Human insulin purity standard substance, its preparation method and application

Abstract

The application discloses a preparation method of a human insulin purity standard substance, which comprises the following steps of raw material selection, qualitative identification, standard substance uniformity, stability test, value determination and uncertainty evaluation. The application uses commercially available high-purity human insulin as a raw material, and after purity analysis, qualitative identification, uniformity test and stability test, isotope-labeled Val, Leu and Phe are used as internal standards, and liquid chromatography-isotope dilution mass spectrometry is used for value determination, so that the interference and influence of possible isoleucine substitution leucine impurities, isomer impurities or other molecular weight protein impurities on the value determination result can be effectively excluded, and the value determination result is more accurate and reliable. The purity standard substance prepared by the application can be used for serum insulin standard substance quality value tracing, quality control of insulin measurement results in in-vitro diagnosis, related method confirmation and evaluation, and evaluation of repeatability, linearity and cross contamination rate of a closed light-emitting immune analyzer.

Description

Technical Field

[0001] This invention belongs to the field of metrology technology, specifically relating to a human insulin purity standard substance, its preparation method, and its application. Background Technology

[0002] Human insulin (hINS) is a protein hormone secreted by pancreatic β-cells in response to stimulation by endogenous or exogenous substances (such as glucose, lactose, ribose, arginine, glucagon, etc.). Its CAS number is 11061-68-0, and its molecular formula is C257H383N65O77S6. Based on the molecular formula, the theoretical monoisotope molecular weight is calculated to be 5803.6375188, and the theoretical average molecular weight is 5807.57882. Human insulin consists of two peptide chains, A and B, with the following amino acid sequence:

[0003] A chain: GIVEQCCTSICSLYQLENYCN;

[0004] B chain: FVNQHLCGSHLVEALYLVCGERGFFYTPKT.

[0005] The physiological functions of insulin mainly involve regulating glucose metabolism, lipid metabolism, and protein metabolism. Among these, insulin's regulation of glucose metabolism is most closely related to diabetes. Insulin promotes the uptake and utilization of glucose by cells throughout the body and inhibits glycogenolysis and gluconeogenesis. It is also the only hormone in the body that can lower blood glucose. Insufficient insulin secretion or lack of insulin receptors often leads to elevated blood glucose. Clinically, insulin testing is mainly used for the diagnosis of diabetes, the observation and follow-up of the disease's condition and treatment effects, and the diagnosis and etiological research of chronic complications of diabetes mellitus (DM) and insulin resistance syndrome (IR).

[0006] The existing GBW09816 insulin (pig) component standard reference material is mainly used for traceability, quality control, method validation and evaluation of porcine insulin content analysis results in drug analysis, and its scope of application is relatively narrow. According to JJF 1752-2019 "Calibration Specification for Fully Automated Closed-Type Chemiluminescence Immunoassay Analyzer", calibration of repeatability, linearity, and cross-contamination rate can be performed using high-purity certified reference materials prepared by gravimetric-volume method. Certified reference materials such as human insulin, growth hormone, or alpha-fetoprotein must be used, and the relative expanded uncertainty should not exceed 5% (k=2). Based on the above uses, the GBW09816 insulin (pig) component standard reference material is no longer applicable. Therefore, it is necessary to develop an accurate, reliable, and stable insulin (human) purity standard reference material for traceability of serum insulin standard values, quality control of insulin measurement results in in vitro diagnostics, and related method validation and evaluation. Summary of the Invention

[0007] The purpose of this invention is to provide a human insulin purity standard substance to solve at least one of the above-mentioned technical problems.

[0008] Another object of the present invention is to provide a human insulin purity standard substance prepared by the above preparation method, so as to solve at least one of the above technical problems.

[0009] Another object of the present invention is to provide the human insulin purity standard material prepared by the above preparation method as a purity standard material for application, so as to solve at least one of the above technical problems.

[0010] According to one aspect of the present invention, a method for preparing a human insulin purity standard substance is provided, comprising the following steps:

[0011] (1) Select human insulin with a nominal purity of not less than 98%, perform purity analysis, and then use high performance liquid chromatography for initial homogeneity testing. Use human insulin that passes the initial homogeneity test as a candidate for human insulin purity standard material, and then package and store it.

[0012] (2) Qualitative identification, homogeneity testing and stability study of candidate human insulin purity standard substances;

[0013] (3) If the structure, homogeneity and stability of the human insulin purity standard material candidate meet the metrological requirements, then quantify the A21 deamino human insulin in the human insulin purity standard material candidate and determine the purity ratio factors of phenylalanine, valine and leucine in the main components of the human insulin purity standard material candidate.

[0014] (4) Hydrolyze the candidate human insulin purity standard substance, and then use isotopically labeled phenylalanine, valine and leucine as internal standards to determine the value of the candidate human insulin purity standard substance by liquid chromatography-isotope dilution mass spectrometry.

[0015] (5) Uncertainty assessment of the candidate human insulin purity standard substance is performed to obtain the human insulin purity standard substance.

[0016] This invention uses commercially available high-purity human insulin as raw material. After purity analysis, qualitative identification, homogeneity testing, and stability investigation, isotope-labeled phenylalanine, valine, and leucine are used as internal standards. Liquid chromatography-isotope dilution mass spectrometry is used for determination, which can effectively eliminate the interference and influence of impurities such as isoleucine replacing leucine, isomer impurities, or other protein impurities of the same molecular weight on the determination results, making the determination results of human insulin purity standard material more accurate and reliable.

[0017] In some embodiments, the qualitative identification of human insulin purity standard material candidates can be selected from at least one of mass spectrometry identification, molecular weight determination, amino acid sequence determination, compositional analysis of leucine and isoleucine, ultraviolet spectroscopy characterization, and infrared spectroscopy characterization.

[0018] In some embodiments, qualitative identification of candidate human insulin purity standards may include: mass spectrometry, molecular weight determination, amino acid sequencing, compositional analysis of leucine and isoleucine, ultraviolet spectroscopy, and infrared spectroscopy. This effectively confirms the presence of impurities such as isoleucine-substituted leucine, isoform impurities, and other protein impurities of the same molecular weight.

[0019] In some implementations, stability testing may include: long-term stability testing, short-term stability testing, and open-bottle stability testing.

[0020] In some implementations, long-term stability studies may include long-term stability studies of principal components and long-term stability studies of moisture content.

[0021] In some implementations, short-term stability testing may include short-term stability tests under three different storage conditions: 4°C, 25°C, and 40°C.

[0022] In some implementations, the open-bottle stability test may include open-bottle stability testing at -20°C and open-bottle stability testing at room temperature.

[0023] In some implementations, when conducting stability studies on candidate human insulin purity standards, the purity of the candidate human insulin purity standard can be determined using a mass balance method, wherein the purity of human insulin is calculated according to the following formula:

[0024] w = P × (1 - ABCD)

[0025] In the formula, w is the purity of the human insulin purity standard candidate, P is the purity of the human insulin purity standard candidate determined by reversed-phase high-performance liquid chromatography, A is the moisture content, B is the total content of ash and anions, C is the residual solvent content, and D is the aggregate content.

[0026] Aggregates do not possess the relevant physiological and chemical functions of insulin and cannot be considered as effective components. Therefore, when using the mass balance method to determine the purity of human insulin purity standard material candidates, the content of aggregates should be deducted for more accurate and reliable purity determination results.

[0027] In some embodiments, when performing homogeneity testing on human insulin purity standard material candidates, the method for determining the purity of human insulin purity standard material candidates may include the following steps: hydrolyzing the human insulin purity standard material candidates, and then using isotopically labeled phenylalanine, valine, and leucine as internal standards, and determining the purity of the human insulin purity standard material candidates by liquid chromatography-isotope dilution mass spectrometry.

[0028] In some embodiments, when determining the purity of human insulin purity standard candidate materials and assigning values ​​to human insulin purity standard candidate materials using liquid chromatography-isotope dilution mass spectrometry, the isotope-labeled internal standards of leucine, phenylalanine, and valine can be D, respectively. 10 -Leucine, 13 C9-phenylalanine and 13 C5-valine. Using three amino acids as internal standards to determine the purity of candidate standards for human insulin or to assign values ​​to them can effectively eliminate the interference and influence of impurities such as isoleucine-substituted leucine, isomer impurities, and other protein impurities of the same molecular weight on the determination and assignment results, making the determination and assignment results more accurate and reliable.

[0029] In some implementations, hydrolyzing human insulin purity standard candidates may include the following steps:

[0030] (1) Prepare a 1 mg / g human insulin solution by using 0.01 mol / L hydrochloric acid to prepare a human insulin purity standard candidate.

[0031] (2) Take human insulin solution, remove water, add 6mol / L hydrochloric acid at a volume ratio of 1:5 (human insulin solution to 6mol / L hydrochloric acid), remove oxygen, and hydrolyze for 60-65 hours under sealed conditions at a temperature of 109.5-110.5℃.

[0032] In some embodiments, in the liquid chromatography-isotope dilution mass spectrometry method, the detection conditions of the liquid chromatography may include: isocratic elution using a mobile phase consisting of 0.1% trifluoroacetic acid, 9% acetonitrile, and the balance water (by volume percentage) for 10 min, and using a Phenomenex kinetex 2.6 μm C column. 18 , 150mm×2.1mm.

[0033] In some implementations, the detection conditions of the mass spectrometer in liquid chromatography-isotope dilution mass spectrometry may include: a collision energy of 14V, an exit voltage of 10V, an inlet voltage of 10V, and a declustering voltage of 50V.

