Martin Michel

High pressure freezing permits the successful cryoimmobilization of thick biological specimens (up to approx. 500 microns). A very high yield of adequately frozen specimens, in which no segregation patterns due to ice crystal formation is apparent after freeze-substitution or freeze-fracturing, is obtained with suspensions of microorganisms as well as plant and animal tissue. This very high yield is attributed to an optimized transfer of pressure and cold to the biological specimen. This is achieved by replacement of extraspecimen water or buffer by 1-hexadecene, a chemically inert, hydrophobic paraffin oil of low viscosity and low surface tension.

For more than 20 years, high-pressure freezing has been used to cryofix bulk biological specimens and reports are available in which the potential and limits of this method have been evaluated mostly based on morphological criteria. By evaluating the presence or absence of segregation patterns, it was postulated that biological samples of up to 600 microns in thickness could be vitrified by high-pressure freezing. The cooling rates necessary to achieve this result under high-pressure conditions were estimated to be of the order of several hundred degrees kelvin per second. Recent results suggest that the thickness of biological samples which can be vitrified may be much less than previously believed. It was the aim of this study to explore the potential and limits of high-pressure freezing using theoretical and experimental methods. A new high-pressure freezing apparatus (Leica EM HPF), which can generate higher cooling rates at the sample surface than previously possible, was used. Using bovine articular cartilage as a model tissue system, we were able to vitrify 150-micron-thick tissue samples. Vitrification was proven by subjecting frozen-hydrated cryosections to electron diffraction analysis and was found to be dependent on the proteoglycan concentration and water content of the cartilage. Only the lower radical zone (with a high proteoglycan concentration and a low water content compared to the other zones) could be fully vitrified. Our theoretical calculations indicated that applied surface cooling rates in excess of 5000 K/s can be propagated into specimen centres only if samples are relatively thin (< 200 microns). These calculations, taken together with our zone-dependent attainment of vitrification in 150-micron-thick cartilage samples, suggest that the critical cooling rates necessary to achieve vitrification of biological samples under high-pressure freezing conditions are significantly their (1000-100,000 K/s) than previously proposed, but are reduced by about a factor of 100 when compared to cooling rates necessary to vitrify biological samples at ambient pressure.