Transcriptomic Profiles of Sepsis in the Human Brain

Identifying hub genes of sepsis-associated and hepatic encephalopathies based on bioinformatic analysis-focus on the two common encephalopathies of septic cirrhotic patients in ICU

BACKGROUND

In the ICU ward, septic cirrhotic patients are susceptible to suffering from sepsis-associated encephalopathy and/or hepatic encephalopathy, which are two common neurological complications in such patients. However, the mutual pathogenesis between sepsis-associated and hepatic encephalopathies remains unclear. We aimed to identify the mutual hub genes, explore effective diagnostic biomarkers and therapeutic targets for the two common encephalopathies and provide novel, promising insights into the clinical management of such septic cirrhotic patients.

METHODS

The precious human post-mortem cerebral tissues were deprived of the GSE135838, GSE57193, and GSE41919 datasets, downloaded from the Gene Expression Omnibus database. Furthermore, we identified differentially expressed genes and screened hub genes with weighted gene co-expression network analysis. The hub genes were then subjected to Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway functional enrichment analyses, and protein-protein interaction networks were constructed. Receiver operating characteristic curves and correlation analyses were set up for the hub genes. Finally, we explored principal and common signaling pathways by using Gene Set Enrichment Analysis and the association between the hub genes and immune cell subtype distribution by using CIBERSORT algorithm.

RESULTS

We identified seven hub genes-GPR4, SOCS3, BAG3, ZFP36, CDKN1A, ADAMTS9, and GADD45B-by using differentially expressed gene analysis and weighted gene co-expression network analysis method. The AUCs of these genes were all greater than 0.7 in the receiver operating characteristic curves analysis. The Gene Set Enrichment Analysis results demonstrated that mutual signaling pathways were mainly enriched in hypoxia and inflammatory response. CIBERSORT indicated that these seven hub genes were closely related to innate and adaptive immune cells.

CONCLUSIONS

We identified seven hub genes with promising diagnostic value and therapeutic targets in septic cirrhotic patients with sepsis-associated encephalopathy and/or hepatic encephalopathy. Hypoxia, inflammatory, and immunoreaction responses may share the common downstream pathways of the two common encephalopathies, for which earlier recognition and timely intervention are crucial for management of such septic cirrhotic patients in ICU.

Identification of IL-8 in CSF as a potential biomarker in sepsis-associated encephalopathy

BACKGROUND

Sepsis-associated encephalopathy (SAE) is frequently present at the acute and chronic phase of sepsis, which is characterized by delirium, coma, and cognitive dysfunction. Despite the increased morbidity and mortality of SAE, the pathogenesis of SAE remains unclear. This study aims to discover the potential biomarkers, so as to clear the pathogenesis potentially contributing to the development of SAE and provide new therapeutic strategies for the treatment of SAE.

METHODS

The GSE135838 dataset was obtained from the Gene Expression Omnibus (GEO) database and utilized for analysis the differentially expressed genes (DEGs). The DEGs were analyzed by limma package of R language and the extracellular protein-differentially expressed genes (EP-DEGs) were screened by the Human Protein Atlas (HPA) and UniProt database. Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were carried out to analyze the function and pathway of EP-DEGs. STRING, Cytoscape, MCODE and Cytohubba were used to construct a protein-protein interaction (PPI) network and screen key EP-DEGs. Key EP-DEGs levels were detected in the cerebrospinal fluid (CSF) of SAE patients and non-sepsis patients with critical illness. ROC curve was used to evaluate the diagnostic of SAE.

RESULTS

We screened 82 EP-DEGs from DEGs. EP-DEGs were enriched in cytokine-cytokine receptor interaction, IL-17 signaling pathway and NOD-like receptor signaling pathway. We identified 2 key extracellular proteins IL-1B and IL-8. We clinically verified that IL-6 and IL-8 levels were increased in CSF of SAE patients and CSF IL-8 (AUC = 0.882, 95 % CI = 0.775-0.988) had a higher accuracy in the diagnosis of SAE than CSF IL-6 (AUC = 0.824, 95 % CI = 0.686-0.961). Furthermore, we found that the IL-8 levels in CSF might not associated with Glasgow Coma Scale (GCS) scores of SAE patients.

CONCLUSION

IL-8 may be the key extracellular cytokine in the pathogenesis of SAE. Bioinformatics methods were used to explore the biomarkers of SAE and validated the results in clinical samples. Our findings indicate that the IL-8 in CSF might be the potential diagnostic biomarker and therapeutic target in SAE.

Microglial Activation: Key Players in Sepsis-Associated Encephalopathy

Sepsis-associated encephalopathy (SAE) is a common brain dysfunction, which results in severe cognitive and neurological sequelae and an increased mortality rate in patients with sepsis. Depending on the stimulus, microglia (resident macrophages in the brain that are involved in SAE pathology and physiology) can adopt two polarization states (M1/M2), corresponding to altered microglial morphology, gene expression, and function. We systematically described the pathogenesis, morphology, function, and phenotype of microglial activation in SAE and demonstrated that microglia are closely related to SAE occurrence and development, and concomitant cognitive impairment. Finally, some potential therapeutic approaches that can prime microglia and neuroinflammation toward the beneficial restorative microglial phenotype in SAE were outlined.

