Article pubs.acs.org/jnp
Glycine-Containing Flaxseed Orbitides Peta-Gaye G. Burnett,† Pramodkumar D. Jadhav,*,† Denis P. Okinyo-Owiti,† Aaron G. Poth,‡ and Martin J. T. Reaney*,†,§ †
Department of Plant Sciences, College of Agriculture and Bioresources, University of Saskatchewan, Saskatoon, SK S7N 5A8, Canada ‡ Institute for Molecular Bioscience, University of Queensland, Brisbane, Queensland 4072, Australia § Guangdong-Saskatchewan Oilseed Joint Laboratory, Department of Food Science and Engineering, Jinan University, Guangzhou, Guangdong 510632, People’s Republic of China S Supporting Information *
ABSTRACT: Five new orbitides, cyclolinopeptides 21−25, were identified in flaxseed oil (Linum usitatissimum) extracts. Their HPLC-ESIMS quasimolecular ion peaks at m/z 1097.7 (21), 1115.6 (22), 1131.6 (23), 1018.6 (24), and 1034.6 (25) [M + H] + corresponded to the molecular formulae C59H89N10O10, C58H87N10O10S, C58H87N10O11S, C53H80N9O9S, and C53H80N9O10S, respectively. Their structures were elucidated by extensive HPLC-ESIMS/MS analyses, and their presence was confirmed by precursor proteins identified in flax genomic DNA sequence data. The amino acid sequences of these orbitides were confirmed as [1−10NαC]-GILVPPFFLI, [1−10-NαC]-GMLIPPFFVI, [1−10NαC]-GOLIPPFFVI, [1−9-NαC]-GMLVFPLFI, and [1−9NαC]-GOLVFPLFI for cyclolinopeptides 21−25, respectively. Previously reported orbitides, [1−9-NαC]-ILVPPFFLI (1), [1−9-NαC]-MLIPPFFVI (2), [1−9-NαC]-OLIPPFFVI (3), [1−8-NαC]-MLVFPLFI (7), and [1−8-NαC]-OLVFPLFI (8), were also present in flaxseed oil. The precursors of orbitides 21, 22, and 24 also produced orbitides 1, 2, and 7 by alternative cyclization. Cyclolinopeptides 3, 8, 23, and 25 contain MetO (O) and are not directly encoded, but are products of posttranslational modification of the Met present in 2, 7, 22, and 24, respectively. Sufficient cyclolinopeptide 23 was isolated for characterization via 1D (1H and 13C) and 2D (NOESY and HMBC) NMR spectroscopy. These compounds have been named as cyclolinopeptides U, V, W, X, and Y for 21, 22, 23, 24, and 25, respectively.
O
methionine can be oxidized to the methionine S-oxide (MetO), which can be further oxidized to the methionine S,S-dioxide (MetO2). MetO may be formed from Met by exposure to reactive oxygen species and by the action of enzymes. However, in fresh flaxseed the levels of MetO are low, but orbitides with MetO accumulate in flaxseed oil with storage.13 It is not known if any oxidation of Met to MetO in flaxseed orbitides is the result of enzymatic processes. Although flax orbitides exhibit a wide range of biological activity, their inherent biological role(s) is unknown.14 In gene sequences that encode orbitides, cyclotides, and other cyclic peptides, the core peptide sequences, whether single or multiple copies, are appended by an amino terminal recognition sequence (leader peptide) and a carboxy terminal recognition sequence.1 These recognition sequences are assumed to be vital for processing reactions such as posttranslational modification, excision, and cyclization.15 The precursor proteins from which cyclotides are derived also
rbitides are plant-derived cyclic peptides that are formed from the cyclization of linear precursor peptides by linkage of the amino and carboxy termini through a peptide bond.1 The molecular weights of these ribosomally synthesized and post-translationally modified N-to-C linked peptides, RiPPs, are less than 10 kDa. Hence, orbitides and cyclotides are both families of homodetic plant cyclic peptides arising from ribosomal synthesis, in which members of the former do not contain disulfide bonds; that is, they lack internal cysteine bonds. Flax (Linum usitatissimum L.) is a member of the plant family Linaceae and is known to contain biologically active hydrophobic orbitides or cyclolinopeptides (CLs) in the seeds, oil, and roots (Figure 1).1−3 The 20 reported flax orbitides are produced by the post-translational modification of eight CL (1, 2, 5, 7, 10, 14, 18, and 20) precursor sequences.4−12 These sequences encode peptides that contain just eight or nine amino acid residues. The eight orbitides are structurally related to the aforementioned 20 CLs (excluding 1 and 20) by a common amino acid sequence with variation caused by the oxidation state of the methionine residue in that the © XXXX American Chemical Society and American Society of Pharmacognosy
Received: October 28, 2014
A
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 1. Structures of known flaxseed orbitides.
