Proc. Natl. Acad. Sci. USA Vol. 76, No. 4, pp. 1901-1905, April 1979
Cell Biology
Lateral and vertical displacement of integral membrane proteins during lipid phase transition in Anacystis nidulans* (plasma membrane/thylakoid membrane/reversible particle migration/freeze-fracture)
PAUL A. ARMONDt AND L. ANDREW STAEHELINt tDepartment of Plant Biology, Carnegie Institution of Washington, Stanford, California 94305; and tDepartment of Molecular, Cellular, and Developmental Biology, University of Colorado, Boulder, Colorado 80309
Communicated by Winslow R. Briggs, January 26, 1979
ABSTRACT Alterations in membrane structure as a result of lipid phase transitions have been studied in Anacystis nidulans, a blue-green alga. Cells grown at 38'C were subjected to temperature transitions of 380C - 21'C and 380C - 100C, previously shown to produce substantial changes in photosynthetic activities, and examined by freeze-fracture electron microscopy. As a result of these treatments, large particle-free regions appeared on the fracture faces of both the plasma and thylakoid membranes. Particle density measurements suggest that the displacement of the integral membrane protein complexes occurs in both lateral and vertical directions. Returning the cells to 380C resulted in the restoration of normal membrane morphology, indicating that the proteins were not lost from the membrane. Such displacement of the integral membrane protein complexes could contribute significantly to the temperature-dependent alterations in the functional activity of membrane-bound enzymatic complexes.
The temperature dependency of the physical state of membranes has been extensively studied in model systems (1-9), intact cells (10-19), and isolated organellar and membrane fractions (10, 19-21). In these studies, it has been shown that lipid species and fatty acid composition have a direct effect upon the phase transition temperature of the membrane. It has also been demonstrated that alterations in the activation energy of the membrane-bound enzymatic systems are correlated with the membrane phase transitions (10, 11, 14, 19-21). Although the protein constituents of the membrane can affect the fluidity of membranes, they apparently play no role in determining the phase transition temperature (7). However, correlations between enzymatic function and the physical state of the membrane stress the fact that protein-lipid interactions are vital for the proper physiological functioning of the membrane. Previous studies of the blue-green alga Anacystis nidulans have shown that a correlation exists between the physical state of the lipids and such parameters as chlorophyll a fluorescence, P700 reduction, and photosynthetic oxygen evolution, parameters that involve integral membrane protein complexes (10). It was demonstrated that for cells grown at 380C, a change in slope in Arrhenius plots of such membrane functions occurred at approximately 220C to 240C. More recently, a second slope change has been detected at approximately 100C to 12'C (22). Linden et al. (14) have interpreted similar observations to indicate the onset (slope change at the higher temperature) and completion (slope change at the lower temperature) of the lateral phase separation of membrane lipids. Recent electron spin resonance experiments (23) have indicated that, in Anacystis lipids from cells grown at 380C, the onset of the lateral phase separation occurs at a temperature somewhat higher than 240C, but the experiments are in agreement with the compleThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
tion of the phase transition at 100C. To learn more about the structural basis of the functional changes, we qualitatively and quantitatively examined the membrane structural organization of A. nidulans by freeze-fracture electron microscopy at temperatures at which the alterations in physiological activities occur.
MATERIALS AND METHODS A. nidulans was grown at 380C in Kratz and Meyers' C medium (24) bubbled with air enriched to 5% CO2. Samples were either maintained under the growth conditions at 38°C or transferred to temperature baths at 21°C or 10°C (210C and 10°C are slightly below the temperatures at which the discontinuities in the slope of the Arrhenius plot for certain enzymatic activities are observed). The overall cooling rate was approximately 1°C/min. An additional sample was transferred to the 21°C bath, maintained at that temperature for 20 min, and returned to the 38°C bath. After the cell suspensions had equilibrated at their respective final temperatures for 15 min, sufficient glutaraldehyde for a final concentration of 2% (wt/vol) was added. The cell suspensions were gently stirred for 5 min, removed from the baths, and centrifuged at 10,000 X g for 10 min. The remaining operations were carried out at 200C. The pelleted cells were resuspended in the growth medium, and glycerol was added slowly over a 1-hr period to a final concentration of 35% (vol/vol). The cells were centrifuged at 10,000 X g for 15 min, and the pelleted cells were prepared for freeze-fracture as described (25). Particle density measurements for each fracture face were made on six to eight micrographs enlarged to X125,000. Measurements were made on the flattest possible membrane areas; sharply curving membrane regions were avoided. Percentage of particle-free membrane areas was determined on the same micrographs used for the particle density measurements. The myelinic preparation was made by allowing a 10 mg/ml suspension of egg leci*thin (obtained from Nutritional Biochemical Corp., Cleveland, OH) in a 20% glycerol/water solution to swell for 1 hr at room temperature. Fracture-face nomenclature is according to Branton et al. (26). RESULTS The freeze-fracture preparations of control (38°C) A. nidulans membranes (Fig. 1) exhibited an even distribution of particles on all fracture faces. Neither the plasma membrane (Fig. 1A) nor the thylakoid membrane (Fig. 1B) exhibited any extensive areas devoid of particles. In contrast, membranes of cells that had undergone a 38°C - 21°C transition showed large partiAbbreviations: PF, protoplasmic fracture face; EF, exoplasmic fracture face. * Carnegie Institution of Washington Department of Plant Biology publication no. 640. 1901
1902
Proc. Natl. Acad. Sci. USA 76 (1979)
Cell Biology: Armond and Staehelin
FIG. 1. Freeze-fractured control cells of A. nidulans grown and fixed with glutaraldehyde at 380C. Neither the plasma membrane (A) nor the thylakoid membrane (B) exhibits any extensive particle-free regions on either the protoplasmic fracture face (PF) or the exoplasmic fracture face (EF). (A, X75,000; B, X90,000.) Bars equal 0.5 ,m.
