A method to simultaneously visualize and quantify cellular accumulation of metal nanoparticles at the single-cell level

Abstract

The present invention discloses a method for simultaneously visualizing and quantifying cell-accumulated metal nanoparticles at a single-cell level, comprising the following steps: first, dividing cell samples that have accumulated a series of metal nanoparticles in different concentrations into two groups, one group using ICP-MS to quantify the metal nanoparticles accumulated in the cells, and the other group using hyperspectral technology to quantify the metal nanoparticles accumulated in the cells; second, establishing an absolute quantitative standard curve based on the corresponding relationship between the two quantitative results in step one; and third, calculating the ratio of the area of ​​the metal nanoparticles in a single cell to the area of ​​the cell in a hyperspectral image, and based on the absolute quantitative standard curve in step two, absolutely quantifying the metal nanoparticles accumulated in the single cell.

Description

Technical Field

[0001] The invention belongs to the technical field of nano material detection, and in particular relates to a method for simultaneously visualizing and absolutely quantifying cell-accumulated metal nanoparticles at a single cell level. Background Art

[0002] AuNPs particles (AuNPs) are widely used in biomedical fields such as bio-optical imaging, biosensors, and drug delivery due to their unique optical properties, excellent biocompatibility, and easy surface modification. In in vitro experimental studies, quantifying and visualizing the accumulated AuNPs in cells plays an important role in evaluating the efficiency of drug delivery, clarifying the mechanism of nanodrug-cell interaction, and the internalization mechanism of AuNPs. In particular, studying the heterogeneity of cell-accumulated AuNPs at the single-cell level can provide deeper information.

[0003] Hyperspectral technology combined with a dark-field microscopy system is a new method for in situ identification of AuNPs in organisms. This is because AuNPs have a unique surface plasmon resonance effect that makes them show strong spectral absorption in the visible-near infrared band of 400-1000nm, and the hyperspectral imaging system equipped with a unique spectrophotometer is also equipped with a charge-coupled device (CCD) camera to record the unique spectral curve of AuNPs in this band. However, due to the sensitivity of the surface plasmon resonance effect of AuNPs, the reproducibility of the AuNPs identification results is low, resulting in inconsistent quantitative results obtained by different laboratories or different technicians. In addition, when quantifying AuNPs in cells, there is no convenient and time-saving method to process the large amount of hyperspectral image data. These limitations prevent hyperspectral technology from becoming a mainstream technology for quantifying cell-accumulated AuNPs like other technologies. Summary of the invention

[0004] Purpose of the invention: The technical problem to be solved by the present invention is to provide a more comprehensive, accurate and simple method to simultaneously visualize and absolutely quantify the accumulation of metal nanoparticles in cells at the single-cell level in response to the shortcomings of the prior art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for simultaneously visualizing and quantifying cell-accumulated metal nanoparticles at the single-cell level, comprising the following steps:

[0007] Step 1: Cell samples that have accumulated a series of metal nanoparticles are divided into two groups, one of which is to quantify the metal nanoparticles accumulated in the cells using ICP-MS, and the other is to quantify the metal nanoparticles accumulated in the cells using hyperspectral technology;

[0008] Step 2: Establish a standard curve for absolute quantification based on the corresponding relationship between the two quantitative results in step 1;

[0009] Step 3: Calculate the ratio of the area of ​​metal nanoparticles in a single cell to the area of ​​the cell in the hyperspectral image, and absolutely quantify the metal nanoparticles accumulated in a single cell based on the absolute quantification standard curve in step 2.

[0010] Specifically, the metal nanoparticles are nanoparticles for which a corresponding reference library can be established and specifically identified under a dark-field hyperspectral imaging system.

[0011] Preferably, the metal nanoparticles are nanoparticles containing metal elements.

[0012] More preferably, the metal nanoparticles include nano-gold, nano-silver or nano-iron oxide.

[0013] Specifically, in step 1, quantifying the metal nanoparticles accumulated in cells using hyperspectral technology includes the following steps:

[0014] (1) Prepare cell samples to be imaged;

[0015] (2) using the sample prepared in step (1), obtaining a corresponding hyperspectral image;

[0016] (3) smoothing the hyperspectral image obtained in step (2), and then using the acquired light source spectrum to perform spectral correction on the smoothed hyperspectral image, and then collecting a spectral library of metal nanoparticles, and using the spectral library to identify metal nanoparticles internalized by cells;

[0017] (4) Calculate the ratio of the metal nanoparticle area to the cell area in each image, and calculate the average value of this ratio in all hyperspectral images of each treatment group to quantify the metal nanoparticles accumulated in the cells.

[0018] In step (2), for each sample, 6 to 8 hyperspectral images of different fields of view are randomly acquired, and these hyperspectral images contain >50 cells in total.

[0019] In step (3), the hyperspectral image is smoothed using the Adjacent Band Averaging function in ENVI 4.8 software.

