Occurrence, characterization and biological role of BXs

Benzoxazinoids (BXs) are protective and allelophatic secondary metabolites found in numerous species belonging to the Poaceae family, including maize, rye, wheat (Niemeyer 1988a; Grün et al. 2005; Nomura et al. 2007; Frey et al. 2009; Chu et al. 2011; Sue et al. 2011), the wheat progenitors Triticum urartu, Aegilops speltoides, Aegilops squarrosa (Niemeyer 1988b), and wild barleys: Hordeum roshevitzii, Hordeum flexuosum, Hordeum brachyantherum, Hordeum lechleri (Grün et al. 2005) and in genera—Chusquea, Elymus, Arudo (Zùñiga et al. 1983), Coix (Nagao et al. 1985), but not in Avena (Hamilton 1964), Hordeum vulgare or its progenitor Hordeum spontaneum (Grün et al. 2005). These compounds are also present in single species within a few dicot families: Acathaceae—Acanthus mollis (Wolf et al. 1985), Aphelandra tetragona, Blepharis edulis and Aphelandra squarrosa (Baumeler et al. 2000); Rannuncuaceae (Gierl and Frey 2001)—Consolida orientalis (Sicker et al. 2000), Plantaginaceae (Gierl and Frey 2001)—Scoparia dulcis (Chen and Chen 1976) and Lamiaceae—Lamium galeobdolon (Sicker et al. 2000).

BXs were first discovered and characterized in rye (Virtanen and Hietala 1955a, b), wheat and maize (Wahlroos and Virtanen 1959), in the 1950s. According to Hanhineva et al. (2011), three classes of compounds comprise the BXs: hydroxamic acids (HAs), lactams and benzoxazolinones, whereas Niemeyer (2009) divided BXs into HAs, lactams and methyl derivatives (Table 1).

Table 1 Classification of benzoxazinoid compounds (according to Niemeyer 2009; Hanhineva et al. 2011)
Full size table

According to Niemeyer (2009), HAs are the most active class of BXs by virtue of a hydroxyl group bound to the heterocyclic nitrogen atom, although studies on pests of maize done by Cambier et al. (2001) and Glauser et al. (2011) showed that the methylated forms of HAs are far more toxic than de-methylated ones. For example, HDMBOA-Glc was proved to have higher toxic impact on aphid Metopolophium dirhodum than DIMBOA-Glc (Cambier et al. 2001).

The biosynthesis of BXs is usually at its highest during the juvenile stage of plant growth, then it subsequently declines and becomes stabilized at a lower level (Ebisui et al. 1998; Nomura et al. 2005, 2008). This pattern of biosynthesis is reflected by the transcript levels of the relevant enzymes (Rad et al. 2001; Nomura et al. 2005; Sue et al. 2006).

It should be pointed out that BXs biosynthesis is influenced by the plant genotype and cultivar (Niemeyer 2009) environmental conditions such as photoperiod (Epstein et al. 1986), light intensity (Manuwoto and Scrrber 1985) and the application of fertilizer (Manuwoto and Scrrber 1985). Moreover, introduction of foreign genes may also have an impact on BX production: in transgenic Bt corn, the level of BXs was found to be significantly lower than in the control non-transgenic plants (Nie et al. 2005). The authors suggested that the Bt gene has adverse effects on the biosynthesis and accumulation of DIMBOA and some phenolic acids, such as ferulic acid. However, under conditions of either water or nitrogen stress, the accumulation of DIMBOA in the leaves of the Bt corns could be enhanced.

In dicot plants, the BX content is comparable with that of monocot seedlings (Frey et al. 2009). However, in monocots, both shoots and roots produce BXs, while dicot species (e.g. A. squarrosa, L. galeobdolon and C. orientalis) synthesize these compounds in the above-ground parts of plants and in roots but at low levels. High levels of BXs have been detected in several adult parts of dicots: flower buds and flowers of C. orientalis, and leaves of L. galeobdolon and monocots: crown roots of maize (Schullehner et al. 2008). In some cases, the elevated biosynthesis of those compounds was observed in older plants; for instance in older whole plants (including the roots) of rye wounded mechanically (Kruidhof et al. 2014).