cle-free regions on the thylakoid membranes (Fig. 2A) as well as on the plasma membrane (Fig. 2B). The extent of such particle-free regions was increased in the membranes of cells that had
experienced
the 380C
100C transition
(Fig.
2 C and
D).
In addition to the extensive particle-free areas, apparent physical breaks in both the plasma membrane and thylakoid membranes were observed in the 380C 100C transition
sample (arrows, Fig. 2D). The membranes of the cell samples that had undergone the 380C 210C 380C treatment were visually indistinguishable from membranes of the control cells shown in Fig. 1. Particle density measurements were made for the PF and EF fracture faces of the plasma and thylakoid membranes of all cell samples (Table 1). The measured particle densities --
--
consider only those membrane areas containing particles; areas devoid of particles or immediately bordering such areas were excluded from these measurements. As shown in Table 1, the appearance of extensive particle-free regions in the membranes of the 380C - 21'C and 380C -- 100C treated cells did not result in large increases in the particle density of the remaining areas. Measurements of the extent of the particle-free areas (Table 1) indicate that the increases in particle density observed are insufficient to account for the particle-free regions by only a lateral movement of the membrane particles. Measurements have been made of the diameter of the particles on the PF face of the thylakoid membrane, the fracture face that exhibited the most extensive particle-free regions at both treatment temperatures. In order to maintain particle density after the 380C
Proc. Nati. Acad. Sci. USA 76 (1979)
Cell Biology: Armond and Staehelin
1903
FIG. 2. Fracture faces of cell membranes subjected to a 380C - 210C (A, B) or a 380C - 100C (C, D) temperature transition. After treatment
at 210C, large particle-free regions are visible. The PF face of the thylakoid membranes (A) shows a clear delineation between particle-free and particle-containing regions. This delineation is not as evident in the PF fracture face of the plasma membrane (B). Note the verrucose texture of the particle-free regions. When cells were subjected to a 380C 100C treatment (C, D), these regions became much more extensive, with large particle-free areas visible on the EF face of the plasma membrane (C) and PF faces of both the plasma (left) and thylakoid (right) membranes (D). In addition, the 380C - 100C treatment produced both verrucose and faint reticulate textured regions in the particle-free areas (see also Fig. 3). Arrows indicate apparent physical breaks in the membranes (D). (X70,000.) Bar equals 0.5 ,m. -
- 21'C transition, the average particle diameter would have to increase from 94 A (measured) to 162 A (calculated on the basis of 67% particle-free regions). An increase of only 15% (average diameter equal to 108 A) was observed in the 380C -- 21'C sample. When the cells were restored to the growth temperature (the 380C -- 21'C -- 380C sample), the particle-free regions were no longer observed, and the overall particle density of the membrane fracture faces returned to that
of the control cells. Closer examination of the micrographs of membrane fracture faces devoid of particles (presumably less fluid lipid regions) reveals that these faces exhibit distinct surface textures. Two basic surface textures can be distinguished (Fig. 3A), a verrucose surface (Sl) and a faint reticulate surface (S2). Only the verru-
cose surface is observed in the 380C - 21'C samples, while both types of textured regions can be recognized in the 380C 100C specimens. In myelinic preparations of pure egg lecithin (Fig. 3d), very similar textures can also be seen on the fracture faces of the bilayer membranes. Both in the particlefree A. nidulans membrane regions and in the egg lecithin bilayers these differently textured areas can exist side by
side. DISCUSSION of the and plasma thylakoid membranes of A. nidulans Analysis by freeze-fracture electron microscopy reveals that extensive alterations in membrane structural organization are associated with the phase transition of the lipid constituents of the mem-
Proc. Nati. Acad. Sci. USA 76 (1979)
Cell Biology: Armond and Staehelin
1904
Table 1. Density of particles on fracture faces and percentage of particle-free areas of the plasma and thylakoid membranes of A. nidulans with various temperature treatments Thylakoid membrane Plasma membrane % particle% particleTemperature Particles/hm2 Particles/,um2 PF free regions EF free regions EF PF treatment
380C (control) 380C - 211C 380C - 100C 38oC 21'C -381C -
1572 i 98 1484 ± 101 1392 I 132 1632 ± 198
3434 ± 119 3853 ± 149 3867 i 132 3277 I 251
None 21 43 None
703 ± 82
3703 ± 190
759 1 156 755 106 723 101
3804 i 169 4278 235 3324 257
None 67 77 None
Density figures are mean ± SD.
