Abstract

Objective: Topical pretreatment with all-transretinoic acid (atRA) is known to improve healing of cutaneous
wounds. We tested the effect of atRA on wound healing of genetically diabetic db/db mice. It is known
that cutaneous wounds of db/db mice show delayed wound healing due to impaired wound contraction, delayed
granulation tissue formation and underexpression of keratinocyte growth factor (KGF). Methods: 0.1%
atRA in 100 mg aqueous gel was applied to the back skin of db/db mice as well as to their normal heterozygous
littermates, db/+ mice, for five consecutive days, and 2 days after completion of the atRA treatment, two
round excisional wounds were created down the panniculus carnosus with a 6-mm punch biopsy on the back
skin of each mouse. Results: After 5 days treatment with 0.1% atRA, significant hypertrophy of the epidermis
and dermis, neovascularization, and inflammatory cell invasion were seen in the skin of the db/db mice, but
these effects were seen only weakly in db/+ mice. Wounds in atRA-treated db/db mice closed more rapidly
than those in vehicle-treated db/db mice. KGF mRNA expression, which is usually significantly lower
in db/db mice than in normal mice, in wounds of atRAtreated db/db mice on day 1 of treatment was as strong
as in db/+ mice. Conclusion: Pretreatment with atRA reversed the impaired wound healing in db/db mice.

Introduction

Retinoic acid (RA) improves the clinical symptoms of acne vulgaris, psoriasis and skin photoaging [1]. Histologically, topical treatment with RA induces hypertrophy in the epidermis and dermis via an increase in dermal vasculature and invasion of inflammatory cells. RA has several contradictory effects on wound healing. For example, systemic application of RA has deleterious effect on healing of fullthickness skin wounds and corneal wounds of diabetic animals [2, 3]. On the other hand, systemic RA reverses delayed wound healing in steroid-treated animals [4-7]. RA is also known to improve wound healing of abraded skin by acceleration of epidermal turnover when it is applied topically prior to wounding [8]. Topical pretreatment with RA also enhances healing of full-thickness excisional wounds of photoaged and normal skin [9, 10]. These results suggest that wound healing may be affected by the route or schedule of retinoid administration, and that, at least in some skin conditions, topical pretreatment may be the most effective method of administration of all-trans-RA (atRA). We hypothesized that impaired wound healing of genetically diabetic db/db mice may be reversed by topical pretreatment with atRA. Db/db mice show clinical symptoms similar to insulin-resistant diabetes mellitus patients, such as severe obesity, hyperglycemia, hyperinsulinemia, and insulin resistance [11-14]. These symptoms were derived from genomic mutation in the leptin receptor. Db/db mice are also known to show impaired cutaneous wound healing, as is universally observed in diabetic patients [11]. Wound healing of db/db mice is characterized by weak contraction, slow inflammatory cell invasion and delayed granulation formation. The characteristic anatomy of the skin of db/db mice, including thick subdermal fat and an extremely thin dermis, may also contribute to impairment of wound healing. Many growth factors and their receptors, such as FGF, FGF receptors, KGF, IGF and VEGF, have been found to be aberrantly expressed in wounds of db/db mice [11, 12] Of these, the factor that is considered to be the most important for delayed wound healing in db/db mice is KGF [15]. In normal mice, the mRNA levels of KGF are acutely upregulated early in wound healing [16, 17]. However, KGF mRNA levels in db/db mice are unchanged throughout the healing process [15]. Since upregulation of KGFmRNA is one of the earliest events of wound healing, KGF is considered to be an important factor that might regulate　downstream events during wound healing.

Materials and methods

Animals
Female C57BLK/ksj db/db mice at 8-12 weeks of age (Saitama Experimental Animals Supply Company, Saitama, Japan) and their　normal littermates (db/+ mice) were used throughout this study. Animal experiments were performed according to the Laboratory　Animal Guidelines of Tokyo University. All animal procedures were performed under general anesthesia with diethyl ether. Three independent　experiments were performed, and 12 to 15 mice were used in each experiment. The back skin of each mouse was shaved and　topically treated for five consecutive days with 100 mg 0.1% atRA (Sigma Aldrich, Tokyo, Japan) in aqueous gel vehicle or 100 mg
aqueous gel without atRA. Scaling was graded by three investigators according to the visual scoring method of Effendy et al. [18]:　0, no scaling; 1, weak scaling; 2, moderate scaling; 3, large flakes, intense peeling.
On the 2nd day after completion of atRA treatment, two 6-mm round excisional wounds, down the panniculus carnosus and its associated　fascia, were created in each mouse. The wound area was calculated as the product of the long and short diameters of each　wound. Mice were killed on days 1, 3, 7 and 14 by cervical dislocation under general anesthesia with diethyl ether. Wound tissue　was processed for histology and RT-PCR.

