Subcellular Compartmentation of the Diterpene ... - Plant Physiology

Subcellular Compartmentation of the Diterpene Carnosic Acid and Its Derivatives in the Leaves of Rosemary1 Sergi Munne´-Bosch and Leonor Alegre* Departament de Biologia Vegetal, Facultat de Biologia, Universitat de Barcelona, Avinguda Diagonal 645, 08028 Barcelona, Spain

The potent antioxidant properties of rosemary (Rosmarinus officinalis) extracts have been attributed to its major diterpene, carnosic acid. Carnosic acid has received considerable attention in food science and biomedicine, but little is known about its function in the plant in vivo. We recently found that highly oxidized diterpenes increase in rosemary plants exposed to drought and high light stress as a result of the antioxidant activity of carnosic acid (S. Munne´-Bosch, K. Schwarz, L. Alegre [1999] Plant Physiol 121: 1047–1052). To elucidate the significance of the antioxidant function of carnosic acid in vivo we measured the relative amounts of carnosic acid and its metabolites in different compartments of rosemary leaves. Subcellular localization studies show that carnosic acid protects chloroplasts from oxidative stress in vivo by following a highly regulated compartmentation of oxidation products. Carnosic acid scavenges free radicals within the chloroplasts, giving rise to diterpene alcohols, mainly isorosmanol. This oxidation product is O-methylated within the chloroplasts, and the resulting form, 11,12-di-O-methylisorosmanol, is transferred to the plasma membrane. This appears to represent a mechanism of a way out for free radicals from chloroplasts. Carnosic acid also undergoes direct O-methylation within the chloroplasts, and its derived product, 12-O-methylcarnosic acid, accumulates in the plasma membrane. O-methylated diterpenes do not display antioxidant activity, but they may influence the stability of the plasma membrane. This study shows the relevance of the compartmentation of carnosic acid metabolism to the protection of rosemary plants from oxidative stress in vivo.

Rosemary (Rosmarinus officinalis) leaf extracts show a very high antioxidant activity and are increasingly used as food additives, proposed as important human dietary factors, and investigated as inhibitors of skin tumorgenesis (Singletary and Nelshoppen, 1991; Schwarz et al., 1992; Huang et al., 1994). The main compound responsible for the antioxidant activity is the diterpene, carnosic acid (Aruoma et al., 1992), which is the most abundant antioxidant found in the leaves of rosemary. Carnosic acid is a lipophilic antioxidant that scavenges singlet oxygen, hydroxyl radicals, and lipid peroxyl radicals, thus preventing lipid peroxidation and disruption of biological membranes (Aruoma et al., 1992; Haraguchi et al., 1995). Its radical scavenging activity follows a mechanism analogous to that of other antioxidants such as ␣-tocopherol and is caused by the presence of two O-phenolic hydroxyl groups found at C11 and C12 of the molecule (Richheimer et al., 1999; Fig. 1). Carnosic acid may give rise to carnosol after enzymatic dehydrogenation or to highly oxidized diterpenes such as rosmanol or isorosmanol after enzymatic dehydrogenation and free radical attack (Luis, 1991; Luis et al., 1994). Oxidative stress in vivo induced by drought or high light stress enhances the formation of highly oxidized diterpenes due to the 1 This work was supported by the Programa Sectorial de Promocioń General del Conocimiento (grant nos. DGICYT PB96 –1257 and BOS2000 – 0560). * Corresponding author; e-mail [email protected]; fax 34 –93– 411–28 – 42.

1094

antioxidant activity of carnosic acid (Munne´-Bosch et al., 1999; Munne´-Bosch and Alegre, 2000). In addition, carnosic acid and isorosmanol can be O-methylated to form 12-O-methylcarnosic acid and 11,12-di-Omethylisorosmanol, respectively. The methylation of the O-phenolic hydroxyl groups eliminates the radical scavenging activity of the molecule and increases its lipid solubility (Brieskorn and Do¨mling, 1969). The biological activity of a compound is determined by its distribution within the plant cell. Diterpenes are synthesized in plastids via a non-mevalonate isopentenyl diphosphate pathway (McGarvey and Croteau, 1995). In this pathway, isopentenyl diphosphate is formed from pyruvate and glyceraldehyde 3phosphate, yielding 1-deoxy-d-xylulose-5-phosphate (Kleinig, 1989; Lichtenthaler, 1999). The action of various phenyltransferases then generates geranylgeranyl pyrophosphate (C20). This undergoes internal addition to form copalyl pyrophosphate from which abietane diterpenes such as carnosic acid are formed (McGarvey and Croteau, 1995). The formation of geranylgeranyl pyrophosphate and copalyl pyrophosphate has been localized in the plastids. However, interaction between organelles may occur in the transfer of metabolites from plastids to sites of secondary transformation such as endoplasmic reticulum-bound Cyt P450 oxygenases, as it occurs in the synthesis of gibberellins (a plant hormone) or abietic acid (a regulator of wound-induced responses in plants; McGarvey and Croteau, 1995). Carnosic acid may, therefore, occur in the chloroplasts or intracellular membranes of rosemary leaves.