branes. For cells grown at 380C, treatment at 21'C results in particle-free areas on the PF and EF fracture faces of both plasma and thylakoid membranes. At 10'C, these particle-free areas become more extensive, and breaks in the membranes appear. An examination of the particle-free areas reveals verrucose and reticulate textured regions on both plasma and thylakoid membrane fracture faces. It is now well documented that bilayer membranes made of different amphipathic lipids can exhibit different kinds of fracture face textures after being cooled below their phase transition temperature (4). Thus, the verrucose and reticulate textured particle-free areas of the A. nidulans membranes may correspond to regions containing different kinds of lipids, or, as in the case of the lecithin bilayers, a possible different ordering of the lipids. Both lateral (15, 16, 18, 21, 27) and vertical (17, 28) displacement of membrane particles have been proposed to explain the formation of particle-free areas in cooled membranes.
FIG. 3. Freeze-fractured A. nidulans cell after a 38°C 10°C temperature transition (A), and a freeze-fractured egg lecithin -
membrane (B). Surface characteristics of the particle-free regions in the cell membranes (A) could be generally categorized as verrucose
(S2). Such surface characteristics are also visible in bilayers (B), indicating that the surface characteristics of the particle-free regions are a function of the lipid constituents of the membrane. (A, X80,000; B, X56,000.) Bars equal 0.5 Am.
(Si) or reticulate pure egg lecithin
These possibilities may be distinguished by examination of the particle densities in the particle-containing areas. If the redistribution of particles involves only lateral movement, a predictable increase in the particle density of particle-containing regions should occur. For the A. nidulans membranes (Table 1), such increases are not evident on the EF faces of the plasma and thylakoid membranes, and are only marginally observed for the remaining fracture faces. The increases in particle density that are observed are insufficient to account for the extent of the observed particle-free regions if lateral displacement is the only mode of particle redistribution. Particle size changes also do not appear to be significant. It would appear that the thermotropic particle redistribution in A. nidulans involves both lateral and vertical displacement of membrane particles. While the pure lateral displacement of particles can easily be explained by their exclusion from crystalline lipid domains, three possible mechanisms may be involved in the apparent vertical movement of particles: (i) The movement of the particles may only be an apparent one, caused by increases in membrane thickness induced by the phase transition (29, 30). Alterations in membrane thickness have been shown to alter the density of particles observed on membrane fracture faces (31). (ii) The movement of the integral membrane protein complexes in a vertical direction may be a direct consequence of the physical events associated with the lipid state change. Alterations in microviscosity (which occur during lipid phase transition) have been correlated with an apparent vertical displacement of proteins in erythrocyte membranes (28). (iii) Certain integral membrane protein complexes could be partly squeezed out of the central plane of the membrane (and therefore not seen as freeze-fracture particles) by the increased lateral pressure on the transmembranous proteins exerted by the expanding regions of lipids in the gel phase. The observed loss of particles from the fracture faces may involve all of these possibilities. As shown in Fig. 4, the loss of particles observed in the transition from a fluid (Fig. 4A) to a mixed-phase state (Fig. 4 B and C) does not have to be the result of a complete loss of the protein complexes from the membranes. Exclusion of all the particles from the fracture plane of less fluid regions of the membrane (Fig. 4B) or maintenance of the number of particles at the fracture plane of fluid regions of the membrane (Fig. 4C) by the exclusion of some particles would be sufficient to explain the observed effect of the fluidphase to mixed-phase transition on membrane structure. Because both of these possible mechanisms allow for a rapid restoration of normal particle densities upon rewarming of the membranes, as was observed for A. nidulans (see Table 1), it is not possible to distinguish between these alternatives at the present time. It should be noted, however, that the alternative presented in Fig. 4B requires that portions of some membrane complexes previously exposed to water are in a hydrophobic region of the membrane after the phase transition. This alternative would require a substantial input of energy or large
Cell Biology: Armond and Staehelin
A
B
f
f
f
f
f
I
npUS0U009URII
i
Proc. Nati. Acad. Sci. USA 76 (1979) 1905 The authors thank M. DeWit and T. Giddings for their technical assistance, Or. 1D. Fork for providing the cell cultures used in this study, and Drs. J. Berry and J. Raison for their helpful suggestions. P.A.A. was supported by a Carnegie Institution of Washington Postdoctoral Fellowship. This work was supported in part by National Institute of General Medical Sciences Grant GM 18639 to L.A.S. 1. Untracht, S. H. & Shipley, G. G. (1977) J. Biol. Chem. 252, 4449-4457. 2. Phillips, M. C., Ladbrooke, B. D. & Chapman, D. (1970) Biochim. Biophys. Acta 196,35-44. 3. Hui, S. W. & Parsons, D. F. (1975) Science 190, 383-384. 4. Gulik-Krzywicki, T. (1975) Biochim. Biophys. Acta 415, 1-28. 5. Lee, A. G. (1977) Biochim. Biophys. Acta 472, 237-281. 6. Lee, A. G. (1977) Biochim. Biophys. Acta 472, 285-344. 7. Papahadjopoulos, D., Moscarello, M., Eylar, E. H. & Isac, T. (1975) Biochim. Biophys. Acta 401, 317-335. 8. Hui, S. W., Parsons, D. F. & Cowden, M. (1974) Proc. Natl. Acad. Sci. USA 71,5068-5072. 9. Shipley, G. G., Green, J. P. & Nichols, B. W. (1973) Biochim. Biophys. Acta 311, 531-544. 10. Murata, N., Troughton, J. H. & Fork, D. C. (1975) Plant Physiol.
