Method for identifying adulteration of exogenous plant protein in pure dairy products and detection system

Abstract

The application discloses a method and a system for identifying exogenous plant protein adulteration in pure dairy products, which comprises the following steps: simultaneously extracting milk protein and lactose from the pure dairy products to be tested; using an elemental analyzer-stable isotope ratio mass spectrometer combined system to determine the carbon stable isotope ratio of the milk protein and the lactose; calculating the carbon isotope characteristic difference; making a qualitative determination of adulteration based on the statistical threshold established for pure cow milk; and quantitatively calculating the adulteration proportion of the exogenous plant protein by using a linear regression model constructed in advance. The system comprises a sample pretreatment module, an isotope detection module, a data processing module, a determination module and a quantitative calculation module. The application uses isoelectric point precipitation combined with organic solvent extraction to extract milk protein, and uses ethanol low-temperature recrystallization to purify lactose, and introduces lactose as an endogenous isotope internal standard, which effectively offsets the system error, can quickly determine the adulteration behavior and accurately calculate the adulteration proportion of the exogenous plant protein.

Description

Technical Field

[0001] This invention relates to the field of dairy product testing technology, and in particular to a method and system for identifying adulteration with exogenous plant proteins in pure dairy products. Background Technology

[0002] Pure dairy products are characterized by bovine casein and whey protein, with lactose as an endogenous characteristic sugar. Both originate from bovine metabolism and have a highly consistent background in carbon isotopes (δ¹³C). To reduce production costs, unscrupulous merchants add cheap plant proteins such as rice, wheat, soybeans, peas, and hydrolyzed wheat protein to pure milk, diluting the bovine milk protein and disrupting the dairy market. Furthermore, some individuals may have allergies to plant proteins, posing a food safety risk.

[0003] Currently, the traditional methods for detecting adulteration of dairy products mainly include ELISA and liquid chromatography-mass spectrometry. When using ELISA, the detection results are easily affected by protein hydrolysis modifications, and hydrolyzed plant proteins are prone to false negatives. When using liquid chromatography, there are problems such as complex procedures and long detection cycles. Summary of the Invention

[0004] The main objective of this invention is to provide a method for identifying adulteration with exogenous plant proteins in pure dairy products, aiming to solve the technical problems of inaccurate detection results, complex procedures, and long detection cycles in traditional detection methods.

[0005] To achieve the above objectives, this invention provides a method for identifying adulteration of exogenous plant proteins in pure dairy products. The method utilizes an elemental analyzer-stable isotope ratio mass spectrometry to determine the carbon stable isotope ratios of milk proteins and lactose, calculates the carbon isotope characteristic differences, performs a hypothetical adulteration determination based on a statistical threshold established using pure cow's milk, and quantitatively calculates the adulteration ratio of exogenous plant proteins using a pre-constructed linear regression model. The carbon isotope characteristic difference is the difference between the carbon stable isotope ratio of milk protein and the carbon stable isotope ratio of lactose. The carbon isotope characteristic difference constitutes a characterization index of the endogenous metabolic isotope coupling relationship of pure milk.

[0006] Optionally, the carbon isotope characteristic difference is obtained through the following steps: The pH of the pure dairy product sample to be tested was adjusted to 4.6 and centrifuged to obtain casein precipitate and supernatant. The casein precipitate was extracted and centrifuged using an ethanol-ether mixture, and then freeze-dried to obtain milk protein powder. Anhydrous ethanol was added to the supernatant and allowed to stand at 4°C to obtain lactose crystals. The lactose crystals were washed with ethanol and dried to obtain lactose powder. The milk protein powder and the lactose powder were placed separately in an elemental analyzer-stable isotope ratio mass spectrometry system and burned at 980°C-1000°C to convert them into carbon dioxide gas. The carbon stable isotope ratios of the milk protein and the lactose were measured, and the carbon isotope characteristic difference was calculated.

[0007] Optionally, the statistical threshold is obtained through the following steps: Collect no fewer than 10 pure milk samples, measure the carbon isotope characteristic difference of each sample, calculate the arithmetic mean and standard deviation of the carbon isotope characteristic difference, and use the arithmetic mean plus twice the standard deviation as the judgment threshold.

[0008] Optionally, the determination of the doping hypothesis includes the following steps: The carbon isotope characteristic difference is compared with the statistical threshold. When the carbon isotope characteristic difference of the sample to be tested is greater than the statistical threshold, it is determined that exogenous plant protein is detected; when the carbon isotope characteristic difference of the sample to be tested is less than or equal to the statistical threshold, it is determined that exogenous plant protein is not detected.

[0009] Optionally, the linear regression model is a univariate linear regression model, which uses the mass ratio of the incorporated exogenous plant protein in the pure dairy product to be tested as the independent variable and the carbon isotope characteristic difference as the dependent variable, and is obtained by fitting using the least squares method. The quantitative calculation includes substituting the carbon isotope characteristic difference of the pure dairy product to be tested as the dependent variable into the linear regression model, and solving the independent variable in reverse to obtain the adulteration ratio of exogenous plant protein.

[0010] Optionally, the coefficient of determination R² of the linear regression model is greater than or equal to 0.99, and the linear regression model is independently established by running gradient-spike samples for a specific exogenous plant protein or a specific detection batch.

