Minoxidil decreases collagen I deposition and tissue-like contraction in clubfoot-derived cells: a way to improve conservative treatment of relapsed clubfoot?

Connective Tissue Research ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/icts20 Minoxidil decreases collagen I deposition and tissue-like contraction in clubfoot-derived cells: a way to improve conservative treatment of relapsed clubfoot? Jarmila Knitlova , Martina Doubkova , Martin Plencner , David Vondrasek , Adam Eckhardt , Martin Ostadal , Jana Musilkova , Lucie Bacakova & Tomas Novotny To cite this article: Jarmila Knitlova , Martina Doubkova , Martin Plencner , David Vondrasek , Adam Eckhardt , Martin Ostadal , Jana Musilkova , Lucie Bacakova & Tomas Novotny (2020): Minoxidil decreases collagen I deposition and tissue-like contraction in clubfoot-derived cells: a way to improve conservative treatment of relapsed clubfoot?, Connective Tissue Research, DOI: 10.1080/03008207.2020.1816992 To link to this article: https://doi.org/10.1080/03008207.2020.1816992 © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. Published online: 20 Sep 2020. Submit your article to this journal Article views: 788 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=icts20 CONNECTIVE TISSUE RESEARCH https://doi.org/10.1080/03008207.2020.1816992 Minoxidil decreases collagen I deposition and tissue-like contraction in clubfootderived cells: a way to improve conservative treatment of relapsed clubfoot? Jarmila Knitlova a*, Martina Doubkova a,b*, Martin Plencner a, David Vondrasek Martin Ostadal d, Jana Musilkova a, Lucie Bacakova a, and Tomas Novotny b,e a,c , Adam Eckhardt a , a Department of Biomaterials and Tissue Engineering, Institute of Physiology of the Czech Academy of Sciences, Prague, Czech Republic; Second Faculty of Medicine, Charles University, Prague, Czech Republic; cFaculty of Physical Education and Sport, Charles University, Prague, Czech Republic; dDepartment of Orthopaedics, First Faculty of Medicine, Charles University and Na Bulovce Hospital, Prague, Czech Republic; e Department of Orthopaedics, Masaryk Hospital, Usti Nad Labem, Czech Republic b ABSTRACT ARTICLE HISTORY Aim: Clubfoot is a congenital deformity affecting the musculoskeletal system, resulting in contracted and stiff tissue in the medial part of the foot. Minoxidil (MXD) has an inhibitory effect on lysyl hydroxylase, which influences the quality of extracellular matrix crosslinking, and could therefore be used to reduce the stiffness and to improve the flexibility of the tissue. We assessed the in vitro antifibrotic effects of minoxidil on clubfoot-derived cells. Methods: Cell viability and proliferation were quantified by xCELLigence, MTS, and LIVE/DEAD assays. The amount of collagen I deposited into the extracellular matrix was quantified using immunofluorescence with subsequent image segmentation analysis, hydroxyproline assay, and Second Harmonic Generation imaging. Extracellular matrix contraction was studied in a 3D model of cell-populated collagen gel lattices. Results: MXD concentrations of 0.25, 0.5, and 0.75 mM inhibited the cell proliferation in a concentration-dependent manner without causing a cytotoxic effect. Exposure to ≥0.5 mM MXD resulted in a decrease in collagen type I accumulation after 8 and 21 days in culture. Changes in collagen fiber assembly were observed by immunofluorescence microscopy and nonlinear optical microscopy (second harmonic generation). MXD also inhibited the contraction of cell-populated collagen lattices (0.5 mM by 22%; 0.75 mM by 28%). Conclusions: Minoxidil exerts an in vitro inhibitory effect on the cell proliferation, collagen accumulation, and extracellular matrix contraction processes that are associated with clubfoot fibrosis. This study provides important preliminary results demonstrating the potential relevance of MXD for adjuvant pharmacological therapy in standard treatment of relapsed clubfoot. Received 27 February 2020 Accepted 25 August 2020 Introduction Idiopathic talipes equinovarus (clubfoot, CTEV) is one of the most common congenital orthopedic deformities affecting the musculoskeletal system of the lower limb. Two methods are frequently chosen for use in the treatment of clubfoot. The Ponseti method is used worldwide and involves serial manipulations combined with a specific cast application technique, which can be supplemented by percutaneous Achilles tenotomy, followed by foot abduction bracing to prevent relapses1. The less widely used French method consists of daily physiotherapy, stimulation of the muscles around the foot, and temporary foot immobilization with elastic and non-elastic adhesive taping2. Idiopathic clubfeet are mainly treated by the Ponseti method with excellent results3,4; however, the long-term successful outcome of the treatment can be spoiled by KEYWORDS Relapsed clubfoot; congenital idiopathic talipes equinovarus; CTEV; fibrosis; minoxidil; collagen type I relapses5. In the case of relapsed clubfoot, there is also the possibility of surgical therapy, which can be applied as described in the literature6. However, studies of the longterm results of these surgeries have reported several complications, such as stiffness of the ankle, arthritis, muscle weakness, pain, and residual deformity3,7. Although several hypotheses about clubfoot pathogenesis have been proposed, there is still no clear understanding of its cause and its early progression on a genetic, molecular, and cellular level. Current hypotheses cover, e. g., environmental factors, including in utero fetus positioning8 or enteroviral infection9; developmental factors like abnormal muscle insertions10 or vascular deficiencies11; and also genetic factors such as homeobox gene abnormalities12. This range of hypotheses suggests that clubfoot pathology is probably of multifactorial origin. CONTACT Martina Doubkova Martina.Doubkova@fgu.cas.cz Institute of Physiology of the Czech Academy of Sciences, Prague 14220, Czech Republic *J. Knitlova and M. Doubkova contributed equally to this work as first authors. © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-ncnd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. 2 J. KNITLOVA ET AL. Older histological and electron microscopy studies have suggested the existence of fibrotic changes in tissue biopsies extracted from the clubfoot deformity13–15 . More recently published proteomic and molecular studies have revealed changes in the amount of collagen type I, III, VI, and profibrotic cytokines, such as transforming growth factor (TGF-β) and plateletderived growth factor (PDGF) in tissue from clubfoot medial side16–18 and also in m. gastrocnemius19. These studies have therefore confirmed the presence of fibrosis in clubfoot. The fibrotic tissue foci of the clubfoot medial side are located between medial malleolus, sustentaculum tali, and the navicular bone. This area is sometimes called “disc-like” tissue20, due to its extreme stiffness and its rigid nature, which is caused by hyperproduction of extracellular matrix (ECM) components. Accumulation of ECM proteins, with collagens being the most abundant component, is characteristic of fibrosis. Collagen production by highly proliferating fibroblasts, tissue architecture, and contraction ability is regulated by growth factors, especially by TGF-β21. Therefore, collagen formation and growth factors regulating fibroblast behavior would be major targets in the development of an antifibrotic therapy. Research on possible antifibrotic therapies for various fibroproliferative diseases is relatively widespread (for a review, see22). For some therapies, licensed pharmacological agents are already available. In Dupuytren’s disease, for example, local administration of clostridial collagenase into the collagen-rich tissue contractures in fingers contributes to the digestion of excess collagen and to tissue release23. Although some possible targets have been suggested in idiopathic clubfoot, very little attention has been given to the idea of using pharmacological agents in the clubfoot tissue, with the exception of studies by Li et al.17 and Poon et al.24. The application of antifibrotic substances directly into the fibrotic tissue on the clubfoot medial side could reduce tissue contracture, and thus could accelerate and/or improve the outcomes of conservative treatment in relapsed patients. Minoxidil is an FDA-approved drug for treatment of hypertension and hair loss25–27. Its main antifibrotic potential lies in the decrease in lysyl hydroxylase mRNA expression, which effectively reduces the activity of the enzyme28,29. Lysyl hydroxylases act on specific lysine residues in the