Determination method of trigonelline and trigonellinic acid in semen mori

Abstract

The present application relates to the technical field of substance detection analysis, in particular to a method for determining the contents of trigonelline and psilocybin in Tribulus terrestris L. seeds. The method comprises: obtaining a first data set of the contents of trigonelline and psilocybin in Tribulus terrestris L. seeds based on UHILIC-MS / MS; obtaining a second data set of near-infrared spectra of trigonelline and psilocybin in Tribulus terrestris L. seeds based on NIRS; establishing a content model corresponding to trigonelline and psilocybin based on the first data set and the second data set based on PLS; and obtaining the contents of trigonelline and psilocybin in Tribulus terrestris L. seed samples by inputting the near-infrared spectra of the samples into the content model. The method can realize rapid detection of trigonelline and psilocybin in Tribulus terrestris L. seeds.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of substance detection analysis, and particularly relates to a method for determining the contents of trigonelline and quisqualic acid in Quisqualis indica L. BACKGROUND

[0002] Quisqualis indica L. is the dried mature fruit of Combretaceae Quisqualis plant, and Quisqualis indica L. is the shelled processed product of Quisqualis indica L. It is sweet and warm in nature, has the effects of killing insects and eliminating accumulation, and is mainly used for roundworm disease and infantile malnutrition. Previous studies have shown that Quisqualis indica L. contains various active ingredients, such as alkaloids, organic acids, and amino acids. The 2020 edition of the People's Republic of China Pharmacopoeia determines trigonelline as the index component for the determination of the content of Quisqualis indica L., which has the effects of antioxidant, antibacterial, and hypoglycemic. Quisqualic acid is a characteristic component, which has the effects of expelling and killing insects. The current detection method for quisqualic acid is a derivatization method, including HPLC-UV detection after derivatization with 2-nitrobenzenesulfonyl chloride (Chang S Y, Li B F, Chen T Y, et al. Determination of Quisqualic Acid in Quisqualis fructus by Pre-column Derivatization and High Performance Liquid Chromatography [J]. JOURNAL OF ANALYTICAL CHEMISTRY, 2020, 75 (8): 1018-1023.) or HPLC-UV detection after derivatization with 2,4-dinitrochlorobenzene (Liao Jiahui, Chu Hongjun, Xie Rui, et al. Comparison of the contents of quisqualic acid in Quisqualis fructus and Quisqualis indica L. before and after processing by column pre-derivatization-high performance liquid chromatography [J]. Chinese Journal of New Drugs and Clinical Remedies, 2021, 27 (11): 60-63.). However, the derivatization method needs to perform sample degreasing and color development processing, and the sample detection throughput is low, the time consumption is long, and the two components of quisqualic acid and trigonelline have large polarity, so that in the pre-experiment, an ordinary C18 chromatographic column is used, and it is found that the two components have poor retention on the chromatographic column, and the peak appears quickly, which causes detection interference. SUMMARY

[0003] In order to overcome the defects of the prior art, the technical problem to be solved by the present application is to provide a method for rapidly detecting the contents of trigonelline and quisqualic acid in Quisqualis indica L.

[0004] In order to solve the above technical problems, the present application provides a method for determining the contents of trigonelline and quisqualic acid in Quisqualis indica L., comprising:

[0005] obtaining a first data set of contents of trigonelline and trigonelline amine in semen trichosanthis based on UHILIC-MS / MS method;

[0006] obtaining a second data set of near infrared spectrum of trigonelline and trigonelline amine in semen trichosanthis based on NIRS method;

[0007] establishing a content model corresponding to trigonelline and trigonelline amine respectively based on the first data set and the second data set based on PLS;

[0008] obtaining the contents of trigonelline and trigonelline amine in semen trichosanthis samples by inputting the near infrared spectrum of the semen trichosanthis samples into the content model.

