Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2015 Aug 21;169(2):1127–1140. doi: 10.1104/pp.15.01240

Isolation and Characterization of O-methyltransferases Involved in the Biosynthesis of Glaucine in Glaucium flavum1

Limei Chang 1, Jillian M Hagel 1, Peter J Facchini 1,*
PMCID: PMC4587479  PMID: 26297140

A subset of multifunctional O-methyltransferases is involved in the formation of the tetra-O-methylated benzylisoquinoline alkaloid glaucine.

Abstract

Transcriptome resources for the medicinal plant Glaucium flavum were searched for orthologs showing identity with characterized O-methyltransferases (OMTs) involved in benzylisoquinoline alkaloid biosynthesis. Seven recombinant proteins were functionally tested using the signature alkaloid substrates for six OMTs: norlaudanosoline 6-OMT, 6-O-methyllaudanosoline 4′-OMT, reticuline 7-OMT, norreticuline 7-OMT, scoulerine 9-OMT, and tetrahydrocolumbamine OMT. A notable alkaloid in yellow horned poppy (G. flavum [GFL]) is the aporphine alkaloid glaucine, which displays C8-C6′ coupling and four O-methyl groups at C6, C7, C3′, and C4′ as numbered on the 1-benzylisoquinoline scaffold. Three recombinant enzymes accepted 1-benzylisoquinolines with differential substrate and regiospecificity. GFLOMT2 displayed the highest amino acid sequence identity with norlaudanosoline 6-OMT, showed a preference for the 6-O-methylation of norlaudanosoline, and O-methylated the 3′ and 4′ hydroxyl groups of certain alkaloids. GFLOMT1 showed the highest sequence identity with 6-O-methyllaudanosoline 4′OMT and catalyzed the 6-O-methylation of norlaudanosoline, but more efficiently 4′-O-methylated the GFLOMT2 reaction product 6-O-methylnorlaudanosoline and its N-methylated derivative 6-O-methyllaudanosoline. GFLOMT1 also effectively 3′-O-methylated both reticuline and norreticuline. GFLOMT6 was most similar to scoulerine 9-OMT and efficiently catalyzed both 3′- and 7′-O-methylations of several 1-benzylisoquinolines, with a preference for N-methylated substrates. All active enzymes accepted scoulerine and tetrahydrocolumbamine. Exogenous norlaudanosoline was converted to tetra-O-methylated laudanosine using combinations of Escherichia coli producing (1) GFLOMT1, (2) either GFLOMT2 or GFLOMT6, and (3) coclaurine N-methyltransferase from Coptis japonica. Expression profiles of GFLOMT1, GFLOMT2, and GFLOMT6 in different plant organs were in agreement with the O-methylation patterns of alkaloids in G. flavum determined by high-resolution, Fourier-transform mass spectrometry.

Glaucine is a benzylisoquinoline alkaloid (BIA) of the aporphine subclass produced in members of the Papaveraceae, including yellow horned poppy (Glaucium flavum; Lapa et al., 2004), Glaucium oxylobum (Morteza-Semnani et al., 2003), and Corydalis yanhusuo (Xu et al., 2004), and in some plants such as Croton lechleri of the Euphorbiaceae (Milanowski et al., 2002). Glaucine functions as a phosphodiesterase-4 inhibitor and calcium channel blocker (Cortijo et al., 1999), displays bronchodilator and antiinflammatory effects, and is used to treat coughs and asthma in Iceland and several eastern European countries (Dargan et al., 2008). Glaucine has also been shown to decrease heart rate, lower blood pressure (Orallo et al., 1995), and relieve pain, although less effectively than other analgesics (Zetler, 1988). Side effects such as sedation, fatigue, and hallucinations have contributed to the increased recreational use of glaucine as a psychoactive drug (Dargan et al., 2008). G. flavum is native to Europe, northern Africa, Macaronesia, western Asia, and the Caucasus, growing exclusively on seashores, and the plant is an introduced species and noxious weed in some parts of North America.

Although the reaction sequence is not known, the biosynthesis of glaucine begins with the central BIA intermediate (S)-norcoclaurine (Hagel and Facchini, 2013), which must undergo (1) 3′-hydroxylation, (2) 6-, 7-, 3′-, and 4′-O-methylations, (3) N-methylation, and (4) oxidative C8-C6′ coupling (Fig. 1A). Initial tracer experiments in wild bleeding heart (Dicentra eximia; Papaveraceae) suggested that glaucine and other aporphine alkaloids were derived from norprotosinomenine (Battersby et al., 1971), which is norcoclaurine substituted via 3′-hydroxylation, and 7- and 4′-methoxylation. However, later tracer studies in bollywood (Litsea glutinosa; Lauraceae) indicated that the key BIA branch-point intermediate (S)-reticuline (Hagel and Facchini, 2013) was preferentially converted by C8-C6′ oxidative coupling to (S)-isoboldine, which is subsequently O-methylated at C3′ and C7 yielding glaucine (Bhakuni and Jain, 1988). Other substituted 1-benzylisoquinolines, including protosinomenine (N-methylated norprotosinomenine), orientaline (6- and 3′-methoxy-substituted norcoclaurine), and laudanosine (6-, 7-, 3′-, and 4′-methoxy-substituted norcoclaurine) were also incorporated at lower levels.

Figure 1.

Figure 1.

Open in a new tab

Substitutions involved in the conversion of (S)-norcoclaurine to (S)-glaucine (A), and known O-methyltransferase reactions in BIA biosynthesis (B). Substitutions are indicated in red. 6OMT, Norcoclaurine/norlaudanosoline 6-O-methyltransferase; 3′OMT, 6-O-methylnorlaudanosoline 3′-O-methyltransferase; N7OMT, norreticuline 7-O-methyltransferase; 7OMT, reticuline 7-O-methyltransferase; 4′OMT, 6-O-methyllaudanosoline 4′-O-methyltransferase; SOMT, scoulerine 9-O-methyltransferase; CoOMT, tetrahydrocolumbamine O-methyltransferase.

Several S-adenosyl-l-Met (SAM)-dependent, and substrate- and regiospecific, O-methyltransferases (OMTs) involved in BIA biosynthesis have been isolated (Fig. 1B), including (1) 6OMT from Japanese goldthread (Coptis japonica; Morishige et al., 2000) and opium poppy (Papaver somniferum; Ounaroon et al., 2003); (2) N7OMT from opium poppy (Pienkny et al., 2009); (3) 7OMT from opium poppy (Ounaroon et al., 2003); (4) 4′OMT from C. japonica (Morishige et al., 2000), opium poppy (Ziegler et al., 2005), and California poppy (Eschscholzia californica; Inui et al., 2007); (5) SOMT from C. japonica (Takeshita et al., 1995) and opium poppy (Dang and Facchini, 2012); and (6) CoOMT from C. japonica (Morishige et al., 2002). In addition, a 6-O -methylnorlaudanosoline 3′-OMT (Fig. 1B) was detected in Argemone platyceras cell cultures (Rueffer et al., 1983). SOMT and CoOMT show a preference for the protoberberine substrates (S)-scoulerine and (S)-tetrahydrocolumbamine, respectively (Fig. 1B), and are involved in the biosynthesis of diverse compounds such as berberine and noscapine (Hagel and Facchini, 2013).

The monophyletic origin of BIA metabolism in members of the order Ranunculales (Liscombe et al., 2005) provides a foundation for the application of biochemical genomics to the isolation of genes encoding enzyme variants with similar or unique catalytic functions. We have established deep transcriptome resources based on a combination of 454 and Illumina sequencing for 20 plant species representing four families in the Ranunculales, including G. flavum (Xiao et al., 2013). The availability of a transcriptome database for yellow horned poppy roots, in which glaucine accumulates to high levels, provided an opportunity to functionally characterize all expressed genes encoding orthologs of known OMTs involved in BIA biosynthesis using the reported signature substrate for each enzyme. We hypothesized that variants of each enzyme would be detected, and that the missing 3′OMT would be identified. In this article, we report the functional characterization of seven OMT orthologs from G. flavum roots with respect to BIA metabolism. Enzymes catalyzing efficient O-methylation at all positions on both 1-benzylisoquinoline and protoberberine scaffolds were identified. However, most active enzymes were multifunctional, accepting several substrates and efficiently targeting more than one position. The efficacy of the detected catalytic activities was confirmed in vivo using various combinations of Escherichia coli strains expressing individual recombinant enzymes. Inclusion of a strain producing recombinant coclaurine N-methyltransferase (CNMT) from C. japonica allowed certain strain combinations to convert norlaudanosoline (3′-hydroxylated norcoclaurine; Fig. 1) to laudanosine, which possesses the same O-substitutions as glaucine, but lacks the C8-C6′ coupling. The primary O-methylation reaction sequence(s) were identified.

