Near infrared spectroscopy is used quickly to measure method and the application of five kinds of saponin contents in pseudo-ginseng
Technical field
The present invention relates to five kinds of saponin content rapid assay methods in a kind of pseudo-ginseng, relate in particular to one
Plant and use near infrared spectroscopy quickly to measure the method for five kinds of drug effect saponin contents in pseudo-ginseng, belong to the traditional Chinese medical science
Medicine studying technological domain.
Background technology
Pseudo-ginseng is dry root and the root of Araliaceae plant Panax notoginseng (Burk) F.H.Chen
Stem, has the hemostasis of the scattered stasis of blood, effect of detumescence ding-tong, is a kind of important common Chinese medicine, goes up in China
Century-old use history.Wherein, the saponin component of pseudo-ginseng is considered as its Chinese drugs, with pseudo-ginseng
Total saposins is the traditional medicine Injectio of raw material, as " Xuesaitong Injection " and thrombus lead to, and annual sales amount the nearlyest 10,000,000,000, be
" cookle " of traditional Chinese medicine circle.
Pseudo-ginseng monomer saponin composition is mostly dammarane type (dammarane) tetracyclic triterpene saponin(e, can be divided into two
Big class, is 20 (S)-protopanaxadiol-type [20 (S)-protopanaxadiol] and 20 (S)-Protopanaxatriol types respectively
[20(S)-protopanaxatriol].Wherein notoginsenoside R, ginsenoside Rg1, ginsenoside Re, ginseng soap
Glycosides Rb1 and ginsenoside Rd are saponin components main in pseudo-ginseng, are proved to be pseudo-ginseng and the anti-cardiac muscle of preparation thereof
The main pharmacodynamics material of ischemic, is the index components of arasaponin class quality of the pharmaceutical preparations control.
Quality of medicinal material analysis is the source of tcm manufacturing process analysis and quality control.It is all pseudo-ginseng, no
The same place of production, different harvest time etc. all can cause the change of quality of medicinal material.Existing Saponins from Panax notoginseng
Content assaying method is mainly high efficient liquid phase analysis method, and the method wastes time and energy, and needs reference substance.Cause
This, it is quite necessary to set up the content assaying method of main pharmacodynamics material in pseudo-ginseng, with meet pseudo-ginseng and
The demand that its quality of the pharmaceutical preparations controls.
In near infrared spectrum (Near Infrared Spectroscopy, NIRS) energy reflected sample, organic molecule contains
The characteristic information of hydrogen group (such as C-H, N-H, O-H etc.).Near-infrared spectrum technique can be used for traditional Chinese medicine quality
Analyze, sample without or only need simple process, do not consume chemical reagent, simple to operate, quick, at Chinese medicine
Field of quality control has huge application potential.Near-infrared spectrum technique is combined chemometric techniques,
The quick detection of main pharmacodynamics material in pseudo-ginseng, to pseudo-ginseng and arasaponin class preparation
Quality control is significant.
Summary of the invention
It is an object of the invention to provide a kind of employing near infrared spectroscopy and quickly measure five kinds of medicines in pseudo-ginseng
The method of effect saponin content and application, control to provide quickly for pseudo-ginseng and the arasaponin class quality of the pharmaceutical preparations,
Efficient technical method, alleviates inspection workload.
It is an object of the invention to be achieved through the following technical solutions:
Near infrared spectroscopy is used quickly to measure the method for five kinds of saponin contents in pseudo-ginseng, including following step
Rapid:
1) collect the pseudo-ginseng of different batches or pseudo-ginseng extract or notoginseng preparations, obtain calibration set sample
This;
2) gather the near infrared spectrum data of described calibration set sample, use single order to lead filter method and pre-process,
Obtain feature near infrared spectrum data;
3) notoginsenoside R, ginsenoside Rg1, ginseng in described calibration set sample is measured by HPLC method
The content of saponin(e Re, ginsenoside Rb1 and ginsenoside Rd;
4) use PLS establishment step 2) described in the feature near infrared spectrum number of calibration set sample
According to step 3) described in quantitative calibration models between five kinds of saponin contents;
5) according to step 2) identical method gathers the near infrared spectrum data of testing sample, and it is described fixed to import
Amount calibration model, obtains five kinds of saponin contents of testing sample.
