Plant Physiol. (1980) 66, 238-241 0032-0889/80/66/0238/04/$00.50/0

Membrane Phospholipid Phase Separations in Plants Adapted to or Acclimated to Different Thermal Regimes1 Received for publication July 16, 1979 and in revised form March 10, 1980 CARL S. PIKE2 AND JOSEPH A. BERRY4 Department ofPlant Biology, Carnegie Institution of Washington, 290 Panama Street, Stanford, California 94305

ABSTRACT The importance of membrane lipids in determining chilling sen- sitivity has recently been questioned (2, 5, 6, 24). No single The phase separation temperatures of total leaf phospholipids from hypothesis is presently able to accommodate these observations, warm and cool climate plants were determined in order to explore the and it seems likely that many factors may distinguish plants that relationship of lipid physical properties to a species' thermal habitat. The have evolved in hot and cool environments. However, these separation temperatures were determined by measuring the fluorescence observations do not necessarily contradict the view that membrane intensity and fluorescence polarization of liposomes labeled with the lipids are involved as one component in this evolutionary process. polyene fatty acid probe trans-parinaric acid. To focus on a single climatic Our approach has been to investigate the thermotropic behavior region, Mojave Desert dicots (chiefly ephemeral annuals) were examined, of lipids from native species which have known ecological pref- with plants grown under identical conditions whenever possible. Winter erence for warm or cool growth conditions. We assume that the active species showed lower phase separation temperatures than the sum- thermal responses of these plants would reflect the constraints of mer active species. A group of warm climate annual grasses showed natural selection operating in their native habitats. If the lipid separuition temperatures distinctly higher than those of a group of cool phase separation temperature is important, natural selection will climate grasses, all grown from seed under the same conditions. Growth at have operated to favor plants with membrane lipid properties that low temperature seems correlated with (and may require) a low phase are appropriate to the thermal regime experienced during its separation temperature. Winter active ephemerals appear genetically pro- growing season. grammed to synthesize a mixture of phospholipids which will not phase The plants selected for this study are, with a few exceptions, separate in the usual growth conditions. When the lipids of desert peren- from the native flora of the Mojave and Sonoran Deserts of North nials were examined in cool and warm seasons, there was a pronounced America. Ephemeral species that are known to grow principally seasonal shift in the phase separation temperature, implying environmental during the summer or winter seasons and some perennials that influences on lipid physical properties. The relationship of these results to are active throughout the year were examined. For the most part high and low temperature tolerance is discussed. the ephemeral species were grown at a common growth tempera- ture in controlled growth facilities. However, the perennial species were sampled during midwinter and in early summer from natural plants growing in Death Valley, California. The mean daily maximum and minimum temperatures in Death Valley for Janu- ary and July are 18 C/3 C and 45 C/32 C, respectively; these temperatures indicate the different thermal regimes experienced The physical state of membrane lipids may bear a significant by plants growing in the winter or summer. relationship to the lower temperature limit for a species' growth We used the fluorescent polyene fatty acid, trans-parinaric acid or survival. In various crop plants there is a correlation between (22, 23), as a probe for determining the phase boundaries of the occurrence of a lipid phase separation at about 10 C, an abrupt liposomes ofmembrane phospholipids extracted from these plants. increase in the activation energy of various membrane-bound Unfortunately, it was not possible to use trans-parinaric acid with reactions at this temperature, and the occurrence of metabolic native membranes of leaf cells or with total lipid preparations dysfunction at temperatures below this point (7, 15, 16). because pigments of the leaf quench the probe's fluorescence. In A change in the temperature dependence of spin label motion interpreting these studies, we assume that differences in the phys- in mung bean chloroplasts and