Article pubs.acs.org/JAFC
Identification and Characterization of Kukoamine Metabolites by Multiple Ion Monitoring Triggered Enhanced Product Ion Scan Method with a Triple-Quadruple Linear Ion Trap Mass Spectrometer Yuan-Yuan Li,† Haixing Wang, Chunyan Zhao, Ye-Qing Huang, Xiaowei Tang, and Hon-Yeung Cheung* Research Group for Bioactive Products, Department of Biomedical Sciences, City University of Hong Kong, 83 Tat Chee Avenue, Hong Kong SAR, China ABSTRACT: Kukoamines are a series of bioactive phytochemicals conjugated by a polyamine backbone and phenolic moieties. Understanding the structural diversity of kukoamine metabolites in plants is meaningful for drug discovery. In this study, an LCMS/MS method was established for kukoamine profiling and characterization from lycii cortex (LyC) via a triple-quadrupole linear ion trap mass spectrometry (Q-TRAP). On the basis of the typical fragmentation of kukoamine, a diagnostic ion, which represents the features of the backbone and phenolic substitute, was chosen as the product ion for precursor ion scan, and then the screened precursor ions were applied to a successive multiple ion monitoring triggered enhanced product ion scan (MIMEPI) to simultaneously present the profile survey and MS/MS acquisition. Because the MIM narrowed the ion scan range in Q1 and the ion trap enhanced the ion fragments passing through Q2, the qualitative capability of quadrupole MS can be greatly improved, especially for capture of the uncommon metabolites. There are 12 kukoamine metabolites identified from LyC, with either spermine or spermidine backbone and with conjugation of one to three dihydrocaffeoyls or other kinds of phenolic moieties. Except for kukoamines A and B, other metabolites were identified in LyC for the first time. This approach can be utilized for metabolite identification in other substrates. KEYWORDS: kukoamines, Q-TRAP, MIM-EPI, fragmentation behavior, structural characterization, molecular diversity
■
INTRODUCTION
difficult to detect the minor metabolites. On the other hand, the inability to detect ion fragments in low abundance makes the MS/MS spectrum less informative for structural elucidation. All of these limitations have made triple-quadrupole MS less competent in qualitative analysis. With the introduction of triple-quadruple linear ion trap mass spectrometry (Q-TRAP), ion fragments can be enriched in the ion trap, which enables a novel function in Q-TRAP MS called enhanced product ion scan (EPI). Compared to the traditional triple-quadrupole MS, the enrichment of fragment ions significantly enhances the ion signals, thus enabling less abundant ion fragments to be shown in the MS/MS spectrum.14−16 Among multiple scan modes based on the QTRAP platform, the multiple ion monitoring (MIM)-dependent MS/MS, which was modified from the conventional MRM mode, was reported to be an enhanced tool for metabolite profiling and characterization in various sample substrates. It is especially useful for discovering uncommon metabolites with low concentrations.14,15,17 The major task of this research is to investigate the kukoamine metabolites in LyC with MIM-EPI based on the Q-TRAP MS platform. More precisely, our work included (1) selecting diagnostic ions for kukoamine metabolites through fragmentation analysis, (2) screening parent ions for kukoamine metabolites in LyC via precursor ion scan, (3) profiling kukoamine metabolites and acquiring their MS/MS spectra via
Kukoamines are a series of plant polyamines composed of a polymethylenepolyamine backbone (i.e., putrescine, spermidine, and spermine) and at least one dihydrocaffeic acid fragment.1,2 They were first discovered in lycii cortex (LyC)3,4 and then subsequently found in other plants of the Solanaceae family, such as potato, tomato, and tobacco.5 Due to their versatile bioactivities, such as antihypertension,3 antitrypanosome,6 antilipid peroxidation and lipoxygenase,7 antisepsis,8,9 and neuroprotection,10 kukoamines have received attention in recent years as functional foods and drug candidates. Because the polymethylenepolyamine backbone may conjugate with different phenolic moieties to form diverse derivatives,5 investigation of its molecular diversity is crucial for understanding kukoamine biosynthesis and metabolism, which are meaningful for drug discovery. The traditional approach for phytochemical identification requires monomer relying on tedious purification steps,3,4 but this approach is unsuitable for the uncommon metabolites present in minute quantity, which are difficult to enrich and purify. Hyphenated instrumental analysis using high-performance liquid chromatography (HPLC) combined with tandem mass spectrometry has become an indispensable tool for profiling and identifying metabolites in plant, microorganism, mammal, and environmental samples.11 Normally, identification of secondary metabolites with LC-MS depends on determining molecular ions and their ion fragments.12−14 Therefore, the investigation widely employs a full-scan MS analysis as a scan survey to trigger the MS/MS acquisition. However, for most of the triple-quadrupole MS, owing to the limited scan speed, the compromise of sensitivity in qualitative analysis makes it © 2015 American Chemical Society
