TY - JOUR
AU - Hashimoto, Takashi
AB - Abstract Nornicotine is formed from nicotine by nicotine N-demethylase, a CYP82E family monooxygenase, and accumulates to high levels in some tobacco (Nicotiana tabacum) cultivars and many wild Nicotiana species. Nicotiana langsdorffii does not form nornicotine, whereas the closely related species N. alata accumulates this alkaloid abundantly. We show here that the two nicotine N-demethylase genes in N. langsdorffii have been inactivated by different molecular mechanisms. We identified four N. alata CYP82E genes that encode functional nicotine N-demethylases. In N. langsdorffii, however, one CYP82E gene encoding a functional enzyme was not expressed at all, whereas the other was weakly expressed but contained a one-nucleotide deletion in the first exon, yielding a truncated protein. Expression analysis of interspecific F1 hybrids between N. alata and N. langsdorffii indicated that cis-acting polymorphisms abolish expression of the otherwise functional CYP82E gene in N. langsdorffii. Segregation analysis of tobacco alkaloids and individual CYP82E alleles in F2 progeny revealed that duplicated CYP82E genes in both species are genetically linked, and provide genetic evidence that CYP82E genes are solely responsible for nornicotine formation in these wild Nicotiana species. Introduction Nicotine, or 1-methyl-2-(3-pyridyl)pyrrolidine, is the main alkaloid found in all species of the genus Nicotiana (Saitoh et al. 1985), and is the principal ingredient in the cigarette that causes relaxation, sharpness, calmness and alertness in smokers by binding to nicotinic cholinergic receptors and releasing neurotransmitters and hormones (Albuquerque et al. 2009). Many pyridine alkaloids that are structurally related to nicotine have been reported, often in low concentrations, in various Nicotiana plants and in the cured tobacco leaf (Leete 1983). Nornicotine, anabasine and anatabine accumulate to considerable concentrations in several wild Nicotiana species and cultivars of commercial tobacco (Nicotiana tabacum). Among them, nornicotine attracts special attention because it is converted to N′-nitrosonornicotine, a potent carcinogen, during the curing and processing of tobacco products. N′-Nitrosonornicotine produces esophageal and nasal cavity tumors in rats, and respiratory tract tumors in mice and hamsters (Hecht 1998). In Nicotiana plants, nicotine is synthesized from aspartate and putrescine by several enzymatic steps (Shoji and Hashimoto 2011), and then demethylated to nornicotine (Leete 1983). Molecular analysis of the enzyme responsible, nicotine N-demethylase, revealed that it is a Cyt P450 monooxygenase belonging to the CYP82E subfamily (Siminszky et al. 2005). In human liver microsomes, the P450 subfamilies CYP2A and CYP2B are involved in nicotine N-demethylation (Yamanaka et al. 2005). Nicotiana tabacum is an allotetraploid that was produced by hybridization of diploid Nicotiana species closely related to modern N. sylvestris and N. tomentosiformis (Murad et al. 2002), and contains at least five CYP82E-related genes: CYP82E4, CYP82E5 and CYP82E10 encode functional nicotine N-demethylases (Siminszky et al. 2005, Gavilano and Siminszky 2007, Lewis et al. 2010), whereas CYP82E2 and CYP82E3 encode inactive enzymes due to degenerate mutations which have occurred since the allotetraploidization event (Chakrabarti et al. 2007, Gavilano et al. 2007). Tobacco plants possessing knockout mutations in all three functional CYP82E genes contain very low amounts of nornicotine, but residual amounts of nornicotine are still found in these triple mutant plants (Lewis et al. 2010). The tobacco genome may contain one or more functional CYP82E genes that have escaped previous identification; alternatively, tobacco N-methylputrescine oxidase can accept putrescine to produce an N-demethylated pyrrolidine ring which may be incorporated into nornicotine, thus entirely by-passing nicotine (Heim et al. 2007, Katoh et al. 2007). Therefore, CYP82E-encoded nicotine N-demethylases are the major, although perhaps not the only, determinants of nornicotine accumulation in tobacco plants. In the Nicotiana genus, which contains >75 species (Knapp et al. 2004), N. alata and N. langsdorffii of the diploid Nicotiana section Alatae (n = 9) are two closely related wild tobacco species native to South America (Lim et al. 2006). Nicotiana langsdorffii accumulates substantial amounts of nicotine and anatabine in the leaf and the root, but does not contain nornicotine, whereas nicotine, anatabine and nornicotine accumulate exclusively in the root of N. alata (Saitoh et al. 1985, Sisson and Severson 1990, Sinclair et al. 2004). In a previous study, we compared the abilities of N. langsdorffi and N. alata to synthesize and translocate tobacco alkaloids by analyzing the expression of alkaloid biosynthesis genes, and by determining alkaloid profiles in root cultures, interspecifically grafted plants and hybrid plants (Pakdeechanuan et al. 2012). We found that N. alata does not accumulate tobacco alkaloids in the leaf due to a lack of long-range translocation of root alkaloids to the leaf, and that the non-translocation trait in N. alata is dominant over the translocation trait in N. langsdorffii. In this study, we focused on the nornicotine phenotype in these two Nicotiana species. We were particularly interested in the molecular mechanisms by which the nicotine N-demethylase genes in N. langsdorffii are inactivated, and we provide genetic evidence that nornicotine formation in wild Nicotina species is governed by the expression of functional CYP82E genes. Results Nornicotine formation Since N. alata accumulates tobacco alkaloids exclusively in the root organ and does not transport them to the aerial parts (Pakdeechanuan et al. 2012), we focused on the tobacco alkaloids