A method for quantitatively determining the content of caffeoyl phenethyl alcohol glycosides in Cistanche tubulosa (Schenk) O. W. Becker 1 A method for quantitatively determining the content of caffeoyl phenethyl alcohol glycosides in Cistanche tubulosa (Schenk) O. W. Becker

Abstract

The application discloses a kind of based on 1 A method for quantitatively determining caffeoyl phenethyl alcohol glycosides in Cistanche by H-NMR, comprising sample pretreatment, internal standard solution preparation, test sample preparation for nuclear magnetic resonance, nuclear magnetic resonance hydrogen spectrum acquisition, automatic spectrum processing, data processing and content calculation. The application is different from evaluating the content of a few CPGs monomers, can directly determine the total content of all CPGs, can comprehensively and truly reflect the intrinsic quality of Cistanche, overcomes the defects that HPLC can only measure a small amount of monomers and ultraviolet method has poor specificity, and is more consistent with the overall evaluation idea of traditional Chinese medicine quality control. The application establishes a phase correction, baseline correction, solvent peak calibration, characteristic peak identification and integral integrated process without manual operation, significantly reduces human error, and can improve detection efficiency and result stability.

Description

Technical Field

[0001] This invention belongs to the field of quality control technology for traditional Chinese medicinal materials, specifically relating to a method based on... 1 A quantitative method for determining caffeoyl phenylethanol glycosides in Cistanche deserticola using ¹H-NMR. Background Technology

[0002] Cistanche deserticola ( Cistanche deserticola Cistanche deserticola (Ma) is a perennial parasitic herb belonging to the genus Cistanche of the family Orobanchaceae. Often called "desert ginseng," it is a traditional and precious Chinese medicinal herb with effects such as tonifying kidney yang, nourishing essence and blood, and moistening the intestines to relieve constipation. It is widely used in clinical practice of traditional Chinese medicine and in the field of health products. Phenylene glycosides are one of the main active components of Cistanche deserticola. They are a class of glycosides formed by the combination of phenylethanol and sugar units. In the plant, they mainly exist in the form of caffeoyl phenylethanol glycosides (hereinafter referred to as CPGs), possessing various pharmacological activities such as antioxidant, anti-aging, neuroprotective, and immunomodulatory effects. Their content is an important indicator for evaluating the quality of Cistanche deserticola.

[0003] Currently, the main methods for quantitative determination of CPGs in Cistanche deserticola are high-performance liquid chromatography (HPLC) and ultraviolet spectrophotometry. HPLC indirectly reflects the quality of Cistanche deserticola by determining the content of a few phenylethanoid glycoside monomers such as echinacoside and verbascoside. However, Cistanche deserticola contains a wide variety of phenylethanoid glycosides, with more than twenty reported CPG components. Traditional HPLC methods can only determine seven of these monomers, as shown in patents CN 107727763 B and CN 121202935A, which cannot comprehensively reflect the content of CPG components in Cistanche deserticola and have certain limitations. Ultraviolet spectrophotometry collects the absorbance of Cistanche deserticola extract at around 330 nm and calculates the content according to Lambert-Beer's law. However, some non-phenylethanoid glycoside components also absorb at this wavelength, therefore this method lacks specificity. In addition, both methods are cumbersome and time-consuming, requiring complex sample pretreatment and chromatographic condition optimization, and have high costs for purchasing standards.

[0004] 1H NMR spectrum ( 1 H-NMR, as a structural analysis technique, has been widely used in the quantitative analysis of components in traditional Chinese medicine in recent years. 1¹H-NMR quantitative analysis offers advantages such as no need for monomer separation, simple operation, rapid and efficient processing, and stable results. It can comprehensively reflect the total content of target components in a sample by selecting characteristic proton signals for quantitative determination. However, current techniques mostly focus on the quantitative analysis of single components, such as CN 106568790 A, and no methods for analyzing a class of structural analogues have been found. Furthermore, existing NMR data processing software (such as TopSpin and MestReNova) is mostly commercial software, with cumbersome operation steps and low automation. Especially for complex mixtures like Cistanche deserticola CPGs, characteristic signal identification and integration often require manual sample processing, resulting in low efficiency. The lack of an integrated method that can automatically preprocess NMR data, identify characteristic peaks, and perform integration calculations further limits its application. 1 The application of H-NMR technology in the rapid quality evaluation of Cistanche deserticola. Summary of the Invention