These authors noted also that orbitide biosynthesis in the Caryophyllaceae family shared some common features with that of kalata, PawS, and cyclic knottin types.21 Particularly, there is a tendency for Gly to occur at the amino terminus of the incipient orbitide sequences of the Caryophyllaceae,19 kalata, PawS, and cyclic knottin types.22 To date, this tendency has not been observed in flax, in which only one of the eight encoded orbitides possesses Gly at the amino terminus (20), another possesses Ile (1), and the remaining six possess Met (2, 5, 7, 10, 14, 18).2,11 The possibility exists that flax orbitides possess a different mode(s) for biosynthetic processing such that the presence of Gly at the amino terminal flank of the precursor peptide is not vital and/or that there are numerous undiscovered flax orbitides that contain Gly at the amino termini of the incipient orbitide sequences. There are only two methods for characterizing orbitides: either (i) prediction via DNA and/or mRNA sequence analyses, followed by observation of the orbitide in mass spectrometry analysis, or (ii) detecting a cyclic peptide by mass spectrometry analysis, followed by the discovery of DNA and or mRNA sequences in a plant that direct its synthesis.11 Through HPLC-ESIMS and HPLC-ESIMS/MS analyses of flaxseed oil extracts, we identified and structurally elucidated five novel orbitide sequences. Additionally, genes were identified that encode for ribosomal biosynthesis of the linear precursors of these orbitides. This investigation also discusses NMR characterization of one of the five novel orbitides.
contain an endoplasmic reticulum signal sequence at the amino terminus of the leader peptide, which may indicate that cyclotides are directed to the endoplasmic reticulum.1 Although many leader peptides of cyclotides tend to form α-helices16 and this folding probably occurs in the endoplasmic reticulum, processing at the amino terminal flank remains undeciphered. On the other hand, many cyclotides possess either an Asn or Asp moiety at the carboxy terminus of the core peptide that has been shown to be cleaved and cyclized by an asparaginyl endoprotease.17,18 The genes encoding orbitides, like cyclotides, comprise a leader peptide, core peptide, and a carboxy terminal recognition sequence. Although common in cyclotides, acidic residues (Asp, Glu) and their amides (Asn, Gln) are among some of the rare amino acids observed in orbitides, and as such, processing by asparaginyl endoproteases is unsupported. Condie et al. showed that orbitide precursor sequences from Saponaria vaccaria exhibit high conservation (i.e., display only minor variation) in the amino and carboxy flanking regions.19 Such a degree of conservation likely indicates that sequences that flank the core peptide sequence play a role in recognition by biosynthetic enzymes.20 Until recently, the actual details of orbitide biosynthesis from linear precursor peptides were unknown. In 2013, Barber et al. discovered via enzyme purification and mass spectrometry that presegetalins in S. vaccaria are processed into mature segetalins in two steps, with each step employing a different enzyme.21 Specifically, the first step involves an oligopeptidase that cleaves at the carboxy terminus of the leader signal, whereas the second step involves a serine protease-like enzyme that cleaves at the carboxy terminus of the precursor peptide and subsequently cyclizes the incipient orbitide sequence.
■
RESULTS AND DISCUSSION Flax orbitides were extracted from flaxseed oil by adsorption onto silica gel, followed by recovery through elution with more polar organic solvents. HPLC-DAD chromatograms of the B
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Table 1. Novel Flaxseed Orbitides mass ([M + H]+, Da) expt
calcd
delta mass (ppm)
ret. time (tR, min)
molecular formula
amino acid sequencea
symbol
proposed new name
1097.6746 1115.6307 1131.6229 1018.5873 1034.5787
1097.6758 1115.6322 1131.6271 1018.5794 1034.5743
−1.1 −1.3 −3.7 7.7 4.3
6.3 5.4 1.6 6.2 2.4
C59H89N10O10 C58H87N10O10S C58H87N10O11S C53H80N9O9S C53H80N9O10S
[1−10-NαC]-GILVPPFFLI [1−10-NαC]-GMLIPPFFVI [1−10-NαC]-GObLIPPFFVI [1−9-NαC]-GMLVFPLFI [1−9-NαC]-GOLVFPLFI
U (21) V (22) W (23) X (24) Y (25)
[1−10-NαC]-CLU [1−10-NαC]-CLV [1−10-NαC],[2-MetOc]-CLV [1−9-NαC]-CLX [1−9-NαC],[2-MetO]-CLX
[1−#-NαC], N to C linkage occurs through the α-amino group between amino acid 1 and amino acid #. bAmino acid symbol O used for methionine S-oxide. cDesignation MetO describes methionine S-oxide.
a
Figure 2. Structures of novel flaxseed orbitides.
Figure 3. Product ion spectra of ions derived from [1−10-NαC]-CLU (21, m/z 1097.6746) at tR 6.3 min (A) and [1−10-NαC],[2-MetO]-CLV (23, m/z 1131.6229) at tR 1.6 min (B).
1). HPLC-ESIMS/MS analyses of the crude extract revealed fragmentation patterns that were similar to those of known flax orbitides.9,11 Building the fragments led to identification of the new orbitides shown in Figure 2. The HPLC-ESIMS/MS showed ring opening at the amide nitrogen of proline followed by type b fragmentation. HRHPLC-ESIMS ([M + H]+) analysis of cyclolinopeptide 21, tR = 6.3 min, showed a quasimolecular ion at m/z 1097.6746. The HPLC-ESIMS/MS of 21 displayed loss of a neutral Val residue as the first amino fragment followed by Leu/Ile (Figure 3a). The next cleavage product was a loss of 170.11 Da, which could be attributed to an amino acid residue pair of either “Gly and Leu/Ile” or “Ala and Val”. Upon closer observation, a less abundant fragment at m/z 772.44 corroborated the first inference, suggesting cleavage of a Leu/Ile residue followed by that of Gly. Further fragmentation of 21 showed sequential
crude extract showed numerous peaks (data not shown), many of which had close retention times and overlapped, both of which limit unequivocal identification of compounds. As such, the crude extract was subjected to HPLC-ESIMS analyses. The crude extract contained known cyclolinopeptides 1−3, 5−8, 10−17, and 20 (Figure 1).4−7,9−12 Extracted ion chromatograms (data not shown) revealed multiple peaks of 11 and 16. This phenomenon was observed by Lao and co-workers.12 The absence of known flax orbitides 4 and 9 suggested that the flaxseed oil had not yet undergone extensive oxidation.8,12,23,24 The remaining known flax orbitides 18 and 19 that were recently identified were not detected in this extract probably due to either their occurrence at low concentration and/or the extraction procedure.11 In addition to the 16 known peptides detected, signals for novel cyclolinopeptides 21, 22, 23, 24, and 25 were also observed (Table C
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 4. Product ion spectra of ions derived from [1−9-NαC]-CLX (24, m/z 1018.5794) at tR 6.2 min (A) and [1−9-NαC],[2-MetO]-CLX (25, m/z 1034.5743) at tR 2.4 min (B).
