Discovery of chemical markers for improving the quality and safety control of Sinomenium acutum stem by the simultaneous determination of multiple alkaloids using UHPLC-QQQ-MS/MS

Sinomenium acutum stem is a popular traditional Chinese medicine used to treat bone and joint diseases. Sinomenine is considered the only chemical marker for the quality control of S. acutum stem in mainstream pharmacopeias. However, higenamine in S. acutum stem is a novel stimulant that was banned by the World Anti-Doping Agency in 2017. Therefore, enhancing the quality and safety control of S. acutum stem to avoid potential safety risks is of utmost importance. In this study, a fast, sensitive, precise, and accurate method for the simultaneous determination of 11 alkaloids in S. acutum stem by ultrahigh-performance liquid chromatography coupled with triple quadrupole tandem mass spectrometry (UHPLC-QQQ-MS/MS) was established. This method successfully analyzed thirty-five batches of S. acutum stem samples. The average contents of sinomenine, magnoflorine, coclaurine, acutumine, higenamine, sinoacutine, palmatine, magnocurarine, columbamine, 8-oxypalmatine, and jatrorrhizine were 24.9 mg/g, 6.35 mg/g, 435 μg/g, 435 μg/g, 288 μg/g, 44.4 μg/g, 22.5 μg/g, 21.1 μg/g, 15.8 μg/g, 9.30 μg/g, and 8.75 μg/g, respectively. Multivariate analysis, including principal component analysis (PCA), orthogonal partial least square method-discriminant analysis (OPLS-DA), and hierarchical cluster analysis (HCA), were performed to characterize the importance and differences among these alkaloids in S. acutum stem samples. As a result, sinomenine, magnoflorine, coclaurine, acutumine, and higenamine are proposed as chemical markers for quality control. Higenamine and coclaurine are also recommended as chemical markers for safety control. This report provides five alkaloids that can be used as chemical markers for improving the quality and safety control of S. acutum stem. It also alerts athletes to avoid the risks associated with consuming S. acutum stem.

Sinomenium acutum stem (Qing Feng Teng in Pinyin) has been a popular Chinese medicine with an extensive and active global market for decades. Alkaloids, dominated by sinomenine, are the principal chemical components of S. acutum stem. Pharmacological research and clinical practice have proven the excellent anti-inflammatory and analgesic effects of sinomenine 1 , which is commonly manufactured as a finished product to treat joint diseases and sports injuries 2 . Therefore, sinomenine has been considered a quality marker for this herb in the Chinese Pharmacopoeia, Japanese Pharmacopoeia, Korean Pharmacopoeia, and European Pharmacopoeia 3-6 .
Nevertheless, using only one marker to reflect the real quality of an herbal medicine containing multiple active ingredients has significant limitations. Phytochemical studies have demonstrated that nearly one hundred alkaloids are present in this herb. It possesses types of morphinans, aporphines, protoberberines, benzylisoquinolines and other compounds, such as sinomenine, magnoflorine, coclaurine, acutumine, and higenamine 2,7 . In addition, the safety of the alkaloids in S. acutum stem, especially higenamine and coclaurine, should be considered. For instance, since 2017, the World Anti-Doping Agency (WADA) has explicitly banned higenamine as a novel stimulant ingredient 8 . However, higenamine is a natural alkaloid found in several herbs, including Aconitum japonicum Thunb., Tinospora crispa (L.) Hook.f. & Thomson, Nandina domestica Thunb., Gnetum parvifolium (Warb.) C.Y. Cheng, and Asarum heterotropoides F. Schmidt 9 , as well as in S. acutum stem. Coclaurine could also be a potential stimulant because it is a metabolite of higenamine with a similar chemical structure [10][11][12] . Coclaurine has also been identified in rat urine by gas chromatography/mass spectrometry after higenamine administration 13 .
Moreover, many athletes inadvertently consumed higenamine-containing products and were sanctioned in recent years [14][15][16][17] . These accidental doping cases have raised global concern. Higenamine has been detected in sports supplements and herbal products. Some such supplements are even unlabeled or inaccurately labeled in North America and Asia 18 . Athletes may use S. acutum stem and related products to treat inflammation and pain, but the health risks of higenamine remain unclear 19 , and little attention is paid to the banned ingredients in this herb. Trace amounts of higenamine and coclaurine can be found in human plasma and urine by extract mass spectrometry after intravenous or oral administration of manufactured herbal products 20,21 . However, the amounts of higenamine and coclaurine in S. acutum stem and whether there is a potential safety hazard of using S. acutum stem by athletes are still unknown. Therefore, we must study potential doping incidents caused by the unintentional use of herbal products. It is of considerable significance to study whether other alkaloids could be accepted as chemical markers to improve the safety control of S. acutum stem.
In this study, a rapid, sensitive, precise and accurate method for the simultaneous quantification of 11 alkaloid compounds in S. acutum stem was established for the first time by using ultrahigh-performance liquid chromatography coupled with triple quadrupole tandem mass spectrometry (UHPLC-QQQ-MS/MS). This strategy can be used to strengthen the quality control of S. acutum stem, thereby ensuring safety.

