Search

CN-121978233-A - Non-targeted extract screening and toxicity driving factor identification method based on IDA-UPLC-QTOF-MS/MS

CN121978233ACN 121978233 ACN121978233 ACN 121978233ACN-121978233-A

Abstract

The invention discloses a non-targeted extract screening and toxicity driving factor identification method based on IDA-UPLC-QTOF-MS/MS, which is characterized in that a real extract is obtained by simulating a real use scene and is subjected to chemical analysis, qualitative chemical characteristics are subjected to three-level screening, potential toxicity effect targets and mechanisms of the real extract are revealed through calculation and simulation, few toxicity driving factors which contribute to health risks to the greatest extent are positioned, and a clear direction is provided for subsequent experimental verification, so that the traditional high-flux toxicity screening is greatly replaced, and a large amount of research, development and detection cost is saved.

Inventors

  • LI DANDAN
  • LIU XIAOYUN
  • HE JINSEN
  • WANG FENGHUA
  • GUAN WANLIN
  • YANG MENGYAO
  • XIONG WENJUN

Assignees

  • 海南大学

Dates

Publication Date
20260505
Application Date
20251230

Claims (10)

  1. 1. A non-targeted extract screening and toxicity driver identification method based on IDA-UPLC-QTOF-MS/MS, comprising the steps of: (1) Preparing a sample into sample powder smaller than 500 mu m, and then sequentially carrying out ultra-pure water leaching, filtering and freeze-drying to obtain freeze-dried powder; (2) Analyzing the solution of the freeze-dried powder by adopting ultra-high resolution liquid chromatography-tandem mass spectrometry based on an information dependent acquisition mode to obtain mass spectrometry data containing all detectable chemical characteristics; (3) Obtaining stable characteristic peaks from the mass spectrum data, carrying out matching qualitative and confidence level classification on the stable characteristic peaks and a database, and obtaining physicochemical information of the stable characteristic peaks; (4) Data cleaning is carried out on the chemical features, interference features and false positive signals are removed through data cleaning, and high-confidence chemical features are obtained for subsequent screening; (5) The chemical characteristics were subjected to the following three-stage screening: (5.1) multivariate statistical screening to screen out the high confidence chemical features for a difference factor that has a significant difference between different samples by orthogonal partial least squares discriminant analysis; (5.2) calculating a toxicology preliminary screening, namely, carrying out toxicity prediction on the differential factors, and screening out high risk factors with high toxicity predicted toxicity grades; And (5.3) molecular docking fine screening, namely taking a toxic pathway which is commonly displayed in an activated state by the high risk factor as a core toxic target, taking all the high confidence chemical characteristics as ligands, carrying out molecular docking simulation on the protein structure of the core toxic target, and taking the chemical characteristics with the binding energy lower than a preset value as the toxicity driving factor.
  2. 2. The non-targeted extract screening and toxicity driver identification method of claim 1, wherein the temperature of the ultrapure water extraction is room temperature and the time of the ultrapure water extraction is 7 days.
  3. 3. The method of claim 1, wherein the stable characteristic peak has a mass variance of 10 ppm, a signal intensity variance of 30%, a signal to noise ratio (S/N) of 3 or more, and a peak width range (S) =c (5, 30).
  4. 4. The method of claim 1, wherein the high confidence chemical profile is selected from the group consisting of (1) a confidence Level of Level2 for the metabonomics standard initiative, (2) a retention of a stable characteristic peak of <30% Relative Standard Deviation (RSD) based on quality control, and (3) a lack of target samples in a blank sample but a lack of <50% in the group, or a target sample abundance of 5-fold or more higher than that of the blank sample.
  5. 5. The method for non-targeted extract screening and toxicity driver identification of claim 1, wherein the differential factors are screened for a threshold of variable importance projection >1.5, effect size >1.5, statistical significance <0.01, and abundance fold difference >2.0.
  6. 6. The non-targeted extract screening and toxicity driver identification method of claim 1, wherein step (5.2) performs toxicology prediction by computing a toxicology platform, and then evaluates the toxicity level of the differential factor according to GHS toxicity classification criteria and defines the factor with a predicted toxicity level of 1-2 as a high risk factor.
  7. 7. The method of non-targeted extract screening and toxicity driver identification of claim 5, wherein said computational toxicology platform is ProTox-3.0.
  8. 8. The method of claim 1, wherein the molecular docking simulation is performed using AutoDock Vina software, the preset binding energy value is-7.0 kcal/mol, and the orthorhombic partial least squares discriminant analysis is performed using ropls packages of R language.
  9. 9. The non-targeted extract screening and toxicity driver identification method of claim 1, wherein the sample is a disposable food package and the core toxicity target is a pregnane X receptor.
  10. 10. The non-targeted extract screening and toxicity driver identification method of claim 9, wherein the disposable food package comprises at least one of a polyolefin, a degradable plastic, and a paper-plastic composite.

