Proteomics - high-tech "fishing"

We live in an era that regularly enriches the language with a mass of new words, terms and concepts. Areas of knowledge and the information that fills them multiply faster than human capabilities allow us to realize this process. Proteomics is a new branch of science that has appeared quite recently. The object of its study are proteins - the "workhorses" of any cell. Interest in these compounds is caused not only by their huge role in the functioning of living organisms, but also by the fact that they can serve as targets in the creation of new drugs.

The word "protein", which is usually used in scientific literature to denote protein(high molecular weight organic compound, which is a chain of amino acids), is derived from the Greek proteios- "original". Its etymology accurately describes the role of proteins in maintaining the life of the cell of any organism, from bacteria to humans.

But the term "proteome" is familiar, perhaps, only to specialists. It is interesting that it was invented in 1994 by an Australian student M. Wilkins, who was trying to find a short name for a complete set of human proteins in his thesis work instead of the clumsy phrase “all proteins expressed by the genome”. The very next year, the term “proteomics” (proteomics) appeared in the scientific press, denoting a new direction in molecular biology (Wasinger et al., 1995).

It should be noted that the very proposal to “create a human molecular protein atlas” was voiced about thirty years ago (Anderson, 1985). But at that time it was rather a wish, technologically impracticable. What has changed over the past decades? First, a genomics program was successfully implemented, including the development of technologies for rapid DNA sequencing and the creation of databases of nucleotide sequences. The second premise is related to the explosive development of instrumental methods: mass spectrometry of proteins and peptides (small chains of amino acids), as well as electrophoresis and chromatography, used to separate organic molecules.

These factors made possible the emergence of a new "high-tech" biological discipline.

Protein abundance

Unlike genomes, proteomes, i.e., the complete sets of cell proteins, are represented by an active set of molecules that are constantly being modified. At the same time, if biochemistry deals with individual isolated molecules, then in the case of the proteome, we are dealing with a huge molecular pool (it is appropriate to draw an analogy with a fish caught on a bait and an abundance of fish brought by a net).

From these provisions follow the goals and objectives of the science of proteomics. First of all, this discipline is responsible for protein "systematics" - an inventory of all proteins encoded in the genome of a particular organism, which also involves the construction of molecular protein atlases of individual cells, organs and tissues.

However, a more interesting and much more significant task (let's call it "physiological"), which consists in determining the principles of interaction between proteins, as well as in establishing the patterns of regulation of their work during the so-called post-translational modification(change) of proteins. This is important for the search for new markers of pathological processes (diseases) in the human body.

The fact is that post-translational modification of proteins occurs in the cell after their synthesis in response to some external disturbance or disease. As a result, the properties of proteins can change rapidly, which affects the rate of their synthesis and degradation. As a result, the result of such processes will be reflected in the overall profile of proteins. By studying it, one can detect proteins whose “production” in a state of illness differs from the “healthy norm”. Such proteins can be used for diagnostic purposes as biomarkers of a particular disease.

Functional, structural and medical

Proteomics is a young science, but research in this area already has good organizational support. In 2001, the Human Proteome Organization (HUPO) was founded, an international organization that unites and directs the efforts of scientists.

Proteomics is a classic example of an interdisciplinary science: it combines biology, chemistry, computer modeling, complex instrumental technology

The official HUPO page details the main areas of research, the list of which can tell a lot even to a non-specialist: human proteome, brain proteomics, study of antibodies, diseases caused by disorders of sugar metabolism, proteomics of cardiovascular diseases, proteomics of stem cells, determination of biomarkers of diseases, study of diseases human on mouse models, etc.

Methodologically, there are several directions in proteomics, the main ones being functional, structural, and medical (clinical) proteomics.

The goals of functional proteomics have already been mentioned above. This is obtaining information about protein-to-protein interactions and their influence on the expression and modulation of gene activity, as well as post-translational modification of proteins within protein complexes.

Structural proteomics, despite being a classic area of ​​protein research, nevertheless continues to develop rapidly due to the improvement of analytical methods, such as new versions of NMR spectroscopy, X-ray diffraction analysis and mass spectrometry.

HOW MOLECULES LEARNED TO "FLY"

Modern mass spectrometric methods are the instrumental basis of proteomics, metabolomics, pharmacokinetics and other “-omics”.


What does a mass spectrometer measure? Strictly speaking, it detects not the mass of the molecule, but the ratio m/z (the ratio of the mass of an ion to its charge). In a mass spectrometer, often the same molecules can have different charges and, accordingly, be detected at different times. It is clear that molecules that do not have a charge (neutral) do not interact with an electric and (or) magnetic field and are invisible to a mass spectrometer.
The whole history of this young science is replete with the names of Nobel laureates. Professor of Cambridge University J. Thompson (Nobel Prize in Physics in 1906) is deservedly considered the creator of the first mass spectrometer. century observed changes in the movement of ions under the influence of an electromagnetic field. Based on these observations, he created a "parabolic spectrograph", in which molecular ions moved in an electric field along parabolic trajectories and were detected by the glow of a luminescent screen.
This work was developed and improved by Thompson's colleague Professor F. Aston, who was awarded the Nobel Prize in Chemistry in 1922 for the creation of a mass spectrograph.


