The Expression of DNMT1 in Pathologic Scar Fibroblasts and the Effect of 5-aza-2-Deoxycytidine on Cytokines of Pathologic Scar Fibroblasts

Objective. This study aimed to investigate the expression and significance of DNA methyltransferase 1 (DNMT1) in pathologic scar fibroblasts, as well as the influence of methylase inhibitor, 5-aza-2-deoxycytidine, on pathological scars. Material and Methods. Samples of 31 keloids, 20 hypertrophic scars, and 25 normal skins were taken to test the expression rate of DNMT1 by immunohistochemistry. Primary fibroblasts were cultured with the monoplast method. Samples were categorized into the keloid group (K group), 5-aza-2-deoxycytidine keloid intervention group (K+ group), normal skin group (N group), hypertrophic scar group (H group), and 5-aza-2-deoxycytidine hypertrophic scar intervention group (H+ group). The expressions of DNMT1, transforming growth factor-β (TGF-β), and Smad7 mRNA in each group were detected with real-time polymerase chain reaction. The effect of 5-aza-2-deoxycytidine on the cell cycle and apoptosis of pathologic scar fibroblasts were analyzed with flow cytometry. Results. The expression rate of DNMT1 was 100% in keloid fibroblasts, 90% in hyperplastic scar fibroblasts, and 8% in normal skin fibroblasts. After the intervention with 5-aza-2-deoxycytidine in the K+ group, the expression of DNMT1 and TGF-β1 mRNA was lower, Smad7 mRNA was elevated in pathological scar fibroblasts, the flow cytometry showed the proportion of cells in G0/G1 phase were increased, and the proportion of apoptosis cells were also increased, with similar changes in the cells in the H and H+ groups. Conclusion. DNMT1 may play a vital role on the generation of pathological scars. Methylaze inhibitors 5-aza-2-deoxycytidine may influence the related cytokines of pathological scars, inhibit proliferation, and promote apoptosis of pathological scar fibroblasts. The generation of pathological scars may be related with methylation of certain genes. 5-aza-2-deoxycytidine may be a new choice for the treatment of pathological scars.

  Pathological scars are common complications in plastic surgery. Aside from their affect on the appearance of the skin, scars are usually accompanied by pruritus, pain, and even functional disability. There are a number of studies about the pathogenesis of pathological scars, but their mechanisms remain unclear. Clinically, there is no effective medication to treat scars. A previous study shows that in the process of cell division and proliferation, DNA methylation potentially may preserve the stemness of cells.¹ DNA methyltransferase 1 (DNMT1) protein was found enriched in undifferentiated cells, where it was a key factor to epidermal stem cells.¹ DNA methyltransferase 1 is involved in many proliferative and fibrosis diseases, especially in cell growth and survival of cancer cells.² Currently there are no reports demonstrating whether or not pathological scar genes are methylated. Protein families of transforming growth factor-β (TGF-β) could induce the apoptosis of epithelial cells, stimulate proliferation of interstitial cells, produce the extracellular matrix, induce fibrosis of all kinds of tissues, and generate scars.³ The dominant factor, TGF-β1, is a multifunctional cytokine involved in all stages of wound healing and scar formation. It is the essential factor in this generation of hyperplastic scar.⁴ The SMADs family of proteins (mothers against decapentaplegic) can mediate the signal pathway of TGF-β/SMADs triggered by TGF-β. It transfers the stimulation from extracellular into nuclei and regulates the intracellular specific gene expression and protein synthesis.⁵ The pathway mediated by SMADs family proteins has positive and negative feedback regulatory mechanisms. Inhibitory SMADs (I-SMADs) SMAD6 and SMAD7 inhibit the TGF-β/SMADs signal transduction pathway by competitive combination with type-I receptor⁶; SMAD7 plays a crucial role in this process. Studies show that there are obvious genetic differences in the fibroblasts that tend to form scars after wounding.⁷ DNA methylation and histone acetylation cause genetically diversified expressions of keloid fibroblasts.⁸ Currently, there are no explicit reports on how the DNMT1 gene in pathological scars is expressed and whether the TGF-β signal pathway and different expressions of SMAD7 are related to gene methylation.