Received: Revised: Accepted: Published: 10785
September 4, 2015 November 4, 2015 December 1, 2015 December 1, 2015 DOI: 10.1021/acs.jafc.5b04321 J. Agric. Food Chem. 2015, 63, 10785−10790
Article
Journal of Agricultural and Food Chemistry MIM-EPI scan, and (4) structural elucidation of novel kukoamine metabolites based on precursor ion and MS/MS spectra. To the best of our knowledge, this is the first work revealing the molecular diversity of kukoamine metabolites in lycii cortex. Our results here could provide important information about the biosynthesis and metabolism of polyamine alkaloids in Solanaceae. The discovery of kukoamine metabolites may also be useful for metabolic and pharmacokinetic study of kukoamines as potential drug or nutrition reagents in cell and animal models. Additionally, our research could provide a practical case for metabolite discovery in herbal samples, which can also be utilized in other kinds of molecules in different sample matrices.
■
Scheme 1. Flowchart of the Screening Method Combined PIS with MIM-EPI Acquisition
MATERIALS AND METHODS
Materials and Reagents. Acetonitrile and methanol (HPLC grade) were purchased from Labscan Asia (Bangkok, Thailand). Formic acid and acetic acid (HPLC grade) were from Fluka (Buchs, Switzerland). Milli-Q water was purified using a Millipore Q-Plus system (Millipore, Bedford, MA, USA). All other chemicals of analytical grade were purchased from Sigma-Aldrich (St. Louis, MO, USA). Kukoamines B and A were purified from lycii cortex by our group with purities >98%. Sample Preparation. The crude drug of lycii cortex was crushed into powder, and then 0.5 g of the powder was extracted twice with 10 mL of 50% methanol each time. The extracted solutions were mixed and then evaporated at 40 °C via a rotary evaporator (Laborota 4000, Heidolph, Germany) under vacuum to reduce the solvent and transferred into a 10 mL volumetric, made up to the mark with 50% methanol. After filtration with a 0.22 μm membrane, the solution was injected to LC-MS/MS for analysis. Liquid Chromatography. The chromatographic analysis was carried out on an Agilent 1200 HPLC system with a binary bump, a vacuum degasser, an autosampler, a diode array detector, and a column oven (Agilent, Santa Clara, CA, USA). The separation was performed on a Waters X Bridge C18 column (2.1 mm i.d. × 10 mm, 3.5 μm) fitted with a C18 guard column (Waters, Ireland) and eluted with a mobile system composed by 0.2% formic acid aqueous solution (phase A) and acetonitrile (phase B). With a flow rate at 0.32 mL min−1, the running program was set as 0−25 min, isocratic elution with 4.3% B; 25−40 min, linear change from 4.3 to 25% B; 40−45 min, isocratic elution with 25% B. The column temperature was 40 °C, and the injection volume was 5 μL. General Parameters of Mass Spectrometry. A Q-TRAP spectrometer equipped with a Turbo V electrospray ionization source (ESI) (API 3200, Foster City, CA, USA) was connected with the HPLC. The software Analyst 1.5.2 was used for system control and data processing. All experiments were conducted in positive mode. The MS was calibrated regularly using polypropylene glycol (PPG) for accuracy. The general parameters for MS were as follows: curtain gas and source gas (gas 1 and gas 2), 20 and 40 psi, respectively; spray voltage, 5500 V; source temperature, 500 °C; interface heater, set as “on”; entrance potential (EP), 10 V; and collision cell exit potential (CXP), 5 V. Nitrogen was used in all cases. Without specific indication, parameter settings in the following scan mode are the same as those in general setting. Precursor Ion Scan Combined Multiple Ion MonitoringTriggered Enhanced Product Ion Scan (MIM-EPI). A flowchart of the scan method is shown in Scheme 1. To investigate the fragmentation, kukoamine B (KB) was subjected to CID via direct infusion and the ion fragments were enhanced and acquired by EPI scanning. First, KB was dissolved in a 5% ACN aqueous solution (v/v) with 0.2% formic acid. The infusion was carried out on a microinfusion syringe pump (Harvard, Quebec, Canada) with the flow rate at 5 μL min−1. The monitored mass range was from 100 to 550 amu. Parameters for MS were set as follows: GS1, 25 psi; GS2, 0 psi; source temperature, 100 °C; DP, 80 V; and CE 50 eV.