For ease of describing, by notoginsenoside R, ginsenoside Rg1, ginsenoside Re, ginseng in the present invention
Saponin(e Rb1 and ginsenoside Rd are referred to as five kinds of saponin(es.
In the present invention, sample collection be by collect the pseudo-ginseng of different batches or pseudo-ginseng extract or
Notoginseng preparations sample, to increase the representativeness of sample set." batch " alleged by the present invention is commonly used in the art one
Individual technical term, specific to the present invention, the difference between the medicinal material of different " batch " is embodied in: the place of production is (the most raw
Long environment), position, main root specification, collecting time, period of storage, arrival entry time, specification etc..
In order to preferably set up quantitative calibration models, the place of production of the present invention includes Wenshan Prefecture, Honghe, elder brother
The places of production such as bright city, Yuxi;Described position includes the positions such as pseudo-ginseng clip, main root, rib, fibrous root;Institute
State main root specification and include at least 10.
In the present invention, it is preferable that before carrying out near infrared spectra collection, sample is pre-processed: by pseudo-ginseng
Medicinal material sample is vacuum dried, and crosses 280 μm sieves, obtain pseudo-ginseng powder after pulverizing.
In the present invention, it is preferable that spectra collection mode is: integrating sphere diffusing reflection is sampled;Preferably pick-up slip
Part: wave-number range 4,000-10,000cm-1, scanning times 64 times, resolution ratio 4cm-1, reference spectrum is
Air.
In the present invention, it is preferable that use single order to lead filtering and spectrum is pre-processed.
In the present invention, it is preferable that the near infrared spectrum data gathered (is being included calibration set sample, is testing
Card collection sample and sample to be tested) pre-process before also include reject spectral singularity sample;It is further preferred that
Chauvenet inspection is used to reject spectral singularity sample.
In the present invention, it is preferable that use partial least-squares regression method (PLSR) to set up described quantitative correction mould
Type.
In the present invention, it is preferable that described five kinds of saponin constituents all with 4,000-10,000cm-1All band is built
Mould.
The present invention is after carrying out preferably spectroscopic data pretreatment and modeling wave band respectively, the most square with cross validation
Root error RMSECV is that index determines optimal number of principal components.Preferably, notoginsenoside R, ginsenoside
Rg1, ginsenoside Re, ginsenoside Rb1, the number of principal components of ginsenoside Rd be respectively 7,7,
7,7,8.
Preferably, step 4 of the present invention) in, also include the described quantitative calibration models set up is pre-
Survey the step that performance is evaluated;Described evaluation index includes coefficient R, predicted root mean square error
RMSEP;Further, if R value is close to 1, RMSEP value is the least, the most described quantitative calibration model
It is applicable to the detection of described testing sample;Otherwise, the most inapplicable.
Specifically, the step bag that the estimated performance of the described described quantitative calibration models to being set up is evaluated
Including: take above-mentioned five kinds of saponin contents known checking collection sample, according to step 2) identical method gathers to be measured
The near infrared spectrum data of sample, imports described quantitative calibration models, is verified five kinds of saponin(es of collection sample
Content;Compared with five kinds of true measured values of saponin content of checking collection sample, evaluation and foreca performance.
The detection of described checking collection five kinds of saponin contents of sample can use art methods, it is possible to uses above-mentioned
HPLC method.
