Plasma free glycated amino acid markers for aiding diagnosis of type 2 diabetes and use thereof

Abstract

The application belongs to the field of biological medicine and clinical testing technology, and discloses a kind of plasma free glycosylated amino acid markers for assisting diagnosis of type 2 diabetes and application thereof, the biomarker is selected from any one or three combinations of the following three plasma free glycosylated amino acids: N-(1-deoxy-1-fructosyl) valine, N-(1-deoxy-1-fructosyl) isoleucine / leucine, N-(1-deoxy-1-fructosyl) phenylalanine.The application efficiently screens out three prediction markers including N-(1-deoxy-1-fructosyl) valine, N-(1-deoxy-1-fructosyl) isoleucine / leucine and N-(1-deoxy-1-fructosyl) phenylalanine for type 2 diabetes.The type 2 diabetes marker of the application can be used for early warning of the risk of type 2 diabetes, as an aid to diagnosis, and research on drug action target, precise medication, pathogenesis, etc.

Description

Technical Field

[0001] This invention belongs to the fields of biomedicine and clinical testing technology, and in particular to a plasma free glycosylated amino acid marker for the auxiliary diagnosis of type 2 diabetes and its application. Background Technology

[0002] Type 2 diabetes mellitus (T2DM) is a metabolic disease characterized primarily by chronic hyperglycemia. Its pathogenesis involves insulin resistance (IR) and progressive decline in pancreatic β-cell function. According to the latest statistics from the International Diabetes Federation (IDF), the number of people with diabetes worldwide has exceeded 537 million, and the disease is showing a significant trend towards affecting younger and earlier age groups. Early and accurate diagnosis, along with continuous and stable blood glucose monitoring, is of crucial clinical significance for delaying disease progression and reducing the risk of cardiovascular and cerebrovascular complications.

[0003] Currently, traditional blood glucose assessment indicators used clinically for the diagnosis and monitoring of type 2 diabetes mellitus (T2DM) mainly include fasting blood glucose (FBG), oral glucose tolerance test (OGTT), glycated hemoglobin (HbA1c), glycated albumin (GA), and 1,5-anhydroglucosidase (1,5-AG). These indicators reflect the glucose metabolism status at different time windows and form the basis of current clinical management of diabetes. However, with the development of precision medicine and the increasing heterogeneity of the diabetic population, these traditional indicators all have inherent limitations in practical application, as specifically manifested as follows: (1) FBG and OGTT: They can only reflect the immediate blood glucose level at the moment of blood collection and are easily affected by factors such as short-term diet, stress or emotional fluctuations. Although OGTT can dynamically reflect the body's ability to respond to glucose load, its detection process is cumbersome, time-consuming (usually 2-3 hours), and patient compliance is poor, making it difficult to be widely used as a daily monitoring method.

[0004] (2) HbA1c: This reflects the average blood glucose level over the past 2-3 months and is considered the "gold standard" for evaluating long-term blood glucose control. However, because its detection principle relies on the lifespan of red blood cells (approximately 120 days), HbA1c exhibits a significant "time lag": for early-stage T2DM patients, their blood glucose abnormalities have not persisted for a sufficient period, and HbA1c often has not yet risen significantly, easily leading to early missed diagnoses; at the same time, for diagnosed patients, adjustments to the treatment plan or short-term improvements in blood glucose control cannot be reflected in HbA1c levels in a timely manner, making it difficult for clinicians to quickly assess short-term efficacy. In addition, HbA1c may also be inaccurate under certain pathological conditions, such as hemoglobinopathies, chronic kidney disease, and liver disease, and may be affected by certain medications, blood loss or transfusions, and pregnancy.

[0005] (3) 1,5-AG: As a short- to medium-term biomarker reflecting blood glucose fluctuations within 1 to 2 weeks, its detection mechanism relies on the competitive inhibition of 1,5-AG renal tubular reabsorption caused by urinary glucose excretion under hyperglycemic conditions. This mechanism determines that 1,5-AG can only produce significant changes when the blood glucose level exceeds the renal threshold for glucose (usually about 10.0 mmol / L). For early-stage T2DM patients with mild to moderate hyperglycemia, the sensitivity of 1,5-AG is significantly reduced, resulting in a clear early detection blind spot. At the same time, in individuals with renal insufficiency or significant individual differences in renal threshold for glucose, the detection results of 1,5-AG cannot reflect the true blood glucose situation, thus limiting its clinical applicability.

[0006] (4) GA: Reflects the average blood glucose level over the past 2-3 weeks and is considered an effective means to fill the gap between the monitoring windows of FBG and HbA1c. However, since GA is essentially a non-enzymatic glycation product of large molecule serum albumin (ALB), its detection results are inevitably strongly interfered with by the synthesis and degradation of albumin in the body. Clinical studies have shown that many common clinical diseases such as obesity, thyroid dysfunction, liver and kidney dysfunction, and malnutrition can lead to significant false abnormalities in GA.

[0007] In recent years, with the development of metabolomics technology, numerous studies have shown that abnormalities in small molecule metabolites in the blood are closely related to the occurrence and development of type 2 diabetes mellitus (T2DM). Existing research indicates that disorders of free amino acid metabolism in the blood are closely related to insulin resistance (IR), especially elevated levels of branched-chain amino acids (BCAAs) and aromatic amino acids (AAAs), and decreased levels of glycine. The aforementioned traditional blood glucose assessment indicators all focus on monitoring the single dimension of "glucose concentration," lacking a composite biomarker that can simultaneously reflect the dual pathological processes of early-stage blood glucose accumulation and free amino acid metabolism disorders.

[0008] Glycation is a non-enzymatic reaction (Maillard reaction) between proteins or amino acids and reducing sugars such as glucose. HbA1c and GA are products formed by the non-enzymatic reaction between hemoglobin and serum albumin in the blood and glucose or other carbohydrate molecules. Given the dual pathological background of persistent hyperglycemia and disordered free amino acid metabolism, small-molecule free glycated amino acids (FGAs), as primary glycation end products directly generated from glucose and free amino acids in the blood via the Maillard reaction, theoretically hold the potential to become a key molecular link connecting these two core pathological processes. Compared to traditional large glycation molecules such as HbA1c and GA, FGAs possess significant potential advantages: their carriers (free amino acids) have a significantly shorter half-life in blood circulation than hemoglobin and serum albumin, and their metabolic clearance pathway is independent of red blood cell lifespan or albumin levels in the body. This unique molecular characteristic theoretically gives it the potential to fill the short-to-medium-term blood glucose monitoring blind spot between FBG and HbA1c, while also having the potential to overcome the existing deficiencies of 1,5-AG (renal threshold dependent) and GA (in vivo albumin turnover interference).

[0009] However, there are many types of free glycated amino acids (GAAs), and different types of GAAs vary in their stability in vivo, plasma concentration levels, correlation with blood glucose, and pathological associations. Some GAAs are structurally unstable, prone to subsequent degradation by Amadori rearrangement products, or have extremely low plasma concentrations, making them difficult to detect and quantify accurately. Therefore, transforming free GAAs from a theoretically promising candidate biomarker into a clinically usable blood glucose assessment tool still faces significant technical barriers in the current technology. To date, there is a lack of systematic, large-sample clinical cohort studies in this field that clearly indicate which single free GAAs or which combination of free GAAs can serve as novel blood glucose assessment biomarkers to supplement the deficiencies of existing traditional blood glucose assessment indicators and provide new clinical tools and theoretical basis for further elucidating the metabolic pathological mechanisms of type 2 diabetes mellitus (T2DM).