present in the root. In the roots of 10-week-old N. alata plants, nicotine was the major alkaloid at a concentration of 4.07 mg g DW−1, followed by nornicotine (2.83 mg g DW−1) and anatabine (1.81 mg g DW−1) (Table 1). The conversion rate from nicotine to nornicotine (see the Materials and Methods) in the N. alata root was 41.0%. However, we only found nicotine (4.33 mg g DW−1) and anatabine (0.61 mg g DW−1) in the roots of 10-week-old N. langsdorffii plants; no nornicotine was detectable. The roots of F1 hybrid plants between N. alata and N. langsdorffii showed a nornicotine conversion rate of 16.6%, intermediate between the two parental values (Table 1; Fig. 1A), indicating that nornicotine formation is a semi-dominant trait. Fig. 1 View largeDownload slide Frequency distributions of the nicotine to nornicotine conversion rate (%) in the roots of 10-week-old plants. (A) N. alata, N. langsdorffii and their F1 hybrid. (B) F2 progeny. Fig. 1 View largeDownload slide Frequency distributions of the nicotine to nornicotine conversion rate (%) in the roots of 10-week-old plants. (A) N. alata, N. langsdorffii and their F1 hybrid. (B) F2 progeny. Table 1 Alkaloid content in the roots of N. alata, N. langsdorffii and their F1 hybrids Plants  Alkaloid content (mg g DW–1)   Conversion (%)    Nicotine  Nornicotine  Anatabine    N. alata  4.07 ± 3.26  2.83 ± 2.22  1.81 ± 0.64  41.0  N. langsdorffii  4.33 ± 0.74  Not detected  0.61 ± 0.61  0  F1  2.27 ± 0.51  0.45 ± 0.21  0.28 ± 0.19  16.6  Plants  Alkaloid content (mg g DW–1)   Conversion (%)    Nicotine  Nornicotine  Anatabine    N. alata  4.07 ± 3.26  2.83 ± 2.22  1.81 ± 0.64  41.0  N. langsdorffii  4.33 ± 0.74  Not detected  0.61 ± 0.61  0  F1  2.27 ± 0.51  0.45 ± 0.21  0.28 ± 0.19  16.6  View Large To examine the genetic locus or loci involved in nornicotine formation, we first analyzed nornicotine conversion rates in the roots of 96 individual F2 plants derived from self-pollination of the F1 hybrid plants (Fig. 1B). Among these plants, 22 (22.9%) did not contain nornicotine and were classified as non-converters, whereas the remainder showed values ranging from <10% conversion to levels that were comparable with or exceeded those in the N. alata roots. The observed results are consistent with the hypothesis (P < 0.05) that a single semi-dominant locus controls nornicotine formation. CYP82E genes in N. alata and N. langsdorffii Because nicotine N-demethylase is encoded by Cyt P450 monooxygenase genes of the CYP82E subfamily in N. tabacum (Siminszky et al. 2005, Lewis et al. 2010), we amplified genomic DNA fragments from N. alata and N. langsdorffii using PCR primers designed from the tobacco CYP82E sequences, and obtained four CYP82E genes from N. alata (NalaCYP82E1, NalaCYP82E2, NalaCYP82E3 and NalaCYP82E4) and two from N. langsdorffii (NlanCYP82E1 and NlanCYP82E2). Reverse transcription–PCR (RT–PCR) was next used to clone full-length cDNA clones for five of these CYP82E genes, the exception being that no cDNA clones corresponding to NlanCYP82E1 were obtained even after cDNA pools were prepared from several different tissues of N. langsdorffii. The six DNA sequences have been deposited in GenBank with the following accession numbers: NalaCYP82E1 (AB709932), NalaCYP82E2 (AB709933), NalaCYP82E3 (AB709934), NalaCYP82E4 (AB709935), NlanCYP82E1 (AB709936) and NlanCYP82E2 (AB709937). Based on alignments of the genomic DNA and cDNA sequences, the exon–intron structure of the CYP82E genes was determined (Fig. 2). The open reading frame of NlanCYP82E1, for which no cDNA was available, was deduced by comparison with other CYP82E sequences. The CYP82E genes of N. alata and N. langsdorffii were composed of two highly conserved exons, separated by a less conserved intron of varying length (Fig. 2), and encoded proteins having amino acid sequences highly similar to CYP82Es of N. tabacum, N. sylvestris and N. tomentosiformis (Fig. 3; Siminszky et al. 2005, Chakrabarti et al. 2007, Gavilano and Siminszky 2007, Gavilano et al. 2007, Lewis et al. 2010). The position of the intron is identical in CYP82E4 of N. tabacum (Xu et al. 2007) and CYP82E2 of N. sylvestris (Chakrabarti et al. 2007). Fig. 2 View largeDownload slide Organization of CYP82E genes. The 5′- and 3′-untranslated regions are shown as open boxes, whereas the filled boxes indicate the protein-coding regions, with numbers indicating their length in base pairs. Introns are shown as lines, with their length (bp) indicated below. (A) Four CYP82E genes in N. alata. (B) Two CYP82E genes in N. langsdorffii. The gene organization of NlanCYP82E1 was deduced from the structures of other CYP82E genes. A single nucleotide deletion (asterisk) in the first exon of NlanCYP82E2 causes a frameshift and creates a premature stop codon. The gray regions of the NlanCYP82E2 exons indicate the predicted protein-coding regions if a full-length CYP82E protein is restored by an insertion of the deleted nucleotide. Fig. 2 View largeDownload slide Organization of CYP82E genes. The 5′- and 3′-untranslated regions are shown as open boxes, whereas the filled boxes indicate the protein-coding regions, with numbers indicating their length in base pairs. Introns are shown as lines, with their length (bp) indicated below. (A) Four CYP82E genes in N. alata. (B) Two CYP82E genes in N. langsdorffii. The gene organization of NlanCYP82E1 was deduced from the structures of other CYP82E genes. A single nucleotide deletion (asterisk) in the first exon of NlanCYP82E2 causes a frameshift and creates a premature stop codon. The gray regions of the NlanCYP82E2 exons indicate the predicted protein-coding regions if a full-length CYP82E protein is restored by an insertion of the deleted nucleotide. Fig. 3 View largeDownload slide Phylogenetic relationship between CYP82Es. The CYP82E sequences reported in this study are