Effects of Temperature on Basal and Insulin-stimulated Glucose Transport Activities in Fat Cells

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 857, No. 23, Isue of December 1, pp Prcnted in L'. S.A Effects of Temperature on Basal and Insulin-stimulated Glucose Transport Activities in Fat Cells FURTHER SUPPORT FOR THE TRANSLOCATION HYPOTHESIS OF INSULIN ACTION* Osamu Ezaki and Tetsuro Konot (Received for publication, July 6, 1982) From the Department of Physiology, School of Medicine, Vanderbilt University, Nashville, Tennessee Effects of temperature on glucose transport in fat Finch (1) that the glucose transport activities in the basal cells were studied. In this system, the basal (no insulin) and plus insulin forms of fat cells show different responses to glucose transport activity was higher at approximately changes in the incubation temperature. More specifically, the 25-3 "C than at 37 "C, as previously reported (Vega, data obtained in our laboratory indicated that the basal (no F. V., and Kono, T. (1979) Arch. Biochem Biophys. 192, insulin) glucose transport activity is higher at 25 "C than at ). The stimulatory effect of low temperature (or 37 "C (7). Because of this latter observation, we suggested the insulin-like effect) was reversible and apparently earlier that translocation of the transport activity may not be required metabolic energy for both its forward and reverse reactions. By lowering the ATP level with 2,4- dinitrophenol, one could separately determine the insulin-like stimulatory effect of low temperature and its inhibitory effect on the transport process itself. The maximum level of stimulation by low temperature was greater at 1 "C than at 25-3 "C, but the rate of stimulation was considerably slower at 1 "C than at 25-3 "C. When cells were exposed to low temperature, the glucose transport activity in the plasma membranerich fraction was increased, while that in the Golgi-rich fraction was decreased. The Arrhenius plot of the basal glucose transport activity determined in the presence of dinitrophenol was apparently linear from 1 to 37 "C and parallel to that of the plus insulin activity measured either in the presence or absence of dinitrophenol. Insulin itself slowly stimulated the glucose transport activity at 1 "C. These results are consistent with the view that (a) low temperature, like insulin, induces translocation of the glucose transport activity from an intracellular storage site to the plasma membrane, (b) insulin stimulates glucose transport activity without changing its activation energy, and (c) subcellular membranes do not entirely stop their movement at a low temperature, e.g. at 1 "C. the only mechanism by which insulin stimulates glucose transport in fat cells (1, 2). In fact, the above temperature data were apparently in agreement with an alternat,e hypothesis, such as that proposed by Amatruda and Finch (1) and Pilch et al. (11) who suggested that insulin might stimulate glucose transport in fat cells by increasing fluidity of the plasma membrane. However, our most recent translocation data indicated that theffect of insulin on glucose transport could be accounted for mostly, if not entirely, by the translocation hypothesis (12). Therefore, our present work was initiated to reinvestigate effects of temperature on glucose transport. We approached this problem with the working hypothesis that low temperature has an insulin-like activity. During our present study, Haring et al. (13) independently reported that the glucose transport activity in fat cells was slowly stimulated at 15 "C. Portions of our present work have been published in abstract form (14). MATERIALS AND METHODS The sources of materials used in our present work were listed in our previous publications (1, 2). Fat cells were prepared by the collagenase method (15) from epididymal adipose tissue of Sprague- Dawley rats weighing approximately 15-25g. The isolated cells were suspended in Krebs-Henseleit 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid buffer, ph 7.4, supplemented with 2 mg/ml of bovine serum albumin (Fraction V) and 2 mm glucose (16). The cell suspension was equilibrated at 37 "C for at least 3 min before the determination of glucose transport activity (7). The data recently published by Suzuki and Kono (l), Kono et al. (2), Cushman and Wardzala (3), Karnieli et al. (4), The glucose transport activity of fat cells was assessed from the rate of uptake of 3-O-methyl-~-['~C]glucose (1 mm). The uptake of Wardzala and Jeanrenaud (5), and Gorga and Lienhard (6) this sugar isomer was routinely determined by the oil flotation method suggest that the glucose transport mechanism in muscle and of Gliemann et al. (17) as specified by Vega and Kono (7); the time fat cells is recycled between the plasma membrane and an for incubation of cells with the labeled sugar was 1 or 6 s. However, unidentified intracellular storage site and that the function of when a rapid plus insulin activity was to be determined accurately, insulin in this system is the recruitment of the transport the incubation time was reduced to 3 s and the oil flotation assay was mechanism from the storage site to the plasma membrane. On carried out as described by Whitesell and Gliemann (9). Since we were concerned about the effects of temperature, all the solutions and the other hand, it was previously reported by Vega and Kono containers that would come in contact with the incubation mixture (7), Czech (8), Whitesell and Gliemann (9), and Amatruda and were carefully equilibrated with the proper temperatures prior to use. The results of the transport assay are reported in terms of the * This work was supported by United States Public Health Service intracellular distribution space of 3-O-methyl-~-glucose (pl/g of cells) Grant 5R1 AM from the National Institutes of Health and by estimated at the end of the indicated incubation period. The intra- Grant 8-R-34 from the Juvenile Diabetes Foundation. The costs of cellular distribution space was calculated by subtracting the publication of this article were defrayed in part by the payment of "extracellular" distribution space of 3-O-methyl-~-glucose, a space page charges. This article must therefore be hereby marked "adver- that was not suppressed in the presence of 2Op~ cytochalasin B, from tisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate the total distribution space of the same sugar isomer. this fact. The glucose transport activity associated with the plasma mem- # To whom all correspondence should be addressed. brane-rich and Golgi-rich fractions was determined in the reconsti- 1436

2 tuted liposome system, as described (2, 12, 181, after the subcellular fractions were partially purified by differential and sucrose density gradient centrifugation (Method B) as described elsewhere (12). The results of this transport assay are shown in terms of the amount of the carrier-mediated D-glucose uptake by the reconstituted liposomes (nmol/mg of protein) during the indicated incubation period. The mediated transport activity was calculated by subtracting the "nonspecific" uptake of ~-['~C]glucose from the total uptake of D-['H~ glucose, as in our previous studies (2, 12, 18). Protein was assayed by the Bradford method (19); the standard used was crystalline bovine serum albumin. The amount of fat cells was estimated from the content of malate dehydrogenase as in our previous studies (7, 16). Since metabolic activities of fat cells were significantly different from batch to batch of the cell preparations, the data to be compared were obtained, if possible, with aliquots of a pooled cell preparation. However, when more than one pooled cell preparation was to be used in one experiment, we included two or more common controls in each preparation and made certain that all the control activities were in good agreement. RESULTS Fig. 1 shows the experimental protocols used in our present study. In Plan I, cells were exposed to various temperatures (1-37 "C) either in the presence or absence of 2,4-dinitrophenol, and the glucose transport activity was measured at the same temperatures. As shown in Fig. 2 (upper curue), when cells that had been equilibrated at 37 "C were exposed to lower temperatures in the absence of dinitrophenol (Plan IA), a biphasic temperature effect was observed. The glucose trans- port activity measured under these conditions was maximum at approximately 25-3 "C, in agreement with our previous data (7). In contrast, when cells were first treated with 1 mm dinitrophenol and then exposed to various low temperatures (Plan IB), a simple monophasic curve was obtained between 1 and 37 "C (Fig. 2, lower curve). Results similar to these presented in Fig. 2 were also obtained when 2 mm KCN or 1 mm sodium azide was substituted for dinitrophenol (data not shown). The results of these experiments with metabolic inhibitors were consistent with the view that low temperature, like insulin, stimulated glucose transport activity by a mechanism which required either ATP or metabolic energy. As reported previously, the above three agents rapidly lower the ATP level in fat cells and concomitantly inhibit the insulin- PLAN I EXPERIMENTAL PROTOCOLS -~ Temperature Effects on Glucose Transport 1437 TRANSPORT ASSAY rnln C.