British journal of Dernmtoiocjii
125. 217-221.
s ()007096391002{)4Q
The metabolism of fibroblasts from normal and fibrotic skin is inhibited by minoxidil in vitro G.C.PRIESTLEY, RUTH LORD AND P.STAVROPOULOS University Department of Dermatology. Royal infirmary. Edinburgh. U.K. Accepted for publication 27 March 1991
Summary
The effects of minoxidil in vitro were studied using fibroblasts grown from the iesional skin of patients with lichen sclero.su,s et atrophicus. morphoea and from the skin of normal individuals. The proliferation of all fibroblast lines over J days was inhibited in proportion to the concentration of minoxidil. heing 20% or less of controls at 1 mM. where cell viability was only marginally reduced (84 ± 2 % vs. 88±2%(SEM) in controls). At 5 mM there was usually a net loss of cells and only 72% of those remaining were viable. In contrast, minoxidil at O-l-l niM stimulated the proliferation of foreskin keratinocytes by up to 1 30%. Contraction of collagen lattices containing the three types of fibroblasts was inhibited by 22-2bYo with 1 mM minoxidil after 5 days and by S()-94% with 5 mM. Secretion of glycosaniinoglycans by normal fibroblasts showed concentration-dependent reduction, being 2 5 ± 6% of that of untreated cultures with 1 mM minoxidil. These findings show that minoxidil has a range of inhibitory effects on both normal and abnormal skin fibroblasts IN vitro, which contrast with its stimulation of skin epithelial cells, aud support suggestions that it may provide a useful topical treatment for keloids and other fibroses.
The hypertrichosis associated with the systemic use of minoxidil to control hypertension' suggested that this drug would be of help in the treatment of baldness.- Cellculture studies have demonstrated that minoxidil has direct effects on epidermal ceils in Wtro'^ independent of the blood supply. More recently Mnrad and Pinneir found a selective inhibition of the enzyme lysyl hydroxylase in cultured skin fibroblasts and that proiyi hydroxylase was unaffected. As hydroxylation of lysine residues is the first step in collagen cross-linking, the authors suggested that minoxidil might be an effective topical treatment for fibroses. such as keloids and hypertrophic scars, where collagen accumulates. We have now extended some of the observations on normal skin fibrohlasts to cells from fibrotic skin.
Methods Materials Minoxidil was provided by Upjohn Limited. Crawley. U,K, The white crystalline powder was dissolved in Correspondence: Dr C.C.Priestley. University Department of Dermatology. Level 4. Lauriston Building. Royal Infirmary. Edinburgh EHJ 9YW. U.K. Presented at the meeting of the British Society for Investigative Dermatology in Newcastle. September. 1990.
culture medium to give a 5 mM solution, filtered to ensure sterility and diluted with fresh culture medium to provide a range of concentrations. Culture media and supplements were obtained from Gibco. Paisley. Scotland, unless specified otherwise. Fibroblast culture Fibroblast lines were established from excision biopsies of the Iesional skin of patients with morphoea. extragenital lichen sclerosus et atrophicus (LSA). and from the normal forearm skin of healthy volunteers. The biopsies were washed, chopped into 1-mm fragments and sandwiched between a glass coverslip and the bottom of a six-well plastic dish (Nunc) in Dulbecco's Eagle medium supplemented with 20% fetal calf serum (FCS). penicillin, streptomycin and glutamine. The dishes were cultured at J7°C in an atmosphere of 5% CO2' 9 "5%) air. The medium was changed twice weekly and after 6-8 weeks, when there was sufficient growth of tibrobfasts. the cells were dispersed with typsinversene and transferred to 2S-cm- fiasks in the same medium with 10% PCS. Samples of third and later passage cells were preserved in liquid nitrogen until needed, and then recovered for use in experiments as 5th to 10th passage cells. 217
218
G.C.PRIESTLEY et al
Fibroblast proliferatiort assay
Twenty-eight cultures were set up by plating 10^ fibroblasts in each 25-cm^ flask (day 0) in medium containing 10% FCS. The medium was changed on day 1 and on day 3 four flasks were used for baseline cell counts. The cultures were washed with phosphatebuffered saline (PBS) and cell suspensions prepared with trypsin-versene were counted electronically (Coulter Counter DN). The remaining cultures were grouped in fours to receive medium containing O i . 0-2. 0-5. 1 and 5 mM minoxidil. leaving a second control group of four cultures untreated. In two experiments shown in Figure 1. the total number of cultures was increased to 52 to accommodate an additional minoxidil concentration. Media were replaced on day 4 and on day 6 the cells in all cultures were counted. The increase in total cells in each group was expressed as a percentage of that in the untreated controls.