t C
I
FIG. 4. Model for the vertical displacement of membrane particles in A. nidulans. The loss of particles from the fracture plane in the transition from a fluid-phase (A) to a mixed-phase state (B and C) may involve different processes, depending upon whether the proteins remain in the less fluid portion of the membrane (B) or laterally migrate to the more fluid region of the membrane (C). If the proteins remain in their original position, integral membrane proteins that are not transmembranous (B, upper leaflet) could become undetectable at the fracture plane because of the change in bilayer thickness induced by the change of the physical state of the lipids. For such nontransmembranous proteins, approximately the same degree of surface exposure could be maintained. Transmembranous proteins also would require some increase in surface exposure to account for nondetectability at the fracture plane (B, lower leaflet). If the proteins are displaced to a more fluid region of the membrane (C), some proteins might be excluded from the fracture plane by the combined lateral pressure of the proteins and lipids. Integral membrane proteins may also partition between the two fracture faces in different manners (arrows indicate possible fracture face affinities). The membrane models depicted in B and C are both consistent with areas of the membrane being devoid of particles, without substantial increases in the particle density of the remaining areas.
conformational changes in the protein not required in Fig. 4C, and may be therefore less likely to occur. Vertical particle displacement could have important physiological consequences, because changing the microenvironment of a protein can alter its conformation. Similarly, the selective vertical displacement of some membrane proteins could affect their ability to interact with their neighbors. Such vertical displacement of membrane proteins associated with thermotropic phase transitions of the lipids could be physiologically as significant as the more frequently discussed lateral redistribution of the membrane proteins.
56,508-517.
11. Fork, D. C. & Murata, N. (1977) Plant Cell Physiol., special issue, pp. 427-436. 12. Overath, P. & Trauble, H. (1973) Biochemistry 12, 26252634. 13. Trauble, H. & Overath, P. (1973) Biochim. Biophys. Acta 307, 491-512. 14. Linden, C. D., Wright, K. L., McConnell, H. M. & Fox, C. F. (1973) Proc. Natl. Acad. Sci. USA 70,2271-2275. 15. James, R. & Branton, D. (1973) Biochim. Biophys. Acta 323, 378-390. 16. Verkleij, A. J., Ververgaert, P. H. J., Van Deenen, L. L. M. & Elbers, P. F. (1972). Biochim. Biophys. Acta 288,326-332. 17. Wunderlich, F., Ronai, A., Speth, V., Seelig, J. & Blume, A. (1975) Biochemistry 14, 3730-3735. 18. Verwer, W., Ververgaert, P. H. J. T., Leunissen-Bijvelt, J. & Verkleij, A. J. (1978) Biochim. Biophys. Acta 504, 231-234. 19. Murata, N. & Fork, D. C. (1976) Biochim. Biophys. Acta 461, 365-378. 20. Nolan, W. G. & Smillie, R. M. (1976) Biochim. Biophys. Acta 440, 461-475. 21. Hochli, M. & Hackenbrock, C. R. (1977) J. Cell Biol. 72, 278291. 22. Murata, N. & Fork, D. (1978) Carnegie Inst. Washington Yearb.
77,289-291. 23. Raison, J., Berry, J., Armond, P. & Pike, C. (1980) in Adaptations of Plants to Water and High Temperature Stress, eds. Kramer, P. & Turner, N. C. (Wiley-Interscience, New York), in press. 24. Kratz, W. & Meyers, J. (1955) Am. J. Bot. 42, 282-287. 25. Miller, K. R. & Staehelin, L. A. (1973) Protoplasma 77,55-78. 26. Branton, D., Bullivant, D., Giluta, N. B., Karnovsky, M. J., Moor, H., Muhlethaler, L., Northcote, D. H., Packer, L., Satir, B., Satir, P., Speth, V., Staehelin, L. A., Steere, R. L. & Weinstein, R. S.
(1975) Science 190,54-56. 27. Duppel, W. & Dahl, G. (1977) Biochim. Biophys. Acta 426, 408-417. 28. Borochov, H. & Shinitzky, M. (1976) Proc. Natl. Acad. Sci. USA
73,4526-4530. 29. Jacobs, R. E., Hudson, B. & Anderson, H. C. (1975) Proc. Natl. Acad. Sci. USA 72,3993-3997. 30. Watts, A., Marsh, D. & Knowles, P. F. (1978) Biochemistry 17, 1792-1801.