At the same time, Professor A. Dempster of the University of Chicago worked on improving the efficiency of molecular ionization in a mass spectrometer. In 1918, he created a mass spectrometer in which molecules were ionized by a directed electron flow. This method is still the main one for the ionization of small, easily volatile molecules. The author himself used a device to determine the isotopic composition of elements. Having become a full participant in the "nuclear race", in 1935 he discovered the uranium isotope 235 U. His follower, professor of the University of Minnesota A. Nier, became one of the key participants in the Manhattan project: the first atomic bomb was created from 235 U, isolated by Nier using masses -spectrometer.
The development and improvement of methods for the separation and identification of molecular ions was carried out in the direction of creating physical models that make it possible to discriminate ions according to their characteristics. The most obvious way is to let the ions fly in a vacuum and separate in velocities as a derivative of the mass of the molecule. For example, in 1946 Professor W. Stephens of the University of Pennsylvania proposed time-of-flight mass spectrometry. This method is based on the fact that all ions are accelerated by an electric field and receive the same kinetic energy. The speed of each of them depends on the m/z ratio, which makes it easy to determine them.
Alternative method for separating molecular ions in the mid-1950s was proposed by the professor of the University of Bonn W. Paul. He developed a method for separating ions in an alternating electric field, initiating another type of mass spectrometer, the ion trap. The quadrupole Paul trap uses constant and alternating (radio frequency) electric fields to confine (separate) ions. Although the accuracy of this method is significantly lower than that of a time-of-flight mass spectrometer, it has advantages in terms of scanning speed and dynamic range. Therefore, ion traps are ideal detectors for analysis in metabolomics, pharmacology, environmental studies, and their creator was also awarded the Nobel Prize in Physics in 1989 (together with H. Dehmelt).

The mass spectrometer, whose design foundations were laid in the 1920s, is still successfully used today. This excellent laboratory tool, capable of very accurately characterizing both chemical elements and small organic molecules, has been and remains an invariable and trouble-free assistant to synthetic chemists.


On the other hand, using traditional mass spectrometers, it turned out to be impossible to analyze biological macromolecules, since biopolymers cannot be ionized and converted into a gaseous state by heating or irradiating electron beams. (Almost every one of us has experienced the truth of this statement from our own experience: scrambled eggs, forgotten on the stove, burn on the pan, and do not evaporate.)
The widespread use of mass spectrometry in the study of the structure and functions of proteins and peptides became possible only due to a technological breakthrough that occurred in the 1980s, when new ionization methods suitable for biomolecules were developed.
In 1983, the team of Yale University professor J. Fenn proposed a method of electrospray ionization, in which the formation of ions was achieved by spraying a sample solution while passing it through a capillary to which a high voltage was applied.
At the same time, the German team of F. Hilenkamp proposed a method for ionizing organic molecules by irradiating their dried samples with an ultraviolet laser. In this case, the molecules evaporated and ionized together with the solid volatile organic matter, the matrix. This was the birth of, perhaps, the most popular and currently in demand method of biological mass spectrometry - MALDI ("Matrix-Assisted Laser Desorption/Ionization").
The choice of the matrix substance and the selection of the optimal substance/matrix ratio determines the sensitivity of the method due to the most efficient transfer of the test substance into the gaseous phase.


Already three years after Hilenkamp's publication, K. Tanaka, an employee of the Japanese company Shimadzu, proposed a ready-made solution for mass spectrometry of proteins weighing up to 100 kDa - a time-of-flight MALDI mass spectrometer. In fairness, it should be noted that the matrices used in it were liquid: proteins were dissolved in glycerol, in which microparticles of metallic cobalt were suspended.
These works were recognized by the Nobel Committee in 2002: Fenn and Tanaka shared a prize in chemistry for the development of soft methods for the ionization of biomolecules in mass spectrometry.

In modern spectrometers, a common analyzer for MALDI is a time-of-flight mass spectrometer. Molecular ions move in an evacuated tube, and they reach the detector in order of increasing mass.
The efficiency of obtaining (registration) of ions in MALDI mass spectrometry depends on many parameters: the method of sample preparation and purification; the quantitative ratio of the matrix to the analyte and the correct choice of substrate material. Important conditions for obtaining a stable signal are the power of laser radiation, the choice of a "hot spot" on the sample, and the pressure in the ionization region.
For mass spectrometers with an electrospray ion source, ion traps are most often used as analyzers. Currently, the combination of a quadrupole with an electrospray ion source is one of the most commonly used tools in biochemistry and metabolomics.
Both mass spectroscopic methods have their advantages and disadvantages. First, they require high chemical purity of the analyte. ESI is a "softer" ionization method than MALDI. With ESI, a continuous flow of ions is formed; with MALDI, a very time-limited (up to 10 ns) packet of ions is formed; at the same time, more than 10 femtomoles of a substance are subject to ESI analysis, MALDI - 10 times less.
But all these remarks are purely technical and only indicate that the methods of mass spectrometry of biological molecules have become available today and are widely used in scientific experiments and clinical studies. In addition, the strengths of these methods complement each other well, serving as a truly powerful analytical tool.
...To date, the history of the development of mass spectrometry has a second hundred years. But only very recently, as John Fenn said in his Nobel lecture, "...molecular elephants were able to fly on ionic wings."

Medical proteomics is a new and promising field of biomedical research that allows to adapt the achievements of functional proteomics, genomics and bioinformatics literally directly “to life”, i.e. to use the available knowledge for the clinical analysis of biological samples taken from patients.

Modern achievements

Researchers from all over the world are intensively working on the search for proteomic markers of significant diseases - not only scientists from academic institutions, but also specialists from the research departments of large and medium-sized pharmaceutical companies.