Preferably, HPLC method chromatographic condition of the present invention: Agilent Zorbax SB-C18 chromatographic column;Stream
Dynamic phase: mobile phase A is water, flowing is B acetonitrile mutually;Gradient elution: 0~22min, 17%~19%B;
22~30min, 19%~27%B;30~35min, 27%B;35~47min, 27%~46%B;47~70
Min, 46%~90%B.
Specifically, HPLC method chromatographic condition of the present invention: Agilent Zorbax SB-C18 chromatographic column
(4.6 × 50mm, 1.8 μm);Flowing phase: mobile phase A is water, flowing is B acetonitrile mutually;Gradient elution:
0~22min, 17%~19%B;22~30min, 19%~27%B;30~35min, 27%B;35~47min,
27%~46%B;47~70min, 46%~90%B;Detection wavelength 203nm;Sampling volume 3 μ L;Stream
Speed 0.8mL min-1;Column temperature 35 DEG C.
It is further preferred that the preparation of described HPLC method need testing solution includes: pseudo-ginseng shreds, and pulverizes,
Cross 280 μm sieves;Accurately weighed Notoginseng Root 0.5g, with 70% methanol-water (v/v) 40mL, ultrasonic (work
Frequency 53kHz) extract 1 hour;Filter, fully wash residue with 70% methyl alcohol, make to shift completely;60℃
Rotary evaporated to dryness;It is transferred in 5.0mL volumetric flask with 70% methyl alcohol, same solvent constant volume;Cross 0.22 μ
M nylon leaching film, to obtain final product.
Present invention additionally comprises above-mentioned employing near infrared spectroscopy and quickly measure five kinds of saponin contents in pseudo-ginseng
Method application in pseudo-ginseng, notoginseng preparations quality testing and control field.
Notoginseng preparations of the present invention includes XUESAITONG JIAONANG, Xuesaitong injection, XUESHUANTONG JIAONANG, thrombus
Logical injection etc..
Near-infrared spectral analysis technology is incorporated into five kinds of main saponins drug effects of pseudo-ginseng and preparation thereof by the present invention
In the detection of material, it is achieved that the quick mensuration to these five kinds of main pharmacodynamics materials, with traditional detection method
Compare, substantially reduce the testing time, save substantial amounts of manpower and materials.The present invention is conducive to improving pseudo-ginseng
Medicinal material and the quality of the pharmaceutical preparations thereof control level, can be in pseudo-ginseng detection and the production of arasaponin class preparation
Journey is promoted the use of.
The present invention is by the spectral band of five kinds of saponin(es in the near infrared spectrum of selection pseudo-ginseng sample, and extraction has
The characteristic spectrum wave band of effect, this feature spectral band has with the five kinds of saponin(es measured according to existing conventional method
Good correlation, can effectively monitor pseudo-ginseng and five kinds of saponin contents of preparation thereof.The present invention is by using
Spectral singularity sample is rejected in Chauvenet inspection, uses single order to lead filtering and pre-processes spectrum, with screening
Information, reduces noise, improves accuracy and the accuracy of the method detection.Five kinds of medicines involved in the present invention
Effect saponin(e is the main pharmacodynamics composition of arasaponin class preparation, and the Quality Control for arasaponin class preparation refers to
Mark thus can be that pseudo-ginseng and the quality of the pharmaceutical preparations thereof control to provide technical method fast and efficiently, alleviate inspection man
Measure.
Accompanying drawing explanation
Fig. 1 is that in experimental example 1, the near-infrared of pseudo-ginseng powder diffuses spectrogram;
The number of principal components of ginsenoside Rd's quantitative model and RMSECV graph of a relation in Fig. 2 experimental example 1;
Fig. 3 is five kinds of saponin(e model predication value figures relevant to reference value in experimental example 1.
Detailed description of the invention
Following example are used for illustrating the present invention, but are not limited to the scope of the present invention.Embodiment is not noted
Bright concrete technology or condition person, according to the technology described by the document in this area or condition, or according to product
Product specification is carried out.Agents useful for same or instrument unreceipted production firm person, be and can be purchased by regular channel
Available conventional products.
