| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
First published online January 11, 2008; 10.1104/pp.107.112979 Plant Physiology 146:1040-1052 (2008) © 2008 American Society of Plant Biologists OPEN ACCESS ARTICLE
Bitterness in Almonds1,[C],[OA]Plant Biochemistry Laboratory, Department of Plant Biology, Center for Molecular Plant Physiology (R.S.-P, K.J., B.L.M.), and Chemistry Department (C.E.O.), Faculty of Life Sciences, University of Copenhagen, DK–1871 Frederiksberg C, Copenhagen, Denmark; and Departamento de Mejora Vegetal, Centro de Edafología y Biología Aplicada del Segura-Consejo Superior de Investigaciones Científicas, E–30100 Murcia, Spain (F.D.)
Bitterness in almond (Prunus dulcis) is determined by the content of the cyanogenic diglucoside amygdalin. The ability to synthesize and degrade prunasin and amygdalin in the almond kernel was studied throughout the growth season using four different genotypes for bitterness. Liquid chromatography-mass spectrometry analyses showed a specific developmentally dependent accumulation of prunasin in the tegument of the bitter genotype. The prunasin level decreased concomitant with the initiation of amygdalin accumulation in the cotyledons of the bitter genotype. By administration of radiolabeled phenylalanine, the tegument was identified as a specific site of synthesis of prunasin in all four genotypes. A major difference between sweet and bitter genotypes was observed upon staining of thin sections of teguments and cotyledons for β-glucosidase activity using Fast Blue BB salt. In the sweet genotype, the inner epidermis in the tegument facing the nucellus was rich in cytoplasmic and vacuolar localized β-glucosidase activity, whereas in the bitter cultivar, the β-glucosidase activity in this cell layer was low. These combined data show that in the bitter genotype, prunasin synthesized in the tegument is transported into the cotyledon via the transfer cells and converted into amygdalin in the developing almond seed, whereas in the sweet genotype, amygdalin formation is prevented because the prunasin is degraded upon passage of the β-glucosidase-rich cell layer in the inner epidermis of the tegument. The prunasin turnover may offer a buffer supply of ammonia, aspartic acid, and asparagine enabling the plants to balance the supply of nitrogen to the developing cotyledons.
The knowledge about hydrogen cyanide (HCN) formation in plants has its origin in antiquity. In ancient Egypt, traitorous priests in Memphis and Thebes were poisoned to death with pits of peaches (Davis, 1991  ). The first known detection of HCN liberated from damaged plant tissue was made in 1802 by the pharmacist Bohm in Berlin upon distillation of bitter almonds (Lechtenberg and Nahrstedt, 1999). In 1830, Robiquet and Boutron-Chalard discovered the structure of the HCN-liberating compound in bitter almonds (Lechtenberg and Nahrstedt, 1999). Because the compound was isolated from Prunus amygdalus (synonym Prunus dulcis), it was named amygdalin. Amygdalin has subsequently been found widespread in seeds of other members of the Rosaceae like in apples (Malus spp.), peaches (Prunus persica), apricots (Prunus armeniaca), black cherries (Prunus serotina), and plums (Prunus spp.; McCarty et al., 1952; Conn, 1980; Frehner et al., 1990; Møller and Seigler, 1991; Swain et al., 1992; Poulton and Li, 1994; Arrázola, 2002; Dicenta et al., 2002). The diglucoside amygdalin was the first member to be isolated of a new class of natural products now known as cyanogenic glucosides. Cyanogenic glucosides are present in more than 2,500 different plant species, including many important crop plants (Seigler and Brinker, 1993; Bak et al., 2006). Upon disruption of plant tissue containing cyanogenic glucosides, these are typically hydrolyzed by β-glucosidases with concomitant release of Glc, an aldehyde or ketone, and HCN. This two-component system, of which each of the separate components is chemically inert, provides plants with an immediate chemical defense against attacking herbivores and pathogens (Conn, 1969; Nahrstedt, 1985; Jones, 1988; Morant et al., 2003; Nielsen et al., 2006; Zagrobelny et al., 2004, 2007a, 2007b). In addition to their possible defense function, accumulation of cyanogenic glucosides in certain angiosperm seeds may provide a storage deposit of reduced nitrogen and sugar for the developing seedlings (Lieberei et al., 1985; Selmar et al., 1988, 1990; Swain et al., 1992).
Caius Plinius Secundus (better known as Pliny the Elder) stated in his 37-volume encyclopedia entitled Naturalis Historia, which was completed shortly after his death in 79 AD that the Romans were proud of knowing how to remove bitterness from almond kernels (Pliny the Elder, 77
Previous studies in almonds have shown that prunasin is transformed into amygdalin during fruit ripening (Frehner et al., 1990
In this study we have investigated prunasin and amygdalin synthesis and turnover in sweet and bitter almond genotypes by direct measurements of the enzyme activities and their tissue and cellular locations. The results point to a difference in β-glucosidase activity in the inner epidermis of the tegument as the main determinant of whether a variety is sweet or bitter.