[0005] The purpose of this invention is to provide a method based on 1 A quantitative NMR method for determining caffeoyl phenylethanol glycosides in Cistanche deserticola is proposed. This method enables high-throughput determination of the total CPG content in Cistanche deserticola without human intervention, with accurate characteristic peak identification, thus improving the quality evaluation level of Cistanche deserticola.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A type based on 1 The method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR includes the following steps: S1. Sample pretreatment: The dried Cistanche deserticola was sliced, crushed, sieved, weighed accurately, and extracted with an organic solvent by ultrasonic extraction. After centrifugation, the supernatant was collected and concentrated to dryness to obtain the sample extract. S2. Preparation of internal standard solution: Accurately weigh 2,3,5-triiodobenzoic acid powder, add deuterated solvent, mix thoroughly, shake to dissolve, and then dilute to the mark to obtain an internal standard solution. Store at 4°C for later use. S3. NMR Sample Preparation: Add the deuterated solvent and the internal standard solution obtained in S2 to the sample extract obtained in S1, mix well, and obtain the NMR test sample; S4, Nuclear Magnetic Resonance Hydrogen Spectrum Acquisition: The NMR test sample prepared by S3 was placed in an NMR spectrometer using a superconducting NMR spectrometer with a MHz or higher. The pulse sequence was zg30, the relaxation delay time was 10-30 s, the temperature was 295-303 K, the number of scans was 32-128, and the acquisition time was 1.5-3.5 s. 1H-NMR spectrum; S5. Automatic spectrum processing: The system automatically performs phase correction, solvent peak calibration, peak identification, characteristic peak identification, and integration on the NMR data obtained from S4. It can autonomously identify the reference peak as a d peak with a chemical shift between 7.40 and 7.49 and a proton number of 1, which is the characteristic proton signal of the internal standard, and normalize the peak area integral to 1.0000. It can also identify d peaks with a target integration range between 7.50 and 7.64 and a proton number of 1, which is the characteristic proton signal of caffeoyl phenylethanol glycosides, and automatically integrate them. S6. Data Processing and Content Calculation: The characteristic proton signals of the internal standard obtained from S5 are non-overlapping and independent with the characteristic proton signals of caffeoyl phenylethanol glycosides. Based on the integral area ratio of the characteristic proton signals of the internal standard and the characteristic proton signals of CPGs, and the amount of internal standard added, the percentage content of caffeoyl phenylethanol glycosides in the sample, calculated as echinacoside, is calculated.

[0007] To further realize the present invention, the organic solvent mentioned in S1 is methanol or a 70% ethanol solution.

[0008] To further realize the present invention, the deuterated solvent in S2 or S3 is one or more of deuterated methanol, deuterated dimethyl sulfoxide, or deuterated pyridine.

[0009] To further realize the present invention, the amount of deuterated solvent added in S3 is 500 μL.

[0010] To further realize the present invention, the concentration of the internal standard solution in S2 is 5-10 mg / mL.

[0011] To further realize the present invention, the amount of internal standard solution added in S3 is 50 μL.

[0012] To further realize the present invention, the formula for calculating the percentage content of the caffeoyl phenylethanol glycosides in S6 is as follows: Among them, A sam for δ H The integral area of ​​the characteristic proton signal in the sample at 7.50-7.64; A is The integral area of ​​the internal standard quantum signal; N is The number of protons in the characteristic signal of the internal standard; N sam The number of protons represented by the sample characteristic signal; M sam The relative molecular mass of echinacoside, the most abundant glycoside in Cistanche deserticola; M is The relative molecular mass of the internal standard; m is The mass of the added internal standard; m sam The mass of the sample taken; w is This represents the mass fraction of the internal standard.

[0013] To further realize the present invention, the Cistanche deserticola mentioned in S1 is Cistanche tubulosa or Cistanche deserticola.

[0014] The advantages of this invention compared to the prior art are as follows: (1) This invention differs from evaluating the content of a few CPGs monomers. It can directly determine the total content of all CPGs, which can comprehensively and truthfully reflect the intrinsic quality of Cistanche deserticola. It overcomes the shortcomings of HPLC, which can only measure a small number of monomers, and the poor specificity of ultraviolet method. It is more in line with the overall evaluation idea of ​​quality control of traditional Chinese medicine.

[0015] (2) In view of the problems of existing NMR methods relying on commercial software, excessive manual intervention and large errors, this invention establishes an integrated process of phase correction, baseline correction, solvent peak calibration, characteristic peak identification and integration that does not require manual operation, which significantly reduces human error and can improve detection efficiency and result stability. Attached Figure Description