Plant Physiology, February 2001, Vol. 125, pp. 1094–1102, www.plantphysiol.org © 2001 American Society of Plant Physiologists

Subcellular Localization of Diterpenes

11,12-di-O-methylisorosmanol was tested in different tissues of rosemary (Fig. 2). The relative amounts of diterpenes were similar in all the tissues tested, with the concentration of carnosic acid 6-fold higher than that of 12-O-methylcarnosic acid and that of carnosol, which were found in similar amounts. Isorosmanol was found at slightly lower concentrations than carnosol, 11,12-di-O-methylisorosmanol was approximately 10 times less abundant than isorosmanol, and very small amounts of rosmanol were detected. Diterpenes were not found in all the tissues tested. The leaf was the tissue showing the highest concentrations of diterpenes. Diterpenes were also present in the flowers of rosemary, although the concentrations found in the sepals were not comparable with those found in the petals. Sepals contained approximately 30% fewer diterpenes than leaves, but 3.2 times more than petals. Diterpenes were also found at low concentrations in seeds and trace amounts were detected in stems. Roots did not contain diterpenes. Subcellular Fractionation

We recently showed that in conditions favoring production of free radicals such as drought and high light stress, the formation of highly oxidized diterpenes (rosmanol, isorosmanol, and 11, 12-di-O-methylisorosmanol) is enhanced in the leaves of rosemary via the antioxidant activity of carnosic acid (Munne´Bosch et al., 1999; Munne´-Bosch and Alegre, 2000). Thus, carnosic acid might play a major role in protecting the plant in vivo from oxidative stress. However, the significance of these results was not fully clear, since studies on the subcellular localization of this function were still lacking. In an attempt to characterize the in vivo function of carnosic acid this study is aimed to elucidate the subcellular compartmentation of carnosic acid and its metabolites in the leaves of rosemary.

Fractions enriched in chloroplasts, endoplasmic reticulum, Golgi apparatus, and plasma membrane were prepared from fully developed young rosemary leaves, which is the tissue containing the highest concentrations of diterpenes. The identity and purity of each fraction were determined by assaying amounts and activities of appropriate markers (Table I). Chloroplasts showed the highest amounts of chlorophylls and no or very little enzymatic activities that are characteristic of other organelles. The identity and purity of chloroplasts were also confirmed further by microscopic observation (data not shown). The endoplasmic reticulum fraction showed the highest activity of its marker enzyme NADPH-Cyt c reductase, but no activity for the other enzymes tested. The Golgi apparatus and the plasma membrane fractions showed the highest activity of their marker enzymes, latent IDPase and vanadatesensitive ATPase, respectively, but little activity for the other enzymes tested. All the fractions tested showed very little activity of the mitochondrial enzyme marker Cyt c oxidase. Setting the specific amount of markers or marker activities at 1 in the leaf homogenate, the following relative enrichments in the organelle fractions were obtained: chloroplast fraction, 10.1 for chlorophyll; endoplasmic reticulum fraction, 8.4 for NADPH-Cyt c reductase; Golgi apparatus fraction, 7.2 for latent IDPase; and plasma membrane fraction, 5.9 for vanadate-sensitive ATPase.

RESULTS


Figure 1. Diterpenes in rosemary leaves. The antioxidant activity of diterpenes is given by the presence of two hydroxyl groups in ortho position at C11 and C12. CA, Carnosic acid; MCA, 12-Omethylcarnosic acid; CAR, carnosol; ROS, rosmanol; ISO, isorosmanol; DMIR, 11,12-di-O-methylisorosmanol.

Distribution of Diterpenes in Different Tissues

The distribution of carnosic acid, 12-O-methylcarnosic acid, carnosol, rosmanol, isorosmanol, and Plant Physiol. Vol. 125, 2001

The localization of diterpenes was studied using subcellular fractions prepared from rosemary leaves. The amount of carnosic acid, 12-O-methylcarnosic 1095

Munne´-Bosch and Alegre

Figure 2. Tissue distribution of the diterpenes carnosic acid (CA), 12-O-methylcarnosic acid (MCA), carnosol (CAR), rosmanol (ROS), isorosmanol (ISO), and 11,12-di-O-methylisorosmanol (DMIR) in rosemary plants. Results, given in milligrams per gram of dry weight, correspond to the means ⫾ SE of triplicate experiments. Sd, Seed; Rt, root; St, stem; Le, leaf; Se, sepal; Pe, petal. aND, Not detected; bTr, trace.

