Planta (1986)167:19(~205
P l a n t a 9 Springer-Verlag 1986
Tissue-distribution of secondary phenolic biosynthesis in developing primary leaves of Avena sativa L. W. Knogge* and G. Weissenb6ck Botanisches Institut der Universiffitzu K61n, GyrhofstraBe 15, D-5000 K61n 41, Federal Republic of Germany
Abstract. Primary leaves of oats ( A r e n a sativa L.) have been used to study the integration of secondary phenolic metabolism into organ differentiation and development. In particular, the tissue-specific distribution of products and enzymes involved in their biosynthesis has been investigated. C-Glucosylflavones along with minor amounts of hydroxycinnamic-acid esters constitute the soluble phenolic compounds in these leaves. In addition, considerable amounts of insoluble products such as lignin and wall-bound ferulic-acid esters are formed. The tissue-specific activities of seven enzymes were determined in different stages of leaf growth. The rate-limiting enzyme of flavonoid biosynthesis in this system, chalcone synthase, together with chalcone isomerase (EC 5.5.1.6) and the terminal enzymes of the vitexin and isovitexin branches of the pathway (a flavonoid O-methyltransferase and an isovitexin arabinosyltransferase) are located in the leaf mesophyll. Since the flavonoids accumulate predominantly (up to 70%) in both epidermal layers, an intercellular transport of products is postulated. In contrast to the flavonoid enzymes, Lphenylalanine ammonia-lyase (EC 4.3.1.5), 4-coumarate: CoA ligase (EC 6.2.1.12), and S-adenosylL-methionine: caffeate 3-O-methyltransferase (EC 2.1.1.-), all involved in general phenylpropanoid metabolism, showed highest activities in the basal leaf region as well as in the epidermis and the vascular bundles. We suggest that these latter * Present address: Max-Planck-Institut ffir Zfichtungsfor-
schung, D-5000 K61n 30, Federal Republic of Germany Abbreviations: CHI = chalcone isomerase; CHS = chalcone synthase; 4CL = 4-coumarate :CoA ligase; CMT = S-adenosyl-gmethionine:caffeate 3-O-methyltransferase; FMT=S-adenosyl-L-methionine:vitexin 2"-O-rhamnoside 7-O-methyltransferase; HPLC = high-performance liquid chromatography; IAT = uridine 5'-diphosphate L-arabinose:isovitexin 2"-O-arabinosyttransferase; PAL = L-phenylalanine ammonia-lyase;
enzymes participate mainly in the biosynthesis of non-flavonoid phenolic products, such as lignin in the xylem tissue and wall-bound hydroxycinnamic acid-esters in epidermal, phloem, and sclerenchyma tissues.
Key words: A v e n a C-Glucosylflavone - Flavonoid biosynthesis - Phenylpropanoid metabolism.
Introduction In definite ontogenetic stages, higher plants are capable of synthesizing an array of secondary products, which are usually restricted to certain organs, tissues, or even single cells (Wiermann 1981). Primary leaves of grass seedlings such as oats and rye provide suitable experimental systems to study secondary phenolic metabolism and its integration into organ differentiation and development (Weissenb6ck and Effertz 1974; Strack et al. 1982). Being essentially equifacial these leaves are composed of three major types of tissue: the autotrophic mesophyll and the heterotrophic epidermal layers and vascular bundles. Flavonoids represent the main soluble phenolic compounds in oat primary leaves (Weissenb6ck and Effertz 1974). In addition, extracellular insoluble materials, lignin and wall-bound esters 1 of hydroxycinnamic acids, are formed (Harris and Hartley 1976). Flavonoid biosynthesis results in the accumulation of three major C-glucosylflavones: an isovitexin arabinoside (F2) and two structurally rei The term wall-bound comprises phenolic acids bound to the carbohydrate moiety of cell walls (Hartley et al. 1976; Fry 1983) as well as to cutin in the leaf cuticle and to suberin described to be present in the walls of grass bundle-sheath cells (Kolattukudy 1980, 1981). In the present paper it was not attempted to discriminate betweenthese fractions
