Membrane proteins. Formation and release of cell transmembrane proteins The end product of the work of transmembrane proteins

Lipids in membranes are primarily responsible for their structural properties - they create a bilayer, or matrix, in which the active components of the membrane - proteins - are located. It is proteins that give various membranes their uniqueness and provide specific properties. Numerous membrane proteins perform the following main functions: they determine the transfer of substances across membranes (transport functions), carry out catalysis, provide the processes of photo- and oxidative phosphorylation, DNA replication, translation and modification of proteins, signal reception and transmission of nerve impulses, etc.

It is customary to divide membrane proteins into 2 groups: integral(internal) and peripheral(external). The criterion for such separation is the degree of strength of binding of the protein to the membrane and, accordingly, the degree of severity of processing required to extract the protein from the membrane. Thus, peripheral proteins can be released into solution even when membranes are washed with buffer mixtures with low ionic strength, low pH values ​​in the presence of chelating substances, for example ethylenediaminetetraacetate (EDTA), which bind divalent cations. Peripheral proteins are released from membranes under such mild conditions because they are associated with lipid heads or with other membrane proteins using weak electrostatic interactions, or with hydrophobic interactions with lipid tails. On the contrary, integral proteins are amphiphilic molecules, have large hydrophobic regions on their surface and are located inside the membrane, so their extraction requires destruction of the bilayer. For these purposes, detergents or organic solvents are most often used. The methods for attaching proteins to the membrane are quite varied (Fig. 4.8).

Transport proteins. The lipid bilayer is an impermeable barrier to most water-soluble molecules and ions, and their transport across biomembranes depends on the activity of transport proteins. There are two main types of these proteins: channels(pores) and carriers. Channels are membrane-crossing tunnels in which binding sites for transported substances are accessible on both membrane surfaces simultaneously. Channels do not undergo any conformational changes during the transport of substances; their conformation changes only when opening and closing. Carriers, on the contrary, change their conformation during the transfer of substances across the membrane. Moreover, at any given time, the binding site of the transported substance in the carrier is accessible only on one surface of the membrane.

Channels, in turn, can be divided into two main groups: voltage-dependent and chemically regulated. An example of a potential-dependent channel is the Na + channel; its operation is regulated by changing the electric field voltage. In other words, these channels open and close in response to change transmembrane potential. Chemically regulated channels

open and close in response to the binding of specific chemical agents. For example, the nicotinic acetylcholine receptor, when a neurotransmitter binds to it, goes into an open conformation and allows monovalent cations to pass through (subsection 4.7 of this chapter). The terms “pore” and “channel” are usually used interchangeably, but pores are more often understood as non-selective structures that distinguish substances mainly by size and allow passage of all sufficiently small molecules. Channels are often understood as ion channels. The transport rate through the open channel reaches 10 6 - 10 8 ions per second.

Transporters can also be divided into 2 groups: passive and active. With the help of passive carriers, one type of substance is transported across the membrane. Passive transporters are involved in facilitated diffusion and only increase the flow of substances along an electrochemical gradient (for example, the transfer of glucose across erythrocyte membranes). Active carriers transport substances across the membrane using energy. These transport proteins accumulate substances on one side of the membrane, transporting them against the electrochemical gradient. The speed of transport using carriers depends very much on their type and ranges from 30 to 10 5 s -1. The terms “permease” and “translocase” are often used to designate individual transporters, which can be considered synonymous with the term “transporter”.

Enzyme functions of membrane proteins. A large number of different enzymes function in cell membranes. Some of them are localized in the membrane, finding there a suitable environment for the transformation of hydrophobic compounds, others, thanks to the participation of membranes, are located in them in strict order, catalyzing successive stages of vital processes, while others require the assistance of lipids to stabilize their conformation and maintain activity. Enzymes were found in biomembranes - representatives of all known classes. They can penetrate the membrane through, be present in it in dissolved form, or, being peripheral proteins, bind to membrane surfaces in response to any signal. The following characteristic types of membrane enzymes can be distinguished:

1) transmembrane enzymes that catalyze coupled reactions on opposite sides of the membrane. These enzymes usually have several active centers located on opposite sides of the membrane. Typical representatives of such enzymes are components of the respiratory chain or photosynthetic redox centers that catalyze redox processes associated with electron transport and the creation of ion gradients on the membrane;

2) transmembrane enzymes involved in the transport of substances. Transport proteins that couple substance transfer with ATP hydrolysis, for example, have a catalytic function;

3) enzymes that catalyze the transformation of membrane-bound substrates. These enzymes are involved in the metabolism of membrane components: phospholipids, glycolipids, steroids, etc.

