研究背景

环状RNA（Circular RNA, circRNA）是前体mRNA（pre-mRNA）反向剪接的产物，是由前体RNA的5’和3’末端首尾相互连接形成的单链环状特殊结构的RNA分子。作为非编码RNA的一种，环状RNA被发现能够作为mRNA的竞争性内源RNA（Completing Endogenous RNA, ceRNA），通过在细胞内结合miRNA，限制miRNA对mRNA的降解，从而对特定基因的表达起到一定的调节作用；也有研究提示，环状RNA在人体的不同组织和不同疾病状态中表达存在巨大差异，检测相应的差异性表达环状RNA（Differentially Expressed Circular RNA, DE-circRNA）能够对某些疾病鉴别诊断及判断其进展状态起到生物标志物（biomarker）的作用。

器官的纤维化可发生在多种脏器中，包括心脏、肝脏、肺、肾脏等，在病理上主要表现为相应部位成纤维细胞、肝星形细胞等多种间充质干细胞的激活，造成脏器细胞间质的细胞外基质成分（Extracellular matrix, ECM）大量堆积，进而出现组织纤维化，最终造成功能衰竭。越来越多的资料表明，环状RNA与多种器官的纤维化疾病存在极为密切的关联。

作为纤维化疾病的一种，瘢痕疙瘩主要的病理表现为皮肤轻度损伤或无明显诱因造成的皮肤内细胞外基质的异常堆积，在临床中则表现为突出于体表、逐渐向周围侵袭且伴随瘙痒疼痛症状的良性肿瘤。瘢痕疙瘩的治疗效果很多，但治疗效果较差，究其原因是现阶段尚未完全了解瘢痕疙瘩的进展机制。

近年来的部分研究发现，瘢痕疙瘩中环状RNA的表达与正常皮肤之间存在明显差异，且其中部分DE-circRNA对瘢痕疙瘩的进展以及瘢痕疙瘩成纤维细胞（Keloid-derived fibroblasts, KFs）的侵袭行为具有一定的调控作用。但现阶段缺乏相关的系统性、大数据的研究，无法对瘢痕疙瘩中差异性表达的环状RNA（Differentially Expressed circRNA, DE-circRNA）进行大量筛选与表达谱构建，也缺乏对环状RNA相关的ceRNA网络及其生物标志物等功能的研究。基于此，我们对瘢痕疙瘩DE-circRNA进行了深入的探索。

研究目的

1. 筛选出瘢痕疙瘩相关的RNA数据库，初步获得人瘢痕疙瘩DE-circRNA表达谱，并进一步筛选出对瘢痕疙瘩进展更具意义的DE-circRNA分子。

2. 构建可能存在“DE-circRNA-miRNA-mRNA”模式的ceRNA网络，进而了解DE-circRNA分子对瘢痕疙瘩进展及重要基因表达起到的调控作用。

3. 从预测出的表达谱中挑选出可能与瘢痕疙瘩进展更为密切的DE-circRNA分子，通过分子生物学的方法，验证所构建的人瘢痕疙瘩中DE-circRNA表达谱的可靠性。

4. 根据靶向mRNA，挑选出可能对瘢痕疙瘩进展更具意义的ceRNA网络，通过分子生物学的方法进行，验证前述ceRNA网络的可靠性。

5. 通过细胞学的方法，验证ceRNA网络在瘢痕疙瘩成纤维细胞中进行行为学调控的可靠性。

6. 探索DE-circRNA作为可否作为瘢痕疙瘩的生物标志物，用于瘢痕疙瘩的鉴别诊断以及了解瘢痕疙瘩进展的指示作用。

研究方法

1. 数据库筛选及DE-circRNA筛选：在国家生物技术信息中心（National Center Library of Medicine，NCBI）中NSE网络生物数据库进行筛选，筛选出瘢痕疙瘩相关的GSE113621数据库。应用R软件中的limma语言包的多阵列平均化语言进行数据标准化，在将探针ID转化为circRNA符号后，对GSE113621数据库中的样本数据进行分析，设置 P Val <0.001 且 log|差异倍数 (Fold Change, FC)| > 2 作为基本的筛选条件，筛选出瘢痕疙瘩倾向患者在受伤后Day42与正常人的明显表达差异的DE-circRNA分子，并通过火山图或热图进行可视化。

2. DE-circRNA富集分析及ceRNA网络预测：使用KEGG和GO分析，对DE-circRNA表达谱进行分析，进而了解DE-circRNA的亲本基因的相关定位与功能，从而了解瘢痕疙瘩中异常表达环状RNA的细胞成分。对所有差异性表达的circRNA与miRNA序列之间、miRNA与mRNA序列之间的进行分子间靶向结合的预测，使用miRDB、Targetscan、mirtarbase等多个数据库进行筛选，构建可能存在的“circRNA-miRNA-mRNA”形式的ceRNA网络。