Description

Non-targeted extract screening and toxicity driving factor identification method based on IDA-UPLC-QTOF-MS/MS Technical Field The invention belongs to the technical field of additive detection, and particularly relates to a non-targeted extract screening and toxicity driving factor identification method based on IDA-UPLC-QTOF-MS/MS. Background Dietary intake is an important route of human exposure to chemical additives. Various chemicals in disposable food packaging materials may migrate into the food product, threatening the health of the consumer. However, the current detection method for the extract has fundamental limitations that firstly, simulation distortion is generated, a conventional organic solvent direct extraction method cannot reflect the migration rule of substances in a real food medium, secondly, a detection blind area depends on a target mode (such as a multi-reaction monitoring mode and a MRM) of a standard substance, and only a small amount of known additives can be covered, so that novel additives, degradation products and unknown impurities cannot be effectively identified. In addition, complex matrix interference seriously affects detection accuracy, and a differential analysis scheme aiming at different materials (such as degradable plastics and paper-plastic composites) is not available. Therefore, there is a need to create an innovative methodology that truly simulates exposure, screens unknown substances comprehensively, and identifies key risk factors accurately. Disclosure of Invention The invention aims to provide a non-targeted extract screening and toxicity driving factor identification method based on IDA-UPLC-QTOF-MS/MS, which is characterized in that real use scenes are simulated, panoramic chemical analysis is carried out on extracts, three-level screening is carried out, potential toxicity effect targets and mechanisms of the extracts are revealed through calculation simulation, few toxicity driving factors which have the greatest contribution to health risks are positioned, and a clear direction is provided for subsequent experimental verification, so that the traditional high-flux toxicity screening is greatly replaced, and a large amount of research, development and detection cost is saved. The technical scheme of the invention is as follows: a non-targeted extract screening and toxicity driver identification method based on IDA-UPLC-QTOF-MS/MS, comprising the steps of: (1) Preparing a sample into sample powder smaller than 500 mu m, and then sequentially carrying out ultra-pure water leaching, filtering and freeze-drying to obtain freeze-dried powder; (2) Analyzing the solution of the freeze-dried powder by adopting a high-resolution liquid chromatography-tandem mass spectrometry based on an information-dependent acquisition mode to obtain mass spectrometry data containing all detectable chemical characteristics; (3) Obtaining stable characteristic peaks from the mass spectrum data, carrying out matching qualitative and confidence level classification on the stable characteristic peaks and a database, and obtaining physicochemical information of the stable characteristic peaks; (4) Data cleaning is carried out on the chemical features, interference features and false positive signals are removed through data cleaning, and high-confidence chemical features are obtained for subsequent screening; (5) The chemical characteristics were subjected to the following three-stage screening: (5.1) multivariate statistical screening, namely screening difference factors with obvious differences among different samples from high-confidence chemical characteristics through orthogonal partial least squares discriminant analysis; (5.2) calculating a toxicology preliminary screening, namely carrying out toxicity prediction on the differential factors, and screening out high risk factors with high toxicity predicted toxicity grades; And (5.3) molecular docking fine screening, namely taking a toxic passage which is commonly shown in an activated state by a high risk factor as a core toxic target, carrying out molecular docking simulation on all high-confidence chemical characteristics serving as a ligand and a protein structure of the core toxic target, and taking the chemical characteristics with the binding energy lower than a preset value as a toxicity driving factor. In some preferred embodiments, the temperature of the ultrapure water extraction is room temperature and the time of the ultrapure water extraction is 7 days. In some preferred embodiments, the stable characteristic peak is extracted by a mass deviation of 10 ppm, a signal intensity deviation of 30%, a signal-to-noise ratio (S/N) > 3, and a peak width range (S) =c (5, 30). In some preferred embodiments, the high confidence chemical features are screened for (1) a Level of confidence of Level2 from the metabonomics standard initiative, (2) retention of a stable characteristic peak of <30% Relative Standard Deviation (RSD)