31. Niedermeyer, W., Parish, G. R. & Moor, H. (1976) Cytobiologie 13,364-379.
Cell Biology
Lateral and vertical displacement of integral membrane proteins during lipid phase transition in Anacystis nidulans* (plasma membrane/thylakoid membrane/reversible particle migration/freeze-fracture)
PAUL A. ARMONDt AND L. ANDREW STAEHELINt tDepartment of Plant Biology, Carnegie Institution of Washington, Stanford, California 94305; and tDepartment of Molecular, Cellular, and Developmental Biology, University of Colorado, Boulder, Colorado 80309
Communicated by Winslow R. Briggs, January 26, 1979
ABSTRACT Alterations in membrane structure as a result of lipid phase transitions have been studied in Anacystis nidulans, a blue-green alga. Cells grown at 38'C were subjected to temperature transitions of 380C - 21'C and 380C - 100C, previously shown to produce substantial changes in photosynthetic activities, and examined by freeze-fracture electron microscopy. As a result of these treatments, large particle-free regions appeared on the fracture faces of both the plasma and thylakoid membranes. Particle density measurements suggest that the displacement of the integral membrane protein complexes occurs in both lateral and vertical directions. Returning the cells to 380C resulted in the restoration of normal membrane morphology, indicating that the proteins were not lost from the membrane. Such displacement of the integral membrane protein complexes could contribute significantly to the temperature-dependent alterations in the functional activity of membrane-bound enzymatic complexes.
The temperature dependency of the physical state of membranes has been extensively studied in model systems (1-9), intact cells (10-19), and isolated organellar and membrane fractions (10, 19-21). In these studies, it has been shown that lipid species and fatty acid composition have a direct effect upon the phase transition temperature of the membrane. It has also been demonstrated that alterations in the activation energy of the membrane-bound enzymatic systems are correlated with the membrane phase transitions (10, 11, 14, 19-21). Although the protein constituents of the membrane can affect the fluidity of membranes, they apparently play no role in determining the phase transition temperature (7). However, correlations between enzymatic function and the physical state of the membrane stress the fact that protein-lipid interactions are vital for the proper physiological functioning of the membrane. Previous studies of the blue-green alga Anacystis nidulans have shown that a correlation exists between the physical state of the lipids and such parameters as chlorophyll a fluorescence, P700 reduction, and photosynthetic oxygen evolution, parameters that involve integral membrane protein complexes (10). It was demonstrated that for cells grown at 380C, a change in slope in Arrhenius plots of such membrane functions occurred at approximately 220C to 240C. More recently, a second slope change has been detected at approximately 100C to 12'C (22). Linden et al. (14) have interpreted similar observations to indicate the onset (slope change at the higher temperature) and completion (slope change at the lower temperature) of the lateral phase separation of membrane lipids. Recent electron spin resonance experiments (23) have indicated that, in Anacystis lipids from cells grown at 380C, the onset of the lateral phase separation occurs at a temperature somewhat higher than 240C, but the experiments are in agreement with the compleThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
tion of the phase transition at 100C. To learn more about the structural basis of the functional changes, we qualitatively and quantitatively examined the membrane structural organization of A. nidulans by freeze-fracture electron microscopy at temperatures at which the alterations in physiological activities occur.
MATERIALS AND METHODS A. nidulans was grown at 380C in Kratz and Meyers' C medium (24) bubbled with air enriched to 5% CO2. Samples were either maintained under the growth conditions at 38°C or transferred to temperature baths at 21°C or 10°C (210C and 10°C are slightly below the temperatures at which the discontinuities in the slope of the Arrhenius plot for certain enzymatic activities are observed). The overall cooling rate was approximately 1°C/min. An additional sample was transferred to the 21°C bath, maintained at that temperature for 20 min, and returned to the 38°C bath. After the cell suspensions had equilibrated at their respective final temperatures for 15 min, sufficient glutaraldehyde for a final concentration of 2% (wt/vol) was added. The cell suspensions were gently stirred for 5 min, removed from the baths, and centrifuged at 10,000 X g for 10 min. The remaining operations were carried out at 200C. The pelleted cells were resuspended in the growth medium, and glycerol was added slowly over a 1-hr period to a final concentration of 35% (vol/vol). The cells were centrifuged at 10,000 X g for 15 min, and the pelleted cells were prepared for freeze-fracture as described (25). Particle density measurements for each fracture face were made on six to eight micrographs enlarged to X125,000. Measurements were made on the flattest possible membrane areas; sharply curving membrane regions were avoided. Percentage of particle-free membrane areas was determined on the same micrographs used for the particle density measurements. The myelinic preparation was made by allowing a 10 mg/ml suspension of egg leci*thin (obtained from Nutritional Biochemical Corp., Cleveland, OH) in a 20% glycerol/water solution to swell for 1 hr at room temperature. Fracture-face nomenclature is according to Branton et al. (26). RESULTS The freeze-fracture preparations of control (38°C) A. nidulans membranes (Fig. 1) exhibited an even distribution of particles on all fracture faces. Neither the plasma membrane (Fig. 1A) nor the thylakoid membrane (Fig. 1B) exhibited any extensive areas devoid of particles. In contrast, membranes of cells that had undergone a 38°C - 21°C transition showed large partiAbbreviations: PF, protoplasmic fracture face; EF, exoplasmic fracture face. * Carnegie Institution of Washington Department of Plant Biology publication no. 640. 1901
1902
Proc. Natl. Acad. Sci. USA 76 (1979)
Cell Biology: Armond and Staehelin
FIG. 1. Freeze-fractured control cells of A. nidulans grown and fixed with glutaraldehyde at 380C. Neither the plasma membrane (A) nor the thylakoid membrane (B) exhibits any extensive particle-free regions on either the protoplasmic fracture face (PF) or the exoplasmic fracture face (EF). (A, X75,000; B, X90,000.) Bars equal 0.5 ,m.