[0010] In summary, existing traditional blood glucose assessment indicators for type 2 diabetes mellitus (T2DM) generally suffer from multiple technical bottlenecks, including susceptibility to interference from non-specific physiological factors, insufficient sensitivity in early-stage and short-term patients, and difficulty in reflecting deeper metabolic pathological dimensions. While free glycated amino acid biomarkers theoretically show potential to overcome these bottlenecks, no specific free glycated amino acid biomarkers with high clinical efficacy, strong anti-interference stability, and independent pathological relevance have yet been systematically screened and identified from blood. Therefore, there is an urgent need to develop one or more specific free glycated amino acid biomarkers based on human blood and their detection and evaluation methods to meet the pressing clinical needs for early diagnosis and refined management of T2DM in the context of precision medicine. Summary of the Invention

[0011] The purpose of this invention is to overcome the shortcomings of the prior art and provide a plasma free glycated amino acid marker for the auxiliary diagnosis of type 2 diabetes and its application.

[0012] The technical solution adopted by this invention to solve its technical problem is: A biomarker for the auxiliary diagnosis of type 2 diabetes mellitus, wherein the biomarker is selected from any one or a combination of three of the following three plasma free glycated amino acids: (1) N-(1-deoxy-1-fructosyl)valine, also known as N-(1-Deoxy-1-fructosyl)valine, N-Fructosyl valine, or Fru-Val; (2) N-(1-deoxy-1-fructosyl)isoleucine / leucine, namely N-(1-Deoxy-1-fructosyl)isoleucine / leucine, N-Fructosyl isoleucine / leucine, Fru-Leu / Ile; Since isoleucine and leucine are isomers, they are difficult to completely separate at baseline under conventional chromatographic conditions. Therefore, they are combined as single detection characteristic peaks for characterization. (3) N-(1-deoxy-1-fructosyl)phenylalanine, namely N-(1-Deoxy-1-fructosyl)phenylalanine, N-Fructosyl phenylalanine, Fru-Phe.

[0013] Further, the molecular formulas of N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are C1, C2, C3, C4, C5, C6, C7, C8, C9 ... 11 H 21 NO7, C 12 H 23 NO7 and C 15 H 21 NO.7

[0014] Furthermore, the three compounds N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine, and N-(1-deoxy-1-fructosyl)phenylalanine all eluted in the polar region on a C18 reversed-phase column, i.e., the retention time was <3 min, which is consistent with the chromatographic behavior characteristics of glycosylated amino acid compounds, which are highly polar and weakly hydrophobic.

[0015] Furthermore, the individual biomarkers among N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine all exhibited extremely high independent diagnostic efficacy. Their individual ROC curve areas (AUCs) remained consistently between 0.845 and 0.855, with sensitivity reaching 85.3% to 86.3%. Alternatively, when N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are used independently to differentiate between patients with short-term T2DM and non-diabetic individuals, their AUC values ​​are consistently above 0.84, specifically 0.851, 0.855, and 0.845, respectively. Among them, N-(1-deoxy-1-fructose)isoleucine / leucine exhibits the highest monomeric diagnostic efficacy, with an AUC of 0.855. At the optimal threshold of 322,649,157.5, its detection sensitivity for short-term T2DM reaches 86.3%, and its specificity reaches 72.7%.

[0016] Furthermore, the N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine all showed a highly significant moderate positive correlation with FPG and HbA1c. P The value <0.0001 ensures its biological reliability as a marker of glucose metabolism; In the primary target population, i.e., those with normal to mildly impaired renal function and an eGFR > 60 mL / min / 1.73 mcg, 2 In this study, the levels of N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine, and N-(1-deoxy-1-fructosyl)phenylalanine biomarkers were not statistically significant with eGFR, and their molecular formation mechanisms did not depend on the renal urinary glucose excretion process at all; they also showed only a very weak clinical association with ALB (r<0.2).

[0017] The application of plasma free glycosylated amino acid markers as described above in the preparation of T2DM auxiliary diagnostic reagents and / or mass spectrometry detection kits and / or early screening and warning models.

[0018] The application of plasma free glycosylated amino acid biomarkers as described above in the preparation of short-course T2DM auxiliary diagnostic reagents and / or mass spectrometry detection kits and / or early screening and warning models.

[0019] The above-mentioned plasma free glycated amino acid biomarkers are used in the preparation of risk warning models for type 2 diabetes and / or to assist in the research of diagnostic kits and / or drug target research, precision medicine, and pathogenesis research.

[0020] The application of plasma free glycated amino acid markers as described above in the preparation of reagents and / or diagnostic kits to reduce the rate of missed diagnoses of early diabetes.

[0021] The application of plasma free glycated amino acid markers as described above in the preparation of early auxiliary diagnostic reagents and / or kits and / or detection devices for short-course type 2 diabetes mellitus (T2DM).

[0022] The advantages and positive effects of this invention are as follows: 1. This invention targets type 2 diabetes mellitus (T2DM) and efficiently screens for three predictive biomarkers: N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine. These T2DM biomarkers can be used for early warning of type 2 diabetes risk; as an adjunct to diagnosis; and for research on drug targets, precision medicine, and pathogenesis.

[0023] 2. This invention reveals for the first time that N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine, and N-(1-deoxy-1-fructosyl)phenylalanine are abnormally elevated in the plasma of patients with type 2 diabetes mellitus (T2DM).

[0024] 3. The invention's single indicator possesses outstanding diagnostic efficacy: Studies have confirmed that the three single biomarkers mentioned above all exhibit extremely high independent diagnostic efficacy. Their individual ROC curve areas (AUC) are consistently between 0.845 and 0.855, with a sensitivity as high as 85.3% to 86.3%. This means that detecting any single one of these substances can significantly reduce the missed diagnosis rate of early diabetes.

[0025] 4. The N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine screened in this invention serve as biomarkers for blood glucose evaluation in type 2 diabetes mellitus (T2DM), possessing the triple advantages of high effectiveness, dimensional independence, and robust anti-interference: they show extremely significant positive correlations with both FPG and HbA1c (P<0.0001), demonstrating sufficient biological reliability; they are significantly correlated with but not collinear with the traditional gold standard for blood glucose, independently reflecting the short- to medium-term gluconeogenic microenvironment and filling a gap in existing monitoring windows; and they are effective in conditions where eGFR > 60 mL / min / 1.73 m 2It showed no correlation with eGFR in the population and only a very weak association with ALB (r<0.2). It broke through the core bottleneck of 1,5-AG renal threshold dependence and glycated albumin turnover interference at the mechanistic level. It is especially suitable for early screening, progression monitoring and personalized efficacy evaluation of short-term T2DM, with outstanding clinical value and industrialization prospects. Attached Figure Description