W. Knogge and G. Weissenb6ck : Localization of secondary phenolic metabolism in oat leaf tissues
lated vitexin rhamnosides (F1, F3; Fig. 5) (Chopin et al. 1977). Depending on leaf age and the type of tissue, flavonoid patterns show distinct quantitative and qualitative differences (Effertz and Weissenb6ck 1980). The two epidermal layers always contain 60-70% of total leaf flavonoids. All compounds are metabolic end products which are not degraded (Effertz and Weissenb6ck 1978; Proksch et al. 1981). Two models can explain the formation of stageand tissue-dependence of flavone patterns in the leaf. First, each tissue is biosynthetically autonomous; thus, product accumulation reflects the distribution of the respective biosynthetic enzymes. Alternatively, biosynthesis occurs in one type of tissue; consequently, products are translocated to other tissues. Our previous investigations on the tissue localization of the initial enzymes of flavonoid biosynthesis, chalcone synthase (CHS) and chalcone isomerase (CHI), had been restricted to that growth stage of plants showing the most rapid flavone accumulation (Fuisting and Weissenb6ck 1980; Weissenb6ck and Sachs 1977). In the present paper we report on the localization of seven enzymes of phenolic metabolism in leaf tissues during organogenesis. The enzymes analyzed were L-phenylalanine ammonia-lyase (PAL), 4-coumarate: CoA ligase (4CL) (Knogge et al. 1981), and caffeate O-methyltransferase (CMT), all involved in general phenylpropanoid metabolism, as well as CHS and CHI and the final enzymes of the vitexin and isovitexin branches of the pathway. These final enzymes included a flavonoid O-methyltransferase (FMT) which specifically catalyzes the transfer of a methyl group to the 7-hydroxyl group of vitexin rhamnoside F3 (Knogge and Weissenb6ck 1984) and an isovitexin O-arabinosyltransferase (IAT) showing high specificity for the flavonoid substrate (Fuisting 1981). Additional results are presented concerning the accumulation of non-flavonoid phenolic products in oat primary leaf tissues. Material and methods Plant material. Oat plants (Arena sativa L. cv. F1/imingskrone; F. von Lochow-Petkus GmbH, Bergen, FRG) were grown under standard conditions in a phytotron under a regime of 7500 lx for 13 h per day (Weissenb6ck and Effertz 1974). For studying tissue-specific enzyme activities as a function of seedling development, 4- to 8-d-old primary leaves were harvested at the end of the dark phase. Chemicals. Polyclar AT, DOWEX AG 1X2 (200-400 mesh), and dithiothreitol (DTT) were obtained from Serva (Heidelberg, FRG), all biochemicals from Boehringer (Mannheim, FRG), all laboratory chemicals (analytical grade) from Merck
197
(Darmstadt, FRG). Cinnamic acids were purchased from Roth (Karlsruhe, FRG). Flavonoids were extracted and purified according to standard methods of this laboratory using polyamide and Sephadex LH-20 column chromatography (Popovici et al. 1977) as well as high-performance liquid chromatography (HPLC). 1-O-4-Coumaroylglucose and 1-O-feruloylglucose from Antirrhinum petals (Harborne and Corner 1961) were kind gifts of Dr. D. Strack (Botanical Institute, University of Cologne). Cellulysin (Triehoderma viride, 10000 units g 1) was purchased from Calbiochem-Behring (La Jolla, Cal., USA). [2-14C] Malonyl CoA (1.56~1.77 GBq mmol 1), uridine 5"-diphosphate-L-[U-14C]arabinose (6.76 GBq mmol 1), and S-adenosyl-L-(methyl-tgC)methionine (1.94 GBq mmo1-1) were obtained from New England Nuclear (Dreieichenhain, FRG), [3l~C]cinnamic acid (1.66-2.03 GBq mmol ~) from CEA (Gifsur-Yvette), France). 4-Coumaroyl CoA and 2',4,4',6"-tetrahydroxychalcone were synthesized according to St6ckigt and Zenk (1975) and Moustafa and Wong (1967), respectively.
