Qualitative and quantitative analysis of the major constituents in traditional Chinese medicine Danmu injection using LC-ESI-MSⁿ and LC-DAD

Abstract

Background:

Danmu injection, a traditional Chinese medicine (TCM) preparation made from Nauclea officinalis, has been commonly used for the treatment of cold, fever, swelling of throat in China. However, the chemical constituents in Danmu injection have not been clarified yet.

Objective:

a HPLC/DAD/ESI-MSⁿ method was developed for qualitative and quantitative analysis of the components in Danmu injection.

Materials and Methods:

The chromatographic separation was performed on a Welch Material XB-C18 (4.6mm × 250mm, 5μm) using gradient elution with acetonitrile (A) and water containing 0.1% formic acid (B) as mobile phase at a flow rate of 1.0 ml/min.

Results:

Twenty-five compounds, including phenolic acid and phenol glycoside, iridous glycoside and glycoalkaloid were identified or tentatively deduced on the base of their retention behaviors, UV absorption, MS and MSⁿ data with those elucidated references or literature. In addition, eleven compounds were simultaneously determined by HPLC–DAD, which was validated and successfully applied for determination of major components in Danmu injection.

Conclusion:

The results suggested that the established qualitative and quantitative method would be a powerful and reliable analytical tool for the characterization of multi-constituent in complex chemical system and quality control of Danmu injection.

INTRODUCTION

Traditional Chinese medicines (TCMs) have been widely used in many oriental countries for thousands of years.[1] In the past decades, the TCM injection has been achieved great development and become a very important formulation. However, with the widespread usage of TCM injection, many serious adverse reactions are reported in recent years,[2,3,4,5,6] mainly due to the lack of a practicable and reliable quality control to monitor the properties changed in the procedure from preparation, transportation and storage to clinic usage.

Danmu injection is a modern formula prepared from Nauclea officinalis (commonly known as Danmu) which is a traditional Chinese medicine growing in the southern part of China and the only one species in genus Nauclea in China. Nauclea officinalis has been used for the treatment of cold, fever, swelling of throat, pink eyes and so on because of its anti-inflammatory effects.[7,8] Although the main constituents in Nauclea officinalis were found to be alkaloids and its glucoside,[9,10,11,12] the chemical ingredient identification method for Danmu injection was still lacking, which resulted in the difficulty in the quality control of Danmu injection. Xie et al., determined five compounds in Nauclea officinalis leaves by HPLC.[13] As for the quantitative analysis of Danmu injection, the single alkaloid strictosamide has been determined,[14,15] and three alkaloids and one phenol acid had been determined by HPLC.[16] However, the above-mentioned methods could not provide a comprehensive chemical profile for Danmu injection.

In the last decades, hyphenated techniques, such as LC–ELSD,[17] LC–DAD,[18] LC–MS,[19] GC–MS,[20] and LC–NMR/MS,[21] have been widely applied for the analysis of the constituents in the botanic extracts and TCMs. Among them, LC–MS becomes more popular for direct identification of multi-component and quality control, due to its wide suitability, sensitivity, and sufficient structural information.[22] In the present study, a sensitive HPLC/DAD/ESI-MSⁿ method was established for rapid separation, reliable identification and quantification of the multiple components in Danmu injection. Twenty-five constituents were identified by comparison of their retention times, UV absorption and MS spectra with those elucidated references or literature data; 11 of the identified ingredients with high content were simultaneously determined by the established HPLC–DAD approach. According to the results, phenolic acid, iridous glycoside and glycoalkaloid were the main constituents in Danmu injection; furthermore, it was found that protocatechuic acid was also the abundant compound except for glycoalkaloid. This study represents the first detailed investigation of the constituents of Danmu injection and provides an applicable method for its quality evaluation.

MATERIALS AND METHODS

Chemicals and materials

HPLC-grade acetonitrile and formic acid were purchased from Tedia (Fairfield, OH, USA). Ultrapure water was prepared by a Milli-Q System (Millipore, Bedford, MA, USA) for preparing samples and mobile solution. Other reagents were of analytical grade. All solvents and samples were filtered through 0.45 μm membrane filters before analysis.