[0034] In some embodiments, the uncertainty may include uncertainty introduced by the setting process, uncertainty introduced by uniformity, uncertainty introduced by short-term stability and long-term stability, wherein the uncertainty introduced by the setting process includes uncertainty introduced by amino acid purity, uncertainty introduced by human insulin hydrolysis efficiency, uncertainty introduced by amino acid purity ratio factor, uncertainty introduced by weighing, uncertainty introduced by mass ratio, uncertainty introduced by peak area ratio and uncertainty introduced by human insulin molecular weight.

[0035] According to another aspect of the present invention, a human insulin purity standard substance prepared by the preparation method provided by the present invention is provided.

[0036] According to another aspect of the present invention, the application of the human insulin purity standard material prepared by the method of the present invention as a purity standard material is provided. The human insulin purity standard material prepared by the present invention has accurate and stable values ​​and can be used as a purity standard material for traceability of serum insulin standard values, quality control of insulin measurement results in in vitro diagnostics, confirmation and evaluation of related methods, and evaluation of repeatability, linearity, and cross-contamination rate of closed-type chemiluminescence immunoassay analyzers. For example, it can be used as a purity standard material for calibration of repeatability, linearity, and cross-contamination rate in JJF1752-2019 "Calibration Specification for Fully Automated Closed-Type Chemiluminescence Immunoassay Analyzers". In use, an appropriate amount of the solid human insulin standard material is weighed and dissolved in a phosphate buffer solution containing BSA, and then the solution is transferred to a volumetric flask and diluted to volume to obtain human insulin solution standard materials of different concentrations. Results show that the standard material prepared by the present invention has good performance, accurate and reliable values, meets the requirements of JJF 1752-2019 "Calibration Specification for Fully Automated Closed-Type Chemiluminescence Immunoassay Analyzers", and can support the calibration work of chemiluminescence immunoassay analyzers and the smooth implementation of insulin detection projects. Attached Figure Description

[0037] Figure 1 For linearity evaluation of reversed-phase high-performance liquid chromatography;

[0038] Figure 2 Reversed-phase chromatography of candidate standards for human insulin purity;

[0039] Figure 3 Mass spectrum for identifying candidate standards for human insulin purity;

[0040] Figure 4 Mass spectra of candidate human insulin purity standard substances for molecular weight determination;

[0041] Figure 5 A-chain mass spectrometry sequencing results of candidate human insulin purity standard substances;

[0042] Figure 6B-chain mass spectrometry sequencing results of candidate human insulin purity standard substances;

[0043] Figure 7 For high-performance liquid chromatography-circular dichroism spectroscopy analysis of Leu and Ile;

[0044] Figure 8 Ultraviolet absorption spectra of candidate human insulin purity standard substances;

[0045] Figure 9 Infrared absorption spectra of candidate human insulin purity standard substances;

[0046] Figure 10 Image of human insulin and its impurities obtained by matrix-assisted laser-induced desorption / desorption-time-of-flight mass spectrometry;

[0047] Figure 11 Mass spectrometry identification of A21 deamination insulin;

[0048] Figure 12 QQ graph showing the value determination data of candidate standards for human insulin purity. Detailed Implementation

[0049] The present invention will be further described in detail below with reference to specific embodiments. These embodiments are for illustrative purposes only and do not limit the invention in any way. Unless otherwise specified, the raw materials and reagents used in the embodiments are conventional products that can be obtained commercially; experimental methods not specified in the embodiments are generally performed under conventional conditions in the art or as recommended by the manufacturer.

[0050] The liquid chromatography-isotope dilution mass spectrometry (HPLC-IDMS) method used in the homogeneity testing and value determination process of this invention is described in detail below:

[0051] 1.1 Method Principles

[0052] This invention hydrolyzes human insulin into amino acids in the presence of 6 mol / L hydrochloric acid. The amino acid content in the hydrolysate is determined by isotope dilution mass spectrometry, and after impurity correction, the purity of human insulin is calculated based on its amino acid composition. By using national standard reference materials for amino acids and measurement methods and instruments that meet metrological requirements, traceability is ensured, and the final value can be traced back to SI units.

[0053] 1.2 Experimental Apparatus

[0054] Includes: High-performance liquid chromatography-triple quadrupole tandem mass spectrometry (LC / MS / MS) systems, including a 1260 HPLC system (Agilent Technologies, USA) and a QTRAP 5500 triple quadrupole mass spectrometer (AB Laboratories, USA); High-performance liquid chromatography-triple quadrupole tandem mass spectrometry (LC / MS / MS) systems, including a 1290 Infinity II HPLC system (Agilent Technologies, USA) and a 6470C triple quadrupole mass spectrometer (Agilent Technologies, USA); Balances (CPA225D, sensitivity 0.01 mg, Sartorius, Germany); Balances (XP56, sensitivity 0.001 mg, Mettler Toledo, Switzerland); Balances (UMX5, sensitivity 0.0001 mg, Mettler Toledo, Switzerland); etc.

[0055] 1.3 Standard substances and related reagents

[0056] Includes: Leucine (GBW09237, Leu, 99.7% ± 0.4%) National Standard Reference Material (National Institute of Metrology, China); Phenylalanine (GBW09235, Phe, 99.8% ± 0.4%) National Standard Reference Material (National Institute of Metrology, China); Valine (GBW09236, Val, 99.4% ± 0.6%) National Standard Reference Material (National Institute of Metrology, China); Insulin (pig) Component Standard Reference Material (GBW09816, 0.892 ± 0.040) g / g, National Standard Reference Material (National Institute of Metrology, China); Labeled Amino Acids (D... 10 -Leucine (isotope-labeled purity >98%) 13 C9-phenylalanine (isotope labeling purity >99%) 13 C5-valine (isotope-labeled purity >99%), all purchased from Cambridge Isotope Laboratory, USA; etc.

[0057] 1.4 Establishment and optimization of isotope dilution mass spectrometry method

[0058] 1.4.1 Hydrolysis of human insulin samples

[0059] Approximately 1 mg of human insulin sample was weighed using a balance with a minimum division of 0.001 mg. It was dissolved in 1 mL of 0.01 mol / L hydrochloric acid, and the mass was accurately measured to prepare a 1 mg / g human insulin solution. 100 μL of the stock solution was added to a 2 mL ampoule, followed by 100 μL of labeled amino acid standard solution, and the mass of each ampoule was accurately measured. The ampoule was then centrifuged for 70 min in a vacuum centrifuge until all moisture was removed. 500 μL of 6 mol / L hydrochloric acid was added, and the sample was deoxygenated by purging with nitrogen for 2 min. The ampoule was then sealed and hydrolyzed in an oven at (110.0 ± 0.5) °C. After hydrolysis, the sample was dried under nitrogen, reconstituted with 500 μL of water (containing 0.1% formic acid), filtered through a 0.22 μm filter, and then analyzed.

[0060] 1.4.2 HPLC-IDMS Analysis

[0061] Chromatographic conditions: mobile phase: 91% water, 9% acetonitrile, 0.1% TFA; gradient: isocratic elution, 10 min; column: Phenomenex kinetex 2.6 μm C18 150 mm × 2.1 mm.

[0062] Mass spectrometry conditions: Ion-pair mass-to-charge ratio: Valine: 118→72 (Val) and 123→76 (labeled Val); Phenylalanine: 166→120 (Phe) and 174→128 (labeled Phe); Leucine: 132→86 (Leu) and 142→96 (labeled Leu).

[0063] 1.4.3 Optimization of hydrolysis time

[0064] To ensure the quantitative hydrolysis of proteins into amino acids, the hydrolysis time needs to be optimized, selecting the peak or plateau phase of the hydrolysis time curve as the appropriate hydrolysis time. The relative proportions of Val, Leu, and Phe in the hydrolysate were investigated at hydrolysis times ranging from 16 h to 96 h. The results showed that the relative proportions of the three amino acids reached a plateau at a hydrolysis time of 63 h; therefore, 63 h was selected as the optimal hydrolysis time for human insulin.

[0065] 1.4.4 Optimization of mass spectrometry detection conditions

[0066] The collision energy (CE), exit voltage (CXP), inlet voltage (EP), and declustering voltage (DP) of valine (118→72, Val) and phenylalanine (166→120, Phe) and leucine (174→128, Phe) were optimized. The optimization ranges for collision energy (0–180 V), exit voltage (0–55 V), inlet voltage (0–15 V), and declustering voltage (0–300 V) were determined using the instrument's automatic optimization function. Based on the final optimization results, the selected values ​​were: collision energy (CE) of 14 V, exit voltage (CXP) of 10 V, inlet voltage (EP) of 10 V, and declustering voltage (DP) of 50 V.

[0067] 1.4.5 Confirmation of Isotope Internal Standards

[0068] As internal standards for isotope dilution mass spectrometry, unlabeled compounds should not be mixed in, otherwise the accuracy of quantification will be affected. To confirm that the isotope-labeled phenylalanine, valine, and leucine used for quantification do not contain unlabeled amino acids, stock solutions of isotope-labeled amino acids were prepared separately and preliminarily analyzed by LC-MS. The results showed that none of the isotope-labeled amino acids used as internal standards contained unlabeled amino acids.

[0069] 1.5 Methodological parameters of isotope dilution mass spectrometry

[0070] The methodological parameters of the optimized isotope dilution mass spectrometry method were investigated.

[0071] 1.5.1 Correctness of the method

[0072] The accuracy of the isotope dilution mass spectrometry method was verified using GBW09816, a national primary standard reference for porcine insulin composition. Using the national amino acid standard reference as the standard, the isotope dilution mass spectrometry method based on amino acid analysis was employed to determine the content of this standard reference reference. Three subsamples were taken, and the results are shown in Table 1.

