探讨人表皮干细胞(ESC)对增生性瘢痕(HS)成纤维细胞的作用。

取2021年7月至2022年1月中山大学附属第一医院小儿外科7~12岁健康小儿包皮手术切除的20例包皮组织，采用快速附着法提取ESC，胶原酶消化法提取正常成纤维细胞(FB)，运用流式细胞学进行鉴定。取2021年10月至2022年5月中山大学附属第一医院烧伤外科HS增生期患者手术切除的6例HS标本，采用胶原酶消化法提取增生性瘢痕来源的成纤维细胞(HSF)。使用差异高速离心法提取人ESC外泌体和FB外泌体，采用透射电子显微镜观察其形态，纳米颗粒跟踪分析仪检测颗粒直径，蛋白质印迹法检测CD9、CD63和α-微管蛋白的蛋白表达。利用PKH67染色，将人ESC外泌体、FB外泌体和HSF细胞分别共培养24 h后，观察HSF对外泌体的吞噬情况。将HSF分为ESC外泌体组(包括5 μg/ml ESC外泌体组、10 μg/ml ESC外泌体组、20 μg/ml ESC外泌体组、30 μg/ml ESC外泌体组)、FB外泌体组(包括20 μg/ml FB外泌体组、30 μg/ml FB外泌体组)和对照组(磷酸盐缓冲液PBS)，每组3孔，处理24、48、72 h后，采用CCK8法检测HSF的数量以确定外泌体对其增殖能力的影响以及最短处理时间和最低有效浓度。将HSF分为对照组、ESC外泌体组和FB外泌体组，每组3孔(下同)，按CCK8实验筛选的最短处理时间和最低有效浓度，采用Ki67免疫荧光染色检测HSF处于增殖期的细胞占所有细胞的比值。采用划痕实验，在处理后第0、12、24、48 h观察细胞划痕面积，计算细胞迁移率。采用胶原收缩实验，在处理后第0、24、48、72 h观察胶原剩余面积，计算胶原收缩面积比值。在处理48 h后，收集HSF细胞RNA、总蛋白，采用实时荧光定量RT-PCR法和免疫印迹法检测Ⅰ型胶原(COLⅠ)和α-平滑肌肌动蛋白(α-SMA)的mRNA和蛋白表达。采用荧光染色法再证COLⅠ和α-SMA的相对荧光表达强度。数据比较采用重复测量方差分析、单因素方差分析、独立样本t检验。

传代后培养24 h，FB呈典型梭形结构，ESC呈"铺路石"样簇状排列，经流式细胞学鉴定为成纤维细胞、表皮干细胞。外泌体呈囊泡状，ESC外泌体平均粒径为123.1 nm，FB外泌体平均粒径为128.9 nm，均表达CD9、CD63而不表达α-微管蛋白。共培养24 h后，外泌体被HSF吞入细胞质。培养24 h后，5 μg/ml ESC外泌体组、10 μg/ml ESC外泌体组、30 μg/ml FB外泌体组CCK8吸光度与PBS组差异无统计学意义(t值分别为0.45、2.04、0.39，P>0.05)，20 μg/ml FB外泌体组CCK8吸光度比PBS组稍高(t＝4.52，P<0.05)，而20 μg/ml ESC外泌体组、30 μg/ml ESC外泌体组CCK8吸光度显著低于PBS组(t值分别为7.06、15.26，P<0.001)，说明该浓度的ESC外泌体能显著抑制HSF的增殖，且抑制增殖能力随处理时间延长更加明显(交互作用F＝5.19，P<0.001)。此外，30 μg/ml ESC外泌体比20 μg/ml ESC外泌体在24、48、72 h的抑制作用均更加显著(t值分别为5.88、5.18、13.64，P<0.05)。由CCK8实验可得，ESC外泌体抑制HSF增殖的最短有效时间为24 h，此时最低有效浓度为20 μg/ml。故选用20 μg/ml的外泌体处理24 h为后续实验条件。刺激24 h后，对比PBS组，ESC外泌体处理后的HSF中Ki67免疫荧光染色阳性的细胞比例显著减少(t＝16.97，P<0.001)，而FB外泌体组差异无统计学意义(t＝1.01, P>0.05)。ESC外泌体培养12、24、48 h后，HSF迁移能力明显下降，细胞迁移率分别为(52.86%±5.02%)、(59.10%±7.45%)、(70.78%±11.33%)，低于对照组(60.87%±3.35%)、(92.15%±3.61%)、(100.00%±0.00%)，差异有统计学意义(t值分别为2.57、10.60、9.37，P<0.05)，作用效果随时间延长而增强(交互作用F＝33.26，P<0.001)。而FB外泌体组的细胞迁移率分别为(58.78±5.46)%、(89.69±5.80)%、(98.40±1.28)%，与对照组相比，差异无统计学意义(t值分别为1.08、1.28、0.83，P>0.05)。胶原收缩实验结果显示，处理24、48、72 h后ESC外泌体能明显抑制胶原凝胶收缩，细胞收缩率分别为(23.07%±8.69%)、(30.68%±6.18%)、(45.92%±3.74%)，低于对照组(46.18%±2.21%)、(66.80%±7.34%)、(76.65%±3.47%)，差异有统计学意义(t值分别为5.68、8.88、7.55，P<0.001)，作用随时间延长而增强(交互作用F＝10.28，P<0.001)。而FB外泌体组的细胞收缩率分别为(41.19%±12.33%)、(63.54%±4.20%)、(73.05%±3.22%)，与对照组相比，差异无统计学意义(t值分别为1.09、0.71、0.79，P>0.05)。ESC外泌体处理48 h后COLⅠ和α-SMA的mRNA相对表达量较对照组显著减少(t值分别为6.25，3.07，P<0.05)，而FB外泌体组COLⅠ的mRNA相对表达量较对照组有轻微提升(t＝3.78，P<0.05)，α-SMA的mRNA相对表达量与对照组相比差异无统计学意义(t＝1.44，P>0.05)。培养48h提取总蛋白进行免疫印迹实验和荧光染色，结果显示ESC外泌体抑制HSF细胞COLⅠ和α-SMA的表达(t值分别为7.00，9.79，P<0.001)，而FB外泌体则无明显作用(t值分别为2.59，3.23，P>0.05)。