Some success has already been achieved in one of the most problematic areas of medicine - the early diagnosis of serious diseases. This is especially true for prostate cancer (Downes et al., 2007; Larkin et al., 2010). Diagnosis of this widespread disease, based on the presence of prostate-specific antigen protein in the patient's urine, is one of the earliest and most accurate to date.

To date, good results have been achieved in the detection of markers of breast cancer (Mathelin et al., 2006; Gast et al., 2009). Due to the wide clinical variability of this disease, it is proposed to use a set of 40 proteins as markers. Such a protein profile allows not only diagnosing a disease with high accuracy, but also predicting the effectiveness of treatment. The main diagnostic proteins of this set - haptoglobin, transferrin, apolipoproteins A-I and C-I - are already used today in the diagnosis of breast cancer.

Research is also underway to detect markers of neurodegenerative diseases such as Alzheimer's disease, sclerosis of various etiologies, etc. (Cedazo-Minguez, Winblad, 2010). In this area, the main prognostic markers are angiogenin (an enzyme that ensures the growth of blood vessels), creatinine kinase, fibrinogen, apolipoprotein E (Bowser, Lacomis, 2009)

In Siberia, the problems of structural and functional proteomics are intensively studied at the Novosibirsk Institute of Chemical Biology and Fundamental Medicine of the Siberian Branch of the Russian Academy of Sciences. As a result of joint work with the Research Institute of Mental Health, TSC SB RAMS, a lot of work has been done to search for protein markers of schizophrenia.

The etiology and pathogenesis of this severe mental illness is unknown. According to one theory of the occurrence of schizophrenia, the disease is based on a violation of protein metabolism. Comparison of proteomic profiles of a statistically significant sample of people with schizophrenia and proteomic profiles of healthy volunteers has already allowed researchers to identify a specific set of proteins as markers: apolipoprotein A-II, phosphomevalonate kinase, and serine (threonine) kinase. Further efforts of scientists will be aimed at clarifying the role of these proteins in the pathogenesis of the disease and introducing these markers into clinical biochemistry.

The work of proteomics researchers has huge potential and prospects. Given that the number of different protein molecules and their variants in the human body can be in the millions, scientists are confident that their high-tech protein "fishing" will continue to bring a guaranteed rich catch in the future. The successes of recent years in this new biomedical field inspire reasonable optimism.

Literature

Cox J., Mann M. Is proteomics the new genomics? // cell. 2007. V. 130(3). P. 395-398.

Capelo J.L., Carreira R., Diniz M. et al. Overview on modern approaches to speed up protein identification workflows relying on enzymatic cleavage and mass spectrometry-based techniques // Anal. Chim. acta. 2009. V. 650. N 2. P. 151-159.

Ulrich-Merzenich G., Panek D., Zeitler H. et al. New perspectives for synergy research with the "omic"-technologies // Phytomedicine. 2009. N. 6-7. P. 495-508.

Feng X., Liu X., Luo Q., Liu B.F. Mass spectrometry in systems biology: an overview // Mass Spectrom. Rev. 2008. V. 6. P. 635-660.

Proteomics

A.A. ZAMYATNIN, Doctor of Biological Sciences, Institute of Biochemistry. A.N. Bach RAS

Our story will be devoted to one of the youngest fundamental sciences (if not the youngest), which was born just a few years ago along with those who are still in elementary school. Unlike many other sciences of proteomics, one can say exactly under what circumstances it arose, indicate the year when its name appeared and who invented it.

Let's start with the circumstances. In the second half of the XX century. the analytical methods of biochemistry, molecular biology and computer technology developed rapidly. The outstanding advances made in these areas have led to the possibility of deciphering huge sequences of nucleic acid bases and to recording the entire genome of a living organism. For the first time, the complete genome was deciphered in 1980 in the bacteriophage phi X-174 (about 5 103 bases), then in the first bacterium, Haemophilus influenzae (1.8 106 bases). And with the end of the XX century. The grandiose work on deciphering the complete human genome was completed - identifying the sequence of approximately 3 billion bases of nucleic acids. Several billion dollars were spent on this work (about one dollar per base). In total, the genomes of several dozen species of living organisms have already been deciphered. It was during this period that two new biological sciences arose: in 1987, for the first time in the scientific press, the word "genomics" was used, and in 1993 - "bioinformatics".

In each biological species, a part of the genome is represented by regions encoding the amino acid sequences of proteins. For example, there are about 100,000 such areas in humans (according to some estimates, this number can reach 300,000, and taking into account chemically modified structures, several million). It would seem that knowing the complete genome and genetic code, it is possible to obtain all information about the structure of proteins by translation. However, everything is not so simple. Gradually, it became apparent that in the given cellular system of the organism under consideration, there is no correlation between the sets of mRNA and proteins. In addition, many proteins synthesized on ribosomes in accordance with the nucleotide sequence undergo chemical modifications after synthesis and can exist in the body in modified and unmodified forms. And it is also important that proteins have a variety of spatial structures that today cannot be determined by linear sequences of nucleotides and even amino acids. Therefore, the direct isolation and determination of the structures of all functioning proteins is still an urgent task (to date, direct structure determination has been carried out only for approximately 10% of human proteins). Thus, in addition to genomics, the term "proteomics" appeared, the object of study of which is the proteome (from the English PROTEins - proteins and genOMe - genome). And in the scientific press, the mention of the proteome first appeared in 1995.