4) enzymes involved in the transformation of water-soluble substrates. With the help of membranes, most often in an attached state, enzymes can concentrate in those areas of the membrane where the content of their substrates is greatest. For example, enzymes that hydrolyze proteins and starch are attached to the membranes of intestinal microvilli, which helps to increase the rate of breakdown of these substrates.

Cytoskeletal proteins . The cytoskeleton is a complex network of protein fibers of various types and is present only in eukaryotic cells. The cytoskeleton provides mechanical support for the plasma membrane and can determine the shape of the cell, as well as the location of organelles and their movement during mitosis. With the participation of the cytoskeleton, such important processes for the cell as endo- and exocytosis, phagocytosis, and amoeboid movement are also carried out. Thus, the cytoskeleton is the dynamic framework of the cell and determines its mechanics.

The cytoskeleton is formed from three types of fibers:

1) microfilaments(diameter ~6 nm). They are thread-like organelles - polymers of the globular protein actin and other proteins associated with it;

2) intermediate filaments (diameter 8-10 nm). Formed by keratins and related proteins;

3) microtubules(diameter ~ 23 nm) - long tubular structures.

They consist of a globular protein called tubulin, the subunits of which form a hollow cylinder. The length of microtubules can reach several micrometers in the cytoplasm of cells and several millimeters in the axons of nerves.

The listed cytoskeletal structures penetrate the cell in different directions and are closely associated with the membrane, attaching to it at some points. These sections of the membrane play an important role in intercellular contacts; with their help, cells can attach to the substrate. They also play an important role in the transmembrane distribution of lipids and proteins in membranes.

Biological membranes, located at the border of the cell and the extracellular space, as well as at the border of membrane organelles of the cell (mitochondria, endoplasmic reticulum, Golgi complex, lysosomes, peroxisomes, nucleus, membrane vesicles) and the cytosol are essential for the functioning of both the cell as a whole and its organelles. Cell membranes have a fundamentally similar molecular organization. In this chapter, biological membranes are discussed primarily using the example of the plasma membrane (plasmolemma), which separates the cell from the extracellular environment.

Any biological membrane(Fig. 2–1) consists of phospholipids(~50%) and proteins (up to 40%). In smaller quantities, the membrane contains other lipids, cholesterol and carbohydrates.

Rice. 2–1. consists of a double layer phospholipids, the hydrophilic parts of which (heads) are directed towards the surface of the membrane, and the hydrophobic parts (tails that stabilize the membrane in the form of a bilayer) into the membrane. I - integral proteins immersed in a membrane. T - transmembrane proteins penetrate the entire thickness of the membrane. P - peripheral proteins located either on the outer or inner surface of the membrane.

Phospholipids. A phospholipid molecule consists of a polar (hydrophilic) part (head) and an apolar (hydrophobic) double hydrocarbon tail. In the aqueous phase, phospholipid molecules automatically aggregate tail to tail, forming the framework of the biological membrane (Figs. 2-1 and 2-2) in the form of a double layer (bilayer). Thus, in the membrane, the tails of phospholipids (fatty acids) are directed into the bilayer, and the heads containing phosphate groups are directed outward.

Arachidonic acid. Arachidonic acid is released from membrane phospholipids - a precursor of Pg, thromboxanes, leukotrienes and a number of other biologically active substances with many functions (inflammatory mediators, vasoactive factors, second messengers, etc.).

Liposomes- membrane vesicles artificially prepared from phospholipids with a diameter of 25 nm to 1 μm. Liposomes used as models of biological membranes, as well as for introducing various substances (for example, genes, drugs) into cells; the latter circumstance is based on the fact that membrane structures (including liposomes) easily merge (due to the phospholipid bilayer).

Squirrels biological membranes are divided into integral (including transmembrane) and peripheral (Fig. 2-1 and 2-2).

Integral membrane proteins (globular) embedded in the lipid bilayer. Their hydrophilic amino acids interact with the phosphate groups of phospholipids, and their hydrophobic amino acids interact with fatty acid chains. Integral membrane proteins include adhesion proteins and some receptor proteins (membrane receptors).