Histology
The mice were injected subcutaneously with 30 μg/g 5-bromo-2-deoxyuridine (BrdU labeling reagent; Amersham Pharmacia, Shinjuku, Tokyo) 3 and 24 h before they were killed. Skin samples were fixed in 10% formalin, dehydrated and embedded in paraffin. Paraffin sections were cut at a thickness of 5 μm. Sections were rehydrated and stained with hematoxylin and eosin, and with Gomori's trichrome stain. Immunohistochemistry to determine the level of BrdU incorporation into cells was performed as follows. Sections were incubated in 4 N HCl for 30 min at 37 °C followed by 0.1 M borax buffer, pH 9.0, for 10 min at room temperature. The specimens
were then treated with anti-BrdU monoclonal antibody (Amersham Pharmacia) followed by biotinylated anti-mouse IgG. The specimens were then incubated in ABC reagent (Vector, Burlingame, Calif.), and developed with 3,3′-diaminobenzidine.

RT-PCR
mRNA was extracted from a 2-mm wound margin using ISOGEN LS (Nippon Gene, Shinjuku, Tokyo). The mRNA (1 μg in a total volume of 20 μl) was transcribed to cDNA by AMV reverse transcriptase (Promega Japan, Higashinihonnbashi, Tokyo) at 37°C for 1 h. For semiquantitative detection of KGF and β-actin cDNA, 1 μl cDNA in a total volume of 50 μl (50 mM KCl, 15 mM Tris-HCl, pH 8.0, 4 mM MgCl2) was amplified with Ampli-Taq Gold (PE Biosystems, Roppongi, Tokyo) in a PC-960G thermal cycler (Cosmo Bio, Toyo, Tokyo). The sequences of the oligonucleotide primers used for amplification of KGF were TTGCAATGAACAAGGAAGGA
and GAATTCTATCTTGCAATGAA, and those for β-actin were TGAGGAGCACCCTATGCTGC and TAGCCCTCGTAGATGGG. Amplification was carried out as follows: pre-PCR step of 10 min at 94 °C; 32 PCR cycles of 30 s at 94 °C, 30 s at 58 °C and 1 min at 72 °C; and a final step of 10 min at 72°C. We had already confirmed in a preliminary study that the PCR products would be within the linear log phase under these conditions.

Results

The effects of topical application of atRA
Scaling was more intense in db/db mice than in db/+ mice following treatment with 0.1% atRA (Fig. 1). Although
0.1% atRA did not induce significant scaling in db/+ mice, higher concentrations of atRA did induce significant scaling in these mice, so the effect of atRA appears to be dose-dependent. Histologically, the skin of db/+ mouse showed only slight dermal and epidermal hypertrophy after treatment with 0.1% atRA. On the other hand, the skin of atRAtreated db/db mice showed significant hypertrophy of epidermis and dermis, scaling of stratum corneum, an increase in cellularity and proliferation of subepidermal vessels (Fig. 2). Elongation of the rete ridges and atrophy of the sebaceous glands were also characteristic of atRAtreated db/db mice. In vehicle-treated db/db mice, BrdU incorporation was observed throughout the basal layer of the epidermis. However, in atRA-treated db/db mice, BrdU incorporation was greater in the follicular epidermis, but not in the interfollicular epidermis (Fig. 3).

Wound healing of atRA- and vehicle-treated db/db mice

Day 0 after wounding
The wounds in vehicle-treated db/db mice were 20-30% larger than the original wounds, but the wounds in atRAtreated db/db mice did not show enlargement just after wounding (Fig. 4).

