There is mounting evidence to indicate that membranes, primarily the plasma membrane, is intimately involved in cold acclimation and freezing injury (11,18,19,(24)(25)(26). Thus, it follows that plant plasma membranes must undergo certain chemical and biophysical changes during cold acclimation in order to withstand freezing stresses.There are several reports on changes in cellular membranes during cold acclimation (6,7,27). However, these are based on analytical data obtained from crude membrane fractions or total membrane extracts. Therefore, it is uncertain whether the reported changes are related to the changes in the plasma membranes per se during cold acclimation. Based on these facts, it is important to know what changes in the plasma membranes are responsible for the development offreezing tolerance during cold acclimation. A major problem encountered has been the difficulty ofisolating the plasma membrane in a high degree ofpurity needed for detailed studies. Recently, we have succeeded in preparing plasma membranes from light-grown winter rye seedlings with a high degree of purity using an aqueous two-polymer phase partition system (29).In the present study, Evaluation of Freezing Resistance. Freezing tolerance of the crown was evaluated by the ion-leakage test. One g fresh weight of tissues were cut into 5-mm-long pieces and were frozen in test tubes at -3°C for 2 h with small pieces of ice. Thereafter, the tissues were cooled in steps of 5°C at 2-h intervals and held at the desired temperatures for 2 h. After thawing at OC, the tissues were immersed in 5 ml of distilled H20 and incubated at room temperature for more than 5 h by gently shaking before measuring the conductivity. The tissues frozen directly in the liquid nitrogen and unfrozen were taken as 100% and 0% in the relative conductivity, respectively.Isolation of Plasma Membranes. Crown tissues were homogenized with a Polytron PT 20 at the medium speed for 90 s in a medium consisting of 0.5 M sorbitol, 50 mm MOPS2-KOH (pH 7.6), 10 mm EGTA neutralized with KOH, 2.5 mm potassium metabisulfite, 4 mM SHAM, 1 mm PMSF, 5% (w/v) soluble PVP (mol wt 24,500), and 1% (w/v) defatted BSA. The homogenate was passed through two layers of gauze and then two layers of Miracloth, and then centrifuged at 14,000g for 15 min and 156,000g for 20 min. The 156,000g pellet (14,000-156,000g fraction) was used for plasma membrane isolation using the aqueous two-polymer phase partition technique as reported previously (29). The purity of the isolated plasma membrane was determined by phosphotungstate-chromate staining and by specific aggregation in the presence ofZnCl2 or at pH 4.5 as reported previously (29). The isolated plasma membranes were suspended in 0.5 M sorbitol-3 mm Tris-maleate at pH 7.3 and kept at -70°C until use.Lipid Extraction. Total lipids were extracted from membrane samples according to the method of Bligh and Dyer (1) with the 2 Abbreviations: MOPS, 34N-morpholino)propanesulfonic acid;