[0026] Figure 1 These are the secondary mass spectra of the three free glycosylated amino acids in this invention; wherein, Figure 1 A is the secondary mass spectrum of N-(1-deoxy-1-fructosyl)valine. Figure 1 B is the secondary mass spectrum of N-(1-deoxy-1-fructosyl)isoleucine / leucine. Figure 1 C is the secondary mass spectrum of N-(1-deoxy-1-fructosyl)phenylalanine; Figure 2 This is a box plot showing the differential expression of three free glycated amino acids in the short-course T2DM group and the control group in this invention; wherein, the first row of the plot is the differential expression box plot of N-(1-deoxy-1-fructose)valine; the second row of the plot is the differential expression box plot of N-(1-deoxy-1-fructose)isoleucine / leucine; and the third row of the plot is the differential expression box plot of N-(1-deoxy-1-fructose)phenylalanine. Figure 3 This is a ROC curve diagram of the diagnosis of diabetes using three free glycosylated amino acids individually and in combination, as described in this invention. Figure 4 The images show scatter plots illustrating the correlation between the three free glycosylated amino acids and FPG and HbA1c in this invention. Specifically, the first row (left) shows the correlation between the relative quantitative peak area of ​​N-(1-deoxy-1-fructosyl)valine and FPG levels, and the right row (right) shows the correlation between the relative quantitative peak area of ​​N-(1-deoxy-1-fructosyl)valine and HbA1c levels. The second row (left) shows the correlation between the relative quantitative peak area of ​​N-(1-deoxy-1-fructosyl)isoleucine / leucine and FPG levels, and the right row (right) shows the correlation between the relative quantitative peak area of ​​N-(1-deoxy-1-fructosyl)isoleucine / leucine and HbA1c levels. The third row (left) shows the correlation between the relative quantitative peak area of ​​N-(1-deoxy-1-fructosyl)phenylalanine and FPG levels, and the right row (right) shows the correlation between the relative quantitative peak area of ​​N-(1-deoxy-1-fructosyl)phenylalanine and HbA1c levels. Detailed Implementation

[0027] The present invention will be further described below with reference to the embodiments. The following embodiments are descriptive and not limiting, and should not be used to limit the scope of protection of the present invention.

[0028] The various experimental operations involved in the specific embodiments are all conventional techniques in the art. For parts not specifically annotated herein, those skilled in the art can refer to various commonly used reference books, scientific and technological literature, or related instructions and manuals prior to the filing date of this invention for implementation. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the invention. Unless otherwise specified, the technical means used in the embodiments are conventional means well known to those skilled in the art, and the raw materials used are all commercially available products.

[0029] A biomarker for the auxiliary diagnosis of type 2 diabetes mellitus, wherein the biomarker is selected from any one or a combination of three of the following three plasma free glycated amino acids: (1) N-(1-deoxy-1-fructosyl)valine, also known as N-(1-Deoxy-1-fructosyl)valine, N-Fructosyl valine, or Fru-Val; (2) N-(1-deoxy-1-fructosyl)isoleucine / leucine, namely N-(1-Deoxy-1-fructosyl)isoleucine / leucine, N-Fructosyl isoleucine / leucine, Fru-Leu / Ile; Since isoleucine and leucine are isomers, they are difficult to completely separate at baseline under conventional chromatographic conditions. Therefore, they are combined as single detection characteristic peaks for characterization. (3) N-(1-deoxy-1-fructosyl)phenylalanine, namely N-(1-Deoxy-1-fructosyl)phenylalanine, N-Fructosyl phenylalanine, Fru-Phe.

[0030] Further, the molecular formulas of N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are C1, C2, C3, C4, C5, C6, C7, C8, C9 ... 11 H 21 NO7, C 12 H 23 NO7 and C 15 H 21 NO.7

[0031] Furthermore, the three compounds N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine, and N-(1-deoxy-1-fructosyl)phenylalanine all eluted in the polar region on a C18 reversed-phase column, i.e., the retention time was <3 min, which is consistent with the chromatographic behavior characteristics of glycosylated amino acid compounds, which are highly polar and weakly hydrophobic.

[0032] Furthermore, the individual biomarkers among N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine all exhibited extremely high independent diagnostic efficacy. Their individual ROC curve areas (AUCs) remained consistently between 0.845 and 0.855, with sensitivity reaching 85.3% to 86.3%. Alternatively, when N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are used independently to differentiate between patients with short-term T2DM and non-diabetic individuals, their AUC values ​​are consistently above 0.84, specifically 0.851, 0.855, and 0.845, respectively. Among them, N-(1-deoxy-1-fructose)isoleucine / leucine exhibits the highest monomeric diagnostic efficacy, with an AUC of 0.855. At the optimal threshold of 322,649,157.5, its detection sensitivity for short-term T2DM reaches 86.3%, and its specificity reaches 72.7%.

[0033] Furthermore, the N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine all showed a highly significant moderate positive correlation with FPG and HbA1c. P The value <0.0001 ensures its biological reliability as a marker of glucose metabolism; In the primary target population, i.e., those with normal to mildly impaired renal function and an eGFR > 60 mL / min / 1.73 mcg, 2 In this study, the levels of N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine, and N-(1-deoxy-1-fructosyl)phenylalanine biomarkers were not statistically significant with eGFR, and their molecular formation mechanisms did not depend on the renal urinary glucose excretion process at all; they also showed only a very weak clinical association with ALB (r<0.2).

[0034] The application of plasma free glycosylated amino acid markers as described above in the preparation of T2DM auxiliary diagnostic reagents and / or mass spectrometry detection kits and / or early screening and warning models.

[0035] The application of plasma free glycosylated amino acid markers as described above in the preparation of short-course T2DM auxiliary diagnostic reagents and / or mass spectrometry detection kits and / or early screening and warning models.

[0036] The above-mentioned plasma free glycated amino acid biomarkers are used in the preparation of risk warning models for type 2 diabetes and / or to assist in the research of diagnostic kits and / or drug target research, precision medicine, and pathogenesis research.

[0037] The application of plasma free glycated amino acid markers as described above in the preparation of reagents and / or diagnostic kits to reduce the rate of missed diagnoses of early diabetes.

[0038] The application of plasma free glycated amino acid markers as described above in the preparation of early auxiliary diagnostic reagents and / or kits and / or detection devices for short-course type 2 diabetes mellitus (T2DM).

[0039] Specifically, the relevant preparation and testing methods are as follows: The three plasma-free glycated amino acids mentioned in this invention are as follows: aN-(1-deoxy-1-fructosyl)valine (N-(1-Deoxy-1-fructosyl)valine, N-Fructosylvaline, Fru-Val) bN-(1-Deoxy-1-fructosyl)isoleucine / leucine (N-(1-Deoxy-1-fructosyl)isoleucine / leucine, N-Fructosyl isoleucine / leucine, Fru-Leu / Ile), Note: Because N-(1-Deoxy-1-fructosyl)isoleucine and N-(1-Deoxy-1-fructosyl)leucine are two isomers of each other, they are difficult to completely baseline separate under conventional chromatographic conditions. Therefore, in this invention, they are combined as single detection characteristic peaks for characterization. cN-(1-deoxy-1-fructosyl)phenylalanine (N-(1-Deoxy-1-fructosyl)phenylalanine, N-Fructosyl phenylalanine, Fru-Phe).