[0016] Figure 1 The extract of Cistanche deserticola after CDnmr treatment in Example 1 of this invention. 1 H-NMR spectrum; Figure 2 The extract of Cistanche deserticola after CDnmr treatment in Example 2 of this invention. 1 H-NMR spectrum; Figure 3 The extract of Cistanche deserticola after CDnmr treatment in Example 3 of this invention. 1 H-NMR spectrum Figure 4 The extract of Cistanche deserticola after CDnmr treatment in Example 4 of this invention. 1 H-NMR spectrum; Figure 5 The extract of Cistanche deserticola after CDnmr treatment in Example 5 of this invention. 1 H-NMR spectrum; Figure 6 The extract of Cistanche deserticola after CDnmr treatment in Example 6 of this invention. 1 H-NMR spectrum; Figure 7 The extract of Cistanche deserticola after CDnmr treatment in Example 7 of this invention. 1H-NMR spectrum; Figure 8 The extract of Cistanche deserticola after CDnmr treatment in Example 8 of this invention. 1 H-NMR spectrum; Figure 9 The extract of Cistanche deserticola after CDnmr treatment in Example 9 of this invention. 1 H-NMR spectrum; Figure 10 The extract of Cistanche deserticola after CDnmr treatment in Example 10 of this invention. 1 H-NMR spectrum; Figure 11 The extract of Cistanche deserticola after CDnmr treatment in Example 11 of this invention. 1 H-NMR spectrum; Figure 12 The extract of Cistanche deserticola after CDnmr treatment in Example 12 of this invention. 1 H-NMR spectrum; Figure 13 The extract of Cistanche deserticola after CDnmr treatment in Example 13 of this invention. 1 H-NMR spectrum; Figure 14 This is a linear relationship graph between different concentrations of echinacoside and the integral ratio in Experimental Example 1 of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0018] Cistanche deserticola samples: Cistanche tubulosa from Yili, Xinjiang (collected in 2023, CD2308001), Cistanche tubulosa from Yutian, Xinjiang (collected in 2023, CD2308002), Cistanche deserticola from Anhui (collected in 2024, CD2406001), and Cistanche deserticola from Alashan, Inner Mongolia (collected in 2025, CD2507001). These samples were identified by Researcher Jin Hui as plants of the genus Cistanche in the family Orobanchaceae. The specimens are deposited at the Center for Natural Products and Chemical Measurement, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences.

[0019] I. Example: Example 1: (1) Sample pretreatment: The dried Cistanche tubulosa sample from Yutian, Xinjiang was pulverized, sieved, and accurately weighed. 0.1005 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 5.0 mg of 2,3,5-triiodobenzoic acid powder, place it in a 1.0 mL volumetric flask, add 500 μL of deuterated methanol, mix thoroughly, shake to dissolve, and then add an appropriate amount of deuterated methanol to make up to the mark to obtain the internal standard solution, which is stored at 4℃ for later use. (3) NMR test sample preparation: After drying the sample, add 500 μL of deuterated methanol and sonicate for 1 min to dissolve it completely. Accurately pipette 50 μL of internal standard solution into the sample solution and shake for 30 s to mix thoroughly to obtain the NMR sample. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: a) Pulse sequence: zg30 b) Relaxation delay time: 15 s c) Temperature: 298.15 K d) Number of scans: 64 e) Data acquisition time: 2.5 s f) Spectral width: 16 ppm g) Center frequency: 5 ppm; (5) Construction of the automated NMR data processing program (CDnmr): Based on the Python environment, the CDnmr program is constructed, and the script is as follows: import nmrglue as ng import numpy as np import matplotlib.pyplot as plt from scipy import signal import warnings warnings.filterwarnings('ignore') data_path = r"E:\licp\nmrai\RCR\1\pdata\1" #Read data dic, data = ng.bruker.read_pdata(data_path) udic = ng.bruker.guess_udic(dic, data) uc = ng.fileiobase.uc_from_udic(udic) ppm = uc.ppm_scale() spec = np.real(data) # Baseline Correction def baseline_correct(spec, deg=2): x = np.arange(len(spec)) poly = np.polyfit(x, spec, deg) return spec - np.polyval(poly, x) spec = baseline_correct(spec) spec = signal.savgol_filter(spec, 7, 2) # Baseline Correction and Normalization spec -= np.min(spec) spec / = np.max(spec) # Deuterated solvent calibration ref_ppm = 4.88 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift # Range Enlargement roi_mask = (ppm_cal>= 6.0)&(ppm_cal<= 8.0) ppm_roi = ppm_cal[roi_mask] spec_roi = spec[roi_mask] # Target proton signal amplification peaks_idx, _ = signal.find_peaks( spec_roi, height=0.0001, prominence=0.0001, width=1 ) if len(peaks_idx) == 0: print("not found") plt.figure(figsize=(12,5)) plt.plot(ppm_roi, spec_roi, 'b-') plt.xlim(6,8) plt.title("6.0-8.0 ppm") plt.show() exit() peak_ppm_list = ppm_roi[peaks_idx] peak_int_list = spec_roi[peaks_idx] # Finding the reference peak proton signal ref_mask = (peak_ppm_list>= 7.40)&(peak_ppm_list<= 7.49) ref_indices = np.where(ref_mask)[0] if len(ref_indices) == 0: print("not found") exit() ref_i = ref_indices[0] # Baseline Peak Integral def get_peak_ranges(peak_indices, max_len): ranges = [] peaks = sorted(peak_indices) n = len(peaks) for i in range(n): left = 0 if i == 0 else (peaks[i]+ peaks[i-1]) / / 2 right = max_len -1 if i == n-1 else (peaks[i]+ peaks[i+1]) / / 2 ranges.append((left, right)) return ranges ranges = get_peak_ranges(peaks_idx, len(spec_roi)) areas = [np.sum(spec_roi[l:r]) for l, r in ranges] # Normalization processing / output signal ref_area = areas[ref_i] ratios = [a / ref_area for a in areas] plt.figure(figsize=(12,5)) plt.plot(ppm_roi, spec_roi, 'black', linewidth=1.5) for p, i, r in zip(peak_ppm_list, peak_int_list, ratios): plt.text(p, i - 0.04, f"{r:.4f}", ha="center", fontsize=10,color="red",weight="bold") plt.xlim(6.0, 8.0) plt.title("6.0-8.0 ppm | 7.40 = 1.0000", fontsize=13) plt.xlabel("Chemical Shift (ppm)") plt.grid(False) plt.tight_layout() plt.show() print("="*70) print(" 6.0~8.0 ppm results(7.40~7.49 ppm = 1.0000)") print("="*70") for i, (p, r) in enumerate(zip(peak_ppm_list, ratios), 1): # Searching for target proton signals # Target peak: 7.50-7.64 ppm if 7.50 <= p <= 7.64: # Output Results print(f" {i:2d}. {p:6.3f} ppm | integration = {r:8.4f} ★CPGs") else: print(f" {i:2d}. {p:6.3f} ppm | integration = {r:8.4f}"); (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 1 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.50-7.64 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method. The calculation formula is as follows: Calculations show that the CPG content in this batch of samples is 5.55%.