Ile-Met-Gly-Leu/Ile-Phe to a dipeptide fragment of m/z 211.14, which can only be attributed to [Leu/Ile+Pro]. On the other hand, the MS/MS of cyclolinopeptide 25 was more complicated in that the third and fourth cleavage products were 111.08 Da and 206.07 Da smaller than the m/z 788.4, respectively, which did not correspond to any one or combination of amino acid residues (Figure 4b). Collectively, the sum of both cleavage events (317.14 Da) corresponded with a tripeptide of [Gly+Leu/Ile+MetO] or [Cys+Leu/Ile +Thr]. As a result of the 15.9914 amu difference between 24 and 25 (observed at the fourth major cleavage product), it was inferred that cyclolinopeptide 25 is derived from 24 by means of Met oxidation to MetO. On the basis of these assumptions and in support of the observed mass fragmentation pattern, the proposed mass fragmentation pattern of 25 is as follows: PheVal-Leu/Ile-MetO-Gly-Leu/Ile-Phe-[Leu/Ile+Pro]. Such a hypothesis would imply also that if cyclolinopeptide 23 contained MetO, then the Met-containing peptide should exist. In order to test this hypothesis, an enriched fraction containing cyclolinopeptides 23 and 25 was treated with NaBH4 and I2. The reduced product yielded cyclolinopeptides 22 (m/z 1115.6307 [M + H]+, tR = 5.4 min) and 24 among other known orbitides. The fragmentation pattern of 22 (Figure 5) resembled that of 23 with the exception of the
loss of individual amino acid residues in the following order: Val-Leu/Ile-Leu/Ile-Gly-Leu/Ile-Leu/Ile-Phe to m/z 342.1823. The final ion observed may be a tripeptide fragment arising from 11 possible combinations of amino acid residues not listed here. However, by eliminating Asp, Asn, Glu, and Gln because of their infrequency in orbitides, the possible options are reduced to four: [Cys+His+Thr], [Gly+His+Phe/MetO], [Leu/Ile+Met+Pro], or [Phe/MetO+Pro+Pro]. Note that combinations of amino acid residues are listed in alphabetical order within square brackets when the fragmentation order cannot be elucidated. The accuracy and precision of a time-offlight mass spectrometer readily enables the differentiation of close masses such that we are able to state that the fragment at m/z 342.1823 can result only from [Phe+Pro+Pro] and the delta mass of 21 is −1.1 ppm (Table 1). The tripeptide fragment [Phe+Pro+Pro] mass was also confirmed using an online mass calculator with an accuracy of 0.002 ppm as compared to other amino acid combinations.25 Cyclolinopeptide 21 is the first flaxseed orbitide reported to contain 10 amino acid residues, the second one to contain Gly, and the third one lacking a methionine residue, with others being 1 and 20.11 Cyclolinopeptide 23 (m/z 1131.6229 [M + H]+), at tR = 1.6 min, showed sequential losses of two Leu/Ile residues or a combination thereof (Figure 3b). Subsequent fragmentation at m/z 701.4 could be attributed to either “Gly and MetO” or “Cys and Thr”. However, a less abundant fragment at m/z 758.44 supported the presence of a Gly residue as the fourth amino acid residue cleaved. Thereafter, sequential fragmentation indicated loss of individual amino acid residues in the order Leu/Ile-Val-Phe to a tripeptide fragment of m/z 342.18 as previously observed for 21. The HPLC-ESIMS/MS of cyclolinopeptides 24 (m/z 1018.5873 [M + H]+, tR = 6.2 min) and 25 (m/z 1034.5787 [M + H]+, tR = 2.4 min) showed similar fragmentation patterns with the exception of the fourth major cleavage product (Figure 4a and b). Note that although a quasimolecular ion corresponding to cyclolinopeptide 25 was detected via HPLC-ESIMS, an MS/MS spectrum was attainable only after enrichment of that compound. The fourth major cleavage product observed in these spectra indicated respective losses of 188.07 Da in 24 and 204.06 Da in 25. A loss of 188.07 Da may be attributed to “Gly and Met” or “Ser and Thr”. As with the previously discussed data, a less abundant daughter ion at m/z 528.33 supported loss of a [Gly-Met] fragment and that Met is cleaved prior to Gly. By accounting for all the daughter ions, 24 undergoes mass fragmentation in the sequence Phe-Val-Leu/
Figure 5. Product ion spectra of ions derived from [1−10-NαC]-CLV (22, m/z 1115.6307) at tR 5.4 min.
