A method for detecting chemical contaminants in breast milk based on ppar gamma nuclear receptor affinity selection mass spectrometry

Abstract

The present application belongs to the technical field of breast milk biological monitoring, and particularly relates to a method for detecting chemical contaminants in breast milk based on PPAR gamma nuclear receptor affinity selection mass spectrometry, which is used for non-target identification of chemical contaminants in human breast milk, and further evaluates potential PPAR gamma ligands through molecular docking and surface plasmon resonance (SPR). The above detection method links chemical detection directly with receptor-mediated activity, promotes breast milk biological monitoring, makes up for the key knowledge gap of evaluating lactation exposure risk, and provides a basis for discovery and identification of novel contaminants in human breast milk biological monitoring.

Description

Technical Field

[0001] This invention belongs to the field of breast milk biomonitoring technology, specifically relating to a method for detecting chemical contaminants in breast milk based on PPARγ nuclear receptor affinity selection mass spectrometry. Background Technology

[0002] Human breast milk biosurveying is crucial for public health monitoring, particularly for protecting newborns and children. By 2025, more than 279 million chemical substances will be registered globally, with approximately 15,000 new substances added daily. Chemical contaminants can enter the maternal circulation through consumer products, contaminated water, or food, leading to potential exposure. Therefore, monitoring chemical contaminants in breast milk can serve as a valuable indicator of exposure to chemical contaminants in vulnerable populations. Previously, breast milk surveillance relied primarily on targeted analyses to detect specific categories of contaminants, including persistent organic pollutants (POPs), per- and polyfluoroalkyl substances (PFAS), plasticizers, and UV absorbers. However, researchers have found that infants are particularly vulnerable to emerging contaminants that preferentially accumulate in maternal adipose tissue, and accidental exposure during critical developmental windows can cause significant public health problems. Therefore, developing new detection methods to identify and detect some emerging contaminants in breast milk is essential.

[0003] Peroxisome proliferator-activated receptors (PPARs) are a class of ligand-activated transcription factors belonging to the nuclear receptor superfamily. There are three isoforms of PPARs: PPARα, PPARβ / δ, and PPARγ, each encoded by different genes and with distinct functions. They must form heterodimers with retinoid X receptors (RXR) and then bind to specific PPREs (peroxisome proliferator response elements) on target genes to regulate the transcription of those genes. Among the three PPAR isoforms, PPARγ is involved in regulating lipid metabolism, glucose homeostasis, anti-inflammatory responses, and cell differentiation. PPARγ is functionally expressed in breast tissue and is associated with lipid metabolism during lactation, making it a highly relevant molecular target for studying the effects of pollutants in the context of breast milk exposure.

[0004] Because endogenous PPARγ-active components in breast milk, such as natural fatty acids, can competitively bind to receptors and obscure signals from exogenous analytes, identifying high-confidence PPARγ interfering agents is challenging. To systematically assess novel pollutants in breast milk that may disrupt the PPARγ signaling pathway, this invention proposes for the first time the use of PPARγ nuclear receptor affinity-selective mass spectrometry (PPARγ-ASMS) for non-targeted screening of chemical pollutants with PPARγ-binding activity in breast milk samples, establishing a reliable affinity-based non-targeted analytical method for the biomonitoring of novel pollutants in human breast milk. Summary of the Invention

[0005] The purpose of this invention is to develop a method for detecting chemical contaminants in breast milk based on PPARγ nuclear receptor affinity-selective mass spectrometry (PPARγ-ASMS). This method is used for the non-targeted identification of chemical contaminants in human breast milk and further evaluates potential PPARγ ligands through molecular docking and surface plasmon resonance (SPR). This detection method advances breast milk biosurveillance by directly linking chemical detection with receptor-mediated activity, bridging a key knowledge gap in assessing lactational exposure risks, and providing a foundation for the discovery and identification of novel contaminants in human breast milk biosurveillance.

[0006] In a first aspect, the present invention provides a method for detecting chemical contaminants in breast milk based on PPARγ nuclear receptor affinity-selective mass spectrometry, characterized in that the method comprises the following steps:

[0007] 1) Sample preprocessing

[0008] Mix breast milk sample with acetonitrile at a volume ratio of 1:(1-3), add sodium chloride and shake to mix, extract by ultrasonication, centrifuge, collect the upper organic phase, evaporate the solvent, and dissolve in methanol to form breast milk sample pretreatment solution.

[0009] 2) PPARγ-ASMS detection

[0010] At 4-6℃, the pretreated breast milk sample solution was incubated with recombinant nuclear receptor PPARγ fusion protein for 1-2 hours to form a chemical contaminant-protein complex. Histidine nickel magnetic beads were added and incubation continued for 1-2 hours. The magnetic beads were fixed and the supernatant was removed. The magnetic beads were washed with washing buffer and then the chemical contaminant-protein complex was eluted from the magnetic beads with elution buffer. The protein was precipitated with acetone in the eluent. The eluent was extracted 3-5 times with ethyl acetate under vigorous shaking. The extract was collected, the solvent was evaporated to dryness, and the extract was dissolved in methanol for subsequent analysis.

[0011] 3) Non-targeted LC-MS analysis

[0012] Non-targeted analysis of chemical pollutants extracted by the PPARγ-ASMS detection method in step 2) was performed using ultra-high performance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS) on an Orbitrap ID-X system. Raw data were processed using CompoundDiscoverer software, and orthogonal partial least squares discriminant analysis (OPLS-DA) was performed on peak area data using SIMCA-P 13.0.3. Statistically significant features were displayed using volcano plots. Compounds were identified using the mzCloud, mzValult, and ChemSpider databases.

