Identifying the roles of hub gene in keloid formation using single-cell transcriptomics

Abstract

Introduction: Keloid, a fibroproliferative tumor characterized by excessive collagen deposition and fibroblast hyperplasia, lacks effective therapeutic strategies due to unclear molecular mechanisms.

Objective: This study aims to elucidate keloid pathogenesis and identify diagnostic biomarkers through multi-omics integration.

Methods: Single-cell RNA sequencing (ScRNA-seq) data (GSE163973) and bulk RNA sequencing datasets (GSE162904/GSE145725) were analyzed. Fibroblast subpopulations were identified using the Seurat R package, and cell–cell interactions were explored using the CellChat R package. Weighted gene co-expression network analysis (WGCNA) was employed to identify key gene modules in fibroblasts. Hub genes were screened using Lasso regression and validated through machine learning algorithms and a gene-immune convolutional neural network (CNN). Immune infiltration patterns were evaluated using the MCP-counter and Immuno-Oncology Biological Research R packages.

Results: ScRNA-seq analysis revealed eight distinct cell subtypes within keloid tissues, with fibroblasts significantly enriched compared to normal skin. Fibroblast clusters 1 and 5 exhibited elevated midkine–low-density lipoprotein receptor-related protein 1-mediated interactions and enhanced differentiation activity. WGCNA identified three critical modules-“brown,” “cyan,” and “yellow”-linked to fibroblast activation. Lasso regression produced an eight-gene signature that effectively distinguished keloid from normal skin (area under the curve = 0.885 – 0.889). Nonnegative matrix factorization classified keloids into four subtypes, each with distinct immune infiltration profiles correlated with hub gene expression. The gene-immune CNN model achieved 100% sensitivity and 88.9% specificity in diagnostic classification.

Conclusion: This study elucidates the molecular mechanisms underlying keloid formation through integrated single-cell and transcriptomic analysis, proposing an eight-gene signature as a potential diagnostic and therapeutic target. The identified keloid subtypes and associated immune infiltration patterns provide novel insights for advancing precision medicine approaches in keloid management.

Keywords

Cell communication / Deep learning / Differentiation trajectory / Immune infiltration / Keloid / Single-cell RNA

Acknowledgements

We thank the authors of GSE163973, GSE162904, and GSE145725 for their contribution and the Metascape database for providing the platform for data analysis.

Funding

This study was funded by the Medical Youth Innovation Research Project Plan of Sichuan Province of the Sichuan Medical Association (S20091).

Conflict of interest

The authors declare no conflicts of interest.

Author contributions

Conceptualization: Yong Zou

Formal analysis: Bing Liu, Jun Yin

Funding acquisition: Zhengqiang Wan

Investigation: Yuzhen Ma

Methodology: Cheng Chen, Jun Yin

Writing–original draft: Xiangbing Zheng, Zhengqiang Wan

Writing–review & editing: Yong Zou, Xiangbing Zheng

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Second People’s Hospital of Yibin (Approval no. 2024-007-01). The data used in this study were obtained from the public Gene Expression Omnibus database; therefore, patients were not required to sign an informed consent form for this study.

Consent for publication

Not applicable.

Availability of data

The authors confirm that the data supporting the findings of this study are available in the article and its supplementary file.

References

[1] Jing P. The Role of Splicing Factor SRSF2 in Keloid Fibroblasts. [Master’s Thesis]; 2019. p. 79.

