Method for quantitatively analyzing metabolites in biological cells based on Raman spectrum technology

Abstract

The invention relates to the field of biological cells, and particularly discloses a method for quantitatively analyzing metabolites in biological cells based on a Raman spectrum technology, and the method comprises the following steps: S1, collecting a Raman spectrum data set; s2, metabolite with uniform distribution and metabolite with stable content in the cells to be compared are selected as internal standards; s3, extracting the Raman characteristic peak area of the target object, and performing Raman distribution imaging in the cells according to the Raman characteristic peak area of the target object to obtain Raman imaging pictures of a plurality of substances; s4, performing Raman imaging normalization on the target object; s5, converting the normalized Raman peak intensity of the target object into a thermal image, and presenting Raman signals with different intensities in different colors to obtain an intracellular distribution diagram of each component of the intracellular metabolite; and S6, spatial distribution and quantitative analysis of intracellular metabolites. The method disclosed by the invention has the advantages of in-situ, nondestructive, high speed, quick effect, capability of simultaneously and quantitatively analyzing various metabolites on the subcellular scale and the like.

Description

Technical Field

[0001] This invention relates to the field of biological cell technology, specifically to the field of measuring changes in the content of intracellular metabolites, and particularly to a method for quantitative analysis of intracellular metabolites based on Raman spectroscopy. Background Technology

[0002] Metabolites, as important signaling molecules in organisms, participate in and regulate numerous physiological and biochemical processes inside and outside cells. Visualizing the distribution and changes in the concentration of metabolites is of great significance for studying metabolic pathways, physiological states, disease mechanisms, and nutrient and drug metabolism. Currently, simultaneously performing qualitative and quantitative calculations of metabolites at the subcellular scale remains a technical challenge. Mass spectrometry imaging is an imaging method that can simultaneously detect multiple molecules; however, its spatial resolution limitations make observing subtle changes within cells quite challenging. Furthermore, sample processing can cause some damage to the sample, which undoubtedly increases the difficulty of studying precious samples, such as deep-sea organisms, and limits its effective application at the subcellular scale.

[0003] Raman spectroscopy, capable of resolving multiple metabolites at the subcellular scale, has been widely applied to imaging biomolecules such as nucleic acids, proteins, tissues, and lipids. However, most Raman quantitative methods for organisms require isotope labeling, which necessitates prior knowledge of intracellular metabolic mechanisms and makes simultaneous quantitative analysis of multiple metabolites difficult. In recent years, the internal standard method adopted in the field of Raman spectroscopy quantitative analysis has opened up a new analytical pathway for related problems; by introducing an internal standard as a reference, it makes the quantitative detection of substances in complex systems possible. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a method for quantitative analysis of metabolites in biological cells based on Raman spectroscopy. This method can identify and visualize the distribution of metabolites in situ, efficiently and simply, while quantitatively assessing the changes in the content of multiple metabolites at the subcellular scale.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for quantitative analysis of intracellular metabolites based on Raman spectroscopy includes the following specific steps:

[0007] S1: Acquisition of Raman spectroscopy dataset: Select the optimal power, integration time, and scan step size to ensure that the acquired spectral data has sufficient accuracy and stability. Use Raman spectroscopy to perform area scans on cells or tissues to acquire the spectral dataset of the research object.

[0008] At the subcellular biological scale, submicron-level spatial resolution is achieved to enable precise acquisition of spectral information of intracellular metabolites;

[0009] S2: Selection of internal standards: Based on Raman imaging, select metabolites that are evenly distributed in the cells to be compared as internal standards; then, based on data obtained by metabolomics and other detection methods, select metabolites that are stable in content in the cells to be compared as internal standards.

[0010] S3: Target Raman Peak Area Distribution Imaging: The acquired spectral data is processed to extract the Raman characteristic peak area of ​​the target. Based on the size of the Raman characteristic peak area of ​​the target, Raman distribution imaging is performed in the cell to obtain Raman images of multiple substances, presenting the spatial distribution of the target in the cell and intuitively showing the existence state of metabolites in different regions of the cell.

[0011] S4: Raman Imaging Normalization of Target Object: The Raman image of the target object is normalized based on the Raman signal of the selected internal standard.

[0012] S5: Raman peak thermal imaging analysis and comparison: The normalized Raman peak intensity of the target substance is converted into a thermal image. Raman signals of different intensities are presented in different colors, clearly showing the distribution differences of intracellular metabolites. The intracellular distribution map of each component of intracellular metabolites is obtained. By comparing and analyzing the thermal images of known standard samples or normal cell metabolites, the abnormality and metabolic state of intracellular metabolites in the target cell can be determined.

[0013] S6: Spatial Distribution and Quantitative Analysis of Intracellular Metabolites: By combining the results of normalized Raman imaging with thermal imaging analysis, the spatial distribution and quantitative analysis of intracellular metabolites are performed; the specific content of metabolites in various regions of the cell, as well as their distribution patterns and trends within the cell, are determined, providing a deeper understanding of cellular metabolic activities and physiological states, and offering important information for biomedical research and disease diagnosis.

[0014] As a further technical solution of the present invention, S2 specifically includes:

[0015] S21: Raman imaging selects uniformly distributed substances: Raman imaging technology is used to observe the distribution of different substances in cells. Substances that are uniformly distributed in cells are selected as internal standards for Raman quantification, such as water, which is widely present and relatively uniformly distributed in cells.

[0016] S22: Screening for substances with relatively stable content using metabolomics and other detection methods: By using metabolomics and other detection methods (such as genomics and proteomics), intracellular substances can be analyzed to screen for substances with relatively stable content as internal standards, such as threonine, which is relatively stable in intracellular metabolic processes.