procollagen alpha chains, which later in the collagen biosynthesis process determine the type and the quantity of crosslinks in newly-formed collagen fibers. For the proper biosynthesis and structural stability of collagen fibers, it is essential to establish collagen crosslinks; however, the quality of mature collagen fibers can vary significantly. Specifically, lysyl hydroxylase 2 isoform activity (for a review, see30) has been found to be responsible for the increased formation of the crosslinks described in various fibrotic tissues, including hypertrophic scars31,32 and Dupuytren’s disease33. In contrast to clostridial collagenase, which is used as a therapeutic agent in Dupuytren’s disease, minoxidil does not degrade collagen fibers already formed in the ECM. This is an important factor for the safety of its medicinal application. Its effects do not entirely prevent abnormal collagen accumulation; however, they may help to lessen it by forming ECM that is more susceptible to degradation. In this way, minoxidil could have a positive impact on the stability and the stiffness of the whole ECM. The aim of our study was to assess the in vitro antifibrotic effects of minoxidil as a candidate for the development of potential adjuvant treatment for relapsed clubfoot. Such antifibrotic treatment may reduce the stiffness and the contraction of the tissue, and thus may improve the outcomes of relapsed clubfoot treatment. Methods Acquisition of biological samples The primary cell cultures used in this study were isolated from tissue samples from 11 patients (8 boys and 3 girls in total; median age of 5 years, SD = 2, Table 1) with idiopathic congenital talipes equinovarus (CTEV, clubfoot). All of these patients had been treated in accordance with the Ponseti method1, as described in Table 1. Tissue samples were acquired during surgeries indicated in the case of a clubfoot relapse; the area of interest was the fibrotic “disc-like” tissue localized between medial malleolus, sustentaculum tali, and the navicular bone on the medial side of the foot. The acquisition of samples and the cell experiments were approved by the ethical committee. Parents and/ or legal guardians of all sample donors gave their informed consent. All procedures were performed in compliance with the Declaration of Helsinki (1934). Isolation, cultivation, and characterization of cells Tissue samples were temporarily stored in cell culture media, and were processed within one hour after surgical excision. The samples (average weight 500 mg) were then cut into very small pieces and were enzymatically digested for at least 3 hours in an enzyme solution (1 ml per 50 mg of tissue) at 37 °C, 5% CO2 on a shaker installed within an incubator. The enzyme solution consisted of clostridial collagenase type 3 (298 U/ml) and neutral protease (0.560 U/ml) (both Worthington Biochemical EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS 3 Table 1. Detailed data of 11 patients (sample donors) with idiopathic congenital clubfoot. Patient Gender 1 M 2 F 3 M 4 M 5 F 6 M 7 M 8 M 9 M 10 F 11 Total M 8 males; 3 females Dimeglio clubfoot classification III III IV II IV III III IV III IV Age during sample acquisition 5 3 1* 5 4 5 5 1* 10 1* Casting (number of plasters) 6 5 10 3 6 5 6 10 5 10 III 1x II; 6x III; 4x IV 7 Median = 5 5 Median = 6 Type of surgery Mc Kay Mc Kay Mc Kay Fowler Mc Kay Mc Kay Mc Kay Mc Kay Mc Kay Mc Kay McKay Mc Kay = 90%; Fowler = 10% Compliance with Previous surgeries abduction bar Achilles tenotomy no Achilles tenotomy no None no Achilles tenotomy yes Achilles tenotomy no None no Achilles tenotomy no None no Achilles tenotomy no Achilles tenotomy no + retenotomy Achilles tenotomy yes 8 in 11 patients 9x no; 2x yes Clubfoot family anamnesis 0 0 0 0 0 1 (mother) 0 0 0 0 Foot (left/ right) L R L R R R R R R R 0 1x L 3x L; 8x R * Patients whose parents were noncompliant, were otherwise unable to maintain proper treatment by application of an abduction bar or failed to attend scheduled examinations. Corporation, USA), and 2.5% antibiotic–antimycotic solution (Sigma-Aldrich, USA) dissolved in Dulbecco’s modified Eagle’s Medium (DMEM; Sigma-Aldrich, USA). The cell suspension was then seeded into poly-L-lysine-coated tissue culture flasks (75 cm2; TPP, Techno Plastic Products, Switzerland) and was cultivated in a 37 °C incubator with 5% CO2 saturation in 20 ml of the DMEM medium, which was supplemented with 20% of fetal bovine serum (FBS; Gibco, Life Technologies, USA) and gentamicin (40 μg/mL, LEK, Slovenia). After approximately 10 days, the cells reached confluence and were passaged. A basic characterization of the primary cell culture was performed in order to identify a subpopulation of cells associated with fibrosis (myofibroblasts, fibrocytes)34. The cell cultures were characterized by a NovoCyte Flow Cytometer (ACEA Biosciences, USA) for the presence of CD90, CD45, and CD34 markers, collagen type I (Col1), smooth muscle α-actin (αSMA), and myosin heavy chain (SMM MYH11). To discriminate the cell types more precisely, the cells were also immunofluorescently stained for Col1, αSMA, SMM, desmin, vimentin, and von Willebrand factor35–37. In order to evaluate the time required for cell spreading and the cell population doubling time, the cells in the 1st passage were fixed in cold (−20 °C) 70% ethanol for 10 min at room temperature (RT), were washed with Phosphate-Buffered Saline (PBS; Sigma-Aldrich Co., USA), and were fluorescently stained for 1 hour (RT, in the dark) with Hoechst #33,258 (5 µg/ml in PBS; 94,403, Sigma-Aldrich, USA) to visualize the cell nuclei, and with Texas Red C2 Maleimide (20 ng/ml in PBS; Sigma-Aldrich Co., USA; red stain) to visualize the cell membrane and the cytoplasmic proteins. The fluorescence signal was evaluated by an Olympus IX51 epifluorescence microscope with a DP74 camera (both Olympus Corp., Japan). The microphotographs that were obtained (10 per well) were analyzed in ImageJ FIJI software (v.1.52 n; open-source)38. The cell spreading area was measured during the first 24 hours of culture (after 1, 3, 5, and 24 hours). The doubling time was determined in the exponential phase of cell growth (i.e. on day 1 and 3 after seeding) according to the following equation: DT ¼ log2 t logNt t0 logNt0 Time intervals after seeding are represented by t0 (earlier interval) and by t (later interval), while Nt0 and Nt represent the number of cells on the sample at these intervals. Preparation of the minoxidil solution Minoxidil (MXD) powder (M4145, Sigma-Aldrich, USA) was dissolved in 96% ethanol (EtOH) to a 119 mM stock solution, was aliquoted, and was stored frozen at −20 °C. Working concentrations were diluted immediately prior to each experiment in DMEM supplemented with 10% FBS. Cell cultivation in the presence of minoxidil The cells for the experiments were seeded into 24-well plates (TPP, Techno Plastic Products, Switzerland) in 1 ml of DMEM with 20% FBS per well. For the Live/ Dead viability/cytotoxicity assay, the cells were seeded at an initial density of 2 × 104 cells per well (ca. 10,730 cells/cm2). For the MTS assay, immunofluorescence staining of collagen, the hydroxyproline assay, qPCR, 4 J. KNITLOVA ET AL. and nonlinear optical microscopy (SHG imaging), the cells were seeded at an initial density of 3 × 104 cells per well (ca. 16,094 cells/cm2). The cell number and viability before seeding were determined using a Vi-CELL XR Analyzer (Beckman Coulter, USA). After 24 hours, the cells were supplied with fresh DMEM with 10% FBS and either MXD (final concentrations of 0 mM, 0.25 mM, 0.5 mM and 0.75 mM, where 0 mM is regarded as untreated cells and is marked as Control), or corresponding volume concentrations of 96% ethanol (vehicle equivalent control). Addition of MXD was considered as the start day (day 0) of treatment for all experiments. In experiments focused on the production and deposition of collagen type I, all samples were also supplemented with L-ascorbic acid-2-phosphate (Asc2P; final concentration of 50 µg/ml). The medium was replaced by the same amount of fresh DMEM with 10% FBS with MXD or vehicle control and Asc-2P on each 5th day of culture. The cell seeding numbers and the cultivation conditions for experiments in the xCELLigence system and in the collagen gel lattices differ from all other experiments and are described in detail in the appropriate method sections. Cytotoxicity testing For the cytotoxicity tests, the cells from the first passage were grown to 80% confluence (ca. 5 day) before seeding for a given test. xCELLigence—real-time cellular analysis (RTCA) The xCELLigence system (ACEA Biosciences, USA) was used to determine