[0073] Table 1. Verification results of GBW09816 porcine insulin component content national primary standard reference material.

[0074]

[0075] The average values ​​of the measurement results are all within the range of the standard reference values, therefore the isotope dilution mass spectrometry method is reliable.

[0076] 1.5.2 Repeatability and Reproducibility of the Method

[0077] The insulin content of nine units was determined by HPLC-IDMS, with each unit analyzed three times. The results are shown in Table 2.

[0078] Table 2. HPLC-IDMS determination results of human insulin content (g / g)

[0079]

[0080]

[0081] The optimized HPLC-IDMS method was used to perform five repeated determinations on the same hINS sample. The average value of the results was 0.861 g / g, and the RSD was 0.3%, indicating that the established method has good repeatability.

[0082] The optimized HPLC-IDMS method was used to measure 11 hINS samples three times each. The average value of the results was 0.872 g / g, and the RSD was 0.5%, indicating that the established method has good reproducibility.

[0083] 1.5.3 Evaluation of Limit of Detection and Limit of Quantification

[0084] Calculate the limit of detection and limit of quantitation for the HPLC-IDMS method according to Formula 1-1:

[0085]

[0086] In the formula, R m-STD R represents the mass ratio of amino acids to isotopically labeled amino acids in the standard solution. h-STD The peak height ratio of amino acids in the standard solution to that of isotopically labeled amino acids; M sample The mass (g) of isotope-labeled amino acids in the sample solution; m sample R represents the mass (g) of amino acids in the sample solution. h-sample This represents the ratio of the noise signal intensity of 3 or 10 times in the unlabeled amino acid channel to the corresponding labeled amino acid signal intensity. The calculated limit of detection and limit of quantitation for hINS are 2.7 × 10⁻⁶. -6 g / g and 9×10 -6 g / g.

[0087] The mass balance method (MB method) used in the stability test of this invention is described in detail below:

[0088] 2.1 Method Principles

[0089] In this invention, the purity of human insulin is calculated according to formula (2-1):

[0090] w = P × (1 - ABCD) (2-1)

[0091] In the formula, w is the purity of the human insulin purity standard candidate, P is the purity of the human insulin purity standard candidate determined by reversed-phase high-performance liquid chromatography, A is the moisture content, B is the total content of ash and anions, C is the residual solvent content, and D is the aggregate content.

[0092] 2.2 Experimental Apparatus

[0093] This includes: Agilent 1200 HPLC (Agilent Technologies, USA); Agilent 1290 HPLC (Agilent Technologies, USA); Karl Fischer moisture analyzer (DL39, Mettler Toledo, Germany); Karl Fischer moisture analyzer (C20, Mettler Toledo, Germany); Simultaneous thermal analyzer (SDT650, TA Instruments, USA); muffle furnace (FP31, Yamato, Japan); gas chromatography-mass spectrometry (ISQ, Thermo Scientific, USA); and so on.

[0094] 2.3 Related Reagents

[0095] Including: anhydrous sodium sulfate (Sinopharm Chemical Reagent Co., Ltd., analytical grade); acetonitrile (JT Baker, USA, purity); phosphoric acid (Alfa Aesar, Alfa Aesar (China) Chemical Co., Ltd.); hydrochloric acid (Zhejiang Shangneng Industrial Co., Ltd.); perfluoroheptanoic acid (PHFA, Sigma-Aldirich, USA); trifluoroacetic acid (TFA, DikmaPure, China); etc.

[0096] 2.4 Determination of the content of principal components in hINS

[0097] 2.4.1 Optimization of purity determination conditions by reversed-phase high-performance liquid chromatography

[0098] In order to clearly determine the separation degree between the main component to be tested and impurities, the conditions of the high performance liquid chromatography-area normalization method in the mass balance method, such as the column, the concentration of sodium sulfate in the mobile phase, the sample concentration and the detection wavelength, were optimized. The optimization results showed that: (1) the Vydac C8 (250mm×4.6mm) column can fully separate insulin, A21 deaminoinsulin and other impurities, while the Agilent SB-Aq (250mm×4.6mm) column cannot fully separate insulin and A21 deaminoinsulin. Therefore, the Vydac column was selected. (1) A C8 (250 mm × 4.6 mm) chromatographic column was used in subsequent experiments; (2) The concentration of sodium sulfate has a significant effect on the retention time and resolution of insulin. As the concentration of sodium sulfate increases, the retention time of the insulin peak gradually decreases. When the concentration of sodium sulfate is 0.2 mol / L, the separation of insulin, A21 deaminoinsulin and impurities is the best; (3) The maximum absorption wavelengths of insulin, A21 deaminoinsulin and impurities at 48 min are all between 210 nm and 240 nm. When the wavelength is 214 nm, the impurity peak can be fully detected; (4) The more suitable detection concentration of the sample is 1 mg / mL (1000 mg / L).

[0099] The final optimized reversed-phase high-performance liquid chromatography (RP-HPLC) conditions are as follows: 1) Sample concentration: 1 mg / mL; 2) Injection volume: 20 μL; 3) Mobile phase: A: 0.2 mol / L sodium sulfate buffer: acetonitrile (82:18); B: acetonitrile: water (50:50); 4) Column: Vydac C8, 250 mm × 4.6 mm; 5) Detection wavelength: 214 nm; 6) Flow rate: 1 mL / min; 7) Column temperature: 40°C; 8) Chromatographic concentration gradient: as shown in Table 3.

[0100] Table 3 Mobile phase gradient for reversed-phase high-performance liquid chromatography

[0101] Time / min A％ B％ 0 78 22 36 78 22 61 33 67 67 33 67

[0102] 2.4.2 Methodological parameters for purity determination by reversed-phase high-performance liquid chromatography

[0103] Under the final optimized conditions, five parallel samples were taken, and each sample was analyzed three times. The average value of the results was 0.9840, and the RSD was 0.19%. This indicates that the reversed-phase high-performance liquid chromatography purity analysis method has good repeatability and can meet the requirements of mass balance determination and homogeneity testing.

[0104] Human insulin solutions with concentrations of 2.128 mg / g, 1.064 mg / g, 0.532 mg / g, 0.266 mg / g, 0.133 mg / g, 0.0665 mg / g, 0.0332 mg / g, 0.0166 mg / g, and 0.00831 mg / g were prepared. The prepared samples were analyzed using an optimized reversed-phase high-performance liquid chromatography (RP-HPLC) method. Linear fitting was performed on the peak areas and the concentrations of the responding samples. The results are as follows: Figure 1 As shown, the linear correlation coefficient r = 0.9996 was calculated, indicating that the method has good linearity.

[0105] Substituting the peak areas of 3 times and 10 times the noise levels into the standard curve obtained in section 2.4.2, the limits of detection and quantitation for insulin determination by reversed-phase high-performance liquid chromatography were found to be 6.4 × 10⁻⁶. -4 mg / g and 1.9×10 -3 mg / g.

[0106] Determination of aggregates in 2.5hINS

[0107] High-performance gel size exclusion liquid chromatography (HPLC) was used to determine aggregates that may be present in hINS samples. The chromatographic conditions of HPLC, including column, detection wavelength, mobile phase, and sample concentration, were optimized. The TSKgel G2000SWxl (7.8 mm × 30 cm, 5 μm), TSKgel SuperSW3000 (4.6 mm × 30 cm, 4 μm), and Yarra 3u columns were compared sequentially. The effects of the SEC-2000 (300×7.8mm) column on the separation efficiency of the system were investigated; the effects of using PBS buffer solution (pH=7.4), pure water, and water:acetonitrile:TFA = 70:30:0.1% (v:v:v) as mobile phases on the separation efficiency of the system were investigated; the effects of detection wavelengths of 214nm, 230nm, 254nm, and 280nm on the detection results of hINS aggregates were investigated; and the effects of sample concentrations of 3.5mg / g, 1.75mg / g, 0.875mg / g, 0.438mg / g, 0.219mg / g, 0.109mg / g, and 0.0545mg / g on the detection results of hINS aggregates were investigated. The finally optimized high-performance gel size exclusion liquid chromatography (HPLC) conditions for the determination of hINS aggregates were determined as follows: 1) Injection volume: 10μL; 2) Column: TSKGel 2000SWxl gel size exclusion column; 3) Gradient: isocratic elution; 4) Flow rate: 0.5 mL / min; 5) Analysis time: 35 min; 6) Detection wavelength: 214 nm; 7) Sample concentration: 3.5 mg / mL.

[0108] Under optimal conditions, three measurements of hINS aggregates were performed, yielding results of 0.540%, 0.531%, and 0.458%, respectively, with an average of 0.510%.

[0109] Determination of moisture content in 2.6hINS

[0110] The moisture content of chemical reagents was determined according to GB / T 606-2003, "General Method for Determination of Moisture in Chemical Reagents - Karl Fischer Method". Since each sample dispensing unit contained only 3 mg of human insulin, which was insufficient to meet the requirements of the Karl Fischer method, moisture was measured after sample pooling. Each pair of samples yielded approximately 6 mg of sample. Therefore, approximately 6 mg of human insulin sample was accurately weighed each time and measured using a Karl Fischer moisture analyzer, with six parallel measurements. To eliminate the influence of atmospheric moisture on the results, the sample addition process, mass, and time were simulated, but the measurement was performed without adding any sample. The blank value of atmospheric moisture was calculated, and the final result was obtained after subtracting the blank value. The results are shown in Table 4.