建立了稳定可靠的人ESC培养和外泌体分离提取体系，验证了ESC外泌体抑制增生性瘢痕来源成纤维细胞的增殖、迁移、收缩，并抑制其向肌成纤维细胞转化的能力。

To investigate the effects of exosomes derived from human epidermal stem cells (ESCs) on fibroblasts of hypertrophic scars (HS).

The foreskin tissues were collected from 20 healthy young children (7~12 years old) who underwent foreskin resection in the Department of Pediatric Surgery, the Hospital, from July 2021 to January 2022. FBs were isolated by the collagenase digestion method, while ESCs were extracted by the rapid attachment method from the foreskin tissues and identified by flow cytometry. Six HS specimens were collected from patients with hyperplasia of HS in the Department of Burn Surgery, the Hospital, from October 2021 to May 2022. Fibroblasts derived from hypertrophic scars (HSFs) were isolated by the collagenase digestion method. ESC exosomes (ESC-Exo) and FB exosomes (FB-Exo) were extracted by differential high-speed centrifugation. The morphology was observed by projection electron microscopy, the particle size was detected by nanoparticle tracking analyzer, and the protein expression of CD9, CD63, and α-tubulin was detected by Western blotting. PKH67-stained ESC-Exo and HSFs were co-cultured for 24 hours to observe the phagocytosis of human ESC exosomes by HSFs. All HSFs were randomly divided into ESC-Exo group (including 5 μg/ml ESC-Exo, 10 μg/ml ESC-Exo, 20 μg/ml ESC-Exo and 30 μg/ml ESC-Exo group), FB-Exo group (including 20 μg/ml FB-Exo, 30 μg/ml FB-Exo group), and control group, with 3 wells in each group. The number of HSFs was measured by CCK8 at 24, 48 and 72 h to determine the effect of exosomes on their proliferative ability, as well as to determine the minimum treatment time and minimum effective concentration. After that, HSFs were randomly divided into the control group, ESC-Exo group, and FB-Exo group, with 3 wells in each group (the same applies to the following). The number and the ratio of HSF cells in the proliferative phase were detected by Ki67 fluorescence staining. Scratch tests were performed according to the lowest effective concentration screened by the CCK8 experiment. The scratch area of cells was observed at 0, 12, 24, and 48 h after treatment, and the cell migration rate was calculated. The collagen contraction experiment wasused to observe the remaining area of collagen at 0, 24, 48, and 72 h after treatment, and the ratio of collagen contraction area was calculated. After 48 hours of treatment, RNA and total protein were collected, and the mRNA and protein expressions of collagen type Ⅰ (COLⅠ) and α-smooth muscle actin (α-SMA) were detected by real-time fluorescence quantitative (RT-PCR) and Western blotting. The relative fluorescence expression intensity of COLⅠ and α-SMA was confirmed by fluorescence staining. Data were compared by repeated measure analysis of variance, one-way analysis of variance, and independent sample t-test.