mitochondria, an indication of a ical properties of isolated phospholipids should provide a good lipid phase separation, corresponded with the lower temperature relative index of differences in the properties of plant membranes limit of etiolated seedling growth (18). In a group of Passif7ora in vivo. Inasmuch as the phospholipids as a group contain a lower species, a high phase separation temperature was associated with proportion ofunsaturated fatty acids than the galacto- or sulfolipid greater chilling sensitivity; a lower separation temperature, with components of the membrane, it seems likely that the phospholip- less sensitivity (11). Habitat preference also correlated with the ids will have the highest phase separation temperature of the point of loss of membrane integrity assayed by ion leakage (10). membrane lipid mixture. In another study we found that the total membrane polar lipids (examined with spin label probes) and l Supported in part by the Science and Education Administration of the phospholipids (examined with trans-parinaric acid) from the same United States Department of Agriculture Grant 5901-0410-8-0128 from lipid preparations had very similar phase separation temperatures the Competitive Research Grants Office. CIW/DPB Publication No. 674. (17). 2 Permanent address: Department of Biology, Franklin and Marshall College, Lancaster, Pennsylvania. MATERIALS AND METHODS 'Supported in part by a National Science Foundation Science Faculty Professional Development Award and by a grant from the Mellon Foun- For plants grown from seed in the laboratory, controlled envi- dation fund of Franklin and Marshall College. ronment chambers were used to grow warm and cool climate 4To whom correspondence should be addressed. plants under the same conditions. For plants collected in Death 238 Plant Physiol. Vol. 66, 1980 PHASE CHANGES AND TEMPERATURE PREFERENCE 239 Valley, the mean daily maximum and minimum temperatures for Temperature, °C the month of collection are indicated. About 5 g of leaf blade was 50 40 30 20 10 0 -10 extracted for 5 min in 50 ml of boiling methanol containing I mg II II II I butylated hydroxytoluene. Then 100 ml ofchloroform was added, 0 and the tissue was ground in a VirTis The extract 0 homogenizer. 0 0 was filtered through Miracloth and partitioned four times with 0o 0.55 M KCI, once with water, and once with 60 mm KCI. After c 0 drying over anhydrous Na2SO4, the extract was concentrated in a rotary evaporator. The total lipid extract was applied to a Bio-Sil II) A (Bio-Rad Laboratories) column (<10 mg lipid/g Bio-Sil) and eluted with chloroform, acetone, and methanol (7 ml/g Bio-Sil). The phospholipid-rich methanol fraction was used in fluorescence studies. A- portion of this fraction was dried onto the walls of a glass vial under N2, held in a vacuum desiccator, and gently sonicated in 100 mm Tris-HCl (pH 7.2) containing 5 mm EDTA. The samples for fluorescence measurements contained 400 ,tg .- lipid and 0.7 ug trans-parinaric acid in 3 ml buffer containing 25 .-0 0 0 *S or 33% (v/v) ethylene glycol (a concentration without effect on -J S 0 . phase separation temperatures). Fluorescence was monitored in a 0 Perkin-Elmer MPF-3L spectrofluorometer. The excitation and emission monochromators were set at 320 and 420 nm, respec- 31 32 33 34 35 36 37 38 39 tively. The excitation beam was passed through a polarizing prism (Karl Lambrecht Corp.); the emitted light was passed through a I/T x 104 OK-' plastic polarizer (Edmund Scientific Co.), and a 350-nm cut-off FIG. 2. Trans-parinaric acid fluorescence polarization of corn and bar- filter. Temperatures in the sample cuvette (contained in a ther- ley phospholipid vesicles. Fluorescence emission was measured with the moregulated holder) were measured with a copper-constantan polarizer parallel (Ii) and perpendicular (IJ) to the orientation of the thermocouple. All temperature scans were made in the ascending excitation polarizer. The polarization ratio is l1/I±. (0), Zea mays; (0), direction. Fluorescence intensity was measured with the emission Hordeum vulgare. polarizer parallel (I 11 ) and perpendicular (1,) to the orientation of the excitation polarizer. Plots were made of log I 11 as a function of reciprocal absolute temperature and of the polarization ratio suggest the occurrence of a change in lipid fluidity. The low (I 11 /I,) as a function of temperature (23). Changes in the temper- polarization ratio above 10 C indicates a fluid probe environment ature dependence (slope) of these parameters were used to