fourth cleavage product, which in this case can be attributed only to Met, hence indicating that 23 arose from modification of the Met present in 22. Hence, the amino acid residues of 22 are sequentially lost in the order Leu/Ile-Leu/Ile-Met-Gly-Leu/ Ile-Val-Phe to a tripeptide fragment of m/z 342.18. High-resolution HPLC-ESIMS and HPLC-ESIMS/MS reveal the molecular weight and formulas of quasimolecular ion signals and their sequential loss of amino acid residues in a mass spectrometer. However, with a cyclic peptide, only genetic D
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 6. Probable translation of GenBank accession AFSQ01016651.1 (positive strand sequence from residue 7078 to 7692) showing one copy each of embedded putative linear protein sequences GMLVFPLFI, GMLIPPFFVI, and GILVPPFFLI for novel flaxseed orbitides 24 ([1−9-NαC]CLX), 22 ([1−10-NαC]-CLV), and 21 ([1−10-NαC]-CLU), respectively.
polyadenylation motif (base 7891 to base 7896), and a promoter (base 6981 to base 7020).29,30 No introns were observed within this GenBank contig. A further search of the flax genome database for the putative precursors of 22 and 24 indicated that they are also encoded by the gene that encodes 21 (Figure 6). One copy of the decapeptide GMLIPPFFVI and the nonapeptide GMLVFPLFI are encoded as linear precursor peptides of 22 ([1−10-NαC]-GMLIPPFFVI) and 24 ([1−9NαC]-GMLVFPLFI), respectively. Cyclolinopeptides 23 ([1− 10-NαC]-GOLIPPFFVI) and 25 ([1−9-NαC]-GOLVFPLFI) are not directly encoded in the DNA sequence, but are derived via post-translational modification of 22 and 24 via Met oxidation to MetO. It is interesting that the gene predicted by GENSCAN 1.0 (AFSQ01016651.1) containing embedded linear precursors for cyclolinopeptides 21, 22, and 24 is the same gene containing those for cyclolinopeptides 1, 2, and 7.2 More intriguing is the fact that cyclolinopeptides 21, 22, and 24 are glycine-containing analogues of 1, 2, and 7 in which the former group possess Gly at the amino termini of their linear precursors. The putative precursor peptides of 21, 22, and 24 are flanked by a conserved “SD” and “FGK” at the amino and carboxy termini, respectively. In contrast, 1, 2, and 7 possess the same carboxy termini flank (“FGK”), but differ in their amino termini flanks of “DG”.2 Thus, the Gly residue of the amino flanking sequence, “DG”, previously reported for 1, 2, and 7 has now been incorporated in the linear precursor peptides of 21, 22, and 24, causing the amino flanking sequence of the glycinecontaining analogues to be “SD”. Cyclolinopeptide 23 was isolated from extract h, previously enriched in this orbitide (see Experimental Section). NMR studies were conducted at 298 K in acetone-d6 to obtain wellresolved spectra (Figures S1−S7, Supporting Information). 1 H-1H COSY, TOCSY, and NOESY experiments were used for sequential assignment of the α, β, γ, and δ protons. In addition, HMQC was performed to assign carbon atoms attached to protons. The protons attached to heteroatoms were assigned from 1H NMR data, and their coupling to amide protons and carbonyl carbons was determined by NOESY and HMBC correlations. The NMR spectroscopic data of 23 (Figure 7, Table 2) showed remarkable similarities to those of 3.6 1H NMR data of 23 displayed most resonances in 3, while 13C NMR data showed 10 amide carbonyl signals (δ 172.56, 172.36, 172.21, 172.10 (2), 170.88, 170.69, 170.07 (2), 169.70) indicating a decapeptide. The sharp proton singlet at δ 2.64 (signal at δ 2.45 for 3 in DMSO) and carbon signal at δ 39.04 (signal at δ 37.6 for 3 in DMSO) in the 1H NMR and 13C NMR spectra of 23, respectively, were thus attributed to the
information is able to determine its linear precursor (core) peptide and, hence, distinguish between Leu/Ile residues and the location of cyclization, i.e., the amino and carboxy termini of the core peptide. Previously, Covello et al. demonstrated that cyclic peptides in the Caryophyllaceae, Linaceae, and Rutaceae families occurred via post-translational modification of mRNAencoded linear peptides.20 Genomic data have already been utilized to unambiguously establish the amino acid sequences of flax orbitides 1, 2, 5, 7, 10, 14, 18, and 20, with the other orbitides resulting from post-translational modifications.2,11 Barber et al. reported the tendency of Gly to appear at the amino termini of incipient orbitide sequences and the lack of such a trend in flax orbitides that instead demonstrate a propensity for incorporation of Met or Ile at the amino terminus of the core peptide.21 However, the possibility exists that whenever a Gly residue is not present, the precursor sequence has a tendency to start with Met or Ile, where present, at the amino terminus. Therefore, based on (i) the fragmentation patterns of the novel flax orbitides; (ii) the homologies between 21 and 1, 22 and 2, 23 and 3, 24 and 7, and 25 and 8; and (iii) the preference for incorporation of Gly, when present, as the first residue in cyclic peptides, the proposed sequences are as follows: 21 as [1−10-NαC]-Gly-Ile-Leu-Val-Pro-Pro-Phe-PheLeu-Ile; 22 as [1−10-NαC]-Gly-Met-Leu-Ile-Pro-Pro-Phe-PheVal-Ile; 23 