(Salvia officinalis)I - Plant Physiology

Plant Physiol. (1990) 93, 1559-1567 0032-0889/90/93/1 559/09/$01 .00/0

Received for publication January 5, 1990 Accepted Aprl 19,1990

Metabolism of Monoterpenes in Cell Cultures of Common Sage (Salvia officinalis)I Biochemical Rationale for the Lack of Monoterpene Accumulation Kimberly L. Falk, Jonathan Gershenzon, and Rodney Croteau* Institute of Biological Chemistry, and Graduate Program in Plant Physiology, Washington State University, Pullman, Washington 99164-6340 ABSTRACT

and considerable evidence indicates that these secretory structures are the primary, if not the exclusive, sites of monoterpene biosynthesis (21, 41, 49). It might seem then that monoterpenes are unlikely to be produced in cell culture systems in the absence of such organized structures. In fact, undifferentiated callus of Mentha piperita (peppermint) showed no trace of monoterpene accumulation (6), whereas M. piperita callus with adventitious shoots, containing leaflets with glandular trichomes, produced significant quantities of monoterpenes (10, 1 1). However, differentiation to the level of glandular trichomes is not always a precondition for monoterpene accumulation in cell culture (17), indicating that, at least under some conditions, monoterpene biosynthesis is possible in less organized systems. At a fundamental level, the general absence of monoterpene accumulation in undifferentiated cultures could be due to the lack of significant biosynthetic activity or to the presence of efficient catabolic processes. De novo monoterpene biosynthesis, as distinct from monoterpene accumulation (1, 17, 44) or the biotransformation of exogenous monoterpenes (19), has rarely been directly measured (5). Several investigators (2-4, 34) have demonstrated the conversion of labeled mevalonic acid to more advanced precursors, such as dimethylallyl, geranyl, and farnesyl pyrophosphate, in cell-free extracts from cultures of diverse essential oil species, but it is not possible to determine with certainty whether these metabolites represent intermediates in the biosynthesis of monoterpenes or in the formation of higher products such as phytosterols. By contrast, the efficient biotransformation of exogenous monoterpenes in cell culture ( 19) implies that at least portions of monoterpene metabolic pathways may be widely present in these systems. The ability of plant cell cultures to catabolize added monoterpenes (1, 7, 15, 20) suggests that degradative capability may be critically important in avoiding the toxic effects of these compounds on the growth and viability of cells in culture ( 14). In this paper, we describe the metabolism of (+)-camphor and other monoterpenes in cell cultures of common sage (Salvia officinalis). Camphor metabolism was emphasized in this study because this bicyclic ketone is one of the major monoterpenoid products of the intact plant and because the pathway and enzymes of camphor biosynthesis from the ubiquitous isoprenoid precursor, geranyl pyrophosphate, are

Leaves of common sage (Salvia officinalis) accumulate monoterpenes in glandular trichomes at levels exceeding 15 milligrams per gram fresh weight at maturity, whereas sage cells in suspension culture did not accumulate detectable levels of monoterpenes (0.3 micrograms per gram fresh weight) at the late log phase of growth. Other monoterpene synthetic enzymes were present as well. In vivo measurement of the ability to catabolize (+)-camphor in these cells indicated that degradative capability exceeded biosynthetic capacity by at least 1000-fold. Therefore, the lack of monoterpene accumulation in undifferentiated sage cultures could be attributed to a low level of biosynthetic activity (relative to the intact plant) coupled to a pronounced capacity for monoterpene catabolism.

The accumulation of terpenoid natural products in plant cell cultures has been successfully demonstrated in the cases of diterpenoids and sesquiterpenoids, but rarely in the case of monoterpenes. Thus, there are reports of the production of diterpenoid substances in culture at levels exceeding those of the intact plant (42, 43) and the induced accumulation of sesquiterpene phytoalexins in culture is well documented (12, 13, 16), whereas most accounts of monoterpene accumulation in cell culture systems (1, 17, 44) describe either very low levels of production or compositional patterns that differ markedly from those of the intact plant. Monoterpenes in intact plants usually accumulate in the extracellular storage spaces of specialized secretory structures, such as glandular trichomes, resin ducts, or resin cavities (35), ' Research supported in part by U.S. Department of Energy grant DE-FG06-88ER13869 and by Project 0268 from the Agricultural Research Center, Washington State University, Pullman, WA 99164.

1559

1 560

Plant Physiol. Vol. 93, 1990

FALK ET AL.

