Genetic Analysis, Expression and Molecular Characterization of BoGSL-ELONG, a Major Gene Involved in the Aliphatic Glucosinolate Pathway of Brassica Species

Corresponding author: Carlos F. Quiros, University of California, Davis, CA 95616., cfquiros{at}ucdavis.edu (E-mail)

Communicating editor: C. S. GASSER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We cloned a major aliphatic glucosinolate (GSL) gene, BoGSL-ELONG in Brassica oleracea, using the Arabidopsis sequence database. We based our work on an Arabidopsis candidate gene forming part of a gene family coding for isopropyl malate synthetase-like enzymes (IPMS). This gene is presumably responsible for synthesis of GSL possessing side chains consisting of four carbons (4C). The similarity of the Brassica homolog IPMS-Bo from broccoli to its Arabidopsis counterpart IPMS-At was on the order of 78%, both sharing the same number of exons. A nonfunctional allele of the BoGSL-ELONG gene from white cauliflower, based on the absence of 4C GSL in this crop, displayed a 30-bp deletion, which allowed us to develop a codominant marker for 4C-GSL. Gene expression analysis based on RT-PCR revealed a splicing site mutation in the white cauliflower allele. This resulted in a longer transcript containing intron 3, which failed to excise. Perfect cosegregation was observed for broccoli and cauliflower alleles at the IPMS-Bo gene and 4C-GSL content, strongly indicating that this gene indeed corresponds to BoGSL-ELONG. Cloning of two other major genes, BoGSL-ALK and BoGSL-PRO, is underway. The availability of these genes and BoGSL-ELONG is essential for the manipulation of the aliphatic GSL profile of B. oleracea.

GLUCOSINOLATES (GSLs) are secondary metabolites synthesized by many species in the order Capparales, including those in the family Brassicaceae. Isothiocyanates, which arise after GSL breakdown by hydrolytic action of the enzyme myrosinase, have diverse and important biological activities including carcinogen detoxification as well as inhibition of pathogenic fungal growth, among others (ROSA et al. 1997 ; MITHEN et al. 2000 ; MITHEN 2001 ). Aliphatic GSL derives from methionine (UNDERHILL 1980 ), which is converted by three major enzymatic pathways, including (1) amino acid side-chain elongation, (2) synthesis of the glycone moiety, and (3) aglycone side-chain modification reactions (HAUGHN et al. 1991 ).

GSL studies in Arabidopsis thaliana (MITHEN et al. 1995 ; MITHEN and CAMPOS 1996 ; MITHEN 2001 ) and Brassica species (MAGRATH et al. 1994 ) provide evidence for the proposed biochemical pathway of these compounds. Genetic analysis indicates that aliphatic GSL synthesis is controlled by a genetic system with two distinct sets of genes, one set controlling side-chain elongation and the second set involved in controlling the modification of side-carbon chains. Carbon chain elongation is probably catalyzed by isopropyl malate synthases (IPMS; CAMPOS DE QUIROS et al. 2000 ). Aliphatic GSL profiles vary considerably in A. thaliana and Brassica species. These compounds are synthesized in the order of methylsulfinylalkyl, alkenyl, and hydroxy types and can be grouped by the size of their side chains, which is determined by the number of carbons in those chains. In Brassica oleracea, the model for biosynthesis of aliphatic GSL and the genes acting in the main steps of this process are shown in Fig 1. In this model, the presence of the dominant allele for the BoGSL-ELONG gene will result in four-carbon (4C) GSL, whereas the presence of the dominant allele for BoGSL-PRO will result in three-carbon (3C) GSL. This expectation is supported by studies of the inheritance of 3C side-chain and 4C side-chain GSL in segregating populations of B. oleracea. Plants carrying both dominant alleles were found to produce both 3C and 4C GSL, whereas plants carrying the null alleles at both loci display only traces, if any, of aliphatic GSL (LI et al. 2001 ).