RESULTS

Selection of G. flavum O-Methyltransferases

Seven orthologs of characterized OMTs involved in BIA biosynthesis were isolated from assembled 454 and Illumina Genome Analyzer databases (Xiao et al., 2013) of G. flavum (GFL). The candidate selection strategy was based on a cutoff of 40% amino acid sequence identity compared with at least one functionally characterized OMT involved in BIA metabolism (Supplemental Table S1). Phylogenetic analysis showed that GFLOMT1 to GFLOMT4 and GFLOMT6 formed separate clades with characterized OMTs (Fig. 2), whereas GFLOMT5 and GFLOMT7 formed a new clade. GFLOMT1 shares 77% and 71% amino acid sequence identities with Ps4′OMT2 from opium poppy and Cj4′OMT from C. japonica, respectively. GFLOMT2 shows 80% and 70% sequence identities with Ps6OMT from opium poppy and Cj6OMT from C. japonica, respectively. GFLOMT3 shares 63% sequence identities with Ps7OMT from opium poppy, and GFLOMT4 shows 38% sequence identity with CjCoOMT from C. japonica. GFLOMT6 shares 60% and 63% sequence identities with PsSOMT from opium poppy and CjSOMT from C. japonica, respectively. In contrast, GFLOMT5 and GFLOMT7 display only 42% and 44% sequence identity with the nearest neighbor CjSOMT from C. japonica, respectively. Phylogenetic relationships based on overall sequence identity were used to formulate an initial hypothesis that 4′OMT, 6OMT, 7OMT, SOMT, and CoOMT functions were associated with GFLOMT1, GFLOMT2, GFLOMT3, GFLOMT4, and GFLOMT6 orthologs in G. flavum. Moreover, the unknown 3′OMT activity required for glaucine biosynthesis was proposed to reside with GFLOMT5 or GFLOMT7, which do not clade with characterized enzymes.

Figure 2.

Figure 2.

Open in a new tab

Unrooted neighbor-joining phylogenetic tree combining functionally characterized OMTs involved in BIA metabolism and seven orthologs from yellow horned poppy. Bootstrap frequencies for each clade were based on 1,000 iterations. Abbreviations and GenBank accessions numbers for each enzyme are listed in the “Materials and Methods.” The scale bar corresponds to 0.2 amino acid substitutions per site.

Purification and Characterization of GFLOMTs

Full-length complementary DNAs (cDNAs) for the seven GFLOMT candidates were cloned into the pRSETA expression vector with an N-terminal His-tag translational fusion. Recombinant GFLOMTs were purified from total protein extracts using a cobalt-affinity resin. All purified recombinant enzymes displayed Mr values corresponding to expected protein sizes, as determined by SDS-PAGE (Supplemental Fig. S1). Enzyme assays were performed on each of the purified, His-tagged recombinant proteins to screen for O-methylation activity using seven compounds that serve as the signature substrates for characterized OMTs involved in BIA metabolism (Fig. 1B).

In the presence of SAM, GFLOMT1 showed differential activity with all seven substrates (Table I; Fig. 3). Norlaudanosoline was the preferred substrate, displaying 96% conversion in the standard OMT assay; however, 6-O-methylnorlaudanosoline (87%) and 6-O-methyllaudanosoline (68%) were also efficiently converted. Scoulerine (32%), tetrahydrocolumbamine (19%), and reticuline (6%) were also accepted, but with relatively lower conversion efficiencies. Norlaudanosoline was also the best substrate for GFLOMT2 (100%), whereas 6-O-methylnorlaudanosoline (1%) and 6-O-methyllaudanosoline (14%) were not efficiently converted (Table I; Fig. 4). In further contrast to GFLOMT1, scoulerine (75%), tetrahydrocolumbamine (32%), and reticuline (22%) were accepted with relatively higher conversion efficiencies. Scoulerine was the preferred substrate for GFLOMT6 (100%), with reticuline (97%) and tetrahydrocolumbamine (90%) also efficiently converted, and norreticuline (36%) and 6-O-methyllaudanosoline (23%) accepted at moderate levels (Table I; Fig. 5). GFLOMT6 did not accept norlaudanosoline or 6-O-methylnorlaudanosoline. GFLOMT7 showed relatively low activity with scoulerine (12%) and tetrahydrocolumbamine (8%), but did not accept other BIAs (Table I; Fig. 5). GFLOMT3, GFLOMT4, and GFLOMT5 did not show activity with any of the tested substrates (Table I).

Table I. Substrate range of GFLOMTs.

Values represent the percentage of each substrate converted to reaction product(s) in a standard assay. Specific enzyme activities are provided in the footnote. nd, Not detected.

Substrate Enzyme
GFLOMT1 GFLOMT2 GFLOMT3 GFLOMT4 GFLOMT5 GFLOMT6 GFLOMT7
(%)
(R,S)-norlaudanosoline 100a 100b nd nd nd nd nd
(R,S)-6-O-methylnorlaudanosoline 90 1 nd nd nd nd nd
(R,S)-6-O-methyllaudanosoline 71 14 nd nd nd 23 nd
(R,S)-norreticuline nd nd nd nd nd 36 nd
(S)-reticuline 6 22 nd nd nd 97 nd
(R,S)-scoulerine 3 75 nd nd nd 100c 100d
(R,S)-tetrahydrocolumbamine 20 32 nd nd nd 91 67
Open in a new tab
a

2,560 nmol min−1 mg−1 protein.

b

2,667 nmol min−1 mg−1 protein.

c

2,667 nmol min−1 mg−1 protein.

d

320 nmol min−1 mg−1 protein.

Figure 3.

Figure 3.

Open in a new tab

Total ion chromatograms showing the O-methylation activity of recombinant GFLOMT1 on various substrates. Numbers in square brackets correspond to identified reaction products based on retention times and ESI[+]-CID spectra (Supplemental Table S3).

Figure 4.

Figure 4.

Open in a new tab

Total ion chromatograms showing the O-methylation activity of recombinant GFLOMT2 on various substrates. Numbers in square brackets correspond to identified reaction products based on retention times and ESI[+]-CID spectra (Supplemental Table S3).

Figure 5.

Figure 5.

Open in a new tab

Total ion chromatograms showing the O-methylation activity of recombinant GFLOMT6 (A–E) and GFLOMT7 (F and G) on various substrates. Numbers in square brackets correspond to identified reaction products based on retention times and ESI[+]-CID spectra (Supplemental Table S3).

Kinetic analyses with preferred substrates yielding single reaction products showed that the three most active GFLOMTs followed the Michaelis-Menten model (Supplemental Fig. S2). GFLOMT1 exhibited a Km of 12 μm for 6-O-methylnorlaudanosoline, GFLOMT2 showed a Km of 15 μm for norlaudanosoline, and GFLOMT6 displayed a Km of 22 μm for scoulerine (Table II). Catalytic efficiencies (kcat/Km) were relatively high for all three conversions.

Table II. Kinetic data for GFLOMT1, GFLOMT2, and GFLOMT6.

Enzyme Substrate Km Vmax kcat kcat/Km
(µm)
 (nmol min−1 mg−1 protein)
 (s−1)
 (m−1 s−1)
GFLOMT1 6-O-methylnorlaudanosoline 17.2 ± 2.8 2,360 ± 204 0.93 ± 0.04 54,100
S-adenosyl-Met 24.3 ± 4.5 2,442 ± 230 0.96 ± 0.05 39,500
GFLOMT2 Norlaudanosoline 18.3 ± 2.5 3,591 ± 255 2.78 ± 0.09 151,900
GFLOMT2 Scoulerine 31.0 ± 5.0 3,220 ± 253 1.25 ± 0.06 40,300
S-adenosyl-Met 27.1 ± 7.1 4,110 ± 529 3.18 ± 0.22 117,300
GFLOMT6 Scoulerine 26.4 ± 4.1 4,343 ± 332 3.68 ± 0.15 139,200
S-Adenosyl-Met 27.4 ± 4.6 4,079 ± 337 3.46 ± 0.16 126,300
Open in a new tab

Reaction Product Identification

Enzyme assays were subjected to positive-mode electrospray ionization (ESI[+]) liquid chromatography (LC)-tandem mass spectrometry (MS/MS) for reaction product characterization, including collision-induced dissociation (CID) fragmentation analysis. ESI[+]-CID of 1-benzylisoquinoline and protoberberine alkaloids at low ionization energy yields isoquinoline and/or benzyl moieties as major ion fragments. Using the ESI[+]-CID spectra of authentic standards (Supplemental Table S2), the identity of recombinant GFLOMT reaction products was determined (Supplemental Table S3). Positions of new O-methyl groups could be inferred from the increased mass-to-charge ratio (m/z; in multiples of 14 D) of dissociated isoquinoline and benzyl ion fragments even in the absence of authentic standards, although most were available.