[0020] Example 2: (1) Sample pretreatment: The dried Cistanche tubulosa sample from Yili, Xinjiang was pulverized, sieved, and accurately weighed. 0.1002 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 10.0 mg of 2,3,5-triiodobenzoic acid powder and place it in a 1.0 mL volumetric flask. Add 500 μL of deuterated dimethyl sulfoxide and mix thoroughly. Shake to dissolve the solution and then add an appropriate amount of deuterated dimethyl sulfoxide to bring the volume to the mark to obtain the internal standard solution. Store at 4℃ for later use. (3) NMR test sample preparation: After the sample to be tested is dried, 500 μL of deuterated dimethyl sulfoxide is added, and the mixture is sonicated for 1 min to dissolve it completely. 50 μL of internal standard solution is accurately pipetted into the sample solution, and the mixture is shaken for 30 s to mix thoroughly to obtain the NMR sample to be tested. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: h) Pulse sequence: zg30 i) Relaxation delay time: 15 s j) Temperature: 298.15 K k) Number of scans: 64 l) Data acquisition time: 2.5 s Spectral width (m): 16 ppm n) Center frequency: 5 ppm; (5) CDnmr construction: Based on the Python environment, a CDnmr program was constructed. The script is the same as in Example 1, except that the deuterated solvent calibration is replaced with: # Deuterated solvent calibration ref_ppm = 2.50 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift; (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 2 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.50-7.64 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method. The calculation formula is as follows: Calculations showed that the CPG content in this batch of samples was 1.90%.

[0021] Although both belong to the Cistanche tubulosa genus, the effective component content of samples from Yili was significantly lower than that from Yutian. This may be due to differences in the host plant species and the parameters of the processing techniques. Meanwhile, literature research has also confirmed the significant impact of geographical origin on content differences. The article "The Significance of Separate Content Limits for Desert Cistanche and Cistanche tubulosa in the Pharmacopoeia of the People's Republic of China Based on Differential Component Analysis" reports that the effective component content of Cistanche tubulosa samples from some producing areas (such as samples T7 and T8 marked in the article) was much lower than that from other regions. This is consistent with the trend of lower content in samples from Yili, further indicating that the combined effects of multiple factors, including the ecological environment of the producing area, host germplasm, and processing methods, are likely the main reasons for this difference.