Day 1 after wounding
Vehicle-treated db/db mice on day 1 after wounding showed only a small number of inflammatory cells in the
wound cleft. Only a small amount of extracellular matrix had formed at this time. However, atRA-treated db/db mice on day 1 showed massive invasion of inflammatory cells in the wound cleft. In addition, a large amount of extracellular matrix staining bluish with trichrome reagent had formed within the granulation tissue, which connected the granulation tissue with the wound margin (Fig. 5a, b).

Day 14 after wounding
Macroscopically, 78% of wounds in vehicle-treated db/db mice had not closed on day 14. Histologically there was a large amount of granulation tissue within the wide wound cleft. Epidermal regeneration had started, but due to the distance between the wound margins, there still was an uncovered area within the cleft. On the other hand, 89% of the wounds in atRA-treated db/db mice had already closed by this time. Histologically the wound margins had come close to each other, and reepithelialization was almost complete (Fig. 5c, d).

Expression of KGF mRNA
mRNA expression of KGF at the time of wounding did not show significant differences between atRA- and vehicletreated db/db mice (data not shown). However, KGF mRNA in the wounds of atRA-treated db/db mice on day 1 after wounding was significantly higher than in vehicleor sodium lauryl sulfate-treated db/db mice (Fig. 6).

Discussion

In the study reported here we demonstrated that (1) topical treatment with atRA had stronger effects in db/db mice than in db/+ mice, and that (2) atRA pretreatment reversed delayed wound healing in db/db mice.
It is noteworthy that atRA induced cutaneous changes in db/db mice much more strongly than in db/+ mice. This
does not mean that db/+ mice are totally refractory to the effects of atRA, but that db/db mice are more sensitive to atRA than db/+ mice, because the latter showed effects similar to those seen in db/db mice when they were treated with higher concentrations of atRA. It is not clear whether this increased sensitivity was due to downstream events resulting from the leptin receptor deficiency, or due to loss of barrier function. Similar model-specific sensitivity has been observed in other experiments. For example, atRA increases the synthesis of collagen and hyaluronic acid in UVB-irradiated mice, but not in non-irradiated hairless mice [19, 20]. It has also been reported that atRA effectively reverses steroid-retarded repair, which may be a result of a mutual interaction between glucocorticoids and retinoids in the process of inflammation, immunity and connective tissue production [4-7]. It is interesting that the sebaceous glands of db/db mice showed extremely strong changes. Sebaceous glands almost completely disappeared in db/db mice after treatment with 0.1% atRA, but those in db/+ mice did not show significant changes after the same treatment. Even with 0.4% atRA, about 30% of sebaceous glands remained intact in db/+ mice, although significant dermal hypertrophy was observed with this concentration of atRA. Pretreatment with 0.1% atRA had little effect on the wound healing in normal mice, but it significantly accelerated wound healing in db/db mice. This may reflect the biological effects of atRA pretreatment in these animals. Delayed wound healing in db/db mice seemed to be enhanced by treatment with atRA. First, full-thickness wounds in db/db mice had enlarged 20-40% just after wounding due to their extremely thin dermis with a small amount of extracellular matrix, which was the primary cause of delayed wound healing in db/db mice during the early stages of wound healing. Treatment with atRA made the dermis thick and stiff preventing enlargement of excisional wounds in db/db mice. Second, granulation tissue and extracellular matrix formation induced by atRA may also accelerate wound healing in db/db mice. This granulation tissue contains huge numbers of inflammatory cells that release cytokines and growth factors that enhance wound healing. In addition, smooth muscle actin fiber involved in wound granulation may exert a contractile force. The mechanism of action of atRA in wound healing is complicated. A study using a dominant-negative retinoic acid receptor mutant strongly indicates that epidermal proliferation and regeneration is initiated by heparin-binding epidermal growth factor (HB-EGF) induced by atRA [21, 22]. Since atRA induction of HB-EGF occurs in suprabasal cells prior to the onset of basal cell hyperproliferation, this mechanism may well explain the effectiveness of atRA pretreatment. On the other hand, atRA may also enhance macrophage function, and thus finally may enhance macrophage-derived growth factors such as FGF, TGF and IGF-I [6]. This is also plausible since macrophage dysfunction is seen in db/db mice [23]. A more sensitive real-time RT-PCR could detect quantitative differences in these macrophagederived growth factors. Although suppression of KGF is characteristic of wound healing in db/db mice, KGF mRNA levels in unwounded skin are almost the same between db/db and db/+ mice [20], which was also confirmed by our RT-PCR (data not shown). In addition, atRA-induced epidermal proliferation in db/db mouse was not associated with KGF upregulation before wounding. Taken together, atRA does not directly induce KGF in unwounded skin but it causes the skin of db/db mice to induce KGF. Previous studies of atRA application to already existing wounds have revealed conflicting results. Some investigators have shown enhanced healing [7, 14, 24], but others have shown retardation of healing [10, 25, 26]. On the other hand, topical pretreatment with atRA has consistently shown improvement in wound healing. In addition, our experimental and clinical experience has revealed that atRA worsens wound healing of full-thickness open wounds (unpublished data). For these reasons we do not recommend atRA application to diabetic ulcers. In practice, atRA may enhance wound healing when it is applied before surgery in diabetic patients. In the present study, we only evaluated the effects of atRA pretreatment in diabetic wounds, but it may be possible to apply this strategy to other healing-impaired wounds, such as those in the skin of the elderly or malnourished, or ischemic skin. Pretreatment with atRA may prevent delayed healing, infection and dehiscence in these potentially healing-impaired wounds.