[0040] Example 1: Non-targeted metabolomics detection of plasma free glycated amino acid levels in patients with short-term type 2 diabetes and non-diabetic individuals 1.1 Study Subjects: This study recruited 194 non-diabetic individuals and 102 patients with short-term type 2 diabetes mellitus (T2DM). All participants were aged between 20 and 70 years. Participants were diagnosed at the First Affiliated Hospital of Dalian Medical University. The diagnosis of T2DM patients followed the diagnostic criteria of the Chinese Guidelines for the Prevention and Treatment of Type 2 Diabetes Mellitus (2020 edition). Inclusion criteria included: patients in the short-term T2DM group (SD-T2DM group, n=102) who met the criteria of disease duration <1 year and had not received hypoglycemic drug treatment; and patients in the non-diabetic group (NDC group, n=194) who met the criteria of no history of abnormal glucose metabolism, normal fasting blood glucose and glycated hemoglobin at enrollment, and age and sex matched with those in the short-term T2DM group. Exclusion criteria were: (1) type 1 diabetes mellitus or other types of diabetes mellitus; (2) patients with ketosis or ketoacidosis, hyperosmolar coma, or severe stress; (3) patients with renal insufficiency: estimated glomerular filtration rate <60 ml / (min·1.73m 2 (4) Patients with liver dysfunction: ALT or AST > 2.5 times the upper limit of the reference value; (5) Patients with heart failure, myocardial infarction or stroke within the past 6 months; (6) Patients with a history of tumors; (7) Pregnant and lactating patients. Baseline data (age, sex, height, weight, waist circumference, blood pressure) of patients meeting the inclusion criteria were collected, as well as routine examination data upon admission (complete blood count, fasting plasma glucose, glycated hemoglobin, liver and kidney function, blood lipids, etc.). Statistical analysis was performed using SPSS 23.0 software. Normally distributed measurement data were expressed as mean ± standard deviation (SD). Normally distributed continuous data are expressed as medians (interquartile ranges, 1st quartile, 3rd quartile), i.e., M(Q1, Q3); non-normally distributed continuous data are expressed as medians (interquartile ranges, 1st quartile, 3rd quartile), i.e., M(Q1, Q3); count data are expressed as percentages. For normally distributed continuous data, independent samples are used for comparisons between two groups. t Tests were performed; nonparametric tests were used for comparisons between groups of non-normally distributed continuous data; chi-square tests were used for categorical data. See Table 1 for detailed statistical results on clinical baseline data.

[0041] Table 1. Comparison of general clinical data between the short-duration T2DM group and the non-diabetic group.

[0042] Studies have found that fasting plasma glucose (FPG) and glycated hemoglobin (HbA1c) levels in the short-duration type 2 diabetes mellitus (T2DM) group were higher than those in the non-diabetic group. P <0.001). In addition, as shown in Table 1, there were significant differences between the two groups in body mass index, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, alanine aminotransferase, aspartate aminotransferase, and albumin levels.

[0043] 1.2 Detection of free glycated amino acids in plasma using non-targeted metabolomics 1.2.1 Sample Collection and Preprocessing Collect 3-5 mL of fasting venous blood and store it in EDTA tubes. In the preparation of samples for non-targeted metabolomics, the collected whole plasma samples were first centrifuged (4℃, 3000 g, 10 min) to separate the plasma supernatant. The supernatant was then aliquoted and quickly stored at -80℃. During sample pretreatment, the samples needed to be thawed at 4℃. The specific steps were as follows: 100 μL of sample was transferred to a 1 mL 96-well plate, followed by the addition of 400 μL of methanol to precipitate proteins. The sample was thoroughly mixed by vortexing (1000 g, 5 min) and then centrifuged (5300 g, 20 min). After centrifugation, the supernatant (200 μL) was transferred to two 450 μL 96-well plates. The plates were then lyophilized and reconstituted with 80 μL of methanol for use in non-targeted metabolomics analysis.

[0044] 1.2.2 Non-targeted metabolomics detection In this invention, we employed a UHPLC-HRMS system, namely the Ultimate™ 3000 ultra-high performance liquid chromatography (UHPLC) coupled with Q Exactive™ quadrupole-electrostatic field orbital trap high resolution mass spectrometry (HRMS) (ThermoScientific, USA), to analyze metabolomics samples.

[0045] UHPLC separation conditions: Metabolites were separated on an Acquity™ HSS C18 reversed-phase column (2.1 × 100 mm, Waters Co., USA). Mobile phase A was 0.1% aqueous formic acid, and mobile phase B was 0.1% formic acid-acetonitrile solution. The elution gradient was set to an initial 2% B, linearly increasing to 98% B within 10 minutes. The liquid flow rate was maintained at 0.4 mL / min, and the column temperature was kept constant at 50 °C.

[0046] HRMS detection conditions: The quadrupole-electrostatic track-trap high-resolution mass spectrometer was operated with the same ionization parameters in both positive and negative ion modes, equipped with a heated electrospray ionization source. Key ion source parameters were set as follows: sheath gas 45 arb, auxiliary gas 10 arb, capillary temperature 320 °C, heater temperature 355 °C, and S-Lens RF level 55%. Mass spectrometry data acquisition employed a full-scan combined with data-dependent secondary mass spectrometry (MS2) scan mode: in full-scan mode, the mass scan range was set to 70–1000 m / z, the resolution to 70,000 FWHM, the automatic gain control (AGC) target value to 1E6, and the maximum ion implantation time to 200 ms. To obtain high-intensity secondary fragment information for structural identification, this invention employed a Top 10 data-dependent acquisition mode to acquire MS2 data through repeated injections of quality control (QC) samples. In MS / MS data acquisition, the resolution was set to 17,500 FWHM.

[0047] 1.2.3 Metabolomics Data Processing This invention employs Compound Discoverer software to process raw data of polar small molecule metabolites acquired by UHPLC-HRMS, including peak extraction, retention time alignment, peak area integration, and metabolite annotation. For known or annotable metabolites, structural alignment and annotation are performed using the NIST 17 tandem mass spectrometry library, mzCloud spectral library, HMDB database, and KEGG database.

[0048] For the three glycosylated amino acid target compounds of interest in this invention, considering the lack of commercially available reference standards for direct comparison, this invention employs a multi-dimensional evidence combination strategy for metabolite annotation and structural determination: high-resolution accurate mass spectrometry, chromatographic retention behavior, secondary mass spectrometry characteristic fragmentation, and comparison with public database spectra. In this strategy, public databases used for manual comparison of the secondary mass spectra of the target compounds include the mzCloud spectral library, the MoNA (Mass Bank of North America) database, and the HMDB database. Searches revealed that corresponding experimental reference secondary mass spectrometry information for two target compounds could be obtained from these databases, which could be used for spectral comparison and confirmation. For the other target compound (N-(1-deoxy-1-fructosyl)valine), no publicly available experimental reference spectrum for direct confirmation has yet been found. Therefore, a comprehensive determination is made by combining computer-predicted spectra from public databases with the recognized characteristic fragmentation patterns of Amadori rearrangement products.

[0049] Based on the above data processing and metabolite annotation workflow, this invention achieves high-confidence structural annotation of three target glycosylated amino acids in the absence of commercially available reference standards. Detailed identification strategies, primary mass spectrometry information, secondary mass spectrometry fragmentation pattern analysis, and spectral comparison verification results with public databases for the target compounds are provided in section 1.2.4 below.

[0050] To obtain relative quantitative information on metabolites, we used TraceFinder software (ThermoScientific) to extract the area under the curve. During data processing, we conducted rigorous manual checks and individual examinations to reduce the false positive rate. Peak area data for all labeled metabolites were exported to Excel for trimming and processing for subsequent statistical analysis.

[0051] To ensure system stability and data reliability during the analysis of large volumes of clinical samples, this embodiment establishes a rigorous quality control process. Specifically, equal volumes of all plasma samples to be tested are mixed to prepare QC samples, which undergo the same polar metabolite extraction steps as the real samples. During instrument sequential injection, a QC sample is inserted every 10 clinical samples. Data shows that under the chromatographic and mass spectrometric conditions described in this invention, the retention times and response intensities of characteristic peaks in the QC samples remain highly consistent. The relative standard deviations (RSDs) of the three free glycosylated amino acids confirmed by this invention in the QC samples are all less than 20% (the RSDs of N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are 13.218%, 11.136%, and 18.999%, respectively), indicating that the entire analytical system has excellent stability and reproducibility, and the clinical diagnostic biomarker data obtained thereby are authentic and reliable.