  In previous studies,^9,10 the authors collected specimens – not those used in the current study – and studied the difference of methylation between keloid and normal skin. The results of that work proved that DNMT1 played an important role in the process of keloid growth and was closely related to infiltrative growth in keloids; therefore, the authors designed this in-depth study, and increased the sample size to further examine the methylation of keloids, hypertrophic scars, and normal skin. The goal of this study is to provide a theoretical basis for the pathogenesis and treatment of keloids and hypertrophic scars, and provide new thinking for drug discovery.

  Subjects. Seventy-six samples of keloids (n = 31), hypertrophic scars (n = 20), and normal skin (n = 25) were collected from selective plastic surgeries performed at the Department of Plastic Surgery of The First Affiliated Hospital of Chongqing Medical University, Chongqing, China. Tissue from keloids and hypertrophic scars were collected from patients during scar removal surgery, and normal skin tissue was taken from patients undergoing circumcision to serve as the control. All patients provided informed written consent. Patients ranged in age from 20 to 40 years. Keloids had a growth period from 1 to 3 years, and were located on the ear lobe (17), neck (3), chest (8), arm (2), and leg (1). Hypertrophic scars had a growth period less than 1 year and were located on the neck (6), chest (5), arm (2), abdomen (5), and back (2). Normal skin was taken from the eyelid (10), ear (3), abdomen (3), and foreskin (9). All samples were tested and found negative for any infection, and were kept in sterile conditions.

  Reagents. The following reagents were obtained: DNMT1 mouse monoclonal antibody (Abcam, Cambridge, MA); second antibody and DAB reagent kit (Hai Kang Life Corp, Ltd, Beijing, China), Dulbecco Modified Eagle’s Medium (DMEM) with 12% fetal bovine serum (FBS) (Thermo Scientific, Waltham, MA); trypsin and dimethyl sulfoxide (Sigma-Aldrich, St. Louis, MO); total RNA extraction reagent Trizol, a reverse transcription kit, and tobacco acid pyrophosphatase enzymes (Takara Biotechnology [Dalian] Co, Ltd, Shiga, Japan); 5-aza-2-deoxycytidine (Sigma-Aldrich, St. Louis, MO).

  Instruments. The authors used a sterile workstation (Suzhou HuaHong Purification Technology Company Ltd, Jiangsu, China); a polymerase chain reaction instrument and gel imaging system (Bio-Rad Laboratories, Hercules, CA); a nucleic acid protein analyzer (ABB, Zurich, Switzerland); a freeze centrifuge (Hermle Machine Company, Franklin, WI); and flow cytometry (Guava Technologies, Hayward, CA).

  Sample grouping and immunohistochemistry. Samples of keloids, hypertrophic scars, and normal skin were collected during plastic surgeries, with patients’ consent having been acquired beforehand. All samples were fixed in paraformaldehyde for 24 hours, embedded in paraffin, and prepared for the immunohistochemical paraffin slices. The procedure was performed according to the immunohistochemistry kit’s instruction, with antigen retrieval methods used as needed. In this case, sodium citrate repair antigen was used. The results were scored according to the positive cell counts of fibroblasts and dyeing degrees as follows: 0 = 0 points; 1%-25% = 1 point; 26%-50% = 2 points; 51%-75% = 3 points; and 76%-100% = 4 points. Points were categorized according to the proportion of positive cells, and assigned scores of stainless (0 points), weak (1 point), medium (2 points), and strong (3 points), according to the dyeing degrees.¹¹ The scores were then integrated into general scores. For example, if, in a piece of glass, strong positive cells are 20%, positive cells are 30%, weakly positive cells are 20%, and negative cells are 30%, then the general score is (3 x 1) + (2 x 2) + (1 x 1) + (0 x 2) = 8 points. Each slice was evaluated individually by 2 pathologists using semiquantitative systemic evaluation and the procedure was double-blinded.