deter- (23). We interpret the slope change at 10 C as indicating the first mine the phase separation temperature. For a given sample the appearance of detectable solid as the temperature is lowered, since two methods usually agreed within 1 C of one another. There was trans-parinaric acid is sensitive to a few per cent solid (23). This a similar reproducibility for replicate experiments. interpretation is supported by experiments with model systems (23) and by experiments in which the phase separation process of Anacystis nidulans membranes was examined using freeze-fracture RESULTS electron microscopy and the properties of extracted phospholipids Phase Separation Processes. The fluorescence intensity (Fig. 1) were studied with trans-parinaric acid (17). The leveling off of the and polarization ratio (Fig. 2) for trans-parinaric acid in maize polarization ratio of the maize phospholipids at about 2.0 (seen at phospholipids illustrate our results with this probe. The changes about -5 C in Fig. 2) should not be interpreted as indicating the in slope occur at about 9 and 10 C, respectively, in good agreement completion of solidification; in model systems such leveling off with electron spin resonance determinations (15). These changes occurred at about 50%1o solid because the probe preferentially partitions into solid phase lipids (23). The intensity (Fig. 1) and polarization ratio (Fig. 2) plots for 2.2 phospholipids from barley (a typical chilling resistant plant) in- dicate rather slight slope changes at about -6 C. The polarization .2 2.0 0 ratio suggests that even below that point the barley lipids are much more fluid than the corn lipids. .8 Phase Separation Temperatures of Warm and Cool Climate Plants. Dicots, chiefly annuals, from several families were selected N16. to represent the winter and summer Mojave Desert ephemeral floras (9, 20). The plants were either collected in Death Valley or *e..~00000o000o0 grown in the laboratory from seeds collected in the Mojave Desert. Q.1.4- The winter active species showed phospholipid phase separations at 3 C or lower (Table I). 1.2 -20 -10 0 10 20 30 40 50 For most of the summer active species, separation tqmperatures of 10 C or were Temperature, higher observed. Tidestromia oblongifolia (a winter 0C dormant perennial) showed a separation temperature of 12 C FIG. 1. Trans-parinaric acid fluorescence intensity of corn and barley when grown under optimal conditions (45 C/32 C [41); for phos- phospholipid vesicles. The phopholipids were dispersed by gentle sonica- pholipids from leaves collected in December (just before they are tion of 100 mM Tris-HCI (pH 7.2), containing 5 mM Na2EDTA and 25 or shed), the separation temperature was 7 C. These results with 33% (v/v) ethylene glycol. A suspension containing 400 ug lipid/3 ml Tidestromia suggest some change in lipid physical properties as a buffer was labeled with 0.7 ILg trans-parinaric acid. Polarized light at 320 function of growth temperature; this acclimation phenomenon nm was used to excite the sample and fluorescence was measured at 420 will be examined below. The low separation temperature for nm, with the excitation polarizer parallel to the emission polarizer. (0), Atriplex elegans elegans, a seeming exception, is also discussed Zea mayr, (0), Hordeum vulgare. below. 240 PIKE AND BERRY Plant Physiol. Vol. 66, 1980

Table I. Phospholipid Phase Separation Temperaturesfor Desert Dicots Table 1II. Phospholipid Phase Separation Temperaturesfor Desert For laboratory-grown plants (L), the growth chamber day and night Perennials temperatures are given (16-h photoperiod). For plants collected in Death Separation Temperature' Valley (DV), the mean daily maximum and minimum temperatures for Species the month of collection are given. The separation temperatures were Januaryh Mayc determined as the average value from trans-parinaric acid fluorescence C intensity and polarization ratio plots of data from two samples. Atriplex hymenelytra -15d 0 Species Growth Conditions Separation Heliotropium curassavicum -2 6 Temperature Larrea divaricata -8 9 C Psathyrotes ramosissima 1 7 Cool climate a Determined as in Table I. Atriplex elegansfasiculata L 28 C/2 1 C 2 b Mean daily maximum and minimum temperatures: 18 C/3 C. Boerhaavia annulata DV 19 C/3 C 2 'Mean daily maximum and minimum temperatures: 39 C/22 C. Camissonia claviformis DV 22 C/6 C -3 d Sample analyzed in fluorometer contained 50%o (v/v) ethylene glycol. Cryptantha angustifolia L 28 C/2 1 C 2 Eriogonum inflatum DV 22 C/6 C -4 Lepidium lasiocarpum L 28 