[0009] In an embodiment, the method further comprises pretreating the second data set before establishing the content model;

[0010] wherein the pretreatment adopted for the second data set of trigonelline is performing multivariate scatter correction, orthogonal signal correction and Savitzky-Golay smoothing;

[0011] the pretreatment adopted for the second data set of trigonelline amine is performing Savitzky-Golay smoothing and standard normal variate transformation.

[0012] In an embodiment, the modeling wavelength for establishing the content model is 1000-1799 nm.

[0013] In an embodiment, the number of principal factors of trigonelline in the content model is 8, and the number of principal factors of trigonelline amine is 12.

[0014] In an embodiment, the chromatographic conditions of the UHILIC-MS / MS method are as follows:

[0015] chromatographic column: Waters ACQUITY BEH HILIC Amide chromatographic column (2.1 mm x 100 mm, 1.7 μm);

[0016] mobile phase: 0.1% (V / V) formic acid aqueous solution (A) - acetonitrile (B);

[0017] gradient elution: 0-0.5 min, 95% B→95% B;

[0018] 0.5-1.0 min, 95% B→60% B;

[0019] 1.0-4.0 min, 60% B→60% B;

[0020] 4.0-4.1 min, 60% B→95% B;

[0021] 4.1-6.0 min, 95% B→95% B;

[0022] Flow rate: 0.20 mL·min -1 ;

[0023] Column temperature: 45℃;

[0024] Injection volume: 2 μL.

[0025] In one embodiment, the mass spectrometry conditions of the UHILIC-MS / MS method are as follows:

[0026] Mode: Electrospray positive / negative;

[0027] Capillary voltage: 3.5 kV;

[0028] Desolvation gas flow: Nitrogen 800 L·h -1 ;

[0029] Desolvation temperature: 500℃;

[0030] Conespray gas flow: Nitrogen 150 L·h -1 ;

[0031] Ion source temperature: 150℃;

[0032] Collision gas: Argon.

[0033] In one embodiment, the mass spectrometry conditions of trigonelline are as follows:

[0034] Quantitative ion pair: 138.30→92.10 m / z;

[0035] Collision energy: 18 eV;

[0036] Ion pair mode: ESI + .

[0037] In one embodiment, the mass spectrometry conditions of trigonelline are as follows:

[0038] Quantitative ion pair: 188.01→101.01 m / z;

[0039] Collision energy: 12 eV;

[0040] Ion pair mode: ESI - .

[0041] The beneficial effects of this invention are as follows: This invention uses UHILIC-MS / MS to simultaneously determine the content of trigonelline and quincetin, and combines it with NIRS to establish a quantitative model for quincetin and trigonelline in Quincetin kernels. It has the advantages of being rapid and non-destructive, and can provide a new technical method for rapid detection of Quincetin kernel quality evaluation. Attached Figure Description

[0042] Figure 1 The image shows the MRM chromatograms of blank (A), reference (B), and sample (C) in a specific embodiment of the present invention; wherein, 1 is acetaminophen; 1 is trigonelline; and 2 is quisqualine.

[0043] Figure 2 The image shows the TIC chromatogram of sample (S15) in a specific embodiment of the present invention; wherein, 1 is acetaminophen; 1 is trigonelline; and 2 is quisqualine.

[0044] Figure 3 The diagram shown is a speculative mass spectrometry fragmentation process of trigonelline, quisqualine, and acetaminophen in a specific embodiment of the present invention; where A represents trigonelline-ESI. + B indicates trigonelline-ESI - C indicates quisqualidine-ESI + D indicates quisqualidine-ESI - E indicates acetaminophen-ESI + F indicates acetaminophen-ESI - ;

[0045] Figure 4 The diagram shown is a cluster analysis tree diagram of 100 batches of Quisqualis indica samples in a specific embodiment of the present invention.

[0046] Figure 5 The diagram shown is a box plot of the trigonelline and quisqualine content in Quisqualis in a specific embodiment of the present invention.