3. 分子生物学实验验证DEcircRNA及ceRNA网络：基于表达谱及富集分析结果，筛选出亲本基因与瘢痕疙瘩进展最可能密切相关的4种DE-circRNA分子，包括circ_065059、circ_064002、circ_071194、circ_053469。同时收集正常皮肤、瘢痕疙瘩皮肤组织，使用RT-qPCR实验探究瘢痕疙瘩的组织内或细胞中相应DE-circRNA的表达情况，验证瘢痕疙瘩DE-circRNA表达谱的可靠性；

基于ceRNA的靶向分子的富集分析结果，筛选出与瘢痕疙瘩成纤维细胞功能可能密切相关的ceRNA网络之一“circ_064002-miR-30a/b-5p-纤连蛋白1（fibronectin1, FN1）”。设计相应双荧光素酶实验验证分子间的相互作用关系，从而证实此ceRNA网络存在的可靠性；

4. 实验验证DE-circ_064002通过ceRNA网络对瘢痕疙瘩成纤维细胞表达的调控作用：根据环状RNA的环化位点序列设计小干扰RNA（small interfering RNA, siRNA），构建miRNA的类似物（mimics）及抑制物（inhibitor），对KFs进行干扰，并通过RT-qPCR，WB等实验，验证KFs中circ_064002表达水平改变时出现FN1的表达变化。

4. 细胞学实验验证circ_064002对于瘢痕疙瘩成纤维细胞行为学功能的影响：使用CCK-8、划痕实验、Transwell实验，验证KFs中circ_064002低表达对成纤维细胞的增殖活性、修复、迁移、侵袭等功能的影响。

5.  DE-circRNA对于瘢痕疙瘩的生物标志物功能验证：收集瘢痕疙瘩、增生性瘢痕、扩张皮肤、正常皮肤组织，使用RT-qPCR实验检测DE-circRNA及其亲本基因最佳转录形成的mRNA表达量，并计算两者表达比值作为DE-circRNA剪接形成的比率，了解DE-circRNA及其形成比率对鉴别瘢痕疙瘩的指示作用。

梯度性地使用瘢痕疙瘩的病损内注射治疗药物——曲安奈德处理KFs，使用RT-qPCR实验探索DE-circRNA以及ceRNA网络中各分子变化，进而了解在曲安奈德注射治疗过程中DE-circRNA对于病情进展的指示作用，以及ceRNA网络在瘢痕疙瘩治疗、进展中的变化规律。

实验结果

1. DE-circRNA：通过生物信息学的方法，在NCBI的GSE113621数据库中筛选出29组在人瘢痕疙瘩中出现显著差异性表达的环状RNA以构建DE-circRNA表达谱；

2. 富集分析及ceRNA网络预测：通过富集分析了解此29组DE-circRNA的亲本基因主要富集在肿瘤代谢、年龄增长等领域；通过分子间相互作用的预测，建立了14组ceRNA网络。

3. DE-circRNA验证：4组DE-circRNA（circ_065059、circ_064002、circ_071194、circ_053469）在瘢痕疙瘩组织中出现高表达，与DE-circRNA表达谱的预测结果一致。

4. ceRNA网络验证：双荧光素酶实验提示细胞内circ_064002和FN1可以不同程度结合miR-30a-5p和miR-30b-5p，形成ceRNA网络的关系。

5. circ_064002对于瘢痕疙瘩成纤维细胞的调控：敲减circ_064002的表达能够抑制KFs中FN1在mRNA以及蛋白水平的表达，并降低KFs增殖活力、迁移、侵袭等行为学功能，与预测的ceRNA网络一致。

6. DE-circRNA对于瘢痕疙瘩的生物标志物作用：瘢痕疙瘩中DE-circRNA的剪接形成的比例明显高于扩张皮肤、增生性瘢痕、正常皮肤，可作为鉴别诊断瘢痕疙瘩的生物标志物之一。DE-cirRNA可以作为瘢痕疙瘩治疗进展的生物标志物。曲安奈德能够剂量依赖性地逆转瘢痕疙瘩KFs中DE-circRNA的异常表达的情况，同时剂量依赖性地降低circ_064002，从而进一步抑制KFs中FN1的表达。

结论

通过生信信息学技术，可以对网络生物数据库进行生物信息学的挖掘和分析，初步构建出瘢痕疙瘩中DE-circRNA表达谱，进而进行分子间结合的预测，构建瘢痕疙瘩成纤维细胞的ceRNA网络，通过对DE-circRNA分子的亲本基因进一步富集分析，可以进一步筛选出可能瘢痕疙瘩进展密切相关的DE-circRNA分子，为进一步的关键基因筛选提供了证据和基础。