It should be added that numerous short fragments of protein precursors, which are called oligopeptides, or simply peptides, play an important role in the life of organisms. It is because of them that there is such a discrepancy in assessing the amount of protein-peptide components in representatives of one biological species. Therefore, along with the terms “proteome” and “proteomics”, such terms as “peptide” and “peptidomics” are already used, which are part of the proteome and proteomics. We have previously described the diversity of the structure and functions of proteins and peptides on the pages of the Biology newspaper.

So, let's formulate the definitions of the new sciences that appeared during the lifetime of the current young generation and which are closely interconnected with each other (Fig. 1).

Genomics is a science that studies the structure and functions of genes (the genome is the totality of all the genes of an organism).

Bioinformatics is a science that studies biological information using mathematical, statistical and computer methods.

Proteomics is a science that studies the totality of proteins and their interactions in living organisms (proteome is the totality of all proteins in an organism).

We also note that proteomics in general includes structural proteomics, functional proteomics, and applied proteomics, which we will consider separately.

Structural proteomics

The most striking feature of biology is diversity. It is visible at all levels of biological organization (biological species, morphology, chemical structure of molecules, network of regulatory processes, etc.). This fully applies to proteins. The scale of their structural diversity has not yet been fully elucidated. Suffice it to say that the number of amino acid residues in one protein can range from two (the minimum structure having a peptide bond) to tens of thousands, and the human titin protein contains 34,350 amino acid residues and today is the record holder - the largest of all known protein molecules.

To obtain information about the proteome, it must first be isolated and purified from other molecules. Since the number of proteins in the entire proteome (i.e., in the entire organism) is very large, usually only a part of the organism (its organ or tissue) is taken and the protein component is isolated by various methods. Over the almost 200-year history of the study of proteins, many methods have been developed for protein isolation - from simple salt precipitation to modern complex methods that take into account the various physical and chemical properties of these substances. After obtaining a pure fraction of an individual protein, its chemical structure is determined.

In structural proteomics, the structure of not one but many proteins is determined at once, and to date, a special cycle of procedures has been developed for this and an arsenal of appropriate high-precision instruments has been created. (A complete set of proteomic research equipment costs more than one million dollars.)

On fig. Figure 2 shows a diagram of the laboratory cycle from sample preparation to determination of its structure. After isolation and purification (the figure shows an already isolated and purified preparation), proteins are separated using two-dimensional electrophoresis. This separation goes in two directions: in one, protein molecules with different masses are separated, in the other, a different total electric charge. As a result of this delicate procedure, identical molecules are grouped on a special carrier, forming macroscopic spots, and each spot contains only identical molecules. The number of spots, i.e. the number of different proteins or peptides can be many thousands (Fig. 3, 4), and automatic devices for processing and analysis are used for their study. Then the spots are selected and the substances contained in them are introduced into the most complex physical device - a mass spectrometer, with the help of which the chemical (primary) structure of each protein is determined.

The primary structure of a protein can also be determined using the results of genomics and bioinformatics. On fig. 5 shows the complete structure of the human serum albumin gene. It contains 1830 nitrogenous bases encoding 610 amino acid residues. This gene, like the vast majority of others, begins with the atg codon encoding the methionine residue and ends with one of the stop codons, in this case taa. Thus, a structure consisting of 609 amino acid residues is encoded (Fig. 6). However, this structure is not yet a serum albumin molecule, but only its precursor. The first 24 amino acid residues are the so-called signal peptide, which is cleaved off during the transition of the molecule from the nucleus to the cytoplasm, and only after that the serum albumin structure is formed, obtained by isolating this protein. As a result, this molecule contains 385 amino acid residues.

However, the amino acid sequence does not reveal the spatial structure of the protein. From the point of view of thermodynamics, an elongated linear structure is energetically unfavorable, and therefore it folds in a specific way for each sequence into a unique spatial structure, which can be determined using two powerful physical methods - X-ray diffraction analysis and nuclear magnetic resonance (NMR spectroscopy). With the help of the first of them, the spatial structures of already several thousand proteins, including human serum albumin, have been determined, the image of which is shown in Fig. 7. This structure, in contrast to the primary (amino acid sequence), is called tertiary, and spiralized sections are clearly visible in it, which are elements of the secondary structure.

Thus, the task of structural proteomics is reduced to the isolation, purification, determination of the primary, secondary, and tertiary structures of all proteins of a living organism, and its main tools are two-dimensional electrophoresis, mass spectrometry, and bioinformatics.

Bioinformatics of proteins

The existence of a huge number of various proteins led to the need to create information arrays - databases (or banks) of data, in which all information known about them would be entered. Currently, there are many general and specialized databases that are available on the Internet to everyone. Common databases contain information about all known proteins of living organisms, i.e. about the global proteome of all living things. An example of such a database is SwissProt-TrEMBL (Switzerland-Germany), which today contains the structures of almost 200,000 proteins identified by analytical methods, and almost 2 million more structures that are determined as a result of translation from nucleotide sequences. On fig. 8 and 9 show the number of existing proteins that are known for each given number of amino acid residues. The abscissa axes in these plots are limited to 2000 residues, but, as mentioned above, although not often, there are also significantly larger molecules. From the data presented in the figures, it follows that the largest number of proteins contains several hundred amino acid residues. These include enzymes and other fairly mobile molecules. Among the larger proteins, there are many that perform supporting or protective functions, holding biological structures together and giving them strength.