well known (Fig. 1) (28, 31 '32). Additionally, the early steps in the catabolism of camphor, via 1,2-campholide and the corresponding glucoside-glucose ester (Fig. 1), in S. officinalis leaves have been documented (25, 26). In the present work, the virtual absence of monoterpene accumulation in S. officinalis cell suspension cultures was shown to result from a low level of biosynthetic activity coupled to a pronounced ability to catabolize these compounds. MATERIALS AND METHODS Plant Materials, Substrates, and Reagents Leaves of common sage (Salvia officinalis L.) were surface sterilized by soaking in 2% aqueous NaOCl containing 0.02% Tween 20 for 10 min followed by rinsing with sterile, distilled water. Discs (5 mm diameter) cut from the sterilized leaves were placed on Murashige-Skoog medium (45) containing 0.2 mg/L 2,4-D, 1 mg/L kinetin, and 0.8% (w/v) Phytagar (Gibco Laboratories), and the resulting callus was subcultured every 4 weeks (since September 1986) onto maintenance medium containing 1.0 mg/L 2,4-D and 1.0 mg/L kinetin, and kept in the dark at 28°C. For the preparation of suspension cultures, flasks containing 50 mL maintenance medium without agar were each inoculated with 1.5 g of callus tissue, and were incubated in the dark at 28°C on an orbital shaker (125 rpm). For time-course studies, cultures were initiated with a 10 mL aliquot (approximately 1 g) of first passage suspension cells harvested at 7 to 10 d (early log phase) and added to fresh medium. The preparation of (+)-[U-'4C]camphor was carried out by exposing approximately 500 sage plants (28 d old) to 1 mCi of '4CO2 (generated from Na214CO3 at 0.3 Ci/mol) in a sealed acrylic chamber under illumination. After 1 h of exposure, the chamber was flushed with air into a KOH trap. The apical buds plus the top leaf pairs were harvested 24 h later, steam distilled, and the [U-'4C]camphor (- 10 mg at 1.1 mCi/mol) was isolated from the distillate by TLC on silica gel G (hexanes:ether, 2:1 (v/v)). The sources of (+)-borneol, (+)-camphor, (+)-bornyl pyrophosphate, and (+)- 1,2-campholide have been described (26, 28, 32). [1-3H]Geraniol and [1-3H] geranyl pyrophosphate (100 Ci/mol) were prepared by standard procedures (31). Tritium-labeled monoterpene olefins were obtained by incubating cell-free extracts from sage leaves with [1-3H]geranyl pyrophosphate as previously described (37). Radio-GLC analysis of this olefin mixture confirmed the presence of a-pinene, camphene, p-pinene, myrcene, limonene, and sabinene (at a combined specific activity of 100 Ci/mol). [U-'4C]Sucrose (671 Ci/mol) was obtained from New England Nuclear. All other reagents and biochemicals were obtained from Aldrich or Sigma Chemical Co. unless otherwise noted. Monoterpene Adsorption by Polystyrene Resin in Sage Cell Culture To test the efficacy of beaded polystyrene resin (Amberlite XAD-4, Rohm and Haas) as a lipophilic trap for volatile monoterpenes in suspension culture, flasks containing 40 mL of the maintenance medium and 500 mg of resin (washed

BIOSYNTHESIS

rSo~PP

Geranyl Pyrophosphate

*

CATABOLISM OPP

Bornyl Pyrophosphate

0

4

Glc

Borneol

*

1 ,2-Campholide

Camphor Figure 1. Pathways for the conversion of geranyl pyrophosphate to camphor, and for the conversion of camphor to the glucoside-glucose ester of 1,2-campholide.

MONOTERPENE METABOLISM IN SAGE CELL CULTURE

exhaustively with 95% ethanol and pentane, then air dried) were inoculated with a 10 mL suspension of early log phase (d 7) cells plus either 1.14 uCi of [1-3H]geraniol or 1.03 uCi of sage-derived [3H]monoterpene olefins. Control flasks contained 50 mL of the maintenance medium (without cells) and 500 mg XAD resin plus either 1.14 uCi of [I -3H]geraniol or 1.03 ,uCi of [3H]olefins. After 14 d of incubation, the cultures were chilled on ice then centrifuged at 200g for 10 min. The XAD resin, which floated after centrifugation, was separated from the mixture of medium and cells, and washed three times with 3 mL aliquots of diethyl ether if the inoculum was [1-3H]geraniol, or twice with pentane (3 mL) and twice with ether (3 mL) if the initial inoculum was [3H]olefins. Cells were filtered from the medium and homogenized in water with a Ten-Broeck homogenizer, and the homogenate was extracted as above, depending on the inoculum. The medium was similarly extracted. In flasks without cells, the resin was separated from the medium by filtration and extracted as described above. The tritium content of each fraction was determined by scintillation spectrometry. Accumulation of Monoterpenes in Culture

For analysis of monoterpene accumulation in suspension culture, eight flasks containing 500 mg XAD resin each were inoculated with 1 g of early log phase suspension cells and allowed to incubate for 1, 4, 8, 12, 14, 16, 18, or 20 d. On the prescribed day, each culture was harvested, the packed cell volume of the culture measured after centrifugation at 200g, and the culture frozen before further analysis. After thawing, the resin was separated from the cells as before and washed with two 3 mL portions of pentane which were passed over a short column of silica gel (type 60A, Mallinckrodt), overlaid with anhydrous Na2SO4, to collect the monoterpene olefins. To obtain the oxygenated monoterpenes, the resin was washed two more times with 3 mL portions of ether and this extract was passed over the same silica gel column. The extracts containing the monoterpene olefins and the oxygenated monoterpenes were concentrated to 1 mL, and an internal standard (25 nmol of menthone) was added to each in preparation for capillary GLC analysis. Accumulation of Monoterpene Glycosides in Culture A 50 mL suspension culture in late stationary phase was used in this experiment. Cells (- 10 g) were separated from the medium and homogenized with a Ten-Broeck homogenizer in 40 mL methanol containing 0.5 g NaHCO3, 8 mmol glucono-.-lactone to inhibit endogenous glucosidase activity (7), and 100 jg [3-3H]menthol glucoside (27,47) as an internal standard. After homogenization, the extract was centrifuged at 27,000g for 30 min, and the supernatant combined with the medium and extracted with pentane:ether (2:1). The organic extract was concentrated under vacuum, lyophilized to near dryness, and then loaded onto a 12 x 100 mm column of Davisil RP- 18 (Alltech Associates) equilibrated with distilled water. The column was washed with 200 mL of distilled water, and the glycosides eluted with 200 mL of methanol. The methanol eluate was concentrated to dryness and then

1561

hydrolyzed sequentially with almond ,B-glucosidase (in 100 mM acetate-Tris, pH 5.0) and porcine esterase (same buffer, adjusted to pH 8.0) using several portions of fresh enzyme over the course of 3 d to ensure complete hydrolysis (48). The liberated aglycones were then extracted from the mixture with several portions of pentane and the extracts analyzed by capillary GLC. An aliquot of the extract was also analyzed for tritium content to determine the recovery of the internal standard. Measurement of in Vivo Rate of Monoterpene Biosynthesis from [U-14C]Sucrose To determine whether monoterpene biosynthesis occurs in cell culture, this activity was measured in d 13 suspension cultures to which 0.45 gCi of [U-_4C]sucrose (671 Ci/mol) was added. Since, as described below, a 50 mL suspension culture could metabolize 0.5 mg (3.2 ,umol) of camphor in 48 h, unlabeled camphor was added as a trap according to the following protocol: 2 ,mol (dissolved in a minimum amount of ethanol) was added 1 h prior to [U-'4C]sucrose addition and 6 Mmol were added with the sucrose. Following incubation (12 h), the cultures were steam distilled with 15 mL of ether, using 200 nmol menthol as internal standard, and the recovered camphor was purified by TLC for determination of radioactivity content.

