Inhibition of Lipid Peroxidation Restores Impaired Vascular … · 2001-02-09 · Inhibition of Lipid Peroxidation Restores Impaired Vascular Endothelial Growth Factor Expression

Inhibition of Lipid Peroxidation Restores ImpairedVascular Endothelial Growth Factor Expression andStimulates Wound Healing and Angiogenesis in theGenetically Diabetic MouseDomenica Altavilla,

1Antonino Saitta,

3Domenico Cucinotta,

3Mariarosaria Galeano,

2

Barbara Deodato,1

Michele Colonna,2

Valerio Torre,4

Giuseppina Russo,3

Aurora Sardella,1

Giuseppe Urna,1

Giuseppe M. Campo,5

Vittorio Cavallari,4

Giovanni Squadrito,3

and Francesco Squadrito1

Impaired wound healing is a well-documented phenom-enon in experimental and clinical diabetes. Experimen-tal evidence suggests that a defect in vascular endo-thelial growth factor (VEGF) regulation might be asso-ciated with wound-healing disorders. We studied theinvolvement of lipid peroxidation in the pathogenesis ofaltered VEGF expression in diabetes-related healingdeficit by using an incisional skin-wound model pro-duced on the back of female diabetic C57BL/KsJ db1/db1 mice and their normal (db1/1m) littermates. Ani-mals were then randomized to the following treatment:raxofelast (15 mg z kg–1 z day–1 i.p.), an inhibitor of lipidperoxidation, or its vehicle (DMSO/NaCl 0.9%, 1:1 vol:vol). The animals were killed on different days (3, 6, and12 days after skin injury), and the wounded skin tissueswere used for histological evaluation, for analysis ofconjugated dienes (CDs), as an index of lipid perox-idation and wound breaking strength. Furthermore,we studied the time course of VEGF mRNA expressionthroughout the skin-repair process (3, 6, and 12 daysafter skin injury), by means of reverse transcriptase–polymerase chain reaction, as well as the mature pro-tein in the wounds. Diabetic mice showed impairedwound healing with delayed angiogenesis, low break-ing strength, and increased wound CD content whencompared with their normal littermates. In healthycontrol mice, a strong induction of VEGF mRNA wasfound between day 3 and day 6 after injury, while nosignificant VEGF mRNA expression was observed at day12 after injury. In contrast, VEGF mRNA levels, after aninitial increase (day 3), were significantly lower in dia-betic mice than in normal littermates, and light induc-tion of VEGF mRNA expression was also present at day12 after injury. Similarly, the wound content of theangiogenic factor was markedly changed in diabeticmice. Administration of raxofelast did not modify the

process of wound repair in normal mice, but signifi-cantly improved the impaired wound healing in diabeticmice through the stimulation of angiogenesis, re-epithe-lization, and synthesis and maturation of extracellularmatrix. Moreover, raxofelast treatment significantly re-duced wound CD levels and increased the breakingstrength of the wound. Lastly, the inhibition of lipidperoxidation restored the defect in VEGF expressionduring the process of skin repair in diabetic mice andnormalized the VEGF wound content. The current studyprovides evidence that lipid peroxidation inhibition re-stores wound healing to nearly normal levels in exper-imental diabetes-impaired wounds and normalizes thedefect in VEGF regulation associated with diabetes-in-duced skin-repair disorders. Diabetes 50:667–674, 2001

Wound healing is a complex programmed se-quence of cellular and molecular processes,including inflammation, cell migration, angio-genesis, provisional matrix synthesis, colla-

gen deposition, and re-epithlization (1). The healingprocess requires a sophisticated interaction among inflam-matory cells, biochemical mediators, extracellular matrixmolecules, and microenvironmental cell population. All ofthese events are stimulated by a number of mitogens andchemotactic factors.

Vascular endothelial growth factor (VEGF) is one of themost potent known angiogenic cytokines and promotes allsteps in the cascade process of angiogenesis. In partic-ular, it induces degeneration of the extracellular matrixof existing vessels by proteases, causes migration andproliferation of capillary endothelial cells, and determinestube proliferation of endothelial cells (2). VEGF action isassociated with a variety of physiological and pathologicalneovascular events, such as embryonic development, tu-mor growth, and wound repair in particular (3). VEGFis related to platelet-derived growth factor and hasfour different isoforms, VEGF121, VEGF165, VEGF189, andVEGF206, which are generated by alternative splicing ofmRNA (4). VEGF is produced by keratinocytes that, to-gether with macrophages, represent the most importantsource of this growth factor during normal wound healing.