Incubation of GFLOMT1 with norlaudanosoline (m/z 288) yielded two major peaks, with m/z 302 at 2.62 min and m/z 316 at 2.95 min (Fig. 3A), suggesting single and double O-methylation events, respectively. The parent ion with m/z 302 produced an ESI[+]-CID spectrum corresponding to authentic 6-O-methylnorlaudanosoline, whereas the parent ion with m/z 316 yielded an ESI[+]-CID spectrum matching that of norreticuline. Assays containing GFLOMT1 and 6-O-methylnorlaudanosoline (m/z 302) generated major and minor products of m/z 316 and m/z 330, with ESI[+]-CID spectra corresponding to norreticuline and norcodamine, respectively (Fig. 3B). Although an authentic standard for norcodamine was not available, compound identity could be inferred. Compared with the ESI[+]-CID spectrum of 6-O-methylnorlaudanosoline, which displays the fragment ions m/z 178 (isoquinoline moiety) and m/z 123 (benzyl moiety), the m/z 330 reaction product yielded major fragment ions of m/z 178 and m/z 151 (increase of 28 D), the latter of which corresponds to a 3′- and 4′-O-methylated benzyl moiety. Incubation of GFLOMT1 with 6-O-methyllaudanosoline (m/z 316) yielded major and minor products with m/z 330 at 2.99 min and m/z 344 at 3.17 min, corresponding to single and double O-methylation events, respectively (Fig. 3C). The m/z 330 parent ion produced an ESI[+]-CID spectrum corresponding to authentic reticuline, whereas the double O-methylated m/z 344 parent ion yielded an ESI[+]-CID spectrum matching that of codamine. In assays containing GFLOMT1 and reticuline, a minor product of m/z 344 with an ESI[+]-CID spectrum corresponding to codamine was also produced (Fig. 3D). The major and minor products resulting from the incubation of GFLOMT1 with scoulerine showed parent ions of m/z 342 and m/z 356, with ESI[+]-CID spectra corresponding to tetrahydropalmatrubine and tetrahydropalmatine (Fig. 3E). An authentic standard for tetrahydropalmatrubine was not available; however, product identification was inferred from the 14-D increase in the isoquinoline moiety of scoulerine (m/z 178) to m/z 192. In assays containing GFLOMT1 and tetrahydrocolumbamine, a minor product was generated with a parent mass of m/z 356 and an ESI[+]-CID spectrum corresponding to tetrahydropalmatine (Fig. 3F).

GFLOMT2 efficiently converted norlaudanosoline (m/z 288) to a product with m/z 302, which yielded an ESI[+]-CID spectrum corresponding to 6-O-methylnorlaudanosoline (Fig. 4A). In contrast, GFLOMT2 incubated with 6-O-methylnorlaudanosoline generated a minor product with m/z 316 (Fig. 4B), which was inferred as nororientaline based on the detection of major fragment ions of m/z 178 (isoquinoline moiety) and m/z 137 (3′-O-methylated benzyl moiety). The different retention time compared with norreticuline (i.e. 4′-O-methylated 6-O-methylnorlaudanosoline) confirmed 3′- rather than 4′-O-methylation. Incubation of GFLOMT2 with 6-O-methyllaudanosoline (m/z 316) yielded three products with m/z 330 at 2.85 min, m/z 330 at 2.98 min, and m/z 344 at 3.15 min, indicating both single and double O-methylation events (Fig. 4C). The identity of the m/z 330 parent ion at 2.85 min was inferred as orientaline based on the detection of major fragment ions of m/z 192 (isoquinoline moiety) and m/z 137 (3′-O-methylated benzyl moiety). The different retention time compared with reticuline (i.e. 4′-O-methylated 6-O-methyllaudanosoline and the m/z 330 parent ion at 2.98 min) confirmed 3′- rather than 4′-O-methylation. The double O-methylation product with m/z 344 yielded an ESI[+]-CID spectrum corresponding to codamine. The minor GFLOMT2 reaction product of reticuline also showed a parent mass of m/z 344 and an ESI[+]-CID spectrum corresponding to codamine (Fig. 4D). Incubation of GFLOMT2 with scoulerine (m/z 328) yielded major and minor products with m/z 342 and m/z 356, identified as tetrahydropalmatrubine and tetrahydropalmatine, respectively (Fig. 4E). The reaction product of GFLOMT2 incubated with tetrahydrocolumbamine showed a parent mass of m/z 356, which was also identified as tetrahydropalmatine (Fig. 4F).

Incubation of GFLOMT6 with 6-O-methyllaudanosoline (m/z 316) yielded a reaction product with m/z 330 (Fig. 5A), which was inferred as 6,7-O,O-dimethyllaudanosoline based on ESI[+]-CID spectrum and a unique retention time. Compared with the ESI[+]-CID spectrum of 6-O-methyllaudanosoline, which displayed major fragment ions of m/z 192 and m/z 123, the m/z 330 reaction product yielded fragment ions of m/z 206 (i.e. the 7-O-methylated isoquinoline moiety) and m/z 123 (i.e. the unchanged benzyl moiety). Incubation of GFLOMT6 with norreticuline (m/z 316) yielded two major and one minor reaction products with m/z 330 at 3.16 min, m/z 330 at 3.25 min, and m/z 344 at 3.43 min (Fig. 5B), corresponding to single and double O-methylation events. Identity of the m/z 330 parent ion at 3.16 min was inferred as norcodamine based on the ESI[+]-CID spectrum and a unique retention time. The m/z 330 parent ion at 3.25 min produced an ESI[+]-CID spectrum corresponding to norlaudanine. The minor double O-methylation product with m/z 344 produced an ESI[+]-CID spectrum matching that of tetrahydropapaverine. GFLOMT6 efficiently converted reticuline (m/z 330) to three products with m/z 344 at 3.17 min, m/z 344 at 3.24 min, and m/z 358 at 3.43 min (Fig. 5C), indicating single and double O-methylation events. The m/z 344 parent ions at 3.17 and 3.24 min yielded ESI[+]-CID spectra corresponding to codamine and laudanine. The double O-methylation product with a parent ion of m/z 358 generated an ESI[+]-CID spectrum matching that of laudanosine. GFLOMT6 efficiently converted scoulerine (m/z 328) to two major reaction products with m/z 342 at 3.29 min and m/z 356 at 3.59 min (Fig. 5D), corresponding to single and double O-methylation events, respectively. The m/z 342 parent ion produced an ESI[+]-CID spectrum corresponding to tetrahydrocolumbamine, whereas the m/z 356 parent ion yielded an ESI[+]-CID spectrum matching that of tetrahydropalmatine. Incubation of GFLOMT6 with tetrahydrocolumbamine (m/z 342) generated a major product with m/z 356 and an ESI[+]-CID spectrum corresponding to tetrahydropalmatine (Fig. 5E).

Incubation of GFLOMT7 with scoulerine (m/z 328) yielded two minor reaction products with m/z 342 at 3.33 min and m/z 342 at 3.44 min (Fig. 5F), with ESI[+]-CID spectra corresponding to tetrahydrocolumbamine (i.e. the methylated benzyl moiety of scoulerine) and tetrahydropalmatrubine (i.e. the methylated isoquinoline moiety of scoulerine), respectively. The reaction product of GFLOMT7 incubated with tetrahydrocolumbamine (m/z 342) generated a reaction product with m/z 356 and an ESI[+]-CID spectrum corresponding to tetrahydropalmatine (Fig. 5G).

Transformations of Norlaudanosoline in E. coli

(R,S)-norlaudanosoline was fed to mixed cultures of E. coli harboring different combinations and permutations of the expression constructs pGFLOMT1, pGFLOMT2, pGFLOMT6, and pCNMT to determine the in vivo efficiency of each OMT using both N-methylated and N-desmethyl 1-benzylisoquinolines (Fig. 6). Norlaudanosoline was not recovered in ethyl acetate extractions or was not detected. The empty vector control showed that E. coli was inherently incapable of transforming norlaudanosoline to other BIAs (Fig. 6A). Transformation product identifications (Supplemental Table S4) were determined using the ESI[+]-CID spectra of authentic standards and inferences described above.

Figure 6.

Figure 6.

Open in a new tab

Total ion chromatograms showing the products of various combinations of E. coli strains harboring individual pGFLOMT constructs, plus or minus a strain harboring pCNMT from C. japonica, and fed 25 µm (R,S)-norlaudanosoline. Cultures were incubated overnight and alkaloid extracts subjected to LC-MS/MS analysis. Product identifications were based on retention times and ESI[+]-CID spectra (Supplemental Table S4). The empty expression vector served as the negative control (A).

Incubation of an E. coli strain harboring pGFLOMT1 with norlaudanosoline (m/z 288) yielded one compound identified as norcodamine (m/z 330; Fig. 6B). In contrast, incubation of an E. coli strain harboring pGFLOMT2 with norlaudanosoline generated three products with m/z 316 at 2.85 min, m/z 330 at 3.15 min, and m/z 344 at 3.42 min (Fig. 6C), corresponding to nororientaline, norcodamine, and tetrahydropapaverine, respectively. Incubation of mixed E. coli strains harboring pGFLOMT1 and pGFLOMT2 with norlaudanosoline yielded compounds with m/z 330 at 3.15 min and m/z 344 at 3.42 min (Fig. 6D), corresponding to norcodamine and tetrahydropapaverine, respectively. Addition of an E. coli strain harboring pCNMT to this series altered the profile of products formed in all pGFLOMT combinations. Incubation of mixed E. coli strains harboring pGFLOMT1 and pCNMT with norlaudanosoline yielded products with m/z 330 at 2.99 min and m/z 344 at 3.17 min, which were identified as reticuline and codamine (Fig. 6E). Incubation of mixed E. coli strains harboring pGFLOMT2 and pCNMT with norlaudanosoline resulted in the production of four compounds with m/z 330 at 2.85 min, m/z 330 at 2.98 min, m/z 344 at 3.16 min, and m/z 358 at 3.4 min, corresponding to orientaline, reticuline, codamine, and laudanosine (Fig. 6F). Incubation of mixed E. coli strains harboring pGFLOMT1, pGFLOMT2, and pCNMT with norlaudanosoline also produced orientaline, reticuline, codamine, and laudanosine, but with an apparently higher yield compared with incubations lacking one of the strains (Fig. 6G).