Preparation and Assay of Monoterpene Biosynthetic Activities in Cell-Free Extracts from Suspension Cultures Each relevant enzyme activity was measured at 10 periods in the growth cycle (d 1, 4, 8, 12, 13, 14, 15, 16, 18, and 20). For each time point, a 50 mL culture was harvested and the packed cell volume determined after centrifugation at 200g. The medium was poured off and the cells were resuspended in buffer (50 mm Mes-5 mm sodium phosphate [pH 6.5] containing 200 mM sucrose, 10 mm Na2S2O5, 10 mm ascorbic acid, and 5 mm dithiothreitol). After centrifugation at 27,000g for 15 min, the supernatant was discarded (this fraction was inactive) and the cells suspended in a minimum amount of the above buffer. Cells were homogenized in a Ten-Broeck homogenizer with 100 mg of polyvinylpolypyrrolidone per gram of cells, and the homogenate slurried with 300 mg of XAD resin per gram of cells for 10 min on ice. The amounts of polyvinylpolypyrrolidone and XAD used were lower than those normally required in extracting monoterpene cyclases from intact plants (24) because cell cultures typically contain much lower levels of resins, phenolics, and monoterpenes than the intact plant. After filtration through eight layers of cheesecloth prewetted with extraction buffer, the filtrate was centrifuged again at 27,000g for 15 min. and the resulting supernatant was used as the enzyme source. Assays were performed as previously described: geranyl pyrophosphate:(+)-pinene cyclase and geranyl pyrophosphate:(-)-pinene cyclase (37, 38); 1,8-cineole cyclase (30); sabinene cyclase (38); (+)-bornyl pyrophosphate cyclase (31); (+)-borneol dehydrogenase (28); and (+)-bornyl pyrophosphate phosphohydrolase (two enzyme activities that account for the sequential hydrolysis of bornyl pyrophosphate to bor-

1 562

FALK ET AL.

neol) (32). The typical reaction mixture for the assay of cyclases (the presumptive rate-limiting enzymes of monoterpene biosynthesis [22, 33, 39]) contained 40 to 150 Ag protein in a 1 mL volume with 30 mM MgCl2 and 20 ,AM [1-3H] geranyl pyrophosphate, overlaid with 1 mL pentane in a Teflon-sealed screw-cap vial. The reaction mixture was incubated for 90 min at 3O°C and, after chilling in ice, the products were isolated by solvent extraction and purified by TLC (24). Protein levels were determined by the method of Bradford (9). Rate of Camphor Catabolism and Assay of Catabolites A preliminary examination of the ability of cell cultures to catabolize camphor under various conditions was carried out to aid in the design of an experiment to trace the metabolic fate of the U-'4C-labeled compound. In this examination, 2 mg of unlabeled camphor were inoculated aseptically into four 50 mL cultures (d 13), two of which contained 500 mg XAD resin and two which lacked this terpene adsorbent. Just prior to inoculation with camphor, one culture of each type was inactivated by autoclaving for 20 min at 121 °C and 15 psi. All cultures were incubated for 48 h at room temperature in the dark and the cultures were then chilled on ice and the XAD resin, when present, was separated from the cells and extracted twice with 3 mL of pentane:diethyl ether (2:1 [v/v]). For each culture, the cells were separated from the medium by filtration and the medium was extracted three times with 10 mL of pentane:ether. The cells were homogenized in 10 mL of water in a Ten-Broeck homogenizer and then centrifuged at 27,000g for 10 min. The supernatant was separated from the cell debris and both fractions were extracted twice with 3 mL portions of pentane:ether. An internal standard (650 nmol menthol) was added to each extract which was then concentrated to 1 mL and analyzed by capillary GLC. To examine the pathway of camphor catabolism in sage cultures, a total of 0.26 ,uCi of [U-'4C]camphor (1.1 mCi/ mol) was divided equally among 16 cultures of d 13 cells and the cultures allowed to incubate for 72 h. The cultures were then chilled at 4°C for 1 h and the medium filtered from the cells. Half of the medium (-400 mL) was frozen and the other half was extracted twice with 400 mL pentane and once with 400 mL pentane:ether (2:1 v/v). An aliquot of the combined organic extract was taken for determination of radioactivity and, following the addition of carrier standards, camphor and 1,2-campholide were separated by TLC (hexanes:ether, 1:2 [v/v]) and analyzed by radio-GLC. An aliquot of the remaining aqueous phase was also taken for determination of radioactivity, and a 1 mL aliquot was subjected to acid hydrolysis (2 N HCl, 30°C, 24 h) to give an indication (by the generation of ether-soluble radioactivity) of the presence of monoterpene glycosides in the medium. The remaining aqueous phase was retained for the analysis of glycosides and esters by enzymatic hydrolysis. The collected cells were ruptured using a Bead Beater (BioSpec Products) with the extraction chamber filled with cold distilled water. Seven 1 min pulses were applied, with the motor controlled by a rheostat set at 10 V. The resulting homogenate was placed in a flask with 1 L of CHCl3:MeOH


(2:1, v/v) and kept at 4°C for 1 week. The cellular debris was then removed by filtration and the CHCl3:MeOH extract washed with 500 mL water to afford an aqueous methanolic fraction. An aliquot of this material was taken for determination of radioactivity and the remainder combined with the original medium for the analysis of glycosides and esters by enzymatic hydrolysis as described above. The products liberated by hydrolysis were extracted into ether as before and analyzed by TLC and radio-GLC. A portion of this material was also methylated with 14% BF3 in methanol for the analysis of campholenic acids (as methyl esters) by radioGLC. The chloroform phase of the cell extract was evaporated to dryness and the residue saponified in 40 mL of 0.15 N KOH in 15% aqueous methanol on a steam bath for 1 h. The reaction mixture was cooled on ice and extracted with three 50 mL portions of diethyl ether to remove nonsaponifiable lipids (primarily phytosterols). The aqueous phase was acidified (to pH 1.0) and extracted with ether to provide the saponifiable lipids (fatty acids) which were methylated with 14% BF3-MeOH as before. The '4C-content ofthe saponifiable and nonsaponifiable lipids was determined by scintillation spectrometry. Analytical Procedures