~ IO TEMPERATURE, FIG. 2. Effects of preliminary treatment of fat cells with 2,4- dinitrophenol on the basal (no insulin) glucose transport activity measured at different temperatures. The glucose transport activity of fat cells was determined at the indicated temperatures according to either Plan IA (upper curue) or Plan IB (Zouw curoe), as specified in the legend to Fig. 1. The data show the mean values & S.E. (n = three separate experiments) of the intracellular distri- bution spaces of 3-O-methyl-~-glucose (3--MG) measured at 6 s of incubation. Note that the results obtained at 37 "C suggest that dinitrophenol (DNP) had a certain injurious effect on fat cells as reflected in a lower transport activity I m TIME mtn FIG. 3. Reversibility of the stimulatory effect of low temperature. Aliquots of a pooled fat cell preparation were divided into three groups. Group A was kept at 37 "C as control. Groups B and C were exposed to 25 "C for 2 min; thereafter, Group B was simply brought back to 37 "C, while Group C was first treated with 1 mm 2,4- dinitrophenol (DNP) for 5 min at 25 "C and then brought back to 37 "C. At the end of the indicated incubation periods, Groups A and B were also mixed with 1 mm 2,4-dinitrophenol, and the glucose transport activities in all of the cell preparations were determined at 37 "C as in Plan I1 (Fig. 1). The data show the mean values +- S.E. (n = 4-5). 3-O-MG, 3-O-methyI-~-glucose. OC I 1 - x 5 1 mln FIG. 1. Experimental protocols. In Plan IA, cells that had been equilibrated at 37 "C were exposed to various temperatures (1-37 "C) for 1 min prior to the transport assay at the same temperatures. In Plan IB, cells were first incubated for 5 min with 1 mm 2,4- dinitrophenol (DNP) which would decrease the cellular level of ATP. Subsequently, the cells were exposed to various temperatures (1-37 "C) and assayed for transport activity as above. In Plan 11, cells were first exposed to various temperatures (37 "C in Plan ITA and 1-3 "C in Plan IIB); they were then mixed with 1 mm 2,4-dinitrophenol and incubated for 5 min. Subsequently, all the cell preparations were brought to 37 "C and assayed for transport activity at 37 "C. As described previously (2), 2,4-dinitrophenol was used as its Tris salt which is highly water-soluble. dependent stimuiation of glucose transport (2-23), presumably by blocking the energy-dependent translocation of the transport mechanism (2). In agreement with our previous data (7), low temperature (25-3 "C) had no atypical effect on the plus insulin activity (see Arrhenius plots presented later in Fig. 6). The stimulatory effect of low temperature on the basal glucose transport activity was reversible (Fig. 3); when cells that had been exposed to 25 "C were brought back to 37 "C, the elevated transport activity was lowered to the initial level within 15 min. This reverse reaction was not seen, however, if the stimulated cells were first treated with 2,4-dinitrophenol at 25 "C and then brought back to 37 "C (Fig. 3). Similar results were also obtained when 2 mm KCN or 1 mm sodium

3 1438 Temperature Effects on Glucose Transport azide was substituted for dinitrophenol (data not shown). These results suggested to us that ATP or metabolic energy was involved in the reversal of the low temperature effect. As reported previously, the above three agents block the reversal of the insulin effect on glucose transport (24-27), possibly by inhibiting the energy-dependent retranslocation of the glucose transport mechanism from the plasma membrane to the intracellular storage site (2). The above observation that reversal of the low temperature effect is blocked with dinitrophenol prompted us to design a new experimental protocol designated as Plan I1 and presented earlier in Fig. 1. In this protocol, we exposed fat cells to various low temperatures, "fixed" the low temperature effect with dinitrophenol treatment, and then measured the TEMPERATURE, "C FIG. 4. Effects of exposure of fat cells to different temperatures for different lengths of Aliquots time. of a fat cell suspension were exposed to the indicated temperatures for the indicated periods. The experiments were carried out according to Plan 11. The data show the mean values f S.E. (n = 3-5). 