Incubation of mixed E. coli strains harboring pGFLOMT1 and pGFLOMT6 with norlaudanosoline yielded two products with m/z 330 at 3.14 min and m/z 344 at 3.43 min, corresponding to norcodamine and tetrahydropapaverine, respectively (Fig. 6H). Incubation of mixed E. coli strains harboring pGFLOMT2 and pGFLOMT6 with norlaudanosoline generated three compounds with m/z 316 at 2.87 min, m/z 330 at 3.14 min, and m/z 344 at 3.43 min, identified as nororientaline, norcodamine, and tetrahydropapaverine, respectively (Fig. 6I). Combining E. coli strains harboring pGFLOMT1, pGFLOMT2, and pGFLOMT6 and incubating with norlaudanosoline only changed the relative abundance of nororientaline, norcodamine, and tetrahydropapaverine (Fig. 6J) compared with the absence of pGFLOMT1 (Fig. 6I). However, addition of an E. coli strain harboring pCNMT to this series altered the profile of products generated by all pGFLOMT combinations. Incubation of mixed E. coli strains harboring pGFLOMT1, pGFLOMT6, and pCNMT with norlaudanosoline produced five compounds with m/z 330 at 2.99 min, m/z 330 at 3.16 min, m/z 344 at 3.18 min, m/z 344 at 3.24 min, and m/z 358 at 3.42 min, which were identified as reticuline, norcodamine, codamine, laudanine, and laudanosine, respectively (Fig. 6K). Mixed E. coli strains harboring pGFLOMT2, pGFLOMT6, and pCNMT converted norlaudanosoline to compounds with m/z 330 at 2.88 min, m/z 330 at 3 min, m/z 344 at 3.16 min, m/z 344 at 3.23 min, and m/z 358 at 3.42 min, corresponding to orientaline, reticuline, codamine, laudanine, and laudanosine, respectively (Fig. 6L). Finally, incubation of mixed E. coli strains harboring pGFLOMT1, pGFLOMT2, pGFLOMT6, and pCNMT with norlaudanosoline produced the same five compounds, but with an apparently higher yield than incubations lacking the E. coli strain harboring pGFLOMT1 (Fig. 6M).

GFLOMT Gene Expression and Alkaloid Accumulation in G. flavum

GFLOMT1, GFLOMT2, and GFLOMT6 showed generally higher levels of expression in different G. flavum organs compared with genes encoding other OMTs (Fig. 7A). Overall, transcripts encoding GFLOMT1, GFLOMT2, and GFLOMT6 were also relatively lower in stems compared with other organs. High-resolution FTMS performed in positive ion mode on corresponding G. flavum organs yielded unbiased exact mass data sets, each containing >200 putative metabolites (Supplemental Data Set S1). Comparison of the retention times and CID spectra of authentic standards (Farrow et al., 2012) facilitated the annotation of putative alkaloids, seven of which could be unambiguously identified (Supplemental Table S5). Glaucine was the predominant alkaloid in aerial organs, occurring at levels up to 100-fold higher than the combined accumulation of all other identified BIAs (Fig. 7B). Conversely, the benzophenanthridine alkaloids sanguinarine and chelerythrine, and the protopine alkaloids, in particular protopine and allocryptopine, predominated in root (Fig. 7C). Levels of the aporphine alkaloids isocorydine and glaucine were similar in root, but isocorydine accumulation was relatively low in aerial organs.

Figure 7.

Figure 7.

Open in a new tab

Quantitative reverse transcription-PCR showing the relative expression of seven GFLOMT genes (A) and high-resolution, Fourier-transform mass spectrometry (FTMS) of BIAs (B and C) in different organs of G. flavum. Bars represent the mean ± sd of four biological replicates.

DISCUSSION

Elucidation of BIA metabolic pathways in plants has advanced rapidly over the past decade with a near complete catalog of biosynthetic genes and corresponding enzymes involved in the formation of major products, including berberine, sanguinarine, morphine, and noscapine (Hagel and Facchini, 2013; Dang et al., 2015). Among the BIA biosynthetic genes not yet isolated is one encoding an enzyme capable of the efficient 3′-O-methylation of 1-benzylisoquinoline and/or aporphine substrates. Several major 1-benzylisoquinoline derivatives, such as glaucine in yellow horned poppy and papaverine in opium poppy, are tetra-O-methylated. A number of BIA OMTs with a wide range of substrate and regiospecificity have been isolated and characterized from opium poppy although none have shown efficient 3′-O-methylation activity with 1-benzylisoquinoline substrates (Ounaroon et al., 2003; Ziegler et al., 2005; Pienkny et al., 2009; Dang and Facchini, 2012). The availability of deep transcriptome resources for several BIA-producing plant species (Xiao et al., 2013) facilitated a more comprehensive approach to the characterization of OMT orthologs involved in the biosynthesis of unique compounds, such as glaucine, with the isolation of a 3′OMT as a principal objective. The monophyletic origin of BIA biosynthetic genes in members of the Ranunculaceae suggests that orthologs in opium poppy and other plant species could then be readily identified (Liscombe et al., 2005).

Despite names suggesting strict functions, characterized BIA OMTs often display a range of substrate and regiospecificities. For example, among several 1-benzylisoquinoline and protoberberine substrates tested, opium poppy 6OMT specifically 6-O-methylated (R,S)-norcoclaurine, whereas (R,S)-reticuline, (R,S)-orientaline, and (R,S)-protosinomenine were 7-O-methylated by opium poppy 7OMT (Ounaroon et al., 2003). However, both enzymes also 7-O-methylated (R,S)-isoorientaline. In contrast, opium poppy N7OMT only accepted (S)-norreticuline among a variety of tested N-methylated and N-desmethyl 1-benzylisoquinolines, including (R,S)-reticuline (Pienkny et al., 2009). Opium poppy 4′OMT displayed its highest activity with (R,S)-6-O-methyllaudanosoline and (R,S)-6-O-laudanosoline, but also less efficiently accepted (R,S)-6-O-methylnorlaudanosoline and (R,S)-norlaudanosoline (Ziegler et al., 2005). 4′-O-methylation of all substrates was assumed, but only 6- and 7-O-methylations could be ruled out. Opium poppy SOMT efficiently 9-O-methylated (S)-scoulerine and showed 2-O-methylation of (S)-tetrahydrocolumbamine, but also weakly 7- and 3′-O-methylated (S)-reticuline and (S)-norreticuline (Dang and Facchini, 2012). Reported orthologs from other species, especially Japanese goldthread (C. japonica; Ranunculaceae), generally showed functional similarity to opium poppy OMTs in terms of signature substrates, but also displayed variances in their range of substrate and regiospecificities. 6OMT from C. japonica efficiently 6-O-methylated (R,S)-norcoclaurine, (R,S)-norlaudanosoline, and (R,S)-laudanosoline, but weakly O-methylated (R,S)-6-O-methylnorlaudanosoline, (S)-scoulerine, and (R,S)-coclaurine, clearly not at the already O-methylated 6-position in each case (Sato et al., 1993; Morishige et al., 2000). 4′OMT from C. japonica preferred (R,S)-6-O-laudanosoline and (R,S)-6-O-methyllaudanosoline, but also 4′-O-methylated the corresponding N-desmethyl compounds (R,S)-norlaudanosoline and (R,S)-6-O-methylnorlaudanosoline, and showed weak activity with (S)-scoulerine (Sato et al., 1993; Morishige et al., 2000). Similar to opium poppy SOMT, the C. japonica ortholog showed low activity with 1-benzylisoquinolines such as (R,S)-norreticuline (Sato et al., 1993; Morishige et al., 2000). CoOMT from C. japonica showed exclusive activity with protoberberines, preferring columbamine over (R,S)-scoulerine (Morishige et al., 2002). Only limited information is available for OMTs from other species. California poppy (E. californica) has been suggested to utilize a single OMT with both 4′- and 6-O-methylation activities (Inui et al., 2007). SOMT from barberry (Berberis wilsonae; Berberidaceae) appeared specific for the 9-O-methylation of (S)-scoulerine and did not accept other tested protoberberines or (R,S)-reticuline (Muemmler et al., 1985). Finally, Berberis koetineuna 4′OMT effectively 4′-O-methylated (S)-6-O-methyllaudanosoline, (R,S)-6-O-laudanosoline, and (R,S)-7-O-methylnorlaudanosoline, and showed weak activity with (S)-norlaudanosoline (Frenzel and Zenk, 1990).

