Summary:

The myrosinase-glucosinolate system is involved in a range of biological activities affecting herbivorous insects, plants and fungi. The system characteristic of the order Capparales includes sulphur containing substrates, the degradative enzymes myrosinases and cofactors. The enzyme catalyzed hydrolysis of glucosinolates initially involves cleavage of the thioglucoside linkage yielding D-glucose and an unstable thiohydroximate-O-sulphonate that spontaneously rearranges, resulting in the production of sulphate and one of a wide range of possible reaction products. The products are generally a thiocyanate, isothiocyanate or nitrile depending on such factors as substrate, pH or the availability of ferrous ions. Glucosinolates in crucifers exemplify components that are often present in food and feed plants and are a major problem in the utilization of products from the plants. Toxic degradation products restrict the use of cultivated plants e.g. belonging to the family Brassicaceae. The myrosinase-glucosinolate system may, however, have several functions in the plant. The glucosinolate degradation products are involved in defence against insects and phytopathogens and potentially in sulphur and nitrogen metabolism and growth regulation. The compartmentalization of the components of the myrosinase-glucosinolate system and the cell-specific expression of the myrosinase represents an unique plant defence system. In this review we summarize earlier results and discuss the organisation and biochemistry of the myrosinase-glucosinolate system.
 

1. Introduction

Oilseed rape (Brassica napus L.) is a major European crop, the most significant oil-producing crop for the temperate regions of central and northern Europe. The oil from this crop is valuable in nutrition (low in polyunsaturates), and as an industrial feedstock - for both specialist industrial use and as a potential fuel. The seed meal remaining after oil extraction is also a valuable animal feed, high in protein, which can be used in place of imported soybeans or other animal feed products. Many other Brassicaceae species are grown in Europe, as human and animal foodstuffs - cabbages, sprouts, kale, turnips, swedes and more. All these species contain a group of secondary metabolites called glucosinolates. Upon mechanical damage, infection or pest attack, cellular breakdown exposes the stored glucosinolates to degradative enzymes (myrosinases).

Plant breeding strategies over the past decade have concentrated on reducing the glucosinolate content of rape seeds, to improve the acceptability of rapeseed meal and meet increasingly stringent requirements from the processing industry. It had earlier proved possible to breed out the erucic acid content of rape seed, to produce the so-called "single low" varieties. The low glucosinolate strategy has been successful, in that current commercial varieties ("double low") have very low seed glucosinolate content, but there is a significant cost in terms of crop protection and nutrition. Reducing or eliminating the synthesis of specific glucosinolates e.g. progoitrin, is a major goal in Brassica breeding. An alternative approach would be to change the amount of myrosinase available for hydrolysis of the glucosinolates.

The myrosinase-glucosinolate system has long been the defining phytochemical character of the order Capparales (Hegnauer 1986). This system and its components have been pre-eminent subjects of chemical investigation of species within this order. Degradation of glucosinolates produces products which affect the value of glucosinolate containing plants when used as food for humans or for feeding animals (Chew 1988). Thiocyanates can inhibit the uptake of iodine in the thyroid and large amounts are known to have negative effects on the liver. More recently, interest has focused on the potential mutagenic, carcinogenic (Dunnick et al. 1982) and anticarcinogenic (Ioannou et al. 1984) properties of naturally-occurring isothiocyanates.

The complexity of the myrosinase-glucosinolate system should indicate an important role in the life of the cruciferous plants. The function of this system may be diverse. The glucosinolates may be a sink for nutrients like nitrogen and sulphur and the products of hydrolysis may have important roles in the defence system of the plants against microorganisms and insects. The number of different glucosinolates and isoenzymes reported, may indicate that specific hydrolysis products are needed for certain situations or developmental stages. Indoleglucosinolates e.g. may be a sink for production of growth factor indole acetic acid (IAA) and thereby be involved in growth regulation.

This review will focus on; How the glucosinolate-thioglucosidase system produces diverse hydrolysis products; how the system is distributed and compartmentalized; how the system is held latent until the products of the hydrolysis are needed; what are the properties of the myrosinases involved; what biological function the myrosinase-glucosinolate system is involved in.
 

3. Glucosinolates

Glucosinolate hydrolysis products

The myrosinase mediated degradation of glucosinolates gives rise to an unstable thiohydroximate-O-sulphonate which on release of sulphate (via a Lossen rearrangement) can result in the production of isothiocyanates, thiocyanates, nitriles and elementary sulphur, depending on the concentration of H+ and/or other factors (Fig. 1). The mechanisms of degradation have been studied in some detail especially isothiocyanate and nitrile formation (Benn 1977) although the mechanism for thiocyanate still requires elucidation (Hasapis and MacLeod 1982).
 

(i) Isothiocyanate and nitrile production

Isothiocyanates are usually produced at neutral pHs while nitrile production occurs at lower pHs. Indole glucosinolates such as glucobrassicin undergo enzyme hydrolysis to give 3-indolemethanol, 3-indoleacetonitrile and 3,3´-diindolylmethane (Labague et al. 1991). The alcohol from the indole glucosinolates is derived from the isothiocyanate which undergoes solvolysis. Evidence for an indole isothiocyanate comes from work by Hanley et al. (1990) who isolated an indole isothiocyanate from gluconeobrassicin degradation under specific experimental conditions. Isothiocyanates with a hydroxy group in the 2 position spontaneously cyclise to give oxazolidine-2-thiones an example of which is goitrin derived from 2-hydroxy-but-3-enylglucosinolate.
 

(ii) Thiocyanate production

Of the naturally ocurring glucosinolates, only three appear to undergo enzymic degradation to thiocyanates. These are allyl-, benzyl- and 4-(methylthio)butylglucosinolates. Various mechanisms have been proposed for thiocyanate formation and the most recent is a combined theory where an isomerase causes Z-E isomerisation of the aglycone but only glucosinolates with stable cations are able to undergo E-aglycone rearrangement to the thiocyanate (Hasapis and MacLeod 1982).
 

(iii) Epithioalkanes

Epithioalkanes are produced from the hydrolysis of alkenyl glucosinolates when myrosinase co-occurs with a small labile protein known as epithiospecifier protein (Tookey 1973).
 
 

4. The myrosinase enzyme system

Occurrence of myrosinases

Myrosinase and glucosinolates were first discovered in mustard seeds by Bussy (1840). For a historical review of myrosinase investigations see e.g. Björkman (1976) and Bones (1991). The myrosinases always appear to be accompanied by one or more glucosinolate. They occur in all Brassicaceae (Cruciferae) species examined and have also been found in Akaniaceae, Bataceae, Bretschneideraceae, Capparaceae, Caricaceae, Drypetes (Euphorbiaceae), Gyrostemonaceae, Limnanthaceae, Moringaceae, Pentadiplantdraceae, Resedaceae, Salvodoraceae, Tovariaceae, Tropaeolaceae (Rodman, 1991). Enzymes with myrosinase activity have also been found in the fungi Aspergillus sydowi (Reese et al. 1958, Othsuru et al. 1969) and Aspergillus niger (Ohtsuru et al. 1973), in the intestinal bacteria Enterobacter cloacae (Tani et al. 1974) and Paracolobactrum aerogenoides (Oginsky et al. 1965), in mammalian tissues (Goodman et al. 1959) and in the cruciferous aphids Brevicoryne brassicae and Lipaphis erisimi (MacGibbon and Beuzenberg 1978). These aphid myrosinases hydrolyze 2-hydroxy-2-phenylethyl glucosinolate in vitro and exhibit electrophoretic mobilities distinct from those of isoenzymes isolated from their hostplants. This distinctive biochemical behaviour does not rule out the possibility that the myrosinases are modified from enzymes acquired from the hostplant, in contrast to the de novo synthesis involved in endogenous myrosinases in fungi and bacteria. However, recent work has shown that antibodies raised to plant myrosinase do not cross react with partially purified aphid myrosinase indicating that the protein is not derived from the plant host (Bridges et al. unpublished).

The amount of myrosinase activity found in seeds from cultivars of Sinapis alba L, Brassica campestris L. and Brassica napus L. has been examined by Henderson and McEwen (1972), Bjørkman and Lønnerdal (1973) and Bones (1990). The myrosinase activity was found to be about ten times higher in S. alba than in B. campestris and the activity in B. napus was slightly higher than in B. campestris (Bones 1990).
 

Myrosinase isoenzymes

The existence of multiple forms of myrosinase has been shown in many plants. By analytical gel electrophoresis various authors have demonstrated the separation of several myrosinase isoenzymes (MacGibbon and Allison 1970, Henderson and McEwen 1972, Buchwaldt et al. 1986). MacGibbon and Allison (1970) examined the isoenzyme pattern after electrophoretic separation of myrosinases of 7 species of Rhoeadales. Each species was found to have a distinct pattern and the number of detectable bands varied from 1 to 4. MacGibbon and Allison (1970) also found different patterns depending on whether the extracts were made from leaf, stem, root or seed. Buchwaldt et al. (1986) reported that the crude extract from seeds of Sinapis alba contained at least fourteen myrosinase isoenzymes. Isoelectric focusing in polyacrylamide gels of ethanol-powder preparations revealed two bands for Brassica nigra and Brassica napus and seven bands for Sinapis alba (Buchwaldt et al. 1986). In a study of the effects of ascorbic acid on myrosinases from different crucifers, Henderson and McEwen (1972) noticed different activation of the various isoenzymes as detected by an increase in bariumsulphate precipitate.