3--MG, 3-O-methyl-~glucose. I / I ' TIME, rn~n FIG. 5. Time courses of stimulation of glucose transport activity. Aliquots of a fat cell suspension were incubated for 4 h at either 1 or 37 "C, either in the presence or absence of 1 nm insulin, which was added at min. The experiments were carried out according to Plan 11. The data show the mean values f S.E. (n = 3-4). The urea space was determined by the oil flotation method (17). Note that the apparent plus insulin activity estimated at 37 "C does not represent the initial rate of transport as the time course of the reaction was curvilinear under the given conditions (7). 3--MG, 3--methyl- D-ghcOSe. glucose transport activity in all the cell preparations at 37 "C. The advantage of this method was that one could determine the insulin-like (stimulatory) effect of low temperature independent of its inhibitory effect on the glucose transport proc- ess itself. This protocol was used in the rest of our present work unless otherwise stated. As shown in Fig. 4, when the incubation period was increased from 1 min to 4 h, the maximum level of stimulation by low temperature was increased, and, at the same time, the point of the maximum stimulation was shifted towards the lower temperature. These results were consistent with the interpretation that the maximum level of stimulation was greater at 1 "C than at 25-3 "C, but the rate of stimulation was substantially less rapid at 1 "C than at 25-3 "C. The actual time courses of the stimulatory effects of low temperature and insulin are presented in Fig. 5, which indicates that the rate of increase in the basal (no insulin) activity at 1 "C was indeed very slow as compared to the insulin-dependent stimulation at 37 "C. Even the insulin-dependent stimulation was slower at 1 "C than at 37 "C. Nevertheless, the effect of insulin was clearly detectable at this low temperature. This effect of insulin at 1 "C, like its effect at 37 "C, was completely blocked by treatment of cells with 1 mm 2,4-dinitrophenol (data not shown). The apparent level of stimulation of the glucose transport activity by a 4-h incubation of cells at 1 "C was approximately 26% (=lo X ( )/( )) of the maximum insulin-dependent stimulation at 37 "C (Fig. 5). However, as is also shown in Fig. 5, the urea space of fat cells was gradually increased during a 4-h incubation to a larger extent at 37 "C than at 1 "C. This indicates that the apparent plus insulin activity observed at 37 "C by incubation of cells with 3-- methyl-d-glucose for 1 s is not directly comparable with the TABLE I Effects of low temperature on the glucose transport activities as observed in the subcellular fractions and in intact fat cells Aliquots of fat cells were incubated for 4 h at 37 "C (Group A, control), at 1 "C (Group B, 1 "C), or at 37 "C with the addition of 1 nm insulin at 3 h, 5 min (Group C, plus insulin). Subsequently, all the preparations were mixed with I mm 2,4-dinitrophenol as in Plan I1 (Fig. 1). In Experiment I, cells treated as above were washed and homogenized, and the plasma membrane-rich (PM) and Golgi-rich (GOLGI) fractions were prepared as described under "Materials and Methods." The glucose transport activity in the two subcellular fractions was assayed in the reconstituted liposome system, as described (2). The data show the mean values f S.E. of eight observations made in eight separately reconstituted liposome preparations that were prepared in two separate experiments. In Experiment 11, the treated cells were subjected to the transport assay as in Figs The incubation periods of cells with labeled 3-O-methyl-~-ghcose were 1 s for assay of the basal (no insulin) activity and 3 s for determination of the plus insulin activity; the results of the latter assay are presented having been divided by.3. The data show the mean value f S.E. of three separate experiments. Note that the results of Experiment I1 show the initial rate of transport (9), but those of Experiment I do not; as reported previously (2), the uptake of o-glucose by reconstituted liposomes was curvilinear. Experiments I and I1 were done on different occasions with different batches of fat cells. Preparation Relative activitv Transnort A. Control B. 1 "C C. +Insulin effect (B - A)/(C - A) Experiment I (D-glucose uptake by reconstituted liposomes)" PM 2.4 f f.2b 13.7 f.25 GOLGI 43.9 f f.7h 26.2 f.7.27 Experiment I1 (3-O-methy~-o-glucose uptake by cells)' Cells 3.5 k t OBh 45.3 f..22 nmol/mg of protein determined at 4 S. hp <.1 (A uersus B) when tested by Student's t test. Fl/g of cells at 1 s.

