Tomografia de hemácias.

Abstract

Transparent objects (phase objects) render visible in a bright-eld optical microscope, if the microscope is defocused. In fact, a defocused bright-eld optical microscope is equivalent to a phase contrast microscope, presenting advantages over the conventional one, as it will be discussed in this thesis. Defocusing causes a phase difference between the transmitted light (non-diffracted) and the light diffracted bythe phase object, which when combined at the detection plane due to light interference, renders the object visible, with an image contrast different than zero. This effect was demonstrated in our laboratory in several articles and previous thesis [1-7]. In this thesis, we improved our optical model of defocusing microscopy, such that its validity was extended to any defocusing, including asymptotically large defocusing,and for larger diffaction wavenumbers. We obtain theoretical expressions for the contras of phase objects composed by one and by two parallel interfaces. In the limit of small defocusing and small diffraction wavenumbers, from the time averaged contrast, we can obtain information about the equilibrium shape of (RBC) red blood cells (erythrocytes), as previously demonstrated [4]. Our new expressions areimportant to establish a relationship between defocused image contrast fluctuations and surface height uctuations on RBCs. With our new optical model we show that by scanning the microscope objective focal plane position (zf ), and since the RBC surfaces are at different positions in relation to the z-axis, there diffracted light are affected by different phase differences, such that it surface can be distinguished by using defocusing microscopy. This fact allows us to study the properties of each interface separately. For fluctuations with wavenumbers larger than 1 um -1, one can approximate the RBC central region by parallel planes, and apply the optical model for two interfaces developed in this thesis. By using the recently developed elastic model by Auth et al. [8], which takes into account the coupling between the lipid bilayer and the spectrin cytoskeleton (and also uses the approximation of parallel planes for the RBC surfaces) and our optical model, we have obtained a very good t for the experimental data. From this t we have obtained the values for the lipid bilayer curvature modulus, for the cytoskeleton shear modulus and for the effective temperature. The RBC height uctuations are not only of thermal origin. Another contribution comes from the conversion of chemical energy into mechanical work driven by ATP (adenosine triphosphate, the cell's combustible) [9,10]. These non-equilibrium effects are lumped into a single parameter, namely the effectivetemperature introduced in an ad-hoc manner. This effect is still contoversial in the literature [11]. We also studied the relaxation of RBC height uctuations through the time autocorrelation function of contrast fluctuations. This relaxation is dictated by the elasticity of the lipid bilayer and the spectrin cytoskeleton, and by the flow of the cytoplasm through the narrow gap of width d ~ 20 to 30 nm, between the bilayer and the cytoskeleton. From the t of this theory to our data, we obtained a gap-width d = 21 +- 1 nm. We showed that defocusing microscopy is an useful and trustable technique to study height proles and static and dynamic roughness of interfaces in a transparent multilayered material, like membranes of living cells.