Protocatechuic acid (D1) and chlorogenic acid (D6) were purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China); Neochlorogenic acid (D3) and cryptochlorogenic acid (D9) were obtained from Chengdu Biopurify Phytochemicals Ltd. (Chengdu, China); Loganin (D7) and sweroside (D13) was purchased from Shanghai Yuanye Bio-Technology Co. Ltd (Shanghai, China); 3, 4-dimethoxyphenol β -D-apiofuranosyl (1 → 6)-β-D-glucopyranoside (D5), kelampayoside A (D8), naucleamide A-10-O-β-D-glucopyranoside (D15), naucleamide G (D17), pumiloside (D18), 3-epi-pumiloside (D19), 3α, 5α-tetrahydrodeoxycordifoline lactam (D23), strictosamide (D24) and vincosamide (D25) were isolated from Nauclea officinalis in our laboratory and their structures were confirmed based on spectroscopic analysis (MS,¹H-NMR and ¹³C-NMR).[23] The purities of all the standards were not less than 98%.

Five batches of Danmu Injection (110601, 110616, 110401, 110801 and 110925) were supplied by Hainan Pharmaceutical Factory Co., LTD. (Wuzhishan, China).

Standard and sample solutions preparation

The standard substances of D1 (20.16 mg), D3 (6.32 mg), D6 (7.96 mg), D9 (6.16 mg), D15 (17.12 mg), D17 (16.00 mg), D18 (16.20 mg), D19 (12.08 mg), D23 (8.48 mg), D24 (60.12 mg) and D25 (60.12 mg) were accurately weighed and dissolved respectively in 10 mL volumetric flask with methanol to give individual stock solutions. The mixed standard solution containing protocatechuic acid (D1, 161.28 μg/mL), neochlorogenic acid (D3, 31.60 μg/mL) , chlorogenic acid (D6, 39.80 μg/mL), cryptochlorogenic acid (D9, 30.80 μg/mL), naucleamide A-10-O-β-D-glucopyranoside (D15, 85.60 μg/mL), naucleamide G (D17, 80.00 μg/mL), pumiloside (D18, 81.00 μg/mL), 3-epi-pumiloside (D19, 60.40 μg/mL), 3α,5α-tetrahydrodeoxycordifoline lactam (D23, 42.40 μg/mL), strictosamide (D24, 601.20 μg/mL) and vincosamide (D25, 41.00 μg/mL) in methanol was prepared and diluted with methanol to six different concentrations with the ranges listed in Table 1. The mixed standard solutions were filtered through a 0.45 μm membrane prior to injection. All solutions were stored at 4°C in refrigerator before analysis.

Table 1.

Linear regression data, LOD and LOQ of the investigated compounds

A sample of 1.0 mL injection was diluted to 10 mL with ultrapure water in a volumetric flask before being filtered through a 0.45 μm membrane filters. An aliquot of each filtrate was injected into the HPLC instrument for analysis.

HPLC-DAD-ESI- MS System

Chromatographic analysis

An Agilent series 1200 HPLC instrument (Agilent, Waldbronn, Germany) equipped with a quaternary pump, a diode-array detector, an auto sampler and a column compartment was used. The samples were separated on a Welch Material XB-C18 (4.6 mm × 250 mm, 5 μm, Welch Material). The mobile phase consisted of acetonitrile (A) and water containing 0.1% formic acid (B). A gradient program was used as follows: 0~70 min, 1%~35% A, 70~75 min, 35%~95% A. Flow rate was 1.0 mL/min, Column temperature was 25°C. The injection volume was 20 μL. Detection wavelength was set at 260 nm for protocatechuic acid, 327 nm for neochlorogenic acid, chlorogenic acid and cryptochlorogenic acid, 245 nm for pumiloside and 3-epi-pumiloside, 226 nm for naucleamide A-10-O-β-D-glucopyranoside, naucleamide G, 3α, 5α-tetrahydrodeoxycordifoline lactam, strictosamide and vincosamide, respectively.

Mass spectrometry

Mass spectrometry was performed using Agilent 6310 mass spectrometers (Agilent, Waldbronn, Germany) equipped with ESI interface and ion trap analyzer. The ESI-MS spectra of samples and reference compounds were acquired in both positive and negative ionization modes. The parameters were as follows: the drying gas temperature was set at 350°C and the capillary voltage was set at 3.5 kV, skimmer voltage at 40 V, Flow rate of nebulizing gas (N₂) 12 L/min; and pressure of drying gas (N₂) 35 psi. Mass spectrometry was performed in the full-scan mode (MS1) and automatic multiple-stage fragmentation-scan modes (MS2-MS4) over an m/z scan range of 50-1000.

Method validation

The method validation was performed after the optimum conditions established. Solutions containing 11 standard substances at six different concentrations were injected in triplicate. Calibration curves were established by plotting the peak area versus concentration of each analyte. The limits of detection (LOD) and limits of quantification (LOQ) were measured with the signal-to-noise ratio of 3~4 and 10~12 as criteria, respectively.

