JXB Advance Access originally published online on July 2, 2004
Journal of Experimental Botany 2004 55(404):1903-1918; doi:10.1093/jxb/erh138
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Biosynthesis of the flavour precursors of onion and garlic
School of Biological Sciences, The Biosciences Building, The University of Liverpool, Liverpool L69 7ZB, UK
* To whom correspondence should be addressed. Fax: +44 (0)151 795 4410. E-mail: m.g.jones{at}liv.ac.uk
Received 12 January 2004; Accepted 26 February 2004
 |
Abstract |
|---|
Top
Abstract
IntroductionOnion (Allium cepa), garlic (A. sativum) and other Alliums are important because of the culinary value of their flavours and odours. These are characteristic of each species and are created by chemical transformation of a series of volatile sulphur compounds generated by cleavage of relatively stable, odourless, S-alk(en)yl cysteine sulphoxide flavour precursors by the enzymes alliinase and lachrymatory-factor synthase. These secondary metabolites are S-methyl cysteine sulphoxide (MCSO, methiin; present in most Alliums, some Brassicaceae), S-allyl cysteine sulphoxide (ACSO, alliin; characteristic of garlic), S-trans-prop-1-enyl cysteine sulphoxide (PECSO, isoalliin; characteristic of onion), and S-propyl cysteine sulphoxide (PCSO, propiin; in onion and related species). Information from studies of the transformation of putative biosynthetic intermediates, radiolabelling, and from measurements of sulphur compounds within onion and garlic have provided information to suggest a biosynthetic pathway. This may involve alk(en)ylation of the cysteine in glutathione, followed by cleavage and oxidation to form the alk(en)yl cysteine sulphoxide flavour precursors. There is also evidence that synthesis of the flavour precursors may involve (thio)alk(en)ylation of cysteine or a precursor such as O-acetyl serine. Both routes may occur depending on the physiological state of the tissue. There are indications from the effects of environmental factors, such as the availability of sulphur, that control of the biosynthesis of each flavour precursor may be different. Cysteine and glutathione metabolism are discussed to indicate parallels with Allium flavour precursor biosynthesis. Finally, possible avenues for exploration to determine the origin in planta of the alk(en)yl groups are suggested.
Key words:
Alliaceae, alliinase, S-alk(en)yl cysteine sulphoxide, cysteine synthase, flavour precursor biosynthesis, garlic, glutathione metabolism, 
-glutamyl alk(en)yl cysteine sulphoxide, onion
|
Introduction |
|---|
The Alliums are a large genus containing about 700 species, including economically important vegetables and flowering ornamentals as well as wild species from Europe, Asia, and the Americas (Fenwick and Hanley, 1985
). Their flavours and odours are easily recognized in vegetables such as Allium cepa L. (onion), Allium sativum L. (garlic), Allium porrum L. (leek), Allium schoenoprasum L. (chives), and Allium fistulosum L. (bunching or Welsh onion). Many other members of the genus are grown for the ornamental value of their flowers. The majority grow as biennials, producing an underground storage bulb at the end of the first growing season, which flowers in the next. Many produce seed, although garlic generally does not. They are all odourless until the tissue is damaged, at which point all generate the volatile and reactive sulphur-containing chemicals that cause their best-known characteristic.
The nature and origin of these flavour compounds, particularly in onion and garlic, have been studied since the 1940s. It soon became apparent that once the plant tissue is damaged, comparatively stable flavour precursors are cleaved to give a series of volatile sulphur compounds that undergo further vapour-phase chemical transformations. There have been extensive investigations and reviews of these gaseous chemicals (Whitaker, 1976; Block, 1992) and they will not be discussed further here. Instead, this review will focus on current knowledge about the biosynthesis of the flavour precursors, the (+)-S-alk(en)yl cysteine sulphoxides (CSOs) and their -glutamyl peptide (GPs) relatives. Although a biosynthetic pathway has been published (Granroth, 1970; Lancaster and Shaw, 1989), there is still considerable uncertainty about several stages, the relationship between the CSOs and GPs, and whether the same pathway is followed in all tissues.
