependent manner and display expressionpatterns associated with anti-inflammatory ef-fects, Th1 response inhibition, and wound healingpromotion. In contrast, M2 Mfs express low levels

TARGET ARTICLES

1. Daley JM, Brancato SK, Thomay AA,Reichner JS, and Albina JE: The phenotype ofmurine wound macrophages. J Leukoc Biol2010; 87: 59.

2. Martinez FO, Helming L, and Gordon S:Alternative activation of macrophages: an im-munologic functional perspective. Annu RevImmunol 2009; 27: 451.

3. Ramanathan M, Luo W, Csoka B, Hasko G,Lukashev D, Sitkovsky M, and Leibovich SJ:Differential regulation of HIF-1a isoforms inmurine macrophages by TLR4 and adenosine A2A

receptor agonists. J Leukoc Biol 2009; 86: 681.4. Macedo L, Pinhal-Enfield G, Alshits V,

Elson G, Cronstein BN, and Leibovich SJ:Wound healing is impaired in MyD88-deficientmice: a role for MyD88 in the regulation ofwound healing by adenosine A2A receptors. Am JPathol 2007; 171: 1774.

MACROPHAGE HETEROGENEITY AND WOUND HEALING 91

____�____0

____þ

AWC-2010-0252-Leibovich_3P.3d 01/27/11 10:04pm 1

of pro-inflammatory mediators and higher levelsof anti-inflammatory mediators and fibrogenicand angiogenic factors (see Table 1 and Fig. 1).M2 Mfs recruit a specific group of cells in theircoordination of immune responses through secre-tion of chemokines that attract monocytes, baso-phils, memory T cells, Th2 cells, and eosinophils.Some M2 Mf markers are listed in Table 1.1,4,8 Inaddition to their central role in immune re-sponses, M2 Mfs also play a role in some patho-logical processes, including vulnerability topathogens due to dampening of Th1 responses byM2 Mfs, allergic responses involving upregulatedM2 gene expression in asthma, tumor progressionassisted by tumor-associated M2-like Mfs, andfibroproliferative complications of infection andinflammation.7–12

M2 Mfs that are activated by IL-4 and IL-13(produced by activated CD4þ Th2 and CD8þ Tc2cells, natural killer cells, basophils, mast cells, andeosinophils) are designated as M2 and are associ-ated with allergic and antiparasitic responses.These Mfs show decreased production of IL-1b andIL-8 and decreased induction of oxygen radicalsthrough the respiratory burst, and express ele-vated levels of CD206 (MRCI=mannose receptor,class C type I) and CD204 (SR-A=scavenger re-ceptor, class A type I=II=IMf scavenger receptor-I[MSRI]).1,4,17,18 Further characterization of Mfshas demonstrated other types of M2-like Mfs in-volved in wound healing that are not dependent onIL-4=IL-13 induction, and M2 Mfs have beensubclassified based on inducing agents and subse-quent expression patterns. M2 Mf subcategoriesinclude the more commonly investigated M2a Mf(activated by IL-4 or IL-13), M2b Mf (induced byimmune complexes [ICs] and IL-1b or TLR ago-nists), and M2c Mf (stimulated by IL-10, TGF-b,or glucocorticoids).1 It should be emphasized thatthere are M2 Mfs that are not associated with aTh2 immune response or IL-4=IL-13 activation, asseen in M2b and M2c Mfs. Further, these acti-vated Mfs show overlapping and nonoverlappingexpression marker patterns. Albina and colleaguesrecently described a dynamic M2 Mf phenotypethat is not dependent on IL-4 or IL-13 activation.In IL-4 receptor-a KO mice and in the presence ofthe IL-13Ra2 decoy receptor, activation occurs inspite of inhibition of IL-13-dependent phosphory-lation of downstream STAT6 and the absence ofIL-4 or IL-13 in the wound environment, andthese mice exhibit Mfs with both M1 and M2 ex-pression patterns. Wound Mfs in early phases ofrepair are more M1-like with elevated expressionof TNF-a and IL-6 and less TGF-b, whereas those

in later phases are more M2 like, with less pro-inflammatory cytokines, no induction of iNOS,and elevated markers of alternative activation,including CD206, b-glucan receptor (dectin-1), ar-ginase-1 (Arg1), and chitinase 3-like 3 (Ym1).2

In contrast to the commonly described M2 Mfsinduced by IL-4 or IL-13, we have demonstrateda new type of M2-like Mf that is induced byphenotypic switching of M1 Mfs in an IL-4=IL-13-independent manner. This novel mechanismof Mf phenotypic switching to produce an M2-like Mf requires initial stimulation of TLRs,which strongly induces A2AR expression, followedby ligation of A2ARs by adenosine. This inductionof A2ARs plays a key role in switching pro-inflammatory M1 to an angiogenic M2-like phe-notype. We have previously shown that whilethe TLR4 agonist LPS (as well as agonists ofTLR2, 7, 9) strongly induce TNF-a and IL-12 ex-pression in Mfs, adenosine present in the micro-environment (as a metabolite formed from thebreakdown of adenosine triphosphate) stronglyinhibits this induction while simultaneously up-regulating IL-10 and VEGF expression.13,14 Blacket al. described upregulated Arg1 expression andactivity and no effect on expression of other M2markers (found in inflammatory zone-1 [Fizz1]and Ym) in Mfs treated with LPS and A2AR ag-onists.19 Thus, to define M2 Mfs in greater depth,we suggest designating these TLR- and A2AR-induced M2-like Mfs as M2d Mfs. A key predic-tor of this proposed pathway is that these M2dMfs should express elevated levels of A2ARs.We are currently investigating this hypothesis indetail (Figure 1).

INNOVATION

We propose the redefinition of M2 Mfs to include anew subtype that is induced by switching of M1Mfs into M2-like Mfs through activation of TLRsand A2ARs. We suggest that these Mfs be calledM2d Mfs. These newly described M2d Mfs expresselevated VEGF and IL-10 and diminished TNF-aand IL-12.13 Analysis of the influence of multiplecostimulators on Mfs allows for the integratedanalysis of pathways downstream of these recep-tors and the resulting phenotypic Mf profiles. Thisanalysis may more accurately mimic true micro-environmental conditions, where adenosine accu-mulation occurs in inflammatory, ischemic, andhypoxic settings, leading to phenotype switchingfrom pro-inflammatory M1 Mfs to wound healing,anti-inflammatory M2d Mfs in a temporally de-fined manner. Singular activation of these recep-

1____

0 ____

1____



tors may not show expression patterns seen incostimulation models. Identification of this new IL-4=IL-13-independent and A2AR-dependent M2dMf shows that Mf activation and its role in woundhealing is a complex event involving numerousmicroenvironmental stimuli.


The observation that both classical and alternativelyactivated Mf phenotypes are found in wounds ofmice where IL-4 and IL-13 signaling is absent

raises the question of the origin of these heteroge-neous Mf populations.2 Our discovery of the A2AR-dependent switch of M1 Mfs to an M2-like phe-notype provides one possible mechanism for thisheterogeneity, and provides a simple, temporallyregulated pathway for the development of M2dMfs. Mfs are initially activated to an M1 phenotypeby PAMPs and DAMPs, which also induce A2ARexpression. When (and if ) extracellular adenosinelevels rise as a result of adenosine triphosphatebreakdown in response to cell stress, inflammation,hypoxia, or ischemia, signaling through the ele-vated A2ARs switches the M1 Mfs into the M2d

FIGURE 1. Pathways of Mf activation. Nonactivated Mjs are induced to express activated phenotypes based on specific inducing agents, and to exhibitexpression patterns of markers that may overlap with those of other phenotypic profiles. Note that the aim of this figure is to provide a general representationof M2 Mf expression. Not all agents are included, and expression profiles for different M2 Mf subtypes vary based on inducing agents.

