P93 Bipolar double-PFG NMR Reveals Pore Morphology in Ran domly Oriented Cylindrical Compartments and in Spherical Yeast Cells

domly Oriented Cylindrical Compartments and in Spherical Yeast Cells Noam Shemesh 1, Evren Ozarslan 2, Peter Basser 2, Yoram Cohen 1 1 Tel Aviv University, 2 National Institute of Health Abstract Diffusion NMR is the most important methodology for non-invasively char­ acterizing pore morphology. Scenarios such as diffusion in spheres or in lo­ cally anisotropic but randomly oriented compartments are extremely difficult to characterize using conventional single-PFG (s-PFG) MR. The angular doublePFG [1] (d-PFG) methodology at long mixing times (tm) offers a novel means to overcome this limitation, since it provides quantitatively different angular dependencies for spheres and randomly oriented cylinders [2]: a flat angular dependence is expected for spheres, while a modulated bell-shaped function is expected for locally anisotropic but randomly oriented pores [2]. Another advantage of angular d-PFG is that it can be conducted at low q-values [3]. Here, we used controlled randomly oriented cylindrical pores, in which the in­ ner diameter (ID) and compartment shape are known a-priori. S-PFG and long tm angular d-PFG experiments were performed. Scanning electron microscope images confirmed that the specimen is indeed comprised of completely ran­ domly oriented cylindrical microtubes. The line width of the water resonance in these pores was ∼ 0.5 kHz, a manifestation of large susceptibility artifacts. Conventional s-PFG experiments surprisingly showed direction-dependant sig­ nal decay, clearly unexpected for randomly oriented compartments. However, when we employed bipolar gradients in the s-PFG sequence (bp-s-PFG), an isotropic decay was observed, as expected. The angular d-PFG experiments yielded inverted angular dependencies, from which microstructural information could not be obtained; therefore, we implemented a d-PFG NMR sequence with bipolar gradients (bp-d-PFG). The angular bp-d-PFG NMR yielded the ex­ pected [2] modulated bell-shaped function, providing insights that could not be inferred from s-PFG methodologies. First, the presence of restricted diffusion was easily inferred, since the non-flat angular dependencies are only observed when restricted diffusion occurs. Second, we could infer on the eccentricity of the pores, a fact we could not infer from s-PFG experiments. To demon­