Results
UHpLc-QQQ-MS/MS optimization. Different solvents and gradient profiles for the mobile phase were compared to achieve a reasonable resolution within 10 min. The separation efficiency and peak symmetry of the analytes were dramatically improved with 0.1% formic acid (A) and acetonitrile (B) as the mobile phase. Because of the wide polarity range of the analytes and the presence of isomers (columbamine and jatrorrhizine), gradient elution was employed and showed a better separation than isocratic elution.
The positive mode provided higher ionization efficiency and sensitivity than negative mode for the MS analysis. The collision energy and fragmentor voltage parameters were optimized to obtain the highest relative abundance of the exclusive ions and product ions in optimized MRM conditions. The final conditions for the collision energy and fragmentor voltage are shown in Table 1. The MS/MS ion spectra of the 11 alkaloids are shown in Fig. 1.

Extraction method optimization.
Since the contents of these 11 compounds are significantly different, the sample extraction method was optimized to achieve efficient extraction for all the compounds based on the Chinese Pharmacopeia method (2015 edition), and five extraction solvents, i.e., 20 mL of 70% ethanol solution, 20 mL of 98% ethanol solution, 20 mL of 0.1 M hydrochloric acid solution, 20 mL of methanol solution and 20 mL of methanol with ammonia (V/V 95:5) solution, were compared. Based on the results (Fig. 2), 20 mL of 70% ethanol solution was selected to efficiently extract the 11 alkaloids from the S. acutum stem.
Method validation. The regression equations, linear ranges, correlation coefficients, limits of detection (LODs), and lower limits of quantification (LLOQs) of the 11 compounds are listed in Table 1. All the calibration curves exhibited excellent linearity with correlation coefficients (R values) in the range of 0.9995-0.9998. The LOD was set to a signal-to-noise ratio (S/N) not less than three, and the LLOQ was not less than approximately ten. www.nature.com/scientificreports/ The precision was determined by continuously testing one concentration of the mixed standard solution six times. Sample Q29 was separated into six portions, and each portion was extracted separately to evaluate the repeatability of this method. The test solution of sample Q29 was employed to evaluate the stability over 0, 2, 4, 6, 8, and 12 h in one day at room temperature. The results for all the compounds, which are summarized in Table 2, indicate that the instrument has good precision, the method is repeatable, and the compounds in the sample solution are sufficiently stable for accurate and precise analysis within 12 h at room temperature (Supplementary information).