In the global proteome, a special place is occupied by small, very mobile molecules containing no more than 50 amino acid residues and possessing a specific spectrum of functional activity. They are called oligopeptides, or simply peptides. For them, i.e. for the global peptidome, a special data bank has been created, which is called EROP-Moscow. This name is an abbreviation of the term Endogenous Regulatory OligoPeptides (endogenous regulatory oligopeptides), and indicates that the bank was created and based in the capital of our country. To date, the structure of almost 6,000 oligopeptides isolated from representatives of all living kingdoms has been deciphered. Like large proteins, the number of oligopeptides with a given number of amino acid residues can be represented graphically (Fig. 10). Judging by the graph, oligopeptides containing approximately 8-10 amino acid residues are most common. Among them, there are mainly molecules that are involved in the regulation of the nervous system, and therefore are called neuropeptides. Obviously, the fastest processes in a living organism are carried out with the participation of the nervous system; therefore, peptide regulators must be mobile and, therefore, small. However, it should be noted that, due to the huge structural and functional diversity of both proteins and peptides, a strict classification has not yet been created for them.

Thus, in this case, the tasks of bioinformatics are the accumulation of information about the physicochemical and biological properties of proteins, the analysis of this information, the cataloging and preparation of an information base and computing tools to identify the mechanisms of their functioning.

Functional proteomics

The presence of this or that protein in the body gives grounds to assume that it has (or had) a certain function, and the entire proteome serves to ensure the full functioning of the entire organism. Functional proteomics deals with the determination of the functional properties of the proteome, and the tasks it solves are much more complicated than, for example, the determination of protein-peptide structures.

Obviously, the functioning of the proteome is carried out in a multicomponent environment, in which there are many molecules of other chemical classes - sugars, lipids, prostaglandins, various ions, and many others, including water molecules. It is possible that after some time such terms as "sugar", "lipid" and the like will appear. Protein molecules interact with other or the same structures surrounding them, which ultimately leads to the emergence of functional reactions, first at the molecular level, and then at the macroscopic level. Many such processes are already known, including those involving proteins. Among them, the interaction of an enzyme with a substrate, an antigen with an antibody, peptides with receptors, toxins with ion channels, etc. (receptors and ion channels are also protein formations). To identify the mechanisms of these processes, both experimental studies of individual participants in the interaction and systematic studies by means of bioinformatics are carried out. Let us consider several examples of such systematic approaches.

On fig. 11 shows representatives of the human proteome (in this case, the peptidome) - various gastrins and cholecystokinins, which are localized in the gastrointestinal tract (when writing amino acid sequences, a standard one-letter code was used, the decoding of which was given by us earlier). The functional parts of the molecules of these peptides are very similar right-hand regions. However, peptides have opposite behavioral properties: gastrins make a person feel hungry, while cholecystokinins make people feel full. Apparently, this difference is due to the fact that in the primary sequence of cholecystokinins, the position of the tyrosine Y residue is shifted by one step compared to gastrins. The same figure shows the primary structure of the cyonine peptide obtained from the protozoan Ciona intestinalis (Fig. 12). Its structure is homologous to both gastrins and cholecystokinins and is characterized by two tyrosine residues located in the same positions as in both of these peptides. Unfortunately, its functional properties have not been studied. And with proper experimental research, it would be possible to answer the question of what is the role of the chemical structure in general and tyrosine residues in particular in the manifestation of opposite physiological effects.

Another example: in Fig. 13 shows the amino acid sequences of very similar molecules, which are also united in a structurally homologous family. These molecules have been found in very evolutionarily distant living organisms - from insects to mammals. The first line gives the primary structure of bradykinin, which contains 9 amino acid residues and is found in many higher organisms, including humans. Over the years, chemists have been synthesizing various non-natural analogues of this molecule in order to answer the question of which part of it is responsible for interacting with the receptor. About 30 years ago, all possible fragments of bradykinin were synthesized - 8 dipeptides, 7 tripeptides, etc. (a total of 36 fragments are possible), the activity value of which was then tested in the same biological test. The result turned out to be trivial: it turned out that only the entire molecule exhibits maximum activity, and each fragment individually has either trace activity or zero. This time-consuming work would not have had to be done if other bradykinins were known at that time, shown in Fig. 13, and by means of bioinformatics they would be isolated from the global proteome. The presented structural-homologous family clearly demonstrates that all molecules have a region that has remained practically unchanged as a result of biological evolution (a quasi-conservative region), and it is a bradykinin molecule of higher living organisms, selected as the most perfect as a result of the evolutionary process. This example demonstrates that proteomics, together with bioinformatics, allows you to quickly (and cheaply) solve fundamental scientific problems.

And, finally, the third example is the structurally homologous family of mammalian endothelins and snake toxins (Fig. 14). Despite the striking similarity of structures, their functional properties are strikingly different from each other: some are very useful regulators of vascular contraction, while others are deadly to life. In this case, we are faced with a situation where the primary structure does not carry sufficient information that can explain the reason for the difference in functions, and a more detailed consideration of the spatial (tertiary) structure is required. On fig. 15 and 16 show the spatial structures of two members of this family, endothelin-1 and sarafotoxin 6b, obtained using NMR spectroscopy. In the figures, they are rotated so as to achieve maximum spatial homology. But complete homology cannot be obtained under any rotation. Consequently, despite the great similarity of the primary structures, their interaction is carried out with different receptor structures, and therefore leads to different physiological effects.

Of course, it is impossible to fully characterize the diversity of functional proteomics with such particular examples. Creating ideas about the huge network of interactions of protein and other molecules in the body requires a lot of work and the use of all the means of modern bioinformatics. In fact, the creation of such representations is just beginning. However, there is reason to believe that every year our knowledge in this area will grow rapidly.