BXs are secondary metabolites that play roles in allelopathy and defense

The most effective allelopathic compounds are DIBOA, DIMBOA and their breakdown products BOA and MBOA (Barnes and Putnam 1987; Tabaglio et al. 2008). Their toxicity is due to the oxidation of cell wall peroxidases with associated production of H2O2, accumulation of lignins, disruption of lipid metabolism and protein synthesis, reduction of transport and/or secretory capabilities, and decreased H+-ATPase activity (Barnes and Putnam 1987; Tabaglio et al. 2008).

Among cereals, rye shows the strongest allelopathic potential with the ability to reduce the germination, growth and development of many weeds (e.g. Lepidium sativum, Amaranthus retroflexus, Echinoehloa crusgalli, Portulaca oleracea) and crop plants (e.g. cucumber, melon, tomato, lettuce, maize, tobacco) by up to 98 % (Barnes and Putnam 1987; Tabaglio et al. 2008). The allellopathic activity of rye depends on the season of the year: in autumn it is far higher than in early spring and, additionally, may increase after mechanical damage (Kruidhof et al. 2014). It seems to have a particular application effect: by wounding-increased allelopathic activity of rye applied as cover crop plant it would be possible to reduce the amount of around-growing weeds more efficiently, although, as underlined by the above mentioned authors, these effects might not be sufficient to compensate for the loss in biomass resulting from wounding.

In many species belonging to the Poaceae family, BXs are a crucial element in their defense mechanisms against pests, e.g. European Corn Borer (ECB, Ostrinia nubilalis) in maize (Klun et al. 1967; Barry et al. 1994), aphids (Sitobion avenae) in wheat (Bohldar et al. 1986) and nematodes in rye (Zasada et al. 2005). BX-based plant defenses against aphids and whorl feeding larvae seem to be connected with the anti-feeding properties of these compounds, namely their inhibition of digestive proteases responsible for detoxification and pest salivation (Feng et al. 1992). In addition, BXs appear to increase plant resistance to viruses transmitted by aphids, e.g. BYDV in wheat transmitted by Rhopalosiphum padi (Givovich and Niemeyer 1991).

Several studies have shown a relationship between BXs and disease resistance. However, the correlation between BXs content and disease resistance is not always positive, and is dependent on their site of synthesis and the character of the pathogen (Long et al. 1978; Søltoft et al. 2008).

Enhanced BXs synthesis can be induced by pathogenic organisms and also by tissue wounding (Basse 2005; Kruidhof et al. 2014). Wounding of maize plants caused increased DIMBOA synthesis, producing levels similar to those after infection by Ustilago maydis (Basse 2005). Persans et al. (2001) showed that the CYP71C1 and CYP71C3 of maize, encoding cytochrome P450s, involved in DIMBOA biosynthesis, were induced in response to wounding and treatment with naphthalic anhydride. After successive defoliations, the BX content in rye shoots was found to decline, but simultaneously it increased in the roots (Collantes et al. 1999). Glauser et al. (2011) showed also that BX derivatives such as HDMBOA-Glc and HDM2BOA-Glc have a toxic impact both on S. littoralis and S. frugiperda despite that S. frugiperda have a strong detoxification capacity in relation to DIMBOA.

BXs have also been shown to play a role in (1) improving plant tolerance to soil salinity (Makleit 2005), (2) the detoxification of triazine derivatives (Marcacci et al. 2005) and aluminum (Poschenrieder et al. 2005), (3) preventing chlorotic symptoms by forming chelates with iron (Pethô 2002), and (4) the inhibition of gibberellin-induced α-amylase activity in barley seeds (Kato-Noguchi 2008).