Collagen lattice contraction
Collagen lattices were made with type 1 collagen from rat-tail tendons, as previously described.'' in the 35-mm diameter wells of bacteriological grade six-well dishes (Linbro. Flow Laboratories). Hach lattice contained 2 x 1 0 ^ fibroblasts and measurements of lattice diameter were made daily using mm-squared graph paper over a light box. Two lattices were used for each concentration of minoxidil. which was Included in the lattice mix at zero time.
Glycosaminoglycans
secretion
Culture media collected on day 6 of fibroblast proliferation assays were dialysed against acetate buffer (pH 5). concentrated twofold with polyacryiamide beads, digested overnight at 3 7°C with testicular hyaluronidase (Sigma Type 1: 2 mg per sample) and acidified with trichloracetic acid to a final 5% to precipitate protein. The uronic acid in 40.000 xgmax supernatants was detected colorimetricaliy with m-hydroxydiphenyl at 520 nm after hoiling with sulphuric acid-borate.' Glycosaminoglycans (GAG) secretion was expressed as /(g uronic acid secreted per 10" tibroblasts in 48 h and calculated as a percentage of that in untreated control cultures."
Keratinocyte proliferation assay
Foreskins obtained from normal infants were treated
with dispase (Sigma, protease type IX) at 4°C overnight to separate the epidermis from the dermis. The epidermis was incubated in 0-25% trypsin in PBS at 37°C for iO min to release the epidermal cells. The cells were collected by centrifugation, counted and plated in 24well plastic dishes (Nunc) in Dulbecco's Eagle medium diluted 3:1 with Ham's F12 medium and containing 10% FCS. penicilfin. streptomycin. 1 0 " ' " M cholera toxin (Sigma). 5 /(g/ml transferrin and 2 x iO"'* M triiodothyronine. Plating densities of the crude epidermal suspension were approximately 7x 10^ cells per well. After 2 days at 3 7 T in 5% CO^: 95% air. the culture medium was supplemented with 10 ng/ml epidermal growth factor (EGF). After a further 2 days, four wells of the dish were trypsinized for baseline cell counts while other wells were replenished with standard medium (four wells) as controls, or with 0-1-5 mM minoxidil (four wells per concentration) dissolved in standard medium without EGF. Media were replaced the next day and after a final 48 h all the cultures were trypsinized for cell counts. Proliferation in minoxidil-treated weUs (i.e. increase in cells above baseline counts) was expressed as a percentage of that in untreated control wells.
Results Cell proliferation
The proliferation of five lines of normal fibroblasts. one line from morphoea and one from LSA was similarly affected by minoxidil. In general, there was concentration-dependent inhibition of proliferation beginning at 0 1 mM (Figs 1 and 2). as described by Murad and Pinnefi."' Proliferation was less than 25% of untreated
100 r
Figure 1. Effect of minoxidil on the proliferation of fibmblasts from lesions of morphoea. LSA and from normal skin IHSI"). Kach point is a mean from four culturesiSIv
IN-VITRO EFFECTS OF MINOXIDIL ON FIBROBLASTS
200 -
HSF
-150 L Figure 2. Comparison of the inhibitory effect of minoxidil on the proliferation of three additional lines of fibroblasts from normal skin (HSF, • means±standard errors, n = 4]. and its stimulation of three different preparations of foreskin keratinocytes (KC. O, means±SE.