  Fibroblast cells culture and grouping. Primary human fibroblast cells were derived from fresh foreskin and scar tissue to provide primary cultures of the fibroblast cells of keloids, hypertrophic scars, and normal skin, with a monocellular method. Cells were maintained in DMEM and 12% FBS. When cell proliferation reached proper numbers, the cells were randomly divided into 5 groups: the keloid drug (K+) intervention group; the keloid fibroblasts (K); hypertrophic scars drugs (H+) intervention group; hypertrophic scars (H) control group; and the normal skin fibroblasts (N) group.

  An intervention of 5-aza-2–deoxycytidine at 5 * 10-5 mol/l was used, with the addition of 2.85 ml 15% FBS of DMEM with high glucose culture medium and 0.15 ml 10-3 mol/l 5-aza-2-deoxycytidine solution per bottle of cells, 5 - aza -2 - deoxycytidine solution prepared with phosphate buffered saline (PBS) for the drug intervention group; and the control group was “treated” with the same amount of PBS instead of the drug that joined 2.85ml 15% FBS of DMEM with high glucose culture medium and 0.15 ml PBS per bottle of cells. The authors changed the liquid every 24 hours and collected cells for experiments after 72 hours.

  The detection of cell period and apoptosis of each group of fibroblasts. The cell cycle distribution and apoptotic state were detected by flow cytometry. Cells from various groups were collected and counted with 500 µl Annexin V plus PI conjunction to 106 cells.

  Each group of cells grew uniformly and counted 106. Flow cytometry analysis was performed using a propidum iodide kit to detect fibroblasts’ apoptosis in 4 study groups (K+, K, H+, and H).

  Detect the expression of DNMT1, TGF-β1, SMAD7 mRNA in each group of fibroblasts. The total mRNA was collected by Trizol extraction from various cell groups. The purity of total mRNA was measured by spectrophotometer. After that, the mRNA were reversed into cDNA by reverse transcription reaction. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used in genes amplification. The authors then compared the difference of mRNA in each group by Agarose gel electrophoresis (Table 1).

  SPSS Statistics for Windows, version 17.0 software (SPSS Inc, Chicago, IL) was used to process and analyze the data. Single factor variance analysis was used for comparing each group. Statistical significance was set at P < 0.05.

  Immunohistochemistry showed the levels of expression in pathologic scars, as they had previously observed the expression of DNMT1 in epidermal stem cells. Protein DNMT1 is mainly expressed inside the fibroblasts’ nucleus. There was no significant expression of DNMT1 in the cytoplasm and cell membrane. The DNMT1 expression rate was 100% in the keloid group (Figure 1a, 1b); 90% in the hypertrophic scar group (Figure 2a, 2b); and 8% in the normal skin group (Figure 3a, 3b). Immunohistochemistry scores are listed in Table 2. The score for keloid and hypertrophic scar fibroblasts is higher than that of normal skin fibroblasts. Mean-variance analysis was undertaken with SPSS Statistics for Windows version 17.0 (SPSS Inc, Chicago, IL). The Newman-Keuls method was used to compared the differences between the drug intervention group and control group. The results show a statistical significant difference between keloids and normal skin and between hypertrophic scars and normal skin; there was no statistical significant difference between hypertrophic scars and keloids (Table 2).

  The effects of 5-aza-2-deoxycytidine on cell cycles and apoptosis of pathological scar fibroblasts. From the flow cytometry results, the authors observed a significantly increased ratio of cells in the G0/G1 phase, which were treated with 5-aza-2-deoxycytidine in the K+ intervention group and the H+ intervention group. One-way analysis of variance was used to compare the difference between each intervention group. The results revealed a statistically significant difference in the proportion of G0/G1 cells in the K+ group compared to the K group (P = 0.001), and the H+ group compared to the H group (P = 0.000) (Table 3).