Intra- and inter-day variations were utilized to assess the precision of the method. The intra-day variation was determined by analyzing six replicate samples within 1 day and the inter-day variation was examined in 3 consecutive days. Recovery was used to evaluate the accuracy of the method. A certain amount of Danmu injection sample was spiked with the mixed standard solution. The mixture was processed and analyzed using the method mentioned above, and three replicates were performed for the analysis. Variations were expressed by relative standard deviation (RSD) in all tests above.

RESULTS

Identification of chemical constituents in Danmu injection

The reference standards and Danmu injection sample were analyzed using the optimized LC–DAD-ESI-MSⁿ method, and 25 peaks were observed in Danmu injection sample [Figure 1]. The chemical structures of the 25 compounds were characterized based on their retention behavior and their UV spectra obtained on-line [Table 2]. Different types of compounds showed different UV absorption characteristics. For MS analysis, both negative and positive modes of ESI mass spectra were examined in this study. Generally, in the positive mode, [M + H]⁺ ions of alkaloids of sufficient abundance could be subjected to MSⁿ analysis and provided more structural information, while in the negative mode, [M − H]⁻ or [M + 46–H]⁻ were observed. Due to the use of formic acid in mobile phase, there were adducted ions of [M + 46–H]⁻ corresponding to [M + HCOOH − H]⁻, which provided valuable information for composition of the constituents.

Figure 1.

HPLC Chromatograms of standard substances (a and b) and sample (c and d) determined at 260 nm (a and c) and 325 nm (b and d); TIC chromatogram of DM injection in negative ionization modes (e)

Table 2.

HPLC-DAD-ESI/MS identification results of 25 chemical compositions in Danmu injection

Twenty-five compounds were tentatively identified on the basis of their retention times, UV absorption, MS data with those of the reference standards and the literature data (chemical structures shown in Figure 2. Among the identified constituents, peaks D1, D3, D5, D6, D7, D8, D9, D13, D15, D17, D18, D19, D23, D24 and D25 were identified as protocatechuic acid, neochlorogenic acid, 3,4-dimethoxyphenol β -D-apiofuranosyl (1→6)-β-D-glucopyranoside, chlorogenic acid, loganin, kelampayoside A, cryptochlorogenic acid, sweroside, naucleamide A-10-O-β-D-glucopyranoside, naucleamide G, pumiloside, 3-epi-pumiloside, 3α-5α-tetrahydrodexycordifoline lactam, strictosamide and vincosamide by comparing the standard substance, respectively.

Figure 2.

The structures of 25 compounds identified in Danmu injection

The identified compounds can be classified into three classes, namely phenolic acid and its phenol glycoside, iridous glycoside and glycoalkaloid. A total of 7 phenolic acids or phenol glycoside (peaks D1, D2, D3, D5, D6, D8 and D9) were identified, most of which produced [M−H]⁻ ion. 4 iridoid glycoside (peaks D4, D7, D11 and D13) and 14 glycoalkaloid (D10, D12, D14, D15, D16, D17, D18, D19, D20, D21, D22, D23, D24 and D25) were characterized by MSⁿ analysis due to their fragmentation pathways, as well as by comparing with reference standards and previous studies. Alkaloids and glycoalkaloids are the main ingredients in Nauclea officinalis.[24,25,26] For MS analysis, glycoalkaloid gave [M + H]⁺ ions and were observed to undergo the Retro–Diels–Alder (RDA) fragmentation reaction [Figures 1 and 2 and Table 2].

The ESI-MS of protocatechuic acid, neochlorogenic acid, chlorogenic acid and cryptochlorogenic acid all produced [M − H]⁻ ion in the negative mode. The fragment ion m/z 109 was found in the MS/MS spectrum of protocatechuic acid, which was followed by neutral loss of CO₂(44 Da). The [M − H]⁻, [2M − H]⁻ ions and the characteristic ions at m/z 191 [quinicacid − H]⁻, 173 [quinic acid − H − H₂O]⁻ and 135 [caffeoyl − H − COO]⁻ were observed in the MSⁿ spectra of chlorogenic acid.[27]

The ESI-MSⁿ of 3,4-dimethoxyphenol β-D-apiofuranosyl (1 → 6)-β-D-glucopyranoside and kelampayoside A

The 3,4-dimethoxyphenol β -D-apiofuranosyl (1 → 6)-β-D-glucopyranoside was found in the genus Nauclea for the first time and kelampayoside A was isolated from Nauclea officinalis for the first time in our lab, respectively.[23] The structures of the two compounds were very similar to each other except for the −OCH₃ at the C-5 position. The analogies of structures indicated the similarities of their MSⁿ fragmentation behaviors. In their ESI-MS spectra, the [M + Na]⁺ ion were obtained in the positive mode, while in the negative mode, the [M − H]⁻, [M + HCOOH − H]⁻ and [M + 2H₂O − H]⁻ ions were simultaneously observed. The MS/MS spectrum showed a significant fragment ion [C₁₁H₁₈O₉ − H]⁻ ion at m/z 293 which was attributed to the loss of a C-1 unit, at the same time, other ions m/z 233, m/z 149 and m/z 131 were observed [Figure 3].