[0052] 1.2.4 Structural identification of three glycosylated amino acid biomarkers 1.2.4.1 Identification Strategy Given that there are currently no commercially available reference standards for the three glycosylated amino acids described in this invention for direct comparison, this invention employs a multidimensional evidence combination strategy based on UHPLC-HRMS technology to annotate and identify the target compounds: (1) High-resolution accurate mass and molecular formula confirmation: The high-resolution accurate mass-to-charge ratio (m / z) of the target compound was obtained by primary mass spectrometry, and the parent ion [M+H] was identified. + The quality error is controlled within ±5 ppm, based on which the elemental composition and molecular formula of the target compound can be inferred. (2) Analysis of characteristic fragments by secondary mass spectrometry: Secondary mass spectrometry (MS / MS) fragmentation spectra of the target precursor ion were acquired under low-energy collision-induced dissociation (CID) conditions. Based on the recognized characteristic fragmentation rules of Amadori rearrangement products (1-deoxy-1-fructosyl amino acids) in positive ion mode (including continuous dehydration and neutral loss of the deoxyfructose skeleton, neutral loss of the sugar ring skeleton, and removal of the intact dehydrated hexose group [-162.0528 Da] to generate the protonated molecular ion [AA+H] of the corresponding free amino acid), + The measured fragment ions were assigned based on the characteristic amino acid imine ions generated by the further removal of formic acid from the ion; a detailed analysis of the characteristic fragmentation patterns and the assignment of neutral loss of each fragment ion are provided in section 1.2.4.3. (3) Chromatographic retention behavior as evidence: Combined with the retention behavior of the target compound on the C18 reversed-phase chromatographic column, it is verified that its chromatographic behavior is consistent with the physicochemical properties of glycosylated amino acid compounds, which are highly polar and weakly hydrophobic. (4) Spectral comparison and verification with public databases: The measured secondary mass spectra were compared with international authoritative public metabolomics databases, including the mzCloud spectral library, experimental reference spectra included in the MoNA database, and spectra included in the HMDB database. For target compounds with experimental reference spectra included in public databases, the structure was confirmed by direct comparison between the measured spectra and the experimental reference spectra; for target compounds without experimental reference spectra included in public databases, the structure was determined by comparing the measured spectra with the computer-predicted spectra, combined with the cross-validation of the characteristic fragmentation patterns of Amadori products and the common fragmentation fingerprints among compounds in the same series. (5) Comprehensive confirmation: Based on the above precise mass number, chromatographic retention behavior, secondary mass spectrometry characteristic fragment attribution and database spectrum comparison results, a complete multidimensional evidence chain is constructed to annotate and confirm the structure of the target compound. The identification level corresponds to the Level 2 standard defined by the Metabolomics Standards Initiative (MSI), which is an inferred annotation based on physicochemical properties and / or reference spectrum comparison.

[0053] 1.2.4.2 Primary mass spectrometry and chromatographic behavior analysis Primary mass spectrometry data of three target compounds were acquired in electrospray ionization (ESI) mode. The molecular formula, retention time, theoretical and experimental [M+H] values ​​of the three compounds are listed below. + The precise mass-to-charge ratio, mass error, and secondary debris information are shown in Table 2.

[0054] Table 2. Chromatographic and mass spectrometric identification information of three glycosylated amino acids

[0055] As shown in Table 2, the deviations between the measured mass-to-charge ratios of the three compounds and the theoretical values ​​in the first-order mass spectrometry are all less than 3 ppm, meeting the accuracy requirements for high-resolution mass spectrometry substance identification. Based on this, the molecular formulas of the three compounds can be determined as C1, C2, C3, C4, C5, C6, C7, C8, C9, C10, C11, C21, C12 ...21< / 11 H 21 NO7, C 12 H 23 NO7 and C 15 H 21 NO7 has a completely consistent theoretical molecular composition with N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine and N-(1-deoxy-1-fructosyl)phenylalanine.

[0056] All three compounds elute in the polar region (retention time <3 min) on a C18 reversed-phase column, consistent with the chromatographic behavior of glycosylated amino acid compounds, which are characterized by strong polarity and weak hydrophobicity.

[0057] 1.2.4.3 Fragmentation Pattern of Secondary Mass Spectrometry According to the internationally accepted theory in mass spectrometry and analytical chemistry (see authoritative literature: Davidek T, Kraehenbuehl K, Devaud S, Robert F, Blank I. Analysis of Amadori compounds by high-performance cation exchange chromatography coupled to tandem mass spectrometry. Anal Chem. 2005 Jan 1;77(1):140-7. doi: 10.1021 / ac048925a.PMID: 15623289), the Amadori rearrangement products (1-deoxy-1-fructosyl amino acids) exhibit a series of highly conserved characteristic cleavage pathways in their deoxyfructosyl skeleton under CID conditions: Pathway I: Continuous dehydration of the sugar ring: The protonated precursor ion successively loses 1-2 water molecules (-18.0106 Da, -36.0211 Da); Pathway II: Sugar ring skeleton breakage: Characteristic neutral loss occurs along with sugar ring cleavage, including loss of formic acid (-46.0055 Da, CH2O2), loss of two water molecules plus one carbon monoxide molecule (-64.0160 Da, CH4O3), and loss of three water molecules plus carbon monoxide (-82.0266 Da, CH6O4), etc. Pathway III: Desugaring to generate free amino acids (decisive diagnostic pathway): Neutral loss of one molecule of intact dehydrated hexose (-162.0528 Da, C6H) from the parent ion 10 O5) generates the protonated molecular ion [AA+H] corresponding to the free amino acid. + The ion further loses one molecule of formic acid (-46.0055 Da) to generate an amino acid-characteristic imine ion.

[0058] Based on the precise mass difference (Δm / z) between the parent ion and each fragment ion, this invention provides a rigorous derivation of the fragmentation of three compounds: (1) Structural identification of compound a: N-(1-deoxy-1-fructosyl)valine The actual parent ion residue of compound a in the secondary mass spectrometry is m / z 280.1387. The derivation of the precise mass difference of each fragment is shown in Table 3.

[0059] Table 3. Derivation of the precise mass difference of each fragment in the secondary mass spectra of compound a.

[0060] Where m / z 118.0864 = [Val+H] + (Free valine C5H) 12 NO2 + The ion is used to determine the type of amino acid to which it is attached; further removal of one molecule of formic acid (46.0055 Da) yields an imine ion with m / z 72.0809, characteristic of valine. Based on these findings, compound a is identified as N-(1-deoxy-1-fructosyl)valine.

[0061] (2) Structural identification of the compound: N-(1-deoxy-1-fructosyl)isoleucine / leucine The actual parent ion residue of compound b in the secondary mass spectrometry is m / z 294.1548. The derivation of the precise mass difference of each fragment is shown in Table 4.

[0062] Table 4. Derivation of the precise mass difference of each fragment in the secondary mass spectrometry of compound b.

[0063] Where m / z 132.1020 = [Leu / Ile+H] + (Free leucine / isoleucine C6H) 14 NO2 +The ion is used to determine the type of amino acid to which it is attached; the m / z 86.0964 generated by further removing one molecule of formic acid is the characteristic imine ion of leucine / isoleucine. Since leucine and isoleucine are isomers and their mass spectrometric behavior is completely consistent, compound b is identified as N-(1-deoxy-1-fructosyl)leucine or isoleucine.

[0064] (3) Structural identification of compound c: N-(1-deoxy-1-fructosyl)phenylalanine The actual parent ion residue of compound c in the secondary mass spectrometry is m / z 328.1392. The derivation of the precise mass difference of each fragment is shown in Table 5.

[0065] Table 5. Derivation of the precise mass difference of each fragment in the secondary mass spectrometry of compound c.

