A method for intracellular temperature measurement based on multi-dimensional fluorescence signal calibration

Abstract

The application discloses a kind of intracellular temperature measurement methods based on multi-dimensional fluorescence signal calibration. For the problem that the existing fluorescence temperature measurement method relies on single calibration curve, is easily interfered by complex microenvironment factors in cell, for example, viscosity, and causes measurement error, the application constructs the multi-parameter response matrix including fluorescence intensity ratio and fluorescence lifetime, by introducing the ratio type organic fluorescent probe with temperature response and viscosity response into living cells, utilize fluorescence intensity ratio and fluorescence lifetime two independent optical parameters as input signal, and combine multi-dimensional calibration model to simultaneously retrieve intracellular temperature and viscosity, to avoid the temperature measurement error caused by viscosity, probe concentration, photobleaching or local microenvironment change in traditional single fluorescence signal temperature measurement. The measurement method provided by the application can effectively eliminate environmental interference without changing the structure of the probe, improve the measurement accuracy and universality, and is suitable for organelle scale temperature distribution imaging and related biological process research.

Description

Technical Field

[0001] This invention relates to the field of fluorescence imaging and biometrics, specifically to a method for measuring intracellular temperature based on multidimensional fluorescence signal calibration. Background Technology

[0002] Intracellular temperature, as a crucial parameter reflecting cellular metabolic state and physiological function, plays a significant role in biological research and disease diagnosis. In recent years, intracellular temperature measurement methods based on fluorescent probes have been widely applied due to their advantages such as high spatial resolution and the ability to achieve subcellular scale imaging. Existing fluorescence thermometry methods typically rely on a single response relationship between fluorescence intensity, ratio, or fluorescence lifetime and temperature, establishing calibration curves in a solution environment and then applying them to intracellular temperature measurement.

[0003] However, the intracellular environment is complex, influenced not only by temperature but also by factors such as viscosity, polarity, and ionic strength. These factors also affect the luminescence behavior of fluorescent probes, especially for molecular rotor-type fluorescent probes, whose fluorescence signals are highly sensitive to changes in viscosity. In actual measurements, the coupling effect of these environmental factors can easily lead to changes in fluorescence signals being misinterpreted as temperature changes, thus introducing significant measurement errors. Therefore, traditional thermometry methods based on a single calibration curve are insufficient to accurately reflect the true intracellular temperature distribution.

[0004] To address the aforementioned issues, there is an urgent need to develop a method for measuring intracellular temperature that can effectively decouple the effects of temperature and environmental factors and improve measurement accuracy. Summary of the Invention

[0005] Purpose of the invention: To address the problem that existing intracellular temperature measurement methods based on fluorescence signals generally rely on a single calibration curve, making it difficult to distinguish the coupling effect of temperature and intracellular microenvironment parameters (including but not limited to viscosity) on the fluorescence signal, thus leading to insufficient measurement accuracy, this invention proposes an intracellular temperature measurement method based on a multidimensional fluorescence response input matrix.

[0006] This invention constructs a multi-parameter response input matrix that includes fluorescence intensity ratio and fluorescence lifetime, and establishes a matrix relationship model between the input matrix and temperature and environmental parameters. By combining a hierarchical calibration strategy to progressively correct the matrix relationship model, the invention achieves decoupling inversion of environmental parameters such as temperature and viscosity, thereby significantly improving the accuracy and stability of intracellular temperature measurement.

[0007] Technical solution: To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0008] A method for measuring intracellular temperature based on multidimensional fluorescence signal calibration, characterized by comprising the following steps:

[0009] Step 1: Construct a fluorescent probe, which includes a temperature-responsive unit, a reference fluorescent unit, and a flexible linker group. The temperature-responsive unit includes a rhodamine structure, the reference fluorescent unit includes a hemicyanine structure, and the temperature-responsive unit and the reference fluorescent unit are linked by the flexible linker group.

[0010] Step 2: Establish the spectral readout channel of the fluorescent probe, characterize the excitation and emission spectra of the fluorescent probe, and collect the fluorescence intensity of two fluorescence channels of the temperature-response unit and the reference fluorescent unit; perform fluorescence lifetime imaging on the rhodamine structure emission channel of the temperature-response unit to obtain the fluorescence lifetime τ of each pixel.

[0011] Step 3: Construct a temperature-viscosity multidimensional calibration system, prepare standard solutions with different viscosities, adjust the temperature of the standard solution to multiple temperature points at each viscosity, measure temperature, viscosity, fluorescence intensity and fluorescence lifetime, establish an initial two-dimensional calibration model, and perform simultaneous quantitative analysis of temperature and viscosity.

[0012] Step 4: Prepare single-cell samples to be tested. After seeding the cells to be tested in a glass-bottomed culture dish and culturing them, wash the cells with PBS, add culture medium containing the fluorescent probe prepared in step 1 for incubation, and wash with PBS after incubation to remove the free fluorescent probe.

[0013] Step 5: Obtain the fluorescence ratio map and lifetime map of the single cell to be tested. Place the single-cell sample processed in Step 4 on a confocal laser scanning microscope equipped with a FLIM module, stabilize the extracellular environment at the set temperature, excite the fluorescent probe with a pulsed laser, and acquire the fluorescence lifetime image. Simultaneously, acquire the fluorescence intensity images of the two fluorescence channels of the temperature-response unit and the reference fluorescence unit to obtain the Ig of each pixel. 590 and I 720 The fluorescence ratio map R(x,y) is calculated; then, the fluorescence decay curve of each pixel is fitted by a double exponential method to obtain the fluorescence lifetime map τ(x,y).

[0014] Step 6: Correct the calibration matrix using a cell simulation system. This correction includes primary calibration of giant membrane vesicles and secondary calibration of fixed cells. The primary calibration of giant membrane vesicles involves preparing giant membrane vesicles containing fluorescent probes, placing these vesicles in an imaging system with a temperature-controlled stage, and acquiring fluorescence intensity channel images and fluorescence lifetime images at multiple temperatures. The fluorescence ratio and lifetime are calculated and substituted into the initial two-dimensional calibration matrix described in Step 3 to obtain the vesicle measurement temperature T. 测 The actual ambient temperature T recorded by the temperature control system 实 The measured temperature is compared with the actual temperature, and a correction is performed using the temperature deviation to ensure that the calculated temperature in the giant membrane vesicles is consistent with the actual set temperature.

[0015] The second calibration of the fixed cells involves fixing the cells after incubation with the fluorescent probe, placing the fixed cells in an imaging system with a temperature-controlled stage, acquiring fluorescence intensity ratio maps R(x,y) and fluorescence lifetime maps τ(x,y) at multiple temperatures, substituting them into the calibration matrix after the first calibration to obtain a pixel-by-pixel temperature map of the fixed cells, calculating the deviation between the measured temperature of the fixed cells and the actual set temperature, and performing a second calibration on the temperature bias term of the calibration matrix to obtain the final calibration matrix after correction by the cell simulation system, which is suitable for the cell environment.

[0016] Step 7: Calculate the single-cell temperature distribution pixel by pixel. After obtaining the fluorescence intensity image and fluorescence lifetime image of a single cell, calculate independently for each pixel to reconstruct the single-cell temperature distribution map. Fit the fluorescence decay curve obtained from the pixel FLIM acquisition to obtain the fluorescence lifetime τ(x,y) of that pixel. Input the R(x,y) and τ(x,y) of each pixel into the final calibration matrix corrected by the cell simulation system as described in step S6. After completing the calculation of all pixels, rearrange the T(x,y) of each pixel according to its original spatial coordinates to obtain the single-cell temperature distribution map; at the same time, rearrange the η(x,y) of each pixel to obtain the single-cell viscosity distribution map.

[0017] Beneficial effects

[0018] This invention presents a method for measuring intracellular temperature based on multidimensional fluorescence signal calibration. It utilizes fluorescence lifetime microscopy and fluorescence confocal microscopy, combined with matrix operation methods, to achieve precise imaging of intracellular temperature within single cells. Compared with existing technologies, the advantages of this invention are:

[0019] (1) Effective decoupling of temperature and microenvironment parameters is achieved in principle, thereby improving measurement accuracy.

[0020] This invention constructs a multidimensional fluorescence response input matrix that includes fluorescence intensity ratios and fluorescence lifetimes, and establishes a matrix relationship model between this input matrix and temperature and environmental parameters. This transforms the fluorescence signal, which is originally coupled with multiple factors such as temperature and viscosity, into a multivariate inversion problem. Compared to traditional one-dimensional calibration curve methods based on a single fluorescence intensity or a single fluorescence lifetime, this invention utilizes the differentiated response characteristics between multidimensional response signals to achieve mathematical decoupling of temperature and viscosity. This fundamentally overcomes the interference of the complex intracellular microenvironment on the measurement results and improves the intrinsic accuracy of temperature measurements.

[0021] (2) The model mismatch error is significantly reduced by the hierarchical calibration strategy, which improves its applicability in the cellular environment.

[0022] This invention employs a hierarchical calibration strategy consisting of a solution system, a cell-like system (such as vesicles), and a fixed cell system to progressively correct the matrix relationship model, making the model parameters gradually approximate the real cellular environment.