[0047] Figure 6 The image shown is a superimposed spectral image of 100 batches of Quisqualis indica kernels in a specific embodiment of the present invention;

[0048] Figure 7 The figure shown is a line graph illustrating the effect of the number of principal factors on the RMSEP of the model in a specific embodiment of the present invention.

[0049] Figure 8 The figure shown is a correlation diagram between the measured and predicted values ​​of trigonelline and quisqualine content in a specific embodiment of the present invention. Detailed Implementation

[0050] To explain the technical contents, purposes and effects of the present application in detail, the following will be described in conjunction with the embodiments and the accompanying drawings.

[0051] 1 Material

[0052] 1.1 Drugs and reagents

[0053] 100 batches of Quisqualis indica L. seeds were purchased from Beijing, Hebei, Anhui and Fujian, and were identified by Professor Fan Shiming, a senior experimentalist of the College of Pharmacy, Fujian University of Traditional Chinese Medicine, as the seeds of Quisqualis indica L. in the Combretaceae family. The reference substances trigonelline (batch number: MUST-18050501, Chengdu Manxite Biological Technology Co., Ltd.), quisqualic acid (batch number: ZZS-20-112-A5, Baoji Chen Guang Biological Technology Co., Ltd.), and the internal standard substance acetaminophen (batch number: 100018-201610, China Institute for Drug Control) were used. Methanol and acetonitrile (mass spectrometry pure, Merck, Germany), formic acid (chromatography pure, Arladin Reagent Co., Ltd., Shanghai), and other reagents were all analytical pure.

[0054] 1.2 Instruments

[0055] Xevo TQS ultra-high performance liquid chromatography tandem triple quadrupole mass spectrometer (Waters, USA), EXPEC1350 portable near-infrared analyzer (Hangzhou Puyu Technology Development Co., Ltd.), CPA225D analytical balance (Sartorius, Germany), Milli-Q ultrapure water machine (Millipore, USA), 5810R centrifuge (Eppendorf, Germany), and Auto Prep200 full-automatic liquid sample processing workstation (Ruike Group Co., Ltd.).

[0056] 2 Methods and results

[0057] 2.1 Determination of the contents of trigonelline and quisqualic acid

[0058] 2.1.1 Chromatography-mass spectrometry conditions

[0059] Chromatography conditions: Waters ACQUITY BEH HILIC Amide column (2.1 mm x 100 mm, 1.7 μm), mobile phase: 0.1% (V / V) formic acid aqueous solution (A)-acetonitrile (B), gradient elution (0-0.5 min, 95% B→95% B; 0.5-1.0 min, 95% B→60% B; 1.0-4.0 min, 60% B→60% B; 4.0-4.1 min, 60% B→95% B; 4.1-6.0 min, 95% B→95% B), flow rate: 0.20 mL·min -1, column temperature was 45℃, injection volume was 2 μL.

[0060] Mass spectrometry conditions: electrospray positive and negative ion mode, capillary voltage was 3.5 kV, desolvation gas flow was nitrogen 800 L·h -1 , desolvation temperature was 500℃, cone gas flow was nitrogen 150 L·h -1 , ion source temperature was 150℃, collision gas was argon. The mass spectrometry optimization parameters of the components to be tested and the internal standard were shown in Table 1, the MRM chromatogram was shown in Figure 1 , the TIC graph of the sample (S15) was shown in Figure 2 , and the secondary mass spectrum of trigonelline, hydroxycitric acid and the internal standard was shown in Figure 3 . Figure 1 The blank (A) was blank solvent, the control (B) was the mixed control in item “2.1.2.1”, and the sample C was the test sample solution prepared in item “2.1.2.2”.