Impaired wound healing may be a consequence of

From the 1Institute of Pharmacology, 2Department of Plastic Surgery, 3De-partment of Internal Medicine, 4Institute of Ultrastructural Pathology, and5Institute of Physiology, University of Messina, Messina, Italy.

Address correspondence and reprint requests to Francesco Squadrito, MD,Institute of Pharmacology, School of Medicine, University of Messina Poli-clinico Universitario, Torre Biologica 5° Piano, Via C. Valeria Gazzi, 98100Messina, Italy. E-mail: [email protected].

Received for publication 30 May 2000 and accepted in revised form 7December 2000.

CD, conjugated diene; ELISA, enzyme-linked immunosorbent assay; PCR,polymerase chain reaction; VEGF, vascular endothelial growth factor.

DIABETES, VOL. 50, MARCH 2001 667

normal aging, metabolic derangement such as diabetes, ortherapeutic intervention.

Genetically diabetic mice (db/db mice) are useful as ananimal model for wound-healing studies, since woundhealing in these animals is markedly delayed when com-pared with nondiabetic littermates (5,6). Healing impair-ment is characterized by delayed cellular infiltration andgranulation tissue formation, reduced angiogenesis, de-creased collagen, and its organization (7–10). The mecha-nism of this alteration is thought to result from diabetesproduction of reactive free radicals that cause lipid per-oxidation, which in turn impairs keratinocyte endothelialcells, fibroblasts, and collagen metabolism (11). Further-more, the presence of a defect in VEGF regulation, char-acterized by an altered expression pattern of VEGF mRNAduring skin repair in db/db mice, has been shown andthereby suggests that an impairment in VEGF regulationmight be associated with wound-healing abnormalitiesseen in these animals (12). Therefore, the aim of ourexperiment was to investigate whether there is a link be-tween altered VEGF regulation and increased lipid peroxi-dation in experimental diabetes-induced skin-repair abnor-malities.

RESEARCH DESIGN AND METHODS

Animals and experimental protocol. All animal procedures were in accor-dance with the Declaration of Helsinki and the Guide for the Care and Use of

Laboratory Animals.Genetically diabetic female C57BL/KsJ db1/db1 mice and their controls

(db1/1m) were obtained from the Jackson Laboratories (Bar Harbor, ME).The animals were 14 weeks old at the start of the experiments. They wereobese, weighing 40–50 g, compared with their nondiabetic littermates, whichweighed 25–32 g. The diabetic mice were markedly hyperglycemic withaverage glucose levels of 527 6 25 mg/dl compared with 205 6 9 mg/dl for thenondiabetic animals. The hyperglycemia produced classic signs of diabetes,including polydipsia, polyuria, and glycosuria.

During the experiments, the animals were housed one per cage, maintainedunder controlled environmental conditions (12-h light/dark cycle, temperature;23°C), and provided with standard laboratory food and water ad libitum. Theanimals were divided into four groups (21 animals each). The first and secondgroups, consisting respectively of diabetic and healthy control mice, weregiven raxofelast, an inhibitor of lipid peroxidation (13), at a dose of 15 mg/kgi.p. for 12 days. The third group of diabetic mice and the fourth group ofhealthy control mice were treated with vehicle (DMSO/NaCl 0.9%, 1:1 vol:volfor 12 days).

After general anesthesia with ketamine hydrochloride (110 mg/kg), hair onthe back was shaved and skin was washed with povidone-iodine solution andwiped with sterile water. Two full-thickness longitudinal incisions (4 cm)were made on the dorsum of the mice, and the wound edges were closed with