controls at 1 ITIM and approached zero or showed a net loss of cells at 5 niM. In a minority of experiments (4/10) there was some (rfm((o/ 1984: 82: 9 0 - i . 5 IVlurad S. Pinnell SR. Suppression of iibroblast proliferalion and lysyi hydroxyiase activily by minoxidil. / Bio! Cheni 1987; 262: 11975-8. 6 Adams i.W. Priestley GC Contraction of ailiiij;eii lattices by skin libroblasts: drug-induced changes. Arch Denmitol Res 1988: 280: 114-8. 7 Blumenkrantz N. Asboe-Hansen C. New method for quantitative determination of uronic acids. Anal Hiochem 197J: 54: 484-9. 8 Oakley CA, Priestley GC. Density-dependent regulation of skin Hbrobiast glycosaminoglycans in vitro: control by a secreted factor. Ari-h Dernmtol Res 1985: 227: 2f)4-9. 9 Cherry G. Hughes MA. Dawber RPR. Ryan T|. Minoxidil-induced changes in the contraction of cDlliiRen lattices by human skin tibroblasts: a new means of control of excessive clinical scar formation. / Invest Dermatol 1 990: 94: 514, 10 l^vene CI. Possibilities for the therapeutic control of librosis. Hr / Demi«to/1985: 112: i 6 J - 7 1 . 11 Spindler JR. Deaths occurring during clinical studies of topical minoxidil. / Am Atvd Iknnatol 1987: 16: 725-9. 12 Priestley GC. Brown |C. Kffects of griseofulvin on the morphology, growth and metabolism of libroblasts in culture. Hr / Dermatol 1978:99: 245-52. I J Priestley GC. Changes in the growth and metabolism of cells cullured from normal, sclerotic and rheumatoid connective tissue brought about by D-penicil!aminc and by sodium saticyiate. / Invest Dermatol 1980: 74: 41 i-7, 14 Priestley CiC. IJrown )C. Tlffecls of potassium para-aminobenzoate on growth unii macromolecule synthesis on (ibroblasts cultures from normal and sclerodermatotis human skin and rheumatoid synovial cells. / Invest Dermatol 1 979: 72: 161 4. I S Priestley (iC. Kffects of corticostt-roids on the growth and metabolism of libroblasts cultured from human .skin, lir I Dermatol 1 978: 99: 253-61. 16 Priestley CC. Glycosaminoglycans production by cultured skin fibroblasts from the Pasini and Cockayne-Touraine forms of dominant dystrophic epidermolysis bullosa. / Irn-rsl l^ermalol 1991: 9ft: 168 71.
125. 217-221.
s ()007096391002{)4Q
The metabolism of fibroblasts from normal and fibrotic skin is inhibited by minoxidil in vitro G.C.PRIESTLEY, RUTH LORD AND P.STAVROPOULOS University Department of Dermatology. Royal infirmary. Edinburgh. U.K. Accepted for publication 27 March 1991
Summary
The effects of minoxidil in vitro were studied using fibroblasts grown from the iesional skin of patients with lichen sclero.su,s et atrophicus. morphoea and from the skin of normal individuals. The proliferation of all fibroblast lines over J days was inhibited in proportion to the concentration of minoxidil. heing 20% or less of controls at 1 mM. where cell viability was only marginally reduced (84 ± 2 % vs. 88±2%(SEM) in controls). At 5 mM there was usually a net loss of cells and only 72% of those remaining were viable. In contrast, minoxidil at O-l-l niM stimulated the proliferation of foreskin keratinocytes by up to 1 30%. Contraction of collagen lattices containing the three types of fibroblasts was inhibited by 22-2bYo with 1 mM minoxidil after 5 days and by S()-94% with 5 mM. Secretion of glycosaniinoglycans by normal fibroblasts showed concentration-dependent reduction, being 2 5 ± 6% of that of untreated cultures with 1 mM minoxidil. These findings show that minoxidil has a range of inhibitory effects on both normal and abnormal skin fibroblasts IN vitro, which contrast with its stimulation of skin epithelial cells, aud support suggestions that it may provide a useful topical treatment for keloids and other fibroses.
The hypertrichosis associated with the systemic use of minoxidil to control hypertension' suggested that this drug would be of help in the treatment of baldness.- Cellculture studies have demonstrated that minoxidil has direct effects on epidermal ceils in Wtro'^ independent of the blood supply. More recently Mnrad and Pinneir found a selective inhibition of the enzyme lysyl hydroxylase in cultured skin fibroblasts and that proiyi hydroxylase was unaffected. As hydroxylation of lysine residues is the first step in collagen cross-linking, the authors suggested that minoxidil might be an effective topical treatment for fibroses. such as keloids and hypertrophic scars, where collagen accumulates. We have now extended some of the observations on normal skin fibrohlasts to cells from fibrotic skin.