[0013] In some embodiments of the present invention, the recombinant nuclear receptor PPARγ fusion protein described in step 2) is prepared by the following method:

[0014] a1) The target gene shown in SEQ ID NO.2 was cloned into the prokaryotic expression vector pET28a to obtain the recombinant plasmid pET28a-PPARγ;

[0015] a2) Transfect the recombinant plasmid pET28a-PPARγ into competent E. coli cells, select positive clones, passage them in LB solid medium, and then transfer them into LB liquid culture to amplify the strain. When the OD600 is 0.5-1.0, add isopropyl-β-D-thiogalactoside to a final concentration of 0.5 mmol / L to induce the expression of recombinant protein.

[0016] a3) The bacterial cells were disrupted by ultrasonication, and the bacterial supernatant was bound to a nickel column. The recombinant nuclear receptor PPARγ fusion protein was obtained by separating and purifying the protein through the specific binding of the recombinant protein's histidine tag to the nickel column.

[0017] Further, the washing buffer formulation in step 2) is: 20 mM Tris-HCl, 250 mM NaCl, pH 8.0; the elution buffer formulation is: 20 mM Tris-HCl, 500 mM NaCl, 20 mM EDTA, pH 8.0.

[0018] In some preferred embodiments of the present invention, step 1) further includes performing at least one of the following treatments on the upper organic phase after centrifugation:

[0019] g1) The upper organic phase was treated using HLB-P / HMR;

[0020] g2) Freeze the upper organic phase at -80 to -60°C for freezing induction, then centrifuge at a pre-cooled 4-6°C to collect the organic phase;

[0021] g3) Extract the upper organic phase with an equal volume of n-hexane to remove the upper n-hexane and collect the lower organic phase.

[0022] In the most preferred embodiment of the present invention, step 1) specifically involves: mixing breast milk sample with acetonitrile at a volume ratio of 1:(1-3), adding sodium chloride and shaking to mix, ultrasonically extracting, centrifuging, collecting the upper organic phase, treating the upper organic phase with HLB-P / HMR, evaporating the solvent, and dissolving it in methanol to form a breast milk sample pretreatment solution.

[0023] Using the above-described detection method, this invention successfully identified 18 novel chemical contaminants with PPARγ binding activity in collected human breast milk samples. Among them, seven compounds were identified with high confidence: triphenyl phosphate (TPHP), piperine (PIP), 4-tert-butylphenol (PTBP), octyl hydrophthalate (OHP), 3,5-di-tert-butyl-4-hydroxybenzaldehyde (BHT-CHO), dipropylene glycol dimethyl ether (DPGDME), and 4-dodecylbenzenesulfonic acid (4-DBSA). The concentration range of these chemical contaminants was 0.67–303.37 ng / mL, indicating the risk of shared exposure of mothers and infants to the chemical mixture. Notably, this invention is the first to identify 4-DBSA in human breast milk at a concentration of 303.37 ng / mL, highlighting its potential risk in maternal-infant exposure assessment.

[0024] In a second aspect, the present invention provides an application of the PPARγ nuclear receptor affinity-selective mass spectrometry method described in the first aspect of the present invention in the detection of chemical contaminants in breast milk.

[0025] Specifically, the chemical pollutant is a compound with PPARγ nuclear receptor binding activity.

[0026] In a third aspect, the present invention provides the application of a recombinant nuclear receptor PPARγ fusion protein in the detection of chemical contaminants in breast milk, wherein the chemical contaminants are compounds having PPARγ nuclear receptor binding activity.

[0027] The recombinant nuclear receptor PPARγ fusion protein was prepared by the following method:

[0028] a1) The target gene shown in SEQ ID NO.2 was cloned into the prokaryotic expression vector pET28a to obtain the recombinant plasmid pET28a-PPARγ;

[0029] a2) Transfect the recombinant plasmid pET28a-PPARγ into competent E. coli cells, select positive clones, passage them in LB solid medium, and then transfer them into LB liquid culture to amplify the strain. When the OD600 is 0.5-1.0, add isopropyl-β-D-thiogalactoside to a final concentration of 0.5 mmol / L to induce the expression of recombinant protein.

[0030] a3) The bacterial cells were disrupted by ultrasonication, and the bacterial supernatant was bound to a nickel column. The recombinant nuclear receptor PPARγ fusion protein was obtained by separating and purifying the protein through the specific binding of the recombinant protein's histidine tag to the nickel column.

[0031] This invention establishes a PPARγ nuclear receptor-driven affinity-selective mass spectrometry (PPARγ-ASMS) method for the determination of 18 novel contaminants in breast milk samples. The detection method provided by this invention is robust in highly complex biological matrices and represents a promising approach for research and regulatory screening. This is the first successful application of a nuclear receptor-driven, activity-based method to identify exogenous PPARγ ligands in human breast milk. This invention also reports three novel PPARγ active compounds (OHP, BHT-CHO, and DPGDME) for the first time and detects 4-DBSA in the breast milk matrix for the first time. By directly linking chemical detection with receptor-mediated activity, this invention advances breast milk biosurveillance, bridges a key knowledge gap in assessing lactational exposure risks, and establishes a reliable non-targeted analytical method for the discovery of novel contaminants in human breast milk biosurveillance. Attached Figure Description

[0032] Figure 1 Workflow for identifying chemical contaminants in breast milk using PPARγ nuclear receptor affinity selection mass spectrometry (PPARγ-ASMS).

