Alkaloids have been considered as the most promising bioactive ingredients in plant species from the genus Zanthoxylum. This study reports on the compositions and contents of the Zanthoxylum alkaloids (ZAs) from three Zanthoxylum species, and their potential anti-proliferation activities.
An HPLC-UV/ESI-MS/MS method was established and employed to analyze the alkaloids in different Zanthoxylum extracts. The common and unique peaks and their relative contents were summarized and compared to evaluate the similarity and dissimilarity of the three Zanthoxylum species. Meanwhile, inhibitory activity tests to four carcinoma cell lines, i.e., stomach tumor cells (SGC-7901), cervical tumor cells (Hela), colon tumor cells (HT-29) and Hepatic tumor cells (Hep G2), were carried out in vitro to evaluate the bioactivities of the ZAs.
Seventy peaks were detected in the crude total alkaloid samples, and 58 of them were identified. As a result, 13 common peaks were found in the extracts of all the three Zanthoxylum species, while some unique peaks were also observed in specific species, with 17 peaks in Z. simulans, 15 peaks in Z. ailanthoides and 11 peaks in Z. chalybeum, respectively. The comparison of the composition and relative contents indicated that alkaloids of benzophenanthridine type commonly present in all the three Zanthoxylum species with high relative contents among the others, which are 60.52% in Z. ailanthoides, 30.52% in Z. simulans and 13.84% in Z. chalybeum, respectively. In terms of activity test, Most of the crude alkaloids extracts showed remarkable inhibitory activities against various tumor cells, and the inhibitory rates ranged from 60.71 to 93.63% at a concentration of 200 μg/mL. However, SGC-7901 cells seemed to be more sensitive to the ZAs than the other three cancer cells.
The alkaloid profiles detected in this work revealed significant differences in both structures and contents among Zanthoxylum species. The inhibitory rates for different cancer cells in this study indicated that the potential anti-cancer activity should be attributed to quaternary alkaloids in these three species, which will provide great guidance for further exploring this traditional medicinal resource as new healthcare products.
Various plants of the genus Zanthoxylum have been used not only as anti-flooding roundworm and tooth pain in ancient China, but also as flavoring and antiseptic additives. The main chemical components of these plants from this genus, such as Z. nitidine, Z. bungeanum, Z. simulans and Z. armatum, could be classified as volatile oils, lignins, flavonoids and alkaloids [1,2,3,4], in which alkaloids from the genus Zanthoxylum (ZAs) were considered to be the most potent anti-tumor ingredients. For some known ZAs, the existing studies showed several mechanisms to further explain the anti-tumor activities, e.g., decarine, 6-acetonyldihydrochelerythrine and zanthocapensine induced colon cancer cells (HCT116) apoptosis by activating casepase-3 pathways ; Sanguinarine, Chelerythrine and Chelidonine exhibited anti-proliferative activity against mouse lymphocytic leukemia cells (L1210) by damaging leukemia DNA ; nitidine and angoline showed inhibitory activity against carcinoma cells by suppress c-Src/FAK  and IL-6/STAT3  pathways. Although the pharmacological studies have boosted the progress in exploring this medicinal resource, the investigation of the chemical basis of different plants of the genera Zanthoxylum, and the screening of new potential bioactive compounds from this genus are still hotspots for further exploring new anti-tumor drugs or healthcare products from the medicinal plants of this genus.
The conventional way to characterize ZAs usually involves the isolation of pure compounds and subsequent structures determination based on their NMR spectra , which is not only time and labor consuming, but also inefficient and insensitive to some ZAs in the characterization process, especially to those in very minute amount. To circumvent these limitations, a series of instrumental methods have been applied for the quantitative and qualitative analysis of ZAs, such as HSCCC (high-speed Countercurrent Chromatography), CE (Capillary electrophoresis), GC-MS, LC-MS, and LC-MS-NMR [10,11,12,13,14,15,16,17,18]. Due to the high performance in separation and sensitivity of alkaloids in the positive mode during the electro-spray ionization, LC-ESI-MS has become one of the most powerful methods for the analysis of alkaloids from plants of these species [19, 20].
In this work, an HPLC-UV-ESI-MS/MS method was developed to analyze the crude alkaloids from Z. ailanthoides, Z. simulans and Z. Chalybeum. Combining with chemical fingerprinting analysis of the crude extracts, the corresponding anti-proliferative activity tests were also conducted for the same samples against hepatic cancer (Hep G2), gastric cancer (SGC-7901), colon cancer (HT-29) and cervical cancer (Hela) cells in order to correlate the contribution of different types of alkaloids to their diverse anti-proliferative activities. This work may lay the groundwork for the further understanding of anti-cancer compositions from these three Zanthoxylum species.
Chemicals and reagents
Formic acid, ammonium, methanol, triethylamine, chloroform, acetic acid, trichloroacetic acid and dimethyl sulfoxide (DMSO) were of analytical grade and purchased from Sino-pharm Chemical Reagent CO. (Beijing, China). Water for HPLC and LC-MS was prepared with EPED (Nanjing Yeap Esselte Technology Development Co., Nanjing, China). The medium, DMEM, Fetal calf Serum and Pancreatic enzymes, for Carcinoma cell culture were obtained from Wuhan Feiyi Science and technology Co. (Wuhan, China).
Fresh plant materials of two Zanthoxylum species (Z. simulans, and Z. ailanthoides) were collected from Wuhan Botanical Garden in April, 2015. Fresh stem bark of Z. chalybeum was collected from Kenya. The authentication and identification of the specimens was kindly assisted by the taxonomist (Prof. Guangwan Hu) of Key Laboratory of Plant Germplasm Enhancement and Specialty Agriculture (Wuhan Botanical Garden), Chinese Academy of Sciences. All of voucher specimens (No. 001–003) were deposited in the herbarium of the Key Laboratory. After dried at 30 °C, the plant materials were crushed and stored at room temperature until use.