acid, carnosol, rosmanol, isorosmanol, and 11,12-diO-methylisorosmanol was determined in chloroplasts, endoplasmic reticulum, Golgi apparatus, and plasma membrane fractions (Fig. 3). Carnosic acid was only found in the chloroplasts of rosemary leaves, along with carnosol, rosmanol, and isorosmanol. In contrast, the O-methylated forms of carnosic acid and isorosmanol, 12-O-methylcarnosic acid, and 11,12-di-O-methylisorosmanol, respectively, were found in chloroplasts, endoplasmic reticulum, Golgi apparatus, and plasma membrane fractions. The relative amounts of carnosic acid, carnosol, rosmanol,

and isorosmanol found in the chloroplasts were similar to those found in the leaves (Fig. 2); carnosic acid was the most abundant diterpene, followed by carnosol, isoromanol, and rosmanol. O-methylated diterpenes accumulated in the plasma membrane. The amount of 12-O-methylcarnosic acid found in the plasma membrane was 2.8-fold higher than that found in the Golgi apparatus, 3.6-fold higher than that found in the endoplasmic reticulum, and 5-fold higher than that found in the chloroplasts. The amount of 11,12-di-O-methylisorosmanol found in the plasma membrane was 6-fold higher than that

Table I. Distribution of marker compounds and marker enzyme activities in subcellular fractions of rosemary leaves Data are shown in absolute values and as a percentage of the maximum for each marker (in brackets). Results correspond to the mean ⫾ of triplicate experiments. Marker a

Chlorophyll NADPH-Cyt c reductaseb Latent IDPased Vanadate-sensitive ATPased Cyt c oxidased

Chloroplast

Endoplasmic Reticulum

185.2 ⫾ 1.0 (100) ND c (0) 1.8 ⫾ 0.4 (⬍1) ND (0) 2.6 ⫾ 0.65

0 (0) 38.9 ⫾ 0.7 (100) ND (0) ND (0) 2.2 ⫾ 0.2

mg (chlorophyll a ⫹ b) g⫺1 protein. protein. a

1096

b

nmol Cyt c min⫺1 mg⫺1 protein.

Golgi Apparatus

Plasma Membrane

0 (0) 11.5 ⫾ 1.0 (30) 153.9 ⫾ 5.7 (100) 1.4 (⬍1) 0.7 ⫾ 0.5 c

ND, Not detected.

d

SE

0 (0) 7.5 ⫾ 2.8 (19) 33.8 ⫾ 5.6 (22) 295.8 ⫾ 60.6 (100) 0.8 ⫾ 0.1

␮mol phosphate min⫺1 mg⫺1

Plant Physiol. Vol. 125, 2001


Figure 3. Subcellular distribution of the diterpenes carnosic acid (CA), 12-O-methylcarnosic acid (MCA), carnosol (CAR), rosmanol (ROS), isorosmanol (ISO), and 11,12-di-O-methylisorosmanol (DMIR) in the leaves of rosemary. Results, given in milligrams per gram of dry weight, correspond to the means ⫾ SE of triplicate experiments. Chl, Chloroplast; ER, endoplasmic reticulum; GA, Golgi apparatus; PM, plasma membrane. aND, Not detected.

found in the Golgi apparatus, 8.3-fold higher than that found in the endoplasmic reticulum, and 16.6-fold higher than that found in the chloroplasts (Fig. 3). Setting the specific amount of diterpenes at 1 in the leaf homogenate, the following relative enrichments in the organelle fractions were obtained: chloroplast fraction, 5.3 for carnosic acid, 4.0 for 12-O-methylcarnosic acid, 5.3 for carnosol, 4.8 for rosmanol, 4.2 for isorosmanol, and 4.0 for 11,12-di-O-methylisorosmanol; endoplasmic reticulum fraction, 5.5 for 12-O-methylcarnosic acid and 8 for 11,12-di-O-methylisorosmanol; Golgi apparatus fraction, 7.3 for 12-O-methylcarnosic acid and 10.5 for 11,12-di-O-methylisorosmanol; and plasma membrane fraction, 18.3 for 12-O-methylcarnosic acid and 63.3 for 11,12-di-O-methylisorosmanol. Occurrence of Diterpene Esters and Glycosides

The possible occurrence of diterpene esters or glycosides in the leaves of rosemary was also tested (Fig. 4). If conjugates were present, the treatment with HCl or KOH might cause a rupture of the bond of the conjugated form resulting in an increase of free diterpenes. However, the acid and alkali treatments resulted in a decrease in the concentration of diterpePlant Physiol. Vol. 125, 2001

nes rather than an increase. This decrease was similar to that observed for pure carnosic acid in solution. Thus, the presence of diterpene conjugates, at least in significant amounts, may be excluded. DISCUSSION

Diterpenes play diverse functional roles in plants, acting as hormones (gibberellins), regulators of woundinduced responses (abietic acid), photosynthetic pigments (the phytyl chain of chlorophylls), and antioxidants (the phytyl moiety of tocopherols; McGarvey and Croteau, 1995; Lichtenthaler et al., 1997). Although tocopherols and carotenoids are, among lipid-soluble antioxidants, the best-characterized groups of compounds in their function of protecting the plant from oxidative stress, plants contain other compounds (i.e. flavonoids and diterpenes) displaying high antioxidant properties (Schwarz and Ternes, 1992; Rice-Evans et al., 1997). In vivo studies have shown that the diterpene carnosic acid may protect biological membranes from oxidative damage (Haraguchi et al., 1995), and that under drought- and high light-induced oxidative stress conditions, the amounts of highly oxidized diterpenes (i.e. rosmanol, isorosmanol, and 11, 12-di-O1097