Transmembrane protein - a protein molecule that passes through the entire thickness of the membrane and protrudes from it on both the outer and inner surfaces. Transmembrane proteins include pores, ion channels, transporters, pumps, and some receptor proteins.

Pores and channels- transmembrane pathways along which water, ions and metabolite molecules move between the cytosol and the intercellular space (and in the opposite direction).

Vectors carry out transmembrane movement of specific molecules (including in combination with the transfer of ions or molecules of another type).

Pumps move ions against their concentration and energy gradients (electrochemical gradient) using the energy released by ATP hydrolysis.

Peripheral membrane proteins (fibrillar and globular) are located on one of the surfaces of the cell membrane (external or internal) and are non-covalently associated with integral membrane proteins.

Examples of peripheral membrane proteins associated with the outer surface of the membrane are - receptor proteins And adhesion proteins.

Examples of peripheral membrane proteins associated with the inner surface of the membrane are - cytoskeleton proteins, second messenger system proteins, enzymes and other proteins.

Lateral mobility. Integral proteins can be redistributed in the membrane as a result of interaction with peripheral proteins, cytoskeletal elements, molecules in the membrane of an adjacent cell, and components of the extracellular matrix.

Carbohydrates(mainly oligosaccharides) are part of the glycoproteins and glycolipids of the membrane, accounting for 2–10% of its mass (Fig. 2–2). Lectins interact with cell surface carbohydrates. Oligosaccharide chains protrude on the outer surface of cell membranes and form the surface shell - glycocalyx.

Glycocalyx has a thickness of about 50 nm and consists of oligosaccharides covalently associated with glycoproteins and glycolipids of the plasmalemma. Functions of the glycocalyx: intercellular recognition, intercellular interactions, parietal digestion (the glycocalyx covering the microvilli of the border cells of the intestinal epithelium contains peptidases and glycosidases that complete the breakdown of proteins and carbohydrates).

Membrane permeability

The membrane bilayer separates the two aqueous phases. Thus, the plasma membrane separates the intercellular (interstitial) fluid from the cytosol, and the membranes of lysosomes, peroxisomes, mitochondria and other membranous intracellular organelles separate their contents from the cytosol. Biological membrane - semi-permeable barrier.

Semi-permeable membrane. A biological membrane is defined as semi-permeable, i.e. a barrier that is not permeable to water, but permeable to substances dissolved in it (ions and molecules).

Semi-permeable tissue structures. Semi-permeable tissue structures also include the wall of blood capillaries and various barriers (for example, the filtration barrier of the renal corpuscles, the aerohematic barrier of the respiratory part of the lung, the blood-brain barrier and many others, although such barriers - in addition to biological membranes (plasmolemma) - also include non-membrane components. The permeability of such tissue structures is discussed in section "Transcellular Permeability" Chapter 4 .

The physicochemical parameters of the intercellular fluid and cytosol are significantly different (see Table 2-1), and the parameters of each membrane intracellular organelle and cytosol are also different. The outer and inner surfaces of a biological membrane are polar and hydrophilic, but the non-polar core of the membrane is hydrophobic. Therefore, nonpolar substances can penetrate the lipid bilayer. At the same time, it is the hydrophobic nature of the core of a biological membrane that determines the fundamental impossibility of direct penetration of polar substances through the membrane.

Non-polar substances(for example, water-insoluble cholesterol and its derivatives) freely penetrate biological membranes. In particular, it is for this reason that steroid hormone receptors are located inside the cell.

Polar substances(for example, Na+, K+ C1-, Ca2+ ions; various small but polar metabolites, as well as sugars, nucleotides, protein and nucleic acid macromolecules) do not penetrate biological membranes by themselves. That is why receptors for polar molecules (for example, peptide hormones) are built into the plasma membrane, and second messengers carry out the transmission of the hormonal signal to other cellular compartments.

Selective permeability- permeability of a biological membrane to specific chemicals) is important for maintaining cellular homeostasis. optimal content of ions, water, metabolites and macromolecules in the cell. The movement of specific substances across a biological membrane is called transmembrane transport (transmembrane transport).

Principles of structural organization of membrane proteins and methods for its prediction for transmembrane proteins

With high resolution, it was possible to establish the structure of only one class of membrane proteins - the reaction center of bacteria, however, even in this case, the position of the protein relative to the lipid bilayer is not unambiguously determined. The principles of its organization should be extended to other membrane proteins with caution. Some clarity can be brought by the use of thermodynamic principles, as well as taking into account the fact that the bulk of experimental data is consistent with the assumption of a high content of α-helices in membrane proteins. Thermodynamic factors impose certain restrictions on what type of protein-lipid structures can be stable.