[0017] As a further technical solution of the present invention, S22 specifically includes the following steps (taking a metabolomics detection method as an example):

[0018] S221: Sample processing and metabolomics detection: Take an appropriate amount of cell samples, pre-process them, extract them by sonication, add internal standard for incubation, centrifuge and filter, and use LC-MS and other equipment to detect them to obtain mass spectrometry data;

[0019] a. Take an appropriate amount of cell sample, such as 1g, and perform pretreatment, such as air drying and grinding into powder, and take a certain amount (such as 60mg) for subsequent operations;

[0020] b. Use a mixed solution with a specific ratio (e.g., 0.6 mL of a mixture of methanol and water in a ratio of 7:3) to extract the sample by ultrasound for 30 minutes to ensure full release of intracellular substances;

[0021] c. Add an internal standard, such as L-2-Cl-Phc, at a concentration of 0.3 mg / mL, and incubate at -20°C for 20 minutes;

[0022] d. Centrifuge the extract at 4℃ and 14000r / min for 10 minutes, filter through a 0.2μm filter membrane, and take 200μL of supernatant for metabolomics detection;

[0023] e. Use equipment such as liquid chromatography-mass spectrometry (LC-MS) to perform metabolomics detection and obtain raw mass spectrometry data;

[0024] S222: Data Preprocessing and Quality Control: Convert mass spectrometry data into mzXML format, extract ion peaks using the XCMS program, match the database to determine metabolite qualitative information, and prepare quality control samples to evaluate system stability;

[0025] a. Convert the raw mass spectrometry data into mzXML format, extract ion peaks using the XCMS program, and obtain the primary and secondary spectra of metabolic ions;

[0026] b. Match the metabolite spectra with those in online public databases to determine the qualitative information of the metabolites;

[0027] c. Prepare quality control (QC) samples, that is, mix all the samples to be tested in the same volume, process them in the same pretreatment method as the samples to be tested, and then inject them for testing; run the QC sample 5 times before the sample is tested to balance the system, and then run the QC sample once for every 3 normal samples to measure the stability of the system.

[0028] S223: Principal Component Analysis (PCA): Principal component analysis was performed on the metabolomics data using R4.2.1 software to gain a preliminary understanding of the sample clustering and dispersion, and to identify batch effects or outliers.

[0029] a. Principal component analysis (PCA) was performed on the metabolomics data using R4.2.1 software. By calculating the scores of the samples on the principal components, the high-dimensional data was projected into a low-dimensional space to visualize and analyze the inherent similarities and differences of the dataset.

[0030] b. By analyzing the PCA score plot, we can gain a preliminary understanding of the clustering and dispersion of the samples and determine whether there are obvious batch effects or outliers;

[0031] S224: Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA): Establish an OPLS-DA model to screen differential metabolites, evaluate the model quality through model parameters, cross-validation, CV-ANOVA and permutation test, calculate the metabolite change fold and significance P value, and determine the differential metabolites between groups;

[0032] a. Qualitative data from metabolomics data are orthogonally projected onto latent structures to establish an orthogonal partial least squares discriminant analysis (OPLS-DA) model, which is used to screen for differential metabolites between different groups;

[0033] b. Evaluate model quality through model parameters, where Q 2 R represents the predictability of the model. 2 To indicate the goodness of model fit, Q is typically required. 2 and R 2 The values ​​are all high, indicating that the model has good predictive ability and fitting effect;

[0034] c. The predictive performance of the model was evaluated using 7-fold cross-validation, cross-validation-based ANOVA, and permutation tests (permutation number = 200). The variable importance value (VIP) in the prediction reflects the importance of terms in the model relative to Y (response variable) and X (predictor variable).

[0035] d. Calculate the fold change (FC) and significance P-value of metabolites between different groups based on the relative expression levels of metabolites; when the VIP of a metabolite is greater than 1, P is less than 0.05, and |log2FC| is greater than 1, the expression level of the metabolite can be considered to have a significant difference between groups.

[0036] S225: Relative Standard Deviation (RSD) Calculation: Calculate the RSD of each metabolite in the QC sample and screen out metabolites with good stability and reproducibility based on the threshold;

[0037] a. Calculate the relative standard deviation (RSD) of each metabolite in the QC sample. For each metabolite, first calculate its mean (μ) and standard deviation (σ) in multiple QC tests, and then calculate the RSD using the formula RSD = σ / μ × 100%.

[0038] b. Set the RSD threshold, such as RSD ≤ 30% for metabolites detected by GC-MS and RSD ≤ 20% for metabolites detected by LC-MS. Screen out metabolites whose RSD meets the requirements in QC samples. These metabolites have good stability and reproducibility in the experiment.

[0039] S226: Screening and validation of internal standards: Screen metabolites that meet the conditions as candidates for internal standards, further validate their stability and applicability, and finally determine suitable internal standards;

[0040] Based on the above analysis, substances with relatively stable content are screened from eligible metabolites as potential internal standard candidates. The internal standard can be external or naturally present (in this invention, the naturally present internal standard is used). The screened internal standard candidates are further verified, for example, by repeated testing under different experimental conditions to evaluate their stability and applicability, and finally, a suitable internal standard is determined.

[0041] Through the above steps, metabolomics detection methods can be used to analyze intracellular substances and screen out substances with relatively stable content as internal standards, providing a reliable reference for subsequent quantitative analysis.

[0042] As a further technical solution of the present invention, S3 specifically includes the following steps:

[0043] S31: Data preprocessing: First, fluorescence background correction is performed on the collected spectral data. The background is obtained by using a polynomial fitting algorithm and subtracted from the original spectrum. Then, baseline correction is performed using the Asymmetric Least Squares (ALS) algorithm to eliminate baseline shift by solving the objective function. Finally, the corrected spectrum is normalized, such as by Min-Max normalization or by mapping the spectral data to the [0,1] interval to eliminate light intensity differences under different measurement conditions.

[0044] S32: Raman characteristic peak extraction: Raman characteristic peaks are identified by calculating the first and second derivatives of the normalized spectral data. When the first derivative crosses zero and the second derivative is negative at that point, the maximum point of the Raman peak is determined. After determining the starting and ending wavenumber ranges of the Raman peak, the area of ​​the Raman peak is calculated using the trapezoidal integral method or the Simpson integral method.