the nontoxic concentration of minoxidil. This system allows real-time monitoring of cell attachment, spreading, and changes in cell numbers, reflecting the cell viability in response to the addition of MXD, by measuring the electrical impedance. This impedance is generated by cells contacting the gold microelectrode arrays incorporated on the bottom of 96-well plates. The changes in the cell-microelectrode contact, i.e. the changes in cell coverage of the well bottom, are reflected by the relative change in the impedance in time, which is calculated by RTCA software (ACEA Biosciences, USA) as the Cell Index (CI). For monitoring CI, the cells were seeded into a 96-well plate (E-Plate 96; ACEA Biosciences, USA) at a density of 3.5 × 103 cells per well (ca. 10,870 cells/cm2), and in 200 µl of DMEM medium supplemented with 20% FBS. The cells were left to adhere for 30 minutes before the plate was mounted into the xCELLigence system and before monitoring began. After 24 hours of monitoring, the cells were treated for 48 hours either with 0, 0.25, 0.5, 1, and 2 mM MXD or with corresponding concentrations of 96% ethanol (EtOH vehicle controls) in 200 µl of DMEM with 10% FBS. After this period, the cells were supplied with fresh DMEM with 20% FBS and were monitored to assess cell recovery from MXD treatment. Since monitoring the CI (which reflects only the coverage of the well bottom with cells) cannot distinguish between the size of the cell spreading area and the number of cells, optical monitoring was also included in the analysis. A microscopic evaluation of the cell morphology was performed in two time intervals, i.e. immediately after the removal of MXD (72 hours of culture) and after an additional 92-hour recovery period (164 hours of culture). MTS assay The cell metabolic activity in the presence of MXD was assessed using an MTS colorimetric assay (G3582, CellTiter 96 AQueous One Solution Cell Proliferation Assay, MTS; Promega Corporation, USA). This assay offers quantification of the activity of mitochondrial enzymes, and it therefore also provides indirect information on cell proliferation and viability. The cells for the experiments were seeded into 24well plates at an initial density of 3 × 104 cells per well (16,094 cells/cm2), and were cultivated and treated as described above (i.e., in the section “Cell cultivation in the presence of minoxidil”). The MTS assay was performed according to the manufacturer’s protocol on day 1 and on day 4 of the treatment. The reduction in MTS was measured using a Synergy HT Microplate Reader spectrophotometer (BioTek Instruments, USA) at 490 nm (with a reference wavelength at 650 nm). The metabolic activity of the MXD-treated samples and the samples treated with EtOH vehicle control was compared with the metabolic activity of untreated control cells. Live/dead assay In order to further test the effect of MXD, the cell viability and proliferation was also evaluated by fluorescence staining on day 1, day 4, and day 8 after the start of MXD treatment, using the LIVE/DEAD Viability/Cytotoxicity kit for mammalian cells (L3224, ThermoFisher Scientific, USA). This method distinguishes between living and dead cells by simultaneous staining in green, for living cells, and in red, for dead cells. The cells were seeded into 24-well plates at an initial density of 2 × 104 cells per well (ca. 10,730 cells/ cm2), and were cultivated and treated as described above (i.e., in the section “Cell cultivation in the presence of minoxidil”) for 8 days. The fluorescence signal was visualized by an Olympus IX51 epifluorescence EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS microscope with a DP74 camera (both Olympus Corp., Japan). The microphotographs that were obtained (10 per well) were analyzed in ImageJ FIJI software (v.1.52 n; open-source)38 by counting the living (green) and dead (red) cells. Viability was then determined as the percentage of the living cells in the culture. The cell numbers were expressed as cell population densities per cm2. Collagen production and deposition Immunofluorescence of collagen type I—short-term collagen deposition Cells for immunofluorescence staining were seeded as described above (i.e., in the section “Cell Cultivation in the presence of minoxidil”) on microscopic glass coverslips inserted into a 24-well plate and cultivated for 8 days. After that, the cells were fixed in cold (−20 ° C) 70% ethanol for 10 min at room temperature (RT), were washed with PBS, and were then treated with 1% bovine serum albumin and 0.1% Triton X-100 in PBS for 30 min at RT in order to block the nonspecific binding sites for antibodies. Subsequently, a primary mouse monoclonal anti-collagen type I antibody (1:200 in PBS; C2456, Sigma-Aldrich, USA), which recognizes the native conformation of collagen I, was applied overnight at 4 °C. After washing with PBS, a secondary antibody Alexa Fluor 488-conjugated F(ab’)2 fragment of goat anti-mouse IgG (1:400 in PBS; A1107, Invitrogen, USA) was added for 1 hour at RT together with Hoechst #33,258 (5 µg/ml in PBS; 94,403, SigmaAldrich, USA) to stain the cell nuclei. The fluorescence signal was visualized by an Olympus IX51 epifluorescence microscope with a DP74 camera (both Olympus Corp., Japan). The microphotographs that were obtained (10 per well) were analyzed by ImageJ FIJI software (v.1.52 n; open-source)38, using a pixel-based segmentation tool. The amount of deposited collagen type I was expressed as a percentage of the stained collagen area in the whole area of a single image, and was normalized to the number of cells on that image. The data are presented as a percentage of fluorescently stained collagen area per cell. Hydroxyproline assay—long-term collagen deposition Cells were seeded as described above and were cultivated for 21 days. First, the cells were lysed in RIPA buffer (R0278, Sigma-Aldrich, USA), and a Bradford assay was performed to measure the total protein content. Then the amount of hydroxyproline was determined by the Hydroxyproline Colorimetric Assay Kit (K555, BioVision, USA), according to the 5 manufacturer’s instructions, and it was measured using a VersaMax Absorbance Microplate Reader (Molecular Devices, USA) at 560 nm. The amount of hydroxyproline in a sample is widely used as a direct measure of native collagen. The data are presented as the relative concentration of hydroxyproline to the total protein content per sample. Second harmonic generation imaging A second harmonic generation (SHG) signal was obtained in a confocal microscope in order to determine the presence of properly structured collagen type I fibers. The cells were seeded on microscopic glass coverslips inserted in a 24-well plate and cultivated as described above. On day 21 of cultivation, the coverslips with cells were taken out of the plate, and the cells were visualized using a 63x water immersion objective (HC PL APO CS 2 63x/1.20 Water), mounted on a Leica DMi8 inverted microscope with a Leica TCS SP8 X confocal unit (Leica Microsystems, Germany). A Chameleon Discovery TPC pulsed femtosecond laser (Coherent Inc., USA; 860 nm tunable output, 80 Mhz, 1.6 W) was used to generate SHG and autofluorescence signals. Leica non-descanned HyD detectors and a 430/ 24 band-pass filter were used for collecting SHG signals, and Leica non-descanned HyD detectors and a 610/75 band-pass filter were used for collecting autofluorescence signals. Real-time PCR Cells were seeded as described above. Collagen type I expression was measured using real-time PCR (qPCR). On day 3 and day 7 of MXD treatment, the cell RNA was isolated using the Total RNA Purification Micro Kit Plus (48,500, Norgen Biotek, Canada) according to the manufacturer’s instructions. RNA in a concentration of 300ng/µl was used for reverse transcription into cDNA by the ProtoScript First Strand cDNA Synthesis Kit (E6300, New England BioLabs, USA), using the oligo-dT primers that were supplied. The reaction ran in a T-Personal Thermocycler (Biometra, Germany). The real-time PCR was performed using 5xHOT FIREPol Probe qPCR Mix Plus (ROX) (08–14-00008, Solis BioDyne, Estonia) and TaqMan Gene Expression Assays (Life Technologies, USA) labeled with the FAM reporter dye specific to target human gene collagen type I alpha chain (COL1A1; Hs00164004_m1, amplicon length 66 bp), and a reference gene glyceraldehyde3-phosphate dehydrogenase (GAPDH; Hs02758991_g1, amplicon length 93 bp). A final reaction was conducted in the Viia 7 Real-time PCR System (Applied Biosystems, USA) in a total reaction volume of 20 μl and the following cycle parameters: incubation at 95 °C 6 J. KNITLOVA ET AL. Results (10 min), followed