[0011] Optionally, the exogenous plant protein is selected from one or more of rice protein, wheat protein, soybean protein, peanut protein, pea protein, and hydrolyzed wheat protein. Furthermore, to achieve the above objectives, the present invention also provides a detection system for exogenous adulterants in dairy products, used to implement the above-mentioned method for detecting exogenous adulterants in dairy products, comprising: The sample pretreatment module is used to simultaneously extract milk protein components and lactose components from the dairy product sample to be tested, wherein the milk protein components include casein and whey protein; The isotope determination module includes an elemental analyzer-stable isotope ratio mass spectrometry system, used to determine the carbon stable isotope ratio of the milk protein component and the carbon stable isotope ratio of the lactose component, respectively. The data processing module is used to calculate the carbon isotope characteristic differences; The judgment and analysis module is used to compare the carbon isotope feature difference with a preset statistical judgment threshold. When the carbon isotope feature difference is greater than the statistical judgment threshold, it is determined that there is an exogenous adulterant in the dairy product to be tested; when the carbon isotope feature difference is less than or equal to the statistical judgment threshold, it is determined that there is no exogenous adulterant in the dairy product to be tested. The quantitative calculation module is used to calculate the adulteration ratio of the exogenous adulterant by substituting the carbon isotope characteristic difference into a pre-constructed linear regression model when the presence of exogenous adulterant is determined. The statistical judgment threshold is determined based on the statistical distribution of carbon isotope characteristic difference in a pure dairy product sample set, and the linear regression model is constructed based on the linear relationship between the carbon isotope characteristic difference of gradient-spike samples and the adulteration ratio. Optionally, the sample pretreatment module includes: The pH adjustment unit is used to adjust the pH value of the dairy product sample to be tested to 4.4-4.8; The first centrifugation unit is used to separate casein precipitate and supernatant by centrifuging at 4000r / min-4500r / min for 8-12 minutes at 3℃-5℃. The protein purification unit is used to add a 1:1 volume ratio ethanol-ether mixture to the casein precipitate for extraction. After vortexing, the upper organic phase is discarded by centrifugation. The extraction operation is repeated 2-3 times. The washed protein precipitate is then freeze-dried under vacuum at -40°C to -60°C for 20-28 hours to obtain milk protein powder. A lactose crystallization unit is used to collect the supernatant, add 3 to 5 times the volume of anhydrous ethanol, and let it stand at 3℃-5℃ for 10 to 14 hours to allow lactose crystals to precipitate. The second centrifugal separation unit is used to collect lactose crystals by centrifugation, wash them 2-4 times with anhydrous ethanol, and dry them to obtain lactose powder.

[0012] Optionally, the coefficient of determination R² of the linear regression model is greater than or equal to 0.99, and the linear regression model is independently established by running gradient-spike samples for a specific exogenous plant protein or a specific detection batch.

[0013] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention effectively removes the cross-interference between fat and free sugar in dairy products by using protein extraction process of isoelectric point precipitation combined with organic solvent extraction and lactose purification process of low temperature recrystallization of ethanol, ensuring the high purity of the powder to be tested, thereby significantly improving the accuracy and repeatability of isotope detection.

[0014] (2) By introducing lactose from cow's own milk as an endogenous isotope internal standard and using the relative difference between Δδ¹³C protein and lactose as a judgment index, the present invention offsets systematic errors such as sample matrix differences and instrument state drift, and greatly enhances the anti-interference ability of the detection method.

[0015] (3) This invention integrates qualitative screening based on statistical thresholds with quantitative analysis based on linear regression models. It can not only quickly determine adulteration behavior, but also accurately calculate the specific adulteration ratio of exogenous plant protein. The coefficient of determination R² of the quantitative model can reach more than 0.99, and the quantitative error is significantly reduced. Attached Figure Description

[0016] Figure 1 This is a flowchart of the method for identifying adulteration of exogenous plant protein in pure dairy products according to an embodiment of the present invention; Figure 2 This is a framework diagram of the detection system for exogenous adulterants in dairy products in an embodiment of the present invention.

[0017] The attached figures are labeled as follows: 1-Sample pretreatment module, 101-pH adjustment unit, 102-First centrifugation unit, 103-Protein purification unit, 104-Lactose crystallization unit, 105-Second centrifugation unit, 2-Isotope determination module, 3-Data processing module, 4-Judgment and analysis module, 5-Quantitative calculation module. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] like Figure 1 As shown, this invention provides a method for identifying adulteration with exogenous plant proteins in pure dairy products. The method includes: S1. Simultaneously extract high-purity milk protein and lactose from the pure dairy products to be tested.

[0020] S2. The carbon stable isotope ratio δ¹³C of milk protein and lactose were determined by elemental analyzer-stable isotope ratio mass spectrometry.

[0021] S3. Calculate the carbon isotope characteristic difference Δδ¹³C protein-lactose between milk protein and lactose.

[0022] S4. Determine the adulteration hypothesis based on statistical thresholds established using pure milk.

[0023] S5. The adulteration ratio of exogenous plant proteins is quantitatively calculated using a pre-constructed linear regression model.

[0024] Among them, the carbon isotope characteristic difference is the difference between the carbon stable isotope ratio of milk protein and the carbon stable isotope ratio of lactose. The carbon isotope characteristic difference constitutes a characterization index of the endogenous metabolic isotope coupling relationship of pure milk.

[0025] The carbon isotope characteristic difference Δδ¹³C represents the characterization index of the endogenous metabolic isotope coupling relationship between protein and lactose in pure cow's milk. Since casein, whey protein, and lactose in pure cow's milk all originate from the metabolic synthesis process of the dairy cow, they have a close coupling relationship in their metabolic pathways, thus exhibiting a high degree of consistency in their stable carbon isotope ratios. When exogenous plant proteins are added to pure cow's milk, the carbon isotope ratio of the exogenous plant proteins differs significantly from that of the endogenous cow's milk proteins. This difference disrupts the original isotopic balance between milk proteins and lactose, leading to a significant change in the carbon isotope characteristic difference, thereby achieving the purpose of adulteration detection.

[0026] In the isotope determination stage, this invention employs a coupled technique of elemental analyzer and stable isotope ratio mass spectrometer to perform high-precision determination of the carbon stable isotope ratio in milk protein and lactose samples. The core equipment of this coupled technique consists of two parts: an elemental analyzer and a stable isotope ratio mass spectrometer, which are connected via an interface device to achieve online transmission of gas components. The working principle of the elemental analyzer is to completely oxidize the organic carbon in the sample into carbon dioxide gas under high temperature conditions. The specific operating steps are as follows: Accurately weigh a certain amount of milk protein powder and lactose powder, and place them in a special tin container for the elemental analyzer. The preferred sample amount of milk protein powder is 0.8 mg to 1.2 mg, and the preferred sample amount of lactose powder is 0.5 mg to 0.8 mg. After carefully folding and sealing the tin container, place it into the automatic sample introduction device of the elemental analyzer. Inside the combustion furnace of the elemental analyzer, the sample reacts fully with oxygen at a high temperature of 980℃ to 1000℃, and the organic carbon is completely oxidized into carbon dioxide. After removing moisture and impurities, the combustion product gas enters a stable isotope ratio mass spectrometer through an interface device for isotope ratio determination. The stable isotope ratio mass spectrometer is used to accurately determine the ratio of carbon-13 to carbon-12 atoms in carbon dioxide gas and reports the measurement results in the form of a thousandth of a percent deviation relative to international standard materials. In this invention, the international standard reference for carbon stable isotope ratios is belemnites fossils from the Pointe Paddy Formation in South Carolina, USA, abbreviated as VPDB. The formula for calculating the carbon stable isotope ratio is: δ¹³C = [Rsample / Rstandard-1]×1000‰, where Rsample is the ratio of the number of ¹³C atoms to ¹²C atoms in the sample, and Rstandard is the ratio of the number of ¹³C atoms to ¹²C atoms in the VPDB standard reference. The measurement accuracy of stable isotope ratio mass spectrometry is typically better than 0.1‰, which meets the accuracy requirements for isotope ratio determination in this invention.