Investigation has centred around the role of GPs in the biosynthetic pathway, and the origin of the alk(en)yl substituents of cysteine. There has also been considerable interest in whether synthesis of the compounds is interlinked, and how it is regulated. Very little is known about some aspects, such as the oxidation step required to form the sulphoxide. There is some information about the subcellular location of flavour biosynthesis, and about the movement of flavour precursors and GPs during plant development. After listing the flavour precursors that have been identified in Alliums, this review will briefly describe similar compounds that have been found in other plants, suggesting that this biosynthetic pathway may be one example of a more widely distributed secondary metabolic pathway. This will be followed by an outline of the evidence for GPs as intermediates in Allium flavour precursor biosynthesis, and the source of the S-alk(en)yl groups of the precursors. Studies into the effects of environmental factors on flavour precursor biosynthesis will also be considered. Finally, information about the biosynthesis of cysteine, reactions in which glutathione participates, and the sources of alkyl donors within other plants will be reviewed, since these are the primary pathways that have parallels with Allium flavour precursor biosynthesis.
The biosynthetic pathways proposed for the Allium flavour precursors are based primarily on chemical analysis and radiotracer studies. Most of the enzyme activities that would be required for the proposed biosynthetic steps are from large protein families where the in vivo tissue, developmental, and substrate specificity of most members has still to be established. The application of molecular genetic techniques and plant transformation has provided insights into many aspects of plant metabolism and, in the future, these approaches may illuminate new avenues to improve knowledge of this group of secondary metabolites of economic, medical, and culinary importance.
|
The flavour precursors of Alliums |
|---|
Four non-volatile, odourless CSOs (Table 1) are the precursors of the flavour and odours of the Alliums. These are S-methyl cysteine sulphoxide (MCSO, methiin; present in most Alliums, some Brassicaceae), S-allyl cysteine sulphoxide (ACSO, alliin; characteristic of garlic), S-trans-prop-1-enyl cysteine sulphoxide (PeCSO, isoalliin; characteristic of onion), and S-propyl cysteine sulphoxide (PCSO, propiin; in onion and related species). The enzyme alliinase (EC 4.4.1.4 [EC] ) cleaves these precursors to give pyruvate, ammonia, and a thiosulphinate. The latter undergoes further reactions so that the smell of Alliums changes over time (Whitaker, 1976
). Degradation of the most widely distributed of these flavour precursors, MCSO, gives odours that are generally described as ‘cabbagy’ or ‘fresh onion’, while the easily distinguished smell of garlic originates in a similar way from ACSO. The lachrymatory effect that is characteristic of onions is caused by the volatile product propanthial S-oxide (Brodnitz and Pascale, 1971). Until recently, this was thought to form spontaneously from the thiosulphinate products of the alliinase reaction. However, it is generated by the activity of a second enzyme, lachrymatory-factor synthase, following alliinase action on PeCSO, the major flavour precursor of onion (Imai et al., 2002). The fourth flavour precursor, PCSO, is also found in onion as well as several other Alliums. Although there are reports of the occurrence of other (+)-S-alk(en)yl cysteine sulphoxides in Alliums, several (e.g. 5-methyl-1,4-thiazan-3-carboxylic acid 1-oxide (cycloalliin)) (Whitaker, 1976) are probably formed by chemical transformation during analytical procedures rather than synthesized in vivo. Others, for example (+)-S-ethyl cysteine sulphoxide (Kubec et al., 2000), are only very minor constituents or have not been detected by most workers. Nevertheless, there are a few reports of further alk(en)yl cysteine sulphoxides as the major flavour precursors in ornamental Alliums such as S-methylthiomethyl cysteine-4-oxide (marasmine) in Tulbaghia violacea Harv. (society garlic) (Kubec et al., 2002a) and S-butyl cysteine sulphoxide in Allium siculum Ucria. (honey garlic) (Kubec et al., 2002b).
|
As well as CSOs, several
-glutamyl peptide (GP) derivatives of these flavour compounds have been detected within the Alliums (Whitaker, 1976). Over 17 types have been isolated (Granroth, 1970) and including -glutamyl-S-alk(en)yl glutathiones, -glutamyl-S-alk(en)yl cysteines and -glutamyl-S-alk(en)yl cysteine sulphoxides, all proposed to derive from glutathione (-glutamyl cysteinyl glycine). Although they do not appear to contribute directly to flavour, the current view is that they are intermediates in biosynthesis and may also act as reserves of nitrogen and sulphur. Their role in the biosynthetic pathway will be discussed below. |
Allium odours from other species |
|---|
Plants produce a very large number of small molecular weight compounds, traditionally called secondary metabolites with functions in key activities including reproduction, defence, pathogenicity, stress resistance, and resource storage. The biosynthetic pathways for most secondary metabolites have not been described in detail and many are obviously complicated (Wittstock and Halkier, 2002
). The suggestion that 15–25% of each plant genome is devoted to secondary metabolism is another indication of its importance (Pichersky and Gang, 2000). One of the challenges in the study of secondary metabolism is, therefore, to unravel which aspects of the synthesis, storage, function, and degradation of a particular secondary metabolite share pathways that are parallel to primary metabolism and common across genera, and which are unique to a species or genus. The areas of metabolism that are involved in biosynthesis of the flavour precursors of the Alliums, namely metabolism of cysteine and glutathione, are ones that are crucial in essential processes such as sulphur uptake by plants, redox homeostasis, and xenobiotic detoxification. The flavour compounds of Alliums may be where this capacity has been exploited to produce new secondary metabolites, and are thus particularly interesting as an example of a group of secondary metabolites that are not dissimilar from several primary metabolites.