by 40 cycles of 95 °C (15 sec) and 60 °C (1 min). The relative mRNA expression was quantified by the 2−ΔΔCt method. The data were normalized according to the gene expression in the control sample (Ctrl, calibrator). Clubfoot primary cell culture characterization First we performed a basic characterization of the primary cell culture with the aim to identify cell subpopulations that are associated with fibrosis (myofibroblasts, fibrocytes)34. The culture contained mainly fibroblasts (96%) with an admixture of myofibroblasts and fibrocytes (<3%), vascular smooth muscle cells (<0.1%) (Table 2), while no endothelial cells (identified by von Willebrand factor positive staining) were present in the culture. Rarely, adipocytes were identified on the basis of their morphology. However, a pure fibroblast culture was obtained in the fourth passage. Therefore only the first, the second, or the third passage was used for the experiments. We refer to these cells as fibroblast-like. The time required for cell spreading and the cell population doubling time were evaluated in the first passage. The cell spreading area was increasing gradually from 1 hour to 24 hours after seeding, when it reached its maximum (ca 1940 μm2). The estimated doubling time of 30.4 ± 1.08 hours (mean ± SEM) was determined in the exponential phase of cell growth (from days 1 to 3 after seeding). Cell-mediated contraction of collagen gel lattices The CytoSelect 24-Well Cell Contraction Assay Kitfloating matrix model (CBA-5020, Cell Biolabs, USA) was used to study the effect of MXD on the cytoskeleton of fibroblast-like cells, and on the contraction of their ECM in vitro. Collagen gel lattices were prepared according to the manufacturer’s instructions in a 24well anti-adhesive plate with 385 × 103 cells per 0.5 ml of collagen gel per well, and were incubated in DMEM with 10% FBS and either MXD (0-1 mM) or corresponding concentrations of 96% ethanol (EtOH vehicle control) with daily changes of 50% of the media for a freshly mixed solution. To exclude the possibility of changes in the pure collagen gel lattices (without addition of cells) due to MXD treatment, all MXD concentrations were also tested with unpopulated lattices. Unpopulated lattices with no treatment served as negative controls where no change should occur. After 2 days of cultivation, the changes in the sizes of the gel lattice areas were photographed and were measured using ImageJ FIJI software (v.1.52 n; open-source)38. An MTS test was carried out at the end of the experiment in order to verify the viability of the cells inside the collagen lattices. Is minoxidil a safe agent for fibroblast-like cells derived from fibrotic relapsed clubfoot tissue? We started by testing the maximum final concentration of MXD that can be added to clubfoot fibroblast-like cells without causing a direct cytotoxic effect. The progression of cell spreading and proliferation throughout the experiment was estimated by the xCELLigence system. Electrical monitoring and also optical monitoring revealed that with an increasing concentration of MXD, the cell spreading and proliferation were suppressed dramatically (Figure 1). The control cells were also grown in DMEM with 10% FBS during the period of MXD treatment. This serum supplementation in the media proved to be enough to support the growth and the viability of the cells, and at the same time, no protective effect of serum against MXD was apparent. Some cells became rounded in the presence of 1 mM MXD, and even more cells were rounded in 2 mM MXD, as Statistical analysis A statistical analysis was performed in SigmaStat 4.0 (Systat Software Inc., USA). Data from all experiments were analyzed either by ANOVA with Dunnett’s post hoc test or by Kruskal-Wallis ANOVA on Ranks with Dunnett’s post hoc test (specified in the respective figure caption). P-values < 0.05 were considered statistically significant. All graphs were created using GraphPad Prism 6 (GraphPad Software, USA). Table 2. Characterization of the cells derived from the relapsed clubfoot, with emphasis on the presence of cell types associated with fibrosis. Analyzed markers Cell type Fibroblasts Myofibroblasts Fibrocytes Vascular smooth muscle cells CD34 CD45 CD90 Col1 αSMA SMM MYH 11 vimentin desmin % in culture + NT + NT + + + NT + + + dim+ + + + + + + - + ~ 96% < 3% < 3% < 0.1% * ± sign = positivity/negativity of markers for the cell to be identified as a particular type. NT = marker not tested. dim+ = marker is at least dim positive. EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS 7 Figure 1. (a). Real-time analysis of the effect of minoxidil (MXD) on the growth of fibroblast-like cells using the xCELLigence system. (b) Morphology of native cells. The cells were seeded 24 hours before the start of the minoxidil (MXD) treatment and were observed immediately after the MXD treatment period (corresponding to a 72-hour culture time interval; 1b first row) and at the end of the recovery period, which followed after the removal of MXD (corresponding to a 164-hour culture time interval; 1b second row). The effect of EtOH vehicle controls was in general negligible. EtOH max indicates the effect of EtOH vehicle control, corresponding to the maximum MXD concentration (2 mM), on the cells. Pictures of native cells, Olympus IX51 microscope, DP71 digital camera, obj. 20x, bar = 50 µm. observed under a phase contrast microscope (Figure 1(b), first row). However, this effect was reversible, as all cells recovered during subsequent cultivation in a fresh DMEM medium with 20% FBS (without MXD), and reached confluence eventually (Figure 1(b), second row). Nevertheless, we excluded the 2 mM MXD concentration from further experiments. Instead, we added a concentration of 0.75 mM MXD to better discriminate the gradual effects of MXD on the cells. In order to determine the possible cytotoxicity of the MXD concentrations that were applied, we performed a standard MTS metabolic assay (Figure 2), and we interpreted the results according to the International Standard for testing in vitro cytotoxicity (International Figure 2. MTS assay. Dose-response relationship for the effect of minoxidil (MXD) concentration on cell proliferation and viability, expressed in terms of cell metabolic activity after 1 and 4 days of treatment. Data presented as a percentage value of the metabolic activity (A490) of cells treated with MXD or cells treated with the ethanol (EtOH) vehicle control, in comparison to the untreated control (regarded as the 100% value). Mean ± SEM. Day 1: One way ANOVA, Dunnett’s test; statistical significance compared to the control: *** p < 0.001. However, the decrease in metabolic activity on day 1 did not reach the limit stated by the ISO standard, and it is therefore considered not to be biologically significant. Day 4: Kruskal-Wallis ANOVA on Ranks, Dunnett’s test; statistical significance compared to the untreated control: ** p < 0.01 (specifically: MXD 0.75 mM, p = 0.002; MXD 1 mM, p ≤ 0.009). 8 J. KNITLOVA ET AL. Figure 3. LIVE/DEAD assay. The effect of minoxidil (MXD) concentration on (a) the cell proliferation rate and (b) cell viability after 1, 4, and 8 days of MXD treatment. The cells were seeded 24 hours before the start of the treatment at an initial density of ca. 10,730 cells/cm2. Proliferation data presented as mean ± SEM. Day 1 and Day 4: One way ANOVA, Dunnett’s test; without statistical significance. Day 8: One Way ANOVA, Dunnett’s test; statistical significance compared to the untreated control (Control): * p < 0.05, ** p < 0.01 (specifically: MXD 0.50 mM, p = 0.01; MXD 0.75 mM, p = 0.003). Viability data presented as the median of the percentage of living cells to the total cell number; the whiskers depict maximum and minimum values. One way ANOVA; without statistical significance. Organization for Standardization, 2009)39, as an assessment of cell viability (measured by the cell metabolic activity) after 24 hours of cultivation with the tested substance. According to the ISO standard, an agent that reduces cell viability to less than 70% of an untreated control sample is considered to have cytotoxic potential. In the case of MTS, this is indicated by an appropriate change in the measured absorbance at 490 nm (A490). The MTS assay was applied to the cells not only on day 1, but also on day 4 of MXD treatment, in order to acquire more complex information about the possible cytotoxic effects. On day 1, none of the MXD concentrations reduced the cell metabolic activity to less than 70% of the control, although both 0.75 mM MXD and 1 mM MXD showed a statistically significant decrease in A490 compared to the