Fruit Development and Ripening
The synthesis, accumulation, and degradation of the cyanogenic glucosides prunasin and amygdalin were studied during the entire growth season from tree flowering to full ripening of the kernel using the four genotypes ‘Ramillete’ (SkSk, sweet), ‘Marcona’ (Sksk, sweet), ‘Garrigues’ (Sksk, slightly bitter), and ‘S3067’ (sksk, bitter). Almond trees flower in the month of February in southern Spain. Fruit development is characterized by an increased size of the cotyledon at the expense of a diminishing nucellus and endosperm (Fig. 2
). The development of the different tissues of the fruit in the four genotypes showed sequential deviations as has previously been reported in other almond cultivars (Serafimov, 1981
Cyanogenic Glucoside Levels from Flowering to Fruit Ripening The levels of prunasin and amygdalin during the entire growth season from tree flowering to fruit ripening was monitored by liquid chromatography-mass spectrometry (LC-MS) analyses in the four genotypes ‘Ramillete’ (SkSk, sweet), ‘Marcona’ (Sksk, sweet), ‘Garrigues’ (Sksk, slightly bitter), and ‘S3067’ (sksk, bitter; Fig. 3 ). In the bitter genotype, prunasin was detected in leaf laminae, petioles, fruit tegument, and nucellus plus endosperm. In leaf laminae and petioles, the prunasin content peaked in April. Prunasin content in tegument from the bitter genotype was the highest found in the fruit tissues. Its increase was from March to June. In the three other genotypes, prunasin was present in much lower amounts in leaf laminae and stems and was not detectable in any of the fruit tissues analyzed. No prunasin was detected in fruit exocarp, mesocarp, and endocarp of the four genotypes (data not shown).
All analyses of teguments from the four genotypes showed the absence of amygdalin in this fruit tissue. Amygdalin was detectable in the nucellus and endosperm from the bitter genotype. Concomitant with the decrease in prunasin content in the nucellus and endosperm of the bitter genotype, the amygdalin content in the cotyledons began to increase to reach a final concentration of 9 µmol/100 mg fresh weight. In the slightly bitter genotype ‘Garrigues’ and in the sweet heterozygous genotype ‘Marcona’, amygdalin was detectable, but at much lower concentrations of 0.03 and 0.007 µmol/100 mg fresh weight, respectively. In the sweet homozygous genotype ‘Ramillete’ no amygdalin was detectable.
To investigate whether prunasin or amygdalin was synthesized in the shoot apex and transported to other parts of the almond tree and whether transport to the developing almond fruit from other parts of the almond tree were occurring, a series of girdling experiments were performed in which the epidermis and cambium cell layers, including the phloem, were removed to prevent transport across the site of incision. Prior to the girdling, analyses of stems and peduncles from ‘Ramillete’ (SkSk, sweet) and ‘S3067’ (sksk, bitter) showed the presence of prunasin and minute amounts of amygdalin (Fig. 4 ). The prunasin level was highest in the stem where it reached 2 µmol/100 mg fresh tissue in the bitter genotype ‘S3067’ and 0.3 µmol/100 mg fresh tissue in the sweet genotype ‘Ramillete’. The corresponding values in peduncles were 0.3 and 0.05 µmols/100 mg tissue, respectively. The absolute prunasin levels in both tissues varied considerably from one experiment to the other. Amygdalin levels never exceeded 0.005 µmol/100 mg fresh tissue in the peduncle and stem and were not significantly different in the sweet and bitter variety. In contrast to the samples taken in April, it was not possible to detect amygdalin in any of the samples taken in May. Upon girdling of the stem beneath the first-year shoot and girdling of the peduncle, neither prunasin nor amygdalin was observed to accumulate in the tissue segments above the incision and no decrease was observed below the incision. These results show that transport of cyanogenic glucosides from the shoot apex to the rest of the almond tree is not the decisive parameter determining whether a fruit becomes sweet or bitter. The experiments with girdled peduncles strongly indicate that cyanogenic glucoside accumulation in the bitter almond fruit reflects de novo synthesis in the fruit and not transport from other parts of the plant. To substantiate the latter conclusion, a second type of trapping experiments was based on collection of exudates from freshly cut peduncle stubs (Fig. 5 ). Prunasin as well as amygdalin was detectable in the exudate, but the level of amygdalin was again approximately 100-fold lower than that of prunasin. The experiments were carried out using both a sweet and a bitter genotype, but the genotype did not influence the exuded amounts of cyanogenic glucosides collected. It is difficult to assess how efficient this experimental setup is with respect to measuring transport of prunasin and amygdalin to the developing fruit. However, because the girdling experiments with the peduncles also provided no indication of transport, we interpret the two series of experiments to demonstrate that prunasin and amygdalin are synthesized de novo in the developing almond fruit.
Measurements of Biosynthetic Activity within the Developing Almond Fruit To investigate which of the tissues in the developing fruit that were biosynthetically active, the different types of tissues were dissected and incubated with either radiolabeled Phe or with radiolabeled UDPG supplemented with either mandelonitrile or prunasin as acceptor.
