Adipose Tissue Heterogeneity - Development and Application of Nonlinear Microscopy Methods
Doctoral thesis, 2018
Although the negative health impacts of obesity have been well documented, the number of overweight or obese patients has and is predicted to continue rising. One of the causal factors of obesity is a constant positive energy balance mainly from excessively high caloric intake from food compared with limited physical activity (caloric expenditure). While this basic concept is well understood, the specific changes that obesity invokes within the body have not been fully uncovered and thus a molecular reversal of obesity has not been achieved to date. Along the way to identify a pharmaceutical target, a new type of fatty tissue, brown adipose tissue (BAT), has moved into the focus with potential therapeutic implications.
In contrast to white adipose tissue (WAT) that serves as long-term fat storage in the form of neutral lipids, BAT uses these molecules as fuel to perform non-shivering thermogenesis, a process seen most commonly in hibernating animals and infants, and used to keep the body core temperature stable. Since BAT function is very different from WAT function, the two tissue types are often studied in conjunction in order to better identify the characteristics and the possibility to increase BAT activity and mass in obese adults (as a way to increase “parasympathetic” caloric expenditure).
This thesis work aims to investigate adipose tissue (AT) physiology on different levels: from interactions of isolated adipose-derived stem cells (ADSC) with matrices to the function of AT in mice. Special focus is directed towards the molecules at the center of adipose tissue physiology: triacylglycerols (TAGs). These are often neglected during analysis due to the difficulty to study them in a cellular context. Classical methods, like gas chromatography, usually rely on extraction of all lipids from a tissue depot requiring extensive sample preparation. Magnetic resonance imaging or matrix-assisted laser desorption/ionization followed by mass spectrometry imaging can be used to visualize lipids in their natural setting but do not offer high enough resolution or require extensive sample preparation. Thus, in this work, I focused on using a label-free chemical imaging approach called coherent anti-Stokes Raman scattering (CARS) microscopy to study TAGs in situ at increasing levels of biological complexity. This method requires almost no sample preparation and can visualize sub-micrometer-sized TAG storage depots based on their intrinsic chemistry.
First, CARS microscopy was used to follow ADSCs during the early stages of attachment and interaction with an extra cellular matrix (ECM); then CARS was employed to follow lipid accumulation during adipogenic differentiation to study how the ECM structure affects that process. Next, mature adipocytes were studied in ex vivo tissue sections. During this study mitochondrial activity was also investigated. In the following studies, not only the volume/number of lipid depots was of interest but also their contents. Therefore, we extended the CARS imaging with a spectral dimension, developing a new method to generate maps of TAG chain length and saturation, which were then be employed to see how high fat diet affects the lipids in BAT and WAT.
least squares decomposition
multivariate spectral analysis
broadband coherent anti-Stokes Raman scattering