Membrane proteins are amphiphilic compounds

Any membrane proteins in direct contact with the hydrophobic core of the lipid bilayer must be amphiphilic. Those regions of the polypeptide that are exposed to the solvent are most likely enriched in polar and ionizable amino acid residues, while the residues in contact with the lylide hydrocarbon chains should be mainly nonpolar. All this logically follows from the energy principles discussed in section. 2.3.1. Charged or polar amino acids can generally be present within the bilayer, but this is subject to certain restrictions.

Let us consider three levels of amphiphilic structures in membrane proteins: primary, secondary and tertiary amphiphilicity.

1. Primary amphiphilic structures contain an extended region of predominantly nonpolar amino acid residues, the length of which is sufficient to cross the bilayer. Such structures have been identified both in the reaction center and in bacteriorhodopsin. In these proteins, all membrane-spanning elements are α-helical. The a-helical structure is preferred because it forms all the hydrogen bonds that can involve hydrogen atoms of the polypeptide backbone. The alternative structure, which lacks one of the hydrogen bonds, is less stable by about 5 kcal/mol. All this allows us to suggest that rotation of the polypeptide chain inside the membrane is unlikely. At the turn sites, three to five amino acid residues would fail to form hydrogen bonds, and this would destabilize the structure by about 15-20 kcal/mol. In globular, water-soluble proteins, the turning regions are located predominantly on the surface of the protein globule, where amide groups can form hydrogen bonds with water; Apparently, in membrane protein molecules, rotations will also occur only in areas exposed to water.

It is possible that the 3-layer can also form transmembrane elements having, for example, the shape of J-cylinders, as in the case of porin. The requirements for the formation of hydrogen bonds by hydrogen atoms of the polypeptide backbone in such structures can be satisfied, but only under the condition of interaction between individual β-chains. How such a structure can be incorporated into a membrane is not entirely clear, and the restrictions imposed by the assembly mechanisms of membrane proteins are generally unknown.

2. Secondary amphiphilic structures. In such structures, hydrophobic residues occur periodically along the chain, and when the polypeptide is folded into a certain secondary structure, they form a continuous surface. The periodicity of some elements of the secondary structure is indicated in Table. 1. An example of proteins in which amphiphilic secondary structures appear to play an important role is porins. In them, polar and nonpolar amino acid residues in each of the /3 chains alternate. All polar residues are located on one side of the folded layer, lining the water-filled pore. Note that everything said about porin is hypothetical.

Table 1. Secondary structure parameters

The α-helix, in which hydrophobic residues occur every second or third monomer unit, must have hydrophobic and polar surfaces. Such structures are often represented as a helical ring with side chains indicated, as shown in Fig. Secondary amphiphilic structures can arise in situations schematically shown in Fig. EVIL.

A. Surface active protein segments; one side of the helix interacts with the hydrophobic region of the lipid bilayer, and the other is in contact with the aqueous phase and polar region of the bilayer. Amphiphilic α-helices are capable of forming many peptide hormones, as well as membrane-disrupting peptides, such as mellitin.

b. Transmembrane elements; the nonpolar surface of the helix faces the lipid phase, and the polar surface lines the water channel penetrating the bilayer. This is a very common model, based mainly on the results of studies of the nicotinic acetylcholine receptor, which functions as a chemically excitable channel. However, the conclusions based on experimental data that it is the amphiphilic helix that penetrates the membrane have raised objections. An amphiphilic 3-chain can also form such a water-filled channel, as in a porin.

V. Transmembrane elements; the nonpolar part of the surface is in contact with lipids, and the polar groups are in contact with the polar groups of other transmembrane elements. It is this principle that underlies the “inverted” structures, which presumably is bacteriorhodopsin. Polar interactions between amphiphilic helices could in principle stabilize interactions between subunits in oligomeric proteins.

3. Tertiary amphiphilic structures. Their existence can only be discussed speculatively. Their hydrophobic surface should be formed at the level of the tertiary structure of residues located in various parts of the polypeptide chain. Similar structures may be characteristic of proteins that bind to the bilayer but do not have clearly defined hydrophobic domains, defined by any of the above criteria. A possible example of this type is α-lactalbumin.