Feeding Experiments with L-[14C]Phe
Microsome Assays with L-[14C]Phe or L-[U-14C]Tyr In other cyanogenic plant species, the conversion of a parent amino acid to the cyanogenic glucoside is known to be catalyzed by two Cyt P450s anchored in the membrane system of the endoplasmatic reticulum and by a soluble UDPG glucosyltransferase. Accordingly, microsomal preparations harboring the two Cyt P450s were shown to catalyze the conversion of the parent amino acid into the corresponding cyanohydrin (McFarlane et al., 1975 ; Fig. 1). When microsomes were prepared from freshly dissected teguments from the two genotypes ‘Ramillete’ (SkSk, sweet) and ‘S3067’ (sksk, bitter) and incubated with L-[14C]Phe in the presence of NADPH, no radiolabeled intermediates or products were formed as monitored by radio-thin layer chromatography (TLC; Fig. 7
). The expected intermediates and final product were phenylacetaldoxime, phenylacetonitrile, and mandelonitrile. The mandelonitrile is labile and would be detected as the dissociation product benzaldehyde on the TLC. The position of all these components on the TLC was determined by coapplication of authentic standards. To assess whether this negative result reflected the presence of inhibitory substances in the tegument that interacted with the Cyt P450s, a new set of experiments were carried out in which almond teguments or almond roots and 2-d-old sorghum seedlings were homogenized together. The microsomes isolated from sorghum seedlings and preparations obtained from coisolation of microsomes from tissues of the two plant species were tested for biosynthetic activity with L-[U-14C]Tyr as substrate (Fig. 7). As expected, the sorghum microsomes showed strong conversion of L-[U-14C]Tyr into [U-14C]p-hydroxybenzaldehyde, the dissociation product of p-hydroxymandelonitrile. p-Hydroxymandelontrile is the aglycon of dhurrin, the Tyr-derived cyanogenic glucoside in sorghum. In contrast, when the experiment was repeated with sorghum microsomes prepared in the presence of almond teguments, no metabolic activity was observed. This documents the presence of inhibitory substances in the almond tegument that upon grinding of the plant material are liberated and inhibit the activity of the sorghum Cyt P450 enzymes. In comparison, almond roots also showed some inhibitory activity on sorghum microsomes, but to a much less extent.
[U-14C]UDP-Glucosyltransferase Assays The distribution of UDPG glucosyltransferase activities able to glucosylate either mandelonitrile (GT1) into prunasin or prunasin into amygdalin (GT2) was followed throughout the growth season from April to July with focus on the activities in leaf laminae and in the different fruit tissues (Fig. 8 ). These experiments were carried out by administration of radiolabeled [U-14C]UDPG in combination with unlabeled aglucon acceptors to young leaf laminae or tissues dissected from developing fruits. Accordingly, the results obtained are not quantitative but indicative of the main distribution of the two glucosyltransferases. Leaf lamina was found to show low UDPG mandelonitrile glucosyltransferase activity over the entire growth phase and independent of the genotype tested. In fruit tissues, the activity of this glucosyltransferase was more dominant in the bitter compared to the sweet variety, but activity was indeed observed in most samples of the sweet genotype. In contrast to these results, UDPG prunasin glucosyltransferase activity was essentially restricted to the cotyledon with similar activities in the sweet and bitter variety (Fig. 8). In the experiments with [U-14C]UDPG and prunasin as the unlabeled acceptor, radiolabeling of prunasin was also observed in the cotyledon. This demonstrates that some of the administered prunasin was degraded into mandelonitrile by endogenous prunasin hydrolase and the mandelonitrile then reconverted into radiolabeled prunasin by the action of UDPG mandelonitrile glucosyltransferase (Fig. 1). Alternatively, the source of mandelonitrile could represent turnover of the newly formed radiolabeled amygdalin or of the endogenous pool of amygdalin by the combined action of amygdalin hydrolase and prunasin hydrolase (Fig. 1).
Localization of β-Glucosidases in the Tegument Although the tegument of the genotype ‘S3067’ (sksk, bitter) showed a high content of prunasin whereas prunasin was barely detectable in ‘Ramillete’ (SkSk, sweet), the radiolabeling experiments did not indicate major differences in biosynthetic capacity between the bitter and sweet genotypes. We therefore focused our study on possible differences in the β-glucosidase activity and location in the tegument of the bitter and sweet varieties. β-Glucosidase activity was assessed in transverse sections of the tegument situated adjacent to the developing cotyledon by staining with Fast Blue BB salt in the presence of the β-glucosidase substrate 6-bromo-2-napthyl-β-D-glucopyranoside (Fig. 9 ). Upon 1-min staining of sections from ‘Ramillete’ (SkSk, sweet), strong β-glucosidase activity was observed in the inner epidermis of the tegument facing the nucellus. Higher magnification of the inner epidermis cells showed that β-glucosidase staining was restricted to the cytosol and the main vacuole present in these cells. When the staining procedure was carried out in the absence of the β-glucosidase substrate 6-bromo-2-napthyl-β-D-glucopyranoside, no or only very weak staining was observed. When comparable sections of the tegument from ‘S3067’ (sksk, bitter) were analyzed in the light microscope, the presence of the inner epidermis cell layer offering strong staining in the sweet variety was clearly visible. However, this cell layer did not stain with the Fast Blue BB salt together with the β-glucosidase substrate 6-bromo-2-napthyl-β-D-glucopyranoside, except for the weak background staining also observed with the sweet variety in the absence of the β-glucosidase substrate. When the staining period of the sections of ‘S3067’ (sksk, bitter) was prolonged from 1 to 10 min, a weak β-glucosidase activity was detectable. However, this activity was derived from the apoplast surrounding the inner epidermis cell layer, and not from the cytosol and central vacuole as observed in the sweet genotype.