One of the first successes along this path is the creation of a map of the metabolism of carboxylic acids at the Institute of Biochemistry. A.N. Bach of the Russian Academy of Sciences (Fig. 17). This map is a network of reactions with a regular periodic structure. This approach proved to be successful due to the fact that functionally similar metabolites undergo similar biochemical transformations, forming functionally similar derivatives. In the map, vertically there are areas containing compounds with the same number of carbon atoms (from 1 to 10), and horizontal rows represent rows of functionally similar metabolites. The chemical structures on the map are connected by numerous arrows indicating which enzymes (proteins) are involved in the corresponding chemical transformations. Isn't it true that such an approach resembles the periodic table of chemical elements of D.I. Mendeleev? And just like the Mendeleev system, this map has predictive power. With its help, a number of new enzymes were predicted, which were subsequently discovered experimentally.

Similar schemes can be extended to other metabolic processes (for example, carbohydrates, amino acids, etc.), and also used to search for new metabolites of biochemical reactions.

Thus, functional proteomics deals with the study of complex relationships between the structure and functions of the proteome.

Practical proteomics

So, the main task of proteomics is to reveal the mechanism of interaction of a huge number of proteins and peptides in one organism. What is the practical significance of this grandiose and costly work? It is obvious that, first of all, pharmacologists and physicians are interested in the results of such work, since very often there is a close relationship between changes in the protein composition and the disease state of a person. Therefore, new data in proteomics will be used (and are already being used) for the rapid development of new drugs and novel treatments for diseases that medicine has struggled with for centuries. Today, 95% of all pharmacological agents act on proteins. Proteomics, with its systematic approach, can help to identify and assess the importance of the emergence of new proteins much more efficiently, which in turn will accelerate the development of new diagnostic tests and therapeutics.

The first practical application of proteomic research took place long before the appearance of the term “proteomics”, at the beginning of the 20th century, when the role of insulin in the development of such a serious disease as diabetes was discovered. The creation of insulin preparations has saved the lives of millions of people.

At present, proteomics, together with genomics and bioinformatics, is focused on the creation of new drugs (Fig. 18), in which certain proteins will serve as molecular targets. The process of finding new targets for the action of drugs is solved with the help of bioinformatics, and the object of analysis is the genome. However, after analyzing the genome, it is necessary to obtain evidence that this protein is intensely expressed and is in a working state in the cell. This problem is solved by proteomics. Thus, the molecular genetic target for the drug is identified.

It should be noted that proteomics alone can solve the problem of finding a target. If you get proteomic maps (similar to those shown in Fig. 3 or 4) of normal and pathological tissues, then by the differences in them you can determine which proteins are important for the development of a particular pathological condition, and select them as targets or use these knowledge for diagnosis. It can be assumed that in the future, the creation of proteomic blood maps will be added to the usual blood test. To do this, polyclinics will need to use special equipment with which patients will periodically take blood. In the event of a disease state, the proteomic map of a sick person will only need to be compared with his own proteomic map, but compiled at the time when he was healthy, and it will be possible to identify changes in the protein composition of the blood and determine the cause of the disease. Such a comparison of the proteomes of tumor and normal cells, cells before and after exposure to certain factors (for example, physical or chemical), the use of biological fluids for diagnostic purposes - all this is of great interest and opens up completely new perspectives for medicine, veterinary medicine, pharmacology, food industry and other application areas. There is a huge and interesting work ahead.

List literature

1. Sanger F., Air G.M., Barrell B.G., Brown N.L. et al. Nucliotide sequence of bacteriophage phi X-174 DNA.//Nature. 1977. V. 265, No. 5596. P. 687-695.

2. Fleischmann R.D., Adams M.D., White O. et al. Whole-genome random sequencing and assembly of Haemophilus influenzae Rd.//Science. 1995. V. 269, No. 5223. P. 496-512.

3. Nature. 2001. 409, no. 6822 (most of the issue of the journal is devoted to deciphering the human genome).

4. Ferguson-Smith A.C., Ruddle F.H. The genomics of human homeobox-containing loci.//Pathol. Immunopathol. Res. 1988. V. 7, No. 1-2. P. 119-126.

5. Franklin J. Bioinformatics changing the face of information.//Ann. NY Acad. sci. 1993. V. 700. P. 145-152.

6. Wasinger V.C., Cordwell S.J., Cerpa-Poljak A. et al. Progress with gene-product mapping of the Mollicutes: Mycoplasma genitalium.//Electrophoresis. 1995. V. 16, No. 7. P. 1090-1094.

7. Zamyatnin A.A. The shining world of proteins and peptides.//Biology. 2002. No. 25-26. P. 8-13.

8. Gorg A., Weiss W., Dunn M.J. Current two-dimensional electrophoresis technology for proteomics.//Proteomics. 2004. V. 4, No. 12. P. 3665-3685.

9. Ramstrom M., Bergquist J. Miniaturized proteomics and peptidomics using capillary liquid separation and high resolution mass spectrometry.//FEBS Lett. 2004. V. 567, No. 1. P. 92-95.

