Citrus pericarp identification based on two-dimensional infrared correlation spectroscopy in combination with chemometrics

Abstract

Rutaceae citrus plants are widely planted, peels of the rutaceae citrus plants have similar chemical components and medicinal characteristics, and adulteration or abuse is easily caused. The citrus peel is identified on the basis of two-dimensional infrared correlation spectroscopy in combination with chemometrics. And drawing the two-dimensional infrared correlation spectrum into a 3D graph, an automatic peak and a cross peak. According to the peak shapes of the 3D image, the peaks can be divided into three types: step peaks, wide and strong peaks and penholder peaks. And the number and the position of each sample automatic peak are different. In addition, the cross peaks of the synchronous spectrums of the samples also have differences. And meanwhile, the similarity degree and the genetic relationship degree between citrus peels are explored in combination with clustering analysis. The method provides reference for better utilization of medicinal values of citrus and other medicinal materials, has certain universality, and can be popularized to other plant identification.

Description

Technical Field

[0001] The invention belongs to the technical field of identification of traditional Chinese medicines, and in particular relates to a method for identifying citrus peels based on two-dimensional infrared correlation spectroscopy combined with chemometrics. Background Art

[0002] Citrus plants in the Rutaceae family are one of the most widely cultivated crops in the world. As they belong to the same family and genus, their chemical composition and medicinal effects are very similar. Fluctuations in market prices can easily lead to adulteration or substitution. However, these medicinal materials have different focuses. Comparative studies on new technologies and methods for peels of different varieties can provide a scientific basis for market supervision and ensure clinical drug safety. This is of great significance for maintaining the market order of Chinese medicinal materials and protecting the health of patients. Therefore, it is necessary to explore an effective method that can easily and quickly identify and distinguish the peels of Citrus fruits in the Rutaceae family.

[0003] The present invention selects 8 commonly used clinical medicines, namely, dried tangerine peel, dried tangerine peel, green peel, citron, citron, citron, red tangerine peel, fructus aurantii and immature tangerine peel, from plants of the genus Citrus in the family Rutaceae to carry out relevant research, and defines the 8 samples to be tested as S1 to S8. Among them, tangerine peel, tangerine peel and green peel are fruits or peels of different maturity of the citrus plant (Citrus reticulata Blanco) and its cultivated varieties of the Rutaceae family; citron is the dried mature fruit of the citron (Citrus medica L.) or the fragrant round (Citrus wilsonii Tanaka) of the Rutaceae family; bergamot is the dried fruit of the bergamot (Citrus medica L.var.sarcodactylis Swingle) of the Rutaceae family; tangerine peel is the immature or nearly mature dried outer peel of the Huazhou pomelo (Citrus grandis'Tomentosa') or the pomelo (Citrusgrandis (L.) Osbeck) of the Rutaceae family; and bitter orange peel is the dried immature fruit of the sour orange (Citrusaurantium L.) and its cultivated varieties of the Rutaceae family. Citrus aurantium is the dried young fruit of the Rutaceae plant sour orange (Citrusaurantium L.) and its cultivated varieties or sweet orange (Citrus sinensis Osbeck), which is cut into two halves and dried in the sun or at a low temperature. The smaller ones are directly dried in the sun or at a low temperature.

[0004] In recent years, people have conducted extensive and in-depth research on the common citrus peel in the Rutaceae family. The research methods cover multiple aspects such as morphological feature observation, physical and chemical analysis technology, and molecular biology technology, providing a scientific basis for the accurate and reliable identification of citrus peel. Morphological feature observation is a traditional and intuitive identification method. By observing the color, shape, size and other characteristics of the fruit, the citrus peel is preliminarily classified. However, there are few morphological markers in the citrus peel, the differences between species are not obvious, and the identification of morphological features is easily affected by people's subjective judgment and experience. With the continuous advancement of physical and chemical testing technology, more and more high-precision instruments provide more accurate means for the accurate identification of citrus peel. For example, high-performance liquid chromatography (HPLC) technology can be used to analyze characteristic components such as hesperidin and naringin in citrus peel. By analyzing the content and proportion of characteristic components in each variety, its type and variety can be further confirmed. However, most of these equipment are expensive or have complicated preparation processes, and the operating and maintenance costs are also high. The detection process usually involves the extraction, separation and concentration of raw materials. When dealing with biomacromolecules such as proteins and polysaccharides, it is necessary to perform derivatization to improve chromatographic performance, which undoubtedly makes the experiment more lengthy and complicated. Molecular biology technology provides a more advanced method for the genetic diversity analysis and variety identification of citrus peel. Although molecular biology techniques such as DNA barcoding and PCR amplification have shown high accuracy in the identification of citrus peel, due to its operational complexity, it often requires operators to have high professional skills and experience. At the same time, the high cost of experimental equipment and reagents required for this technology limits its wide application. Two-dimensional infrared correlation spectroscopy is a rapid and non-destructive method for the identification of medicinal materials. In addition, it has high sensitivity and high accuracy, and can detect small changes in chemical composition and accurately identify complex herbal mixtures. Summary of the invention