BXs importance is not only limited to the plant defense strategies, but also, like in case of majority plant secondary metabolites, they have strong, positive influence on human health. They are considered to be capable of lowering cancer risk (Zhang et al. 1995; Roberts et al. 1998) and insulin secretion (Landberg et al. 2010). Moreover, they have anti-allergic properties (Otsuka et al.1988; Poupaert et al. 2005), appetite suppression and weight reduction effects (Rosenfeld and Forsberg 2009). However, the content of BXs in final food products strongly depends on the type of grain processing (Tanwir et al. 2013; Hanhineva et al. 2014).

The biosynthesis of BXs

The BXs biosynthetic pathway has been studied in maize (Frey et al. 1997; Rad et al. 2001; Jonczyk et al. 2008), diploid, tetraploid, hexaploid Triticales (Nomura et al. 2002, 2005, 2008) and wild barley (Grün et al. 2005), but so far not in dicots (Schullehner et al. 2008). The most detailed study was performed in maize.

The first step, a branchpoint in BXs biosynthesis that occurs in chloroplasts, is the conversion of indole-3-glycerolphosphate to indole. The products of the next four reactions that take place in endoplasmic reticulum are indolin-2-one, 3-hydroxy-indolin-2-one, HBOA and DIBOA (Frey et al. 1997, 2009; Gierl and Frey 2001; Grün et al. 2005; Chu et al. 2011). Subsequently, the glucosylation of DIBOA to 2-O-β-glucoside (which is stored in vacuole) occurs prior to hydroxylation and O-methylation reactions in the cytoplasm that produce TRIBOA-Glc and DIMBOA-Glc, respectively (Gierl and Frey 2001; Grün et al. 2005; Jonczyk et al. 2008). Then DIMBOA-Glc and DIBOA-Glc can be transported to the vacuole. After hydroxylation DIBOA-Glc and DIMBOA-Glc are converted into DIBOA and DIMBOA, respectively, and released from vacuole to cytosol. Recently, Meihls et al. (2013) discovered other step: O-methylation reaction catalyzed by O-methyl transferases ZMBX10a÷ZMBX10c resulting in converse of DIMBOA-Glc into HDMBOA-Glc (Fig. 1). In several dicot plant species, DIBOA-Glc is the final product of BXs biosynthesis (Sicker et al. 2000).

Fig. 1
figure 1

BX biosynthesis pathway. Based on Sue et al. (2011), Dutartre et al. (2012) and Meihls et al. (2013)

Full size image

Upon disintegration of the cell due to pathogen or pest attack and mobilization of jasmonic acid and/or its methyl ester, glucosidases stored in the chloroplast are activated to produce toxic aglucons (Oikawa et al. 2002; Niemeyer 2009). DIMBOA is the main aglucon in maize and wheat (Niemeyer 1988a; Frey et al. 1997), whereas in rye (Niemeyer 1988a; Gierl and Frey 2001; Zasada et al. 2005) and wild barley (Niemeyer et al. 1992; Grün et al. 2005) it is DIBOA.

Aglucons with an open ring structure generate the highly reactive α-oxo-aldehyde, which reacts with numerous nucleophiles occurring in amino acid residues (particularly thiols and amines) of proteins including catalytic enzymes, and this may account for the inhibition of various metabolic processes, such as electron transport in mitochondria and chloroplasts or NADH oxidation by the cell wall (Niemeyer 2009). DIBOA and DIMBOA are unstable compounds, and are spontaneously degraded in the soil to BOA and MBOA, respectively (Zasada et al. 2005; Meyer et al. 2009).

