Alpha-structures_Branden-Tooze

Alpha-structures_Branden-Tooze - The first globular protein...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: The first globular protein structure that was determined, myoglobin, belongs to the class of alpha- (Ge) domain structures. The structure illustrated in Figure 2.9 is called the globin foici and is a representative example of one class of or domains in proteins; short or helices, the building blocks, are connected by loop regions and packed together to produce a hydrophobic core. Packing interactions within the core hold the helices together in a stable globular structure, while the hydrophilic residues on the surface make the protein so]— uble in water. In this chapter we will describe some of the different iii-domain structures in soluble proteins. Alpha helices are sufficiently versatile to produce many very different classes of structures. In membrane-bound proteins, the regions inside the membranes are frequently or. helices whose surfaces are covered by hydropho- bic side chains suitable for the hydrophobic environment inside the mem- branes. Membrane-bound proteins are described in Chapter 12. Alpha helices are also frequently used to produce structural and motile proteins with vari- ous different properties and functions. These can be typical fibrous proteins such as keratin, which is present in skin, hair, and feathers, or parts of the cellular machinery such as fibrinogen or the muscie proteins rnyosin and dystrophin. These cit-helical proteins will be discussed in Chapter 14. Coiled-coil or helices contain a repetitive heptad amino acid sequence pattern Despite its frequent occurrence in proteins an isoiated or helix is only mar- ginally stable in solution. Alpha helices are stabilized in proteins by being packed together through hydrophobic side chains. The simpiest way to achieve such stabilization is to pack two :1 helices together. As early as 1953 Francis Crick showed that the side-chain interactions are maximized if the two or helices are not straight rods but are wound around each other in a supercoilr a so-caIled coiled-coil arrangement (Figure 3.1). Coiled-coils are the basis for some of the fibrous proteins we shall discuss in Chapter 14. Coiled—coils in fibers can extend over many hundreds of amino acid residues 14 am Figure 3.1 Schematic diagram of the coiled- coil structure. Two or helices are intertwined and gradually coii around each other. 35 a b c d e .f 9 NH; — Mel — Lys - Gin — Len — Glu — Asp — Lys Val - Glu — Glu - Len - Len - Ser - Lys — Asa - Tyr — His — Len — Glu - Asa - Glu Val - Ala — Arg - Len - Lys - Lys - Lari I COOH (If!) Figure 3.2 Repetitive pattern of amino acids in a coiled-coil 0L helix. (a) The amino acid sequence of the transcription factor GCN4 showing a heptad repeat of leucine residues. Within each heptad the amino acids are labeled avg. (1:) Schematic diagram of one heptad repeat in a coiled- coii structure showing the backbone of the poiypeptide chain. The or helices in the coiled-coil are slightly distorted so that the helical repeat is 3.5 residues rather than 3.6, as in a regular helix. There is therefore an integral repeat of seven residues along the helix. to produce long, flexible dimers that contribute to the strength and flexibiL ity of the fibers. Much shorter coiled-coils are used in some transcription fac- tors to promote or prevent formation of homo— and heterodimers, as we shall discuss in Chapter 10. Crick showed that a leftwhanded supercoil of two right-handed or helices reduces the number of residues per turn in each helix from 3.6 to 3.5 so that the pattern of side-chain interactions between the helices repeats every seven residues, that is, after two turns. This is reflected in the amino acid sequences of polypeptide chains that form tit-helical coiled-coils. Such sequences are repetitive with a period of seven residues, the heptad repeat. The amino acid residues within one such heptad repeat are usually labeled a—g (Figure 3.2a), and one of these, the d-residue, is hydrophobic, usually a leucine or an isoleucine. When two or helices form a coiled—coil structure the side chains of these d—residues pack against each other every second turn of the CL helices (Figure 3.21)). The hydrophobic region between the a helices is completed by the a-residues, which are frequently hydrophobic and also pack against each other (Figure 3.3). Residues "e" and "g," which border the hydrophobic core (see Figure 3.2b), frequently are charged residues. The side chains of these residues provide ionic interactions (salt bridges) between the or helices that define the relative chain alignment and orientation (Figure 3.4}. The repetitive heptad amino acid sequence pattern required for a coiled~ coil structure can be identified in computer searches of amino acid sequence databases. Heptad repeats provide strong indications of tic-helical coiled-coil structures, and they have been found in a number of different proteins with very diverse functions. Fibrinogen, which plays an essential role in blood coagulation; some RNA- and DNA-binding proteins; the class of cell-surface recognition proteins called collectins; both spectIin and dystrophin, which link actin molecules; and the muscle protein myosin all contain heptad repeats and therefore coiled-coil or helices. An illustrative example is pro- vided by GCN4, a DNA-binding protein. GCN4 contains one region of or helix, the leucine zipper region, and its dimerization is