Precursor ion scan (PIS) was fulfilled with the connection of HPLC with the separation condition as 2.3. Ion m/z 222 was selected as the product ion of PIS, and the monitored precursor ions were in a range from 250 to 1000 amu. The DP was set at 80 V, and the collision energy (CE) was set at 30 eV. MIM-EPI scan was conducted after PIS, and the selection of ion pairs was based on the precursor ions captured in the PIS. The MIM scan was carried out on the MRM mode targeting the same ions in Q1 and Q3, respectively, with the minimal CE (5 eV) in Q2. The threshold for IDA triggered for the EPI acquisition was set to 500 cps. Each MIM transition was performed with a 50 ms dwell time and a 5 ms pause time. DP was set at 80 V, and the CE was set at 50 eV. Full-Scan Trigged Enhanced Product Ion Scan (FS-EPI). The enhanced full-scan survey was conducted at a mass range from 100 to 1000 amu. The IDA threshold for triggered EPI was set to 500 000 cps, and the monitoring mass range of MS/MS spectra was from 100 to 1000 amu. The DP was set at 80 eV, and the collision energy (CE) was set at 30 eV. HPLC conditions for this scan are the same as given under Liquid Chromatography.
■
RESULTS AND DISCUSSION Fragmentation Behavior of Kukoamines. To investigate the diagnostic ion for metabolite identification, kukoamine B (KB) was adopted as a model compound for fragmentation analysis. Because the amine group is easily charged with a proton to leave an ion, [M + H]+, the positive mode was adopted throughout the study. Figure 1 shows the MS/MS spectrum of KB (Figure 1A) and its fragmentation (Figure 1B), respectively. The parent ion at m/z 531 was first split at the peptide bond to produce a product ion at m/z 165 and a neutral unit. This neutral unit donated one proton to form a positive ion at m/z 367. This ion was unstable, and it continued to lose a 1,3-diaminopropane unit (Δm = 74) to yield a further fragment ion at m/z 293. Then, the ion m/z 293 continued to eliminate a 3-aminopropylene (Δm = 71) to form a more stable ion at m/z 222. The further cleavage of m/z 222 was through eliminating a CO (Δm = 28) and a propylene (C3H6, Δm = 42) with a typical collision rearrangement to yield an abundant ion at m/z 123. Ion m/z 123 was also the product ion of the fragment at m/z 165, one of the major fragments formed in peptide bond cleavage. The collision energy (CE) is a key factor influencing the fragmentation. When the CE was set at 30 eV, only two major fragments, m/z 293 and 222, were found in the spectrum. When the CE was increased to 50 eV, ion fragments in lower m/z value became much more abundant, 10786
DOI: 10.1021/acs.jafc.5b04321 J. Agric. Food Chem. 2015, 63, 10785−10790
Article
Journal of Agricultural and Food Chemistry
have been reported in LyC, whereas the other metabolites were found for the first time in this species. MIM-EPI for Metabolite Profiling and Characterization. To further study the structure of the metabolites, a scan mode named MIM-EPI was adopted to perform profile survey and MS/MS acquisition for the concerned 12 compounds. The MIM scan setting for profile survey is based on a multiple reaction monitoring (MRM) mode in triplequadruple MS. Unlike the routine setting in MRM, the ions in Q1 and Q3 for each transition pair were kept consistent, and the CE was set to the lowest value of 5 eV to prevent the precursor ion from undergoing fragmentation