Table 1. Phenotypic profiles of activated macrophages

M1 M2a, M2b, M2c, M2d

Inducing agents IFN-g, TLR agonists (e.g., LPS) M2a: IL-4, IL-13M2b: ICþ IL-1b or TLR agonistsM2c: GC, IL-10, TGF-bM2d: TLR agonistsþ adenosine

High expression TNF-a, IL-12, IL-1b, IL-6, IL-8, IL-23, iNOS, CCL3 CD206, CD204, dectin-1, IL-10, TGF-b, Arg1, Ym1=Ym2, Fizz1, CCL18Low expression CD206, IL-10, TGF-b TNF-a, IL-12, IL-1b, IL-6, IL-8, IL-23, iNOSGeneral phenotypic profile Inflammatory Th1 response Anti-inflammatory, wound healing Th2 response

Shown in the table is a partial list of activating agents and subsequent marker expression patterns in M1 and M2 Mfs. Note that some markers may be specificto certain Mf types or may overlap in some profiles.

____�____0

____þ



phenotype, which then participates in the promo-tion of angiogenesis and repair.13,14 Future studieswill determine the contribution of this pathway toaberrant wound healing, fibrosis in disease, andcancer development. Analysis of signaling path-ways that mediate the M1=M2d switch will providefocus for the development of pharmaceutical ap-proaches to regulating this switch.

FUTURE DEVELOPMENTS

The discovery of a novel A2AR-dependentM2 Mf subtype (M2d) that is induced in-dependently of IL-4=IL-13 provides a po-tential target for modulating inflammationand wound healing by pharmacologicalregulation of the M1=M2d switch. Our in-vestigation continues with characteriza-tion of M2d Mfs and signaling pathwaysthat mediate their induction. Subsequentstudies will examine Mf phenotypes inwound healing, chronic inflammatory, andtumor models, to determine when andwhere the M2d phenotype is expressed.While most studies of this M1=M2d switchare in murine Mfs, human monocytes alsoundergo a similar LPS=A2AR-dependentphenotypic switch in vitro. Future studiesshould characterize human monocytes=Mfs and the role of the A2AR-mediatedswitch in regulating their properties. Thesestudies will help phenotype these M2 Mfsand will allow to the testing of agonists andantagonists of signaling pathways involvedin acquisition of these phenotypes, with theaim of developing therapeutics for theirregulation. Maintenance of an M1 pheno-type might promote clearance of pathogenicorganisms and wound debridement; pro-motion of an M2 phenotype might enhanceangiogenesis and granulation tissue for-mation, whereas reverting M2 Mfs to anM1 phenotype could diminish excessivegranulation and scarring.

ACKNOWLEDGEMENTS

Macrophage biology research in S.J.L.’slaboratory was supported by National In-stitutes of Health grant RO1-GM068636.The contents of this chapter are solelythe responsibility of the authors and do

not necessarily represent the official views of theNational Institute of General Medical Science(NIGMS) or National Institutes of Health.


The authors have no conflicts of interest. No ghost-writers were used in the writing of this chapter.

TAKE-HOME MESSAGE


� Mfs exhibit different phagocytic and secretory phenotypes based on theiractivation status. Further, changes in the microenvironment provide differentsets of stimulating agents, which allow for Mfs to switch phenotypes.

� M2 Mfs make up a broad category of anti-inflammatory, wound healingMfs that downregulate effects of pro-inflammatory M1 Mfs and haveoften been described in the context of IL-4 and IL-13 activation.

� While IL-4 and IL-13 can induce an M2 alternatively activated phenotype,wounds in mice where IL-4 and IL-13 signaling is absent still exhibit bothM1 and M2-like Mf phenotypes.

� We have defined a new IL-4- and IL-13-independent subtype of M2-likeMfs that is induced by switching of M1 into M2-like Mfs through the actionof adenosine. This switch requires the induction of A2AR expression, whichoccurs during M1 activation by TLR-mediated signaling. We suggest thatthese Mfs should be called M2d Mfs. M2d Mfs are anti-inflammatory andpro-angiogenic, and exhibit a phenotypic profile characterized by down-regulation of TNF-a and IL-12 and upregulation of VEGF and IL-10 expression.

� M2d Mfs are potential candidates for therapies for regulation of in-flammation and repair. These Mfs may also be involved in pathologicalconditions such as abnormal repair, fibroproliferative diseases, and tumorprogression.


� Identification of Mf subpopulations that regulate both the inflammatoryand fibroproliferative, angiogenic phases of wound repair provides insightinto potential mechanisms for therapeutic modulation of aberrant woundhealing.

� The identification of the role of A2ARs in the switching of Mfs from anM1 to an M2d phenotype provides insight into the role of adenosine inthe regulation of inflammation and angiogenesis (see chapter by Chanand Cronstein in this volume).

� While heterogeneity of Mfs has been shown in numerous situations,including chronic wounds, fibrotic livers, lungs, kidneys, atheroscleroticplaques, and developing tumors, the mechanistic basis of this heteroge-neity is not always clear. The identification of the A2AR-mediated pathwayfor the induction of an M2 Mf subpopulation provides a new paradigm forthe analysis of Mfs, and for potential therapeutic approaches.

1____

0 ____

1____



REFERENCES

1. Martinez FO, Helming L, and Gordon S: Alternativeactivation of macrophages: an immunologic func-tional perspective. Annu Rev Immunol 2009; 27: 451.

2. Daley JM, Brancato SK, Thomay AA, Reichner JS,and Albina JE: The phenotype of murine woundmacrophages. J Leukoc Biol 2010; 87: 59.

3. Porcheray F, Viaud S, Rimaniol AC, Leone C, Sa-mah B, Dereuddre-Bosquet N, Dormont D, andGras G: Macrophage activation switching: an as-set for the resolution of inflammation. Clin ExpImmunol 2005; 142: 481.

4. Classen A, Lloberas J, and Celada A: Macrophageactivation: classical versus alternative. MethodsMol Biol 2009; 531: 29.

5. Taylor PR, Martinez-Pomares L, Stacey M, Lin HH,Brown GD, and Gordon S: Macrophage receptorsand immune recognition. Annu Rev Immunol 2005;23: 901.

6. Rubartelli A and Lotze MT: Inside, outside, upsidedown: damage-associated molecular-pattern mole-cules (DAMPs) and redox. Trends Immunol 2007;28: 429.

7. Van Ginderachter JA, Movahedi K, GhassabehGH, Meerschaut S, Beschin A, Raes G, and Baet-selier P: Classical and alternative activation ofmononuclear phagocytes: picking the best of bothworlds for tumor promotion. Immunobiology 2006;211: 487.

8. Mantovani A: Macrophage diversity and polari-zation: in vivo veritas. Blood 2006; 108: 408.

9. Mantovani A, Sozzani S, Locati M, Allavena P,and Sica A: Macrophage polarization: tumor-as-sociated macrophages as a paradigm for polarizedM2 mononuclear phagocytes. Trends Immunol2004; 25: 677.

10. Luo Y, Zhou H, Krueger C, Kaplan S, Lee SH,Dolman C, Markowitz D, Wu W, Liu C, Reisfeld RA,and Xiang R: Targeting tumor-associated macro-phages as a novel strategy against breast cancer.J Clin Invest 2006; 116: 2132.

11. Yang HZ, Cui B, Liu HZ, Chen ZR, Yan HM, Hua F,and Hu ZW: Targeting TLR2 attenuates pulmonaryinflammation and fibrosis by reversion of sup-pressive immune microenvironment. J Immunol2009; 182: 692.

12. Porta C, Rimoldi M, Raes G, Brys L, Ghezzi P, DiLiberto D, Dieli F, Ghisletti S, Natoli G, DeBaetselier P, Mantovani A, and Sica A: Toleranceand M2 (alternative) macrophage polarization arerelated processes orchestrated by p50 nuclearfactor B. Proc Natl Acad Sci U S A 2009; 106:14978.

13. Macedo L, Pinhal-Enfield G, Alshits V, Elson G,Cronstein BN, and Leibovich SJ: Wound healing isimpaired in MyD88-deficient mice: a role forMyD88 in the regulation of wound healing by

adenosine A2A receptors. Am J Pathol 2007; 171:1774.