Methods Materials Minoxidil was provided by Upjohn Limited. Crawley. U,K, The white crystalline powder was dissolved in Correspondence: Dr C.C.Priestley. University Department of Dermatology. Level 4. Lauriston Building. Royal Infirmary. Edinburgh EHJ 9YW. U.K. Presented at the meeting of the British Society for Investigative Dermatology in Newcastle. September. 1990.
culture medium to give a 5 mM solution, filtered to ensure sterility and diluted with fresh culture medium to provide a range of concentrations. Culture media and supplements were obtained from Gibco. Paisley. Scotland, unless specified otherwise. Fibroblast culture Fibroblast lines were established from excision biopsies of the Iesional skin of patients with morphoea. extragenital lichen sclerosus et atrophicus (LSA). and from the normal forearm skin of healthy volunteers. The biopsies were washed, chopped into 1-mm fragments and sandwiched between a glass coverslip and the bottom of a six-well plastic dish (Nunc) in Dulbecco's Eagle medium supplemented with 20% fetal calf serum (FCS). penicillin, streptomycin and glutamine. The dishes were cultured at J7°C in an atmosphere of 5% CO2' 9 "5%) air. The medium was changed twice weekly and after 6-8 weeks, when there was sufficient growth of tibrobfasts. the cells were dispersed with typsinversene and transferred to 2S-cm- fiasks in the same medium with 10% PCS. Samples of third and later passage cells were preserved in liquid nitrogen until needed, and then recovered for use in experiments as 5th to 10th passage cells. 217
218
G.C.PRIESTLEY et al
Fibroblast proliferatiort assay
Twenty-eight cultures were set up by plating 10^ fibroblasts in each 25-cm^ flask (day 0) in medium containing 10% FCS. The medium was changed on day 1 and on day 3 four flasks were used for baseline cell counts. The cultures were washed with phosphatebuffered saline (PBS) and cell suspensions prepared with trypsin-versene were counted electronically (Coulter Counter DN). The remaining cultures were grouped in fours to receive medium containing O i . 0-2. 0-5. 1 and 5 mM minoxidil. leaving a second control group of four cultures untreated. In two experiments shown in Figure 1. the total number of cultures was increased to 52 to accommodate an additional minoxidil concentration. Media were replaced on day 4 and on day 6 the cells in all cultures were counted. The increase in total cells in each group was expressed as a percentage of that in the untreated controls.
Collagen lattice contraction
Collagen lattices were made with type 1 collagen from rat-tail tendons, as previously described.'' in the 35-mm diameter wells of bacteriological grade six-well dishes (Linbro. Flow Laboratories). Hach lattice contained 2 x 1 0 ^ fibroblasts and measurements of lattice diameter were made daily using mm-squared graph paper over a light box. Two lattices were used for each concentration of minoxidil. which was Included in the lattice mix at zero time.
Glycosaminoglycans
secretion
Culture media collected on day 6 of fibroblast proliferation assays were dialysed against acetate buffer (pH 5). concentrated twofold with polyacryiamide beads, digested overnight at 3 7°C with testicular hyaluronidase (Sigma Type 1: 2 mg per sample) and acidified with trichloracetic acid to a final 5% to precipitate protein. The uronic acid in 40.000 xgmax supernatants was detected colorimetricaliy with m-hydroxydiphenyl at 520 nm after hoiling with sulphuric acid-borate.' Glycosaminoglycans (GAG) secretion was expressed as /(g uronic acid secreted per 10" tibroblasts in 48 h and calculated as a percentage of that in untreated control cultures."