in Q2.14 The MS/MS acquisition was triggered depending on the signals in MIM profiling via the IDA mode, and then the product ions formed by CID fragmentation were enhanced by the ion trap to get highly abundant signals in spectra. Figure 2 shows the profile of kukoamine metabolites in LyC and the MS/MS spectra of compound E8 under different acquisition modes. Compared to the conventional FS-EPI, which showed only three major peaks in profile survey (Figure 2A1) and very few fragment signals in spectrum (Figure 2A2), the MIM-EPI displayed much better performance in both metabolite profiling (Figure 2B1) and MS/MS spectrum acquisition (Figure 2B2). We considered the advantages of MIM-EPI to lie in the following aspects: (1) it allows the MRM mode to be used for qualitative analysis for unknown compounds, as the optimization of ion pairs and fragmentation energy with standard reference is unnecessary; (2) it improves the sensitivity of quadruple MS for untargeted analysis, owing to the greatly minimized ion scan range; and (3) it increased the screening efficiency, as multiple isomers with the same molecular weight can be analyzed in one transition. Identification of Kukoamine-Related Derivatives. Information about the relevant compounds, such as retention time, quasi-molecular ions, and major fragment ions, is summarized in Table 1. All of the metabolites yield common fragments, those being m/z 222, 293, and 165, in MS/MS spectra. They were all considered as kukoamine-related metabolites, due to sharing fragmentation with the model compound (KB) in CID. Therefore, structures of these compounds can be deduced through comparison with the model compound in MS/MS spectrum and molecular weight. On the basis of the polyamine skeleton, kukoamine metabolites can be generally classified into two types, spermine and spermidine.5 In light of the “nitrogen rule”, the spermine type containing an even number of nitrogen atoms normally yields an odd-numbered quasi-molecular ion, whereas the spermidine type with an odd number of nitrogens yields an even-numbered quasi-molecular ion.5 Therefore, nine metabolites in our study were indentified with a spermine backbone (E1, E2, E3, E4, E5, E7, E9, E10, and E12) and two were with a spermidine backbone (E8 and E11). Another compound (E6), which yielded an even-numbered quasi-molecular ion, was considered as a backbone with one more aminopropyl on the spermine. According to the fragmentation of KB in Figure 1, the fragment ion at m/z 165 is responsible for the dihydrocaffeoyl moiety. In the profile of LyC, five compounds (E3, E4, E8, E11, and E12) yielded a unique and abundant signal at m/z 165 within the range of m/z 100−200, indicating the only existence of dihydrocaffeoyl moieties in molecule. The different molecular weights (MW) of these compounds were dependent on the number of dihydrocaffeoyl groups in conjugation and
Figure 1. MS/MS spectrum of kukoamine B (A) and its fragmentation in CID (B). The spectrum was acquired by the enhanced precursor ion (EPI) scan via direct infusion mode in Q-TRAP. Signal m/z 531 is the quasi-molecular ion of KB in positive mode. Other signals are ion fragments formed by collision-induced dissociation (CID). The fragmentation in panel B was proposed according to the ion fragments observed in panel A. The symbol ⊕ indicates the possible positive charge formed in the molecule.