At 24 hours of culture, FBs showed a typical fusiform structure, and ESCs arranged in "paving stone" like clusters. ESCs and FBs were identified by flow cytometry. The exosomes were vesicular and positively expressed CD9 and CD63 but did not express α-tubulin. After 24 hours of co-culture, human ESC-Exo was swallowed into HSFs. After 24 to 72 hours of culture, CCK8 tests showed the proliferation of HSFs in groups with 20 μg/ml and 30 μg/ml ESC-Exo were significantly lower than in the control group. After 24 hours of culture, the absorbance of CCK8 in the 5 μg/ml ESC-Exo group, 10 μg/ml ESC-Exo group, and 30 μg/ml FB-Exo group were not different from that in the PBS group (t-values were 0.45, 2.04, 0.39, P>0.05). The absorbance of CCK8 in the 20 μg/ml FB-Exo group was slightly higher than that in the PBS group (t=4.52, P<0.05), while the absorbance of CCK8 in 20 μg/ml ESC-Exo and 30 μg/ml ESC-Exo groups was significantly lower than that in PBS group (t-values were 7.06, 15.26, P<0.001). These results indicated that the ESC-Exo at this concentration could significantly inhibit the proliferation of HSF, and the inhibitory ability became more obvious with the extension of treatment time (interaction F=5.19, P<0.001). In addition, 30 μg/ml ESC-Exo showed more significant inhibition than 20 μg/ml ESC-Exo at 24, 48, and 72 h (t-values were 5.88, 5.18, 13.64, P<0.05). The shortest effective time was 24 hours, and the lowest effective concentration was 20 μg/ml. Therefore, 20 μg/ml was selected as the following experimental concentration. After 24 hours of treatment, the ratio of Ki67-positive cells in the ESC-Exo group was obviously decreased (t=16.97, P<0.001), while there was no significant difference in the FB-Exo group (t=1.01, P>0.05). After a co-culture with ESC-Exo for 12, 24, and 48h, the migration ability of HSF decreased significantly. The migration rates of scratched cells in the ESC-Exo group were (52.86%±5.02%), (59.10%±7.45%), and (70.78%±11.33%), respectively. They were lower than the control group (60.87%±3.35%), (92.15%±3.61%), and (100.00%±0.00%). The difference was statistically significant (t-values were 2.57, 10.60, 9.37, P<0.05), and the effect increased with the extension of time (interaction F=33.26, P<0.001). The cell mobility of the FB-Exo group was (58.78%±5.46%), (89.69%±5.80%), and (98.40%±1.28%), respectively, which showed no significant difference compared with the control group (t-values were 1.08, 1.28, 0.83, P>0.05). The results of the collagen contraction experiment showed that after treatment for 24, 48, and 72 h, the ESC-Exo inhibited collagen gel shrinkage significantly, and the cell shrinkage rates were (23.07%±8.69%), (30.68%±6.18%), (45.92%±3.74%), respectively. They were lower than the control group (46.18%±2.21%), (66.80%±7.34%), and (76.65%±3.47%). The difference was statistically significant (t-values were 5.68, 8.88, 7.55, P<0.001), and the effect increased with the extension of time (interaction F=10.28, P<0.001). The cell shrinkage rates in the FB-Exo group were (41.19%±12.33%), (63.54%±4.20%), and (73.05%±3.22%), respectively, showing no significant difference compared with the control group (t-values were 1.09, 0.71, 0.79, P>0.05). The relative mRNA expressions of COLⅠ and α-SMA in ESC-Exo treated for 48 hours were significantly lower than those in the control group (t=6.25, 3.07, P<0.05). Nevertheless, the relative mRNA expression of COLⅠ in the FB-Exo group was slightly higher than that of the control group (t=3.78, P<0.05), while there was no significant difference in α-SMA mRNA between FB-Exo group and control group (t=1.44, P>0.05). Western blot and fluorescence staining results also showed that ESC-Exo inhibited the expression of COLⅠ and α-SMA in HSFs (t-values were 7.00, 9.79, P<0.001), while FB-Exo didn′t (t-values were 2.59, 3.23, P>0.05).

A stable and reliable system for human ESCs culture and exosome isolation and extraction was established to verify the ability of ESC-Exo to inhibit the proliferation, migration, and contraction of fibroblasts derived from hypertrophic scars and suppress their transformation into myofibroblasts.