[0033] Figure 2 The figure shows the average recovery rate of the PPARγ group in the matrix compared to the positive control (rosiglitazone) to evaluate the specificity of the protein.

[0034] Figure 3 To assess the matrix effect, the average recovery of rosiglitazone spiked in skim milk and whole milk matrices was evaluated using PPARγ nuclear acceptor affinity selection mass spectrometry (PPARγ-ASMS).

[0035] Figure 4 The volcano plot shows the statistical filtering results for rosiglitazone, which shows significant enrichment (≥3-fold change, p < 0.05) under the applied screening criteria.

[0036] Figure 5Identification of PPARγ active contaminants in human breast milk. (A) Volcano plot of PPARγ-ASMS significantly enriched characteristic ions in breast milk samples. Red and blue dots represent features of PPARγ significantly enriched compared to the control group in positive and negative ionization modes (≥3-fold, p < 0.05), respectively; gray dots indicate insignificant features. (B) Orthogonal partial least squares discriminant analysis (OPLS-DA) scoring plot, showing the separation between the recombinant nuclear receptor PPARγ fusion protein treated sample 1 (red) and the control (blue). (C-F) Chemical structures and MS data of candidate PPARγ active contaminants in breast milk sample 1 from the mzCloud library. 2 Spectra: (C) muscone, (D) n-vinyl-2-pyrrolidone (NVP), (E) dipropylene glycol dimethyl ether (DPGDME) and (F) 4-dodecylbenzenesulfonic acid (4-DBSA).

[0037] Figure 6 The image shows the OPLS-DA score of breast milk. Red represents breast milk treated with PPARγ-ASMS, blue represents the same sample treated with the control group, and AI represents different breast milk samples 2-10.

[0038] Figure 7 A volcano map showing the diverse chemical characteristics of breast milk, AI representing different breast milk samples 2-10.

[0039] Figure 8 High-resolution mass spectrometry data of confirmed PPARγ ligands in human breast milk were used to identify seven compounds in breast milk samples with corresponding MS² spectra (top) and matching real chemical standards (bottom).

[0040] Figure 9 The retention time of the detected exogenous compound is consistent with that of its reference standard.

[0041] Figure 10 The image shows the SPR sensor diagram, illustrating the concentration-dependent binding response of OHP to the immobilized recombinant nuclear receptor PPARγ fusion protein.

[0042] Figure 11 To predict the molecular interactions between the screened compounds and the PPARγ ligand binding domain. (A) TPHP, (B) OHP, (C) PTBP, (D) DPGDME, (E) PIP, (F) 4-DBSA, (G) BHT-CHO.

[0043] Figure 12The results of Coomassie Brilliant Blue staining on SDS-PAGE gels are shown. M: Protein Marker; 1. Bacterial protein before induction of expression; 2. Bacterial protein after induction of expression; 3. Protein in the supernatant after lysis; 4. Precipitate after lysis; 5. Ni2+ column-bound protein; 6. Desalted and purified protein. The target protein (PPARγ) is indicated by the red box.

[0044] Figure 13 The average recovery rate of rosiglitazone spiked under different pretreatment methods is given. Detailed Implementation

[0045] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] Chemicals and reagents

[0047] *E. coli* BL21(DE3) pLysS chemicompetent cells were purchased from TransGen Biotech Ltd., Beijing, China. Histidine nickel magnetic agarose beads were purchased from Xintian Biotechnology Co., Ltd., Shanghai, China. HLB-P / HMR columns were provided by Beijing Nao Technology Co., Ltd. (Beijing). HisTrap HP (5 mL) and HiTrap desalting columns (5 mL) for protein purification were purchased from Cytiva (Uppsala, Sweden). B-PER bacterial protein extraction reagent was purchased from Thermo Fisher Scientific (Shanghai) Co., Ltd.

[0048] Preparation of recombinant nuclear receptor PPARγ fusion protein

[0049] 1) Construction of recombinant plasmid pET28a-PPARγ: The amino acid sequence of the human nuclear receptor PPARγ ligand-binding domain is shown in SEQ ID NO.1. After codon optimization, the target gene shown in SEQ ID NO.2 was obtained and synthesized by a third-party institution. The target gene was cloned into the prokaryotic expression vector pET28a. The recombinant protein expressed by this expression system has a 6×histidine tag fused to its N-terminus.

[0050] SEQ ID NO.1

[0051] HMDLRALAKHLYDSYIKSFPLTKAKARAILTGKTTDKSPFVIYDMNSLMMGEDKIKFKHITPLQEQSKEVAIRIFQGCQFRSVEAVQEITEYAKSIPGFVNLDLNDQVTLLKYGVHEIIYTMLASLMNKDGVLISEGQGFMTREFLKSLRKPFGDFMEPKFEFAVKFNALELDDSDLAIFIAVIILSGDRPGLLNNVKPIEDIQDNLLQALELQLKLNHPESSQLFAKLLQKMTDLRQIVTEHVQLLQVIKKTETDMSLHPLLQEIYKDLE

[0052] SEQ ID NO.2

[0053]

[0054] 2) Expression of recombinant plasmid pET28a-PPARγ in Escherichia coli: The recombinant plasmid was transfected into competent E. coli BL21(DE3)pLysS cells. Positive clones were selected, passaged in LB solid medium, and then transferred to LB liquid medium to amplify the bacterial strain. When the OD600 was about 0.5-1.0, IPTG was added to a final concentration of 0.5 mmol / L to induce the expression of recombinant protein. The supernatant and precipitate of the recombinant bacteria before and after induction, and after induction, were collected by sonication and centrifugation. 2×SDS loading buffer was added for SDS-PAGE electrophoresis.