[0061] Table 1

[0062]

[0063] 2.1.2 Preparation of solutions

[0064] 2.1.2.1 Control solution and internal standard solution: accurately weighed the control of trigonelline and hydroxycitric acid, dissolved in methanol to prepare a single control stock solution with a mass concentration of 0.52 mg·mL -1 and 0.85 mg·mL -1 , respectively; accurately measured the control stock solution into a volumetric flask, added 50% (V / V) methanol water to constant volume, and shook well to obtain a mixed control stock solution with a mass concentration of 2.60 μg·mL -1 and 4.25 μg·mL -1 , respectively;

[0065] accurately weighed the acetaminophen, dissolved in methanol to prepare an internal standard stock solution with a mass concentration of 0.47 mg·mL -1 ; and accurately measured the internal standard stock solution, diluted with 50% (V / V) methanol water, and shook well to obtain an internal standard solution with a mass concentration of 18.8 ng·mL -1 .

[0066] 2.1.2.2 Preparation of test sample solution: accurately weighed about 0.1 g of the powdered Terminalia fruit (passed through a 60 mesh sieve) into a conical flask with a stopper. Accurately added 50% (V / V) methanol water solution 50 mL, tightly stoppered, weighed, treated with ultrasonic (power 250 W, frequency 50 kHz) for 30 min, cooled, weighed again, supplemented with 50% methanol water to make up the weight loss, and shook well at 10000 r·min -1Centrifuge for 10 min, take 0.25 mL of supernatant into 20 mL volumetric flask, dilute to the mark with 50% methanol water, filter through 0.22 μm filter membrane, take the filtrate, add internal standard solution at a volume ratio of 1:1, and obtain.

[0067] 2.1.3 Methodology investigation

[0068] 2.1.3.1 Linear relationship investigation: Using the automatic liquid sample processing workstation, 0.2, 1, 2, 5, 8, 10 mL of the mixed control solution in item 2.1.2.1 was precisely pipetted into a 100 mL volumetric flask, diluted to the mark with 50% methanol water, added with internal standard solution at a volume ratio of 1:1, and prepared into a series of control mixed solution with gradient concentrations, which was analyzed by sample injection under the conditions in item 2.1.1. The peak area ratio (y) of the tested component to the internal standard and the mass concentration (x) of the tested component were used as the vertical and horizontal coordinates for linear regression, and a standard curve was drawn, from which a regression equation and a correlation coefficient (r) were obtained. The lower limit of detection (LLOD) and the lower limit of quantification (LLOQ) were S / N≥3 and S / N≥10, respectively, and the results are shown in Table 2.

[0069] Table 2

[0070]

[0071]

[0072] 2.1.3.2 Precision test: The mixed control solution in item 2.1.3.1 (in which the concentrations of trigonelline and hydroxycitric acid were 130 ng·mL -1 and 212.5 ng·mL -1 , respectively) was analyzed by sample injection under the conditions in item 2.1.1, the peak area was recorded, and the RSD of the concentration was calculated. The intraday precision RSD of trigonelline and hydroxycitric acid was 1.67% and 2.80% (n=6) respectively, which was measured by 6 times of sample injection within 1 d. The interday precision RSD of trigonelline and hydroxycitric acid was 1.91% and 3.15% (n=9) respectively, which was measured by 3 times of sample injection per day for 3 consecutive days. The experimental results showed that the precision of the instrument was good.

[0073] 2.1.3.3 Stability test: The sample solution was prepared according to the method in item 2.1.2.2, and analyzed by sample injection under the conditions in item 2.1.1 at 0, 2, 4, 6, 8, 10, 12, and 24 h, the peak area was recorded, the mass concentration was calculated according to the standard curve, and the RSD of trigonelline and hydroxycitric acid was 1.21% and 2.71% (n=6) respectively. It showed that the sample solution was stable within 24 h.

[0074] 2.1.3.4 Reproducibility test: 6 portions of the same batch of Premnae Semen (S15) were precisely weighed, and the test solution was prepared according to the method of “2.1.2.2”. The sample was injected under the conditions of “2.1.1”, and the average contents of trigonelline and premniamine were calculated to be 7.888 mg·g -1 and 8.924 mg·g -1 , respectively, with RSDs of 0.80% and 2.15% (n = 6), respectively. This indicated that the method had good reproducibility.

