Biomolecules and biological membranes

Every standard amino acid has a central alpha carbon bonded to an amino group, a carboxyl group, a hydrogen atom and a side chain (R). Proteins are built from 20 standard amino acids. Every one except glycine has a chiral alpha carbon, and in proteins it is always the L configuration; glycine (R = H) is achiral. Proline's side chain loops back and bonds to its own alpha-amino nitrogen, making it a secondary amine (traditionally called an imino acid); its ring fixes the backbone angle $\varphi$, so proline often puts a kink or bend in a chain. Beyond the standard 20, a few proteins contain selenocysteine (encoded by a recoded UGA) and, in some archaea and bacteria, pyrrolysine (a recoded UAG).

In solution near neutral pH an amino acid exists as a zwitterion: the alpha-carboxyl group is $-{\text{COO}}^{-}$ and the alpha-amino group is $-{\text{NH}}_3^{+}$. For an amino acid without an ionisable side chain the net charge is then zero; at pH 7 Asp and Glu carry about -1 and Lys and Arg about +1. Typical pKa values are about 2 for the alpha-carboxyl and 9 to 10 for the alpha-amino group. Side chains with ionisable groups (Asp, Glu, His, Cys, Tyr, Lys, Arg) add a third pKa.

The peptide bond is an amide link between the alpha-carboxyl carbon of one residue and the alpha-amino nitrogen of the next, formed with the loss of water. Resonance gives the C-N bond partial double-bond character (it is shorter than a C-N single bond), so the six atoms of the peptide group (C$\alpha$, C, O, N, H, C$\alpha$) lie in one plane and rotation about the C-N bond is restricted. The trans arrangement ($\omega \approx 180^{\circ }$) is strongly favoured; the cis form occurs appreciably only in X-Pro bonds (the peptide bond to proline's nitrogen). Rotation is possible about N-C$\alpha$ (angle $\varphi$) and C$\alpha$-C (angle $\psi$); the sterically allowed ($\varphi$, $\psi$) combinations are shown on a Ramachandran plot, where right-handed alpha helices and beta strands fall in separate allowed regions. Peptides are written and numbered from the N-terminus to the C-terminus.

The right-handed alpha helix has 3.6 residues per turn, a rise of 0.15 nm per residue and a pitch of 0.54 nm. The C=O of residue i hydrogen bonds to the N-H of residue i+4, so the hydrogen bonds run nearly parallel to the helix axis, and the side chains point outward. Within a helix proline is disruptive because its nitrogen has no hydrogen to donate and its ring prevents the required $\varphi$ angle; glycine also destabilises helices because its flexibility favours other conformations. In a beta sheet, extended strands (rise about 0.33 to 0.35 nm per residue) lie side by side and hydrogen bond between strands, with side chains pointing alternately above and below the sheet. Strands may run parallel or antiparallel; in antiparallel sheets the hydrogen bonds are more nearly linear. A beta turn reverses chain direction over four residues, with the C=O of residue i bonded to the N-H of residue i+3, and often contains Gly or Pro.

Folding is governed by the amino acid sequence: a denatured protein can often refold in vitro to its native, active state, which shows that the sequence contains the information that specifies the native fold. In cells, chaperones help many proteins fold by preventing misfolding and aggregation, but they do not supply that information. In water-soluble proteins hydrophobic side chains are mostly buried and polar ones exposed. Disulfide bonds (Cys-Cys) form in oxidising environments and are the only covalent bonds among these stabilising forces.

Denaturation is loss of secondary, tertiary and quaternary structure without breaking peptide bonds, so the primary structure is unchanged. Heat, extreme pH, urea and guanidinium chloride (which disrupt hydrogen bonds and weaken the hydrophobic effect), and detergents all denature proteins. Reducing agents such as beta-mercaptoethanol or dithiothreitol break disulfide bonds. Denaturation is usually accompanied by loss of function and decreased solubility.