[0055] 3) Isolation and purification of target protein: After ultrasonic disruption of bacterial cells, the bacterial supernatant was bound to a nickel column. The target protein was obtained by the specific binding of the histidine tag of the recombinant protein to the nickel column and then subjected to SDS-PAGE electrophoresis.

[0056] Specifically, the harvested bacterial pellets underwent mild protein extraction using the B-PER™ bacterial extraction reagent. After centrifugation at 9000 rpm for 20 minutes, the supernatant was purified on a 5 mL HisTrap HP column. Histidine-labeled proteins were eluted using a BioRad NGC chromatography system (Bio-Rad, USA) with elution buffer (20 mM Tris-HCl, 500 mM NaCl, 500 mM imidazole, pH 8.0), followed by desalting using a HiTrap desalting column. Protein expression and purity were verified by SDS-PAGE, and the purified PPARγ protein was used for PPARγ-ASMS detection.

[0057] The results are as follows Figure 12 As shown, SDS-PAGE analysis revealed that the expressed protein has a molecular weight of approximately 34 kDa, with nearly half existing in soluble protein form. The purified protein appeared as a single band on SDS-PAGE, and gel imaging software analysis indicated a purity greater than 90%.

[0058] Source of breast milk samples

[0059] The research protocol involving human subjects was reviewed and approved by the Ethics Committee of the Beijing Center for Disease Control and Prevention (Approval No.: 202037). Breast milk was collected from lactating mothers in Beijing, China. Prior to collection, participants' breasts were thoroughly rinsed with clean water. Breast milk was expressed using a pre-cleaned electric breast pump and quickly transferred to sterile brown glass bottles for storage.

[0060] Breast milk sample pretreatment

[0061] Mix 2 mL of breast milk with 5 mL of acetonitrile, then add 2 g of sodium chloride. After shaking, extract the sample thoroughly by sonication. Centrifuge the sample and clean the upper organic phase using HLB-P / HMR. Collect the organic phase, evaporate it to near dryness under a gentle nitrogen stream, dissolve it in 200 μL of methanol, and store it at -20 °C for subsequent analysis. Prepare the blank control sample using the same procedure, without adding breast milk.

[0062] Detection methods

[0063] PPARγ-ASMS assays were used to screen for chemical contaminants with PPARγ binding activity in breast milk pretreatment solutions. In short, 10 μL of breast milk pretreatment solution was incubated with 1 mg / mL of the recombinant nuclear receptor PPARγ fusion protein prepared according to this invention in an Epppendorf tube at 4°C for 1 hour. Then, 200 μL of nickel histidine magnetic beads (Beyotime, China) were added, and the incubation was continued at 4°C for another hour. The magnetic beads were fixed with a magnetic rack, and the supernatant was removed. The sample was washed three times with an equal volume of wash buffer (20 mM Tris-HCl, 250 mM NaCl, pH 8.0). The protein-ligand complex was eluted from the magnetic beads for 30 minutes with an equal volume of elution buffer (20 mM Tris-HCl, 500 mM NaCl, 20 mM EDTA, pH 8.0). The protein precipitate was then denatured by precipitation with 500 μL of cold acetone. Lipophilic ligands were extracted three times with 3 mL of ethyl acetate under vigorous shaking, each extraction lasting 20 minutes. The organic extracts were combined, evaporated to near dryness under a gentle nitrogen stream, and dissolved in 200 μL of methanol for non-targeted LC-MS analysis. A negative control was prepared by parallel treatment with protein expressed in the empty pET-28a(+) vector.

[0064] The performance of the PPARγ-ASMS assay was validated using the known PPARγ agonist rosiglitazone as a positive control. Specifically, 10 ng / mL rosiglitazone was incubated with 1 mg of the recombinant nuclear receptor PPARγ fusion protein prepared in this invention and processed according to the method described above. To assess the matrix effect during breast milk screening, we performed recovery experiments by injecting rosiglitazone into two different breast milk matrices and calculated the recovery rate of rosiglitazone in each matrix to evaluate the robustness of the method and matrix interference.

[0065] Non-targeted analysis and data processing of chemical pollutants

[0066] Non-targeted analysis of chemical contaminants extracted by PPARγ-ASMS was performed using ultra-high performance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS) on an Orbitrap ID-X system (ThermoFisher™ Scientific, Waltham, MA, USA). Chromatographic separation was performed using ACQUITY Premier CSH C 18 Column (100 mm × 2.1 mm, 1.7 μm; Waters, Milford, USA), temperature 40 °C.

[0067] Raw data were processed using Compound Discoverer software (v.3.3.1.111, Thermo Fisher Scientific). Peak detection, calibration, picking, and deconvolution were performed with the following parameters: retention time (RT) ranging from 0.5 to 20 minutes, mass range of 100 to 1000 Da (m / z), maximum RT shift of 0.5 minutes, mass tolerance of 5 ppm, signal-to-noise ratio (S / N) threshold >3, and minimum peak intensity of 100,000. The adduct ions in the positive ion mode were [M+H]. + [M+Na] + [M+K] + [M+H-HO2] + [M+NH4] + and [M+H+MeOH] + The adduct ions in the negative ion mode are [MH]. − [MH-H2O] − [M-H+HAc] − and [M-H+FA] − Compound identification and molecular formula prediction were performed using the mzCloud, mzValult, and ChemSpider databases.