cle-free regions on the thylakoid membranes (Fig. 2A) as well as on the plasma membrane (Fig. 2B). The extent of such particle-free regions was increased in the membranes of cells that had
experienced
the 380C
100C transition
(Fig.
2 C and
D).
In addition to the extensive particle-free areas, apparent physical breaks in both the plasma membrane and thylakoid membranes were observed in the 380C 100C transition
sample (arrows, Fig. 2D). The membranes of the cell samples that had undergone the 380C 210C 380C treatment were visually indistinguishable from membranes of the control cells shown in Fig. 1. Particle density measurements were made for the PF and EF fracture faces of the plasma and thylakoid membranes of all cell samples (Table 1). The measured particle densities --
--
consider only those membrane areas containing particles; areas devoid of particles or immediately bordering such areas were excluded from these measurements. As shown in Table 1, the appearance of extensive particle-free regions in the membranes of the 380C - 21'C and 380C -- 100C treated cells did not result in large increases in the particle density of the remaining areas. Measurements of the extent of the particle-free areas (Table 1) indicate that the increases in particle density observed are insufficient to account for the particle-free regions by only a lateral movement of the membrane particles. Measurements have been made of the diameter of the particles on the PF face of the thylakoid membrane, the fracture face that exhibited the most extensive particle-free regions at both treatment temperatures. In order to maintain particle density after the 380C
Proc. Nati. Acad. Sci. USA 76 (1979)
Cell Biology: Armond and Staehelin
1903
FIG. 2. Fracture faces of cell membranes subjected to a 380C - 210C (A, B) or a 380C - 100C (C, D) temperature transition. After treatment
at 210C, large particle-free regions are visible. The PF face of the thylakoid membranes (A) shows a clear delineation between particle-free and particle-containing regions. This delineation is not as evident in the PF fracture face of the plasma membrane (B). Note the verrucose texture of the particle-free regions. When cells were subjected to a 380C 100C treatment (C, D), these regions became much more extensive, with large particle-free areas visible on the EF face of the plasma membrane (C) and PF faces of both the plasma (left) and thylakoid (right) membranes (D). In addition, the 380C - 100C treatment produced both verrucose and faint reticulate textured regions in the particle-free areas (see also Fig. 3). Arrows indicate apparent physical breaks in the membranes (D). (X70,000.) Bar equals 0.5 ,m. -
- 21'C transition, the average particle diameter would have to increase from 94 A (measured) to 162 A (calculated on the basis of 67% particle-free regions). An increase of only 15% (average diameter equal to 108 A) was observed in the 380C -- 21'C sample. When the cells were restored to the growth temperature (the 380C -- 21'C -- 380C sample), the particle-free regions were no longer observed, and the overall particle density of the membrane fracture faces returned to that
of the control cells. Closer examination of the micrographs of membrane fracture faces devoid of particles (presumably less fluid lipid regions) reveals that these faces exhibit distinct surface textures. Two basic surface textures can be distinguished (Fig. 3A), a verrucose surface (Sl) and a faint reticulate surface (S2). Only the verru-
cose surface is observed in the 380C - 21'C samples, while both types of textured regions can be recognized in the 380C 100C specimens. In myelinic preparations of pure egg lecithin (Fig. 3d), very similar textures can also be seen on the fracture faces of the bilayer membranes. Both in the particlefree A. nidulans membrane regions and in the egg lecithin bilayers these differently textured areas can exist side by
side. DISCUSSION of the and plasma thylakoid membranes of A. nidulans Analysis by freeze-fracture electron microscopy reveals that extensive alterations in membrane structural organization are associated with the phase transition of the lipid constituents of the mem-
Proc. Nati. Acad. Sci. USA 76 (1979)
Cell Biology: Armond and Staehelin
1904
Table 1. Density of particles on fracture faces and percentage of particle-free areas of the plasma and thylakoid membranes of A. nidulans with various temperature treatments Thylakoid membrane Plasma membrane % particle% particleTemperature Particles/hm2 Particles/,um2 PF free regions EF free regions EF PF treatment
380C (control) 380C - 211C 380C - 100C 38oC 21'C -381C -
1572 i 98 1484 ± 101 1392 I 132 1632 ± 198
3434 ± 119 3853 ± 149 3867 i 132 3277 I 251
None 21 43 None
703 ± 82
3703 ± 190
759 1 156 755 106 723 101
3804 i 169 4278 235 3324 257
None 67 77 None
Density figures are mean ± SD.