The distinctive Allium smell has been reported from plants in dicotyledonous genera, and also in fungi. In several instances, chemical analysis has identified compounds that are similar to the flavour precursors of Alliums, suggesting that they may be one particularly well-studied and exploited example of a more widespread secondary metabolic activity. MCSO is present in members of the Brassicaceae. It has been identified in all the Brassica oleracea L. vegetables, such as cabbage, broccoli, and cauliflower (Stoewsand, 1995) and the genus Raphanus L. as well as the model plant Arabidopsis thaliana Heynh. (Kubec et al., 2001).
Compounds related to the Allium flavour precursors have also been identified in several tropical plants. A smell of garlic emanates from the leaves, and especially the roots, of Petiveria alliacea L. (family Phytolaccaceae; anamu, garlic weed), a perennial shrub indigenous to the Amazon rainforest and other tropical regions. The compound S-benzyl cysteine sulphoxide has been isolated from fresh roots of this plant (Kubec and Musah, 2001). The tree Scorodocarpus borneensis Becc. (family Olacaceae; sindu, wood garlic), grows in Malaysia and Borneo and also emits a strong smell of garlic when damaged. The amino acid derivatives S-methylthiomethyl cysteine-4-oxide and S-methylthiomethyl cysteine were isolated from its fruit at levels similar to that of MCSO in Brassica sp. (Kubota et al., 1998). They could be cleaved by a broccoli cysteine sulphoxide lyase (C-S lyase) to yield pyruvate and volatile sulphur compounds with a garlic odour. Interestingly, marasmine, and -glutamyl-marasmine, were first isolated from fruiting bodies of several species of basidiomycete (Marasmius alliaceus (Jacq. ex Fr.) Fr., M. scorodonius Fr., and M. prasiosmus Fr.) that develop a strong garlic-like odour when crushed (Gmelin et al., 1976) and marasmine has also been identified in one species of Allium, Tulbaghia violacea, although the configuration of the sulphoxide group was different in the basidiomycete (Kubec et al., 2002a).
The presence of S-methyl cysteine and -glutamyl-methyl cysteine, but not their sulphoxides, has been noted in Fabaceae species (Ellis and Salt, 2003), including members of the Phaseolus L. and Vigna Savi. genera as well as Melilotus indicus L. These genera may use S-methyl cysteine to transport reduced sulphur and -glutamyl-methyl cysteine for sulphur storage in seeds, since their seed proteins are low in sulphur-containing amino acids. The selenium hyperaccumulator Astragalus bisulcatus A. Grey produces a selenium analogue of S-methyl cysteine to prevent non-specific incorporation of selenocysteine into proteins or conversion into selenomethionine, thereby conferring selenium tolerance on the plant. The seeds accumulate -glutamyl-methyl selenocysteine.
|
Allium genomes |
|---|
Genomes within the Alliums are particularly large. Estimates of the 2C DNA amounts per genome in 75 Allium species ranged from 16.93–63.57 pg (Ohri and Pistrick, 2001
). Both onion and garlic are diploid (2n=16). The nuclear genome of onion is estimated to be around 15 290 Mbp per 1C, of which at most 6% is single-copy DNA, and garlic is a similar size (c. 15 901 Mbp). The GC content of onion DNA is 32%, one of the lowest for any angiosperm, but within coding regions is higher, approximately equal to Arabidopsis thaliana (Kuhl et al., 2004). There is no cytogenetic evidence for a polyploid origin, and molecular studies of Allium genomes are limited. In one study, a low-density genetic map of onion with 116 markers, based primarily on restriction fragment length polymorphisms (King et al., 1998), indicated that duplicated loci were present more frequently than would be expected for a diploid. Among the duplicated genes identified and mapped by King and co-workers were an alliinase and a glutathione-S-transferase as well as members of other gene families. McCallum et al. (2002) have more recently cloned onion homologues of key genes in the sulphur assimilation pathway, and a set of over 10 000 onion expressed sequence tags has recently become available (Kuhl et al., 2004). Although assignments can be made based on shared motifs, it is at present impossible to assign the exact function of most members of gene families without individual biochemical and molecular biological investigation. Application of information from other plants to the biosynthesis of Allium flavour precursors may, therefore, be useful, but will require validation within Alliums. |
The role of flavour compounds in Alliums |
|---|
One of most intriguing questions about the flavour compounds is their role within Alliums. The levels of non-protein cysteine and glutathione derivatives amount to 1–5% dry weight (Lancaster and Kelly, 1983
), indicating that this is a major biosynthetic activity within the plant. The two roles that have been ascribed are for defence against pests and predation, particularly in the overwintering bulb, and for carbon, nitrogen, and sulphur storage and transport (Lancaster and Boland, 1990). There has been extensive investigation of the antimicrobial properties of Allium extracts in vitro, reviewed by Ankri and Mirelman (1999), but there are fewer in vivo examples, especially studies addressing how the flavour compounds confer advantage on Alliums. Disease resistance has not been a priority during breeding of current commercial onion varieties so the impact of flavour compounds is difficult to assess (B Smith, personal communication). In general, mild flavoured onions are reported to have poorer storage properties.
