Surgical Trauma in Mice Modifies the Content of Circulating Extracellular Vesicles

Abstract

Surgical interventions rapidly trigger a cascade of molecular, cellular, and neural signaling responses that ultimately reach remote organs, including the brain. Using a mouse model of orthopedic surgery, we have previously demonstrated hippocampal metabolic, structural, and functional changes associated with cognitive impairment. However, the nature of the underlying signals responsible for such periphery-to-brain communication remains hitherto elusive. Here we present the first exploratory study that tests the hypothesis of extracellular vesicles (EVs) as potential mediators carrying information from the injured tissue to the distal organs including the brain. The primary goal was to investigate whether the cargo of circulating EVs after surgery can undergo quantitative changes that could potentially trigger phenotypic modifications in the target tissues. EVs were isolated from the serum of the mice subjected to a tibia surgery after 6, 24, and 72 h, and the proteome and miRNAome were investigated using mass spectrometry and RNA-seq approaches. We found substantial differential expression of proteins and miRNAs starting at 6 h post-surgery and peaking at 24 h. Interestingly, one of the up-regulated proteins at 24 h was α-synuclein, a pathogenic hallmark of certain neurodegenerative syndromes. Analysis of miRNA target mRNA and corresponding biological pathways indicate the potential of post-surgery EVs to modify the extracellular matrix of the recipient cells and regulate metabolic processes including fatty acid metabolism. We conclude that surgery alters the cargo of circulating EVs in the blood, and our results suggest EVs as potential systemic signal carriers mediating remote effects of surgery on the brain.

Keywords: alpha-synuclein; circulating extracellular vesicles; miRNA; proteomics; surgery.

Figure 1

Flowchart of the study design. Mice were subjected to tibia surgery, blood was collected from the control and surgery animals after 6, 24, and 72 h post-surgery. Isolated blood serum was used for the purification of EVs using the ExoQuick ULTRA kit. EVs were characterized by Nanoparticle Tracking Analysis (NTA), transmission electron microscopy (TEM), and identification of EV enriched proteins by western blot. EVs’ proteome and miRNAome were investigated by LC/MS and RNA-seq and findings were validated by qPCR and western blot.

Figure 2

Characterization of circulating EVs. (A) TEM image of the isolated particles, scale bar, 500 nm. Encircled are representative spherical particles. White arrows indicate putative lipid bilayers of EVs. The dotted insert is represented in higher magnification on the right panel. (B) Representative size distribution profile of isolated particles and their concentrations estimated by NTA. However, no differences in the control vs. S6h and the control vs, S24h and S72h comparisons were detected. (C) Western blot identification of EV enriched proteins and albumin. SDS-PAGE for CD63 western blot was run under non-reducing conditions. dEV, serum depleted of EVs, S, serum.

Figure 3

Principal component analysis (PCA) of the proteomic data of circulating EVs. LC/MS produced quantitative proteomic data (protein abundances) were quantile normalized, log2-transformed, and used by Qlucore software for creating the PCA plots. (A) Control and 6 h post-surgery (S6h) experimental group form individual clusters. (B) Control and S24h groups are completely separated whereas the S72h group overlaps with the control samples.

Figure 4

Differential expression of proteins in the post-surgery circulating EVs. (A) Bar diagram representing the number of differentially expressed proteins in the 6, 24, and 72 h post-surgery groups compared to the control values as assessed by GraphPad Prism (v. 9.1.1) using multiple unpaired t-tests analysis with FDR < 0.05 and fold change (FC) > 1.5. (Materials and Methods). (B) Volcano plots showing the differentially expressed proteins at 6h (left) and 24h (right) post-surgery compared to their corresponding control groups (C6h and CS24-72h). Dotted lines represent FC > 1.5 threshold (log2-FC) and FDR <0.05 (-log10FDR), respectively. (C) Differential expression of α-synuclein in the EVs from S24h and S72h groups using log2-transformed LC/MS data. (D) α-synuclein expression validation by western blot. Mouse brain (hippocampus) homogenate was used as a positive control. SDS-PAGE was run under non-reducing conditions. dEV, serum depleted of EVs, S, serum. Snca in the 24h Volcano plot is the official gene symbol for α-synuclein (α-syn). **p < 0.005, ***p < 0.0005.

Figure 5

RNA mapping and miRNA PCA. (A) The averaged RNA-seq data from the control samples were used to map RNA categories [sRNAbench webserver (19)]. (B) Principal component analysis of the circulating EVs’ miRNA data. Normalized miRNA read counts [Counts Per Million reads (CPM)] were log2-transformed and used by Qlucore software to create a PCA plot. 24 h and 72 h post-surgery groups cluster together and are separated from the control whereas the 6 h group clusters with the control.

Figure 6

Differential expression of miRNAs in the post-surgery circulating EVs. (A) Bar diagram representing the number of differentially expressed miRNAs in the 6, 24, and 72 h post-surgery groups compared to control values as assessed by DESeq2 analysis (FDR<0.05 and fold change (FC)>1.5; see Materials and Methods). (B) Post-surgery temporal kinetics of the differentially expressed miRNAs. Each line represents an individual differentially expressed miRNA. (C) Volcano plots showing the differentially expressed proteins at 24- (left) and 72 h (right) post-surgery compared to the control group (Control). The dotted lines represent FC > 1.5 threshold (log2-FC) and FDR <0.05 (-10logFDR), respectively. (D) Venn diagram showing the overlap between the differentially expressed miRNAs at 24 and 72 h post-surgery. (E) The log2-transformed CPM values of differentially expressed miRNAs at 24 and 72 h and the same miRNAs from the control and S6h groups were used to generate heatmap (Qlucore). Cluster 1 (blue) and cluster 2 (red) include down-regulated and up-regulated miRNAs at 24 and 72 h, respectively.

Figure 7

Pathway analysis of differentially expressed miRNAs. Down- (cluster 1 from Figure 6E ) and up-regulated (cluster 2 from Figure 6E ) at 24 h post-surgery miRNAs were analyzed by the miRPath online tool (see Materials and Methods) that determines the miRNA target mRNAs (based on experimental data) and analyses the overrepresentation of these genes in the different KEGG pathways.

Figure 8

Correlation analysis between the differentially expressed protein and miRNA data sets. Temporal kinetics of the differential expression of EVs’ proteins and miRNAs demonstrates a high degree of correlation as to the Pearson’s linear correlation coefficient (GraphPad Prism).