[0023] Experimental results show that:

[0024] In solution systems, the temperature measurement error can be controlled within approximately ±0.08 ℃;

[0025] In the cell-like system, the measurement error was reduced to approximately 0.06 °C after calibration;

[0026] In fixed cells, the measurement error can reach ±0.05 ℃ after further calibration.

[0027] Hierarchical environmental calibration effectively reduces systematic errors caused by environmental differences, significantly improving the applicability and stability of the model in complex biological systems.

[0028] (3) Integrate multi-parameter information to improve measurement accuracy and anti-interference capability.

[0029] This invention utilizes two physical quantities, fluorescence intensity ratio and fluorescence lifetime, as input parameters:

[0030] The fluorescence intensity ratio can eliminate probe concentration fluctuations and some photobleaching effects;

[0031] Fluorescence lifetime is sensitive to environmental changes but does not depend on excitation intensity;

[0032] By fusing multiple parameters, the measurement results have higher redundancy and robustness, effectively reducing the error caused by single signal fluctuations, thereby improving the accuracy and repeatability of temperature measurement.

[0033] (4) High-precision measurement can be achieved without changing the probe structure, and it has good versatility.

[0034] This invention achieves signal decoupling through algorithmic modeling and calibration strategies, eliminating the need for additional design or modification of existing fluorescent probe molecular structures. Therefore, this method is directly applicable to various existing fluorescent probe systems, reducing experimental complexity and cost, and demonstrating good versatility and scalability.

[0035] (5) Achieve subcellular scale temperature imaging and expand the dimensions of biological research.

[0036] This invention combines fluorescence imaging with pixel-by-pixel matrix operations to obtain spatial distribution information of intracellular temperature, enabling subcellular-scale temperature imaging. Based on this method, it is possible to further obtain temperature differences among different organelles (such as mitochondria, lysosomes, and cytoplasm), and use this information to study intracellular heat transfer processes, mitochondrial metabolism and thermogenic behavior, and the thermal response mechanisms of cells under stimulation or pathological conditions, thus providing new technical means for the study of cell thermodynamics and related life processes.

[0037] (6) Through comparative experiments, it is found to be significantly superior to traditional single-parameter measurement methods.

[0038] Compared with traditional calibration methods that use a single fluorescence intensity or a single fluorescence lifetime, this invention can effectively avoid systematic deviations caused by environmental factors such as viscosity in complex cellular environments.

[0039] Traditional methods suffer from significant measurement biases in cellular environments. For example, conventional thermometry methods have measured mitochondrial temperatures as high as 50 °C, causing considerable controversy. The method of this invention can control the error within a sub-Celsius range, revealing that the upper limit of mitochondrial temperature is between 42-43 °C. Excessively high temperature readings are interferences caused by errors such as viscosity.

[0040] This demonstrates that the measurement accuracy and reliability of the present invention in complex biological systems are significantly superior to those of existing technologies. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the method of the present invention;

[0042] Figure 2 This is a matrix operation diagram of the present invention;

[0043] Figure 3 This is a schematic diagram of the matrix calibration method of the present invention. Similar to the principle of neural networks, the final temperature data needs to be calibrated in the environments of solution, vesicles, and fixed cells. Only after calibration can the matrix equation be further used for live cell temperature measurement. In the diagram, (a) is a schematic diagram of the calibration process, (b) shows the measurement of solution temperature, (c) shows the measurement of vesicle temperature, (d) shows the measurement of fixed cell temperature, and (e) shows the optical signal measurement results of the corresponding region.

[0044] Figure 4 The temperature imaging method of this invention uses a pixel-by-pixel method to perform matrix operations;

[0045] Figure 5 It involves measuring the temperature of mitochondria, lysosomes, and cytoplasm. By combining co-localization methods, different organelles are distinguished within the cell, and the measured temperatures are then correlated with those of different organelles to obtain the temperatures of those organelles.

[0046] Figure 6 The changes in temperature and viscosity of cells in the face of bacterial infection were measured, and it was found that as the infection time increased, both cell temperature and viscosity gradually increased until they eventually died.

[0047] Figure 7 This invention illustrates temperature imaging under Rotenone stimulation according to an embodiment of the present invention;

[0048] Figure 8 This illustration shows mitochondrial region temperature imaging stimulated by Mdivi-1 and STS according to an embodiment of the present invention.

[0049] Figure 9 This illustration shows temperature imaging without external stimuli according to an embodiment of the present invention.

[0050] Figure 10 This illustrates temperature imaging under calcium salt stimulation according to an embodiment of the present invention. Detailed Implementation

[0051] The preferred embodiments of the present invention will now be described in detail with reference to specific examples. It should be understood that the following examples are given for illustrative purposes only and are not intended to limit the scope of the invention. Those skilled in the art can make various modifications and substitutions to the present invention without departing from its spirit and essence.

[0052] Unless otherwise specified, the experimental methods used in the following examples are conventional methods.

[0053] Unless otherwise specified, all materials and reagents used in the following examples are commercially available.

[0054] One embodiment of this application provides a method for measuring intracellular temperature based on multidimensional fluorescence signal calibration, characterized by comprising the following steps:

[0055] Step 1: Construct a fluorescent probe, which includes a temperature-responsive unit, a reference fluorescent unit, and a flexible linker group. The temperature-responsive unit includes a rhodamine structure, the reference fluorescent unit includes a hemicyanine structure, and the temperature-responsive unit and the reference fluorescent unit are linked by the flexible linker group.

[0056] Step 2: Establish the spectral readout channel of the fluorescent probe, characterize the excitation and emission spectra of the fluorescent probe, and collect the fluorescence intensity of two fluorescence channels of the temperature-response unit and the reference fluorescent unit; perform fluorescence lifetime imaging on the rhodamine structure emission channel of the temperature-response unit to obtain the fluorescence lifetime τ of each pixel.

[0057] Step 3: Construct a temperature-viscosity multidimensional calibration system, prepare standard solutions with different viscosities, adjust the temperature of the standard solution to multiple temperature points at each viscosity, measure temperature, viscosity, fluorescence intensity and fluorescence lifetime, establish an initial two-dimensional calibration model, and perform simultaneous quantitative analysis of temperature and viscosity.

[0058] Step 4: Prepare the single-cell sample to be tested. After seeding the cells to be tested in a glass-bottomed culture dish, wash the cells with PBS, add culture medium containing the fluorescent probe prepared in step 1 for incubation, and wash with PBS after incubation to remove the free fluorescent probe.

[0059] Step 5: Obtain the fluorescence ratio map and lifetime map of the single cell to be tested. Place the single-cell sample processed in Step 4 on a confocal laser scanning microscope equipped with a FLIM module, stabilize the extracellular environment at the set temperature, excite the fluorescent probe with a pulsed laser, and acquire the fluorescence lifetime image. Simultaneously, acquire the fluorescence intensity images of the two fluorescence channels of the temperature-response unit and the reference fluorescence unit to obtain the Ig of each pixel. 590 and I 720 The fluorescence ratio map R(x,y) is calculated; then, the fluorescence decay curve of each pixel is fitted by a double exponential method to obtain the fluorescence lifetime map τ(x,y).

[0060] Step 6: Correct the calibration matrix using a cell simulation system. This correction includes primary calibration of macromembrane vesicles and secondary calibration of fixed cells. The primary calibration of macromembrane vesicles involves preparing macromembrane vesicles containing fluorescent probes, placing these vesicles in an imaging system with a temperature-controlled stage, acquiring fluorescence intensity channel images and fluorescence lifetime images at multiple temperatures, calculating the fluorescence ratio and lifetime, and substituting these values ​​into the initial two-dimensional calibration matrix described in Step 3 to obtain the vesicle measurement temperature T. 测 The actual ambient temperature T recorded by the temperature control system 实 The measured temperature is compared with the actual temperature, and a correction is performed using the temperature deviation to ensure that the calculated temperature in the giant membrane vesicles is consistent with the actual set temperature.

[0061] The second calibration of the fixed cells involves fixing the cells after incubation with the fluorescent probe, placing the fixed cells in an imaging system with a temperature-controlled stage, acquiring fluorescence intensity ratio maps R(x,y) and fluorescence lifetime maps τ(x,y) at multiple temperatures, substituting them into the calibration matrix after the first calibration to obtain a pixel-by-pixel temperature map of the fixed cells, calculating the deviation between the measured temperature of the fixed cells and the actual set temperature, and performing a second calibration on the temperature bias term of the calibration matrix to obtain the final calibration matrix after correction by the cell simulation system, which is suitable for the cell environment.

[0062] Step 7: Calculate the single-cell temperature distribution pixel by pixel. After obtaining the fluorescence intensity image and fluorescence lifetime image of a single cell, calculate independently for each pixel to reconstruct the single-cell temperature distribution map. Fit the fluorescence decay curve obtained by FLIM to obtain the fluorescence lifetime τ(x,y) of the pixel. Input R(x,y) and τ(x,y) of each pixel into the final calibration matrix corrected by the cell simulation system as described in step S6. After completing the calculation of all pixels, rearrange T(x,y) of each pixel according to its original spatial coordinates to obtain the single-cell temperature distribution map; at the same time, rearrange η(x,y) of each pixel to obtain the single-cell viscosity distribution map.

[0063] In one embodiment, the molecular formula of the synthesized small molecule fluorescent probe was determined using NMR and mass spectrometry. The synthesized organic fluorescent small molecule probes include, but are not limited to, the molecules described above; other temperature-sensitive small molecules with calibrated structures may also be used.

[0064] In one embodiment, in step 2, the probe is characterized by excitation and emission spectra. The fluorescence intensities of two fluorescence channels of the temperature-response unit and the reference fluorescence unit are collected. The first channel is the temperature-response channel, and the emission peak of the temperature-response unit is denoted as I. 590 The second channel is the reference channel, representing the emission peak position of the fluorescent reference unit, denoted as I. 720 Calculate the fluorescence intensity ratio: R = I 590 / I 720 Fluorescence lifetime imaging was performed on the rhodamine structure emission channel of the temperature-responsive unit to obtain the fluorescence lifetime τ of each pixel. The fluorescence lifetime τ can be expressed by a double exponential fit as follows:

[0065] I(t) = A1 exp(-t / τ1) + A2exp(-t / τ2)

[0066] Where t is time, A1 and A2 are the amplitudes of the decay components, and τ1 and τ2 are the corresponding fluorescence lifetimes. The weighted average fluorescence lifetime is calculated as the fluorescence lifetime parameter τ of the pixel.

[0067] In one embodiment, in step 3, the initial two-dimensional calibration model is established for simultaneous temperature and viscosity quantification. For each standard sample, the following parameters are measured: true temperature T, actual viscosity η, fluorescence intensity ratio R, and fluorescence lifetime τ. R and τ are used as inputs, and T and η are used as outputs to establish the initial two-dimensional calibration model.

[0068] (T, η) = f(R, τ), using a cubic polynomial feature extension model, the input signal is extended to:

[0069] [R, R², R³, τ, τ², τ³]

[0070] And by solving the calibration coefficient matrix W and the bias term b through least squares regression, we obtain:

[0071] [T, η] = [R, R², R³, τ, τ², τ³] · W + b

[0072] The initial two-dimensional calibration model separates the temperature information coupled in fluorescence ratio and fluorescence lifetime from the viscosity information, enabling simultaneous quantification of temperature and viscosity.

[0073] In one embodiment, in step 4, the step of seeding the cells to be tested in a glass-bottomed culture dish means culturing the cells in a high-glucose DMEM medium containing 10% fetal bovine serum, 100 U / mL penicillin and 100 μg / mL streptomycin at 37 ℃ and 5% CO2; the step of incubation means incubating at 37 ℃ and 5% CO2 for about 30 minutes until the cells adhere to the dish and are in a good growth state.

[0074] In one embodiment, in step 5, the temperature is set to a range of 20 ℃-40 ℃, and the pulsed laser excitation probe is set to use a 561 nm pulsed laser with a repetition frequency of about 20 MHz to acquire a fluorescence lifetime image in 512×512 pixel format.

[0075] In one embodiment, in step 6, the preparation of fluorescently labeled giant membrane vesicles involves seeding cells in a glass-bottomed culture dish and culturing them under CO2 conditions for 24 hours. After the cells adhere to the dish and reach approximately 70% confluence, they are washed with PBS buffer, and DMEM culture medium containing the fluorescent probe is added for incubation. This allows the probe to enter the cell membrane and intracellular environment. After incubation, the cells are washed with GPMV buffer, and then vesicle induction solution is added for incubation to induce the cell membrane to detach and form giant membrane vesicles, thus preparing fluorescently labeled giant membrane vesicles.

[0076] In one embodiment, the vesicle induction solution may be prepared from 4% paraformaldehyde solution and dithiothreitol, and incubated at 37 °C and 5% CO2.

[0077] In one embodiment, step 6, the temperature deviation correction is performed by comparing the measured temperature with the actual ambient temperature T recorded by the temperature control system, and calculating the deviation at each temperature point: ΔT = T 实 - T 测 Based on the average deviation obtained from multiple temperature points, the temperature output bias term in the calibration matrix is ​​corrected once to ensure that the calculated temperature in the giant membrane vesicles is consistent with the actual set temperature. The calibration model is expressed as follows:

[0078] [T, η] = [R, R², R³, τ, τ², τ³] · W + b

[0079] Keeping the coefficient matrix W unchanged, we first correct the components of the bias term b corresponding to the temperature output, so that the corrected model satisfies: T 修正1 = T 测 + ΔT 平均 .

[0080] In one embodiment, step 6, fixing the cells after incubation with the fluorescent probe, involves washing the cells with PBS, adding freshly prepared 4% paraformaldehyde solution for fixation, washing the cells again with PBS after fixation, and adding culture medium or PBS to keep the sample moist. The secondary correction involves calculating the deviation between the measured temperature of the fixed cells and the actual set temperature, and performing a secondary correction on the temperature bias term of the calibration matrix.

[0081] ΔT 固定 = T 实 - T 测

[0082] T 修正2 = T 测 + ΔT 固定平均。

[0083] In one embodiment, in step 7, the fluorescence decay curve of the pixel obtained by FLIM is fitted as follows: let the fluorescence intensity of the temperature response channel at the (x,y)th pixel be I. 590 (x,y), the fluorescence intensity of the reference channel is I 720 Given (x, y), the fluorescence ratio of that pixel is:

[0084] R(x,y) = I 590 (x,y) / I 720 (x,y).

[0085] In one embodiment, in step 7, the step of inputting R(x,y) and τ(x,y) of each pixel into the final calibration matrix corrected by the cell simulation system in step S6 is to perform cubic polynomial feature expansion on R(x,y) and τ(x,y) to construct the input vector of that pixel:

[0086] X(x,y) = [R(x,y), R(x,y)², R(x,y)³, τ(x,y), τ(x,y)², τ(x,y)³]

[0087] Then, calculate the temperature and viscosity at that pixel based on the calibration model:

[0088] [T(x,y), η(x,y)] = X(x,y) · W + b

[0089] Where W is the multidimensional calibration coefficient matrix, b is the bias term corrected by giant membrane vesicles and fixed cells, T(x,y) is the temperature at the pixel, and η(x,y) is the viscosity at the pixel; the pixel rearrangement is to rearrange the T(x,y) of each pixel according to its original spatial coordinates to obtain a single-cell temperature distribution map; at the same time, the η(x,y) of each pixel is rearranged to obtain a single-cell viscosity distribution map.

[0090] In one embodiment, the temperature graph may be displayed in pseudo-color, where colors correspond to different temperature values.

[0091] In one embodiment, the fluorescent probe is prepared by first preparing an intermediate containing a rhodamine temperature-responsive unit: compound 1 (rhodamine B), methyl 5-aminovalerate, HATU, and triethylamine are dissolved in tetrahydrofuran and stirred at room temperature for about 4 hours; after the reaction, the solvent is removed under reduced pressure, and the mixture is purified by silica gel column chromatography. The eluent is a dichloromethane / methanol mixture, and the volume ratio can be gradually adjusted from 50:1 to 10:1 to obtain an orange solid compound 2 (rhodamine BN-methyl-β-alanine methyl ester amide hydrochloride). Subsequently, compound 2 is dissolved in methanol, and an aqueous sodium hydroxide solution is slowly added. The mixture is stirred at room temperature for about 2 hours to hydrolyze the ester group to a carboxyl group; after the reaction, the mixture is purified by semi-preparative high-performance liquid chromatography. The eluent is an acetonitrile / water system containing 0.1% trifluoroacetic acid, and the acetonitrile ratio can be gradually increased from 40% to 90% within 20 minutes to obtain the carboxylated intermediate compound 3 (rhodamine BN-methyl-γ-aminobutyric acid amide chloride). Finally, compound 3, compound 4 (1-(3-aminopropyl)-2-[(E)-2-(7-chloro-6-hydroxy-2,3-dihydro-1H-xanthon-4-yl)vinyl]-3,3-dimethyl-3H-indolium chloride containing a hemicyanine reference fluorescent unit), HBTU, and N,N-diisopropylethylamine were dissolved in tetrahydrofuran and reacted at room temperature with stirring for about 4 hours to couple the rhodamine temperature-responsive unit and the hemicyanine reference unit via a flexible linker. After the reaction, the product was purified by semi-preparative high-performance liquid chromatography (HPLC) using an acetonitrile / water system containing 0.1% trifluoroacetic acid. The acetonitrile ratio increased from 40% to 90% within 20 minutes, ultimately yielding a blue solid product, i.e., a ratiometric fluorescent probe. This probe simultaneously contains both the rhodamine temperature-responsive structure and the hemicyanine reference structure, and can be used for subsequent simultaneous imaging analysis of single-cell temperature and viscosity.

[0092] In one embodiment, in step 3, a series of standard solutions with different viscosities are prepared by adjusting the volume ratio of glycerol to the total solution, sequentially preparing glycerol / water mixed solutions with a glycerol volume fraction of 0%-60%, and measuring the solution viscosity at an environment of 20-50 °C using a viscometer.

[0093] In one embodiment, the fluorescence spectrometer used in step 3 is an FLS980 spectrometer with heating and temperature measurement functions.

[0094] In one embodiment, the measurement method using a fluorescence spectrometer in step 3 is as follows: Step 31: The excitation wavelength of the fluorescent probe used is 550 nm, and the emission wavelength is 590 nm. These are the absorption and emission wavelengths of the rhodamine structure. The emission wavelength intensity is affected by temperature and viscosity. Another excitation and emission wavelength is 690 nm and 720 nm, respectively. These are the absorption and emission wavelengths of the hemifluorescence structure. The emission wavelength at this position is mainly used for calibration to eliminate the influence of dye concentration and fluorescence quenching. The ratio of the two wavelengths is the desired fluorescence ratio. Step 32: Fluorescent photons at 590 nm are collected. The number of photons is generally greater than 5000. The fluorescence decay is obtained by the number of collected photons, and the fluorescence lifetime is fitted using a double exponential function.

[0095] In one embodiment, in step 5, a double exponential fitting is performed on the fluorescence decay curve of each pixel. Origin can be used for fitting, or other software or programs can be used for fitting. However, the coefficient of determination should generally be greater than 0.95. The larger the coefficient, the better the fitting effect and the more accurate the subsequent data processing.

[0096] One embodiment of this application provides a detection device for running the temperature measurement method. The device includes a spectral excitation module for acquiring fluorescence signals, a signal detection module for acquiring fluorescence intensity and fluorescence lifetime information, and a data processing module for constructing a multidimensional response matrix and realizing decoupled calculation of temperature and viscosity.

[0097] The detection principle of one embodiment of this application is as follows: Figure 1 As shown, other environmental factors affecting solution temperature measurement, mainly viscosity, are also considered in the calculation process. To solve for these two variables, the fluorescence ratio and fluorescence lifetime, two independent variables, are measured simultaneously. The traditional one-dimensional function calculation method is changed, and matrix operations are used to obtain accurate temperature values.

[0098] One embodiment of this application provides a method for measuring intracellular temperature based on multidimensional fluorescence signal calibration, the method comprising the following steps:

[0099] S1. Construct a fluorescent probe to prepare a ratiometric organic fluorescent probe. The probe includes a temperature-responsive unit, a reference fluorescent unit, and a linker group.

[0100] The temperature-responsive unit is preferably a rhodamine structure, where the diethylamino group on the xanthracene ring undergoes intramolecular rotation. As temperature increases, the intramolecular rotation and nonradiative transitions intensify, reducing the fluorescence intensity and lifetime of the rhodamine channel. The reference fluorescence unit is preferably a hemicyanine structure, whose emission signal is relatively stable, used to construct a ratiometric readout. The two parts are connected by a flexible linker, enabling the probe to simultaneously possess temperature response, ratiometric readout, and cell imaging capabilities. The probe as a whole is preferably positively charged, allowing it to accumulate in the mitochondrial region dependent on the mitochondrial membrane potential, thereby achieving mitochondrial and organelle-scale temperature measurements.

[0101] S2. Establish the spectral readout channel for the probe.

[0102] The probes were characterized by excitation and emission spectra. The excitation peak of the rhodamine unit was located at approximately 560 nm, and the emission peak was located at approximately 590 nm; the excitation peak of the hemicyanine unit was located at approximately 700 nm, and the emission peak was located at approximately 720 nm.

[0103] Two fluorescence channels were acquired during the measurement:

[0104] The first channel is the temperature response channel, denoted as I. 590 ;

[0105] The second channel is the reference channel, denoted as I. 720 .

[0106] Therefore, the fluorescence intensity ratio can be calculated as: R = I 590 / I 720

[0107] Simultaneously, fluorescence lifetime imaging was performed on the rhodamine emission channel to obtain the fluorescence lifetime τ of each pixel.

[0108] The fluorescence lifetime τ, when fitted by a bi-exponential method, can be expressed as:

[0109] I(t) = A1 exp(-t / τ1) + A2exp(-t / τ2)

[0110] Where A1 and A2 are the amplitudes of the attenuation components, and τ1 and τ2 are the corresponding lifetimes. The average lifetime can be further calculated as the lifetime parameter τ for this pixel.

[0111] S3. Construct a temperature-viscosity multidimensional calibration system

[0112] Since the fluorescence ratio R and fluorescence lifetime τ of this fluorescent probe are affected by both temperature T and viscosity η, a traditional one-dimensional temperature calibration curve cannot be used. Therefore, a series of standard solutions with different viscosities are first prepared, such as glycerol / water mixed solutions with glycerol volume fractions of 0%-60%.

[0113] For each viscosity condition, the standard solution temperature was adjusted to multiple temperature points, preferably within the range of 25-45 °C. For each standard sample, the following parameters were measured:

[0114] True temperature T;

[0115] Actual viscosity η;

[0116] Fluorescence intensity ratio R;

[0117] Fluorescence lifetime τ.

[0118] Then, using R and τ as inputs and T and η as outputs, a two-dimensional calibration model is established:

[0119] (T, η) = f(R, τ)

[0120] A cubic polynomial feature extension model is preferred, which extends the input signal to:

[0121] [R, R², R³, τ, τ², τ³]

[0122] And by solving the calibration coefficient matrix W and the bias term b through least squares regression, we obtain:

[0123] [T, η] = [R, R², R³, τ, τ², τ³] · W + b

[0124] This model can separate the temperature information coupled with the viscosity information in fluorescence ratio and fluorescence lifetime, and achieve simultaneous quantification of temperature and viscosity.

[0125] S4. Preparation of single-cell samples for testing

[0126] The cells to be tested were seeded and cultured in glass-bottomed culture dishes. Taking HeLa cells as an example, the cells were cultured in high-glucose DMEM medium containing 10% fetal bovine serum, 100 U / mL penicillin and 100 μg / mL streptomycin at 37 ℃ and 5% CO2.

[0127] Once the cells have adhered and are in good growth condition, wash them with PBS, add culture medium containing fluorescent probes, and incubate at 37 °C and 5% CO2 for about 30 minutes. After incubation, wash the cells three times with PBS to remove any unbound or non-entered probes, and then add fresh culture medium for imaging.

[0128] S5. Obtain single-cell fluorescence ratio maps and lifetime maps.

[0129] The stained live cells are placed on a confocal laser scanning microscope equipped with a FLIM module. It is preferable to use a microscope stage with temperature control to stabilize the extracellular environment temperature at a set temperature, such as 27 ℃, 32 ℃, 37 ℃ or 42 ℃.

[0130] The probe is excited using a pulsed laser, preferably a 561 nm pulsed laser with a repetition frequency of about 20 MHz, to acquire fluorescence lifetime images in 512×512 pixel format.

[0131] Simultaneously acquire fluorescence intensity images of the temperature response channel and the reference channel to obtain the Ig of each pixel. 590 and I 720 The fluorescence ratio map R(x,y) is calculated; then, the fluorescence decay curve of each pixel is fitted by a double exponential method to obtain the fluorescence lifetime map τ(x,y).

[0132] S6. Correct the calibration matrix using a cell simulation system.

[0133] To improve the applicability of the calibration matrix in a real cellular environment, after the initial calibration in the solution system, the calibration matrix was further corrected using a cell simulation system that does not have metabolic heat production capabilities. This correction included two steps: a primary calibration of giant cell membrane vesicles and a secondary calibration of fixed cells.

[0134] First, fluorescently labeled giant membrane vesicles were prepared. HeLa cells were seeded in glass-bottomed culture dishes and cultured at 7°C and 5% CO2 for 24 hours. After the cells adhered and reached approximately 70% confluence, they were washed three times with PBS buffer. Then, DMEM culture medium containing the fluorescent probe was added, and the cells were incubated at 37°C and 5% CO2 for 30 minutes to allow the probe to enter the cell membrane and intracellular environment. After incubation, the cells were washed three times with GPMV buffer; the GPMV buffer preferably contained 10 mmol / L HEPES, 150 mmol / L NaCl, and 2 mmol / L CaCl2, and the pH was adjusted to 7.4. Then, vesicle induction medium was added, which could be prepared from 4% paraformaldehyde solution and dithiothreitol, for example, by adding 2 μL of 1 mol / L DTT to 1 mL of 4% PFA solution. Incubation was continued at 37°C and 5% CO2 for approximately 1 hour to induce cytoplasmic membrane detachment and the formation of giant membrane vesicles.

[0135] Since giant membrane vesicles lack active thermogenic processes such as mitochondrial respiration and ATP metabolism, their internal temperature can be considered to be consistent with the external culture medium temperature, thus serving as an isothermal calibration target in a cell membrane environment. Giant membrane vesicles containing fluorescent probes were placed in an imaging system with a temperature-controlled stage. The external ambient temperature was set to multiple temperature points, such as 27 ℃, 32 ℃, 37 ℃, and 42 ℃. After each temperature point stabilized, images of the two fluorescence intensity channels and fluorescence lifetime of the probe within the vesicle were acquired. The fluorescence ratio R and lifetime τ were calculated pixel-by-pixel and substituted into the initial calibration matrix obtained from the solution system to obtain the vesicle measurement temperature T. 测 (For specific measurement methods, see S7. Pixel-by-pixel calculation of single-cell temperature distribution). Compare the measured temperature with the actual ambient temperature T recorded by the temperature control system, and calculate the deviation at each temperature point: ΔT = T 实 - T 测

[0136] Based on the average deviation obtained from multiple temperature points, the temperature output bias term in the calibration matrix is ​​corrected to ensure that the calculated temperature in the giant membrane vesicles matches the actual set temperature. If the calibration model is expressed as:

[0137] [T, η] = [R, R², R³, τ, τ², τ³] · W + b

[0138] Then, keeping the coefficient matrix W unchanged, we first correct the components of the bias term b corresponding to the temperature output, so that the corrected model satisfies:

[0139] T 修正1 = T 测 + ΔT 平均

[0140] After the initial calibration of the giant membrane vesicles, a second calibration was performed using fixed cells. Specifically, HeLa cells incubated with the fluorescent probe were washed three times with PBS and fixed for approximately 30 minutes with freshly prepared 4% paraformaldehyde solution. After fixation, the cells were washed again with PBS, and culture medium or PBS was added to keep the sample moist. Although fixed cells lose their metabolic activity and no longer generate endogenous heat, they still better preserve the complex microenvironment of the cell membrane, organelles, proteins, and intracellular macromolecules, thus more closely resembling the internal environment of real living cells than giant membrane vesicles.

