Mass spectrometric proteomic profiling of acute wound healing processes in the porcine model

Wounds are a common occurrence and generally heal well within a short time depending on whether tissue needs to be replaced or not. Although there is an understanding of the physiological processes taking place during wound healing, the changes taking place at the molecular level as well as the causes of chronic wounds are not well defined. The purpose of this study was a proof of concept making use of complementary current mass spectrometric techniques to better define the spatial and kinetic changes occurring during healing of well-defined acute wounds across proteomic, metabolomic and lipidomic bases. Several wound tissue preparation techniques, analytical approaches, data conversion methodologies and statistical tools were assessed to best visualise, characterise and identify molecular features in the complex and continuously changing environment of the acute wound at six pre-selected time points over 16 days of healing. Different mass spectrometric approaches to analyse snap frozen histological sections of wound tissue collected from a porcine wound healing model allowed ‘omic’ analyses to provide a better understanding of the acute wound healing process. Unidentified features found at key time points were compared to provide a foundation to develop an acute wound healing model, which could be applied to non-healing chronic wounds. An initial animal study involved full-skin thickness wounds being surgically created in the porcine model using strictly controlled wound formation and the subsequent healing thereof accurately investigated. Photographic and histological analyses of resected wound tissue were used to track the sequential yet overlapping phases of the acute wound healing pathway. The spatial resolution capabilities of matrix-assisted laser desorption ionisation imaging mass spectrometry (MALDI-IMS) was linked to histological staining and localised healthy reference tissue to successfully characterise changes in wound tissue. Optimised multivariate analyses and statistical analysis could statistically characterise the kinetic changes and revealed molecular features and key time points during acute wound healing from proteomic, metabolomic and lipidomic assay data collected individually from sequential tissue sections from the same wound. Methods for extraction of soluble protein from histological wound tissue sections were compared and further analysed using hydrophilic interaction-based bead capture with in-solution tryptic digestion followed by classic proteomic liquid chromatography tandem mass spectrometry (ESI-LC-MS/MS) analysis. The extracted proteins were also assessed by sodium dodecyl sulfate polyacrylamide gel electrophoresis where protein mass fingerprints were visualised and major protein bands excised, followed by standard in-gel digestion and LC-ESI-MS/MS analysis. All mass spectral proteomic data was subject to bioinformatic analysis using open source packages. A comparative analysis of in-solution versus on-tissue sample preparation was performed to determine optimal methodology for tissue protein extraction and identification. The optimised protocol was used to identify proteins across six time points and 16 days showing potential as predictive acute wound healing phase features. A total of 18 proteins showed potential as acute wound healing phase markers through relative quantitative change and functional healing relation. Biological functions of these proteins included involvement in adaptive and innate immunity, inflammation, cellular adhesion and various biosynthetic processes, and may be used as validating features of acute wound healing in clinical settings or as new drug targets for healing therapies. This study revealed the best time range and individual features that could be used to most accurately represent essential molecular environmental changes in the progression of healing of acute wounds across proteomic, metabolomic and lipidomic bases. These approaches may aid in comparative healing studies, such as those seeking to better define differences that can be linked to chronic wounds. Using a well-defined but reduced sample set of acute wound molecular environment data as a fixed comparable model, as well as potential markers of successful healing, may simplify the approach and aid in the identification of potential drug targets in non-healing chronic wounds.Thesis (PhD)--University of Pretoria, 2019.PharmacologyPhDUnrestricte