skin clips placed at 1-cm intervals. Seven animals for each group were killedafter 3, 6, and 12 days, respectively, and the wounds were divided into threesegments (0.8 cm wide). The caudal and cranial strips were used for histology,while the central one was used for biochemical and molecular analysis andwound breaking strength measurements (only day 12).Histological evaluation. The samples were fixed in 10% buffered formalinfor light microscopic examination. After fixation, sections perpendicular tothe anterior-posterior axis of the wound were dehydrated with gradedethanols and embedded in paraffin. Sections 5-mm thick of paraffin-embeddedtissues were mounted on glass slides, rehydrated with distilled water, andstained with hematoxylin and eosin. As part of the histological evaluation, allslides were examined by a pathologist without knowledge of the previoustreatment, using masked slides under the microscope at 320 to 3100magnification. The parameters measured were epidermal and dermal regen-eration, granulation tissue thickness, and angiogenesis. The margins of thewound in each of the sections, as well as normal control wounds, were usedas comparisons for scoring (Table 1). Concerning angiogenesis, only maturevessels that contained erythrocytes were counted. To evaluate well-formedfrom poorly formed capillary vessels, the following parameters were consid-ered: presence or absence of edema, congestion, hemorrhage, thrombosis,and intravascular or intervascular fibrin formation.Breaking strength. The maximum load (breaking strength) tolerated bywounds was measured blindly on coded samples using a calibrated tensom-eter (Instron, Canton, MA) as previously described (14). The ends of the skinstrip were pulled at a constant speed (20 cm/min), and breaking strength wasexpressed as the mean maximum level of tensile strength (grams permillimeter) before separation of wounds.Conjugated dienes evaluation. Estimation of the tissue content of conju-gated dienes (CDs) was carried out to evaluate the extent of lipid peroxidationin wounds. Samples (0.2 mg tissue) were collected in polyethylene tubes andthen washed with 1 ml butylated hydroxytoluene (BHT) (1 mg/ml in phos-phate buffer).

The samples, after drying in absorbant paper, were frozen at 4°C until theanalysis. The biochemical assay of CDs required previous lipid extractionfrom the tissue samples by chloroform/methanol (2:1). The lipid layer wasdried under nitrogen atmosphere and then dissolved in ciclohexane. Woundcontents of CDs were measured at 232 nm by using a spectrophotometrictechnique. The amount of wound CDs was expressed as D ABS per milligram.VEGF expression. Total cellular RNA was extracted from incisional fullthickness wounds at different intervals after wounding. In brief, ;50 mg tissuewas homogenized with 800 ml RNAZOL STAT (Teltest, Friendswood, TX) in amicrofuge tube, after which 80 ml chloroform was added. After vortexing andcentrifugation, the aqueous phase was transferred to a new microfuge tubecontaining an equal volume of cold isopropanol, and the RNA was recoveredby precipitation by chilling at –80°C for 15 min. The pellet was washed withcold ethanol (70%), centrifuged, dried in a speed vacuum, centrifuged a secondtime, and then dissolved in 20 ml buffer. A 2-mg portion of total RNA wassubjected to first-strand cDNA synthesis in a 20-ml reaction mixture containingthe AMV, reverse transcriptase (Superscript II; BRL), each dNTP, the specificprimers, Tris-HCl, and MgCl2.

After dilution of the product with distilled water, 5 ml was used for eachpolymerase chain reaction (PCR), which contained the Taq polymerase(Perkin Elmer), the buffer as supplied with the enzyme, each dNTP, and thespecific primers designed to cross introns and to avoid confusion between

TABLE 1Histological scores of wounds

ScoresEpidermal and dermal

regenerationGranulation tissue

thickness Angiogenesis

16 Little epidermal anddermal organization.

Thin granular layer. Altered angiogenesis (1–2 vessels per site) characterized bya high degree of edema, hemorrhage, and occasional con-gestion and thrombosis.

26 Moderate epidermal anddermal organization.

Moderate granulation layer. Few newly formed capillary vessels (3–6 per site); moderatedegree of edema and hemorrhage. Occasional congestionand intervascular fibrin deposition; absence of thrombosis.

36 Complete remodeling ofepidermis and dermis.

Thick granulation layer. Newly formed capillary vessels (7–10 per site); moderatedegree of perivascular and interstitial edema and conges-tion. Absence of thrombosis and hemorrhage.

46 Very thick granulation layer. Newly formed and well-structured capillary vessels (.10 persite) vertically disposed toward the epithelium and at thewound margins. Slight degree of perivascular edema.

INHIBITION OF LIPID PEROXIDATION STIMULATES WOUND HEALING

668 DIABETES, VOL. 50, MARCH 2001

mRNA expression and genomic contamination. The following primers wereused: mouse VEGF121 and mouse b-actin. The optimal cycle number for mouseVEGF121 was 25, and we used a PCR-negative and a PCR-positive controlwithout cDNA or with a known cDNA, respectively. A portion of the PCRproduct was electrophoresed and transferred to a nylon membrane, whichwas prehybridized with oligonucleotide probes and radiolabeled with[32P]ATP by a T4 oligonucleotide kinase. After hybridization, filters underwentautoradiography in a darkroom with a fixed camera. The captured image,which was sent for image analysis (Bio-Profil software; Celbio, Milan, Italy),was subjected to a densitometric analysis.Determination of VEGF in wounds. The amount of VEGF in wounds wasdetermined by an enzyme-linked immunosorbent assay (ELISA). Briefly,tissues were homogenized in 1.0 ml of 13 phosphate-buffered saline contain-ing complete protease inhibitor cocktail (Boehringer Mannhein, Indianapolis,IN). Homogenates were centrifuged to remove debris and were filteredthrough a 1.2-mm pore syringe filter. Analysis was performed with a commer-cially available murine VEGF-specific ELISA kit (R&D Systems). The amountof VEGF was expressed as picograms per wound.Drug. Raxofelast was supplied by Biomedica Foscama Research Center,Ferentino, Italy. The compound was administered intraperitoneally in DMSO/NaCl 0.9% (1:1 vol:vol). All substances were prepared fresh daily and admin-istered in a volume of 1 ml/kg.Statistical analysis. All data were analyzed by Student’s paired t test. Theresults were expressed as means 6 SE. The level for statistical significancewas set at P , 0.05.

