Mingjie Shi,1,2 Yue Wei,3 Runmin Guo,1,2 Fei Luo1,2 

1Key Laboratory of Research in Maternal and Child Medicine and Birth Defects, Guangdong Medical University, Foshan, Guangdong, People’s Republic of China; 2Matenal and Child Research Institute, Shunde Women and Children’s Hospital (Maternity and Child Healthcare Hospital of Shunde Foshan), Guangdong Medical University, Foshan, People’s Republic of China; 3Department of Ultrasound, Shunde Women and Children’s Hospital (Maternity and Child Healthcare Hospital of Shunde Foshan), Guangdong Medical University, Foshan, People’s Republic of China

Correspondence: Fei Luo, Shunde Women and Children’s Hospital (Maternity and Child Healthcare Hospital of Shunde Foshan), Guangdong Medical University, Foshan, People’s Republic of China, Email luofei_00@163.com

Background: Sepsis is a major contributor to morbidity and mortality among hospitalized patients. This study aims to identify markers associated with the severity and prognosis of sepsis, providing new approaches for its management and treatment.
Methods: Data were mined from the Gene Expression Omnibus (GEO) databases and were analyzed by multiple statistical methods like the Spearman correlation coefficient, Kaplan-Meier analysis, Cox regression analysis, and functional enrichment analysis. Candidate indicator’ associations with immune infiltration and roles in sepsis development were evaluated. Additionally, we employed techniques such as flow cytometry and neutral red staining to evaluate its impact on macrophage functions like polarization and phagocytosis.
Results: Twenty-eight genes were identified as being closely linked to the severity of sepsis, among which transforming growth factor beta induced (TGFBI) emerged as a distinct marker for predicting clinical outcomes. Notably, reductions in TGFBI expression during sepsis correlate with poor prognosis and rapid disease progression. Elevated expression of TGFBI has been observed to mitigate abnormalities in sepsis-related immune cell infiltration that are critical to the pathogenesis and prognosis of the disease, including but not limited to type 17 T helper cells and activated CD8 T cells. Moreover, the protein-protein interaction network revealed the top ten genes that interact with TGFBI, showing significant involvement in the regulation of the actin cytoskeleton, extracellular matrix-receptor interactions, and phagosomes. These are pivotal elements in the formation of phagocytic cups by macrophages, squaring the findings of the Human Protein Atlas. Additionally, we discovered that TGFBI expression was significantly higher in M2-like macrophages, and its upregulation was found to inhibit lipopolysaccharide-induced polarization and phagocytosis in M1-like macrophages, thereby playing a role in preventing the onset of inflammation.
Conclusion: TGFBI warrants additional exploration as a promising biomarker for assessing illness severity and prognosis in patients with sepsis, considering its significant association with immunological and inflammatory responses in this condition.

Keywords: sepsis, septic shock, TGFBI, prognosis, biomarker