Figure 3.

MS, MS/MS spectrum of 3,4-Dimethoxyphenol β-D-apiofuranosyl(1→6)-β-D-glucopyranoside in the negative mode

The ESI-MSⁿ of sweroside

The [M − H]⁻, [M + 2H₂O − H]⁻ and [M + HCOOH − H]⁻ ions were shown in the negative mode. In the MS/MS spectrum analysis of [M + HCOOH − H]⁻ ion, the ion caused by the losses of glucose and other units successively were observed [Figure 4].

Figure 4.

MS/MS spectrum of sweroside

In common with sweroside, loganin gave the [M − H]⁻, [M + 2H₂O − H]⁻ and [M + HCOOH − H]⁻ ions in the negative mode. The MS/MS spectrum of m/z 435 revealed a radical aglycone ion [M − glu − H]⁻ at m/z 227 as the base peak, resulting from the loss of glucose (162 Da). In the further MSⁿ analysis of [M-glu − H]⁻ ion of loganin, the obvious [M-glu-H₂O − H]⁻ ion at m/z 209 was observed, meanwhile, other ions m/z 127 and m/z 101 were observed [Figure 5]. In the positive mode, the ESI-MSⁿ of loganin was consistent with literature.[28]

Figure 5.

MS2, MS3 spectrum of loganin

The ESI-MSⁿ of naucleamide A-10-O-β-D-glucopyranoside

The ESI-MSⁿ of the naucleamide A-10- O-β- D-glucopyranoside has been well investigated.[29] The fragmentation pathway of [M + H]⁺ ion m/z 519 was shown in the positive mode. Firstly, the [M + H]⁺ ion produced a prominent ion m/z 357 in the MS/MS spectrum, which revealed a neutral loss of 10-O-glucose residue. And then the obtained ion produced the prominent [M − glu − H₂O + H]⁺ ion at m/z 339. Then the [M-glu − H₂O + H]⁺ ion of m/z 339 was selected for further MS⁴ analysis to generate ions at m/z 321, m/z 309 and m/z 160 were also observed, which were attributed to the losses of H₂O, CH₄O and RDA cleavage of ring C, respectively.

The ESI-MSⁿ of naucleamide G

The fragmentation of naucleamide G exhibited a little difference from that of naucleamide A-10-O-β-D-glucopyranoside because of the CH₂OH at the C-16 position. In the MS/MS spectrum, the [M − glu + H]⁺ ion at m/z 327 was found as the base peak. And then the obtained ion produced the prominent [M − glu − H₂O + H]⁺ ion at m/z 309. Successively, in the MS⁴ spectrum of m/z 309, two ions at m/z 186 and 160 were observed, which were assigned as the RDA cleavage of ring D and C, respectively [Figure 6].

Figure 6.

Fragmentations pathways of naucleamide G

The ESI-MSⁿ of pumiloside and its isomer 3-epi-pumiloside

Pumiloside and 3-epi- pumiloside, with the same MS spectra, UV, molecular formulae, are isomers. Pumiloside and its isomer 3-epi- pumiloside are the only two quinoline alkaloids in Nauclea officinalis,[24,25,26] and pumiloside is higher content in Danmu injection. The ESI-MSⁿ of the pumiloside and 3-epi- pumiloside had been studied.[29,30] In their ESI-MS spectra, the [M + H]⁺ ion was observed, while the significant [M − glu + H]⁺ ion was found in the MS/MS spectrum. The [M − glu + H]⁺ ion was triggered by the loss of H₂O and CO₂ in the MS³ spectrum, at the same time, the ion at m/z 281 caused by the loss of neutral fragments C₄H₆O₄ of RDA cleavage of ring E could be observed in the MS³ spectrum. Then the ion m/z 281 was selected for further MS⁴ analysis to generate two ions at m/z 185and 158, which were assigned as the RDA cleavage of ring D and C, respectively [Figure 7].

Figure 7.