especially for m/z 165 and 123. To the best of our knowledge, the fragmentation behavior of kukoamine in CID was determined in our study for the first time. Ion fragments at m/z 293, 222, 165, and 123 can be used to recognize kukoamine metabolites, because they are not only abundant in MS/MS spectrum but also capable of representing the structural feature of kukoamine in ion fragments, such as dihydrocaffeoyl and polyamine backbone. Selection of Product Ion for Precursor Ion Scan (PIS). Among numerous fragment ions in the MS/MS spectrum, ion m/z 222 was considered as the most suitable product ion for PIS, for the following reasons: (1) it was stable in fragmentation, enabling abundant signal in MS/MS spectra; and (2) it contains both moieties of dihydrocaffeoyl and polyamine backbone, allowing better selectivity in screening. With ion m/z 222 as product ion, 12 precursor ions in LyC containing this ion in fragmentation were screened out, which were 547, 605, 531, 531, 529, 543, 687, 695, 474, 638, and 588 (Table 1 and Figure 3). According to the literature,3−5 the two 531 signals referred to kukoamines A and B, respectively, which 10787
DOI: 10.1021/acs.jafc.5b04321 J. Agric. Food Chem. 2015, 63, 10785−10790
Article
Journal of Agricultural and Food Chemistry Table 1. Kukoamine Metabolites Detected and Structural Characterization in LyC via Q-TRAP LC-MS no.a
tR (min)
E1
3.8
E2
8.2
E3e
20.2
E4e E5
22.0 29.8
N-dihydrocaffeoyl-N′-(3,4,5-trihydroxycinnamoyl) spermine N-dihydrocaffeoyl-N′-(3,4,5-trihydroxycinnamoyl)-N″hydroxyacetylspermine N1,N14-bis(dihydrocaffeoyl)spermine (kukoamine A)3,5,18,19 N1,N10-bis(dihydrocaffeoyl)spermine (kukoamine B)4 N-dihydrocaffeoyl-N′-dihydrocaffeoyl o-quinonespermine
E6 E7 E8 E9 E10 E11 E12
30.6 32.1 32.5 35.8 36.1 40.1 45.8
N,N′-bis(dihydrocaffeoyl)-N″-n-propylaminospermine N-dihydrocaffeoyl-N′-dihydroxyferuloylspermine1,5,18 N1,N10-bis(dihydrocaffeoyl)spermidine5,18,19 N-dihydrocaffeoyl-N′-feruloylspermine1,5,18 N,N′-bis(dihydrocaffeoyl)-N″-shikimoylspermine N1,N5,N10-tris(dihydrocaffeoyl)spermindine5,18,19 N1,N5,N14-tris(dihydrocaffeoyl)spermine5,18,19
identification or tentative identificationd
MW (Da)
parent ionb (MH+, m/z)
546
547
547, 539, 367, 293, 222, 181, 165
604
605
605, 547, 367, 293, 222, 181, 165
530
531
531, 513, 367, 293, 222, 194, 165, 123
530 528
531 529
587 544 473 542 686 637 694
588 545 474 543 687 638 695
531, 513, 367, 293, 222, 194, 165, 529, 511, 458, 440, 383, 341, 293, 164, 163, 152,137, 135 588, 515, 367, 293, 222, 165 545, 307, 293, 236, 222, 179, 165, 474, 455,293, 236, 222, 194, 165 543, 525, 305, 293, 234, 222, 177, 687, 448, 293, 222, 165, 157 638, 621, 581, 568, 293, 222, 165 695, 474, 293, 222, 165
fragment ionsc (m/z)
123 222, 194, 220, 166,
137, 123 165
a
Number of peaks is consistent with that in Figures 2 B1 and 3. bParent ions were the screened precursor ions using m/z 222 as product ion via PIS. Fragment ions were acquired by MIM-EPI mode. dStructures with N, N′, or N″ indicate that multiple isomers may exist in one specific molecular weight eCompounds were further identified with standard substances. c
Figure 2. Comparison of FS-EPI and MIM-EPI in peak profiling and MS/MS spectra acquisition for kukoamine metabolites. A1 and A2 are metabolite profiles obtained by FS-EPI and MIM-EPI; B1 and B2 are MS/MS spectra of compound E8 (m/z 474, [M + H]+) acquired by FS-EPI and MIM-EPI. Details of conditions for acquisition are described.