Genetic basis of BX biosynthesis

Several genes controlling the biosynthesis of BXs have been isolated and characterized. The BX biosynthetic genes of maize are ZmBx1–ZmBx10a÷c, Zmglu1 and Zmglu2. These encode the following enzymes: ZmBx1–indole-3-glycerol phosphate lyase; ZmBx2–ZmBx5—members of the CYP71C subfamily of cytochrome P450 monooxygenases; ZmBx6–2-oxoglutarate dependent dioxygenase, ZmBx7–O-methyltransferase; ZmBx8, ZmBx9—glucosyltransferases; ZmBx10a÷ZmBx10c–O-methyltransferase; Zmglu1, Zmglu2–glucosylglucosidases (Frey et al. 1997; Gierl and Frey 2001; Rad et al. 2001; Jonczyk et al. 2008; Meihls et al. 2013).

Orthologs of the maize genes Bx1–Bx5 have also been identified in hexaploid Triticum aestivum (Nomura et al. 2002, 2003), diploid wheat Triticum boeoticum (Nomura et al. 2007), Secale cereale (Rakoczy-Trojanowska et al. 2013; Bakera et al. 2015) and Hordeum lechleri (Grün et al. 2005). Genes encoding glucosylglucosidases (Taglu1a, Taglu1b, Taglu1c, Taglu1d, Scglu) and glucosyltransferases (TaGTa, TaGTb, TaGTc, TaGTd, ScGT) have been identified in hexaploid wheat (Sue et al. 2006) and rye (Nikus et al. 2003; Sue et al. 2011).

For the majority of Bx genes, only the sequences of their mRNAs are known (Table 2). The exceptions are: maize ZmBx1÷ZmBx5 and wheat TaBx3 and TaBx4, for which cds, introns, 3′ UTRs, 5′ UTRs and putative promoters sequences have been described (Kramer and Koziel 1995; Frey et al.1997; Nomura et al. 2005, 2008) and in rye: ScBx1÷ScBx5 sequences, including cds, introns, 3′ UTRs (http://www.ncbi.nlm.nih.gov/nuccore: KF636828, KF620524, KF636827, KF636826, KF636825, Table 2). La Hovary 2012 published the complete cDNA with 3′ UTR of ScBx1 gene. Promoter sequences and 5′ UTRs have also been reported for the genes TuBx3, TuBx4 (Triticum urartu), AtBx3, AtBx4 (Aegilops tauschii), AsBx3 and AsBx4 (Aegilops speltoides), (Nomura et al. 2008). However, the availability of genomic sequence data means that full sequences of the other Bx genes of maize and wheat may readily be predicted. For example, the ZmBx1 gene sequence was used as the query sequence in a BLAST screen of the B73 maize genome and a homologous gene was identified (http://www.ncbi.nlm.nih.gov/nuccore/AC200309.3:82,911–85,155,GRMZM2G085381), (Butrón et al. 2010). These authors then used an identical approach to obtain maize gene orthologs of Bx2–Bx5 and Bx8. By the same method, TAC clones of T. aestivum the full-length sequences of the TaBx3 and TaBx4 genes (http://www.ncbi.nlm.nih.gov/nuccore/AB298184.1-AB298186.1) were identified (Nomura et al. 2008).

Table 2 GenBank accession numbers of Bx genes of selected Poaceae species
Full size table

Interestingly, the genes Bx6 and Bx7, which encode enzymes catalyzing sequential 7-hydroxylation and 7-O-methylation of DIBOA-Glc to DIMBOA-Glc, have only been identified in maize, and appear to be absent from wheat, Hordeum lechleri and rye (Sue et al. 2011). Therefore, the identity of the genes controlling the transformation of DIBOA-Glc to DIMBOA-Glc in these cereal species is unknown. Recently, however, Tanwir et al. isolated Bx6-like gene from rye cv. Picasso (http://www.ncbi.nlm.nih.gov/nuccore/HG380520.1) which showed, on the nucleotide level, more than 78 % identity to Zea mays 2-oxoglutarate dependent dioxygenase (Bx6) (http://www.ncbi.nlm.nih.gov/nuccore/AF540907).