After being immersed in 70% ethanol (with 0.5% HCl) for 12 h, plant materials of these three species were ultrasonically extracted for 30 min in thrice. The extracts were combined and condensed to form dark residues. Then, the resulted residues were dispersed in water, and extracted with chloroform to remove chlorophyll after the pH of water phase was adjusted to 9–11. Finally, chloroform phase was combined and concentrated to afford crude alkaloids.
HPLC and LC-MS conditions
A Thermo Accela 1250 HPLC equipped with an auto-sampler and a UV-visible detector (Thermo Fisher Scientific, San Jose, CA, USA) was employed for the analysis of crude alkaloids. A 10 μL sample was analyzed on a Phenomenex Kinetex column (2.6 μm, C18, 100 × 2.1 mm) at 25 °C. The flow rate was 0.2 mL/min and the chromatograms were recorded at the wavelength of 280 nm. 0.5% formic acid solution (adjusted to pH = 4.5 by ammonium, A) and acetonitrile (B) were selected as mobile phase, and the gradient was set as follows: 0–25 min, 5–20% (B); 25–40 min, 20% (B); 40–55 min, 20–35% (B); 55–65 min, 35–80% (B).
For ESI-MS/MS analysis, a Thermo Accela 600 HPLC system with both UV detector and TSQ Quantum Access MAX mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA) was used for the LC-MS analysis in the positive mode. The mass condition was set as follows: mass range from 200 to 800 Da; Spray Voltage, 3.0 kV; Capillary temperature, 350 °C; Sheath gas pressure, 40 psi; Aux gar pressure, 10 psi.
Qualitative and quantitative analysis
The alkaloids were identified based on the comparison of detected MS/MS spectra with reported literatures, MS/MS spectra of reference standards and their fragment pathways. Finally, they were semi-quantified by the relative peak areas in the HPLC chromatograms.
In vitro anti-proliferative activity assay
Four Human cancer cell lines i.e., stomach tumor cells (SGC-7901), cervical tumor cells (Hela), colon tumor cells (HT-29) and hepatic tumor cells (Hep G2), obtained from the China Center for Type Culture Collection (CCTCC), were used for anti-proliferative activity tests with the crude extracts using Sulforhodamine B (SRB). These cancer cells were incubated into 96-well plate under 5% CO2 at 37 °C for 24 h and treated with crude ZAs at a concentration of 200 μg/mL in three duplicates, and DMSO was used as negative control. To make sure the security of the cells, the final concentration of DMSO was less than 0.1%. After incubation for 72 h, the supernatant was discarded, and 100 μL 10% trichloroacetic acid (TCA) was added into the wells and washed after the cells were fixed for 1 h. SRB was used to dye cells, and 1% acetic acid was used to remove the excess amount of SRB. The optical density (OD) values were determined at 565 nm [21, 22]. The inhibitory rate (%) equals to (ODC-ODT)/ODC × 100%, where ODT and ODC are the OD values of the blank and the alkaloids extracts, respectively.
Results and discussion
Optimization of HPLC conditions
The previous phytochemical investigation of ZAs indicated that the structure diversity and complexity may cause much difficulty in HPLC separation. To get better chromatograms of the crude alkaloids, the HPLC conditions were optimized for the analysis of the extract of Z. ailanthoides. Using one-variable method, the chromatographic column was firstly optimized by successively separating the crude extract on phenomenex kinetex C18 (100 × 2.1 mm, 2.6 μm) and Phenomenex ODS column (150 × 2.00 mm, 5 μm). As a result, phenomenex kinetex C18 (100 × 2.1 mm, 2.6 μm) was found to have higher column efficiency, and thus selected for the further sample analysis. Considering the significant effect of mobile phase and additives on alkaloids separation and peak shape, the mobile phase conditions based on the reported systems, such as 0.1% formic acid - 0.1% formic acid acetonitrile, 50 mM ammonium acetate-methanol, 0.5% acetic acid and 0.1% triethylamine-acetonitrile, 0.1% formic acid-acetonitrile and 0.5% formic acid (pH adjusted to 4.5 by ammonia)-methanol-acetontrile [13, 15, 23,24,25], were also investigated. As preferred, 0.5% formic acid (pH adjusted to 4.5 by ammonia)-acetonitrile was found to be the best mobile phase system for the analysis of the crude alkaloids extracted from Z. ailanthoides. At last, the final optimized conditions were achieved for the fingerprinting analysis of ZAs from these three species.
Under the optimized HPLC conditions, the crude alkaloids of these three species (Z. ailanthoides, Z. simulans and Z. chalybeum) were analyzed, in which the chromatograms were shown in Fig. 1. It was observed that 70 peaks in the three extracts have been well separated and detected, indicating that the operation conditions adopted were suitable for the analysis of ZAs. Meanwhile, a preliminary study of the chromatograms revealed that the alkaloids compositions and contents exhibited significant differences among the three species.
Identification of Zanthoxylum alkaloids
Among the detected peaks, 58 alkaloids were identified based on the reported literatures, standard spectra or fragment pathways, 31 of which (labeled as “b” in Table 1) were identified by comparing the MS/MS spectra with reported literatures. Based on the identification, the structures of the 58 ZAs were shown in Fig. 2. To the best of our knowledge, 15 of the ZAs (labeled as “a” in Table 1) were firstly reported in the genus zanthoxylum, and six (peaks 4, 6, 13, 31, 35 and 37) were found to be alkaloids glycoside. Since different types of ZAs have distinguishing fragment pathways in ESI-MS/MS, we classified the alkaloids into eight types to simplify the structures interpretations, i.e., quinolone and iso-quinolone type, benzyltetrahydroisoquinoline type, benzophenanthridine type, dihydrobenzophenanthridine type, berberine type, tetrahydroproberberine type, protopine type and aporpine type [26,27,28]. The interpretations on their MS/MS data were discussed as below.
Seven ZAs (peaks 31, 48, 51, 54, 60, 62 and 68) were identified and classified as benzophenanthridine type. With four benzene rings conjugated into the major skeleton, alkaloids of this type produced abundant fragments by neutral loss of substituents in the positive ESI-MS/MS .