[0020] In step (4), the calculation of the cell area includes the following steps:

[0021] S1: First, the cell area is obtained by manually processing the hyperspectral images of a certain number of cells;

[0022] S2: Perform machine learning modeling on the data in step S1, where the data-label pair used for modeling is a hyperspectral image of cells and an image of cells after manual processing;

[0023] S3: Use the machine learning model obtained in step S2 to automatically identify new cell hyperspectral images, obtain processed cell images, and then represent the cell area with the number of pixels.

[0024] Specifically, in step S1, the hyperspectral images of cells are manually processed using ImageJ to obtain the cell area, and at least 30 images are processed using this method.

[0025] Specifically, in step S2, machine learning modeling is performed on the data of step S1 based on the sklearn package of python 3.6; the machine learning method used is decision tree DecisionTree.

[0026] Beneficial effects:

[0027] The present invention establishes a method for simultaneously quantifying and visualizing the accumulation of label-free gold nanoparticles in single cells. A large amount of hyperspectral image data is efficiently and accurately processed using a machine learning method. The results of absolute quantification using ICP-MS and the standard curve of the hyperspectral quantitative results are used to absolutize the results of quantifying the accumulation of gold nanoparticles in cells using the hyperspectral technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0029] Figure 1 It is a hyperspectral image of cells.

[0030] Figure 2 is the cell area hand-drawn by ImageJ.

[0031] Figure 3 is the area of ​​the cell identified by machine learning.

[0032] Figure 4 is the normal distribution of machine learning recognition errors.

[0033] Figure 5 It is the standard curve of ICP-MS and hyperspectral quantitative results.

[0034] Figure 6 is the cell outline drawn by ImageJ.

[0035] Figure 7 is to statistically analyze the differential distribution of accumulated AuNPs between individual cells. DETAILED DESCRIPTION

[0036] The present invention can be better understood with reference to the following examples.

[0037] Example 1

[0038] This example provides a method for simultaneously visualizing and absolutely quantifying cellular accumulation of gold nanoparticles (AuNPs) at the single cell level, comprising the following steps:

[0039] Step 1: Place 5 sterile coverslips in a six-well plate, inoculate approximately 400,000 HeLa cells in these 5 wells, and grow on the coverslips for 24 hours to adhere to the wall, then replace the culture medium containing 60nm citric acid-modified AuNPs at concentrations of 0.3, 1, and 3 mg / L, respectively, with exposure times of 2, 4, 6, and 8 hours. After the cells absorb AuNPs, they are washed 5 times with 2mL PBS buffer. The cells in the 5 wells are divided into two groups, of which the cells in 3 wells are one group for quantification of cell-accumulated AuNPs using ICP-MS; the cells in the other 2 wells are one group for quantification of cell-accumulated AuNPs using hyperspectral technology;

[0040] For the hyperspectral technique to quantify the cellular accumulated AuNPs, the specific steps are as follows:

[0041] (1) In the two wells used for imaging, add PBS solution containing 2% paraformaldehyde to fix the cells for 20 min, then wash twice with PBS, cover the cover slip with cells on the slide, and seal with nail polish to keep the cells in a liquid state.

[0042] (2) Using the sample, dark-field hyperspectral microscopy was used to image the cells that had absorbed AuNPs. For each coverslip, 6-8 hyperspectral images of different fields of view were randomly acquired, and these hyperspectral images contained >50 cells in total.

[0043] (3) After obtaining the hyperspectral images, all hyperspectral images were smoothed using the Adjacent Band Averaging function in ENVI 4.8 software, and then the smoothed hyperspectral images were spectrally corrected using the acquired light source spectrum. Subsequently, the spectral library of AuNPs was collected and used to identify AuNPs internalized by cells.

[0044] (4) Calculate the ratio of the area of ​​AuNPs identified in each image to the cell area, and take the average of the ratio in all hyperspectral images of each treatment group to quantify the AuNPs accumulated in the cells. The calculation process of the cell area is:

[0045] S1Use ImageJ to convert the hyperspectral images (such as Figure 1) is set to 8bit, and the outline of the cell is manually drawn. Then, the fine outline of the cell is drawn by manual outlining, and finally the cell part is set to black and the other parts are set to white, such as Figure 2 As shown. Then count the cell area represented by the number of black pixels. Use this method to process 40 pictures.

[0046] S2 performs machine learning modeling on the data from step 1 based on the sklearn package of python 3.6. The data-label pairs used for modeling are the hyperspectral images of cells and the images of cells after manual processing, and the machine learning method used is decision tree. 75% of the data-label pairs are used as training sets, and the remaining 25% are used as test sets. The accuracy of the trained model is 95.7% on the training set and 95.6% on the test set. Compare the cell area obtained by ImageJ plus manual processing and the cell area obtained by the machine learning model (such as Figure 3 ), the error distribution of both conforms to the normal distribution (such as Figure 4 ), the error rate is within 15%.