View larger version (15K): In this window In a new window Download PPT slide   Figure 1. Model for aliphatic glucosinolate biosynthesis in B. oleracea, including the inferred major genes controlling this process (MITHEN et al. 1995 ; LI et al. 2001 ).

In Arabidopsis, the GS-ELONG locus was mapped on chromosome V (CAMPOS DE QUIROS et al. 2000 ). There were two duplicated and contiguous isopropylmalate synthase-like genes (IPMS-At1 and IPMS-At2), located on two flanking bacterial artificial chromosome (BAC) clones, T20O9 and MYJ24, respectively. These genes were identified as candidate genes of GS-ELONG (CAMPOS DE QUIROS et al. 2000 ; KROYMANN et al. 2001 ). We report in this article the identification of the Brassica homolog for this gene, BoGSL-ELONG, determining the presence of 4C GSL in B. oleracea. We also sequenced this gene and developed molecular markers that can be used to do marker-based selection to increase glucoraphanin content, an important 4C GSL, in B. oleracea crops, such as cauliflower, cabbage, and broccoli. This and two other genes, BoGSL-ALK and BoGSL-PRO, are essential for the manipulation of the aliphatic GSL profile of B. oleracea.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plant materials:
Twenty-one commercial varieties of three different B. oleracea crops, broccoli, cauliflower, and collard, and four doubled-haploid (DH) lines of broccoli and cauliflower were included in this study (Table 1).

DNA amplification and library screening:
Using the sequence of the GS-ELONG candidate genes from Arabidopsis [IPMS-At1(At5g23010), IPMS-At2 (At5g23020)], a pair of primers (IPM1, 5'-GCCATCTTCGCACCCAAA-3' and IPM2, 5'-GTGACGGTGAACAATCTCCT-3') was designed to amplify the corresponding region of the B. oleracea homolog. These primers were designed to amplify part of exon 1, exon 2, and the intervening intron between these two exons. For this purpose we used broccoli genomic DNA, extracted as reported by LI and QUIROS 2001 . The PCR conditions for amplification for 35 cycles were: 94°, 1 min; 56°, 1 min; 72°, 2 min. The resulting amplified DNA was confirmed to correspond to the IPMS genes by sequencing.

Primers IPM1 and IPM2 were then used to screen a BAC library constructed with the broccoli doubled-haploid line "Early Big-10" (QUIROS et al. 2001 ) for clones harboring IPMS Brassica homologs. Two rounds of PCR were used for the library screening by 3-D pooling of the clones following the strategy of KOES et al. 1995 .

Partial BAC sequencing was done using the SRAP protocol as described by LI and QUIROS 2001 . Plasmid DNA from BAC clones was prepared following the plasmid minipreparation protocol as described by SAMBROOK et al. 1989 . DNA was fingerprinted using the SRAP protocol. Procedures for DNA collection from the gels and sequencing were as reported by LI and QUIROS 2001 .

Cosegregation analysis:
An F₂ population of 450 plants, generated by crossing doubled-haploid lines from cauliflower (An-Nan-83) and broccoli (Early Big-10), was used for cosegregation analysis between 4C GSL and the IPMS candidate genes. This population was previously used for genetic analysis of the aliphatic GSL biosynthesis (LI et al. 2001 ). A third primer (IMP9), based on the Brassica IPMS homolog together with primer IPM2, was used to amplify DNA from individual plants of the F₂ population. The sequence of IPM9 was 5'-GTAGTATTCTCAAAATCTTGT-3'. The PCR conditions were the same as those described above. The amplified products were separated using a LI-COR (Lincoln, NE) IR2 sequencer.

Gene expression analysis:
We used reverse-transcription (RT)-PCR to do gene expression analysis. We designed primers located in exon 3 (5'-AAGCGATCAAAGCGGGTG-3) and exon 4 (5'-CTTCAAGCGGTGCATTCC-3'), where a splicing site change in the candidate B. oleracea gene IPMS BoGLS-ELONG occurs in white cauliflowers, as explained in the results. For RT-PCR, total RNA was prepared as described by SAMBROOK et al. 1989 . Ten micrograms of RNA was used to do reverse transcription using the RT-PCR kit from Life Technologies.