Peaks 51, 54, 60 and 62 could be identified as 8-O-demethylchelerythrine, sanguinarine, nitidine and chelerythrine, respectively, by comparing their MS/MS spectra with the published literatures [11, 15, 26]. For peak 51, the most intensive fragment at m/z 319 was observed due to the neutral loss of ∙CH3 from the molecular ion ([M + H]+ at m/z 334) at N moiety. The successive loss of CH3 (15 Da) and CO (28 Da) from the fragment ion at m/z 319 led to the fragments at m/z 304 and 291, respectively. Combined with the further loss of CO (28 Da) from fragment at m/z 304, peak 51 can be definitely identified as 8-O-demethylchelerythrine . As to sanguinarine (peak 54), its fragment ion at m/z 317 was produced by the loss of CH3 from the molecular ion ([M + H]+ at m/z 332). The most abundant fragment ion at m/z 274 was observed due to the neutral loss of CO from fragment ion at m/z 304. Therefore, peak 54 can be definitely identified as sanguinarine [11, 26, 27]. With same parent ion [M + H]+ at m/z 348, and the same fragment ions with relative intensity at m/z 332, 318, 304 and 290, peak 60 and 62 could be identified as nitidine or chelerythrine based on the investigation and comparison of their MS/MS spectra with those reported literatures [11, 15, 26, 27]. However, the distinguished retention time of peak 60 and 62, which are 45.22 min and 48.20 min, can be used to differentiate them. Separated by Reversed-Phase Chromatography, the different retention time provided further evidence for identifying peaks 60 and 62 as nitidine and chelerythrine, respectively .
For peak 48, its fragment ions at m/z 350, 320 and 292 were 2 Da more than fragment ions at m/z 348, 318 and 290 of nitidine. The different fragment ions at m/z 335 and 307 could be deduced by successive neutral loss of CH3 (15 Da) from N and CO (28 Da) from A ring of the fragment ion at m/z 350. Thus, peak 48 could be identified as fagaronine [25, 29]. With the same fragment ions as peak 48 at m/z 350, 335, 320, 307 and 292, the parent ion ([M + H]+) for peak 31 at m/z 512 indicated the presence of a glycoside. Thus, peak 31 was identified as fagaronine-3-glucoside . In the MS/MS spectrum of peak 68, fragment ions at m/z 305 and 277 were 30 Da less than those at m/z 335 and 307 of fagaronine. Thus, peak 68 could be proposed as 7,8-dimethylchelerythrine .
In this study, 12 peaks (peaks 28, 30, 33, 43, 56, 58, 61, 63, 64, 66, 69 and 70) were identified and classified into dihydrobenzophenanthridine type. Compared with alkaloids of benzophenanthridine type, the N7 - C8 double bond of B ring was reduced to be a single bond in the skeleton of this type. Thus, the characteristic fragment ion was always observed to be [M + H-CH4]+ in the positive mode from the MS/MS spectra .
Compared the MS/MS data with those from authentic standards, peaks 43, 61, 63, 69 and 70 were identified as 6,10-dihydroxyl-3- demethyldihydrobenzophenanthridinium, dihydrochelirubine, oxychelerythrine, dihydrosanguinarine and 6-hydroxy-10-methylsanguinarine , respectively. With [M + H]+ at m/z 384, peak 43 produced the most abundant fragment ion at m/z 352 by neutral loss CH3OH (32 Da) of from D ring, and the further successive loss of CH3 (15 Da) and OH (17 Da) led to fragment ions at m/z 337 and 320. Thus, peak 43 was identified as 6,10-dihydroxyl-3-demethyldihydrobenzophenanthridinium . For peaks 61 and peak 63, same molecular ion [M + H]+ at m/z 364 and fragment ions at m/z 348 and 334 were observed in the MS/MS spectra of these two peaks. However, in terms of the different fragments at m/z 332, 320, 304 and 290 for peak 61 and m/z 316, 306, 289 and 270 for peak 63, peaks 61 and 63 were identified as dihydrochelirubine  and oxychelerythrine , respectively. For peak 69, the most abundant fragment ion at m/z 318 was produced corresponding to the loss of CH4 from N moiety, due to the double bond on B ring reduced. With the same fragment ions at m/z 304 and 290 as peak 61, the loss of CO (28 Da) from fragment ion at m/z 304 resulted in the different fragment at m/z 276. Therefore, peak 69 was assigned as dihydrosanguinarine .
As for peak 70, a neutral loss of H2O from the molecular ion at m/z 380 [M + H]+ resulted in fragment ion at m/z 362. The cleavage of B ring led to the most abundant fragment at m/z 289 by loss of C2H5N (43 Da). Thus, peak 70 was definitely identified as 6-hydroxy-10-methylsanguinarine . With the same molecular ion [M + H]+ at m/z 380 as peak 70, fragment ions of peak 58 at m/z 349, 306 and 292 were observed. The neutral loss of substitute CO2 (44 Da) from fragment ion at m/z 364 yielded the most intensive fragment ion at m/z 320. Thus, peak 58 was deduced as 8-methoxy-dihydrochelerythrine or 8-methoxy-dihydronitidine . In the positive mode, the alkaloid at 55.71 min (peak 64) yielded fragment ions at m/z 364, 336, 306, 291 and 277, which were 2 Da more than fragment ions at m/z 362, 334, 304, 289 and 275 in peak 70. Thus, this alkaloid (peak 64) could be proposed as 6-methoxy-dihydronorchelerythrine N-oxide or 6-methoxy-dihydrorhoidoline B . According to the MS/MS spectra of peak 33, fragment ions at m/z 350 and 334 were produced by the loss of CH4 (14 Da) and CH4O (32 Da) from B ring, respectively. By B ring rearrangement and further neutral loss of C2H4O (44 Da), fragment ion at m/z 322 was produced. Thus, peak 33 was deduced as 6-methyl-5,6-dihydrofagaridine .
