RU2756994C1 - Method for detection and classification of protein molecules and their interactions - Google Patents

Abstract

FIELD: analytical chemistry.

SUBSTANCE: invention relates to analytical chemistry, in particular to methods for the detection and classification of protein molecules and products of their interactions in solutions based on optical sensor devices without the use of analytical labels. A method for detecting interactions of protein molecules with a specific binding site includes preparing model samples, which are at least one solution of a detected protein, a solution of a specific binding site and a mixture of a protein solution with solutions of specific binding sites with different concentrations, introducing solutions and mixtures into the internal cavity of a microstructured optical fibers, detection of transmission spectra of solutions and mixtures, which are used to consider the presence of interactions, in this case, the detection of the transmission spectra is carried out in the spectral range from 250 nm to 3000 nm, the data of the transmission spectra of the model samples are used to construct a mathematical model of the interaction of the detected proteins, for this, the data of the spectra are presented in the form of a matrix, the rows of which correspond to the sample numbers, and the columns - to the intensities of the spectral signal, the matrix is decomposed by the method for principal components, graphs of scores and loads are obtained, on which groups of points characterizing the presence or absence of interaction are distinguished, a test solution and mixtures of a test solution with a specific binding site with different concentrations are prepared, the test solution and mixtures are introduced into the internal cavities of the microstructured optical fiber and detect the transmission spectra, which are compared with the spectra of the model samples, and when the spectral data of the tested solutions fall into the corresponding group of points of the model samples, the presence or absence of interaction is considered.

EFFECT: reduction of the analysis time, increase in the accuracy and reliability of the detection and classification of protein molecules and products of their interactions in solutions.

1 cl, 2 dwg

Description

The invention relates to analytical chemistry and, in particular, to methods for the detection and classification of protein molecules and products of their interactions in solutions based on optical sensor devices without the use of analytical labels. Optical sensor devices can be made on the basis of various optical fibers (waveguides): planar, microstructural, etc.

To date, known sources have proposed a large number of sensors using microstructured optical fibers (MF) as a sensitive element for analyzing gases and liquids. Common to all the proposed schemes is the use of the effect of the interaction of radiation with the analyte introduced into the MF, both in the hollow volumes of the fiber and on its surface, and the registration of changes in the optical properties of the MF corresponding to this interaction.

The detection and classification of specific interactions of protein molecules play an essential role in the functioning of living systems, scientific research and medical practice.

A known method of detecting protein molecules by functionalizing the surface of an optical fiber, which makes it possible to create biosensors for the analysis of specific interactions between target proteins (see Huang Y., Yu B., Guo T., Guan BO Ultrasensitive and in situ DNA detection in various pH environments based on a microfiber with a graphene oxide linking layer // RSC Advances 2017. V. 7. No. 22. P. 13177). The method uses the modification of the MW surface with carboxylated graphene oxide capable of π – π interaction with DNA molecules in a wide pH range (4.3–8.5), which is promising for in situ determination of single-stranded DNA in the human body.

However, this approach has drawbacks associated with a long and complex procedure for forming a sensitive layer on the surface of the MB, which has the stability of properties on the entire surface of the MB and is stable during storage.

A known method for detecting the cardiac biomarker troponin cTnI without the use of analytical labels using photonic crystal fibers with an open microcavity (see Zhang, B., Morales, AW, Peterson, R., Tang, L., & Ye, JY (2014). Label-free detection of cardiac troponin I with a photonic crystal biosensor Biosensors and Bioelectronics, 58, 107-113 . ). The sensor structure contains an open cavity that facilitates the immobilization of cTnI-specific antibodies, which are immobilized on the sensor surface for specific detection of cTnI in a wide concentration range.

The disadvantage of this method is, first of all, the duration, complexity and laboriousness of creating a homogeneous layer on the inner surface of the MW, which is specific to the sorption of the target analyte.

There is also known a method of detecting and classifying protein molecules without analytical labels using plastic sensor elements obtained by microreplication (see Japanese patent No. 4504686 according to IPC class
B01L3 / 00, publ. 02.06.2005). The possibility of detecting various protein molecular complexes, including oligonucleotides, antibody-antigen interactions, hormone-receptor interactions, and enzyme-substrate interactions, is described. The sensor device consists of two parts: a specific recognition element, an analytical signal detection system including a transducer that converts a molecular recognition event into a quantifiable signal. A specific recognition element is made on the basis of an optical grating containing a material with a high refractive index, a substrate layer supporting the optical grating, and substances immobilized on the surface of the optical grating capable of specifically binding the analyzed molecules. When radiation interacts with an optical grating, a detectable resonance effect arises in the reflected radiation spectrum.