as [1−10-NαC]-Gly-MetO-Leu-Ile-Pro-Pro-PhePhe-Val-Ile; 24 as [1−9-NαC]-Gly-Met-Leu-Val-Phe-Pro-LeuPhe-Ile; and 25 as [1−9-NαC]-Gly-MetO-Leu-Val-Phe-ProLeu-Phe-Ile. Using this information, the National Center for Biotechnology Information (NCBI) GenBank was searched for genome sequences of all possible linear precursors. On the understanding that other combinations regarding the positions of Leu and Ile residues and other circular permutations of the mature orbitide sequences could not be excluded, the putative linear transcripts GILVPPFFLI, GMLIPPFFVI, and GMLVFPLFI encoding for 21, 22, and 24, respectively, were used as the initial queries in TBLASTN searches26,27 of the whole genome shotgun (WGS) assembly of Linum usitatissimum (var. CDC Bethune).28 The TBLASTN searches using the initial queries returned matching linear sequences for their biosynthetic precursors, together with those of other known peptides. Specifically, a sequence encoded within GenBank accession AFSQ01016651.1 contained one copy of an embedded linear decapeptide GILVPPFFLI, a putative precursor of 21 ([1−10NaC]-GILVPPFFLI, Figure 6). The exon predicted on the plus strand by GENSCAN 1.0 (setting Arabidopsis) identified a putative open reading frame (base 7078 to base 7692), a E
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
NαC]-linusorb E1 (24); [1−9-NαC],[2-(Rs,Ss)-MetO]-linusorb E1 (25).31 Future flax research should be directed at investigating how flax orbitides are biosynthetically processed, as currently the requirements for flanking residues in the flax orbitide linear precursors remain unclear. An additional important focus of further studies should be to determine whether cyclic nonapetides and decapeptides with and without proto-Nterminal glycine, respectively, arise from a given flax precursor protein via the combined proto-C-terminal cleavage and cyclization reactions of a singular enzyme following proto-Nterminal cleavage or whether two distinct enzymes perform the proto-C-terminus-related biosynthetic cyclization steps. Once understood, the innate capability of flax to produce an extended array of cyclic peptides might be exploited for the large-scale agronomic production of custom cyclopeptides. It is proposed that the five new cyclolinopeptides 21, 22, 23, 24, and 25 should be named U, V, W, X, and Y, respectively.
■
EXPERIMENTAL SECTION
General Experimental Procedures. NMR spectra were recorded on a 600 MHz Bruker Avance spectrometer (Broadband Observed probe, TXI, 5 mm; TopSpin 3.2 software). All NMR spectra were recorded in acetone-d6. The 1H NMR spectral (600 MHz) chemical shift (δ) values are reported in parts per million (ppm). The δ values are referenced to residual solvent signal at 2.05 ppm, and multiplicities are indicated by the following symbols: s = singlet, d = doublet, dd = doublet of doublets, m = multiplet, and br = broad. For the 13C NMR (125.8 MHz) spectra, the δ values were referenced to residual solvent signal at 29.84 ppm. HRESIMS-HPLC was performed on an Agilent HPLC 1200 series directly connected to a micrOTOF-Q II mass spectrometer (Hybrid Quadrupole-TOF MS/MS, Bruker Daltonik GmbH, Bremen, Germany) with an Apollo II electrospray ionization (ESI) ion source at a capillary voltage of −4500 V, nebulizer gas at 4.0 bar, and dry gas temperature held at 200 °C. Chromatographic separation was achieved at ambient temperature using a Chromolith FastGradient RP-18e column (50 × 2.0 mm i.d., Merck KGaA, Darmstadt, Germany). The mobile phase consisted of a linear gradient of 0.1% formic acid in H2O and 0.1% formic acid in MeCN (60:40 for 2 min, to 10:90 in 8 min, to 60:40 in 0.5 min, to equilibration for 5.5 min) at a flow rate of 0.40 mL/min.11 Data acquisition was carried out in positive polarity mode per LC run. Data processing was carried out with Bruker Compass DataAnalysis software. HPLC-MS/MS analyses were conducted on a Bruker micrOTOF-Q II mass spectrometer using identical parameters as described for HR-HPLC-MS. HPLC-DAD analysis was performed with an Agilent 1200 series HPLC system (Agilent Technologies Canada, Mississauga, ON, Canada) equipped with a quaternary pump, autosampler, photodiode array detector (wavelength range 190−300 nm), degasser, and Chromolith SpeedRod RP-18e column (50 × 4.6 mm i.d., Merck KGaA) equipped with an in-line filter. The mobile phase consisted of a linear gradient of H2O−MeCN (70:30 to 30:70 in 4 min, to 0:90 in 0.5 min, to 70:30 in 0.5 min, to equilibration for 1 min) at a flow rate of 2 mL/min.32 Preparative reversed-phase chromatography was performed on a BioCAD SPRINT Perfusion chromatography workstation (Perspective Biosystems Inc., MA, USA) equipped with a Chromolith SemiPrep RP-18e column (100 × 10 mm) and a UV/vis detector operating at a wavelength of 214 nm. The mobile phase consisted of H2O−MeCN (60:40 for 2 min, to 10:90 in 3 min, to 60:40 in 0.5 min, to equilibration for 2 min) at a flow rate of 5 mL/min. All analytical analysis and preparative separations were performed at ambient temperature. Plant Material. Flaxseed oil was obtained from Bioriginal Food & Science Corporation (lot no. 804961, Saskatoon, Canada).