[0022] Example 3: (1) Sample pretreatment: The dried Cistanche deserticola sample from Inner Mongolia was pulverized, sieved, and accurately weighed. 0.1008 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 5.0 mg of 2,3,5-triiodobenzoic acid powder and place it in a 1.0 mL volumetric flask. Add 500 μL of deuterated pyridine, mix thoroughly, shake to dissolve, and then add an appropriate amount of deuterated pyridine to make up to the mark to obtain the internal standard solution. Store at 4℃ for later use. (3) NMR test sample preparation: After the sample to be tested is dried, 500 μL of deuterated pyridine is added, and the mixture is sonicated for 1 min to dissolve it completely. 50 μL of internal standard solution is accurately pipetted into the sample solution, and the mixture is shaken for 30 s to mix thoroughly to obtain the NMR sample to be tested. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: o) Pulse sequence: zg30 p) Relaxation delay time: 15 s q) Temperature: 298 K r) Number of scans: 64 s) Acquisition time: 2.5 s t) Spectral width: 16 ppm u) Center frequency: 5 ppm; (5) CDnmr construction: Based on the Python environment, a CDnmr program was constructed. The script is the same as in Example 1, except that the deuterated solvent calibration is replaced with: # Deuterated solvent calibration ref_ppm = 8.74 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift; (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 3 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.50-7.64 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method. The calculation formula is as follows: Calculations showed that the CPG content in this batch of samples was 1.45%.

[0023] Example 4: (1) Sample pretreatment: The dried Cistanche deserticola sample from Anhui was pulverized, sieved, and accurately weighed. 0.1006 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 5.0 mg of 2,3,5-triiodobenzoic acid powder, place it in a 1.0 mL volumetric flask, add 500 μL of deuterated methanol, mix thoroughly, shake to dissolve, and then add an appropriate amount of deuterated methanol to make up to the mark to obtain the internal standard solution, which is stored at 4℃ for later use. (3) NMR test sample preparation: After the sample to be tested is dried, 500 μL of deuterated dimethyl sulfoxide is added and sonicated for 1 min to dissolve it completely. 50 μL of internal standard solution is then precisely pipetted into the sample to be tested and shaken for 30 s to mix thoroughly to obtain the NMR test sample. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: v) Pulse sequence: zg30 w) Relaxation delay time: 15 s x) Temperature: 298.15 K y) Number of scans: 64 z) Data acquisition time: 2.5 s aa) Spectral width: 16 ppm ab) Center frequency: 5 ppm; (5) CDnmr construction: Based on the Python environment, a CDnmr program was constructed. The script is the same as in Example 1, except that the deuterated solvent calibration is replaced with: # Deuterated solvent calibration ref_ppm = 2.50 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift; (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 4 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.50-7.64 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method. The calculation formula is as follows: Calculations showed that the CPG content in this batch of samples was 0.80%.

[0024] Example 5: (1) Sample pretreatment: The dried Cistanche tubulosa sample from Yutian, Xinjiang was pulverized, sieved, and accurately weighed. 0.1004 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 5.0 mg of 2,3,5-triiodobenzoic acid powder, place it in a 1.0 mL volumetric flask, add 500 μL of deuterated methanol, mix thoroughly, shake to dissolve, and then add an appropriate amount of deuterated methanol to make up to the mark to obtain the internal standard solution, which is stored at 4℃ for later use. (3) NMR test sample preparation: After the sample to be tested is dried, 500 μL of deuterated pyridine is added and sonicated for 1 min to dissolve it completely. 50 μL of internal standard solution is then precisely pipetted into the sample to be tested and shaken for 30 s to mix thoroughly to obtain the NMR test sample. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: AC) Pulse sequence: ZG30 (ad) Relaxation delay time: 15 s ae) Temperature: 298.15 K (af) Number of scans: 64 (ag) Collection time: 2.5 s spectral width (ah): 16 ppm ai) Center frequency: 5 ppm; (5) CDnmr construction: Based on the Python environment, a CDnmr program was constructed. The script is the same as in Example 1, except that the deuterated solvent calibration is replaced with: # Deuterated solvent calibration ref_ppm = 4.88 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift; (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 5 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.50-7.64 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method. The calculation formula is as follows: Calculations showed that the CPG content in this batch of samples was 5.42%.

[0025] Example 6: (1) Sample pretreatment: The dried Cistanche deserticola sample from Inner Mongolia was pulverized, sieved, and accurately weighed. 0.1005 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 5.0 mg of 2,3,5-triiodobenzoic acid powder and place it in a 1.0 mL volumetric flask. Add 500 μL of dimethyl sulfoxide, mix thoroughly, shake to dissolve, and then add an appropriate amount of deuterated dimethyl sulfoxide to bring the volume to the mark to obtain the internal standard solution. Store at 4℃ for later use. (3) NMR test sample preparation: After the sample to be tested is dried, 500 μL of deuterated pyridine is added and sonicated for 1 min to dissolve it completely. 50 μL of internal standard solution is then precisely pipetted into the sample to be tested and shaken for 30 s to mix thoroughly to obtain the NMR test sample. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: aj) Pulse sequence: zg30 (ak) Relaxation delay time: 15 s al) Temperature: 298.15 K (am) Number of scans: 64 (a) Data collection time: 2.5 s αo) spectral width: 16 ppm (ap) Center frequency: 5 ppm; (5) CDnmr construction: Based on the Python environment, a CDnmr program was constructed. The script is the same as in Example 1, except that the deuterated solvent calibration is replaced with: # Deuterated solvent calibration ref_ppm = 8.47 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift; (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 6 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.5-7.7 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method, as shown in the following formula: Calculations showed that the CPG content in this batch of samples was 1.46%.