However, several species of insect and fungi have acquired specialist interactions with Alliums where the flavour volatiles are attractants to specialized feeders and pathogens. Field experiments on Sclerotium cepivorum Berk., the causal agent of white rot, one of the most economically significant fungal diseases of onions and garlic, indicated a strong dependence on the presence of the flavour volatiles for germination of sclerotia (Coley-Smith, 1986). Similarly, the leek moth Acrolepiopsis assectella Zeller and the onion fly Delia antiqua Meigen are both attracted to their respective host plants by the flavour volatiles (Romeis et al., 2003). In the case of white rot, there are no truly resistant onion varieties and experimental infections can be difficult to establish. Some of this variability can be attributed to the onion, but also to environmental conditions and the fungus (Earnshaw et al., 2000).
One interesting observation on the role of the flavour precursors comes from an experiment by Parkin and Thomas (1996) on the effect of elicitors on flavour precursor levels in differentiating onion callus cultures. Elicitors, which are fragments of plant pathogens, plant tissue, or signalling molecules, can trigger the production of defence compounds when presented to tissue cultures (Collin, 2001). In their experiments, onion and fungal cell wall fragments, as well as the signal molecule salicylic acid, suppressed production of flavour precursors. This is the opposite of what might be expected if the flavour compounds are involved in plant defence.
|
Biosynthesis of the flavour precursors of Alliums |
|---|
Investigation of the biosynthesis of Allium flavour compounds started with the discovery by Stoll and Seebeck (1947)
that ACSO was present within garlic tissue as a stable source of diallyl thiosulphinate (allicin), the major flavour volatile. They demonstrated that the conversion of ACSO to allicin required the enzyme alliinase. Two further cysteine sulphoxide derivatives, MCSO and PCSO, were identified in onion by Virtanen and Matikkala (1959) and a fourth cysteine sulphoxide, PeCSO, was soon identified within onion (Virtanen and Spåre, 1961). It also became apparent that Alliums contained significant amounts of -glutamyl derivatives of these compounds (reviewed in Whitaker, 1976).
Lancaster and colleagues have proposed a pathway (Lancaster and Shaw, 1989; Lancaster et al., 1989; Randle et al., 1995) requiring GPs as intermediates, which has become accepted as the biosynthetic route (Block, 1992; Prince et al., 1997). The pathway is based primarily on radiolabelling studies and analysis of Allium tissues. It proposes (Fig. 1A) that the biosynthesis of the flavour precursors in Alliums proceeds via S-alk(en)ylation of the cysteine in glutathione, followed by transpeptidation to remove the glycyl group, oxidation to the cysteine sulphoxide, and, finally, removal of the glutamyl group to yield CSOs. This pathway has parallels with glutathione degradation (Leustek et al., 2000) that will be discussed later. An alternative biosynthetic route (Fig. 1B) omits glutathione in favour of direct alk(en)ylation of cysteine or thioalk(en)ylation of O-acetyl serine followed by oxidation to a sulphoxide. This also has parallels, in the production of several secondary metabolites mediated by cysteine synthase (CS, EC 4.2.99.8
[EC]
also known as O-acetyl serine thiol lyase) (Ikegami and Murakoshi, 1994). In both of the pathways, few of the proposed biosynthetic enzymes from Alliums have been studied in detail and their roles are inferred from other systems. The relative contribution of both pathways in all tissues and throughout the life history is also unclear and the source of the alk(en)yl groups remains to be resolved.