[0111] Table 4. Moisture content of human insulin purity standard material determined by Karl Fischer method.

[0112]

[0113] Ash content determination in 2.7hINS

[0114] The ash content was determined according to the "General Rules and Guidelines" of Part III of the 2020 edition of the Chinese Pharmacopoeia, specifically the "0841 Residue on Ignition Test Method". Since the sample volume of a single packaging unit was small, two units were combined for ash content determination. First, a clean crucible was ignited in a muffle furnace until constant weight was achieved. After constant weight, approximately 5 mg of accurately weighed human insulin sample was added and precisely weighed. The sample was then ignited at 700–800℃ until completely ashed. It was then transferred to a desiccator, cooled, and precisely weighed again. It was then ignited again at 700–800℃ until constant weight was achieved. The ash content was calculated based on the residual mass after ignition. The determination was performed in triplicate, and the results are shown in Table 5.

[0115] Table 5 Ash content (%) of human insulin purity standard material determined by ignition method

[0116]

[0117] Due to the high ash content, the composition of ash impurities was verified by testing common inorganic elements. The method used was inductively coupled plasma mass spectrometry (ICP-MS). Based on the test results, the inorganic element content of the human insulin purity standard residue was calculated to be 1.57%. The most abundant common inorganic elements were Zn, Al, Fe, Na, and Mg, with a total content of 1.43%. Standard solutions of Zn, Al, Fe, Na, and Mg were prepared with concentrations of 0.1, 1.0, 2.0, 3.0, and 4.0 μg / L, respectively. The five elements were quantified using the ICP-MS external standard method, and the results were: Zn: 0.51%, Al: 0.30%, Fe: 0.18%, Na: 0.17%, Mg: 0.11%, with a total content of 1.27%. Since ICP-MS can only detect cations in the sample, while the muffle furnace calcination method can detect all inorganic impurities, the inorganic impurity content of human insulin was finally calculated based on the results of the muffle furnace calcination method.

[0118] Determination of anions in 2.8hINS

[0119] 2.8.1 Experimental Procedure

[0120] Ion chromatography was used to determine the anions that may be present in human insulin, such as phosphate, sulfate, nitrite, trifluoroacetate, carbonate, chloride, and fluoride ions. Specific experimental conditions are shown in Table 6. Quantification was performed using a standard curve method, with analytical grade chemical reagents used as standards.

[0121] Table 6 Ion Chromatography Analysis Conditions

[0122]

[0123]

[0124] 2.8.2 Experimental Results

[0125] The methodological parameters and measurement results are shown in Table 7.

[0126] Table 7. Results of anion concentration analysis in human insulin.

[0127] Repeatability linear LOD Concentration in the sample phosphate 2.93％ 0.999 0.0038 mg / g N / A Fluoride ions 1.45％ 0.997 0.0008mg / g N / A Nitrite 1.05％ 0.997 0.0005mg / g N / A chloride ions 2.31％ 0.999 0.0039 mg / g <![CDATA[4.93×10 -4 g / g<!-- 9 --> ]]> sulfate 2.94％ 0.999 0.0010 mg / g N / A Trifluoroacetate 2.89％ 0.999 0.028mg / g N / A carbonate 2.40％ 0.999 0.0010 mg / g N / A

[0128] Therefore, the calculated anion content in the sample is 0.049%.

[0129] Determination of volatile organic compounds in 2.9hINS

[0130] 2.9.1 Experimental Procedure

[0131] Volatile organic components that may be present in human insulin were determined using headspace sampling gas chromatography-mass spectrometry (GC-MS). Analytical chemical reagents were used as standards, and quantification was performed using the external standard method. Specific experimental conditions were as follows: 1) Gas chromatographic column: Agilent DB-17MS 30m × 0.250mm; 2) Carrier gas: Helium; 3) Carrier gas flow rate: 1.0 mL / min; 4) Injection volume: 1000 μL; 5) Split ratio: 1:30; 6) Temperature program: initial temperature 30℃ held for 10 min, increased to 140℃ at a rate of 7℃ / min, held for 2 min; then increased to 220℃ at a rate of 15℃ / min, held for 10 min; 7) Transfer line temperature: 240℃.

[0132] Headspace gas chromatography was used to test human insulin dissolved in DMSO and water, respectively.

[0133] 2.9.2 Experimental Results

[0134] First, DMSO was selected as the solvent, and headspace analysis was performed by gas chromatography-mass spectrometry. The results showed that no other solvents were detected except for DMSO.

[0135] Subsequently, 3 mg of human insulin was weighed and diluted to a 5 mL sample tube with DMSO. Gas chromatography-mass spectrometry analysis was performed, and no other solvents were detected except for DMSO.

[0136] Meanwhile, to eliminate residual DMSO solvent and to prepare human insulin of the same concentration, 0.01 mol / L HCl was added to increase the stability of the solution. First, the solvent was analyzed by headspace mass spectrometry with a 0.01 mol / L HCl aqueous solution. Due to the small molecular weights of H2O and HCl, the remaining solvents were not detected.

[0137] Finally, headspace gas analysis of human insulin in 0.01 mol / L HCl was performed, and no solvent residue was detected. Quantification was performed using analytical-grade chemical reagents via external standard method. Methodological parameters and results are shown in Table 8. The results indicate that none of the 30 solvents listed in the table were detected in the human insulin sample.

[0138] Table 8. Results of volatile organic compound analysis in human insulin.

[0139]

[0140]

[0141] Finally, the test results for moisture content, ash content, anion content, solvent residue content, and aggregate content are shown in Table 9.

[0142] Table 9. Quantitative results of each impurity component

[0143] Impurities content / % Moisture 10.8 Ash 2.1 anions 0.049 Solvent residue 0 Aggregates 0.51

[0144] Example 1

[0145] 1. Raw material selection

[0146] In this embodiment, the human insulin raw material used is a commercially available high-purity human insulin purification biochemical reagent (CAS No.: 11061-68-0) from a bioproducts company.

[0147] The purity of the purchased human insulin raw materials in this batch was analyzed by reversed-phase high-performance liquid chromatography. The chromatographic conditions were as follows: the injection concentration was prepared at 1 mg / mL, the injection volume was 20 μL, the mobile phase A was 0.2 mol / L sodium sulfate buffer: acetonitrile (82:18), the mobile phase B was acetonitrile: water (50:50), the chromatographic column was Vydac C8, the detection wavelength was 214 nm, the flow rate was 1 mL / min, and the column temperature was 40 ℃.

[0148] The results are as follows Figure 2 As shown in the figure. The results indicate that the normalized peak area of ​​the human insulin raw material used in this embodiment is approximately 98.4%, indicating a high purity value.

[0149] 2. Initial inspection of uniformity

[0150] High-performance liquid chromatography (HPLC) was used for preliminary homogeneity testing. Samples were randomly taken from the top, middle, and bottom sections of the premixed human insulin raw material vials. The samples were dissolved in 0.01 mol / L hydrochloric acid to prepare a solution containing approximately 3.5 mg hINS per mL. After filtration through a 0.22 μm microporous membrane, the samples were injected. The mobile phase consisted of 78% mobile phase A (0.2 mol / L sodium sulfate buffer: acetonitrile (82:18)) and 22% mobile phase B (acetonitrile: water (50:50)). The flow rate was 1 mL / min, the column temperature was 40 °C, and the detection wavelength was 214 nm. The preliminary homogeneity test data are shown in Table 10.

[0151] Table 10 Initial data on the uniformity of insulin in humans

[0152]

[0153] As shown in Table 10, the standard deviation of the initial test for uniformity of human insulin raw materials was 0.01%, F = 3.09. α =5.14, F <F α This indicates that there is no significant inhomogeneity in the premixed human insulin raw material from different locations within the same batch, making it a candidate for human insulin purity standard material.

[0154] The candidate human insulin purity standard substance was aliquoted into brown glass tubes, 3 mg per tube. After aliquoting, the glass tubes were sealed and stored in a -20°C refrigerator.

[0155] 3. Qualitative identification of candidate human insulin purity standard substances

[0156] 3.1 Experimental Apparatus

[0157] These include: matrix-assisted laser-induced desorption / desorption-time-of-flight mass spectrometer (MALDI-TOF, Bruker, USA); 1290 ultra-high performance liquid chromatograph (HPLC, Agilent Technologies, USA); Spectrum BX Fourier transform infrared spectrometer (PE, PE, USA); CD-2095Plus circular dichroism chromatograph (JASCO, Japan); and so on.

[0158] 3.2 Related Reagents

[0159] Including: ammonium bicarbonate (Xilong Chemical Co., Ltd., analytical grade); dithiothreitol (DTT, SIGMA, ≥99.5%); etc.

[0160] 3.3 Mass Spectrometry Identification

[0161] 3.3.1 Experimental Procedure

[0162] Accurately weigh 1 mg of human insulin purity standard candidate, dissolve it in 0.01 mol / L hydrochloric acid to prepare a 1 mg / g human insulin purity standard candidate stock solution, and then dilute it to 10. -3 mg / g, dispensed in 6×100μL solutions. Dissolve 0.197g of ammonium bicarbonate solid in 45mL of pure water to prepare a 50mmol / L ammonium bicarbonate solution for later use. Dissolve 0.154g of dithiothreitol solid in 0.85mL of the 50mmol / L ammonium bicarbonate solution to prepare a 1mol / L dithiothreitol solution for reducing human insulin. Take 10μL of human insulin solution, add 5μL of dithiothreitol solution, reduce in a 60℃ water bath for 15min, desalt using a ZipTip pipette tip, then mix with CHCA matrix 1:1, and analyze using matrix-assisted laser desorption / ionization time-of-flight mass spectrometry.