The validity of the method for development of myrosinase bands after electrophoretic separation should be controlled. Henderson and McEwen (1972) stated that it was possible to locate active isoenzymes in the Crambe abyssinica and the two Brassica juncea preparations, even though activity could not be detected by a direct spectrophotometric method. A possible explanation for this very surprising result is that the conditions were changed as a result of the electrophoretic run. However, another possibility may also be that this assay is not as specific as it should be. One should also remember that only active myrosinase is detected in this assay.

Recently, James and Rossiter (1991) have separated two myrosinases from cotyledons of five days old Brassica napus seedlings. The most striking difference between these two myrosinases was the degree of glycosylation. Myrosinase I was similar to myrosinase C reported by Lönnerdal and Janson (1973) and Bones and Slupphaug (1989), while the myrosinase II was reported to be considerably less glycosylated. By immunoblots with polyclonal antibodies against myrosinase and by periodic acid Schiff staining these two myrosinases were shown to be different. When tested for substrate specificity against five glucosinolates, the degree of activation by ascorbic acid was also different (James and Rossiter 1991).

Little is known about the substrate specifisity of myrosinase isoenzymes. The data available is from James and Rossiter (1991) and Björkman and Janson (1973). The two myrosinases examined by James and Rossiter degraded different glucosinolates at a different rate. However, both isoenzymes showed highest activity against aliphatic glucosinolates and least activity against indole glucosinolates. From their results it is evident that members of a given class of glucosinolates are apparently degraded at approximately the same rate in vitro. An exception being the significant difference in degradation of sinigrin. Similar results was also obtained by Björkman and Janson (1973). They also noted that the activity against an aliphatic glucosinolate (progoitrin) was dependent on the iosoenzyme tested. The results from Arabidopsis thaliana could indicate that myrosinases accept a wide range of glucosinolate substrates. Xue et al. (1992, 1995) and Chadchawan et al. (1993) have all reported that there are three myrosinase genes in Arabidopsis thaliana. Active transcripts have only been reported for two of these genes, and they are then supposed to be degrading the 23 different glucosinolates reported to be present in Arabidopsis thaliana (Haughn et al. 1991). When considering substrate specifisity one should also be aware of the possibility that the specifisity could be affected by associated factors like epithiospecifier protein, myrosinase binding protein or other myrosinase associated proteins or components. To clarify the substrate specifisity question it will be necessary to examine isoenzymes from different developmental stages in combination with the known associated factors.
 

Distribution of myrosinase activity

Distribution of the myrosinase isoenzymes seems to be both organ-specific and species-specific. Electrophoretic examination of isoenzymes from many plants, organs or tissues demonstrates that the distinct patterns may vary with species, organ and age of the plant (MacGibbon and Allison 1970, Henderson and McEwen 1972, Buchwaldt et al. 1986). Little is known about the physiological reason for this difference. It has been postulated that the particular isoenzymes correspond to endogenous conditions found in that plant, or to conditions found in the target organism, or to particular glucosinolates that dominate the profile of that tissue. In fact nobody knows that all these bands after electrophoretic separation are real isoenzymes with a distinct amino acid sequence. As a consequence more effort should be placed on determining whether or not these are isoenzymes. A combination of immunological detection, amino acid sequencing and characterization of myrosinase genes will be used to solve this problem.

There have been few reports with a systematic analysis of the variation in myrosinase activity in plants at different developmental stages and organs. Bones (1990) examined the myrosinase activity at different deveolopmental stages and in different parts of plant in a full life cycle. A comparison between plants regenerated from protoplasts and explants and seed germinated plants was also included. The results given in Bones (1990) show that hypocotyls contain the highest specific myrosinase activity of the seedling organs. The specific activity in hypocotyls was always more than twice the activity in cotyledons and normally several times the activity in seedling roots. Although the highest specific myrosinase activity was observed in hypocotyls, cotyledons contain most myrosinase activity based on total activity. But since the total protein content in cotyledons was several times the amount in roots and hypocotyls the specific activity was lower (Bones 1990). In a study of the myrosinase expression in Sinapis alba L., Xue et al. (1993) detected myrosinase mRNA´s in cotyledon, hypocotyl, root and leaves of 6, 8 and 14 day old seedlings. Highest levels of myrosinase mRNA accumulation were observed for hypocotyls at 6 and 8 day after germination, and in leaves 14 days after germination.

The reported activities in different tissues varies, but all tested organs/tissue contains some myrosinase activity (Bones 1990). Myrosinase activity could also be detected in callus cultures and in in vitro cultured plants (Bones 1990). Except for the roots of fully grown plants where a high activity was observed, other organs of mature plants normally contained low myrosinase activity (Bones 1990). References to several other studies of myrosinase activity can be found in Bones (1990, 1991). James and Rossiter (1991) reported that the total potential myrosinase activity increased during early seedling growth when the activity was measured in the presence of 0.3 mM ascorbic acid.

Using a slot-blot technique, Falk et al. (1992) showed that myrosinase mRNA also was produced in all tested organs of B. napus. Xue et al. (1993) have shown that myrosinase mRNA in seedling organs decreases during early seedling growth. Recently, Xue et al. (1993) examined the transcript levels from members of two subfamilies of myrosinase genes in developing seedlings of Sinapis alba. One of the myrosinase genes (MB=Myr2) was transcribed in all organs of the seedlings, while the other myrosinase gene (MA=Myr1) showed no transcription in the investigated organs of the seedling. Expression of both MA and MB myrosinase genes was observed during embryo development of Sinapis alba.

To examine the possible correlation between myrosinase and myrosin cells the myrosinase activity in calli was measured and ultrastructural analysis of calli was performed. Although low, a significant myrosinase activity was found. In common with other observations where similar low myrosinase activity was detected, no myrosin cells could be detected with conventional LM- and TEM-techniques. It was stressed that lack of observable myrosin cells seemed to be followed by a low myrosinase activity, indicating a positive relationship between the enzyme and the myrosin cells (Bones 1990).

Springett and Adams (1989) investigated the distribution of myrosinase in Brussel sprouts (B. oleracea L. var. bullata subvar. gemmifera). Myrosinase activity was 4-5 times higher in the outer leaves of the Brussel sprouts than in other regions (stalk, inner leaves and centre).
 

Myrosinase isolation and purification

Myrosinase isoenzymes from plant sources show substantial diversity in physico-chemical characteristics (Tab. 1). All myrosinases isolated and purified so far have been reported to be glycoproteins. The carbohydrate content varies from 9 to 23% of the total molecular mass. James and Rossiter (1991) reported that they have purified a myrosinase from 5 day old Brassica napus seedlings with low levels of glycosylation. Molecular mass of purified myrosinases normally range from 125 kDa to over 150 kDa. One exception is the myrosinase from Wasabia japonica which was reported to have molecular mass of 580 kDa (Ohtsuru and Kawatani 1979). Isoelectric points vary between 4.6 and 6.2 (Björkman 1976).

Several reports have described the isolation and characterization of myrosinase from white mustard (Björkman and Janson 1972, Björkman and Lönnerdal 1973, Buchwaldt et al. 1986) and oilseed rape (Lönnerdal and Janson 1973, Björkman and Lönnerdal 1973, Buchwaldt et al. 1986, Bones and Slupphaug 1989). Ohtsuru and Hata (1972) purified four myrosinase isoenzymes from mustard powder; three had a molecular mass of 153 kDa, and the fourth 125 kDa. From SDS-polyacrylamide gel electrophoresis the number of subunits was estimated to be 4 (Tab. 1). Björkman and Janson (1972) purified one myrosinase to apparent homogeneity and obtained a partial purification of two other isoenzymes. The completely purified myrosinase was shown to contain 2 subunits of similar size. A new purification procedure for myrosinase from B. napus L. was presented by Bones and Slupphaug (1989). When comparing the results it should been noticed that there are few reports of apparently homogenous myrosinase purifications. Lenman et al. (1990) recently reported that myrosinases could form complexes with molecular masses in the range 140-800 kDa.
 

Genes encoding myrosinase.

A significant amount of work has been put into the cloning of myrosinase genes recently. As a result several myrosinase genes from Sinapis alba, Brassica napus and Arabidopsis thaliana have been isolated and characterized (Xue et al. 1992, Falk et al. 1992, 1995, Chadchawan et al. 1993; Thangstad et al. 1993). The results show that myrosinases in Brassica napus are encoded by a multigene family consisting of three subgroups (Thangstad et al. 1993, Falk et al. 1995). Thangstad et al. (1993) have suggested that the subgroups should be named according to the recommendations from the International Society for Plant Molecular Biology, Commision on Plant Gene Nomenclature (Plant Mol. Biol. Reporter 11: 3-9, 1993).

The first cDNA gene encoding myrosinase was isolated from Sinapis alba by Xue et al. (1992). The gene was isolated from an immature seed cDNA library by screening with degenerated oligonucleotides obtained from amino acid sequence information. One full-length clone of the Myr2-type (MB3) was isolated with a probe obtained from a polymerase chain reaction based on sequence information from a truncated clone of the Myr1-type, that was first isolated (MA1). The clone MB3 consisted of 1878 bp, started with a 118 bp 5´untranslated region and had a 20 amino acid signal sequence. The clones were grouped into MA(Myr1)- and MB(Myr2)-type based on their sequence similarity. MB(Myr2) clones showed 92-94% sequence identity, whereas MA(Myr1) and MB(Myr2) showed 62-64 % similarity. These data are, however, based on only one full-length cDNA clone. CDNA encoding myrosinases from Brassica napus and Arabidopsis thaliana has also been reported (Falk et al. 1992, Chadchawan et al. 1993). The B. napus gene was found to be of 91.2% identity to the MB3 clone from Sinapis alba and of 67.5% identity to the MA1 clone, at the amino acid level, and therefore to belong to the Myr2 subtype (=MB) of the gene family.