Monosaccharides are polyhydroxy aldehydes (aldoses) or ketones (ketoses). Glucose is an aldohexose; fructose is a ketohexose. A sugar with n chiral carbons in its open-chain form has $2^n$ stereoisomers, provided none is a meso form. D and L refer to the configuration at the chiral carbon farthest from the carbonyl, and natural sugars are mostly D. Epimers differ at only one chiral carbon: glucose and galactose differ at C4, glucose and mannose at C2.

In solution, pentoses and hexoses cyclise: an aldose forms a hemiacetal and a ketose a hemiketal, giving six-membered pyranose or five-membered furanose rings. The former carbonyl carbon (C1 of an aldose, C2 of a ketose) becomes the anomeric carbon, and the two configurations at it, alpha and beta, are anomers. For a D-sugar drawn as a Haworth projection, alpha has the anomeric OH on the opposite side of the ring from the terminal CH2OH group (below the ring) and beta has it on the same side (above). The anomers interconvert through the open-chain form, and the resulting change in optical rotation of a freshly dissolved sample is called mutarotation; at equilibrium D-glucose is about one-third alpha and two-thirds beta. A sugar whose anomeric carbon is free (not in a glycosidic bond) can open to the carbonyl form and reduce mild oxidising agents such as alkaline Cu2+ reagents; it is a reducing sugar. Ketoses such as fructose also test as reducing, because under the alkaline test conditions they isomerise to aldoses.

Starch (plants) and glycogen (animals) are storage polysaccharides; cellulose and chitin are structural. Humans make no enzyme that hydrolyses the beta(1-4) bond between two glucose units, so cellulose is not digested (lactase cleaves the beta(1-4) galactose-glucose bond of lactose but does not act on cellulose). Glycogen's dense branching gives many non-reducing ends, where enzymes add or remove glucose units, so it can be built up and mobilised rapidly.

Lipids are biological molecules that are poorly soluble in water and soluble in nonpolar solvents. Fatty acids are long hydrocarbon chains (usually an even number of carbons, 12 to 24) ending in a carboxyl group. In the notation 18:1(${\Delta }^9$), 18 is the number of carbons, 1 the number of double bonds, and ${\Delta }^9$ places the double bond between C9 and C10 counting from the carboxyl carbon (C1); omega ($\omega$) numbering counts from the methyl end instead. Natural unsaturated fatty acids are mostly cis. Saturated chains pack tightly; each cis double bond kinks the chain, so melting point falls as unsaturation rises (and rises with chain length). Triacylglycerols (triglycerides) are glycerol esterified with three fatty acids and store energy. Phospholipids have a polar head and two hydrophobic tails, making them amphipathic; in water they form bilayers.

A nucleoside is a base plus a pentose sugar; a nucleotide is a nucleoside plus one or more phosphate groups (adenosine is a nucleoside; AMP, ADP and ATP are nucleotides). Purines (adenine, guanine) have two fused rings; pyrimidines (cytosine, thymine, uracil) have one. DNA uses 2'-deoxyribose and the bases A, G, C, T; RNA uses ribose and has U in place of T. The base joins the sugar's C1' by a beta-N-glycosidic bond (through N9 of a purine or N1 of a pyrimidine), and the phosphate is esterified to C5'. The 2'-OH of ribose makes RNA susceptible to alkaline hydrolysis, which DNA resists.

Nucleotides are linked by 3'-5' phosphodiester bonds, so a strand has a 5' end (usually carrying a phosphate) and a 3' end (free 3'-OH). Sequences are written 5' to 3'. In the double helix the two strands are antiparallel and complementary, with the negatively charged sugar-phosphate backbone outside. They are held together by hydrogen bonds between bases (A with T by 2 bonds, G with C by 3) and by stacking interactions between neighbouring base pairs, which contribute most of the helix's stability.

The fluid mosaic model describes a membrane as a fluid phospholipid bilayer in which proteins float, and lipids and proteins diffuse laterally. Hydrophobic tails face inward and polar heads face the water. The two leaflets differ in composition (asymmetry). Lateral diffusion within a leaflet is rapid, while transverse movement of a lipid from one leaflet to the other (flip-flop) is very slow unless catalysed by flippases, floppases or scramblases. Carbohydrates, attached to lipids and proteins, lie only on the extracellular face of the plasma membrane (the side facing away from the cytosol).