[0045] S33: Raman distribution imaging: Determine the spatial extent of the cells and divide them into pixels and grids; associate the Raman characteristic peak area of ​​each sub-region with the corresponding pixel, and distribute the Raman characteristic peak area to each pixel through interpolation to form a Raman distribution imaging map;

[0046] S34: Acquisition of multi-substance Raman imaging: Repeat step S32 to process different substances within the cell and obtain the set of Raman characteristic peak areas for each substance; then use the weighted average method or principal component analysis (PCA) to fuse the Raman distribution images of different substances to form a multi-substance Raman imaging.

[0047] As a further technical solution of the present invention, S32 specifically includes the following steps:

[0048] S321: Raman Peak Identification: Identifying Raman characteristic peaks by calculating the first derivative of the normalized spectral data. Find the point where the derivative crosses zero; this is the candidate location for the bursting peak. Simultaneously, calculate the second derivative. To determine the shape of the Raman peak, if the second derivative is negative at a certain point and the first derivative near that point is zero, then that point can be determined as the maximum point of the Raman peak.

[0049] S322: Calculation of characteristic peak area: After determining the starting and ending wavenumber ranges of the Raman peak, the area of ​​the Raman peak can be calculated. Integration methods include trapezoidal integration and Simpson integration.

[0050] The formula for the trapezoidal integral method is: The formula for Simpson's integral method is: Where: N is the number of data points within the Raman peak range. It is the wavenumber of the i-th data point. It is the interval between adjacent wavenumbers.

[0052] As a further technical solution of the present invention, S33 specifically includes the following steps:

[0053] S331: Pixel-to-Mesh Division: Determine the spatial extent D of the cell and divide it into an M×N pixel-to-mesh grid. Each pixel corresponds to a sub-region, and each sub-region can be represented as P. j (x,y), where j=1,2,…,K,K=M×N is the total number of pixels, and x and y are the x and y coordinates of the pixel in the spatial coordinate system, respectively;

[0054] S332: Feature Peak Area Mapping: Associate the Raman feature peak area in each sub-region with the corresponding pixel; distribute the Raman feature peak area to each pixel using interpolation methods such as nearest neighbor interpolation or bilinear interpolation to form a Raman distribution imaging map;

[0055] The formula for nearest neighbor interpolation is: The formula for bilinear interpolation is: Where: A1, A2, A3, A4 are the Raman characteristic peak areas of the four surrounding points, and x1, x2, y1, y2 are the horizontal and vertical coordinates of the four surrounding points.

[0056] As a further technical solution of the present invention, in step S4, by comparing the Raman signal intensity of the target object and the internal standard, the influence of systematic errors such as instrument drift and sample inhomogeneity is eliminated, thereby improving the quantitative accuracy of Raman imaging and making the measurement results at different times and locations comparable.

[0057] As a further technical solution of the present invention, S5 specifically includes the following steps:

[0058] S51: Thermal Image Conversion: Define a color mapping table, commonly a thermographic color mapping, to map Raman intensity to different colors. The range of the color mapping table starts from the minimum value of the normalized Raman intensity, I. min and maximum value I max The color value C and the normalized Raman intensity I are determined. normalized The relationship is: Where: C min and C max These are the minimum and maximum color values ​​in the color map, corresponding to cool colors (such as blue) and warm colors (such as red);

[0059] S52: Comparative Analysis: Let I sample (x,y) represents the Raman intensity of the target cell at position (x,y), I standard (x,y) represents the Raman intensity of a known standard sample or normal cell at location (x,y); the difference is calculated in Figure I. diff (x,y), the formula is: I diff (x,y)=I sample (x,y)―I standard (x,y);

[0060] Perform statistical analysis on the difference plot and calculate the average difference D. avg (Measures the overall difference between target cells and standard samples or normal cells) and standard deviation D std (Reflecting the dispersion of Lacun intensity differences in the difference map): Where N is the total number of pixels;

[0062] S53: Result Presentation and Interpretation: The color-mapped thermal imaging image can visually display the differences in the distribution of metabolites within cells. Colors from cold to hot (e.g., blue to red) indicate Raman intensity from low to high, meaning the content of metabolites increases from low to high. If the average difference D... avg Significant deviation from zero, or standard deviation D std A larger value indicates an abnormality in the metabolites within the target cell, suggesting that the metabolic state may differ significantly from that of the standard sample or normal cells.

[0063] As a further technical solution of the present invention, in step S6, by statistically analyzing the changes in the average brightness of different regions in the thermal imaging image, that is, by performing statistical analysis based on the brightness values ​​in the image, qualitative and semi-quantitative analysis is further performed at the subcellular scale, and the changes in the content of target metabolites are evaluated based on the numerical differences of different samples, thereby achieving the purpose of quantitative analysis of intracellular metabolites.

[0064] As a further technical solution of the present invention, the quantitative analysis of intracellular metabolites in S6 specifically includes:

[0065] S61: According to the quantitative analysis principle of Raman spectroscopy, the intensity of Raman scattering is linearly related to the concentration of the analyte, which can be expressed by the following formula: R = IKPσC, where: R represents the Raman scattering intensity, I represents the excitation light intensity, K represents the coefficient related to the equipment and sample, P represents the optical path length, σ represents the Raman scattering cross section of the analyte, and C represents the dissolved concentration of the analyte.

[0066] S62: Based on the Raman spectral intensity normalization theory, a quantitative analysis method is established using the internal standard method and the relationship between component concentration and its Raman characteristic peak intensity. The formula can be expressed as follows: A 目标物 / A 内标物 =(C 目标物 / C 内标物 )(σ 目标物 / σ 内标物 )(η 目标物 / η 内标物 )=(C 目标物 / Cinternal standard)(F target / Finternal standard), where: A represents the intensity of the Raman characteristic peak of the analyte (usually determined by the peak area), η represents the photoluminescence efficiency, and F is the quantization factor.