Glucosinolate determination:
Glucosinolate profiles in leaves were determined by high-performance liquid chromatography (HPLC) using the method described by KRALING et al. 1990 with some modifications. For this purpose we grind 2 g of fresh leaves collected from 6-week-old seedlings in liquid nitrogen. Ground tissue was extracted twice with 70% methanol at 80° for 10 min. After applying the supernatant to a DEAE-Sephadex A-25 (Sigma, St. Louis) column, the glucosinolates were converted into desulfoglucosinolates with 0.5% sulfatase H-1 (Sigma) in water for 16 hr at room temperature. The desulfoglucosinolates were then eluted by adding 1.5 ml water. The resulting desulfoglucosinolates were separated by HPLC in a gradient of acetonitrile. The HPLC chromatographs were compared to the chromatograph of "Linetta," a rapeseed variety widely used as a standard for glucosinolate identification. Qualitative assessment of GSL was done visually by the presence or absence of the specific peaks. GSL content was quantified with glucotropaeolin (E.M. Science, Gibbstown, NJ) as an internal standard. Glucosinolate content was expressed as micromoles of GSL per gram of fresh leaves. We corrected the data for UV response factors for different types of glucosinolates (WATHELET et al. 2001 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Amplification of broccoli DNA with the IPMS primers produced one band displaying sequence identities of 86.3 and 85.0% with exons 1 and 2 of IPMS-At1 and IPMS-At2, respectively.

In total we isolated 16 BAC clones from the broccoli library with the IPMS-designed primers. These BAC clones were divided into three putative cistronic groups according to their sequence similarity to IPMS-At genes and to their BAC-end sequences. One of these three groups consisted of five BAC clones, B5B10, B11I7, B13D10, B19N3, and B39I16. In addition to the conserved portion of IPMS-At genes, all five clones had one end sequence that matched that next to the IPMS-At gene in Arabidopsis. Furthermore, the end sequences of B5B10, B19N3, and B39I16 were similar to Arabidopsis gene MYJ24.14, and those of B11I7 and B13D10, to gene MKD15.5. BAC clone MKD15, containing this gene, is contiguous to clone MYJ24. In total, 15 fragments of broccoli BAC B19N3 were sequenced using the SRAP protocol. After BLAST analysis, we found 1 fragment that matched Arabidosis gene MYJ24.2, which is next to IPMS-At2, (MYJ24.1). These results indicated that these five BAC clones contained the IPMS-At homolog (IPMS-Bo), likely matching the IPMS genes in Arabidopsis. Through direct BAC sequencing, we obtained the complete sequence of IPMS-Bo (GenBank accession no. AF399834). Similar to IPMS-At1 and IPMS-At2, IPMS-Bo also contains 10 exons. Except for exons 1 and 10, all others share the same size in all three genes. At the amino acid level, IPMS-Bo shares 78 and 75% identity to IPMS-At1 and IPMS-At2, respectively. The size of intron 1 of IPMS-Bo is considerably larger, being twice the size of the corresponding intron in IPMS-At1 and four times that of IPMS-At2. On the basis of this analysis, IPMS-Bo has higher similarity to IPMS-At1 than to IPMS-At2 (Table 2).

To confirm that candidate gene IPMS-Bo corresponded to the BoGSL-ELONG, we amplified with primers IPM9 and IPM2 the parental lines of the segregating F₂ population resulting from crossing cauliflower and broccoli. Using these primers, we successfully developed a codominant marker, which detected a 30-bp deletion in intron 1 in cauliflower. Among 383 plants of the F₂ population, 89 plants lacked 4C GSL and all these plants were homozygous for the smaller-size cauliflower marker. All plants with 4C GSL carried at least one broccoli allele. Therefore, there was complete cosegregation between 4C GSL content and the IPMS-based marker (Fig 2).