Figure 7. Structure of [1−10-NαC],[2-MetO]-CLV (23). Double arrows show selected NOESY correlations. Half-arrows show selected HMBC correlations.
three protons and a carbon of a MetO moiety.6 This singlet indicated that 23 existed as a single conformer under the NMR experimental conditions because no other conformers were detected. Additionally, strong NOESY correlations were observed between the α-proton of Ile4 and both δ-protons of Pro5, indicating that the Ile4-Pro5 amide bond possesses a transconfiguration, while a strong correlation between the α-protons of Pro5 and Pro6 indicates that Pro5 and Pro6 are cis-configured. This strong NOESY correlation between the α protons of the two prolyl groups signifies rigidity within the molecule maintaining 23 in one stable conformation. The 1H NMR spectrum (Table 2) also showed seven amide proton signals that were observed at δ 6.89, 7.78, 7.82, 7.94, 8.05, and 8.17(2). The amino acid sequence of 23 was further corroborated by 1 H−1H COSY data. The NOESY experiment shows the presence of only one cis-amide bond between two prolines and eight trans-amide bonds. Overall, the structure of 23 was confirmed as [1−10-NαC]-GOLIPPFFVI using NMR spectroscopic data. Herein we report the discovery of five novel glycinecontaining flax orbitides (21−25) from flaxseed oil and their structural analysis via HPLC-ESIMS/MS, with one peptide (orbitide 23) having been isolated and subjected to additional characterization by 1D and 2D NMR spectroscopy. The assigned cyclopeptide sequences were further supported by finding their putative corresponding gene sequences. Our data suggest that two distinct N-to-C linked peptides can be produced by the variable cyclization of common and overlapping precursor sequences by including different amino acids, which is unprecedented for members of both orbitide and cyclotide classes of plant cyclopeptides. This suggests novel flexibility in the biosynthetic capabilities within flax in terms of the production of cyclic peptides. We have recently proposed systematic nomenclature for flax orbitides namely: [1−10NαC]-linusorb E3 (21); [1−10-NαC]-linusorb E2 (22); [1−10-NαC],[2-(Rs ,S s)-MetO]-linusorb E2 (23); [1−9F
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Table 2. 1H and 13C NMR Assignments for 23 ([1−10-NαC],[2-MetO]-CLV) in Acetone-d6 at 298 K δH
assignment
δC
δC
Pro6 α NH CO
4.27, 1H, m 3.42, 1H, m 8.05, 1H, m
43.19
170.69
MetO2 α β γ εMe NH CO
4.47, 1.94, 2.78, 2.64, 8.17,
α β γ δMe
4.59, 1.83, 1.72, 0.92, 0.91, 7.82,
1H, 2H, 2H, 3H, 1H,
m m m s m
172.21
NH CO
1H, 2H, 1H, 3H, 3H, 1H,
m m m dd (6.7) dd (6.7) m
53.07 40.75 24.98 23.79 21.11 170.07
NH CO
4.59, 1.63, 1.27, 0.94, 0.86, 8.17,
1H, 1H, 2H, 3H, 3H, 1H,
m m m dd (6.7) dd (6.7) m
4.47, 1.54, 1.94, 3.31,
α β γ δ ε ς NH CO
4.37, 1H, m 3.1, 2H, m 7.36−7.06, 5H, m
α β γ δ ε ς NH CO
4.74, 1H, m 3.1, 2H, m 7.36−7.06, 5H, m
59.39 38.16 30.35 15.84 11.73
α β δMe
170.07
NH CO
δ NH CO
4.02, 1.72, 1.39, 1.04, 3.31,
2H, 2H, 1H, 1H, 2H,
m m m m m
m m m m
61.33 28.77 26.51 47.11 172.56 56.34 39.09 130.47 129.81 129.10 127.04
nda 172.1 55.39 39.09 131.07 129.46 128.76 126.55
7.94, 1H, m 172.1
Val9
Pro5 α β γ
2H, 2H, 2H, 2H,
Phe8
Ile4 α β γ δMe
α β γ δ CO Phe7
54.16 25.64 50.43 39.02
Leu3
a
δH
assignment
Gly1
4.63, 1.72, 1.03, 0.97, 6.89,
1H, 2H, 3H, 3H, 1H,
dd (4.9) m dd (7) dd (7) m
55.31 28.11 19.72 19.58 169.70
Ile10
61.25 31.22 23.00
α β γ δMe
48.23 170.88
NH CO
4.37, 2.02, 1.54, 0.99, 0.83, 7.78,
1H, 1H, 2H, 3H, 3H, 1H,
m m m dd (6.7) dd (6.7) m
56.34 36.39 28.71 15.59 11.45 172.36
Not detected. NaHCO3 (500 mL), followed by EtOAc and brine.33 The organic phase was subsequently concentrated to yield a solvent-free orbitide extract that was subjected to flash column chromatography on silica gel 60 to obtain multiple fractions enhanced with different flax orbitides. Sequential elution with 50% EtOAc in n-hexane to 100% EtOAc using 10% EtOAc increments (a to f), 5% MeOH in CH2Cl2 (g), and 10% MeOH in CH2Cl2 (h) were conducted. The final fraction (h) containing cyclolinopeptides 23 and 25, in addition to other known orbitides, was subjected to HPLC-ESIMS/MS analyses. A portion of fraction h was reduced with