[0066] Where m / z 166.0864 = [Phe+H] + (Free phenylalanine C9H) 12 NO2 + The ion is used to determine the type of amino acid to which it is attached; further removal of one molecule of formic acid produces an imine ion with m / z 120.0809, characteristic of phenylalanine. Based on comprehensive identification, compound c is determined to be N-(1-deoxy-1-fructosyl)phenylalanine.

[0067] (4) Cross-validation of the common cleavage features of the three core biomarkers Of particular note is that the three target compounds (a, b, and c) discovered in this invention not only independently and accurately detected the desugaring diagnostic ion (Δm / z = 162.0528 Da) that determines the type of amino acid to which they are linked in their secondary mass spectra, but more importantly, these three compounds perfectly reproduced a set of completely consistent neutral loss spectra of the deoxyfructose backbone: -18.01 Da (-H2O) → -36.02 Da (-2H2O) → -46.01 Da (-HCOOH) → -64.02 Da (-2H2O-CO) → -82.03 Da (-3H2O-CO) This highly conserved and uniform common cleavage fingerprint, exhibited on different amino acid derivatives, is completely consistent with the characteristic cleavage patterns of Amadori compounds reported in the cited authoritative literature. This excludes the possibility of accidental cleavage of a single substance and constitutes a strong network of cross-validation evidence. It irrefutably confirms at the molecular skeleton level that these three markers belong to the same core structure: namely, Amadori rearrangement products of N-(1-deoxy-1-fructosyl) amino acids.

[0068] 1.2.4.4 Comparison and verification with public database maps To further verify the reliability of the above structural identification results, this invention compared the measured secondary mass spectra of the three compounds with authoritative international public metabolomics databases: (1) The measured secondary mass spectrum of compound b (N-(1-deoxy-1-fructosyl)isoleucine / leucine) is in high agreement with the experimental spectrum of N-Fructosyl isoleucine (Accession No. PR310824) in the MoNA database. This reference spectrum was collected by the research group of Tetsuya Mori et al. at RIKEN, Japan, and the relevant methods and data were published in Nature Methods (Tsugawa H., et al., 2019, doi:10.1038 / s41592-019-0358-2).

[0069] (2) The measured secondary mass spectrum of compound c (N-(1-deoxy-1-fructosyl)phenylalanine) is highly consistent with the experimental spectrum of N-Fructosyl phenylalanine (Accession No. PR310826) in the MoNA database. This reference spectrum is also derived from the high-resolution experimental data published by the RIKEN team mentioned above.

[0070] (3) Compound a (N-(1-deoxy-1-fructosyl)valine): Currently, there is no experimental spectrum of this compound in the MoNA database. In this invention, the measured secondary mass spectrum is compared with the computer-predicted spectrum of N-Fructosylvaline in the HMDB database (HMDB ID: HMDB0037844). The main characteristic fragment ions (such as [Val+H]) are compared. + The m / z values ​​(118.0864 and valine imine ion m / z 72.0809) are consistent with the predicted spectra. Furthermore, the fragmentation behavior of this compound completely follows the same Amadori product fragmentation pattern as compounds b and c, forming a complete chain of structural evidence.

[0071] The measured secondary mass spectra of the three compounds are as follows: Figure 1 As shown in Figure 1A: compound a; Figure 1B: compound b; Figure 1 C: Compound C).

[0072] 1.2.4.5 Expert Conclusion Based on the above, the first-level high-resolution accurate mass numbers (mass errors <3 ppm) and the second-level mass spectrometry characteristic fragment assignments (all detected deglycosylation diagnostic ions [AA+H]) were analyzed. + Based on the analysis of the characteristic fragmentation patterns of Amadori products and the comparison results with the spectral data of authoritative international public databases, this invention confirms the structures of three plasma free glycosylated amino acid biomarkers as follows: Compound a: N-(1-deoxy-1-fructosyl)valine (N-(1-Deoxy-1-fructosyl)valine, N-Fructosyl valine, Fru-Val).

[0073] Compound b: N-(1-deoxy-1-fructosyl)isoleucine / leucine (N-(1-Deoxy-1-fructosyl)isoleucine / leucine, N-Fructosyl isoleucine / leucine, Fru-Leu / Ile).

[0074] Compound c: N-(1-deoxy-1-fructosyl)phenylalanine (N-(1-Deoxy-1-fructosyl)phenylalanine, N-Fructosyl phenylalanine, Fru-Phe).

[0075] Based on the Substance Identification Grading Standards of the Metabolomics Standards Initiative (MSI), this invention achieves Level 2 identification for the three compounds mentioned above.

[0076] Example 2: Significant differences were found in the levels of three free glycated amino acids between the short-duration T2DM group and the non-diabetic group. To assess whether there were differences among the three free glycated amino acids in 194 non-diabetic individuals and 102 patients with short-term type 2 diabetes mellitus (T2DM), independent univariate statistical analysis was performed on the three glycated amino acids between the two groups after the extraction of relative quantitative peak area data for the target metabolites. Since normality tests indicated a skewed distribution of the relative peak area data from mass spectrometry quantification, the median and interquartile range (M[Q1, Q3]) were used for statistical description. Nonparametric tests (Mann-Whitney U test) were employed using SPSS 23.0 software to assess differences between groups. Two-tailed tests were used. P A value <0.05 is set as the threshold standard for statistically significant differences.

[0077] Meanwhile, to intuitively reflect the overall trend of biomarkers in the body during the disease pathogenesis, this embodiment calculated the fold change (FC) between groups based on the arithmetic mean of each group of samples. The analysis results are shown in Table 6: the three glycosylated amino acids mentioned above not only showed extremely significant differences between groups ( P The values ​​are all less than 0.0001. Figure 2 The chart shows the differential expression of three free glycated amino acids between the two groups, exhibiting a highly fold upregulation trend in the short-duration T2DM group. Specifically, compared to non-diabetic healthy individuals, the relative levels of these three glycated amino acids in the plasma of short-duration T2DM patients were more than 2-fold higher (FC values ​​were 2.51, 2.27, and 2.45, respectively). These statistical results demonstrate that these three substances are not random background noise, but rather highly specific pathological characteristic molecules, possessing significant application value and translational potential as independent clinical auxiliary diagnostic biomarkers for reagent kit development. Figure 2 As shown.

[0078] Table 6. Relative quantitative characteristics of three glycated amino acids in the two groups and statistical differences between groups.

[0079] Example 3: Clinical diagnostic efficacy evaluation of three free glycated amino acids and their combination To further verify the practical application value of the three free glycated amino acids of this invention in distinguishing short-term T2DM patients from non-diabetic individuals in clinical settings, this embodiment uses receiver operating characteristic (ROC) curve analysis on SPSS 23.0 statistical software to systematically evaluate the independent and combined diagnostic efficacy of these three biomarkers. The combined diagnostic model (three combined panels) is constructed based on a binary logistic regression algorithm, using the relative quantitative values ​​of the three free glycated amino acids as covariates input into the model to extract their combined predictive probabilities, which are then used to plot the combined ROC curve (see...). Figure 3 Meanwhile, this embodiment adopts the Youden exponent maximization principle to determine the optimal diagnostic threshold for each model, and calculates the sensitivity and specificity corresponding to the threshold. The specific diagnostic efficacy results are detailed in Table 7.

[0080] Table 7. Clinical diagnostic efficacy assessment of three single free glycosylated amino acids and three combined Panel diagnostic models.