the type of polyamine backbone. For example, E3 and E4, showing the same MS/MS spectra as the model compound (Figure 1), are identified as KA and KB, which are spermine alkaloids with two dihydrocaffeoyls conjugated. Compound E12, which yields an additional dihydrocaffeoyl than KB in MW, was identified as N1,N5,N14-tris(dihydrocaffeoyl)spermine containing three dihydrocaffeoyls. In a similar way, E8 and E11, which share the same spermidine backbone but with different dihydrocaffeoyls in conjugation, were identified as N1,N10bis(dihydrocaffeoyl)spermidine and N 1 ,N 5 ,N 1 0 -tris(dihydrocaffeoyl)spermidine.5 In addition to the dihydrocaffeoyl, other phenolic moieties were found in some metabolite molecules. This fact was reflected by the appearance of ions in addition to m/z 165 within the range from m/z 100 to 200. For instance, compound E7 yielded two prominent fragment ions at m/z 165 and 179,
respectively, which indicated an additional dihydroferuloyl attaching on the spermine backbone. In a similar way, with indication of the additional characteristic ions, 3,4,5-trihydroxycinnamoyl, feruloyl, and shikimoyl were considered to be conjugated E1, E9, and E10, respectively. With the exception of E7 and E9, which contain dihydroferuloyl and feruloyl, respectively, and have been mentioned in the Solanaceae family before,1,5,18,19 all of these metabolites with hybrid phenolic moieties in conjugation were discovered in LyC for the first time. It is interesting to mention compound E5, which had a quasimolecular ion just 2 Da less than that of KB. The presenting fragment ions at m/z 222, 220, 164, 163, 137, and 135 suggested two types of phenolic moieties conjugated with the spermine. We considered that one moiety is dihydrocaffeoyl and the other one is dihydrocaffeoyl o-quinone, which was 10788
DOI: 10.1021/acs.jafc.5b04321 J. Agric. Food Chem. 2015, 63, 10785−10790
Article
Journal of Agricultural and Food Chemistry
Figure 3. Structures of kukoamine metabolites identified from lycii cortex via PIS combined MIM-EPI screening method. Compounds were classified into three types according to the polyamine backbone: (A) spermine; (B) spermidine; (C) aminopropyl spermine. Compounds inside the dashed line box are in conjugation with dihydrocaffeoyls, whereas others are in conjugation with hybrid phenolic moieties. Different types of phenolic moieties are marked with different colors. Codes E1−E12 are consistent with those in Figure 2. Symbols: (▲) assignment of compounds confirmed with standard references; (△) structures are speculative on the basis of MS/MS spectra. Letters (a−e) in parentheses indicate that the identification of compounds is supported by the literature: (a) Funayama et al.;4; (b) Parr et al.;5 (c) Funayama et al.;3 (d) Nárvez-Cuenca et al.;18 (e) Nárvez-Cuenca et al.;19 (f) Rogoza et al.1
formed through oxidizing one of the phenolic hydroxyls on dihydrocaffeoyl. The assumption of E5 as an oxidative product of KB was supported by a fact that E5 was increased when KB was exposed to air for long periods. Results obtained from 10 batches of LyC samples suggested that kukoamine metabolites were widespread in LyC, even though the kind and abundance of each compound varied remarkably in samples from batch to batch. KA and KB were predominant in all samples, and KB was more abundant than KA. 20,21 Compound E8, N 1 ,N 10 -bis(dihydrocaffeoyl)spermidine, was also found in all batches of samples with a content close to kukoamines A and B. However, other kinds of kukoamine metabolites are very rare in LyC. Some of them can be found in only one or several batches. It is necessary to point out that kukoamine metabolites in one specific MW may have different isomers yielding the same MS/MS spectra (e.g., kukoamines A and B), because the conjugation position may vary among multiple amino groups in