Discussion
In this study, thirty-five batches of S. acutum stem samples were acquired from local hospitals or pharmacies in China and were determined by the optimized UHPLC-QQQ-MS/MS method. Hua Zhou authenticated all the samples, and the voucher specimens are stored at the State Key Laboratory of Quality Research in Chinese Medicine (Macau University of Science and Technology). The sample number, sample origin, and collection location for each sample are shown in Table 3, and the samples are highly representative of this herb.
The average total content of these 11 alkaloids was 32.6 ± 7.17 mg/g. These alkaloids can be empirically divided into three groups (Fig. 7) based on their abundance. Sinomenine and magnoflorine, which are present at the milligram level and account for 76.6 and 19.5%, respectively, of total content, can be classified as high-abundance compounds. Coclaurine, acutumine, and higenamine, which are present at the microgram level and account for 1.34, 1.34, and 0.885%, respectively, of the total content, can be classified as moderate-abundance compounds. Moreover, the remaining six alkaloids, sinoacutine, palmatine, magnocurarine, columbamine, 8-oxypalmatine, and jatrorrhizine, although at the microgram level, account for less than 0.374% of the total content and therefore can be classified as low-abundance compounds. Magnoflorine is a homolog of sinomenine, and they have similar phenanthrene structures and similar biosynthetic pathways 22 , which could be the reason that they are in the same group. A similar explanation can be applied in the case of higenamine and coclaurine 11 . www.nature.com/scientificreports/ PCA can simplify complex information by replacing the original variable index with a small number of comprehensive indicators utilizing dimension reduction. In the present study, PCA was used to analyze the distribution pattern of the 11 alkaloids in the S. acutum stem. The result of the cumulative contributions meant that the original information of the dataset was basically retained. The 2D loading scatter plot (Fig. 4B), which provided useful information to identify the essential features in the PC1 and PC2 dimensions, showed that sinomenine, magnoflorine, coclaurine, acutumine, and higenamine were located at the edges of the axes, demonstrating the greater correlations with PC1 and PC2 and implying that these five compounds were essential.
OPLS-DA is a supervised discrimination method. Figure 5B shows the modeling used to predict the specific components with the most significant influence on the samples. It demonstrated the considerable quality of the model. Samples with more substantial VIP (> 1.0) are generally more related to sample classification. We found that the VIP values of sinomenine and magnoflorine were more than 1.0, and those of higenamine and coclaurine were very close to 1, and these compounds were in the top four. Therefore, as indicated by the plots in Fig. 5C,D, sinomenine, magnoflorine, higenamine, and coclaurine played an essential role in the S. acutum stem samples.
The HCA results were also consistent with the results of content determination, PCA, and OPLS-DA above. Therefore, multivariate analysis verified that sinomenine, magnoflorine, higenamine, coclaurine, and acutumine were the main chemical components of the S. acutum stem.
According to the Chinese Pharmacopeia, the following three requirements should be used for the selection of potential chemical markers. The content of the marker in the medicinal material should be higher than 0.02%. The corresponding specific or active ingredients selected as markers for content determination should be involved in the function or bioactivity of the Chinese medicines. A multicomponent detection method should be used when a single component cannot reflect the medicinal materials' overall activity 23 . The pharmacological activities of sinomenine 2,24-26 , magnoflorine 14,27 , coclaurine 28 , acutumine 29 , and higenamine 30,31 well represent the main indications or bioactivities of S. acutum stem, which are anti-inflammation, analgesia, anti-hypertension, anti-arrhythmia, anticancer, and immunomodulation 32 . Therefore, based on the multivariate analysis results and the bioactivities of these components, it was reasonable and representative to identify these specific alkaloids, i.e., sinomenine, magnoflorine, higenamine, acutumine, and coclaurine, as chemical markers of S. acutum stem. These alkaloids are recommended for improving the quality control of S. acutum stem in pharmacopeias and for use as relevant standards in the future. The use of sinomenine and magnoflorine as chemical markers was also consistent with the published literature conclusions 2,33 . The other three alkaloids proposed together with sinomenine and magnoflorine better reflect the overall quality of this herb. www.nature.com/scientificreports/ For the safety control of S. acutum stem, higenamine and coclaurine are proposed as chemical markers. Before discussing the safety, we determined that the sinomenine content in each of the samples tested in this study met the requirements of the pharmacopoeias 3,4 . These samples were qualified under the existing standards.
In this study, higenamine and coclaurine were found in S. acutum stem at high levels. Therefore, athletes should consider this herbal medicine and its products with caution. The WADA stipulates that all selective and nonselective beta-2 agonists, including all optical isomers, are prohibited. Higenamine is a β-androgenic receptor agonist. It possesses lipolytic activity and can improve cardiac left ventricular function. Therefore, it can promote the growth of skeletal muscle. This is the main reason that higenamine is explicitly banned as a novel stimulant ingredient 8 .
The average contents of higenamine and coclaurine in the S. acutum stem were 288 μg/g and 435 μg/g, respectively, in the present study. The recommended oral dosage of S. acutum stem is 6-12 g per day according to the Chinese Pharmacopoeia. A previous study reported that the maximum urinary concentration of higenamine in humans was 0.2-0.4 ng/mL within 10-12 h after oral administration of the herbal product containing 19.8 μg higenamine, and the maximum urinary concentration of coclaurine was 0.3-1.0 ng/mL (corresponding to 4.5 μg of coclaurine) 21 . Therefore, we initially speculated that the maximum urinary concentration ranges of higenamine and coclaurine that could be detected after oral S. acutum stem treatment for 10-12 h were 17.5-69.9 ng/mL and 39.5-264 ng/mL, respectively, if the ingredients were completely extracted. The anti-doping organization required that the concentration of higenamine in urine should not be more than 10 ng/mL 34 . Apparently, when athletes take regular doses of S. acutum stem, the urinary concentration of higenamine exceeds the stimulant detection threshold. Even worse, because of the enzymatic conversion between coclaurine and higenamine, the total concentration of stimulants may also increase in the human body. Therefore, according to the existing www.nature.com/scientificreports/ standards, these qualified herbs still pose a notable safety risk. With the widespread use of S. acutum stem and its products, there is a high risk that unintentional higenamine doping could be detected. This situation also reminds industry and regulatory bodies that higenamine and coclaurine can be used as chemical markers for the safety control for S. acutum stems. It is better to warn athletes to use caution on the packaging of related products. This study also alerts athletes that they should be very cautious when they are seeking assistance from herbal medicines to relieve inflammation and pain caused by sports injuries.