C/21 C -I DISCUSSION Perityle emoryi L 28 C/21 C 3 On the basis of data in Figures I and 2, and considering the Warm climate work with model systems (23), we have suggested that the phase Atriplex elegans elegans L 28 C/2 1 C -1 separation observed at about 10 C in typical chilling sensitive Boerhaavia coccinea L 28 C/21 C 12 plants represents the beginning of the appearance of ordered (gel Mollugo verticillata L 28 C/21 C 17 phase) lipid (14). This solidification is not completed until well Pectispapposa L 28 C/21 C 13 below 0 C. Our proposal differs from the assignment of slope Portulaca oleracea L 28 C/21 C 11 changes based on spin label studies (15, 18). Tidestromia oblongifolia L 45 C/32 C 12 The low phase separation temperatures of the winter active DV 19 C/3 C 7 ephemerals (Table I) and cool season grasses (Table II) suggest that the membranes of these plants would seldom contain any solid lipid in the typical thermal regime of the Death Valley area in winter where the mean daily minimum temperature in the Table II. Phospholipid Phase Separation Temperaturesfor Annual coldest month is 3 C. Wherever possible these plants have been Grasses grown from seed at the same growth temperature so that we could All plants were grown at constant 27 C and a 16-h photoperiod. The focus upon genetically based rather than environmentally induced separation temperatures were determined as in Table I. differences in lipid properties. Winter ephemeral species collected Species Separation Temperature from natural populations in Death Valley were, however, similar to the other winter ephemerals grown under laboratory conditions. C The separation temperatures for the summer active species are Cool climate generally higher than 7 C and range to 17 C (Tables I and II). A venafatua -9 Minimum temperatures in the deserts in summer would seldom A vena sativa -11 be low enough to cause the membrane lipids of these plants to Bromus rigidus -10 phase separate. Two exceptions to the general pattern are ob- Hordeum vulgare -6 served. Atriplex elegans elegans (Table I) and Chloris virgata Warm climate (Table II) have significantly lower phase separation temperatures Chloris virgata 4 than the other summer active species. These exceptions may be Digitaria sanguinalis 8 accommodated if it is postulated that it may only be necessary Panicum texanum 7 that the phase separation temperature of the lipids be as low or Zea mays 9 lower than the normal minimum temperatures during the plant's growing season. It is also relevant that although these two species are characterized as part of the summer flora, A. elegans elegans Table II shows the separation temperature for phospholipids may grow during the winter in the Mojave Desert and C. virgata from grasses considered warm or cool temperature active. Both may flower by April in the Sonoran Desert (9). wild and cultivated species are included; some of the former are Perennial species active throughout the year experience the from the Sonoran Desert (9, 20). The plants were all grown from contrasting thermal regimes to which the respective summer and seed at 27 C. The warm season species showed considerably higher winter ephemeral species are genetically adapted. It is known that separation temperatures. Atriplex lentiformis (13), N. oleander (3), and Larrea divaricata (1, Phase Separation Temperatures in Death Vafley Perennials. 8) exhibit substantial seasonal (or growth temperature-induced) We have shown (14) that phospholipids from Nerium oleander changes in the temperature response of photosynthesis. These leaves developed at 20 C/ 15 C or 45 C/32 C had phase separation changes improve photosynthetic performance at the prevailing temperatures of -3 C and 7 C, respectively. To test the generality temperature and increase the capacity to tolerate high temperature of this observation on acclimation, phospholipids were extracted during the hot months. The acclimation of these perennials may from leaves collected from four Death Valley perennials in early in part be based upon environmentally induced alterations in lipid winter and late spring (Table III). When sampled in January, the properties, since substantial shifts in lipid phase separation tem- separation temperatures were all at C or below (cf. the winter peratures were observed between the summer and winter seasons ephemerals in Table I). In late spring, when the mean daily (Table III). temperature was about 20 C higher, all the plants exhibited an Growth temperature-induced changes in the fatty acid compo- increase in