the backbone. Information from MS/MS spectra can explain only the type of polyamide backbone and the phenolic groups in conjugation. The absolute structures still require purification and further elucidation based on other technical means, for example, NMR. Nevertheless, our method can provide an easy and fast way to reveal the molecular diversity of kukoamine metabolites, particularly those with low abundance. The wide existence of bisdihydrocaffeoyl and trididydrocaffeoyl spermines or spermidines in LyC also supports the viewpoint that kukoamines are widespread in plants of the Solanaceae family. Conclusions. A qualitative screening method combining PIS and MIM-EPI acquisition based on Q-TRAP has been developed to investigate the kukoamine-related metabolites in LyC. This method is superior to the conventional acquisition mode in triple-quadrupole MS for qualitative analysis, particularly for metabolites with low abundance. In this study, 12 kukoamine metabolites with either spermine or spermidine backbones were identified from LyC samples. Metabolites with 10789
DOI: 10.1021/acs.jafc.5b04321 J. Agric. Food Chem. 2015, 63, 10785−10790
Article
Journal of Agricultural and Food Chemistry
neurons via activating the PI3-K/Akt/GSK3β pathway by kukoamine from lycii cortex. J. Funct. Foods 2015, 17, 709−721. (11) Seger, C.; Sturm, S.; Stuppner, H. Mass spectrometry and NMR spectroscopy: modern high-end detectors for high resolution separation techniques-state of the art in natural product HPLC-MS, HPLC-NMR, and CE-MS hyphenations. Nat. Prod. Rep. 2013, 30, 970−987. (12) Clifford, M. N.; Wu, W.; Kirkpatrick, J.; Kuhnert, N. Profiling the chlorogenic acids and other caffeic acid derivatives of herbal chrysanthemum by LC-MSn. J. Agric. Food Chem. 2007, 55, 929−936. (13) Yang, M.; Wang, X.; Guan, S.; Xia, J.; Sun, J.; Guo, H.; Guo, D. Analysis of triterpenoids in Ganoderma lucidum using liquid chromatography coupled with electrospray ionization mass spectrometry. J. Am. Soc. Mass Spectrom. 2007, 18, 927−939. (14) Yao, M.; Ma, L.; Humphreys, W. G.; Zhu, M. Rapid screening and characterization of drug metabolites using a multiple ion monitoring-dependent MS/MS acquisition method on a hybrid triple quadrupole-linear ion trap mass spectrometer. J. Mass Spectrom. 2008, 43, 1364−1375. (15) Yan, Z.; Lin, G.; Ye, Y.; Wang, Y.; Yan, R. Triterpenoid saponins profiling by adducts-targeted neutral loss triggered enhanced resolution and product ion scanning using triple quadrupole linear ion trap mass spectrometry. Anal. Chim. Acta 2014, 819, 56−64. (16) Yang, W.; Ye, M.; Liu, M.; Kong, D.; Shi, R.; Shi, X.; Zhang, K.; Wang, Q.; Zhang, L. A practical strategy for the characterization of coumarins in radix glehniae by liquid chromatography coupled with triple quadrupole-linear ion trap mass spectrometry. J. Chromatogr. A 2010, 1217, 4587−4600. (17) Kim, K. H.; Ahn, Y. H.; Ji, E. S.; Lee, J. Y.; Kim, J. Y.; An, H. J.; Yoo, J. S. Quantitative analysis of low-abundance serological proteins with peptide affinity-based enrichment and pseudo-multiple reaction monitoring by hybrid quadrupole time-of-flight mass spectrometry. Anal. Chim. Acta 2015, 882, 38−48. (18) Nárvez-Cuenca, C. E.; Vincken, J. P.; Gruppen, H. Identification and quantification of (dihydro) hydroxycinnamic acids and their conjugates in potato by UPLC-DAD-ESI-MSn. Food Chem. 2012, 130, 730−738. (19) Nárvez-Cuenca, C. E.; Vincken, J. P.; Zheng, C.; Gruppen, H. Diversity of (dihydro) hydroxycinnamic acid conjugates in Colombian potato tubers. Food Chem. 2013, 139, 1087−1097. (20) Li, Y. Y.; Di, R.; Baibado, J. T.; Cheng, Y. S.; Huang, Y. Q.; Sun, H.; Cheung, H. Y. Identification of kukoamines as the novel markers for quality assessment of lycii cortex. Food Res. Int. 2014, 55, 373−380. (21) Li, Y. Y.; Di, R.; Hsu, W. L.; Huang, Y. Q.; Sun, H.; Cheung, H. Y. Sensitivity improvement of kukoamine determination by complexation with dihydrogen phosphate anions in capillary zone electrophoresis. Electrophoresis 2015, 36, 1801−1807.