RESULTS

Histological results. Figure 1 shows the histologicalscores of wounds throughout the experiment according tothe criteria in Table 1.

Qualitative data regarding histological evaluation at day12 are summarized in Table 2. In diabetic (db/db) micetreated with vehicle, edema and hemorrhage were prom-inent because of extensive endothelial cell damage withfew incomplete newly formed capillary vessels (Fig. 2A).Most of endothelial cells showed swelling. The endothe-lium of capillaries was prominent, and extravasation oferythrocytes was observed. Histological sections showedlittle dermal and epidermal organization. Significant re-duction of granulation tissue formation and incompletematrix maturation and remodeling characterized by looseconnective tissue in an irregular fashion were noted (Fig.3A). These effects were probably related to vascularalterations and to subsequent prominent edema and hem-orrhage. Scattered intravascular thrombi or interstitialfibrin were also observed. Moreover, generalized vascularcongestion was present in all samples, whereas endothe-lial cells exhibited wide eosinophilic cytoplasm and irregu-larly shaped nuclei with prominent nucleoli.

On the other hand, in diabetic mice treated with raxo-felast, re-epithlialization was moderate to complete withepidermal elongation spreading over two-thirds of thewound surface (Fig. 3B). Dermal regeneration was char-acterized by granulation tissue rich in fibroblasts, gener-ally oriented parallel to the epidermal layer. A moderateamount of collagen fibrils and collagen bundles were orga-nized in a more regular fashion than that seen in the db/db

mice treated with vehicle (Fig. 2B). Newly formed capillaryvessels were observed in moderate numbers in the dermisof the entire wound area. The number of profiles of smallvessels was higher than that in diabetic nontreated mice.The differences between these two groups also consistedof a different structural preservation of the capillary walland a different degree of edema, hemorrhage, and throm-bosis. Scattered intervascular fibrin was sometimes seen.Thin capillary vessels were lined by lightly swollen endo-thelial cells with prominent nuclei and eosinophilic cyto-plasm.

In the nontreated normal mice (db/1), epidermal regen-eration and remodeling of the dermis was complete andalmost similar to that in diabetic treated mice (Fig. 3C).New well-formed capillary vessels were disposed verti-cally toward the wound surface and were identical to

FIG. 1. Time course of histological score in db/1 and db/db micetreated with vehicle (1 ml z kg–1 z day–1 i.p.) and raxofelast (15 mg z kg–1

z day–1 i.p.). Each point represents the mean 6 SE of seven experi-ments. *P < 0.01 vs. db/1 mice; #P < 0.01 vs. db/db mice treated withvehicle.

TABLE 2Main morphological characteristics of wounds

Mice Treated db/db mice Nontreated db/db mice Nontreated and treated db/1 mice

Angiogenesis Well-oriented and well-formedcapillary vessels in the entirewound area.

Few altered capillaries scatteredin the entire wound area.

Vessels disposed vertically towardthe epithelial surface in theedge site of the wound.

Fibroblasts Oval- and spindle-shaped fibroblastsparallel to the surface of thewound.

Stellate or spindle-shapedfibroblasts scattered throughoutthe granulation tissue.

Numerous spindle fibroblastsparallel to the surface of thewound.

Epithelium Almost completely remodeled. Disarranged. Complete remodeling.Dermis Slight edema and well-formed

collagen matrix disposed in aregular fashion.

Few collagen fibrils or collagenbundles disposed in an irregularfashion; moderate edema.

Complete remodeling.