control. In the case of 0.75 mM MXD, the decrease was to 83.5% of the control value (p < 0.001), and in the case of 1 mM MXD, the decrease was to 85.7% of the control value (p < 0.001) (Figure 2). However, these decreases did not reach the limit stated by the ISO standard, and they can therefore be considered biologically non-significant. Similarly, on day 4, statistically significant decreases in A490 compared to the control were detected only for MXD 0.75 mM and 1 mM. The cell viability decreased to 53.9% of the control value in the case of 0.75 mM MXD (p = 0.002), and decreased even to 36% of the control value in the case of 1 mM MXD (p ≤ 0.009). The metabolic activity of cells treated with 0.5 mM MXD dropped to 71.5% of the control value, but the statistical test used for the data analysis did not classify this decrease as statistically significant (Figure 2). In the samples treated with ethanol (EtOH) vehicle equivalents to the tested MXD concentrations, the cell metabolic activity never dropped below 82%, and this drop showed no statistical significance. Although the ISO standard does not specifically state an acceptable reduction in MTS-measured viability for 4 days of cultivation, the reduction after treatment with 1 mM MXD to 36% of the control was so enormous that we excluded this concentration from further experiments. The MTS assay, i.e. a test of the activity of mitochondrial enzymes, is stated to indirectly assess cell viability and, at the same time, to reflect cell proliferation. However, this assay does not distinguish whether the cells are dying or are only proliferating more slowly. We therefore performed a third type of cytotoxic test, the LIVE/DEAD assay, for concentrations of 0–0.75 mM MXD. In all samples, including those treated with the tested MXD concentrations and the corresponding EtOH vehicle controls, the number of viable cells per sample never dropped below 98%. However, the growth in cell numbers per cm2 between the experimental intervals declined slowly over time (Figure 3). It even appeared that cells treated with 0.50 mM MXD and with 0.75 mM MXD almost reached the stationary phase of cell growth during the time period between day 4 and day 8. These results clearly indicate that minoxidil has an antiproliferative effect, rather than a cytotoxic effect, on fibroblast-like cells. The influence of the EtOH vehicle equivalent on cell proliferation and viability in these experiments was negligible in comparison with the untreated Controls. The cytotoxicity and proliferation tests revealed that MXD reduces cell proliferation in a dose-dependent manner, while it does not have a direct cytotoxic effect. On the basis of these results, we chose three concentrations for further testing of their antifibrotic activity (0.25, 0.5, and 0.75 mM MXD). These concentrations are safe for the cells and do not inhibit cell proliferation completely. In the following experiments, the antiproliferative effect of minoxidil was taken into account and, when applicable, the results that were obtained were normalized to the number of cells on the sample EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS or in the area of interest, or to the total protein content in each sample (depending on the type of test). Does minoxidil reduce the amount of collagen type I in the extracellular matrix of relapsed clubfootderived cells? We investigated the assumed antifibrotic effect of minoxidil by measuring collagen type I deposition into the ECM after 8 days of cultivation. An analysis of microphotographs from fluorescence microscopy showed that a concentration of 0.25 mM MXD had no significant effect on collagen deposition (median = 43.9x10−3, IQR = 0.024, p = 0.135). However, both of the higher concentrations significantly reduced the collagen content: 0.5 mM MXD (median = 14 x10−3, IQR = 0.009, 9 p < 0.001) and 0.75 mM MXD (median = 9 x10−3, IQR = 0.009, p < 0.001) in comparison with the control (median = 79x10−3, IQR = 0.038) (Figure 4). The collagen I production pathway is naturally regulated at various levels on its way from mRNA expression via procollagen secretion to the formation of mature collagen fibers, all in reaction to the current cell state and condition. We investigated whether minoxidil treatment can influence procollagen type I expression in reaction to the reduced cell population density or in reaction to the reduced amount of collagen type I in the ECM. We measured the level of procollagen type 1 alpha-1 chain (a procollagen type I structural component, encoded by COL1A1 gene) mRNA by real-time PCR (Figure 5). We found that there was no concentration-dependent effect of Figure 4. The short-term collagen deposition is lower in the minoxidil (MXD)-treated samples after 8 days of treatment, as represented by the smaller area covered by fluorescently stained collagen. (a) Changes in collagen type I deposition visualized by immunofluorescence staining. The amount of collagen type I (green signal, upper panel) after treatment with increasing concentrations of MXD decreased gradually, but not in proportion to the lower number of cells in culture (blue signal, lower panel, merged with the upper panel). Olympus IX51 microscope, obj. 10x, bar = 200 µm. (b) Percentage of the fluorescently stained collagen area expressed per single cell. Image analysis of microphotographs obtained via fluorescence microscopy. Data presented as a box plot with a median line, the outer edges represent the 1st and the 3rd quartile, the whiskers depict maximum and minimum values. Kruskal-Wallis ANOVA on Ranks, Dunnett’s test; statistical significance compared to the untreated Control: *** p < 0.001. 10 J. KNITLOVA ET AL. Figure 5. COL1A1 gene expression after 3 and 7 days of minoxidil (MXD) treatment. The data are normalized according to the gene expression in the untreated control sample (Control, calibrator; represented by the dashed line), and are presented as a box plot with a median line, the outer edges represent the 1st and the 3rd quartile, the whiskers depict maximum and minimum values. Kruskal-Wallis ANOVA on Ranks, Dunnett’s test; without statistical significance. minoxidil on collagen type 1 alpha 1 chain mRNA expression after 3 or 7 days of cultivation in the presence of MXD. This result indicates that the inhibitory effect on collagen is realized downstream from the COL1A1 expression. The influence of the vehicle equivalent (EtOH) on the cells in these experiments was not statistically significant (data not shown). To assess the long-term effect of minoxidil treatment on collagen type I deposition, we measured the content of collagen type I in ECM after a 3-week cultivation period by means of a hydroxyproline assay. As in the case of shortterm cultivation, only 0.5 mM and 0.75 mM MXD had a significant effect on collagen deposition into the ECM: 0.5 mM MXD (median = 1.04x10−3, IQR = 0.46x10−3, p ≤ 0.004), 0.75 mM MXD (median = 0.848x10−3, IQR = 0.234x10−3, p ≤ 0.004) in comparison with the untreated control (median = 2.50x10−3, IQR = 0.89x10−3) (Figure 6(a)). After a 3-week cultivation period, the collagen fibers were so abundant that it was not possible to visualize them by conventional fluorescence microscopy. We therefore used second harmonic generation (SHG) imaging, i.e. label-free confocal microscopy, to acquire qualitative information on the collagen type I molecular assembly. The representative pictures in Figure 6(b) depict the decrease in the SHG signal generated by properly assembled collagen fibers with increasing concentration of minoxidil, while they also illustrate the reduction in cell numbers after long-term cultivation. The LIVE/DEAD assay performed in parallel revealed that there is no cytotoxicity, even within 3 weeks of cultivation and repeated treatment with MXD in all concentrations (data not shown here). Can minoxidil reduce the tissue-like contraction of the extracellular matrix caused by relapsed clubfoot-derived cells? In order to explore the effect of minoxidil on cellmediated contraction of ECM, we used 3D collagen gel lattices as an in vitro model of ECM. The area of the gel seeded with cells without the addition of MXD (control) was the most contracted, and was therefore regarded as a 100% value of the area shrinkage. The percentage values of the MXD-treated samples express by how many percent the gels are less contracted in comparison to the control (Figure 7(a)). No significant changes were observed in cell-populated gel lattice areas, when only the vehicle equivalent (EtOH) was added (data not shown here). Gels without cells, with or without MXD treatment, did not exhibit any shrinkage at all during the