[0026] Example 7: (1) Sample pretreatment: The dried Cistanche tubulosa sample from Yutian, Xinjiang was pulverized, sieved, and accurately weighed. 0.1008 g of the sample was then passed through a 60-mesh sieve and placed in a 1.5 mL centrifuge tube. 1.0 mL of methanol was added, and the tube was sealed and placed in an ultrasonic cleaner. Ultrasonic extraction was performed at 40 Hz and 120 W for 60 min. After extraction, the tube was centrifuged at 7000 r / min for 10 min. The supernatant was transferred to a 1.5 mL centrifuge tube and dried in a 40℃ constant temperature drying oven to obtain the sample extract. (2) Preparation of internal standard solution: Accurately weigh 5.0 mg of 2,3,5-triiodobenzoic acid powder, place it in a 1.0 mL volumetric flask, add 500 μL of deuterated methanol, mix thoroughly, shake to dissolve, and then add an appropriate amount of deuterated methanol to make up to the mark to obtain the internal standard solution, which is stored at 4℃ for later use. (3) NMR test sample preparation: After the sample to be tested is dried, 250 μL of deuterated dimethyl sulfoxide and 250 μL of deuterated pyridine are added, and the mixture is sonicated for 1 min to dissolve it completely. 50 μL of internal standard solution is precisely pipetted into the sample to be tested, and the mixture is shaken for 30 s to mix thoroughly to obtain the NMR test sample. (4) NMR sample testing: Transfer the test sample to a 5 mm NMR tube, label it, and then acquire the proton NMR spectrum as follows: aq) Pulse sequence: zg30 (ar) Relaxation delay time: 15 s Temperature: 298.15 K (at) Number of scans: 64 (au) Data collection time: 2.5 s AV spectral width: 16 ppm (aw) Center frequency: 5 ppm; (5) CDnmr construction: Based on the Python environment, a CDnmr program was constructed. The script is the same as in Example 1, except that the deuterated solvent calibration is replaced with: # Deuterated solvent calibration ref_ppm = 2.50 peak_idx = np.argmin(np.abs(ppm - ref_ppm)) shift = ppm[peak_idx] - ref_ppm ppm_cal = ppm – shift; (6) Data processing and calculation: Run the CDnmr program in the Python environment and import the collected spectral data. Figure 7 Phase correction, baseline balancing, characteristic peak identification, and integration were performed, and the integration results within the range of 7.50-7.64 ppm were recorded. Using echinacoside (relative molecular mass: 786.72) as the calculation standard, the percentage content of CPGs in the sample was calculated using the internal standard method. The calculation formula is as follows: Calculations showed that the CPG content in this batch of samples was 5.48%.

[0027] Example 8: The method is the same as in Example 1. In sample pretreatment, the sample was *Cistanche deserticola* from the Alashan Desert in Inner Mongolia, and 70% ethanol was used as the organic solvent. Deuterated methanol was used in the preparation of the internal standard solution and the NMR test sample. The spectrum is shown below. Figure 8 The As / Ai ratio was 4.1943, and the CPG content was 1.61%.

[0028] Example 9: The method is the same as in Example 1. In sample pretreatment, the sample was *Cistanche deserticola* from the Alashan Desert in Inner Mongolia, and 70% ethanol was used as the organic solvent. Deuterated dimethyl sulfoxide was used in the preparation of the internal standard solution and the NMR sample. The spectrum is shown below. Figure 9 The As / Ai ratio was 4.0813, and the CPG content was 1.57%.

[0029] Example 10: The method is the same as in Example 1. In sample pretreatment, the sample was *Cistanche deserticola* from the Alashan Desert in Inner Mongolia, and 70% ethanol was used as the organic solvent. Deuterated pyridine was used in the preparation of the internal standard solution and the NMR test sample. The spectrum is shown below. Figure 10 The As / Ai ratio was 4.2498, and the CPG content was 1.64%.

[0030] Example 11: The method is the same as in Example 1. In sample pretreatment, the sample was *Cistanche tubulosa* from Yutian, Xinjiang, and 70% ethanol was used as the organic solvent. Deuterated methanol was used in the preparation of the internal standard solution and the NMR test sample. The spectrum is shown below. Figure 11 The As / Ai ratio was 14.1521, and the CPG content was 5.45%.