D. ALTAVILLA AND ASSOCIATES

DIABETES, VOL. 50, MARCH 2001 669

those of normal dermis (Fig. 2C). Treatment with raxo-felast did not significantly change the process of woundrepair in normal (db/1) mice.Breaking strength. The wound breaking strengths foreach group at day 12 are depicted in Fig. 4. The breakingstrength of incisional wounds from diabetic mice treatedwith raxofelast was higher than that of diabetic micetreated with vehicle. As a result of raxofelast administra-tion, breaking strength of wounds from db/db mice treatedwith raxofelast was approximately the same as that innondiabetic mice. No significant differences in breakingstrength were observed between nondiabetic mice treatedwith raxofelast or vehicle.Conjugated dienes. CDs were evaluated throughout thestudy. Very low CD levels, which were investigated as anindex of lipid peroxidation, were found in db/1 mice dur-ing the wound-healing process; in addition, the adminis-tration of raxofelast in these animals did not cause anymodification of this parameter (Fig. 5). In contrast, in-creased CD levels were observed in diabetic mice treatedwith vehicle, and raxofelast treatment succeeded in reduc-ing the increased lipid peroxidation (Fig. 5).VEGF expression. The top of Fig. 6 shows representativeautoradiograms highlighting mRNA expression for VEGFin control and diabetic mice treated with vehicle or raxo-felast. The bottom of the figure depicts quantitative dataand indicates the relative amount of VEGF mRNA. Verylow VEGF expression was found in unwounded skin ofseveral groups of mice (data not shown). In the wounds ofhealthy control mice, a strong induction of VEGF mRNAwas found between day 3 and day 6 after injury, whileVEGF expression was not detectable at day 12; these dataare in agreement with the evidence of a completed histo-logical wound repair process in these mice. Administra-tion of the lipid peroxidation inhibitor did not modifythe pattern of VEGF expression in the wounds of healthycontrol mice.

In the wounds of diabetic mice, VEGF mRNA wasmarkedly reduced at day 3 and day 6 after injury. Further-more, mRNA levels for the growth factor were still detect-able at day 12. In accordance with these findings, the his-tological results pointed out a markedly delayed wound-healing process in diabetic mice with altered angiogenesisand insufficient vessel formation. The inhibition of lipidperoxidation restored the pattern of VEGF mRNA expres-sion in the wounds of diabetic mice (Fig. 6).VEGF production in wound. To determine whetherlevels of the VEGF protein were altered in diabetic mice,wound homogenates were assayed for the presence of theangiogenic factor. The amounts of VEGF in uninjured skinfrom both normoglycemic and diabetic animals were lowor undetectable (data not shown). In normal mice, levelsof VEGF were increased at days 3 and 6, declining there-after to baseline by day 12 (Fig. 7). The administration ofraxofelast, an inhibitor of lipid peroxidation, did not changeVEGF protein levels in the wounds of healthy control mice(Fig. 7). At day 3 and day 6, VEGF levels in wounds fromdiabetic mice were substantially diminished compared withnondiabetic mice, and the angiogenic factor was also slightlypresent in the wound of untreated diabetic rats at day 12 (Fig.7). The administration of raxofelast significantly increasedthe levels of VEGF in the wounds from diabetic mice at days

FIG. 2. Hematoxilin- and eosin-stained sections of wound specimens atday 12 from diabetic (db/db) mice treated with vehicle (A), a diabeticmouse treated with raxofelast (B), and a nondiabetic (db/1) non-treated mouse (C). (Original magnification 3100.) A: Vascular changesare characterized by swollen endothelial cells, poorly formed capillarychannels, and evident hemorrhage. B: A good re-epithlialization andwell-formed granulation tissue is shown. Spindle-shaped and ovalfibroblasts are oriented parallel to the epithelial surface. Neovascular-ization is characterized by well-structured capillary vessels andabsence of hemorrhage. C: Well-formed collagen matrix with scat-tered newly formed capillary vessels lined by a single layer ofendothelial cells with round-to-oval nuclei and absence of patholog-ical alterations.

INHIBITION OF LIPID PEROXIDATION STIMULATES WOUND HEALING

670 DIABETES, VOL. 50, MARCH 2001

3 and 6; furthermore, the levels of the angiogenic factor de-clined to baseline by day 12 (Fig. 7) after raxofelast treat-ment.

DISCUSSION

Decreased healing capacity in diabetes is the result ofmultiple factors, including elevated blood glucose levels,suppressed cell-mediated immunity, local ischemia, andfree radical generation. Inadequate oxygenation, such asthat seen in local ischemia, causes production of extreme-ly reactive metabolites (called free oxygen radicals) thatimpair normal wound healing by damaging keratinocyteendothelial cells, capillary permeability, and collagen me-tabolism (15).