Bitterness in almond is determined by the content of the cyanogenic diglucoside amygdalin. It has previously not been clear whether amygdalin accumulation in the bitter kernel reflected transport from other parts of the almond tree or de novo synthesis in the kernel. To address this issue, the occurrence of amygdalin and its precursor prunasin was monitored during the entire growth season from tree flowering to fruit ripening with focus on the developing stems, leaf laminae, petioles, and different tissues of the developing fruit. During the entire season, prunasin was detected in the vegetative part of all four genotypes ‘Ramillete’ (SkSk, sweet), ‘Marcona’ (Sksk, sweet), ‘Garrigues’ (Sksk, slightly bitter), and ‘S3067’ (sksk, bitter), but the content was always severalfold higher in the bitter genotype. In all four genotypes, prunasin content in leaf laminae, petioles, and stems varied in a similar manner, probably reflecting the physiological development of the tree. At the beginning of the growth season, the content of prunasin increases in leaf laminae and stems. This may be related to mobilization of reserves stored in lignified tissues throughout the winter. The prunasin content in stems, petioles, and leaf laminae decreases in the subsequent period (April). This coincides with the initiation of secondary growth, formation of terminal shoot sprouts resulting in a significant expansion of the tree crown and with the formation of the different fruit tissues (exocarp, mesocarp, endocarp, and kernel; Girona and Marsal, 1995 ). Between March and April the fruit reaches maximum size (Fig. 2) and prunasin levels increase in all four genotypes although not simultaneously. Previous studies in several Rosaceae (Seigler, 1981), in other Prunus species (Selmar et al., 1988) and in almond (Frehner et al., 1990; Dicenta et al., 2002) indicated similar patterns of prunasin and amygdalin accumulation although other studies using leaf lamina of different taxa or hybrids between Prunus species, showed that the ratios of prunasin to amygdalin were independent of the time of harvest and to vary somewhat erratically (Santamour, 1998). The high but quite variable differences in prunasin content between sweet and bitter genotypes indicate that the control of prunasin level is a polygenic trait (Dicenta et al., 2002). In contrast to the results with vegetative tissues, prunasin was only detected in the fruits of the bitter genotype (Fig. 3). Most remarkable is the prunasin level in the tegument that rises constantly to a level of 1.6 µmol/100 g fresh weight at the end of June, where a rapid decline followed by complete disappearance is observed. Amygdalin is first detected in the nucellus and endosperm where transient accumulation is observed at the end of March. From April, the amygdalin level in the cotyledons rises constantly to a level of 9 µmol/100 mg fresh weight in the month of August when the fruit is mature. The sequential appearance and decline of prunasin level in the tegument and parallel accumulation of amygdalin in the cotyledon suggested that prunasin produced or imported into the tegument might serve as a direct precursor for amygdalin formation in the cotyledons.
Girdling experiments using the two genotypes ‘Ramillete’ (SkSk, sweet) and ‘S3067’ (sksk, bitter) demonstrated that the prunasin and amygdalin observed to accumulate in the tegument and cotyledons, respectively (Fig. 4), were de novo synthesized in these tissues. In undamaged stems and peduncles, both prunasin and amygdalin were detected in low but varying amounts. The prunasin content was always higher in the ‘S3067’ (sksk, bitter) compared to ‘Ramillete’ (SkSk, sweet), whereas the amygdalin content was about the same in the bitter and sweet genotype. In no case did analysis of stem internodes beneath the first-year shoots and of peduncle segments above and below the incisions show significant changes in the content of prunasin and amygdalin. Likewise, in the sap exuded from freshly cut peduncles, no differences in the amounts of prunasin and amygdalin between the sweet and bitter genotype was observed (Fig. 5). This strongly argues that the prunasin and amygdalin present in the fruit is formed by de novo synthesis in the fruit in course of fruit development. Similar girdling experiments have previously been carried out in cassava and demonstrated the significance of transport of cyanogenic glucosides from the site of production in the actively growing shoot apexes and leaf laminae to the tuber. Thus in cassava, the cyanogenic glucoside content in the internode above the incision zone in the shoot apex increased by a factor of 75 (Jørgensen et al., 2005
A number of experiments were carried out to assess in which fruit tissues the synthesis of prunasin and amygdalin takes place. Upon administration of radiolabeled Phe to different excised tissues of the fruit, the only tissue that showed capacity to produce radiolabeled prunasin was the tegument. No major differences in the ability of the four tested sweet, slightly bitter, and bitter genotypes to produce radiolabeled prunasin was observed (Fig. 6). To obtain a quantitative measure of the biosynthetic capacity, microsomes were prepared from each of the four genotypes. Unfortunately, this approach was negative, because of the presence of a Cyt P450 inhibitor in the tegument tissue that completely inactivated the microsomal preparations as shown by cohomogenization of the almond teguments with sorghum seedlings that actively synthesize dhurrin. This resulted in inactivation of the sorghum Cyt P450 enzyme system (Fig. 7). We have previously reported that the tegument in cassava seeds contains a potent Cyt P450 inhibitor (Koch et al., 1992
In cassava, the two cyanogenic glucosides linamarin and lotaustralin contribute to the bitterness of the tubers. However, other bitter constituents like isopropyl-β-D-apiofuranosyl-(1