Five functional genomic myrosinase genes have so far been sequenced, two from Brassica napus (Thangstad et al. 1993), two from Arabidopsis thaliana (Xue et al. 1995) and one from Brassica campestris (Machlin et al. 1993). The first two genomic myrosinase genes which were characterized, Myr1.Bn1 and Myr2.Bn1, from Brassica napus , are about 3 kb in length, with 11 small introns and 12 exons (Thangstad et al. 1993). The homology is 83.2 % at the amino acid and 77.4 % at the nucleic acid level (Thangstad et al. 1993). A signal peptide and a variable number of sites for N-linked glycosylation indicate transport and glycosylation through the ER-Golgi complex. Thangstad et al. (1993) concluded that Myr1.Bn1 most likely encodes a major seed myrosinase, and that Myr2.Bn1 encodes a vegetative type myrosinase. A sequence comparison of the five functional myrosinase genomic genes has been presented by Xue et al. (1995).

A myrosinase pseudogene has also been reported (Lenman et al. 1993). The pseudogene was found to span more than 5 kb. A portion of the 5´-end was missing and it contained probably 12 exons. The uncertainty was due to missing exon-intron splice sites. Sequence comparisons with myrosinase cDNA grouped this gene to the Myr1 (MA) subgroup of the gene family. A multiple sequence alignment with the cDNA from Falk et al. (1992) and several other b-glycosidases and b-galactosidases revealed that myrosinases belonged to the BGA family of b-glycosidases (Lenman et al. 1993).

Myrosinases are encoded by a large gene family in both Sinapis alba and Brassica napus. Thangstad et al. (1993) estimated the number of myrosinase genes in B. napus to 14-17. A similar result was also reported by Lenman et al. (1993) which estimated that there were 13 Myr2 (MB) genes and 3-4 Myr1(MA) genes in Sinapis alba. Some of these genes may be pseudogenes (Lenman et al. 1993). Present results indicate that these estimates may be to low (Thangstad et al. unpublished, Falk et al. 1995). In Arabidopsis thaliana a small myrosinase gene family with only three genes have been found (Chadchawan et al. 1993, Xue et al. 1995).

Falk et al. (1992) found highest expression of myrosinase mRNA in young leaves, cotyledons and developing seed, where it was detected from day 15 to day 30 after pollination. In a later report Xue et al. (1993) reported that two myrosinase genes show a tissue preferential expression during embryo and seedling development in Sinapis alba. Northern blots with mRNA from seeds and young leaves of Brassica napus probed with MA (Myr1) and MB(Myr2) specific probes showed that the two subgroups of the myrosinase gene family were differentially expressed (Lenman et al. 1993). The highest level of expression of MA(Myr1) was found in seeds, whereas MB(Myr2) expression was found to be highest in leaves. The large number of genes which are transcriptionally active indicate that the myrosinase-glucosinolate system must play an important role in the life of the plants of Brassicaceae (Winge et al. unpublished).
 

Myrosinase gene structure

Based on conserved regions in cDNA from three species, PCR (polymerase chain reaction) primers were made, and used to amplify and characterize the structure of the myrosinase genes in Brassica napus, B. chinensis (B. campestris var. chinensis), B. campestris, B. oleracea, Cheiranthus cheiri, Raphanus sativus and Sinapis alba. The strong similarity of nucleotide sequence, intron-exon structure and gene copy number between the seven species compared, indicate that the myrosinase genes are similar organised in these species. Work in our lab (Winge et al. unpublished) show that the structure of myrosinase genes in Arabis alpina, Tropaeolum majus, Iberis umbellata and Lepidum sativum are similar with the results reported by Thangstad et al. (1993).
 
 

5. Effects of ascorbic acid on myrosinase activity

Ascorbic acid has been shown to modulate myrosinase activity in some species (Nagashima and Uchiyama 1959). A model for the action of ascorbic acid was postulated by Tsuruo and Hata (1968). Ascorbic acid does not participate in the reaction catalyzed by mustard myrosinase (Ettlinger et al. 1961, Tsuruo and Hata 1968), nor is it involved in the association of the enzyme subunits (Ohtsuru and Hata 1973). The activation appears to be due to a conformational change in the protein structure leading to an enhanced reaction rate when the effector binding sites are occupied (Tsuruo and Hata 1968, Ohtsuru and Hata 1973). Tsuruo and Hata (1968) postulated the presence of one site of action for the substrate, and two sites for the ascorbic acid. The substrate site has two moieties, one for the glycone and one for the aglycone part of the glucosinolate. The conformation of the aglycone moiety is altered when the ascorbic acid site is occupied, so that the glucosinolate fits better in its site. Because the second site for ascorbic acid is the same as the substrate site, high concentrations of ascorbic acid have an inhibitory effect. When p-NPG is used as a substrate high concentrations of ascorbic acid inhibit the reaction due to competition at the binding site. Moderate concentrations have no effect, since p-NPG does not occupy the aglycone moiety of the glucosinolate site. Considering that a weaker association of glucosinolates to the myrosinase may give an increased reaction rate this may also explain the increased Km in the presence of ascorbic acid. When ascorbic acid is added both the Km and the Vmax are increased. The decreased affinity (Km) of the enzyme for the substrate is not the normal behaviour of a positive effector. However, as postulated by Tsuruo and Hata (1968) this may be due to a common binding site for the effector and the substrate.

The lack of activation of p-NPG hydrolysis in the presence of ascorbic acid has, however, been questioned recently (Durham and Poulton 1990). A 2.6 fold activation of hydrolysis of pNPG was found in the presence of 1 mM ascorbic acid using principally the same methodology. If this is correct, the model of activation by ascorbic acid has to be reevaluated. The possibility exists that only the glucose and sulphate group of the substrate is involved in binding to the active site, and that the R-group is outside the binding site, and contributing only by electrostatic effects with the areas surrounding the binding site. This would also imply that the conformational change is in the active site.

Ohtsuru and Hata (1973) investigated the inhibition of myrosinase from brown mustard by trinitrobenzenesulfonic acid (TNBS) in the presence and absence of ascorbic acid. In the presence of 1 mM ascorbic acid and TNBS, myrosinase activity was reduced by 55% while TNBS alone gave no inactivation. These results supported the theory that the activation mechanism of myrosinase by ascorbic acid depends on a slight conformational change of the protein, and is not due to the dissociation and association mechanism of myrosinase. L-ascorbic acid has been reported to be solely localized in vacuoles (Matile 1980, Grob and Matile 1980).
 
 

6. Metal ions and myrosinase activity.
 
 

7. Myrosinase binding proteins

Recently it has been shown that several polypeptides are associated with myrosinases by virtue of their co-precipitation with anti-myrosinase antibodies (Lenman et al. 1990). However, it is not known at this stage if they have a role in the myrosinase-glucosinolate system and the biochemistry of the phenomenom has yet to be fully explored. The function of these polypeptides are unknown, but Rask and coworkers in Uppsala have shown that wounding induces the expression of these polypeptides (Falk et al. 1995). A series of ongoing experiments will probably yield more information about the proteins bound to or associated to myrosinase.

Falk et al. (1995) characterized myrosinase-binding proteins (MBP´s) first observed by Lenman et al. (1990). At least ten proteins in extracts from Brassica napus reacted with a monoclonal antibody against MBP. The proteins were classified as myrosinase-binding proteins (MBP) and myrosinase-binding protein-related proteins (MBPRP). MBPRP´s were present in almost all parts of the plant, whereas MBP´s were exclusively found in seeds. Two MBP with molecular mass of 50 and 52 kDa and two MBPRP´s with molecular mass of 80 and 100 kDa were purified and characterized and were found to display structural similarities. Based on differences observed in e.g. solubility Falk et al. (1995) suggested that they could have different function. Falk et al. (1995) hypotesised that they could have some role in response to tissue damage since the myrosinase-glucosinolate system is activated upon tissue damage. Lenman et al. (1990) concluded that the possibility exists that MBP50 and 52 are identical to epithiospecifier protein. The lack of relation between the ESP activity (Rossiter et al. unpublished) and expression of MBP´s (Falk et al. 1995) do not support that ESP are similar to MBP´s. The distribution of MBP and MBPRP in different species is not known. The function, cellular and subcellular localization of MBP´s and MBPRP´s remains so far speculative. It should also be mentioned that another group of polypeptides named MAP (myrosinase associated proteins) with affinity for myrosinase complexes recently have been described (Falk pers. comm.). These polypeptides is based on amino acid sequence information and immuno-reactivity unrelated to the MBP and MBPRP polypeptides.
 
 

8. Myrosinase-associated proteins (MyAP`s)

Another protein exhibiting affinity for myrosinase complexes. The myrosinase associated protein is derived from a gene family with wound- and MeJA-inducable members (Taipalensuu et al. 1996). Sequence determination and immunoreactivity show that the MyAP is unrelated to the previous characterized MBP`s and MBPRP`s (Falk et al. 1995). MyAP do have some similarity to an early nodulin from Medicago sativa and to anther specificPro-rich proteins from B. napus and Arabidopsis thaliana. The protein has been named MyAP to distinguish it from the MBP family of proteins (Taipalensuu et al. 1996).
 