[0027] This invention addresses the key problems in existing technologies, such as severe sample matrix interference, susceptibility to instrument fluctuations in detection results, and the inability to simultaneously perform qualitative and quantitative analysis. By employing a protein extraction process combining isoelectric point precipitation with organic solvent extraction, and a lactose purification process using low-temperature recrystallization of ethanol, the cross-interference between fat and free sugars in dairy products is effectively removed, ensuring the high purity of the analyte powder and significantly improving the accuracy and repeatability of isotope detection. By introducing lactose from bovine milk as an endogenous isotope internal standard and using the Δδ¹³C protein-lactose relative difference as a criterion, systematic errors such as sample matrix differences and instrument drift are offset, greatly enhancing the anti-interference capability of the detection method.

[0028] In other embodiments, the carbon isotope characteristic differences can be obtained through the following steps: The pH of the pure dairy product sample to be tested was adjusted to 4.6 and centrifuged to obtain casein precipitate and supernatant. The casein precipitate was extracted and centrifuged using an ethanol-ether mixture, and then freeze-dried to obtain milk protein powder. Anhydrous ethanol was added to the supernatant and allowed to stand at 4°C to obtain lactose crystals. The lactose crystals were washed with ethanol and dried to obtain lactose powder. The milk protein powder and lactose powder were placed separately in an elemental analyzer-stable isotope ratio mass spectrometry system and burned at 980°C-1000°C to convert them into carbon dioxide gas. The carbon stable isotope ratios of milk protein and lactose were measured, and the carbon isotope characteristic difference was calculated.

[0029] In the above steps, adjusting the pH of the pure dairy product sample to the isoelectric point of casein (4.6) aims to ensure sufficient precipitation of casein, thereby achieving effective separation of milk protein and lactose. The preferred volume ratio of the ethanol-ether mixture is 1:1. This mixture effectively dissolves and removes milk fat and free lactose without damaging the chemical structure of the protein. The preferred conditions for vacuum freeze-drying are -40°C to -60°C for 20-28 hours, which effectively removes solvent residue while maintaining the chemical integrity of the protein. During lactose crystallization, the preferred amount of anhydrous ethanol added is 3-5 times the volume of the supernatant, the preferred settling temperature is 3°C-5°C, and the preferred settling time is 10-14 hours. These conditions ensure sufficient precipitation of lactose crystals and a high yield. The combustion temperature of the elemental analyzer is set to 980°C-1000°C to ensure complete combustion of organic carbon in the sample into carbon dioxide, while avoiding isotope fractionation due to excessively high temperatures.

[0030] In other embodiments, the statistical threshold can be obtained through the following steps: First, collect no fewer than ten milk samples that have been verified as pure using authoritative methods. These samples should originate from healthy dairy cows of different batches, origins, or farms, and be reliably confirmed to be free of any added exogenous plant proteins, exogenous sugars, whey powder, or fat substitutes. The collection and preservation of samples must strictly adhere to relevant standard operating procedures to avoid sample deterioration or changes in isotopic composition due to improper storage. Then, milk protein and lactose were extracted simultaneously from each pure milk sample using the aforementioned method. The carbon stable isotope ratios of milk protein and lactose in each sample were determined using an elemental analyzer-stable isotope ratio mass spectrometry system, and the carbon isotope characteristic differences of each sample were calculated. After the measurements were completed, the carbon isotope characteristic difference data of all pure milk samples were collected, and the arithmetic mean and standard deviation of these differences were calculated. Finally, the arithmetic mean plus twice the standard deviation was used as the statistical threshold. This threshold setting method is based on the statistical principle of normal distribution: under normal circumstances, the carbon isotope characteristic differences of pure milk fluctuate within a small range around a certain mean, following a normal or approximately normal distribution. Using the arithmetic mean plus twice the standard deviation as the threshold can distinguish pure milk from adulterated samples at a 95% confidence level, thus achieving a high-confidence determination of adulteration.

[0031] The principle behind setting the statistical threshold is that the carbon isotope characteristic difference of pure milk follows a certain statistical distribution law. Under normal circumstances, this difference should fluctuate within a small range around a certain mean. When exogenous plant proteins are introduced, the carbon isotope characteristic difference will significantly deviate from this normal fluctuation range. Using the arithmetic mean plus twice the standard deviation as the threshold can distinguish pure milk from adulterated samples at a 95% confidence level, thus achieving a high-confidence determination of adulteration.

[0032] In other embodiments, the doping hypothesis determination includes the following steps: The carbon isotope characteristic difference is compared with a statistical threshold. When the carbon isotope characteristic difference of the sample is greater than the statistical threshold, it is determined that exogenous plant protein is detected; when the carbon isotope characteristic difference of the sample is less than or equal to the statistical threshold, it is determined that exogenous plant protein is not detected.

[0033] The above-mentioned judgment rule is based on the fact that the carbon stable isotope ratio of exogenous plant proteins is usually significantly lower than that of endogenous proteins in cow's milk. When exogenous plant proteins are added to cow's milk, the carbon stable isotope ratio of milk proteins decreases, while the carbon stable isotope ratio of lactose remains unchanged (because lactose originates from the metabolism of dairy cows and is not affected by exogenous plant proteins). Therefore, the difference in carbon isotope characteristics increases. When this difference increases to exceed the statistical judgment threshold established based on pure cow's milk, it can be determined that exogenous plant protein adulteration exists.