bisdihydrocaffeoyl conjugations, such as kukoamines A and B, were predominant. Compounds with tridihydrocaffeoyl or hybrid phenolic conjugations were identified from LyC for the first time as the minority compounds. Our research reveals for the first time the molecular diversity of kukoamine metabolites in LyC. Our results could provide information about the biosynthesis and metabolism of polyamines in plants. The approach coupling PIS with MIM-EPI significantly enhances the qualitative capability of triple-quadrupole MS in metabolite analysis. It can also be utilized for metabolite identification in other substrates.
■
AUTHOR INFORMATION
Corresponding Author
*(H.-Y.C.) Phone: +852 34427746. Fax: +852 27887406. Present Address †
System and Translational Science Center, RTI International, Research Triangle Park, NC, USA, 27709. Funding
We are Health, (CityU Medica
thankful for funding support from the Department of Hong Kong Government SAR, China, to a project No. 9210029) on the Hong Kong Chinese Materia Standards (HKCMMS).
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank William T. Mahan for his proofreading of the manuscript and Lijun Li from SCIEX for his technical support in LC-MS.
■
REFERENCES
(1) Rogoza, L. N.; Salakhutdinov, N. F.; Tolstikov, G. A. Plant alkaloids of the polymethyleneamine series. Russ. Chem. Rev. 2005, 74, 381−396. (2) Bassard, J. E.; Ullmann, P.; Bernier, F.; Werck-Reichhart, D. Phenolamides: Bbridging polyamines to the phenolic metabolism. Phytochemistry 2010, 71, 1808−1824. (3) Funayama, S.; Yoshida, K.; Konno, C.; Hikino, H. Structure of kukoamine A, a hypotensive principle of Lycium chinense root barks. Tetrahedron Lett. 1980, 21, 1355−1356. (4) Funayama, S.; Zhang, G. R.; Nozoe, S. Kukoamine B, a spermine alkaloid from Lycium chinense. Phytochemistry 1995, 38, 1529−1531. (5) Parr, A. J.; Mellon, F. A.; Colquhoun, I. J.; Davies, H. V. Dihydrocaffeoyl polyamines (kukoamine and allies) in potato (Solanum tuberosum) tubers detected during metablite profiling. J. Agric. Food Chem. 2005, 53, 5461−5466. (6) Ponasik, J. A.; Strickland, C.; Faerman, C.; Savvides, S.; Karplus, P. A.; Ganem, B. Kukoamine A and other hydrophobic acylpolyamines: potent and selective inhibitors of Crithidia fasciculate trypanothione reductase. Biochem. J. 1995, 311, 371−375. (7) Hadjipavlou-Litina, D.; Garnelis, T.; Athanassopoulos, C. M.; Papaioannou, D. Kukoamine A analogs with lipoxygenase inhibitory activity. J. Enzyme Inhib. Med. Chem. 2009, 24, 1188−1193. (8) Liu, X.; Zheng, X. C.; Wang, N.; Cao, H. W.; Lu, Y. L.; Long, Y. P.; Zhao, K. C.; Zhou, H.; Zheng, J. Kukoamine B, a novel dual inhibitor of LPS and CpG DNA, is a potential candidate for sepsis treatment. Br. J. Pharmacol. 2011, 162, 1274−1290. (9) Liu, X.; Zheng, X. C.; Long, Y. P.; Cao, H. W.; Wang, N.; Lu, Y. L.; Zhao, K. C.; Zhou, H.; Zheng, J. Dual targets guided screening and isolation of kukoamine B as a novel natural anti-sepsis agent from traditional Chinese herb cortex lycii. Int. Immunopharmacol. 2011, 11, 110−120. (10) Li, Y. Y.; Hu, S. Q.; Huang, Y. Q.; Han, Y.; Cheung, H. Y. Preventing H2O2-induced toxicity in primary cerebellar granule 10790
DOI: 10.1021/acs.jafc.5b04321 J. Agric. Food Chem. 2015, 63, 10785−10790