[0067] Therefore, quantification is achieved by using the ratio of the characteristic peak intensity / peak area of ​​the target analyte to that of the internal standard in the imaging diagram.

[0068] The beneficial effects of this invention are as follows: It utilizes confocal Raman spectroscopy to achieve submicron-level spatial resolution at the subcellular biological scale, enabling qualitative and distribution analysis of biological metabolites. It also uses uniformly distributed intracellular substances (such as water and threonine) as internal standards for Raman quantification to quantitatively analyze intracellular metabolites. No sample preparation is required; rapid, non-contact detection can be performed without damaging the sample, achieving non-destructive testing. Compared with traditional mass spectrometry and metabolomics methods, it significantly improves the spatial resolution of intracellular analysis, allowing for more detailed analysis of intracellular substances. It enables spatial visualization of multiple intracellular metabolites while also analyzing their content changes, offering a simple, efficient, and low-cost solution. Attached Figure Description

[0069] Figure 1 This is a flowchart of a method for quantitative analysis of intracellular metabolites based on Raman spectroscopy proposed in this invention;

[0070] Figure 2 The average intracellular spectral pattern and the peak position of phenylalanine;

[0071] Figure 3 This is a schematic diagram showing the distribution of phenylalanine in cells when threonine is used as an internal standard peak. Detailed Implementation

[0072] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0073] Please see the appendix Figure 1 - Appendix Figure 3 A method for quantitative analysis of intracellular metabolites based on Raman spectroscopy includes the following specific steps:

[0074] S1: Acquisition of Raman spectroscopy dataset: Select the optimal power, integration time, and scan step size to ensure that the acquired spectral data has sufficient accuracy and stability. Use Raman spectroscopy to perform area scans on cells or tissues to acquire the spectral dataset of the research object.

[0075] At the subcellular biological scale, submicron-level spatial resolution is achieved to enable precise acquisition of spectral information of intracellular metabolites;

[0076] S2: Selection of internal standards: Based on Raman imaging, select metabolites that are evenly distributed in the cells to be compared as internal standards; then, based on data obtained by metabolomics and other detection methods, select metabolites that are stable in content in the cells to be compared as internal standards.

[0077] S21: Raman imaging selects uniformly distributed substances: Raman imaging technology is used to observe the distribution of different substances in cells. Substances that are uniformly distributed in cells are selected as internal standards for Raman quantification, such as water, which is widely present and relatively uniformly distributed in cells.

[0078] S22: Screening for substances with relatively stable content using metabolomics and other detection methods: By using metabolomics and other detection methods (such as genomics and proteomics), intracellular substances can be analyzed to screen for substances with relatively stable content as internal standards, such as threonine, which is relatively stable in intracellular metabolic processes.

[0079] Taking metabolomics detection methods as an example:

[0080] S221: Sample processing and metabolomics detection: Take an appropriate amount of cell samples, pre-process them, extract them by sonication, add internal standard for incubation, centrifuge and filter, and use LC-MS and other equipment to detect them to obtain mass spectrometry data;

[0081] a. Take an appropriate amount of cell sample, such as 1g, and perform pretreatment, such as air drying and grinding into powder, and take a certain amount (such as 60mg) for subsequent operations;

[0082] b. Use a mixed solution with a specific ratio (e.g., 0.6 mL of a mixture of methanol and water in a ratio of 7:3) to extract the sample by ultrasound for 30 minutes to ensure full release of intracellular substances;

[0083] c. Add an internal standard, such as L-2-Cl-Phc, at a concentration of 0.3 mg / mL, and incubate at -20°C for 20 minutes;

[0084] d. Centrifuge the extract at 4℃ and 14000r / min for 10 minutes, filter through a 0.2μm filter membrane, and take 200μL of supernatant for metabolomics detection;

[0085] e. Use equipment such as liquid chromatography-mass spectrometry (LC-MS) to perform metabolomics detection and obtain raw mass spectrometry data;

[0086] S222: Data Preprocessing and Quality Control: Convert mass spectrometry data into mzXML format, extract ion peaks using the XCMS program, match the database to determine metabolite qualitative information, and prepare quality control samples to evaluate system stability;

[0087] a. Convert the raw mass spectrometry data into mzXML format, extract ion peaks using the XCMS program, and obtain the primary and secondary spectra of metabolic ions;

[0088] b. Match the metabolite spectra with those in online public databases to determine the qualitative information of the metabolites;

[0089] c. Prepare quality control (QC) samples, that is, mix all the samples to be tested in the same volume, process them in the same pretreatment method as the samples to be tested, and then inject them for testing; run the QC sample 5 times before the sample is tested to balance the system, and then run the QC sample once for every 3 normal samples to measure the stability of the system.

[0090] S223: Principal Component Analysis (PCA): Principal component analysis was performed on the metabolomics data using R4.2.1 software to gain a preliminary understanding of the sample clustering and dispersion, and to identify batch effects or outliers.

[0091] a. Principal component analysis (PCA) was performed on the metabolomics data using R4.2.1 software. By calculating the scores of the samples on the principal components, the high-dimensional data was projected into a low-dimensional space to visualize and analyze the inherent similarities and differences of the dataset.

[0092] b. By analyzing the PCA score plot, we can gain a preliminary understanding of the clustering and dispersion of the samples and determine whether there are obvious batch effects or outliers;

[0093] S224: Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA): Establish an OPLS-DA model to screen differential metabolites, evaluate the model quality through model parameters, cross-validation, CV-ANOVA and permutation test, calculate the metabolite change fold and significance P value, and determine the differential metabolites between groups;

[0094] a. Qualitative data from metabolomics data are orthogonally projected onto latent structures to establish an orthogonal partial least squares discriminant analysis (OPLS-DA) model, which is used to screen for differential metabolites between different groups;

[0095] b. Evaluate model quality through model parameters, where Q 2 R represents the predictability of the model. 2 To indicate the goodness of model fit, Q is typically required. 2 and R 2 The values ​​are all high, indicating that the model has good predictive ability and fitting effect;

[0096] c. The predictive performance of the model was evaluated using 7-fold cross-validation, cross-validation-based ANOVA, and permutation tests (permutation number = 200). The variable importance value (VIP) in the prediction reflects the importance of terms in the model relative to Y (response variable) and X (predictor variable).