  From the flow cytometry results, the authors also observed there were more cells in an apoptotic state in the intervention groups located in lower right area than in the control groups. Single factor analysis of variance was used to compare the difference between each group. The result showed a significant difference between the K+ group and K group (P = 0.002), and between the H + group and H group (P = 0.003) (Table 4).

  The effect of 5-aza-2-deoxycytidine on expression of pathological scar fibrocyte DNMT1, TGF-β1, and SMAD7 mRNA that are cultured in vitro. The authors used real time-polymerase chain reaction (PCR) to measure levels of DNMT1, TGF-β1, and SMAD7 mRNA expression in each group. Agarose gel electrophoresis showed that the expression of DNMT1 and TGF-β1 mRNA on pathological scar fibroblasts (cultured in vitro) significantly increased, and the expression of SMAD7 mRNA on pathological scar fibroblasts (cultured in vitro) was lower than in the normal skin fibroblasts. The expression of DNMT1 and TGF-β1 mRNA on pathological scar fibroblasts (cultured in vitro and treated with 5-aza-2-deoxyctidine) decreased (Tables 4 and 5), but the expression of SMAD7 mRNA increased (Table 6). ImageJ software was used to contrast and analyze the PCR diagrams gray value; the results showed a statistical difference.

  The pathological scar is a superficial fibrosis disease resulting from excessive fibroblast proliferation and collagen deposition mainly in the extracellular matrix during wound-healing and repairing.¹² Keloids are more prevalent in certain ethnic groups than in whites.¹³ Recent studies on the pathology of scars have not found an effective treatment. The SMADs family-mediated TGF-β/SMADs pathways play essential roles in scar formation. Epigenetic modifications have been shown to play a role in the pathogenesis of cancer, as well as autoimmune and inflammatory disorders. Studies have shown that a potential epigenetic mechanism for the cellular memory is needed to preserve the somatic progenitor state through repeated cell divisions provided by DNA methylation,^1,14-16 and DNA methylation patterns after cellular replication has been maintained by DNMT1.^1,17,18 Abnormal DNA methylation is always accompanied with abnormal expression of DNMTs.¹⁹ DNMT1 is one of 3 types of DNA methylation enzymes, the other 2 being DNMT3a and DNMT3b, which were also coded by DNMT genes and have recently been identified.²⁰ Keloids are benign dermal tumors that form during wound healing in genetically susceptible individuals. Bechtel et al²¹ found that renal fibrosis is highly correlated with gene methylation. Their findings support an altered program of DNA methylation and histone acetylation that could account for the stable pattern of differential gene expression in keloid fibroblasts in culture.¹⁸ Studies have explored if, in the wound healing environment of genetically predisposed individuals, the body produced or selected these epigenetically distinct fibroblasts.^8,21

  Some studies have shown the histone deacetylase inhibitors trichostatin A^22,23 could reduce the synthesis of collagen and hyperplasia of keloid fibroblasts by inhibiting the TGF-β pathways.²⁴ This proves that the formation of pathological scars is related to epigenetics, since histones deacetylased of certain factors were involved. However, whether or not there are overexpressed DNMT1 in pathological scars and methylation of related genes through TGF-β/SMADs pathways in human beings has yet to be reported.

  In the current study, the authors found DNMT1 is highly expressed in the nuclei of the cicatricial fibroblasts, while in normal skin, the expression is weak or negative. The immunohistochemical score of DNMT1 in pathological scar fibroblasts is also higher than that of normal skin. Analysis of variance test showed a significant difference between hypertrophic scars and normal skin, and between keloids and normal skin, but there were no significant differences between the hypertrophic scars and keloids. Due to changes on mRNA levels, the study also showed that DNMT1 mRNA expression is significantly higher in the cultured fibroblasts of pathological scars in vitro than in normal skins performed by RT-PCR. Thus, the results illustrate that the higher expression of DNMT1 in the pathological process plays a key role in scar formation.