boxed; the restored NlanCYP82E2 sequence is used. The tree was constructed by the UPGMA method based on a distance matrix calculated by the ClustalW program. The number at each node indicates the bootstrap value based on 100 samplings. The scale bar indicates the distance corresponding to one difference per 100 positions. Nala, N. alata; Nlan, N. langsdorffii; Nsyl, N. sylvestris; Ntab, N. tabacum; Ntom, N. tomentosiformis. Fig. 3 View largeDownload slide Phylogenetic relationship between CYP82Es. The CYP82E sequences reported in this study are boxed; the restored NlanCYP82E2 sequence is used. The tree was constructed by the UPGMA method based on a distance matrix calculated by the ClustalW program. The number at each node indicates the bootstrap value based on 100 samplings. The scale bar indicates the distance corresponding to one difference per 100 positions. Nala, N. alata; Nlan, N. langsdorffii; Nsyl, N. sylvestris; Ntab, N. tabacum; Ntom, N. tomentosiformis. Interestingly, we found that a nucleotide at position 345 is deleted in the first exon of NlanCYP82E2, which causes a frameshift in the downstream coding sequence and is predicted to result in a truncated CYP82E protein (Fig. 2; Supplementary Fig. S1). If a cytosine is inserted into the deleted nucleotide position, the restored NlanCYP82E2 protein sequence shows very high similarity to other CYP82Es of N. alata and N. langsdorffii. N. alata CYP82Es and N. langsdorffii CYP82E1 are functional nicotine N-demethylases Since a small number of degenerate mutations in CYP82E sequences are known to result in inactivation of nicotine N-demethylase (Chakrabarti et al. 2007, Lewis et al. 2010), we examined whether the CYP82E genes found in this study encode functional enzymes. For NlanCYP82E1, for which no cDNA was available, we constructed a full-length cDNA by removing the intron from the genomic DNA fragment. Four full-length NalaCYP82E cDNAs and the constructed NlanCYP82E1 cDNA were individually expressed in yeast strain WAT11, which co-expressed an Arabidopsis P450 reductase gene (Pompon et al. 1996). Microsomal fractions isolated from yeast cells that expressed any one of the five CYP82E clones catalyzed the conversion of nicotine to nornicotine (Fig. 4A). Further catalytic analysis showed that these nicotine N-demethylases possess a relatively high affinity for nicotine, with Km values for nicotine ranging from 20.3 to 118.6 µM (Fig. 4B). These Km values are comparable with those reported for other functional Nicotiana CYP82E enzymes expressed in yeast (Gavilano et al. 2007, Gavilano and Siminszky 2007, Xu et al. 2007, Lewis et al. 2010). We thus conclude that the four NalaCYP82E genes and NlanCYP82E1 encode functional enzymes. Fig. 4 View largeDownload slide CYP82Es from N. alata and N. langsdorffii are functional nicotine N-demethylases. (A) Gas–liquid chromatograms of the reaction products are shown. Microsomes were prepared from yeast cells expressing empty vector (control microsomes) and NalaCYP82E1. Dodecane was included in the samples as an internal standard (IS). (B) Double reciprocal plots of enzyme activity (nornicotine formed) vs. nicotine concentration. The Km values for nicotine are indicated as the means ± SD for NalaCYPE1, NalaCYPE2, NalaCYPE3, NalaCYPE4 and NlanCYPE1. Fig. 4 View largeDownload slide CYP82Es from N. alata and N. langsdorffii are functional nicotine N-demethylases. (A) Gas–liquid chromatograms of the reaction products are shown. Microsomes were prepared from yeast cells expressing empty vector (control microsomes) and NalaCYP82E1. Dodecane was included in the samples as an internal standard (IS). (B) Double reciprocal plots of enzyme activity (nornicotine formed) vs. nicotine concentration. The Km values for nicotine are indicated as the means ± SD for NalaCYPE1, NalaCYPE2, NalaCYPE3, NalaCYPE4 and NlanCYPE1. NlanCYP82E1 is not expressed due to defects in cis-activation Because NlanCYP82E1 encodes a functional nicotine N-demethylase, nornicotine formation in N. langsdorffii is likely to depend on whether or not this gene is expressed. Our previous RNA gel blot analysis of CYP82E transcripts using an N. tabacum CYP82E4 probe indicated that CYP82E genes are expressed strongly in the root of N. alata but negligibly in the root of N. langsdorffii (Pakdeechanuan et al. 2012). In this study, we amplified CYP82E cDNAs from the root of N. langsdorffii by using a common PCR primer set which would anneal perfectly to all six CYP82E genes identified in this study, and then sequenced 100 individual cDNA clones to reveal their identity. All sequenced clones were found to be NlanCYP82E2; no NlanCYP82E1 clone was present (Fig. 5A), which is consistent with our inability to obtain any NlanCYP82E1 cDNAs by RT–PCR based on its genomic DNA sequence information. Similar expression analysis of N. alata indicated that all four NalaCYP82E genes are expressed in the root of N. alata (Fig. 5B). Fig. 5 View largeDownload slide Expression levels of CYP82E genes in the roots of 10-week-old plants. CYP82E cDNA fragments were amplified from RNA samples prepared from the roots of four independent plants. One hundred CYP82E cDNA clones were sequenced each from N. langsdorffii (A), N. alata (B) and their F1 hybrids (C). The numbers of identified cDNAs are shown in parentheses. ND, not detected. Fig. 5 View largeDownload slide Expression levels of CYP82E genes in the roots of 10-week-old plants. CYP82E cDNA fragments were amplified from RNA samples prepared from the roots of four independent plants. One hundred CYP82E cDNA clones were sequenced each from N. langsdorffii (A), N. alata (B) and their F1 hybrids (C). The numbers of identified cDNAs are shown in parentheses. ND, not detected. F1 hybrids provide a useful tool to distinguish between differences in cis- or trans-activation of species-specific