experiments. A concentration of 0.25 mM MXD (SEM = 0.02, p = 0.067) had no significant effect on the contraction of cell-populated collagen gel lattices, as the shrinkage of the gel area was only 9% less than the shrinkage of the untreated control. However, the gel lattices incubated in 0.5 mM, 0.75 mM, and 1 mM MXD were significantly less contracted. The contraction was by 22%, by 28% and by 35% less than in the control (SEM = 0.27, 0.28, and 0.32, respectively, p ≤ 0.001) (Figure 7(a)). The 3D model of the collagen gel lattices resembles the environment of connective tissue ECM more closely than the situation when cells are grown in a monolayer. We repeated the MTS tests for cells in a gel with a concentration of 0.25–1 mM MXD because of the different culture conditions in these experiments. MTS confirmed that the cells EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS 11 Figure 6. (a) The long-term collagen type I deposition after 3 weeks is lower in the minoxidil(MXD)-treated samples than in the untreated control, as represented by a lower relative hydroxyproline content in these samples. Data presented as a box plot with a median line, the outer edges represent the 1st and the 3rd quartile, the whiskers depict the maximum and minimum values. KruskalWallis ANOVA on Ranks, Dunnett’s test; statistical significance compared to the untreated control: ** p < 0.01, *** p < 0.001 (specifically: MXD 0.5 mM, p < 0.001; MXD 0.75 mM, p = 0.004). (b) SHG imaging of collagen type I. The decrease in the SHG signal in clubfoot fibroblast-like cells treated with minoxidil demonstrates its effect on the correct structural assembly of collagen type I (upper panel, green signal). The persistent effect of minoxidil concentration-dependent suppression of cell proliferation is still apparent after 3 weeks of cultivation by the lower number of cells in the samples (autofluorescence of cells, lower panel, red signal). Leica DMi8 confocal microscope, obj. 63x water immersion, bar = 50 µm. thrive in these conditions, as the viability of all samples did not drop below 95% of the control (Figure 7(b)). The influence of minoxidil on the cell-mediated contraction of collagen gel lattices populated by fibroblastlike cells is illustrated in representative pictures taken after 2 days of treatment (Figure 7(c)). Discussion Minoxidil (MXD) has previously been tested in vitro in human dermal fibroblasts and in keratinocytes,40–42 and also in vivo in mice with induced pulmonary fibrosis43. However, its antifibrotic effect on clubfoot cells, derived directly from the contracted part of the tissue, has been assessed for the first time in this study. 12 J. KNITLOVA ET AL. Figure 7. (a) The contraction of collagen gel lattices by clubfoot fibroblast-like cells after 2 days of minoxidil (MXD) treatment is presented as a percentage of the decrease in the gel lattice contraction in comparison with the control sample. Mean ± SEM. One Way ANOVA, Dunnett’s test; statistical significance in comparison with the untreated control: *** p < 0.001. (b) Effect of minoxidil (MXD) concentration on the proliferation and viability of cells in collagen lattices, expressed in terms of the cell metabolic activity after 2 days of treatment (MTS assay). Data presented as a percentage value of the metabolic activity (A490) of cells treated with MXD to the untreated control cells (regarded as 100%). Mean ± SEM. One Way ANOVA, Dunnett’s test; without statistical significance. (c) Pictures representing the changes in collagen gel areas after 2 days of MXD treatment. Admittedly, there are some limitations to our investigation that should be addressed here. As all our experiments were performed in vitro, it is uncertain how well our findings would be translated into in vivo conditions. The extracellular matrix formed in vitro is produced by cells which may lack important factors, e.g. those coming from blood vessels or secreted by other cell types in the surrounding tissue. An exact simulation of all these signals in vitro, however, is almost impossible to achieve. Although fibrotic markers could be modeled by adding profibrotic cytokines (e.g. TGF-β, PDGF) to the cell culture, it still would not fully represent the complex interplay among soluble factors and cells in vivo. It is also likely that cells originating from relapsed clubfoot tissue lose their fibrotic characteristics with prolonged in vitro cultivation. Taking this into consideration, we performed all experiments only within the first, the second, or the third passage of cells isolated from each tissue sample. For example, we found that the amount of myofibroblasts in culture declined, and this cell type became almost absent in the third passage in vitro. We are also aware that the selected effective concentrations of MXD relate to the in vitro situation. For potential adjuvant treatment of relapsed clubfoot, local delivery directly into the area of medial side contracted tissue seems to be the most logical administration route. However, in the literature there are no data on MXD pharmacokinetics in connective tissue for administration of this type. In vitro testing is the first step in analyzing the antifibrotic actions of MXD in clubfoot, and we consider that animal models of fibrosis could be introduced in the not-so-distant future. Two knockout mouse models of clubfoot were recently established.44,45 Further research is necessary in order to achieve the desired sustainable antifibrotic effect in vivo, probably with MXD bound to a biomacromolecular carrier with affinity to the affected tissue. Such a targeted carrier can increase the efficacy of the treatment, can simultaneously reduce possible side-effects, and can lessen the immunogenicity of the drug (for a review, see46). Another limitation of our study is the relatively small number of patients used as tissue sample donors (n = 11; Table 1). Operative treatment of idiopathic relapsed clubfoot is very rare, so there are limited opportunities to obtain tissue samples from this exact region of the foot. Nevertheless, samples obtained from operative treatment have provided a good opportunity to study the stiff tissue of clubfoot. We have shown in this study that MXD is not cytotoxic to clubfoot-derived fibroblast-like cells, but that it lowers their proliferation in a dose-dependent manner. This finding is in agreement with the previous work of Pinnell and Murad29,40. The administration of 1 mM MXD in Pinnell and Murad’s study caused almost complete cessation of skin fibroblast growth in subconfluent cultures, but no cytotoxic effect was reported. Tissue fibrosis is associated with hyperproliferation of fibroblasts, and a moderate decline in cell proliferation is therefore even desirable. EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS The main consequence of fibrotic processes in ECM is a massive net accumulation of collagen. Antifibrotic strategies involve inhibiting either growth factors or cytokines, driving the cells into contractile phenotype, which is accompanied by massive collagen production (for a review, see47), or targeting the biosynthetic pathway of collagen itself (for a review, see48). To the best of our knowledge, only one study to date has examined the use of antifibrotics in clubfoot-derived cells in vitro17. Li et al.17 also reported decreased cell proliferation in addition to decreased collagen type III mRNA expression in clubfoot-derived cells when they were treated with neutralizing antibodies to TGF-β and PDGF. However, Li et al.17 do not provide data on collagen deposition into the matrix. We took a different strategy, and targeted the post-translational modifications/processing which occurs during collagen biosynthesis. We observed that concentrations of 0.5 mM and 0.75 mM MXD lowered the deposition and the structural maturation of collagen type I into ECM in both studied time intervals (after 8 days and after 3 weeks of MXD treatment). This took place without the procollagen type I alpha-1 chain mRNA (encoded by COL1A1 gene) expression being affected. The expression of procollagen type I alpha-2 chain mRNA (encoded by COL1A2 gene), i.e. the other chain needed for forming a collagen triple-helix, was also not affected by MXD in human dermal fibroblasts and cells from keloid scars in vitro42. However, this may not be the case for all types of cells and tissues. In contradiction to our findings for clubfoot, the expression of COL1A1 was reduced by MXD in vivo in the lung tissue of mice with bleomycin-induced fibrosis43. The exact effect of MXD on the mRNA expression of genes encoding procollagen chains should therefore be further tested on a larger