Transcriptome analysis of six tissues obtained post-mortem from sepsis patients

Septic shock is a life-threatening clinical condition characterized by a robust immune inflammatory response to disseminated infection. Little is known about its impact on the transcriptome of distinct human tissues. To address this, we performed RNA sequencing of samples from the prefrontal cortex, hippocampus, heart, lung, kidney and colon of seven individuals who succumbed to sepsis and seven uninfected controls. We identified that the lungs and colon were the most affected organs. While gene activation dominated, strong inhibitory signals were also detected, particularly in the lungs. We found that septic shock is an extremely heterogeneous disease, not only when different individuals are investigated, but also when comparing different tissues of the same patient. However, several pathways, such as respiratory electron transport and other metabolic functions, revealed distinctive alterations, providing evidence that tissue specificity is a hallmark of sepsis. Strikingly, we found evident signals of accelerated ageing in our sepsis population.

Brain Epitranscriptomic Analysis Revealed Altered A-to-I RNA Editing in Septic Patients

Recent studies suggest that RNA editing is associated with impaired brain function and neurological and psychiatric disorders. However, the role of A-to-I RNA editing during sepsis-associated encephalopathy (SAE) remains unclear. In this study, we analyzed adenosine-to-inosine (A-to-I) RNA editing in postmortem brain tissues from septic patients and controls. A total of 3024 high-confidence A-to-I RNA editing sites were identified. In sepsis, there were fewer A-to-I RNA editing genes and editing sites than in controls. Among all A-to-I RNA editing sites, 42 genes showed significantly differential RNA editing, with 23 downregulated and 19 upregulated in sepsis compared to controls. Notably, more than 50% of these genes were highly expressed in the brain and potentially related to neurological diseases. Notably, cis-regulatory analysis showed that the level of RNA editing in six differentially edited genes was significantly correlated with the gene expression, including HAUS augmin-like complex subunit 2 (HAUS2), protein phosphatase 3 catalytic subunit beta (PPP3CB), hook microtubule tethering protein 3 (HOOK3), CUB and Sushi multiple domains 1 (CSMD1), methyltransferase-like 7A (METTL7A), and kinesin light chain 2 (KLC2). Furthermore, enrichment analysis showed that fewer gene functions and KEGG pathways were enriched by edited genes in sepsis compared to controls. These results revealed alteration of A-to-I RNA editing in the human brain associated with sepsis, thus providing an important basis for understanding its role in neuropathology in SAE.

Brain innate immune response via miRNA-TLR7 sensing in polymicrobial sepsis

Sepsis-associated encephalopathy (SAE) occurs in sepsis survivors and is associated with breakdown of the blood-brain barrier (BBB), brain inflammation, and neurological dysfunction. We have previously identified a group of extracellular microRNAs (ex-miRNAs), such as miR-146a-5p, that were upregulated in the plasma of septic mice and human, and capable of inducing potent pro-inflammatory cytokines and complements. Here, we established a clinically relevant mouse model of SAE and investigated the role of extracellular miRNAs and their sensor Toll-like receptor 7 (TLR7) in brain inflammation and neurological dysfunction. We observed BBB disruption and a profound neuroinflammatory responses in the brain for up to 14 days post-sepsis; these included increased pro-inflammatory cytokines production, microglial expansion, and peripheral leukocyte accumulation in the CNS. In a battery of neurobehavioral tests, septic mice displayed impairment of motor coordination and neurological function. Sepsis significantly increased plasma RNA and miRNA levels for up to 7 days, such as miR-146a-5p. Exogenously added miR-146a-5p induces innate immune responses in both cultured microglia/astrocytes and the intact brain via a TLR7-dependent manner. Moreover, mice genetically deficient of miR-146a showed reduced accumulation of monocytes and neutrophils in the brain compared to WT after sepsis. Finally, ablation of TLR7 in the TLR7^-/- mice preserved BBB integrity, reduced microglial expansion and leukocyte accumulation, and attenuated GSK3β signaling in the brain, but did not improve neurobehavioral recovery following sepsis. Taken together, these data establish an important role of extracellular miRNA and TLR7 sensing in sepsis-induced brain inflammation.

Commonly expressed key transcriptomic profiles of sepsis in the human circulation and brain via integrated analysis

BACKGROUND

Sepsis is the leading cause of death in intensive care units and is characterized by multiple organ failure, including dysfuction of the immune system and brain. This study aims to determine the differential effect of sepsis on specific circulating immune cell subsets compared with brain transcriptome and identify the genes co-expressed by them, so as to identify key genes and regulatory factors involved in the pathogenesis of sepsis induced brain injury and identify novel therapeutic targets.

METHODS

The GSE133822 and GSE135838 datasets were obtained from the Gene Expression Omnibus (GEO) database and utilized for bioinformatics analyses. Functional enrichment analysis was used to identify commonly expressed genes that were differentially expressed between sepsis patients and non-sepsis patients with critical illness; protein-protein interaction (PPI) networks were also generated. Then, key transcriptomic biomarkers were further validated in an external dataset from the GEO. We also investigated the expression of key mRNAs in peripheral blood mononuclear cells (PBMCs) from sepsis patients by quantitative PCR (qPCR) and an in-vitro model stimulated by lipopolysaccharide (LPS) was generated in brain cell lines.

RESULTS

The transcriptomic profiles of brain tissue were relatively similar as those of specific immune cells. In addition, our validation showed that these key genes were up regulated both in PBMCs in sepsis patients and LPS-treated brain cells.

CONCLUSION

Brain injury in sepsis was correlated with circulating immune responses, and the expression of DEFA3, MMP8, MMP9 and LCN2 might be potential diagnostic biomarkers as well as therapeutic target in septic brain dysfunction.