Figure 4. Influence of acid and alkali treatments on the extraction of carnosic acid (CA) from rosemary leaves. The extraction of diterpenes in methanol (Me) was compared with their extraction in the presence of HCl (⫹ HCl) and KOH (⫹ KOH) to study the possible occurrence of diterpenoid esters or glycosides. A control experiment using pure carnosic acid in solution was performed. Similar changes to those obtained for carnosic acid were also observed with the other diterpenes. Results are given in absolute values (milligrams per gram of dry weight or mililiter, bars) and as a percentage of the initial values (%, black symbols). Results correspond to the means ⫾ SE of three independent replicates.

methylisorosmanol) in rosemary leaves increase due to the antioxidant activity of carnosic acid (Munne´-Bosch et al., 1999; Munne´-Bosch and Alegre, 2000). This study shows the subcellular compartmentation of carnosic acid and its oxidation products in rosemary leaves (Fig. 5). This allows us to fully explain previous results and better characterize the in vivo function of carnosic acid in rosemary plants. Carnosic acid was only found in chloroplasts. Copalyl pyrophosphate, the immediate precursor of the diterpene carnosic acid, has also been found in these organelles (McGarvey and Croteau, 1995), thus suggesting that carnosic acid is synthesized in the chloroplasts. The oxidation products of carnosic acid, rosmanol, and isorosmanol were also only found in the chloroplasts, thus indicating that carnosic acid functions as an antioxidant in the chloroplasts of 1098

rosemary leaves. This demonstrates that the antioxidant function of carnosic acid is linked to the chloroplasts, which are the organelles most exposed to oxidative damage in plant cells. Chloroplasts are organelles particularly liable to oxygen toxicity, since they function under high oxygen tensions and in the light. As a result of a stress-induced de-regulation of the photosynthetic activity of chloroplasts, more free radicals (singlet oxygen, superoxide anion, hydroxyl radicals, and peroxyl radicals) may be generated in this organelle, which may cause disruption of membranes and cell death (Foyer et al., 1994; Osmond et al., 1997; Asada, 1999). Environmental constraints such as drought and high light stress may inhibit the normal functioning of the photosynthetic apparatus and cause oxidative damage to the chloroplasts if the plants are not protected by antioxidant defenses (Bjo¨rkman, 1987; Smirnoff, 1993). Therefore, the compartmentation of the antioxidant function of carnosic acid in chloroplasts might be an adaptive mechanism that rosemary plants have evolved to withstand drought and high light stress typical of the Mediterranean climate. The lipophilic nature of carnosic acid and its antioxidant properties suggest that carnosic acid may play a role similar to ␣-tocopherol in scavenging free radicals formed as a result of the photosynthetic activity of chloroplasts. As occurs with ␣-tocopherol (Fryer, 1992; Shintani and DellaPenna, 1998), diterpenes are found at their highest concentrations in photosynthetic tissues (i.e. leaves and sepals), and appreciable amounts of these compounds are also found in seeds and petals. The leaves of rosemary contain amounts of ␣-tocopherol comparable with other species (Munne´-Bosch and Alegre, 2000), which indicates that this species is not deficient in this antioxidant. Thus, carnosic acid may cooperate with ␣-tocopherol in chloroplasts, rather than replace its activity. Some authors have suggested that carnosic acid could interact synergically with ␣-tocopherol by reducing the tocopheryl radicals to active ␣-tocopherol (Hopia et al., 1996). This might explain the great tolerance of rosemary to drought and high light stress, but such an interaction needs to be examined in vivo. The results also indicate that the oxidation product of carnosic acid, isorosmanol, is O-methylated within the chloroplasts and that the derived diterpene, 11,12-di-O-methylisorosmanol, is transferred from the chloroplasts to the plasma membrane. Although we did not find 11,12-di-O-methylrosmanol, the O-methylated form of rosmanol, it is likely to occur in rosemary leaves, as it has been described in other species (Luis et al., 1994). Thus, the mechanism accounting for isorosmanol may also apply for its isomer, rosmanol. The formation of O-methylated diterpenes from carnosic acid oxidation products within the chloroplasts may represent, therefore, a mechanism of a way out of free radicals from chloroplasts. Plant Physiol. Vol. 125, 2001


Figure 5. Subcellular compartmentation of carnosic acid and its metabolites in the leaves of rosemary. CA, Carnosic acid; CAR, carnosol; MCA, 12-O-methylcarnosic acid; ROS, rosmanol; ISO, isorosmanol; DMIR, 11,12-di-O-methylisorosmanol; frs, free radical scavenging.