10. http://au.expasy.org/sprot/

11. http://erop.inbi.ras.ru/

12. Malygin A.G. Metabolism of carboxylic acids (periodic chart). - M .: "International Education Program", 1999.

13. Archakov A.I. What is genomics? - Proteomics.//Vopr. honey. chemistry. 2000. V. 46, No. 4. S. 335-343.

So, the main task of proteomics is to reveal the mechanism of interaction of a huge number of proteins and peptides in one organism. What is the practical significance of this grandiose and costly work? It is obvious that, first of all, pharmacologists and physicians are interested in the results of such work, since very often there is a close relationship between changes in the protein composition and the disease state of a person. Therefore, new data in proteomics will be used (and are already being used) for the rapid development of new drugs and novel treatments for diseases that medicine has struggled with for centuries. Today, 95% of all pharmacological agents act on proteins. Proteomics, with its systematic approach, can help to identify and assess the importance of the emergence of new proteins much more efficiently, which in turn will accelerate the development of new diagnostic tests and therapeutics.

The first practical application of proteomic research took place long before the appearance of the term “proteomics”, at the beginning of the 20th century, when the role of insulin in the development of such a serious disease as diabetes was discovered. The creation of insulin preparations has saved the lives of millions of people.

At present, proteomics, together with genomics and bioinformatics, is focused on the creation of new drugs (Fig. 18), in which certain proteins will serve as molecular targets. The process of finding new targets for the action of drugs is solved with the help of bioinformatics, and the object of analysis is the genome. However, after analyzing the genome, it is necessary to obtain evidence that this protein is intensely expressed and is in a working state in the cell. This problem is solved by proteomics. Thus, the molecular genetic target for the drug is identified.

It should be noted that proteomics alone can solve the problem of finding a target. If you get proteomic maps (similar to those shown in Fig. 3 or 4) of normal and pathological tissues, then by the differences in them you can determine which proteins are important for the development of a particular pathological condition, and select them as targets or use these knowledge for diagnosis. It can be assumed that in the future, the creation of proteomic blood maps will be added to the usual blood test. To do this, polyclinics will need to use special equipment with which patients will periodically take blood. In the event of a disease state, the proteomic map of a sick person will only need to be compared with his own proteomic map, but compiled at the time when he was healthy, and it will be possible to identify changes in the protein composition of the blood and determine the cause of the disease. Such a comparison of the proteomes of tumor and normal cells, cells before and after exposure to certain factors (for example, physical or chemical), the use of biological fluids for diagnostic purposes - all this is of great interest and opens up completely new perspectives for medicine, veterinary medicine, pharmacology, the food industry and other application areas. There is a huge and interesting work ahead.

Bibliography

1.Sanger F., Air G.M., Barrell B.G., Brown N.L. et al. Nucliotide sequence of bacteriophage phi X-174 DNA.//Nature. 1977. V. 265, No. 5596. P. 687–695.

2.Fleischmann R.D., Adams M.D., White O. et al. Whole-genome random sequencing and assembly of Haemophilus influenzae Rd.//Science. 1995. V. 269, No. 5223. P. 496–512.

3. Nature. 2001. 409, no. 6822 (most of the issue of the journal is devoted to deciphering the human genome).

4. Ferguson-Smith A.C., Ruddle F.H. The genomics of human homeobox-containing loci.//Pathol. Immunopathol. Res. 1988. V. 7, no. 1–2. P. 119–126.

5.Franklin J. Bioinformatics changing the face of information.//Ann. NY Acad. sci. 1993. V. 700. P. 145–152.

6. Wasinger V.C., Cordwell S.J., Cerpa-Poljak A. et al. Progress with gene-product mapping of the Mollicutes: Mycoplasma genitalium.//Electrophoresis. 1995. V. 16, No. 7. P. 1090–1094.

7. Zamyatnin A.A. The shining world of proteins and peptides.//Biology. 2002. No. 25–26. P. 8–13.

8.Gorg A., Weiss W., Dunn M.J. Current two-dimensional electrophoresis technology for proteomics.//Proteomics. 2004. V. 4, No. 12. P. 3665–3685.

9. Ramstrom M., Bergquist J. Miniaturized proteomics and peptidomics using capillary liquid separation and high resolution mass spectrometry.//FEBS Lett. 2004. V. 567, No. 1. P. 92–95.

10. http://au.expasy.org/sprot/

11 http://erop.inbi.ras.ru/

12. Malygin A.G. Metabolism of carboxylic acids (periodic chart). - M .: "International Education Program", 1999.

13. Archakov A.I. What is genomics? – Proteomics.//Vopr. honey. chemistry. 2000. V. 46, No. 4. S. 335–343.

14. en.wikipedia.org/wiki/Proteomics

15. www.biomed.spbu.ru/equipment/proteomics/

16. thesaurus.rusnano.com/wiki/article1579

17. www.inbi.ras.ru/ubkh/49/Terentiev.pdf

18. www.strf.ru/material.aspx?CatalogId=352&d_no=11979

19. www.textronica.com/lcline/proteomics/proteomics.html

20. www.bionet.nsc.ru/bioinf/files/proneomika.pdf

21. www.bionet.nsc.ru/bioinf/files/proneomika.pdf

Functional genomics is closely related and in fact overlaps with the latest direction in biology, called " proteomics"- the science of proteomes. The word "proteome" is formed from the word "protein" (protein) and the end of the word "genome", so that in the name itself, "protein" and "genome" (DNA) seem to be merged. This emphasizes their closest However, there is one fundamental difference between genomics and proteomics, between the genome and the proteome, which brings to life completely new methods of research, new strategies.