[0081] Note: The regression coefficients and diagnostic thresholds described in this embodiment are calculated based on a specific clinical cohort sample of this invention. When using the three free glycated amino acids described in this invention for combined diagnosis, those skilled in the art can adaptively recalibrate the regression coefficients and diagnostic thresholds based on their actual detection platform, sample source, and population distribution characteristics, while retaining the combination of the three free glycated amino acid biomarkers. Such adaptive modifications are all within the protection scope of this invention.

[0082] As shown in Table 7, the three free glycated amino acid biomarkers described in this invention, when used independently to differentiate between patients with short-term type 2 diabetes mellitus (T2DM) and non-diabetic individuals, all exhibited AUC values ​​consistently above 0.84 (0.851, 0.855, and 0.845, respectively). Among them, N-(1-deoxy-1-fructosyl) isoleucine / leucine demonstrated the highest monomeric diagnostic efficacy (AUC = 0.855), achieving a detection sensitivity of 86.3% and a specificity of 72.7% for short-term T2DM at the optimal threshold (322649157.5). In the field of diagnostic medicine, a single blood biomarker with an AUC value above 0.8 is generally considered to have good clinical application value. Therefore, any of the glycated amino acids described in this invention can be used as independent biomarkers for developing early auxiliary diagnostic reagents, kits, or detection devices for short-term T2DM.

[0083] The steps of the logistic regression diagnostic model based on the Z-score standardized values ​​of three glycosylated amino acids are as follows: 1. Data Preprocessing Before performing model calculations, it is necessary to first analyze the peak areas of the three detected glycosylated amino acids. Z-score standardization is performed, and the calculation formula is as follows: in, and These represent the mean and standard deviation of the corresponding metabolites in the reference population, respectively.

[0084] 2. Diagnostic model formula The standardized Z-scores of the three glycated amino acids were used as independent variables in a binary logistic regression model, with whether the subject had short-term type 2 diabetes mellitus (T2DM) as the dependent variable (non-T2DM = 0, T2DM = 1). The model yielded the predicted probability of diabetes. Value, will The value was used as a joint diagnostic indicator for ROC analysis. Disease probability ( The calculation formula for ) is as follows: in, e Represents the natural constant. The regression equation is calculated as follows: In the above equation: Z-score for N-(1-deoxy-1-fructosyl)valine; The Z-score for N-(1-deoxy-1-fructosyl)isoleucine / leucine is: Z-score for N-(1-deoxy-1-fructosyl)phenylalanine.

[0085] The results show (see Table 5 and...) Figure 3 The AUC of the combined diagnostic model rose to 0.857 (95% CI: 0.814-0.900), higher than the AUC of any single indicator (0.851, 0.855, 0.845; the combined panel showed no statistically significant difference in ROC compared to other individual indicators). Simultaneously, its specificity (0.737) was also higher than the specificity levels of all three single indicators (0.732, 0.727, 0.691), indicating that the combined panel, while maintaining high sensitivity (0.843), can more effectively reduce the false positive rate. Furthermore, the lower limit of the 95% confidence interval for the combined panel (0.814) was also higher than the lower limit of the confidence interval for any single glycosylated amino acid, demonstrating that the combined diagnostic model has better stability and repeatability in different sample populations.

[0086] Example 4: Correlation analysis of three free glycosylated amino acid markers with clinical biochemical indicators To comprehensively evaluate the clinical independence, efficacy, and stability of the three free glycated amino acids provided in this invention as novel diabetes biomarkers, this embodiment conducted a correlation analysis with clinical biochemical indicators on blood samples from 296 subjects. Spearman correlation analysis was used to explore the intrinsic association between the relative levels of the three biomarkers and traditional gold standard indicators of blood glucose such as FPG and HbA1c, as well as renal function indicators (eGFR) and protein metabolism indicators (ALB). The results are shown in Table 8. The scatter plot of the correlation between the relative quantitative peak areas of the three free glycated amino acids and FPG and HbA1c levels is shown in Table 8. Figure 4 .

[0087] Table 8. Correlation analysis of three plasma free glycated amino acids with multiple clinical indicators.

[0088] Table 8 and Figure 4 As shown, the relative quantitative peak area levels of the three free glycosylated amino acids were all statistically highly significantly positively correlated with FPG (r=0.5673~0.5739) and HbA1c (r=0.5303~0.5484). P The values ​​are all <0.0001. This result strongly confirms, from the perspective of a large sample cohort, that the three biomarkers screened in this invention are closely related to the glucose metabolism disorder status of subjects, and their biological basis as biomarkers for diabetes is solid and reliable. It is particularly noteworthy that the above correlation coefficients are concentrated in the range of 0.53 to 0.57, which is a moderate positive correlation rather than a high degree of collinearity (r>0.8). This non-obvious feature indicates that the biomarkers of this invention are not simple redundant substitutes for FPG or HbA1c, but rather reflect an independent and complementary new physiological dimension in glucose metabolism disorders. Since the in vivo half-life of free amino acids is significantly shorter than that of hemoglobin (the carrier of HbA1c), the biomarkers of this invention are very likely to fill the short-to-medium-term monitoring blind spot between instantaneous blood glucose and long-term average blood glucose, providing incremental diagnostic information that traditional indicators cannot replace.

[0089] eGFR > 60 mL / min / 1.73 m 2 This study covered individuals with normal renal function (eGFR ≥ 90) and those with mild renal impairment (eGFR 60-89, i.e., CKD stages 1-2). This population constitutes the vast majority of patients seeking treatment for type 2 diabetes mellitus (T2DM) and is the primary target population for this invention. In a cohort of 296 subjects who had been pre-excluded from those with moderate to severe renal insufficiency (eGFR < 60), the correlation coefficients (r values) between the three free glycosylated amino acids and eGFR approached 0 (range -0.089 to 0.002). PThe values ​​were all much greater than 0.05 (0.126~0.966), and there was no statistical correlation. The results show that in the main target population of the present invention, the detection results of the biomarker of the present invention are not affected by the physiological fluctuations of eGFR within this range, and the detection results are stable and reproducible. This discovery has significant and unexpected technical advantages over 1,5-AG in the prior art, specifically in the following two aspects. (1) Mechanism level: In the prior art, the blood glucose biomarker ability of 1,5-AG is based on the indirect mechanism of "increased blood glucose → urinary glucose excretion → competitive inhibition of renal tubular reabsorption of 1,5-AG → decrease in blood concentration", which is based on renal excretion. Therefore, it is naturally constrained by the renal glucose threshold and renal tubular function, and there is a risk of missed diagnosis and misjudgment in patients with short-term diabetes with low early blood glucose load and patients with abnormal renal function. In stark contrast, the three free glycated amino acid markers of this invention are primary glycation products directly generated from glucose and free amino acids in the blood through a non-enzymatic Maillard reaction. Their generation process occurs entirely within the blood itself, and at the molecular level, it does not involve the renal process of glucose excretion or reabsorption. Therefore, even in early-stage diabetic patients who have not physiologically reached the renal threshold for glucose, the markers of this invention can still sensitively reflect changes in the glycemic microenvironment, fundamentally overcoming the inherent "renal threshold dependence" bottleneck of 1,5-AG at the mechanistic level. This mechanistic advantage is universal and not limited to the specific eGFR range included in this invention. (2) Stability verification: In individuals with normal to mildly impaired renal function (eGFR>60mL / min / 1.73m... 2 In the study, the correlation analysis results between the levels of the three biomarkers of this invention and eGFR showed that the correlation coefficient r ranged from -0.089 to 0.002. P All values ​​were >0.05, and no statistically significant correlation was observed. This empirical data further confirms that the physiological fluctuations in eGFR within the main target population of this invention are insufficient to constitute meaningful interference with the biomarker detection results, which corroborates the aforementioned mechanistic advantages (1) and jointly supports the robustness of the biomarkers of this invention in routine clinical applications. Based on the comprehensive mechanistic analysis and empirical data, the biomarker combination provided by this invention exhibits detection stability and clinical applicability in its main target population that are unmatched by 1,5-AG.