The accumulation of O-methylated diterpenes in the plasma membrane indicates that these compounds may have a function apart from being antioxidants, since the O-methylation of one of two hydroxyl groups found at C11 and C12 leads to a decrease in the free radical scavenging activity of the molecule (Brieskorn and Do¨mling, 1969). O-methylated diterpenes accumulating in the plasma membrane may have a similar function to sterols. Although O-methylated diterpenes do not display antioxidant activity, they retain at least one hydroxyl group in the molecule. This hydroxyl group may form hydrogen bonds with the head group of a phospholipid and its dimensions may allow cooperative van der Waals attractive forces to reinforce and stabilize the lipid chain (Havaux, 1998). As occurs with sterols, the amount of O-methylated diterpenes accumulated in the plasma membrane may affect the integrity of the cells and influence plant growth and development (Schaller et al., 1998; Brown and Goldstein, 1999). The same may apply for diterpenes found in the chloroplasts. It is well known that the lipid-soluble antioxidants ␣-tocopherol and ␤-carotene may significantly influence the stability of membranes in chloroplasts (Havaux, 1998). The leaves of rosemary contain 100 times more carnosic acid than ␣-tocopherol and 35 times more carnosic acid than ␤-carotene (Munne´-Bosch and Alegre, 2000). Thus, carnosic acid is thought to affect the stability of membranes in the chloroplasts more than these lipid-soluble antioxidants. We conclude that carnosic acid metabolism plays a dual role in rosemary plants. First, the localization of the antioxidant function of carnosic acid in the chloroplasts and the enhanced formation of oxidation products in stress conditions indicates that carnosic acid protects rosemary plants from environmental constraints by scavenging free radicals within the chloroplasts. Second, the large amounts of antioxidant diterpenes and nonantioxidant diterpenes found in Plant Physiol. Vol. 125, 2001

chloroplasts and intracellular membranes, respectively, suggest that carnosic acid metabolism may also play a role in the stability of cell membranes. MATERIALS AND METHODS Plant Material Rosemary (Rosmarinus officinalis) plants grown at the experimental fields of the University of Barcelona were used for this study. Six plants of the same genetic origin, age (2 years old), and height (1 m) were chosen. Diterpenes were determined in plant tissues (i.e. roots, stems, leaves, sepals, and petals) and subcellular fractions of leaves (i.e. chloroplasts, endoplasmic reticulum, Golgi apparatus, and plasma membrane) from material collected at predawn during January and February 2000. Diterpenes were also determined in seeds obtained from Semillas Fito´ (Barcelona). For the determination of diterpenes in different tissues, the plant material was collected, immediately frozen in liquid nitrogen, and freeze-dried prior to extraction. For the determination of diterpenes in subcellular fractions, the leaves were collected and cells were immediately fractionated as described below. The subcellular fractions obtained were immediately frozen in liquid nitrogen and freezedried prior to extraction. Isolation of Chloroplasts Chloroplasts were isolated by an adaptation of the method of Walker and Weinstein (1991). The abaxial epidermis of leaves was mechanically removed with the help of a scalpel to avoid the interference of trichomes in the isolation procedure. Leaf tissue (5 g) was ground with an ice-chilled mortar and pestle with a 20-mL isolation buffer, pH 7.8, containing 0.5 m sorbitol, 50 mm Tricine, 1 mm dithiothreitol, 1 mm MgCl2, 1 mm butylated hydroxytoluene, and 0.1% (w/v) bovine serum albumin (BSA). The homogenate was filtered through four layers of cheesecloth 1099