accomplished by the formation of an tic-helical coiled-coil with the leucine zipper regions of two subunits. The structure and DNA-binding function of this protein are described in Chapter 10. Detailed structure determinations of GCN4 and other coiled»coil pro- teins have shown that the 0'. helices pack against each other according to the "knobs in holes’' model first suggested by Francis Crick (Figure 3.5). Each side chain in the hydrophobic region of one of the or helices can contact four side chains from the second on helix. The side chain of a residue in position "d" Figure 3.3 Schematic diagram showing the packing of hydrophobic side chains between the two or helices in a coiled-coil structure. Every seventh residue in both or helices is a leucine, labeled "d." Due to the heptad repeat, the d-residues pack against each other along the coiled- coil. Residues labeled “a” are also usually hydrophobic and participate in forming the hydrophobic core along the coiled-coil. 36 (Eli [Ill ionic interactions fiydrapfiabic ionic interactions in one helix is directed into a hole at the surface of the second helix sur— rounded by one d-residue, two a-residues, and one e-residue, with numbers n, n — 3, n + 4, and n + 1, respectively. The two helices are aligned in such a way that the two d-residues, frequently leucines or isoleucines, face each other (see Figure 3.3). The four-helix bundle is a common donmin structure in 0: proteins Two 0!. helices packed together into a coiled-coil are building blocks within a domain or a fiber but are not sufficient to form a complete domain. The sim- plest and most frequent (ii-helical domain consists of four or helices arranged in a bundle with the helical axes almost parallel to each other. A schematic representation of the structure of the four-helix bundle is shown in Figure 3.6a. The side chains of each helix in the four-helix bundle are arranged so that hydrophobic side chains are buried between the helices and hydro- philic side chains are on the outer surface of the bundle (Figure 3.6b). This arrangement creates a hydrophobic core in the middle of the bundle along its length, where the side chains are so closely packed that water is excluded. The four-helix bundle occurs in several widely different proteins, such as myohemerythrin (an oxygen-transport protein in marine worms that does not contain heme iron), cytochrome c’ and cytochrome 13552 (heme-contain- ing electron carriers) (Figure 3.7a), ferritin (a storage molecule for iron atoms in eucaryotic cells), and the coat protein of tobacco mosaic virus. In these examples, sequentially adjacent or helices are always antiparallel. However, four—helix bundles can also be formed with different topological arrange— ments of the a helices. In human growth hormone (Figure 3.713), a four-helix bundle is formed from two pairs of parallel Ct helices that are joined in an (I!) helix 1 it} Figure 3.4 Salt bridges can stabilize coiled-coil structures and are sometimes important for the formation of heterodimeric coiled-coil structures. The residues labeled "e" and “g” in the heptad sequence are close to the hydrophobic core and can form salt bridges between the two or helices of a coiled-coil structure, the e-residue in one helix with the g—residue in the second and vice versa. (a) Schematic view from the top of a heptad repeat. (1)) Schematic View from the side of a coiled-coil structure. Figure 3.5 Schematic diagram of packing side chains in the hydrophobic core of coiled-coil structures according to the "knobs in holes” model. The positions of the side chains along the surface of the cylindrical or helix is projected onto a plane parallel with the heli- cal axis for both 0'. helices of the coiled-coil. (a) Projected positions of side chains in helix 1. (b) Proiected positions of side chains in helix 2. (c) Superposition of (a) and (b) using the relative orientation of the helices in the coiled-coil structure. The side-chain positions of the first helix, the "knobs," superimpose between the side-chain positions in the second helix, the "holes." The green shading outlines a d-residue (leucine) from helix 1 surrounded by four side chains from helix 2, and the brown shading outlines an a-residue (usually hydrophobic) from helix 1 surrounded by four side chains from helix 2. 37 antiparallel fashion. The interaction of this hormone with its receptor is described in Chapter 13. in most four-helix bundle structures, including those shown in Figure 3.7, the at helices are packed against each other according to the “ridges in grooves" model discussed later in this chapter. However, there are also exam- ples where coiled-coil dimers packed by the “knobs in holes" model partici- pate in four-helix bundle structures. A particularly simple illustrative exam— ple is the Rop protein, a small RNA-binding protein that is encoded by cer- tain plasmids and is involved in plasmid replication. The monomeric sub- unit of Rop is a polypeptide chain of 63 amino acids built up from two {fl} {{1} Figure 3.6 Four-helix bundles frequently occur as domains in 0'. proteins. The arrangement of the u helices is such that adjacent helices in the amino acid sequence are also adiacent in the three-dimensional structure. Some side chains from all four helices are buried in the middle of the bundle, where they form a hydrophobic core. (a) Schematic representation of the path of the polypeptide chain in a four— helix—bundle domain. Red cylinders are or helices. (b) Schematic