[0068] To distinguish compounds from the background signal, orthogonal partial least squares discriminant analysis (OPLS-DA) was performed on the peak area data of PPARγ-treated and control samples using SIMCA-P 13.0.3 (Umetrics, Ume, Sweden). A stepwise filtering strategy was employed to reduce false positives caused by nonspecific binding. First, features showing a less than 3-fold increase in abundance (p < 0.05, Student's t-test) in the recombinant nuclear receptor PPARγ fusion protein sample compared to the protein-free blank were excluded. Second, features not showing a ≥3-fold increase in abundance (p < 0.05) compared to the empty vector control protein group were removed, and statistically significant features were displayed using volcano plots.

[0069] After statistical filtering, the features are manually reviewed. Quality errors exceeding ±5 ppm or lacking interpretable MS data are considered defects. 2 The characteristics were excluded. The confidence level for compound identification was assessed using a scaling mechanism established by Schymanski et al. The results were compared with those obtained using real standards (RT and MS). 2 The confirmed match is assigned a first-level confidence level. When MS 2 Assign a second-degree confidence level when the spectrum and spectral library (such as mzCloud) show a high-quality match (≥2 major diagnostic fragments). No MS. 2 Features with insufficient data or spectral matching were not assigned a reliable structure and were considered provisional annotations.

[0070] Molecular docking

[0071] The crystal structure of the PPARγ ligand-binding domain (PDB ID: 3OZ0) was retrieved from the RCSB protein database. The 3D structures of exogenous compounds identified in breast milk samples were obtained from PubChem. Receptors and ligands were prepared in AutoDock Tools. Deleted hydrogen atoms were added, and water of crystallization and non-essential heteroatoms were removed. The processed structures were exported as pdbqt files. Molecular docking was performed using a genetic algorithm with AutoDock 4. A grid box centered at coordinates x=12.379, y=3.803, z=26.641 enclosed the orthogonal binding pocket. The obtained conformations were sorted according to predicted binding energies, and the highest-scoring conformations were visualized in PyMOL 4.6 software for further interaction analysis.

[0072] Surface Plasmon Resonance Measurement

[0073] The binding interaction between the ligand and the purified recombinant nuclear receptor PPARγ fusion protein was analyzed at 25 °C using the Biacore 8K+ system (Cytiva, Uppsala, Sweden). The recombinant nuclear receptor PPARγ fusion protein was immobilized on a Series S CM5 sensor chip (Cytiva) via standard amine conjugation. The running buffer was phosphate-buffered saline (PBS, 20 mM Na2HPO4, 150 mM NaCl, pH 7.4) containing 5% (v / v) dimethyl sulfoxide. The carboxymethyl dextran surface on one side of the flow cell was activated by injection for 7 min with a 1:1 mixture of 0.4 M EDC and 0.1 M NHS. Subsequently, the recombinant nuclear receptor PPARγ fusion protein was covalently conjugated by dilution to a target immobilization level of 10,000 response units (RU) with 10 mM sodium acetate buffer (pH 5.5) and injected for 7 min. The remaining active ester was inactivated by injection of 1.0 M ethanolamine (pH 8.5) for 7 min. Biosensor experiments were conducted at 10 different concentrations, starting at 5 μM and decreasing by a factor of 2 each time until all 10 concentrations were obtained. The binding phase was monitored for 120 seconds, followed by monitoring the dissociated phase for 120 seconds using the run buffer. Since all tested analytes completely dissociated back to baseline within 120 seconds, no surface regeneration step was required between cycles. Sensing maps were processed using Biacore evaluation software (Cytiva) with dual references (subtracting the reference flow cell and buffer blank signals). A 1:1 incorporation kinetic model was fitted to estimate Kd.

[0074] 3. Results and Discussion

[0075] 3.1 Workflow of the PPARγ-ASMS method

[0076] To screen for PPARγ active contaminants, this invention develops a PPARγ nuclear receptor affinity-selective mass spectrometry (PPARγ-ASMS) detection method. The recombinant nuclear receptor PPARγ fusion protein prepared in this invention is co-incubated with breast milk extract to form receptor-ligand complexes. These complexes are then separated using nickel magnetic beads, the bound ligands are eluted, and analyzed by non-targeted high-resolution mass spectrometry (HRMS). Figure 1 The performance of the PPARγ-ASMS assay was validated using the known agonist rosiglitazone as a positive control. The results showed that approximately 75% of the spiked rosiglitazone was successfully recovered, confirming the functional activity of the recombinant nuclear receptor PPARγ fusion protein prepared in this invention. Figure 2 ).

[0077] The matrix effect was assessed by adding rosiglitazone to skim milk and whole milk. The recovery rate was significantly lower in whole milk. Figure 3This indicates that endogenous milk components compete for the PPARγ binding site, a finding consistent with many endogenous ligands we later identified. To mitigate this interference, the inventors optimized the pretreatment process for breast milk samples multiple times.

[0078] 3.2 Non-targeted identification of chemical contaminants in human breast milk by PPARγ-ASMS

[0079] To enhance the identification of potential PPARγ ligands in the complex breast milk matrix, we rigorously filtered out non-targeted screening data. This step-by-step approach was implemented to minimize false positives and increase confidence in candidate compounds. The effectiveness of the screening conditions is as follows: Figure 4 As shown, spiked positive control rosiglitazone was successfully and specifically detected in the milk matrix. These results validate the PPARγ-ASMS platform as a reliable method for the non-targeted discovery of exogenous bioactive substances in human breast milk samples.