14. Ramanathan M, Luo W, Csoka B, Hasko G, Lu-kashev D, Sitkovsky M, and Leibovich SJ: Differ-ential regulation of HIF-1a isoforms in murinemacrophages by TLR4 and adenosine A2A receptoragonists. J Leukoc Biol 2009; 86: 681.

15. Stout RD: Macrophage functional phenotypes: noalternative in dermal wound healing? J LeukocBiol 2010; 87: 19.

16. Menzies FM, Henriquez FL, Alexander J, and Ro-berts CW: Sequential expression of macrophageanti-microbial=inflammatory and wound healingmarkers following innate, alternative and classicalactivation. Clin Exp Immunol 2010; 160: 369.

17. Stein M, Keshav S, Harris N, and Gordon S: In-terleukin 4 potently enhances murine macrophagemannose receptor activity: a marker of alternativeimmunologic macrophage activation. J Exp Med1992; 176: 287.

18. Abramson SL and Gallin JI: IL-4 inhibits superox-ide production by human mononuclear phago-cytes. J Immunol 1990; 145: 1435.

19. Black SG, Wilson JM, Ernst PB, and Smith MF: A2A

adenosine receptor stimulation enhances arginaseI expression in macrophages resulting in a pheno-typically unique macrophage. FASEB J 2008; 22:1065.25 (Abstract).

____�____0

____þ



REVIEW DIGEST

Sustained Release and Dual DeliveryStrategies for Platelet-Derived Growth Factor

Scott A. Guelcher*

Department of Chemical and Biomolecular Engineering, Vanderbilt University, Nashville, Tennessee.

Background: Strategies for improving impaired wound healing are clinicallyrelevant areas of investigation. Many of the consequences from failed healingmay be caused by altered intracellular signaling, which implies that admin-istration of signaling molecules to sites of tissue repair is a viable strategy forimproving healing.The Problem: Local delivery of recombinant human platelet-derived growthfactor BB (PDGF-BB) from a carboxymethylcellulose gel is a Food and DrugAdministration–approved therapy for treatment of diabetic foot ulcers. Recentstudies have aimed to overcome the limitations of the bolus release of growthfactor from topical gels by designing drug delivery systems and scaffolds ca-pable of sustained release and dual delivery with other factors and cells.Basic/Clinical Science Advances: Dual delivery of PDGF-BB and transforminggrowth factor-beta was reported to increase the tensile strength of incisionalwounds relative to no treatment and treatment with a single growth factor atvery low dosages (<5 mg cm�3). Another study reported that dual delivery offibroblasts and PDGF-BB enhanced ingrowth of granulation tissue and woundresurfacing relative to delivery of fibroblasts alone. Other studies cited in thisarticle show that the sustained release of PDGF-BB enhances wound healingrelative to the bolus release associated with topical gels.Clinical Care Relevance: Alternative delivery approaches that achieve pre-dictable wound healing at lower PDGF-BB doses are anticipated to reduce theneed for repeated administration associated with a high-concentration bolusrelease.Conclusion: Sustained local delivery of PDGF-BB, dual delivery of PDGF-BBand transforming growth factor beta, and dual delivery of fibroblasts andPDGF-BB are promising approaches for improving the healing of impairedwounds relative to currently available therapies.

BACKGROUND

WOUND HEALING IS a complex series ofevents regulated by intercellular sig-naling. Failure leads to negative pa-tient outcomes, such as increases inhospitalization and additional surgi-cal procedures. Therefore, strategiesfor improving impaired wound heal-ing are clinically relevant areas ofinvestigation. Wound healing pro-ceeds through a highly controlledsequence of events regulated by in-

tercellular communication via cyto-kines and growth factors. Many of theconsequences from failed healingmay be caused by altered intercellu-lar signaling, which implies that ad-ministration of signaling molecules tosites of tissue repair is a viable strat-egy for improving healing. Local de-livery of recombinant human platelet-derived growth factor BB (PDGF-BB)has been shown to enhance thehealing of cutaneous defects under

Scott A. Guelcher

*Correspondence: Department of Chemical

and Biomolecular Engineering, Vanderbilt Uni-

versity, VU Station B #351604, 2301 Vanderbilt

Place, Nashville, Tennessee 37235. (e-mail:

[email protected]).


FDA¼ Food and Drug Adminis-tration

FS¼ fibrin sealant

PDGF-BB¼ platelet-derivedgrowth factor BB

TGF-b¼ transforming growthfactor-beta

VEGF¼ vascular endothelialgrowth factor

____�____0

____þ

review-article


AWC-2010-0250-Guelcher_3P.3d 01/27/11 10:04pm 3

impaired healing conditions. In recent years, sev-eral studies have aimed to overcome the limitationsof the bolus release of growth factor from topical gelsby designing drug delivery systems and scaffoldscapable of sustained release. The target articles re-viewed in this article1–3 investigate the strategiesfor improving the efficacy of locally delivered PDGF-BB by a sustained release delivery system, codeliv-ery of another factor associated with wound healing,or codelivery with fibroblasts.


Regranex�, a carboxymethylcellulose gel incorpo-rating 100mg cm�3 PDGF-BB, is a Food and DrugAdministration (FDA)–approved therapy for treat-ing foot ulcers. Because of the bolus release ofgrowth factor, the gel must be applied at least dailyat high concentrations. A retrospective study founda fivefold increased risk of cancer mortality inpatients exposed to three or more tubes of Regranexcompared with patients who had not been exposed.Although this is a controversial unpublished report,it nevertheless prompted FDA to mandate a blackbox warning on the product label. Sustained releaseof PDGF-BB is anticipated to reduce the amount ofdrug delivered, thereby reducing the frequency ofcomplications and improving patient outcomes.


Implanting or injecting a biodegradable, biocom-patible scaffold into a tissue defect is an establishedapproach for regenerating tissue and restoringits architecture. By adding an appropriate growth

factor(s), the scaffold can also function as a localdelivery system in addition to providing mechani-cal support. The scaffold must be biocompatible,support the infiltration of cells and new tissue,degrade to nontoxic decomposition products at arate comparable to that of new tissue ingrowth, andrelease biologics at a time scale that is relevant tothe biology of the wound site. PDGF is a usefulgrowth factor for wound healing because of itschemotactic and mitogenic properties for a widevariety of cell types,4,5 as well as the fact that it is acofactor with vascular endothelial growth factor(VEGF) for angiogenesis.6 PDGF-BB, the mostwidely investigated isoform for tissue regenera-tion, has been reported to stimulate the formationof new tissue in both healthy7 and diabetic rats.8 Itis desirable to lower the dose of PDGF to reduce thefrequency of complications and improve patientoutcomes. Several approaches have been taken toimprove the safety and efficacy of PDGF-BB, in-cluding development of sustained release vehicles,2

codelivery with other growth factors associatedwith wound healing (such as transforming growthfactor-beta [TGF-b]1), and codelivery with fibro-blasts.3 The effects of each of these strategies onwound healing in preclinical models are reviewedin this article.


In one of the reviewed studies, the effects of PDGFand TGF-b, delivered either alone or in combina-tion, on the healing of 6-cm incisional wounds wereinvestigated in a cyclophosphamide-induced im-paired wound healing model in rats.1 The deliverysystem comprised two 0.7�15 mm ethylene-vinylacetate copolymer rods that have been shown toachieve sustained release of growth factors.9 Thedoses were 3.0 mg cm�3 PDGF-BB and=or 2.0mg cm�3 TGF-b.9 The effects of dual drug deliveryon healing were evaluated using histology andbiomechanical testing on days 4, 7, and 14 post-implantation.

In another study, the effects of dual delivery ofPDGF-BB and fibroblasts on the healing of full-thickness biopsy wounds in a rabbit ear model wereinvestigated.3 The delivery system comprised fourfibrinogen–thrombin formulations of fibrin sealant(FS) combined with fibroblasts and PDGF-BB. Thedose of PDGF-BB was 3 mg in a 7�0.27 mm full-thickness wound (*300,000mg cm�3). The extentof wound resurfacing and ingrowth of granulationtissue were measured histologically on day 7 post-injection.