view of a projection down the bundle axis. Large circles repreSent the main chain of the a helices; small circles are side chains. Green circles are the buried hydrophobic side chains; red circles are side chains that are exposed on the surface of the bundle, which are mainly hydrophilic. [{a) Adapted from PC. Weber and ER. Salernme, Nature 287: 82—84, 1930.] Figure 3.7 The polypeptide chains of cytochrome bfifig and human growth hormone both form four-helix—bundle structures. In cytochrome bfifiz (a) adjacent helices are antiparaliel, whereas the human growth hormone (b) has two pairs of parallel or helices joined in an antiparalle] fashion. Figure 3.8 Schematic diagram of the dimeric Rop molecule. Each subunit comprises two 0t helices arranged in a coiled-coil structure with side chains packed into the hydrophobic core according to the “knobs in holes" model. The two subunits are arranged in such a way that a bundle of four a helices is formed antiparallel or helices joined by a short loop of three amino acids. The struc- ture of Rop was determined by David Banner at EMBL, Heidelberg, Germany. The two at helices of the Rop subunit are arranged as an antiparallel coiled-coil in which the hydrophobic side chains are packed against each other according to the “knobs in holes" model. Two such subunits, each with the same structure, form the dimeric Rop molecule in which the subunits are arranged as a bundle of four or helices with their long axes aligned (Figure 3.8). The two dimers pack against each other according to the “ridges in grooves" model. The helix-loop-helix (HLH) family of transcription factors, discussed in Chapter 10, is another example of a four-helix bundle structure invoiving coiled-coil helices. Alpha—helical domains are sometimes large and complex The structures of several enzymes are known in which a long polypeptide chain of 300—400 amino acids is arranged in more than 20 0t helices packed together in a complex pattern to form a globular domain. One such enzyme is a bacterial muramidase that is involved in the metabolism of peptidogly- cans, which form part of the bacterial cell wall. The structure of this enzyme was determined by Bauke Dijkstra and colleagues in Groningen, Netherlands, as a basis for the design of specific inhibitors to the enzyme, which might lead eventually to novel types of antibacterial drugs. The polypeptide chain of this monomeric enzyme has 618 amino acids, of which the N-terminal 450 residues form one int-helical domain. This domain is built up from 27’ a helices arranged in a two-layered ring with a right-handed superhelical twist (Figure 3.9). The ring has a large central hole, like in a doughnut, with a diameter of about 30 it. The remaining residues form the catalytic domain that lies on top of the ring. The function of the Figure 3.9 Schematic diagram of the structure of one domain of a bacterial muramiclase, comprising 450 amino acid residues. The structure is built up from 27 a helices arranged in a two-layered ring. The ring has a large central hole, like a doughnut, with a diameter of about 30 A. doughnut-shaped domain is not known, but its shape may be required for the specificity of the catalytic reaction in vivo. The globin fold is present in myoglobin and hemoglobin One of the most important or structures is the globin fold. This fold has been found in a large group of related proteins, including myoglobin, hemoglo- bins, and the light—capturing assemblies in algae, the phycocyanins. The functional and evolutionary aspects of these structures will not be discussed in this book; instead, we will examine some features that are of general struc— tural interest. The pairwise arrangements of the sequential or helices in the globin fold are quite different from the antiparallel organization found in the four-helix- bundle :1 structures. The globin structure is a bundle of eight or helices, usu- ally labeled A—H, connected by rather short loop regions and arranged so that the helices form a pocket for the active site, which in myoglobin and the hemoglobins binds a heme group (Figure 3.10). The lengths of the or helices vary considerably, from 7 residues in the shortest helix (C) to 28 in the longest helix (H) in myoglobin. In the globin fold the or helices wrap around the core in different directions so that sequentially adjacent or helices are usu- ally not adjacent to each other in the structure. The only exceptions are the last two a helices (G and H), which form an antiparallei pair with extensive packing interactions between them. All other packing interactions are formed between pairs of or helices that are not sequentially adjacent. Because the globin fold is not built up from an assembly of smaller motifs, it is quite difficult to visualize conceptually in spite of its relatively small size and simplicity. Geometric considerations determine car—helix packing When we compare the arrangements of the a helices in coiled-coil structures (see Figure 3.1), in the four~helix—bundle structure (see Figure 3.8), and in the globin fold (see Figure 3.10), it is obvious that the geometry of (it-helix pack- ing is quite different. We described earlier the way that the side chains of coiled-coil or helices pack according to the “knobs in holes" model. In con- trast, other or-helical structures pack their cc helices according to a "ridges in grooves" model. In the four-helix bundle the or helices pack almost parallel, or antiparallel, to each other, with