[0005] In order to solve the problem that the genetic backgrounds of common Rutaceae Citrus plant varieties are highly similar and difficult to distinguish and identify, the present invention provides a method for identifying common Rutaceae Citrus peel based on two-dimensional infrared correlation spectroscopy combined with chemometrics, the purpose of which is to use Fourier transform infrared spectroscopy and two-dimensional infrared correlation spectroscopy technology to achieve non-destructive, simple and fast differential analysis of common Rutaceae Citrus peel without the need for a separation process.

[0006] In order to achieve the above-mentioned purpose of the present invention, the following technical scheme is adopted:

[0007] (1) Sample preparation: All samples were crushed using a pulverizer, passed through a 120-mesh pharmacopoeia sieve, dried in an oven at 40±5°C to constant weight, and sealed for later use.

[0008] (2) Collection of two-dimensional infrared spectra: The sample powder and KBr powder were accurately weighed at a ratio of 1:150, mixed and ground in an agate mortar, and pressed into tablets. Air was used as the background, and the interference of water and carbon dioxide was subtracted. The scanning range was set to 4000-400 cm -1 , resolution 4cm -1 Each spectrum was scanned 8 times cumulatively, the temperature rise range was 40-150°C, the heating rate of the infrared temperature variable accessory was 2°C / min, and sampling was performed every 5°C.

[0009] (3) Data processing of two-dimensional infrared spectra: The raw data in (2) were smoothed at 13 points and baseline corrected using OMNIC 9.2 software and then imported into 2DCS software to obtain synchronous correlation spectra and automatic peak data. The 3D graph and automatic peak map were then obtained using Origin 2021 software.

[0010] (4) Data processing for cluster analysis: IBM SPSS Statistics 26 was used to perform a systematic cluster analysis on the automatic peak values ​​of the two-dimensional infrared correlation spectra of S1 to S8. The inter-group linkage method was selected as the clustering method, and the Euclidean distance was selected as the interval. The automatic peak values ​​of the two-dimensional infrared correlation spectra of S1 to S8 were used to generate an average inter-group dendrogram.

[0011] The present invention adopts advanced spectral technology, two-dimensional infrared correlation spectrum combined with cluster analysis method, to conduct direct or mutual comparative analysis on 8 clinical commonly used drugs in the peel of citrus of Rutaceae family, which greatly simplifies the analysis process and eliminates tedious purification or separation steps, making the analysis process more efficient and convenient. At the same time, this technology can quickly and comprehensively obtain the overall information of traditional Chinese medicine, realize non-destructive differential analysis, and provide new ideas and methods for the research and application of traditional Chinese medicine.

[0012] Beneficial effects of the present invention:

[0013] 1. The samples of the present invention are pre-treated by low-temperature drying, which reduces the influence of moisture on the chemical components in the peel of citrus of the Rutaceae family without damaging the effective ingredients.

[0014] 2. The present invention sets the heating speed of the infrared temperature variable accessory to 2°C / min, performs sampling every 5°C, and the program heating speed is slow, so more detailed information can be obtained.

[0015] 3. The present invention uses a 3D graph of two-dimensional infrared correlation spectroscopy to classify samples into three categories, and then the peels of eight species of citrus in the Rutaceae family can be easily and quickly identified based on the number, position and cross-peak position of the automatic peaks and their positive and negative combinations.

