This dissertation is intended to highlight the recent advancements in the field of lipidomics, particularly in reference to the discovery of novel biomarkers of traumatic brain injury (TBI). Lipidomics falls under the umbrella of the related ‘omics technique known as metabolomics and refers specifically to the study of the lipidome, or collection of all lipophilic metabolites in a biological system. Lipids play critical roles in cellular structure and function, including acting as signaling molecules, regulating energy storage, and maintaining membrane integrity. The use of omics’ techniques has led to the proposal of a number of biomarkers for the study of TBI, though most have stemmed from targeted proteomic approaches, with researchers focusing on a few proteins in the brain and central nervous system that are upregulated as a consequence of TBI. The response of the lipidome to TBI is less understood, so a non-targeted approach is suggested here for the purpose of generating hypotheses regarding lipidomic responses to TBI using data-driven, discovery-oriented techniques.
Research interest into TBI has been accelerated due to heightened awareness of the growing concussion epidemic in the military, sports, and general population. With diagnosis of subtle cognitive dysfunction typically based on the presence of subjective self-reported symptoms, the number mild traumatic brain injuries (mTBIs) is vastly underestimated, as many injuries may go unreported or misdiagnosed due to lack of perceived symptoms. Chapter 1 introduces the commonly utilized techniques for clinical diagnosis of TBI and outlines the known pathophysiology associated with the onset of injury. Ranging from simple grading schemes that measure a patient’s verbal, motor and xvieye-opening responses to more advanced radiologic imaging and neuropsychological testing modalities, the field of mTBI diagnostics is growing rapidly. However, a substantial gap remains between the literature and clinical translation, indicating a need for further scientific collaboration.
Biomarker discovery studies have been enabled by the advent of sophisticated instrumentation, fueling the growth of the fields of metabolomics and lipidomics. Such studies often involve careful measurement of a large number of molecular targets, so advanced techniques are required beyond simple single-target assays. Mass spectrometry (MS) and nuclear magnetic resonance (NMR) are the two most commonly utilized platforms for metabolomics analysis, and each offers complementary advantages to research efforts. MS-based techniques are used throughout this dissertation in order to study the lipidomic response within the serum of rodents incurring TBI induced by controlled cortical impact.
Two distinct severities of injury are investigated and modeled in chapters 2 and 3. Chapter 2 first focuses on an injury of moderate severity in an effort to measure the response of lipids to injury. A standard metabolomics workflow was utilized, and a number of novel biomarker candidates for TBI were identified in the serum lipidome of adult male Sprague-Dawley rats in the first week following injury. Serum samples were analyzed in positive and negative modes by Ultra Performance Liquid Chromatography Mass Spectrometry (UPLC-MS). A predictive panel for the classification of injured and uninjured sera samples, consisting of 26 dysregulated species belonging to a variety of lipid classes, was developed with a cross-validated accuracy of 85.3% using omniClassifier software to optimize feature selection. Polyunsaturated fatty acids xvii(PUFAs) and PUFA-containing diacylglycerols were found to be upregulated in sera from injured rats, while changes in sphingolipids and other membrane phospholipids were also observed, many of which map to known secondary injury pathways. Cholesterol sulfate, a sterol sulfate that plays a stabilizing role in cellular membranes and serves as a precursor in the synthesis of other sulfonated adrenal steroids, was found to be significantly decreased in the TBI cohort. Overall, the identified biomarker panel presents a number of viable molecular candidates representing lipids affected by of TBI pathophysiology.
In chapter 3, we proposed to study mild injury at an early acute time point (t=24 h). Using a matched pair study design to improve statistical power, a panel of 16 lipid metabolites was developed with a classification accuracy of 88.5%, a small improvement from the previous study despite the use of a less complex metabolite panel (reduction by 50% of the number species studied) and the occurrence of a milder form of injury in which the average effect size of lipid alterations was reduced. These promising results demonstrate the feasibility of utilizing non-targeted lipidomics to detect mild, concussive events in serum.
The broad applicability of non-targeted lipidomics experiments is then demonstrated in chapter 4 through the study of phytoplankton competitor responses to exposure to Karenia brevis, the dinoflagellate responsible for forming harmful algal blooms commonly referred to as red tides. MS-based analysis of phytoplankton lipidomes led to the identification of 80 distinct lipid metabolites whose concentrations differed significantly in T. pseudonana following exposure to K. brevis allelopathy. These 80 metabolites represent nine major lipid classes, of which members of five xviii(phosphatidylcholines [PCs], sulfoquinovosyldiacylglycerols [SQDGs], monogalactosyldiacylglycerols [MGDGs], digalactosyldiacylglycerols [DGDGs], and phosphatidylglycerols [PGs]) were generally less abundant when T. pseudonana was subjected to K. brevisallelopathy, whereas members of four classes (non-SQDG sulfonated lipids [SULFs], free fatty acids [FFAs], primary fatty acid amides [PFAAs], and phosphatidylethanolamines [PEs]) were generally more abundant. In contrast, for the other competitor, A. glacialis, concentrations of only six metabolites were significantly affected by allelopathy, reinforcing the hypothesis that A. glacialis maintains a more robust metabolism in response to K. brevis allelopathy due to an evolved resistance stemming from periods of prior co-habitation.
A majority of the lipids formed (PFAAs, FFAs, and SULFs) by allelopathic exposure were either metabolic breakdown products or metabolic precursors of PCs and SQDGs, whose pools shrunk in T. pseudonana upon exposure to allelopathy. Globally, concentrations of membrane-associated lipids (MGDG, DGDG, SQDG, PC) were significantly suppressed for T. pseudonana exposed to allelopathy, leading membranes of living cells to become more permeable. Increased membrane permeability as well as decreased photosynthetic capability both likely occurred due to decreases in the concentrations of membrane- and thylakoid-associated lipids. K. brevis allelopathy appears to target lipid anabolism, affecting multiple physiological pathways. This suggests that exuded compounds have the ability to significantly alter competitor physiology through effects on metabolism, giving K. brevis an edge over sensitive species.
inally, chapter 5 explores the potential future work associated with the projects contained within this dissertation. Critical points of exploration include validation of proposed biomarkers in another cohort of rats and examination of altered response patterns based on sex, strain, and age of the animals. Use of human samples for the purposes of clinical translation is also proposed. Limitations of the current work are examined, and suggestions are made for future experiments to better understand efflux pathways for biomarker clearance and answer questions regarding specificity of the proposed biomarkers to the brain and brain injury.