and centrifuged at 2,500g for 4 min. The pellet was resuspended in 10 mL of isolation buffer and then centrifuged at 200g for 1 min. The chloroplasts in the supernatant were sedimented by centrifugation at 2,500g for 4 min. Chloroplasts were purified by resuspending the pellets in 2 mL of isolation buffer, layering onto 10 mL of 25% (v/v) Percoll (in isolation buffer), and centrifuging at 15,800g for 20 min. The chloroplasts pellets were resuspended in 10 mL of isolation buffer lacking BSA, centrifuged at 2,500g for 4 min, and used immediately. The whole procedure was carried out in dimmed room light at 4°C. Isolation of Endoplasmic Reticulum, Golgi Apparatus, and Plasma Membrane Fractions A Suc-density gradient and a polymer two-phase system were used for the isolation of plasma and intracellular membranes (Briskin et al., 1987; Moreau et al., 1998). Leaf tissue (5 g) was homogenized at high speed for 60 s with a Polytron probe (Kinematica, Luzern, Switzerland) in 20 mL of buffer consisting of 10 mm KH2PO4, pH 8.2, with 0.5 m sorbitol, 5% (w/v) polyvinylpyrrolidone (40,000), 0.5% (w/v) BSA, 2 mm salicylhydroxamic acid, 1 mm butylated hydroxytoluene, and 1 mm phenylmethylsulfonyl fluoride. The homogenate was filtered through six layers of cheesecloth and was centrifuged at 1,000g for 10 min, 10,000g for 10 min, and 150,000g for 60 min. The resulting microsomal pellet was resuspended in 10 mL of buffer consisting of 10 mm KH2PO4, pH 8.2, with 0.5 m sorbitol. One-half of the suspension was loaded onto a discontinuous Suc-density gradient consisting of 2.5 mL of 37% (w/v) Suc, 3.5 mL of 25% (w/v) Suc, and 3.5 mL of 18% (w/v) Suc. After centrifuging at 80,000g for 150 min, membranes at the 18%/ 25% (endoplasmic reticulum fraction) and 25%/37% (Golgi apparatus fraction) Suc interface were collected, diluted with phosphate buffer, pH 8.2, centrifuged at 100,000g for 60 min, and used immediately. The other one-half of the microsomal suspension was mixed with a polymer (polyethylene glycol [PEG] 4000/Dextran T500 mixture) in 0.5 m sorbitol containing 10 mm KH2PO4 and 40 mm NaCl, pH 7.8, to obtain final PEG and Dextran concentrations of 6% (w/w). The solution (final volume of 28 mL) was centrifuged for 15 min at 1,000g and the PEG-enriched upper phase (12 mL) was recovered without disturbing the interface. The upper phase was repartitioned twice to produce a chlorophyll-free preparation. Plasma membranes were recovered after centrifuging at 150,000g for 60 min, and were used immediately. The whole procedure was carried out in dimmed room light at 4°C. Assays of Markers for Subcellular Fractions Specific subcellular compartments were identified by assays for the following markers: chloroplasts, chlorophyll; endoplasmic reticulum, NADPH-Cyt c reductase; Golgi apparatus, latent IDPase; and plasma membrane, vanadate-sensitive ATPase. Marker assays were performed immediately after isolation of the corresponding fraction. Chlorophylls were measured spectrophotometrically in 1100

80% (v/v) acetone extracts (Lichtenthaler and Wellburn, 1983). The NADPH-Cyt c reductase assay was performed at 25°C in a 3-mL reaction volume containing 0.1 mL of membrane suspension (10 ␮g of protein), 0.1 mL of 50 mm NaCN, 0.2 mL of 0.45 mm Cyt c, and 2.5 mL of 50 mm sodium phosphate buffer, pH 7.5. The reaction was started by the addition of 0.1 mL of 3 mm NADPH, and the reduction of Cyt c was followed spectrophotometrically as an absorbance increase at 550 nm (Lord, 1987). Latent IDPase was assayed in a 3-mL reaction volume containing 0.1 mL of membrane suspension (10 ␮g of protein), 0.1 mL of 3 mm MgSO4, 0.1 mL of 50 mm KCl, and 2.6 mL of 30 mm Tris-MES [2-(N-morpholino)-ethanesulfonic acid] buffer (MES titrated with Tris to pH 7.5). The reaction was started by the addition of 0.1 mL of 3 mm IDP (sodium salt), and the released Pi was determined on freshly isolated membranes and after 6 d of storage at 2°C to 4°C by using ammonium molybdate (Briskin et al., 1987). Vanadatesensitive ATPase was assayed in a 3-mL reaction volume containing 0.1 mL of membrane suspension (10 ␮g of protein), 0.1 mL of 3 mm MgSO4, 0.1 mL of 50 mm KCl, and 2.5 mL of 30 mm Tris-MES buffer (MES titrated with Tris to pH 6.5), in the presence or absence of 0.1 mL of 50 ␮m Na3VO4. The reaction was started by the addition of 0.1 mL of 3 mm ATP. The ATP substrate was present as Tris salt after treatment with Dowex 50-W exchange resin (H⫹ form). The assay was performed at 38°C for 30 min, and the released Pi was determined by using ammonium molybdate (Briskin et al., 1987; Fan et al., 1999). The specific activity of Cyt c oxidase (a mitochondria marker enzyme) was assayed at 25°C in a 3-mL reaction volume containing 0.1 mL of membrane suspension (10 ␮g of protein), 0.1 mL of 0.3% (w/v) Triton X-100, and 2.7 mL of 50 mm sodium phosphate buffer, pH 7.5. The reaction was started by the addition of 0.1 mL of 0.45 mm dithionite-reduced Cyt c, and the decrease at A550 was monitored spectrophotometrically (Fan et al., 1999). Protein concentration was determined by the method of Bradford (1976) using a kit (Bio-Rad, Hercules, CA) with BSA as a standard. Control Experiments To test the possible re-distribution of diterpenes between different subcellular organelles during subcellular fractionation, the following experiment was performed. Carnosic acid (90% purity, 5 mg) was added to the initial homogenate. After isolation of the corresponding subcellular fractions, the amount of diterpenes was determined. The amounts of diterpenes obtained in chloroplast-, endoplasmic reticulum-, Golgi apparatus-, and plasma membraneenriched fractions were not significantly different (P ⱕ 0.05, ANOVA) from those obtained when carnosic acid was not added to the homogenate. Hydrolisis of Esters and Glycosides To investigate the presence of diterpenoid esters or glycosides, freeze-dried leaves were hydrolized by heating at 60°C for 1 h with 5 mL of 6% (w/v) KOH in methanol or Plant Physiol. Vol. 125, 2001