[0034] In other embodiments, the linear regression model is a univariate linear regression model, using the mass proportion of the added exogenous plant protein in the tested pure dairy product as the independent variable and the carbon isotope characteristic difference as the dependent variable, and is established by least squares fitting. Quantitative calculation involves substituting the carbon isotope characteristic difference of the tested pure dairy product as the dependent variable into the linear regression model, and solving the independent variable in reverse to obtain the adulteration proportion of the exogenous plant protein.

[0035] The principle behind the aforementioned quantitative calculation method is that when exogenous plant protein is added to pure milk in different proportions, there is an approximately linear relationship between the carbon isotope characteristic difference and the adulteration ratio. By preparing a series of gradient-spiked samples with known adulteration ratios, measuring the carbon isotope characteristic difference of each spiked sample, and using the least squares method for linear regression fitting, a quantitative relationship model between the carbon isotope characteristic difference and the adulteration ratio can be established. In actual testing, simply substituting the carbon isotope characteristic difference of the sample to be tested into this model will allow the calculation of the corresponding adulteration ratio.

[0036] In other embodiments, the coefficient of determination R² of the linear regression model is greater than or equal to 0.99, and the linear regression model is independently established by running gradient-spike samples for a specific exogenous plant protein or a specific test batch.

[0037] The coefficient of determination (R²) is used to evaluate the goodness of fit of a linear regression model. The closer R² is to 1, the better the model fits the data. When R² ≥ 0.99, it indicates a highly significant linear relationship between the carbon isotope characteristic difference and the adulteration ratio, and the model has high predictive accuracy. Because the background carbon isotope values ​​differ among different types of plant proteins, separate linear regression models need to be constructed for specific types of exogenous plant proteins or specific batches of samples to ensure the accuracy of quantitative calculations.

[0038] Constructing a linear regression model requires preparing a series of gradient-spiked samples with known adulteration ratios for a specific type of exogenous plant protein. The specific steps are as follows: First, a sample of milk that has been confirmed to be pure is selected as the base sample, and the carbon stable isotope ratio of its milk protein and lactose is measured. The carbon isotope characteristic difference is calculated as the background value.

[0039] Then, known amounts of a specific type of exogenous plant protein are added in a gradient to the substrate sample to prepare at least five spiked samples with different adulteration ratios. The gradient setting of the adulteration ratio should cover possible adulteration ranges, such as 0 mg / 10 mL, 20 mg / 10 mL, 50 mg / 10 mL, 80 mg / 10 mL, 100 mg / 10 mL, 200 mg / 10 mL, etc. Next, the samples spiked at each gradient were simultaneously extracted and isotope measured using the aforementioned method to obtain the carbon isotope characteristic differences for each sample. Using the adulteration ratio as the independent variable and the carbon isotope characteristic differences as the dependent variable, a univariate linear regression was performed using the least squares method. The general form of the linear regression equation is: Y = aX + b, where Y is the carbon isotope characteristic difference, X is the adulteration ratio of the exogenous plant protein, a is the slope, and b is the intercept. Finally, the goodness of fit of the linear regression model is evaluated. The coefficient of determination (R²) is the core indicator for evaluating the goodness of fit of the linear regression model, and its calculation formula is: R² = SSR / SST, where SSR is the regression sum of squares and SST is the total sum of squares. The value of R² ranges from 0 to 1; the closer it is to 1, the better the model fits the data. This invention requires that the coefficient of determination R² of the linear regression model must be greater than or equal to 0.99 to ensure that the model has sufficient predictive accuracy. In practical quantitative calculations, the carbon isotope characteristic difference of the pure dairy product being tested is substituted as the dependent variable Y into a pre-established linear regression equation Y = aX + b. The independent variable X is then solved in reverse to obtain the adulteration ratio of the exogenous plant protein. Because the background carbon isotope values ​​differ among different types of plant proteins, a separate linear regression model needs to be constructed for each specific type of exogenous plant protein or specific batch of testing to ensure the accuracy of the quantitative calculations.

[0040] In other embodiments, the exogenous plant protein is selected from one or more of rice protein, wheat protein, soybean protein, peanut protein, pea protein, and hydrolyzed wheat protein. The above six plant proteins are the most common and inexpensive sources of plant protein in dairy product adulteration. Their carbon stable isotope ratios are all significantly lower than those of endogenous bovine milk proteins, which can produce obvious isotopic difference signals, thereby ensuring that the method of the present invention can effectively cover the main types of adulteration currently.

[0041] like Figure 2 As shown, the present invention provides a detection system for exogenous adulterants in dairy products, used to implement the above-mentioned detection method for exogenous adulterants in dairy products, comprising: Sample pretreatment module 1 is used to simultaneously extract milk protein components and lactose components from the dairy product sample to be tested. The milk protein components include casein and whey protein.

[0042] The isotope determination module 2 includes an elemental analyzer-stable isotope ratio mass spectrometry system, used to determine the carbon stable isotope ratios of milk protein components and lactose components, respectively.