If no special instructions, following experimental example and embodiment instrument and reagent:
Antaris near infrared spectrometer (Thermo company, the U.S.), is equipped with integrating sphere diffusing reflection sampling apparatus
With RESULT 3.0 spectra collection software, the near infrared spectrum in collecting sample.
Agilent 1100 Series HPLC System (Agilent company, the U.S.), be furnished with quaternary solvent pump, from
Dynamic injector, column oven, variable wavelength UV-detector.Notoginsenoside R, ginsenoside Rg1, people
Ginseng saponin(e Re, ginsenoside Rb1 and five kinds of saponin(e standard items of ginsenoside Rd (upper Hiroad standing grain medical sci-tech
Development Co., Ltd, purity >=98%).
Liquid phase water is ultra-pure water (Millipore), chromatogram acetonitrile (Merck, Germany), and other reagent is for dividing
Analyse pure.
Experimental example 1
1 sample collection: take 173 groups of pseudo-ginseng different parts and (include pseudo-ginseng clip, main root, rib, fibrous root
At position), different sources (including the places of production such as Wenshan Prefecture, Honghe, Kunming, Yuxi), different big
The pseudo-ginseng sample of small dimension, vacuum drying, cross 280 μm sieves after pulverizing.
2 methods
2.1 measure five kinds of drug effect saponin contents in sample by HPLC method
2.1.1 prepared by need testing solution: accurately weighed Notoginseng Root 0.5g, with 70% methanol-water (v/v) 40mL,
Ultrasonic (operating frequency 53kHz) extracts 1 hour.Filter, fully wash residue with 70% methyl alcohol, made
Total transfer.Rotary evaporated to dryness at 60 DEG C.It is transferred in 5.0mL volumetric flask with 70% methyl alcohol, identical molten
Agent constant volume.Cross 0.22 μm nylon leaching film, to obtain final product.
2.1.2 chromatographic condition: Agilent Zorbax SB-C18 chromatographic column (4.6 × 50mm, 1.8 μm);Stream
Dynamic phase water (A)-acetonitrile (B), gradient elution (0~22min, 17%~19%B;22~30min, 19%~27%
B;30~35min, 27%B;35~47min, 27%~46%B;47~70min, 46%~90%B);
Detection wavelength 203nm;Sampling volume 3 μ L;Flow velocity 0.8mL min-1;Column temperature 35 DEG C.Bent according to standard
Five kinds of drug effect saponin contents in line computation pseudo-ginseng sample.
2.2 samples and near infrared spectra collection
Spectra collection condition: wave-number range 4,000-10,000cm-1, scanning times 64 times, resolution ratio 4cm-1,
Reference spectrum is air.The near-infrared of pseudo-ginseng powder diffuses spectrogram as shown in Figure 1.
3 results and discussion
3.1 near infrared spectrum data preprocess methods
Spectral singularity sample, then random division calibration set 152 example and checking is rejected according to Chauvenet inspection
Collect 16 examples.Using the modeling of all band PLSR, lead filtering with single order and pre-process spectrum, each saponin(e is all
Model for all band.Notoginsenoside R, ginsenoside Rg1, ginsenoside Re, ginsenoside Rb1 and people
The number of principal components of ginseng saponin(e Rd is respectively 7,7,7,7 and 8.
The method for building up of 3.2 quantitative calibration models
In order to remove the noise near infrared spectrum and interference.Use the modeling of all band PLSR, be respectively compared
Become without Pretreated spectra, first derivative, second dervative, multiplicative scatter correction (MSC) and standard canonical
Change (SNV) these five kinds of preprocessing procedures.Result is as shown in table 1-table 5, and result display single order leads filter
Ripple is preferable to five kinds of saponin(e quantitative results.Wherein RMSEC represents calibration set root-mean-square error;RMSEP table
Show forecast set root-mean-square error.