Amygdalin is formed from prunasin by the action of a UDPG prunasin glucosyltransferase (Fig. 8). Radiolabelling experiments demonstrated that this enzyme activity was restricted to the nucellus, endosperm, and cotyledon with by far the strongest activity observed in the cotyledons. This would imply that prunasin synthesized in the tegument is transported to the nucellus, endosperm or cotyledon where it becomes glucosylated and finally is stored in the cotyledon. Amygdalin only accumulates in the bitter genotype ‘S3067’ and this matches very well the observed specific accumulation of the precursor prunasin in the tegument of ‘S3067’. But from the radiolabeling experiments, the sweet genotypes would also be predicted to be able to produce amygdalin because all biosynthetic enzyme activities were present, except for the UDPG mandelonitrile glucosyltransferase in amounts similar to those found in the bitter genotype. This prompted us to investigate whether the difference between the ability of the bitter and sweet genotypes to accumulate amygdalin resided in their ability to transfer the prunasin precursor molecule from the tegument to the nucellus, endosperm, or cotyledon. This issue was addressed by studying thin sections of whole seeds (tegument, nucellus, endosperm, and cotyledon) at the developmental stage where prunasin content declined in the tegument of the bitter genotype. Thin sections (6 µm) were stained with Fast Blue BB salt together with the β-glucosidase substrate 6-bromo-2-napthyl-β-D-glucopyranoside to monitor β-glucosidase activity. In the sweet variety ‘Ramillete’, the staining showed the existence of a continuous β-glucosidase-rich cell layer in the inner epidermis of the tegument, facing the nucellus. The β-glucosidase activity in these cells was restricted to the cytoplasm and the main central vacuole. The similar cell layer was also present in the bitter genotype but in this genotype staining with the Fast Blue BB salt together with the β-glucosidase substrate was weak and was restricted to the apoplastic space. If transport of prunasin from tegument to cotyledon proceeds via the symplast, this would indicate that the prunasin synthesized in the tegument of the sweet variety (SkSk) is degraded when it has to pass this cell layer. The tegument is the thin layer of mother-derived tissue that encapsulates the nucellus and the developing endosperm and embryo and this mechanism for control of bitterness would comply with the notion that the genetic control of this trait is mother-tissue dependent (Werner and Creller, 1997
But why would prunasin first be synthesized in the tegument tissue of the developing almond fruit in the sweet variety and subsequently be subjected to degradation? In sorghum seedlings, the Cyt P450 system catalyzing dhurrin synthesis constitutes about 0.5% of the total membrane proteins (Sibbesen et al., 1994
Cassava contains the two cyanogenic glucosides linamarin and lotaustralin (Andersen et al., 2000
Bitterness in almond is inherited as a single recessive gene (sksk, bitter) and controlled by the genotype of the seed mother (Heppner, 1923
Plant Material Almond (Prunus dulcis) branches of the following four genotypes ‘Ramillete’ (SkSk, sweet), ‘Marcona’ (Sksk, sweet), ‘Garrigues’ (Sksk, slightly bitter), and ‘S3067’ (sksk, bitter) were provided by the Almond Breeding Program of Centro de Edafología y Biología Aplicada del Segura-Consejo Superior de Investigaciones Científicas (CEBAS-CSIC). Every second week during the growth season from tree flowering to kernel maturity (March to August), plant material was sent to the Plant Biochemistry Laboratory at the University of Copenhagen by courier shipment. The branches were placed in water and used as a source of leaf laminae, petioles, stems, peduncles, roots, and fruits. Fruits were separated into mesocarp, endocarp, tegument, nucellus, endosperm, and cotyledon as required for the different sets of experiments. All analyses were carried out on the day of arrival of the plant material, i.e. the day after the branches were cut off the tree. Roots analyzed were from 9-year-old plants. Girdling experiments were carried out using almond trees growing in the orchard at CEBAS-CSIC. Sorghum seeds (Sorghum bicolor ‘SS1000’) were purchased from Agripro.
Plant material (leaf lamina, petiole, stem, tegument, nucellus plus endosperm, and cotyledon; three sample specimens of each) was weighed and immersed in boiling MeOH (80%, 500 µL, 5 min). The material was ground with a small pestle and filtered (0.22 µm low-binding Durapore membrane) after addition of lotaustralin (10 µg) as an internal standard. Analytical LC-MS was carried out on an Agilent 1100 Series LC (Agilent Technologies) coupled to a HCTplus ion trap mass spectrometer (Bruker Daltonics). The column was a Synergy Fusion-RP column (Phenomenex; 2.5 µm, 100 A, 2 x 50 mm), and the flow rate was 0.3 mL min–1. The mobile phases were as follows: (1) 0.1% (v/v) formic acid and 50 µM NaCl in water; and (2) 0.1% (v/v) formic acid in acetonitrile. The gradient program was as follows: (1) 0 to 7.5 min, linear gradient 6% to 19% (v/v); and (2) 7.5 to 10 min, linear gradient 19% to 100%. The mass spectrometer was run in positive ion mode. Traces of total ion current and of extracted ion currents for specific [M + Na]+ adduct ions were used to identify peaks. The retention time for lotaustralin, amygdalin, and prunasin was 1.9, 4.1, and 4.8 min, respectively.
Phloem transport of cyanogenic glucosides in peduncles of developing fruits and in the stems beneath the first-year shoots was monitored in girdling experiments. The epidermis and cambium cell layers including the phloem were removed by scalpel incisions (2-mm wide). The experiments (five replicates) were carried out in April and May using almond trees of the genotype ‘Ramillete’ (SkSk, sweet) and ‘S3067’ (sksk, bitter) growing at the experimental field at CEBAS-CSIC. Three days after the girdling took place, the sections positioned above and below the incision site (0.5-cm peduncle segments; 1.0-cm stem segments) were excised and boiled separately in MeOH (80%, 5 min). After filtering (0.22 µm low-binding Durapore membrane), the prunasin and amygdalin content of the MeOH extract was determined by LC-MS. As a reference (two replicates), corresponding segments were excised from peduncles and stems that had not been girdled. In parallel to the girdling experiments, the peduncles of developing fruits were cut and each of the peduncle stubs remaining on the tree were immediately immersed into a septum-covered Eppendorf tube filled with an agar (0.9%, w/v)/EDTA (20 mM, pH 6.0). After 3 d (five replicates), the cyanogenic glucoside content in the agar was extracted in MeOH (80%) and measured using LC-MS as previously described.
