Local burn wound environment versus systemic response: Comparison of proteins and metabolites

Abstract

In this study, paired blood plasma (BP) and blister fluid (BF) samples from five paediatric burn patients were analysed using mass spectrometry to compare their protein and metabolite composition. The relative quantification of proteins was achieved through a label-free data independent acquisition mode. The relative quantification of metabolites was achieved using a Shimadzu Smart Metabolite Database gas chromatography mass spectrometry (GCMS) targeted assay. In total, 562 proteins and 141 individual metabolites were identified in the samples. There was 81% similarity in the proteins present in the BP and BF, with 50 and 54 unique proteins found in each sample type respectively. BF contained keratinocyte proliferation-related proteins and blood plasma contained abundant blood clotting proteins and apolipoproteins. BF contained more carbohydrates and less alpha-hydroxy acid metabolites than the BP. In this study, there were unique proteins and metabolites in BF and BP which were reflective of the local wound environment and systemic environments respectively. The results from this study demonstrate that the biomolecule content of BF is mostly the same as blood, but it also contains information specific to the local wound environment.

Keywords: burns; child; metabolite; paediatric; plasma; protein; wound fluid.

FIGURE 1

Comparison of the proteins and metabolites in BF and BP. (A) Venn diagram of proteins identified in matched BF versus BP samples, n = 12. There are 81% common proteins between BF and BP. (B) The fraction (%) of subclasses of metabolites identified in BF and BP samples, based on the abundance of the metabolites, n = 12. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 2

BF and BP exhibit distinct protein and metabolite profiles. PCA plots based on the protein and metabolite abundance within each sample. (A) Unsupervised PCA plot based on the VSN normalised protein abundance between BF and BP, n = 12. (B) Unsupervised PCA plot based on the VSN normalised metabolite abundance between BF and BP, n = 12. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 3

The proteins and metabolites which most discriminate between the BF and BP samples in supervised partial least‐squares discriminant analysis (PLS‐DA). (A) Top 15 proteins with highest variable importance in projection (VIP) scores, fibrinogen alpha chain (FGA), haemoglobin subunit alpha (HBA1), histone H2B type 2‐F (H2BC18), fibrinogen beta chain (FGB), platelet glycoprotein V (GP5), histone H2A type 2‐C (H2AC20), platelet factor 4 variant (PF4V1), haemoglobin subunit beta (HBB), histone H1.3 (H1‐3), platelet basic protein (PPBP), fibrinogen gamma chain (FGG), histone H2A type 2‐A (H2AC18), 60S acidic ribosomal protein P1 (RPLP1), 14‐3‐3 protein epsilon (YWHAE), and thrombospondin‐1 (THBS1). (B) Top 15 metabolites with highest VIP scores. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 4

Most significant protein and metabolite abundance changes in each individual sample, based on p value. Red in the heatmap indicates higher abundance and blue indicates lower abundance. Yellow in the top bar represents BF samples and red represents BP samples. (A) Top 25 most significant proteins. Heat shock protein beta‐1 (HSPB1), histone H4 (H4C1), histone H2A type 2‐C (H2AC20), histone H2B type 2‐F (H2BC18), histone H2A type 2‐A (H2AC18), histone H1.3 (H1‐3), vitamin D‐binding protein (GC), triosephosphate isomerase (TPI1), serine protease inhibitor Kazal‐type 5 (SPINK5), 14‐3‐3 protein sigma (SFN), histone H2B type 1‐M (H2BC14), 14‐3‐3 protein epsilon (YWHAE), complement C1s subcomponent (C1S), apolipoprotein C‐I (APOC1), platelet factor 4 variant (PF4V1), haemoglobin subunit alpha (HBA1), haemoglobin subunit beta (HBB), fibrinogen gamma chain (FGG), fibrinogen beta chain (FGB), fibrinogen alpha chain (FGA), immunoglobulin J chain (JCHAIN), thrombospondin‐1 (THBS1), vitamin K‐dependent protein S (PROS1), platelet basic protein (PPBP) and apolipoprotein L1 (APOL1). (B) Top 25 most significant metabolites. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 5

Significant proteins and metabolites which also showed large fold change differences in abundance between BF (n = 5) and BP (n = 5) samples. Red points have higher abundance in BF while blue points have a higher abundance in BP. Grey points are p > 0.05. (A) Significant proteins with ≥2‐fold change. Heat shock protein beta‐1 (HSPB1), histone H4 (H4C1), histone H2A type 2‐C (H2AC20), histone H2B type 2‐F (H2BC18), histone H2A type 2‐A (H2AC18), histone H1.3 (H1‐3), vitamin D‐binding protein (GC), triosephosphate isomerase (TPI1), serine protease inhibitor Kazal‐type 5 (SPINK5), 14‐3‐3 protein sigma (SFN), histone H2B type 1‐M (H2BC14), 14‐3‐3 protein epsilon (YWHAE), complement C1s subcomponent (C1S), apolipoprotein C‐I (APOC1), platelet factor 4 variant (PF4V1), haemoglobin subunit beta (HBB), fibrinogen gamma chain (FGG), fibrinogen beta chain (FGB), fibrinogen alpha chain (FGA), immunoglobulin J chain (JCHAIN), thrombospondin‐1 (THBS1) and vitamin K‐dependent protein S (PROS1). (B) Significant metabolites with ≥1.8‐fold change. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 6

The integrated proteomic and metabolomic results showed that several biological pathways were enriched, and these pathways contained proteins and metabolites that were significantly upregulated in BF or BP. (A) The enriched pathways of significant proteins and metabolites, the colour of the dots represent significance based on the combined p‐value, with red more significant and yellow less significant, and the size of the circle represents the pathway impact factor. (B) The pathway of cysteine and methionine metabolism, pyruvate metabolism and glycolysis or gluconeogenesis. Squares are identified proteins, circles are identified metabolites, and text which is not within a square/circle are proteins and metabolites included in the pathways but not identified in this study. Green indicates proteins and metabolites which showed no significant differences between the two sample types, red represents significantly higher abundance in BF, and blue represents significantly lower abundance in BF. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 7

Individual protein and metabolite abundance changes between BF and BP. Box–Whisker plots of significant metabolites and proteins, y‐axis represents the raw MS intensity. p‐value is calculated based on raw data, *<0.05, **<0.01, ***<0.001 and ****<0.0001. Metabolites identified in the enriched pathways (A) cholesterol, (B) lactic acid, (C) pyruvic acid, (D) cystine and (E) 2‐ketobutyric acid. Box–Whisker plots of significant proteins with p‐value <0.01 and more than 2‐fold‐change, (F) apolipoprotein C‐I, (G) immunoglobulin J chain, (H) platelet basic protein, (I) vitamin K‐dependent protein S, (J) platelet factor 4 variant, (K) complement C1s subcomponent and (L) 14‐3‐3 protein sigma. [Color figure can be viewed at wileyonlinelibrary.com]