[0090] The correlation coefficients (r values) between the three free glycosylated amino acids and ALB were only 0.120–0.167. Although the statistical effect is due to the large sample size, the correlation analysis of ALB concentrations at the levels of the three biomarkers in this invention is limited. PThe values ​​showed weak statistical significance, but the Spearman correlation coefficients (r values) were all <0.2. According to the consensus of medical statistics and clinical epidemiology, r < 0.2 is clearly defined as "very weak correlation" or "clinically negligible correlation." This result is also groundbreaking: In existing technologies, GA, as a mid-term blood glucose monitoring indicator, is highly susceptible to interference from albumin synthesis and degradation. This invention innovatively targets "free small-molecule glycosylated amino acids," fundamentally eliminating dependence on the body's macromolecular protein metabolism at the molecular level. Therefore, in T2DM patients with obesity, abnormal liver and kidney function, and other factors affecting albumin levels, it exhibits stability and accuracy unmatched by GA.

[0091] Innovative summary: Based on the above correlation analysis, the present invention has achieved the following unexpected technical effects, thus possessing outstanding substantive features and significant progress: This invention surprisingly discovers that the combination of three free glycated amino acids screened—N-hexosylvaline, N-(1-deoxy-1-fructose) isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine—simultaneously satisfies three key clinical indicators that are difficult to achieve simultaneously with existing technologies: ① High validity: It showed a highly significant moderate positive correlation with both FPG and HbA1c. P The value <0.0001 ensures its biological reliability as a marker of glucose metabolism; ② Dimensional independence: It has a significant but non-collinear correlation with the traditional gold standard for blood glucose, and can provide a new independent diagnostic dimension that reflects the "short-to-medium-term glucose toxicity microenvironment", filling the clinical gap in the existing monitoring window; ③ Robustness against interference: In the main target population of this invention (normal to mildly impaired renal function, eGFR > 60 mL / min / 1.73 mcg), 2 In this invention, the biomarker levels show no statistically significant correlation with eGFR and are not dependent on renal glucose excretion at the molecular level; they also exhibit only a negligible clinical association with ALB (r < 0.2). These characteristics fundamentally overcome the core bottlenecks limiting the clinical application of existing mid-stage blood glucose biomarkers, namely 1,5-AG (renal threshold dependent) and glycated albumin (GA) (albumin turnover interference), at the mechanistic level. Therefore, the biomarker combination provided by this invention demonstrates superior comprehensive performance compared to any single existing indicator in large-sample T2DM populations, making it particularly suitable for early screening, disease progression monitoring, and personalized efficacy evaluation of short-term T2DM, possessing irreplaceable clinical application value and broad industrialization prospects.

[0092] Although embodiments of the invention have been disclosed for illustrative purposes, those skilled in the art will understand that various substitutions, variations, and modifications are possible without departing from the spirit and scope of the invention and the appended claims. Therefore, the scope of the invention is not limited to the contents disclosed in the embodiments.

Claims

1. A plasma free glycated amino acid marker for the auxiliary diagnosis of type 2 diabetes, characterized in that: The biomarker is selected from any one or a combination of three of the following three plasma free glycated amino acids: (1) N-(1-deoxy-1-fructosyl)valine, also known as N-(1-Deoxy-1-fructosyl)valine, N-Fructosylvaline, or Fru-Val; (2) N-(1-deoxy-1-fructosyl)isoleucine / leucine, namely N-(1-Deoxy-1-fructosyl)isoleucine / leucine, N-Fructosyl isoleucine / leucine, Fru-Leu / Ile; Since isoleucine and leucine are isomers, they are difficult to completely separate at baseline under conventional chromatographic conditions. Therefore, they are combined as single detection characteristic peaks for characterization. (3) N-(1-deoxy-1-fructosyl)phenylalanine, namely N-(1-Deoxy-1-fructosyl)phenylalanine, N-Fructosyl phenylalanine, Fru-Phe.

2. The plasma free glycated amino acid marker according to claim 1, characterized in that: The molecular formulas of N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are C1, C2, C3, C4, C5, C6, C7, C8, C9 ... 11 H 21 NO7, C 12 H 23 NO7 and C 15 H 21 NO.7 3. The plasma free glycated amino acid marker according to claim 1, characterized in that: The three compounds, N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine, all eluted in the polar region on a C18 reversed-phase column, i.e., with retention times < 3 min, which is consistent with the chromatographic behavior characteristics of glycosylated amino acid compounds, which are highly polar and weakly hydrophobic.

4. The plasma free glycated amino acid marker according to claim 1, characterized in that: The individual biomarkers of N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine all exhibited extremely high independent diagnostic efficacy. Their individual ROC curve areas (AUCs) remained consistently between 0.845 and 0.855, with sensitivity reaching 85.3% to 86.3%. Alternatively, when N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine are used independently to differentiate between patients with short-term T2DM and non-diabetic individuals, their AUC values ​​are consistently above 0.84, specifically 0.851, 0.855, and 0.845, respectively. Among them, N-(1-deoxy-1-fructose)isoleucine / leucine exhibits the highest monomeric diagnostic efficacy, with an AUC of 0.

855. At the optimal threshold of 322,649,157.5, its detection sensitivity for short-term T2DM reaches 86.3%, and its specificity reaches 72.7%.

5. The plasma free glycated amino acid marker according to any one of claims 1 to 4, characterized in that: The N-(1-deoxy-1-fructose)valine, N-(1-deoxy-1-fructose)isoleucine / leucine, and N-(1-deoxy-1-fructose)phenylalanine all showed a highly significant, moderate positive correlation with FPG and HbA1c. P The value <0.0001 ensures its biological reliability as a marker of glucose metabolism; In the primary target population, i.e., those with normal to mildly impaired renal function and an eGFR > 60 mL / min / 1.73 mcg, 2 In this study, the levels of N-(1-deoxy-1-fructosyl)valine, N-(1-deoxy-1-fructosyl)isoleucine / leucine, and N-(1-deoxy-1-fructosyl)phenylalanine biomarkers were not statistically significant with eGFR, and their molecular formation mechanisms did not depend on the renal urinary glucose excretion process at all; they also showed only a very weak clinical association with ALB (r<0.2).

6. The use of plasma free glycosylated amino acid markers as described in any one of claims 1 to 5 in the preparation of T2DM auxiliary diagnostic reagents and / or mass spectrometry detection kits and / or early screening and warning models.

7. The use of plasma free glycosylated amino acid markers as described in any one of claims 1 to 5 in the preparation of short-course T2DM auxiliary diagnostic reagents and / or mass spectrometry detection kits and / or early screening and warning models.

8. The application of plasma free glycated amino acid biomarkers as described in any one of claims 1 to 5 in the preparation of a risk warning model for type 2 diabetes and / or in assisting in the research of diagnostic kits and / or drug target studies, precision medicine, and pathogenesis studies.

9. The use of the plasma free glycated amino acid markers as described in any one of claims 1 to 5 in the preparation of reagents and / or detection kits for reducing the missed diagnosis rate of early diabetes.

10. The use of plasma free glycated amino acid markers as described in any one of claims 1 to 5 in the preparation of early auxiliary diagnostic reagents and / or kits and / or detection devices for short-course type 2 diabetes mellitus (T2DM).

Images