Ionizable amino acid residues in transmembrane segments

Many models of membrane proteins assume that their transmembrane segments contain ionizable residues. These residues likely play important functional and/or structural roles. In some cases, this role is clearly established: 1) lysine residues in bacteriorhodopsin and rhodopsin form Schiff bases with the prosthetic group of retinal, which is necessary for light excitation of the molecule; 2) histidine residues in the polypeptides of the bacterial reaction center are involved in binding to photosynthetic pigments; 3) charged residues in lactose permease from E. coli are involved in the implementation of transport functions by this protein; perhaps these residues form a network of hydrogen bonds within the protein molecule.

The transfer of charged groups from water to the low dielectric constant environment inside the membrane is energetically very unfavorable, and these groups must be stabilized in some way. It has been repeatedly suggested that the formation of ion pairs is sufficient for stabilization, and this principle was used in constructing a three-dimensional model of bacteriorhodopsin. However, calculations have shown that the free energy of transfer of an ion pair from water to a medium with low dielectric constant is also very high. For further stabilization, additional polar interactions are required, possibly involving other polar groups or via hydrogen bonds.

In principle, even a single charged group within a membrane can be stabilized through interactions with polar groups and with the participation of hydrogen bonds, which effectively delocalize the charge. There are several examples of isolated, desolvated ions stabilized by interactions in water-soluble proteins. Similar principles apparently operate in the case of charged residues of transmembrane segments of integral proteins.

However, it seems more likely that ionizable amino acids are neutralized within the membrane by protonation or deprotonation. The free energy of neutralization of charged amino acids is estimated to be approximately 10–17 kcal/mol. In the absence of specific conditions for polar interactions to stabilize the charged residue in the transmembrane segment, it is likely to be neutralized.

Charged amino acids in segments exposed to an aqueous environment

As we have already said, charged residues are distributed asymmetrically between the two sides of the bacterial reaction center. This asymmetry is also characteristic of some other internal membrane proteins of bacteria. Thus, the basic Lys and Arg residues are four times more common in those regions connecting transmembrane elements that are located on the inner side of the membrane rather than on the outer side. For acidic residues Asp and Glu, a similar trend is not detected. It is possible that this asymmetry is related to the mechanism of membrane protein assembly, but how exactly is unclear. Moreover, it is unknown whether this observation can be generalized or whether it has any predictive value.

In globular, water-soluble proteins, proline residues are rarely found in the middle part of the α-helix. Based on studies of 58 proteins containing 331 α-helices, 30 such cases were identified. In half of them, proline was located in places where the helix was damaged, and in the remaining cases it was in the area of ​​curvature or irregularity of the structure.

At the same time, in bacteriorhodopsin, proline residues are located in the middle part of three of the seven transmembrane helices, and in rhodopsin - in five of seven such spirals. A similar trend was revealed for other transmembrane segments of integral proteins, especially transport ones. The significance of this phenomenon is unknown. It should be noted, however, that due to the presence of a cyclic side chain, proline does not form hydrogen bonds with residues located on the previous turn of the α-helix. This may promote the formation of structures in which a hydrogen bond is formed through a specific interaction with a residue located in another membrane-spanning region. Such polar interactions within the bilayer could stabilize the three-dimensional structure of membrane proteins.

Methods for identifying primary amphiphilic structures

Unambiguous structural information about membrane proteins has been obtained in only a few cases, but researchers have at their disposal extensive amino acid sequence data based on DNA sequencing results. To identify transmembrane α-helices, presumably 20 residues long and consisting predominantly of hydrophobic amino acids, several amino acid sequence analysis methods have been developed. Each of them is based on the arrangement of amino acids in a row in accordance with a certain parameter, which reflects the probability of finding this residue in the transmembrane segment.

There are two types of scales. In one case, amino acids are classified by their relative polarity or "hydrophobicity". These scales are thermodynamic in nature and are based on the magnitude of the free energy change when an amino acid is transferred from an aqueous solution to a hydrocarbon medium. However, the number of methods for quantitatively assessing the hydrophobicity of amino acids is very large, and they do not always agree with each other. Data related to more than one physical characteristic at a time are often used. An example of this is the Kite and Doolittle "hydropathy" scale, which is based on hydrophobicity, measured by hydration potential, and the likelihood of residues being found within the globule.

