Protein

• 1. Introduction

  ›

  Protein is the molecule that built a false idea before it built the truth. In 1838, the Dutch chemist Gerardus Johannes Mulder ran an elemental analysis of common proteins and found that nearly all of them shared the same empirical formula, C400H620N100O120P1S1. From this he drew a wrong conclusion. He decided they might all be a single type of very large molecule. His associate, the Swedish chemist Jöns Jacob Berzelius, gave these substances the name we still use. It comes from the Greek proteios, meaning primary, in the lead, or standing in front. The name fit better than Mulder knew. Proteins are large biomolecules built from one or more long chains of amino acid residues, and they carry out a vast array of jobs inside living things. They catalyse metabolic reactions. They copy DNA. They respond to stimuli, give cells their shape, and ferry molecules from place to place. So how does a chain of small units fold itself into a working machine? How did scientists pry apart something this small and this essential? And why does a single methyl group sometimes decide whether two proteins ever touch?

• 2. From Albumins To Insulin

  ›

  Antoine Fourcroy and others had been studying these substances since the 1700s, often lumping them together as albumins, or in German, Eiweisskörper. In 1789, Fourcroy named three distinct varieties of animal protein: albumin, fibrin, and gelatin. Gluten had already been separated from wheat in published research around 1747, and was later found in many plants. Vegetable proteins studied in this era included gluten, plant albumin, gliadin, and legumin.

  Carl von Voit, a German nutritional scientist, believed protein was the single most important nutrient for the body, because people generally held that flesh makes flesh. Around 1862, Karl Heinrich Ritthausen isolated the amino acid glutamic acid. Thomas Burr Osborne compiled a detailed review of vegetable proteins at the Connecticut Agricultural Experiment Station. Working with Lafayette Mendel, Osborne fed laboratory rats and established which amino acids were nutritionally essential. Rats on a diet missing an essential amino acid simply stopped growing, in line with Liebig's law of the minimum. The last essential amino acid to be found, threonine, was identified by William Cumming Rose.

  The understanding of proteins as polypeptides, chains of amino acids, came through Franz Hofmeister and Hermann Emil Fischer in 1902. Yet the idea that proteins acted as enzymes was not fully accepted until 1926, when James B. Sumner showed that the enzyme urease was itself a protein. The first protein to have its amino acid chain read out was insulin, sequenced by Frederick Sanger in 1949. Sanger proved that proteins are linear polymers of amino acids, not branched chains, colloids, or cyclols, and he won the Nobel Prize for it in 1958.

• 3. Reading The Shape

  ›

  Linus Pauling is credited with successfully predicting regular protein secondary structures from hydrogen bonding, an idea first floated by William Astbury in 1933. Walter Kauzmann's work on denaturation, building on earlier studies by Kaj Linderstrøm-Lang, showed how folding is driven by hydrophobic interactions. Christian Anfinsen studied the oxidative folding of ribonuclease A and won the Nobel Prize in 1972, supporting the thermodynamic hypothesis: a protein's folded form sits at its free energy minimum.

  X-ray crystallography opened a door no light microscope could, since proteins are far too small to see directly. The first protein structures solved were hemoglobin by Max Perutz and myoglobin by John Kendrew, both in 1958. In 1999, Roger Kornberg worked out the highly complex structure of RNA polymerase using high intensity X-rays from synchrotrons.

  Cryo-electron microscopy came next, freezing protein samples rather than crystallising them and using beams of electrons instead of X-rays. It damages the sample less, which lets scientists read larger structures and gather more information. A protein has up to four named levels of structure, and biochemists describe a fifth. Primary structure is the amino acid sequence. Secondary structure is the local repeating shapes held by hydrogen bonds, such as the alpha-helix, the beta-sheet, and turns. Tertiary structure is the overall shape of one molecule, often stabilised by a hydrophobic core, and it controls the protein's basic function. Quaternary structure is several chains acting as one complex. Quinary structure is the surface signature that organises the crowded interior of a living cell. The Protein Data Bank, the public archive of solved structures, holds 181,018 X-ray, 19,809 EM, and 12,697 NMR protein structures.

• 4. The Grip Of Binding

  ›

  The ribonuclease inhibitor protein binds to human angiogenin with a sub-femtomolar dissociation constant, below 10 to the minus 15 molar. Yet it does not bind at all to angiogenin's amphibian relative, onconase, where the constant runs above 1 molar. That gap shows the trait that gives proteins their range: the ability to bind other molecules tightly and specifically. The region that does the binding is often a pocket or depression on the molecular surface, shaped by the tertiary structure and the side chains around it.

  Even a tiny chemical change can switch binding off. Adding a single methyl group to a partner molecule can nearly eliminate the interaction. The aminoacyl tRNA synthetase that handles the amino acid valine manages to discriminate against the very similar side chain of isoleucine. Proteins also bind copies of themselves, oligomerising into fibrils, a process common in structural proteins where globular monomers self-associate into rigid fibers.

  Protein-protein interactions regulate enzyme activity, steer progress through the cell cycle, and assemble large complexes that run many related reactions at once. Because these interactions are reversible and depend on which partners are available, studying them is a key to understanding what makes one cell type differ from another. The ability of a binding partner to force a conformational change is what allows enormously complex signaling networks to be built.