Leaf laminae, petioles, stem, tegument, nucellus plus endosperm, and cotyledons from the almond genotypes ‘Ramillete’ (SkSk, sweet) and ‘S3067’ (sksk, bitter) were obtained at different developmental stages throughout March to May and incubated in L-[14C]Phe (0.125–1.25 µCi, 321 mCi/mmol; Amersham Biosciences), NADPH (0.1 mM), and dithiothreitol (DTT; 1 mM). At the end of the incubation period (12 h, 20°C), the material was immersed in boiling MeOH (80%, 500 µL, 5 min), filtered (0.22-µm low-binding Durapore membrane), and aliquots (10 µL) were applied to silica gel 60 F254 TLC plates (Merck). Radiolabeled cyanogenic glucosides formed were separated by development in EtOAc/HOAc/MeOH/H2O (8:2.5:2.5:1, v/v) and monitored using a Storm 860 PhosphorImager (Molecular Dynamics). The position of prunasin and amygdalin was defined by the UV absorption of coapplied unlabeled authentic standards.
Plant material (3–100 g fresh weight of leaf laminae, roots, and fruit exocarp, endocarp, and mesocarp, tegument, nucellus plus endosperm, and cotyledon) from the genotypes ‘Ramillete’ (SkSk, sweet) and ‘S3067’ (sksk, bitter) was harvested in April and May and homogenized with 0.1 mass of polyvinylpolypyrrolidine in a buffer composed of 250 mM Suc, 100 mM Tricine (pH 7.9), 50 mM NaCl, 2 mM EDTA, and 2 mM DTT using mortar and pestle or mechanical food chopper as required. The homogenate was filtered through a nylon cloth (50-µm mesh) and centrifuged (10 min, 12,000 rpm, 4°C). Microsomes were recovered from the supernatant by centrifugation (60 min, 46,000 rpm) and resuspended in 50 mM Tricine (pH 7.9)/2 mM DTT. Microsomes were incubated (30 min, 30°C, total volume 20 µL) with 0.05 µCi L-[14C]Phe (321 mCi/mmol; Amersham Biosciences) in the presence or absence of 1 mM NADPH. An aliquot (10 µL) was applied to the silica gel 60 F254 TLC plates and radiolabeled metabolites formed were separated by development in toluene/EtOAc (5:1, v/v) and monitored using the Storm 860 PhosphorImager. Phenylacetaldoxime, phenylacetonitrile, and benzaldehyde were applied to the TLCs as reference compounds and their position located by their UV absorbance.
To analyze whether the almond plant material was containing inhibitors of the microsomal Cyt P450 system, microsomes were prepared from cohomogenized 3-d-old etiolated sorghum seedlings (12 g) and almond tissue (roots and teguments; 3 g) of the genotypes ‘Ramillete’ and ‘S3067’. In these experiments, microsomes were incubated using 0.05 µCi L-[U-14C]Tyr (443 mCi/ mmol; Amersham Biosciences) as substrate. A separate microsomal preparation from sorghum was used as a control. p-Hydroxyphenylacetaldoxime, p-hydroxyphenylacetonitrile, and p-hydroxybenzaldehyde were used as reference compounds (Møller et al., 1977
Enzyme extracts from leaf laminae, petioles, peduncles, nucellus plus endosperm, and cotyledons (100- to 500-mg tissue) of the following four genotypes ‘Ramillete’ (SkSk, sweet), ‘Marcona’ (Sksk, sweet), ‘Garrigues’ (Sksk, slightly bitter), and ‘S3067’ (sksk, bitter) were prepared every second week throughout the entire growth season (March to August) by homogenization (Eppendorff tube, pestle, 4°C) in 250 mM Suc, 100 mM Tris-HCl (pH 7.5), 50 mM NaCl, 2 mM EDTA, 5% (w/v) polyvinylpolypyrrolidone, 200 µM phenylmethylsulfonyl fluoride, and 6 mM DTT (total volume, 0.5 mL). The supernatants were collected after centrifugation (20,000g, 20 min) and aliquots (5 µL) were incubated (total volume, 20 µL) with 20 mM acceptor (prunasin or mandelonitrile), 0.025 µCi [U-14C] UDPG (200 mCi/mmol; Amersham Biosciences), and 25 mM
In April, fruit samples (tegument, nucellus, and cotyledon) from the genotypes ‘Ramillete’ and ‘S3067’ were imbedded in plastic according to the manufacturers manual for Technovit 8100 (Heraeus) with minor alterations. The tissues were dehydrated in a graded series of acetone solutions (25%, 50%, and 100%, v/v, 1 h each) and left overnight in the filtration solution. Sections (6 µm) were cut on a Reichert-Jung 2030 rotary microtome (Reichert-Jung).
The sections were stained for different time periods (1–10 min) with Fast Blue BB salt with and without the substrate 6-bromo-2-napthyl-β-D-glucopyranoside (Spielman and Mowshowitz, 1982
We thank Teresa Cremades Rosado and Mariano Gambín for technical help. Received November 13, 2007; accepted December 29, 2007; published January 11, 2008.