  The TGF-β/SMADs pathways are important in the process of scar genesis. The main stimulating factor of scar formation is TGF-β1, while the SMADs family is the main intracellular regulative factor.²⁵ The SMADs family can be differentiated into 3 main categories based on function and structure: The receptor-regulated SMADs (R-SMADs)²⁶ including SMAD1, 2, 3, 5, 8; the common-mediator SMADs (only Smad4 at present); and the suppressor SMADs, mainly SMAD6 and 7. In the scar pathogenesis, the suppressor SMADs are a down-regulating factor; SMAD7 can block the activation of R-Smads, negatively regulate the TGF-β pathways, then inhibit the formation of scars.²⁵

  As an inhibitor of DNA methylation, 5-aza-2-deoxycytidine prevents DNA methylation by inhibiting the activity of DNMT enzymes. It has been reported that treatment of donor cells²⁷ or early somatic cell nuclear transfer (SCNT) embryos²⁸ with 5-aza-2-deoxycytidine does not increase the developmental potential of cloned embryos.⁸ Wang et al²⁹ demonstrated that 5-aza-2-deoxycytidine and trichostatin A treatment treatment significantly reduced the DNA methylation level of the satellite I region in SCNT blastocysts.²⁹ In the current study, the authors used a methyltransferase inhibitor 5-aza-2-deoxycytidine as an intervention on the cultured cicatricial fibroblasts in vitro. The proportion of apoptosis and the G0/G1 cells increased, demonstrating 5-aza-2-deoxycytidine could inhibit the proliferation of scar fibroblasts and promote apoptosis. They also demonstrated the hyperplasia in the fibroblasts is highly correlated with gene methylation.

  In the current study, after the intervention of 5-aza-2-deoxycytidine, the results of RT-PCR illustrated that factors related to scar formation changed simultaneously, and the expression of DNMT1 mRNA and TGF-β1 mRNA decreased while the expression of SMAD7 mRNA increased. The authors conclude that 5-aza-2-deoxycytidine inhibited the methylation through the inhibition of DNMT1 mRNA expression. After the inhibition of the mRNA expression, the essential TGF-β/SMADs pathway may be restrained, then the level of SMAD7 rebounds to more than its original level. A previous study³⁰ on the methylation of keloids proved there is no significant effect on cell proliferation of normal skin fibroblast after 5-aza-2-deoxycytidine was used. Also, the expression of scar-related factors such as TGF-β1 and Smad7 had no significant changes before or after drug intervention. This phenomenon may be related to hypomethylation in normal skin.³⁰

  The pathogenesis of pathological scars is considerably complex. This study demonstrates that methylation is one of the factors inducing scar pathogenesis in which genes are involved. Further studies are needed to study methylation in scar formation. As this research was confined to the in vitro study of fibroblasts, further elucidation is needed to clarify the intricate in vivo mechanism. The inhibition of methylation could possibly become a new approach to scar treatment, but more research is needed to confirm the effects of methyltransferase inhibitor 5-aza-2-deoxycytidine on scar therapy in vivo.

The authors are from the Department of Plastic and Burn Surgery, The First Affiliated Hospital of Chongqing Medical University, Yuanjiagang, Yuzhong District, Chongqing City, China.

Address correspondence to:
Zhang Hengshu, MD
Department of Plastic and Burn Surgery
The First Affiliated Hospital of Chongqing Medical University
Yuanjiagang, Yuzhong District
Chongqing City, China
512442489@qq.com

Disclosure: The authors disclose this research was supported by the Scientific and Technological Research Program of the Chongqing Municipal Education Commission.