Two slightly different cDNA-clones corresponding to the wound-inducible form of a previously characterized seed myrosinase associated protein (MyAP) have been isolated from Brassica napus L. (Taipalensuu et al. 1997). MyAP transcripts have been found to developmentally regulated in various vegetative organs. Induction by wounding or MeJA have been shown to be antagonized by salicylic acid (Taipalensuu et al. 1997).
 

Taipalensuu et al (1997) proposed a function for MyAP`s in liberating acetylated glucosinolates from their acyl-group and thereby making them available for hydrolysis by the myrosinase enzymes.
 
 

9. Epithiospecifier protein (ESP)

Epithiospecifier protein is a small protein (30-40 kD) first isolated from Crambe abyssinica seeds (Tookey 1973). This protein does not have myrosinase activity, but it interacts with myrosinase to promote sulphur transfer from the S-glucose moiety to the terminal alkenyl moiety (Fig. 1.). Degradation of progoitrin (2-hydroxy-3-butenyl glucosinolat) in the absence of ESP produces mainly oxazolidine-2-thione. The same degradation in the presence of ESP, produces mainly epithionitrile. ESP is unique in that it specifies the reaction products and not the substrate. The function of ESP in vivo is unknown. The effects of epithionitriles in mammals have been evaluated by e.g. Wallig et al. (1988) and Van Steenhouse et al. (1991).

Formation of epithionitriles represents an alternative pathway for alkenyl glucosinolates with terminally unsaturated carbon (e.g. progoitrin, sinigrin, gluconapin and glucobrassicanapin). This degradation step is mediated by epithiospecifier protein (ESP) and the conditions under which the thioglucoside cleavage occurs (MacLeod and Rossiter 1987). Combinations of myrosinase and epithiospecifier protein from various plant sources produced the same products, demonstrating that epithionitriles are formed from glucosinolates by the same mechanism in all crucifers tested (Petroski and Tookey 1982). Similar results were also obtained with the myrosinase producing fungus Aspergillus sydowi (Petroski and Kwolek 1985). Cole (1975) investigated the distribution of aglucones upon autolysis of glucosinolates in eight week old plants in 79 species. Cyano-epithioalkanes were detected in 15 of these 79 species indicating the presence of ESP. Species with ESP activity includes Crambe abyssinica (Tookey 1973), Alyssum perenne, Alyssum saxatile, Arabidopsis thaliana, Berteroa incana, Brassica chinensis, B. juncea, B. napus, B. oleracea, B. rapa, Cakile maritima, C. pratensis, Hirchfeldia incana, Lepidum sativum, Lubularia maritima, Sisymbrium altissimum and Turritis glabra (Cole 1975, MacLeod and Rossiter 1985).

Kinetic studies show that epithiospecifier protein inhibits myrosinase activity non-competitively (Rossiter 1983, MacLeod and Rossiter 1985), providing evidence that the epithiospecifier protein interacts at a site on the myrosinase other than the substrate binding site. Ferrous ions are essential to epithionitrile formation (MacLeod and Rossiter 1985). Epithiospecifier protein in B. napus was shown to be inactive in the absence of ferrous ions (Rossiter 1983, Macleod and Rossiter 1985). Addition of ferrous ions changed the major products of the hydrolysis from oxazolidine-2-thione to epithionitrile.
 
 

10. Myrosin cells

The term myrosin cell was first used by Guignard (1890) and has later been used to describe this special type of cell discovered by Heinricher (1884) assumed to contain myrosinase. Myrosin cells were confined to parenchymatous tissue of the green parts of different plants of Brassicaceae, especially in epidermal cells of leaves (for review of distribution, occurrence, morphology reported in early studies see Bones, 1991, Bones and Iversen, 1985).

The importance of the myrosin cells as a taxonomic tool has been generally accepted and is noticed by numerous authors (see e.g. Dahlgren, 1980; Jørgensen, 1981). Jørgensen (1981) used occurrence and distribution of myrosin cells as one of several criteria for the classification of the order Capparales.

As a consequence of the taxonomic investigations some information about the occurrence and distribution of myrosin cells was generated. Myrosin cells were observed in seeds, parenchymatic tissue, epidermis and in guard cells (for references see Bones, 1991; Bones and Iversen, 1985).

The first reports on the morphology of myrosin cells were based on light microscopic observations. Although the resolution obtained in these studies was limited, some interesting observations were reported. Spatzier (1893) observed that the grains in myrosin cells differed from the grains in the surrounding cells. The grains in myrosin cells had a different refractivity and were more homogenous. Therefore he suggested these grains to be named myrosin grains. The myrosin grains did not have inclusions (globoids), had a higher refractivity and gave an intense reaction with general protein stains.

Later reports on the structure of myrosin cells have mainly used the higher resolution obtained in the electron microscope. Rest and Vaughan (1972) followed the development of protein bodies and oleosomes (spherosomes) during nine stages of embryo development in Sinapis alba L. Special emphasis was paid to the cells in the developing cotyledons and the accumulation of proteins in the protein bodies. The myrosin cells could be distinguished from the aleurone cells 24 days after petal fall. At this stage the central vacuole has subdivided. In aleuron cells proteins were first observed as electron dense lumps with globoids. The initial accumulation of protein in myrosin cells was reported to include vacuolar structures which had more irregular outlines and were filled with a homogenous fibrillar material (Rest and Vaughan 1972).

Bones and Iversen (1985) presented an analysis of the distribution of myrosin cells in plants of Rhapanus sativus and Sinapis alba. In seeds of Rhapanus sativus L. and Sinapis alba L., imbibed for two hours, the relative area occupied by myrosin cells was measured to be in the range 0.9-1.6% and 4.5-6.5%, respectively (Bones and Iversen 1985). Twenty five days after seeding, the identification of myrosin cells was speculative and solely dependent on the different reaction after staining.

The morphology of myrosin cells varies and depends on the organ and tissue in which they are present as well as the age of the tissue (Bones and Iversen 1985). In cotyledons two morphologically distinct types of myrosin cells have been observed. In pallisade tissue of cotyledons elongated myrosin cells were frequently observed. In the parenchyma tissue both elongated and isodiametric cells were observed (see for example Bones and Iversen 1985). In roots and hypocotyls myrosin cells are generally found in the elongated form in longitudinal sections. Compared to normal cortex cells they are 1-6 times the size of surrounding cells.

Myrosin cells in seeds and young seedlings are characterized by homogenous protein bodies in contrast to the protein bodies in aleuron-like cells which always contain electron opaque globoids. In the electron microscope myrosin grains appear moderately electron dense and with a finely granular content (Bones and Iversen 1985). Changes in the ultrastructure of myrosin cells of Sinapis alba and Rhapanus sativus have been followed systematically both during early seedling growth and embryo development (Bones and Iversen 1985). The myrosin grains of myrosin cells in hypocotyls and roots change dramatically during the first days after soaking. The myrosin grains undergo an active period which first seems to include division of the protein bodies and later fusions to form the big central vacuole.
 
 

Dilated cisternae of the endoplasmic reticulum.

Several attempts have been made to correlate the presence of dilated cisternae of the endoplasmic reticulum with myrosinase (see references in Bones et al. 1989). Evidence for such a correlation has not been presented.
 
 

11. Compartmentation of components in the myrosinase-glucosinolate system.

"The myrosinase-glucosinolate bomb".

Matile (1980) concluded that the stability of glucosinolates in the intact root tissues of horseradish (Armoracia rusticana) appeared to be due to the location of glucosinolates and myrosinase in distinct subcellular compartments of the same cell. Evidence was presented for the presence of myrosinase in extracellular compartments (cell walls) and associated to the cytosolic side of internal membranes, while glucosinolates were localized in vacuoles. Activation of this system would be induced by permeabilization of membranes resulting in the release of glucosinolates and L-ascorbic acid from the central vacuoles of parenchyma cells. In a later report from Lüthy and Matile (1984) a rectified analysis of the subcellular organisation of the myrosinase system is presented. The main difference between the model of Matile (1980) and the model presented by Lüthy and Matile (1984), was that myrosinase in the latter report was reported to be a cytosolic enzyme which tended to bind to membranes. This rectified "Mustard bomb" model is presented in Fig. 2. Lüthy and Matile (1984) concluded that the association of myrosinase with smooth ER, plasmalemma, dictyosomes and mitochondria as reported by Iversen (1970) could reflect true localization of myrosinase in situ.
 
 

Cellular and subcellular localization of myrosinase.

Attempts to localize myrosinase in plants have been reported for more than 100 years and involve either morphological, anatomical and histochemical observations or cytochemical, cell fractionation and biochemical studies. Using the former approach, several workers have considered myrosinase to occur in myrosin cells that stain specifically with Millon´s reagent, orcein solution and concentrated hydrochloric acid, and lactophenol aniline blue (see for example Rest and Vaughan 1972, Bones and Iversen 1985). Staining techniques have demonstrated that the grains in myrosin cells and aleurone-like cells are of different chemical origin. However, since no histochemical reagents were found which react specifically with either myrosinases or glucosinolates definitive conclusion about the myrosin cell content could not be made.