Homology searches of combined 454- and Illumina-generated transcriptome databases for G. flavum resulted in the isolation of seven candidate cDNAs encoding proteins with >40% amino acid identity to at least one characterized OMT involved in BIA metabolism (Fig. 2; Supplemental Table S1). The characterized OMT, or group of functionally related enzymes, displaying the maximum amino acid sequence identity was the basis for predicting substrate- and regiospecific activity for each GFLOMT. Partially purified recombinant protein obtained for each GFLOMT (Supplemental Fig. S1) was assayed with the signature substrate for each of seven characterized OMTs (Fig. 1). In one case, 6-O-methylnorlaudanosoline was the only substrate accepted by an apparently regiospecific enzyme partially purified from A. platyceras cell cultures (Rueffer et al., 1983), although the corresponding gene has not been isolated. In general, the signature substrate of the nearest phylogenetic ortholog was accepted by each active G. flavum OMT. GFLOMT1 exhibited the highest amino acid identity with characterized 4′OMTs and catalyzed the 4′-O-methylation of 6-O-methyllaudanosoline to reticuline (Fig. 3; Table I). GFLOMT2 showed the highest amino acid sequence identity with characterized 6OMTs (and with opium poppy N7OMT) and catalyzed the 6-O-methylation of norlaudanosoline yielding 6-O-methylnorlaudanosoline (Fig. 4; Table I). GFLOMT6 was the closest SOMT ortholog and catalyzed the 9-O-methylation of scoulerine to tetrahydrocolumbamine, but also efficiently 2-O-methylated tetrahydrocolumbamine yielding tetrahydropalmatine (Fig. 5; Table I). However, the functions of the remaining GFLOMTs did not correspond to those of characterized enzymes. GFLOMT3 displayed substantial amino acid sequence identity with 7OMTs, but did not accept reticuline or any other tested 1-benzylisoquinoline or protoberberine substrates (Table I). Similarly, GFLOMT4 was the closest ortholog to an uncharacterized enzyme from opium poppy (SOMT2; Dang and Facchini, 2012; Winzer et al., 2012), and GFLOMT5 showed low amino acid sequence identity with characterized SOMTs. However, neither enzyme was active with any tested substrates. GFLOMT7 also showed low amino acid sequence identity with characterized SOMTs and displayed relatively weak activity with scoulerine (Table I). However, both the 9- and 2-hydroxyl positions were independently targeted (Fig. 5). Interestingly, despite the depth of transcriptome resources from the alkaloid-rich roots of G. flavum, no ortholog of C. japonica CoOMT was detected (Supplemental Table S1).

An initial hypothesis for the identification of a 3′OMT was that orthologs displaying relatively low amino acid sequence identity with functionally characterized OMTs, none of which display efficient 3′-O-methylation activity, could represent a unique substrate and regiospecific catalyst. The only indication that a 3′OMT might be highly substrate specific was based on the characterization of a partially purified A. platyceras enzyme, which accepted only 6-O-methylnorlaudanosoline (Rueffer et al., 1983). However, GFLOMT3, GFLOMT4, GFLOMT5, or GFLOMT7 failed to accept this substrate. Further examination of the range of products produced by the five tested 1-benzylisoquinoline substrates showed that GFLOMT1, GFLOMT2, and GFLOMT6 catalyzed additional regiospecific O-methylations with differential efficiency, in addition to the aforementioned signature reactions. The least multifunctional enzyme was GFLOMT2, which appears to serve as an efficient 6OMT in G. flavum. Nevertheless, GFLOMT2 also performed 3′- and 4′-O-methylations on various 1-benzylisoquinolines, and functioned as a regiospecific scoulerine 2-O-methyltransferase yielding tetrahydropalmatrubine (Fig. 4).

GFLOMT1 and GFLOMT6 were more versatile enzymes, but also displayed striking functional differences. With 1-benzylisoquinoline substrates, GFLOMT1 was relatively efficient at the 6- and 4′-O-methylation of norlaudanosoline and 6-O-methylnorlaudanosoline, respectively (Fig. 3), whereas GFLOMT6 did not accept these substrates (Table I). With less apparent efficiency, GFLOMT1 also catalyzed 3′-O-methylations on substrates yielding low levels of codamine (Fig. 3). In contrast, GFLOMT6 catalyzed the efficient 7- and 3′-O-methylation of norreticuline and reticuline, although there was a preference for N-methylated substrates with, at minimum, a 6-O-methyl and ideally a 4′-O-methyl group (Fig. 5). Taking an unbiased and comprehensive approach to the functional characterization of OMT candidates from an unexplored BIA-producing plant led to both (1) corroboration of previously reported activities for several orthologs and (2) unique and important insights into the associations between amino acid sequence identity, and substrate and regiospecificity. In G. flavum, we conclude that only three OMTs are responsible for four and two O-methylations on 1-benzylisoquinoline and protoberberine substrates, respectively.

To test whether GFLOMT1, GFLOMT2, and GFLOMT6 were capable of the efficient tetra-O-methylation of the 1-benzylisoquinoline scaffold, individual E. coli strains producing each enzyme (Supplemental Fig. S1) were combined in various permutations and combinations, with or without a strain producing a functional CNMT, and fed norlaudanosoline. Analysis was primarily qualitative, although equivalent scaling allowed an estimate of relative product accumulation. Results from in vivo feeding (Fig. 6) were generally consistent with in vitro assays (Figs. 3–5). As expected from the range of substrate and regiospecificity shown by GFLOMT1, GFLOMT2, and GFLOMT6 in vitro, the inclusion of strains producing all three enzymes together with the introduced capacity for N-methylation was the most effective combination for the production of the 6,7,4′-, 6,3′,4′-, and 6,7,3′,4′-O-methylated compounds laudanine, codamine, and laudanosine, respectively. Fewer products were obtained at lower levels in the absence of CNMT, demonstrating the importance of N-methylation to complete the tetra-O-methylation process, at least in G. flavum. Differences in substrate preference would be expected in species, including opium poppy, that produce high levels of an N-desmethyl 1-benzylisoquinoline, such as papaverine. Although a hitherto untested OMT might catalyze efficient 3′-O-methylation of 1-benzylisoquinolines in opium poppy, the search for this functionality should not preclude previously characterized enzymes (Hagel and Facchini, 2013).

GFLOMT1, GFLOMT2, and GFLOMT6 showed relatively low Km values and high catalytic efficiencies as determined using substrates yielding a single product, and thus allowing the calculation of reliable kinetic data (Table II). Based primarily on the apparent in vitro conversion efficiency of each enzyme, supported by in vivo feeding experiments, the major routes for the transformation of the tetra-hydroxylated and N-desmethylated norlaudanosoline to the tetra-methoxylated and N-methylated laudanosine were mapped (Fig. 8). Among all possible O-methylation sequences involving both N-desmethylated and N-methylated intermediates (Supplemental Fig. S3), only routes through norreticuline and reticuline appear active, consistent with the proposed glaucine biosynthetic pathway based on tracer feeding experiments (Bhakuni and Jain, 1988). For the formation of (S)-glaucine, C8-C6′ oxidative coupling of (S)-reticuline to (S)-isoboldine is also consistent with substrate specificity of corytuberine synthase (CYP80G2), which leads to the aporphine magnoflorine in C. japonica (Ikezawa et al., 2008). The structural similarity of the 1-benzylisoquinoline and aporphine scaffolds suggests that GFLOMTs also use certain aporphine derivatives as glaucine pathway intermediates. Unfortunately, the key aporphine (S)-isoboldine was not available to test this hypothesis.

Figure 8.

Figure 8.

Open in a new tab

Proposed major and minor O-methylation routes involved in the O-methylation of (R,S)-norlaudanosoline based on the detected functions of GFLOMT1, GFLOMT2, and GFLOMT6. Diagonal blue arrows represent N-methylation reactions. Major substrates and reaction products are highlighted in yellow. Enzymes in bold catalyze relatively efficient reactions compared with other conversions.

Alkaloid content in G. flavum has been reported to show substantial variation with glaucine, in particular, occurring as a major alkaloid in some ecotypes, but apparently absent in others (Peled et al., 1988). Previously detected BIAs generally belong to the protopine, benzophenanthridine, or aporphine subgroups. HPLC-UV analysis showed that glaucine predominates in aerial organs, whereas protopine occurs mostly in roots (Bournine et al., 2013). Isocorydine, an aporphine alkaloid with one free hydroxyl group, has also been reported in G. flavum (Daskalova et al., 1988; Peled et al., 1988; Kintsurashvili and Vachnadze 2000; Petitto et al., 2010; ). Our high-resolution, FTMS analysis provided unequivocal confirmation of the alkaloid profile in G. flavum, with glaucine found predominantly in aerial organs (Fig. 7, B and C). Gene expression analysis shows that GFLOMT1, GFLOMT2, and GFLOMT6 are expressed in all plant organs (Fig. 7A) in support of their proposed roles in the biosynthesis of glaucine above ground and other alkaloids in the root.

In conclusion, we have characterized all apparent OMTs involved in BIA metabolism from the roots of G. flavum, and shown that three multifunctional enzymes are responsible for four substrate and regiospecific O-methylations. In addition to improving our understanding of plant alkaloid metabolism, functional OMT variants are important to the synthetic biology objective of reconstituting BIA biosynthetic pathways in microorganisms (Hawkins and Smolke, 2008; Nakagawa et al., 2011; Fossati et al., 2014). A comprehensive characterization of enzyme functionality is clearly essential for the rational engineering of metabolic pathways in any organism.