Keratinocyte proliferation assay
Foreskins obtained from normal infants were treated
with dispase (Sigma, protease type IX) at 4°C overnight to separate the epidermis from the dermis. The epidermis was incubated in 0-25% trypsin in PBS at 37°C for iO min to release the epidermal cells. The cells were collected by centrifugation, counted and plated in 24well plastic dishes (Nunc) in Dulbecco's Eagle medium diluted 3:1 with Ham's F12 medium and containing 10% FCS. penicilfin. streptomycin. 1 0 " ' " M cholera toxin (Sigma). 5 /(g/ml transferrin and 2 x iO"'* M triiodothyronine. Plating densities of the crude epidermal suspension were approximately 7x 10^ cells per well. After 2 days at 3 7 T in 5% CO^: 95% air. the culture medium was supplemented with 10 ng/ml epidermal growth factor (EGF). After a further 2 days, four wells of the dish were trypsinized for baseline cell counts while other wells were replenished with standard medium (four wells) as controls, or with 0-1-5 mM minoxidil (four wells per concentration) dissolved in standard medium without EGF. Media were replaced the next day and after a final 48 h all the cultures were trypsinized for cell counts. Proliferation in minoxidil-treated weUs (i.e. increase in cells above baseline counts) was expressed as a percentage of that in untreated control wells.
Results Cell proliferation
The proliferation of five lines of normal fibroblasts. one line from morphoea and one from LSA was similarly affected by minoxidil. In general, there was concentration-dependent inhibition of proliferation beginning at 0 1 mM (Figs 1 and 2). as described by Murad and Pinnefi."' Proliferation was less than 25% of untreated
100 r
Figure 1. Effect of minoxidil on the proliferation of fibmblasts from lesions of morphoea. LSA and from normal skin IHSI"). Kach point is a mean from four culturesiSIv
IN-VITRO EFFECTS OF MINOXIDIL ON FIBROBLASTS
200 -
HSF
-150 L Figure 2. Comparison of the inhibitory effect of minoxidil on the proliferation of three additional lines of fibroblasts from normal skin (HSF, • means±standard errors, n = 4]. and its stimulation of three different preparations of foreskin keratinocytes (KC. O, means±SE.
controls at 1 ITIM and approached zero or showed a net loss of cells at 5 niM. In a minority of experiments (4/10) there was some (rfm((o/ 1984: 82: 9 0 - i . 5 IVlurad S. Pinnell SR. Suppression of iibroblast proliferalion and lysyi hydroxyiase activily by minoxidil. / Bio! Cheni 1987; 262: 11975-8. 6 Adams i.W. Priestley GC Contraction of ailiiij;eii lattices by skin libroblasts: drug-induced changes. Arch Denmitol Res 1988: 280: 114-8. 7 Blumenkrantz N. Asboe-Hansen C. New method for quantitative determination of uronic acids. Anal Hiochem 197J: 54: 484-9. 8 Oakley CA, Priestley GC. Density-dependent regulation of skin Hbrobiast glycosaminoglycans in vitro: control by a secreted factor. Ari-h Dernmtol Res 1985: 227: 2f)4-9. 9 Cherry G. Hughes MA. Dawber RPR. Ryan T|. Minoxidil-induced changes in the contraction of cDlliiRen lattices by human skin tibroblasts: a new means of control of excessive clinical scar formation. / Invest Dermatol 1 990: 94: 514, 10 l^vene CI. Possibilities for the therapeutic control of librosis. Hr / Demi«to/1985: 112: i 6 J - 7 1 . 11 Spindler JR. Deaths occurring during clinical studies of topical minoxidil. / Am Atvd Iknnatol 1987: 16: 725-9. 12 Priestley GC. Brown |C. Kffects of griseofulvin on the morphology, growth and metabolism of libroblasts in culture. Hr / Dermatol 1978:99: 245-52. I J Priestley GC. Changes in the growth and metabolism of cells cullured from normal, sclerotic and rheumatoid connective tissue brought about by D-penicil!aminc and by sodium saticyiate. / Invest Dermatol 1980: 74: 41 i-7, 14 Priestley CiC. IJrown )C. Tlffecls of potassium para-aminobenzoate on growth unii macromolecule synthesis on (ibroblasts cultures from normal and sclerodermatotis human skin and rheumatoid synovial cells. / Invest Dermatol 1 979: 72: 161 4. I S Priestley (iC. Kffects of corticostt-roids on the growth and metabolism of libroblasts cultured from human .skin, lir I Dermatol 1 978: 99: 253-61. 16 Priestley CC. Glycosaminoglycans production by cultured skin fibroblasts from the Pasini and Cockayne-Touraine forms of dominant dystrophic epidermolysis bullosa. / Irn-rsl l^ermalol 1991: 9ft: 168 71.