|
One technical problem in experiments that utilize damaged Allium tissues, such as cut leaves or bulb segments, is that alliinase will be liberated, and may cleave newly formed flavour precursors. In addition, peptidases and
-glutamyl transpeptidases (EC 2.3.2.2
[EC]
) are present in onion and garlic tissues (Lancaster and Shaw, 1991; Ceci et al., 1992; Hanum et al., 1995) providing further opportunities for interchange among the biosynthetic intermediates of the flavour precursors. This may explain some of the divergence in results between different research groups. Some studies have attempted to avoid tissue damage by following the progress of 35S-sulphate supplied to the roots of seedlings or sprouting bulbs. Another strategy uses tissue culture, where the thin cuticle allowed potential biosynthetic intermediates to be administered to undamaged, but undifferentiated, cells. |
The role of -glutamyl peptides in Allium flavour precursor biosynthesis
|
|---|
Lancaster and Shaw (1989)
investigated the biosynthetic relationship between the GPs and CSOs through following the fate of a short pulse of 35S-sulphate in leaves excised from onion seedlings, sprouting garlic cloves, and bulbs of the ornamental Allium siculum. This study formed the basis of the proposal that GPs are intermediates in CSO biosynthesis. Glutathione, -glutamyl cysteine, -glutamyl-propenyl cysteine sulphoxide, methyl glutathione, and S-2-carboxypropyl glutathione, along with several unidentified compounds, became labelled within the first 15 min while it took 6 h for the radiolabel to appear in the CSOs. Although some 35S was detected in glutathione throughout the experiment, no radiolabel was detected in the GPs after the first day. In onion leaves, MCSO behaved differently from PeCSO and PCSO in that the amount of label remained unchanged in measurements made after 7 d, while, in PeCSO, labelling had increased 8-fold and in PCSO 5-fold by this time. 35S-labelled S-2-carboxypropyl glutathione was detected in onion and garlic but not in A. siculum.
The pulse of radioactive sulphur was therefore incorporated first into glutathione and GPs and then the CSOs. The appearance of radiolabelled sulphate in glutathione is not unexpected since sulphate is assimilated into cysteine and it has been estimated that one-third of this is immediately channelled into reactions of the glutathione cycle. Although MCSO is present in onion and garlic, it was the major flavour precursor detected in A. siculum by Lancaster and Shaw (1989). The presence of labelled S-2-carboxypropyl glutathione in onion and garlic but not A. siculum suggested that it was an intermediate in the biosynthesis of the allyl, propenyl, and propyl groups of the CSOs (Fig. 2), in agreement with Granroth (1970). A similar study (Edwards et al., 1994) using intact sprouting onion sets indicated that the pattern of incorporation was different in undamaged tissues, with the majority of the radioactivity occurring in the CSOs with little indication that the label had been in GPs prior to entering the CSO fraction. The differences seen during these two studies may be caused by metabolic differences between intact plants and detached senescent leaves. These experiments, nevertheless, provided a persuasive biosynthetic route and an explanation of the relationship between several of the sulphur compounds found in Alliums. However, the biosynthesis in different tissues and developmental states, the origin of the alk(en)yl groups, and also the relationship between CSOs and GPs warranted further investigation.
|
|
The origin of the alk(en)yl substituents |
|---|
Cysteine is known to react chemically with methacrylic acid to give S-2-carboxypropyl cysteine, so providing a route to the alk(en)yl groups in ACSO, PeCSO, and PCSO. Granroth (1970)
had already shown that 14C-valine gave rise to radiolabelled methacrylic acid, which could react with glutathione to produce S-2-carboxypropyl glutathione. In vivo, methacrylyl CoA is an intermediate in valine catabolism (Zolman et al., 2001), providing a source of this material within the cell. The suggestion from these data was that all the alk(en)yl side-chains in the CSOs (except methyl) were derived from a S-2-carboxypropyl group through decarboxylation and reduction as indicated in Fig. 2. As further evidence for this route in vivo, both Granroth (1970) and Suzuki et al. (1962) identified incorporation of 14C-valine into S-2-carboxypropyl cysteine and S-2-carboxypropyl glutathione. In addition, Granroth (1970) confirmed that onion could hydrolyse S-2-carboxypropyl glutathione and that, in onion leaf tissue, S-2-carboxypropyl cysteine was rapidly converted to PeCSO, although without incorporation of radiolabel in S-2-carboxypropyl glutathione.