Fig. 1 Scaling score of db/db mice treated with 0.1% atRA, and of db/+ mice treated with 0.1% or 0.4% atRA. Three
mice in each group were evaluated. The values are means ± SD of duplicate measurements from three independent experiments.

Fig. 2a-d H&E staining of the back skin of db/+ mice (a vehicletreated mouse, b 0.1% atRA-treated mouse) and db/db mice (c vehicle-treated mouse, d 0.1% atRA-treated mouse). Arrows indicate sebaceous glands.

Fig. 3a-d BrdU staining of back skin of db/db mice treated with either of vehicle (a) or 0.1% atRA (b) for 5 days. c, d High-power magnification of a and b (a, b ×100; c, d ×200)

Fig. 4 Wound area of db/+ and db/db mice. Round wounds of diameter 6 mm were made on the back skin of mice pretreated with either 0.1% atRA or vehicle. Wounds of db/+ mice had almost closed by day 7 whether they had been treated with 0.1% atRA or vehicle. On the other hand, vehicle-treated db/db mice showed enlargement
of the wound area on day 1, which was followed by delayed wound contraction. Pretreatment with atRA prevented
the wounds of db/db mice from expanding, which may have contributed to the earlier wound closure. Data represent means ± SD of duplicate measurements from three independent experiments.

Fig. 5a-c Wound sections of db/db mice pretreated with vehicle (a, c) or atRA (b, d) were stained with Gomori's trichrome. Arrows indicate the wound margin (a, b 1 day after wounding; c, d 14 days after wounding). a, b ×200; c, d ×50

Fig. 6 Expression of KGF mRNA in day 1 wounds of db/db and
db/+ mice treated with atRA (RA), sodium lauryl sulphate (SLS), or
vehicle (V). A representative result of RT-PCR from three independent
experiments is shown (KGF KGF mRNA, BA β-actin mRNA)