The first attempt to localize myrosinase at the cellular level was reported by Guignard in 1890. Since then a lot of attempts have been performed which have given strikingly different results (Peche 1913, Iversen 1970, Pihakaski and Iversen 1976, Matile 1980, Lüthy and Matile 1984). Techniques which have been used to localize myrosinase have included light - and electron-microscopy and combination of microscopy and detection of enzyme activity (Iversen, 1970) or microscopy combined with histochemical stains (Bones and Iversen 1985), cell- and subcellular fractionation (Pihakaski and Iversen 1976, Lüthy and Matile 1984), organell separation by centrifugation techniques, comparison of the number of myrosin cells with the detectable myrosinase activity, membrane fractionation and purification of myrosin cell protoplasts (Bones and Iversen 1985).

The first localization based on a cytochemical method was reported by Peche (1913). The method employed was based on enzymatic hydrolysis of the substrate sinigrin, which produced sulphate in situ on fresh sections of Rhapanus sativus. By including barium chloride in the substrate a precipitate of insoluble barium sulphate was deposited on the site of the enzyme reaction. This cytochemical method showed that myrosinase could be localized in myrosin cells. Iversen (1970) concluded that most of the myrosinase activity was confined to the dilated cisternae of the endoplasmic reticulum and in a limited extent to the mitochondria. Using the same cytochemical technique Maheswari et al. (1981) concluded that myrosinase was largely associated with the plasmalemma of most cells. Lüthy and Matile (1984) using subcellular fractionation suggested that most of the myrosinase activity was associated with the tonoplast, plasmalemma and endoplasmic reticulum, and that although myrosinase was present in the cytosol it has a remarkable tendency to adhere to membrane surfaces. Matile (1980) suggested that myrosinase was a cytosolic enzyme, while the glucosinolates are compartmentalized in the vacuoles. Pihakaski and Iversen (1976) conclude that the myrosinase activity is mainly found in dictyosomes and smooth endoplasmic reticulum.

The initial light- and electron microscope investigations indicated a connection between myrosin cells and myrosinase. However, evidence for the localization could not be obtained solely based on microscopy. The development of techniques based on specific antibodies opened up the possibility of performing an immunohistochemical localization of myrosinase. The combination of knowledge on myrosin cells, their distribution and specific antibodies finally enabled the localization of myrosinases to myrosin cells (Thangstad et al. 1990). Further immunogold-EM studies proved the subcellular localization of myrosinase to the protein bodies/vacuoles in myrosin cells (Thangstad et al. 1991). Examples of immunocytochemical and immunogold-EM localization of myrosinase is shown on Fig. 3. The cellular and subcellular localization was verified by Höglund et al. (1992). The possibility that minor amounts of myrosinase may be present in other compartments is still open (Thangstad et al. 1991). The cell and tissue specific localization of myrosinase has also been verified by in situ hybridization experiments with subfamily specific myrosinase probes (Xue et al. 1993).

As shown by Bones et al. (1991) myrosin cells do undergo considerable developmental changes during the first 14 days after seeding. This seems to include a dramatic change of the compartmentation due to a reorganization of the protein bodies/vacuoles (Bones et al., 1991).
 
 

Localization of glucosinolates and ascorbic acid

Glucosinolates and ascorbic acid have been reported to be localized in vacuoles of non-specific cells (Grob and Matile 1979, Matile 1980, Helminger et al. 1983). Recently the glucosinolate sinigrin was located to vacuoles of non-myrosin cells in cotyledons of Brassica juncea seeds/seedlings (Kelly et al. 1998) using immunogold-EM and antibodies raised against sinigrin conjugated to bovine serum albumin. As shown by several authors (e.g. Nagashima and Uchiyama 1959, Ettlinger et al. 1961, James and Rossiter 1991) the degradation rate of glucosinolate increases considerably in the presence of ascorbic acid. Grob and Matile (1980) examining root tissue of Armoracia rusticana found that 99.5% of the ascorbic acid was compartmentilized in vacuoles. In samples of root tissue they measured an ascorbic acid concentration of approximately 2.0 mM. Considering that they also state that nearly all ascorbic acid was localized in vacuoles the concentration of ascorbic acid in this organelle should be considerably higher.
 
 

Organisation of the myrosinase-glucosinolate system

More evidence for the localization of glucosinolates are needed, but is hard to obtain because of the water soluble characteristics of this group of compounds. One unsolved question is whether other types of glucosinolates are localized in the same vacuoles as sinigrin. Since glucosinolates are hydrolyzed only after the plant is injured, at least the following three alternatives are possible for the location of substrate and enzyme: In different cells; In different compartments of the same cell; And inside the same compartment of the same cell, but in an inactive form.

Myrosinase, glucosinolates and ascorbic acid may reside in the same compartment, the myrosin grains/vacuoles of myrosin cells. High concentrations of ascorbic acid inhibit and low concentrations activate myrosinase activity (Bones and Slupphaug 1989). Optimal ascorbic acid concentration for myrosinase from Armoracia rusticana has been reported to be 1.8 mM (Ohtsuru and Hata 1979). Together with the value for the total ascorbic acid concentration in Armoracia rusticana of 2.0 mM as reported by Grob and Matile (1980), this can be considered to support the hypothesis suggested above. If all ascorbic acid is compartmentilized in vacuoles, disruption of the tonoplast membranes should give a maximum activation of the myrosinase and therefore a maximum response after wounding. As discussed in Bones and Slupphaug (1989) ascorbic acid is an unusual activator. It inhibits at high concentrations and activates at lower concentrations. In a system like the one just described this would be an ideal effector molecule. Based on the results of Grob and Matile (1980) the ascorbic acid concentration in vacuoles of Armoracia rusticana must be considerably higher than 2.0 mM. At this concentration of ascorbic acid the myrosinase enzyme system would be inactive. The co-localization of myrosinase and glucosinolates would be most likely if this is a defence system. In this case, a co-localization makes the system a real "toxic mine" which would be activated simply by disruption of the tonoplast membranes. After disruption the ascorbic acid concentration will drop due to dilution and myrosinase will be activated.

A localization of glucosinolates and myrosinase in different vacuoles in the same cell is not likely. The vacuoles of a myrosin cell do undergo considerable changes after seeding (Bones et al. 1991) of which the fisions and fusions of the vacuoles in the cells are the most remarkable. To maintain the stability of such a system, a hitherto unknown mechanism for sorting and fusion of glucosinolate or myrosinase containing vacuoles is necessary. A more likely system would include compartmentation of glucosinolates in vacuoles of some cells and myrosinases in vacuoles of myrosin cells. The demolition of subcellular compartmentation by mechanical disruption or by micro- or macro-organisms feeding on the plants would cause the necessary contact between enzyme and substrate.

Another possiblity is that myrosinase is localized in myrosin cells and other components of the system in separate cells. To activate such a system enzymes or substrate must be transported or the organisation of a tissue broken.

The three sub-groups of myrosinases recently reported (Thangstad et al. 1993, Falk et al. 1995) may also have relevance to the organisation of the enzyme system. The two genes belonging to the Myr1 and Myr2 subgroups of the gene family were reported to have 3 and 9 potential glycosylation sites (Thangstad et al. 1993). Since glycosylation may be a requisite for a vacuolar localization due to the hydrophilic properties added with the carbohydrates, this could indicate that myrosinase also can be localized in other places than the vacuole/myrosin grain.

From the abovementioned it can be concluded that the cellular organisation of the myrosinase-glucosinolate system is unclear. Evidence shows that myrosinases, glucosinolates and ascorbic acid are localized in vacuoles, but with the exception of myrosinase which is localized in myrosin cells, the cellular localization of the other components including myrosinase associated proteins are unknown.
 
 

12. Plant-fungi interactions

Work in our laboratories have shown that infection with fungi can induce a local synthesis of myrosinase (Karapapa et al. unpublished). The possibility exists that other stress responses also induce a similar response. Initial results did not reveal any increase in total myrosinase activity, but this could be due to a high constitutive synthesis of myrosinase.
 
 

13. Plant-insect interactions

Mechanical wounding and infestation of oilseed rape (Koritsas et al. 1991) with cabbage stem flea beetle (Psylliodes chrysocephala) has been shown to induce indole glucosinolate biosynthesis while the aliphatic glucosinolates decreased. Similar effects have also been noticed in leaves of oilseed rape (Doughty et al. 1991) challenged with the dark leaf spot pathogen (Alternaria brassicae) roots (Birch et al. 1990) to the field slug (Deroceras reticulatum).
 
 

14. Plant-microorganism interactions
 
 

15. Responses to nutrients available

Since glucosinolates contain a significant proportion of sulphur and nitrogen it might be expected that fertilisers will influence the concentrations of glucosinolates in Brassica crops. It has been suggested that under conditions of sulphur deficiency sulphur bound in glucosinolates of Brassica species can be remobilised by enzymatic cleavage with myrosinase (Schnug et al. 1993). The mechanism for this is thought to involve the control of myrosinase activity by ascorbate /glutathione cycle. Studies have shown (Schnug et al. 1995) a close relationship between sulphur status and glucosinolate concentrations and glutathione although inter-dependencies for ascorbate are less apparent.