[0097] d. Calculate the fold change (FC) and significance P-value of metabolites between different groups based on the relative expression levels of metabolites; when the VIP of a metabolite is greater than 1, P is less than 0.05, and |log2FC| is greater than 1, the expression level of the metabolite can be considered to have a significant difference between groups.

[0098] S225: Relative Standard Deviation (RSD) Calculation: Calculate the RSD of each metabolite in the QC sample and screen out metabolites with good stability and reproducibility based on the threshold;

[0099] a. Calculate the relative standard deviation (RSD) of each metabolite in the QC sample. For each metabolite, first calculate its mean (μ) and standard deviation (σ) in multiple QC tests, and then calculate the RSD using the formula RSD = σ / μ × 100%.

[0100] b. Set the RSD threshold, such as RSD ≤ 30% for metabolites detected by GC-MS and RSD ≤ 20% for metabolites detected by LC-MS. Screen out metabolites whose RSD meets the requirements in QC samples. These metabolites have good stability and reproducibility in the experiment.

[0101] S226: Screening and validation of internal standards: Screen metabolites that meet the conditions as candidates for internal standards, further validate their stability and applicability, and finally determine suitable internal standards;

[0102] Based on the above analysis, substances with relatively stable content are screened from eligible metabolites as potential internal standard candidates. The internal standard can be external or naturally present (in this invention, the naturally present internal standard is used). The screened internal standard candidates are further verified, for example, by repeated testing under different experimental conditions to evaluate their stability and applicability, and finally, a suitable internal standard is determined.

[0103] Through the above steps, metabolomics detection methods can be used to analyze intracellular substances and screen out substances with relatively stable content as internal standards, providing a reliable reference for subsequent quantitative analysis.

[0104] S3: Target Raman Peak Area Distribution Imaging: The acquired spectral data is processed to extract the Raman characteristic peak area of ​​the target. Based on the size of the Raman characteristic peak area of ​​the target, Raman distribution imaging is performed in the cell to obtain Raman images of multiple substances, presenting the spatial distribution of the target in the cell and intuitively showing the existence state of metabolites in different regions of the cell.

[0105] S31: Data preprocessing: First, fluorescence background correction is performed on the collected spectral data. The background is obtained by using a polynomial fitting algorithm and subtracted from the original spectrum. Then, baseline correction is performed using the Asymmetric Least Squares (ALS) algorithm to eliminate baseline shift by solving the objective function. Finally, the corrected spectrum is normalized, such as by Min-Max normalization or by mapping the spectral data to the [0,1] interval to eliminate light intensity differences under different measurement conditions.

[0106] S311: Original data correction:

[0107] a. Fluorescence background correction: Raw Raman spectral data Fluorescent background often exists and needs to be removed using algorithms based on polynomial fitting; assuming the fluorescent background can be represented as... Select the region in the spectrum without Raman peaks and obtain the result using polynomial fitting. Then subtract the background from the original spectrum:

[0108] b. Baseline Correction: Raman spectral data are often affected by baseline shift; this can be corrected using the Asymmetric Least Squares (ALS) algorithm, whose objective function is... Where: y i It is the raw spectral data, z i The objective function represents the corrected baseline, n is the number of spectral data points, and λ is a smoothing parameter used to control the degree of correction. The corrected spectral data are obtained by solving this objective function.

[0109] S312: Spectral Normalization: To eliminate differences in light intensity under different measurement conditions, the corrected spectrum needs to be normalized. Normalization methods include Min-Max normalization and mean-standard deviation normalization, etc. The Min-Max normalization formula is: The normalization formula for mean-standard deviation is: Among them: I min and I max These are the minimum and maximum values ​​in the spectral data, and μ and σ are the mean and standard deviation of the spectral data;

[0110] S32: Raman characteristic peak extraction: Raman characteristic peaks are identified by calculating the first and second derivatives of the normalized spectral data. When the first derivative crosses zero and the second derivative is negative at that point, the maximum point of the Raman peak is determined. After determining the starting and ending wavenumber ranges of the Raman peak, the area of ​​the Raman peak is calculated using the trapezoidal integral method or the Simpson integral method.

[0111] S321: Raman Peak Identification: Identifying Raman characteristic peaks by calculating the first derivative of the normalized spectral data. Find the point where the derivative crosses zero; this is the candidate location for the bursting peak. Simultaneously, calculate the second derivative. To determine the shape of the Raman peak, if the second derivative is negative at a certain point and the first derivative near that point is zero, then that point can be determined as the maximum point of the Raman peak.

[0112] S322: Calculation of characteristic peak area: After determining the starting and ending wavenumber ranges of the Raman peak, the area of ​​the Raman peak can be calculated. Integration methods include trapezoidal integration and Simpson integration.