alleles of a gene (Wittkopp et al. 2004). Differences in trans-regulatory activity can be inferred by comparing the ratio of allelic expression in hybrids with the ratio of gene expression between species, an approach which has been used, for example, to analyze the difference in expression of a transcription factor gene for floral scent production in two Petunia species (Klahre et al. 2011). Examination of the CYP82E cDNAs prepared from the roots of four F1 hybrid plants between N. alata and N. langsdorffii revealed that NalaCYP82E clones predominated, whereas NlanCYP82E2 clones represented only 3% and NlanCYP82E1 was absent (Fig. 5C). The nonsense-mediated mRNA decay pathway, which targets mRNAs having premature termination codons for rapid degradation (Rebbapragada and Lykke-Andersen 2009), may contribute to the low abundance of NlanCYP82E2 mRNA. In our experiments, the ratios of NalaCYP82E1, NalaCYP82E2 and NalaCYP82E3 clones were similar between N. alata and the F1 hybrids, while the ratio of NalaCYP82E4 was considerably higher in the former than in the latter. The likely cause of this difference is discussed below. Aborted expression of NlanCYP82E1 in both N. langsdorffi and the F1 hybrids is consistent with a model whereby defects in cis-activation, rather than in trans-activation, cause silencing of this gene. Segregation of CYP82E genes in F2 progeny To determine the presence or absence of the six CYP82E genes in individual F2 plants derived from self-pollination of >10 F1 plants, we developed a capillary electrophoresis-based analytical system that detected fragment length polymorphism of the CYP82E introns (Fig. 6A; see the Materials and Methods). Genomic DNA samples from N. langsdorffii yielded PCR fragments of 690 and 726 bp, which corresponded to NlanCYP82E1 and NlanCYP82E2, respectively, whereas N. alata DNA samples gave fragments of 369 bp (NalaCYP82E3), 603 bp (NalaCYP82E4), 637 bp (NalaCYP82E1) and 642 bp (NalaCYP82E2). In the chromatogram of an F1 hybrid sample, all six CYP82E fragments were clearly detectable and well separated. Fig. 6 View largeDownload slide Segregation analysis of CYP82E genes in 109 F2 progeny plants. (A) Capillary electrophoresis-based fragment length polymorphism analysis of the CYP82E introns in N. alata, N. langsdorffii and their F1 hybrids. DNA size is shown in base pairs on the x-axis, while the y-axis indicates relative fluorescence intensity. Amplified DNA fragments showed the expected sizes: NalaCYP82E1, 637 bp; NalaCYP82E2, 642 bp; NalaCYP82E3, 369 bp; NalaCYP82E4, 603 bp; NlanCYP82E1, 690 bp; and NlanCYP82E2, 726 bp. (B) Genetic linkage maps of CYP82E genes in N. alata and N. langsdorffii. Fig. 6 View largeDownload slide Segregation analysis of CYP82E genes in 109 F2 progeny plants. (A) Capillary electrophoresis-based fragment length polymorphism analysis of the CYP82E introns in N. alata, N. langsdorffii and their F1 hybrids. DNA size is shown in base pairs on the x-axis, while the y-axis indicates relative fluorescence intensity. Amplified DNA fragments showed the expected sizes: NalaCYP82E1, 637 bp; NalaCYP82E2, 642 bp; NalaCYP82E3, 369 bp; NalaCYP82E4, 603 bp; NlanCYP82E1, 690 bp; and NlanCYP82E2, 726 bp. (B) Genetic linkage maps of CYP82E genes in N. alata and N. langsdorffii. With this method, we next analyzed segregation of CYP82E genes in 109 F2 plants (Supplementary Table S1). When fragment peaks were ambiguous, we confirmed the genotype of these samples by standard genomic PCR amplification of the CYP82E genes and by the derived cleaved amplified polymorphic sequences (dCAPS) assay (see the Materials and Methods). NalaCYP82E1, NalaCYP82E2, NalaCYP82E3, NlanCYP82E1 and NlanCYP82E2 were present in approximately 75% of the F2 plants, as expected for a simple Mendelian segregation (P < 0.05), whereas NalaCYP82E4 was inherited in only 27.5% of the F2 population (Table 2). To explain this unexpectedly low segregation frequency, we explored the possibility that NalaCYP82E4 is present in some, but not all, of the original N. alata plant population used in our experiments, and thus that some F1 plants did not inherit this gene. Indeed, when 12 N. alata individuals were tested for the presence of NalaCYP82E4 by genomic PCR analysis, we found this gene in eight plants; the remaining four plants did not have it (data not shown). Because DNA samples of the F1 plants were not available at this stage, we could not directly test whether some of them lacked NalaCYP82E4. In conclusion, the non-universal occurrence of NalaCYP82E4 in the N. alata plants used in this study appears to account for the abnormally low frequency of this gene in the F2 segregation analysis. Table 2 Allele frequencies of CYP82E genes among F2 plants Phenotypea  n  Number of plantsb       NalaE1  NalaE2  NalaE3  NalaE4  NlanE1  NlanE2  Non-converter  26  0  0  0  0  26  26  Medium converter  73  71  70  69  24  60  54  High converter  10  9  10  9  6  0  0  Total  109 (100%)  80 (73.4%)  80 (73.4%)  78 (71.6%)  30 (27.5%)  86 (78.9%)  80 (73.4%)  Phenotypea  n  Number of plantsb       NalaE1  NalaE2  NalaE3  NalaE4  NlanE1  NlanE2  Non-converter  26  0  0  0  0  26  26  Medium converter  73  71  70  69  24  60  54  High converter  10  9  10  9  6  0  0  Total  109 (100%)  80 (73.4%)  80 (73.4%)  78 (71.6%)  30 (27.5%)  86 (78.9%)  80 (73.4%)  a Non-converters did not contain detectable amounts of nornicotine, whereas medium converters and high converters showed nicotine to nornicotine conversion rates of less than and more than 50%, respectively. b The number of plants which contained the indicated CYP82E alleles. CYP82E genes are shown in abbreviated forms; NalaE1, for example, indicates NalaCYP82E1. View Large We next investigated co-segregation of the CYP82E genes by linkage analysis. This indicated that NalaCYP82E1, NalaCYP82E2 and