the phase separation temperature, ranging from 6 to sition ofthe lipids ofthe perennial A. lentiformis indirectly indicate 17 C. seasonal changes in lipid physical properties (12). The shifts in Plant Physiol. Vol. 66,1980 PHASE CHANGES AND TEMPERATURE PREFERENCE 241 lipid physical properties of the membranes lipids observed here category rather than the pathway. Many of the C4 plants were may also be caused by changes in fatty acid composition of the summer ephemerals; the thermal properties of their lipids may lipids, but lipid composition was not measured in this study. In limit their growth at low temperature. This observation may terms of the properties pertinent to plant functions, direct mea- explain the generalization that C4 plants do poorly at low temper- surements ofthe lipid physical properties are clearly more inform- ature (9). ative than extrapolations from lipid chemical composition data. Acknowledgments-We thank R. Simoni for the trans-parinaric acid and J. Ehler- Given the evergreen nature of the desert perennials sampled, we inger and M. Nobs for seeds. surmise that their mature leaves, like those of N. oleander, are capable of changing the chemical composition of the membranes (Raison, Pike, and Berry, in preparation). LITERATURE CITED Growth in membrane temperature-induced changes properties 1. ARMOND PA, U SCHREIBER, 0 BJORKMAN 1978 Photosynthetic acclimation to of bacteria have been noted. When the growth temperature is temperature in the desert shrub, Larrea divaricata. II. Light-harvesting effi- changed, the phospholipids of Escherichia coli exhibit complete ciency and electron transport. Plant Physiol 61: 411-415 viscosity compensation (21) (homeoviscous adaptation). There is 2. BAGNALL DJ, JA WOLFE 1978 Chilling sensitivity in plants. Do the activation energies of growth processes show an abrupt change at a critical temperature? considerable difference between the higher plant species in the J Exp Bot 29: 1231-1242 extent to which they modified the phase separation temperature 3. BJORKMAN 0. M BADGER, PA ARMOND 1978 Thermal acclimation of photosyn- (Table III). Given the approximately 20 C degree shift in habitat thesis: effect of growth temperature on photosynthetic characteristics and temperature between summer and winter it is evident that no components of the photosynthetic apparatus in Nerium oleander. Carnegie Inst Wash Year Book 77: 262-276 species compensated fully for the change in thermal regime. 4. BJORKMAN 0, B MAHALL, M NOBS, W WARD, F NICHOLSON, H MOONEY 1974 The elevated phase separation temperatures seen with either the An analysis of the temperature dependence of growth under controlled con- summer ephemeral species or with perennial species in the summer ditions. Carnegie Inst Wash Year Book 73: 757-767 could reflect the absence of low the 5. CRITCHLEY C, RM SMILLIE, BD PATTERSON 1978 Effect of temperature on simply temperatures during photoreductive activity of chloroplasts from passion fruit species of differing summer period; but the general correlation, particularly for the chilling sensitivity. Aust J Plant Physiol 5: 443-448 perennials, suggests that there may be a selective advantage to this 6. GRAHAM D, DG HOCKLEY, BD PATTERSON 1980 Temperature effects on PEP property. In a study of N. oleander leaves, we have shown that carboxylases from chilling-sensitive and chilling-resistant plants. In JM Lyons. to JK Raison, eds, Low Temperature Stress in Crop Plants: The Role of the adaptive changes causing the photosynthetic membranes be Membrane. Academic Press, New York. In press more stable to high temperatures are associated with a shift in 7. LYONS, JM 1973 Chilling injury in plants. Annu Rev Plant Physiol 24: 445-466 membrane lipid fluidity and an upward shift in the phase sepa- 8. MOONEY HA, 0 BJORKMAN, GJ COLLATZ 1978 Photosynthetic acclimation to ration temperature (1, 14, 17). temperature in the desert shrub, Larrea divaricata. I. Carbon dioxide exchange characteristics of intact leaves. Plant Physiol 61: 406-410 Generalizing from the data presented here, adaptation of 9. MULROY TW, PW RUNDEL 1977 Annual plants: Adaptations to desert environ- ephemeral plant species to growth at low temperature is accom- ments. BioScience 27: 109-114 panied by (and may require) a low phospholipid phase separation 10. PATTERSON BD, J KENRICK, JK RAISON 1978 Lipids of chilling-sensitive and temperature, whereas adaptation to growth at high temperature is -resistant Passiflora species: Fatty acid composition and temperature depend- ence of spin label