[0163] 3.3.2 Experimental Results

[0164] Human insulin was reduced with DTT and then analyzed by MALDI-TOF. The results are as follows: Figure 3 As shown. From Figure 3 The spectrum shown clearly displays the A and B chains of human insulin, and the [M+H] junctions of the A and B chains. + The mass-to-charge ratios (z=1) are 2384.884 and 3431.950, respectively, which are consistent with the theoretical molecular weights of chains A and B (the average molecular weights of A and B are 2383.7015 and 3429.92496, respectively).

[0165] 3.4 Molecular weight determination

[0166] Human insulin was prepared into a 0.1 mg / g solution using 0.01 mol / L hydrochloric acid. A small amount was mixed with α-cyano-4-hydroxycinnamic acid (CHCA matrix) in the same proportion, and then 1 μL was spotted onto a TLC plate. After drying, the molecular weight was determined by mass spectrometry. The determination was performed in parallel eight times. The results are shown in Table 11.

[0167] Table 11 Results of Molecular Weight Determination

[0168]

[0169] The average molecular weight determination result was 5808.702, which is the result of [M+H]+, and is consistent with the molecular formula of human insulin, C0. 257 H 383 N 65 O 76 The S6 calculation results are consistent (the average molecular weight of human insulin is 5807.57882 Da). A typical mass spectrum for molecular weight determination is shown below. Figure 4 As shown.

[0170] 3.5 Sequencing

[0171] 3.5.1 MALDI-TOF mass spectrometry sequencing

[0172] First, the sequence was identified and analyzed using matrix-assisted laser-induced desorption / ionization time-of-flight mass spectrometry. 1 mg of the human insulin purity standard candidate was accurately weighed and dissolved in 0.01 mol / L hydrochloric acid to prepare a 1 mg / g human insulin purity standard candidate stock solution, which was then diluted to 10⁻⁶. -3 mg / g, dispensed in 6×100μL solutions. Dissolve 0.197g of ammonium bicarbonate solid in 45mL of pure water to prepare a 50mmol / L ammonium bicarbonate solution for later use. Dissolve 0.154g of dithiothreitol solid in 0.85mL of the 50mmol / L ammonium bicarbonate solution to prepare a 1mol / L dithiothreitol solution for reducing human insulin. Add 5μL of dithiothreitol solution to 10μL of human insulin solution, reduce in a 60℃ water bath for 15min, then desalt using a ZipTip pipette tip. Mix with CHCA matrix 1:1, and mount on a matrix-assisted laser desorption / ionization time-of-flight mass spectrometer (LIFT mode). Perform secondary mass spectrometry analysis on the molecular ion peaks of chains A and B to determine the amino acid sequences of chains A and B.

[0173] 3.5.1.1 Experimental results of hINS A-strand sequencing

[0174] hINS A-strand sequencing results are as follows Figure 5 As shown.

[0175] Sequencing of the A chain based on the b-ion series generated by peptide fragmentation yielded a partial sequence of SI(L)CSL(I)YQL(I)ENY; sequencing of the B chain based on the y-ion series generated by peptide fragmentation yielded a partial sequence of NQHL(I)CGSHL(I)VEAL(I)YL(I)VCGERGFF. The sequencing results perfectly match the theoretical sequences of the A and B chains from hINS (the theoretical sequence of the A chain is GIVEQCCTSICSLYQLENYCN, and the theoretical sequence of the B chain is FVNQHLCGSHLVEALYLVCGERGFFYTPKT).

[0176] Sequence alignment using De-novo sequencing revealed that the A-chain sequencing hits P30410.1 INS_PANTR, Q8HXV2.1 INS_PONPY, P01308.1 INS_HUMAN, P30406.1 INS_MACFA, and P30407.1 INS_CHLAE. A search for the complete insulin sequences of these species yielded the results shown in Table 12. The A-chain sequencing hits proteins from all five species, each containing only two peptide chains, A and B, with identical amino acid sequences for both chains, indicating they are the same compound. Therefore, the A-chain sequencing results confirm that the candidate standard material was identified as human insulin.

[0177] Table 12 Comparison of insulin sequences from five species hit by A-strand sequencing

[0178]

[0179]

[0180] 3.5.1.2 Experimental Results of hINS B-strand Sequencing

[0181] hINS B-strand sequencing results are as follows Figure 6 As shown.

[0182] Sequence alignment using De-novo sequencing revealed B-chain sequencing hits of P30410.1 INS_PANTR, Q8HXV2.1 INS_PONPY, P01308.1 INS_HUMAN, P30406.1 INS_MACFA, and P30407.1 INS_CHLAE. Searching for the complete insulin sequences of these species yielded results identical to those in Table 12. Further investigation revealed that the B-chain sequencing hits of proteins from all five species, each containing only two peptide chains, A and B, with identical amino acid sequences for both chains, indicating they are the same compound. Therefore, the B-chain sequencing results confirm the identification of the standard material candidate as human insulin. Combined with matrix-assisted laser-induced desorption / ionization time-of-flight mass spectrometry (MALAMS) sequencing of both peptide chains A and B, the standard material candidate is definitively identified as human insulin.

[0183] 3.5.2 Amino acid composition analysis of Leu and Ile in human insulin

[0184] Since mass spectrometry sequencing cannot distinguish between Leu and Ile, this invention analyzes the amino acid composition of Leu and Ile in candidate standard substances to determine whether Leu and Ile have substituted for each other. Human insulin was hydrolyzed to obtain an amino acid solution, and then the peaks of Leu and Ile in the hydrolysate were measured using high-performance liquid chromatography-circular dichroism spectroscopy. The results are as follows: Figure 7 As shown. The ratio of Leu to Ile was calculated using standard solutions prepared from national standard reference materials of Leu and Ile. The results of five measurements for the two samples were Leu / Ile = 3.07, 3.16, 3.15, 3.13, and 3.12, with an average value of 3.12:1, which is close to the theoretical amino acid composition of 3:1, proving that there is no substitution between Leu and Ile.

[0185] 3.5.3 Edman Degradation Method for N-Terminal Sequencing of Proteins

[0186] By commissioning a third-party laboratory, phenyl isothiocyanate (PITC) was coupled to the N-terminal α-amino group of a protein under weakly alkaline (TMA) conditions to generate phenylaminothiocarbamoyl peptide (PTC-protein). Then, under anhydrous strong acid (TFA) conditions, the first residue at the N-terminus was cleaved from the intact polypeptide chain as 2-phenylaminothiazolinone (ATZ-AA). Subsequently, under dilute acid (25% TFA) conditions, ATZ-AA was converted into a more stable phenylhydantoin N-urea derivative, namely PTH-amino acid. The generated PTH-AA was sent to high-performance liquid chromatography (HPLC) for online analysis. The remaining polypeptide samples could be repeatedly processed to generate various PTH-AAs. Separation using a liquid chromatography column could determine the N-terminal amino acid sequence of the tested polypeptide protein sample. Finally, the N-terminal sequence of the A-chain of the sample was analyzed using a Shimadzu fully automated protein and peptide sequencer (PPSQ-33A) to determine the N-terminal sequence of human insulin.

[0187] The sequencing results of all amino acids in the B chain of the human insulin purity standard candidate are: FVNQHLCGSHLVEALYLVCGERGFFYTPKT. This is consistent with the theoretical amino acid sequence of the insulin B chain (FVNQHLCGSHLVEALYLVCGERGFFYTPKT).

[0188] The sequencing results of all amino acids in the A chain of the human insulin purity standard candidate were: N-terminal GIVEQCCTSICSLYQLENYCN, which matches the standard amino acid sequence GIVEQCCTSICSLYQLENYCN of the A chain of standard human insulin.

[0189] 3.6 Ultraviolet Spectroscopic Characterization of Human Insulin

[0190] Accurately weigh 1 mg of human insulin purity standard candidate and dissolve it in 0.01 mol / L hydrochloric acid to prepare a 1 mg / g human insulin purity standard candidate stock solution. Perform high-performance liquid chromatography (HPLC) purity analysis under the following chromatographic conditions to obtain an online UV absorption spectrum from 210 nm to 400 nm. Specific chromatographic conditions are as follows: Injection volume: 20 μL; Mobile phase: A: 0.2 mol / L sodium sulfate buffer: acetonitrile (82:18); B: acetonitrile: water (50:50); Column: Vydac C8, 250 mm × 4.6 mm; Flow rate: 1 mL / min; Column temperature: 40 °C; Gradient settings are shown in Table 13.

[0191] Table 13 Mobile Phase Gradient for High Performance Liquid Chromatography

[0192] Time / min A％ B％ 0 78 22 36 78 22 61 33 67 67 33 67

[0193] The ultraviolet absorption spectrum of hINS is as follows: Figure 8 As shown.

[0194] Depend on Figure 8 As can be seen, the ultraviolet absorption spectrum shows that the maximum absorption peak is located near 280 nm, which is a protein-specific absorption peak. As the wavelength moves towards the ultraviolet direction, the absorption increases rapidly.

[0195] 3.7 Infrared Spectroscopic Characterization of Human Insulin

[0196] Weigh approximately 1 mg of human insulin purity standard candidate material and determine its concentration at 400–4000 cm⁻¹ using the KBr pellet method. -1 The infrared absorption spectrum, the results are as follows Figure 9 As shown.

[0197] The peak positions of the above figure were searched and compared with the standard human insulin spectrum in the NIST Chemistry WebBook. The results are shown in Table 14.