TARGET ARTICLES

1. Ashraf A, Lee PH, Kim K, Zaporojan V,Bonassar L, Valentini R, et al.: Effect ofsustained-release PDGF and TGF-beta oncyclophosphamide-induced impaired woundhealing. Plast Reconstr Surg 2009; 124: 1118.

2. Li B, Davidson JM, and Guelcher SA: Theeffect of the local delivery of platelet-derivedgrowth factor from reactive two-componentpolyurethane scaffolds on the healing in rat skinexcisional wounds. Biomaterials 2009; 30: 3486.

3. Mogford JE, Tawil B, Jia S, and Mustoe TA:Fibrin sealant combined with fibroblasts andplatelet-derived growth factor enhance woundhealing in excisional wounds. Wound Repair Re-gen 2009; 17: 405.

1____

0 ____

1____


AWC-2010-0250-Guelcher_3P.3d 01/27/11 10:04pm Page 144

In the sustained release approach, the effects ofPDGF-BB delivered from a biodegradable poly-urethane scaffold on the healing of full-thicknessexcisional wounds in rats were investigated.2

PDGF-BB doses of 31.6 and 316mg cm�3 were tes-ted. Ingrowth of new granulation tissue and scaffolddegradation were measured by histomorphometryon days 3, 7, and 14 postimplantation.


The results from the three target reports showthat the efficacy of locally delivered PDGF-BB canbe enhanced by (a) sustained release up to day 7,(b) codelivery with another factor, such as TGF-b,or (c) by codelivery with fibroblasts. In a study in-vestigating the effects of PDGF-BB released frombiodegradable polyurethane scaffolds, PDGF-BB(31.6mg cm�3) was observed to accelerate cellularinfiltration, ingrowth of new tissue, and scaffolddegradation relative to blank scaffold controls atdays 3, 7, and 14 in a rat excisional wound model.2

Interestingly, no differences were observed at anorder of magnitude higher dose (316mg cm�3).These observations are consistent with a previousstudy reporting that sustained release of PDGF-BBfrom poly(lactic-co-glycolic) acid microspheres em-bedded in poly(L-lactic acid) nanofibrous scaffoldsimplanted subcutaneously in rats increased theformation of new tissue at days 7 and 14 relative toa bolus release of PDGF-BB.7 Interestingly, in thenanofiber scaffolds, dose-dependent effects on tis-sue ingrowth were observed when the PDGF-BBdose was increased from 31 to 310mg cm�3. Further,at a low dose, ingrowth of new tissue was signifi-cantly higher for the slow-release scaffolds (day 1burst release <5%) compared with the fast-releasescaffolds (day 1 burst release¼ 30%). Surprisingly,for the polyurethane scaffolds characterized by aday 1 burst release of 60%, formation of new gran-ulation tissue was significantly higher relative tothe empty scaffold controls as early as day 3, com-pared with day 7 for the nanofiber scaffolds withlinear release of PDGF-BB.

In the PDGF-BB=TGF-b dual-delivery study,cyclophosphamide treatment was found to signif-icantly decrease the tensile strength at break of6-mm incisional wounds relative to untreatedcontrols at all time points, thus validating theimpaired healing model. Interestingly, delivery ofeither PDGF-BB or TGF-b alone did not signifi-cantly increase the tensile strength of the woundsat all time points and, in fact, resulted in a de-crease in wound strength relative to the control

(no growth factor treatment). However, the combi-nation of PDGF-BB (3mg cm�3) and TGF-b (2mgcm�3) was observed to significantly increase woundstrength at days 7 and 14. The ethylene-vinyl ace-tate copolymer delivery system exhibited a linearrelease of growth factor with no burst.9 In anotherstudy investigating the effects of PDGF-BB deliv-ered from a fibrin gel on the healing of flexor digi-torum profundus tendons in dogs, PDGF-BB (3.3mgcm�3) was observed to significantly improve func-tional properties, but not tensile properties.10 Thefailure to improve mechanical properties was at-tributed to suboptimal PDGF-BB dose or releasekinetics. Taken together, these results suggestthat the failure of PDGF-BB to improve the ten-sile strength of incisional wounds at a dose of *3mgcm�3 could be attributed to a suboptimal dose and=orrelease kinetics. The incisional wound study dem-onstrates that enhanced healing can be achieved at arelatively low dose of PDGF-BB through codeliveryof another factor associated with wound healing,which is conjectured to more closely mimic the invivo wound healing environment.

Another approach to improving the efficacy ofPDGF is codelivery with fibroblasts. Delivery offibroblasts from FS was reported to increase boththe area of granulation tissue as well as the per-centage of wound coverage at day 7 postinjection.Interestingly, incorporation of fibroblasts in the FSwas observed to elicit migration of the epithe-lium over the material, as well as incorporation ofgranulation tissue within the material. The au-thors conjectured that secretion of cytokines by fi-broblasts delivered in the FS altered the behaviorof and=or protease secretion by the host cells,thereby modifying the chemical composition of theFS and accelerating infiltration of granulation tis-sue. The authors further speculated that this en-hanced integration of the FS could lead to moreeffective delivery of therapeutic agents as well asless scarring. Addition of PDGF-BB to the fibrino-gen component with fibroblasts significantlyincreased both granulation area and surface cov-erage relative to delivery of fibroblasts alone. Theseresults are consistent with another study reportingthat delivery of a plasmid encoding for VEGF froma fibrin gel increased VEGF-A protein expression,tissue perfusion, and flap survival in a rat model at7 days postimplantation.11 In the rabbit ear study,the effects of PDGF-BB alone were not investi-gated and so it is difficult to identify the relativecontributions of the fibroblasts and PDGF-BB.Further, the dose of PDGF-BB was estimated to be*300,000mg cm�3, which is substantially higherthan that used clinically or in other preclinical

____�____0

____þ

DELIVERY STRATEGIES FOR PDGF 145


studies. Taken together, these studies suggest thatdelivery of cells in combination with PDGF-BBenhances wound healing and is a potentially ef-fective approach for improving the efficacy ofPDGF for the healing of cutaneous defects.

INNOVATION

Several significant innovations have been re-ported, which contribute to substantially movingthe field forward. Sustained release of PDGF-BBfrom biodegradable polymeric scaffolds has beenshown to enhance the healing of cutaneouswounds relative to a bolus release of the growthfactor. Second, codelivery of PDGF-BB with TGF-benhances the healing of incisional wounds at dosesof 2 orders of magnitude lower than that incorpo-rated in FDA-approved topical gels. These resultsshow that dual delivery of growth factors enhanceshealing relative to delivery of one factor alone. In

another study, dual delivery of fibroblasts andPDGF-BB has been reported to significantly in-crease the area of granulation tissue and woundresurfacing. Taken together, these studies suggestthat the sustained release and dual-delivery ap-proaches improve wound healing outcomes by moreaccurately mimicking the in vivo microenviron-ment, potentially providing the significant benefitof requiring a lower dose of growth factor.