The Goldman, Engelman and Steitz scale is based on a quantitative assessment of the free energy of transfer of α-helices from an aqueous environment into the membrane. In Fig. The Kite-Doolittle scale is compared with the Goldman-Engelman-Steitz scale.

Engelman et al. estimate that the free energy change upon insertion of a 20-residue polyanionic α-helix into the membrane is 30 kcal/mol. The calculation is based on estimating the surface area of ​​the helix exposed to the solvent. The contribution to the energy of each side group was estimated taking into account the surface area exposed to the aqueous environment inside the helix. The free energy of transfer of polar groups into the bilayer was also taken into account. For example, it was assumed that glutamine would be protonated when transferred to the bilayer and the free energy of this process would be 10.8 kcal/mol. Likewise, the transfer of hydroxyls will "cost" approximately 4.0 kcal/mol.

All of the above shows how the gain in interaction energy when transferring an α-helix into the bilayer can be used to “pull” polar side groups into the bilayer. For example, one arginine residue can be incorporated into the bilayer as part of a nonpolar transmembrane helix if it is deprotonated; this requires 16.7 kcal/mol at pH 7.0. The total free energy of α-helix transfer will still remain negative. However, the situation will change if two arginine residues need to be inserted into the bilayer or if the arginine is positively charged. Of course, polar residues can be stabilized within the bilayer due to specific interactions, but it is very difficult to actually take this into account in calculations. For example, serine, cysteine, and threonine side groups can form hydrogen bonds with the polypeptide backbone, and acidic and basic residues can form ion pairs; the appearance of such pairs is possible if these residues are located four or five monomer units from each other.

The second type of scale used to classify amino acids is based on the frequency with which amino acids actually occur in membrane-spanning segments.

In this case, hydrophobicity is empirically taken into account, as well as many other factors that cannot be quantified as hydrophobicity. The disadvantage of this semi-empirical approach is the lack of precise data on the boundaries of the transmembrane regions. However, such scales can be as useful as scales based on thermodynamic parameters. As an example, one can cite the scale of “propensity” to the membrane of Kuhn and Leigh or the scale of “immersion of the helix into the membrane” of Rao and Argos. The four most hydrophobic residues on the Goldman-Engelman-Steitz scale are also the four residues with the highest parameter value on the Rao and Argos scale.

In Fig. profiles of three different membrane proteins obtained using different scales are presented. When constructing these profiles, the average values ​​of the numbers on the scales assigned to each amino acid within the selected “window” are taken into account; this average is plotted relative to the residue number in the polypeptide. For example, if the window is 19 residues, the value assigned to position 40 will be the average number on the scale for all amino acids from 31 to 49 inclusive. The value assigned to position 41 will be the average of residues 32 to 50, etc. The peaks in the profile correspond to hydrophobic regions or those regions that are more likely to form transmembrane helices. To build a profile, the size of the window is important; most of the curves in Fig. were built with a window size of 19 residues.

Let's try to interpret the constructed profiles. According to the Goldman-Engelman-Steitz scale, peaks with values ​​close to zero correspond to transmembrane helices. A value of 1.25 on the Kite-Doolittle scale is the smallest value corresponding to a known transmembrane helix in the L subunit of the reaction center R . viridis . In all three cases presented in Fig. 3.12, the profiles for the subunits of the reaction center are similar.

In Fig. two profiles for cytochrome P450 from microsomes are shown. This protein was chosen because data on its primary structure suggest the presence of eight transmembrane helices. However, available experimental data indicate the existence of only one N- koh- ring anchor in the membrane. Both the Kite-Doolittle profile and the Goldman-Engelman-Steitz profile identify the N-terminal region, but they also indicate the presence of one or more additional transmembrane segments, which is not true. Note that many of the constructed models of membrane proteins, which are based only on amino acid sequence data, may be incorrect.

In Fig. Three profiles for bacteriorhodopsin are given. Despite their similarity, differences are visible in the shape of the peaks corresponding to the seven transmembrane segments. The Goldman-Engelm-on-Steitz algorithm does not take into account the stabilizing effect associated with the formation of an ion pair from closely spaced charged residues within the same helix. Taking this factor into account, the division between the last two spirals becomes clearer.

One of the problems encountered in the application of all the algorithms described above is to exclude hydrophobic segments in known globular proteins that are not transmembrane, but are located within the protein. However, when we search for fairly extended areas, this problem does not arise.