The regulation of Bx gene expression has been studied in Triticum. In hexaploid wheat, the expression of the TaBx1–TaBx5 genes is co-regulated, but despite this, their level of transcription depends on the genome (A or B or D) on which the particular genes are located (Nomura et al. 2005). All genes located on chromosomes in genome B (TaBx1B–TaBx5B, TaGTa–TaGTb, Taglu1a and Taglu1b) are transcribed at a higher level than those from genomes A and D (Nomura et al. 2005; Sue et al. 2011). The same phenomenon has been observed in diploid progenitors of hexaploid wheat: T. urartu (genome A) and A. tauschii (genome D) display lower level Bx gene transcription than A. speltoides, the donor of genome B (Sue et al. 2011).

Phylogenetic relationships between Bx genes

As mentioned above, the biosynthesis of BXs is catalyzed by enzymes encoded by Bx genes: Bx1–Bx9. Bx1 is considered a gene at the cluster branch point (whose evolution can be traced back to the duplication and functionalization of an ancestor encoding the alpha-subunit of tryptophan synthase—TSA) because the BX biosynthesis pathway is initiated by the indole-3-glycerol phosphate lyase encoded by this gene, which mediates the transformation of indole-3-glycerol-phosphate into indole (Frey et al. 2009). In Hordeum spontaneum and most Hordeum vulgare varieties, in which the defensive system is based on the indole alkaloid gramine [3-(dimethylaminomethyl)-indole], a Bx1 ortholog is absent. Analysis of a phylogenetic tree of Bx1 sequences from maize, wheat, rye and wild barley drawn according to Saitou and Nei (1987) indicates that they share a monophyletic origin (Fig. 2). However, there is relatively high structural dissimilarity between HlBx1 and ZmBx1. To explain this difference, Grün et al. (2005) proposed that HlBx1 should be evaluated separately. Phylogenetic analysis of the Bx1-encoded enzymes in three BX-producing dicots, C. orientalis, L. galeobdolon and A. squarrosa, led to the conclusion that BX biosynthesis evolved independently in dicot and monocot plants. Surprisingly, despite little similarity between their amino acid sequences, the BX1 enzymes of maize and wheat, and CoBX1, their ortholog from C. orientalis, have comparable catalytic properties (Schullehner et al. 2008).

Fig. 2
figure 2

Phylogenetic tree of Bx1 sequences in chosen Poaceae species. The phylogenetic tree is generated by Mega 6 software (Tamura et al. 2013) based on the Neighbor Joining (NJ) algorithm (Saitou and Nei 1987) and Maximum Composite Likelihood substitution method with pairwise deletion and bootstrap analysis of 500. cDNA sequences from GenBank were translated into protein sequences followed by alignment done by means of ClustalW (attached to Mega6). After that, the protein sequences were transformed again into cDNA sequences and used for further analysis. The bootstrap values are indicated at the branch points. Scale bar indicates number of substitutions per site. Sequences from Arabidopsis thaliana TSA genes were used as outgroup. Individual labels were used for each Bx1 cds from organisms: Sc: Secale cereale, Ta: Triticum aestivum, Zm: Zea mays, Zl: Zea luxurians, Zn: Zea nicaraguensis, Zd: Zea diploperennis, Zp: Zea perennis, Sb: Sorghum bicolor, Si: Setaria italica, Os: Oryza sativa, Bd: Brachypodium distachyon, Hl: Hordeum lechleri, Tt: Triticum turgidum, At: Arabidopsis thaliana

Full size image

A second group of genes controlling BX biosynthesis is composed of Bx2–Bx5, encoding four CYP71 monooxidases that are responsible for the sequential introduction of four oxygen atoms into the indole moiety, yielding DIBOA. Structurally, the Bx2–Bx5 genes of wheat, maize, rye and wild barley share high homology and form four clades on the phylogenetic tree (Fig. 3). Their conserved structure suggests that their progenitor evolved before the divergence of the Triticeae and the Panicoideae (Grün et al. 2005; Frey et al. 2009). The CYP71 family proteins encoded by the Bx2–Bx5 genes are highly substrate specific, which most probably is a result of duplication of a common ancestor gene followed by neofunctionalization (Frey et al. 2009; Chu et al. 2011).