Table 1 R1 Pretreated spectra results contrast
Table 2 Rg1 Pretreated spectra results contrast
Table 3 Re Pretreated spectra results contrast
Table 4 Rb1 Pretreated spectra results contrast
Table 5 Rd Pretreated spectra results contrast
The selection of 3.3 spectral bands
Although PLSR method can to a certain degree eliminate the synteny of spectrum, still it is necessary modeling wave band is entered
Row optimizes.By comparing the modeling effect of different-waveband, the results are shown in Table 6-table 10.Five kinds of saponin(es are all entirely
Wave band modeling is more excellent.
Table 6 R1 wave band modeling effectiveness comparison
Table 7 Rg1 wave band modeling effectiveness comparison
Table 8 Re wave band modeling effectiveness comparison
Table 9 Rb1 wave band modeling effectiveness comparison
Table 10 Rd wave band modeling effectiveness comparison
The determination of 3.4 optimal number of principal components
Judge the number of principal components of PLSR according to cross validation root-mean-square error RMSECV, Fig. 2 is ginseng soap
The number of principal components of glycosides Rd quantitative model and cross validation root mean square residual (RMSECV) graph of a relation, when main one-tenth
When mark is 8, RMSECV tends towards stability.In like manner, the number of principal components of other saponin constituents is respectively as follows: R1,
7;Rg1,7;Re, 7;Rb1,7.
3.5 model prediction results
PLSR method is used to set up quantitative calibration models, with prediction related coefficient R and predicted root mean square error
RMSEP is metrics evaluation model prediction performance.
Use partial least-squares regression method (PLSR) by the NIR spectra of 152 example calibration set samples and HPLC
Five kinds of saponin contents that method records carry out returning association, set up the optimum near-infrared quantitative correction mould of five kinds of saponin(es
Type.By the near infrared spectrum input quantitative calibration models of remaining 16 example checking collection sample, calculate five kinds of saponin(es
Content, and compare with HPLC method measured value, with predicted root mean square error RMSEP for metrics evaluation mould
Type estimated performance, the RMSEP of notoginsenoside R, ginsenosides Rg1, Re and Rb1 and Rd is respectively
2.50,3.73,1.41,3.10 and 1.00.Relevant figure between predicted value and the reference value of each regression model
As shown in Figure 3.In Fig. 3, R1, Rg1, Re, Rb1, Rd represent notoginsenoside R, ginseng respectively
Saponin(e Rg1, ginsenoside Re, ginsenoside Rb1 and ginsenoside Rd;Calibration is calibration set,
Validation is checking collection.
Notoginsenoside R, ginsenosides Rg1, Re and Rb1 and the Rd prediction phase relation in calibration model
Number (R) is respectively 0.9463,0.9718,0.9347,0.9596 and 0.9892.
The application of 3.6 Quantitative Analysis Model
Under the conditions of selected Pretreated spectra and wave band, in pseudo-ginseng five kinds of drug effect saponin contents are carried out
Prediction.This pseudo-ginseng diffusing reflection near-infrared quantitative model, has higher precision of prediction, and five kinds of saponin(es become
The prediction related coefficient dividing (notoginsenoside R, ginsenosides Rg1, Re and Rb1 and Rd) is all higher than 0.93.
And Testing index is consistent with the quality control index of arasaponin class preparation, adds the practicality of method.Should
Method is suitable for by arasaponin class preparation production line feeding intake fast and accurately the content information of medicinal material, for system
The production of agent provides reference.Meanwhile, the method can be used for the quality testing of pseudo-ginseng acquisition phase.
Although, the present invention is described in detail the most with a general description of the specific embodiments,
But on the basis of the present invention, can make some modifications or improvements it, this is to those skilled in the art
Obviously.Therefore, these modifications or improvements without departing from theon the basis of the spirit of the present invention, all
Belong to the scope of protection of present invention.