TLC was performed on 1 mm layers of silica gel G. Developed plates were sprayed with a 0.2% ethanolic solution (w/ v) of 2,7-dichlorofluorescein and viewed under long-wave UV light to locate components which were eluted from the gel with ether. For scintillation spectrometry, 15 mL of a cocktail consisting of 0.4% (w/v) Omnifluor (New England Nuclear) dissolved in 30% ethanol in toluene was employed (3H efficiency = 40%; 14C efficiency = 96%). Capillary GLC analyses were performed on a HewlettPackard 5890A gas chromatograph with 3392 integrator using bonded-phase, fused-silica open-tubular columns (30 m x 0.25 mm i.d.) coated with either a 0.2 um film of SuperoxFA or a 1 ,um film of RSL- 150 (Alltech Associates), and operated using H2 as carrier (2 mL/min) and RD2 (250°C) with on-column or split injection modes. For borneol dehydrogenase assays, the Superox FA column was programmed from 45°C (5 min hold) at 10°C/min to 220°C. For the analysis of monoterpene accumulation, the RSL-150 column was programmed from 70°C (5 min hold) at 10°C/min to 250°C. For the analysis of camphor catabolites, the Superox FA column was programmed from 50°C (5 min hold) at 10°C/ min to 220°C. Radio-GLC was performed on a GOW-MAC 550P gas chromatograph (TCD, He flow rate of 45 mL/min) attached to a Nuclear Chicago 7357 gas proportional counter. Both thermal conductivity and radioactivity output channels were monitored with a SICA 7000A chromatogram processor, and the system was externally calibrated with [3H]toluene or [14C] toluene. For the analysis of 3H-labeled monoterpene olefins, the chromatographic column was 12 feet x 0.125 inch o.d. 2Abbreviations: FID, flame ionization detector; TCD, thermal conductivity detector.


stainless steel containing 15% Silar lOC on 80/100 mesh Chromosorb WHP and was programmed from 70°C (15 min hold) at 5°C/min to 1 10°C. For the analysis of 3H-labeled oxygenated monoterpenes, the column used was 12 feet x 0.125 inch o.d. stainless steel containing 15% AT-1000 on Gas-Chrom Q and was programmed from 1 30°C (5 min hold) at 5°C/min to 180°C. For the analysis of [U-'4C]camphor, 1,2-campholide and related catabolites, the column used was 12 feet x 0.125 inch o.d. stainless steel containing 15% SE30 on Chromosorb WHP and was programmed from 90°C (10 min hold) at 3°C/min to 1 30°C. RESULTS AND DISCUSSION Monoterpene Accumulation in Suspension Cultures

Monoterpene production was examined in suspension cultures of Salvia officinalis that had been generated from callus initiated from leaf tissue. Preliminary experiments indicated that a lipophilic organic phase in the suspension medium, like that employed by Berlin and Witte (8), would be necessary to trap monoterpenes synthesized by the culture, especially the more volatile olefins. After unsuccessful trials with mineral oil, Miglylol (a mixture of triglycerides), and various gas chromatographic stationary phases, it was found that Amberlite XAD-4, a beaded polystyrene resin, was very efficient at trapping exogenously applied monoterpenes while giving a low background of extractable contaminants when analyzed by gas chromatography, and that this material was only slightly inhibitory to cell growth at a concentration of 1% (w/ v). The low density resin beads were also easy to separate from the cells and medium, since they could be removed by flotation after low speed centrifugation. The recoveries of 3H-labeled monoterpenes (1.95 jig geraniol or 1.40 Ag mixed olefins) added to culture flasks containing XAD resin were evaluated after a 14-d incubation period. Trials were conducted using both active cultures and flasks containing medium only to evaluate the effect of living cells on recovery, and in either case the recovery was negligible in the absence of the resin. Solvent extraction of the resin recovered 85% of the geraniol added to the culture without cells (with negligible levels in the medium), whereas only 15% of this monoterpene was recovered from the resin in the culture with cells (with 10% of the initial radioactivity recovered in the medium; most of which was not extractable in organic solvent). This result suggests that the cells played a role in the disappearance of geraniol, possibly by transformation to more volatile substances, or to water-soluble materials. Evidence indicates that plant cells in culture secrete into the medium high levels of hydrolytic and oxidative enzymes which are capable of degrading primary and secondary metabolites (51). The recovery of labeled monoterpene olefins (a mixture of a-pinene, f-pinene, camphene, myrcene, limonene, and sabinene) from the culture without cells was 64% (most was bound to the XAD resin with negligible amounts remaining in the medium), whereas in the presence of cells, 58% of the olefins were recovered from the XAD resin (with 10% of the initial radioactivity remaining in the medium, most of which was extractable in organic solvent).