研究证明circ_065059、circ_064002、circ_071194、circ_05346在瘢痕疙瘩中出现高表达；其中circ_064002可作为miR-30a/b-5p的分子海绵，对KFs中FN1的表达起到正向的调节作用，敲减circ_064002的表达可抑制KFs的中FN1的mRNA与蛋白的表达，同时抑制成纤维细胞增殖、迁移、细胞活性等功能状态。同时也证明生物信息学方法具有一定的可靠性。

DE-circRNA及其与亲本基因mRNA的比值可作为鉴别瘢痕疙瘩与增生性瘢痕以及反映瘢痕疙瘩治疗进展的生物指示物的生物标志物。而曲安奈德可以通过ceRNA网络，通过抑制circ_064002进一步下调瘢痕疙瘩进展的关键分子FN1的表达。

Background

Circular RNA (circRNA) is a product of reverse splicing of pre-mRNA, which is an RNA molecule with a single-stranded circular structure formed by the 5' and 3' ends of the precursor RNA linked to each other. As a type of non-coding RNA, circular RNA has been found to be able to act as a competitive endogenous RNA (Completing Endogenous RNA, ceRNA), which restricts the degradation of mRNA by miRNA by binding miRNA in cells, thereby playing a certain role in regulating the expression of specific genes. DE-circRNA) can play a role as a biomarker in the differential diagnosis of certain diseases and the determination of their progression status.

Organic fibrosis can occur in a variety of organs, including the heart, liver, lungs, kidneys, etc., and is pathologically manifested as the activation of various mesenchymal stem cells such as fibroblasts and hepatic astrocytes in the corresponding parts, resulting in a large accumulation of extracellular matrix (ECM) components in the interstitium of organ cells, and then tissue fibrosis occurs, and eventually causes functional failure. A growing body of data suggests that circular RNA is closely associated with fibrotic diseases in a variety of organs.

As a kind of fibrotic disorder, keloids are mainly pathological manifestations of mild skin damage or abnormal accumulation of extracellular matrix in the skin caused by no obvious cause, and in clinical practice, they are benign tumors that protrude on the body surface, gradually invade the periphery, and are accompanied by itching and painful symptoms. There are many therapeutic effects of keloids, but the treatment effect is poor, and the reason for this is that the progression mechanism of keloids is not fully understood at this stage.

In recent years, some studies have found that there are significant differences between the expression of circular RNA in keloids and normal skin, and some DE-circRNAs have a certain regulatory effect on the progression of keloids and the invasion of keloid-derived fibroblasts (KFs). However, at this stage, there is a lack of relevant systematic and big data research, which makes it impossible to screen and construct a large number of differentially expressed circular RNAs (DE-RNAs) in keloids, and there is also a lack of further exploration of the functions of cyclic RNA-related ceRNA networks and their biomarkers. Based on this situation, we conducted an in-depth exploration of keloid DE-circRNA.

Objective

1. The RNA database related to keloids was screened out, and the expression profile of DE-circRNA from human keloids was preliminarily obtained, and the DE-circRNA molecules that were more meaningful to the progression of keloids were further screened.

2. To construct a ceRNA regulatory network that may have the pattern of "DE-circRNA-miRNA-mRNA", and to understand the regulatory effect of DE-circRNA on keloid progression.

3. Molecules that may be more meaningful for keloid progression were selected from the predicted DE-circRNAs, and the reliability of the DEcircRNA expression profile in the aforementioned human keloids was verified by molecular biology methods.

4. The ceRNA regulatory network that may be more meaningful to the progression of keloids was selected, and the reliability of the aforementioned ceRNA regulatory network was verified by molecular biology methods.

5. Verify the reliability of the ceRNA regulatory network in the behavioral regulation of keloid fibroblasts by cytological methods.

6. To explore the role of DE-circRNA expression in the identification of keloids and the role of indicators in understanding the progression of keloids.

Methods and meterials

1. The NSE network biological database in the National Center Library of Medicine (NCBI) was screened to screen out the GSE113621 databases. This dataset is information collected from human skin wound samples 0 and 42 days post-injury in normal and people with a tendency to keloids. The multi-array averaging language of the limma language pack in R software was used for data standardization, and after converting the probe ID into circRNA symbols, the sample data in the GSE113621 database were analyzed, and P Val was set <0.001 and log|Fold Change, FC| > 2 was used as a basic screening condition to screen out DE-circRNA molecules with obvious differences in expression between Day42 and normal people in patients with keloid tendencies after injury, and visualized by volcano map or heat map. KEGG and GO analysis were used to analyze the expression profile of DE-circRNA to understand the relevant localization and function of the parental gene of DE-circRNA, so as to understand the cellular components of abnormally expressed circular RNA in keloids. The prediction of intermolecular target binding between all differentially expressed circRNAs and miRNA sequences, and between miRNA and mRNA sequences, was screened by multiple databases such as miRDB, Targetscan, and mirtarbase, and the possible ceRNA network in the form of "circRNA-miRNA-mRNA" was constructed.