(A) Tensile strengths required to break 6-cm inci-sional wounds in rats at days 4, 7, and 14 post-wounding (n¼ 4 per time point) are shown. Dorsalskin specimens were tested for tensile breakingstrength. Error bars represent standard deviations.Groups were found to be significantly different bytwo-way analysis of variance ( p< 0.0001). For eachtime point, the Tukey post hoc multiple comparison

test was used to evaluate differencesamong groups ( p< 0.05). On days 7 and14, delivery of either PDGF-BB or TGF-bsignificantly reduced wound strength,whereas dual delivery significantly in-creased wound strength relative to thecontrol.1 (B) Granulation tissue areameasured in 7-mm full-thickness exci-sional wounds in rabbit ears on day 7postoperation is shown. Wounds weretreated with one of four FS formulationsalone or FS with embedded rabbit dermalfibroblasts. Inclusion of dermal fibro-blasts (3�106 cells=wound) in the fibrin-ogen component of the sealant increasedthe measured area of granulation tissuefor formulations 2–4, although no differ-ences were observed between formula-tions. Granulation tissue formation wasdramatically enhanced by addition ofPDGF-BB as tested with formulation 4.Data are shown as mean�SEM. Back-ground lines indicate saline control lev-els (n¼ 12). *p< 0.05 and **p< 0.01versus FS alone for each formulation bytwo-way t-test. {Versus FSþ cell groupsby one-way factorial analysis of variancewith Tukey’s highly significantly differ-ent test.3 (C) Histology and histomor-phometric analysis of biodegradablepolyurethane scaffolds with and withoutPDGF-BB implanted in 6-mm excisionalwounds in Sprague–Dawley rats wasperformed. Implants were harvestedon days 3, 7, and 14 postimplantation,

TAKE-HOME MESSAGE


� Several new approaches were proposed for enhancing the effects oflocally delivered PDGF-BB on cutaneous wound healing. In one study,dual delivery of PDGF-BB and TGF-b was reported to increase the tensilestrength of incisional wounds relative to no treatment and treatment witha single growth factor at very low dosages (<5 mg cm�3).1 Another studyreported that dual delivery of fibroblasts and PDGF-BB enhanced theingrowth of granulation tissue and wound resurfacing relative to deliveryof fibroblasts alone.3 Other studies show that sustained release of PDGF-BB enhances wound healing relative to the bolus release associated withtopical gels.2,7

Clinical science advance

� Local delivery of PDGF-BB from a topical carboxymethylcellulose gel is anFDA-approved therapy for diabetic foot ulcers. However, the gel must beapplied at least daily at high concentrations. A retrospective study found afivefold increased risk of cancer mortality in patients exposed to three ormore tubes of gel incorporating PDGF-BB compared with patients who hadnot been exposed. Alternative delivery approaches that achieve predict-able wound healing at lower PDGF-BB doses are anticipated to reduce theincidence of complications associated with a high-concentration bolusrelease.


� The studies reviewed in this article present compelling preclinical data thatsustained release of PDGF-BB and dual-delivery approaches are potentialtherapies for treatment of diabetic foot ulcers, which reduce the risk ofsecondary complications. Potentially useful future therapies include in-jectable or implantable devices that achieve sustained release of PDGF-BBand=or devices for dual delivery of autologous cells and PDGF-BB. Aparticularly useful characteristic of the polyurethane sustained releasedelivery systems is that they can be mixed with PDGF-BB at the point of useand injected into irregularly shaped cutaneous defects.

1____

0 ____

1____



embedded in paraffin, cut into thin (*5 mm) sec-tions, and stained with trichrome. In the repre-sentative histological sections (left), the white,red, and green sections represent the presence ofpolymer, tissue, and collagen deposition, respec-tively. The area percentages of polymer and newgranulation tissue within the implants at eachtime point are shown in the plots on the right.Polyurethane scaffolds incorporating PDGF-BBshowed significantly higher percentages of newtissue area and lower percentages of scaffold arearelative to the control at each time point.2


It is important to consider that the studies reviewedin this article were performed using varying dos-

ages of PDGF-BB. Although the FS incorporatingfibroblasts exhibited enhanced healing whenPDGF-BB was delivered with fibroblasts, the doseof PDGF-BB was considerably higher than thatused clinically and the effects of PDGF-BB inde-pendent of fibroblasts were not evaluated. Futurestudies investigating the independent contribu-tions of fibroblasts and PDGF-BB at lower doseswould be of interest.

FUTURE DEVELOPMENTS

Although topical gels incorporating PDGF-BB arean FDA-approved therapy for treatment of dia-betic foot ulcers, there are concerns regardingcomplications and patient morbidity due to thebolus release of growth factor, which necessitatesa high dose and repeated applications. A recent

____�____0

____þ

DELIVERY STRATEGIES FOR PDGF 147


study has shown that sustained release of PDGF-BB for at least 7 days enhances healing relative toa bolus release,7 suggesting that effective woundhealing can be achieved at lower dosages in asustained release delivery system.2 A potentiallyuseful therapy is envisioned to comprise a scaf-fold=delivery system that delivers both autologouscells as well as PDGF-BB. The dual-delivery ap-proach is anticipated to yield a device that im-proves wound healing at a substantially lower

dose than that can be achieved through delivery ofPDGF-BB alone.

ACKNOWLEDGMENT

Funding from the National Institutes of Health(R01 AR056138-01A2) is acknowledged.


The author has nothing to disclose.

REFERENCES

1. Ashraf A, Lee PH, Kim K, Zaporojan V, Bonassar L,Valentini R, et al.: Effect of sustained-releasePDGF and TGF-beta on cyclophosphamide-inducedimpaired wound healing. Plast Reconstr Surg2009; 124: 1118.

2. Li B, Davidson JM, and Guelcher SA: The effect ofthe local delivery of platelet-derived growth factorfrom reactive two-component polyurethane scaf-folds on the healing in rat skin excisional wounds.Biomaterials 2009; 30: 3486.

3. Mogford JE, Tawil B, Jia S, and Mustoe TA: Fibrinsealant combined with fibroblasts and platelet-derived growth factor enhance wound healing inexcisional wounds. Wound Repair Regen 2009;17: 405.

4. Centrella M, McCarthy T, Kusik W, and CanalisE: Isoform-specific regulation of platelet-derivedgrowth factor activity and binding in osteoblast-

enriched cultures from fetal rat bone. J Clin Invest1992; 89: 1076.

5. Gruber R, Varga F, Fischer M, and Watzek G:Platelets stimulate proliferation of bone cells:involvement of platelet-derived growth factors,microparticles and membranes. Clin Oral ImplantsRes 2002; 13: 529.

6. Heldin CH and Westermark B: Mechanism of ac-tion and in vivo role of platelet-derived growthfactor. Physiol Rev 1999; 70: 1283.

7. Jin Q, Wei G, Lin Z, Sugai JV, Lynch SE, Ma PX, et al.:Nanofibrous scaffolds incorporating PDGF-BB micro-spheres induce chemokine expression and tissueneogenesis in vivo. PLoS ONE 2008; 3: e1729.

8. Grotendorst GR, Martin GR, Pencev D, Sodek J,and Harvey AK: Stimulation of granulation tissueformation by platelet-derived growth factor in

normal and diabetic rats. J Clin Invest 1985; 76:2323.

9. Kim HD and Valentini RF: Human osteoblast re-sponse in vitro to platelet-derived growth factorand transforming growth factor-beta deliveredfrom controlled-release polymer rods. Biomaterials1997; 18: 1175.

10. Gelberman RH, Thomopoulos S, Sakiyama-ElbertSE, Das R, and Silva MJ: The early effects ofsustained platelet-derived growth factor adminis-tration on the functional and structural propertiesof repaired intrasynovial flexor tendons: an in vivobiomechanic study at 3 weeks in canines. J HandSurg Am 2007; 32: 373.

11. Michlits W, Mittermayr R, Schafer R, Redl H, andAharinejad S: Fibrin-embedded administration ofVEGF plasmid enhances skin flap survival. WoundRepair Regen 2007; 15: 360.

1____

0 ____

1____



REVIEW DIGEST

Management of Scars with Lasers

Jennifer L. MacGregor* and Tina S. AlsterWashington Institute of Dermatologic Laser Surgery and Georgetown University Medical Center,

Washington, District of Columbia.