Field work with Brassica carinata, B. juncea and B. napus examining irrigation and nitrogen levels has been carried out and quality parameters such as protein and sinigrin content measured (Singh et al. 1994). Irrigation had no effect on these parameters although increases of nitrogen up to 120 kg N ha-1 increased sinigrin content. Work by Fieldsend and Milford (1994) with single and double low cultivars of oilseed rape demonstrated large differences in the ability to synthesise glucosinolates and showed these differences to be related to developing pods rather than vegetative tissues. Sulphur measurements showed that the glucosinolates contained only a small proportion of the crops sulphur and were unlikely to be a major source of recyclable sulphur even under conditions of severe sulphur deficiency. Glucosinolate profiles (Fieldsend and Milford 1994b) have been determined throughout the life cycle of four oilseed rape cultivars (Bienvenue, Ariana, Cobra and Capricorn) at defined stages of development. It was shown that substantial differences developed in the profiles of these compounds during vegetative growth. Changes in the profiles of glucosinolates throughout the plant´s development are thought to have implications for pests and diseases.

A decreasing sulphur supply to the plants results in a decrease in free sulphate and glucosinolate concentrations and an increase in myrosinase activity (Underhill 1980, Schnug 1990). This implies that the increase in myrosinase activity during sulphur stress could have the function of a remobilization of sulphate sulphur from glucosinolates, because sulphate and isothiocyanates can be utilized as sulphur sources in the primary metabolism of the plants (Machev and Schraudolf 1978). Sulphate has recently been shown to induce differential expression of myrosianses (Bones et al. 1994). This represents a hitherto unkown mechanism of activation by an inorganic compound. A possible function of the sulphate induction may be that an initial degradation caused by mechanical disruption of the tissue (and thereby the myrosin cells) release sulphate from degradation of glucosinolates which again signals that more myrosinase (and glucosinolates?) is needed.

It appears that various environmental stresses will affect myrosinase levels and glucosinolate concentrations and composition and more work is required to establish the underlying biochemical mechanisms that control this biosynthesis.
 
 

16. Plant growth hormones from glucosinolates?

In most higher plants de novo indole-3-acetic acid (IAA) biosynthesis proceeds from L-tryptophan via indole-3-pyruvic acid or tryptamine to indole-3-acetaldehyde, which is finally oxidized to IAA (Schneider and Weightman 1974). Species of Brassicaceae have, however, a complex indole metabolism where IAA biosynthesis may follow different pathways. In Brassicaceaea, indole-3-acetaldoxime is metabolized to indole-3-methylglucosinolate (glucobrassicin) and several of its derivatives which have been shown to accumulate in the vacuole (Helmlinger et al. 1983). The presence of in vivo conversion of indole-3-methylglucosinolate has been discussed, and it has been established that over the pH range from 4 to 7 indole-3-acetonitrile (IAN) may be formed by enzymatic degradation of indole-3-methylglucosinolate by myrosinase in vitro (Searle et al. 1982). One possibility is therefore that the indole glucosinolate pool can act as a sink for production of indole acetic acid (IAA). By a step to IAN the indole glucosinolates can be converted by a nitrilase (Bestwick et al. 1993) to IAA which in addition to its normal hormone action may also be involved in the response action after e.g. infestation, i.e. a part of the defence system of the plant. Four Arabidopsis thaliana cDNA´s encoding nitrilases with key roles in the biosynthesis of IAA has recently been cloned (Bartel and Fink 1994).
 
 

17. Isoforms with different functions?

It has been shown that the expression of specific myrosinases can be both organ specific (James and Rossiter 1991, Xue et al. 1993, Bones et al. 1994) and that some can be induced by specific components like sulphate (Bones et al. 1994) or jasmonic acid (L. Rask pers. comm.). James and Rossiter (1991) discussed the existence of two myrosinases in 5 day old cotyledons, of which one was glycosylated and one with low glycosylation. It was suggested that the glycosylated myrosinase could be involved in the defence system, because the enzyme is located throughout the seedling, and because it was twice as active in degrading indole glucosinolates as was the non-glycosylated myrosinase (myrosinase II). This explanation is supported by the results presented by Koritsas et al. (1991). They showed that the infestation of adult B. napus plants by the cabbage stem flea (Psylliodes chrysocephala L.) resulted in an accumulation of indole glucosinolates in damaged tissues.

It has been shown for B. napus and three other crucifers that while total glucosinolate levels drop during early seedling growth, endogenous ascorbate concentration increases (Sukhija et al. 1985). Furthermore, James and Rossiter (1991) have shown that the total "potential" myrosinase activity in seedlings, i.e. the enzyme activity in the presence of 0.3 mM ascorbate, increases during day 2-6 after seeding. This may indicate that myrosinase II is involved in hydrolysis of aliphatic glucosinolates. The exclusive cotyledonary location of myrosinase II supports this hypothesis. Glucosinolates can perhaps be considered to be a storage of reserve compounds that may be mobilized in the developing seedlings. A better knowledge of the sub-cellular localization of the compounds involved would, however, make these interpretations easier.
 
 

18. Future research.

There is an increase in the interest for the myrosinase-glucosinolate system. This is at least partially because some of the tools which are necessary for molecular investigation of the system have now been developed. Future research will focus on different topics of the system of which some will be briefly mentioned here.

The physiological role of glucosinolates will be investigated in detail. These studies will probably include experiments which will determine the role of glucosinolates for growth of the plants including the role at different stages such as the vegetative growth period, germination and flowering. There are already some indications that double zero oilseed rape is more sensitive to sulphur defiency than single zero plants. Although there have been attempts to correlate the levels of glucosinolates with e.g. growth and resistance, future experiments should be more specific and e.g. determine the direct effect of a low level of indole-glucosinolates on specific growth periods like flowering.

The myrosinase sequence data available show that there is at least three subgroups of the myrosinase gene family in Brassica napus. This may indicate that the different genes are involved in different processes in the plant. Information from expression studies using probes from different genes have already shown that myrosinases can have a tissue specific expression (Xue et al. 1993). Transformation of plants containing myrosinase and also plants which do not contain myrosinase will be used to study the effects of over-expression and under-expression (anti-sense). Localization of the glucosinolates and the myrosinase associated proteins at a sub-cellular level will help greatly in the work to determine the functions of the system.

The goal will be to understand the molecular mechanisms and thereby get a better understanding of biological processes where the myrosinase-glucosinolate system are involved. Another goal will be to use these results in breeding programs. The amount of, distribution of and mixture of specific glucosinolates are targets for breeding programs. For example high concentrations of glucosinolates in parts of plants which are not used for feed. Modification of plants to obtain the optimal combination of myrosinases and glucosinolates is a final goal. The results wanted will probably differ and depend on e.g. where the plants are to be used. Control of compounds working as insect repellants and attractants will probably be important also because one must expect a decline in the amounts of plant pesticides which can be used in the future. However, the glucosinolate-myrosinase system consists of more than 100 substrates and several enzyme forms. To engineer this system one needs specific information about the components and genes involved, and the function of the system, as well as how the elements involved are compartmentalized, the synthesis of glucosinolates and distribution of the compounds. A combination of traditional plant breeding and genetic engeenering could possibly produce some of the wanted traits.
 

19. Reference list
 

Bartel, B. & Fink, G.R. 1994. Differential regulation of an auxin-producing nitrilase gene family in Arabidopsis thaliana. - Proc. Natl. Acad. Sci. USA 91: 6649-6653.

Benn, M. 1977. Glucosinolates. - Pure and Appl. Chem. 49: 197-210.

Bennett, R.N. & Wallsgrove, R.M. 1994. Secondary metabolites in plant defence mechanisms. Tansley Review No. 72. - New Phytol. 127: 617-633.

Bestwick, L.A., Grønning, L.M., James, D.C., Bones, A. & Rossiter, J.T. 1993. Purification and characterization of a nitrilase from Brassica napus. - Physiol. Plant. 89: 811-816.

Birch, A.N.E, Griffiths, D.W. & MacFarlane Smith, W.H. 1990. Changes in forage and oilseed rape (Brassica napus) root glucosinolates in responce to attack by turnip root fly (Delia floralis). - J. Sci. Food Agric. 51: 309-320.

Björkman, R. 1976. Properties and function of plant myrosinases. - In The biology and chemistry of Cruciferae (Vaughan, J.G., A.J. MacLeod and B.M.G. Jones, eds.), pp. 191-205, Academic Press, London, ISBN 0-12-715150-8.

Björkman, R. & Janson, J.C. 1972. Studies on myrosinases. I. Purification and characterization of a myrosinase from white mustard seed (Sinapis alba L.). - Biochim. Biophys. Acta 276: 508-518.

Björkman, R. & Lönnerdal, B. 1973. Studies on myrosinases. III. Enzymatic properties of myrosinases from Sinapis alba and Brassica napus seeds. - Biochim. Biophys. Acta 327: 121-131.

Bones, A.M. 1990. Distribution of b-thioglucosidase activity in intact plants, cell and tissue cultures and regenerated plants of Brassica napus L. - J. Exp. Bot. 41: 737-744.

Bones, A.M. 1991. Compartmentation and molecular properties of thioglucoside glucohydrolase (myrosinase). Dr-thesis. Unigen Center for Molecular Biology, Department of Botany, University of Trondheim, pp. 1-195.

Bones, A.M & Iversen, T.-H. 1985. Myrosin cells and myrosinase. - Isr. J. Bot. 34: 351-375.

Bones, A.M. & Slupphaug, G. 1989. Purification, characterization and partial amino acid sequencing of b-thioglucosidase from Brassica napus L. - J. Plant Physiol. 134: 722-729.