4 data obtained at 1 "C. Therefore, in subsequent experiments, we measured the plus insulin activity at 37 "C by exposing cells to the substrate sugar for 3 s (rather than 1 s). The data obtained under those conditions are affect.ed not by the change in the urea (i.e. water) space mentioned above (9). We next examined whether it could be considered that low temperature, like insulin, induces translocation of the glucose transport activity from the Golgi-rich fraction to the plasma membrane-rich fraction. As summarized in Table I, the results of this experiment were affirmative. When cells were exposed to 1 "C for 4 h, the glucose transport activity in the plasma membrane-rich fraction was increased while the activity in the Golgi-rich fraction was decreased. The insulin-like effects of low temperature observed in the two subcellular fractions were approximately 25-27%; of the maximum insulin effect recorded in the same fractions. The corresponding value estimated in experiments with intact fat cells was approximately 22% (Table I). We interpreted these results (Table I and Fig. 5) as indicating that the insulin-like effect of low temperature on the glucose transport activity in fat cells was largely, if not quantitatively, explained by the above mentioned translocation hypothesis. : -INSULIN,-DNP, 6s :-INSULIN,+DNP,6s Temperature Effects on Glucose Transport 1439 treatment by Plan I, either in the presence or in the absence of dinitrophenol. Since the slope of an Arrhenius plot represents the activation energy of the reaction, these results indicate that the "true" activation energy of the basal glucose transport activity (i.e. the activity determined in the presence of dinitrophenol) is indistinguishable from the activation energy of the plus insulin activity, which is not significantly affected by the presence of dinitrophenol. The exact values of the activation energy calculated from the above Arrhenius plots for the transport of 3-O-methyl-~-glucose were 1.5 kcal/ mol for the basal transport activity in the presence of dinitrophenol and 1.7 kcal/mol for the plus insulin activity either in the presence or absence of dinitrophenol. DISCUSSlON InsuZin-like Effect of Low Temperature-Our present data indicate that (a) the basal (no insulin) glucose transport activity in fat cells is increased when cells are exposed to 1-35 "C (Figs. 2-5), (b) this low temperature effect is reversible (Fig. 3), (e) ATP or metabolic energy appears to be involved both in the development and in the reversal of this low temperature effect (Figs. 2 and 3), and (d) low temperature, Fig. 6 presents Arrhenius plots of the glucose transport like insulin, appears to induce translocation of the glucose activities determined under several different conditions. The transport activity from the intracellular storage site to the results show that the Arrhenius plot of the basal transport plasma membrane (Table I). It could be speculated, as a activity measured in the absence of dinitrophenol (Plan 1A in possible underlying mechanism, that low temperature might Fig. 1) was curvilinear, as might be expected from the data facilitate the generation of an insulin-like signal (which is shown earlier in Fig. 2. In contrast, the Arrhenius plot of the unknown) in the cell; alternatively, it is also conceivable that basal activity determined in the presence of dinitrophenol the changes in the incubation temperature simply shift the (and presented earlier in Fig. 2) was apparently linear between steady state of distribution of the transport activity between 1 and 37 "C and parallel to the plots of the plus insulin the plasma membrane and the storage site. activities of fat cells that had been exposed to insulin prior to The time course of stimulation by low temperature observed in our present study (Fig. 5) is basically in agreement with TEMPERATURE, OC that previously reported by Haring et al. (13), although their data do not clearly distinguish the insulin-like effect of low temperature from its inhibitory effect on the transport process itself. The insulin-like effect of low temperature is probably the cause of