Experiments to characterize the ability of onion to convert S-2-carboxypropyl cysteine to PeCSO were also carried out by Parry and colleagues. Radiolabelling of the sulphur atom and propyl group in S-2-carboxypropyl cysteine showed that onion plants converted it to PeCSO without significant alteration (Parry and Sood, 1989). These experiments also provided information about the oxidative decarboxylation of S-2-carboxypropyl cysteine. A tritium at C-2 of the carboxy-propyl group was retained in PeCSO, while tritium at C-3 was lost. This indicated that a 3-keto-S-2-carboxypropyl cysteine derivative was not an intermediate in the decarboxylation reaction. Subsequent experiments, where S-2-carboxypropyl cysteine was stereospecifically tritiated at C-3, proved that the configuration of the reaction was the same as in decarboxylations associated with porphyrin and terminal alkene biosynthesis in plants (Parry and Lii, 1991).
There are indications that the incorporation of sulphur into the CSOs and oxidation to the sulphoxide are carried out by enzymes with a broad substrate specificity. The biosynthesis of cysteine, by addition of sulphur to O-acetyl serine, requires members of the cysteine synthase family, several members of which are known to have a role in secondary metabolism (Ikegami and Murakoshi, 1994). Although cysteine may be alk(en)ylated while within the glutathione tripeptide, there is strong evidence that it can also occur during synthesis of the amino acid alone. Incubation of onion or garlic leaves with 14C-serine and a series of thiols (including ethanethiol, benzenemethanethiol, and 2-thioethanol) led to the synthesis of the corresponding radiolabelled substituted cysteines (Granroth, 1970). In general agreement with this work, experiments by Turnbull et al. (1981) showed that infiltration of both 14C-serine and 14C-cysteine into sprouting onion leaves led to incorporation of radioactivity into both MCSO and PeCSO. Addition of serine and either 2-propenethiol or ethanethiol to onion root cultures resulted in the synthesis of ACSO and S-ethyl CSO (Prince et al., 1997), neither of which would normally be synthesized by onion. However, the observation that supplying radiolabelled cysteine to onion tissue (Granroth, 1970) always resulted in a rapid formation of PeCSO, suggests that this originated directly from cysteine. If thioalkyl transfer to O-acetyl serine were the only source for CSO biosynthesis, radiolabels would be unlikely to enter CSOs from cysteine.
Evidence for the general nature of the oxidation step comes from several studies that demonstrate oxidation of exogenous S-alk(en)yl cysteines to the corresponding sulphoxide. Granroth (1970) showed that S-methyl cysteine, S-ethyl cysteine, S-propyl cysteine, and S-propenyl cysteine could all be oxidized to the corresponding sulphoxide by onion leaf tissue. More recent experiments reinforce this finding. For example, tissues from onion, chives, and bunching onion were all able to convert S-allyl cysteine to ACSO, even though these species do not produce ACSO (Ohsumi et al., 1993), with indications that the oxidation was mediated stereospecifically by an enzyme. The oxidase proposed by Lancaster and Shaw (1989) to act on -glutamyl-S-alk(en)yl cysteines may, therefore, also recognize S-alk(en)yl cysteines.
|
Control of the biosynthesis of flavour precursors: biosynthesis in tissue culture |
|---|
The biosynthetic pathway for the flavour precursors in onion and garlic does not appear to be confined to a particular tissue or type of cell. Synthesis of secondary metabolites by the undifferentiated cells in many plant cell or tissue cultures is generally much lower than in the intact plant, but manipulation of growth conditions can stimulate significant increases in yield (Collin, 2001
). Studies using onion and garlic tissue cultures may, therefore, provide insight into whether addition of any proposed biosynthetic intermediates can stimulate steps in CSO biosynthesis. All studies indicate that alliinase is present in onion tissue cultures (Davey et al., 1974), but the major flavour precursors are absent, or present at only low levels in undifferentiated callus. Undifferentiated, colourless callus of onion and garlic contained less than 10% of the level of flavour precursors in onion bulbs and was almost entirely MCSO (Selby et al., 1979; Lancaster et al., 1988). Trace amounts of PeCSO could be detected in onion callus after incubation with 14C-cysteine or serine (Turnbull et al., 1981) indicating that this biosynthetic pathway was only active to a very limited extent. There have been reports of low levels of ACSO and PCSO in undifferentiated garlic callus (Madhavi et al., 1991) and of both PCSO and PeCSO in chive callus (Mellouki et al., 1996), suggesting that Allium species in tissue culture may differ in biosynthetic capacity.