View larger version (17K): In this window In a new window Download PPT slide   Figure 2. Broccoli x cauliflower F2 population segregating for codominant marker (intense bands): top band from broccoli and bottom band from cauliflower. Change in band size is due to 30-bp deletion in cauliflower haplotype. + indicates presence of 4C GSL, and - indicates absence of 4C GSL.

Initially, our genetic analysis was mainly focused on the parental lines and their derived segregating population. In this preliminary survey we had observed that white cauliflower varieties did not have 4C GSL. To confirm this observation, we extended our GSL survey to the varieties and doubled-haploid lines listed in Table 1. Among the varieties, there were 15 white cauliflowers, 3 purple cauliflowers, four broccolis, and a collard (Table 1). The glucosinolate composition of this material is presented in Table 3. All white cauliflower varieties had phenotype BoGLS-ELONG-/BoGSL-PRO⁺/BoGSL-ALK⁺ since the alkyl GSLs they contained were exclusively 3C GSLs (glucoiberin and sinigrin or only glucoiberin). All 3 purple cauliflower varieties contained either the 4C GSL (glucoraphanin) or both 4C and 3C GSL (glucoiberin and glucoraphanin). Accessions B314 and B485 containing 3C and 4C GSL had phenotype BoGSL-ELONG⁺/GSL-PRO^-+/GSL-ALK^- whereas B265 had phenotype BoGSL-ELONG⁺/BoGSL-PRO^-/BoGSL-ALK^-. On the other hand, the broccoli varieties had exclusively 4C GSL, glucoraphanin, that is, phenotype BoGSL-ELONG⁺/GSL-PRO^-+/GSL-ALK^-, whereas in the collard variety, sinigrin (3C) and progoitrin (4C) were the predominant GSLs (>90% of total aliphatic GSL). Therefore the phenotype of this crop was BoGSL-ELONG⁺/BoGSL-PRO^-/BoGSL-ALK⁺.

GSL analysis was also performed in two white cauliflower doubled-haploid line populations, in two broccoli doubled-haploid line populations, and in their original F₁ hybrid parental varieties. The same results were obtained as those described above, where white cauliflower had only 3C glucosinolates and broccoli 4C glucosinolates. Noteworthy is the fact that sinigrin segregated among the cauliflower doubled-haploid lines, indicating that the desaturation gene BoGSL-ALK was heterozygous in both F₁ hybrid parental varieties of these lines. The broccoli DH lines had only glucoraphanin, indicating that BoGSL-ALK was null in this material.

With the sequence of candidate gene BoGSL-ELONG in hand, we proceeded to determine whether this gene was expressed in the B. oleracea varieties. All white cauliflower varieties tested were considered phenotypically as BoGSL-ELONG^- since they lacked 4C GSL and therefore were expected to carry the null allele for this gene. On the other hand, the broccoli varieties tested had 4C GSL, thus being BoGSL-ELONG⁺. When we performed RT-PCR with cDNA from broccoli and white cauliflower with the primers on the basis of the sequence of BoGSL-ELONG, we detected a polymorphism resulting in bands of two different sizes for each crop type. After sequencing both bands, we found that a mutation in the white cauliflower allele caused a splicing site change (intron 3 failed to excise), resulting in a larger-size cDNA (Table 2, Fig 3). The larger-size cDNA band cosegregated with absence of 4C GSL in the F₂ population of broccoli x cauliflower (Fig 4). This allele was present in all white cauliflower varieties and DH lines we tested.

View larger version (13K): In this window In a new window Download PPT slide   Figure 3. Diagrammatic representation at approximate scale of IPMS genes in Arabidopsis and the two Brassica alleles homologous to these genes. Boxes represent exons. In the cauliflower allele exons 3 and 4 are fused due to a splicing mutation (diagram based on KROYMANN et al. 2001 ).

