Integrative analysis of metabolomics and transcriptomics to uncover biomarkers in sepsis

Abstract

To utilize metabolomics in conjunction with RNA sequencing to identify biomarkers in the blood of sepsis patients and discover novel targets for diagnosing and treating sepsis. In January 2019 and December 2020, blood samples were collected from a cohort of 16 patients diagnosed with sepsis and 11 patients diagnosed with systemic inflammatory response syndrome (SIRS). Non-targeted metabolomics analysis was conducted using liquid chromatography coupled with mass spectrometry (LC-MS/MS technology), while gene sequencing was performed using RNA sequencing. Afterward, the metabolite data and sequencing data underwent quality control and difference analysis, with a fold change (FC) greater than or equal to 2 and a false discovery rate (FDR) less than 0.05.Co-analysis was then performed to identify differential factors with consistent expression trends based on the metabolic pathway context; KEGG enrichment analysis was performed on the crossover factors, and Meta-analysis of the targets was performed at the transcriptome level using the public dataset. In the end, a total of five samples of single nucleated cells from peripheral blood (two normal controls, one with systemic inflammatory response syndrome, and two with sepsis) were collected and examined to determine the cellular location of the essential genes using 10× single cell RNA sequencing (scRNA-seq). A total of 485 genes and 1083 metabolites were found to be differentially expressed in the sepsis group compared to the SIRS group. Among these, 40 genes were found to be differentially expressed in both the metabolome and transcriptome. Functional enrichment analysis revealed that these genes were primarily involved in biological processes related to inflammatory response, immune regulation, and amino acid metabolism. Furthermore, a meta-analysis identified four genes, namely ITGAM, CD44, C3AR1, and IL2RG, which were highly expressed in the sepsis group compared to the normal group (P < 0.05). Additionally, scRNA-seq analysis revealed that the core genes ITGAM and C3AR1 were predominantly localized within the macrophage lineage. The primary genes ITGAM and C3AR1 exhibit predominant expression in macrophages, which play a significant role in inflammatory and immune responses. Moreover, these genes show elevated expression levels in the plasma of individuals with sepsis, indicating their potential as valuable subjects for further research in sepsis.

Keywords: Biomarkers; Metabolomics; RNA sequencing; Sepsis; Single-cell RNA sequencing.

Figure 1

Genomic data quality control and differential screening. (a) The y-axis of the box plot represents the logarithmised FPKM (number of fragments per thousand exon patterns per million mapped fragments), also known as log10(FPKM), which indicates that the data for each sample are homogenised, distributed at the same level and comparable; (b) principal component analysis shows that the two groups can be clearly differentiated and have no outliers; (c) the bar graphs show that the differential analysis screened out up-regulated (yellow) 184 and down-regulated (green) 301 genes; (d) volcano plot showing the up-regulated (yellow) and down-regulated (green) genes screened by differential analysis, with horizontal coordinates for gene expression in the sepsis group and vertical coordinates for gene expression in the SIRS group.

Figure 2

Metabolomic data allegation and difference screening. (a) Principal component analysis showed that the two groups were clearly distinguishable with no outliers; (b) OPLS-DA analysis showed a clear separation between sepsis patients and SIRS subjects, indicating significant differences in their serum non-targeted metabolomics profiles; (c) volcano plots showed that the differential analysis screened for up-regulated (yellow) 831 and down-regulated (green) 1083 metabolites, with the horizontal coordinates being the sepsis group's gene expression and vertical coordinates for the SIRS group.

Figure 3

Intergroup analysis (a) intergroup correlation heatmap showing that sepsis patients differ significantly from SIRS subjects in terms of genomic and non-targeted metabolomic data; genes characterised based on metabolic pathway (b) co-regulation are mainly involved in haematopoietic cell profiles, Th17 cell differentiation, Th1 and Th2 cell differentiation, cellular adhesion molecules, Staphylococcus aureus infection, graft-versus-host disease (b) metabolites are mostly enriched in glutathione metabolism, mucin-type O-glycan biosynthesis, fructose and mannose metabolism, and cysteine and methionine metabolism. It has an important role in energy production and human immunity.

Figure 4

(A–D) represents ITGAM, CD44, C3AR1, IL2RG genes in GSE6535, GSE12624, GSE28750, GSE63042, GSE74224 of SIRS and sepsis groups respectively, the four core genes were lowly expressed in the normal group, and highly expressed in the sepsis group, the difference was statistically significant (P < 0.05).

Figure 5

Single-cell RNA sequencing. (a) General diagram of mixed sample sequencing. Cell populations 1, 2, 6 and 8 are T cells, 3 and 5 are macrophages, 4 are NK cells, 7 are B cells and 9 are platelets. (c, e) Suggests that the core genes ITGAM and C3AR1 are mainly localised in cell populations 3 and 5, i.e. the macrophage lineage. (b, d) The core genes CD44 and IL2RG are widely localised to various cell populations and are expressed to varying degrees in all types of cell populations.