[0043] Data processing module 3 is used to calculate the carbon isotope characteristic differences. The judgment analysis module 4 is used to compare the carbon isotope characteristic difference with the preset statistical judgment threshold. When the carbon isotope characteristic difference is greater than the statistical judgment threshold, it is determined that there is an exogenous adulterant in the dairy product to be tested; when the carbon isotope characteristic difference is less than or equal to the statistical judgment threshold, it is determined that there is no exogenous adulterant in the dairy product to be tested. The quantitative calculation module 5 is used to calculate the adulteration ratio of exogenous adulterants by substituting the carbon isotope characteristic difference into a pre-constructed linear regression model when it is determined that there are exogenous adulterants. The statistical judgment threshold is determined based on the statistical distribution of carbon isotope characteristic difference in the pure dairy product sample set, and the linear regression model is constructed based on the linear relationship between the carbon isotope characteristic difference of the gradient spiked samples and the adulteration ratio. Specifically, sample pretreatment module 1 is responsible for separating and purifying milk protein and lactose, two target components, from the complex dairy matrix, a crucial step in ensuring the accuracy of subsequent isotope detection. Isotope determination module 2 employs a combined elemental analyzer and stable isotope ratio mass spectrometer, enabling quantitative combustion conversion of carbon and high-precision determination of isotope ratios. Data processing module 3 is responsible for the acquisition, calculation, and storage of raw data. Judgment analysis module 4 performs binarization judgment based on preset statistical judgment thresholds, outputting qualitative detection conclusions. Quantitative calculation module 5, based on the qualitative judgment, uses a pre-established linear regression model to perform quantitative calculations, outputting the specific value of the adulteration ratio. In other embodiments, the sample pretreatment module 1 includes: pH adjustment unit 101 is used to adjust the pH value of the dairy product sample to be tested to 4.4-4.8; The first centrifugal separation unit 102 is used to separate casein precipitate and supernatant by centrifuging at 4000-4500 r / min for 8-12 minutes at 3-5℃. The protein purification unit 103 is used to add a 1:1 volume ratio ethanol-ether mixture to the casein precipitate for extraction. After vortexing, the upper organic phase is discarded by centrifugation. The extraction operation is repeated 2 to 3 times. The washed protein precipitate is then freeze-dried under vacuum at -40 to -60°C for 20 to 28 hours to obtain milk protein powder. Lactose crystallization unit 104 is used to collect the supernatant, add 3 to 5 times the volume of anhydrous ethanol, and let it stand at 3℃-5℃ for 10 to 14 hours to allow lactose crystals to precipitate. The second centrifugal separation unit 105 is used to collect lactose crystals by centrifugation, wash them 2-4 times with anhydrous ethanol, and dry them to obtain lactose powder.

[0044] Specifically, in the sample pretreatment module 1, the pH adjustment unit 101 typically uses a precision pH meter for real-time monitoring. The pH of the system is adjusted to the range of 4.4 to 4.8, preferably 4.6, by adding dilute hydrochloric acid solution. At this pH, casein has the lowest solubility, maximizing precipitation. The first centrifugation unit 102 uses a high-speed refrigerated centrifuge. Preferred centrifugation conditions are a temperature of 3℃-5℃, a rotation speed of 4000r / min-4500r / min, and a centrifugation time of 8-12 minutes. These conditions effectively separate the casein precipitate from the lactose-containing supernatant. In the protein purification unit 103, the volume ratio of the ethanol-ether mixture is preferably 1:1. The extraction operation is preferably repeated 2-3 times to ensure sufficient removal of fat and lactose impurities. The vacuum freeze-drying conditions are preferably -40℃ to -60℃ for 20-28 hours. The amount of anhydrous ethanol added in the lactose crystallization unit 104 is preferably 3 to 5 times the volume of the supernatant, and the settling conditions are preferably settling at 3°C ​​to 5°C for 10 to 14 hours. The washing operation in the second centrifugation unit 105 is preferably performed 2 to 4 times to ensure the purity of the lactose crystals. The present invention will be further described below with reference to specific embodiments.

[0045] Example 1 This embodiment collected ten batches of confirmed pure sterilized milk samples, labeled M1 to M10. These samples came from different dairy farms in different regions and were confirmed by high-performance liquid chromatography and enzyme-linked immunosorbent assay (ELISA) to have no added exogenous plant proteins. Milk protein and lactose were extracted simultaneously from each sample according to the method of this invention. The specific operating steps are as follows: Take 10 ml of each pure milk sample and place it in a centrifuge tube. Adjust the pH to 4.6 with dilute hydrochloric acid solution and let it stand at room temperature for 20 minutes to allow casein to precipitate completely. Centrifuge at 4000 rpm for 10 minutes at 4°C to separate the casein precipitate and supernatant. Add 3 volumes of ethanol-ether mixture (1:1 volume ratio) to the casein precipitate, vortex for 2 minutes, and centrifuge at 3500 rpm for 5 minutes to remove the upper organic phase. Repeat the above extraction operation twice. Freeze-dry the protein precipitate after the last centrifugation at -50°C for 24 hours to obtain milk protein powder. Add 4 volumes of anhydrous ethanol to the supernatant and let it stand at 4°C for 12 hours to allow lactose crystals to fully separate. Collect the lactose crystals by centrifugation at 4000 rpm for 10 minutes, wash three times with anhydrous ethanol, and then vacuum-dry at 45°C for 3 hours to obtain lactose powder. The obtained milk protein powder and lactose powder were placed in an elemental analyzer-stable isotope ratio mass spectrometry system for carbon stable isotope ratio determination. The combustion temperature of the elemental analyzer was set to 990℃, and the samples were completely combusted into carbon dioxide gas under high temperature conditions. The stable isotope ratio mass spectrometry determined the atomic ratio of ¹³C / ¹²C in the carbon dioxide gas and calculated the δ¹³C value relative to the VPDB standard. The test results of ten pure milk samples are as follows: The milk protein δ¹³C value of sample M1 was -17.521‰, the lactose δ¹³C value was -17.554‰, and the carbon isotope characteristic difference was 0.033‰. The milk protein δ¹³C value of sample M2 was -18.012‰, the lactose δ¹³C value was -17.675‰, and the carbon isotope characteristic difference was -0.337‰. The milk protein δ¹³C value of sample M3 was -17.398‰, the lactose δ¹³C value was -17.605‰, and the carbon isotope characteristic difference was 0.207‰. The milk protein δ¹³C value of sample M4 was -17.865‰, the lactose δ¹³C value was -17.711‰, and the carbon isotope characteristic difference was -0.154‰. The M5 sample had a milk protein δ¹³C value of -18.234‰, a lactose δ¹³C value of -17.801‰, and a carbon isotope characteristic difference of -0.433‰. The milk protein δ¹³C value of sample M6 was -17.712‰, the lactose δ¹³C value was -17.609‰, and the carbon isotope characteristic difference was -0.103‰. The milk protein δ¹³C value of sample M7 was -17.521‰, the lactose δ¹³C value was -17.521‰, and the carbon isotope characteristic difference was 0.000‰. The milk protein δ¹³C value of sample M8 was -18.155‰, the lactose δ¹³C value was -17.821‰, and the carbon isotope characteristic difference was -0.334‰. The milk protein δ¹³C value of sample M9 was -17.688‰, the lactose δ¹³C value was -17.609‰, and the carbon isotope characteristic difference was -0.079‰. The milk protein δ¹³C value of sample M10 was -17.463‰, the lactose δ¹³C value was -17.505‰, and the carbon isotope characteristic difference was 0.042‰. Based on the above measurement results, the arithmetic mean of the carbon isotope characteristic differences of ten pure milk samples was calculated. The arithmetic mean was calculated by adding the ten carbon isotope characteristic differences and then dividing by the sample size of 10. The calculated arithmetic mean was -0.1158‰. The standard deviation was calculated using the sample standard deviation formula. First, the square of the difference between each sample's carbon isotope characteristic difference and the arithmetic mean was calculated. Then, the arithmetic mean of these squared values ​​was calculated. Finally, the square root of this arithmetic mean was taken. The calculated standard deviation was 0.201325‰. The arithmetic mean plus twice the standard deviation was used as the statistical decision threshold, resulting in a decision threshold of 0.28685‰. This decision threshold was calculated by adding the arithmetic mean (-0.1158‰) to twice the standard deviation (0.40265‰), resulting in 0.28685‰. Based on this threshold, the following qualitative judgment rule is established: when the carbon isotope characteristic difference of the sample is greater than 0.28685‰, it is judged as adulteration with exogenous plant protein; when the carbon isotope characteristic difference is less than or equal to 0.28685‰, it is judged as pure milk, and no exogenous plant protein is detected. The theoretical basis of this judgment rule is that when exogenous plant protein is added to pure milk, the carbon stable isotope ratio of milk protein will decrease due to the addition of exogenous plant protein, while the carbon stable isotope ratio of lactose remains unchanged. Therefore, the carbon isotope characteristic difference will show an increasing trend. When this difference increases to exceed the threshold of 0.28685‰, which is determined based on the statistical distribution of pure milk, it can be reliably judged that exogenous plant protein adulteration exists.