The disadvantages of this method are the use of a reflected signal, which significantly limits the possibilities when analyzing the interactions of protein molecules, as well as a complex, time-consuming and laborious procedure for manufacturing microreplications of optical gratings with a resolution of 0.1 μm with substances immobilized on the surface of the optical grating capable of specific binding of the analyzed molecules.

A common disadvantage of the methods described above for the detection and classification of protein molecules and the products of their interactions is the use of specific binding elements (centers) immobilized on a solid support, for example, antibodies, that is, analytical determinations in a heterogeneous format, which involves long and difficult to reproduce stages of synthesis of a modified solid support. and a long-term procedure of incubation of the analyzed solutions on a support for specific binding of the determined protein molecules and obtaining a significant analytical signal.

An alternative to the heterogeneous format of determinations is the homogeneous one, in which the detection of protein molecules and the products of their interaction is carried out by introducing specific binding sites directly into the analyzed solution.

Closest to the proposed invention is a method for determining the interactions of antibodies with antigens in microstructured optical fibers (Pavel S. Pidenko, Andrey A. Shuvalov, Anastasia A. Zanishevskaya, Natalia A. Burmistrova, "Detection of antigen-antibody interactions in microstructured optical fibers," Proc. SPIE 11457, Saratov Fall Meeting 2019 :), including the preparation of model samples representing a solution of a detectable protein, a solution of a specific binding site, and a mixture of a protein solution with solutions of specific binding sites with different concentrations, the introduction of solutions and mixtures into the internal cavity of a microstructured optical fiber, detection of transmission spectra of solutions and mixtures, which are used to judge the presence of interactions. The prototype shows the fundamental possibility of detecting the interaction of antigens with antibodies using the example of a pair of mouse antibodies (mAb) and rabbit anti-mouse antibodies (rAmAb) in microstructured optical fibers in a homogeneous format. A rAmAb solution (100 µg / L) was used to bind various concentrations of mAbs (1-100 µg / L). MAb / rAmAb mixtures were prepared by adding the mAb solution to the rAmAb solution in a 1: 1 ratio. The mixtures were incubated for 30 minutes and the transmission spectrum of the fiber was detected. Based on the change in the intensity of the characteristic maximum in the transmission spectrum of the fiber, the degree of antibody interaction was assessed.

The disadvantages of the prototype include the use of the visible spectral range of wavelengths (from 400 to 700 nm), which does not cover the near ultraviolet (UV) and infrared (IR) regions, in which significant absorption and scattering of radiation by protein molecules is carried out, as well as the use of a one-dimensional approach ( estimation of the signal intensity at one wavelength) when processing spectral data, which, due to the complex nature of these transmission spectra of MF, does not allow separating similar and overlapping changes in the transmission spectra of MF, which significantly reduces the accuracy of determinations.

The technical problem of the claimed invention is the creation of a method for the detection and classification of protein molecules and products of their interactions in solutions using microstructured optical fibers in a homogeneous format without using the stage of immobilizing specific binding sites on the fiber surface and without using analytical labels.

The technical result consists in reducing the time and increasing the accuracy and reliability of detection, and classification of protein molecules and products of their interactions in solutions.

To achieve a technical result in a method for detecting interactions of protein molecules with a specific binding site, including the preparation of model samples, which are at least one solution of a detected protein, a solution of a specific binding site, and a mixture of a protein solution with solutions of specific binding sites with different concentrations, introduction of solutions and mixtures into the inner cavity of a microstructured optical fiber, detection of the transmission spectra of solutions and mixtures, which are used to judge the presence of interactions,according to the invention, the transmission spectra are detected in the optical spectral range from 250 nm to 3000 nm, the transmission spectra data of the model samples are used to construct a mathematical model of the interaction of the detected proteins, for this, the spectral data are presented in the form of a matrix, the rows of which correspond to the sample numbers, and the columns - to the spectral signal intensities , the matrix is decomposed by the method of principal components, graphs of scores and loads are obtained, on which groups of points characterizing the presence or absence of interaction are distinguished, a test solution and mixtures of a test solution with a specific binding site with different concentrations are prepared, the test solution and mixtures are introduced into the internal cavities microstructured optical fiber and detects transmission spectra that compared with the spectra of the model samples, and when the spectral data of the tested solutions fall into the corresponding group of points of the model samples, the presence or absence of interaction is judged.

The invention is illustrated by drawings, where:

- Fig. 1 shows the distribution of point groups of model samples: a solution of a model protein, in particular immunoglobulin G (region 1), a solution of a specific binding site, in particular protein A (region 2), a mixture of solutions of specific binding sites with a solution of a model protein, in particular - a protein A / immunoglobulin G complex (region 3);

- in Fig. 2, the distribution of groups of points of the test solutions relative to the model samples is represented by symbols in the form of crosses: the presence of immunoglobulin G in the test fluid - in area 2, the formation of a protein A / immunoglobulin G complex - in area 3;

The method is carried out as follows.