number of relapsed clubfoot samples. Minoxidil reduces the levels of individual lysyl hydroxylase isoforms to a varying extent (LH1≫LH2>LH3).42 To reduce fibrosis, it would be advantageous to inhibit predominantly the hydroxylation of telopeptidic lysin residues, which is catalyzed by lysyl hydroxylase isoform 2b (LH2b). However, no specific inhibitor of LH2 is currently available, despite efforts that are being made to identify such compounds.49 Another important posttranslational modification of collagen, which takes place after hydroxylation, is the glycosylation of hydroxylysyl residues. It is possible that by limiting the supply of hydroxylysin residues, minoxidil indirectly inhibits glycosylation of collagen and thus modulates collagen fibrillogenesis at different in its biosynthesis. It has been shown that the LH3 isoform of lysyl hydroxylase also possesses a modest glucosyltransferase and galactosyltransferase activity.50 13 However the LH3 isoform is affected only to a limited extent by minoxidil.42 Interestingly, the ability of MXD to reduce collagen deposition was also observed in bleomycin-induced lung fibrosis in mice, and was suggested to be at least partially linked to the downregulation of TGF-β/Smad3 signaling.43 As Merl-Pham et al.51 have recently shown in their study on human lung fibrosis, TGF-β can also modulate collagen glycosylation. It is probable that the decrease in collagen deposition in MXD-treated samples observed in the present study is due to the effect of MXD on some steps in the posttranslational modification of collagen. The observed changes in the diameters of cell-populated collagen gel lattices indicated that minoxidil can also modulate cell-matrix mechanical forces, and thus it can lessen the contraction in a 3D tissue-like model of ECM. We speculate that the inhibition of collagen gel contraction is not due to reduced collagen deposition or decreasing cell numbers. Shrinkage of cell-populated collagen gels was observable within hours after it had been prepared, but inhibition of these processes by MXD does not occur so quickly. Dallon and Ehrlich52 reviewed the possible mechanism of collagen lattice contraction, and pointed out that after seeding within the gel, the cells attach and elongate. This process generates forces that translocate the collagen fibrils and cause shrinkage of the collagen gel. A microscopic examination of the cells in specific intervals (72 and 164 hours of culture) during real-time xCELLigence monitoring showed that MXD affects the cell morphology in a dose-dependent manner (Figure 1(b)). With increasing MXD concentration, the cells gradually changed their morphology from elongated well-spread cells with long processes to less-elongated smaller cells, and often to completely rounded cells at 2 mM MXD (Figure 1(b)). It is probable that the effect of MXD on collagen gel contraction is associated with changes in cell morphology. Another possible explanation comes from the work of Shao et al.43, who proved that minoxidil inhibits the signal transduction of TGF-β/Smad3 and reduces collagen formation, thus reducing the lung fibrosis previously induced by bleomycin in mice. The association between TGF-β and increased cell contractility is well known, and its regulation by minoxidil in clubfoot-derived cells would be of great importance. It has also been reported by Priestley et al.41 that minoxidil decreased the secretion of glysosaminoglycans (GAG) in skin fibroblasts, an important component of ECM in direct interaction with collagen. This subsequently resulted in decreased contractility of collagen gel lattices. 14 J. KNITLOVA ET AL. Conclusion Minoxidil inhibits the cell proliferation of clubfoot fibroblast-like cells in vitro in a concentration-dependent manner, but does not cause cytotoxicity per se. Concentrations smaller than or equal to 1 mM can be considered safe for use with clubfoot fibroblast-like cells in vitro. However, a concentration of 0.5 mM seems to be enough to elicit significant changes in extracellular matrix collagen content. Minoxidil lowers both short-term and long-term deposition and the structural maturation/assembly of collagen type I fibers. At the same time, it has no considerable influence on the expression of the COL1A1 gene, indicating that the inhibitory effect on collagen type I production and on structural maturation is realized downstream from COL1A1 expression. In addition, minoxidil reduces the cell-mediated contraction of cell-populated collagen gel lattices in a concentration-dependent manner. Our study provides important preliminary data for the potential utilization of minoxidil as an adjuvant pharmacological treatment for reducing fibrosis in the clubfoot medial side after local administration. This treatment could potentially reduce tissue contracture in relapsed patients and thus accelerate and/or improve the outcomes of current conservative methods, especially in cases when the relapsed patients are not adequately responsive to Ponseti treatment. There is ongoing research on antifibrotic drugs for use in various fibrotic diseases (especially pulmonary and liver fibrosis), but to the best of our knowledge little or no attention is being paid to their similar potential for treating clubfoot. Further investigation is needed with a view to harnessing this potential. Acknowledgments We would also like to express our thanks to Daniel Hadraba, Ph.D. for consultations on statistical methods, to Marta Vandrovcova, Ph.D. for proofreading the article, and to the personnel of Bulovka Hospital for their cooperation in sample acquisition and storage. Mr. Robin Healey is gratefully acknowledged for his language revision of the manuscript. Disclosure statement The authors report no conflict of interest. Funding This work was supported by the Ministry of Health of the Czech Republic [AZV 17-31564A] and by Charles University [project GA UK No. 336218]. Some of the images used in this study were obtained from the confocal microscope unit, which is supported by the Ministry of Education, Youth and Sports of the Czech Republic [LM2015062 CzechBioImaging] and by the European Regional Development Fund [project OPPK BrainView CZ.2.16/3.1.00/21544]. Author contributions J.K.: Conceptualization, Methodology, Investigation, Validation, Project administration, Writing – Original draft. M.D.: Conceptualization, Investigation, Validation, Formal analysis, Visualization, Funding acquisition, Writing – Original draft. M. P.: Conceptualization, Investigation. D.V.: Validation, Formal analysis. A.E.: Conceptualization, Supervision, Funding acquisition. M.O.: Resources, Writing – Review. J.M.: Investigation, Writing – Review. L.B.: Supervision, Writing – Review and Editing. T.N.: Supervision, Writing – Review and Editing. ORCID Jarmila Knitlova http://orcid.org/0000-0003-4607-4497 Martina Doubkova http://orcid.org/0000-0002-4955-4093 Martin Plencner http://orcid.org/0000-0002-9540-9188 David Vondrasek http://orcid.org/0000-0001-6432-1012 Adam Eckhardt http://orcid.org/0000-0003-4757-5226 Martin Ostadal http://orcid.org/0000-0002-3282-9847 Jana Musilkova http://orcid.org/0000-0001-9605-9821 Lucie Bacakova http://orcid.org/0000-0002-1818-9484 Tomas Novotny http://orcid.org/0000-0002-3855-0038 References 1. Ponseti IV, Smoley EN.Congenital Club Foot: the Results of Treatment. JBJS. 1963;45(2):261. 2. Dimeglio A, Canavese F. The French functional physical therapy method for the treatment of congenital clubfoot. J Pediatr Orthop B. 2012;21(1):28–39. doi:10.1097/BPB.0b013e32834ee5f8 3. Dobbs MB, Gurnett CA. Update on Clubfoot: etiology and Treatment. Clin Orthop Relat Res. 2009;467 (5):1146–1153. doi:10.1007/s11999-009-0734-9 4. The RC. Ponseti method for the treatment of congenital club foot: review of the current literature and treatment recommendations. Int Orthop. 2013;37(9):1747– 1753. doi:10.1007/s00264-013-2031-1 5. Hosseinzadeh P, Kiebzak G, Dolan L, Zionts L, Morcuende J. Management of Clubfoot Relapses With the Ponseti Method: results of a Survey of the POSNA Members. PubMed - NCBI. 2019;39(1):38–41. doi:10.1097/BPO.0000000000000953 6. Ošťádal M, Chomiak J, Dungl P, Frydrychová M, Burian M. Comparison of the short-term and longterm results of the Ponseti method in the treatment of idiopathic pes equinovarus. Int Orthop. 2013;37 (9):1821–1825. doi:10.1007/s00264-013-2033-z 7. Jeans KA, Karol LA, Erdman AL, Stevens WR. Functional Outcomes Following Treatment for Clubfoot: ten-Year Follow-up. J Bone Joint Surg. 2018;100(23):2015–2023. doi:10.2106/JBJS.18.00317 8. Dunn PM.Congenital postural deformities: perinatal associations. Proc R Soc Med. 1972;65(8):735–738. EFFECT OF MINOXIDIL ON CLUBFOOT-DERIVED CELLS 9. Robertson WW, Corbett D. Congenital clubfoot. Month of conception. Clin Orthop Relat Res. 1997;338:14–18. doi:10.1097/00003086-199705000-00004. 