View larger version (15K): In this window In a new window Download PPT slide   Figure 4. Sample of segregation in a broccoli x cauliflower F2 population for cDNA bands of two sizes resulting from the splicing site mutation illustrated in Fig 3. The larger band (top) corresponds to the mutant allele from white cauliflower. + indicates presence of 4C GSL, and - indicates absence of 4C GSL.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Now that the sequence of the Arabidopsis genome is available, many genes have been annotated at a high rate of speed in this species due to the extensive input and effort from many laboratories throughout the world. To transfer this information to crops for their improvement, finding Arabidopsis homologs of genes of economic importance has become a research priority. Arabidopsis gene sequences are now being used to clone useful genes controlling important agronomic traits in crop plants. This is illustrated by our work where we successfully used the Arabidopsis candidate gene GS-ELONG to clone and characterize its corresponding homolog in B. oleracea. Through cosegregation analysis and gene expression, we further confirmed that the IPMS-At genes are indeed the best candidate genes for 4C GSL synthesis, although final confirmation still will have to come in the future by complementary transformation. Our strategy was based on the conservation of gene sequence and gene order along the chromosome between Arabidopsis and Brassicas. At the nucleotide level, we (QUIROS et al. 2001 ) have found 80–90% identity levels in the coding region of both species. The linear arrangement of genes is fairly well conserved, although rearrangements are common (QUIROS et al. 2001 ). Taking as a starting point the Arabidopsis candidate gene for 4C-GSL synthesis, we picked a fairly large number of clones containing sequences matching the IPMS-At genes. This was due to the fact that IPMS is actually a gene family consisting of four members in Arabidopsis (CAMPOS DE QUIROS et al. 2000 ; KROYMANN et al. 2001 ). In spite of this complication, we were able to identify the clones harboring the GS-ELONG homolog. After sequencing the Brassica homolog IPMS-Bo, we successfully developed a codominant marker, which made it easier to do cosegregation analysis. The complete linkage of this marker with the presence of 4C GSL in the F₂ population strongly indicated that candidate gene IPMS-Bo is likely to correspond to BoGSL-ELONG. Further confirmation was provided by the cDNA marker on the basis of the splicing mutation, which cosegregated with 4C GSL content in the same F₂ population. By sequence comparison, we found that IPMS-Bo is more similar to IPMS-At1 than to IPMS-At2 at the translation and structural levels. In Arabidosis, IPMS-At1 was identified as functional whereas IPMS-At2 was nonfunctional in a heterologous expression system using Escherichia coli (KROYMANN et al. 2001 ). However, complementary transformation with different constructs may provide final evidence for this assessment. For the first time, we used differential gene expression to match phenotypic differences in GSL biosynthesis. The splicing mutation of the IPMS-Bo allele present in all white cauliflower varieties and breeding lines that we tested associated perfectly with the absence of 4C GSL. Therefore, it should be possible to easily introduce a functional BoGSL-ELONG allele into white cauliflower to create new types containing glucoraphanin, a 4C GSL, which releases an isothiocyanate possessing the ability to upregulate carcinogen detoxification enzymes in mammalian cells (MITHEN 2001 ). The purple cauliflower varieties, which are classified as cauliflowers, are actually intermediate types more closely resembling broccoli in inflorescence type (SMITH and KING 2000 ). Therefore, it was not surprising to find that they contain glucoraphanin, which is a 4C GSL typical of broccoli (KUSHAD et al. 1999 ). SMITH and KING 2000 provided a historic scenario on the origin of broccoli and cauliflower, postulating that cauliflower derived from broccoli via Sicilian purple cauliflower is an intermediate in a two-step process. This conclusion was reached according to inflorescence morphology and presence of specific alleles determining this trait, such as those at gene BoCAL-a and BoAP1 loci. The broccoli-to-cauliflower inflorescence transformation, discounting color, occurred by two successive mutations in these genes, the first originating purple cauliflower and the second white cauliflower. Putting this in the GSL gene context, most broccoli with phenotype BoGSL-ELONG⁺/BOGSL-PRO^-/BoGSL-ALK^- will require a single gain-of-function mutation at gene BoGSL-ALK to produce the purple cauliflower phenotype BoGSL-ELONG⁺/BOGSL-PRO^-/BoGSL-ALK⁺. The passage from purple cauliflower to white cauliflower of phenotype BoGSL-ELONG^-/BoGSL-PRO⁺/BoGSL-ALK⁺ will require two additional mutations, a loss-of-function mutation at BoGSL-ELONG and a gain-of-function mutation at BoGSL-PRO, to result in 3C GSL synthesis. Unless there are some other varieties of purple cauliflower that we have not tested, which do not have the BoGSL-ALK⁺ allele, a simpler alternative is that broccoli and white cauliflower had independent origins of domestication and purple cauliflower is the result of the hybridization of these two crops.