[0031] Example 12: The method is the same as in Example 1. In sample pretreatment, the sample was *Cistanche tubulosa* from Yutian, Xinjiang, and 70% ethanol was used as the organic solvent. Deuterated methanol was used in the preparation of the internal standard solution and the NMR test sample. The spectrum is shown below. Figure 12 The As / Ai ratio was 14.2033, and the CPG content was 5.47%.

[0032] Example 13: The method is the same as in Example 1. In sample pretreatment, the sample was *Cistanche tubulosa* from Yutian, Xinjiang, and 70% ethanol was used as the organic solvent. Deuterated methanol was used in the preparation of the internal standard solution and the NMR test sample. The spectrum is shown below. Figure 13 The As / Ai ratio was 14.1073, and the CPG content was 5.43%.

[0033] II. Experimental Example: Experimental Example 1: Optimization and Selection of Key NMR Parameters Since the active ingredients of desert Cistanche and tubular Cistanche are the same, only the content is different, and the chemical shift, coupling constant and peak shape of the characteristic peaks in the NMR spectrum are the same, the main purpose of optimizing the NMR conditions is to clearly present the characteristic signals of CPGs. Therefore, choosing one of them is representative.

[0034] Following the method in Example 1, NMR samples of Cistanche deserticola from Inner Mongolia were prepared, and the relaxation delay time, number of scans, acquisition time, and temperature were verified. The following experimental groups were set up.

[0035] Experiment 1: The number of scans was 64, the acquisition time was 2.5 s, and the temperature was 298.15 K. Relaxation delay times were set to 5, 10, 15, 20, and 30 s, and spectra were acquired and the content was calculated sequentially. The results showed that when d1 < 10 s, the peak shape was unstable, leading to an unstable integral area ratio and an underestimation of the content; when d1 ≥ 10 s, the content calculation results tended to be stable (RSD < 1%).

[0036] Experiment 2: The acquisition time was set to 2.5 s, the temperature to 298.15 K, and the relaxation delay time to 15 s. The number of scans was set to 16, 32, 64, and 128, respectively. Spectra were acquired sequentially, and the content was calculated. The results showed that when the number of scans was greater than 32, the signal-to-noise ratio was greater than 150, meeting the quantitative requirements.

[0037] Experiment 3: The temperature was set at 298.15 K, the relaxation delay time was 15 s, and the number of scans was 64. Acquisition times were set at 1.5, 2.5, and 3.5 s, and spectra were acquired and the content calculated sequentially. The results showed that the peak shape was stable within the acquisition time range of 1.5–3.5 s, and the signal-to-noise ratio was greater than 150, indicating that quantitative analysis was feasible. Considering all factors, 2.5 s was the optimal condition balancing accuracy and efficiency.

[0038] Experiment 4: The acquisition time was set to 2.5 s, the relaxation delay time to 15 s, and the number of scans to 64. Temperatures were set to 295, 298, and 303 K, and spectra were acquired and the content was calculated sequentially. The results showed that the peak shape of the sample was stable at different temperatures, and the calculated content was stable, with the sample content ranging from 1.56% to 1.57% and RSD < 1%.

[0039] The experimental results are shown in Table 1. According to the results, the RSD is less than 1%. Therefore, the relaxation delay time is set to 10-30 s, the temperature is 295-303 K, the number of scans is 32-128, and the acquisition time is within 1.5-3.5 s. The spectrum quality is good, the peak shape is relatively stable, the proton signal is independent, and the integration result is reliable, which can meet the requirements of quantitative analysis.

[0040] Experimental Example 2, Methodological Validation Accurately weigh 15.0 mg of echinacoside powder into a volumetric flask, dissolve and dilute to volume with deuterated methanol to prepare standard solutions of 15.0, 7.5, 3.75, 1.875, 0.9375, 0.46875, and 0.234 g∙L⁻¹. Accurately pipette 500 μL of each of the above standard solutions, add 50 μL of internal standard solution (5 mg / mL), mix well, and transfer to an NMR tube. Proceed according to the parameters of Example 1. 1 ¹H-NMR determination. The peak area ratio (A0) of the sample quantitative proton signal to the internal standard proton signal is plotted on the x-axis as the standard solution concentration (C, g·L⁻¹). sam / A is Using y=1.3871x + 0.1295 as the ordinate, a standard curve is plotted, yielding the regression equation: y = 1.3871x + 0.1295, with a correlation coefficient of 0.1295. R 2 = 0.9992, indicating a good linear relationship in the range of 0.2344-15.0 g・L⁻¹. Figure 14 ).according to LOD = 3.3 σ / S , LOQ =10 σ / SThe detection limit was calculated to be 0.3081 g·L⁻¹, and the quantitation limit was 0.9336 g·L⁻¹.