Note that algorithms used to detect α-helical structures in soluble globular proteins, such as the Chow-Fasman algorithm, are not suitable for detecting transmembrane elements. These algorithms are not applicable to describe the structure of non-globular regions, such as segments located inside the bilayer.

Algorithms designed to identify transmembrane regions cannot be used in the case of segments that are secondary amphiphilic structures or cross the membrane in the form of a /3-layer. In the first case, this region is excluded from consideration due to the presence of polar residues in it, and in the second, the transmembrane segment turns out to be too short, since only 10-12 amino acid residues in the /3 structure are needed to cross the bilayer. Some algorithms were designed to detect β-turns rather than the transmembrane elements themselves. Although this avoids some of the problems associated with distinguishing different classes of transmembrane elements, it is unclear how acceptable they will be if they are used more widely.

Methods for identifying secondary amphiphilic structures

Several approaches have been developed to detect secondary amphiphilicity or asymmetry in the distribution of hydrophobic residues in segments of a polypeptide chain. Quite often, α-helices and β-sheets in globular proteins are characterized by periodicity in the distribution of hydrophobic residues. The use of the spiral ring as a qualitative indicator is not always justified; more quantitative approaches are needed. The main one is the determination of periodicity in the distribution of hydrophobic residues using Fourier transform methods. An example is the hydrophobic moment.

1. Hydrophobic moment. This parameter was proposed by Eisenberg et al. It is defined as

and represents a certain vector sum of the hydrophobicity of residues in a segment of N elements. The hydrophobicity of each residue is represented as a vector, which is characterized by the angle , formed by the side chain and the axis of the polypeptide backbone. For a-helix 6 = 100°. In Fig. 3.9, B The hydrophobicity “vectors” are presented in projection onto the plane of the helical ring, and the hydrophobic moment is equal to their vector sum. The hydrophilic residue is represented by a vector with a negative directionality. For a random sequence the value qi due to the random distribution, there will be very few hydrophobic residues. At the same time, in the mellitin peptide, hydrophobic residues are located on one side of the structure, and polar ones on the other. The numerical value of the hydrophobic moment is assigned to the amino acid located in the center of the analyzed segment. Therefore, you can “scan” the sequence and assign an average hydrophobicity to each position, and also find ^n-

Eisenberg et al analyzed 11-residue segments from many proteins and peptides, determining the hydrophobic moment

and average hydrophobicity for each of the studied segments. Polypeptide segments of globular proteins are characterized by low values ​​of both qi - Transmembrane elements of a hydrophobic nature have high pH values, but low pH values, being mostly non-polar. Peptides and protein regions classified as “surfactant” have high values tsn due to the strong asymmetry in the distribution of polar and nonpolar residues. Using this algorithm, some segments of surfactant proteins were identified, for example, regions of diphtheria toxin and pyruvate oxidase from E. coli

The hydrophobic moment serves as a quantitative measure of the periodicity in the distribution of hydrophobic residues in different regions of the polypeptide. An important role is played by choice 6. The hydrophobic moment is essentially one of the parameters of the Fourier transform of the hydrophobicity function. The more general methods described below allow all Fourier components to be analyzed and any possible periodicity to be identified.

2. Periodicity of the sequence. Many methods have been developed for identifying regions of protein molecules that are characterized by periodic changes in hydrophobicity along the chain. All of them involve a Fourier transform function that depends on the hydrophobicity of amino acid residues along the polypeptide. The presence of a peak with a period of 3.6 indicates that a hydrophobic residue in this segment of the analyzed polypeptide occurs on average every 3.6 residues. This means that the segment is an α-helix, on one side of which there are predominantly hydrophobic residues. This method has been used to identify amphiphilic regions in several transport and channel-forming proteins; Examples include acetylcholine receptor, sodium channel, glucose transporter, mitochondrial uncoupler protein, and erythrocyte band 3 protein, an anion transporter. However, there is no clear indication that these putative amphiphilic helices are transmembrane.

These methods have also been used to analyze membrane surface interacting peptides and apolipoproteins.

Peptides - models of membrane proteins

Peptides began to be used to study protein-lipid interactions many years ago. In most cases, these were natural membrane-active peptides, primarily gramicidin A, alamethicin and mellitin. Currently, synthetic peptides are more often used as model systems. In this case, it is necessary to remember two points: 1) when binding a peptide to a membrane, both primary and secondary amphiphilicity are significant; 2) peptides often have polymorphism, i.e. the ability to change conformation depending on the environment. Not; It is possible that in the future, with the help of synthetic peptides, it will be possible to study protein-lipid interactions in detail, but for now we are still very far from this.