NalaCYP82E3 are aligned in a single chromosomal region with a genetic distance of 5.5 cM between NalaCYP82E1 and NalaCYP82E2, and of 1.8 cM between NalaCYP82E2 and NalaCYP82E3 (Fig. 6B). In an N. langsdorffii chromosomal region, NlanCYP82E1 and NlanCYP82E2 are linked with a distance of 5.5 cM. Finally, we examined the relationship between phenotype (nornicotine accumulation) and genotype (CYP82E) (Supplementary Table S1). F2 plants were classified into non-converters (no nornicotine detected), medium converters (<50% conversion) and high converters (conversion rates exceeding 50%) (Fig. 1), and allele frequencies of CYP82E genes were scored for each group (Table 2). Non-converters (n = 26) lacked all four of the N. alata CYP82E genes, but instead had both CYP82E genes of N. langsdorffii. In medium converters (n = 73), two or more N. alata CYP82E genes were always present, and NlanCYP82E1 and NlanCYP82E2 were also found in most plants. High converters (n = 10) also contained two or more N. alata CYP82E genes but lacked both NlanCYP82E1 and NlanCYP82E2. While we need to interpret the data for NalaCYP82E4 with caution as explained above, these genetic data indicate that the clustered CYP82E genes of N. alata and N. langsdorffii are located at the identical locus of the corresponding chromosome in these two diploid Nicotiana species. Whereas the N. alata CYP82E genes act semi-dominantly and individually to enhance conversion of nicotine to nornicotine, the N. langsdorffii CYP82E genes are both non-functional. Discussion In this study, we provided genetic, molecular and biochemical evidence that CYP82E genes encoding functional nicotine N-demethylases are responsible for the accumulation of nornicotine in N. alata. In N. langsdorffii, where two CYP82E genes are inactive, and in F2 progeny plants which did not inherit functional CYP82E genes from N. alata, nornicotine was below the detection limit of our sensitive capillary gas–liquid chromatography assay. An N-methylputrescine oxidase-mediated and nicotine-independent pathway of nornicotine formation (Heim et al. 2007, Katoh et al. 2007) has been postulated to explain residual amounts of nornicotine in triple tobacco mutants for CYP82E genes (Lewis et al. 2010), but such a by-pass pathway, if it exists in planta, does not contribute to nornicotine formation in N. alata. Moreover, the conversion of nicotine to nornicotine in plants appeared to depend on the dosage of functional CYP82E genes: F1 hybrid plants showed approximately half the conversion rate of N. alata, and the low converters in the F2 progeny possessed both N. alata and N. langsdorffii CYP82E alleles, indicative of the heterozygosity of the functional N. alata alleles. Thus, nicotine N-demethylase activity is rate limiting for nornicotine formation in N. alata and its hybrid progeny. Allotetraploid N. tabacum generally does not accumulate nornicotine, whereas its evolutionary progenitors N. sylvestris and N. tomentosiformis convert nicotine to nornicotine in the senescing leaf (Wernsman and Matzinger 1968). In tobacco, three CYP82E genes are inactive. The nicotine N-demethylase activity of the CYP82E2 gene, derived from N. sylvestris, is inactivated by the E375K and W420L mutations (Chakrabarti et al. 2007), whereas the N. tomestosiformis-derived CYP82E3 gene encodes a non-functional enzyme due to the W330C mutation (Gavilano et al. 2007). The CYP82E4 gene, from N. tomentosiformis, is transcriptionally silenced, but this silencing is unstable and the gene is often reactivated to produce converter tobacco lines. Allopolyploidization is associated with transcriptional silencing of a considerable proportion of the transcriptome, which may be caused by epigenetic changes, transposon activation, and small RNA-mediated and RNA interference-mediated interactions (Adams and Wendel 2005). The transcriptional inactivation mechanism of CYP82E4 in tobacco may be distinct from that observed in this study, since the inactivation of NlanCYP82E1 is stable and is not triggered by polyploidization. Loss of critical transcription factor-binding sites in the NlanCYP82E1 promoter may underlie the cis-inactivation of this gene. In N. tomentosiformis, the conversion locus contains at least two nicotine N-demethylase genes, NtomCYP82E3 and NtomCYP82E4 (Chakrabrti et al. 2007, Gavilano et al. 2007). In this study, we showed that NalaCYP82E1, NalaCYP82E2 and NalaCYP82E3 are clustered within a 7.3 cM chromosomal interval of N. alata, while NlanCYP82E1 and NlanCYP82E2 are linked at a genetic distance of 5.5 cM. Therefore, duplication of CYP82E genes at the conversion locus appears to be a general feature in diploid Nicotiana species. Duplicated genes may undergo rapid diversification, leading to differences in expression profiles or protein function (Lynch and Conery 2000). NtomCYP82E3 is expressed strongly in the green tobacco leaf and is down-regulated during senescence, whereas expression of NtomCYP82E4 is low in the green leaf and is strongly induced during senescence (Gavilano et al. 2007). The four N. alata CYP82E genes may be expressed in distinct patterns during growth and development, and in response to environmental stimuli, but we were unable to analyze the expression patterns of each CYP82E gene because of difficulties in designing gene-specific PCR primers for the quantitative RT–PCR assay. Nicotaina alata is strongly self-incompatible, and usually propagates by out-crossing. In such plant species, a population may exhibit high polymorphic variation at many genetic loci. We found that NalaCYP82E4 is present in only eight out of 12 N. alata plants tested in our population, suggesting that the NalaCYP82E4 DNA sequence represents a polymorphic allele of NalaCYP82E4. The PCR primers we used to amplify NalaCYP82E4 may have failed to detect other NalaCYP82E4 alleles in the