Model solutions are prepared: a solution of a detected protein, a solution of a specific binding site, and a mixture of a protein solution with solutions of specific binding sites with a concentration (plus and minus 30% of the maximum and minimum possible).

A microstructured optical fiber with a hollow, suspended, solid core, or a fiber of another design with hollow microcapillaries is used. (see, for example, Pavel S. Pidenko, Andrey A. Shuvalov, Anastasia A. Zanishevskaya, Natalia A. Burmistrova, "Detection of antigen-antibody interactions in microstructured optical fibers," Proc. SPIE 11457, Saratov Fall Meeting 2019 :) ...

Solutions and mixtures are injected into hollow microcapillaries of a microstructured optical fiber under the action of capillary forces, a pump, or by any other available method.

Since individual protein molecules and their interaction products absorb and scatter radiation in the optical wavelength range from 250 nm to 3000 nm, a radiation source and a detector operating in this spectral range are used to obtain the transmission spectra of MW.

The data of the transmission spectra of the model samples are used to construct a mathematical model of the interaction of the detected proteins. When building a model, multidimensional methods of analysis are used, for example, the method of principal components.

For processing, the data of the transmission spectra of the MW are presented in the form of a matrix X consisting of n objects (rows) and p characteristics (columns). Thus, an n × p data matrix is represented as follows:

(1)

(1)

At the first stage, the data matrix is checked for completeness and, if necessary, the gaps in the data are filled by the mean value of the column or row, or the nearest neighbors. In cases where the data is highly correlated with each other or redundant and constant, it is removed from the sample.

центрируется вычитанием среднего значения столбца

в соответствии с уравнением (2):Next, the data is centered. Centering is the subtraction of the matrix M from the original matrix X. The method uses average centering, where each variable

centered by subtracting the mean of the column

according to equation (2):

(2)

(2)

Further, the data are processed by the method of principal components in order to reduce the dimension of the spectral data without significant loss of information due to the latent data structure. The data matrix is decomposed, in which the number of rows corresponds to the number of samples, and the number of columns corresponds to the number of independent variables (in the particular case, analytical signals), into the product of two matrices T and P with lower dimensions and a matrix of residuals E, which contains background information (noise ) (Rodionova O.E. Chemometrics in analytical chemistry // 2006.61p.).

Next, a sample of biological or other analyzed fluid is taken. The necessary sample preparation and dilution is carried out in order to obtain a visually transparent initial solution. When preparing solutions, preparing samples and performing measurements, the following conditions are observed: ambient air temperature (20 ± 5) ° С; atmospheric pressure 84.0-106.7 kPa (630-800 mm Hg); relative humidity at 25 ° C below 85%. The introduction of specific binding sites with a known concentration into the solution is carried out.

The samples prepared in this way are introduced into the hollow microcapillaries of a microstructured optical fiber and the transmission spectra of the MW filled with various samples are detected. The MW transmission spectrum is used as an analytical signal. In this case, the detection and classification of protein analyte molecules and the products of their interactions is determined by the change in the characteristics of the transmission spectrum of MW. The detection and classification of protein analyte molecules and the products of their interactions is carried out on the basis of the above constructed mathematical model by the method of principal components.

Example. Application of protein A for specific recognition of immunoglobulins G in human blood plasma.

At the first stage, a multidimensional mathematical model of a model dataset is constructed. To do this, prepare a set of at least 15 solutions of protein A, immunoglobulin G in various concentrations and ratios. Transmittance spectra of MW filled with prepared solutions are obtained. The transmission spectra data are presented in the form of a matrix consisting of n objects (rows) and p characteristics (columns). where the lines correspond to the number of the sample, and the columns to the intensities of the spectral signal. Check the data matrix for completeness and, if necessary, fill in the gaps in the data by the mean value of the column or row, or the nearest neighbors. If necessary, remove redundant data and perform centering. Next, the matrix is decomposed by the method of principal components, the graphs of accounts and loads are obtained and analyzed.

In this case, the graph of accounts (see Fig. 1) contains three characteristic groups of points: spectral data of blood plasma samples containing immunoglobulins G (region 1), protein A (region 2) and protein complexes protein A / immunoglobulin G (region 3).