1 This work was supported by the Danish National Research Foundation (to the Center for Molecular Plant Physiology) and the Spanish Ministry of Education and Science, including a postdoctoral fellowship (to R.S.-P.).  The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Birger Lindberg Møller (blm{at}life.ku.dk).
[C] Some figures in this article are displayed in color online but in black and white in the print edition.
[OA] Open Access articles can be viewed online without a subscription. www.plantphysiol.org/cgi/doi/10.1104/pp.107.112979 * Corresponding author; e-mail blm{at}life.ku.dk.
Akazawa T, Miljanich P, Conn EE (1960) Studies on cyanogenic glucoside of Sorghum vulgare. Plant Physiol 35: 535–538 Álvarez J (1798) Diccionario Universal de Agricultura. Imprenta Real, Madrid Andersen MD, Busk PK, Svendsen I, Møller BL (2000) Cytochromes P-450 from cassava (Manihot esculenta Crantz) catalyzing the first steps in the biosynthesis of the cyanogenic glucosides linamarin and lotaustralin. Cloning, functional expression in Pichia pastoris, and substrate specificity of the isolated recombinant enzymes. J Biol Chem 275: 1966–1975 Arrázola G (2002) Análisis de glucósidos cianogénicos en variedades de almendro: implicaciones en la mejora genética. PhD thesis. Universidad de Alicante, Alicante, Spain Bak S, Kahn RA, Nielsen HL, Møller BL, Halkier BA (1998) Cloning of three A-type cytochromes P450, CYP71E1, CYP98, and CYP99 from Sorghum bicolor (L.) Moench by a PCR approach, and identification by expression in Escherichia coli of CYP71E1 as the oxime-metabolizing cytochrome P450 in the biosynthesis of the cyanogenic glucoside dhurrin. Plant Mol Biol 36: 393–405[CrossRef][ISI][Medline] Bak S, Olsen CE, Halkier BA, Møller BL (2000) Transgenic tobacco and Arabidopsis plants expressing the two multifunctional sorghum cytochrome P450 enzymes, CYP79A1 and CYP71E1, are cyanogenic and accumulate metabolites derived from intermediates in dhurrin biosynthesis. Plant Physiol 123: 1437–48 Bak S, Paquette SM, Morant M, Rasmussen AV, Saito S, Bjarnholt N, Zagrobelny M, Jørgensen K, Hamann T, Osmani S, et al (2006) Cyanogenic glycosides; a case study for evolution and application of cytochromes P450. Phytochem Rev 5: 309–329[CrossRef] Bliss FA, Arulsekar S, Foolad MR, Becerra AM, Gillen A, Warburton ML, Dandekar AM, Kocsisne GM, Mydin KK (2002) An expanded genetic linkage map of Prunus based on an interspecific cross between almond and peach. Genome 45: 520–529[Medline] Conn EE (1969) Cyanogenic glycosides. J Agric Food Chem 17: 519–526[CrossRef][ISI] Conn EE (1980) Cyanogenic compounds. Annu Rev Plant Physiol 31: 433–451[ISI] Davis RH (1991) Cyanogens. In JPF D'Mello, CM Duffus, JH Duffus, eds, Toxic Substances in Crop Plants. Royal Society of Chemistry, Cambridge, UK, pp 202–225 Dicenta F, García JE (1993) Inheritance of the kernel flavour in almond. Heredity 70: 308–312[CrossRef][ISI] Dicenta F, Martínez-Gómez P, Grané N, Martín ML, León A, Cánovas JA, Berenguer V (2002) Relationship between cyanogenic compounds in kernels, leaves, and roots of sweet and bitter kernelled almonds. J Agric Food Chem 50: 2149–2152[CrossRef][Medline] Dicenta F, Ortega E, Martinez-Gomez P (2007) Use of recessive homozygous genotypes to assess the genetic control of kernel bitterness in almond. Euphytica 153: 221–225[CrossRef][ISI] Floss HG, Hadwiger L, Conn EE (1965) Enzymatic formation of β-cyanoalanine from cyanide. Nature 208: 1207–1208[CrossRef][Medline] Frehner M, Scalet M, Conn EE (1990) Pattern of the cyanide-potential in developing fruits. Plant Physiol 94: 28–34 Girona J, Marsal J (1995) Estrategias de RDC en almendro. In M Zapata, P Segura, eds, Riego Deficitario Controlado. Fundamentos y Aplicaciones. MundiPrensa, Madrid, pp 99–118 Halkier BA, Møller BL (1989) Biosynthesis of the cyanogenic glucoside dhurrin in seedlings of Sorghum bicolor (L.) Moench and partial purification of the enzyme system involved. Plant Physiol 90: 1552–1559 Hansen KS, Kristensen C, Tattersall DB, Jones PR, Olsen CE, Bak S, Møller BL (2003) The in vitro substrate regiospecificity of recombinant UGT85B1, the cyanohydrin glucosyltransferase from Sorghum bicolor. Phytochemistry 64: 143–151[CrossRef][Medline] Heppner J (1923) The factor for bitterness in the sweet almond. Genetics 8: 390–392 Heppner J (1926) Further evidence on the factor for bitterness in the sweet almond. Genetics 11: 605–606 Jenrich R, Trompetter I, Bak S, Olsen CE, Møller BL, Piotrowski M (2007) Evolution of heteromeric nitrilase complexes in Poaceae with new functions in nitrile metabolism. Proc