1 563

The apparent lack of metabolic transformation of the olefins may be a consequence of the fact that these compounds, being more hydrophobic than geraniol, are more favorably partitioned into the polystyrene resin, and are therefore less accessible to degradative enzymes. In general, the lower recovery of monoterpene olefins compared to geraniol in the cultures without cells may be attributed to the higher volatility of these compounds relative to geraniol. For the purpose of evaluating the production of monoterpenes in sage cultures, these data allow prediction that, in the presence of XAD resin, approximately 85% of the geraniol (and other oxygenated monoterpenes) and roughly 65% of the monoterpene olefins synthesized in the culture can in theory be recovered, in the absence of cellular catabolism. With the anticipated recoveries as a guide, suspension cultures containing XAD resin were harvested periodically throughout a growth cycle of 21 d and examined for the presence of endogenous monoterpenes by capillary-GLC analysis of pentane:ether extracts of the resin trap. No measurable amounts of monoterpenes were found at any day in the growth cycle. The use of internal standards showed that the limits of detection were 3 ng of monoterpene product per 50 mL culture. If cultures were synthesizing monoterpenes at a level comparable to that of leaves on the intact plant, approximately 150 mg of product would be expected to accumulate per 50 mL culture, given that the monoterpene content of sage leaves on a fresh weight basis is usually 1.5% (29) and that stationary phase cultures had a wet weight of about 10 g. An examination of glycosidically-bound or esterified monoterpenes also failed to detect accumulation at greater than 35 Ag per culture, which was the limit of detection of this method based on enzymatic hydrolysis. Monoterpene Biosynthetic Capacity in Culture: Synthesis from [U-14C]Sucrose

Despite the negligible recovery of monoterpenes from sage cell cultures, the fact that cells could degrade a significant proportion of added monoterpenes suggested that the biosynthesis of these products might take place without net accumulation. A culture was therefore supplied with 0.45 ,uCi of [U-'4C]sucrose (700 pmol) during the period of peak monoterpene biosynthetic activity in early stationary phase (as determined by in vitro assay; see below). The production of labeled camphor was examined since this monoterpene ketone is one of the principal products of the intact plant (31). Because of the potential for monoterpene degradation in culture, unlabeled camphor (1.2 mg) was also added in an attempt to trap the labeled biosynthetic product. No XAD resin was employed in this experiment. After a 13 h incubation period, which straddled the peak of biosynthetic activity, 74% of the unlabeled camphor was recovered, but radio-TLC analysis showed only 62 pCi (-0. 1 pmol, based on the specific activity of the starting material) of labeled camphor to be present. Although the results of this experiment suggest that the level of camphor biosynthetic activity in vivo is extremely low, a higher rate of camphor biosynthesis might have been obscured either by dilution of the precursor (since it is unlikely that unlabeled sucrose originally present in the medium had

1 564


FALK ET AL.

been fully depleted) or by very rapid catabolism of the monoterpene product (since exogenously applied camphor may not have fully equilibrated with that generated endogenously). Monoterpene Biosynthetic Capacity in Culture: In Vitro Measurement of Enzyme Activities In an attempt to determine if monoterpene biosynthesis was occurring at a significant rate in cultured cells, the activities of several enzymes of monoterpene biosynthesis were measured, including all of those required for the formation of camphor from the ubiquitous precursor geranyl pyrophosphate (28, 31, 32) (Fig. 1). Cell-free extracts were prepared from cultures at several points in the growth cycle, and the ability to cyclize geranyl pyrophosphate to monoterpene olefins, 1,8-cineole, and bornyl pyrophosphate (Fig. 1) was assayed. The cyclization of geranyl pyrophosphate represents the first committed reaction leading to monoterpenes, and this enzymatic transformation is considered to be a regulatory step in monoterpene biosynthesis (22, 39). Figure 2A depicts the growth of sage suspension cultures over a 2 l-d period as measured by the packed cell volume of the entire culture after a low speed centrifugation. This overall pattern of growth was confirmed by both fresh weight and dry weight measurements of the cultures made at the same time points. The stationary phase, which is often found to be the stage during which the most active synthesis of secondary metabolites occurs in cell culture (1, 18, 43, 50), is reached at d 13. The time courses of enzyme activity are illustrated in Figure 2 for monoterpene olefin cyclases (38), 1,8-cineole cyclase (30), and bornyl pyrophosphate cyclase (31). Cyclase activity of all types was virtually absent throughout most of the cell culture growth cycle, except for a brief period (48 h) near the beginning of stationary phase at which time these enzymes showed a pronounced increase in activity (on either a per culture or per g fresh weight basis). The maximum activity observed for each of these cyclases was considerably lower than that noted in intact plant tissue. For example, bornyl pyrophosphate cyclase, which had the highest activity noted for any cyclase in sage cultures, registered a peak activity of 13 fmol/s per g fresh weight of cells, approximately 5% of the level noted for young, expanding sage leaves (33, 40). Radio-GLC analysis of the products generated by the monoterpene olefin cyclases indicated the presence of ,3-pinene (50%), myrcene (36%), and terpinolene (13%). The appearance of a measurable quantity of terpinolene is surprising, considering that this olefin is normally a trace component of the olefin mixture produced by the intact plant. Sabinene, camphene, and a-pinene, which are normal components of sage essential oil (38), were not detected. This distribution of olefinic products underscores a phenomenon previously observed in cell cultures of monoterpene producing species: most cultures do not synthesize the same mixture of monoterpenes as that found in the intact plant ( 18, 46, 52). Activities of the subsequent steps in the biosynthesis of camphor following formation of bornyl pyrophosphate (Fig. 1) were also examined by in vitro assay. Bornyl pyrophosphate phosphohydrolase activity (the summation of two hydrolase activities leading to borneol (32]) was consistently higher than cyclase activity throughout the growth curve. This enzyme

activity peaks at a maximum of 140 pmol/s per culture at d 13, corresponding to a level of 14 pmol/s per g fresh weight, which is about the same as that observed in extracts of the intact plant (32, 33). Borneol dehydrogenase catalyzes the final step of camphor biosynthesis (Fig. 1), and the activity of this enzyme peaks at nearly 6 pmol/s per culture at d 13, corresponding to a level of 0.6 pmol/s per g fresh weight, which is about half the level of the intact plant (28, 33). The maximum activity per culture of the dehydrogenase coincides with that of the bornyl pyrophosphate cyclase and bornyl pyrophosphate hydrolases. Since the dehydrogenase and phosphohydrolases are present at all stages of culture development, plots of activity on a per g fresh weight basis exhibit less variation than do plots on a per culture basis. Nevertheless, a peak of activity per g was also noted in both cases at d 13 (i.e., the period when the cyclases are present). It is interesting that all of the enzymes of monoterpene

120

E

-20

90

E

C _

0

60

0

toa)

._..

0

10

E 30

0C

N

w

CO

%-

_0 a)

E

N

c

w

Days Figure 2. Growth curve and in vitro measurement of the levels of monoterpene biosynthetic enzymes in sage suspension cultures. Packed cell volume (0), monoterpene olefin synthase (cyclase) activity (0), and 1,8-cineole cyclase activity (A) are plotted in panel A. Bomyl pyrophosphate cyclase activity x 10° (0), bornyl pyrophosphate phosphohydrolase activity x 10-3 (U) and borneol dehydrogenase activity x 10-2 (A) are plotted in panel B. The enzyme assays are described in "Materials and Methods." 1 Unit = fmol/s -culture.


biosynthesis studied in sage cells exhibit a coordinately regulated burst of activity in the culture near the end of the logarithmic phase of growth and the beginning of stationary phase. This pattern is frequently observed in enzymological studies of natural product metabolism in cell culture and is believed to be a function of the depletion of some essential nutrient from the culture medium (1, 17, 44). The relative activities of the enzymes of camphor biosynthesis measured are consistent with the cyclization of geranyl pyrophosphate being the rate-limiting step of this pathway (33). In cell culture, the maximum cyclase activity is considerably less than the activities of the other two enzymes (2% of the dehydrogenase and 0.08% of the phosphohydrolase, compared with 40% and 3%, respectively, in the intact plant), suggesting that low cyclase activity may be an important constraint on monoterpene biosynthesis in culture. Monoterpene Catabolic Capacity in Culture: Rate and Pathway of Camphor Catabolism Although the level of monoterpene biosynthesis in culture was low as judged by in vivo and in vitro measurements, calculation based on the levels of cyclases measured in vitro indicated that about 50 nmol of monoterpenes (22 nmol of camphor) would have been produced in a single culture in the 2 d during which the cyclases were most active. Since this level of product would have been easily detected, the lack of monoterpene accumulation actually observed might be due to catabolic processes. To assess the extent of catabolism in culture, a series of camphor feeding experiments were conducted. In the first experiment, 2 mg of camphor were administered to both live cells and heat inactivated cells. After a 48 h incubation period, 1.06 mg of the original camphor was recovered from the flask of inactivated cells, whereas only 0.56 mg of camphor was recovered from the flask of live cells. These data suggest that the loss of camphor due to volatilization is about 50%, and that sage cell cultures are capable of degrading about half of the remaining camphor (i.e., about 0.5 mg in 48 h). Given the extent of camphor loss to catabolic processes, it was of interest to examine the pathway of camphor degradation. For this purpose, 0.26 ,OCi of [U-'4C]camphor (0.24 mmol) was distributed among 16 cultures at d 13. After incubation for 72 h, only 0.01 gCi (-4%) of the initial radioactivity was recovered as camphor and it was determined that 0.12 gCi (46%) of the camphor applied was degraded, the remainder having been lost by volatilization. Based on the rate of camphor loss, it can be estimated that catabolic capacity exceeds the maximum cyclase activity in sage cultures by at least two orders of magnitude. A pathway for the degradation of camphor, via 1,2-campholide and the corresponding glucoside-glucose ester (Fig. 1), has been described in the intact sage plant (25, 26). However, in the present experiment, neither of these two intermediates were detected as metabolites of [U-_4C]camphor in extracts of the medium or the cells, nor were chemical degradation products of these metabolites (such as campholenic acids [26]) observed. Thus, if 1,2-campholide or its conjugates are intermediates in the degradation of camphor in culture, they are very rapidly turned over.

1 565

The greatest amount of label from exogenous ['4C]camphor (46%) was recovered as water-soluble components of the medium and cells, from which label was not appreciably liberated by ,B-glucosidase and esterase hydrolysis. No attempt was made to identify these labeled products, but it is likely that they represent a wide range of cellular metabolites. Acid hydrolysis of the total water-solubles from medium and cells released 20% of the '4C-label as unidentified ether-soluble constituents. Small amounts of the total radioactivity applied to the cells as camphor were recovered in phytosterols (0.3%) and fatty acids (1%). There are significant differences between the pathway of camphor degradation previously demonstrated in the intact plant and that in cell culture. As mentioned, 1,2-campholide was not detected in culture, nor was the corresponding glucoside-glucose ester. Since sucrose in the medium is nearly depleted at this point of the growth cycle, degradation of the applied camphor may not necessarily proceed through glycosylated intermediates as in the intact plant. The glucosideglucose ester serves as a phloem transport derivative between the site of monoterpene accumulation in leaves and the site of catabolism in the roots (25, 26). Transport to a remote site for catabolism seems unnecessary in culture, and it appears that camphor may be degraded directly (probably via 1,2campholide to accomplish ring cleavage) to basic metabolites without the intermediacy of glycoconjugates. In the intact plant, the ultimate products of camphor degradation are acyl and isoprenoid lipids (25). The lack of significant labeling of these compounds in culture likely indicates that these stationary phase cells are using camphor as a source of energy rather than as a carbon source for synthesis of new membrane constituents.

CONCLUSION In this investigation, we have shown that undifferentiated cell suspension cultures of sage exhibit no measurable accumulation of either free monoterpenes or conjugated forms. In theory, the lack of observable accumulation and the low apparent rate of monoterpene production from [U-'4C]sucrose could be due either to the absence of significant biosynthetic activity or to the presence of efficient catabolic processes. Studies with cell-free extracts of cultures indicated that several enzymes of monoterpene biosynthesis are present at activity levels comparable to those measured in the intact plant, although the cyclases, which are often thought to catalyze the rate-limiting step of the pathway (22, 33, 39), are present at significantly lower levels than those in the intact plant. Nevertheless, sufficient amounts of enzyme activity appear to be present in culture to produce readily detectable levels of monoterpenes. The lack of observable accumulation thus indicates that these suspension cultures must readily degrade monoterpenes, and the efficient degradation of exogenous camphor to water-soluble metabolites was, in fact, demonstrated. Catabolism of exogenous monoterpenes has been shown in cell cultures of a variety of other species (7, 15, 20). Undifferentiated cell cultures lack organized structures for the extracellular storage of monoterpenes, such as resin ducts or the subcuticular space of glandular trichomes. Monoter-

FALK ET AL.

1 566

penes that are secreted into the medium would appear to be much more susceptible to enzymatic degradation than those sequestered in extracellular compartments since, in culture, plant cells typically excrete large amounts of hydrolytic and oxidative enzymes into the medium (51). If cells are unable to store monoterpenes in discrete structures, both extra- and intracellular degradation may, in fact, be critically important in order to avoid the toxic effects of monoterpenes on growth and viability (14, 36). In intact sage plants, catabolism has been shown to represent a mechanism for the salvage of carbon from monoterpene defense compounds in older leaves (23). In culture, catabolism may result from the need to detoxify monoterpenes and provide substrate for cell growth, or could simply be a consequence of the greater accessibility of monoterpenes to catabolic enzymes in undifferentiated cells. ACKNOWLEDGMENTS We thank Margaret Duffy-Riggle, Henry Fisk, and D. Michael Satterwhite for technical assistance, Greg Wichelns for raising the plants, and Karen Maertens for typing the manuscript. LITERATURE CITED 1. Banthorpe DV (1988) Monoterpenes and sesquiterpenes. In F Constabel, IK Vasil, eds, Cell Culture and Somatic Cell Genetics, Vol 5, Phytochemicals in Plant Cell Cultures. Academic Press, San Diego, CA, 143-157 2. Banthorpe DV, Barrow SE (1983) Monoterpene biosynthesis in extracts from cultures of Rosa damascena. Phytochemistry 22: 2727-2728 3. Banthorpe DV, Branch SA, Njar VCO, Osborne MG, Watson DG (1986) Ability of plant callus cultures to synthesize and accumulate lower terpenoids. Phytochemistry 25: 629-636 4. Banthorpe DV, Greg TJ, Poots I, Fordham WD (1986) Monoterpene metabolism in cultures of Rosa species. Phytochemistry 25: 2321-2326 5. Banthorpe DV, Njar VCO (1984) Light-dependent monoterpene synthesis in Pinus radiata cultures. Phytochemistry 23: 295299 6. Becker H (1970) Studies on the formation of volatile substances in plant tissue cultures. Biochem Physiol Pflanzen 161: 425441 7. Berger RG, Drawert F (1988) Glycosylation of terpenols and aromatic alcohols by cell suspension cultures of peppermint (Mentha piperita L.). Z Naturforsch 43c: 485-490 8. Berlin J, Witte L (1988) Formation of mono- and diterpenoids by cultured cells of Thuja occidentalis. Phytochemistry 27: 127-132 9. Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Chem 72: 248-254 10. Bricout J, Garcia-Rodriquez M-J, Paupardin C, Saussay R (1978) Biosynthese de composes monoterpeniques par les tissus de quelques especes de Menthes cultivees in vitro. C R Acad Sci (Paris) Ser D 287: 611-613 11. Bricout J, Paupardin C (1975) Sur la composition del'huile essentielle de Mentha piperita L. cultivee in vitro: influence de quelques facteurs sur sa synthese. C R Acad Sci (Paris) Ser D 281: 383-386 12. Brindle PA, Kuhn PJ, Threlfall DR (1983) Accumulation of phytoalexins in potato-cell suspension cultures. Phytochemistry 22: 2719-2721 13. Brooks CJW, Watson DG, Freer IM (1986) Elicitation of capsidiol accumulation in suspended callus cultures of Capsicum annuum. Phytochemistry 25: 1089-1092


14. Brown JT, Hegarty PK, Charlwood BV (1987) The toxicity of monoterpenes to plant cell cultures. Plant Sci 48: 195-201 15. Carriere F, Gil G, Tapie P, Chagvardieff P (1989) Biotransformation of geraniol by photoautotrophic, photomixotrophic and heterotrophic plant cell suspensions. Phytochemistry 28: 1087-1090 16. Chappell J, Nable R, Fleming P, Anderson RA, Burton HR (1987) Accumulation of capsidiol in tobacco cell cultures treated with fungal elicitor. Phytochemistry 26: 2259-2260 17. Charlwood BV, Brown JT, Moustou C, Morris GS, Charlwood KA (1988) The accumulation of isoprenoid flavour compounds in plant cell cultures. In P Schreier, ed, Bioflavour '87: Analysis, Biochemistry, Biotechnology. Walter de Gruyter, Berlin, pp 303-314 18. Charlwood BV, Charlwood KA (1983) The biosynthesis of monoand sesquiterpenes in tissue culture. Biochem Soc Trans 11: 592-593 19. Charlwood BV, Hegarty PK, Charlwood KA (1986) The synthesis and biotransformation of monoterpenes by plant cells in culture. In P Morris, A Scragg, A Stafford, M Fowler, eds, Secondary Metabolism in Plant Cell Cultures. Cambridge University Press, Cambridge, pp 15-34 20. Cormier F, Ambid C (1987) Extractive bioconversion of geraniol by a Vitis vinifera cell suspension employing a two-phase system. Plant Cell Rep 6: 427-430 21. Croteau R (1977) Site of monoterpene biosynthesis in Majorana hortensis leaves. Plant Physiol 59: 519-520 22. Croteau R (1987) Biosynthesis and catabolism of monoterpenoids. Chem Rev 87: 929-954 23. Croteau R (1988) Catabolism of monoterpenes in essential oil plants. In BM Lawrence, BD Mookherjee, BJ Willis, eds, Flavors and Fragrances: A World Perspective. Elsevier, Amsterdam, pp 65-84 24. Croteau R, Cane DE (1985) Monoterpene and sesquiterpene cyclases. Methods Enzymol 110: 383-405 25. Croteau R, El-Bialy H, DehalSS (1987) Metabolism of monoterpenes: metabolic fate of(+)-camphor in sage (Salvia officinalis). Plant Physiol 84: 649-653 26. Croteau R, El-Bialy H, El-Hindawi S (1984) Metabolism of monoterpenes: lactonization of (+)-camphor and conversion of the corresponding hydroxy acid to the glucoside-glucose ester in sage (Salvia officinalis). Arch Biochem Biophys 228: 667-680 27. Croteau R, Hooper CL (1978) Metabolism of monoterpenes: acetylation of (-)-menthol by a soluble enzyme preparation from peppermint (Mentha piperita) leaves. Plant Physiol 61: 737-742 28. Croteau R, Hooper CL, Felton M (1978) Biosynthesis of monoterpenes: partial purification and characterization of a bicyclic monoterpenol dehydrogenase from sage (Salvia officinalis). Arch Biochem Biophys 188: 182-193 29. Croteau R, Karp F (1976) Biosynthesis of monoterpenes: enzymatic conversion of neryl pyrophosphate to 1,8-cineole, aterpineol, and cyclic monoterpene hydrocarbons by a cell-free preparation from sage (Salvia officinalis). Arch Biochem Biophys176: 734-736 30. Croteau R, Karp F (1977) Biosynthesis of monoterpenes: partial purification and characterization of 1,8-cineole synthetase from sage (Salvia officinalis). Arch Biochem Biophys 179: 257265 31. Croteau R, Karp F (1979) Biosynthesis of monoterpenes: preliminary characterization of bomyl pyrophosphate synthetase from sage (Salvia officinalis) and demonstration that geranyl pyrophosphate is the preferred substrate for cyclization. Arch Biochem Biophys 198: 512-522 32. Croteau R, Karp F (1979) Biosynthesis of monoterpenes: hydrolysis of bomyl pyrophosphate, an essential step in camphor biosynthesis, and hydrolysis of geranyl pyrophosphate, the acyclic precursor of camphor, by enzymes from sage (Salvia oficinalis). Arch Biochem Biophys 198: 523-532 33. Croteau R, Karp F, Felton M, Kjonaas R(1981) Relationship of


34. 35. 36.

37.

38.

39. 40.

41.

42.

camphor biosynthesis to leaf development in sage (Salvia officinalis). Plant Physiol 67: 820-824 Downing MR, Mitchell ED (1975) Mevalonate-activating enzymes in callus culture cells from Nepeta cataria. Phytochemistry 14: 369-371 Fahn A (1979) Secretory Tissues in Plants. Academic Press, London Fischer NH (1986) The function of mono- and sesquiterpenes as plant germination and growth regulators. In AR Putnam, C-S Tang, eds, The Science of Allelopathy. John Wiley and Sons, New York, pp 203-218 Gambliel H, Croteau R (1982) Biosynthesis of (±)-a-pinene and (-)-fl-pinene from geranyl pyrophosphate by a soluble enzyme system from sage (Salvia officinalis). J Biol Chem 257: 23362342 Gambliel H, Croteau R (1984) Pinene cyclases I and II: Two enzymes from sage (Salvia officinalis) which catalyze stereospecific cyclizations of geranyl pyrophosphate to monoterpene olefins of opposite configuration. J Biol Chem 259: 740-748 Gershenzon J, Croteau R (1990) Regulation of monoterpene biosynthesis in higher plants. Recent Adv Phytochem 24: (in press) Gershenzon J, Duffy MA, Karp F, Croteau R (1987) Mechanized techniques for the selective extraction of enzymes from plant epidermal glands. Anal Biochem 163: 159-164 Gershenzon J, Maffei M, Croteau R (1989) Biochemical and histochemical localization of monoterpene biosynthesis in the glandular trichomes of spearmint (Mentha spicata). Plant Physiol 89: 1351-1357 Kutney JP, Hewitt GM, Kurihara T, Salisbury PJ, Sindelar RD, Stuart KL, Townsley PM, Chalmers WT, Jacoli GG (1981) Cytotoxic diterpenes triptolide, tripdiolide, and cytotoxic tri-

43. 44.

45.

46. 47.

48. 49.

50.

51.

51.

1 567

terpenes from tissue cultures of Tripterygium wilfordii. Can J Chem 59: 2677-2683 Miyasaka H, Nasu M, Yamamoto T, Yoneda K (1985) Production of ferruginol by cell suspension cultures of Salvia miltiorrhiza. Phytochemistry 24: 193 1-1933 Mulder-Krieger T, Verpoorte R, Baerheim Svendsen A, Scheffer JJC (1988) Production of essential oils and flavours in plant cell and tissue cultures. A review. Plant Cell Tissue Organ Cult 13: 85-154 Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15: 473-497 Reichling VJ, Becker H (1976) Calluskulturen von Matricaria chamomilla: Mitteilung: Anlage und Wachstumsverhalten der Calluskultur und erste phytochemische Untersuchungen. Planta Med 30: 258-268 Sakata I, Iwamura H (1979) Synthesis and properties of menthyl glycosides. Agric Biol Chem 43: 307-312 Schwab W, Schreier P (1988) Simultaneous enzyme catalysis extraction: a versatile technique for the study of flavor precursors. J Agric Food Chem 1238-1242 Tanaka S, Yamaura T, Tabata M (1988) Localization and photoregulation of monoterpenoid biosynthesis in thyme seedlings. In P Schreier, ed, Bioflavour '87: Analysis, Biochemistry, Biotechnology. Walter de Gruyter, Berlin, pp 237-241 Whitehead IM, Threllfall DR, Ewing DF (1987) cis-9,0O-Dihydrocapsenone: a possible catabolite of capsidiol from cell suspension cultures of Capsicum annuum. Phytochemistry 26: 1367-1369 Wink M (1984) Evidence for an extracellular lytic compartment of plant cell suspension cultures: the cell culture medium. Naturwissenschaften 71: 635-637 Witte L, Berlin J, Wray V, Schubert W, Kohl W, Hofle G, Hammer J (1983) Mono- and diterpenes from cell cultures of Thuja occidentalis. Planta Med 49: 216-221