Determination of the presence of immunoglobulins G using protein A is carried out according to the following method. Initially, biological material (whole blood) is taken into vacuum tubes with lithium heparin. Within 24 hours after sampling, the necessary sample preparation procedures are carried out: centrifugation of samples to isolate blood plasma, dividing (aliquoting) a blood plasma sample into equal volumes for parallel studies (n = 5) and calculating the standard deviation. Immediately before the introduction of a specific binding site (Protein A (Sigma-Aldrich, USA, cat. # P 6031), blood plasma samples are serially diluted using phosphate buffered saline until the maximum optical density (A = 0.8 a.u) is reached. range 200-1200 nm. Control of optical density is carried out on a spectrophotometer Shimadzu UV-1800 (Shimadzu Europa GmbH, Duisburg, Germany). To the prepared blood plasma sample, add a solution of Protein A at a concentration of 2 g / l. After adding Protein A, the mixture incubated at a temperature of 36.6 degrees Celsius, with constant stirring (200 rpm) for 15 minutes.Then, the mixture (100 μl) is injected with a syringe pump into the internal cavities of the MV, for the stability of the optical signal, a cuvette with an installed sample of MV, similarly filled with a mixture (200 μl) after filling with the mixture, the transmission spectra of the MW are detected.After recording the transmission spectra of the mixture in MW, the spectral the data normalized to the spectral data of the radiation source is used as a test dataset. The test dataset is related to the model dataset. The result is a determination in which of the three characteristic regions of the groups of points of the model dataset (Fig. 1) the obtained spectral data of the test set (Fig. 2) fall. If the formation of the protein A / immunoglobulin G protein complex is observed in the samples of the test dataset, then they will be located in the region of 3 characteristic group, if absent, in the region of group 2, respectively.

Thus, the introduction of protein A into a biological fluid sample and subsequent detection and classification of its possible interactions can be used to quantify the content of immunoglobulins G. The concentration burst of immunoglobulins G in a biological object obtained from a patient indicates the presence of a pronounced immune response. Since the half-life for IgG1, IgG2 and IgG4 is 21 days, the detection and classification of this type of interaction is applicable when creating rapid tests for recognizing the immune response in the development and testing of vaccines, as well as for the formation of risk groups in epidemic conditions.

Detection and classification of interactions between specific IgG and protein antigens is applicable in the development of rapid test systems and streaming monitoring systems, instead of the methods of classical enzyme-linked immunosorbent assay, which can significantly reduce the time of analysis and interpretation.

The following examples can be cited as protein molecules and products of their interactions that can be detected and classified by the proposed method.

1. Biotin ( X ) Biotinase - Carboxylases (combined with biotin) - Streptavidin / Avidin ( Y ). Biotin is a vitamin, deficiency of which contributes to the development of a number of diseases. Streptavidin and avidin are proteins that selectively bind biotin. The introduction of streptavidin and avidin proteins into a sample of biological fluid (blood serum) and subsequent detection of BBV is applicable for the rapid determination of the level of free biotin in the treatment and prevention of biotin deficiency in a patient.

2. Polymerase, specific and nonspecific products of its reactions during the polymerase chain reaction procedure in various versions.

3. Protein pRb ( X ) - transcription factor E2F ( Y ). Retinoblastoma protein (pRb) is a cell cycle inhibitor and acts as a tumor suppressor by binding to transcription factors of the E2F family. pRb inactivates the transcription of the E2F-DP complex (consisting of dimers of the E2F protein and a dimerized partner of the DP protein), which activates cell cycle progression. In the event of a mutation in both alleles of the RB1 gene (encoding pRB), pRB is inactivated, which leads to the development of retinoblastoma. Detection of the interaction between the patient's pRb and the introduced transcription factor E2F allows detecting a genetic mutation and initiating therapy at an early stage of retinoblastoma development, which increases the patient's chance of survival and preserving the eye and its function as an organ of vision.

Thus, the claimed method improves the accuracy and reliability of the detection and classification of protein molecules and products of their interactions in a homogeneous format using microstructured optical fibers without using the stage of immobilizing specific binding centers on the fiber surface and without using analytical labels.

Claims (1)

A method for detecting interactions of protein molecules with a specific binding site, including the preparation of model samples, which are at least one solution of the detected protein, a solution of a specific binding center and a mixture of a protein solution with solutions of specific binding sites with different concentrations, introducing solutions and mixtures into the internal cavity of a microstructured optical fiber, the detection of transmission spectra of solutions and mixtures, which are used to judge the presence of interactions, characterized in that the detection of transmission spectra is carried out in the spectral range from 250 nm to 3000 nm, the data of transmission spectra of model samples are used to construct a mathematical model of the interaction of detected proteins, for the data of the spectra are presented in the form of a matrix, the rows of which correspond to the numbers of the sample, and the columns to the intensities of the spectral signal; plots of counts and loads are calculated, on which groups of points characterizing the presence or absence of interaction are distinguished, a test solution and mixtures of a test solution with a specific binding site with different concentrations are prepared, the test solution and mixtures are introduced into the internal cavities of a microstructured optical fiber, and transmission spectra are detected, which are compared with the spectra of the model samples, and when the spectral data of the tested solutions fall into the corresponding group of points of the model samples, the presence or absence of interaction is judged.

Images