MATERIALS AND METHODS

Plants and Chemicals

Glaucium flavum plants were obtained from the collection at the Jardin botanique de Montréal (espacepourlavie.ca/jardin-botanique). (R,S)-6-O-methylnorlaudanosoline was purchased from Toronto Research Chemicals. (R,S)-6-O-methyllaudanosoline was generated from (R,S)-6-O-methylnorlaudanosoline using purified recombinant CNMT Coptis japonica (Choi et al., 2002). Enzymatic conversion was performed at 37°C with 1.2 mm (R,S)-6-O-methylnorlaudanosoline, 200 μm SAM, and 150 μg of purified CjCNMT. Reactions were extracted by ethyl acetate three times. The purity and identity of 6-O-methyllaudanosoline was confirmed by LC-MS/MS. (R,S)-norreticuline was produced from (R,S)-6-O-methylnorlaudanosoline using purified recombinant C. japonica 4′OMT (Morishige et al., 2000). (S)-tetrahydrocolumbamine was prepared from (R,S)-scoulerine using purified recombinant Papaver somniferum SOMT1 (Dang and Facchini, 2012). Other alkaloids were obtained as described previously (Liscombe and Facchini, 2007; Hagel and Facchini, 2010; Dang and Facchini, 2012). SAM was purchased from Sigma-Aldrich. All other chemicals were purchased from Bioshop.

Phylogenetic Analysis

Amino acid alignments were performed using ClustalW (Larkin et al., 2007), and a phylogenetic tree was built by the neighbor-joining method using the Geneious (Biomatters) software package. Abbreviations and GenBank accession numbers for sequences used to construct the phylogenetic tree are as follows: Cj4′OMT, C. japonica SAM:3′-hydroxy-N-methylcoclaurine 4′-O-methyltransferase (BAB08005); Cj6OMT, C. japonica SAM:norcoclaurine 6-O-methyltransferase (BAB08004); CjCoOMT, C. japonica columbamine O-methyltransferase (BAC22084); CjSOMT, C. japonica SAM:scoulerine 9-O-methyltransferase (BAA06192); Ec4′OMT, Eschscholzia californica 3′-hydroxy-N-methylcoclaurine-4′-O-methyltransferase (BAM37633); Ec6OMT, E. californica O-methyltransferase (BAM37634); Ec7OMT, E. californica reticuline-7-O-methyltransferase (BAE79723); Ps4′OMT1, P. somniferum SAM:3′-hydroxy-N-methylcoclaurine 4′-O-methyltransferase 1 (AAP45313); Ps4′OMT2, P. somniferum SAM:3′-hydroxy-N-methylcoclaurine 4’-O-methyltransferase 2 (AAP45314); Ps6OMT, P. somniferum SAM:norcoclaurine 6-O-methyltransferase (AAP45315); Ps7OMT, P. somniferum reticuline 7-O-methyltransferase (AAQ01668); PsN7OMT, P. somniferum norreticuline-7-O-methyltransferase (ACN88562); PsSOMT, P. somniferum scoulerine-9-O-methyltransferase (AFK73709); PsSOMT2, P. somniferum O-methyltransferase 2 (AFK73710); PsSOMT3, P. somniferum O-methyltransferase 3 (AFK73711); Tf4′OMT, Thalictrum flavum 3′-hydroxy-N-methylcoclaurine 4′-O-methyltransferase (AAU20768); Tf6OMT, T. flavum norcoclaurine 6-O-methyltransferase (AAU20765); and TfSOMT, T. flavum scoulerine 9-O-methyltransferase (AAU20770).

RNA Extraction and cDNA Synthesis

Total RNA was isolated from G. flavum tissues ground to a fine powder under liquid nitrogen using a Tissue Lyser (Qiagen) and extracted using the cetyl-trimethyl-ammonium bromide method (Gambino et al., 2008). Reverse transcription was performed in a 20-µL reaction containing approximately 1.5 µg of total RNA using Moloney murine leukemia virus reverse transcriptase (Invitrogen) according to the manufacturer’s instructions.

Cloning and Expression of GFLOMT cDNAs

Full-length GFLOMT coding regions were amplified from cDNA derived from total G. flavum RNA using Phusion High-Fidelity Taq DNA polymerase (New England Biolabs) and the primer sets listed in Supplemental Table S6. For heterologous expression of His-tagged GFLOMT proteins, PCR products were ligated into pRSETA (Invitrogen) and transformed in Escherichia coli strain Rosetta (DE3) pLysS (EMD Millipore). For the production of recombinant GFLOMT proteins, bacteria were cultured in Luria-Bertani medium to an optical density at 600 nm of 0.6 and subsequently induced with 1 mm isopropyl β-d-thiogalactopyranoside (IPTG). Cultures were incubated at 30°C on a gyratory shaker at 200 rpm. Cells were harvested and sonicated in binding buffer (50 mm potassium phosphate [pH 7.5], 100 mm NaCl, 10% [v/v] glycerol, and 1 mm β-mercaptoethanol). Cleared lysates obtained after centrifugation at 10,000g for 10 min were loaded onto Talon cobalt affinity resin (Clontech). The resin was washed three times in binding buffer and eluted using aliquots of binding buffer containing increasing concentrations of imidazole to obtain purified proteins. Purified and His-tagged recombinant GFLOMT proteins were desalted on PD10 columns (GE Healthcare) in storage buffer (50 mm potassium phosphate [pH 7.5], 10% [v/v] glycerol, and 1 mm β-mercaptoethanol), and protein concentrations were determined using the Bradford assay (Bio-Rad) using bovine serum albumin as the standard. The purity of recombinant proteins was evaluated by SDS-PAGE.

Enzyme Assays and Characterization of GFLOMTs

The standard enzyme assay for OMT activity was performed at 37°C in 50 μL of 50 mm potassium phosphate (pH 7.0), 25 mm sodium ascorbate, 100 µm SAM, 100 µm potential alkaloid substrate, and 5 µg of the purified recombinant enzyme. Reactions were incubated for 2 h and stopped by adding 500 μL of methanol. Proteins were precipitated by centrifugation at 20,000g for 20 min, and supernatants were subjected to LC-MS/MS. Controls were performed with denatured purified His-tagged proteins prepared by boiling in water for 20 min. Conversion rates were calculated based on substrate loss. Kinetic parameters were determined at 37°C for 30 min by (1) varying alkaloid substrate concentrations from 1 to 300 µm at a fixed concentration of 100 µm SAM and (2) varying SAM concentrations from 1 to 300 µm at a fixed concentration of 100 µm for each alkaloid substrate. Kinetic constants were determined by fitting initial velocity versus substrate concentration to the Michaelis-Menten equation using GraphPad Prism 5 (www.graphpad.com).

LC-MS/MS Analysis of Enzyme Assays

Enzyme assays were analyzed using a 6410 Triple Quadrupole LC-MS/MS (Agilent Technologies) for the identification and quantification of alkaloids. LC was carried out using a Poroshell 120 SB C18 column (2.1 × 50 mm, 2.7-μm particle size; Agilent Technologies) at a flow rate of 0.7 mL min−1. LC was initiated at 100% solvent A (1% [v/v] formic acid), ramped to 60% (v/v) solvent B (acetonitrile) using a linear gradient over 6 min, further ramped to 99% (v/v) solvent B using a linear gradient over 1 min, held constant at 99% (v/v) solvent B for 1 min, and returned to original conditions over 0.1 min for a 3.9-min equilibration period. Eluate was applied to the mass analyzer using an electrospray ionization probe operating in positive mode with the following conditions: capillary voltage, 4,000 V; fragmentor voltage, 100 V; source temperature, 350°C; nebulizer pressure, 50 pounds per square inch; gas flow, 10 L min−1. For full-scan analysis, quadrupole 1 and 2 were set to radio frequency only, whereas the third quadrupole scanned from 200 to 700 m/z. ESI[+]-CID spectra were analyzed, the precursor m/z was selected in quadrupole 1, and a collision energy of 30 eV and an argon collision gas pressure of 1.8 × 10−3 torr were applied in quadrupole 2. The resulting ion fragments were resolved in quadrupole 3 scanning from 40 m/z to 2 m/z greater than the precursor ion m/z. Compounds were identified based on retention times and ESI[+]-CID spectra compared with authentic standards or compared with previously published spectral data (Desgagné-Penix et al., 2012; Winzer et al., 2012).

Bioconversions

IPTG-induced E. coli cultures harboring pGFLOMT1, pGFLOMT2, pGFLOMT6, and/or pCNMT were combined at equal ratios, and 25 µm (R,S)-norlaudanosoline was added to 1 mL of the mixture. The mixed cultures were incubated overnight at 30°C and subsequently quenched with methanol. Cell debris was removed by centrifugation, and the supernatant was dried and extracted three times using ethyl acetate. Pooled extracts were dried, subsequently dissolved in 50 µL of 1% (v/v) formic acid, and subjected to LC-MS/MS analysis. IPTG-induced cultures expressing empty vector pRSETA were used as the negative control.