[2] Asilian A, Darougheh A, Shariati F. New combination of triamcinolone, 5-fluorouracil, and pulsed-dye laser for treatment of keloid and hypertrophic scars. Dermatol Surg. 2006; 32: 907-915. doi: 10.1111/j.1524-4725.2006.32195.x

[3] Shih B, Bayat A. Genetics of keloid scarring. Arch Dermatol Res. 2010; 302: 319-339. doi: 10.1007/s00403-009-1014-y

[4] Supp DM, Hahn JM, Glaser K, McFarland KL, Boyce ST. Deep and superficial keloid fibroblasts contribute differentially to tissue phenotype in a novel in vivo model of keloid scar. Plast Reconstr Surg. 2012; 129: 1259-1271. doi: 10.1097/PRS.0b013e31824ecaa9

[5] Lim IJ, Phan TT, Bay BH, et al. Fibroblasts cocultured with keloid keratinocytes: Normal fibroblasts secrete collagen in a keloidlike manner. Am J Physiol Cell Physiol. 2002; 283: C212-C222. doi: 10.1152/ajpcell.00555.2001

[6] Iqbal SA, Sidgwick GP, Bayat A. Identification of fibrocytes from mesenchymal stem cells in keloid tissue: A potential source of abnormal fibroblasts in keloid scarring. Arch Dermatol Res. 2012; 304: 665-671. doi: 10.1007/s00403-012-1225-5

[7] Ueda K, Furuya E, Yasuda Y, Oba S, Tajima S. Keloids have continuous high metabolic activity. Plast Reconstr Surg. 1999; 104: 694-698. doi: 10.1097/00006534-199909030-00012

[8] Van De Water L, Varney S, Tomasek JJ. Mechanoregulation of the myofibroblast in wound contraction, scarring, and fibrosis: Opportunities for new therapeutic intervention. Adv Wound Care (New Rochelle). 2013; 2: 122-141. doi: 10.1089/wound.2012.0393

[9] Tholpady SS, DeGeorge BR Jr., Campbell CA. The effect of local rho-kinase inhibition on murine wound healing. Ann Plast Surg. 2014; 72: S213-S219. doi: 10.1097/SAP.0000000000000150

[10] Kao HK, Chen B, Murphy GF, Li Q, Orgill DP, Guo L. Peripheral blood fibrocytes: Enhancement of wound healing by cell proliferation, re-epithelialization, contraction, and angiogenesis. Ann Surg. 2011; 254: 1066-1074. doi: 10.1097/SLA.0b013e3182251559

[11] Wolfram D, Tzankov A, Pulzl P, Piza-Katzer H. Hypertrophic scars and keloids-a review of their pathophysiology, risk factors, and therapeutic management. Dermatol Surg. 2009; 35: 171-181. doi: 10.1111/j.1524-4725.2008.34406.x

[12] Van Leeuwen MC, Stokmans SC, Bulstra AJ, Meijer OW, Van Leeuwen PA, Niessen FB. High-dose-rate brachytherapy for the treatment of recalcitrant keloids: A unique, effective treatment protocol. Plast Reconstr Surg. 2014; 134: 527-534. doi: 10.1097/PRS.0000000000000415

[13] Halim AS, Emami A, Salahshourifar I, Kannan TP. Keloid scarring: Understanding the genetic basis, advances, and prospects. Arch Plast Surg. 2012; 39: 184-189. doi: 10.5999/aps.2012.39.3.184

[14] Wu Y, Wang B, Li YH, et al. Meta-analysis demonstrates association between arg72pro polymorphism in the p53 gene and susceptibility to keloids in the Chinese population. Genet Mol Res. 2012; 11: 1701-1711. doi: 10.4238/2012.June.29.2

[15] Messadi DV, Doung HS, Zhang Q, et al. Activation of nfkappab signal pathways in keloid fibroblasts. Arch Dermatol Res. 2004; 296: 125-133. doi: 10.1007/s00403-004-0487-y

[16] Lu F, Gao J, Ogawa R, Hyakusoku H, Ou C. Fas-mediated apoptotic signal transduction in keloid and hypertrophic scar. Plast Reconstr Surg. 2007; 119: 1714-1721. doi: 10.1097/01.prs.0000258851.47193.06

[17] Hu Z, Lou L, Luo S. [Experimental study of the expression of c-myc, c-fos and proto-oncogenes on hypertrophic and scars]. Zhonghua Zheng Xing Wai Ke Za Zhi. 2002; 18: 165-167.