[0113] The formula for the trapezoidal integral method is: The formula for Simpson's integral method is: Where: N is the number of data points within the Raman peak range. It is the wavenumber of the i-th data point. It is the interval between adjacent wavenumbers;

[0114] S33: Raman distribution imaging: Determine the spatial extent of the cells and divide them into pixels and grids; associate the Raman characteristic peak area of ​​each sub-region with the corresponding pixel, and distribute the Raman characteristic peak area to each pixel through interpolation to form a Raman distribution imaging map;

[0115] S331: Pixel-to-Mesh Division: Determine the spatial extent D of the cell and divide it into an M×N pixel-to-mesh grid. Each pixel corresponds to a sub-region, and each sub-region can be represented as P. j (x,y), where j=1,2,…,K,K=M×N is the total number of pixels, and x and y are the x and y coordinates of the pixel in the spatial coordinate system, respectively;

[0116] S332: Feature Peak Area Mapping: Associate the Raman feature peak area in each sub-region with the corresponding pixel; distribute the Raman feature peak area to each pixel using interpolation methods such as nearest neighbor interpolation or bilinear interpolation to form a Raman distribution imaging map;

[0117] The formula for nearest neighbor interpolation is: The formula for bilinear interpolation is: Where: A1, A2, A3, A4 are the Raman characteristic peak areas of the four surrounding points, and x1, x2, y1, y2 are the horizontal and vertical coordinates of the four surrounding points;

[0118] S34: Acquisition of multi-substance Raman imaging: Repeat step S32 to process different substances within the cell and obtain the set of Raman characteristic peak areas for each substance; then use the weighted average method or principal component analysis (PCA) to fuse the Raman distribution images of different substances to form a multi-substance Raman imaging.

[0119] S341: Extraction of characteristic peaks of multiple substances: Repeat step S32 to extract Raman characteristic peaks and calculate their areas for different substances within the cell, obtaining the Raman characteristic peak area set A for each substance. k (k = 1, 2, ..., T, where T is the number of types of substances);

[0120] S342: Image Fusion: This involves fusing Raman distribution images of different substances to form a multi-substance Raman image. Fusion methods include weighted averaging and principal component analysis (PCA). The formula for weighted averaging is... Among them: I k (x,y) is the burst intensity distribution of the k-th substance at pixel (x,y), w k It represents the weight of the k-th substance; the PCA method calculates the covariance matrix of the distribution image of each substance to obtain the principal components, and then performs weighted fusion of the image according to the contribution rate of the principal components.

[0121] S4: Raman Imaging Normalization of Target Object: The Raman image of the target object is normalized based on the Raman signal of the selected internal standard.

[0122] By comparing the Raman signal intensity of the target object and the internal standard, the influence of systematic errors such as instrument drift and sample inhomogeneity is eliminated, thereby improving the quantitative accuracy of Raman imaging and making the measurement results at different times and locations comparable.

[0123] S41: Data Preparation: Let I target (x,y) represents the Raman intensity of the target object at position (x,y); let I internal (x,y) represents the Raman intensity of the internal standard at position (x,y);

[0124] S42: Normalization Processing: Pixel-by-pixel normalization; for each pixel (x, y), calculate the normalized Raman intensity of the target object. Among them: I target (x,y) represents the original Raman intensity of the target object at position (x,y); I internal (x,y) represents the Raman intensity of the internal standard at position (x,y). The selection of the internal standard should meet conditions such as uniform distribution within the cell and chemical stability; I normalized (x,y) represents the Raman intensity of the target at position (x,y) after normalization. This value reflects the intensity change of the target relative to the internal standard and can be used for quantitative analysis.

[0125] This step eliminates systematic errors caused by factors such as instrument drift, laser intensity variation, or sample inhomogeneity by calculating the ratio of the Raman intensity of the target object to the Raman intensity of the internal standard.

[0126] S43: Results Presentation: Normalized Raman images can more accurately reflect the true distribution of the target in cells. Raman images obtained at different time points or under different experimental conditions will be comparable, providing a reliable data basis for subsequent quantitative analysis.

[0127] The above normalization process can effectively improve the accuracy and reliability of Raman imaging, making the application of Raman imaging technology more widespread and in-depth in biomedical research.

[0128] S5: Raman peak thermal imaging analysis and comparison: The normalized Raman peak intensity of the target substance is converted into a thermal image. Raman signals of different intensities are presented in different colors, clearly showing the distribution differences of intracellular metabolites. The intracellular distribution map of each component of intracellular metabolites is obtained. By comparing and analyzing the thermal images of known standard samples or normal cell metabolites, the abnormality and metabolic state of intracellular metabolites in the target cell can be determined.

[0129] S51: Thermal Image Conversion: Define a color mapping table, commonly a thermographic color mapping, to map Raman intensity to different colors. The range of the color mapping table starts from the minimum value of the normalized Raman intensity, I. min and maximum value I max The color value C and the normalized Raman intensity I are determined. normalized The relationship is: Where: C min and C max These are the minimum and maximum color values ​​in the color map, corresponding to cool colors (such as blue) and warm colors (such as red);

[0130] S52: Comparative Analysis: Let I sample (x,y) represents the Raman intensity of the target cell at position (x,y), I standard (x,y) represents the Raman intensity of a known standard sample or normal cell at location (x,y); the difference is calculated in Figure I. diff (x,y), the formula is: I diff (x,y)=I sample (x,y)―I standard (x,y);

[0131] Perform statistical analysis on the difference plot and calculate the average difference D. avg (Measures the overall difference between target cells and standard samples or normal cells) and standard deviation D std (Reflecting the dispersion of Lacun intensity differences in the difference map): Where N is the total number of pixels;

[0132] S53: Result Presentation and Interpretation: The color-mapped thermal imaging image can visually display the differences in the distribution of metabolites within cells. Colors from cold to hot (e.g., blue to red) indicate Raman intensity from low to high, meaning the content of metabolites increases from low to high. If the average difference D... avg Significant deviation from zero, or standard deviation D stdA larger value indicates an abnormality in the metabolites within the target cell, suggesting that the metabolic state may differ significantly from that of the standard sample or normal cells.

[0133] S6: Spatial Distribution and Quantitative Analysis of Intracellular Metabolites: By combining the results of normalized Raman imaging with thermal imaging analysis, the spatial distribution and quantitative analysis of intracellular metabolites are performed; the specific content of metabolites in various regions of the cell, as well as their distribution patterns and trends within the cell, are determined, providing a deeper understanding of cellular metabolic activities and physiological states, and offering important information for biomedical research and disease diagnosis.