[0075] 2.1.3.5 Sample addition recovery test: 6 portions of the same batch of Premnae Semen powder were precisely weighed, about 0.05 g, and the control was added according to the approximate 1:1 of the reproducibility result. The test solution was prepared according to the method of “2.1.2.2”, and the sample was analyzed under the conditions of “2.1.1”. The recovery rate was calculated, and the results are shown in Table 3 (n = 6), which indicated that the method had good recovery rate.

[0076] Table 3

[0077]

[0078]

[0079] 2.1.3.6 Sample content determination: 0.1 g of each batch of Premnae Semen powder was precisely weighed, and the test solution was prepared according to the method of “2.1.2.2”. The sample was analyzed under the conditions of “2.1.1”, and the content of the two components in the sample (mg·g -1 , n = 3) was calculated by internal standard method. The results are shown in Table 4.

[0080] Table 4

[0081]

[0082]

[0083]

[0084] SPSS 26.0 data processing software was used for analysis: the mass fraction of the two components was used as the variable, the data conversion method was principal component conversion, the clustering distance was Euclidean distance, and the clustering method was Ward's method. The clustering analysis of 100 batches of Premnae Semen samples was performed, and the results are shown in Figure 4 . Further, Origin 2019b software was used for content statistical analysis of the 4 types of Premnae Semen samples, and the box plot distribution of the contents of the two components is shown in Figure 5 .

[0085] 2.2 Establishment of near-infrared spectroscopy quantitative model of trigonelline and premniamine

[0086] 2.2.1 Sample processing and near infrared spectral information collection

[0087] 100 batches of semen cassiae medicinal materials were placed in a 60°C oven for drying for 6h. The dried medicinal materials were crushed and passed through a 60-mesh sieve for standby use. An appropriate amount of semen cassiae powder was placed in a measuring cup and evenly spread. The built-in background of the instrument was used as the reference. The collection parameters were as follows: each spectrum was obtained at a wavelength range of 1000-1799 nm with a resolution of 10 cm -1 -1, the collection mode was diffuse reflection, the scanning number was 64, each sample was continuously scanned 3 times, and the average spectrum was obtained. The superimposed graph of the original spectra of 100 batches of semen cassiae is shown in Figure 6 .

[0088] 2.2.2 Division of calibration set and validation set

[0089] The obtained spectral data of 100 batches of semen cassiae were randomly grouped using RIMP ClientV2.0 software. 60 samples were selected to form the calibration set, 30 samples were used as the validation set, and the remaining 10 samples were used as the external validation. It was necessary to ensure that the content range of the samples in the validation set was within the content range (mg·g -1 ) of the samples in the calibration set. See Table 5.

[0090] Table 5

[0091]

[0092]

[0093] 2.2.3 Screening of different spectral pretreatment methods

[0094] During the process of near infrared spectrum collection, it is easy to be interfered by noise, light and baseline drift, etc. The pretreatment of the collected spectral information can eliminate or reduce the influence of the above factors on the near infrared model to a certain extent. The near infrared spectrum of semen tribuli scanned was imported into RIMP Client V2.0 software, PLS was used as the modeling algorithm, and the original spectrum was pretreated by combining Savitzky-Golay smoothing (S-G smoothing), multivariate scatter correction (MSC), mean centering, standard normal variate (SNV), orthogonal signal correction (OSC) and various combinations of pretreatment methods, and a calibration model was established. The performance of the calibration model was evaluated by four indicators of root mean square error of calibration (RMSEC), root mean square error of prediction (RMSEP), correlation coefficient of calibration (Rc) and correlation coefficient of prediction (Rp). When the correlation coefficient is closest to 1, and the RMSEC and RMSEP are the smallest, the model is the best. After comparison, the optimal pretreatment method for the quantitative model of trigonelline is MSC+OSC+S-G smoothing, and the optimal pretreatment method for the quantitative model of semen tribuli amine is S-G smoothing+SNV, as shown in Table 6.