[0046] Comparative Example 1 This comparative example uses pure milk (M7 grade) as the base sample, with a background carbon isotope characteristic difference of 0‰. Six groups of simulated adulterated samples were prepared by gradually adding hydrolyzed wheat protein to the samples, with addition amounts of 0 mg / 10 mL, 20 mg / 10 mL, 50 mg / 10 mL, 80 mg / 10 mL, 100 mg / 10 mL, and 200 mg / 10 mL, respectively. Each group of samples was pretreated and subjected to isotope analysis according to the method of this invention, and the carbon isotope characteristic difference of each group was calculated. The sample pretreatment and isotope determination procedures for the comparative examples were exactly the same as in Example 1. The results for the zero-addition control group M7-0 showed: milk protein δ¹³C value -17.521‰, lactose δ¹³C value -17.521‰, and carbon isotope characteristic difference of 0.000‰. This value was less than the judgment threshold of 0.28685‰, and was determined to be pure cow's milk. The results for the 20 mg addition group M7-20 showed: milk protein δ¹³C value -18.215‰, lactose δ¹³C value -17.498‰, and carbon isotope characteristic difference of -0.717‰. The results for the 50 mg addition group M7-50 showed: milk protein δ¹³C value -18.756‰, lactose δ¹³C value -17.512‰, and carbon isotope characteristic difference of -1.244‰. The results for the 80 mg supplemented group (M7-80) showed: milk protein δ¹³C value -19.234‰, lactose δ¹³C value -17.489‰, and carbon isotope characteristic difference -1.745‰. The results for the 100 mg supplemented group (M7-100) showed: milk protein δ¹³C value -19.567‰, lactose δ¹³C value -17.525‰, and carbon isotope characteristic difference -2.042‰. The results for the 200 mg supplemented group (M7-200) showed: milk protein δ¹³C value -20.498‰, lactose δ¹³C value -17.534‰, and carbon isotope characteristic difference -2.964‰. Comparing the above comparative results with the judgment threshold of 0.28685‰, it can be found that the carbon isotope characteristic difference values ​​of all sample groups with added hydrolyzed wheat protein are less than the judgment threshold. According to the judgment rules of this invention, a carbon isotope characteristic difference value less than the judgment threshold should be judged as pure milk, with no detectable exogenous plant protein. However, the opposite is true; these samples were actually adulterated with hydrolyzed wheat protein. This phenomenon indicates that when exogenous plant protein is added to pure milk, the carbon stable isotope ratio of milk protein decreases significantly, leading to a negative increasing trend in the carbon isotope characteristic difference value, rather than a positive increasing trend. In-depth analysis revealed that the reason for the above phenomenon lies in the fact that the carbon stable isotope ratio of lactose does not remain constant as expected. In fact, when exogenous plant proteins are incorporated into the milk system, even though lactose itself originates from the metabolism of the cow's body and is not directly affected by the exogenous plant proteins, the carbon stable isotope ratio of lactose will still change to some extent due to the dilution effect of the system. More importantly, there is a close metabolic coupling between the carbon stable isotope ratios of milk proteins and lactose in pure milk; their values ​​are very close. When exogenous plant proteins with significantly lower carbon isotope ratios than endogenous milk components are incorporated, the carbon isotope ratio of milk proteins will decrease significantly, while the carbon isotope ratio of lactose, although also affected to some extent, usually changes less than that of milk proteins. Therefore, the carbon isotope characteristic difference shows a negative increasing trend, that is, the absolute value of the difference between δ¹³C protein and δ¹³C lactose increases, and the sign shifts towards the negative. Based on the above comparative analysis, this invention adaptively adjusts the judgment rule: comparing the absolute value of the carbon isotope characteristic difference with a statistical judgment threshold. When the absolute value of the carbon isotope characteristic difference is greater than the judgment threshold, it is judged as adulteration with exogenous plant protein. This adjusted judgment rule can simultaneously cover both positive and negative increases in the carbon isotope characteristic difference, greatly improving the applicability and detection coverage of the method of this invention.

[0047] Example 2 Based on the measurement results of Example 1 and Comparative Example 1, this example establishes a linear regression model for quantitatively calculating the adulteration ratio of wheat hydrolyzed protein. Using the amount of hydrolyzed wheat protein added as the independent variable X and the absolute value of the carbon isotope characteristic difference as the dependent variable Y, a univariate linear regression was performed using the least squares method. The data pairs for the five addition gradient samples are: X1=20, Y1=0.717; X2=50, Y2=1.244; X3=80, Y3=1.745; X4=100, Y4=2.042; X5=200, Y5=2.964.