[0198] Table 14 Infrared Absorption Analysis of Candidate Standard Substances for Human Insulin Purity

[0199]

[0200] The infrared absorption spectra of the candidate human insulin purity standard material basically correspond to the NIST standard spectrum.

[0201] 4. Uniformity test

[0202] 4.1 Principles and Statistical Methods for Homogeneity Testing

[0203] The properties of a standard reference material should be homogeneous, meaning that its properties remain unchanged within a specified subdivision range. According to the technical requirements for assessing the homogeneity of standard reference materials in the national metrological technical specification JJF 1343-2012 "General Principles and Statistical Principles for the Determination of Standard Reference Materials," a certain number of the smallest packaging units are randomly selected (sampling can be done using the method shown in the random number table). A highly precise test method is used to measure each sample under the same controlled experimental conditions, thus ensuring that the differences between samples are fully reflected by the sample inhomogeneity.

[0204] 4.2 Uniformity Test Methods and Results

[0205] According to JJF 1343-2012, when the total number of units is ≤500, the number of units sampled should not be less than 11. A total of 200 units were prepared for this batch of purity standard materials; therefore, 11 units were randomly selected for homogeneity testing. Using the `randbetween` function in Excel, 11 random numbers were generated. Eleven units were then selected from the candidate purity standard materials based on these random numbers. Each candidate unit was measured three times using liquid chromatography-isotope dilution mass spectrometry. For each analysis, 0.5 mg of solid was taken from the sample unit, and the extracted hINS solid was prepared into a 1 mg / g solution for analysis. To eliminate errors caused by instrument drift during the measurement process, the samples were analyzed three times in an alternating order of 1→11, 11→1. The results are shown in Table 15.

[0206] Table 15. Homogeneity test data of candidate human insulin purity standard substances (g / g)

[0207] unit 1 time 2 times 3 times Observations Summation average variance 1 0.862 0.857 0.869 3 2.588 0.863 0.0060 2 0.857 0.848 0.859 3 2.564 0.855 0.0059 3 0.862 0.871 0.867 3 2.600 0.867 0.0045 4 0.851 0.854 0.865 3 2.570 0.857 0.0074 5 0.857 0.859 0.865 3 2.581 0.860 0.0042 6 0.864 0.863 0.868 3 2.595 0.865 0.0026 7 0.869 0.875 0.865 3 2.609 0.870 0.0050 8 0.854 0.862 0.872 3 2.588 0.863 0.0090 9 0.855 0.859 0.870 3 2.584 0.861 0.0078 10 0.861 0.853 0.851 3 2.565 0.855 0.0053 11 0.859 0.852 0.862 3 2.573 0.858 0.0051

[0208] The homogeneity of the candidate human insulin purity standard was tested using analysis of variance, and the results are shown in Table 16.

[0209] Table 16. Results of one-way ANOVA on homogeneity data of candidate human insulin purity standard substances.

[0210] Source of difference SS df MS F <![CDATA[F crit ]]> Between groups <![CDATA[0.0007055(Q1)]]> <![CDATA[10(v1)]]> <![CDATA[7.055×10 -5 (s1 2 )]]> 1.9811 2.2967 within the group <![CDATA[0.0007820(Q2)]]> <![CDATA[22(v2)]]> <![CDATA[3.555×10 -5 (s2 2 )]]>

[0211] The homogeneity variance analysis results of the human insulin purity standard candidate were F < F crit If there is no significant difference between the data groups, it can be considered that the samples are homogeneous.

[0212] 4.3 Minimum Sampling Size

[0213] During the homogeneity test, 0.5 mg of sample was taken from each sample unit for analysis. Therefore, the minimum sampling amount for the human insulin purity standard is 0.5 mg.

[0214] 5. Stability test

[0215] 5.1 Principles and Statistical Methods for Stability Assessment

[0216] The stability of reference materials is also an important parameter for evaluating them. The short-term stability of reference materials is related to external factors during sample transportation, while the long-term stability is related to storage conditions.

[0217] The stability of the human insulin purity standard material candidate was tested in accordance with the technical requirements, methods and result judgment methods for standard material stability assessment in the national metrological technical specification JJF 1343-2012 "General Principles and Statistical Principles for Standard Material Value Determination".

[0218] 5.2 Long-term stability assessment

[0219] Considering that the human insulin standard was aliquoted into amber glass tubes, and these tubes were sealed to prevent changes in volatile substances such as moisture during storage, reversed-phase high-performance liquid chromatography (RP-HPLC) was used to determine the results after deducting various impurities (i.e., the mass balance method) to evaluate the potential changes in the main components of the standard during storage. Simultaneously, due to the high moisture content of the sample, the Karl Fischer method was used to investigate the moisture content before and after 12 months of storage.

[0220] 5.2.1 Long-term stability study of principal components

[0221] First, the average results of the stability study were obtained by taking three candidate samples of human insulin purity standard at different time intervals and characterizing them using the mass balance method. The stability study results are shown in Table 17.

[0222] Table 17 Average results (g / g) of long-term stability studies of candidate human insulin purity standard substances

[0223] Time / Month 1 2 3 average 0 0.8511 0.8525 0.8513 0.8516 1 0.8528 0.8502 0.8532 0.8521 2 0.8510 0.8525 0.8516 0.8517 4 0.8524 0.8514 0.8511 0.8516 7 0.8523 0.8497 0.8499 0.8506 10 0.8527 0.8510 0.8520 0.8519 12 0.8530 0.8505 0.8502 0.8512

[0224] Fit the data in the table to a straight line, with x representing time and y representing the characteristic values ​​of the standard substance. The slope is b = -4.291 × 10⁻⁶. -5 The intercept b0 = 0.852.

[0225] The standard deviation of a straight line can be calculated using the following formula:

[0226]

[0227] Taking its square root s = 0.000587, the uncertainty of the slope is calculated using the following formula:

[0228]

[0229] The t-distribution factor with 5 degrees of freedom and p = 0.95 (95% confidence level) is 2.57, because...

[0230] |b|=-4.29×10 -5 <t 0.95,n-2 ·s(b)=0.000112

[0231] The t-test showed that the slope was not significant, indicating that the value of the human insulin purity standard material stored at -20℃ was stable over one year, meeting the requirements of national standard materials.

[0232] 5.2.2 Investigation of Moisture Content Stability

[0233] In this embodiment, the purity standard material was packaged into glass tubes and sealed to prevent changes in volatile substances such as water during storage. The main impurity component was water, and due to its high water content (approximately 10%), the change in water content before and after a 12-month stabilization period was investigated. The results are shown in Table 18.

[0234] Table 18 Average Results of Water Content Stability Study for Candidate Human Islet Purity Standard Reference Materials

[0235] Time / Month 1 2 3 average 0 10.76 10.61 11.04 10.80 12 11.11 10.25 11.28 10.88

[0236] Before and after the 12-month stabilization period, the average water content of the human islet purity standard material candidates did not change significantly, indicating that the water content was stable.

[0237] 5.3 Short-term stability assessment

[0238] The short-term stability of the standard material was tested using the mass balance method. The short-term stability test temperatures were 4℃, 25℃ and 40℃. The test period was 7 days at each temperature. Table 19 lists the average results of the stability test of 3 samples taken at different intervals.

[0239] Table 19 Results of short-term stability studies of human insulin purity standard candidates at different temperatures.

[0240]

[0241] The data in the table were fitted with a straight line, with x representing time and y representing the characteristic values ​​of the standard substance. The results are shown in Table 20.

[0242] Table 20 Results of Short-Term Stability Testing for Candidate Human Insulin Purity Standard Substances

[0243] Temperature / °C 0 days 1 day 2 days 4 days 7 days b <![CDATA[b0]]> s(b) s(b)·t in conclusion 4 0.8516 0.8520 0.8522 0.8513 0.8502 <![CDATA[-2.40×10 -4 ]]> 0.852 <![CDATA[8.51×10 -5 ]]> <![CDATA[2.71×10 -4 ]]> Stablize 25 0.8516 0.8506 0.8507 0.8497 0.8503 <![CDATA[-1.77×10 -4 ]]> 0.851 <![CDATA[1.03×10 -4 ]]> <![CDATA[3.28×10 -4 ]]> Stablize 40 0.8516 0.8506 0.8486 0.8479 0.8481 <![CDATA[-4.94×10 -4 ]]> 0.851 <![CDATA[1.95×10 -4 ]]> <![CDATA[6.20×10 -4 ]]> Stablize

[0244] The t-test showed that the slope was not significant, indicating that the value of the human insulin purity standard substance was stable over a period of 7 days under three different storage conditions: 4℃, 25℃, and 40℃.

[0245] 5.4 Stability after opening

[0246] The open-bottle stability of the standard reference material was tested using the mass balance method. One unit of standard reference material was randomly selected, and 0.5 mg of sample was taken from the bottle for analysis each time after opening. After analysis, the bottle was sealed with sealing film and stored at -20°C. This process was repeated every one day for a total of four days. The results are shown in Tables 21-22.

[0247] Table 21 Results of Open-Bottle Stability Study of Candidate Human Insulin Purity Standards

[0248]

[0249] Table 22 Results of Open-Bottle Stability Test for Candidate Human Insulin Purity Standards

[0250] Temperature / °C 0 days 1 day 2 days 4 days b <![CDATA[b0]]> s(b) s(b)·t in conclusion -20 0.8499 0.8512 0.8505 0.8509 <![CDATA[1.63×10 -4 ]]> 0.850 <![CDATA[2.02×10 -4 ]]> <![CDATA[8.69×10 -4 ]]> Stablize

[0251] Meanwhile, the open-bottle stability of the standard substance at room temperature was tested using the mass balance method. One unit of standard substance was randomly selected, and 0.5 mg of sample was taken from the bottle for analysis each time after opening. After analysis, the bottle was sealed with sealing film and stored at room temperature. The above process was repeated every one day for a total of four days. The results are shown in Tables 23-24.