plant population used in this study. Results involving NalaCYP82E4 (Fig. 5, Table 2) are complicated by this likely polymorphism. We conclude that CYP82E genes encoding nicotine N-demethylase are duplicated at closely linked chromosomal regions in the two diploid Nicotiana species of the Alatae section, and that the two CYP82E genes in N. langsdorffii are rendered non-functional by either transcriptional inactivation or premature translational termination, resulting in the absence of nornicotine in this species. While many tobacco herbivores show higher susceptibility toward nicotine than nornicotine, some are more susceptible to nornicotine (Siegler and Bowen 1946, Soloway 1976). Contrasting ratios of these tobacco alkaloids may influence the interactions between herbivores and N. alata and N. langsdorffii. Materials and Methods Plant materials Seeds of N. alata Link & Otto and N. langsdorffii Weinmann were obtained from the Leaf Tobacco Research Center, Japan Tobacco. Since N. alata plants showed strong self-incompatibility, they were maintained by outcrossing. Plants were grown as described previously (Pakdeechanuan et al. 2012). Alkaloid analysis Roots were harvested, processed, and analyzed for tobacco alkaloids by gas–liquid chromatography, as described previously (Pakdeechanuan et al. 2012). The conversion rate from nicotine to nornicotine was calculated by dividing the nornicotine content by the sum of the nicotine and nornicotine content. When necessary, we grouped plants into three categories: non-converters (no nornicotine detected), medium converters (detectable nornicotine but <50% conversion) and high converters (exceeding 50% conversion). Cloning of CYP82E cDNAs Total RNA was extracted from leaves and roots using the RNeasy Mini kit according to the manufacturer’s instructions (QIAGEN), and was used to synthesize first-strand cDNA with the SuperScript II Reverse Transcriptase kit (Invitrogen). cDNA fragments of CYP82E were PCR-amplified using primers which were designed based on the conserved nucleotide sequences of the published CYP82E subfamily. Full-length CYP82E cDNA clones were obtained by amplification of the 5′ and 3′ ends of the cDNA, according to the protocol supplied with the SMART RACE cDNA Amplification kit (Clontech). The conditions for the nested PCRs were 94°C for 3 min followed by 30 cycles of 94°C for 30 s, 65°C for 30 s, 72°C for 3 min, and a final extension at 72°C for 7 min. PCR products were purified, cloned into pGEM-T Easy (Promega), and sequenced using an ABI PRISM 3100 automated DNA sequencer. All the PCR primers used for cloning CYP82E cDNAs are listed in Supplementary Table S2. Cloning of CYP82E genomic fragments Genomic DNA was extracted from leaves using the PureLink Plant Total DNA Purification kit (Invitrogen), following the manufacturer’s instructions. Genomic fragments containing introns and part of the exons of the CYP82E genes were obtained by amplifying genomic DNA with PCR primer sets which were specific to each CYP82E gene. The 5′ and 3′ regions of NlanCYP82E1 were obtained by inverse PCR. Genomic DNA of N. langsdorffii was digested with XbaI and self-ligated with T4 DNA ligase (TAKARA) to form circular DNA fragments. The circularized DNA was amplified using two primer sets specific to NlanCYP82E1. The PCR products were cloned, and sequenced to verify their identity as NlanCYP82E1. All the PCR primers used for cloning CYP82E genomic fragments are listed in Supplementary Table S2. According to the nomenclature of Dr. David Nelson (http://drnelson.uthsc.edu/CytochromeP450.html), NaCYP82E1 (AB709932), NaCYP82E2 (AB709933), NaCYP82E3 (AB709934), NaCYP82E4 (AB709935), NlCYP82E1 (AB709936) and NlCYP82E2 (AB709937) are designated as CYP82E15, CYP82E16, CYP82E17, CYP82E18, CYP82E19 and CYP82E20P, respectively. Genotyping by capillary electrophoresis PCR primers were designed to obtain distinct intron length polymorphisms for each CYP82E gene (Supplementary Table S3). The 5′ end of the forward primers was fluorescently labeled with 6-FAM (Invitrogen) and VIC (Applied Biosystems). PCR conditions were 5 min at 94°C and then eight cycles of touchdown PCR (30 s at 94°C; 30 s at 64°C, with a 1°C drop after each cycle; 1 min at 72°C), followed by 32 cycles of regular PCR (30 s at 94°C; 30 s at 57°C; 1 min at 72°C), with a final extension for 7 min at 72°C. For NalaCYP82E4, the annealing temperature of touchdown PCR was set at 66°C. The M13-tail PCR method (Schuelke 2000) was used to amplify the intron of NalaCYP82E3. The unlabeled forward primer contained a universal M13 sequence at the 5′ end, while the universal M13 primer was labeled with VIC. The annealing temperature of touchdown PCR was set at 60°C. Fluorescently labeled PCR products (1 µl) were mixed with 8.5 µl of Hi-Di formamide (Applied Biosystems) and 0.5 µl of size markers (Applied Biosystems) comprising 68 single-stranded labeled fragments ranging from 20 to 1,200 bp. Samples were denatured by heating for 5 min at 95°C, cooled on ice for at least 3 min and then applied to an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems) equipped with 50 cm capillaries containing POP-6 polymer. The results were analyzed by GeneMapper Software version 4.0 (Applied Biosystems). Allelic data scores for each individual were entered manually into a spreadsheet (Supplementary Table S1), which was then used to calculate recombination frequencies between CYP82E genes. Derived cleaved amplified polymorphic sequences The dCAPS procedure was based on a published protocol (Neff et al. 1998). PCR primers were designed to detect single-nucleotide polymorphisms in the CYP82E genes, using the dCAPS Finder 2.0 program (http://helix.wustl.edu/dcaps/dcaps.html), and are listed in Supplementary Table S4. PCR-amplified products were digested with appropriate restriction enzymes, and then