[0080] Taking breast milk sample 1 as an example, its non-targeted HRMS analysis initially yielded 11,706 molecular features (7,151 in positive ion mode and 4,555 in negative ion mode); the OPLS-DA model showed that the down fraction of the recombinant nuclear receptor PPARγ fusion protein was significantly separated from the control in the chemical spectrum. Figure 5 B). After applying a strict statistical filter (≥3-fold enrichment, p<0.05), 1279 features (653 positive, 626 negative) were retained as candidate PPARγ binders. Figure 5 Visualization of the volcano plot shown in Figure A. Subsequently, MS was analyzed using the mzCloud database. 2 The spectra were manually processed to obtain credible identifications of 25 compounds (Schymanski confidence level 2), as shown in Table 1.

[0081] Table 1. Matching MS of characteristic components in human breast milk 2 RT and corresponding chemical information

[0082]

[0083] Based on MS 2 High-quality matching of spectra with the mzCloud database identified four exogenous compounds in breast milk sample 1 (Schymanski confidence level 2): ​​muscone, n-vinyl-2-pyrrolidone (NVP), dipropylene glycol dimethyl ether (DPGDME), and 4-dodecylbenzenesulfonic acid (4-DBSA). Figure 5(CF). These annotations provide targeted candidates for subsequent confirmatory analyses and exposure quantification.

[0084] The OPLS-DA model showed a significant separation between the PPARγ-LBD experimental group and the control group in breast milk samples 2-10. Figure 6 This confirmed the significant differences in their respective chemical spectra. Subsequent volcano plot analysis identified 451, 232, 990, 1462, 1667, 2624, 2087, 900, and 886 significantly enriched features (≥3-fold enrichment, p < 0.05) in samples 2-10, respectively. Figure 7 These results confirm that the PPARγ-ASMS detection method provided by this invention can consistently achieve selective affinity enrichment and HRMS-based characterization of potential receptor active ligands in different sample sets, laying a solid foundation for compound identification. Then, MS data with these significant enrichment features were queried in the mzCloud database. 2 Spectrum, and provide structural annotations.

[0085] MS of 18 putative exogenous bioactive compounds detected in mzCloud after manual processing to consolidate redundant features. 2 A high-confidence match was observed. These 18 candidate ligands were then validated using real chemical standards. For seven compounds—triphenyl phosphate (TPHP), piperine (PIP), 4-tert-butylphenol (PTBP), octyl phthalate hydrogen ester (OHP), 3,5-di-tert-butyl-4-hydroxybenzaldehyde (BHT-CHO), dipropylene glycol dimethyl ether (DPGDME), and 4-dodecylbenzenesulfonic acid (4-DBSA)—retention times (RTs) and MS were measured. 2 Spectral matching ( Figure 8 and Figure 9 The concentrations of these confirmed compounds, quantified by single-point external standard correction, are as follows: TPHP (4.61 ng / mL), PIP (0.67 ng / mL), PTBP (qualitative detection, not quantified), OHP (125.67 ng / mL), BHT-CHO (37.37 ng / mL), DPGDME (56.41 ng / mL), and 4DBSA (303.37 ng / mL).

[0086] TPHP is a well-known organophosphorus flame retardant (OPFR) frequently detected in human placenta and breast milk. TPHP accumulates in placental tissue, activates PPARγ in human trophoblast cells, and disrupts lipid metabolism and steroid production. Considering the approximately 3.5% fat content in breast milk, the TPHP concentration measured in this study reached 131.7 ng / g lipids, exceeding the highest level reported in European breast milk samples from 1997–2007 (11 ng / g lipids). This elevated level reflects increased environmental and human exposure to OPFRs and underscores the need for continued biosurveillance and risk assessment.

[0087] This invention marks the first detection of 4-dodecylbenzenesulfonic acid (4-DBSA) in human breast milk, a key component of linear alkylbenzenesulfonic acid (LAS) surfactants used in detergents and cleaning products. The estimated concentration of 303.37 ng / mL indicates substantial maternal exposure, possibly from the use of household products or indoor dust. The presence of 4-DBSA in milk suggests its ability to cross biological barriers and transfer to infants, highlighting a previously unrecognized exposure route that warrants further toxicological studies.

[0088] OHP, BHT-CHO, and DPGDME are the first compounds reported to possess PPARγ activity. OHP is the major hydrolytic metabolite of dioctyl phthalate (DnOP). Due to the widespread use of DnOP in polyvinyl chloride (PVC) flooring, sealants, coatings, and consumer products, human exposure to OHP is primarily indirect. Major routes of exposure include dietary intake through food contact materials, ingestion of contaminated dust, inhalation of indoor emissions, skin contact, and occupational handling. While toxicological assessments have historically focused on the maternal plasticizer DnOP, data on its monoester metabolite OHP remain limited. The detection of elevated OHP levels (125.67 ng / mL) in breast milk not only confirms maternal and infant exposure to DnOP but also provides important biomarkers for assessing the toxicokinetics of human DnOP and the lactational transfer of its bioactive metabolites.

[0089] BHT-CHO is a transformation product of the hindered phenolic antioxidant BHT (2,6-di-tert-butyl-4-methylphenol). Human exposure to BHT-CHO is second only to BHT in its widespread use in polymers, rubber, coatings, lubricants, food contact materials, and personal care products. The high level of BHT-CHO detected in the Beijing samples (37.37 ng / mL) indicates potentially high exposure and underscores the need for expanded biosurveillance and source allocation of BHT-CHO in lactating populations.