References
1. Zouboulis ChC, Orfanos CE (2000) Retinoids. In: Millikan LE (ed) Drug therapy in dermatology. Marcel Dekker, New York Basel, pp 171-233
2. Frosch PJ, Czarnetzki BM (1989) Effect of retinoids on wound healing in diabetic rats. Arch Dermatol Res 281:424-426
3. Hatchell DL, Ubels JL, Stekiel T, Hatchell MC (1985) Corneal epithelial wound healing in normal and diabetic rabbits treated with tretinoin. Arch Ophthalmol 103:98-100
4. Wicke C, Halliday B, Allen D, Roche NS, Scheuenstuhl H, Spencer MM, Roberts AB, Hunt TK (2000) Effects of steroids and retinoids on wound healing. Arch Surg 135:1265-1270
5. Anstead GM (1998) Steroids, retinoids, and wound healing. Adv Wound Care 11:277-285
6. Hunt TK (1986) Vitamin A and wound healing. J Am Acad Dermatol 15:817-821
7. Hunt TK, Ehrlich HP, Garcia JA, Dunphy JE (1969) Effect of vitamin A on reversing the inhibitory effect of cortisone on healing of open wounds in animals and man. Ann Surg 170:633-641
8. Mandy SH (1986) Tretinoin in the preoperative and postoperative management of dermabrasion. J Am Acad Dermatol 15: 878-879, 888-889
9. Popp C, Kligman AM, Stoudemayer TJ (1995) Pretreatment of photoaged forearm skin with topical tretinoin accelerates healing of full-thickness wounds. Br J Dermatol 132:46-53
10. Klein P (1975) Vitamin A acid and wound healing. Acta Derm Venereol Suppl (Stockh) 74:171-173
11. Greenhalgh DG, Sprugel KH, Murray MJ, Ross R (1990) PDGF and FGF stimulate wound healing in the genetically diabetic mouse. Am J Pathol 136:1235-1246 520
12. Tsuboi R, Rifkin DB (1990) Recombinant basic fibroblast growth factor stimulates wound healing in healing-impaired db/db mice. J Exp Med 172:245-251
13. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, Lakey ND, Culpepper J, Moore KJ, Breitbart RE, Duyk GM, Tepper RI, Morgenstern JP (1996) Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell 84:491-495
14. Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG, Lee JI, Friedman JM (1996) Abnormal splicing of the leptin receptor in diabetic mice. Nature 379: 632-635
15. Werner S, Peters KG, Longaker MT, Fuller-Pace F, Banda MJ, Williams LT (1992) Large induction of keratinocyte growth factor expression in the dermis during wound healing. Proc Natl Acad Sci U S A 89:6896-6900
16. Werner S, Breeden M, Hubner G, Greenhalgh DG, Longaker MT (1994) Induction of keratinocyte growth factor expression is reduced and delayed during wound healing in the genetically diabetic mouse. J Invest Dermatol 103:469-473
17. Nakagawa K, Zhang Y, Tsuji H, Yoshizumi M, Kasahara T, Nishimura H, Nawroths PP, Nakagawa M (1998). The angiogenic effect of tissue factor on tumors and wounds. Semin Thromb Hemost 24:207-210
18. Effendy I, Weltfriend S, Kwangsukstith C, Singh P, Maibach HI (1996) Effects of all-trans retinoic acid and sodium lauryl sulphate on the permeability of human skin in vitro. Br J Dermatol 135:428-432
19. Margelin M, Medaisko C, Lombard D, Picard J, Fountanier A (1996) Hyaluronic acid and dermatan sulfate are selectively stimulated by retinoic acid in irradiated and nonirradiated hairless mouse skin. J Invest Dermatol 106:505-515
20. Varani J, Fisher GJ, Kang S, Voorhees JJ (1998) Molecular mechanisms of intrinsic skin aging and retinoid-induced skin repair. J Invest Dermatol Symp Proc 3:57-60
21. McCarthy DW, Downing MT, Brigstock DR, Luquette MH, Brown KD, Abad MS, Besner GE (1996) Production of heparin- binding epidermal growth factor (HB-EGF) at sites of thermal injury in pediatric patients. J Invest Dermatol 106:49-56
22. Xiao JH, Feng X, Di W, Peng ZH, Li LA, Chambon P, Voorhees JJ (1999) Identification of heparin-binding EGF-like growth factor as a target in intercellular regulation of epidermal basal cell
growth by suprabasal retinoic acid receptors. EMBO J 18:1539-1548
23. Zykova SN, Jenssen TG, Berdal M, Olsen R, Myklebust R, Seljelid R (2000) Altered cytokine and nitric oxide secretion in vitro by macrophages from diabetic type II-like db/db mice. Diabetes 49:1451-1458
24. Herrmann JB, Woodward SC (1972) An experimental study of wound healing accelerators. Am J Surg 38:26-34
25. Hung VC, Lee JY, Zitelli JA, Hebda PA (1989) Topical tretinoin and epithelial wound healing. Arch Dermatol 125:65-69
26. Watcher MA, Wheeland RG (1989) The role of topical agents in the healing of full-thickness wound. J Dermatol Surg Oncol 15:1188-1195