the atypical temperature effects that were observed by Czech (8), Whitesell and Gliemann (9), and Ama- truda and Finch (1) on the basal glucose transport activity in fat cells. We further suggest that the insulin-like effect of low temperature may also affect the glucose transport activity in muscle because it was reported by Kipnis and Cori (28) that, unlike the plus insulin activity, the basal glucose transport activity in rat diaphragm was almost constant between 17 and 37 "C. In addition, Brown et al. (29) and Yu and Gould (3) noted that the glucose transport activities in rat diaphragm and soleus muscle were stimulated when the tissue prepar.1- tions were chilled to "C prior to incubation at 3'7 "C. It should be noted, however, that certain adipocyte preparations, such as those used by Olefsky (31) and Vinten (32), apparently did not respond to the insulin-like effect of low temperature. The reason for this is not clear although several possibilities were considered by Amatruda and Finch (1). Activation Energy of Glucose Transport-Our present data indicate that (a) the development of the insulin-like effect of' low temperature can be blocked if the cells are deprived of ATP by treatment with 2,4-dinitrophenol, KCN, or sodium azide (Fig. 2), (b) the Arrhenius plot of the basal transport /T K" FIG. 6. Arrhenius plots of the transport data. The lower two curves show the Arrhenius plots of the basal (no insulin) glucose (3- -methyl-d-glucose (3--MG)) transport activities determined after treatment of cells by Plan I (Fig. 1) either in the presence or in the absence of 1 mm 2,4-dinitrophenol (DNP)(and presented earlier in Fig. 2). The upper two curves are the Arrhenius plots of the plus insulin activities of fat cells that had been exposed to 1 nm insulin for 1 min prior to treatment by Plan I. The incubation periods of cells activity measured in the presence of dinitrophenol is apparwith labeled 3-O-methyl-~-glucose were 6 s for determination of the ently linear and parallel to those of the plus insulin activities basal activity, 1 s for measurement of the plus insulin activity at 1- measured either in the presence or in the absence of dinitro- 25 "C, and 3 s for assay of the plus insulin activity at 3-37 "C; the phenol (Fig. 61, and (c) the true activation energy of the basal results of the last determinations are presented having been divided by.3. The data show the mean values f S.E. (n = 3-5). Note that transport activity is indistinguishable from that of the plus the slopes of the curves in this figure (a semilogarithmic chart) are insulin activity. unaffected regardless of whether the transport activities are shown in Previously, Ludvigsen and Jarett (33) reported that no units/l s or units/6 s. effect of insulin (added to cells) was detectable on the acti-

5 1431 Temperature Effects on Glucose Transport vation energy (i.e. on the temperature coefficient) of the 2. Kono, T., Suzuki, K., Dansey, L. E., Robinson, F. W., and Blevins, glucose transport activity when the latter was measured in a T. L. (1981) J. Biol. Chem. 256, cell-free system consisting of the resealed plasma membrane 3. Cushman, S. W., and Wardzala, L. J. (198) J. Biol. Chem. 255, vesicles. Likewise, we previously reported that no insulin Karnieli, E., Zarnowski, M. J., Hissin, P. J., Simpson, I. A., Salans, effect was detectable in the apparent temperature coefficient L. B., and Cushman, S. W. (1981) J. Biol. Chem. 256, of the glucose transport activity when the latter was solubi lized from the isolated plasma membrane fraction and assayed in the reconstituted liposome system (2). These two negative 5. Wardzala, L. J., and Jeanrenaud, B (1981) J. Biol. Chern. 256, observations are consistent with the view that low tempera- 6. Gorga, J. C., and Lienhard, G. E. (1982) Fed. Proc. 41, ture modulates only the distribution of the glucose transport Vega, F. V., and Kono, T. (1979) Arch. Biochem. Biophys. 192, activity between the plasma membrane and the storage site 8. Czech, M. P. (1976) Mol. Cell. Biochem. 