an angle of about 20" between the helical axes. In the globin fold the angles between the helical axes are usually larger, in most cases around 50". These are the two main ways that or helices pack against each other in the "ridges in grooves" model, a packing motif dictated by the geometry of the surfaces of or helices. Ridges of one o: helix fit into grooves of an ndiacent helix Since the side chains of an a helix are arranged in a helical row along the sur- face of the helix, they form ridges separated by shallow furrows, or grooves, on the surface. Alpha helices pack with the ridges on one helix packing into the grooves of the other and vice versa. The ridges and grooves are formed by amino acids that are usually three or four residues apart. This is illustrated in Figure 3.11, which shows slices through the surface of a polyalanine a helix on which the directions of the ridges are marked. In contrast to the ridges and grooves of the DNA double helix described in Chapter 7, which are formed by the sugar-phosphate main-chain atoms, those of an or helix are formed by the amino acid side chains. The detailed geometry of the ridges and grooves of an or helix is thus dependent not only on the geometry of the helix but also on the actual amino acid sequence. The most common way of packing a helices is by fitting the ridges formed by a row of residues separated in sequence by four in one helix into the same type of grooves in the other helix. In this case the ridges and 40 Figure 3.10 Schematic diagram of the globin domain. The eight or helices are labeled A—H. A—D are red, E and F green, and G and H blue. The heme group is shown in white. (Adapted from originals provided by A. Lesk.) grooves form an angle of about 25° to the helical axis. In order to pack the two helices shown in Figure 3.12a (red and blue) against each other, one of these (the blue in Figure 3.12a) must be turned around 180“ out of the plane of the paper and placed on top of the other (red). In the interface between the two or helices the directions of the ridges and grooves are then on oppo- site sides of the vertical axis, as illustrated in Figure 3.123. The on helices must thus be inclined by an angle of about 50" (25° + 25°) in order for the ridges of one helix to fit into the grooves of the other and vice versa. This is the type of packing of several of the helix-helix interactions in the globin fold, and in many other helical structures. In the second frequently occurring packing mode the ridges formed by amino acids three residues apart fit into the grooves of amino acids four residues apart and vice versa. The direction of the first type of ridge forms an angle of about 45“ to the helical axis, whereas the other type makes an angle of about 25° to the axis in the opposite direction (Figure 3.121)). In the inter- face, hOWever, after one helix has been rotated 180“, these directions are on the same side of the helical axis. Thus an inclination of about 20“ (45°——25“) between the two a helices will fit these ridges and grooves into each other. Some four-helix-bundle structures (see Figure 3.8b) have this mode of packing. These two rules for fitting ridges into grooves are quite general: they apply to most packing interactions between or helices, and they explain the geometrical arrangements of adjacent or helices observed in many protein structures. The globin fold has been preserved during evolution The three-dimensional structures of globin domains from many diverse sources, including mammals, insects, and plant root nodules, have been determined independently of each other. All these domains have amino acid Figure 3.11 The side chains on the surface of an or helix form ridges separated by grooves, as schematically illustrated here. (a) An a helix with each residue represented by the first atom in the side chain, C5. (b) The surface relief of a polyalanine or helix in the orientation shown in (a). Sections are cut through a space-filling model and superimposed. The residue numbers are placed on the side-chain atom. The ridges caused by the side chains separated by four residues are shown as lines. (c) The same as (b), but here the ridges are caused by side chains separated by three residues. 41 (a) helix [ helix ll Sequence homologies that range from 99% to 16% in pairwise comparisons, but they all share the same essential features of the globin fold. This family of structures is thus the prime example of a situation where natural seiection has produced proteins whose amino acid sequences have diverged widely (although some homology is usually still recognizable) but whose three- dimensional snucture has been essentialiy preserved. Arthur Lesk and Cyrus Chothia at the MRC Laboratory of Molecular Biol- ogy in Cambridge, UK, compared the family of globin structures with the aim of answering two general questions: How can amino acid sequences that are very different form proteins that are very similar in their three-dimensional structure? What is the mechanism by which proteins adapt to mutations in the course of their evolution? The hydrophobic interior is preserved To answer the first question, Lesk and Chothia examined in detail residues at structurally equivalent positions that are involved in helix-heme contacts and in packing the or helices against each other. After comparing the nine glo- bin structures then known, the 59 positions they found that fulfilled these criteria were divided into 31 positions buried in the interior of