Bones, A.M., Evjen, K. & Iversen, T.-H. 1989. Characterization and distribution of dilated cisternae of the endoplasmic reticulum in intact plants, protoplasts, and calli of Brassicaceae. Isr. J. Bot. 38: 177-192.

Bones, A.M., Visvalingam, S. & Thangstad, O.P. 1994. Sulphate can induce differential expression of thioglucoside glucohydrolases (myrosinases). - Planta 193: 558-566.

Bones, A.M., Thangstad, O.P., Haugen, O.A. & Espevik, T. 1991. Fate of myrosin cells: Characterization of monoclonal antibodies against myrosinase. - J. Exp. Bot. 42: 1541-1549.

Buchwaldt, L., Larsen, L.M., Plöger, A. & Sørensen, H. 1986. Fast polymer liquid chromatography isolation and characterization of plant myrosinase, b-thioglucoside glucohydrolase, isoenzymes. - J. Chromatography 363: 71-80.

Bussy, A. 1840. Sur la formation de l`huile essentielle de moutarde. - J. Pharm. 27: 464-471.

Chadchawan, S., Bishop, J., Thangstad, O.P., Bones, A.M., Mitchell-Olds, T. & Bradley, D. 1993. Arabidopsis cDNA sequence encoding myrosinase. - Plant Physiol. 103: 671-672.

Chew, F.S. 1988a. Biological effects of glucosinolates. - In Biologically active natural products for potential use in agriculture (Cutler, H.G., ed.), pp. 155-181. Am. Chem. Soc., Washington.

Chew, F.S 1988b. Searching for defensive chemistry in the cruciferae, or, do glucosinolates always control interactions of cruciferae with their potential herbivores and symbionts? No! (Spencer, K.A. ed.), pp. 81-112. Chemical mediation of coevolution. Academic Press, San Diego.

Cole, R.A. 1975. Cyanoepithioalkanes: nitriles and thiocyanate as products of autolysis of glucosinolates in Cruciferae. - Phytochemistry 15: 759-762.

Dahlgren, R. 1980. A revised system of classification of the angiosperms. - Bot. J. Linn. Soc. 80: 91-124.

Doughty, K.J., Porter, A.J.R., Morton, A.M., Kiddle, G., Bock, C.H. & Wallsgrove, R. 1991. Variation in the glucosinolate content of oilseed rape (Brassica napus L.) leaves II. Response to infection by Alternaria brassicae (Berk.) Sacc. - Ann. Appl. Biol. 118: 469-477.

Dunnick, J.K., Prejean, J.D., Haseman, J., Thompson, R.B., Giles, H.D. & McConnel, E.E. 1982. Carcinogenesis bioassay of allyl isothiocyanate. - Fundam. Appl. Toxicol. 2: 114-120.

Durham, P.L. & Poulton, J.E. 1990. Enzymic properties of purified myrosinase from Lepidium sativum seedlings. Z. Naturforsch. 45: 173-178.

Ettlinger, M.G., Dateo, G.P., Harrison, B.W., Mabry, T.J. & Thompson, C.P. 1961. Vitamin C as a coenzyme: The hydrolysis of mustard oil glucosides. - Proc. Nat. Acad. Sci. USA 47: 1875-1880.

Falk, A., Xue, J., Lenman, M.& Rask, L. 1992. Sequence of a cDNA clone encoding the enzyme myrosinase and expression of myrosinase in different tissues of Brassica napus. - Plant Sc. 83: 181-186.

Falk, A., Ek, B. & Rask, L. 1995. Characterization of a new myrosinase in Brassica napus. - Plant Mol Biol. 27: 863-874.

Falk, A., Taipalensuu, Ek, B., Lenman, M. & Rask, L. 1995. Characterization of rapeseed myrosinase-binding protein. - Planta 195: 387-395.

Fielsend, J. & Milford, G.F.J. 1994a. Changes in glucosinolates during crop development in single- and double-low genotypes of winter oilseed rape (Brassica napus). I. Production and distribution in vegetative tissues and developing pods and potential role in recycling of sulphur within the crop. - Annal. Applied. Biol. 124: 531-542.

Fieldsend, J. & Milford, G.F.J. 1994b. Changes in glucosinolates during crop development in single- and double-low genotypes of winter oilseed rape (Brassica napus). II. Profiles and tissue-water concentration in vegetative tissues and developing pods. - Annal. Applied. Biol. 124: 543-555.

Fenwick, G.R., Heaney, R.K. & Mullin, W.J. 1983. Glucosinolates and their breakdown products in food and food plants. - CRC Crit. Rev. Fd. Sci. Nutr. 18: 123-201.

Geshi, N. & Brandt, A. 1998. Two jasmonate-inducible myrosinase-binding proteins from Brassica napus L. seedlings with homology to jacalin. Planta 204: 295-304.

Goodman, I., Fouts, J.R. Bresnick, E., Menegas, R. & Hitchings, G.H. 1959. A mammalian thioglycosidase. - Science 130: 450-451.

Grob, K. & Matile, Ph. 1979. Vacuolar location of glucosinolates in horseradish root cells. - Plant Sci. Lett. 14: 327-335.

Grob, K. & Matile, Ph. 1980. Compartmentation of ascorbic acid in vacuoles of horseradish root cells. Note on vacuolar peroxidase. - Z. Pflanzenphysiol. 98: 235-243.

Guignard, L. 1890. Sur la localisation des principes qui fournissent les essences sulfurées des Crucifères. C.R. Acad.Hebd. - Seances III: 249-251.

Haughn, G.W., Davin, L., Giblin, M. & Underhill, E.W. 1991. Biochemical genetics of plant secondary metabolites in Arabidopsis thaliana. The glucosinolates. - Plant Physiol. 97: 217-226.

Hanley, B.A., Parsley, K.R., Lewis, J.A. & Fenwick, R.G. 1990. Chemistry of indole glucosinolates: Intermediacy of indole-3-ylmethyl isothiocyanates in the enzyme hydrolysis of indole glucosinolates. - J. Chem. Soc. Perkin Trans 1. 2273-2276

Hasapis, X. & MaCleod, A.J. 1982. Benzylglucosinolate degradation in heat-treated Lepidium sativum seeds and detection of a thiocyanate-forming factor. - Phytochemistry 21: 1009-10013.

Hegnauer, R. 1986. Phytochemistry and plant taxonomy - An essay on the chemotaxonomy of higher plants. - Phytochemistry 25: 1519-1535.

Heinricher, E. 1884. Ueber eiweissstoffe führende idioblasten bei einigen Cruciferen. - Ber. Deutsch. Bot. Ges. II: 463-467.

Helmlinger, J., Rausch, T. & Hilgenberg, W. 1983. Localization of newly synthesized indole-3-methylglucosinolate (=glucobrassicin) in vacuoles from horseradish (Armoracia rusticana). - Physiol. Plant. 58: 302-310.

Henderson, H.M. & McEwen, T.J. 1972. Effect of ascorbic acid on thioglucosidases from different crucifers. - Phytochemistry 11: 3127-3133.

Hillebrand, H., Bartling, D & Weiler, E.W. 1998. Structural analysis of the nit2/nit1/nit3 gene cluster encoding nitrilases, enzymes catalyzing the terminal activation step in indole-acetic acid biosynthesis in Arabidopsis thaliana. Plant Mol. Biol. 36: 89-99.

Höglund, AS., Lenman, M. & Rask, L. 1992. Myrosinase is localized to the interior of myrosin grains and is not associated to the surrounding tonoplast membrane. - Plant Sci. 85: 165-170.

Ioannou, Y., Burka, L.T. & Matthews, H.B. 1984. Allyl isothiocyanate: comparative disposition in rats and mice. - Toxicol. and Appl. Pharmacol. 75: 173-181.

Iversen, T.-H. 1970. Cytochemical localization of myrosinase (b-thioglucosidase) in root tips of Sinapis alba L. - Protoplasma 71: 451-466.

James, D. & Rossiter, J.T. 1991. Development and characteristics of myrosinase in Brassica napus during early seedling growth. - Physiol. Plant. 82: 163-170.

Jørgensen, L.B. 1981. Myrosin cells and dilated cisternae of the endoplasmic reticulum in the order Capparales. - Nord. J. Bot. 1: 433-445.

Koritsas, V.M., Lewis, J.A. & Fenwick, G.R. 1991. Glucosinolate responses oilseed rape, mustard and kale to mechanical wounding and infestation by cabbage stem flea beetle (Psylliodes chrysocephala). - Ann. Appl. Biol. 118: 209-221.

Labague, L., Gardrat, G., Coustille, J.L., Viaud, M.C. & Rollin, P. 1991. Identification of enzymatic degradation products from synthesised glucobrassicin by gas chromatography-mass spectrometry. - J. Chromatography 586: 166-170.

Lenman , M., Falk, A., Xue, J. & Rask, L. 1993. Characterization of a Brassica napus myrosinase pseudogene: myrosinases are members of the BGA family of b-glycosidases. - Plant Mol. Biol. 21: 463-474.

Lenman, M., Rödin, M., Josefsson, L-G. & Rask, L. 1990. Immunological characterization of rapeseed myrosinase. - Eur. J. Biochem. 194: 747-753.

Lüthy. B. & Matile, Ph. 1984. The mustard oil bomb: Rectified analysis of the subcellular organisation of the myrosinase system. - Biochem. Physiol. Pflanzen 179: 5-12.