Materials and methods
Reagents and chemicals. Sinomenine, magnoflorine, sinoacutine, columbamine, and coclaurine as chemical reference standards were purchased from Chengdu Chroma Biotechnology Co., Ltd. (Chengdu, P.R. China). Higenamine, acutumine, magnocurarine, jatrorrhizine, palmatine, and 8-oxypalmatine as reference standards were purchased from Shanghai Chenyi Biotechnology Co., Ltd. (Shanghai, P.R. China). All compounds were at a purity of ≥ 98.0%, and the chemical structures are given in Fig Table 2. The other parameters were as follows: drying gas (N 2 ) flow rate, 11.0 L/min; drying gas temperature, 300 °C; nebulizer, 15 psig; and capillary voltage, 4,000 V.   www.nature.com/scientificreports/ QQQ-MS/MS system, and the extracted ion chromatogram (EIC) of the 11 reference standards is shown in Fig. 9. All the standard solutions were stored at 4 °C until use.

Method validation and test solution preparation.
The method was validated in Supplementary information for linearity, precision, repeatability, stability, and recovery following the Guidelines for the Validation of Quality Standard of Traditional Chinese Medicine (Chinese Pharmacopoeia, 2015 Edition, Volume 1) and the United States Food and Drug Administration Bioanalytical Method Validation (US Food and Drug Administration, 2001). The sample powder (0.5 g passed through a 50 mesh sieve) was accurately weighed in a conical flask with a stopper. The flask was supplemented with 20.0 mL of 70% ethanol, well stoppered, and accurately weighed. Next, the flask was ultrasonicated for 20 min (power 150 W, frequency 20 kHz), cooled to room temperature, weighed again, and the weight was restored to its initial value with 70% ethanol. The sample solution was shaken thoroughly and filtered through 0.22-μm microporous filtration membranes. Two hundred microliters of the obtained filtrate was diluted to 1.0 mL with 70% ethanol to afford the test solution.