[0252] Table 23 Results of the Open-Bottle Stability Study of Candidate Human Insulin Purity Standards

[0253]

[0254] Table 24 Results of Open-Bottle Stability Test for Candidate Human Insulin Purity Standards

[0255] Temperature / °C 0 days 1 day 2 days 4 days b <![CDATA[b0]]> s(b) s(b)·t in conclusion room temperature 0.850 0.851 0.853 0.852 <![CDATA[5.14×10 -4 ]]> 0.851 0.0003918 0.001686 Stablize

[0256] Conclusion: The prepared human insulin purity standard can be stored for at least 1 year at -20℃, at least 7 days at 4℃, 25℃ and 40℃, and at least 4 days when stored at -20℃ or room temperature after opening.

[0257] 6. Identification and quantification of impurities in hINS

[0258] Analysis of hINS by reversed-phase high-performance liquid chromatography (RP-HPLC) showed that the area-normalized purity of the sample was 98.4%. The largest impurity was located immediately adjacent to the main peak, with an area normalization ratio of approximately 1.1%. According to the Chinese Pharmacopoeia (2020 edition), this peak is the peak of A21 deaminized insulin. The area normalization ratio of impurities (groups) around 48 min was approximately 0.4%. A21 deaminized insulin was quantified using fractional collection-HPLC-isotope dilution mass spectrometry (ICMMS), and the purity ratio factors of the main components, leucine, valine, and phenylalanine, were determined.

[0259] 6.1 Identification of impurities in hINS

[0260] The main insulin peak and subsequent impurity peaks at a concentration of 1.1% were collected using optimized reversed-phase high-performance liquid chromatography (RP-HPLC) conditions. After desalting, the molecular weight was determined using matrix-assisted laser-induced desorption / desorption-time-of-flight mass spectrometry (MA-ALS).

[0261] To determine which peptide chain of insulin underwent deamination, insulin was reduced with DTT and then desalted. MS analysis was then performed on both the A and B chains of insulin. 2 The obtained b and y ions were analyzed to determine the location of the modification.

[0262] The measured mass spectrum is as follows Figure 10 As shown.

[0263] Depend on Figure 10 As can be seen, the mass number difference of 17 between the impurity peak and the main insulin peak indicates the presence of deamination products in insulin. Insulin and impurities were subjected to DTT reduction together, desalted, and then analyzed by MS. 2 Analysis revealed a clear deamination peak on the MS2 mass spectrum of chain A, such as... Figure 11 As shown, deamination occurs on the A chain. Sequence analysis of the deamination peak of the A chain yielded the tag YQLENY. The insulin A chain peptide sequence is GIVEQCCTSICSLYQLENYCN. Following YQLENY, there are two amino acid residues, CN. Deamination of C is impossible; only N (Asn) could potentially undergo cyclization to form a carbodiimide structure, losing NH3 (theoretical molecular weight 17.0265). This confirms that the impurity is A21 deamination insulin, consistent with records in the Chinese Pharmacopoeia and other publications.

[0264] Quantification of impurities in 6.2hINS

[0265] The main component and impurity proteins of hINS were separated using the same chromatographic conditions as those used for determining the purity of hINS by reversed-phase high-performance liquid chromatography (RP-HPLC). The main peak, A21 deaminoinsulin, and other impurity components were collected separately and labeled F1, F2, and F3, with masses m1, m2, and m3, respectively. A known concentration of isotopically labeled amino acids was quantitatively added to F1, F2, and F3, followed by mixing, centrifugation, concentration, drying, and hydrolysis according to the hINS sample hydrolysis method. After hydrolysis, each component was dried under nitrogen, reconstituted, and the amino acid concentrations (c1, c2, c3) of each component were determined using HPLC-IDMS. The impurity protein correction factor for each amino acid was calculated using formula 6-1; the mass fraction of A21 deaminoinsulin was calculated using formula 6-2.

[0266]

[0267] In the formula, R AAc1 is the correction factor for a specific amino acid; m1 is the mass of the main peak collected (g); m2 is the mass of the A21 deaminated insulin peak collected (g); m3 is the mass of other impurity peaks collected (g); c1 is the mass concentration of a specific amino acid in the main peak (g / g); c2 is the mass concentration of a specific amino acid in the A21 deaminated insulin peak (g / g); c3 is the mass concentration of other impurity peaks (g / g).

[0268]

[0269] In the formula, c A21-hINS The mass fraction of deaminated insulin is expressed in g / g; c AA R represents the mass fraction of amino acids, in g / g. AA The purity ratio factor of A21 deamination insulin for a certain amino acid; N AA M represents the number of a specific amino acid in A21 deamination insulin; AA M represents the molecular weight of a specific amino acid, expressed in g / mol. A21-hINS ρ represents the molecular weight of deaminized insulin, in g / mol.

[0270] The purity ratio factor of each amino acid in each component was calculated according to Equation 6-1, and the results are shown in Table 25.

[0271] Table 25 Purity ratio factors of amino acids in each component

[0272]

[0273] According to Equation 6-2 and Table 25, the mass fraction of A21 deamination insulin is 0.034 g / g.

[0274] 7. Constant value

[0275] The determination of protein content was performed using liquid chromatography-isotope dilution mass spectrometry (HPLC-IDMS). Five parallel samples were taken, and each sample was analyzed three times. The protein content was obtained by dividing the measured amino acid content by the mass fraction of that amino acid in insulin. The result after impurity correction based on the amino acid ratio factor was used as the determination result of the HPLC-IDMS method, as shown in Table 26.

[0276] Table 26 Results of the Assignment of Purity Standard Reference Materials for Human Insulin (g / g)

[0277]

[0278] The normality of the mass balance determination results was tested using IBM SPSS 20 software, and the results are shown in Table 27.

[0279] Table 27 Normality Test of Candidate Values ​​for Human Insulin Purity Standard Material

[0280]

[0281]

[0282] Due to the small sample size, the KS result is used as the standard. Both statistical methods yielded sig. values ​​greater than 0.05, indicating a normal distribution. Further confirmation can be obtained by examining the QQ graph (e.g.). Figure 12 As shown in the figure, the expected standard is basically distributed around the straight line, which can be considered to follow a normal distribution.

[0283] Outlier detection was performed on the candidate values ​​of human insulin purity standard material in Table 26 using Dixon and Grubbs outlier detection methods, respectively. The results are shown in Table 28.

[0284] Table 28. Inspection of Outliers in Fixed Value Data

[0285]

[0286] The average value of each standard substance unit was tested using both Grubbs and Dixon outlier detection methods, and no outliers were found. Therefore, the arithmetic mean of the five units was taken as the final value of the human insulin purity standard substance isotope dilution mass spectrometry method, which is 0.857 g / g.

[0287] 8. Uncertainty assessment and result presentation

[0288] The uncertainty of the determination result of the purity standard substance of human insulin comes from the uncertainty introduced in the determination process, the uncertainty introduced by the homogeneity of the standard substance, and the uncertainty introduced by the short-term stability and long-term stability.

[0289] 8.1 Uncertainty introduced by the determination process

[0290] The calculation model for the normal values ​​of liquid chromatography-isotope dilution mass spectrometry is as follows:

[0291]

[0292] In the formula: C is the mass fraction of human insulin (in g / g); M is the mass of the human insulin sample used for hydrolysis (in g); R1 is the peak area ratio of amino acids to labeled amino acids in the high-standard solution; R2 is the peak area ratio of amino acids to labeled amino acids in the low-standard solution; P is the purity of amino acids (in g / g); m 标 R represents the mass (in g) of labeled amino acids in the hydrolyzed sample; 样 W1 is the peak area ratio of amino acids to labeled amino acids after sample hydrolysis; W2 is the mass ratio of amino acids to labeled amino acids in the high-standard solution; P is the mass ratio of amino acids to labeled amino acids in the low-standard solution. HFor hydrolysis efficiency; R AA M is the amino acid purity ratio factor (unit: g / g); INS N is the molecular weight of insulin. AA M represents the number of amino acids in insulin. AA This represents the molecular weight of insulin.

[0293] According to the mathematical model of human insulin measurement results, the uncertainty mainly comes from the uncertainty components of weighing, peak area ratio measurement repeatability, hydrolysis efficiency, standard substance purity, insulin molecular weight, amino acid molecular weight, and amino acid purity ratio factor.

[0294] In the experiment, Val, Leu, and Phe were used to measure insulin levels, and their average values ​​were taken. The uncertainty of each amino acid in the insulin quantification result was calculated. The measurement uncertainty of Phe is calculated as an example below.