[0048] The calculation process for least squares linear regression is as follows: First, calculate the arithmetic mean of the independent and dependent variables. Then, calculate the sum of the products of the deviations of each data point and the sum of the squares of the deviations of the independent variables, and finally calculate the regression coefficients. The calculated slope is a = 0.01175, and the intercept is b = 0.466. Therefore, the linear regression equation is: Y = 0.01175X + 0.466. The calculation process for the coefficient of determination is as follows: first, calculate the arithmetic mean of the dependent variable; then, calculate the total sum of squares, regression sum of squares, and residual sum of squares. The final calculated R² = 0.9928, satisfying the requirement of R² ≥ 0.99. This linear regression model can be used to quantitatively calculate the amount of wheat hydrolyzed protein added to unknown adulterated samples. To verify the predictive accuracy of the linear regression model, the following verification experiment was conducted: A sample with an unknown adulteration ratio was taken and pretreated and isotope determined according to the method of this invention. The carbon isotope characteristic difference was measured to be -1.520‰, and the absolute value was 1.520‰. Substituting 1.520‰ into the linear regression equation Y = 0.01175X + 0.466, and solving for X in reverse: X = (1.520 - 0.466) / 0.01175 = 89.70 mg per 10 mL. The actual amount added was 90 mg per 10 mL. The relative error between the calculated adulteration amount and the actual amount added was 0.33%, which is less than 3%, indicating that the quantitative result is accurate and reliable.

[0049] Example 3 This embodiment measured the background carbon stable isotope ratios of six common exogenous plant proteins to verify the effectiveness of the method of the present invention in detecting these common adulteration types. The six exogenous plant protein samples were rice protein, wheat protein, soybean protein, peanut protein, pea protein, and hydrolyzed wheat protein. All samples were purchased from legitimate sources and their protein content was confirmed to be greater than 80% by high-performance liquid chromatography (HPLC). Each plant protein sample was pretreated and isotope analyzed according to the method of this invention. Each sample was analyzed in triplicate, and the average value was taken as the final result. The results showed that the average carbon isotope ratio of rice protein was -29.8875‰; the average carbon isotope ratio of wheat protein was -28.141‰; the average carbon isotope ratio of soybean protein was -25.9225‰; the average carbon isotope ratio of peanut protein was -25.9865‰; the average carbon isotope ratio of pea protein was -26.3275‰; and the average carbon isotope ratio of hydrolyzed wheat protein was -28.172‰. The above measurement results were compared with the isotope ratio range of pure milk protein. The carbon isotope ratio range of pure milk protein is -16.71‰ to -18.321‰. It can be seen that the carbon isotope ratios of all six plant proteins are significantly lower than that of pure milk protein, with differences exceeding 7‰, and the difference for rice protein exceeding 10‰. These background isotope data fully confirm the significant isotopic differences between plant proteins from different sources and milk proteins, demonstrating that the method of this invention can effectively distinguish pure milk from adulterated samples. The table below shows a comparison of the isotopic characteristics of six common exogenous plant proteins and pure milk proteins:

[0050] Example 4 This embodiment tests a commercially available sterilized milk sample labeled as 100% pure cow's milk to verify the detection effect of the method of the present invention in a practical application scenario. Sample pretreatment and isotope determination were performed according to the method of this invention, with the specific operating steps being the same as in Example 1. The determination results showed that the carbon stable isotope ratio of the milk protein in the sample was -19.620‰, the carbon stable isotope ratio of lactose was -18.100‰, and the calculated carbon isotope characteristic difference was -1.520‰. The carbon isotope characteristic difference value of -1.520‰ was compared with the judgment threshold of 0.28685‰. According to the adjusted judgment rule, the absolute value of the carbon isotope characteristic difference value of 1.520‰ was compared with the judgment threshold of 0.28685‰. Since 1.520‰ is greater than 0.28685‰, it was preliminarily determined that the sample contained exogenous plant protein adulteration. Further quantitative calculations of the adulteration ratio were performed. The absolute value of the carbon isotope characteristic difference, 1.520‰, was substituted into the linear regression equation for wheat hydrolyzed protein established in Example 2, Y = 0.01175X + 0.466, and the result was X: X = (1.520 - 0.466) / 0.01175 = 89.70 mg per 10 ml. Considering systematic and random errors in the detection process, the final reported result was approximately 92 mg per 10 ml of milk. To verify the repeatability and stability of the method of the present invention, three parallel repeated experiments were conducted. The carbon isotope characteristic differences of the three parallel detections were -1.520‰, -1.495‰, and -1.538‰, respectively, with a fluctuation range not exceeding 0.05‰. The doping amounts calculated based on the results of the three parallel detections were 92 mg / 10 mL, 88 mg / 10 mL, and 94 mg / 10 mL, respectively, with relative deviations of less than 2.5%. The above parallel experimental results show that the method of the present invention has good repeatability and stability and can meet the requirements of practical detection work. Based on the verification results of the above embodiments and comparative examples, it is fully demonstrated that the method and detection system for identifying exogenous plant protein adulteration in pure dairy products provided by this invention have the following technical features: First, by using the endogenous metabolic isotope coupling relationship between pure milk protein and lactose as the basis for judgment, it can fundamentally distinguish pure dairy products from adulterated products, exhibiting high specificity at the principle level; Second, by simultaneously extracting milk protein and lactose and measuring them separately, it effectively eliminates systematic errors caused by differences in sample matrix, improving the accuracy and reliability of the detection results; Third, by using elemental analyzer-stable isotope ratio mass spectrometry for isotope determination, it has the advantages of simple sample pretreatment, fast detection speed, and high sensitivity, meeting the practical needs of large-scale rapid screening; Fourth, by using statistical thresholds for qualitative judgment and combining linear regression models for quantitative calculation, it achieves both qualitative and quantitative judgment of adulteration behavior; Fifth, the method of this invention has effective detection capabilities for various common types of plant protein adulteration, covering the main forms of adulteration in current dairy products.