the protein and 28 in contact with the heme group. These positions are the principal determinants of both the function and the three—dimensional structure of the globin family. One might expect these positions to exhibit a higher degree of amino acid conservation and hence sequence identity than the rest of the mole- cule. This is not, however, the case for distantly related molecules that have low sequence identity and derive from distantly related species. The sequence identity of these residues is no greater than in the rest of the r 42 Figure 3.12 By fitting the ridges of side chains from one helix into the grooves between side chains of the other helix and vice versa, or helices pack against each other. (a) Two (1 helices, I and II, with ridges from side chains separated by four residues marked in red and blue, respectively. Panels 1 and 2 are the same view of the two or helices. In panel 3 the blue a helix is turned over through 180“ in order to form an interface with the red or helix. In panei 4 the orientation of the helices has been rotated 50° in order to pack the ridges of one a helix into the grooves of the other. (b) in the red or helix the ridges are formed by side chains separated by four residues and in the blue CL helix by three residues. The O: helices are rotated 20“ in order to pack ridges into grooves, in a direction opposite that in (a). (Adapted from C. Chothia et al., Proc. Natl. Acad. Sci. USA 74: 4130—4134, 1977.) molecule. Since the important residues involved in packing the or helices are not conserved, we used to assume that the changes which have occurred compensate each other in size. This is not the case either. The volumes occu- pied by the 31 buried residues vary considerably between individual mem- bers. Thus neither conserved sequence nor size-compensatory mutations in the hydrophobic core are important factors in preserving three-dimensional structure during evolution. We now know that this is also true for other pro- teins, such as the immunoglobulins. Lesk and Chothia did find, however, that there is a striking preferential conservation of the hydrophobic character of the amino acids at the 59 buried positions, but that no such conservation occurs at positions exposed on the surface of the molecule. With a few exceptions on the surface, hydrophobic residues have replaced hydrophilic ones and vice versa. How- ever, the case of sickle-cell hemoglobin, which is described below, shows that a charge balance must be preserved to avoid hydrophobic patches on the surface. in summary, the evolutionary divergence of these nine globins has been constrained primarily by an almost absolute conservation of the hydro— phobicity of the residues buried in the helix—to-helix and heiix-to-heme contacts. Helix movements accommodate interior side-chain mutations Lesk and Chothia also found a simple answer to the question of how proteins adapt to changes in size of buried residues. The mode of packing the or helices is the same in all the globin structures: the same types of packing of ridges into grooves occur in corresponding or helices in all these structures. How- ever, the relative positions and orientations of the a helices change to accommodate changes in the volume of side chains involved in the packing. The proteins thus adapt to mutations of buried residues by changing their overall structure, which in the globins involves movements of entire or helices relative to each other. The structure of loop regions changes so that the movement of one or helix is not transmitted to the rest of the structure. Only movements that preserve the geometry of the heme pocket are ac— cepted. Mutations that cause such structural shifts are tolerated because many different combinations of side chains can produce well-packed helix- helix interfaces of similar but not identical geometry and because the shifts are coupled so that the geometry of the active site is retained. Sickle-cell hemoglobin confers resistance to malaria Sickle-cell anemia is the classic example of an inherited disease that is caused by a change in a protein's amino acid sequence-Linus Pauling proposed in 1949 that it was caused by a defect in the hemoglobin molecule; he thus coined the term molecular disease. Seven years later Vernon Ingram showed that the disease was caused by a single mutation, a change in residue 6 of the 5 chain of hemoglobin from Glu to Val. Hemoglobin is a tetrarner built up of two copies each of two different polypeptide chains, or- and B—globin chains in normal adults. Each of the four Chains has the globin fold with a heme pocket. Residue 6 in the [3 chain is on the surface of Ct helix A, and it is also on the surface of the tetrameric mole- cule (Figure 3.13). The hemoglobin concentration in red blood cells, erythrocytes, is extremely high, 340 mg/m]. This is almost as high as in the crystalline state: the hemoglobin molecules, which are spheroids of dimension 50 x 55 x 65 A, are on average only 10 it apart in the cells. It is thus surprising that they can nevertheless rotate and flow past one another. The mutation in sickle-cell hemoglobin converts a charged residue to a hydrophobic residue, and as a result, it produces a hydrophobic patch on the surface. This patch happens to fit and bind a hydrophobic pocket in the deoxygenated form of another 43 ...
View Full Document

Page1 / 9

Alpha-structures_Branden-Tooze - The first globular protein...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online