[0095] Table 6

[0096]

[0097]

[0098] 2.2.4 Selection of modeling wavelength

[0099] The selection of modeling wavelength plays an important role in filtering redundant information, simplifying data and improving model stability. Different wavelengths contain different information, so selecting appropriate wavelengths for modeling can obtain an accurate quantitative model. In this experiment, the full wavelength and multiple wavelength combinations were compared to investigate the quantitative model, and finally the modeling wavelength of trigonelline and semen tribuli amine was determined as 1000-1799 nm, as shown in Table 7.

[0100] Table 7

[0101]

[0102] 2.4.5 Determination of the number of principal components of the model

[0103] The number of principal components is an important factor in the quantitative model. When the number of principal components is too small, some original information is lost, which leads to the decrease of the prediction accuracy of the model. If the number of principal components is too large, the model will appear over-fitting phenomenon, and the prediction ability of the model will decrease. In this experiment, the number of principal components from 1 to 20 was compared to investigate the quantitative model, and RMSEP was used as an index. When RMSEP was the smallest, the selected number of principal components was the best. The best number of principal components of trigonelline and psicose were 8 and 12, respectively, as shown in Table 2. Figure 7 .

[0104] 2.4.6 Model establishment and test

[0105] The PLS combined with the optimized pretreatment method was used to establish the quantitative model of the two components under the above modeling wavelength and the best number of principal components. The Rc of the quantitative model of trigonelline and psicose was 0.9770 and 0.9718, respectively, the Rp was 0.9553 and 0.9464, respectively, the RMSEC was 0.3739 and 0.2848, respectively, and the RMSEP was 0.4902 and 0.3635, respectively. It can be seen from Table 3 that the measured value and the near-infrared predicted value had good correlation. Figure 8

[0106] Ten batches of psidium were selected for external validation, which were not involved in modeling. The contents of trigonelline and psicose in 10 batches of samples were predicted by the quantitative model, and compared with the measured values. The results were shown in Table 8. The average relative deviation of the NIR predicted value and the measured value of trigonelline and psicose was 0.242% and 0.252%, respectively. The established model had good prediction ability.

[0107] Table 8

[0108]

[0109] 3 Results and analysis

[0110] 3.1 Color, mass spectrum condition optimization

[0111] ​Trigonelline and Quisqualic acid are the main active components of Semen Terminaliae chebulae for expelling parasites. Trigonelline can exert anti-inflammatory and antioxidant effects by inhibiting oxidative stress and inflammation in hippocampus and properly regulating NF-κB / TLR4 and AChE activity signaling. Pai et al. (Pai K S, Ravindranath V. Quisqualic acid-induced neurotoxicity is protected by NMDA and non-NMDA receptor antagonists [J]. Neurosci Lett, 1992, 143(1-2): 177-180.) found that Quisqualic acid can induce toxic reactions by mediating the activation of non-N-methyl-D-aspartate receptors, thereby exerting the efficacy of expelling parasites by paralyzing roundworms. Therefore, both Trigonelline and Quisqualic acid are important quality markers (Q-Markers) of Semen Terminaliae chebulae. Currently, the detection method for Quisqualic acid is a derivatization method, which includes HPLC-UV detection after derivatization with 2-nitrobenzenesulfonyl chloride or HPLC-UV detection after derivatization with 2,4-dinitrochlorobenzene. However, the derivatization method requires sample defatting and color development processing, and has low sample detection throughput and long time consumption. In addition, Quisqualic acid and Trigonelline have large polarity, and in the pre-experiment, it was found that the two components have poor retention on a common C18 chromatographic column, and the peak is out quickly, which leads to detection interference. Accordingly, the inventors used hydrophilic interaction chromatography coupled with triple quadrupole mass spectrometry (UHILIC-MS / MS method) to test different mobile phase systems of methanol-acid water and acetonitrile-acid water, respectively. The acetonitrile system is better, and then the inventors investigated different proportions (V / V) of acid water in the acetonitrile system, including 0.1% formic acid water, 0.5% formic acid water, 0.1% acetic acid water and 0.5% acetic acid water. The results showed that 0.1% formic acid water is the best, and finally the inventors determined that the acetonitrile-0.1% formic acid water system is the best for the separation of each component.