Cloning of the BoGSL-ELONG gene opens new avenues for Brassica breeding. This is the second major gene cloned in the aliphatic GSL pathway of Brassica. The chain modification gene BoGSL-ALK has been already cloned in this species (G. LI and C. F. QUIROS, unpublished data). With these two genes in hand, it might be possible to utilize other varieties or wild forms of B. oleracea with high glucoraphanin content or high level of GSL in general to maximize the content of glucoraphanin or specific GSL of interest in any Brassica crop and not only in purple cauliflower and broccoli. The use of wild species has been already explored by FAULKNER et al. 1997 . The tools are now in place to improve Brassica crops for type and content of GSL by either marker-assisted selection or genetic transformation.

	FOOTNOTES

Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under accession no. AF399834.

	ACKNOWLEDGMENTS

We are indebted to Vincent D'Antonio, Bo Yang, Muquiang Gao, and Pierina Benavente for technical assistance. Our thanks also to Drs. Elizabeth Earle and Peter McVetty for critical reading of the manuscript and Steffen Abel for providing access to his HPLC. This work was supported by United States Department of Agriculture Institute of Food and Agricultural Sciences grant "Development of Genomic Tools for Brassica oleracea."

Manuscript received April 23, 2002; Accepted for publication September 17, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

CAMPOS DE QUIROS, H., R. MAGRATH, D. MCCALLUM, J. KROYMANN, and D. SCNABELRAUCH et al., 2000  -Keto acid elongation and glucosinolate biosynthesis in Arabidopsis thaliana. Theor. Appl. Genet. 101:429-437.

FAULKNER, K., R. MITHEN, and G. WILLIAMSON, 1997  Selective increase of the potential anticarcinogenic 4-methylsinphinylbutyl glucosinolate in broccoli. Carcinogenesis 19:605-609.[Abstract/Free Full Text]

HAUGHN, G. W., L. DAVIN, M. GIBLIN, and E. W. UNDERHILL, 1991  Biochemical genetics of plant secondary metabolites in Arabidopsis thaliana. The glucosinolates. Plant Physiol. 97:217-226.[Abstract/Free Full Text]

KOES, R., E. SOUER, A. VAN HOUWELINGEN, L. MUR, and C. SPELT et al., 1995  Targeted gene inactivation in petunia by PCR-based selection of transposon insertion mutants. Proc. Natl. Acad. Sci. USA 92:8149-8153.[Abstract/Free Full Text]

KRALING, K., G. ROBBELEN, W. THIES, M. HERRMANN, and M. R. AHMADI, 1990  Variation of seed glucosinolates in lines of Brassica napus.. Plant Breed. 105:33-39.