[0134] By statistically analyzing the changes in the average brightness of different regions in the thermal imaging image, that is, by performing statistical analysis based on the brightness intensity in the image, further qualitative and semi-quantitative analysis is performed at the subcellular scale. The changes in the content of target metabolites are evaluated based on the numerical differences of different samples, thereby achieving the goal of quantitative analysis of intracellular metabolites.

[0135] S61: According to the quantitative analysis principle of Raman spectroscopy, the intensity of Raman scattering is linearly related to the concentration of the analyte, which can be expressed by the following formula: R = IKPσC, where: R represents the Raman scattering intensity, I represents the excitation light intensity, K represents the coefficient related to the equipment and sample, P represents the optical path length, σ represents the Raman scattering cross section of the analyte, and C represents the dissolved concentration of the analyte.

[0136] S62: Based on the Raman spectral intensity normalization theory, a quantitative analysis method is established using the internal standard method and the relationship between component concentration and its Raman characteristic peak intensity. The formula can be expressed as follows: A 目标物 / A 内标物 =(C 目标物 / C 内标物 )(σ 目标物 / σ 内标物 )(η 目标物 / η 内标物 )=(C 目标物 / Cinternal standard)(F target / Finternal standard), where: A represents the intensity of the Raman characteristic peak of the analyte (usually determined by the peak area), η represents the photoluminescence efficiency, and F is the quantization factor.

[0137] Therefore, quantification is achieved by using the ratio of the characteristic peak intensity / peak area of ​​the target analyte to that of the internal standard in the imaging diagram.

[0138] As can be seen from the above description, the embodiments of the present invention achieve the following technical effects: They utilize confocal Raman spectroscopy to achieve submicron-level spatial resolution at the subcellular biological scale, enabling qualitative and distribution analysis of biological metabolites. Furthermore, they use uniformly distributed intracellular substances (such as water and threonine) as internal standards for Raman quantification to quantitatively analyze intracellular metabolites. This eliminates the need for sample preparation, allowing for rapid, non-contact detection without damaging the sample. Compared to traditional mass spectrometry and metabolomics methods, this significantly improves the spatial resolution of intracellular analysis, enabling more detailed analysis of intracellular substances. It also allows for spatial visualization of multiple intracellular metabolites and analysis of their content changes, offering a simple, efficient, and low-cost solution.

[0139] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.

[0140] This invention is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this specification. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for quantitative analysis of intracellular metabolites based on Raman spectroscopy, characterized in that, The specific steps include the following: S1: Acquisition of Raman spectroscopy dataset: Select the optimal power, integration time, and scanning step size, and use Raman spectroscopy to perform a surface scan on cells or tissues to acquire the spectral dataset of the research object; S2: Selection of internal standards: Based on Raman imaging, uniformly distributed metabolites in the cells to be compared were selected as internal standards; then, based on the data obtained by the metabolomics detection method, metabolites with stable content in the cells to be compared were selected as internal standards. S3: Target Raman Peak Area Distribution Imaging: The acquired spectral data is processed to extract the Raman characteristic peak area of ​​the target. Based on the size of the Raman characteristic peak area of ​​the target, Raman distribution imaging is performed in the cell to obtain Raman images of multiple substances. S4: Raman Imaging Normalization of Target Object: The Raman image of the target object is normalized based on the Raman signal of the selected internal standard. S5: Raman peak thermal imaging analysis and comparison: The normalized Raman peak intensity of the target substance is converted into a thermal image. Raman signals of different intensities are presented in different colors to obtain the intracellular distribution map of each component of the metabolites. By comparing and analyzing the thermal images of known standard samples or normal cell metabolites, the abnormality and metabolic state of the metabolites in the target cells can be determined. S6: Spatial distribution and quantitative analysis of intracellular metabolites: By combining the results of normalized Raman imaging with the results of thermal imaging analysis, the spatial distribution and quantitative analysis of intracellular metabolites are performed.

2. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 1, characterized in that, S2 specifically includes: S21: Raman imaging selects uniformly distributed substances: Raman imaging technology is used to observe the distribution of different substances in cells, and substances that are uniformly distributed in cells are selected as internal standards for Raman quantification. S22: Metabolomics detection methods to screen substances with relatively stable content: Using metabolomics detection methods, intracellular substances are analyzed, and substances with relatively stable content are screened out as internal standards.

3. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 2, characterized in that, S22 specifically includes the following steps: S221: Sample processing and metabolomics detection: Take an appropriate amount of cell samples, pre-process them, extract them by sonication, add internal standard for incubation, centrifuge and filter, and use LC-MS and other equipment to detect them to obtain mass spectrometry data; S222: Data Preprocessing and Quality Control: Convert mass spectrometry data into mzXML format, extract ion peaks using the XCMS program, match the database to determine metabolite qualitative information, and prepare quality control samples to evaluate system stability; S223: Principal Component Analysis: Principal component analysis was performed on the metabolomics data using R4.2.1 software to gain a preliminary understanding of the sample clustering and dispersion, and to identify batch effects or outliers. S224: Orthogonal Partial Least Squares Discriminant Analysis: An OPLS-DA model was established to screen differentially expressed metabolites. The model quality was evaluated through model parameters, cross-validation, CV-ANOVA, and permutation test. The metabolite change fold and significance P-value were calculated to determine the differentially expressed metabolites between groups. S225: Relative Standard Deviation Calculation: Calculate the RSD of each metabolite in the QC sample, and screen out metabolites with good stability and reproducibility based on the threshold; S226: Screening and validation of internal standards: Screen metabolites that meet the conditions as candidates for internal standards, further validate their stability and applicability, and finally determine the appropriate internal standard.

4. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 1, characterized in that, S3 specifically includes the following steps: S31: Data preprocessing: First, fluorescence background correction is performed on the collected spectral data. The background is obtained by using a polynomial fitting algorithm and subtracted from the original spectrum. Next, baseline correction is performed using the Asymmetric Least Squares algorithm to eliminate baseline shift by solving the objective function. Finally, the corrected spectrum is normalized. S32: Raman characteristic peak extraction: Raman characteristic peaks are identified by calculating the first and second derivatives of the normalized spectral data. When the first derivative crosses zero and the second derivative is negative at that point, the maximum point of the Raman peak is determined. After determining the starting and ending wavenumber ranges of the Raman peak, the area of ​​the Raman peak is calculated using the trapezoidal integral method or the Simpson integral method. S33: Raman distribution imaging: Determine the spatial extent of the cells and divide them into pixels and grids; associate the Raman characteristic peak area of ​​each sub-region with the corresponding pixel, and distribute the Raman characteristic peak area to each pixel through interpolation to form a Raman distribution imaging map; S34: Acquisition of multi-substance Raman imaging: Repeat step S32 to process different substances within the cell and obtain the set of Raman characteristic peak areas for each substance; then use the weighted average method or principal component analysis to fuse the Raman distribution images of different substances to form a multi-substance Raman imaging.

5. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 1, characterized in that, S32 specifically includes the following steps: S321: Raman Peak Identification: Identifying Raman characteristic peaks by calculating the first derivative of the normalized spectral data. Find the points where the derivative crosses zero; these are the candidate locations for Raman peaks. Simultaneously, calculate the second derivative I. norm " To determine the shape of the Raman peak, if the second derivative is negative at a certain point and the first derivative near that point is zero, then that point can be determined as the maximum point of the Raman peak. S322: Calculation of characteristic peak area: After determining the starting and ending wavenumber ranges of the Raman peak, the area of ​​the Raman peak can be calculated. Integration methods include trapezoidal integration and Simpson integration. The formula for the trapezoidal integral method is: The formula for Simpson's integral method is: Where: N is the number of data points within the Raman peak range. It is the wavenumber of the i-th data point. It is the interval between adjacent wavenumbers.

6. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 5, characterized in that, S33 specifically includes the following steps: S331: Pixel-to-Mesh Division: Determine the spatial extent D of the cell and divide it into an M×N pixel-to-mesh grid. Each pixel corresponds to a sub-region, and each sub-region can be represented as P. j (x,y), where j=1,2,…,K,K=M×N is the total number of pixels, and x and y are the x and y coordinates of the pixel in the spatial coordinate system, respectively; S332: Feature Peak Area Mapping: Associate the Raman feature peak area in each sub-region with the corresponding pixel; distribute the Raman feature peak area to each pixel using interpolation methods such as nearest neighbor interpolation or bilinear interpolation to form a Raman distribution imaging map; The formula for nearest neighbor interpolation is: The formula for bilinear interpolation is: Where: A1, A2, A3, A4 are the Raman characteristic peak areas of the four surrounding points, and x1, x2, y1, y2 are the horizontal and vertical coordinates of the four surrounding points.

7. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 1, characterized in that, In step S4, by comparing the Raman signal intensity of the target object and the internal standard, the influence of systematic errors such as instrument drift and sample inhomogeneity is eliminated, thereby improving the quantitative accuracy of Raman imaging and making the measurement results at different times and locations comparable.

8. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 1, characterized in that, S5 specifically includes the following steps: S51: Thermal Image Conversion: Define a color mapping table, commonly a thermographic color mapping, to map Raman intensity to different colors. The range of the color mapping table starts from the minimum value of the normalized Raman intensity, I. min and maximum value I max The color value C and the normalized Raman intensity I are determined. normalized The relationship is: Where: C min and C max These are the minimum and maximum color values ​​in the color map, corresponding to cool and warm colors, respectively. S52: Comparative Analysis: Let I sample (x,y) represents the Raman intensity of the target cell at position (x,y), I standard (x,y) represents the Raman intensity of a known standard sample or normal cell at location (x,y); the difference is calculated in Figure I. diff (x,y), the formula is: I diff (x,y)=I sample (x,y)―I standard (x,y); Perform statistical analysis on the difference plot and calculate the average difference D. avg and standard deviation D std : Where N is the total number of pixels; S53: Result Presentation and Interpretation: The color-mapped thermal imaging image can visually display the differences in the distribution of metabolites within cells. Colors from cold to hot indicate Raman intensity from low to high, meaning the content of metabolites increases from low to high. If the average difference D... avg Significant deviation from zero, or standard deviation D std A larger value indicates an abnormality in the metabolites within the target cell, suggesting that the metabolic state may differ significantly from that of the standard sample or normal cells.

9. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 1, characterized in that, In step S6, the changes in the average brightness of different regions in the thermal imaging image are statistically analyzed, that is, the brightness values ​​in the image are statistically analyzed, and further qualitative and semi-quantitative analysis is performed at the subcellular scale. The changes in the content of target metabolites are evaluated based on the numerical differences of different samples, so as to achieve the purpose of quantitative analysis of intracellular metabolites.

10. The method for quantitative analysis of intracellular metabolites based on Raman spectroscopy according to claim 9, characterized in that, In step S6, the quantitative analysis of intracellular metabolites specifically includes: S61: According to the quantitative analysis principle of Raman spectroscopy, the intensity of Raman scattering is linearly related to the concentration of the analyte, which can be expressed by the following formula: R = IKPσC, where: R represents the Raman scattering intensity, I represents the excitation light intensity, K represents the coefficient related to the equipment and sample, P represents the optical path length, σ represents the Raman scattering cross section of the analyte, and C represents the dissolved concentration of the analyte. S62: Based on the Raman spectral intensity normalization theory, a quantitative analysis method is established using the internal standard method and the relationship between component concentration and its Raman characteristic peak intensity. The formula can be expressed as follows: A 目标物 / A 内标物 =(C 目标物 / C 内标物 )(σ 目标物 / σ 内标物 )(η 目标物 / η 内标物 )=(C 目标物 / Cinternal standard)(F target / Finternal standard), where: A represents the intensity of the Raman characteristic peak of the analyte (usually determined by the peak area), η represents the photoluminescence efficiency, and F is the quantization factor.

Images