Integral proteins are embedded in the bilayer and need detergents to release them; many span it with hydrophobic alpha helices of about 20 residues, and some (porins, in the outer membranes of bacteria, mitochondria and chloroplasts) form beta barrels. Peripheral proteins bind the surface by ionic and hydrogen bonds and can be released with high salt or a pH change. Fluidity increases with more unsaturated and shorter fatty acid chains and with higher temperature. Cholesterol acts as a buffer: it reduces fluidity at high temperature and prevents tight packing (keeps the membrane fluid) at low temperature.

The Na+/K+-ATPase is a P-type ATPase: during its cycle it is transiently phosphorylated on an aspartate residue. In each cycle it pumps 3 Na+ out and 2 K+ in using 1 ATP, so it is electrogenic (net one positive charge leaves per cycle). Ouabain and digoxin inhibit it. It maintains the high intracellular K+ and low intracellular Na+ that set the resting potential and drive secondary active transport. The SERCA Ca2+ pump and the gastric H+/K+-ATPase are also P-type. Other ATP-driven pumps include V-type ATPases, which pump protons to acidify lysosomes and vesicles and form no phosphorylated intermediate, and ABC transporters (for example multidrug-resistance pumps that export drugs). F-type ATPases (the F1Fo ATP synthase) normally run the other way, making ATP as protons flow down their gradient; they can also hydrolyse ATP to pump protons.

Unequal ion concentrations and selective permeability produce a voltage across the membrane. Approximate values for a mammalian cell (they vary with cell type) are 140 mM K+ inside and 5 mM outside, 10 mM Na+ inside and 145 mM outside, and 10 mM Cl- inside and 110 mM outside. The equilibrium potential of one ion is the voltage at which its electrical and chemical driving forces balance, given by the Nernst equation. Convention: membrane potentials are inside relative to outside.

The resting potential (about -70 mV in many neurons) is close to E_K but less negative, because the resting membrane is mostly permeable to K+ through leak channels, with a small Na+ leak. The action potential is an all-or-none, self-propagating voltage change in excitable cells:

The spike is all-or-none: once threshold is reached its amplitude does not depend on stimulus strength, and a stronger stimulus raises the firing frequency instead. During the absolute refractory period (Na+ channels inactivated) no stimulus can trigger a second spike, which keeps the impulse travelling away from where it started. In the relative refractory period a stronger-than-normal stimulus is needed. Conduction velocity rises with axon diameter and with myelination: myelin insulates the axon so the impulse jumps between nodes of Ranvier (saltatory conduction), which is faster and uses less energy than continuous conduction. Tetrodotoxin blocks voltage-gated Na+ channels; tetraethylammonium blocks voltage-gated K+ channels. Channels can be voltage-gated, ligand-gated or mechanically gated.

Motor proteins convert the chemical energy of ATP hydrolysis into directed movement along cytoskeletal filaments. Filaments have polarity, with a plus end and a minus end (for actin, the barbed and pointed ends), and each motor moves in one direction along its track.

Most kinesins move to the plus end; the kinesin-14 family is an exception and moves to the minus end. The cross-bridge cycle of myosin has four steps. ATP binds and the myosin head detaches from actin. ATP is hydrolysed (ADP and Pi stay bound), which cocks the head into a high-energy state. The head binds actin and releases Pi, triggering the power stroke. ADP is released, leaving the rigor state until a new ATP arrives. Kinesin is a dimer that walks hand over hand along a microtubule in steps of about 8 nm, hydrolysing about one ATP per step; myosin V walks along actin in steps of about 36 nm. In skeletal muscle, rising Ca2+ binds troponin (troponin C), which shifts tropomyosin off the myosin-binding sites on actin; thick and thin filaments then slide past each other (sliding filament model), shortening the sarcomere without shortening the filaments: the A band keeps its length while the I band and H zone shorten.

Rotary motors are also molecular motors: the F1Fo ATP synthase rotates as protons flow down their gradient, and the bacterial flagellar motor is driven by a proton (in some species Na+) gradient rather than by ATP.