[0090] DPGDME is a p-series glycol ether used as a coagulating solvent in water-based paints, inks, and cleaning agents, and can potentially enter the human body through inhalation or skin contact in indoor environments. To our knowledge, this is the first report of DPGDME in human breast milk or any human biological matrix, and the contamination level of 56.41 ng / mL indicates that this solvent can enter the systemic circulation and be transferred to infants through breastfeeding. This novel finding underscores the urgent need to identify sources of exposure for breastfeeding women (e.g., home or occupational product use), characterize its toxicokinetics and lactational transfer, and assess the potential health risks of early-life exposure.

[0091] Furthermore, this invention utilized SPR to confirm the binding activity of OHP, BHT-CHO, and DPGDME for PPARγ. OHP exhibited strong binding, with an equilibrium dissociation constant (Kd) of 0.149 µM ( Figure 10 This suggests that they may interact with regulatory sites such as the activating function-2 (AF2) domain. In contrast, neither BHT-CHO nor DPGDME yielded measurable binding signals under experimental conditions, indicating that their binding affinity is below the detection limit of SPR.

[0092] To further evaluate potential binding modes, this invention used AutoDock Vina to perform molecular docking simulations on the seven identified exogenous compounds, allowing them to enter the crystal structure of the PPARγ ligand binding domain (PDB: 3OZ0). Results Figure 11 As shown, all compounds dock within the canonical ortho-pocket, interacting through hydrogen bonding, hydrophobic contacts, and π-π / π-cation stacking. The docking scores range from -3.6 to -7.8 kcal / mol, in the following order: PIP (-7.8) > OHP (-7.3) > TPHP (-7.1) > BHT-CHO (-6.6) > PTBP (-6.4) > 4-DBSA (-5.1) > DPGDME (-3.6).

[0093] Combined with SPR (Kd = 0.149 µM) data, the predicted docking of OHP (-7.3 kcal / mol) confirmed its role as a direct ligand for PPARγ. Despite limited interaction, the intermediate score of BHT-CHO was consistent with weak or transient binding under SPR detection. The consistency between the predicted interaction modes and experimental SPR data provides a reasonable framework for prioritizing compounds in further functional and toxicological evaluations.

[0094] Optimization of breast milk sample pretreatment process

[0095] Preprocessing method 1

[0096] Take 2 ml of breast milk, add 10 μl of 2 μg / ml rosiglitazone standard solution, shake well, let stand for 10 min, mix the sample with 5 ml of acetonitrile, add 2 g of sodium chloride, shake well, extract by sonication, centrifuge, collect the upper organic phase, evaporate the solvent to dryness, and dissolve in methanol to form a breast milk sample pretreatment solution.

[0097] Preprocessing method 2

[0098] Take 2 ml of breast milk, add 10 μl of 2 μg / ml rosiglitazone standard solution, shake well, let stand for 10 minutes, mix the sample with 5 ml of acetonitrile, add 2 g of sodium chloride, shake well, extract by sonication, centrifuge, collect the upper organic phase, treat the upper organic phase with HLB-P / HMR, evaporate the solvent, and dissolve in methanol to form a breast milk sample pretreatment solution.

[0099] Preprocessing method 3

[0100] Take 2 ml of breast milk, add 10 μl of 2 μg / ml rosiglitazone standard solution, shake well, let stand for 10 minutes, mix the sample with 5 ml of acetonitrile, add 2 g of sodium chloride, shake well, extract by sonication, centrifuge, collect the upper organic phase, freeze the organic phase at -80℃ for 1-2 hours for freezing induction, centrifuge at 10000 rpm for 10 minutes in a pre-cooled 4℃ centrifuge, collect the organic phase, evaporate the solvent to dryness, and dissolve in methanol to form a breast milk sample pretreatment solution.

[0101] Preprocessing method 4

[0102] Take 2 ml of breast milk, add 10 μl of 2 μg / ml rosiglitazone standard solution, shake well, let stand for 10 minutes, mix the sample with 5 ml of acetonitrile, add 2 g of sodium chloride, shake well, extract by sonication, centrifuge, collect the upper organic phase, extract the upper organic phase with an equal volume of n-hexane, remove the upper n-hexane after extraction, evaporate the lower organic solvent to dryness, and dissolve in methanol to form a breast milk sample pretreatment solution.

[0103] The samples obtained by the pretreatment methods 1-4 described above were used as breast milk sample pretreatment solutions. 1 mg / mL of the recombinant nuclear receptor PPARγ fusion protein prepared according to the aforementioned method was added and incubated. The recovery rate of rosiglitazone in the breast milk samples under the above four pretreatment methods was calculated. The results are as follows: Figure 13As shown, different pretreatment methods for breast milk samples affect the recovery rate of rosiglitazone. Experimental data show that the HLB-P / HMR treatment method (pretreatment method 2) not only achieved a near 100% highest recovery rate but also exhibited minimal error, demonstrating excellent extraction precision. In contrast, hexane extraction (pretreatment method 4) easily leads to the loss of the target substance along with lipids, resulting in the lowest recovery rate. While the low-temperature induction method (pretreatment method 3) achieved a higher recovery rate, its reproducibility was poor. The non-defatted group (pretreatment method 1) showed significant matrix effect loss. In summary, when processing complex dietary samples rich in lipids, HLB-P / HMR treatment perfectly balances efficient defatting with efficient retention of the target active substance, making it the ideal pretreatment method to ensure the accuracy and sensitivity of subsequent non-targeted screening methods.