analyzed by electrophoresis on a 4% (w/v) NuSieve GTG Agarose gel (FMC BioProducts). Heterologous expression of CYP82E in yeast The Saccharomyces cerevisiae strain WAT11 (MAT a; ade 2-1; his 3-11, −15; leu 2-3, −112; canR; cyr+), which expresses Arabidopsis thaliana ATR1 NADPH-P450 reductase, and the yeast replicative plasmid pYeDP60 (Pompon et al. 1996) were kindly provided by Dr. P. Urban (Centre de Génétique Moléculaire, CNRS, Gif-sur-Yvette, France). Yeast cells were transformed with the vector constructs by electroporation as described (Thompson et al. 1998), cultured, and then induced for expression of CYP82E as described (Pompon et al. 1996). The protein-coding sequences of NalaCYP82E genes were amplified using the appropriate primer sets based on the corresponding cDNA clone sequences. The protein-coding sequence of NlanCYP82E1 was obtained by joining the coding sequence from the two exons of a genomic fragment by PCR. Each forward PCR primer contained a BamHI restriction site, and the yeast Kozak sequence (Hamilton et al., 1987), whereas each reverse primer contained an EcoRI restriction site after the stop codon. The amplified fragments were double-digested with BamHI and EcoRI, and then inserted into pYeDP60. Primer sequences are listed in Supplementary Table S2. After induction of CYP82E, yeast cells were harvested by centrifugation and microsomes were isolated using a published enzymatic breaking procedure (Pompon et al. 1996), with minor modifications. Yeast cells were treated with zymolyase (Zymolyase 20T; Seigakaku) with shaking at 90 r.p.m. for 1.5 h. The resulting spheroplasts were lysed by sonication, and the lysate was centrifuged at 20,000 × g for 20 min at 4°C. The supernatant was then centrifuged at 100,000 × g for 1 h at 4°C, and the microsome pellet was resuspended in buffer (20% glycerol, 1 mM EDTA, 1 mM dithiothreitol and 50 mM Tris–HCl, pH 7.5). Protein concentration was determined by the Coomassie Protein Assay Kit (Thermo Scientific). In vitro enzyme assays for nicotine N-demethylase The nicotine N-demethylase activity was measured in a reaction mixture of 100 µl containing 50 mM Tris–HCl (pH 7.5), 1 mM NADPH, 50 µg of microsomes, and (S)-nicotine at various concentrations. The assays were initiated by the addition of NADPH. Enzyme activity was determined during 15 min at 25°C by measuring the formation of nornicotine. After adding 0.9 ml of 0.1 N H2SO4 and 0.1 ml of 25% (w/v) NH4OH, the mixture was loaded onto an Extrelut-1 column (Merck) and eluted with 6 ml of chloroform. The chloroform elutant was evaporated to dryness at 37°C, and the dry residues were dissolved in ethanol containing 0.1% (v/v) dodecane as an internal standard. The measurement of alkaloids was carried out by gas–liquid chromatography as described (Pakdeechanuan et al., 2012). Kinetic constants were calculated by linear regression of the double-reciprocal analysis, using Prism 5 software (GraphPad Software). Supplementary data Supplementary data are available at PCP online. Funding This study was supported by the Japan Society for the Promotion of Science [Grant-in-aid for scientific research (C), No. 23570055 to T.S.], the Ministry of Education, Sports, Science and Technology, Japan [Global COE program of Nara Institute of Science and Technology (Frontier Biosciences: Strategies for survival and adaptation in a changing global environment)]; Royal Thai Government [scholarship to P.P.]. Acknowledgments We thank the Leaf Tobacco Research Center, Japan Tobacco for the tobacco seeds used in this study, and Philippe Urban for the WAT11 yeast strain and the pYeDP60 plasmid. We are grateful to Xintian Lao and Lai Kok Song for useful advice on the capillary electrophoresis-based genotyping, Noriyoshi Yagi for technical support in the molecular biological experiments, David Nelson for nomenclature of the CYP82E genes, and Ian Smith for English editing of the text. Abbreviations Abbreviations CYP82 Cyt P450 family 82 dCAPS derived cleaved amplified polymorphic sequences RT–PCR reverse transcription–PCR. References Adams KL,  Wendel JF.  Novel patterns of gene expression in polyploid plants,  Trends Genet. ,  2005, vol.  21 (pg.  539- 543) Google Scholar CrossRef Search ADS PubMed  Albuquerque EX,  Pereira EFR,  Alkondon M,  Rogers SW.  Mammalian nicotinic acetylcholine receptors: from structure to function,  Physiol. Rev. ,  2009, vol.  89 (pg.  73- 120) Google Scholar CrossRef Search ADS PubMed  Chakrabarti M,  Meekins KM,  Gavilano LB,  Siminszky B.  Inactivation of the cytochrome P450 gene CYP82E2 by degenerative mutations was a key event in the evolution of the alkaloid profile of modern tobacco,  New Phytol. ,  2007, vol.  175 (pg.  565- 574) Google Scholar CrossRef Search ADS PubMed  Gavilano LB,  Coleman NP,  Bowen SW,  Siminszky B.  Functional analysis of nicotine demethylase genes reveals insights into the evolution of modern tobacco,  J. Biol. Chem. ,  2007, vol.  282 (pg.  249- 256) Google Scholar CrossRef Search ADS PubMed  Gavilano LB,  Siminszky B.  Isolation and characterization of the cytochrome P450 gene CYP82E5v2 that mediates nicotine to nornicotine conversion in the green leaves of tobacco,  Plant Cell Physiol. ,  2007, vol.  48 (pg.  1567- 1574) Google Scholar CrossRef Search ADS PubMed  Hamilton R,  Watanabe CK,  de Boer HA.  Compilation and comparison of the sequence context around the AUG start codons in Saccharomyces cerevisiae mRNAs,  Nucleic Acids Res. ,  1987, vol.  15 (pg.  3581- 3593) Google Scholar CrossRef Search ADS PubMed  Hecht SS.  Biochemistry, biology, and carcinogenicity of tobacco-specific N-nitrosoamines,  Chem. Res. Toxicol. ,  1998, vol.  11 (pg.  559- 603) Google Scholar CrossRef Search ADS PubMed  Heim WG,  Sykes KA,  Hildreth SB,  Sun J,  Lu RH,  Jelesko JG.  