Synthesis of the full spectrum of flavour precursors may resume if the callus is allowed to redifferentiate or regain a capacity for phototrophic metabolism. Once garlic callus had redifferentiated into green shoots or roots (Lancaster et al., 1988), the tissues contained ACSO, MCSO, and PCSO at levels comparable with intact plants. The resumption of flavour precursor synthesis after redifferentiation of onion callus into shoots and roots following removal of the auxin 2,4-dichlorophenoxyacetic acid (2,4,D) had been recorded earlier by Turnbull et al. (1981), and the auxin 4-amino-3,5,6-trichloropicolinic acid (picloram) induced PeCSO even at high levels that suppressed overt differentiation (Musker et al., 1988).
Addition of S-alk(en)yl donors or presumed biosynthetic intermediates aids CSO synthesis in tissue cultures. Very low levels of MCSO and ACSO were detected in shoot-forming callus from garlic (90 µM CSOs compared with 6.0 mM in bulbs) along with trace amounts of S-allyl cysteine and S-methyl cysteine (Ohsumi et al., 1993), but addition of 2-propenethiol to the culture medium resulted in detectable levels of both S-allyl cysteine and ACSO within a few hours. Recent experiments in our laboratory have also demonstrated that both onion and garlic tissue cultures are able to synthesize ACSO when provided with 2-propenethiol or S-allyl-L-cysteine, and PCSO when supplied with either propyl thiol or S-propyl-L-cysteine (J Hughes, unpublished work). Synthesis of PeCSO could be similarly restored to substantial levels in onion callus culture by S-2-carboxypropyl cysteine or S-propenyl cysteine (Selby et al., 1980). Potential intermediates from earlier in the biosynthetic pathway did not have this effect. This suggests that the absence of flavour compound biosynthesis in onion tissue cultures is due to an inhibition of S-2-carboxypropyl cysteine formation.
There is, therefore, experimental evidence for the synthesis of CSO proceeding via an alk(en)ylated free cysteine and via S-alk(en)ylated glutathione. The question of whether both paths operate at all times in all tissues remains to be resolved, but information about the enzymes required for both pathways would contribute to an answer.
|
Peptidases and transpeptidases |
|---|
The relationship between the
-glutamyl peptides and alk(en)yl cysteine sulphoxides proposed by Lancaster and co-workers (Lancaster and Shaw, 1989; Lancaster et al., 1989; Randle et al., 1995), where the former are biosynthetic precursors of the latter, requires the activity of enzymes to remove the glycyl and -glutamyl residues from the nascent alk(en)yl sulphoxide. -Glutamyl transpeptidase catalyses the transfer of the -glutamyl group from -glutamyl peptides to either amino acids or other peptides. This enzyme may also act as a -glutamyl peptidase, requiring only water as an acceptor. In mammals and microorganisms it is involved in the ‘-glutamyl cycle’ (Fig. 3) that degrades glutathione and participates in cysteine transport (Noctor et al., 2002). Both of these activities have been detected in aqueous extracts from the cloves and leaves of sprouting garlic cultivar Red (Ceci et al., 1992). The enzyme activities were higher in the rapidly growing leaves. -Glutamyl peptidase activity has been recorded in chive leaves, and was reduced to 5% of this level in cell cultures (Mellouki et al., 1996). Addition of exogenous peptidase elevated the yield of the alliinase reaction from chive cell cultures, suggesting that lack of -glutamyl transpeptidase may also be a reason of the low levels of flavour precursors in tissue culture.
|
-Glutamyl peptidase activity has been detected in leaves, roots, and bulbs of growing onions, but not in dormant bulbs of variety Southport White Globe (Lancaster and Shaw, 1991). However, it has been detected in stored bulbs of onion variety Spartan Banner when they were stored at 20 °C for 8 months. Activity gradually increased from the fourth month in store (Hanum et al., 1995). Measurement of -glutamyl propenyl cysteine sulphoxide and PeCSO in bulbs of seven onion varieties during cool storage for 4 months, showed a gradual decrease in levels of -glutamyl propenyl cysteine sulphoxide, and an increase in levels of PeCSO (Kopsell et al., 1999). Examination of the bulbs indicated that some remained dormant during storage, while other varieties broke dormancy after the first month. It is apparent that onion varieties vary in their storage characteristics, and this may explain the differing results for -glutamyl transpeptidase activity found in these studies.