The formation of transmembrane proteins must include the stages of recognition of transmembrane domains and their integration into the lipid bilayer

Transmembrane domains extend laterally from the translocon through the protein-lipid interface

Positioning on the translocation channel and start transfer of secretory and transmembrane proteins happen the same way. However, the translocation of membrane proteins must be combined with their integration or insertion into the ER lipid bilayer. Integration occurs at the moment when the transmembrane domains are recognized by the translocon, then their translocation into the ER lumen stops, and they begin to be transferred from the channel into the lipid bilayer in the lateral direction. Many different types of transmembrane proteins are synthesized and integrated in this way, including those that span the membrane several times.

The first step on the protein integration pathway into the membrane is the recognition of transmembrane domains by the translocon. These domains extend over a distance of about twenty hydrophobic amino acids. Due to their hydrophobic nature, some transmembrane domains are recognized by SRP as signal sequences. These so-called signal anchor sequences first position the nascent protein on the ER and are then directed into the channel like normal signal sequences.

However signal anchor sequences are not cleaved from the protein, but are integrated into the membrane. As shown in the figure below, unlike the signal anchor sequence, most transmembrane domains are recognized by the translocon as soon as they leave the ribosome, after completion of targeting by a conventional N-terminal signal sequence. Information that the transmembrane domain has already been synthesized must be transferred to the translocon by a route other than transfer from SRP.

The simplest sign, indicating that the transmembrane domain is located in the translocon is the hydrophobicity of the domain itself. Due to the peculiarity of its structure, the translocation channel exhibits this hydrophobicity. As shown in Fig. 3.21, the structure of the translocon suggests that the channel is capable of opening like a clam shell, allowing the transmembrane domain to simultaneously contact the channel and the lipid bilayer. Apparently, signal sequences and transmembrane domains bind to the Sec61a protein located on the side of the opening valves, and this binding then causes the lateral opening of the channel.

This scheme has been proposed based on experimental data, according to which the transmembrane domains in the channel contact Sec61a and . As a result, although the translocon contains a water channel, there are enough hydrophobic channels in the membrane in which translocated polypeptides can enter the lipid environment. It is expected that regions containing polar amino acids should move through the channel without stopping, while hydrophobic domains, due to strong interactions with lipids, will remain associated with the side walls of the channel, preventing translocation.

In some cases translocon may identify the transmembrane domain differently. For example, sometimes during the synthesis of a transmembrane domain, changes in the interaction between the ribosome and the translocon occur before the domain leaves the ribosome. These changes serve as a signal to the translocon about the imminent appearance of a transmembrane domain. How the transmembrane domain causes changes in the ribosome and transfers them to the translocon remains unclear. Sometimes polar elements of the newly formed chain adjacent to the transmembrane domain are also required for recognition. This suggests that, at least in some cases, the recognition process must involve more than just a hydrophobic interaction between the domain and the lipid channel environment.

Boundary of the channel and the lipid layer, apparently, serves as a pathway for the exit of transmembrane domains from the channel after their recognition. However, the mechanism for domain exit from the translocon varies somewhat from substrate to substrate. Some domains leave the translocon almost immediately after being recognized in the channel. In these cases, the transmembrane domain first contacts Sec61a and lipids, and then only lipids; it is assumed that the domain has already penetrated into the lipid bilayer.

For integration of such domains of other proteins, with the exception of the Sec61 complex, is not required. Other transmembrane domains integrate more slowly and, after recognition, do not leave the translocon for a long time, sometimes even remaining there until the end of translation. As they exit the channel into the bilayer, these transmembrane domains come into contact with the TRAM protein, although its further role remains unclear.

Degree of hydrophobicity partly determines whether the transmembrane domain is integrated immediately or occurs at a later stage in protein synthesis. More hydrophobic domains can move more quickly into the lipid bilayer, but less hydrophobic ones can remain at the interface and require additional transport factors. It is possible that TRAM and other proteins serve as chaperones for some transmembrane domains. They promote the integration of domains whose hydrophobicity is insufficient for movement.

It's obvious that at least group of transmembrane proteins may be multiple forms containing a specific domain, which in some cases is integrated into the membrane, and in others remains unrecognized. Proteins such as TRAM may determine under what conditions such substitutions will be integrated.