Fig. 3
figure 3

Phylogenetic tree of Bx2÷5 sequences. Mega 6 software (Tamura et al. 2013) based on the Neighbor Joining (NJ) algorithm (Saitou and Nei 1987) and Maximum Composite Likelihood substitution method with pairwise deletion and bootstrap analysis of 500. cDNA sequences from GenBank were translated into protein sequences followed by alignment done by means of ClustalW (attached to Mega6). After that, the protein sequences were transformed again into cDNA sequences and used for further analysis. The bootstrap values are indicated at the branch points. Scale bar indicates number of substitutions per site. Sequences from Arabidopsis thaliana CYP76 genes were used as outgroup. Individual labels were used for each Bx2÷5 cds from organisms: Sc: Secale cereale, Ta: Triticum aestivum, Zm: Zea mays, Os: Oryza sativa, Hl: Hordeum lechleri, Tm: Triticum monococcum subsp. aegilopoides, At: Arabidopsis thaliana

Full size image

Orthologs of Bx2–Bx5 have so far not been identified in rice or sorghum. The rice genes Cyp71c16 and Cyp71c17 appear to be related to Bx2 of other Poaceae, but their function is unclear. The remaining BX enzymes: BX3÷BX5 have also not been detected in these species (Frey et al. 2009).

A third group of Bx genes comprises those encoding glucosyl transferases and glucoside glucosidases. From the studies of von Rad et al. (2001) and Sue et al. (2011), and our own unpublished analysis, it may be concluded that the wheat and rye GT and glu genes correspond to maize ZmBx8 and Zmglu, respectively. The ZmBx9 gene seems to have arisen by the duplication of ZmBx8, or it may be a remnant of paleotetraploidy of the maize genome (Swigoňová et al. 2004).

The structural polymorphism of Bx genes has so far been investigated in only two species: maize and rye. Using a study population consisting of 281 maize diverse inbred lines, Butrón et al. (2010) identified 45 INDELs and 44 SNPs in four amplicons for ZmBx1, 6 INDELs and 11 SNPs in one amplicon for ZmBx2, 3 INDELs and 19 SNPs in one amplicon for ZmBx3, 7 INDELs and 2 SNPs in two amplicons for ZmBx4, 10 INDELs and 35 SNPs in three amplicons in ZmBx5, and no polymorphisms of greater than 5 % frequency for ZmBx8. Twenty-eight polymorphisms of ZmBx1 and one polymorphism of ZmBx2 were significantly associated with the leaf content of DIMBOA and DIMBOA-Glc. Numerous SNPs and INDELs were found within fragments (675–832 bp) of the ScBx5 gene (comprising part of first and second exon, and first intron) in winter rye (inbred line L318) and its wild relatives: Secale cereale ssp. africanum, S. cereale ssp. ancestrale, S. cereale ssp. dighoricum, S. cereale ssp. segetale, S. strictum, S. strictum ssp. africanum, S. strictum ssp. anatolicum, S. strictum ssp. ciliatoglume, S. strictum ssp. kuprianovii S. strictum ssp. strictum, S. sylvestre and S. vavilovii (Rakoczy-Trojanowska et al. 2013). The analyzed sequence is partially collinear with the cytochrome P450 domain, including CYP cysteine heme–iron ligand signature. The total number of SNPs was 201, on average 2.7 SNP occurred per 10 nt and the total number of INDELs-50, averagely 0.7 INDELs per 10 nt. It should be emphasized that the majority of SNPs (5 times more abundant in intron than in two analyzed exons) and all INDELs were present in the intron. The most repeated SNP type in exons was C/G when in intron A/T and the most rare A/T and C/G, respectively. The amount of SNPs and INDELs differed between the examined species and ranged from 1 (S. cereale ssp. dighoricum and S. strictum ssp. ciliatoglume) to 99 (S. strictum) and from 1 (S. cereale ssp. africanum) to 13 (S. strictum), respectively. The longest insertion (106 bp) was identified in S. strictum and the longest deletion (51 bp)—in S. strictum ssp. kuprijanovii. No INDELs were detected in amplicons of S. cereale ssp. segetale, S. vavilovii, S. cereale ssp. dighoricum and S. strictum ssp. ciliatoglume.