Background: Cutaneous scars can cause functional impairment, discomfort,and significant aesthetic disfigurement.The Problem: Choice of laser modality depends on scar characteristics, loca-tion, and previous treatments. Hypertrophic scars and keloids respond best to585-nm pulsed-dye laser irradiation, whereas atrophic scars (e.g., acne orsurgical scars, striae distensae) require laser skin resurfacing. Deep atrophicscars, scars on nonfacial sites, and those in darker skin are traditionally moredifficult to treat.Basic/Clinical Science Advances: The 585-nm pulsed-dye laser selectivelytargets cutaneous blood vessels, modulates transforming growth factor-b1expression, and induces regression in hypertrophic scars and keloids. Laserskin resurfacing is useful for blending scar texture with the surrounding skin.Ablative and nonablative lasers have essentially been replaced by newer,fractional laser devices. This latter technology delivers energy to tissue inmicroscopic columns in a grid pattern, leaving intervening islands of un-treated skin to promote rapid healing. Shorter recovery times have improvedthe safety profile of laser skin resurfacing.Relevance to Clinical Care: Optimized treatment protocols and fractional lasertechnology have led to improved safety and efficacy for the treatment of scarsin sensitive or delicate body locations and in patients with darker skin types.Conclusion: Advances in laser technology have improved the prognosis forvirtually all types of cutaneous scars.

BACKGROUND

SCARRING IS A COMMON SEQUELA of cuta-neous injury that can cause obviousphysical disfigurement, functional im-pairment, pain, and dysesthesia. Pa-tients are usually most impacted by thepsychological burden and stigma as-sociated with the appearance oftheir scars. Although it is not possibleto remove a scar completely with anytherapy, advances in laser technologyhave improved treatment outcomes forall types of scarring.

The earliest attempts to treatscars with continuous-wave argon,neodymium:yttrium–aluminum–gar-net, and carbon dioxide (CO2) laserdevices resulted in a high risk of tis-

sue necrosis and scar recurrence. Inthe 1980s, pulsed laser systems weredeveloped for selective absorption oflaser light by target chromophores(e.g., hemoglobin, melanin, water)in the skin to induce temperature-controlled, target-specific injury with-out damage to surrounding healthytissue.1 By the early 1990s, the firstseries of clinical studies showed sus-tained improvements in erythema-tous, hypertrophic scars and keloidsusing the pulsed-dye laser (PDL), arefined vascular laser system thatis widely used today.2–6 Treatment ofatrophic scars has evolved from tra-ditional pulsed CO2 and erbium lasersystems to safer ablative and non-

Jennifer L. MacGregor

*Correspondence: Washington Institute of

Dermatologic Laser Surgery 1430 K St., NW Suite

200, Washington, District of Columbia 20005

(e-mail: [email protected]).


CO2¼ carbon dioxide

PDL¼ pulsed-dye laser

TGF-b1¼ transforming growthfactor-b1


____�____0

____þ

review-article

AWC-2010-0269-MacGregor_3P.3d 01/27/11 10:06pm Page 185

ablative fractional technologies. These lasers limitdelivery of energy in microscopic columns within is-lands of nontreated skin, thereby maximizing clini-cal efficacy and improving the recovery time andside-effect profile of laser skin resurfacing.7–10


Characterization of the scar type and location areessential to selection of the appropriate treatmentmodality and protocol (see Summary Illustrationsection).

Hypertrophic scars and keloids (Fig. 1A, B)present as red, thickened, firm scars that occurwhen collagen deposition is exuberant and exceedscollagen breakdown during the remodeling phase ofwound healing. Hypertrophic scars remain confinedto the area of original injury and may slowly im-prove over time, whereas keloids continue to growbeyond the site of the original injury and tend tocause obvious disfigurement. Histologically, keloidscan be distinguished from hypertrophic scars by aunique pattern of thickened, hyalinized collagenarranged in whorls, which is attributed to an in-

herited alteration in fibroblast response to stimuliand continued production of excessive collagen.11

Atrophic scars (Fig. 2A, B) appear thinner thanthe surrounding skin because of inadequate colla-gen replacement in the scarred areas. Atrophicscars typically follow inflammatory skin diseasessuch as acne or varicella and present as indenta-tions in the skin, wrinkled areas resembling ‘‘ciga-rette paper,’’ or bulges where the subcutaneous fatherniates through the thinned dermis. Acne scarsare commonly atrophic and are best classified ac-cording to the system described by Jacob et al.12 intothree types: rolling, boxcar, and icepick. Rollingscars are wide undulating depressions with slopingborders where the depression is often associatedwith fibrous tethering in the depth. Boxcar scarshave sharply marginated borders with a ‘‘step-off,’’and icepick scars are 1–2-mm-diameter, narrow,tapered pits. Deep boxcar and icepick scars arelikely to require surgical revision in addition to laserresurfacing.

Striae distensae (Fig. 3A, B) are linear, atro-phic pink, red, or white streaks overlying stretchedskin, located most commonly on the hips, buttocks,abdomen, or breasts in women.

Early wounds in scar-prone skin appear pinkwith mild textural change. These prescars (Fig. 4A,B) have the potential to form unfavorable scars andshould be considered for intervention with earlylaser therapy.8,13–15


Modern lasers are based on the theory of selectivephotothermolysis, which describes the use ofspecific laser wavelengths to effect a controlled,temperature-mediated change in a specific cutaneous

TARGET ARTICLES

1. Alster TS and Zaulyanov-Scanlon L: Laserscar revision: a review. Dermatol Surg 2007; 33:131.

2. Sobanko JF and Alster TS: Laser treat-ment for scars and wounds. G Ital DermatolVenereol 2009; 144: 583.

3. Tanzi EL and Alster TS: Skin resurfacing:ablative lasers, chemical peels, and dermabra-sion. In: Fitzpatrick’s Dermatology in GeneralMedicine (8th Edition), 2010.

Figure 1. Keloid before (A) and after pulsed-dye laser treatment (B).


1____

0 ____

1____


target.1 The three target chromophores in skin in-clude hemoglobin, melanin, and water. The thermalenergy applied to the cutaneous chromophore is lim-ited so that the target is heated=destroyed, buttransfer of energy to surrounding tissue is. Para-meters can be calculated based on the size=shape and

absorption characteristics of each target. In 2004, theconcept of fractional delivery in a novel beam patternwas described.7 The regular, pixilated pattern of thelaser beam is delivered in a grid where the energy islimited to regularly spaced microthermal zones orphotocoagulated columns within areas of untreated

Figure 3. Striae distensae on the thigh before (A) and after nonablative fractional 1550-nm laser resurfacing (B). Note subtle improvement in skin colorand texture.

Figure 2. Atrophic acne scars before (A) and after fractional carbon dioxide laser skin resurfacing (B).

MANAGEMENT OF SCARS WITH LASERS 187

____�____0

____þ


skin. Fractional ablative systems vaporize tissue inthe treated microthermal zones, whereas fractionalnonablative resurfacing systems deliver enough heatto denature collagen and cause cell necrosis, but thetreated tissue remains intact.16

EXPERIMENTAL MODEL OF MATERIAL—ADVANTAGES AND LIMITATIONS

Although lasers have been used for decades, mo-lecular effects following laser irradiation of scartissue are only just beginning to be understood.Recent work by Kuo et al. provides evidence that the585-nm PDL alters signaling pathways involvedwith keloid formation by suppressing activatorprotein-1 transcription and transforming growthfactor-b1 (TGF-b1) expression via the mitogen-activated protein kinase pathway, thus reducingproliferation of fibroblasts and inducing keloid re-gression via fibroblast apoptosis during the re-modeling phase of wound healing.17–19 PDLirradiation also selectively reduces scar vasculatureand increases mast cells and mediators within thescar, but it is not clear if these effects have any rolein scar improvement following treatment.2,4 Otherstudies have demonstrated laser- and light-inducedmodulation of TGF-b. Arany et al. demonstratedthat low-power 904-nm laser irradiation activateslatent TGF-b1 in early wounds in vivo.20 Infraredlight irradiation also upregulates TGF-b1 in cul-tured human fibroblasts and may speed woundhealing following multiple exposures.21 It is unclearwhat role these effects will play in the clinicalmanagement of scars with lasers.

The action of skin resurfacing lasers variesaccording to the device, wavelength, pulse energy,and density, with the overall goal being reinitiation ofthe wound healing response in a controlled fashion toinduce collagen remodeling. In experimental models

of normal skin, depth of penetration and width ofmicrothermal zones increases in a fairly predictablepattern with increasing energy.16 It is not clear howthese models correlate with scar tissue.


