Thesis Abstract - Rantamäki Antti H.

Tear Film Lipid Layer - from Composition to Function, Implications of the Anti-Vaporative Effect

The tear film lining the ocular surface consists of three qualitatively different layers. The hydrated (i) mucous matrix, which is composed of the epithelial glycocalyx and secreted gel-like mucins, continues as a concentration gradient into the overlaying protein-rich (ii) aqueous layer, which is largely responsible for the hydration, nutrition, and host defence of the ocular surface. Finally, the air-tear interface is lined with a thin (iii) tear film lipid layer (TFLL), which is considered to stabilise the entire tear film and is thought to retard evaporation from the air-tear interface. Meibum, an oily secretion produced by meibomian glands, is considered largely as the source of the TFLL lipids. The integrity of the tear film is vital for the ocular surface, and disturbances in any of the aforementioned sections typically result in dry eye symptoms. Despite the extremely low evaporation rates measured from the ocular surface in vivo, no evaporation-retarding mechanism of TFLL has been shown in vitro. Altogether, due to the co-operative character of lipids, the function and behaviour of the TFLL and similar lipid layers are largely defined by their lipid composition. To understand the behaviour of complex lipid layers on the molecular level, the composition of the layer should be determined. Therefore, the first aim of this study was to analyse the lipid composition of tear fluid. Because most of the lipids are considered to be located in the TFLL, the aim was to create TFLL-like lipid compositions for in vitro experiments. Finally, the aim was to investigate the evaporation-retarding effect of such lipid layers to better understand the potential mechanism of evaporation retardant in vivo TFLL.

A modern mass-spectrometric platform, namely ultra performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry, was employed for the tear fluid lipid analysis. In contrast to the widely recognised meibum lipid composition, the tear fluid samples contained a considerable amount of phospholipids. Unfortunately, most of the non-polar lipids, such as the cholesteryl esters and the wax esters typically present in meibum, could not be detected using this mass spectrometric platform; however, they were detected with enzymatic assays and thin layer chromatography. In addition, we demonstrated that the phospholipids function as a spreading aid facilitating the uniform spreading of the hydrophobic non-polar lipids at the air-water interface.

Based on the tear fluid lipid composition, we created TFLL-like lipid compositions and investigated their ability to retard evaporation from the air-water interface in vitro. A custom-built system was assembled for evaporation rate determination, and Brewster angle microscopy was employed to observe the interfacial organisation of the lipid layer at the air-water interface. It was found that very specific lipids and very definite lipid compositions are needed to retard evaporation from the air-water interface. None of the complex TFLL-like lipid compositions retarded evaporation. However, a specific class of TFLL lipids, namely wax esters (WEs), were shown to be efficient evaporation retardants, but layers composed of WEs mixed with large proportions of other TFLL lipids did not retard evaporation.

WEs, however, did not retard evaporation under all conditions, but only in a defined phase of the layer. This phase and therefore the evaporation-retarding effect was dependent on the melting point of the specific WE and the temperature of the air-water interface. The WEs that were close to their bulk melting temperature retarded evaporation, whereas the WEs in their solid or liquid states lacked this property. In their solid state, the WEs did not spread as a uniform layer at the air-water interface, whereas in their liquid state, the WEs formed very fluid layers, and therefore, the water molecules diffused through the loosely packed lipid layer. We also investigated the surface-active properties of WEs and noted that WEs are expectedly poor surfactants compared to phospholipids; they are extremely prone to aggregation and are poorly compressible, especially when the layer is in the evaporation-retarding phase. These results support the theory suggested in our lipid composition study: hydrophobic lipids need to be mixed with a certain amount of amphiphilic lipids to form rapidly spreading films at the air-water interface.

In summary, this thesis project concentrated on providing an in vitro model for linking certain functions, properties, and behaviours of TFLL to the composition of such layers. In short, amphiphilic phospholipids seem to be a vital component of the TFLL, although possibly in a smaller proportion than originally hypothesised, providing aid for non-polar lipid spreading. The evaporation-retarding effect is largely dependent on the composition. Pure WEs turned out to be effective evaporation-retarding TFLL lipids, but only in a certain temperature-dependent phase. Therefore, WEs are most likely the lipids that provide the evaporation-retarding effect of the TFLL; however, such complex evaporation-retarding lipid composition is yet to be determined in vitro. The results of this thesis suggest that the compositional changes affect the behaviour, such as spreading and the evaporation-retarding effect, of TFLL. Defects in such properties may result in accelerated evaporation from the ocular surface and, consequently, in dry eye symptoms.