LTQ-Orbitrap Analysis of Plant Extracts

G. flavum organs from four individual plants were flash frozen and stored at −80°C. Frozen tissues were ground to fine powder using a TissueLyser II (Qiagen), extracted with 5 mL of methanol:chloroform (50:50), and centrifuged at 8,000g for 10 min. Liquid phases (aqueous and organic) were aspirated from insoluble debris and pooled. Insoluble material was extracted twice more with methanol:chloroform (50:50), all liquid phases were pooled, and the remaining debris was dried and weighed. Liquid phases were reduced to dryness under vacuum, resuspended in 1.5 mL of methanol:chloroform (50:50), and centrifuged at 10,000g for 10 min to remove particulates. Dilutions of 1:100 and 1:1,000 were prepared for (1) the acquisition of exact mass data and quantification of low-abundance alkaloids, and (2) the quantification of high-abundance alkaloids, respectively. Ten microliters of each sample was fractionated on a Zorbax C18 column (2.1 × 50 mm, 1.8 μm; Agilent) using an Accela HPLC system (Thermo Scientific) at a flow rate of 0.5 mL min−1 and a solvent (10 mm ammonium acetate, pH 4.5) gradient of 100% to 80% over 5 min, 80% to 50% over 3 min, 50% to 0% over 3 min, isocratic at 0% for 2 min, 0% to 100% over 0.1 min, and isocratic at 100% for 1.9 min. The second solvent was 100% acetonitrile. Total run time was 15 min, but data were collected for only 10 min. Heated ESI source and interface conditions were operated in positive ion mode as follows: vaporizer temperature, 400°C; source voltage, 3 kV; sheath gas, 60 au, auxiliary gas, 20 au; capillary temperature, 380°C; capillary voltage, 6 V; tube lens, 45 V. LTQ-Orbitrap-XL (Thermo Scientific) instrumentation was performed as three scan events in data-dependent, parallel detection mode. The first scan consisted of high-resolution FTMS from 200 to 700 m/z with ion injection time of 500 ms and scan time of approximately 1.5 s. The second and third scans (approximately 0.5 s each) collected CID spectra in the ion trap, where the parent ions represented the first- and second-most abundant alkaloid masses, respectively, as determined by fast Fourier transform preview using a parent ion mass list corresponding to exact masses of known alkaloids. Dynamic-exclusion and reject-ion-mass-list features were enabled. External and internal calibration procedures ensured <2 ppm. FTMS data were processed via R i386 v. 3.2.0 using the centWave, obiwarp, group, and fillPeaks features of Package XCMS v.1.44.0 (Tautenhahn et al., 2008). Exact mass, retention times, and CID spectra of authentic standards were used to identify alkaloids, and quantification was performed using standard curves.

Gene Expression Analysis

After denaturing the RNA-primer mix for 15 min at 70°C, cDNA synthesis was performed at 37°C for 50 min using 2.5 µm anchored oligo(dT) primer (dT20VN), 0.5 mm deoxynucleotide triphosphate mix, 500 ng G. flavum RNA, and 5 microunits µL−1 reverse transcriptase (Invitrogen). Quantitative reverse transcription-PCR was performed using SYBR Green detection on an Applied Biosystems 7300 real-time PCR system. Each 10-µL reaction included 1 µL of synthesized cDNA, 300 nm forward and reverse primers (Supplemental Table S6), and 1× Power SYBR Green PCR Master Mix (Applied Biosystems). Thermal cycling conditions for relative quantification included 40 cycles of template denaturation, primer annealing, and primer extension. To evaluate PCR specificity, the amplified products of all primer pairs were subjected to melt curve analysis using the dissociation method suggested by Applied Biosystems. The Inline graphic method was used for the analysis of relative gene expression (Livak and Schmittgen, 2001) with β-actin as the internal control. Values were calculated based on three biological replicates for each GFLOMT gene.

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers KP176693 to KP176699.

Supplemental Data

The following supplemental materials are available.

  • Supplemental Figure S1. Purification of recombinant proteins from E. coli expressing pGFLOMT constructs.

  • Supplemental Figure S2. Steady-state enzyme kinetics of affinity-purified recombinant GFLOMT1, GFLOMT2, and GFLOMT6 for (R,S)-6-O-methylnorlaudanosoline, (R,S)-norlaudanosoline, (R,S)-scoulerine, and SAM.

  • Supplemental Figure S3. Possible permutations in the sequence of O- and N-methyl transfer reactions and the detected functions of GFLOMT1, GFLOMT2, and GFLOMT6 in the conversion of (R,S)-norlaudanosoline to (R,S)-laudanosine.

  • Supplemental Table S1. Amino acid sequence identity matrix for GFLOMTs with functionally characterized OMTs involved in BIA biosynthesis.

  • Supplemental Table S2. Chromatographic and mass spectral data for authentic BIA standards.

  • Supplemental Table S3. Chromatographic and mass spectral data for the reaction products of recombinant GFLOMTs assayed with various substrates.

  • Supplemental Table S4. Chromatographic and mass spectral data for compounds produced from various combinations of E. coli strains harboring individual pGFLOMT constructs, plus or minus a strain producing pCNMT from C. japonica, and fed (R,S)-norlaudanosoline.

  • Supplemental Table S5. Primers used to amplify open reading frames of GFLOMTs for insertion into pRSETA expression vector.

  • Supplemental Table S6. Primers used to amplify open reading frames of GFLOMTs for insertion into pRSETA expression vector.

  • Supplemental Data Set S1. Exact mass data obtained by high-resolution Fourier-transform mass spectrometry performed in positive mode on extracts from different organs of G. flavum.

Supplementary Material

Supplemental Data
supp_169_2_1127__index.html (1.1KB, html)

Acknowledgments

We thank Scott Farrow for assistance with the preparation of norreticuline.

Glossary

BIA

benzylisoquinoline alkaloid

SAM

S-adenosyl-l-Met

cDNA

complementary DNA

LC

liquid chromatography

MS/MS

tandem mass spectrometry

CID

collision-induced dissociation

m/z

mass-to-charge ratio

FTMS

Fourier-transform mass spectrometry

IPTG

isopropyl β-d-thiogalactopyranoside

Footnotes

1

This work was supported by Genome Canada, Genome Alberta, the Government of Alberta, the Canada Foundation for Innovation, and the Natural Sciences and Engineering Research Council of Canada (to P.J.F.).