10. Bonnell J, Cruess R.Anomalous Insertion of the Soleus Muscle as a Cause of Fixed Equinus Deformity: A Case Report. J Bone Joint Surg. 1969;51(5):999–1000. 11. Hootnick DR, Levinsohn EM, Crider RJ, Packard DS. Congenital arterial malformations associated with clubfoot. A report of two cases. Clin Orthop Relat Res. 1982;167:160–163. 12. Sadler BA, Gurnett CB, Dobbs M. The genetics of isolated and syndromic clubfoot. J Children’s Orthop. 2019;13:238–244. doi:10.1302/1863-2548.13.190063. 13. Fukuhara K, Schollmeier G, Uhthoff HK. The pathogenesis of club foot. A histomorphometric and immunohistochemical study of fetuses. J Bone Joint Surg. British Volume. 1994;76(3):450–457. 14. Ippolito E, Farsetti P, Caterini R, Tudisco C. Longterm comparative results in patients with congenital clubfoot treated with two different protocols. J Bone Joint Surg. American Volume. 2003;85(7):1286–1294. doi:10.2106/00004623-200307000-00015. 15. Zimny ML, Willig SJ, Roberts JM, D’Ambrosia RD. An electron microscopic study of the fascia from the medial and lateral sides of clubfoot. J Pediatr Orthop. 1985;5 (5):577–581. doi:10.1097/01241398-198509000-00014 16. Eckhardt A, Novotny T, Doubkova M, Hronkova L, Vajner L, Pataridis S, Hadraba D, Kulhava L, Plencner M, Knitlova J, et al. Novel contribution to clubfoot pathogenesis: the possible role of extracellular matrix proteins. J Orthop Res. 2019;37(3):769–778. doi:10.1002/jor.24211. 17. Li C, Nguyen Q, Cole WG, Alman BA.Potential treatment for clubfeet based on growth factor blockade. J Pediatr Orthop. 2001;21(3):372–377. 18. Ošt’ádal M, Eckhardt A, Herget J, Mikšík I, Dungl P, Chomiak J, Frydrychová M, Burian M. Proteomic analysis of the extracellular matrix in idiopathic pes equinovarus. Mol Cell Biochem. 2015;401(1–2):133–139. doi:10.1007/s11010-014-2300-3 19. Kerling A, Stoltenburg-Didinger G, Grams L, Tegtbur U, Horstmann H, Kück M, Mellerowicz H. The congenital clubfoot - immunohistological analysis of the extracellular matrix. Orthop Res Rev. 2018;10:55–62. doi:10.2147/ORR.S156449. 20. Hersh A. The role of surgery in the treatment of club feet. J Bone Joint Surg. American Volume. 1967;49(8):1684– 1696. 21. Branton MH, Kopp JBTGF-Β. and fibrosis. Microbes Infect. 1999;1(15):1349–1365. doi:10.1016/S1286-4579 (99)00250-6 22. McVicker BL, Bennett RG. Novel Anti-fibrotic Therapies. Front Pharmacol. 2017;8:318. doi:10.3389/ fphar.2017.00318. 23. Collagenase DI. Treatment in Dupuytren Contractures: A Review of the Current State Versus Future Needs. Rheumatol Ther. 2016;3(1):43–51. doi:10.1007/s40744016-0027-1 24. Poon R, Li C, Alman BA. Beta-catenin Mediates Soft Tissue Contracture in Clubfoot. Clin Orthop Relat Res. 2009;467(5):1180–1185. doi:10.1007/s11999-008-0692-7 25. Food and Drug Administration. 2017 First Generic Drug Approvals; 2017 [accessed 2020 Feb 6]. https:// 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 15 www.fda.gov/drugs/first-generic-drug-approvals/2017first-generic-drug-approvals Clinic M. Minoxidil (Topical Route) Side Effects; 2020 [accessed 2020 Feb 6]. https://www.mayoclinic.org/ drugs-supplements/minoxidil-topical-route/sideeffects/drg-20068750?p=1 RxList. Minoxidil (Minoxidil Tablets): uses, Dosage, Side Effects, Interactions, Warning; 2018 Oct 29 [accessed 2020 Feb 6]. https://www.rxlist.com/minoxi dil-drug.htm Hautala T, Heikkinen J, Kivirikko KI, Myllylä R. Minoxidil specifically decreases the expression of lysine hydroxylase in cultured human skin fibroblasts. Biochem J. 1992;283(Pt 1):51–54. Murad S, Pinnell SR.Suppression of fibroblast proliferation and lysyl hydroxylase activity by minoxidil. J Biol Chem. 1987;262(25):11973–11978. Piersma B, Bank RA. Collagen cross-linking mediated by lysyl hydroxylase 2: an enzymatic battlefield to combat fibrosis. Essays Biochem. 2019;63(3):377–387. doi:10.1042/EBC20180051 Bailey AJ, Bazin S, Sims TJ, Le Lous M, Nicoletis C, Delaunay A. Characterization of the collagen of human hypertrophic and normal scars. Biochimica et Biophysica Acta (BBA) - Protein Struct. 1975;405 (2):412–421. doi:10.1016/0005-2795(75)90106-3 Moriguchi T, Fujimoto D. Crosslink of Collagen in Hypertrophic Scar. J Invest Dermatol. 1979;72(3):143– 145. doi:10.1111/1523-1747.ep12530609 van der Slot AJ, Zuurmond A-M, van den Bogaerdt AJ, Ulrich MMW, Middelkoop E, Boers W, Karel RH, DeGroot J, Huizinga TWJ, Bank RA. Increased formation of pyridinoline cross-links due to higher telopeptide lysyl hydroxylase levels is a general fibrotic phenomenon. Matrix Biol. 2004;23(4):251–257. doi:10.1016/j.matbio.2004.06.001 Keeley E, Mehrad B, Strieter R. The role of circulating mesenchymal progenitor cells (fibrocytes) in the pathogenesis of fibrotic disorders. Thromb Haemost. 2009;101(4):613–618. doi:10.1160/TH08-11-0726 Hinz B, Phan SH, Thannickal VJ, Galli A, Bochaton-Piallat M-L, Gabbiani G. The Myofibroblast. Am J Pathol. 2007;170(6):1807–1816. doi:10.2353/ajpath.2007.070112 Sidney LE, Branch MJ, Dunphy SE, Dua HS, Concise Review: HA. Evidence for CD34 as a Common Marker for Diverse Progenitors: CD34 as a Common Marker for Diverse Progenitors. STEM CELLS. 2014;32 (6):1380–1389. doi:10.1002/stem.1661 Peng H, Herzog EL. Fibrocytes: emerging effector cells in chronic inflammation. Curr Opin Pharmacol. 2012;12(4):491–496. doi:10.1016/j.coph.2012.03.002 Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9(7):676– 682. doi:10.1038/nmeth.2019. International Organization for Standardization. ISO 10993-5:2009(E): Biological evaluation of medical devices - Part 5: Tests for in vitro cytotoxicity; 2009. Pinnell SR, Murad S.Effects of minoxidil on cultured human skin fibroblasts. Dermatologica. 1987;175(Suppl 2):12–18. 16 J. KNITLOVA ET AL. 41. Priestley GC, Lord R, Stavropoulos P. The metabolism of fibroblasts from normal and fibrotic skin is inhibited by minoxidil in vitro. Br J Dermatol. 1991;125(3):217– 221. doi:10.1111/j.1365-2133.1991.tb14743.x 42. Zuurmond A, Vanderslotverhoeven A, Vandura E, Degroot J, Bank R. Minoxidil exerts different inhibitory effects on gene expression of lysyl hydroxylase 1, 2, and 3: implications for collagen cross-linking and treatment of fibrosis. Matrix Biol. 2005;24(4):261–270. doi:10.1016/j.matbio.2005.04.002 43. Shao S, Zhang X, Duan L, Fang H, Rao S, Liu W, Guo B, Zhang X. Lysyl Hydroxylase Inhibition by Minoxidil Blocks Collagen Deposition and Prevents Pulmonary Fibrosis via TGF-β1/Smad3 Signaling Pathway. Med Sci Monit. 2018;24:8592–8601. doi:10.12659/MSM.910761. 44. Alvarado DM, McCall K, Aferol H, Silva MJ, Garbow JR, Spees WM, Patel T, Siegel M, Dobbs MB, Gurnett CA. Pitx1 haploinsufficiency causes clubfoot in humans and a clubfoot-like phenotype in mice. Hum Mol Genet. 2011;20(20):3943–3952. doi:10.1093/hmg/ ddr313 45. Collinson JM, Lindström NO, Neves C, Wallace K, Meharg C, Charles RH, Ross ZK, Fraser AM, Mbogo I, Oras K, et al. The developmental and genetic basis of ‘clubfoot’ in the peroneal muscular atrophy mutant mouse. Development (Cambridge, England). 2018;145:3. 46. Kakkar A, Traverso G, Farokhzad OC, Weissleder R, Langer R. Evolution of macromolecular complexity in 47. 48. 49. 50. 51. 52. drug delivery systems. Nature Rev Chem. 2017;1 (8):0063. doi:10.1038/s41570-017-0063 Wynn TA. Common and unique mechanisms regulate fibrosis in various fibroproliferative diseases. J Clin Invest. 2007;117(3):524–529. doi:10.1172/JCI31487 Chen CZ, Raghunath M. Focus on collagen: in vitro systems to study fibrogenesis and antifibrosis state of the art. Fibrogenesis Tissue Repair. 2009;2:7. doi:10.1186/1755-1536-2-7. Devkota AK, Veloria JR, Guo HF, Kurie JM, Cho EJ, Dalby KN. Development of a High-Throughput Lysyl Hydroxylase (LH) Assay and Identification of SmallMolecule Inhibitors against LH2. SLAS Discovery. 2019;24(4):484–491. doi:10.1177/2472555218817057 Wang C, Luosujärvi H, Heikkinen J, Risteli M, Uitto L, Myllylä R. The third activity for lysyl hydroxylase 3: galactosylation of hydroxylysyl residues in collagens in vitro. Matrix Biol. 2002;21(7):559–566. doi:10.1016/ s0945-053x(02)00071-9 Merl-Pham J, Basak T, Knüppel L, Ramanujam D, Athanason M, Behr J, Engelhardt S, Eickelberg O, Hauck SM, Vanacore R, et al. Quantitative proteomic profiling of extracellular matrix and site-specific collagen post-translational modifications in an in vitro model of lung fibrosis. Matrix Biol Plus. 2019;1:100005. Dallon JC, Ehrlich HP. A review of fibroblast-populated collagen lattices. Wound Repair Regenerat. 2008;16 (4):472–479. doi:10.1111/j.1524-475X.2008.00392.x