With molecular ion [M + H]+ at m/z 352 for peak 28 and 354 for peak 30, same fragment ions at m/z 322, 308 and 280 were observed, while fragments at m/z 339 and 293 in the MS/MS spectra of peak 30 were 2 Da more than the fragment ions of peak 28 at m/z 337 and 291. The loss of H2O from fragment ion at m/z 337 resulted in fragment ion at m/z 319, indicating the presence of an OH on the parent molecular ion. In addition to the 2 Da discrepancy of the molecular and fragment ions, peak 28 and 30 were deduced as buegenine and 2-dimethoxy-3- hydroxybeugenine . Peak 66 had [M + H]+ at m/z 336 and fragment ions at m/z 321, 293, 278 and 264 were 2 Da more than fragment ions at m/z 319, 291, 276 and 262 of 8-O-demethylchelerythrine (peak 51) . As the MS/MS spectra and proposed fragment pathways shown in Fig. 3, peak 66 was identified as 8-O-demethyl-dihydrochelerythrine . As to peak 56, the same fragment ions to peak 66 indicated the similarity of their structures. The loss of a glycosyl (162 Da) implied the compound to be an alkaloid glucoside. Hence, peak 56 had a molecular ion [M + H]+ at m/z 498 and was identified as 8-O-demethyl-dihydrochelerythrine-9-O-glucoside .
Compared with the benzophenanthridine type alkaloids, the structural skeleton of berberine type of alkaloids had double bond of C5 - C6 reduced. In the MS/MS spectra of berberine type of alkaloids, most fragment ions were successively generated by neutral loss of substituents . Based on the reported literatures, five alkaloids (peaks 39, 44, 52, 53 and 65) were identified and classified into this group.
By comparing with the reported literatures, four alkaloids (peaks 44, 52, 53 and 65) were identified as jatrorrhizine (peak 44) , palmatine (peak 52) [32, 33],berberine (peak 53) [26, 27] and berberrubine (peak 65) , respectively. In terms of peak 39, fragment ions at m/z 368, 352, 338 and 310 were 30 Da more than fragment ions at m/z 338, 322, 308 and 280. The gap between the two peaks indicated the structural similarity and might be attributed to a CH2O (30 Da) lose. In addition to the investigation of substituents on this skeleton, such as 13-hydroxylberberruine and 13-methylberberine , C-13 was the most proposed position. Thus, peak 39 could be assigned as 10-hydroxy-2,3,9,12-tetramethoxy-jatrorrhizine .
Alkaloids of tetrahydroberberine type produced their characteristic and most intensive fragment ions typically via Retro-Diels-Alder (RDA) cleavages . Based on the detected MS/MS data, six alkaloids (peaks 23, 26, 27, 40, 45 and 47) were identified and classified into this type.
Based on the reported MS/MS spectra, four alkaloids were identified as tetrahydropalmatine (peak 23), N-methylcanadine or its isomers (peak 40 and 47) and N-methyltetrahydropalmatine (peak 45), respectively [25, 26]. In the MS/MS spectra of peak 23, fragment ions at m/z 192 and 163 were observed due to the C ring cleavage. The further neutral loss of CH2 (14 Da) from fragment ion at m/z 163 led to the fragment ion at m/z 149. In addition, another abundant fragment ion at m/z 177 was produced by RDA cleavage of B ring. Thus, this alkaloid (peak 23), with molecular ion [M + H]+ at m/z 356, was definitely identified as tetrahydropalmatine . In terms of peaks 40 and 47, the characteristic fragment ions at m/z 190 and 175 were 2 Da less than fragment ions at m/z 192 and 177 of peak 23. With [M + H]+ at m/z 354, peaks 40 and 47 had same fragment ions and could be identified as N-methylcanadine or its isomers . For peak 45, the most abundant fragment ion at m/z 206 was 14 Da more than fragment ion at m/z 192 of peak 23, which indicated the presence of a CH2 on A or B ring. Therefore, peak 45, with parent ion [M + H]+ at m/z 370, was identified as N-methyltetrahydropalmatine .
Peak 26 had the same fragment ions at m/z 192, 177 and 149 as peak 23. The further cleavage of B ring from fragment ion at m/z 192 led to the distinguished fragment ion at m/z 151, which corresponded to a CH3 combined with N moiety. Thus, peak 26 with parent ion [M + H]+ at m/z 356 was identified as N-methyltetrahydrocolumbamine . Peak 27 produced characteristic fragment ions at m/z 190 and 149 by RDA cleavage at C ring. As a result of further loss of CH2O (30 Da) from fragment ion at m/z 149, fragment ion at m/z 119 was observed. The successive neutral loss of CH3 and CH6O from the molecular ion [M + H]+at m/z 340 resulted in fragment ions at m/z 325 and 291. Therefore, peak 27 was deduced as 2-methoxy-5-dehydroscoulerine . The MS/MS spectra and proposed fragment pathway of peak 27 was shown in Fig. 4.
Compared with tetrahydroproberberine type alkaloids, alkaloids of this type have carbonyl at C14, other than N7-C14. As for MS/MS spectra in ESI-MS/MS, major fragment ions were produced by the loss of H2O (18 Da) from parent ion [M + H]+and further RDA cleavage . By comparing with reported literatures, peaks 9 and 42 were classified into protopine type and identified as hunnemannine  and α-allocryptopine [26, 37], respectively. Peaks 9 and 42 had the same fragment ions at m/z 206, 188 and 149, which were abundant and generated mainly by RDA cleavage. To distinguish the two peaks, fragment ion at m/z 338 of peak 9 yielded fragment ion at m/z 323 by neutral loss of CH3 (15 Da) from N. As to peak 42, the loss of CH4O (32 Da) from fragment ion at m/z 352 led to fragment ion at m/z 320. Thus, peaks 9 and 42, with molecular ion [M + H]+ at m/z 356 and 370, were identified as hunnemannine and α-allocryptopine, respectively.
Due to the loss of substituents on A and D ring and N vicinal moieties through B ring cleavage, most abundant fragment ions of aporphine type were produced in the MS/MS spectra. By comparing with reported literatures, five peaks (peaks 11, 20, 29, 32 and 34) were classified into aporphine type and identified as magnoflorine , laurifoline , menisperine , 10-demethyl-magnoflorine  and Xanthoplanine , respectively.
