Pressure Cycling Technology Assisted Mass Spectrometric Quantification of Gingival Tissue Reveals Proteome Dynamics during the Initiation and Progression of Inflammatory Periodontal Disease

Abstract

Understanding the progression of periodontal tissue destruction is at the forefront of periodontal research. The authors aimed to capture the dynamics of gingival tissue proteome during the initiation and progression of experimental (ligature-induced) periodontitis in mice. Pressure cycling technology (PCT), a recently developed platform that uses ultra-high pressure to disrupt tissues, is utilized to achieve efficient and reproducible protein extraction from ultra-small amounts of gingival tissues in combination with liquid chromatography-tandem mass spectrometry (MS). The MS data are processed using Progenesis QI and the regulated proteins are subjected to METACORE, STRING, and WebGestalt for functional enrichment analysis. A total of 1614 proteins with ≥2 peptides are quantified with an estimated protein false discovery rate of 0.06%. Unsupervised clustering analysis shows that the gingival tissue protein abundance is mainly dependent on the periodontitis progression stage. Gene ontology enrichment analysis reveals an overrepresentation in innate immune regulation (e.g., neutrophil-mediated immunity and antimicrobial peptides), signal transduction (e.g., integrin signaling), and homeostasis processes (e.g., platelet activation and aggregation). In conclusion, a PCT-assisted label-free quantitative proteomics workflow that allowed cataloging the deepest gingival tissue proteome on a rapid timescale and provided novel mechanistic insights into host perturbation during periodontitis progression is applied.

Keywords: experimental periodontitis; gingival inflammation; gingival tissue; label-free quantitation; pressure cycling technology; proteome.

Figure 1:. Experimental flow of the PCT label-free quantification of gingival tissue.

(A) Tissue lysis and protein digestion. The gingival tissues from mouse (typically around 0.5 mg) were lysed in barocycles for 60 min. The extracted proteins were reduced, alkylated and diluted for Lys-C digestion in the barocycler for another 45 min. The protein solutions were further diluted prior to trypsin digestion for an additional process of 90 min in the barocycler. The resultant peptide solutions were desalted and dried for LC-MS/MS analysis. (B) Mass spectrometry and data analysis. The peptides were separated by liquid chromatography in an 80-min gradient and then analyzed by Oritrap fusion mass spectrometer. The raw data were quantified by Progenesis QI and processed with various functional analyses. (C) Overview of the experiment

Figure 2:. Dynamic of gingival tissue proteome in ligature-induced periodontitis model over consecutive time points.

(A) Representative images of time-dependent bone loss in unligated and ligated sites in ligature-induced periodontitis model over consecutive time points. (B) The heatmap of normalized abundance for all quantified proteins among four different time points (n=1 for each day, Table S1). The distance matrix of the heatmap was shown in Table S2. Dot plots of log 2 (FC) for all identified proteins in baseline (C) or consecutive time points (D). Upregulated, downregulated, and unregulated proteins were plotted in red, blue, and grey, respectively. Venn diagrams showed the overlap of quantified proteins in baseline (E) or consecutive time points (F) (Table S3). The numbers of upregulated protein and downregulated protein were labelled in red and blue color, respectively. The numbers of protein that did not regulate in any of condition were list on the right side of each Venn diagram with black color.

Figure 3:. Alveolar bone loss of ligature-induced periodontitis model after five days of ligation.

(A) Kinetics of total bone loss “and representative images of time-dependent bone loss in unligated and ligated sites. *p≤0.05 (B) The normalized abundance of all these quantified proteins among 10 samples (Table S4). Samples from the unligated side and the ligated side were highlighted in green and red, respectively. The distance matrix of the heatmap was shown in Table S5. (C) Volcano plots of log 2(FC) and log 10 p-value of ANOVA test for all identified proteins. Upregulated (p ≤ 0.05, Log2(FC) ≥ 1), downregulated (p ≤ 0.05, Log2(FC) ≤ −1), and unregulated proteins (p ≥ 0.05) were plotted in orange, green, and grey in periodontitis compared to healthy gingival tissues, respectively; proteins that significant regulated (p ≤ 0.05) but not reaches stringent cut off (FC ≥ 1) were labelled in red or blue. Regulated proteins that also had Benjamini-Hochberg p-value less than 0.05 were plotted in crosses. The labels of regulated proteins were shown in Figure S1. Proteins involved in top 10 pathways based on Reactome database were labelled with black plots and their Swissprot IDs (Table S13). (D) Fold change distribution of regulated proteins in mouse gingival tissues. (E) Network established using STRING 10.5 (Table S11) based on the highest confident score (0.9) of regulated tissue proteins with stringent cut-off. Lines indicated different types of protein-protein interactions. Blue and purple lines indicated known interaction determined from the curated database and experimental results, respectively. Green, red, and dark blue lines indicated predicated interaction determined from gene neighbourhood, gene fusions, and gene co-occurrence, respectively. Yellow, black and light blue lines indicated interactions were from textmining, co-expression and protein homology, respectively. An enlarged version of this protein–protein interaction network was updated as Figure S2.