Proteome- the concept is dynamic, while the genome is stable and constant, otherwise it would be impossible to transfer hereditary properties from generation to generation, to ensure the conservation of species, etc. The variability of the genome always occurs against the background of its high stability and in no way cancels it. Proteome - a set of proteins of a given cell in a given phase of its development at a given time, i.e. less than the genome in terms of the total amount of information. In any cell of the human body, all approximately 80 thousand genes never function, only a part of them work - sometimes smaller, sometimes larger. Although it is still difficult to give exact figures, in an ordinary specialized cell, for example, in a liver or lung cell, probably no more than 10 thousand proteins are simultaneously present, and in sharply different quantities - from a few molecules per cell to several percent of the total cellular squirrel. The set of proteins is constantly changing depending on the phase of cell division, the tissue specialization of the cell, the stage of its differentiation, belonging to normal or malignant cells, stress or rest, exposure to extracellular physiologically active substances, and so on ad infinitum. Therefore, the protein "portrait" of the cell depends on many factors and influences, is subject to almost continuous changes, which makes its study especially difficult.

There is a "bouquet" of proteomic technologies; each has its own merits and demerits. Let's focus on the two most effective. A complex mixture of proteins extracted from a cell can be separated on a carrier (usually a polyacrylamide gel) in two directions: in one, the proteins will be divided by size (molecular weight), in the other, by electric charge (isoelectric point). The result is a two-dimensional map containing many hundreds of points, each corresponding to one or more proteins.

If a researcher is interested in a certain group of proteins, it can be isolated on a “map” and subjected to re-separation under slightly modified conditions with higher resolution. Databanks now store information on many different types of cells whose proteins have been subjected to electrophoretic separation in two directions. The computer is able to compare such two-dimensional "protein maps" and isolate what these types of cells have the same, and in what proteins they differ.

The method of "two-dimensional maps" is constantly being improved, and most of the individual protein dots that are visible on these "maps" have already been identified or are in the process of identification.

The most modern method of protein identification is that the protein under study is cleaved into fragments (peptides) with the help of one or another enzyme (protease). The resulting peptides are then separated, usually by high pressure chromatography, and then each of the individual peptides is placed in a mass spectrometer and its mass is determined. Comparison of the obtained results with those available in protein databases makes it possible to reliably identify a protein if its structure is known. For an unknown protein, this method helps to find "relatives" and, consequently, to formulate a preliminary idea of ​​its possible function.

The variability of the proteome is connected not only with the fact that one part of the genes works at a given moment of time, and another part at another moment. The set of proteins strongly depends on the processes occurring on the way from DNA to messenger RNA (mRNA). Here, most of the primary gene products (RNA) undergo the so-called "alternative splicing", the essence of which is that before the formation of mature messenger RNA, different parts of the molecule are removed from it. As a result, one gene can give rise to many proteins that differ in primary structure. Thus, it became obvious that one of the old dogmas of biochemistry and molecular biology - "One gene - one enzyme" - needs to be modernized. For very many cases, the formula is valid: "One gene - many proteins."

In this regard, it should be noted that after synthesis, proteins undergo many chemical changes (modifications), which create their great diversity, although initially they are encoded by one gene. These modifications include reactions of phosphorylation, acetylation, methylation, glycosylation, and many others. Considering that there are many places on a large protein where these modifications can occur, it is easy to imagine what an almost infinite variety of forms of the same protein molecule can arise. The vast majority of modifications significantly affect the biological activity of a given protein molecule, as well as its ability to interact with other protein molecules. As a result, we come to the conclusion that when, say, 10% of all genes work in a cell - let's say 8 thousand - then the number of different proteins can exceed this value by 10 times. Researchers have previously guessed that such a situation is possible, however, only now they really represent its true scale.

An extremely important branch of proteomics, of course, should be considered the study of protein-protein and protein-nucleic acid interactions. During the life of a cell, almost every protein in its functioning interacts with a variety of macromolecules, as well as low molecular weight ligands.

In recent years, the method of the so-called "yeast double hybrids" has become widely used to study protein-protein interactions. With the help of genetic engineering, a construct is created that consists of a DNA region that interacts with a transcription factor and a DNA region that encodes a “reporter gene”, which in turn encodes an enzyme protein whose activity is easy to measure. The transcription factor consists of two dominants and works only when the dominants interact with each other. If you need to find out whether two proteins under study interact with each other, you need to isolate the transcription factor and attach the protein of interest to each of the dominants. When they interact, the transcription factor will restore its activity, which will allow the “reporter gene” to work, and then you will find the activity of the “reporter protein”. If the studied proteins do not interact, the protein-enzyme is not formed.

The application of the two-hybrid system to the proteins of humans and other organisms made it possible to prove that there are a huge number of protein-protein contacts of various types and, in addition, to discover many previously unknown protein-protein interactions. This information is extremely important for identifying the components of the signaling pathways in the cell. As a rule, “intermediary proteins” are involved in the transmission of signals from the cell surface to the nucleus, which are often present in the cell in negligible concentrations, so the analysis of signaling pathways for experimenters is very difficult. The discovery of protein-protein interactions dramatically changed the situation.

In the analysis of protein-nucleic acid interactions, the methods of "chemical cross-linking" of these components are widely used (for example, the staff of Academician A.A. Bogdanov revealed many important interactions within ribosomal particles, where protein biosynthesis occurs in the cell).

Another convenient method is the change in electrophoretic mobility during complexation, by which a variety of DNA-protein and RNA-protein contacts have been analyzed. The original version of this method in combination with "chemical crosslinking" was developed by Academician A.D. Mirzabekov and used to reveal the structure of the nucleosome - an elementary structural unit consisting of DNA and histone proteins, from which all chromosomes are built.