With molecular ion [M + H]+ at m/z 342, fragment ions at m/z 282, 265, 237, 222 and 191 were observed both in the MS/MS spectra of peak 11 and 20. Different fragment ion at m/z 233 was generated by the neutral loss of CH4O (32 Da) from fragment ion at m/z 265. Another fragment ion [M + H-C2H7N-2CH4O-CO]+ at m/z 205 was found to be 2 Da less than fragment ion at m/z 207 in the MS/MS spectra of peak 11. Therefore, peaks 11 and 20 were identified as magnoflorine and laurifoline by comparing the fragments relative intensity and differences, respectively . As to peaks 29 and 34, same fragment ions at m/z 311, 296, 219, 192 and 177 were observed. With intensive and same fragment ions at m/z 279, 264, 248 and 236, peak 29 had retention time at 21.71 min and was identified as menisperine . For peak 34, the loss of N vicinal moieties and substituents from A and D ring led to fragment ions at 280, 265, 253 and 237. Thus, peak 34 had retention time at 24.70 min and was identified as Xanthoplanine . In reference to peak 32, fragment ions at m/z 251, 208 and 191 were 14 Da less than fragment ions at m/z 265, 222 and 205 in the MS/MS spectra of peak 20. The most intensive fragment ion at m/z 121 indicated the presence of 2OH on D ring. Thus, peak 32 was proposed as 10-demethyl-magnoflorineby comparing with the reported data .
Biosynthesis analysis of ZAs provided evidence of benzyltetrahydroisoquinoline type alkaloids as precursor of alkaloids, such as alkaloids of benzophenanthridine type, berberine type and aporphine type . As to the MS/MS spectra of benzyltetrahydroisoquinoline type of alkaloids, fragment ions with most abundant intensity were produced by the neutral loss of N vicinal moieties and substituents on iso-quinoline ring . Among the detected alkaloids, ten peaks (4, 6, 7, 8, 12, 15, 18, 21, 22 and 36) were identified and classified into benzyltetrahydroisoquinoline type.
By comparing with published MS/MS data, peaks 7, 8, 18, 21 and 36 were identified. Same fragment ions and relative intensity were observed in the MS/MS spectra of peaks 7, 21 and 36. The loss of benzyl from B ring led to fragment ions at m/z 209 and 107. Due to the cleavage of B ring, fragment ions at m/z 166 and 137 were also observed. The successive loss of C2H7N (45 Da) and CH4O (32 Da) from the parent ion [M + H]+ at m/z 314 generated fragment ions at m/z 269 and 237. Thus, the three compounds (peaks 7, 21 and 36) were identified as magnocurarine or its isomers . In reference to peak 6, the alkaloids had molecular ion [M + H]+ at m/z 476 and same fragment ions with peak 7. The mass gap between parent ion and fragment ion at m/z 314 indicated the possible presence of a glycosyl. Therefore, with retention time at 11.67 min, peak 6 was deduced as magnocurarine-6-glucoside . Due to the loss of benzyl in the MS/MS of peak 8, the characteristic fragment ion at m/z 107 was observed. Moreover, the same fragment ions to peak 7 at m/z 269 and 237 showed the structural discrepancy at N vicinal moieties. Thus, peak 8, with the molecular ion at m/z 286, was identified as 7-methoxyhigenamine based on its reported mass data . As to peak 18, the loss of benzyl and its substituents yielded fragment ions at m/z 192 and 137, and the further loss of CH3 (15 Da) from isoquinoline ring produced the fragment ion at m/z 177. Due to the loss of substituents and N vicinal moieties, fragment at m/z 207 [M + H-C7H7O2]+ generated fragment ions at m/z 175, 143 and 115 by similar fragment pathway with peak 8. Thus, peak 18, with molecular ion [M + H]+ at m/z 330, was identified as reticuline .
With retention time at 12.28 min, peak 12 had fragment ions at m/z 269, 254, 237, 160, 143 and 107, which was almost the same to peak 8. In contrary to the loss of NH3 (17 Da), fragment ion at m/z 269 was produced by the loss of CH3-NH2 (31 Da). To display the fragment pathway of this type of alkaloids in detail, the MS/MS spectra and fragment pathway of peak 12 was shown in Fig. 5. Thus, peak 12, with the molecular ion [M + H]+ at m/z 300, was deduced as 7-methoxy-2-methylhigenamine . The same fragment ions of MS/MS spectra from peak 4 and peak 12 indicated the structure similarity between the two alkaloids. In addition to the molecular ion [M + H]+ at m/z 462, fragment ion at m/z 300 was produced by the loss of a glycosyl. Thus, peak 4, with retention time at 8.83 min, was deduced as 7-methoxy-2- methylhigenamine-4′-glucoside or 7-methoxy-2-methylhigenamine-6-glucoside by comparing with the mass data of 7-methoxy-2-methylhigenamine . For the fragment ions of peak 15, the characteristic loss of benzyl and its substituents led to fragment ions at m/z 207 and 137. Moreover, the N vicinal moiety loss generated fragment ion at m/z 301, of which the further substituents loss yielded fragment ions at m/z 269 [M + H-C2H5N-CH4O]+ and 239 [M + H-C2H5N-CH4O-CH2O]+. Therefore, peak 15 was speculated as isotembetarine or xylopinidine [25, 35]. In reference to peak 22, the characteristic loss of benzyl generated the fragment ions at m/z 221 and 137. The molecular ion [M + H]+ at m/z 358 and fragment ion at m/z 221 in the MS/MS spectra were 14 Da more than [M + H]+ at m/z 344 and fragment ion at m/z 207 of peak 15, respectively. The mass gap between the two compounds indicated a methyl substituent discrepancy on the isoquinoline ring. Thus, with retention time at 19.92 min, peak 22 could be deduced as 8-methoxy-isotembetatrine .