Skin ischemia provides favorable conditions for forma-tion of oxygen-derived free radicals by means of leuko-cytes, which are activated during ischemia. The release ofoxygen radicals by adhered activated leukocytes causesadditional damage because more leukocytes are attractedand the process is amplified (16). Under normal condi-tions, the generation of free radicals is counterbalanced bythe presence of adequate endogenous antioxidant defenses(17), but when the generation of free radicals exceeds thecapacity of the defenses, these highly active radicals mayproduce structural changes that may contribute to revers-ible or irreversible cell injury. Oxygen radicals cause tis-sue damage by lipid peroxidation of cellular and organellemembranes, disruption of the intracellular matrix, and al-teration of important protein enzymatic processes (16,18).These agents not only damage the lipids but also producelipid hydroperoxides, secondary intermediates that canlead to a chain reaction of lipid peroxidation (19).

Experimental evidence has demonstrated expression ofVEGF and its receptors during wound healing (20). Highlevels of VEGF mRNA were detected in keratinocytes atthe wound edge and in keratinocytes that migrated tocover the wound surface (21). These findings suggest animportant role of keratinocytes in wound angiogenesis.Because VEGF is highly specific for endothelial cells, it islikely to act in a paracrine manner on the sproutingcapillaries of the wound edge and the granulation tissue.The exclusive detection of VEGF receptors in these cellssupports this hypothesis (22). In our experiment, wecompared the time course of VEGF expression duringwound healing of healthy control mice and geneticallydiabetic db/db mice; the latter are characterized by asignificant delay in the skin-repair process and have beenwidely used as a model for wound-healing disorders. Inhealthy control mice, a marked induction of VEGF expres-sion was observed between day 3 and day 6. Expressionreturned to the basal level after the completion of theskin-repair process (day 12). In the wounds of db/db mice,the mRNA levels for VEGF were severely depressed duringthe first phase of the healing process, and low VEGFmRNA expression was also detectable at day 12. The dataregarding VEGF protein in the wound indicated an over-lapping alteration in diabetic mice. Furthermore, in agree-ment with this result, db/db mice showed an incompleteand altered skin-repair process at day 12.

The reason for the wound-healing defect in db/db mice isstill not completely understood. The present data supportthe hypothesis that an altered pattern of VEGF mRNA ex-

pression might be, at least in part, one of the mechanismsunderlying the diabetes-induced disorder in wound repair.

CD measurement is an indicative method for evaluatinglipid peroxidation (23). The large amount of CDs found in

FIG. 3. Hematoxilin- and eosin-stained sections of wound specimens atday 12 from diabetic (db/db) mice treated with vehicle (A), a diabeticmouse treated with raxofelast (B), and a nondiabetic (db/1) non-treated mouse (C). (Original magnification 325.) A: Evident edemaand poorly formed granulation tissue, with altered epidermal anddermal organization and extravasation of erythrocytes. B: A goodre-epithlialization and well-organized granulation tissue is shown.Connective tissue is highly cellular, mainly composed of fibroblastsorganized in a regular fashion. C: Complete remodeling of epitheliumand connective tissue.

D. ALTAVILLA AND ASSOCIATES

DIABETES, VOL. 50, MARCH 2001 671

the wound tissue of diabetic mice is consistent with theoccurrence of free radical–mediated wound-healing dam-age. Lipid peroxidation is considered responsible for theimpairment of endothelial cells, keratinocyte capillarypermeability, and fibroblast and collagen metabolism.Therefore, we hypothesized that the increased lipid per-oxidation might be one of the factors causing the defect inVEGF expression and finally producing the impairment inthe wound-healing process. To test such a hypothesis, wetreated diabetic mice with raxofelast, an inhibitor of lipidperoxidation.

The antioxidant activity of raxofelast and its deacety-lated active metabolite IRFI 005 has been described inprevious in vitro and in vivo studies (13). In addition, IRFI005 has been shown to be a scavenger of superoxideanion, with a linear dose-response curve starting from 5mmol/l. After systemic administration of raxofelast to rats,dogs, and humans, the plasma concentrations of the par-ent compound were very low, whereas high levels of IRFI005 were found in plasma and tissue (24,25).

In our model, raxofelast was able to reverse the effectsof diabetes on wound healing by reducing lipid perox-idation and edema and by stimulating re-epithelization,neovascularization, proliferation of fibroblasts, and syn-thesis and maturation of extracellular matrix. Thus, thedegree of wound healing in db/db mice treated with raxo-felast was approximately the same as that in control het-erozygous (db/1) mice.