[0016] 4. The present invention combines cluster analysis with changes in Euclidean distances of 1 to 11 to explore the similarities and genetic relationships between the peels of various citrus species. At the same time, two-dimensional infrared correlation spectra can be directly identified based on the number and position of the graph and the use of automatic peaks, which is simpler and faster. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 S1~S8(1700~800cm -1 )'s 3D spectrum of two-dimensional infrared correlation spectra Figure 2 S1~S8(1700~800cm -1 ) of the two-dimensional infrared correlation spectrum (A) and synchronous spectrum (B)

[0018] Figure 3 Cluster analysis graph of S1 to S8 DETAILED DESCRIPTION

[0019] The present invention is described in detail below with reference to diagrams and specific implementation operations.

[0020] The following experiment uses 8 commonly used clinical drugs from the genus Citrus in the family Rutaceae, including dried tangerine peel, tangerine peel, green peel, citron, citron, citron, citrus fruit, tangerine peel, fructus aurantii and immature tangerine peel, recorded as S1, S2, S3, S4, S5, S6, S7 and S8. The instruments used are Fourier transform infrared spectrometer (FTIR-650) and infrared temperature variable accessory (GS20730).

[0021] Embodiment 1

[0022] This embodiment provides a method for identifying citrus peel based on two-dimensional infrared spectroscopy, and the operating steps are as follows:

[0023] All samples were crushed with a pulverizer, passed through a 120-mesh pharmacopoeia sieve, and dried in an oven at 40±5°C to constant weight. The sample powder and spectrally pure KBr powder were accurately weighed at a ratio of 1:150, mixed and ground in an agate mortar, and pressed into tablets. With air as the background, the interference of water and carbon dioxide was deducted, and the scanning range was set to 4000-400cm -1 , resolution 4cm -1 Each spectrum was scanned 8 times cumulatively, the temperature rise range was 40-150°C, the temperature rise rate of the infrared temperature variable accessory was 2°C / min, and sampling was performed every 5°C. The obtained data were smoothed and baseline corrected by OMNIC 9.2 software at 13 points, and then imported into the 2DCS software independently developed by Sichuan University to obtain synchronous correlation spectra and automatic peak data, and then the 3D map and automatic peak spectrum were obtained by Origin2021 software.

[0024] According to the peak shape of the 3D graph, Figure 1 As shown. The eight samples can be divided into three categories: step peaks, broad and strong peaks, and pen holder peaks. Among them, the peaks of S1 to S2 increase in a step-like manner, with a narrow peak width, and can be classified as step peaks. The 3D graph shows that the S1 cross peak at (1580,1000) is weaker than the S2 cross peak. The peaks of S3 to S7 are broad and strong, showing broad and strong peaks. S8 is thin, high, numerous and dense, and belongs to the pen holder peak type.

[0025] The automatic peaks represent the responsiveness of the sample chemical composition to temperature changes. -1 Automatic peaks within the range are further analyzed, such as Figure 2 (A) is shown. Table 1 shows the distribution of automatic peaks in eight samples. The automatic peak counts of samples S1 to S8 are as follows: S1 is 11, S2 is 17, S3 is 15, S4 is 17, S5 is 20, S6 is 17, S7 is 16, and S8 is 22. According to 1674cm -1 and 1193cm -1 The presence or absence of the automatic peak can be used to classify samples into three categories. The first category is S1 to S3 samples, which have a peak at 1674 cm -1 and 1193cm -1 The second type is S4 and S5, which are different from other samples in that the peak at 1674 cm -1 The third category is S6~S8 samples, which have an automatic peak at 1193cm -1 An automatic peak appeared at

[0026] In the first category, S1 can be detected by 1371, 1155, 1113, 980 and 847 cm -1 S2 can be identified by the automatic peaks at 1510, 1063 and 885 cm -1 Only S3 at 1531cm -1 and 879cm -1 In the second category, S5 has peaks at 1589, 1248, 1115, and 1024 cm -1 In the third category, according to S8, there are four more automatic peaks at 1633, 1529, 1444, 1365, 1338, 1255, 1138, 1076, 1030, 895 and 826 cm -1 The automatic peak at 1601 cm-1 can quickly identify S8. Comparison of S6 and S7 revealed that S6 has a peak at 1601 cm-1. -1 There is an automatic peak at 1700-800cm, while S7 does not, so the two can be distinguished based on this feature. By analyzing the number and position of automatic peaks, the eight samples can be distinguished and the -1The number of cross peaks and strong automatic peaks within the range was analyzed in more detail for the eight samples.