branes. For cells grown at 380C, treatment at 21'C results in particle-free areas on the PF and EF fracture faces of both plasma and thylakoid membranes. At 10'C, these particle-free areas become more extensive, and breaks in the membranes appear. An examination of the particle-free areas reveals verrucose and reticulate textured regions on both plasma and thylakoid membrane fracture faces. It is now well documented that bilayer membranes made of different amphipathic lipids can exhibit different kinds of fracture face textures after being cooled below their phase transition temperature (4). Thus, the verrucose and reticulate textured particle-free areas of the A. nidulans membranes may correspond to regions containing different kinds of lipids, or, as in the case of the lecithin bilayers, a possible different ordering of the lipids. Both lateral (15, 16, 18, 21, 27) and vertical (17, 28) displacement of membrane particles have been proposed to explain the formation of particle-free areas in cooled membranes.
FIG. 3. Freeze-fractured A. nidulans cell after a 38°C 10°C temperature transition (A), and a freeze-fractured egg lecithin -
membrane (B). Surface characteristics of the particle-free regions in the cell membranes (A) could be generally categorized as verrucose
(S2). Such surface characteristics are also visible in bilayers (B), indicating that the surface characteristics of the particle-free regions are a function of the lipid constituents of the membrane. (A, X80,000; B, X56,000.) Bars equal 0.5 Am.
(Si) or reticulate pure egg lecithin
These possibilities may be distinguished by examination of the particle densities in the particle-containing areas. If the redistribution of particles involves only lateral movement, a predictable increase in the particle density of particle-containing regions should occur. For the A. nidulans membranes (Table 1), such increases are not evident on the EF faces of the plasma and thylakoid membranes, and are only marginally observed for the remaining fracture faces. The increases in particle density that are observed are insufficient to account for the extent of the observed particle-free regions if lateral displacement is the only mode of particle redistribution. Particle size changes also do not appear to be significant. It would appear that the thermotropic particle redistribution in A. nidulans involves both lateral and vertical displacement of membrane particles. While the pure lateral displacement of particles can easily be explained by their exclusion from crystalline lipid domains, three possible mechanisms may be involved in the apparent vertical movement of particles: (i) The movement of the particles may only be an apparent one, caused by increases in membrane thickness induced by the phase transition (29, 30). Alterations in membrane thickness have been shown to alter the density of particles observed on membrane fracture faces (31). (ii) The movement of the integral membrane protein complexes in a vertical direction may be a direct consequence of the physical events associated with the lipid state change. Alterations in microviscosity (which occur during lipid phase transition) have been correlated with an apparent vertical displacement of proteins in erythrocyte membranes (28). (iii) Certain integral membrane protein complexes could be partly squeezed out of the central plane of the membrane (and therefore not seen as freeze-fracture particles) by the increased lateral pressure on the transmembranous proteins exerted by the expanding regions of lipids in the gel phase. The observed loss of particles from the fracture faces may involve all of these possibilities. As shown in Fig. 4, the loss of particles observed in the transition from a fluid (Fig. 4A) to a mixed-phase state (Fig. 4 B and C) does not have to be the result of a complete loss of the protein complexes from the membranes. Exclusion of all the particles from the fracture plane of less fluid regions of the membrane (Fig. 4B) or maintenance of the number of particles at the fracture plane of fluid regions of the membrane (Fig. 4C) by the exclusion of some particles would be sufficient to explain the observed effect of the fluidphase to mixed-phase transition on membrane structure. Because both of these possible mechanisms allow for a rapid restoration of normal particle densities upon rewarming of the membranes, as was observed for A. nidulans (see Table 1), it is not possible to distinguish between these alternatives at the present time. It should be noted, however, that the alternative presented in Fig. 4B requires that portions of some membrane complexes previously exposed to water are in a hydrophobic region of the membrane after the phase transition. This alternative would require a substantial input of energy or large
Cell Biology: Armond and Staehelin
A
B
f
f
f
f
f
I
npUS0U009URII
i
Proc. Nati. Acad. Sci. USA 76 (1979) 1905 The authors thank M. DeWit and T. Giddings for their technical assistance, Or. 1D. Fork for providing the cell cultures used in this study, and Drs. J. Berry and J. Raison for their helpful suggestions. P.A.A. was supported by a Carnegie Institution of Washington Postdoctoral Fellowship. This work was supported in part by National Institute of General Medical Sciences Grant GM 18639 to L.A.S. 1. Untracht, S. H. & Shipley, G. G. (1977) J. Biol. Chem. 252, 4449-4457. 2. Phillips, M. C., Ladbrooke, B. D. & Chapman, D. (1970) Biochim. Biophys. Acta 196,35-44. 3. Hui, S. W. & Parsons, D. F. (1975) Science 190, 383-384. 4. Gulik-Krzywicki, T. (1975) Biochim. Biophys. Acta 415, 1-28. 5. Lee, A. G. (1977) Biochim. Biophys. Acta 472, 237-281. 6. Lee, A. G. (1977) Biochim. Biophys. Acta 472, 285-344. 7. Papahadjopoulos, D., Moscarello, M., Eylar, E. H. & Isac, T. (1975) Biochim. Biophys. Acta 401, 317-335. 8. Hui, S. W., Parsons, D. F. & Cowden, M. (1974) Proc. Natl. Acad. Sci. USA 71,5068-5072. 9. Shipley, G. G., Green, J. P. & Nichols, B. W. (1973) Biochim. Biophys. Acta 311, 531-544. 10. Murata, N., Troughton, J. H. & Fork, D. C. (1975) Plant Physiol.