Quinoline or Isoquinoline type
Ten peaks (1, 2, 3, 5, 13, 19, 24, 35, 37 and 67) were identified and classified into quinoline or isoquinoline type. In addition to the relative fragment ions intensity of peak 1, fragment ions at m/z 177 and molecular ion [M + H]+ at m/z 192 were also observed in the MS/MS spectra of tetrahydropalmatine. The successive neutral loss of substituents led to fragment ions at m/z 162 [M + H-CH2O]+ and 148 [M + H-CH2O-CH2]+. Therefore, peak 1 was deduced as 6,7-dimethoxy- 3,4-dihydroisoquinoline . With retention time at 6.11 min, fragment ions of peak 2 at m/z 190, 162 and 132 were observed due to the successive loss of CH4 (16 Da),CO (28 Da) and CH3OH (31 Da) from parent ion. The further loss of N moiety and substituent from fragment ion at m/z 190 led to fragment ions at m/z 173 and 145. Therefore, peak 2, with molecular ion at m/z 206, was tentatively identified as edulitine . For peak 3, the cleavage of isoquinoline ring resulted in fragment ions at m/z 169 and 145 by the loss of N vicinal moiety from fragment ion at m/z 167. Intensive fragment at m/z 117 was generated by further loss of CO (28 Da) from fragment ion at m/z 145. As a result, peak 3, with retention time at 7.39 min, was tentatively assigned as thalifoline . Compared with peak 3, the discrepancy fragment ion at m/z 149 obtained from peak 5 explored the differences of the two alkaloids on the isoquinoline ring. Taking the neutral loss of NH3 (17 Da) from fragment ion at m/z 162 into consideration, the lack of substituent on N was proved. Thus, peak 5, with molecular ion [M + H]+ at m/z 194, was deduced as Des-N-methyl-thalifoline .
As to peak 67, the successive loss of CH3 (15 Da),CH2O (30 Da) and C6H4 (76 Da) from parent ion led to fragment ions at m/z 257, 242 and 196, and fragment ions at m/z 164 and 149 were produced by successive loss of CH3OH (32 Da) and CH3 (15 Da) from fragment ion at m/z 196. Therefore, peak 67, with molecular ion [M + H]+ at m/z 272, was deduced as 1,2-dihydroxy-3- methoxy-10-methyl-9-acridone . With same fragment ions as peak 67, peak 13 had molecular ion [M + H]+ at m/z 434, which revealed the presence of a glycosyl in this alkaloids. Thus, with retention time at 15.85 min, peak 13 was deduced as 2-dihydroxy-3-methoxy-10-methyl-9-acridone-1-glucoside or 1-dihydroxy-3-methoxy-10-methyl-9-acridone-2-glucoside . For peak 24, RDA cleavage of C ring produced the most intensive fragment ion at m/z 188, and the further loss of CO (28 Da) yielded fragment at m/z 160. In addition, fragment ions at m/z 134 and 106 were observed due to the cleavage of B ring and substituent loss. Thus, with molecular ion [M + H]+ at m/z 260, peak 24 could be tentatively identified as ribalinine . Another alkaloid (peak 35) shared same fragment ions to peak 24. Fragment ion at m/z 260 indicated a loss of a glycosyl from parent ion [M + H]+ at m/z 422. Thus, with retention time at 25.33 min, peak 35 could be tentatively deduced as ribalinine 3′-glucoside . Peak 19 had the same fragment ions at m/z 242, 188 and 160 as peak 24, indicating the structure similarity of the two compounds. The cleavage of C ring led to fragment ions at m/z 218 [M + H-C4H8O]+ and 160 [M + H-C6H10O3]+, which confirmed the difference on C-4′. Therefore, with molecular ion [M + H]+ at m/z 290, peak 19 could be deduced as 4′-methoxy-ribalinine . Peak 37 was detected at 25.92 min, and displayed same fragment ions with peak 19. As a result of the loss of glycosyl, fragment ion at m/z 290 was observed. Thus, peak 37, with molecular ion [M + H]+ at m/z 452, was deduced as 4′-methoxy-ribalinine-3′-glucoside .
Comparison of ZAs detected among the three species
The fingerprinting profiles of ZAs in Fig. 1 showed a preliminary difference in the alkaloids among the three species (Z. ailanthoides, Z. simulans and Z. chalybeum). Since both qualitative and quantitative information may help to understand the most bio-active and potential alkaloids, the common and unique peaks for the three species were summarized in Fig. 6a. As a result, 13 common peaks (i.e., 3, 5, 11, 17, 18, 34, 39, 42, 46, 51, 62, 68 and 70) were found in the extracts of all the three Zanthoxylum species, while some unique peaks were also observed in specific species, which are 17 peaks in Z. simulans, 15 peaks in Z. ailanthoides and 11 peaks in Z. chalybeum, respectively. Moreover, three alkaloids (peaks 7, 14 and 22) were only found in Z. ailanthoides and Z. simulans, six alkaloids (peaks 2, 21, 28, 30, 60 and 64) only in Z. ailanthoides and Z. chalybeum, and five alkaloids (peaks 6, 15, 16, 47 and 53) only in Z. simulans and Z. chalybeum. To some extent, these unique peaks may be used as chemical markers to differentiate these three species for quality control purposes.
Based on the alkaloids identification, the peak areas of each type of alkaloids from the three species (Z. ailanthoides, Z. simulans and Z. chalybeum) were summarized to explore quantitative differences of the ZAs, as shown in Fig. 6b. For alkaloids from Z. ailanthoides, benzophenanthridine type (60.52%) and quinoline or isoquinoline type (20.66%) were the most abundant alkaloids, followed by benzyltetrahydroisoquinoline type (4.88%), dihydrobenzophenanthridine type (2.94%), berberine type (2.17%), aporphine type (2.15%), tetrahydroproberberine type (0.42%) and protopine type (0.25%). For Z. simulans, benzophenanthridine type and benzyltetrahydroisoquinoline type contributed to 30.52% and 23.49%, followed by aporphine type (15.22%), berberine type (11.10%), tetrahydroproberberine type (9.28%), quinoline or isoquinoline type (8.30%), protopine type (1.59%) and dihydrobenzophenanthridine type (0.49%), respectively. For the alkaloids in Z. Chalybeum, the major alkaloids were in protopine type at relative content of 59.88%, together with benzyltetrahydroisoquinoline type at 14.42%, benzophenanthridine type at 13.84%, dihydrobenzophenanthridine type at 8.09%, tetrahydroproberberine type at 1.35%, quinoline or isoquinoline type at 1.02%, protopine type at 0.76% and aporphine type at 0.64%, respectively. Notably, based on the analysis of composition and contents of ZAs in the three species, alkaloids of benzophenanthridine type were found to be the most common ZAs, which accounts for over 14% in all the three species.
Anti-proliferative activities of ZAs against carcinoma cells
To explore the correlation between ZAs and their potential bioactivities, in vitro inhibitory activity tests against cancer cells have been performed on the crude extracts of alkaloids from the three Zanthoxylum species, using Sulforhodamine B (SRB) method to test four cancer cells, i.e., human gastric cancer cells (SGC-7901), cervical cancer cells (Hela), human colorectal adenocarcinoma cells (HT-29) and human hepatocyte carcinoma cells (Hep G2). As shown in Fig. 7, ZAs showed remarkably proliferative inhibition to all the four cancer cell lines, which ranged from 60.71 to 93.63% at a concentration of 200 μg/mL. An exception was found on the crude alkaloids of Z. ailanthoides with Hep G2 inhibitory rate at 16.83%. To some extent, SGC-7901 seemed to be more sensitive to the ZAs than the other three cancer cells.
It was well documented that the cancer cell inhibitory rate was not only relevant to the composition and content of the testing chemicals, but also to the water solubility of the chemicals , which inspired us to focus on the water-soluble ZAs. Quaternary alkaloids in Zanthoxylumgenus have been recognized as promising anti-cancer candidates by penetrating through the carcinoma cell membrane  and attracting the negative electron on DNA . Therefore, in this study, the percentage of quaternary alkaloids from the three Zanthoxylum species has been further evaluated for their contribution to the cancer cell inhibitory rate. As a result, with positive charge on the molecule, 27 ZAs have been classified into quaternary alkaloids from the three Zanthoxylum species, i.e., peaks 6, 7, 11, 15, 20, 21, 22, 26, 29, 31, 32, 34, 36, 39, 40, 44, 45, 47, 48, 51, 52, 53, 54, 60, 62, 65 and 68, which belongs to benzophenanthridine type, aporphine type, berberine type, benzyltetrahydroisoquinoline type and tetrahydroproberberine type according to their structural skeleton. As shown in Fig. 7b, other than non-quaternary and unidentified alkaloids, the percentages of quaternary alkaloids were 83.4%, 58.6%, 52.4% for Z. Chalybeum, Z. ailanthoides, Z. simulans, respectively. Since high content of quaternary alkaloids and high cancer cell inhibitory rates were simultaneously found on bioactivity tests of ZAs, the quaternary alkaloids could be proposed as the most potential bioactive compounds in the genus Zanthoxylum, which will provide great guidance to the further exploitation of this Chinese herb medicine as new drugs or healthcare products.
Based on the HPLC-UV/ESI-MS/MS method developed in this work, a fingerprinting comparison was conducted on ZAs from Z. ailanthoides, Z. simulans and Z. chalybeum. As a result, remarkable differences were shown on the fingerprints in both the compositions and contents of alkaloids among the three Zanthoxylum species. In more details, a total of 70 peaks corresponding to alkaloids were detected, of which 58 peaks were identified by comparing their MS/MS spectra with the reported literatures. More interestingly, 27 quaternary alkaloids and 6 alkaloid-glycosides were reported for the first time. Moreover, carcinoma cell inhibitory rates to four cancer cell lines were investigated to evaluate the contribution of some distinct compositions. Consequently, quaternary alkaloids of ZAs were proposed as the most potent proliferative inhibitors against cancer cells and crude ZAs showed more sensitive to SGC-7901 cells than the other cancer cells. To the best of our knowledge, this work provides the most details on ZAs and their potential activities, and offered valuable information to further explore Zanthoxylum alkaloids in this genus as new healthcare products in the near future.
Guo T, Xie H, Deng YX, Pan SL. A new lignan and other constituents from Zanthoxylum armatum DC. Nat Prod Res. 2012;26:859–64.
Mansoor TA, Borralho PM, Luo X, Mulhovo S, Rodrigues CM, Ferreira MJ. Apoptosis inducing activity of benzophenanthridine-type alkaloids and 2-arylbenzofuran neolignans in HCT116 colon carcinoma cells. Phytomedicine. 2013;20:923–9.
Kaminskyy V, Lin KW, Filyak Y, Stoika R. Differential effect of sanguinarine, chelerythrine and chelidonine on DNA damage and cell viability in primary mouse spleen cells and mouse leukemic cells. Cell Biol Int. 2008;32:271–7.
Pan X, Han H, Wang L, Yang L, Li R, Li Z, Liu J, Zhao Q, Qian M, Liu M, Du B. Nitidine chloride inhibits breast cancer cells migration and invasion by suppressing c-Src/FAK associated signaling pathway. Cancer Lett. 2011;313:181–91.
Wangensteen H, Ho GTT, Tadesse M, Miles CO, Moussavi N, Mikolo B, Malterud KE. A new benzophenanthridine alkaloid and other bioactive constituents from the stem bark of Zanthoxylum heitzii. Fitoterapia. 2016;109:196–200.
Gathungu RM, Oldham JT, Bird SS, Lee-Parsons CW, Vouros P, Kautz R. Application of an integrated LC-UV-MS-NMR platform to the identification of secondary metabolites from cell cultures: benzophenanthridine alkaloids from elicited Eschscholzia californica (california poppy) cell cultures. Anal Methods-UK. 2012;4:1315–25.
Klejdus B, Lojkova L, Kosina P, Ulrichova J, Simanek V, Kuban V. Liquid chromatographic/electrospray mass spectrometric determination (LC/ESI-MS) of chelerythrine and dihydrochelerythrine in near-critical CO(2) extracts from real and spiked plasma samples. Talanta. 2007;72:1348–56.
Kulp M, Bragina O, Kogerman P, Kaljurand M. Capillary electrophoresis with LED-induced native fluorescence detection for determination of isoquinoline alkaloids and their cytotoxicity in extracts of Chelidonium majus L. J Chromatogr A. 2011;1218:5298–304.
Liang M, Zhang W, Hu J, Liu R, Zhang C. Simultaneous analysis of alkaloids from Zanthoxylum nitidum by high performance liquid chromatography-diode array detector-electrospray tandem mass spectrometry. J Pharm Biomed Anal. 2006;42:178–83.
Matsushima Y, Minami H, Hori K, Sato F. Pathway engineering of benzylisoquinoline alkaloid biosynthesis in transgenic California poppy cells with ectopic expression of tetrahydroberberine oxidase from Coptis japonica. Plant Biotechnol-NAR. 2012;29:473–81.
Yang C, Xiao Y, Wang Z, Wang S, Chen L, Wu L, Liu G. UHPLC-ESI-MS/MS determination and pharmacokinetic study of two alkaloid components in rat plasma after oral administration of the extract of Corydalis bungeana Turcz. J Chromatogr B. 2014;960:59–66.
Ju R, Guo L, Li J, Zhu L, Yu X, Chen C, Chen W, Ye C, Zhang D. Carboxyamidotriazole inhibits oxidative phosphorylation in cancer cells and exerts synergistic anti-cancer effect with glycolysis inhibition. Cancer Lett. 2016;370:232–41.
Luo XB, Chen B, Yao SZ. Rapid determination of protopine, allocryptopine, sanguinarine and chelerythrine in fruits of Macleaya cordata by microwave-assisted solvent extraction and HPLC-ESI/MS. Phytochem Anal. 2006;17:431–8.
Qing ZX, Cheng P, Liu XB, Liu YS, Zeng JG. Systematic identification of alkaloids in Macleaya microcarpa fruits by liquid chromatography tandem mass spectrometry combined with the isoquinoline alkaloids biosynthetic pathway. J Pharm Biomed Anal. 2014;103C:26–34.
Qing ZX, Cheng P, Liu XB, Liu YS, Zeng JG, Wang W. Structural speculation and identification of alkaloids in Macleaya cordata fruits by high-performance liquid chromatography/quadrupole-time-of-flight mass spectrometry combined with a screening procedure. Rapid Commun Mass SP. 2014;28:1033–44.
Ouchani F, Jeanne A, Thevenard J, Helesbeux JJ, Wahart A, Letinois I, Duval O, Martiny L, Charpentier E, Devy J. Ethoxyfagaronine, a synthetic analogue of fagaronine that inhibits vascular endothelial growth factor-1, as a new anti-angiogeneic agent. Invest New Drug. 2015;33:75–85.
Ding B, Zhou T, Fan G, Hong Z, Wu Y. Qualitative and quantitative determination of ten alkaloids in traditional Chinese medicine corydalis yanhusuo W.T. Wang by LC-MS/MS and LC-DAD. J Pharm Biomed Anal. 2007;45:219–26.
Liao C, Chang S, Yin S, Wang Z, Meng Y. A HPLC-MS/MS method for the simultaneous quantitation of six alkaloids of Rhizoma Corydalis Decumbentis in rat plasma and its application to a pharmacokinetic study. J Chromatogr B. 2014;944:101–6.
Ai Y, Wu Y, Wang F, Ma W, Bian Q, Lee DY, Dai R. A UPLC-MS/MS method for simultaneous quantitation of three monoterpene glycosides and four alkaloids in rat plasma: application to a comparative pharmacokinetic study of Huo Luo Xiao Ling Dan and single herb extract. J Mass Spectrom. 2015;50:567–77.
Nikolic D, Godecke T, Chen SN, White J, Lankin DC, Pauli GF, van Breemen RB. Mass spectrometric dereplication of nitrogen-containing constituents of black cohosh (Cimicifuga racemosa L.). Fitoterapia. 2012;83:441–60.
Ngoumfo RM, Jouda JB, Mouafo FT, Komguem J, Mbazoa CD, Shiao TC, Choudhary MI, Laatsch H, Legault J, Pichette A, Roy R. In vitro cytotoxic activity of isolated acridones alkaloids from Zanthoxylum leprieurii Guill. et Perr. Bioorg Med Chem. 2010;18:3601–5.
Yung BC, Li J, Zhang M, Cheng X, Li H, Yung EM, Kang C, Cosby LE, Liu Y, Teng L. Lipid Nanoparticles composed of quaternary amine-tertiary amine cationic lipid combination (QTsome) for therapeutic delivery of AntimiR-21 for lung cancer. Mol Pharm. 2016;13:653–62.
Bai LP, Zhao ZZ, Cai Z, Jiang ZH. DNA-binding affinities and sequence selectivity of quaternary benzophenanthridine alkaloids sanguinarine, chelerythrine, and nitidine. Bioorgan Med Chem. 2006;14:5439–45.
This work was jointly supported by the Natural Science Foundation of China (Grant No. 81673580 and 31601800), the Natural Science Foundation of Hubei Province of China (Grant No. 2016CFB285). “the Hundred Talents Program” from Chinese Academy of Sciences (Grant No. 29Y429291a0129), and the Sino-Africa Joint Research Project (Grant No. SAJC20160233). These funders played no roles in the study design, data collection and analysis, and decision to publish.
Availability of the data and materials
All data and materials are described within the article.
Authors and Affiliations
Key Laboratory of Plant Germplasm Enhancement and Specialty Agriculture, Wuhan Botanical Garden, Chinese Academy of Sciences, Moshan, Wuchang, Wuhan, 430074, China
Yongqiang Tian, Chunyun Zhang & Mingquan Guo
Graduate University of Chinese Academy of Sciences, Beijing, 100049, China
The Sino-Africa Joint Research Center, Chinese Academy of Sciences, Wuhan, 430074, China
All authors have contributed to the work. MG conceived, designed, and supervised the study. YT conducted the experiments, data analysis, and manuscript drafting. CZ drafted and revised the manuscript. All the authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Tian, Y., Zhang, C. & Guo, M. Comparative study on alkaloids and their anti-proliferative activities from three Zanthoxylum species.
BMC Complement Altern Med17, 460 (2017). https://doi.org/10.1186/s12906-017-1966-y