References

  1. Battersby AR, McHugh JL, Staunton J, Todd M (1971) Biosynthesis of the apparently “directly coupled” aporphine alkaloids. J Chem Soc Chem Commun 1971: 985–986 [Google Scholar]
  2. Bhakuni DS, Jain S (1988) The biosynthesis of glaucine in Litsea glutinosa. J Chem Soc Perkin Trans 1: 1447–1449 [Google Scholar]
  3. Bournine L, Bensalem S, Wauters JN, Iguer-Ouada M, Maiza-Benabdesselam F, Bedjou F, Castronovo V, Bellahcène A, Tits M, Frédérich M (2013) Identification and quantification of the main active anticancer alkaloids from the root of Glaucium flavum. Int J Mol Sci 14: 23533–23544 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Choi KB, Morishige T, Shitan N, Yazaki K, Sato F (2002) Molecular cloning and characterization of coclaurine N-methyltransferase from cultured cells of Coptis japonica. J Biol Chem 277: 830–835 [DOI] [PubMed] [Google Scholar]
  5. Cortijo J, Villagrasa V, Pons R, Berto L, Martí-Cabrera M, Martinez-Losa M, Domenech T, Beleta J, Morcillo EJ (1999) Bronchodilator and anti-inflammatory activities of glaucine: In vitro studies in human airway smooth muscle and polymorphonuclear leukocytes. Br J Pharmacol 127: 1641–1651 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dang TTT, Chen X, Facchini PJ (2015) Acetylation serves as a protective group in noscapine biosynthesis in opium poppy. Nat Chem Biol 11: 104–106 [DOI] [PubMed] [Google Scholar]
  7. Dang TTT, Facchini PJ (2012) Characterization of three O-methyltransferases involved in noscapine biosynthesis in opium poppy. Plant Physiol 159: 618–631 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dargan PI, Button J, Hawkins L, Archer JR, Ovaska H, Lidder S, Ramsey J, Holt DW, Wood DM (2008) Detection of the pharmaceutical agent glaucine as a recreational drug. Eur J Clin Pharmacol 64: 553–554 [DOI] [PubMed] [Google Scholar]
  9. Daskalova E, Iskrenova E, Kiryakov HG, Evstatieva L (1988) Minor alkaloids of Glaucium flavum. Phytochemistry 27: 953–955 [Google Scholar]
  10. Desgagné-Penix I, Farrow SC, Cram D, Nowak J, Facchini PJ (2012) Integration of deep transcript and targeted metabolite profiles for eight cultivars of opium poppy. Plant Mol Biol 79: 295–313 [DOI] [PubMed] [Google Scholar]
  11. Farrow SC, Hagel JM, Facchini PJ (2012) Transcript and metabolite profiling in cell cultures of 18 plant species that produce benzylisoquinoline alkaloids. Phytochemistry 77: 79–88 [DOI] [PubMed] [Google Scholar]
  12. Fossati E, Ekins A, Narcross L, Zhu Y, Falgueyret JP, Beaudoin GA, Facchini PJ, Martin VJJ (2014) Reconstitution of a 10-gene pathway for synthesis of the plant alkaloid dihydrosanguinarine in Saccharomyces cerevisiae. Nat Commun 5: 3283. [DOI] [PubMed] [Google Scholar]
  13. Frenzel T, Zenk MH (1990) S-adenosyl-L-methionine:3′hydroxyl-N-methyl-(S)-coclaurine-4′-O-methyltransferase, a regio- and stereoselective enzyme of the (S)-reticuline pathway. Phytochemistry 29: 3505–3511 [Google Scholar]
  14. Gambino G, Perrone I, Gribaudo I (2008) A rapid and effective method for RNA extraction from different tissues of grapevine and other woody plants. Phytochem Anal 19: 520–525 [DOI] [PubMed] [Google Scholar]
  15. Hagel JM, Facchini PJ (2010) Dioxygenases catalyze the O-demethylation steps of morphine biosynthesis in opium poppy. Nat Chem Biol 6: 273–275 [DOI] [PubMed] [Google Scholar]
  16. Hagel JM, Facchini PJ (2013) Benzylisoquinoline alkaloid metabolism: a century of discovery and a brave new world. Plant Cell Physiol 54: 647–672 [DOI] [PubMed] [Google Scholar]
  17. Hawkins KM, Smolke CD (2008) Production of benzylisoquinoline alkaloids in Saccharomyces cerevisiae. Nat Chem Biol 4: 564–573 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ikezawa N, Iwasa K, Sato F (2008) Molecular cloning and characterization of CYP80G2, a cytochrome P450 that catalyzes an intramolecular C-C phenol coupling of (S)-reticuline in magnoflorine biosynthesis, from cultured Coptis japonica cells. J Biol Chem 283: 8810–8821 [DOI] [PubMed] [Google Scholar]
  19. Inui T, Tamura K, Fujii N, Morishige T, Sato F (2007) Overexpression of Coptis japonica norcoclaurine 6-O-methyltransferase overcomes the rate-limiting step in benzylisoquinoline alkaloid biosynthesis in cultured Eschscholzia californica. Plant Cell Physiol 48: 252–262 [DOI] [PubMed] [Google Scholar]
  20. Kintsurashvili LG, Vachnadze VY (2000) Alkaloids of Glaucium corniculatum and G. flavum growing in Georgia. Chem Nat Compd 36: 225–226 [Google Scholar]
  21. Lapa GB, Sheichenko OP, Serezhechkin AG, Tolkachev ON (2004) HPLC determination of glaucine in yellow horn poppy grass (Glaucium flavum Crantz). Pharm Chem J 38: 441–442 [Google Scholar]
  22. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947–2948 [DOI] [PubMed] [Google Scholar]
  23. Liscombe DK, Facchini PJ (2007) Molecular cloning and characterization of tetrahydroprotoberberine cis-N-methyltransferase, an enzyme involved in alkaloid biosynthesis in opium poppy. J Biol Chem 282: 14741–14751 [DOI] [PubMed] [Google Scholar]
  24. Liscombe DK, MacLeod BP, Loukanina N, Nandi OI, Facchini PJ (2005) Evidence for the monophyletic evolution of benzylisoquinoline alkaloid biosynthesis in angiosperms. Phytochemistry 66: 2501–2520 [DOI] [PubMed] [Google Scholar]
  25. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCt method. Methods 25: 402–408 [DOI] [PubMed] [Google Scholar]
  26. Milanowski DJ, Winter RE, Elvin-Lewis MP, Lewis WH (2002) Geographic distribution of three alkaloid chemotypes of Croton lechleri. J Nat Prod 65: 814–819 [DOI] [PubMed] [Google Scholar]
  27. Morishige T, Dubouzet E, Choi KB, Yazaki K, Sato F (2002) Molecular cloning of columbamine O-methyltransferase from cultured Coptis japonica cells. Eur J Biochem 269: 5659–5667 [DOI] [PubMed] [Google Scholar]
  28. Morishige T, Tsujita T, Yamada Y, Sato F (2000) Molecular characterization of the S-adenosyl-L-methionine:3′-hydroxy-N-methylcoclaurine 4′-O-methyltransferase involved in isoquinoline alkaloid biosynthesis in Coptis japonica. J Biol Chem 275: 23398–23405 [DOI] [PubMed] [Google Scholar]
  29. Morteza-Semnani K, Amin G, Shidfar MR, Hadizadeh H, Shafiee A (2003) Antifungal activity of the methanolic extract and alkaloids of Glaucium oxylobum. Fitoterapia 74: 493–496 [DOI] [PubMed] [Google Scholar]
  30. Muemmler S, Rueffer M, Nagakura N, Zenk MH (1985) S-adenosyl-L-methionine: (S)-scoulerine 9-O-methyltransferase, a highly stereo- and regio-specific enzyme in tetrahydroprotoberberine biosynthesis. Plant Cell Rep 4: 36–39 [DOI] [PubMed] [Google Scholar]
  31. Nakagawa A, Minami H, Kim JS, Koyanagi T, Katayama T, Sato F, Kumagai H (2011) A bacterial platform for fermentative production of plant alkaloids. Nat Commun 2: 326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Orallo F, Fernández Alzueta A, Campos-Toimil M, Calleja JM (1995) Study of the in vivo and in vitro cardiovascular effects of (+)-glaucine and N-carbethoxysecoglaucine in rats. Br J Pharmacol 114: 1419–1427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Ounaroon A, Decker G, Schmidt J, Lottspeich F, Kutchan TM (2003) (R,S)-Reticuline 7-O-methyltransferase and (R,S)-norcoclaurine 6-O-methyltransferase of Papaver somniferum - cDNA cloning and characterization of methyl transfer enzymes of alkaloid biosynthesis in opium poppy. Plant J 36: 808–819 [DOI] [PubMed] [Google Scholar]
  34. Peled B, Waisel Y, Carmeli S (1988) Alkaloid content in various chemoecotypes of Glaucium flavum from Israel. Phytochemistry 27: 1021–1024 [Google Scholar]
  35. Petitto V, Serafini M, Gallo FR, Multari G, Nicoletti M (2010) Alkaloids from Glaucium flavum from Sardinia. Nat Prod Res 24: 1033–1035 [DOI] [PubMed] [Google Scholar]
  36. Pienkny S, Brandt W, Schmidt J, Kramell R, Ziegler J (2009) Functional characterization of a novel benzylisoquinoline O-methyltransferase suggests its involvement in papaverine biosynthesis in opium poppy (Papaver somniferum L). Plant J 60: 56–67 [DOI] [PubMed] [Google Scholar]
  37. Rueffer M, Nagakura N, Zenk MH (1983) A highly specific O-methyltransferase for nororientaline synthesis isolated from Argemone platyceras cell cultures. Planta Med 49: 196–198 [DOI] [PubMed] [Google Scholar]
  38. Sato F, Takeshita N, Fitchen JH, Fujiwara H, Yamada Y (1993) S-adenosyl-L-methionine:scoulerine-9-O-methyltransferase from cultured Coptis japonica cells. Phytochemistry 32: 659–664 [PubMed] [Google Scholar]
  39. Takeshita N, Fujiwara H, Mimura H, Fitchen JH, Yamada Y, Sato F (1995) Molecular cloning and characterization of S-adenosyl-L-methionine:scoulerine-9-O-methyltransferase from cultured cells of Coptis japonica. Plant Cell Physiol 36: 29–36 [PubMed] [Google Scholar]
  40. Tautenhahn R, Böttcher C, Neumann S (2008) Highly sensitive feature detection for high resolution LC/MS. BMC Bioinformatics 9: 504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Winzer T, Gazda V, He Z, Kaminski F, Kern M, Larson TR, Li Y, Meade F, Teodor R, Vaistij FE, et al. (2012) A Papaver somniferum 10-gene cluster for synthesis of the anticancer alkaloid noscapine. Science 336: 1704–1708 [DOI] [PubMed] [Google Scholar]
  42. Xiao M, Zhang Y, Chen X, Lee EJ, Barber CJ, Chakrabarty R, Desgagné-Penix I, Haslam TM, Kim YB, Liu E, et al. (2013) Transcriptome analysis based on next-generation sequencing of non-model plants producing specialized metabolites of biotechnological interest. J Biotechnol 166: 122–134 [DOI] [PubMed] [Google Scholar]
  43. Xu XH, Yu GD, Wang ZT (2004) Resource investigation and quality evaluation on wild Corydalis yanhusuo. Zhongguo Zhong Yao Za Zhi 29: 399–401 [PubMed] [Google Scholar]
  44. Zetler G. (1988) Neuroleptic-like, anticonvulsant and antinociceptive effects of aporphine alkaloids: bulbocapnine, corytuberine, boldine and glaucine. Arch Int Pharmacodyn Ther 296: 255–281 [PubMed] [Google Scholar]
  45. Ziegler J, Diaz-Chávez ML, Kramell R, Ammer C, Kutchan TM (2005) Comparative macroarray analysis of morphine containing Papaver somniferum and eight morphine free Papaver species identifies an O-methyltransferase involved in benzylisoquinoline biosynthesis. Planta 222: 458–471 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data
supp_169_2_1127__index.html (1.1KB, html)
supp_pp.15.01240_PP2015-01240_Supplemental_Material.pdf (2.4MB, pdf)
supp_pp.15.01240_1240Supplemental_Data_Set_1.xlsx (148.9KB, xlsx)

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES