1.1 Configuration, conformation, and the backbone #
The study of Configuration, conformation, and the backbone provides fundamental principles governing cellular architecture, molecular interactions, and physiological regulation.

Protein Backbone Phipsiomega
1.1.1 Configuration versus conformation #
Configuration is fixed by covalent bonds: connectivity, stereochemistry, and, once formed, which cysteines are paired. Changing it means breaking a bond. The amino acids in proteins are L, except glycine, which is not chiral; a D residue is a different molecule, and its backbone map is the mirror image. Side-chain chemistry is Biochemistry. Conformation is the arrangement reached by rotation about single bonds. One configuration has many conformations, and folding is a search among them. Side chains cluster into rotamers, the Ο angles. The backbone has fewer free variables than that list suggests.
1.1.2 The peptide bond and the two free angles #
The peptide bond is partly double, about 1.33 Γ rather than about 1.45 Γ for a CβN single bond, and prefers to be planar. Its torsion, Ο, is usually trans (near 180Β°). Cis (near 0Β°) clashes the flanking Ξ±-carbons. About 5% of bonds before proline are cis, against well under 0.5% of other peptide bonds, because the ring narrows the energy gap. Cisβtrans isomerisation is slow and can limit folding. With Ο fixed, Ο is rotation about NβCΞ± and Ο about CΞ±βC. By convention 180Β° is extended. Backbone conformation is a choice of Ο and Ο along the chain.
1.1.3 The Ramachandran plot #
Carbonyl oxygen, amide hydrogen, and the Ξ²-carbon collide in most of the (Ο, Ο) map that Ramachandran drew. The origin, near 0Β°, 0Β°, is impossible. Real proteins occupy two basins: a right-handed Ξ± region around Ο β β60Β°, Ο β β45Β°, and a broad Ξ² region around Ο β β120Β°, Ο β +120Β°. A smaller left-handed Ξ± region, near +60Β°, +45Β°, is strained wherever there is a Ξ²-carbon.

6 Ramachandran Plots Of Wwpdb Validation
Glycine has no Ξ²-carbon, so it is allowed where others clash, including much of the left-hand side. Backbone atoms still collide near the origin, which is why glycine has its own validation plot. Proline is the opposite: the ring fixes Ο near β65Β°, and there is no amide hydrogen to donate. Proline breaks a helix in the middle and is tolerated at the N-terminus, where that hydrogen bond is missing anyway. Empirical plots divide favoured, allowed, and outlier regions. An outlier is either real strain or an error in the model. A cluster of them in a badly ordered active site is a warning about the coordinates, not evidence of exotic chemistry, until the density or the restraints have been checked.
1.2 Secondary structure #
1.2.1 The Ξ± helix #
Pauling and Corey predicted the Ξ± helix from peptide geometry. It repeats near Ο, Ο = β60Β°, β45Β°, with 3.6 residues per turn, a rise of about 1.5 Γ , and a pitch of about 5.4 Γ . The carbonyl of residue i hydrogen-bonds to the NH of i+4. That register is the definition used here. An i to i+3 bond is a short 3ββ helix, often at the end of an Ξ± helix; an i to i+5 Ο helix is a rare defect. Protein helices are right-handed. Side chains point outwards, and the aligned peptide dipoles leave the N-terminus partially positive and the C-terminus partially negative. Proline cannot donate the i+4 hydrogen bond and kinks the axis. A helix guessed from sequence alone is not a structure. Collagen is a different periodic fold: three left-handed polyproline II helices in a right-handed triple helix, needing glycine every third residue. The chemistry of hydroxyproline is Biochemistry.

Alpha Helix Wheel
1.2.2 Ξ² sheets #
A Ξ² strand is extended, in the Ξ² basin, with side chains alternating above and below. In an antiparallel sheet, neighbours run in opposite directions and the hydrogen bonds between them are nearly perpendicular to the strands, linear, and strong. A hairpin is the simplest unit; silk is a macroscopic antiparallel sheet. In a parallel sheet the strands run the same way, the hydrogen bonds are angled, and an isolated pair is less favourable. Parallel strands are usually buried inside Ξ±/Ξ² domains and joined by helices or long loops. Real sheets twist right-handedly. An exposed edge can satisfy its hydrogen bonds by sticking to another molecule, so natural sheets curl, cover, or charge their edges.