Mapping

All identified genes controlling the biosynthesis of BXs in maize, wheat and rye have been mapped. Most of them are organized in clusters, especially the maize Bx genes. Gene clustering is a common feature in bacteria (Zheng et al. 2002), but less so in plant genomes, particularly in the case of genes encoding enzymes participating in secondary metabolism, which are mostly unlinked (Frey et al. 2009). However, recent studies have shown several examples of occurrence of genes clusters in plants and all of them are involved in secondary metabolite biosynthesis; besides BXs, to this group belong linamarin and lotaustralin in Lotus japonicus, diterpenes in Oryza sativa, avenacin in Avena spp., thianol and marneral in Arabidopsis thaliana, noscapine in Papaver somniferum, steroidal glycoalkaloid in Solanum lycopersicum and Solanum tuberosum (reviewed in Boycheva et al. 2014).

It is generally agreed that the maize genes encoding the oxidative enzymes of BX biosynthesis (ZmBx1–ZmBx6) as well as the O-methyl transferase (ZmBx7) and glucosyl transferase (ZmBx8) are localized on the short arm of chromosome 4, while ZmBx9, a close homolog of ZmBx8, is situated on chromosome 1, and two genes encoding glucosyl glucosidases (Zmglu1, Zmglu2) have been mapped to the short arm of chromosome 10 (Frey et al. 1997; Rad et al. 2001; Jonczyk et al. 2008). The chromosomal location of Bx6 gene in maize genome is controversial. Although Dutartre et al. (2012) failed to locate the ZmBx6 gene in the maize genomic sequence and they found only a close paralog on the long arm of chromosome 2, our recent analysis applying B73 genome revealed that the sequence deposited in NCBI data base (http://www.ncbi.nlm.nih.gov/nuccore/AF540907.1) is identical with 1,252,700–1254047 nt fragment of chromosome 4 (http://www.ncbi.nlm.nih.gov/nuccore/NC_024462.1) and similar to the 231,954,236–231,953,295 nt fragment of chromosome 2 (http://www.ncbi.nlm.nih.gov/nuccore/NC_024460.1) at 84 %. The newest discovered genes in maize-ZmBx10a÷ZmBx10c are located on chromosome 1, mapped to bin 1.04, approximately 66,304,872–66,500,692 bp (Meihls et al. 2013).

In hexaploid wheat, the Bx gene cluster is divided between group 4 (TaBx1 and TaBx2) and group 5 (TaBx3÷TaBx5), with an additional copy of TaBx3 on genome B (Nomura et al. 2003). The genes Taglus and TaGTs have been mapped to chromosomes belonging to groups 2 and 7, respectively (Sue et al. 2011). The order of genes TaBx3÷TaBx5 is still not agreed. According to Nomura et al. (2008) these genes are ordered in the following manner (in relation to centromere): Bx4–Bx3–Bx5 whereas from the figure published by Nomura et al. (2003) and Sue et al. (2011) there might be concluded that these genes are located on the short arms of chromosomes from group 5 in another way: Bx3–Bx4–Bx5.

In rye, genes ScBx1 and ScBx2 are located on chromosome 7R, ScBx3÷ScBx5—on chromosome 5R (Nomura et al. 2003) and ScGT (an ortholog of ZmBx8/ZmBx9 coding for BX glucosyltransferase) and Scglu (an ortholog of Zmglu1/Zmglu2 coding for BX glucosylglucosidase) are located separately on chromosomes 4R and 2R, respectively (Sue et al. 2011). Similarly as in wheat, the chromosomal arrangement of ScBx3÷ScBx5 has not been determined yet.

The differences between the locations of the Bx genes in the genomes of maize, wheat and rye can be explained by numerous rearrangements of monocotelydoneous proto-chromosomes during the course of evolution (Salse et al. 2009). This hypothesis might explain the ancestry of individual chromosomes of modern cereals and the Bx genes they carry: (1) wheat chromosome group 4 (with TaBx1, TaBx2) and 5 (with TaBx3÷TaBx5), rye chromosomes 5 (with ScBx3÷ScBx5) and 7 (with ScBx1, ScBx2), and maize chromosome 4 (with ZmBx1÷ZmBx8) originated from protochromosome A11; (2) wheat chromosome group 2 (with Taglua, Taglub, Tagluc, Taglud), rye chromosome 2 (with Scglu) and maize chromosome 10 (with Zmglu1, Zmglu2) originated from protochromosome A4, and (3) wheat chromosome group 7 (with TaGTa, TaGTb TaGTc TaGTd), rye chromosome 4 (with ScGT) and maize chromosomes 1 (with ZmBx9) and 4 (ZmBx8) originated from protochromosome A8 (Sue et al. 2011; Dutartre et al. 2012). Rye chromosomes 7R and 5R show synteny with the long arms of hexaploid wheat group-4 chromosomes where TaBx1 and TaBx2 are located, and the short arms of group 5 chromosomes (with Tabx3, TaBx4), respectively (La Hovary 2012 upon Devos et al. 1993).

Although 5 Bx genes (HlBx1–HlBx5) have been isolated from Hordeum lechleri (Grün et al. 2005), their genomic location is currently unknown (Sue et al. 2011).

Bx genes have not so far been identified in cultivated barley (Grün et al. 2005). It is possible that Hordeum vulgare has “lost” these genes during the domestication process (Gierl and Frey 2001) or through chromosomal rearrangement (Grün et al. 2005).

Several QTLs controlling resistance to leaf feeding by a native American lepidopteran species (Bohn et al. 2001), European corn borer (ECB), (Cardinal et al. 2006; Butrón et al. 2010) and the Asian corn borer (ACB), (Li et al. 2010) have been identified in maize on chromosome 4, close to the Bx gene cluster (Cardinal et al. 2006; Li et al. 2010). Furthermore, two of these QTLs (umc123 and php200713) have been mapped to the region of the Bx cluster (Cardinal et al. 2006). Further studies are required to prove the proposed correlation between the level of DIMBOA biosynthesis and resistance to leaf feeding damage by ACB and ECB (Li et al. 2010).

Outstanding questions

Despite the growing body of data on the genetic mechanisms controlling the biosynthesis of BXs, there are still many unanswered or poorly resolved questions: (1) which enzymes mediate the transformation of DIBOA-Glc into DIMBOA-Glc in wheat, rye and other cereals besides maize, and which genes encode these enzymes? (2) Is the ScBx6-like gene of rye a functional ortholog of maize ZmBx6? (3) On which chromosome(s) are located genes Bx6 and Bx7 of rye, wheat and other cereals accumulating BX? (4) How are Bx3, Bx4 and Bx5 loci ordered in wheat and rye? (5) On which arm of chromosome 1 are mapped genes ZmBx10a÷ZmBx10c? (6) What is the location of ZmBx6 gene in maize genome and if there is only one or two copies of this gene? (7) How has cultivated barley lost the ability to synthesize BXs and has the proposed elimination of all Bx loci due to degeneration of the coding sequences, silencing, or loss of one Bx locus activated the loss of all other Bx loci, as proposed by Nomura et al. (2007)? (8) How do different biotic and abiotic stresses regulate the expression of Bx genes? (9) Which polymorphisms are functionally relevant to BX biosynthesis?

Author contribution statement

Contribution to this publication inserted through the co-authors was equal.