Cloning and characterization of a Nicotiana tabacum methylputrescine oxidase transcript,  Phytochemistry ,  2007, vol.  68 (pg.  454- 463) Google Scholar CrossRef Search ADS PubMed  Katoh A,  Shoji T,  Hashimoto T.  Molecular cloning of N-methylputrescine oxidase from tobacco,  Plant Cell Physiol. ,  2007, vol.  48 (pg.  550- 554) Google Scholar CrossRef Search ADS PubMed  Klahre U,  Gurba A,  Hermann K,  Saxehofer M,  Bossolini E,  Guerin PM, et al.  Pollinator choice in Petunia depends on two major genetic loci for floral scent production,  Curr. Biol. ,  2011, vol.  21 (pg.  730- 739) Google Scholar CrossRef Search ADS PubMed  Knapp S,  Chase MW,  Clarkson JJ.  Nomenclatural changes and a new sectional classification in Nicotiana (Solanaceae),  Taxon ,  2004, vol.  53 (pg.  73- 82) Google Scholar CrossRef Search ADS   Leete E.  Pelletier SW.  Biosynthesis and metabolism of the tobacco alkaloids,  Alkaloids: Chemical and Biological Perspectives ,  1983, vol.  Vol. 1  New York John Wiley & Sons(pg.  85- 152) Lewis RS,  Bowen SW,  Keogh MR,  Dewey RE.  Three nicotine demethylase genes mediate nornicotine biosynthesis in Nicotiana tabacum L.: functional characterization of the CYP82E10 gene,  Phytochemistry ,  2010, vol.  71 (pg.  1988- 1998) Google Scholar CrossRef Search ADS PubMed  Lim KY,  Kovarik A,  Matyasek R,  Chase MW,  Knapp S,  McCarthy E, et al.  Comparative genomics and repetitive sequence divergence in the species of diploid Nicotiana section Alatae,  Plant J. ,  2006, vol.  48 (pg.  907- 919) Google Scholar CrossRef Search ADS PubMed  Lynch M,  Conery JS.  The evolutionary fate and consequences of duplicated genes,  Science ,  2000, vol.  290 (pg.  1151- 1155) Google Scholar CrossRef Search ADS PubMed  Murad L,  Lim KY,  Christopodulou V,  Matyasek R,  Lichtenstein C,  Kovarik A, et al.  The origin of tobacco’s T genome is traced to a particular linage within Nicotiana tomentosiformis (Solanaceae),  Amer. J. Bot. ,  2002, vol.  89 (pg.  921- 928) Google Scholar CrossRef Search ADS   Neff MM,  Neff JD,  Chory J,  Pepper AE.  dCAPS, a simple technique for the genetic analysis of single nucleotide polymorphisms: experimental applications in Arabidopsis thaliana genetics,  Plant J. ,  1998, vol.  14 (pg.  387- 392) Google Scholar CrossRef Search ADS PubMed  Pakdeechanuan P,  Shoji T,  Hashimoto T.  Root-to-shoot translocation of alkaloids is dominantly suppressed in Nicotiana alata,  Plant Cell Physiol. ,  2012, vol.  53 (pg.  1247- 1254) Google Scholar CrossRef Search ADS PubMed  Pompon D,  Louerat B,  Bronine A,  Urban P.  Yeast expression of animal and plant P450s in optimized redox environments,  Methods Enzymol. ,  1996, vol.  272 (pg.  51- 64) Google Scholar PubMed  Rebbapragada I,  Lykke-Andersen J.  Execution of nonsense-mediated mRNA decay: what defines a substrate?,  Curr. Opin. Cell Biol. ,  2009, vol.  21 (pg.  394- 402) Google Scholar CrossRef Search ADS PubMed  Saitoh F,  Noma M,  Kawashima N.  The alkaloid content of sixty Nicotiana species,  Phytochemistry ,  1985, vol.  24 (pg.  477- 480) Google Scholar CrossRef Search ADS   Schuelke M.  An economic method for the fluorescent labeling of PCR fragments,  Nat. Biotechnol. ,  2000, vol.  18 (pg.  233- 234) Google Scholar CrossRef Search ADS PubMed  Shoji T,  Hashimoto T.  Ashihara H,  Crozier A,  Komamine A.  Nicotine biosynthesis,  Plant Metabolism and Biotechnology ,  2011 New York John Wiley & Sons(pg.  191- 216) Siegler EH,  Bowen CV.  Toxicity of nicotine, nornicotine, and anabasine to codling moth larvae,  J. Econ. Entomol. ,  1946, vol.  39 (pg.  673- 674) Google Scholar CrossRef Search ADS   Siminszky B,  Gavilano L,  Bowen SW,  Dewey RE.  Conversion of nicotine to nornicotine in Nicotiana tabacum is mediated by CYP82E4, a cytochrome P450 monooxygenase,  Proc. Natl Acad. Sci. USA ,  2005, vol.  102 (pg.  14919- 14924) Google Scholar CrossRef Search ADS   Sinclair S,  Johnson R,  Hamill JD.  Analysis of wound-induced gene expression in Nicotiana species with contrasting alkaloid profiles,  Funct. Plant Biol. ,  2004, vol.  31 (pg.  721- 729) Google Scholar CrossRef Search ADS   Sisson VA,  Severson RF.  Alkaloid composition of the Nicotiana species,  Beit. Tabakforsch. Int. ,  1990, vol.  14 (pg.  327- 339) Soloway SB.  Naturally occurring insecticides,  Environ. Health Perspect. ,  1976, vol.  14 (pg.  109- 117) Google Scholar CrossRef Search ADS PubMed  Thompson JR,  Register E,  Curotto J,  Kurtz M,  Kelly R.  An improved protocol for the preparation of yeast cells for transformation by electroporation,  Yeast ,  1998, vol.  14 (pg.  565- 571) Google Scholar CrossRef Search ADS PubMed  Wernsman EA,  Matzinger DF.  Time and site of nicotine conversion in tobacco,  Tobacco Sci. ,  1968, vol.  12 (pg.  226- 228) Wittkopp PJ,  Haerum BK,  Clark AG.  Evolutionary changes in cis and trans gene regulation,  Nature ,  2004, vol.  430 (pg.  85- 88) Google Scholar CrossRef Search ADS PubMed  Xu D,  Shen Y,  Chappell J,  Cui M,  Nielsen M.  Biochemical and molecular characterizations of nicotine demethylase in tobacco,  Physiol. Plant. ,  2007, vol.  129 (pg.  307- 319) Google Scholar CrossRef Search ADS   Yamanaka H,  Nakajima M,  Fukami T,  Sakai H,  Nakamura A,  Katoh M, et al.  CYP2A6 and CYP2B6 are involved in nornicotine formation from nicotine in humans: interindividual differences in these contributions,  Drug Metab. Dispos. ,  2005, vol.  33 (pg.  1811- 1818) Google Scholar PubMed  © The Author 2012. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com
TI - Non-Functionalization of Two CYP82E Nicotine N-Demethylase Genes Abolishes Nornicotine Formation in Nicotiana langsdorffii
JF - Plant and Cell Physiology
DO - 10.1093/pcp/pcs139
DA - 2012-10-03
UR - https://www.deepdyve.com/lp/oxford-university-press/non-functionalization-of-two-cyp82e-nicotine-n-demethylase-genes-loZK3saH2M
SP - 2038
EP - 2046
VL - 53
IS - 12
DP - DeepDyve
ER -