Compared with older destructive methods (e.g.,dermabrasion, deep chemical peels), modern lasersystems are more precise and controlled and rep-resent a superior method of initiating the woundremodeling process to improve scars. Managementvaries considerably depending upon scar type,location, and patient characteristics.

Management of scars with lasers

Hypertrophic scars=keloids. The vascularPDL remains the first-line laser for treatment ofhypertrophic scars and keloids. Two decades of studydemonstrates remarkable improvements in scarvascularity, color, height, pliability, texture, andsymptomatology after one to several treatmentsusing low energy densities and short pulse dura-tions.2–6 Studies have confirmed the 585 nm wave-length as the superior wavelength for thisapplication.22 Following PDL irradiation, there isselective heating of cutaneous blood vessels. Shortpulses confine the heating to the intended target,2–6

but no clinical differences have been demonstratedbetween 0.45 and 1.5 ms pulse durations.23 Treat-ment parameters used should be adjusted accord-ing to the scar type, location, and previoustreatment response. Despite abundant literatureavailable on the subject, determining the appro-priate settings comes with experience. In general,fluences are decreased for patients with darkerskin phototypes and in sensitive scar-prone areassuch as the neck and chest. Immediate mild-to-

Figure 4. Prescar at 1 month after trauma to upper lip (A) and after pulsed-dye laser therapy (B).

1____

0 ____

1____



moderate purpura is typically produced, along withtransient erythema and edema. After treatment,the skin is cooled and protected from the sun. Pur-pura resolves in 7–10 days. The patient can bereevaluated and treated at 6–8-week intervals de-pending upon response. Adjunctive use of intra-lesional corticosteroids, 5-fluorouracil, or surgicaldebulking should be considered for extremely thick(>1 mm), nodular, or rapidly proliferative scars. Useof adjunctive therapy should be considered early forthese large=aggressive scars because laser energywill not penetrate the lesion. Facial acne scars withand without hypertrophic and erythematous com-ponents should be considered for laser resurfacingcombined with 585-nm PDL irradiation.

Prescars. The 585-nm PDL can also be used totreat early scars (prescars) within the first fewweeks after wounding. Studies confirm that scarstreated early will ultimately heal more favorablythan if left untreated8,13–15 and new fractional la-ser systems may also be successful at preventingunfavorable scar formation in surgical patients.

Atrophic scars and striae. Earlypink=red striae respond to low-energy,short-pulse PDL treatment, with clinicalimprovements in skin color and texture.8

Older white striae can be treated withfractional laser technology, but furtherstudy is warranted to validate this ap-proach.

The goal of treatment for atrophic sur-gical and acne scars is to soften the scarborders and blend the texture with thesurrounding skin. Modern laser systemsare superior to older modalities (e.g.,dermabrasion, deep chemical peels) be-cause they allow precise, controlled heatingor vaporization of tissue to stimulate neo-collagenesis, scar remodeling, and skintightening. The 10,600-nm CO2 and 2,940-nm erbium-doped yttrium–aluminum–garnet lasers are the ‘‘gold standard’’ablative resurfacing lasers. They have well-proven clinical efficacy in recontouring fa-cial atrophic scars,24,25 but are associatedwith prolonged recovery and risk of com-plications during healing. Nonablative in-frared devices were developed to thermallyalter dermal collagen and induce remodel-ing while preserving epidermal integrity.After a series of treatments, the 1,064- and1,320-nm neodymium:yttrium–aluminum–garnet and 1,450-nm diode lasers have

produced mild improvement in scars, with virtuallyno postoperative recovery.26

Fractional skin resurfacing has changed theprognosis for virtually all types of scars, and thesedevices have essentially replaced earlier ablativeand nonablative lasers. The fractionated approachallows for rapid recovery as the microscopic woundsheal quickly from surrounding normal skin. Clin-ical efficacy approaches that of ablative techniques,particularly when repeat treatments are applied.

Nonablative fractional lasers (1,440-, 1,540-, and1,550-nm devices) are delivered in an outpatientsetting with topical anesthesia. Posttreatment re-covery involves 2–3 days of sunburn-like redness andedema. Despite short recovery times, collagen re-modeling and scar improvement continues for 6months or longer following a series of three or moremonthly treatments. Clinical studies show that themajority of patients achieve significant improvementin atrophic acne scars after a treatment series.27

Similar success has been achieved in hypopigmentedand surgical scars on the face and body.

Fractional ablative lasers (10,600-, 2,940-, and2,790-nm devices) have most recently been used for

TAKE HOME MESSAGE


� Lasers selectively alter dermal collagen to induce collagen remodeling inunfavorable scar tissue.

� Fractional delivery of laser energy in a grid pattern allows for faster, safer(and potentially more effective) treatment of scars. This concept repre-sents a major advance in laser technology over the last 2 decades.


� The 585-nm PDL remains the preferred laser treatment for hypertrophicscars, keloids, new surgical scars, and erythematous striae distensae.PDL treatment protocols have been refined to optimize efficacy, safety,and predictability.

� Traditional ablative and nonablative resurfacing lasers have been es-sentially replaced by newer, fractional devices.

� Fractional laser skin resurfacing is a safe and effective treatment foratrophic and hypopigmented surgical and acne scars in facial and non-facial sites. The 1,550-nm and 10,600-nm devices are currently the mostwidely used for this application.


� The prognosis for all types of scars is changing rapidly with the devel-opment of new fractional laser technology.

� Successful management of scars with laser depends upon proper patientselection, realistic expectations, and appropriate pre- and postoperativemanagement.

� Operator experience with selecting parameters and familiarity withavailable evidence-based literature is essential to successful outcomesand prevention of complications.

____�____0

____þ



scar resurfacing. Emerging studies demonstrate ef-ficacy for various types of acne, surgical, and burnscars.27,28 The ablative wavelengths vaporize micro-scopic columns of tissue and induce surrounding ar-eas of collagen denaturation. For more fibroticmoderate-to-deep boxcar, atrophic, and icepick scars,these ablative devices have the potential to effect agreater change than nonablative resurfacing be-cause of their superior ability to tighten the skin andstimulate collagen remodeling. They do, however,have a greater risk of complications.29 Anesthesiarequirements vary according to treatment area—forisolated scars, local anesthesia is sufficient, whereasa full facial treatment for acne scars would requirenerve blockade and systemic agents for pain andanxiety. Recovery times involve 5–7 days of ser-osanguinous discharge during the reepithelializationprocess, followed by a week or more of erythema.

INNOVATION

Fractional technology can be applied to virtuallyany laser or optical device. The operator canmanipulate the energy density (or % coverage),width of the microscopic beams, and energy=depthof delivery. This technologic advance enhancesthe safety of many skin resurfacing procedures,allowing treatment of scars not only on facialskin, but also on nonfacial sites and in patients withdarker skin types. With repeat treatments, even atnonablative wavelengths, the clinical efficacy is si-milar to that of traditional ablative resurfacing.


It is important to note that many of the laserdevices discussed in this article can themselves

cause dyspigmentation and scarring. Appropriatepatient selection and operator experience are crucialvariables that ultimately determine treatment ef-ficacy, patient satisfaction, and overall outcome.The risk of complications with fractional resurfa-cing is lower than with traditional ablative modal-ities, but serious complications have beenreported.29 Hypertrophic scars, ectropion formation,dyspigmentation, and viral or bacterial infectionare rare; however, side effects such as acne, contactdermatitis, and prolonged erythema are occasion-ally observed. As many laser technologies arerelatively new, delayed reactions may also bediscovered.

Prior to treatment, a complete discussion of theserisks must be weighed against the patient’s goalsand expectations. The consultation must include adocumented discussion of the patient’s prognosis,projected number of treatments, and understand-ing of medical=social factors that may impact theirability to tolerate the procedures and associatedrecovery. The provider should pay specific attentionto skin phototype, concurrent infection or inflam-matory skin disorders, medication use, and otherconditions that would impact healing and present acontraindication to treatment (such as previousradiation). The provider must also be familiar with,and able to communicate, evidence-based literatureso that patients understand and accept circum-stances where new devices are being used withoutstudies of long-term follow-up.