motion. Phytochemistry 17: 1089-1092 usually associated with a higher phase separation temperature. It 11. PATTERSON DB, T MURATA, D GRAHAM 1976 Electrolyte leakage induced by appears that plants normally growing in a certain environment chilling in Passiflora species tolerant to different climates. Aust J Plant Physiol synthesize a mixture of phospholipids which will not phase sepa- 3: 435-442 12. PEARCY RW 1978 Effect of growth temperature on the fatty acid composition of rate in the usual thermal regime. For ephemeral species the lipid the leaf lipids in Atriplex lentuformis (Torr.) Wats. Plant Physiol 61: 484-486 properties appear to be in large part genetically determined, 13. PEARCY RW, JA BERRY, DC FORK 1977 Effects of growth temperature on the whereas for perennial plants adapted to a changing environment, thermal stability of the photosynthetic apparatus ofAtriplex lentiformis (Torr.) the properties of their lipids appear to respond to environmental Wats. Plant Physiol 59: 873-878 14. PIKE CS, JA BERRY, JK RAISON 1980 Fluorescence polarization studies of changes such that their lipid properties at any time of the year are membrane phospholipid phase separations in warm and cool climate plants. In similar to those of ephemeral species which grow at that time. JM Lyons, JK Raison, eds, Low Temperature Stress in Crop Plants: The Role We do not know what properties of the membrane lipids are of the Membrane. Academic Press, New York. In press can 15. RAISON JK 1974 A biochemical explanation of low-temperature stress in tropical acted upon by natural selection. Since present techniques only and sub-tropical plants. R Soc NZ Bull 12: 487-497 study the bulk lipid properties, there is no evidence to determine 16. RAISON JK 1980 Membrane lipids-structure and function. In PK Stumpf, ed. if temperature preference depends at least in part on the mainte- Biochemistry of Plants, Vol 7. Academic Press, New York. In press nance of a proper fluidity in the bulk phospholipids or if the bulk 17. RAISON JK, JA BERRY, PA ARMOND, CS PIKE 1980 Membrane properties in relation to the adaptation of plants to high and low temperature stress. In P phase behavior is a chance reflection of some other properties of Kramer, N Turner, eds, Adaptations of Plants to Water and High Temperature a crucial fraction. The correlation between lipid phase separation Stress. Wiley-Interscience, New York. In press temperature and a minimum habitat temperature observed in this 18. RAISON JK, EA CHAPMAN 1976 Membrane phase changes in chilling-sensitive study suggests that a rough lower limit for normal function may Vigna radiata and their significance to growth. Aust J Plant Physiol 3: 291-299 19. RAISON JK, EA CHAPMAN, SWL JACOBS, LC WRIGHT 1980 Membrane lipid be set by the phase separation. In addition, the correlation between transitions: Their correlation with the climatic distribution of plants. In JM a high lipid phase separation temperature and high habitat tem- Lyons, JK Raison, eds, Low Temperature Stress in Crop Plants: The Role of peratures suggests that the capacity to tolerate high temperature the Membranes. Academic Press, New York. In press may be related to that property. A survey of Australian plants has 20. SHREVE F, IL WIGGINS 1964 Vegetation and Flora of the Sonoran Desert. Stanford Univ Press, Stanford, pp 127-142 indicated a correlation, comparable to that reported here, between 21. SINENSKY M 1974 Homeoviscous adaptation: a homeostatic process that regulates habitat preference and phase separation temperature measured the viscosity of membrane lipids in Escherichia coli. Proc Nat Acad Sci USA with spin label probes (19). These correlations provide support for 71: 522-525 22. SKLAR LA, BS HUDSON, RD SIMONI 1976 Conjugated polyene fatty acids as theories on the importance of lipid physical properties in the membrane probes. Preliminary characterization. Proc Nat Acad Sci USA 72: adaptation and acclimation of plants to temperature. 1649-1653 There were 14 species with the C3 and 12 with the C4 pathways 23. SKLAR LA, GP MILJANICH, EA DRATZ 1979 Phospholipid lateral phase separa- of photosynthetic CO2 fixation. Both pathways occurred in each tion and the partition of cis-parinaric acid and trans-parinaric acid among aqueous, solid lipid and fluid lipid phases. Biochemistry 18: 1707-1716 category-winter ephemeral, summer ephemeral, and perennial. 24. WOLFE J 1978 Chilling injury in plants-the role of membrane lipid fluidity. The lipid properties ofthese plants appeared to reflect the thermal Plant Cell Environ 1: 241-247