Lönnerdal, B. & Janson, J.-C. 1973. Studies on myrosinases. II. Purification and characterization of a myrosinase from rapeseed (Brassica napus L.). - Biochim. Biophys. Acta 315: 421-429.

MacGibbon, D.B. & Allison, R.M. 1970. A method for the separation and detection of plant glucosinolases (myrosinases). - Phytochemistry 9: 541-544.

MacGibbon, D.B. & Beuzenberg, E.J. 1978. Location of glucosinolase in Brevicoryne brassicae and Lipaphis erysimi (Aphididae). - New Zealand J. Sci. 21:389-392.

Machev, N.P. & Schraudolf, H. 1978. Thiocyanate as predecessor of asparagine in Sinapis alba L. - Plant Physiol. (Bulgaria) 4: 26-33.

Machlin, S., Mitchell-Olds, T. & Bradley, D. 1993. Sequence of a Brassica campestris myrosinase gene. - Plant Physiol. 102: 1359-1360.

MacLeod, A.J. & Rossiter, J.T. 1985. The occurrence of activity of epithiospecifier protein in some Cruciferae seeds. - Phytochemistry 24: 1895-1898.

MacLeod, A.J. & Rossiter, J.T. 1987. Degradation of 2-hydroxybut-3-enylglucosinolate (progoitrin). - Phytochemistry 26: 669-673.

Matile, Ph. 1980. "Die senfölbombe": Zur Kompartimentierung des Myrosinasesystems. - Biochem. Physiol. Pflanz 175: 722-731.

Nagashima, Z. & Uchiyama, M. 1959. Possibility that myrosinase is a single enzyme and mechanism of decomposition of mustard oil glucoside by myrosinase. - Bull. Agric. Biol. Chem. Soc. Japan 23: 6: 555-556.

Oginsky, E.L., Stein, A.E. & Greer, M.A. 1965. Myrosinase activity in bacteria as demonstrated by the conversion of progoitrin to goitrin. - Soc. Exp. Biol. Med. Proc. 119: 360-364.

Ohtsuru, M. & Hata, T. 1972. Molecular properties of multiple forms of plant myrosinase. - Agric. Biol. Chem. 36: 2495-2503.

Ohtsuru, M. & Hata, T. 1973. Studies on the activation mechanism of the myrosinase by L-ascorbic acid. - Agr. Biol. Chem. 37: 1971-1972.

Ohtsuru, M. & Kawatani, H. 1979. Studies of the myrosinase from Wasabia japonica: Purification and some properties of Wasabi myrosinase. - Agric. Biol. Chem. 43: 2249-2255.

Ohtsuru, M., Tsuruo, I. & Hata, T. 1973. The production and stability of intracellular myrosinase from Aspergillus niger. - Agr. Biol. Chem. 37: 967-971.

Peche, K. 1913. Microchemical detection of myrosin. - J. Chem. Soc. 106: II, 79-80.

Petroski, R.J. & Kwolek, W.F. 1985. Interactions of a fungal thioglucoside glucohydrolase and cruciferous plant epithiospecifier protein to form 1-cyanoepithioalkanes, implications of an allosteric mechanism. - Phytochemistry 24: 213-216.

Petroski, R.J. & Tookey, H.L. 1982. Interactions of thioglucoside glucohydrolase and epithiospecifier protein of cruciferous plants to form 1-cyanoepithioalkanes. - Phytochemistry 21: 1903-1905.

Pihakaski, K. & Iversen, T.-H. 1976. Myrosinase in Brassicaceae. I. Localization of myrosinase in cell fractions of roots of Sinapis alba L. - J. Exp. Bot. 27: 242-258.

Reese, E. T., Clapp, R.C. & Mandels, M. 1958. A thioglucosidase in fungi. - Arch. Biochem. Biophys. 75: 228-242.

Rest, J.A. & Vaughan, J.G. 1972. The development of protein and oil bodies in the seed of Sinapis alba L. - Planta 105: 245-262.

Rodman, J.E. 1991. A taxonomic analysis of glucosinolate-producing plants, Part 1: Phenetics. Systematic Bot. 16: 598-618.

Rossiter, J.T. 1983. Synthesis and degradative studies of progoitrin. PhD-thesis. Queen Elisabeth College, University of London, pp. 1-255.

Schnug, E. 1990. Glucosinolates - fundamental, environmental and agricultural aspects. - In: Sulfur nutrition and sulfur assimilation in higher plants (Rennenberg, H., C. Brunold, L.J. de Kok and I. Stulen, eds.), pp. 97-106. SPB Academic Publishing, Hague. ISBN 90-5103-038-X.

Schnug, E., Haneklaus, S., Borchers, A. & Polle, A. 1995. Relations between sulphur supply and glutathione and ascorbate concentrations in Brassica napus. - Zeitschrift für Pflanzenernahrung und Bodenkunde 158: 67-69.

Searle, L.M., Chamberlain, K. Rausch, T. & Butcher, D.N. 1982. The conversion of 3-indolylmethylglucosinolate to 3-indolylacetonitrile by myrosinase, and its relevance to the clubroot disease of the Cruciferae. - J. Exp. Bot. 33: 935-942.

Spatzier, W. 1893. Ueber das Auftreten und die physiologische Bedeutung des Myrosins in der Pflanze. - Jahrb. Wiss. Bot.. 25: 39-79.

Springett, M.B. & Adams, J.B. 1989. Properties of brussels sprouts thioglucosidase. - Food Chemistry 33: 173-186.

Sukhija, P.S., Loomba, A., Ahuja, K.L. & Munshi, S.K. 1985. Glucosinolate and liquid content in developing and germinating cruciferous seeds. - Plant Sci. 40: 1-6.

Taipalensuu, J., Eriksson, S. & Rask, L. 1997. The myrosinase-binding protein from Brassica napus seeds possesses lectin activity and has a highly similar vegetatively expressed wound-inducible counterpart. Eur. J. Biochem. 250: 680-688.

Taipalensuu, J., Falk, A. & Rask, L. 1996. A wound- and methyl jasmonate- inducible transcript coding for a myrosinase-associated protein with similarities to an early nodulin. Plant Physiol. 110: 483-491.

Taipalensuu, J., Lundgren, S. & Rask, L. 1995. No evidence for the ascorbic activation site of myrosinase being encoded by end of exon-9, beginning of exon-10. Hereditas 122: 95-98.

Taipalensuu, J., Andreasson, S., Eriksson, S. & Rask, L. 1997. Regulation of the wound-induced myrosinase-associated protein transcript in Brassica napus plants. Eur. J. Biochem 247: 963-971.

Taipalensuu, J., Falk, A., Ek, B. & Rask, L. 1997. Myrosinase-binding proteins are derived from a large wound-inducible and repetitive transcript. Eur. J. Biochem. 243: 605-611.

Tani, N., Ohtsuru, M. & Hata, T. 1974. Isolation of myrosinase producing microorganism. - Agr. Biol. Chem. 38: 1617-1622.

Thangstad, O.P., Evjen, K. & Bones, A. 1991. Immunogold-EM localization of myrosinase in Brassicaceae. Protoplasma 161: 85-93.

Thangstad, O.P., Iversen, T.-H., Slupphaug, G. & Bones, A. 1990. Immunocytochemical localization of myrosinase in Brassica napus L. - Planta 180: 245-248.

Thangstad, O.P., Winge, P., Husebye, H. & Bones, A. 1993. The thioglucoside glucohydrolase (myrosinase) gene family in Brassicaceae. - Plant Mol Biol. 23: 511-524.

Tookey, H.L. 1973. Crambe thioglucoside glucohydrolase (EC 3.2.3.1): Separation of a protein required for epithiobutane formation. - Can. J. Biochem. 51: 1654-1660.

Tsuruo, I. & Hata, T. 1968. Studies on myrosinase in mustard seeds. Part IV. Sugars and glucosides as competitive inhibitors. - Agr. Biol. Chem. 32: 1420-1424.

VanSteenhouse, J.L., Fettman, M.J. & Gould, D.H. 1991. The effect of glutathione depletion by buthione sulphoximine on 1-cyano-3,4 epithiobutane toxicity. - Food Chem. Toxicol. 29: 153-157.

Wallig, M.A., Gould, D.H., Fettman, MJ. & Willhite, C.C. 1988. Comparative toxicities of the naturally ocurring nitrile 1-cyano-3,4-epithiobutane and the synthetic nitrile n-valeronitrile in rats: Differences in target organs, metabolism and toxic mechanisms. - Food Chem. Tox. 26: 149-157.

Yusuf, S.W. & Collins, G.G. 1998. Effect of soil sulphur levels on feeding preferences of Brevicoryne brassicae on brussels sprouts. J. Chem. Ecology 24: 417-424.

Xue, J., Philgren, U. & Rask, L. 1993. Temporal, cell-specific, and tissue-preferential expression of myrosinase genes during embryo and seedling development in Sinapis alba. - Planta 191: 95-101.

Xue, J., Jørgensen, M., Philgren, U. & Rask, L. 1995. The myrosinase gene family in Arabidopsis thaliana: gene organisation, expression and evolution. - Plant Mol. Biol. 27: 911-922.

Xue, J., Lenman, M., Falk, A. & Rask, L. 1992. The glucosinolate-degrading enzyme myrosinase in Brassicaceae is encoded by a gene family. - Plant Mol. Biol. 18: 387-398.