[0141] Immobilized cells are placed in the same imaging system and imaged at multiple set temperatures, preferably 27℃, 32℃, 37℃, and 42℃. At each temperature point, after the sample temperature reaches equilibrium, fluorescence intensity ratio map R(x,y) and fluorescence lifetime map τ(x,y) are acquired. These are then substituted into a first-correction calibration matrix to obtain a pixel-by-pixel temperature map of the immobilized cells. Since immobilized cells are inactive, the true temperature of each region within the cell should equal the external set temperature. The deviation between the measured temperature of the immobilized cells and the actual set temperature is then calculated, and the temperature bias term of the calibration matrix is ​​corrected a second time.

[0142] ΔT 固定 = T 实 - T 测

[0143] T 修正2 = T 测 + ΔT 固定平均

[0144] After secondary correction, a final calibration matrix suitable for the cellular environment is obtained. This final calibration matrix retains the decoupling capability of temperature and viscosity in the solution system while compensating for systematic biases caused by cell membrane structure, protein binding, local refractive index, background fluorescence, and the subcellular microenvironment on the probe fluorescence signal. Preferably, the corrected matrix can reduce the absolute error of temperature measurement in fixed cells to within 0.1 °C, thus enabling subsequent pixel-by-pixel quantitative imaging of live cell temperature and viscosity.

[0145] S7. Calculate single-cell temperature distribution pixel by pixel.

[0146] After obtaining the fluorescence intensity and fluorescence lifetime images of a single cell, the temperature distribution map of the single cell is reconstructed by independently calculating the fluorescence intensity and fluorescence lifetime images of each pixel. Specifically, two fluorescence intensity images within the same cell region are first registered to ensure a one-to-one spatial correspondence between the temperature response channel and the reference channel. Let the fluorescence intensity of the temperature response channel at the (x,y)th pixel be I. 590 (x,y), the fluorescence intensity of the reference channel is I 720Given (x, y), the fluorescence ratio of that pixel is:

[0147] R(x,y) = I 590 (x,y) / I 720 (x,y)

[0148] Meanwhile, the fluorescence decay curve obtained by FLIM at this pixel is fitted to obtain the fluorescence lifetime τ(x,y) of this pixel.

[0149] Subsequently, the R(x,y) and τ(x,y) of each pixel are input into the multidimensional calibration matrix corrected by the cell simulation system. Preferably, R(x,y) and τ(x,y) are extended by a cubic polynomial feature to construct the input vector for that pixel:

[0150] X(x,y) = [R(x,y), R(x,y)², R(x,y)³, τ(x,y), τ(x,y)², τ(x,y)³]

[0151] Then, calculate the temperature and viscosity at that pixel based on the calibration model:

[0152] [T(x,y), η(x,y)] = X(x,y) · W + b

[0153] Where W is the multidimensional calibration coefficient matrix, b is the bias term after correction by giant membrane vesicles and fixed cells, T(x,y) is the temperature at the pixel, and η(x,y) is the viscosity at the pixel.

[0154] In the actual calculation process, background subtraction and cell region segmentation are first performed on the cell image. Specifically, cell-free regions can be selected as background regions, and the background intensities of the temperature response channel and reference channel are calculated separately and subtracted from the original image. Then, cell boundaries are determined based on probe fluorescence signals or bright-field images, and a cell mask is generated. Temperature calculation is performed only on pixels within the mask. Pixels with low signal-to-noise ratios, reference channel intensities close to the background, or large lifetime fitting residuals can be discarded to avoid invalid pixels causing noise in the temperature map.

[0155] After calculating all pixels, the T(x,y) values ​​of each pixel are rearranged according to their original spatial coordinates to obtain a single-cell temperature distribution map; simultaneously, the η(x,y) values ​​of each pixel are rearranged to obtain a single-cell viscosity distribution map. The temperature maps can be displayed in pseudo-color, where colors correspond to different temperature values. Furthermore, the average temperature, maximum temperature, minimum temperature, standard deviation, and temperature distribution range of all effective pixels within the cell region can be statistically analyzed to evaluate the overall temperature level and spatial temperature heterogeneity of the single cell.

[0156] For example, at an ambient temperature of 37 ℃, fixed cells, after the above pixel-by-pixel calculation, should exhibit a nearly uniform temperature distribution, with their average temperature basically consistent with the ambient temperature. Live cells, on the other hand, can exhibit a temperature distribution with both localized high-temperature and low-temperature regions, reflecting the processes of metabolic heat production, heat transfer, and heat dissipation within the cell. This pixel-by-pixel calculation method can convert the original fluorescence image into a quantitative temperature map, rather than simply obtaining the average temperature of a single cell.

[0157] Example 1

[0158] Temperature Measurement Based on Multidimensional Fluorescence Signal Calibration under Viscosity Interference

[0159] This embodiment illustrates how to achieve accurate temperature measurement using multidimensional fluorescence signals under conditions of viscosity interference.

[0160] 1. Construction of the Standards System

[0161] First, standard systems with different viscosities and temperatures are configured.

[0162] (1) Prepare a series of solution systems with different viscosities by mixing glycerol and deionized water in different proportions; (2) Obtain samples covering multiple viscosity ranges by changing the volume fraction of glycerol (e.g., 0%–90%); (3) Determine the viscosity value η corresponding to each solution using a rotational viscometer or by consulting a standard data table; (4) Adjust the solution temperature to different set values ​​(e.g., 25–45 °C) using a fluorescence spectrometer with a temperature control module; thereby obtaining a series of standard samples with known parameters: (T i ,η i )

[0163] 2. Fluorescence signal measurement

[0164] Optical signal acquisition was performed on the fluorescent probe under the aforementioned combinations of temperature and viscosity conditions. Specifically, this included:

[0165] (1) Add a fluorescent probe with a concentration of 1 μmol / L to each system;

[0166] (2) Measure its emission spectrum using a fluorescence spectrometer and record it:

[0167] The fluorescence intensity I1 of the first emission channel (e.g., 590 nm)

[0168] The fluorescence intensity I2 of the second emission channel (e.g., 720 nm)

[0169] (3) Calculate the fluorescence intensity ratio: R = I1 / I2

[0170] The fluorescence lifetime τ is measured using a time-resolved fluorescence measurement module (such as TCSPC); thus obtaining the optical response parameter: (Ri , τ i )

[0171] 3. Establishment of the multidimensional response matrix

[0172] The above experimental data are summarized as follows:

[0173] Input matrix:

[0174] Output matrix;

[0175] Based on experimental data, a matrix relationship model between input and output is established: Y = f(X)

[0176] In this embodiment, the relationship is represented in matrix form as follows:

[0177] in:

[0178] W is the coefficient matrix.

[0179] b is the bias term.

[0180] R is the fluorescence ratio.

[0181] τ is the fluorescence lifetime.

[0182] The model parameters are obtained through least squares fitting or orthogonal distance regression.

[0183] 4. Model Validation

[0184] To verify the effectiveness of the established model, several test samples with unknown temperatures were selected. For example, a test solution with known viscosity and temperature was prepared, and these two physical quantities were simultaneously measured using optical signals to verify the feasibility of the method. The fluorescence ratio R and fluorescence lifetime τ of the solution were measured. The measured (R, τ) were then substituted into the matrix model to calculate the predicted temperature T. pred The actual temperature T recorded by the temperature control system. real Compare;

[0185] 5. Results Analysis

[0186] Experimental results show that:

[0187] (1) Under different viscosity conditions: The fluorescence intensity ratio R changes with temperature, but is also affected by viscosity; the fluorescence lifetime τ is also affected by both temperature and viscosity.

[0188] (2) In traditional single-parameter methods: it is difficult to accurately deduce temperature based solely on the fluorescence intensity ratio R or fluorescence lifetime τ; viscosity changes can lead to significant systematic errors;

[0189] (3) In this embodiment: multidimensional input is constructed by using R and τ together; inversion calculation is performed using a matrix relationship model; temperature information can be accurately recovered under different viscosity conditions; the experimentally measured temperature error can be controlled within the range of approximately ±0.08 K.

[0190] This embodiment demonstrates that by constructing a multidimensional fluorescence response input matrix that includes the fluorescence intensity ratio and fluorescence lifetime, and establishing a matrix relationship model between this matrix and temperature and viscosity, accurate temperature measurement can be achieved under viscosity interference conditions. This method overcomes the measurement error problem caused by the coupling of temperature and viscosity in traditional single calibration methods, and realizes temperature decoupling inversion under multi-parameter conditions.

[0191] Example 2

[0192] Intracellular temperature imaging calibrated by multidimensional fluorescence signals

[0193] This embodiment illustrates how, based on the multidimensional fluorescence response input matrix model established in Example 1, spatial distribution characterization of intracellular temperature can be achieved through hierarchical calibration and imaging measurements.

[0194] 1. Hierarchical calibration of matrix relation models

[0195] The matrix relationship model established in Example 1 was obtained in a solution system. To improve its applicability in a cellular environment, further calibration is required.

[0196] The specific steps are as follows:

[0197] (1) Optical signal measurements were performed in the following three types of systems: Solution systems (see Figure 3 (b) Cell-like system (cell-derived vesicles) (see...) Figure 3 c), fixed cell system (see Figure 3 (d) to obtain the optical signal measurement results of the corresponding system (see d) Figure 3 (e).

[0198] (2) In each of the above systems, set a known ambient temperature (e.g., 27–42 °C) and measure the corresponding fluorescence intensity ratio R and fluorescence lifetime τ.

[0199] (3) Compare the measured temperature value with the set temperature and calculate the error;

[0200] (4) Based on error feedback, the coefficient matrix in the matrix relationship model is iteratively corrected until the deviation between the model output temperature and the actual temperature meets the preset accuracy requirements;

[0201] Through the above steps, a calibration matrix model suitable for the cellular environment is obtained.

[0202] 2. Preparation of live cell samples

[0203] (1) Seed the cells to be tested in a culture dish and culture them to an appropriate density;

[0204] (2) Add the fluorescent probe to the culture system and incubate with the cells for about 30 minutes;

[0205] (3) After incubation, change the culture medium and rinse three times with buffer to remove probes that have not entered the cells;

[0206] (4) Place the cells on the microscope stage and stabilize the culture system at the set temperature (e.g., 37 °C) using the temperature control system.

[0207] 3. Multidimensional fluorescence signal acquisition

[0208] Cells were measured using confocal fluorescence lifetime imaging microscopy (CLSM-FLIM), and fluorescence intensity images I1(x,y), I2(x,y), and fluorescence lifetime image τ(x,y) of different wavelength channels were acquired simultaneously.

[0209] The fluorescence intensity ratio is calculated pixel by pixel: R(x,y)=I1(x,y) / I2(x,y)

[0210] This results in the following: ratio matrix diagram and lifetime matrix diagram.

[0211] 4. Pixel-by-pixel temperature inversion calculation

[0212] Substitute the multidimensional input parameters [R(x,y),τ(x,y)] obtained in step three into the calibrated matrix relation model:

[0213] [T(x,y), η(x,y)] = f (R(x,y), τ(x,y))

[0214] By performing pixel-by-pixel matrix operations, intracellular temperature distribution maps T(x,y) and viscosity distribution maps η(x,y) are obtained, achieving cell-scale temperature imaging (see...). Figure 4 ).

[0215] 5. Organelle temperature analysis (colocalization analysis)

[0216] To further obtain the temperature distribution of different organelles, fluorescence co-localization technology was employed, using the mitochondrial-specific dye Mito-Tracker and the lysosome-specific dye LysoSensor to co-incubate cells. Organelle-labeled images were then acquired using confocal microscopy, and the temperature distribution map and organelle-labeled map were spatially registered and overlaid to extract temperature information from different regions (mitochondria, lysosomes, and cytoplasm) (see [link to study]). Figure 5).

[0217] 6. Dynamic temperature measurement during physiological processes

[0218] To verify the applicability of this method in complex physiological processes, external stimuli were introduced into the cell culture system:

[0219] (1) Add Staphylococcus aureus to the culture medium containing cells;

[0220] (2) Record changes in intracellular fluorescence signals over time;

[0221] (3) Calculate the temperature and viscosity distribution in real time using the method described above;

[0222] Experimental results are as follows Figure 6 As shown, during bacterial infection, both intracellular temperature and viscosity increased significantly, and these changes occurred during cell damage and death, indicating that this method can effectively characterize the thermodynamic response during changes in cellular physiological state.

[0223] This embodiment demonstrates that by hierarchically calibrating a multidimensional fluorescence response model and combining pixel-by-pixel analysis of fluorescence intensity ratios and fluorescence lifetimes, high-precision temperature imaging can be achieved in complex cellular environments. This method not only effectively suppresses interference from environmental factors such as viscosity on temperature measurements but also enables subcellular-scale characterization of temperature distribution, providing a reliable technical means for studying intracellular heat transfer and metabolic processes.

[0224] Example 3: Dynamic temperature changes under different drug stimuli

[0225] This embodiment illustrates the measurement of the dynamic temperature changes of a single living cell under different drug stimuli using the method of the present invention. The drug stimuli include calcium ion stimulation to promote cellular thermogenicity, FCCP stimulation to enhance mitochondrial uncoupling thermogenicity, Rotenone stimulation to inhibit the mitochondrial respiratory chain, and Mdivi-1 and STS stimulation to regulate mitochondrial fusion / division states.

[0226] HeLa cells in logarithmic growth phase were seeded in glass-bottomed culture dishes and cultured in high-glucose DMEM medium containing 10% fetal bovine serum, 100 U / mL penicillin, and 100 μg / mL streptomycin at 37 °C and 5% CO2 for 24 hours. After cell attachment, the cells were washed three times with PBS buffer, and DMEM medium containing fluorescent probes was added. The cells were incubated at 37 °C and 5% CO2 for 30 minutes to allow the probes to enter the cells and accumulate in the mitochondrial region. After incubation, the cells were washed three more times with PBS buffer to remove any unexploded probes, and then fresh DMEM medium was added for imaging.

[0227] The stained cells were placed on a confocal laser scanning microscope equipped with a fluorescence lifetime imaging module, and the extracellular temperature was stabilized at 37 °C using a temperature-controlled stage. Fluorescent probes were excited using a pulsed laser, and fluorescence intensity images of the temperature-responsive channel and reference channel were acquired, along with fluorescence lifetime images. Continuous imaging of individual target cells was performed. Baseline images before drug stimulation were acquired first, and the fluorescence ratio map R(x,y) and fluorescence lifetime map τ(x,y) of the cell in the unstimulated state were recorded. The temperature distribution map T0(x,y) before stimulation was calculated using a corrected multidimensional calibration matrix.

[0228] Subsequently, drug stimulants were added to the culture dish, and fluorescence intensity and lifetime images of the same cell were continuously acquired at fixed time intervals after the drug was added, and the average temperature of a single cell over time was obtained.

[0229] Calcium ion stimulation

[0230] Iomycin calcium salt was added to the cell culture medium to induce Ca²⁺ shock and enhance cellular metabolic activity. The concentration of iomycin calcium salt could be set to 0.1-1.0 μM, preferably 1.0 μM. After adding iomycin calcium salt, temperature images of the same cell were continuously acquired. The results are as follows: Figure 5 As shown, the average temperature of single cells gradually increased with increasing Ca²⁺ stimulation; when 1.0 μM ionomycin calcium salt was added, the average cell temperature increased by more than 1 °C. Further analysis of organelle regions revealed that the temperature increase was most significant in the mitochondrial region, indicating that Ca²⁺ stimulation first enhances mitochondrial metabolism and heat production, and then the heat diffuses to the cytoplasm and other organelles.

[0231] FCCP stimulation enhances mitochondrial uncoupling thermogenesis

[0232] FCCP was added to the cell culture medium to enhance mitochondrial uncoupling thermogenesis. FCCP disrupts the proton gradient across the inner mitochondrial membrane, allowing the electron transport chain to continue operating while reducing ATP synthesis efficiency, thus releasing more energy as heat. Temperature images of the same cell or the same mitochondrial region were continuously acquired after the addition of FCCP. Results are as follows: Figure 7 The results showed that the temperature in the mitochondrial region gradually increased over time, indicating that enhanced mitochondrial respiratory metabolism can lead to increased local heat production. Further changes in FCCP concentration revealed that mitochondrial temperature increased with increasing FCCP concentration, but at higher concentrations, the temperature rise tended to saturate or slightly decrease, suggesting that excessive FCCP may impair mitochondrial function and reduce metabolic heat production capacity.

[0233] Rotenone stimulation that inhibits the mitochondrial respiratory chain

[0234] Rotenone was added to the cell culture medium to inhibit mitochondrial respiratory chain complex I. Continuous temperature imaging was performed on the same cells after the addition of Rotenone. Results are as follows: Figure 7 The results show that the temperature in the mitochondrial region gradually decreased, indicating that the inhibition of electron transport in the respiratory chain led to a decline in oxidative phosphorylation efficiency and a reduction in mitochondrial heat production. This result can be used to validate the ability of this method to detect the temperature response to changes in mitochondrial metabolic state.

[0235] Mdivi-1 and STS stimulation regulate mitochondrial fusion / division states

[0236] Mdivi-1 and STS were used to regulate mitochondrial morphology, and the temperature of single cells under different morphological states was measured. Specifically, four experimental groups were set up: Group 1 was a blank control group; Group 2 was treated with Mdivi-1 to inhibit mitochondrial division and promote mitochondrial fusion; Group 3 was treated with STS to induce excessive mitochondrial division; and Group 4 was treated with both Mdivi-1 and STS to inhibit STS-induced excessive division. After drug treatment, temperature imaging was performed on cells in each group, and the temperature of mitochondrial regions was obtained by combining mitochondrial fluorescent labeling or image segmentation. Results are as follows: Figure 8 The results showed that the average temperature of mitochondria in the fused state was slightly lower than that in the normal state, while the temperature of mitochondria in the highly fissuring state was significantly higher; when Mdivi-1 inhibited STS-induced excessive fissuring, the mitochondrial temperature decreased relative to the STS-only treatment group. These results indicate that mitochondrial morphological remodeling is related to the process of localized heat release in the cell.

[0237] As demonstrated by the above embodiments, the method of the present invention can monitor temperature changes induced by different drug stimuli in real time at the level of a single living cell, and can further distinguish the thermal responses of mitochondria, cytoplasm, and other subcellular regions. Compared with traditional methods that rely solely on a single fluorescence intensity or lifetime, this method, through dual parameter input of fluorescence ratio and fluorescence lifetime, combined with a temperature / viscosity multidimensional calibration matrix, can reduce temperature measurement errors caused by factors such as intracellular viscosity changes, probe concentration changes, and photobleaching, thereby achieving quantitative imaging of single-cell temperature under drug stimulation conditions.

[0238] Example 4: Temperature distribution in mitochondrial and lysosomal regions within a single living cell.

[0239] This embodiment illustrates the method of the present invention for measuring the temperature distribution in the mitochondrial and lysosomal regions within a single living cell. This embodiment utilizes multi-channel co-localization imaging with fluorescent probes, mitochondrial-labeled probes, and lysosomal-labeled probes, combined with image segmentation and pixel-by-pixel temperature calculation, to achieve the separation and quantification of temperatures in different organelle regions.

[0240] HeLa cells in the logarithmic growth phase were seeded in glass-bottomed culture dishes and cultured for 24 hours at 37 °C and 5% CO2 in high-glucose DMEM medium containing 10% fetal bovine serum, 100 U / mL penicillin, and 100 μg / mL streptomycin. After cell attachment, the cells were washed three times with PBS buffer, and DMEM medium containing fluorescent probes was added. The cells were then incubated at 37 °C and 5% CO2 for 30 minutes to allow the fluorescent probes to enter the cells. The fluorescent probes have a cationic structure and preferentially accumulate in the mitochondrial region depending on the mitochondrial membrane potential, while some probes can also be distributed in lysosomes and cytoplasm.

[0241] After incubation with the fluorescent probes, the cells were washed three times with PBS buffer to remove free probes. Then, Mito-Tracker Green (mitochondrial labeling probe) and LysoSensor Green DND-189 (lysosome labeling probe) were added to co-localize mitochondria and lysosomes in the same cells. After labeling, the cells were washed again with PBS buffer and fresh DMEM medium was added for imaging.

[0242] The stained cells were placed on a confocal laser scanning microscope equipped with a FLIM module, and the extracellular temperature was stabilized at 37 °C using a temperature-controlled stage. Temperature response channel images, reference channel images, and fluorescence lifetime images of the fluorescent probes were acquired. Simultaneously, mitochondrial fluorescence images corresponding to Mito-Tracker Green and lysosomal fluorescence images corresponding to LysoSensorGreen DND-189 were acquired. Since each fluorescent probe has different excitation and emission wavelengths, independent excitation light and detection windows were set during imaging to reduce channel crosstalk.

[0243] Subsequently, spatial registration was performed on the acquired multi-channel images to ensure that the fluorescent probe images, mitochondrial label images, and lysosomal label images corresponded pixel-by-pixel in the same coordinate system. For the fluorescent probe images, the fluorescence ratio at each pixel was obtained based on the temperature response channel and the reference channel, and combined with the fluorescence lifetime information acquired by FLIM. These two values ​​were then input into a multidimensional calibration matrix corrected for macromembrane vesicles and fixed cells, and the temperature distribution map of the entire single cell was calculated pixel-by-pixel.

[0244] Next, region segmentation was performed based on organelle labeling channels. In the mitochondrial labeled image, the mitochondrial region was identified based on the Mito-Tracker Green signal; in the lysosome labeled image, the lysosome region was identified based on the LysoSensor Green DND-189 signal. Fluorescent probe regions overlapping with the Mito-Tracker Green signal were defined as mitochondrial temperature measurement regions; fluorescent probe regions overlapping with the LysoSensor Green DND-189 signal were defined as lysosome temperature measurement regions; and regions containing fluorescent probe signals but not belonging to the mitochondrial or lysosome labeled regions were defined as cytoplasmic regions.

[0245] After dividing the cells into regions, the mitochondrial, lysosomal, and cytoplasmic regions were mapped to single-cell temperature distribution maps. Temperature information for each region was extracted, and the average, maximum, minimum, and temperature distribution ranges of the mitochondria, lysosomes, and cytoplasm were statistically analyzed. This allows us to obtain the temperature differences between different organelle regions within the same single cell.

[0246] No external stimulation

[0247] When the ambient temperature was maintained at 37 °C and no external stimulation was applied, significant subcellular temperature differences were observed within living cells. The average temperature in the mitochondrial region was approximately 39.0 °C, higher than the average temperature in the cytoplasm region (approximately 37.6 °C); the average temperature in the lysosomal region was approximately 36.7 °C, lower than the mitochondrial region and slightly lower than the cytoplasm region. These results are as follows... Figure 9 As shown, mitochondria, as the main site of oxidative phosphorylation and ATP production, can form a local high-temperature region, while the lysosomal region may exhibit a relatively low temperature due to hydrolysis, acidification processes, or local heat dissipation.

[0248] Ionomycin calcium salt induces Ca²⁺ shock

[0249] Isomycin calcium salt was added to the culture medium to induce Ca²⁺ shock, and the above-described organelle temperature measurement procedure was repeated. After Ca²⁺ stimulation, the average temperature of the mitochondrial region further increased, for example, to about 40.1 °C; the temperature of the cytoplasmic region increased to about 39 °C; and the temperature of the lysosomal region increased to about 38.3 °C. Figure 10 This indicates that after Ca²⁺ stimulation enhances mitochondrial metabolic heat production, heat can diffuse from the mitochondria to the cytoplasm and other organelles, forming a dynamic heat transfer process.

[0250] As demonstrated in this embodiment, the method of the present invention can not only obtain the overall temperature distribution of a single cell, but also separately resolve mitochondrial temperature, lysosomal temperature, and cytoplasmic temperature through organelle colocalization markers and image segmentation. This method is suitable for studying thermal differences between different organelles, mitochondrial heat production, lysosomal heat dissipation, and inter-organelle heat transfer behavior.

[0251] Example 5

[0252] This embodiment uses Staphylococcus aureus infection of HeLa cells as an example to illustrate how the method of the present invention can be used to detect temperature changes in a single cell during bacterial infection.

[0253] HeLa cells in the logarithmic growth phase were seeded in glass-bottomed culture dishes and cultured at 37 °C with 5% CO2 for 24 hours. After cell attachment, the cells were washed with PBS, and culture medium containing a fluorescent probe was added. The cells were incubated at 37 °C for 30 minutes to allow the probe to enter the cells. After incubation, the cells were washed with PBS and fresh culture medium was added for later use.

[0254] Separately, collect Staphylococcus aureus and inoculate it into LB liquid medium, incubating at 37 °C for approximately 12 hours. After incubation, collect the bacterial cells by centrifugation, wash with PBS, and redisperse them in cell culture medium to obtain the bacterial infection solution. Adjust the bacterial concentration according to experimental needs to achieve the preset multiplicity of infection.

[0255] HeLa cells labeled with fluorescent probes were placed in a confocal microscope equipped with a fluorescence lifetime imaging module, and the culture environment was maintained at 37 °C. Before adding bacteria, fluorescence intensity and fluorescence lifetime images of individual cells were acquired, and the cell temperature distribution before infection was calculated as the initial temperature.

[0256] Subsequently, Staphylococcus aureus infection solution was added to the cell culture system, and fluorescence intensity and fluorescence lifetime images of the same cell were continuously acquired during the infection process. At each time point, the single-cell temperature distribution map was calculated by substituting the fluorescence ratio and fluorescence lifetime of the fluorescent probe into a corrected multidimensional calibration matrix.

[0257] The results showed that the average temperature of uninfected cells at 37 °C was approximately 37.6 °C; after the addition of Staphylococcus aureus, the cell temperature rose rapidly, reaching an average of approximately 39.6 °C about 10 minutes after infection. This result indicates that bacterial infection can induce enhanced cellular metabolism and stress response, thereby causing an increase in single-cell temperature.

[0258] As can be seen from this embodiment, the method of the present invention can be used to monitor the temperature changes of individual living cells in real time during bacterial infection, and can be used to evaluate the cellular thermal response caused by pathogenic microorganism stimulation.

[0259] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention. Equivalent substitutions or improvements made by those skilled in the art without departing from the principles of the present invention should all fall within the scope of protection of the present invention.