[0104] The above specific embodiments are merely illustrative of the present invention and do not represent a limitation thereof. Those skilled in the art will recognize that other variations of the specific embodiments of the present invention are possible.

Claims

1. A method for detecting chemical contaminants in breast milk based on PPARγ nuclear receptor affinity-selective mass spectrometry, characterized in that, The method includes the following steps: 1) Sample preprocessing Mix breast milk sample with acetonitrile at a volume ratio of 1:(1-3), add sodium chloride and shake to mix, extract by ultrasonication, centrifuge, collect the upper organic phase, evaporate the solvent, and dissolve in methanol to form breast milk sample pretreatment solution. 2) PPARγ-ASMS detection At 4-6℃, the pretreated breast milk sample solution was incubated with recombinant nuclear receptor PPARγ fusion protein for 1-2 h to form a chemical contaminant-protein complex. Histidine nickel magnetic beads were added and incubated for another 1-2 h. The magnetic beads were fixed and the supernatant was removed. The magnetic beads were washed with washing buffer and then the chemical contaminant-protein complex was eluted from the magnetic beads with elution buffer. The protein was precipitated with acetone in the eluent. The eluent was extracted 3-5 times with ethyl acetate under vigorous shaking. The extract was collected, the solvent was evaporated to dryness, and the extract was dissolved in methanol for subsequent analysis. 3) Non-targeted LC-MS analysis Non-targeted analysis of chemical pollutants extracted by the PPARγ-ASMS detection method in step 2) was performed using ultra-high performance liquid chromatography-high resolution mass spectrometry on an Orbitrap ID-X system. Raw data were processed using Compound Discoverer software, and orthogonal partial least squares discriminant analysis (OPLS-DA) was performed on peak area data using SIMCA-P 13.0.

3. Statistically significant features were displayed using volcano plots. Compounds were identified using the mzCloud, mzValult, and ChemSpider databases.

2. The method according to claim 1, characterized in that, The recombinant nuclear receptor PPARγ fusion protein described in step 2) was prepared by the following method: a1) The target gene shown in SEQ ID NO.2 was cloned into the prokaryotic expression vector pET28a to obtain the recombinant plasmid pET28a-PPARγ; a2) Transfect the recombinant plasmid pET28a-PPARγ into competent E. coli cells, select positive clones, passage them in LB solid medium, and then transfer them into LB liquid culture to amplify the strain. When the OD600 is 0.5-1.0, add isopropyl-β-D-thiogalactoside to a final concentration of 0.5 mmol / L to induce the expression of recombinant protein. a3) The bacterial cells were disrupted by ultrasonication, and the bacterial supernatant was bound to a nickel column. The recombinant nuclear receptor PPARγ fusion protein was obtained by separating and purifying the protein through the specific binding of the recombinant protein's histidine tag to the nickel column.

3. The method according to claim 1, characterized in that, The washing buffer formulation described in step 2) is: 20 mM Tris-HCl, 250 mM NaCl, pH 8.

0.

4. The method according to claim 1, characterized in that, The elution buffer formulation described in step 2) is: 20 mM Tris-HCl, 500 mM NaCl, 20 mM EDTA, pH 8.

0.

5. The method according to claim 1, characterized in that, Step 1) further includes performing at least one of the treatments shown in g1)-g3) on the upper organic phase after centrifugation: g1) The upper organic phase was treated using HLB-P / HMR; g2) Freeze the upper organic phase at -80 to -60°C for freezing induction, then centrifuge at a pre-cooled 4-6°C to collect the organic phase; g3) Extract the upper organic phase with an equal volume of n-hexane to remove the upper n-hexane and collect the lower organic phase.

6. The method according to claim 1, characterized in that, Step 1) Specifically, the breast milk sample is mixed with acetonitrile at a volume ratio of 1:(1-3), sodium chloride is added and shaken to mix, ultrasonic extraction is performed, centrifugation is performed, the upper organic phase is collected, the upper organic phase is treated with HLB-P / HMR, the solvent is evaporated, and the solution is dissolved in methanol to form a breast milk sample pretreatment solution.

7. The application of the PPARγ nuclear receptor affinity-selective mass spectrometry method according to any one of claims 1-6 in the detection of chemical contaminants in breast milk.

8. The application according to claim 7, characterized in that, The chemical pollutant is a compound with PPARγ nuclear receptor binding activity.

9. Application of recombinant nuclear receptor PPARγ fusion protein in the detection of chemical contaminants in breast milk, wherein the recombinant nuclear receptor PPARγ fusion protein is prepared by the following method: a1) The target gene shown in SEQ ID NO.2 was cloned into the prokaryotic expression vector pET28a to obtain the recombinant plasmid pET28a-PPARγ; a2) Transfect the recombinant plasmid pET28a-PPARγ into competent E. coli cells, select positive clones, passage them in LB solid medium, and then transfer them into LB liquid culture to amplify the strain. When the OD600 is 0.5-1.0, add isopropyl-β-D-thiogalactoside to a final concentration of 0.5 mmol / L to induce the expression of recombinant protein. a3) The bacterial cells were disrupted by ultrasonication, and the bacterial supernatant was bound to a nickel column. The recombinant nuclear receptor PPARγ fusion protein was obtained by separating and purifying the protein through the specific binding of the recombinant protein's histidine tag to the nickel column.

10. The application according to claim 9, characterized in that, The chemical pollutant is a compound with PPARγ nuclear receptor binding activity.

Images