The beneficial effects of raxofelast on wound healingwere also stressed by the increase in breaking-strengthmeasurements. Finally, the inhibition of lipid peroxidationnormalized the pattern of VEGF mRNA expression andsecretion in diabetic mice, thus strongly supporting theidea that there might exist a close link between thedeleterious phenomenon of lipid peroxidation and a defectin VEGF production. Indeed, the improvement in VEGFexpression after raxofelast administration does not seemto be a consequence of a direct effect of the drug on theangiogenic factor. In fact, the vitamin E analog did notenhance VEGF expression in nondiabetic mice; further-

FIG. 4. Tensile strength (g/mm), evaluated atday 12, in wounds obtained from db/1 anddb/db mice treated either with vehicle (1 mlz kg–1 z day–1 i.p.) or raxofelast (15 mg z kg–1 zday–1 i.p.). Bar heights represent the mean 6SE of seven experiments. *P < 0.01 vs. db/1mice; #P < 0.01 vs. db/db mice treated withvehicle.

FIG. 5. CD levels in wound specimens col-lected at different time points from db/1 anddb/db mice treated either with vehicle (1 mlz kg–1 z day–1 i.p.) or raxofelast (15 mg z kg–1 zday–1 i.p.). Each point represents the mean 6SE of seven experiments. *P < 0.01 vs. db/1mice; #P < 0.01 vs. db/db mice treated withvehicle.

INHIBITION OF LIPID PEROXIDATION STIMULATES WOUND HEALING

672 DIABETES, VOL. 50, MARCH 2001

more, in vitro raxofelast (50 mmol/l) did not change theability of murine macrophages to secrete VEGF in re-sponse to lipopolysaccharide (F.S. et al., unpublishedobservations). The mechanism by which increased lipidperoxidation impairs VEGF expression in diabetic miceremains, at the moment, a matter of speculation. Onemay speculate that the large production of unstablereactive intermediates and hydroxyperoxides that oc-curs during lipid peroxidation could cause structuralDNA changes that lead to an impairment in the trans-duction mechanism.

Besides VEGF, other import growth factors, such asplatelet-derived growth factor and fibroblast growth fac-tor, have been shown to be severely impaired during thewound-healing process in diabetes (26,27). We must fur-ther investigate whether enhanced lipid peroxidation alsoplays a role in causing this dysfunction.

In conclusion, these results suggest that lipid peroxida-tion and an altered pattern of VEGF mRNA expressionmay contribute to deficient wound repair in geneticallydiabetic mice.

FIG. 6. VEGF mRNA expression in woundspecimens collected at different time pointsfrom db/1 and db/db mice treated either withvehicle (1 ml z kg–1 z day–1 i.p.) or raxofelast(15 mg z kg–1 z day–1 i.p.). The top panel showsrepresentative autoradiograms highlightingVEGF mRNA expression. The bottom panelshows quantitative data and represent themean 6 SE of seven experiments. *P < 0.01vs. db/1 mice; #P < 0.01 vs db/db micetreated with vehicle.

FIG. 7. VEGF levels in wound specimens collected at different timepoints from db/1 and db/db mice treated either with vehicle (1 ml zkg–1 z day–1 i.p.) or raxofelast (15 mg z kg–1 z day–1 i.p.). Each pointrepresents the mean 6 SE of seven experiments. *P < 0.01 vs. db/1mice; #P < 0.01 vs. db/db mice treated with vehicle.

D. ALTAVILLA AND ASSOCIATES

DIABETES, VOL. 50, MARCH 2001 673

ACKNOWLEDGMENTS

This work was supported, in part, by a grant from theUniversity of Messina (Fondi Ricerca d’Ateneo). There isno financial interest held by any of the investigators in thesaid company.

We thank Biomedica Foscama (Italy) for the generoussupply of raxofelast.

REFERENCES

1. Clark RAF: Cutaneous tissue repair. I. Basic biologic consideration. J Am

Acad Dermatol 13:701–725, 19852. Ferrara N, Davis-Smyth T: The biology of vascular endothelial growth

factor. Endocr Rev 18:4–25, 19973. Ferrara N: Role of vascular endothelial growth factor in the regulation of

angiogenesis. Kidney Int 56:794–814,19994. Ferrara N: Molecular and biological properties of vascular endothelial

growth factor. J Mol Med 77:527–543, 19995. Greenhalgh DG, Sprugel KH, Murray MJ, Ross R: PDGF and FGF stimulate

wound healing in the genetically diabetic mouse. Am J Pathol 136:1235–1246, 1990

6. Matuszewska B, Keogan M, Fisher DM, Soper KA, Heo CM, Huber AC,Bondi JV: Acidic fibroblast growth factor: evaluation of topical formula-tions in a diabetic mouse wound healing model. Pharm Res 11:65–71, 1994