[0112] In the mass spectrometry condition optimization, in order to improve the sensitivity of mass spectrometry detection, the MRM of Quisqualic acid and Trigonelline was optimized respectively, and the mass spectrometry detection condition of Quisqualic acid was reported for the first time. First, in the positive ion mode, the parent ion of Quisqualic acid is 190.11, which can produce m / z 100.85 [M+H-C3H6NO2] + daughter ion through secondary cleavage, and m / z 100.85 is [C2HN2O3] + peak with the highest abundance; the parent ion of Trigonelline is 138.05, which can produce m / z 64.01 [M+H-C3H6O2]+, 78.18 [M+H-C2H4O2] + , 92.24 [M+H-CH2O2] +94.06 [M+H-CO2] + Four daughter ions, of which [C6H6N] has a m / z of 92.24. + The peak represents the most abundant daughter ion. Secondly, in negative ion mode, the precursor ion of quisqualine is 188.21, of which 101.08 is [C₂HN₂O₃]. - The peak was the most abundant; the precursor ion of trigonelline was 135.92, of which 92.12 were [C6H6N]. - The peak represents the most abundant daughter ion. Third, comparing the response under the total ion chromatogram and the mode response under MRM positive / negative switching revealed that quinceidine was superior in negative ion mode to its positive ion mode, while trigonelline was superior in positive ion mode. Therefore, the optimal quantitative ion pair for trigonelline was selected as 138.30→92.10, and for quinceidine as 188.01→101.01. Fourth, for the internal standard acetaminophen, the precursor ion in positive ion mode was 152.02, and the optimal quantitative daughter ion was m / z 110.06 [M+H-C2H2O]. + In negative ion mode, the parent ion is 150.20, and the optimal quantitative product ion is m / z 107.20 [M+H-C2H2O]. - All showed good responses and had similar peak values, which can meet the internal standard detection requirements for trigonelline and quisqualine.

[0113] 3.2 Analysis of Content Determination Results

[0114] Content determination showed that the contents of trigonelline and quisqualine in 100 batches of Quisqualis indica seed kernels ranged from 5.057 to 11.123 mg / g. -1 and 7.081~12.846mg·g -1 SPSS 26.0 software was used for analysis. Hierarchical clustering was employed, with the mass fractions of two components as variables. Principal component transformation was used as the data transformation method, Euclidean distance was used as the cluster distance, and Wald's method was used for cluster analysis. Cluster analysis was performed on 100 batches of Quisqualis indica samples. When the Euclidean distance was 15, the 100 batches of samples were well distinguished, and could be divided into two chemical phenotype categories (I-II). Further analysis of the box plot distribution of the two component contents for the two phenotypes revealed differences in content between the different chemical phenotypes: the content of trigonelline was significantly higher in phenotype II than in phenotype I; for quisqualine, there was no significant difference between phenotypes I and II. This suggests that the main reason for the differences in the chemical phenotypes of Quisqualis indica kernels is the difference in trigonelline content.

[0115] 3.3 Model Optimization of NIR Method

[0116] The present study constructed a method for rapid determination of trigonelline and psicamine in Fructus Tribuli by NIRS. After near infrared spectrum scanning, the effects of different spectral pretreatment methods, modeling wavelengths and principal factor numbers on the quantitative analysis models of the two compounds were investigated. After comparison, the optimal pretreatment methods for the quantitative models of trigonelline and psicamine were MSC+OSC+S-G smoothing and S-G smoothing+SNV, respectively, the modeling wavelengths were both 1000-1799 nm, and the optimal principal factor numbers were 8 and 12, respectively. Then, PLS was used to establish the quantitative models of psicamine and trigonelline in 90 batches of Fructus Tribuli. The Rc of trigonelline and psicamine was 0.9770 and 0.9718, respectively, the Rp was 0.9553 and 0.9464, respectively, the RMSEC was 0.3739 and 0.2848, respectively, and the RMSEP was 0.4902 and 0.3635, respectively. In addition, 10 batches of samples outside the modeling set were used for external validation. The average relative deviation of the NIR predicted values and the measured values of the external validation set was 0.242% and 0.252%, respectively, indicating that there was no systematic error between the NIRS method and the UHILIC-MS / MS method. The method is simple and feasible.