NaBH4 and I2 following a reported protocol.34 Cyclolinopeptides 22 and 24 were detected among other known peptides in the crude reduced product. The remaining portion of fraction h was purified on Sep-Pak Vac 35 cc (10 g) tC18 cartridges (Waters, Ireland). Sequential elutions of the cartridges with increasing concentration of MeOH in H2O (50% to 100% MeOH using 10% increments) were conducted. H2O was purified to 18.2 MΩcm on a Milli-Q Integral system (Millipore, Molsheim, France). Fractions enhanced with cyclolinopeptide 23 were combined and purified using preparative reversed-phase HPLC chromatography as described in the HPLC experiments. The structure
Extraction and Isolation. Flax orbitides were extracted from flaxseed oil by adsorption onto silica gel 60 (40−63 μm particle size, EMD Chemicals), followed by recovery through solvent elution. Specifically, flaxseed oil (15 L) was mixed with silica gel 60 for 2 h, and after settling overnight the flaxseed oil was carefully decanted. A slurry of the orbitide-laden silica residue was prepared with n-hexane (1 L) and transferred to a sintered glass funnel for solvent removal by filtration. The recovered solid residue was further extracted with nhexane (1 L × 5) to remove as much entrained oil as possible from the orbitide-laden silica. The oil-free residue was sequentially extracted with (i) 50% (v/v) EtOAc in n-hexane (1 L × 6) and (ii) EtOH (1 L × 4). An aliquot of each filtrate was concentrated in vacuo, resuspended in MeOH, and filtered through 0.45 μm PTFE membranes for HPLC-DAD analyses. Flax orbitides were detected only in the EtOH filtrate, and as such, the entire extract (ii) was freed of solvent using a rotary evaporator. Additional solvent was removed by subjecting the crude extract to high vacuum overnight to yield a solvent-free crude orbitide extract that was analyzed by high-resolution HPLC-ESIMS (HPLC-HRESIMS) and tandem HPLC-ESIMS. The solvent-free crude orbitide extract was enriched by solvent/ solvent partitioning between EtOAc and a saturated solution of G
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
of 23 was completely elucidated by MS and 1D (1H and 13C) and 2D (NOESY and HMBC) NMR spectroscopy. Cyclolinopeptide 21: HPLC-ESIMS m/z 1097.6746 [M + H]+ (calcd for C59H89N10O10, 1097.6758). Cyclolinopeptide 22: HPLC-ESIMS m/z 1115.6307 [M + H]+ (calcd for C58H87N10O10S, 1115.6322). Cyclolinopeptide 23: HPLC-ESIMS m/z 1131.6229 [M + H]+ (calcd for C58H87N10O11S, 1131.6271); 1H and 13C NMR data, see Table 2. Cyclolinopeptide 24: HPLC-ESIMS m/z 1018.5873 [M + H]+ (calcd for C53H80N9O9S, 1018.5794). Cyclolinopeptide 25: HPLC-ESIMS m/z 1034.5787 [M + H]+ (calcd for C53H80N9O10S, 1034.5743).
■
(10) Stefanowicz, P. Eur. J. Mass Spectrom. 2004, 10, 665−671. (11) Okinyo-Owiti, D. P.; Young, L.; Burnett, P.-G. G.; Reaney, M. J. T. Biopolymers 2014, 102, 168−175. (12) Lao, Y. W.; Mackenzie, K.; Vincent, W.; Krokhin, O. V. J. Sep. Sci. 2014, 37, 1788−1796. (13) Brühl, L.; Matthäus, B.; Fehling, E.; Wiege, B.; Lehmann, B.; Luftmann, H.; Bergander, K.; Quiroga, K.; Scheipers, A.; Frank, O.; Hofmann, T. J. Agric. Food Chem. 2007, 55, 7864−7868. (14) Shim, Y. Y.; Gui, B.; Arnison, P. G.; Wang, Y.; Reaney, M. J. T. Trends Food Sci. Technol. 2014, 38, 5−20. (15) Oman, T. J.; van der Donk, W. A. Nat. Chem. Biol. 2010, 6, 9− 18. (16) Dutton, J. L.; Renda, R. F.; Waine, C.; Clark, R. J.; Daly, N. L.; Jennings, C. V.; Anderson, M. A.; Craik, D. J. J. Biol. Chem. 2004, 279, 46858−46867. (17) Gillon, A. D.; Saska, I.; Jennings, C. V.; Guarino, R. F.; Craik, D. J.; Anderson, M. A. Plant J. 2008, 53, 505−515. (18) Saska, I.; Gillon, A. D.; Hatsugai, N.; Dietzgen, R. G.; HaraNishimura, I.; Anderson, M. A.; Craik, D. J. J. Biol. Chem. 2007, 282, 29721−29728. (19) Condie, J. A.; Nowak, G.; Reed, D. W.; Balsevich, J. J.; Reaney, M. J. T.; Arnison, P. G.; Covello, P. S. Plant J. 2011, 67, 682−690. (20) Covello, P. S.; Datla, R. S. S.; Stone, S. L.; Balsevich, J. J.; Reaney, M. J.; Arnison, P. G.; Condie, J. A. U.S. Patent application 2012/0058905A1, 2012. (21) Barber, C. J. S.; Pujara, P. T.; Reed, D. W.; Chiwocha, S.; Zhang, H.; Covello, P. S. J. Biol. Chem. 2013, 288, 12500−12510. (22) Mylne, J. S.; Chan, L. Y.; Chanson, A. H.; Daly, N. L.; Schaefer, H.; Bailey, T. L.; Nguyencong, P.; Cascales, L.; Craik, D. J. Plant Cell 2012, 24, 2765−2778. (23) Aladedunye, F.; Sosinska, E.; Przybylski, R. J. Am. Oil Chem. Soc. 2013, 90, 419−428. (24) Jadhav, P. D.; Okinyo-Owiti, D. P.; Ahiahonu, P. W. K.; Reaney, M. J. T. Food Chem. 2013, 138, 1757−1763. (25) http://www.colby.edu/chemistry/NMR/scripts/tripeptides. html. (26) Gertz, E. M.; Yu, Y.