7. Coleman DL: Diabetes-obesity syndromes in mice. Diabetes 31 (Suppl.1):1–6, 1982

8. Bohlen HG, Niggl BA: Adult microvascular disturbances as a result ofjuvenile-onset diabetes in db/db mice. Blood Vessels 16:269–276, 1979

9. Yue DK, McLennan S, Marsh M, Mai YW, Spaliviero J, Delbridge L, ReeveT, Turtle JR: Effects of experimental diabetes, uremia and malnutrition onwound healing. Diabetes 36:295–299, 1987

10. Goodson WH, Hunt TK: Studies of wound healing in experimental diabe-tes. J Surg Res 22:221–227, 1977

11. Silhi N: Diabetes and wound healing. J Wound Care 7:47–51, 198812. Frank S, Hubner G, Breier G, Longaker MT, Grenhalgh DG, Werner S:

Regulation of vascular endothelial growth factor expression in culturedkeratinocytes. J Biol Chem 12607–12613, 1995

13. Campo GM, Ceccarelli S, Squadrito F, Altavilla D, Dorigotti L, Caputi AP:Raxofelast (IRFI 016): a new hydrophilic vitamin E-like antioxidant agent.Cardiovasc Drug Rev 15:157–173, 1997

14. Pierce GF, Mustoe TA, Senior RM, Reed J, Griffin GL, Thomason A, DeuelTF: In vivo incisional wound healing augmented by platelet-derived growthfactor and recombinant c-cis gene homodimeric proteins. J Exp Med

167:974–987, 198815. Senel O, Cetinkale O, Ozbay G, Ahcioglu F Bulan R: Oxygen free radicals

impair wound healing in ischemic rat skin. Ann Plast Surg 39:517–523, 199716. Niki E, Yamamoto Y, Komuro E, Sato K: Membrane damage due to lipid

peroxidation. Am J Clin Nutr 53:201S–205S, 199117. Hall ED, Yonkers PA, Horan KL: Correlation between attenuation of

post-traumatic spinal cord ischemia and preservation of tissue vitamin Eby the 21-aminosteroid U-Z4006F: evidence for an in vivo antioxidantmechanism. J Neurotrauma 6:169–176, 1989

18. White MJ, Heckler FR: Oxygen free radicals and wound healing. Clin Plast

Surg 17:473–484, 199019. Martin A: The use of antioxidants in healing. Dermatol Surg 22:156–160,

199620. Brown LF, Yeo KT, Berse B, Yeo TK, Senger DR, Dvorak HF, van de Water

L: Expression of vascular permeability factor (vascular endothelial growthfactor) by epidermal keratinocytes during wound healing. J Exp Med

176:1375–1379, 199221. Werner S, Peters KG, Longaker MT, Fueller-Pace F, Banda MJ, Williams

LT: Large induction of keratinocyte growth factor expression in dermisduring wound healing. Proc Natl Acad Sci U S A 89:6896–6900, 1992

22. de Vries C, Escobedo JA, Ueno H, Houck K, Ferrara N, Williams LT: Thefms-like tyrosine kinase, a receptor for vascular endothelial growth factor.Science 255:989–991, 1992

23. Moore K, Roberts LJ 2nd: Measurement of lipid peroxidation. Free Radic

Res 28:659–671, 199824. Urso R, Moltoni L, Billarelli C, Segre G: Pharmacokinetics of IRFI 016 and

IRFI 005 in rats. Pharmacol Res 26 (Suppl. 1):265, 199225. Centore R, Bianco T, Ceccarelli S, Dorigotti L: Pharmacokinetic profile of

raxofelast, a novel antioxidant agent chemically related to alpha-toco-pherol. Pharmacol Res 31(Suppl. 2):38, 1995

26. Beer HD, Longaker MT, Werner S: Reduced expression of PDGF and PDGFreceptors during impaired wound healing. J Invest Dermatol 109:132–138,1997

27. Werner S, Breeden M, Hubner G, Greenhalgh, Longaker MT: Induction ofkeratinocyte growth factor expression is reduced and delayed duringwound healing in the genetically diabetic mouse. J Invest Dermatol 103:469–473, 1994

INHIBITION OF LIPID PEROXIDATION STIMULATES WOUND HEALING

674 DIABETES, VOL. 50, MARCH 2001