[0047] S3 uses the obtained machine learning model to automatically identify new cell hyperspectral images, quickly obtain processed cell images, and then represent the cell area in terms of the number of pixels.

[0048] Step 2: According to the corresponding relationship between the two quantitative results in step 1, establish a standard curve for absolute quantification. Figure 5 shown.

[0049] Step 3: If Figure 6 As shown, ImageJ software was used to semi-automatically outline the contours of individual cells, and the ratio of the area of ​​AuNPs and the area of ​​cells identified in the hyperspectral images was calculated simultaneously. 25 photos of cells were statistically analyzed at three time points: 2h, 4h, and 6h. According to the absolute quantitative standard curve in step 2, the results of hyperspectral quantification were converted into absolute quantitative results, as shown in Figure 7 This result shows that there is a certain degree of heterogeneity in the ability of cells to accumulate AuNPs.

[0050] The present invention provides a method and idea for simultaneously visualizing and quantifying the accumulation of metal nanoparticles in cells at the single cell level. There are many methods and approaches to implement the technical solution. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A method to simultaneously visualize and quantify cellular accumulation of metal nanoparticles at the single-cell level, It is characterized in that The steps include: Step 1: Cell samples that have accumulated a series of metal nanoparticles are divided into two groups, one of which is to quantify the metal nanoparticles accumulated in the cells using ICP-MS, and the other is to quantify the metal nanoparticles accumulated in the cells using hyperspectral technology; Step 2: Establish a standard curve for absolute quantification based on the corresponding relationship between the two quantitative results in step 1; Step 3: Calculate the ratio of the area of ​​metal nanoparticles in a single cell to the area of ​​the cell in the hyperspectral image, and absolutely quantify the metal nanoparticles accumulated in a single cell according to the absolute quantification standard curve in step 2; In step 1, quantifying the accumulation of metal nanoparticles in cells using hyperspectral technology includes the following steps: (1) Prepare the cell sample to be imaged; (2) Using the sample prepared in step (1), obtain the corresponding hyperspectral image; (3) Smoothing the hyperspectral image obtained in step (2), and then using the acquired light source spectrum to perform spectral correction on the smoothed hyperspectral image, and then collecting a spectral library of metal nanoparticles, and using the spectral library to identify metal nanoparticles internalized by cells; (4) Calculate the ratio of the metal nanoparticle area to the cell area in each image, and calculate the average value of this ratio in all hyperspectral images of each treatment group to quantify the metal nanoparticles accumulated in the cells.

2. The method for simultaneously visualizing and quantifying cellular accumulation of metal nanoparticles at the single cell level according to claim 1, It is characterized in that The metal nanoparticles are nanoparticles for which a corresponding reference library can be established and specifically identified under a dark-field hyperspectral imaging system.

3. The method for simultaneously visualizing and quantifying cell accumulation of metal nanoparticles at the single cell level according to claim 1, It is characterized in that The metal nanoparticles are nanoparticles containing metal elements.

4. The method for simultaneously visualizing and quantifying cell accumulation of metal nanoparticles at the single cell level according to claim 1, It is characterized in that The metal nanoparticles are nano-gold, nano-silver or nano-iron oxide.

5. The method for simultaneously visualizing and quantifying cell accumulation of metal nanoparticles at the single cell level according to claim 1, It is characterized in that In step (2), for each sample, 6 to 8 hyperspectral images of different fields of view are randomly acquired, and these hyperspectral images contain >50 cells in total.

6. The method for simultaneously visualizing and quantifying cell-accumulated metal nanoparticles at the single cell level according to claim 1, It is characterized in that In step (3), the hyperspectral image is smoothed using the Adjacent Band Averaging function in ENVI 4.8 software.

7. The method for simultaneously visualizing and quantifying cell accumulation of metal nanoparticles at the single cell level according to claim 1, It is characterized in that In step (4), the calculation of cell area includes the following steps: S1: First, the cell area is obtained by manually processing the hyperspectral images of a certain number of cells; S2: Perform machine learning modeling on the data in step S1. The data-label pairs used for modeling are the hyperspectral images of cells - the manually processed cell images. S3: Use the machine learning model obtained in step S2 to automatically identify new cell hyperspectral images, obtain the processed cell images, and then represent the cell area in terms of the number of pixels.

8. The method for simultaneously visualizing and quantifying cell-accumulated metal nanoparticles at the single-cell level according to claim 7, characterized in that in step S1, use ImageJ to manually process the hyperspectral images of cells, obtain the cell area, and process at least 30 images using this method.

9. The method for simultaneously visualizing and quantifying cell-accumulated metal nanoparticles at the single-cell level according to claim 7, characterized in that in step S2, perform machine learning modeling on the data in step S1 based on the sklearn package of python 3.6; the machine learning method used is DecisionTree.

Images