[0117] In conclusion, the present study firstly simultaneously determined the contents of trigonelline and psicamine by UHILIC-MS / MS, and established the quantitative models of psicamine and trigonelline in Fructus Tribuli by NIRS. The method is rapid and non-destructive, and can provide a new technical method for the rapid detection of the quality evaluation of Fructus Tribuli.

[0118] The above only describes the embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent transformation or direct or indirect application in related technical fields based on the content of the present application specification and drawings is also included in the patent protection scope of the present application.

Claims

1. A method for determining the content of trigonelline and quisqualine in Quisqualis in kernels, characterized in that, include: The first dataset on the content of trigonelline and quisqualine in Quisqualis kernel was obtained using UHILIC-MS / MS method; A second dataset of near-infrared spectra of trigonelline and quisqualine in Quisqualis kernel was obtained based on the NIRS method; The second dataset is preprocessed; The preprocessing used for the second dataset on trigonelline included multivariate scattering correction, orthogonal signal correction, and Savitzky-Golay smoothing. The preprocessing used for the second dataset on quisqualine was Savitzky-Golay smoothing and standard normal transformation; Based on PLS, a content model for trigonelline and quisqualine was established based on the first and second datasets, respectively. The near-infrared spectrum of the Quisqualis indica kernel sample was obtained and then input into the content model to obtain the content of trigonelline and quisqualine in the Quisqualis indica kernel sample; The modeling wavelength for establishing the content model is 1000~1799nm. The number of principal factors for trigonelline in the content model is 8, and the number of principal factors for quisqualine is 12. The mass spectrometry conditions for the UHILIC-MS / MS method are as follows: Mode: Electrospray positive and negative ions; Capillary voltage: 3.5 kV; The desolventizing gas stream is 800 L·h of nitrogen. -1 ; Desolventization temperature: 500 ℃; Conical orifice gas flow: Nitrogen 150 L·h -1 ; Ion source temperature: 150 ℃; Collision gas: Argon; Trigonelline was processed using an electrospray positive ion mode, while quisqualine was processed using an electrospray negative ion mode. The chromatographic conditions for the UHILIC-MS / MS method include: Chromatographic column: Waters ACQUITY BEH HILIC Amide column, 2.1 mm × 100 mm, 1.7 µm; Mobile phase: A is 0.1% formic acid aqueous solution - B is acetonitrile; Gradient elution: 0–0.5 min, 95%B → 95%B; 0.5~1.0 min, 95%B→60%B; 1.0~4.0 min, 60%B→60%B; 4.0~4.1 min, 60%B→95%B; 4.1~6.0 min, 95%B→95%B.

2. The content determination method according to claim 1, characterized in that, The chromatographic conditions for the UHILIC-MS / MS method also include: Flow rate: 0.20 mL·min -1 ; Column temperature: 45 ℃; Injection volume: 2 μL.

3. The content determination method according to claim 1, characterized in that, The mass spectrometry determination conditions for trigonelline are as follows: Quantitative ion pair: 138.30→92.10 m / z; Collision energy: 18 eV; Ion pair mode: ESI + .

4. The content determination method according to claim 1, characterized in that, The mass spectrometry conditions for determining quisqualidine are as follows: Quantitative ion pair: 188.01→101.01 m / z; Collision energy: 12 eV; Ion pair mode: ESI - .

Images