[0295] 8.1.1 Uncertainty u(P) introduced by amino acid purity

[0296] The experiment used GBW09235 L-phenylalanine purity standard material. According to the standard material certificate, the purity value is (0.998±0.004) g / g, k=2. Therefore, the expanded uncertainty introduced by the purity of the standard material is:

[0297]

[0298] 8.1.2 Uncertainty u(P) introduced by hydrolysis efficiency H )

[0299] Based on the international comparison results of CCQM P55.1 peptide-protein isotope dilution mass spectrometry and the joint verification project of ACRM Action#1 insulin (pig) content standard material between China, Japan and South Korea, the hydrolysis efficiency was estimated at 1%. The measured result of human insulin isotope dilution mass spectrometry was 0.861 g / g. Therefore, the uncertainty u(P) introduced by the hydrolysis efficiency... H Calculate according to the following formula:

[0300] u(P H = 0.857 × 1% = 0.00857 g / g

[0301] 8.1.3 Uncertainty u(R) introduced by the amino acid purity scaling factor AA )

[0302] Based on the results of the determination method for the amino acid purity ratio factor, the range of the measurement results is R = 0.005 g / g, n = 3, and C = 1.69 is obtained from the table. Therefore, the uncertainty component introduced by the amino acid purity ratio factor is:

[0303]

[0304] 8.1.4 Uncertainty introduced by weighing

[0305] The labeled amino acid and human insulin samples were weighed using a balance. According to the balance's calibration certificate, the expanded uncertainty for weighing 1–500 mg was U. r =0.2%, k=2, so the standard uncertainty component introduced by weighing can be calculated by the following formula:

[0306]

[0307] 8.1.5 Uncertainty introduced by mass ratio

[0308] The mass ratio is the ratio of the masses measured in two weighings. Therefore, the uncertainty introduced by the mass ratio is calculated using the following formula:

[0309]

[0310] 8.1.6 Uncertainty introduced by peak area ratio

[0311] The uncertainty introduced by the peak area ratio is assessed using a Type A uncertainty evaluation method. Six measurements are performed on a standard or sample to obtain the area A of labeled and unlabeled amino acids. The relative standard deviation of the area ratio from the six repeated measurements is calculated, and the uncertainty introduced by the peak area ratio is calculated using the following formula:

[0312]

[0313] 8.1.7 Uncertainty introduced by molecular weight

[0314] Based on the molecular formula of insulin C 257 H 383 N 65 O 77 S6. Calculate the uncertainty introduced by the molecular weight using the following formula:

[0315]

[0316] Where: N i The number of atoms of the element; u i M represents the uncertainty in the relative atomic mass of an element; INS This represents the molecular weight of insulin.

[0317] According to the IUPAC International Atomic Weights Table, the relative atomic masses and uncertainties of the five atoms are as follows: C: 12.0107±0.0008; H: 1.00794±0.00007; O: 15.9994±0.0003; N: 14.00674±0.00007; S: 32.065±0.005.

[0318] For each element, its standard uncertainty is derived from the reference uncertainty according to a uniform distribution:

[0319] therefore:

[0320]

[0321] Similarly:

[0322]

[0323] 8.1.8 Uncertainty in calculating the mass fraction of human insulin from phenylalanine

[0324] The uncertainty introduced when calculating the mass fraction of human insulin from phenylalanine is calculated using the following formula:

[0325]

[0326] 8.1.9 Standard uncertainty of mass fraction isotope dilution mass spectrometry determination results of human insulin

[0327] Similarly, the uncertainties introduced when calculating the mass fraction of human insulin using valine and phenylalanine were calculated as follows:

[0328] u(c,Val)=0.016g / g; u(c,Leu)=0.024g / g.

[0329] Since the mass fraction isotope dilution mass spectrometry value of human insulin is obtained by taking the arithmetic mean of the results of three amino acids, therefore:

[0330]

[0331] 8.2 Uncertainty introduced by sample homogeneity

[0332] According to JJF1343-2012, the uncertainty introduced by the homogeneity of the standard substance is equal to the standard deviation of the homogeneity between bottles. When the mean square between groups is greater than the mean square within groups, it is calculated according to formula (8-2).

[0333]

[0334] When the mean square between groups is less than the mean square within groups, it is calculated according to formula (8-3).

[0335]

[0336] Uncertainty introduced by sample homogeneity:

[0337]

[0338] 8.3 Uncertainty introduced by sample stability

[0339] The short-term and long-term stability of the standard material are calculated using formulas (8-4) and (8-5):

[0340] u sts =s k,sts ·t Formula 8-4

[0341] u lts =s k,lts ·t Formula 8-5

[0342] The uncertainty components introduced by stability are:

[0343] u Stability，r =s(k)×12=5.17×10 -5 ×12=0.00062

[0344] therefore:

[0345] u Stability =u Stability ×0.857g / g=0.00062×0.857=0.000531g / g

[0346] 8.4 Uncertainty Assessment Results

[0347] The standard combined uncertainty is as follows:

[0348]

[0349] Taking the expansion factor k = 2, corresponding to a 95% confidence level, the expanded uncertainty can be expressed as:

[0350] U = k × u c =2 × 0.012 = 0.024 g / g

[0351] Therefore, the determination result of the standard substance can be expressed as: (0.857±0.024) g / g.

[0352] The above descriptions are merely some embodiments of the present invention. Those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these all fall within the scope of protection of the present invention.

Claims

1. A method for the preparation of a human insulin purity reference material, characterized in that, The method comprises the following steps: (1) selecting human insulin with a nominal purity of not less than 98%, performing purity analysis, and then performing homogeneity preliminary inspection by high performance liquid chromatography, taking the human insulin passing the homogeneity preliminary inspection as a human insulin purity standard material candidate, and performing packaging and storage; (2) performing qualitative identification, homogeneity inspection and stability investigation on the human insulin purity standard material candidate; (3) if the structure, homogeneity and stability of the human insulin purity standard material candidate all meet the measurement requirements, quantifying A21 desamino human insulin in the human insulin purity standard material candidate, and measuring purity proportion factors of phenylalanine, valine and leucine of the main component of the human insulin purity standard material candidate; the calculation formula of the purity proportion factors is as follows: wherein, R AA is the purity proportion factor for a certain amino acid; m 1, m 2, m 3 are the masses of the main peak of human insulin, the A21 desamino insulin peak and other impurity peaks collected by reverse phase high performance liquid chromatography, respectively, g; c 1 is the mass concentration of a certain amino acid in the main peak, g / g; c 2 is the mass concentration of a certain amino acid in the A21 desamino insulin, g / g; c 3 is the mass concentration of other impurity peaks, g / g; (4) hydrolyzing the human insulin purity standard material candidate, taking isotopically labeled phenylalanine, valine and leucine as internal standards, and performing value determination on the human insulin purity standard material candidate by liquid chromatography-isotope dilution mass spectrometry; wherein, the determined amino acid content is divided by the mass fraction of the amino acid in the insulin to obtain the protein content, and the result after the amino acid purity proportion factor is corrected is taken as the value determination result; (5) performing uncertainty evaluation on the human insulin purity standard material candidate, and obtaining the human insulin purity standard material.

2. The production method according to claim 1, characterized by, The qualitative identification on the human insulin purity standard material candidate is at least one selected from mass spectrometry identification, molecular weight determination, amino acid sequence determination, composition analysis of leucine and isoleucine, ultraviolet spectrum characterization and infrared spectrum characterization.

3. The production method according to claim 1 or 2, characterized by, When the stability of the human insulin purity standard material candidate is investigated, the purity of the human insulin purity standard material candidate is measured by the mass balance method, wherein the purity of the human insulin is calculated according to the following formula: w = P ×(1- A - B - C - D ) wherein w the purity of the human insulin purity reference material candidate, P the purity of the human insulin purity reference material candidate as determined by reversed phase high performance liquid chromatography, A the water content, B the total content of ash and anion content, C the residual solvent content, D the aggregate content.

4. The production method according to claim 3, characterized by, When the homogeneity of the human insulin purity standard material candidate is inspected, the purity of the human insulin purity standard material candidate is measured by the following steps: hydrolyzing the human insulin purity standard material candidate, taking isotopically labeled phenylalanine, valine and leucine as internal standards, and measuring the purity of the human insulin purity standard material candidate by liquid chromatography-isotope dilution mass spectrometry.

5. The production method according to claim 1 or 4, characterized by, The hydrolysis of the human insulin purity standard material candidate comprises the following steps: (1) the human insulin purity standard material candidate is prepared into a human insulin solution with a concentration of 1 mg / g by using 0.01 mol / L hydrochloric acid; (2) the human insulin solution is taken, water is removed, 6 mol / L hydrochloric acid is added in a volume ratio of 1:5 with the taken human insulin solution, oxygen is removed, and then the solution is hydrolyzed under the conditions of sealing and temperature of 109.5-110.5℃ for 60-65 h.

6. The production method according to claim 5, wherein In the liquid chromatography-isotope dilution mass spectrometry method, the detection conditions of the liquid chromatography include: using a solvent composed of 0.1% trifluoroacetic acid, 9% acetonitrile and the balance of water by volume percentage as a mobile phase for isocratic elution, an elution time of 10 min, and a chromatographic column of Phenomenex kinetex 2.6 μm C 18 18, 150 mm x 2.1 mm.

7. The production method according to claim 6, wherein In the liquid chromatography-isotope dilution mass spectrometry, the detection conditions of the mass spectrometry include: a collision energy of 14 V, an outlet voltage of 10 V, an inlet voltage of 10 V and a de-clustering voltage of 50 V.

8. The preparation method according to claim 7, characterized in that, The uncertainty includes a constant value process introduced uncertainty, a homogeneity introduced uncertainty, a short-term stability and a long-term stability introduced uncertainty, wherein the constant value introduced uncertainty includes an amino acid purity introduced uncertainty, a human insulin hydrolysis efficiency introduced uncertainty, an amino acid purity proportionality factor introduced uncertainty, a weighing introduced uncertainty, a mass ratio introduced uncertainty, a peak area ratio introduced uncertainty and a human insulin molecular weight introduced uncertainty.

9. A human insulin purity reference material characterized in that, which is prepared by the method according to any one of claims 1-8.

10. Use of the human insulin purity reference material according to claim 9 as a purity reference material.

Images