Zhiyong Tan,1– 3,* Xiaorong Chen,4,* Shi Fu,1– 3,* Yinglong Huang,1– 3,* Haihao Li,1– 3,* Chen Gong,1– 3 Dihao Lv,1– 3 Chadanfeng Yang,1– 3 Jiansong Wang,1– 3 Mingxia Ding,1– 3 Haifeng Wang1– 3 

1Department of Urology, The Second Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, People’s Republic of China; 2Urological Disease Clinical Medical Center of Yunnan Province, The Second Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, People’s Republic of China; 3Scientific and Technological Innovation Team of Basic and Clinical Research of Bladder Cancer in Yunnan Universities, The Second Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, People’s Republic of China; 4Department of Kidney Transplantation, The Third Hospital of Sun Yat-Sen University, Guangzhou, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Mingxia Ding, Department of Urology, The Second Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, People’s Republic of China, Email dingmingxia@kmmu.edu.cn Haifeng Wang, Department of Urology, The Second Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, People’s Republic of China, Email t13085348360@163.com

Introduction: Bladder cancer (BCa) is one of the top ten most common cancers, yet its underlying mechanisms remain unclear. This study aimed to explore the potential molecular mechanisms of BCa using multi-omics and single-cell analysis.
Methods: First, differential analysis of transcriptome data related to BCa from public databases was performed, and a risk model was then developed using 101 different machine learning algorithms to determine prognostic genes, followed by independent prognostic analysis to construct a nomogram. Immune infiltration analysis was performed to explore the impact of prognostic genes on the tumor microenvironment. Metabolomics, proteomics, and post-translational modification data from BCa tumor and adjacent non-tumor tissues were used to explore the relationships between prognostic genes and various omics levels. Finally, single-cell analysis identified key cells involved in BCa pathogenesis, and in vitro experiments validated the expression and function of key genes.
Results: The risk model constructed by 8 prognostic genes identified using 101 algorithms effectively predicted the survival outcomes of BCa patients. Furthermore, risk scores, pathological T stage, and pathological N stage were confirmed as independent prognostic factors for the nomogram construction. Interestingly, high-risk patients showed a significantly lower response to PD-L1 treatment, with higher TIDE scores. Omics analysis revealed a close relationship between prognostic genes and proteomics, metabolomics, and post-translational modifications. Specifically, FLNC and MYH11 may influence BCa progression through phosphorylation and succinylation. Single-cell analysis identified fibroblasts as key cells in BCa. Functional experiments showed that MYH11 knockdown promoted cell proliferation, migration, and invasion.
Conclusion: This study identified 8 prognostic genes to construct a risk model, and suggest that MYH11 is a potential diagnostic and prognostic biomarker for BCa.

Keywords: bladder cancer, MYH11, multi-omics, single-cell sequencing analysis, therapeutic target