-K.; Agarwala, R.; Schaffer, A. A.; Altschul, S. F. BMC Biol. 2006, 4, 41. (27) Altschul, S. F.; Madden, T. L.; Schaffer, A. A.; Zhang, J.; Zhang, Z.; Miller, W.; Lipman, D. J. Nucleic Acids Res. 1997, 25, 3389−3402. (28) Wang, Z.; Hobson, N.; Galindo, L.; Zhu, S.; Shi, D.; McDill, J.; Yang, L.; Hawkins, S.; Neutelings, G.; Datla, R.; Lambert, G.; Galbraith, D. W.; Grassa, C. J.; Geraldes, A.; Cronk, Q. C.; Cullis, C.; Dash, P. K.; Kumar, P. A.; Cloutier, S.; Sharpe, A. G.; Wong, G. K. S.; Wang, J.; Deyholos, M. K. Plant J. 2012, 72, 461−473. (29) Burge, C.; Karlin, S. J. Mol. Biol. 1997, 268, 78−94. (30) Burge, C. B. In Computational Methods in Molecular Biology; Salzberg, S. L., Searls, D. B., Kasif, S., Eds.; Elsevier: Amsterdam, 1998; pp 127−163. (31) Shim, Y. Y.; Young, L. G.; Arnison, P. G.; Gilding, E.; Reaney, M. J. T. J. Nat. Prod. 2015, DOI: 10.1021/np500802p. (32) Olivia, C. M.; Burnett, P.-G. G.; Okinyo-Owiti, D. P.; Shen, J.; Reaney, M. J. T. J. Chromatogr. B 2012, 904, 128−134. (33) Okinyo-Owiti, D. P.; Burnett, P.-G. G.; Reaney, M. J. T. J. Chromatogr. B 2014, 965, 231−237. (34) Reaney, M. J.; Burnett, P.-G. G.; Jadhav, P. D.; Okinyo-Owiti, D. P.; Shen, J.; Shim, Y. Y. WIPO WO/2013/091070A1, 2013.
ASSOCIATED CONTENT
S Supporting Information *
NMR data (1H, 13C, COSY, HMBC, HMQC, NOESY, and TOCSY) for compound 23 in Figures S1−S7. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Authors
*(P. D. Jadhav) Tel: +1-306-966-8840. Fax: +1-306-966-5015. E-mail:
[email protected]. *(M. J. T. Reaney) Tel: +1-306-966-5027. Fax: +1-306-9665015. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Financial support was provided by Genome Canada (TUFGEN) and the Agricultural Development Fund from the Saskatchewan Ministry of Agriculture. We would like to thank K. Gui and V. Jadhav for laboratory assistance and D. Craik for help in preparing this manuscript for publication.
■
REFERENCES
(1) Arnison, P. G.; Bibb, M. J.; Bierbaum, G.; Bowers, A. A.; Bugni, T. S.; Bulaj, G.; Camarero, J. A.; Campopiano, D. J.; Challis, G. L.; Clardy, J.; Cotter, P. D.; Craik, D. J.; Dawson, M.; Dittmann, E.; Donadio, S.; Dorrestein, P. C.; Entian, K.-D.; Fischbach, M. A.; Garavelli, J. S.; Goeransson, U.; Gruber, C. W.; Haft, D. H.; Hemscheidt, T. K.; Hertweck, C.; Hill, C.; Horswill, A. R.; Jaspars, M.; Kelly, W. L.; Klinman, J. P.; Kuipers, O. P.; Link, A. J.; Liu, W.; Marahiel, M. A.; Mitchell, D. A.; Moll, G. N.; Moore, B. S.; Mueller, R.; Nair, S. K.; Nes, I. F.; Norris, G. E.; Olivera, B. M.; Onaka, H.; Patchett, M. L.; Piel, J.; Reaney, M. J. T.; Rebuffat, S.; Ross, R. P.; Sahl, H.-G.; Schmidt, E. W.; Selsted, M. E.; Severinov, K.; Shen, B.; Sivonen, K.; Smith, L.; Stein, T.; Suessmuth, R. D.; Tagg, J. R.; Tang, G.-L.; Truman, A. W.; Vederas, J. C.; Walsh, C. T.; Walton, J. D.; Wenzel, S. C.; Willey, J. M.; van der Donk, W. A. Nat. Prod. Rep. 2013, 30, 108− 160. (2) Gui, B.; Shim, Y. Y.; Datla, R. S. S.; Covello, P. S.; Stone, S. L.; Reaney, M. J. T. J. Agric. Food Chem. 2012, 60, 8571−8579. (3) Morita, H.; Takeya, K. Heterocycles 2010, 80, 739−764. (4) Kaufmann, H. P.; Tobschirbel, A. Chem. Ber. 1959, 92, 2805− 2809. (5) Morita, H.; Shishido, A.; Matsumoto, T.; Takeya, K.; Itokawa, H.; Hirano, T.; Oka, K. Bioorg. Med. Chem. Lett. 1997, 7, 1269−1272. (6) Morita, H.; Shishido, A.; Matsumoto, T.; Itokawa, H.; Takeya, K. Tetrahedron 1999, 55, 967−976. (7) Matsumoto, T.; Shishido, A.; Morita, H.; Itokawa, H.; Takeya, K. Phytochemistry 2001, 57, 251−260. (8) Matsumoto, T.; Shishido, A.; Takeya, K. Tennen Yuki Kagobutsu Toronkai Koen Yoshishu 2001, 43, 407−412. (9) Stefanowicz, P. Acta Biochim. Pol. 2001, 48, 1125−1129. H
DOI: 10.1021/np5008558 J. Nat. Prod. XXXX, XXX, XXX−XXX