Preparation of tissue fractions. Preparation of epidermal peels and mesophyll protoplasts has been described previously (Weissenb6ck and Sachs 1977; Haas etal. 1979). The vascular strands resistant to cellulolytic digestion were collected and washed twice with protoplast isolation medium. This method did not allow tissue separation in the basal region of the leaves. This part was defined by the break point of the epidermal peels and comprised one (4-d-old leaves) to 2 cm (8-d-old leaves). It was analyzed, therefore, as a whole. For routine tests of enzyme distribution the leaves were separated into epidermal peels and the corresponding leaf complements (compare Weissenb6ck and Sachs 1977). Extraction of enzymes. Tissue fractions from 40 leaves were ground in a mortar at 0 4 ~ C in the presence of sufficient quartz sand, DOWEX, Polyclar AT and buffer (see below). The homogenates were squeezed through Miracloth (Chicopee Mills, Milltown, N.J., USA) and centrifuged for 10 rain at 48000 g. The clear supernatants served as enzyme sources. Extraction procedures were optimized for the individual enzymes, and different extraction buffers were, therefore, used: PAL, 0A M 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1 or 0.1M glycine-NaOH pH 8.5 [5mM 2-mercaptoethanol (2-ME)I; 4CL, 0.1 M K-phosphate pH 7.3 (40 mM dithiothreitol); CMT and FMT, 0.1 M K-phosphate pH 7.5 (5 mM ethylenediaminetetraacetate, 10 mM 2-ME); CHS and CHI, 0.1 M K-phosphate pH 8.0 (4.3 mM 2-ME); IAT, 0.1 M K-phosphate pH 7.5 (8 mM 2-ME; Fuisting 1981). In the case of CHS, before centrifugation the filtrate was again treated with DOWEX in a batch procedure followed by filtration through Miracloth. Enzyme assays. All enzyme assays were carried out at 30~ C; PAL, 4CL, and CHI were measured photometrically; for all other enzymes, radioactive assays were used. Assay conditions have been optimized starting from published methods as indi'cated. Assay mixtures were as follows PAL, 0.1 M Tris-HC1 pH 8.5, 5 mM 2-ME, 5 mM L-phenylalanine and 50-100 gl of enzyme extract in a total volume of I ml (Weissenb6ck 1975); 4CL: 50 mM K-phosphate pH 7.3, 2 mM dithiothreitol, 1 mM ATP, 10 mM MgC12, 0.2 mM CoA, 2 mM 4-coumaric acid and 10-50 gl of enzyme extract in a total volume of 0.5 ml (Gross and Zenk 1974); CHI, 0.1 M K-phosphate pH 8.0, 75 ~tM 2',4,4"6'-tetrahydroxychalcone, 50 mM KCN and 10 20 gl of enzyme extract in a total volume of 0.5 ml (Boland and Wong 1975); CMT, 0.1 M K-phosphate pH 7.5, 1 mM caffeic acid, 1.5 mM MgClz, 20 mM Na-ascorbate, 0.17 mM S-adenosyl-L(methyl-14C)methionine (37 MBq mmol 1), and 10 50 gl of enzyme extract in a total volume of 0.2 ml (Poulton and Butt
198
W. Knogge and G. Weissenb6ck: Localization of secondary phenolic metabolism in oat leaf tissues
1975); IAT, 0.1 M K-phosphate pH 7.5, 8 m M 2-ME, 0.3 mM isovitexin, 9 g M uridine 5'-diphosphate-L-[U-14C]arabinose (6.76 GBq mmol-1), and I0 ill of enzyme extract in a total volume of 30 gl (Fuisting 1981). Chalcone synthase was measured according to Schr6der et al. (1979). The FMT assay is described elsewhere (Knogge and Weissenb6ck 1984). Labelled reaction products were isolated by thin-layer chromatography (TLC) in the solvent systems I (CMT), II (CHS), and IV (IAT), respectively.
Extraction offlavonoids. Flavonoids were extracted from leaves or tissue fractions by homogenization in 80% methanol. Tissue debris was removed by centrifugation (10rain, 5000g). The supernatants were subjected to HPLC analysis.
Analysis of non-flavonoidphenolic compounds Microscopy. Lignificafion of cell walls was demonstrated in free-hand cross sections through different regions of the developing leaf by staining with the phloroglucinol-HC1 reagent after bleaching the tissues with chloral hydrate. For UV-fluorescence microscopy, small leaf pieces were fixed in 4% glutaraldehyde in 0.1 M cacodylate buffer pH 7.0, dehydrated and embedded in an epoxy resin (Spurr 1969). Semi-thin sections were prepared using a microtome. Treatment of the sections with 0.1 M NH4OH resulted in a general intensification of the fluorescence and a shift of the blue fluorescence of wall-bound hydroxycinnamic acid esters to green (Harris and Hartley 1976).
600 : 262 : 63 (by vol.), mixed 1 : 1 with n-propionic acid/water 427:496 (v/v) (Feige et al. 1969).
High-performance liquid chromatography. The quarternary HPLC system of DuPont Instruments (Bad Nauheim, FRG) was used in combination with the Spectroflow 773 absorbance detector (Kratos Analytical Instruments, Trappenkamp, FRG) and the SP4270 integrator (Spectra-Physics, Santa Clara, Cal., USA). Separation of flavonoids was achieved on a Zorbax C-8 colmnn (DuPont) by an isocratic solvent system using a mixture of 78% water (1% H3PO4, 9% methanol, 7% acetonitrile and 6% tetrahydrofuran (by vol). The separated compounds were quantified by external standardization. Separation of cinnamic acids was achieved either by a gradient system on an RP-8 column (Merck, Darmstadt, FRG) (Proksch et al. 1981) or by an isocratic system on a Zorbax C-8 column using a mixture of 67% water (1% H3PO4), 21% methanol and 12% tetrahydrofuran (by vol.).
Determination of radioactivity. Labelled products were localized on TLC plates using a TLC-scanner with linear analyzer (Berthold, Wildbad, FRG), respectively. After scraping off the spots, radioactivity was determined in a liquid scintillation spectrometer (Packard Instruments, Frankfurt, F R G ) using a toluene cocktail containing 0.5% (w/v) of 2,5-diphenyloxazole and 0.03% (w/v) of 1,4-bis(5-phenyloxazoyl)benzene.
Extraction. Analysis of soluble phenolic esters was carried out
Results
after extraction of tissue fractions with 80% methanol and alkaline hydrolysis (1 N NaOH, 30 ~ C, 1-3 h). The insoluble material remaining after extraction was washed with methanol and chloroform and then also hydrolyzed. Hydrolysates were acidified and extracted with ethyl acetate. The resulting solutions of aglyca, mainly hydroxycinnamic acids, were analyzed by HPLC. In this paper, it was not attempted to extract and quantify lignin.
Separation of leaf tissues. Enzymatic digestion of leaf cell walls after removal of one epidermal layer enabled the separation of mesophyll protoplasts and vascular bundles, the latter being resistant to cellulolytic digestion. Approximately 90% of the leaf mesophyll could be isolated as intact protoplasts. Contamination of vascular strands with adhering mesophyll cells was negligible. Preparations of mesophyll protoplasts were not contaminated with epidermal protoplasts since the latter remained in the supernatant during the sedimentation steps. Efforts to prepare ~pidermal protoplasts by repeated floating resulted in suspensions still containing significant amounts of mesophyll protoplasts (10-30%). Routine analyses of enzyme activities, therefore, were carried out with epidermal peels which never showed more than 10% contamination with mesophyll cells.
Feeding experiments. Ten leaves were cut close to the caryopsis and placed in 20 gM [3-14C]cinnamic acid dissolved in 2% ethanol. In pulse-chase experiments, five leaves were extracted with 80% methanol after I h of incubation. The remaining five leaves were transferred to a 20-gM solution of unlabelled cinnamic acid and extracted after a further 2 h of incubation. The insoluble residues were submitted to alkaline hydrolysis (see above). The methanolic extracts were fractionated by chromatography on a polyamide column with water, methanol and 0.01% NH4OH in methanol (Knogge et al. 1981). The flavonoids were eluted with methanol. Both other fractious mainly contained different hydroxycinnamic-acid esters. Fractions were cochromatographed with authentic 1-O-feruloyl- and l-O4-coumaroylglucose as well as with the oat flavones F1. 3 in solvent system III. The hydrolysates were acidified and extracted with ethyl acetate. Cochromatography with ferulic and 4coumaric acid was performed in solvent system I.
Determination of chlorophyll and protoplast yield. Chlorophyll was determined according to Bruinsma (1961). Chlorophyll recovery in isolated protoplasts was used to calculate protoplast yield.
Thin-layer chromatography was performed on microcristalline cellulose with the following solvent systems: I, toluene/acetic acid 2:1 (v/v), saturated with water; II, benzene/acetic acid/ water 115 : 72: 3 (by vol.); III, chloroform/acetic acid 3:2 (v/v), saturated with water; IV, n=butanol/n=propanol/water
Flavonoid biosynthesis. Figure 1 demonstrates activities of four specific enzymes along with flavonoid accumulation as a function of plant age. After the initial rapid increase, the enzymes, with the exception of CHS, continue to exhibit high activities even in leaf stages with already diminished flavonoid biosynthesis. Studies on the distribution of enzymes throughout the leaf revealed highest activities in the upper parts in which the flavonoid products predominantly accumulate (results not shown; compare Weissenb6ck and Sachs 1977; Effertz and
W. Knogge and G. Weissenb6ck: Localization of secondary phenolic metabolism in oat leaf tissues
loo-
|
,'/,,
-0
o33
0.66
-J
/\
50-
Incorporation of Radioactivity into Flavonoids [kaq]
Fig. 2. Incorporation of [14C]cinnamic acid into C-glucosylfla-
vones Ft_3 in four independent experiments. Suspensions (1 ml) containing similar numbers of epidermal protoplasts (o) but increasing numbers of mesophyll protoplasts (o) as contaminants were incubated for 2 h at 30 ~ C with 15 nmol (28.6 kBq) of [3-14C]cinnamic acid. The bars mark SDs of protoplast numbers as counted in a hemacytometer. The coincidence of number of mesophyll protoplasts present and radioactivity incorporated into flavonoids was calculated by linear regression. The insert shows the mean values (with SDs) of radioactivity incorporated into the single flavonoids in the four experiments. Note the high labelling of F~ and F3 both accumulating predominantly in the epidermal layers of the leaf
>
nr
z,
,
l
,
§
6
7
8
5 6 7 Plont Age [ d ]
8
loo- 9 A
/
>
u
P l a n t a 9 Springer-Verlag 1986
Tissue-distribution of secondary phenolic biosynthesis in developing primary leaves of Avena sativa L. W. Knogge* and G. Weissenb6ck Botanisches Institut der Universiffitzu K61n, GyrhofstraBe 15, D-5000 K61n 41, Federal Republic of Germany
Abstract. Primary leaves of oats ( A r e n a sativa L.) have been used to study the integration of secondary phenolic metabolism into organ differentiation and development. In particular, the tissue-specific distribution of products and enzymes involved in their biosynthesis has been investigated. C-Glucosylflavones along with minor amounts of hydroxycinnamic-acid esters constitute the soluble phenolic compounds in these leaves. In addition, considerable amounts of insoluble products such as lignin and wall-bound ferulic-acid esters are formed. The tissue-specific activities of seven enzymes were determined in different stages of leaf growth. The rate-limiting enzyme of flavonoid biosynthesis in this system, chalcone synthase, together with chalcone isomerase (EC 5.5.1.6) and the terminal enzymes of the vitexin and isovitexin branches of the pathway (a flavonoid O-methyltransferase and an isovitexin arabinosyltransferase) are located in the leaf mesophyll. Since the flavonoids accumulate predominantly (up to 70%) in both epidermal layers, an intercellular transport of products is postulated. In contrast to the flavonoid enzymes, Lphenylalanine ammonia-lyase (EC 4.3.1.5), 4-coumarate: CoA ligase (EC 6.2.1.12), and S-adenosylL-methionine: caffeate 3-O-methyltransferase (EC 2.1.1.-), all involved in general phenylpropanoid metabolism, showed highest activities in the basal leaf region as well as in the epidermis and the vascular bundles. We suggest that these latter * Present address: Max-Planck-Institut ffir Zfichtungsfor-
schung, D-5000 K61n 30, Federal Republic of Germany Abbreviations: CHI = chalcone isomerase; CHS = chalcone synthase; 4CL = 4-coumarate :CoA ligase; CMT = S-adenosyl-gmethionine:caffeate 3-O-methyltransferase; FMT=S-adenosyl-L-methionine:vitexin 2"-O-rhamnoside 7-O-methyltransferase; HPLC = high-performance liquid chromatography; IAT = uridine 5'-diphosphate L-arabinose:isovitexin 2"-O-arabinosyttransferase; PAL = L-phenylalanine ammonia-lyase;
enzymes participate mainly in the biosynthesis of non-flavonoid phenolic products, such as lignin in the xylem tissue and wall-bound hydroxycinnamic acid-esters in epidermal, phloem, and sclerenchyma tissues.
Key words: A v e n a C-Glucosylflavone - Flavonoid biosynthesis - Phenylpropanoid metabolism.
Introduction In definite ontogenetic stages, higher plants are capable of synthesizing an array of secondary products, which are usually restricted to certain organs, tissues, or even single cells (Wiermann 1981). Primary leaves of grass seedlings such as oats and rye provide suitable experimental systems to study secondary phenolic metabolism and its integration into organ differentiation and development (Weissenb6ck and Effertz 1974; Strack et al. 1982). Being essentially equifacial these leaves are composed of three major types of tissue: the autotrophic mesophyll and the heterotrophic epidermal layers and vascular bundles. Flavonoids represent the main soluble phenolic compounds in oat primary leaves (Weissenb6ck and Effertz 1974). In addition, extracellular insoluble materials, lignin and wall-bound esters 1 of hydroxycinnamic acids, are formed (Harris and Hartley 1976). Flavonoid biosynthesis results in the accumulation of three major C-glucosylflavones: an isovitexin arabinoside (F2) and two structurally rei The term wall-bound comprises phenolic acids bound to the carbohydrate moiety of cell walls (Hartley et al. 1976; Fry 1983) as well as to cutin in the leaf cuticle and to suberin described to be present in the walls of grass bundle-sheath cells (Kolattukudy 1980, 1981). In the present paper it was not attempted to discriminate betweenthese fractions
W. Knogge and G. Weissenb6ck : Localization of secondary phenolic metabolism in oat leaf tissues
lated vitexin rhamnosides (F1, F3; Fig. 5) (Chopin et al. 1977). Depending on leaf age and the type of tissue, flavonoid patterns show distinct quantitative and qualitative differences (Effertz and Weissenb6ck 1980). The two epidermal layers always contain 60-70% of total leaf flavonoids. All compounds are metabolic end products which are not degraded (Effertz and Weissenb6ck 1978; Proksch et al. 1981). Two models can explain the formation of stageand tissue-dependence of flavone patterns in the leaf. First, each tissue is biosynthetically autonomous; thus, product accumulation reflects the distribution of the respective biosynthetic enzymes. Alternatively, biosynthesis occurs in one type of tissue; consequently, products are translocated to other tissues. Our previous investigations on the tissue localization of the initial enzymes of flavonoid biosynthesis, chalcone synthase (CHS) and chalcone isomerase (CHI), had been restricted to that growth stage of plants showing the most rapid flavone accumulation (Fuisting and Weissenb6ck 1980; Weissenb6ck and Sachs 1977). In the present paper we report on the localization of seven enzymes of phenolic metabolism in leaf tissues during organogenesis. The enzymes analyzed were L-phenylalanine ammonia-lyase (PAL), 4-coumarate: CoA ligase (4CL) (Knogge et al. 1981), and caffeate O-methyltransferase (CMT), all involved in general phenylpropanoid metabolism, as well as CHS and CHI and the final enzymes of the vitexin and isovitexin branches of the pathway. These final enzymes included a flavonoid O-methyltransferase (FMT) which specifically catalyzes the transfer of a methyl group to the 7-hydroxyl group of vitexin rhamnoside F3 (Knogge and Weissenb6ck 1984) and an isovitexin O-arabinosyltransferase (IAT) showing high specificity for the flavonoid substrate (Fuisting 1981). Additional results are presented concerning the accumulation of non-flavonoid phenolic products in oat primary leaf tissues. Material and methods Plant material. Oat plants (Arena sativa L. cv. F1/imingskrone; F. von Lochow-Petkus GmbH, Bergen, FRG) were grown under standard conditions in a phytotron under a regime of 7500 lx for 13 h per day (Weissenb6ck and Effertz 1974). For studying tissue-specific enzyme activities as a function of seedling development, 4- to 8-d-old primary leaves were harvested at the end of the dark phase. Chemicals. Polyclar AT, DOWEX AG 1X2 (200-400 mesh), and dithiothreitol (DTT) were obtained from Serva (Heidelberg, FRG), all biochemicals from Boehringer (Mannheim, FRG), all laboratory chemicals (analytical grade) from Merck
197
(Darmstadt, FRG). Cinnamic acids were purchased from Roth (Karlsruhe, FRG). Flavonoids were extracted and purified according to standard methods of this laboratory using polyamide and Sephadex LH-20 column chromatography (Popovici et al. 1977) as well as high-performance liquid chromatography (HPLC). 1-O-4-Coumaroylglucose and 1-O-feruloylglucose from Antirrhinum petals (Harborne and Corner 1961) were kind gifts of Dr. D. Strack (Botanical Institute, University of Cologne). Cellulysin (Triehoderma viride, 10000 units g 1) was purchased from Calbiochem-Behring (La Jolla, Cal., USA). [2-14C] Malonyl CoA (1.56~1.77 GBq mmol 1), uridine 5"-diphosphate-L-[U-14C]arabinose (6.76 GBq mmol 1), and S-adenosyl-L-(methyl-tgC)methionine (1.94 GBq mmo1-1) were obtained from New England Nuclear (Dreieichenhain, FRG), [3l~C]cinnamic acid (1.66-2.03 GBq mmol ~) from CEA (Gifsur-Yvette), France). 4-Coumaroyl CoA and 2',4,4',6"-tetrahydroxychalcone were synthesized according to St6ckigt and Zenk (1975) and Moustafa and Wong (1967), respectively.
Preparation of tissue fractions. Preparation of epidermal peels and mesophyll protoplasts has been described previously (Weissenb6ck and Sachs 1977; Haas etal. 1979). The vascular strands resistant to cellulolytic digestion were collected and washed twice with protoplast isolation medium. This method did not allow tissue separation in the basal region of the leaves. This part was defined by the break point of the epidermal peels and comprised one (4-d-old leaves) to 2 cm (8-d-old leaves). It was analyzed, therefore, as a whole. For routine tests of enzyme distribution the leaves were separated into epidermal peels and the corresponding leaf complements (compare Weissenb6ck and Sachs 1977). Extraction of enzymes. Tissue fractions from 40 leaves were ground in a mortar at 0 4 ~ C in the presence of sufficient quartz sand, DOWEX, Polyclar AT and buffer (see below). The homogenates were squeezed through Miracloth (Chicopee Mills, Milltown, N.J., USA) and centrifuged for 10 rain at 48000 g. The clear supernatants served as enzyme sources. Extraction procedures were optimized for the individual enzymes, and different extraction buffers were, therefore, used: PAL, 0A M 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1 or 0.1M glycine-NaOH pH 8.5 [5mM 2-mercaptoethanol (2-ME)I; 4CL, 0.1 M K-phosphate pH 7.3 (40 mM dithiothreitol); CMT and FMT, 0.1 M K-phosphate pH 7.5 (5 mM ethylenediaminetetraacetate, 10 mM 2-ME); CHS and CHI, 0.1 M K-phosphate pH 8.0 (4.3 mM 2-ME); IAT, 0.1 M K-phosphate pH 7.5 (8 mM 2-ME; Fuisting 1981). In the case of CHS, before centrifugation the filtrate was again treated with DOWEX in a batch procedure followed by filtration through Miracloth. Enzyme assays. All enzyme assays were carried out at 30~ C; PAL, 4CL, and CHI were measured photometrically; for all other enzymes, radioactive assays were used. Assay conditions have been optimized starting from published methods as indi'cated. Assay mixtures were as follows PAL, 0.1 M Tris-HC1 pH 8.5, 5 mM 2-ME, 5 mM L-phenylalanine and 50-100 gl of enzyme extract in a total volume of I ml (Weissenb6ck 1975); 4CL: 50 mM K-phosphate pH 7.3, 2 mM dithiothreitol, 1 mM ATP, 10 mM MgC12, 0.2 mM CoA, 2 mM 4-coumaric acid and 10-50 gl of enzyme extract in a total volume of 0.5 ml (Gross and Zenk 1974); CHI, 0.1 M K-phosphate pH 8.0, 75 ~tM 2',4,4"6'-tetrahydroxychalcone, 50 mM KCN and 10 20 gl of enzyme extract in a total volume of 0.5 ml (Boland and Wong 1975); CMT, 0.1 M K-phosphate pH 7.5, 1 mM caffeic acid, 1.5 mM MgClz, 20 mM Na-ascorbate, 0.17 mM S-adenosyl-L(methyl-14C)methionine (37 MBq mmol 1), and 10 50 gl of enzyme extract in a total volume of 0.2 ml (Poulton and Butt
198
W. Knogge and G. Weissenb6ck: Localization of secondary phenolic metabolism in oat leaf tissues
1975); IAT, 0.1 M K-phosphate pH 7.5, 8 m M 2-ME, 0.3 mM isovitexin, 9 g M uridine 5'-diphosphate-L-[U-14C]arabinose (6.76 GBq mmol-1), and I0 ill of enzyme extract in a total volume of 30 gl (Fuisting 1981). Chalcone synthase was measured according to Schr6der et al. (1979). The FMT assay is described elsewhere (Knogge and Weissenb6ck 1984). Labelled reaction products were isolated by thin-layer chromatography (TLC) in the solvent systems I (CMT), II (CHS), and IV (IAT), respectively.
Extraction offlavonoids. Flavonoids were extracted from leaves or tissue fractions by homogenization in 80% methanol. Tissue debris was removed by centrifugation (10rain, 5000g). The supernatants were subjected to HPLC analysis.
Analysis of non-flavonoidphenolic compounds Microscopy. Lignificafion of cell walls was demonstrated in free-hand cross sections through different regions of the developing leaf by staining with the phloroglucinol-HC1 reagent after bleaching the tissues with chloral hydrate. For UV-fluorescence microscopy, small leaf pieces were fixed in 4% glutaraldehyde in 0.1 M cacodylate buffer pH 7.0, dehydrated and embedded in an epoxy resin (Spurr 1969). Semi-thin sections were prepared using a microtome. Treatment of the sections with 0.1 M NH4OH resulted in a general intensification of the fluorescence and a shift of the blue fluorescence of wall-bound hydroxycinnamic acid esters to green (Harris and Hartley 1976).
600 : 262 : 63 (by vol.), mixed 1 : 1 with n-propionic acid/water 427:496 (v/v) (Feige et al. 1969).
High-performance liquid chromatography. The quarternary HPLC system of DuPont Instruments (Bad Nauheim, FRG) was used in combination with the Spectroflow 773 absorbance detector (Kratos Analytical Instruments, Trappenkamp, FRG) and the SP4270 integrator (Spectra-Physics, Santa Clara, Cal., USA). Separation of flavonoids was achieved on a Zorbax C-8 colmnn (DuPont) by an isocratic solvent system using a mixture of 78% water (1% H3PO4, 9% methanol, 7% acetonitrile and 6% tetrahydrofuran (by vol). The separated compounds were quantified by external standardization. Separation of cinnamic acids was achieved either by a gradient system on an RP-8 column (Merck, Darmstadt, FRG) (Proksch et al. 1981) or by an isocratic system on a Zorbax C-8 column using a mixture of 67% water (1% H3PO4), 21% methanol and 12% tetrahydrofuran (by vol.).