[0027] Table 1 Rutaceae Citrus fruit peel 1700~600cm -1 Automatic peak parameters for bands

[0028]

[0029] S1 at 1655, 1606, 1466 and 980 cm -1 There are four strong automatic peaks at Figure 2 (B) As shown. There are five groups of negative cross peaks at (848, 1601), (881, 1608), (1209, 1467), (1209, 1602) and (1209, 1658), indicating that there is a strong synergistic effect between them. The spectral intensity of each group of negative cross peaks increases while the other decreases; 848, 881 and 1209 cm -1 The spectral intensity changes at 1601, 1608, 1467, 1602 and 1658 cm -1 The synchronous spectra of samples S2, S3, S6 and S7 have certain similarities as a whole. These four samples have two positive cross peaks at (1602, 1471) and (1658, 1598), and the rest at 1602 cm -1 and 1658cm -1 The cross peaks at 1471 cm -1 The cross peaks at all locations are negative. Among them, S2 has five strong automatic peaks, which are 1599, 1219, 1200, 997 and 941 cm -1 S3 has 1653cm -1 and 1603cm -1 The strong automatic peaks of S6 are located at 1464, 1217, 1198, 1155, 998 and 943 cm -1 The strong automatic peaks of S7 are located at 1655, 1589, 1398, 1219, 1196 and 943 cm -1 S4 has 1597, 1466, 1213 and 1041cm -1 There are 4 strong automatic peaks at (848, 1469), (848, 1591), (891, 1471), (891, 1592), (939, 1471) and (939, 1591), and there are 6 groups of negative cross peaks at (848, 1469), (848, 1591), (891, 1471), (891, 1592), (939, 1471) and (939, 1591), and the rest of the cross peaks are positive. The cross peaks of S5 and S8 are all positive cross peaks, indicating that the spectral intensity increases and decreases at the same time. However, the number of cross peaks of S8 is much higher than that of S5. In addition, S5 has 1589, 1471, 1211 and 943 cm -1There are 4 strong automatic peaks at 1655, 1603, 1529 and 1444 cm -1 The synchronous spectrum can clearly observe the similarities and differences between each sample. The experimental results show that according to the 3D graph of the sample's two-dimensional infrared correlation spectrum, the automatic peak and cross peak can distinguish 8 citrus medicinal materials, providing a simple and effective method for the identification of easily confused citrus medicinal materials.

[0030] Embodiment 2

[0031] This embodiment provides a method for analyzing the two-dimensional infrared correlation spectrum of citrus peel based on a cluster analysis method, and the operation steps are as follows:

[0032] Cluster analysis is a chemometric method for studying “birds of a feather flock together”. The entire clustering process can be intuitively represented by a dendrogram. Figure 3 As shown in the figure, the 4000~900cm -1 The band range automatic peak data is examined and compared. As the Euclidean distance increases, all samples tend to belong to the same category.

[0033] From the pedigree diagram, we can see that when the Euclidean distance is 1, samples S4 and S5 are grouped into one category, indicating that they are closely related. Sample S5 belongs to a special type or variant of S4 in plant classification. Samples S6 and S7 are grouped into one category when the Euclidean distance is 5, indicating that sample S6 may be very similar to S7 in terms of chemical composition and content. When the Euclidean distance is 11, samples S1, S2 and S3 are grouped into one category, where samples S1 and S3 belong to different medicinal parts of plants of the same family (S1 is the peel of mature fruit, and S3 is the peel of immature fruit), with similar biological characteristics and genetic background. Sample S2, as a local medicinal material, belongs to a special type of S1, specifically the peel of the cultivated variety of orange, the tea-branch orange. There may be some differences in the content and type of active ingredients between S2 and S1, so S1 and S3 are more closely related. From the cluster analysis dendrogram, it can be seen that the pattern recognition method of cluster analysis can quickly and accurately distinguish different varieties of citrus peels from each other, and intuitively reflects the similarity between the spectra of each peel. The pattern recognition method of two-dimensional infrared correlation spectroscopy cluster analysis can be used as a basis for judging the affinity of citrus peels from the Rutaceae family.