KROYMANN, J., S. TEXTOR, J. G. TOKUHISA, K. L. FALK, and S. BARTRAM et al., 2001  A gene controlling variation in Arabidopsis glucosinolate composition is part of the methionine chain elongation pathway. Plant Physiol. 127:1077-1088.[Abstract/Free Full Text]

KUSHAD, M. M., A. F. BROWN, A. C. KURILICH, J. A. JUVIK, and B. P. KLEIN et al., 1999  Variation of glucosinolates in vegetable crops of Brassica oleracea.. J. Agric. Food Chem. 47:1541-1548.[Medline]

LI, G. and C. F. QUIROS, 2001  Sequence-related amplified polymorphism (SRAP), a new marker system based on a simple PCR reaction: its application to mapping and gene tagging in Brassica. Theor. Appl. Genet. 103:455-461.

LI, G., A. RIAZ, S. GOYAL, S. ABEL, and C. F. QUIROS, 2001  Inheritance of three major genes involved in the synthesis of aliphatic glucosinolates in Brassica oleracea. J. Am. Soc. Hort. Sci. 126:427-431.

MAGRATH, R., M. MORGEN, F. BANO, I. PARKIN, and A. SHARPE et al., 1994  Genetics of aliphatic glucosinolates. I. Side chain elongation in Brassica napus and Arabidopsis thaliana. Heredity 72:290-299.

MITHEN, R., 2001  Glucosinolates—biochemistry, genetics and biological activity. Plant Growth Reg. 34:91-103.

MITHEN, R. and H. CAMPOS, 1996  Genetic variation of aliphatic glucosinolates in Arabidopsis thaliana and prospects for map based cloning. Entomol. Exp. Appl. 80:202-205.

MITHEN, R. F., J. CLARKE, C. LISTER, and C. DEAN, 1995  Genetics of aliphatic glucosinolates. III. Side chain structure of aliphatic glucosinolates in Arabidopsis thaliana. Heredity 74:210-215.

MITHEN, R. F., M. DEKKER, R. VERKERK, R. RABOT, and I. T. JOHNSON, 2000  The nutritional significance, biosynthesis and bioavailability of glucosinolates in human foods. J. Sci. Food Agric. 80:967-984.

QUIROS, C. F., F. GRELLET, J. SADOWSKI, G. LI, and T. SUZUKI et al., 2001  Arabidopsis and Brassica comparative genomics: sequence, structure and gene content in the ABI1-Rps2-Ck1 chromosomal segment and related regions. Genetics 157:1321-1330.[Abstract/Free Full Text]

ROSA, E. A., R. K. HEANEY, G. R. FENWICH, and C. A. M. PORTAS, 1997  Glucosinolates in crop plants. Hort. Rev. 19:99-215.

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SMITH, L. B. and G. J. KING, 2000  The distribution of BoCAL-a alleles in Brassica oleracea is consistent with a genetic model for curd development and domestication of the cauliflower. Mol. Breed. 6:603-613.

UNDERHILL, E. W., 1980 Glucosinolates, pp. 493–511 in Encyclopedia of Plant Physiology (New Series), Vol. 8. Springer Verlag, Berlin.

WATHELET, J-P., R. IORI, N. MABON, S. CALMIERI, O. LEONI et al., 2001 Determination of the response factors of several desulfo-glucosinolates used for quantitative analysis of Brassicaceae. International Rapeseed Congress Technical Meeting, June 5–7, Poznan, Poland.

This article has been cited by other articles:

				Y.-X. Yuan, J. Wu, R.-F. Sun, X.-W. Zhang, D.-H. Xu, G. Bonnema, and X.-W. Wang A naturally occurring splicing site mutation in the Brassica rapa FLC1 gene is associated with variation in flowering time J. Exp. Bot., March 1, 2009; 60(4): 1299 - 1308. [Abstract] [Full Text] [PDF]

				A. J. Heidel, M. J. Clauss, J. Kroymann, O. Savolainen, and T. Mitchell-Olds Natural Variation in MAM Within and Between Populations of Arabidopsis lyrata Determines Glucosinolate Phenotype Genetics, July 1, 2006; 173(3): 1629 - 1636. [Abstract] [Full Text] [PDF]