FUTURE DEVELOPMENTS

Studies are needed to evaluate the safety andefficacy of new laser technologies for the treat-ment of difficult-to-treat scars such as scars on

Scar

Prescar

Hypertrophic scars and keloids

Pulsed-dye laser Fractional nonablative resurfacing

Pulsed-dye laser +/ Fractional nonablative resurfacing +/ Fractional ablative resurfacing

Atrophic scars

Boxcar

Icepick

Rolling

Striae distensae

Shallow/Moderate: Fractional nonablative or ablative resurfacing

Deep: Surgical revision, fillers, laser resurfacing

Subcision, fillers, fractional ablative resurfacing

Surgical revision followed by fractional ablative resurfacing

Red: Pulsed-dye laser White: Fractional nonablative resurfacing

1____

0 ____

1____



nonfacial sites, burn scars, striae distensae, andthose in darker skin types. Unlike the face, skinon the neck, trunk, and extremities has rela-tively few adnexal structures and less capacityto replace the epidermis following resurfacingprocedures. Fractional technology reduces thepostoperative risk of prolonged healing, infection,scarring, and pigmentary alterations. Appro-priate treatment protocols and safety assess-ment for these new applications have yet to becharacterized.


Relating scar type to preferred laser treatmentsmodality:

ACKNOWLEDGMENT

The authors have not received funding for this work.


The authors have no disclosures relevant to thiswork.

REFERENCES

1. Anderson RR and Parrish JA: Selective photo-thermolysis: precise microsurgery by selective ab-sorption of pulsed radiation. Science 1983; 220: 524.

2. Alster TS: Improvement of erythematous and hy-pertrophic scars by the 585 nm flashlamp-pumpedpulsed dye laser. Ann Plast Surg 1994; 32: 186.

3. Dierickx C, Goldman MP, and Fitzpatrick RE: Lasertreatment of erythematous=hypertrophic and pig-mented scars in 26 patients. Plast Reconstr Surg1995; 95: 84.

4. Alster TS and Williams CM: Treatment of keloidsternotomy scars with the 585 nm flashlamp-pumped pulsed dye laser. Lancet 1995; 345: 1198.

5. Alster TS: Laser scar revision: comparison study of585 nm pulsed dye laser with and without intrale-sional corticosteroids. Dermatol Surg 2003; 29: 25.

6. Alster TS and Nanni CA: Pulsed dye laser treat-ment of hypertrophic burn scars. Plast ReconstrSurg 1998; 102: 2190.

7. Manstein D, Herron GS, Sink RK, Tanner H, andAnderson RR: Fractional photothermolysis: a newconcept for cutaneous remodeling using micro-scopic patterns of thermal injury. Lasers SurgMed 2004; 34: 426.

8. Alster TS and Zaulyanov-Scanlon L: Laser scarrevision: a review. Dermatol Surg 2007; 33: 131.

9. Sobanko JF and Alster TS: Laser treatment forscars and wounds. G Ital Dermatol Venereol 2009;144: 583.

10. Tanzi EL and Alster TS: Skin resurfacing: ablativelasers, chemical peels, and dermabrasion. In: Fitzpa-trick’s Dermatology in General Medicine (8th Edition);New york: Mcgraw Hill Professional 2010.

11. Wolfram D, Tzankov A, and Pulzl P: Hypertrophicscars and keloids—review of their pathophysi-ology, risk factors, and therapeutic management.Dermatol Surg 2009; 35: 171.

12. Jacob CI, Dover JS, and Kaminer MS: Acnescarring: a classification system and review of

treatment options. J Am Acad Dermatol 2001;45: 109.

13. Bowes LE and Alster TS: Treatment of facial scarringand ulceration resulting from acne excoriee with585-nm pulsed dye laser irradiation and cognitivepsychotherapy. Dermatol Surg 2004; 30: 934.

14. Nouri K, Jimenez GP, Harrison-Balestra C, andElgart GW: 585-nm pulsed dye laser in thetreatment of surgical scars starting on the sutureremoval day. Dermatol Surg 2003; 29: 65.

15. Alam M, Pon K, Van Laborde S, Kaminer MS,Arndt KA, and Dover JS: Clinical effect of a singlepulsed dye laser treatment of fresh surgical scars:randomized controlled trial. Dermatol Surg 2006;32: 21.

16. Bedi VP, Chan KF, Sink RK, Hantash BM, HerronGS, Rahman Z, Struck SK, and Zachary CB: Theeffects of pulse energy variations on the dimen-sions of microscopic thermal treatment zones innonablative fractional resurfacing. Lasers SurgMed 2007; 39: 145.

17. Kuo YR, Wu WS, and Wang FS: Flashlamp pulsed-dye laser suppressed TGF-beta1 expression andproliferation in cultured keloid fibroblasts is me-diated by MAPK pathway. Lasers Surg Med 2007;39: 358.

18. Kuo YR, Wu WS, Jeng SF, et al.: Activation of ERKand p38 kinase mediated keloid fibroblast apo-ptosis after flashlamp pulsed-dye laser treatment.Lasers Surg Med 2005; 36: 31.

19. Kuo YR, Wu WS, Jeng SF, et al.: Suppressed TGF-beta1 expression is correlated with up-regulationof matrix metalloproteinase-13 in keloid regres-sion after flashlamp pulsed-dye laser treatment.Lasers Surg Med 2005; 36: 38.

20. Arany PR, Nayak RS, Hallikerimath S, et al.: Ac-tivation of latent TGF-b1 by low-power laser invitro correlates with increased TGF-b1 levels inlaser-enhanced oral wound healing. Wound Re-pair Regen 2007; 15: 866.

21. Danno K, Mori N, Toda K, et al.: Near-infraredirradiation stimulates cutaneous wound repair:laboratory experiments on possible mechanisms.Photodermatol Photoimmunol Photomed 2001;17: 261.

22. Nouri K, Rivas MP, Stevens M, Ballard CJ, SingerL, Ma F, Vejjabhinanta V, Elsaie ML, and ElgartGW: Comparison of the effectiveness of thepulsed dye laser 585-nm versus 595-nm in thetreatment of new surgical scars. Lasers Med Sci2009; 24: 801.

23. Nouri K, Elsaie ML, Vejjabhinanta V, Stevens M,Patel SS, Caperton C, and Elgart GW: Comparisonof the effects of short- and long-pulse durationswhen using a 585-nm pulsed dye laser in thetreatment of new surgical scars. Lasers Med Sci2010; 25: 121.

24. Alster TS and Tanzi EL: Laser skin resurfacing:ablative and nonablative. In: Surgery of the Skin,edited by Robinson J, Sengelman R, Siegel DM,and Hanke CM. Philadelphia: Elsevier, 2005, pp.611–624.

25. Alster TS: Cutaneous resurfacing with CO2 anderbium: YAG laser: preoperative, intraoperative,and postoperative considerations. Plast ReconstrSurg 1999; 103: 619.

26. Tanzi EL and Alster TS: Comparison of a 1450-nmdiode laser and a 1320-nm Nd:YAG laser in thetreatment of atrophic facial scars: a prospectiveclinical and histologic study. Dermatol Surg 2004;30(2 pt 1): 152.

27. Alster TS, Tanzi EL, and Lazarus M: The use offractional photothermolysis for the treatment ofatrophic scars. Dermatol Surg 2007; 33: 295.

28. Tanzi EL, Alster TS, and Wanitphakdeecha R:Fraxel laser indications and long-term follow-up.Aesthet Surg J 2008; 28: 675.

29. Metelitsa AI and Alster TS: Fractionated laser skinresurfacing treatment complications: a review.Dermatol Surg 2010; 36: 299.

____�____0

____þ



REVIEW DIGEST In Vitro Reconstructed Human Skin Equivalents for ...

Documents

Transcript of REVIEW DIGEST In Vitro Reconstructed Human Skin Equivalents for ...