[0034] This study used two-dimensional infrared correlation spectroscopy combined with chemometrics to identify citrus peels. Due to the great influence of temperature on the chemical composition of samples, the 3D graph of two-dimensional infrared correlation spectroscopy, the number and position of automatic peaks, and the position and positive and negative of cross peaks can distinguish different samples more easily and more effectively. Cluster analysis based on the automatic peak data of two-dimensional infrared correlation spectroscopy obtained an intuitive and clear dendrogram, which can quickly show the closeness of the relationship between different varieties of Rutaceae citrus peels, and intuitively reflects the similarity between the spectra of each peel. At the same time, the 3D graph, automatic peaks and cross peaks of two-dimensional infrared spectra are intuitive and clear, easy to analyze, and have better identification effects, which are worth recommending. This method will provide a reference for better utilization of the medicinal value of citrus medicinal materials and other plants.

[0035] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and the inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. Identification of citrus peel based on two-dimensional infrared correlation spectroscopy combined with chemometrics, characterized by: The two-dimensional raw data of the sample is collected by Fourier infrared spectrometer and variable temperature accessory, and then the two-dimensional infrared correlation spectrum of the sample is constructed by software and correlation analysis is performed.

2. The use according to claim 1, characterized in that The sample to be tested was crushed, passed through a 120-mesh pharmacopoeia sieve, dried at 40±5℃ to constant weight, and sealed for later use.

3. The use according to claim 1, characterized in that With air as the background, the interference of water and carbon dioxide was subtracted, and the scanning range was set to 4000~400cm -1 , resolution 4cm -1 Each spectrum was scanned 8 times cumulatively, the temperature rise range was 40-150°C, the heating rate of the infrared temperature variable accessory was 2°C / min, and sampling was performed every 5°C.

4. The use according to claim 1, characterized in that The obtained data were smoothed and baseline corrected at 13 points using OMNIC 9.2 software and then imported into 2DCS software to obtain synchronized spectra and automatic peak data. The relevant 3D spectra and automatic peak spectra were obtained using Origin 2021 software with a wavelength of 4000–400 cm -1 The band contains all the effective information on the changes in chemical composition in the peel of citrus plants of the Rutaceae family under temperature disturbance.

5. The use according to claim 1, characterized in that According to the peak shape of the 3D graph, the eight samples can be divided into three categories: step peak, broad and strong peak, and pen holder peak. Among them, the peak shape of S1 to S2 increases in a step-like manner with a narrow peak width, and can be classified as a step peak; the peaks of S3 to S7 are wide and strong, showing a broad and strong peak; S8 is thin, high, numerous and dense, and belongs to the pen holder peak type. The 3D graph can intuitively see the differences in the peak shapes of each sample.

6. The use according to claim 1, characterized in that Samples at 1700~800cm -1 The synchronized spectra within the range showed obvious differences in the number, position and cross-peaks of automatic peaks. The samples can be divided into three categories through the 3D graph of the two-dimensional infrared correlation spectrum. The number, position and cross-peak positions of automatic peaks combined with their positive and negative values ​​can be used to intuitively and accurately perform differential analysis on the peel of citrus plants in the Rutaceae family.

7. The use according to claim 1, characterized in that For 4000~400cm -1 Cluster analysis of the automatic peak data of two-dimensional infrared correlation spectra within the band range showed that when the Euclidean distance was 1, S4 and S5 were classified into one class; when the Euclidean distance was 5, S6 and S7 were clustered into one class; when the Euclidean distance was 11, S1, S2 and S3 were classified into one class. The pattern recognition method of cluster analysis can intuitively display the degree of genetic relationship between samples; the pattern recognition method of cluster analysis of two-dimensional infrared correlation spectra can be used as a basis for determining the genetic relationship of citrus peel in the Rutaceae family.

Images