MS/MS, MS3and MS4 spectrum of pumiloside and 3-epipumiloside

The ESI-MSⁿ of strictosamide and vincosamide

Strictosamide and vincosamide are isomers. Strictosamide is the main ingredient and the highest content in Nauclea officinalis[24,25,26] and Danmu injection. In the positive mode, the [M + H]⁺ ion at m/z 499 was shown in full scan mass spectrum. [M + H]⁺ ion generated the base peak at m/z 337, originating from the loss of glucose residue in the MS/MS spectrum. In the MS³ spectrum of at m/z 337, [M − glu + H] ⁺ ion yielded fragment ion [M − glu-C₄H₆O₄+ H]⁺ at m/z 267, which was assigned as the RDA cleavage of ring E of loss neutral fragments (70Da), at the same time, the ions at m/z 319, m/z 171 and m/z 141 were observed, which were attributed to the losses of H₂O, C₉H₁₀O₃ and RDA cleavage of ring C, respectively. Successively, in the MS⁴ spectrum of m/z 267, the ions at m/z171 was observed, which were attributed to the losses of RDA cleavage of ring D.[29,30] In the MS⁵ spectrum of m/z 171, the ions at m/z118, 130 and 154 were observed. Vincosamide is the same fragmentation pathway with strictosamide [Figures 8 and 9].

Figure 8.

MS/MS, MS3,MS4,MS5 spectrum of strictosamide and vincosamide

Figure 9.

Fragmentations pathways of strictosamide and vincosamide

The ESI-MSⁿ of 3α, 5α-tetrahydrodeoxycordifoline lactam

The structures of 3α, 5α-tetrahydrodeoxycordifoline lactam and strictosamide were very similar to each other except for “−COOH” groups at the C-5 position. In the positive mode, the CO₂ lost in the MS/MS spectrum, and then [M-glu-CO2^-+H]⁺ ion m/z 335 was observed in the MS³ spectrum, successively, in the MS⁴ spectrum of m/z 265 and m/z 169 were observed, which were attributed to the losses of RDA cleavage of ring E and D, respectively [Figure 10].

Figure 10.

Fragmentations pathways of 3α,5α-tetrahydrodeoxycordifoline lactam

On the chromatogram, D6, D7 and D8 are in one peak, but their molecular mass is different. We can extract their molecular mass and ion fragments in the TIC chromatogram, and deduced their structures. By the same token, D4, D5, D12 and D13 were identified.

Method validation of the quantitative analysis

The calibration curves, linear ranges, LOD and LOQ of 11 analyses were performed using the developed HPLC–DAD method described in section 2.4 [Table 1]. Reasonable correlation coefficient values (r²> 0.9998) indicated good correlations between investigated standards concentrations and their peak areas within the ranges tested. The ranges of LOD and LOQ for all the analytes were from 1.06 ng to 20.03 ng, and 3.42 ng to 80.13 ng, respectively. The intra- and inter-day variations (RSD) of the 11 analyses were in the range from 1.31 to 2.42 and 1.14 to 2.82% [Table 3], respectively. The developed method has good accuracy with the recoveries were between 95.97 and 109.30%. It was also found that the analyses in the sample solution were stable for 24 h with a RSD less than 3.1%. All these values fall within acceptable limits, which indicates this method, is reliable with significant repeatability, recovery and precision. The results proved that HPLC is appropriate for analyzing and assessing the quality of Danmu injection.

Table 3.

Precisions and recoveries of 11 analyses (110801)

Quantitative determination of Danmu injection

Typical chromatograms for the quantitative determination of 11 main compounds in Danmu injection were shown in Figure 1. A total of five different batches of Danmu injection were tested using the developed HPLC–DAD method. The contents of 11 investigated compounds were determined and the results were listed in Table 4. It was found that strictosamide (D24) was the most abundant compound, protocatechuic acid (D1), naucleamide A-10-O-β-D-glucopyranoside (D15), naucleamide G (D17) and pumiloside (D18) are higher content, while other 6 compounds were found to be lower in Danmu injection.

Table 4.

The content of 11 compounds in five batches of Danmu injection (n=3, μg/mL)

CONCLUSION

Chemical constituents of Danmu injection was systematically and thoroughly investigated by HPLC–DAD–ESI-MSⁿ, which provided full-scale qualitative and quantitative information for analysis of chemical constituents in Danmu injection. Twenty-five compounds including phenolic acid, phenol glycoside, iridous glycoside and alkaloid glycoside were identified; eight (D3, D5, D6, D8, D9, D11, D17 and D23) of them were never reported previously in Nauclea officinalis.

The quantitative method was proved to have excellent linearity, good accuracy, sensitivity and repeatability. The results would provide the chemical support for the further pharmacokinetic studies and for effective quality assessment of Danmu injection, which would be of significant importance for the safety use and modern research of Danmu injection.