• 5. The Cell's Chief Actors

  ›

  Proteins make up half the dry weight of an Escherichia coli cell, while DNA and RNA account for only 3% and 20%. A typical bacterial cell such as E. coli or Staphylococcus aureus is estimated to hold about 2 million proteins. Smaller bacteria like Mycoplasma or spirochetes carry fewer, on the order of 50,000 to 1 million. Yeast cells hold roughly 50 million protein molecules, and human cells on the order of 1 to 3 billion.

  The best-known job of a protein is as an enzyme, and about 4,000 reactions are known to be catalysed by enzymes. The speed-up can be staggering. Orotate decarboxylase delivers as much as a 10 to the 17th-fold increase in rate, turning a reaction that would take 78 million years without the enzyme into one that finishes in 18 milliseconds. Although an enzyme may run to hundreds of amino acids, usually only three to four residues are directly involved in catalysis, gathered at the active site.

  Not every protein-coding gene is switched on in a given cell. Of the roughly 20,000 proteins encoded by the human genome, only 6,000 are detected in lymphoblastoid cells. The most abundant protein in nature is thought to be RuBisCO, the enzyme that pulls carbon dioxide into organic matter during photosynthesis. Plants can be as much as 1% RuBisCO by weight.

• 6. Assembling A Chain

  ›

  AUG, the sequence adenine-uracil-guanine, is the code for the amino acid methionine. The genetic code is built from three-nucleotide sets called codons, and because DNA uses four nucleotides, there are 64 possible codons. That leaves redundancy, with some amino acids spelled by more than one codon. In general the code specifies 20 standard amino acids, though some organisms add selenocysteine, and certain archaea add pyrrolysine.

  Translation is the process of building a protein from an mRNA template. The mRNA is loaded onto the ribosome and read three nucleotides at a time, each codon matched to the anticodon on a transfer RNA that carries the right amino acid. The enzyme aminoacyl tRNA synthetase charges each tRNA with its correct amino acid. Proteins are always biosynthesized from the N-terminus to the C-terminus, and synthesis can reach up to 20 amino acids per second, faster in prokaryotes than in eukaryotes.

  Size is measured in amino acids and in mass, reported in daltons or kilodaltons. The average protein grows from Archaea to Bacteria to Eukaryote: 283, 311, and 438 residues, at 31, 34, and 49 kilodaltons. Yeast proteins average 466 amino acids and 53 kilodaltons. The largest known proteins are the titins, part of the muscle sarcomere, at a molecular mass of almost 3,000 and a length of almost 27,000 amino acids. Short proteins can also be built in the lab by chemical synthesis, though that route is inefficient beyond about 300 amino acids and usually runs C-terminus to N-terminus, opposite the biological direction.

• 7. Structure As Strength

  ›

  Keratin sits at 1.5 to 10 gigapascals of stiffness, and collagen at 5 to 7.5 gigapascals, several orders of magnitude stiffer than elastin at about 1 megapascal. Young's modulus, the axial stress over the resulting strain, measures this stiffness, and in proteins it often tracks directly with biological function. Collagen is the major component of connective tissue, bone, and cartilage. Keratin shows up in hair, nails, feathers, hooves, claws, and some animal shells. Elastin is thought to give elasticity to blood vessels, pulmonary tissue, and bladder tissue.

  Proteins fall informally into three classes that match their tertiary shapes: globular, fibrous, and membrane. Almost all globular proteins are soluble and many are enzymes, with bovine serum albumin floating in the cytosol at a far lower modulus, roughly 2.5 to 15 kilopascals. Beta-barrel outer membrane proteins, by contrast, run from 20 to 45 gigapascals. Some globular proteins still build structure: actin and tubulin are soluble as monomers but polymerise into the long, stiff fibers of the cytoskeleton that hold a cell's shape.

  Motor proteins such as myosin, kinesin, and dynein generate mechanical force. They drive the motility of single-celled organisms and the sperm of sexually reproducing animals, power contracting muscle, and carry cargo inside the cell. A single protein's stiffness can be calculated by stretching it in a molecular dynamics simulation using force-fields like CHARMM, GROMOS, or the coarse-grained Martini, or measured directly with atomic force microscopy.

• 8. Life And Death Of A Protein

  ›

  An average protein in a mammalian cell lives 1 to 2 days, though lifespans range from minutes to years. Once a protein has served its time it is degraded and recycled through protein turnover, its lifespan measured as a half-life. Abnormal or misfolded proteins are broken down faster, often by the proteasome, which is itself a large protein assembly. A protein is marked for destruction when ubiquitin ligases tag it, or simply because it has become unstable or damaged.

  Digestion runs the same teardown in the gut. Proteolysis, the breakdown of proteins into small peptides and amino acids, depends on enzymes called proteases or peptidases, which are themselves proteins. Pepsin is an endopeptidase that works in the stomach. After the stomach, the pancreas secretes trypsin and chymotrypsin to finish the job, and the products are absorbed in the small intestine. In animals, the diet must supply the essential amino acids that the body cannot make for itself.

  The same chemistry is run on an industrial scale. Protein hydrolysis is used commercially to produce amino acids from bulk sources such as blood meal, feathers, and keratin, treated with hot hydrochloric acid to break the peptide bonds. The same nitrogen that fills those amino groups is what the total Kjeldahl nitrogen test measures across wastewater, soil, food, and feed, naming proteins by the element they carry in abundance.