11, and that it does not change the physicochemical properties of 9. Whitesell, R. R., and Gliemann, J. (1979) J. Biol. Chem. 254, the transport activity Earlier, Olefsky (31) and Vinten (32) also reported that 1. Amatruda, J. M., and Finch, E. D. (1979) J. BioE. Chem. 254, insulin did not change the activation energy of glucose trans port; however, as mentioned in the previous section, the 11. Pilch, P. F., Thompson, P. A., and Czech, M. P. (198) Proc. Natl. Acad. Sci. U. S. A. 77, characteristics of their cell preparations were apparently dif- 12. Kono, T., Robinson, F. W., Blevins, T. L., and Ezaki,. (1982) J. ferent from those of the preparations used by others (7-1). Biol. Chem Action of Insulin ut 1 "C-It is now clear that insulin 13. Hiiring, H. U., Biermann, E., and Kemmler, W. (1981) Am. J. slowly, but definitely, stimulates glucose transport in fat cells Physiol. 24, E556-E565 either at 1 "C as observed in our present study (Fig. 5) or at 14. Ezaki, O., and Kono, T. (1982) Fed. Proc. 41, "C as reported by Haring et al. (13). These observations 15. Rodbell, M. (1964) J. Biol. Chem. 239, Kono, T., and Barham, F. W. (1971) J. Biol. Chem. 246, 621- seem to raise a new question since it was previously suggested 6216 that the glucose transport activity is translocated by a mech- 17. Gliemann, J., Osterlind, K., Vinten, J., and Gammeltoft, S. (1972) anism that involves endocytotic and exocytotic reactions (1, Biochim. Biophys. Acta , 4) which, as reviewed by Silverstein et al. (34), would not 18. Robinson, F. W., Blevins, T. L., Suzuki, K., and Kono, T. (1982) occur below the threshold temperature of approximately Anal. Biochem. 122, "C. However, it was recently reported by Weigel and Oka 19. Bradford, M. M. (1976) Anal. Biochem. 72, Kono, T., Robinson, F. W., Sarver, J. A., Vega, F. V., and Pointer, (35) that isolated rat hepatocytes showed a significant level of R. H. (1977) J. Biol. Chem. 252, endocytosis below the transition temperature of 2 "C, at least 21. Chandramouli, V., Milligan, M., and Carter, J. R., Jr. (1977) down to 1 "C. In addition, it was noted by Steinman et al. Biochemistry 16, (36) that cultured L-cells exhibited a weak, but significant, 22. Yu, K. T., and Gould, M. K. (1978) Am. J. Physiol. 234, E47- pinocytotic activity at 2 "C. Therefore, we suggest (a) that the E416 membrane structures in mammalian cells do not entirely stop 23. Siegel, J., and Olefsky, J. M. (198) Biochemistry 19, Kono, T., Vega, F. V., Raines, K. B., and Shumway, S. J. (1977) their movement below the threshold, or transition, ternpera- Fed. Proc. 36, 341 ture, and (b) that the glucose transport system may be trans- 25. Vega, F. V., Key, R. J., Jordan, J. E., and Kono, T. (198) Arch. located slowly at temperatures below 2 "C. Biochem. Biophys. 23, In conclusion, our present results indicate that the apparent 26. Ciaraldi, T. P., and Olefsky, J. M. (1981) Arch. Biochem. Biophys. temperature coefficient of the glucose transport activity is 28, affected, not by insulin, but by the insulin-like effect of low 27. Laursen, A. L., Foley, J. E., Foley, R., and Gliemann, J. (1981) Biochim. Biophys. Acta 673, temperature. Therefore, it could be postulated as the working 28. Kipnis, D. M., and Cori, C. F. (1957) J. Biol. Chem. 224, hypothesis that insulin stimulates glucose transport in fat cells 29. Brown, D. H., Park, C. R., Daughaday, W. H., and Cornblath, M. mainly, if not entirely, by facilitating translocation of the (1952) J. Biol. Chem. 197, glucose transport activity from the storage site to the plasma 3. Yu, K. T., and Gould, M. K. (1981) Diabetologia 21, membrane without changing the physicochemical characteristics of the transport activity. 31. Olefskv. J. M. (1978) Biochem. J. 172, Acknowledgments-We are grateful to Dr. Sidney P. Colowick and Dr. Charles R. Park for reading this manuscript. REFERENCES I. Suzuki, K., and Kono, T. (198) Proc. Natl. Acad. Sci. U. S. A. 77, Vintei,'J. (1978) Biochim. Biophys. Acta 511, Ludvigsen, C., and Jarett, L. (198) Diabetes 29, Silverstein, S. C., Steinman, R. M., and Cohn, 2. A. (1977) Annu. Rev. Biochem. 46, Weigel, P. H., and Oka, J. A. (1981) J. Biol. Chem. 256, Steinman, R. M., Silver, J. M., and Cohn, Z. A. (1974) J. Cell Biol. 63,