t C
I
FIG. 4. Model for the vertical displacement of membrane particles in A. nidulans. The loss of particles from the fracture plane in the transition from a fluid-phase (A) to a mixed-phase state (B and C) may involve different processes, depending upon whether the proteins remain in the less fluid portion of the membrane (B) or laterally migrate to the more fluid region of the membrane (C). If the proteins remain in their original position, integral membrane proteins that are not transmembranous (B, upper leaflet) could become undetectable at the fracture plane because of the change in bilayer thickness induced by the change of the physical state of the lipids. For such nontransmembranous proteins, approximately the same degree of surface exposure could be maintained. Transmembranous proteins also would require some increase in surface exposure to account for nondetectability at the fracture plane (B, lower leaflet). If the proteins are displaced to a more fluid region of the membrane (C), some proteins might be excluded from the fracture plane by the combined lateral pressure of the proteins and lipids. Integral membrane proteins may also partition between the two fracture faces in different manners (arrows indicate possible fracture face affinities). The membrane models depicted in B and C are both consistent with areas of the membrane being devoid of particles, without substantial increases in the particle density of the remaining areas.
conformational changes in the protein not required in Fig. 4C, and may be therefore less likely to occur. Vertical particle displacement could have important physiological consequences, because changing the microenvironment of a protein can alter its conformation. Similarly, the selective vertical displacement of some membrane proteins could affect their ability to interact with their neighbors. Such vertical displacement of membrane proteins associated with thermotropic phase transitions of the lipids could be physiologically as significant as the more frequently discussed lateral redistribution of the membrane proteins.
56,508-517.
11. Fork, D. C. & Murata, N. (1977) Plant Cell Physiol., special issue, pp. 427-436. 12. Overath, P. & Trauble, H. (1973) Biochemistry 12, 26252634. 13. Trauble, H. & Overath, P. (1973) Biochim. Biophys. Acta 307, 491-512. 14. Linden, C. D., Wright, K. L., McConnell, H. M. & Fox, C. F. (1973) Proc. Natl. Acad. Sci. USA 70,2271-2275. 15. James, R. & Branton, D. (1973) Biochim. Biophys. Acta 323, 378-390. 16. Verkleij, A. J., Ververgaert, P. H. J., Van Deenen, L. L. M. & Elbers, P. F. (1972). Biochim. Biophys. Acta 288,326-332. 17. Wunderlich, F., Ronai, A., Speth, V., Seelig, J. & Blume, A. (1975) Biochemistry 14, 3730-3735. 18. Verwer, W., Ververgaert, P. H. J. T., Leunissen-Bijvelt, J. & Verkleij, A. J. (1978) Biochim. Biophys. Acta 504, 231-234. 19. Murata, N. & Fork, D. C. (1976) Biochim. Biophys. Acta 461, 365-378. 20. Nolan, W. G. & Smillie, R. M. (1976) Biochim. Biophys. Acta 440, 461-475. 21. Hochli, M. & Hackenbrock, C. R. (1977) J. Cell Biol. 72, 278291. 22. Murata, N. & Fork, D. (1978) Carnegie Inst. Washington Yearb.
77,289-291. 23. Raison, J., Berry, J., Armond, P. & Pike, C. (1980) in Adaptations of Plants to Water and High Temperature Stress, eds. Kramer, P. & Turner, N. C. (Wiley-Interscience, New York), in press. 24. Kratz, W. & Meyers, J. (1955) Am. J. Bot. 42, 282-287. 25. Miller, K. R. & Staehelin, L. A. (1973) Protoplasma 77,55-78. 26. Branton, D., Bullivant, D., Giluta, N. B., Karnovsky, M. J., Moor, H., Muhlethaler, L., Northcote, D. H., Packer, L., Satir, B., Satir, P., Speth, V., Staehelin, L. A., Steere, R. L. & Weinstein, R. S.
(1975) Science 190,54-56. 27. Duppel, W. & Dahl, G. (1977) Biochim. Biophys. Acta 426, 408-417. 28. Borochov, H. & Shinitzky, M. (1976) Proc. Natl. Acad. Sci. USA
73,4526-4530. 29. Jacobs, R. E., Hudson, B. & Anderson, H. C. (1975) Proc. Natl. Acad. Sci. USA 72,3993-3997. 30. Watts, A., Marsh, D. & Knowles, P. F. (1978) Biochemistry 17, 1792-1801.
31. Niedermeyer, W., Parish, G. R. & Moor, H. (1976) Cytobiologie 13,364-379.