Determination of radioactivity. Labelled products were localized on TLC plates using a TLC-scanner with linear analyzer (Berthold, Wildbad, FRG), respectively. After scraping off the spots, radioactivity was determined in a liquid scintillation spectrometer (Packard Instruments, Frankfurt, F R G ) using a toluene cocktail containing 0.5% (w/v) of 2,5-diphenyloxazole and 0.03% (w/v) of 1,4-bis(5-phenyloxazoyl)benzene.
Extraction. Analysis of soluble phenolic esters was carried out
Results
after extraction of tissue fractions with 80% methanol and alkaline hydrolysis (1 N NaOH, 30 ~ C, 1-3 h). The insoluble material remaining after extraction was washed with methanol and chloroform and then also hydrolyzed. Hydrolysates were acidified and extracted with ethyl acetate. The resulting solutions of aglyca, mainly hydroxycinnamic acids, were analyzed by HPLC. In this paper, it was not attempted to extract and quantify lignin.
Separation of leaf tissues. Enzymatic digestion of leaf cell walls after removal of one epidermal layer enabled the separation of mesophyll protoplasts and vascular bundles, the latter being resistant to cellulolytic digestion. Approximately 90% of the leaf mesophyll could be isolated as intact protoplasts. Contamination of vascular strands with adhering mesophyll cells was negligible. Preparations of mesophyll protoplasts were not contaminated with epidermal protoplasts since the latter remained in the supernatant during the sedimentation steps. Efforts to prepare ~pidermal protoplasts by repeated floating resulted in suspensions still containing significant amounts of mesophyll protoplasts (10-30%). Routine analyses of enzyme activities, therefore, were carried out with epidermal peels which never showed more than 10% contamination with mesophyll cells.
Feeding experiments. Ten leaves were cut close to the caryopsis and placed in 20 gM [3-14C]cinnamic acid dissolved in 2% ethanol. In pulse-chase experiments, five leaves were extracted with 80% methanol after I h of incubation. The remaining five leaves were transferred to a 20-gM solution of unlabelled cinnamic acid and extracted after a further 2 h of incubation. The insoluble residues were submitted to alkaline hydrolysis (see above). The methanolic extracts were fractionated by chromatography on a polyamide column with water, methanol and 0.01% NH4OH in methanol (Knogge et al. 1981). The flavonoids were eluted with methanol. Both other fractious mainly contained different hydroxycinnamic-acid esters. Fractions were cochromatographed with authentic 1-O-feruloyl- and l-O4-coumaroylglucose as well as with the oat flavones F1. 3 in solvent system III. The hydrolysates were acidified and extracted with ethyl acetate. Cochromatography with ferulic and 4coumaric acid was performed in solvent system I.
Determination of chlorophyll and protoplast yield. Chlorophyll was determined according to Bruinsma (1961). Chlorophyll recovery in isolated protoplasts was used to calculate protoplast yield.
Thin-layer chromatography was performed on microcristalline cellulose with the following solvent systems: I, toluene/acetic acid 2:1 (v/v), saturated with water; II, benzene/acetic acid/ water 115 : 72: 3 (by vol.); III, chloroform/acetic acid 3:2 (v/v), saturated with water; IV, n=butanol/n=propanol/water
Flavonoid biosynthesis. Figure 1 demonstrates activities of four specific enzymes along with flavonoid accumulation as a function of plant age. After the initial rapid increase, the enzymes, with the exception of CHS, continue to exhibit high activities even in leaf stages with already diminished flavonoid biosynthesis. Studies on the distribution of enzymes throughout the leaf revealed highest activities in the upper parts in which the flavonoid products predominantly accumulate (results not shown; compare Weissenb6ck and Sachs 1977; Effertz and
W. Knogge and G. Weissenb6ck: Localization of secondary phenolic metabolism in oat leaf tissues
loo-
|
,'/,,
-0
o33
0.66
-J
/\
50-
Incorporation of Radioactivity into Flavonoids [kaq]
Fig. 2. Incorporation of [14C]cinnamic acid into C-glucosylfla-
vones Ft_3 in four independent experiments. Suspensions (1 ml) containing similar numbers of epidermal protoplasts (o) but increasing numbers of mesophyll protoplasts (o) as contaminants were incubated for 2 h at 30 ~ C with 15 nmol (28.6 kBq) of [3-14C]cinnamic acid. The bars mark SDs of protoplast numbers as counted in a hemacytometer. The coincidence of number of mesophyll protoplasts present and radioactivity incorporated into flavonoids was calculated by linear regression. The insert shows the mean values (with SDs) of radioactivity incorporated into the single flavonoids in the four experiments. Note the high labelling of F~ and F3 both accumulating predominantly in the epidermal layers of the leaf
>
nr
z,
,
l
,
§
6
7
8
5 6 7 Plont Age [ d ]
8
loo- 9 A
/
>
u