with 5 mL of 20% (v/v) 6 m HCl in methanol (Hartmann and Benveniste, 1987; Hertog et al., 1992). Controls were carried out by heating the leaves at 60°C for 1 h with 5 mL of methanol and by testing the degradation of carnosic acid (90% purity) under the same conditions. Determination of Diterpenes For the determination of diterpenes, the corresponding samples were freeze-dried and immediately analyzed. Diterpenes were determined as described (Munne´-Bosch et al., 1999). In short, the samples were extracted with methanol containing 0.005% (w/w) citric acid and 0.005% (w/w) isoascorbic acid, and were sonicated for 20 s (Vibra-Cell Ultrasonic Processor, Sonics and Materials, Danbury, CT). Diterpenes were separated on an octadecyl (C18) silica Hypersil 5-␮m column (250 ⫻ 4 mm, Teknokroma, St. Cugat, Spain) during 52 min at a flow rate of 0.6 mL min⫺1. The eluants consisted of A, 51% (w/v) acetonitrile and 49% (w/v) water, containing 0.83% (w/v) 2 m citric acid; and B, 97% (w/v) acetonitrile and 3% (w/v) water, containing 0.5% (w/v) 2 m citric acid. The following gradient was used: 0 to 20 min, 100% A and 0% B; 20 to 34 min decreasing to 50% A and 50% B; 34 to 40 min decreasing to 0% A and 100% B; 40 to 48 min, increasing to 100% A and 0% B; and 48 to 52 min, 100% A and 0% B. Individual diterpenes were identified by their characteristic mass spectra. Carnosic acid (98% purity) was used for calibration. All diterpenes were quantified relative to carnosic acid at 230 nm (Spectralphotometer 430 Kontron, Zurich), because the UV spectra of other diterpenes are similar to that of carnosic acid. ACKNOWLEDGMENTS We thank the Serveis Cientı´fico-te`cnics from the University of Barcelona for technical assistance and Dr. Karin Schwarz (University of Kiel) for kindly providing us with carnosic acid. We also thank the University of Barcelona for the grant given to S.M. Received June 19, 2000; returned for revision September 5, 2000; accepted October 22, 2000. LITERATURE CITED Aruoma OI, Halliwell B, Aeschbach R, Lo¨liger J (1992) Antioxidant and pro-oxidant properties of active rosemary constituents: carnosol and carnosic acid. Xenobiotica 22: 257–268 Asada K (1999) The water-water cycle in chloroplasts: scavenging of active oxygens and dissipation of excess photons. Annu Rev Plant Physiol Plant Mol Biol 50: 601–639 Bjo¨rkman O (1987) Responses to different quantum flux density. In J Biggings, ed, Progress in Photosynthesis Research. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 11–18 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing Plant Physiol. Vol. 125, 2001

the principle of protein-dye binding. Anal Biochem 72: 248–254 Brieskorn CH, Do¨mling H (1969) Carnosolsaüre, der wichtige antioxidativ wirksame Inhaltsstoff des Rosmarinund Salbeiblattes. Z Lebens Unter Fors 141: 10–16 Briskin DP, Leonard RT, Hodges TK (1987) Isolation of the plasma membrane: membrane markers and general principles. Methods Enzymol 148: 543–558 Brown MS, Goldstein JS (1999) A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA 96: 11041–11048 Fan L, Zheng S, Cui D, Wang X (1999) Subcellular distribution and tissue expression of phospholipase D␣, D␤, and D␥ in Arabidopsis. Plant Physiol 119: 1371–1378 Foyer CH, Descourvie`res P, Kunert KJ (1994) Protection against oxygen radicals: an important defense mechanism studied in transgenic plants. Plant Cell Environ 17: 507–523 Fryer MJ (1992) The antioxidant effects of thylakoid vitamin E (␣-tocopherol). Plant Cell Environ 15: 381–392 Haraguchi H, Saito T, Okamura N, Yagi A (1995) Inhibition of lipid peroxidation and superoxide generation by diterpenoids from Rosmarinus officinalis. Planta Med 61: 333–336 Hartmann MA, Benveniste P (1987) Plant membrane sterols: isolation, identification and biosynthesis. Methods Enzymol 148: 632–650 Havaux M (1998) Carotenoids as membrane stabilizers in chloroplasts. Trends Plant Sci 3: 147–151 Hertog MGL, Hollman PCH, Venema DP (1992) Optimization of a quantitative HPLC determination of potentially anticarcinogenic flavonoids in vegetables and fruits. J Agri Food Chem 40: 1591–1598 Hopia AI, Huang S, Schwarz K, German JB, Frankel EN (1996) Effect of different lipid systems on antioxidant activity of rosemary constituents carnosol and carnosic acid with and without ␣-tocopherol. J Agri Food Chem 44: 2030–2036 Huang M, Ho C, Wang ZY, Ferraro T, Lou Y, Stauber K, Ma W, Georgiadis C, Laskin JD, Conney AH (1994) Inhibition of skin tumorgenesis by rosemary and its constituents carnosol and ursolic acid. Cancer Res 54: 701–708 Kleinig H (1989) The role of plastids in isoprenoid biosynthesis. Annu Rev Plant Physiol Plant Mol Biol 40: 39–59 Lichtenthaler HK (1999) The 1-deoxy-d-xylulose-5phosphate pathway of isoprenoid biosynthesis in plants. Annu Rev Plant Physiol Plant Mol Biol 50: 47–65 Lichtenthaler HK, Schwender J, Disch A, Rohmer M (1997) Biosynthesis of isoprenoids in higher plant chloroplasts proceeds via a mevalonate-independent pathway. FEBS Lett 400: 271–274 Lichtenthaler HK, Wellburn AR (1983) Determination of total carotenoids and chlorophylls a and b of leaf extracts in different solvents. Biochem Soc Trans 11: 591–592 Lord JM (1987) Isolation of endoplasmic reticulum: general principles, enzymatic markers, and endoplasmic reticulum-bound polysomes. Methods Enzymol 148: 576–584 1101