[0041] Experimental Example 3: Comparison of this method with the commonly used detection method HPLC *Cistanche deserticola* from Anhui and Inner Mongolia, and *Cistanche tubulosa* from Yili and Yutian in Xinjiang were collected. Samples were processed according to the method in Example 1, and CPG content was determined. Simultaneously, referring to the content determination method for *Cistanche deserticola* in the 2020 edition of the Chinese Pharmacopoeia, the total content of echinacoside and verbascoside in the same batch of samples was determined. The experimental results are shown in Table 2. The CPG content in *Cistanche deserticola* from different sources all met the pharmacopoeia requirements. 1 The concentration detected by ¹H-NMR was higher than that detected by HPLC. This is because Cistanche deserticola contains other CPGs besides echinacoside and verbascoside, therefore, ¹H-NMR was used to detect the higher concentrations. 1 ¹H-NMR can more comprehensively detect the content of CPGs in Cistanche deserticola.

[0042] Note: a Content is calculated as echinacoside; b The HPLC method, based on the 2020 edition of the Chinese Pharmacopoeia, only detects echinacoside and verbascoside.

Claims

1. A method based on 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that... Includes the following steps: S1. Sample pretreatment: The dried Cistanche deserticola was sliced, crushed, sieved, weighed accurately, and extracted with an organic solvent by ultrasonic extraction. After centrifugation, the supernatant was collected and concentrated to dryness to obtain the sample extract. S2. Preparation of internal standard solution: Accurately weigh 2,3,5-triiodobenzoic acid powder, add deuterated solvent, mix thoroughly, shake to dissolve, and then dilute to the mark to obtain an internal standard solution. Store at 4°C for later use. S3. NMR Sample Preparation: Add the deuterated solvent and the internal standard solution obtained in S2 to the sample extract obtained in S1, mix well, and obtain the NMR test sample; S4, Nuclear Magnetic Resonance Hydrogen Spectrum Acquisition: The NMR test sample prepared by S3 was placed in an NMR spectrometer using a superconducting NMR spectrometer with a MHz or higher. The pulse sequence was zg30, the relaxation delay time was 10-30 s, the temperature was 295-303 K, the number of scans was 32-128, and the acquisition time was 1.5-3.5 s. 1 H-NMR spectrum; S5. Automatic spectrum processing: The system automatically performs phase correction, solvent peak calibration, peak identification, characteristic peak identification, and integration on the NMR data obtained from S4. It can autonomously identify the reference peak as a d peak with a chemical shift between 7.40 and 7.49 and a proton number of 1, which is the characteristic proton signal of the internal standard, and normalize the peak area integral to 1.0000. It can also identify d peaks with a target integration range between 7.50 and 7.64 and a proton number of 1, which is the characteristic proton signal of caffeoyl phenylethanol glycosides, and automatically integrate them. S6. Data Processing and Content Calculation: The characteristic proton signals of the internal standard obtained from S5 are non-overlapping and independent with the characteristic proton signals of caffeoyl phenylethanol glycosides. Based on the integral area ratio of the characteristic proton signals of the internal standard and the characteristic proton signals of CPGs, and the amount of internal standard added, the percentage content of caffeoyl phenylethanol glycosides in the sample, calculated as echinacoside, is calculated.

2. The method based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The organic solvent mentioned in S1 is methanol or a 70% ethanol solution.

3. Based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The deuterated solvent mentioned in S2 or S3 is one or more of deuterated methanol, deuterated dimethyl sulfoxide, or deuterated pyridine.

4. The method based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The amount of deuterated solvent added in S3 is 500 μL.

5. The method based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The concentration of the internal standard solution mentioned in S2 is 5-10 mg / mL.

6. The method based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The amount of internal standard solution added in S3 is 50 μL.

7. The method based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The formula for calculating the percentage content of caffeoyl phenylethanol glycosides mentioned in S6 is as follows: Among them, A sam for δ H The integral area of ​​the characteristic proton signal in the sample at 7.50-7.64; A is The integral area of ​​the internal standard quantum signal; N is The number of protons in the characteristic signal of the internal standard; N sam The number of protons represented by the sample characteristic signal; M sam The relative molecular mass of echinacoside, the most abundant glycoside in Cistanche deserticola; M is The relative molecular mass of the internal standard; m is The mass of the added internal standard; m sam The mass of the sample taken; w is This represents the mass fraction of the internal standard.

8. The method based on claim 1 1 A method for quantitative determination of caffeoyl phenylethanol glycosides in Cistanche deserticola by ¹H-NMR, characterized in that: The Cistanche deserticola mentioned in S1 refers to either Cistanche tubulosa or Cistanche deserticola.

Images