In sprouting onion bulbs it has been associated with hydrolysis and remobilization of GPs. A -glutamyl transpeptidase was partially purified from onion (Lancaster and Shaw, 1994) that exhibited both peptidase activity, which was independent of pH, and transpeptidase activity that increased substantially above pH 8.0. It exhibited Km values between 0.4 mM and 2.0 mM for several -glutamyl derivatives and a Km value for glutathione of 5 mM. Its substrates included -glutamyl methyl cysteine, -glutamyl propenyl cysteine, 2-carboxyglutathione, and -glutamyl propenyl cysteine sulphoxide, where for the latter two the Km values were below the estimated cellular concentrations (Lancaster and Shaw, 1994). If hydrolysis of the -glutamyl moiety from -glutamyl alk(en)yl cysteine sulphoxides was required during CSO synthesis in these tissues, this enzyme might be involved, and the in vitro experiments indicated that it had a broad substrate specificity. There is also evidence for a different glutathione degradation pathway in some plants (Fig. 3B) that commences with removal of the glycine residue from glutathione, rather than the glutamyl group (Leustek et al., 2000).
|
Control of the biosynthesis of flavour precursors: effects of sulphur and nitrogen supply and storage |
|---|
Biosynthesis of the flavour compounds clearly requires substantial amounts of sulphur. In commercial onion production, it is generally considered that pungency is affected by sulphur, with a higher availability of sulphur generally resulting in greater flavour intensity. There is commercial interest in growing mild onions, and this generally involves using appropriate onion varieties in a low sulphur environment (Randle et al., 1995
). Studies on the effects of these growth regimens on flavour provide some information about flavour biosynthesis. The effects on sulphur accumulation and flavour compound levels in onion on altering the growth environment suggests that MCSO behaves differently from PeCSO.
Although PeCSO is the major CSO in onion, MCSO synthesis can be enhanced by manipulating growing conditions outside the ranges normally encountered in agriculture. These include low sulphur (S) and high nitrogen (N) levels. A study of 16 short-day onion cultivars grown in an artificial soil medium at two levels of sulphur (as sulphate, 2.0 mM and 0.05 mM) showed that although bulbs grown at the higher level of sulphur contained more flavour precursors (Randle and Bussard, 1993), there was little correlation between levels of sulphur in the soil and flavour precursors (measured as enzymatically generated pyruvate). However, a later study of only three onion varieties (Rio Grande, Savannah Sweet, and Southport White Globe) gave a more detailed picture. These onions were grown in sand maintained at five levels of sulphate (Randle et al., 1995). At the lowest level of sulphate (0.05 mM) all varieties showed symptoms of sulphur deficiency, but almost 95% of total bulb sulphur could be accounted for as MCSO, PCSO, and PeCSO and -glutamyl propenyl cysteine sulphoxide, indicating that flavour precursor biosynthesis was a strong sink for sulphur even in sulphur deficiency. At the highest level of sulphate used (1.55 mM), less than 40% of the bulb sulphur could be attributed to these compounds and, instead, must have been present in components of the flavour biosynthetic pathway that were not measured or other organic sulphur compounds.
A more recent study has detected significant amounts of sulphate in onion bulbs of these three varieties that increased as the level of sulphate in the artificial growth medium was increased (Randle et al., 1999). The total amount of sulphur in bulbs of all three cultivars was similar at each level of sulphate in the medium and increased as the levels were increased. Level of CSOs (measured as enzymatically generated pyruvate) also increased in response to increasing sulphate in the growth medium, but the levels of individual CSOs were different in the three cultivars. MCSO dominated at the lowest levels of sulphate in all three cultivars and remained similar or decreased as sulphate in the medium increased. Above 0.425 mM the level of PeCSO increased to match or exceed that of MCSO. Southport White Globe had a lower level of PeCSO than the other two varieties, which was unaffected by increases in sulphate. Levels of PeCSO in Rio Grande were higher, but also unaffected by environmental sulphate. The levels in Savannah Sweet, however, decreased steadily as sulphate was increased. McCallum et al. (2002) have observed changes in transcript levels of several genes required for sulphate uptake and assimilation (APS reductase, ATP sulphurylase, high affinity sulphate transporter, and sulphite reductase) in response to transient or long-term sulphur deprivation of onion varieties Canterbury Longkeeper and Houston Grano indicating that regulation of these initial stages in sulphur assimilation also underlie differences in onion flavour.
For production of over-wintered mild-flavoured onions in the sandy loam soil of the Vidalia region of the USA, applications of S-containing fertilizers are discouraged from spring onwards, so that the onions experience an environment of decreasing available sulphur. Sulphate is applied during winter to ensure good root and leaf growth. In an investigation of the time-schedule for sulphate withdrawal, variety Sweet Vidalia was grown in an artificial medium where the initial level of 1 mM sulphate was reduced to 0.05 mM for six groups of plants at successive fortnightly intervals (Randle et al., 2002