[18] Teofoli P, Barduagni S, Ribuffo M, Campanella A, De Pita’O, Puddu P. Expression of bcl-2, p53, c-jun and c-fos protooncogenes in keloids and hypertrophic scars. J Dermatol Sci. 1999; 22: 31-37. doi: 10.1016/s0923-1811(99)00040-7

[19] Jin Z. Increased c-met phosphorylation is related to keloid pathogenesis: Implications for the biological behaviour of keloid fibroblasts. Pathology. 2014; 46: 25-31. doi: 10.1097/PAT.0000000000000028

[20] Xiaoyang M. Study on hypoxia/HIF-1α-induced EMT in keloid keratinocytes and its effect on invasiveness [Doctoral dissertation]. Peking Union Medical College; 2015.

[21] Aran D, Looney AP, Liu L, et al. Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. Nat Immunol. 2019; 20: 163-172. doi: 10.1038/s41590-018-0276-y

[22] Morabito S, Reese F, Rahimzadeh N, Miyoshi E, Swarup V. Hdwgcna identifies co-expression networks in high-dimensional transcriptomics data. Cell Rep Methods. 2023; 3: 100498. doi: 10.1016/j.crmeth.2023.100498

[23] Jin S, Guerrero-Juarez CF, Zhang L, et al. Inference and analysis of cell-cell communication using cellchat. Nat Commun. 2021; 12: 1088. doi: 10.1038/s41467-021-21246-9

[24] Trapnell C, Cacchiarelli D, Grimsby J, et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat Biotechnol. 2014; 32: 381-386. doi: 10.1038/nbt.2859

[25] Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization. Nature. 1999; 401: 788-791. doi: 10.1038/44565

[26] van Merriënboer B, Bahdanau D, Dumoulin V, et al. Blocks and Fuel: Frameworks for Deep Learning. arXiv preprint arXiv:1506.00619; 2015. doi: 10.48550/arXiv.1506.00619

[27] Heping H. Study on the Expression of Hedgehog Signaling Pathway and EMT-related lncRNAs in Keloids [Doctoral Dissertation]. Nanchang University; 2018.

[28] Vincent AS, Phan TT, Mukhopadhyay A, Lim HY, Halliwell B, Wong KP. Human skin keloid fibroblasts display bioenergetics of cancer cells. J Invest Dermatol. 2008; 128: 702-709. doi: 10.1038/sj.jid.5701107

[29] Murray JC. Keloids and hypertrophic scars. Clin Dermatol. 1994; 12: 27-37. doi: 10.1016/0738-081x(94)90254-2

[30] Slemp AE, Kirschner RE. Keloids and scars: A review of keloids and scars, their pathogenesis, risk factors, and management. Curr Opin Pediatr. 2006; 18: 396-402. doi: 10.1097/01.mop.0000236389.41462.ef

[31] Yahong C, Xiaoli W. Research progress in the mechanism of invasive growth of keloid. J Tissue Eng Reconstr Surg. 2015; 11: 335-338.

[32] Gang Z, Shaojun L, Shaoming T, Jie L, Qiang Z. Expression of BCL-2 in different parts of keloid and its significance. Guangdong Med. 2006: 1811-1812.

[33] Sidgwick GP, Bayat A. Extracellular matrix molecules implicated in hypertrophic and keloid scarring. J Eur Acad Dermatol Venereol. 2012; 26: 141-152. doi: 10.1111/j.1468-3083.2011.04200.x

[34] Placik OJ, Lewis VJ Jr. Immunologic associations of keloids. Surg Gynecol Obstet. 1992; 175: 185-193.

[35] Niessen FB, Schalkwijk J, Vos H, Timens W. Hypertrophic scar formation is associated with an increased number of epidermal langerhans cells. J Pathol. 2004; 202: 121-129. doi: 10.1002/path.1502

[36] Han B. Study on the Effect and Molecular Mechanism of Paracrine Factors from Adipose-Derived Mesenchymal Stem Cells against Fibrosis of Pathological Scar Fibroblasts [Doctoral Dissertation]. Peking Union Medical College; 2018.