Claims

1. A method for measuring intracellular temperature based on multidimensional fluorescence signal calibration, characterized in that, Includes the following steps: Step 1: Construct a fluorescent probe, which includes a temperature-responsive unit, a reference fluorescent unit, and a flexible linker group. The temperature-responsive unit includes a rhodamine structure, the reference fluorescent unit includes a hemicyanine structure, and the temperature-responsive unit and the reference fluorescent unit are linked by the flexible linker group. Step 2: Establish the spectral readout channel of the fluorescent probe, characterize the excitation and emission spectra of the fluorescent probe, and collect the fluorescence intensity of two fluorescence channels of the temperature-response unit and the reference fluorescent unit; perform fluorescence lifetime imaging on the rhodamine structure emission channel of the temperature-response unit to obtain the fluorescence lifetime τ of each pixel. Step 3: Construct a temperature-viscosity multidimensional calibration system, prepare standard solutions with different viscosities, adjust the temperature of the standard solution to multiple temperature points at each viscosity, measure temperature, viscosity, fluorescence intensity and fluorescence lifetime, establish an initial two-dimensional operation matrix, and obtain temperature and viscosity simultaneously through matrix operations; Step 4: Prepare single-cell samples to be tested. After seeding the cells to be tested in a glass-bottomed culture dish and culturing them, wash the cells with PBS, add culture medium containing the fluorescent probe prepared in step 1 for incubation, and wash with PBS after incubation to remove the free fluorescent probe. Step 5: Obtain the fluorescence ratio map and lifetime map of the single cell to be tested. Place the single-cell sample processed in Step 4 on a confocal laser scanning microscope equipped with a FLIM module, stabilize the extracellular environment at the set temperature, excite the fluorescent probe with a pulsed laser, and acquire the fluorescence lifetime image. Simultaneously, acquire the fluorescence intensity images of the two fluorescence channels of the temperature-response unit and the reference fluorescence unit to obtain the Ig of each pixel. 590 and I 720 The fluorescence ratio map R(x,y) is calculated; then, the fluorescence decay curve of each pixel is fitted by a double exponential method to obtain the fluorescence lifetime map τ(x,y). Step 6: Correct the calibration matrix using a cell simulation system. This correction includes primary calibration of macromembrane vesicles and secondary calibration of fixed cells. The primary calibration of macromembrane vesicles involves preparing macromembrane vesicles labeled with fluorescent probes, placing these vesicles in an imaging system with a temperature-controlled stage, acquiring fluorescence intensity channel images and fluorescence lifetime images at multiple temperatures, calculating the fluorescence ratio and lifetime, and substituting these values ​​into the initial two-dimensional calibration matrix described in Step 3 to obtain the vesicle measurement temperature T. 测 The actual ambient temperature T recorded by the temperature control system 实 The measured temperature is compared with the actual temperature, and a correction is performed using the temperature deviation to ensure that the calculated temperature in the giant membrane vesicles is consistent with the actual set temperature. The second calibration of the fixed cells involves fixing the cells after incubation with the fluorescent probe, placing the fixed cells in an imaging system with a temperature-controlled stage, acquiring fluorescence intensity ratio maps R(x,y) and fluorescence lifetime maps τ(x,y) at multiple temperatures, substituting them into the calibration matrix after the first calibration of the giant membrane vesicles, obtaining a pixel-by-pixel temperature map of the fixed cells, calculating the deviation between the measured temperature of the fixed cells and the actual set temperature, and performing a second calibration on the temperature bias term of the calibration matrix to obtain the final calibration matrix after correction by the cell simulation system, which is suitable for the cell environment. Step 7: Calculate the single-cell temperature distribution pixel by pixel. After obtaining the fluorescence intensity image and fluorescence lifetime image of a single cell, calculate the temperature distribution map of each pixel independently. Fit the fluorescence decay curve obtained by FLIM to obtain the fluorescence lifetime τ(x,y) of the pixel. Input the R(x,y) and τ(x,y) of each pixel into the final calibration matrix corrected by the cell simulation system as described in step S6. After completing the calculation of all pixels, rearrange the T(x,y) of each pixel according to its original spatial coordinates to obtain the two-dimensional temperature distribution map of the single cell; at the same time, rearrange the η(x,y) of each pixel to obtain the viscosity distribution map of the single cell.

2. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 2, the fluorescence probe is characterized by excitation and emission spectra. The fluorescence intensities of two fluorescence channels of the temperature-responsive unit and the reference fluorescence unit are collected. The first channel is the temperature-responsive channel, and the emission peak of the temperature-responsive unit is denoted as I. 590 The second channel is the reference channel, representing the emission peak position of the fluorescent reference unit, denoted as I. 720； Calculate the fluorescence intensity ratio: R = I 590 / I 720 Fluorescence lifetime imaging was performed on the rhodamine structure emission channel of the temperature-responsive unit to obtain the fluorescence lifetime τ of each pixel. The fluorescence lifetime τ can be expressed by a bi-exponential fit as follows: I(t) = A1 exp(-t / τ1) + A2exp(-t / τ2) Where t is time, A1 and A2 are the amplitudes of the decay components, and τ1 and τ2 are the corresponding fluorescence lifetimes. The weighted average fluorescence lifetime is calculated as the fluorescence lifetime parameter τ of the pixel.

3. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 3, an initial two-dimensional calibration model is established for simultaneous temperature and viscosity quantification. For each standard sample, the following parameters are measured: true temperature T, actual viscosity η, fluorescence intensity ratio R, and fluorescence lifetime τ. R and τ are used as inputs, and T and η are used as outputs to establish the initial two-dimensional calibration model. (T, η) = f(R, τ), using a cubic polynomial feature extension model, the input signal is extended to: [R, R², R³, τ, τ², τ³] And by solving the calibration coefficient matrix W and the bias term b through least squares regression, we obtain: [T, η] = [R, R², R³, τ, τ², τ³] · W + b The initial two-dimensional calibration model separates the temperature information coupled in fluorescence ratio and fluorescence lifetime from the viscosity information, enabling simultaneous quantification of temperature and viscosity.

4. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 4, the step of seeding the cells to be tested in a glass-bottomed culture dish involves culturing the cells in high-glucose DMEM medium containing 10% fetal bovine serum, 100 U / mL penicillin, and 100 μg / mL streptomycin at 37 ℃ and 5% CO2. The step of incubation involves incubating at 37 ℃ and 5% CO2 for about 30 minutes until the cells adhere to the dish and are in a good growth state.

5. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 5, the temperature range is set to 20 ℃-40 ℃, the pulsed laser excitation probe is set to use a 561 nm pulsed laser with a repetition frequency of about 20 MHz, and a fluorescence lifetime image in 512×512 pixel format is acquired.

6. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 6, the preparation of fluorescently labeled giant membrane vesicles involves seeding cells in a glass-bottomed culture dish and culturing them under CO2 conditions. After the cells adhere to the dish and reach approximately 70% confluence, they are washed with PBS buffer, and DMEM culture medium containing the fluorescent probe is added for incubation. This allows the fluorescent probe to enter the cell membrane and intracellular environment. After incubation, the cells are washed with GPMV buffer, and then vesicle induction solution is added for incubation to induce the cell membrane to detach and form giant membrane vesicles, thus preparing fluorescently labeled giant membrane vesicles.

7. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 6, a temperature deviation correction is performed by comparing the measured temperature with the actual ambient temperature T recorded by the temperature control system, and calculating the deviation at each temperature point: ΔT = T 实 - T 测 Based on the average deviation obtained from multiple temperature points, the temperature output bias term in the calibration matrix is ​​corrected once to ensure that the calculated temperature in the giant membrane vesicles is consistent with the actual set temperature. The calibration model is expressed as follows: [T, η] = [R, R², R³, τ, τ², τ³] · W + b Keeping the coefficient matrix W unchanged, we first correct the components of the bias term b corresponding to the temperature output, so that the corrected model satisfies: T 修正1 = T 测 + ΔT 平均 .

8. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 6, the fixation of cells incubated with the fluorescent probe involves washing the cells with PBS, adding 4% paraformaldehyde solution for fixation, washing the cells again with PBS, and adding culture medium or PBS to keep the sample moist. The secondary correction involves calculating the deviation between the measured temperature of the fixed cells and the actual set temperature, and performing a secondary correction on the temperature bias term of the calibration matrix. ΔT 固定 = T 实 - T 测 T 修正2 = T 测 + ΔT 固定平均。 9. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 7, the fluorescence decay curve of the pixel obtained by FLIM is fitted, and the fluorescence intensity of the temperature response channel at pixel (x,y) is set as I. 590 (x,y), the fluorescence intensity of the reference channel is I 720 Given (x, y), the fluorescence ratio of that pixel is: R(x,y) = I 590 (x,y) / I 720 (x,y), where (x,y) are pixel coordinates.

10. The intracellular temperature measurement method based on multidimensional fluorescence signal calibration according to claim 1, characterized in that, In step 7, the input of R(x,y) and τ(x,y) of each pixel into the final calibration matrix corrected by the cell simulation system in step S6 is as follows: R(x,y) and τ(x,y) are expanded using a cubic polynomial feature extension to construct the input vector for that pixel. X(x,y) = [R(x,y), R(x,y)², R(x,y)³, τ(x,y), τ(x,y)², τ(x,y)³] Then, calculate the temperature and viscosity at that pixel based on the calibration model: [T(x,y), η(x,y)] = X(x,y) · W + b Where X is the measured fluorescence parameter, W is the multidimensional calibration coefficient matrix, b is the bias term corrected by giant membrane vesicles and fixed cells, T(x,y) is the temperature at the pixel, and η(x,y) is the viscosity at the pixel; the pixel rearrangement is to rearrange the T(x,y) of each pixel according to its original spatial coordinates to obtain a single-cell two-dimensional temperature distribution map; at the same time, the η(x,y) of each pixel is rearranged to obtain a single-cell two-dimensional viscosity distribution map.

Images