Luis JG (1991) Chemistry, biogenesis, and chemotaxonomy of the diterpenoids of Salvia. In JB Harborne, FA TomasBarberan, eds, Ecological Chemistry and Biochemistry of Plant Terpenoids. Clarendon Press, Oxford, pp 63–82 Luis JG, Quinõnes W, Grillo TA, Kishi MP (1994) Diterpenes from the aerial part of Salvia columbariae. Phytochemistry 35: 1373–1374 McGarvey DJ, Croteau R (1995) Terpenoid metabolism. Plant Cell 7: 1015–1026 Moreau P, Hartmann M, Perret M, Sturbois-Balcerak B, Cassagne C (1998) Transport of sterols to the plasma membrane of leek seedlings. Plant Physiol 117: 931–937 Munne´-Bosch S, Alegre L (2000) Changes in carotenoids, tocopherols and diterpenes during drought and recovery, and the biological significance of chlorophyll loss in Rosmarinus officinalis plants. Planta 210: 925–931 Munne´-Bosch S, Schwarz K, Alegre L (1999) Enhanced formation of ␣-tocopherol and highly oxidized abietane diterpenes in water-stressed rosemary plants. Plant Physiol 121: 1047–1052 Osmond B, Badger M, Maxwell K, Bjo¨rkman O, Leegod R (1997)) Too many photons: photorespiration, photoinhibition and photooxidation. Trends Plant Sci 2: 119–120 Rice-Evans CA, Miller NJ, Paganga G (1997) Antioxidant properties of phenolic compounds. Trends Plant Sci 2: 152–159 Richheimer SL, Bailey DT, Bernart MW, Kent M, Vininski JV, Anderson LD (1999) Antioxidant activity and oxidative degradation of phenolic compounds isolated from rosemary. Recent Res Dev Oil Chem 3: 45–58

1102

Schaller H, Bouvier-Nave´ P, Benveniste P (1998) Overexpression of an Arabidopsis cDNA encoding a sterol-C241methyltransferase in tobacco modifies the ratio of 24methyl cholesterol to sitosterol and is associated with growth reduction. Plant Physiol 118: 461–469 Schwarz K, Ternes W (1992) Antioxidative constituents of Rosmarinus officinalis and Salvia officinalis: I. Isolation of carnosic acid and formation of other phenolic diterpenes. Z Lebens Unter Fors 195: 99–103 Schwarz K, Ternes W, Schmauderer E (1992) Antioxidative constituents of Rosmarinus officinalis and Salvia officinalis: III. Stability of phenolic diterpenes of rosemary extracts under thermal stress as required for technological processes. Z Lebens Unter Fors 195: 104–107 Shintani D, DellaPenna D (1998) Elevating the vitamin E content of plants through metabolic engineering. Science 282: 2098–2100 Singletary KW, Nelshoppen JM (1991) Inhibition of 7,12dimethylbenzanthracene- (DMBA) induced mammary tumorgenesis and of in vivo formation of mammary DMBA-DNA adducts by rosemary extract. Cancer Lett 60: 169–175 Smirnoff N (1993) Tansley review no. 52: the role of active oxygen in the response of plants to water deficit and desiccation. New Phytol 125: 27–58 Walker CJ, Weinstein JD (1991) In vitro assay of the chlorophyll biosynthetic enzyme Mg-chelatase: resolution of the activity into soluble and membrane-bound fractions Proc Natl Acad Sci USA 88: 5789–5793

Plant Physiol. Vol. 125, 2001