Natl Acad Sci USA 104: 18848–18853 Jones DA (1988) Cyanogenesis in animal-plant interactions. In D Evered, S Harnett, eds, Cyanide Compounds in Biology. Ciba Foundation Symposium. John Wiley & Sons, Chichester, UK, pp 151–170 Jones PR, Møller BL, Hoj PB (1999) The UDP-glucose:p-hydroxymandelonitrile-O-glucosyltransferase that catalyzes the last step in synthesis of the cyanogenic glucoside dhurrin in Sorghum bicolor. Isolation, cloning, heterologous expression, and substrate specificity. J Biol Chem 274: 35483–35491 Joobeur T, Viruel MA, de Vicente MC, Jáuregui B, Ballester J, Dettori MT, Verde I, Truco MJ, Messeguer R, Battle I, et al (1998) Construction of a saturated linkage map for Prunus using an almond x peach F2 progeny. Theor Appl Genet 97: 1034–1041[CrossRef][ISI] Jørgensen K, Bak S, Busk PK, Sørensen C, Olsen CE, Puonti-Kaerlas J, Møller BL (2005) Cassava plants with a depleted cyanogenic glucoside content in leaves and tubers. Distribution of cyanogenic glucosides, their site of synthesis and transport, and blockage of the biosynthesis by RNA interference technology. Plant Physiol 139: 363–374 Kahn RA, Bak S, Svendsen I, Halkier BA, Møller BL (1997) Isolation and reconstitution of cytochrome P450ox and in vitro reconstitution of the entire biosynthetic pathway of the cyanogenic glucoside dhurrin from sorghum. Plant Physiol 115: 1661–1670[Abstract] Kester DE, Gradziel M (1996) Almonds. In J Janick, JN Moore, eds, Fruit Breeding, Vol 3. J Wiley and Son, New York, pp 1–97 King NLR, Bradbury JH (1995) Bitterness of cassava: identification of a new apiosyl glucoside and other compounds that affect its bitter taste. J Sci Food Agric 68: 223–230[CrossRef][ISI] Koch B, Nielsen VS, Halkier BA, Olsen CE, Møller BL (1992) The biosynthesis of cyanogenic glucosides in seedlings of cassava (Manihot esculenta Crantz). Arch Biochem Biophys 292: 141–50[CrossRef][ISI][Medline] Kriechbaumer V, Park WJ, Piotrowski M, Meeley RB, Gierl A, Glawischnig E (2007) Maize nitrilases have a dual role in auxin homeostasis and β-cyanoalanine hydrolysis. J Exp Bot 58: 4225–4233 Kristensen C, Morant M, Olsen CE, Ekstrøm CT, Galbraith DW, Møller BL, Bak S (2005) Metabolic engineering of dhurrin in transgenic Arabidopsis plants with marginal inadvertent effects on the metabolome and transcriptome. Proc Natl Acad Sci USA 102: 1779–1784 Lechtenberg M, Nahrstedt A (1999) Cyanogenic glycosides. In R Ikan, ed, Naturally Occurring Glycosides. John Wiley and Sons, Chichester, UK, pp 147–191 Lieberei R, Selmar D, Biel B (1985) Metabolism of cyanogenic glucosides in Hevea brasiliensis. Plant Syst Evol 150: 49–63[CrossRef] McCarty CD, Leslie JW, Frost HB (1952) Bitterness of kernels of almond x peach hybrids and their parents. Proc Am Soc Hortic Sci 59: 254–258 McFarlane IJ, Lees EM, Conn EE (1975) The in vitro biosynthesis of dhurrin, the cyanogenic glycoside of Sorghum bicolor. J Biol Chem 250: 4708–4713 Mentzer C, Favre-Bonvin J (1961) Sur la biogenèse du glucoside cyanogénétique des feuilles de laurier-cerise (Prunus lauro-cerasus). C R Acad Sci Ser III Sci Vie 253: 1072–1074 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
RESULTS
), ‘Marcona’ (Sksk, sweet; 
), ‘Garrigues’ (Sksk, slightly bitter; 
), and ‘S3067’ (sksk, bitter; x). Measurements were made every second week and initiated March 14, 2007 and ended August 15, 2007. Marcona ripened last and therefore this genotype was the only one analyzed on August 15. B, Fruit anatomy of the four almond genotypes during the growth period: In March, exocarp (x), green mesocarp (m), a soft endocarp (n), tegument (t), and the nucellus (u) are visible. In April, the formation of cotyledons (c) is apparent in the earliest varieties. In May, cotyledons (c) and endosperm (n) are being formed at the expense of the nucellus (u). In August, at the end of the ripening season, the cotyledons (c) occupy the entire seed; the endocarp had fully hardened and fragmented during opening. This is the reason the fruits shown appear smaller. [See online article for color version of this figure.]
6)-β-D-glucopyranoside are also main contributors to the bitterness (King and Bradbury, 1995
-gluconolactone (β-glucosidase inhibitor) in 100 mM Tris-HCl (pH 7.5). After incubation (10 min, 30°C), aliquots (10 µL) were applied to the silica gel 60 F254 TLC plates and radiolabeled products separated by development EtOAc/HOAc/MeOH/H2O (8:2.5:2.5:1, v/v) and monitored using the Storm 860 PhosphorImager. The position of prunasin and amygdalin on the TLCs was defined by coapplication of the unlabeled authentic standards.