A longitudinal study highlights shared aspects of the transcriptomic response to cardiogenic and septic shock

Abstract

Background: Septic shock (SS) and cardiogenic shock (CS) are two types of circulatory shock with a different etiology. Several studies have described the molecular alterations in SS patients, whereas the molecular factors involved in CS have been poorly investigated. We aimed to assess in the whole blood of CS and SS patients, using septic patients without shock (SC) as controls, transcriptomic modifications that occur over 1 week after ICU admission and are common to the two types of shock.

Methods: We performed whole blood RNA sequencing in 21 SS, 11 CS, and 5 SC. In shock patients, blood samples were collected within 16 h from ICU admission (T1), 48 h after ICU admission (T2), and at day 7 or before discharge (T3). In controls, blood samples were available at T1 and T2. Gene expression changes over time have been studied in CS, SS, and SC separately with a paired analysis. Genes with p value < 0.01 (Benjamini-Hochberg multiple test correction) were defined differentially expressed (DEGs). We used gene set enrichment analysis (GSEA) to identify the biological processes and transcriptional regulators significantly enriched in both types of shock.

Results: In both CS and SS patients, GO terms of inflammatory response and pattern recognition receptors (PRRs) were downregulated following ICU admission, whereas gene sets of DNA replication were upregulated. At the gene level, we observed that alarmins, interleukin receptors, PRRs, inflammasome, and DNA replication genes significantly changed their expression in CS and SS, but not in SC. Analysis of transcription factor targets showed in both CS and SS patients, an enrichment of CCAAT-enhancer-binding protein beta (CEBPB) targets in genes downregulated over time and an enrichment of E2F targets in genes with an increasing expression trend.

Conclusions: This pilot study supports, within the limits of a small sample size, the role of alarmins, PRRs, DNA replication, and immunoglobulins in the pathophysiology of circulatory shock, either in the presence of infection or not. We hypothesize that these genes could be potential targets of therapeutic interventions in CS and SS.

Trial registration: ClinicalTrials.gov, NCT02141607. Registered 19 May 2014.

Keywords: Cardiogenic shock; Circulatory shock; Critical illness; Immunoglobulin; PRR; RNA-Seq; Septic shock.

Fig. 1

PCA plot of CS and SS patients. PCA plot of CS patients at three time points (a). PCA plot of SS patients at three time points (b). PCA plot of SC patients at T1 and T2 (c). PCs were adjusted in order to remove the patient effect

Fig. 2

Inflammatory pathway, pattern recognition receptors, and DNA replication. Expression trends of significantly modulated genes in CS and SS, but not in SC patients. Data are normalized on T1; log₂FoldChanges are plotted

Fig. 3

Gene expression trends of enriched transcription factor targets in CS and SS. CEBPB and E2F target genes are shown. Data are normalized on T1; log₂FoldChanges are plotted

Fig. 4

Boxplots of C-reactive protein serum measurements. C-reactive protein measurements (mg/L) in cardiogenic and septic shock patients measured at three time points. The lower and upper hinges correspond to the 25th and 75th percentiles, respectively