[0051] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for identifying adulteration with exogenous plant proteins in pure dairy products, comprising simultaneously extracting milk protein and lactose from the pure dairy product to be tested, characterized in that, The carbon stable isotope ratios of milk protein and lactose were determined by elemental analyzer-stable isotope ratio mass spectrometry, and the carbon isotope characteristic difference was calculated. The adulteration hypothesis was determined based on the statistical threshold established by pure cow's milk, and the adulteration ratio of exogenous plant protein was quantitatively calculated by a pre-constructed linear regression model. The carbon isotope characteristic difference is the difference between the carbon stable isotope ratio of milk protein and the carbon stable isotope ratio of lactose. The carbon isotope characteristic difference constitutes a characterization index of the endogenous metabolic isotope coupling relationship of pure milk.

2. The method for identifying adulteration with exogenous plant protein in pure dairy products as described in claim 1, characterized in that, The carbon isotope characteristic differences are obtained through the following steps: The pH of the pure dairy product sample to be tested was adjusted to 4.6 and centrifuged to obtain casein precipitate and supernatant. The casein precipitate was extracted and centrifuged using an ethanol-ether mixture, and then freeze-dried to obtain milk protein powder. Anhydrous ethanol was added to the supernatant and allowed to stand at 4°C to obtain lactose crystals. The lactose crystals were washed with ethanol and dried to obtain lactose powder. The milk protein powder and the lactose powder were placed separately in an elemental analyzer-stable isotope ratio mass spectrometry system and burned at 980°C-1000°C to convert them into carbon dioxide gas. The carbon stable isotope ratios of the milk protein and the lactose were measured, and the carbon isotope characteristic difference was calculated.

3. The method for identifying adulteration with exogenous plant protein in pure dairy products as described in claim 1, characterized in that, The statistical threshold is obtained through the following steps: Collect no fewer than 10 pure milk samples, measure the carbon isotope characteristic difference of each sample, calculate the arithmetic mean and standard deviation of the carbon isotope characteristic difference, and use the arithmetic mean plus twice the standard deviation as the judgment threshold.

4. The method for identifying adulteration with exogenous plant protein in pure dairy products as described in claim 1, characterized in that, The determination of the doping hypothesis includes the following steps: The carbon isotope characteristic difference is compared with the statistical threshold. When the carbon isotope characteristic difference of the sample to be tested is greater than the statistical threshold, it is determined that exogenous plant protein is detected; when the carbon isotope characteristic difference of the sample to be tested is less than or equal to the statistical threshold, it is determined that exogenous plant protein is not detected.

5. The method for identifying adulteration with exogenous plant protein in pure dairy products as described in claim 1, characterized in that, The linear regression model is a univariate linear regression model, which uses the mass ratio of the added exogenous plant protein in the pure dairy product to be tested as the independent variable and the carbon isotope characteristic difference as the dependent variable, and is obtained by fitting the model using the least squares method. The quantitative calculation includes substituting the carbon isotope characteristic difference of the pure dairy product to be tested as the dependent variable into the linear regression model, and solving the independent variable in reverse to obtain the adulteration ratio of exogenous plant protein.

6. The method for identifying adulteration with exogenous plant protein in pure dairy products as described in claim 5, characterized in that, The coefficient of determination R² of the linear regression model is greater than or equal to 0.99, and the linear regression model is independently established by running gradient-spike samples for a specific exogenous plant protein or a specific batch of tests.

7. The method for detecting exogenous adulterants in dairy products as described in any one of claims 1-6, characterized in that, The exogenous plant protein is selected from one or more of rice protein, wheat protein, soybean protein, peanut protein, pea protein, and hydrolyzed wheat protein.

8. A detection system for exogenous adulterants in dairy products, used to implement the detection method for exogenous adulterants in dairy products according to any one of claims 1-7, characterized in that, include: The sample pretreatment module is used to simultaneously extract milk protein components and lactose components from the dairy product sample to be tested, wherein the milk protein components include casein and whey protein; The isotope determination module includes an elemental analyzer-stable isotope ratio mass spectrometry system, used to determine the carbon stable isotope ratio of the milk protein component and the carbon stable isotope ratio of the lactose component, respectively. The data processing module is used to calculate the carbon isotope characteristic differences; The judgment and analysis module is used to compare the carbon isotope feature difference with a preset statistical judgment threshold. When the carbon isotope feature difference is greater than the statistical judgment threshold, it is determined that there is an exogenous adulterant in the dairy product to be tested; when the carbon isotope feature difference is less than or equal to the statistical judgment threshold, it is determined that there is no exogenous adulterant in the dairy product to be tested. The quantitative calculation module is used to calculate the adulteration ratio of the exogenous adulterant by substituting the carbon isotope characteristic difference into a pre-constructed linear regression model when the presence of exogenous adulterant is determined. The statistical judgment threshold is determined based on the statistical distribution of carbon isotope characteristic difference in a pure dairy product sample set, and the linear regression model is constructed based on the linear relationship between the carbon isotope characteristic difference of gradient-spike samples and the adulteration ratio.

9. The detection system for exogenous adulterants in dairy products as described in claim 8, characterized in that, The sample pretreatment module includes: The pH adjustment unit is used to adjust the pH value of the dairy product sample to be tested to 4.4-4.8; The first centrifugal separation unit is used to separate casein precipitate and supernatant by centrifuging at 4000r / min-4500r / min for 8-12 minutes at 3℃-5℃. The protein purification unit is used to add a 1:1 volume ratio ethanol-ether mixture to the casein precipitate for extraction. After vortexing, the upper organic phase is discarded by centrifugation. The extraction operation is repeated 2-3 times. The washed protein precipitate is then freeze-dried under vacuum at -40°C to -60°C for 20-28 hours to obtain milk protein powder. A lactose crystallization unit is used to collect the supernatant, add 3 to 5 times the volume of anhydrous ethanol, and let it stand at 3℃-5℃ for 10 to 14 hours to allow lactose crystals to precipitate. The second centrifugal separation unit is used to collect lactose crystals by centrifugation, wash them 2-4 times with anhydrous ethanol, and dry them to obtain lactose powder.

10. The detection system for exogenous adulterants in dairy products as described in claim 8, characterized in that, The coefficient of determination R² of the linear regression model is greater than or equal to 0.99, and the linear regression model is independently established by running gradient-spike samples for a specific exogenous plant protein or a specific batch of tests.

Images