Beta Sheet Bonding Parallel-Color
1.2.3 Turns #
A Ξ² turn uses four residues, with a hydrogen bond from i to i+3, and reverses the chain. In a type II turn the third residue sits where glycine is comfortable and a Ξ²-carbon is strained, so glycine is strongly enriched there. A type I turn often has proline second. A Ξ³ turn is three residues, i to i+2. Longer loops have no single geometry. They carry many binding and catalytic groups, and they are the first part of a weak map to get wrong. “Loop” means neither “unimportant” nor, without evidence, “disordered”.
| Element | Geometry | Hydrogen bonds |
|---|---|---|
| Ξ± helix | 3.6 residues/turn; rise about 1.5 Γ ; Ο, Ο about β60Β°, β45Β° | i to i+4 |
| 3ββ helix | 3 residues/turn | i to i+3 |
| Antiparallel Ξ² | Ο, Ο near β120Β°, +120Β° | nearly perpendicular bonds between strands |
| Parallel Ξ² | same Ξ² basin | angled bonds between strands |
| Type II Ξ² turn | four residues | i to i+3; glycine favoured at residue three |
1.3 Motifs, domains, and quaternary structure #
1.3.1 Supersecondary motifs #
A shared local packing is geometry, not by itself common ancestry. A Ξ² hairpin is two antiparallel strands and a short turn. A Greek key lets neighbours in a four-strand sheet be non-neighbours in sequence. A Ξ²βΞ±βΞ² unit connects parallel strands by a helix and is almost always a right-handed crossover; repeats build the Rossmann nucleotide-binding fold and the TIM barrel, whose chemistry sits in the loops at the C-terminal ends of the strands. A coiled coil winds helices on a seven-residue repeat, hydrophobic side chains at positions one and four packing as knobs into holes. That motif returns as the leucine zipper. A helixβturnβhelix is the geometric unit that, in Chapter 9, places one helix in the major groove of DNA.
1.3.2 Domains #
A domain folds alone, often about 100β200 residues, with its own core. Longer chains are usually several domains, packed together or hinged. A crystal shows one hinge angle. Some domains are discontinuous in sequence. Folds are described as all Ξ±, all Ξ², Ξ±/Ξ², or Ξ±+Ξ². The immunoglobulin fold is a Ξ² sandwich that can display many different loops; how antibodies generate that variety is Immunology. A fold is not a function. The TIM barrel is used for many reactions, and seven transmembrane helices will later pose the same warning.
1.3.3 Quaternary assembly and symmetry #
Quaternary structure is the packing of separate chains. Symmetry is common because one gene then builds the object and a binding surface evolves once. Cyclic symmetry, Cn, is a single rotation axis: subunits in a ring, all pointing the same way. C2 is a twofold dimer. Dihedral symmetry, Dn, adds perpendicular twofold axes, a second copy related by a flip. A D2 tetramer has three mutually perpendicular twofold axes. Haemoglobin, Ξ±2Ξ²2, is only approximately 222, because Ξ± and Ξ² differ: effectors can bind between the Ξ² chains, and the tetramer can still switch state.
Closed symmetry fixes the number of subunits. Helical symmetry, rotation plus rise, does not. Actin, sickle-cell fibres, and amyloid keep adding subunits because the contact is open. A D2 tetramer cannot become a fibre without a new kind of contact. Molecular symmetry is not crystal symmetry, and a crystal contact is not automatically biological. The ribosome has no point-group symmetry and does not need one.
1.4 When there is no single structure #
1.4.1 Intrinsically disordered regions #
An intrinsically disordered region does not adopt one tertiary structure under native conditions. It is not a chain an experimenter has forcibly unfolded. Such sequences lack a hydrophobic core and are rich in charged and polar residues, often proline and glycine. A charge-versus-hydrophobicity plot separates many of them from globular proteins, as a tendency, not a diagnosis. Disorder can be the functional state. Short motifs bind reversibly, and phosphorylation can switch them. Some segments fold on binding: the transactivation region of p53 is disordered alone and partly helical on MDM2, whereas its DNA-binding domain is a stable fold. The entropy cost of folding at the moment of binding keeps the affinity modest, which suits signalling. A few metamorphic proteins, lymphotactin among them, populate two unrelated folds.
1.4.2 Why one coordinate set is the wrong picture #
Even a folded domain is a family of nearby conformations at room temperature. T and R haemoglobin, or an open and a closed domain, are farther apart and are still one protein. A disordered region is a broad ensemble, from which a partner may select one member. A coordinate file is one state, one set of conditions, and one average over motion. Missing crystal density, sharp random-coil NMR signals, fast hydrogen exchange, and an unexpectedly large hydrodynamic radius are the signs. None alone assigns a function. Together they mean that one cartoon has dropped the biology. Low confidence in a prediction, defined in Chapter 10, is the same warning from software.
1.5 Key terms, summary, and self-check #
Key terms. Configuration; conformation; Ο; Ο; Ο; Ramachandran plot; Ξ± helix; Ξ² sheet; Ξ² turn; domain; cyclic symmetry; dihedral symmetry; intrinsically disordered region.
Summary. Configuration is covalent; conformation is rotation about the bonds that remain. With the peptide bond planar, Ο and Ο are restricted by clashes to the Ξ± and Ξ² basins, glycine being allowed more and proline fixed in Ο. The Ξ± helix hydrogen-bonds i to i+4; antiparallel and parallel sheets differ in direction and hydrogen-bond geometry; turns reverse the chain. Domains fold alone, and quaternary symmetry is cyclic, dihedral, or open and helical. Disorder, alternative states, and thermal motion make one coordinate set a sample, not the molecule.
Self-check.
1. Why is cisβtrans isomerisation not free during ordinary folding, whereas a change in Ο is?
2. A residue sits near Ο = +60Β°, Ο = +45Β°. Which amino acid is most plausible, and why?
3. State the hydrogen-bond register of an Ξ± helix and of a Ξ² turn. Why does a type II turn favour glycine?
4. What does dihedral symmetry add to cyclic symmetry, and why can helical symmetry make a fibre?
5. Give two signs that a stretch is disordered rather than merely missing from a crystal, and say why it can still function.
—
—
—