2.  Based on the results of expression profile and enrichment analysis, four DE-circRNA molecules, including circ_065059, circ_064002, circ_071194 and circ_053469, were screened out for the most likely close relationship between parental genes and keloid progression. At the same time, normal skin and keloid skin tissues were collected, and the expression of the corresponding DE-circRNA in the tissues or cells of keloids was explored, and the reliability of the expression profile of keloid DE-circRNA was verified. Based on the results of enrichment analysis of the targeted molecules of ceRNA, one of the ceRNA networks that may be closely related to the function of keloid fibroblasts, "circ_064002-miR-30a/b-5p-fibronectin1 (FN1)", was screened. The corresponding dual-luciferase assay was designed to verify the interaction between molecules, so as to confirm the reliability of the existence of this ceRNA network.

3. Design small interfering RNA (siRNA), construct mimics and inhibitors of miRNA, interfere with KFs, and verify the expression of FN1 when the expression level of circ_064002 in KFs changes through RT-qPCR, WB and other experiments.

4. CCK-8, scratch assay and Transwell assay were used to verify the effect of low circ_064002 expression in KFs on the proliferative activity, repair, migration, invasion and other functions of fibroblasts.

5. Keloids, hypertrophic scars, dilated skin, and normal skin tissues were collected, and the mRNA expression levels of DE-circRNA and its parental genes were detected for optimal transcription, and the expression ratio of the two was calculated as the ratio of DE-circRNA splicing, so as to understand the indicative effect of DE-circRNA and its formation ratio in identifying keloids.

6. Triamcinolone acetonide was used to explore the molecular changes of DE-circRNA and ceRNA network, so as to understand the indicative effect of DE-circRNA on disease progression, and the changes of ceRNA network in the treatment and progression of keloids.

Results

1. Through the method of bioinformatics, we screened 29 groups of circular RNAs with significant differential expression in human keloid in the GSE113621 database of NCBI to construct the expression profile of DE-circRNA, and through enrichment analysis, we found that the parental genes of these 29 groups of DE-circRNA were mainly enriched in the fields of tumor metabolism and aging, and 14 groups of ceRNA networks were established through the prediction of intermolecular interactions.

2. 4 groups of DE-circRNA (circ_065059, circ_064002, circ_071194, circ_053469) were highly expressed in keloid tissues, which was consistent with the prediction results of DE-circRNA expression profile.

3. Dual luciferase assay showed that intracellular circ_064002 and FN1 could bind to miR-30a-5p and miR-30b-5p to varying degrees to form a ceRNA network.

4. The expression of knockdown circ_064002 can inhibit the expression of FN1 in KFs at the mRNA and protein levels, and reduce the behavioral functions such as proliferation, migration and invasion of KFs, which is consistent with the predicted ceRNA network.

5. The proportion of DE-circRNA splicing in keloids is significantly higher than that of dilated skin, hypertrophic scars, and normal skin, which can be used as one of the biomarkers for differential diagnosis of keloids. DE-cirRNA can be used as a biomarker for keloid treatment progression.

6. Triamcinolone acetonide can dose-dependently reverse the abnormal expression of DE-circRNA in keloid KFs, while decreasing the circ_064002 in a dose-dependent manner, thereby further inhibiting the expression of FN1 in KFs.

Conclusions

Through bioinformatics technology, bioinformatics mining and analysis can be carried out on the network biological database, and the expression profile of DE-circRNA in keloids can be preliminarily constructed, and then the intermolecular binding can be predicted, and the ceRNA network of keloid fibroblasts can be constructed。

Studies have shown that circ_065059, circ_064002, circ_071194 and circ_05346 are highly expressed in keloids, and circ_064002 can be used as molecular sponges for miR-30a/b-5p, which can positively regulate the expression of FN1 in KFs circ_ The expression of 064002 can inhibit the expression of FN1 mRNA and protein in KFs, and inhibit the proliferation, migration, cell viability and other functional states of fibroblasts. At the same time, it also proves that the bioinformatics method has a certain degree of reliability.

DE-circRNA and its ratio to parental mRNA can be used as biomarkers to distinguish keloids from hypertrophic scars, as well as biomarkers that reflect progress in keloid treatment. Triamcinolone acetonide can further downregulate the expression of FN1, a key molecule in keloid progression, by inhibiting circ_064002 through the ceRNA network.