How Drugs Are Designed: Receptors, Binding and Shape

A modern medicine is a molecule that has been shaped, atom by atom, to do one job inside the body and as little else as possible. Getting there typically takes more than ten years and involves thousands of compounds being made and tested for each one that reaches a pharmacy shelf. The chemistry behind that process — medicinal chemistry — draws on almost everything you learn about bonding, intermolecular forces, acids and bases, and isomerism. This article explains the main ideas.

Step 1: choose a target

Most drugs work by binding to a protein and changing what it does. The protein is called the target. Common targets are:

• Enzymes, which the drug blocks (an inhibitor). Examples: aspirin and ibuprofen block cyclooxygenase enzymes; statins block HMG-CoA reductase, a key enzyme in making cholesterol.
• Receptors, proteins that receive signals from hormones or neurotransmitters. A drug can mimic the natural signal (an agonist) or block it (an antagonist). Salbutamol, used in asthma inhalers, is an agonist at β₂-adrenergic receptors; beta-blockers such as propranolol are antagonists at β receptors.
• Ion channels and transporters, which control what crosses cell membranes. Many antidepressants block the transporter that recycles serotonin.
• Microbial proteins, which differ from ours. Penicillin attacks an enzyme bacteria use to build their cell walls — an enzyme human cells don’t have.

A good target is one that plays a clear role in the disease and can be affected without causing too much harm elsewhere.

Step 2: understand the binding site

Proteins fold into precise 3D shapes (see protein folding). On the surface, or in a cleft, is a binding site: a pocket lined with amino-acid side chains. A drug works by fitting into this pocket and sticking there long enough to have an effect.

“Sticking” means forming non-covalent interactions — the same forces you meet in intermolecular bonding:

Interaction	Typical example in a binding site
Hydrogen bonds	Drug –OH or –NH donating to a backbone C=O
Ionic interactions (salt bridges)	Protonated amine on drug with aspartate –COO⁻
Hydrophobic interactions	Benzene ring of drug in a pocket of leucine and phenylalanine side chains
π-stacking	Aromatic ring of drug stacking on tryptophan or tyrosine
Van der Waals forces	Close contact across the whole surface

Each interaction is weak on its own, but a drug that makes many of them in the right places binds tightly and selectively. This is the same logic as enzyme–substrate binding: shape and chemistry together decide the fit.

A few drugs bind covalently. Aspirin transfers an acetyl group to a serine in its target enzyme, blocking it permanently; penicillin’s strained four-membered β-lactam ring opens and bonds to its bacterial enzyme. Covalent drugs can be very effective but need careful design to avoid reacting with the wrong proteins.

Binding strength: why ΔG matters

Chemists describe how tightly a drug binds with a dissociation constant, K_d. A smaller K_d means tighter binding. Most successful drugs bind with K_d values in the nanomolar range (10⁻⁹ mol dm⁻³) or better.

Binding is governed by the Gibbs energy change, ΔG = −RT ln K. At 37 °C, each tenfold improvement in binding corresponds to about 5.9 kJ mol⁻¹ more favourable ΔG. A single well-placed hydrogen bond can be worth a few kJ mol⁻¹, so small changes to a molecule can shift potency by orders of magnitude.

Both enthalpy and entropy count. Forming good contacts releases heat (favourable ΔH), but locking a floppy molecule into one shape costs entropy. That’s why chemists often make drugs more rigid — for example by closing a chain into a ring — so less entropy is lost on binding.

Shape and chirality

Because binding sites are built from L-amino acids, they are chiral. Two mirror-image forms (enantiomers) of a drug can therefore behave very differently:

• Salbutamol: the (R)-enantiomer is responsible for the bronchodilating effect.
• Ibuprofen: the (S)-form is the active one, although the body converts some of the (R)-form into (S).
• Thalidomide: the tragic case. It was sold in the late 1950s and early 1960s as a sedative and was given to pregnant women for morning sickness, causing severe birth defects in thousands of children. One enantiomer is associated with the harmful effect, but the two forms interconvert in the body, so giving a single enantiomer wouldn’t have made it safe.

Regulators now require that the activity and safety of each enantiomer be studied, and many modern drugs are sold as single enantiomers. If you need a refresher on why mirror images can’t be superimposed, read about amino acids, whose chirality underlies all of this.

Step 3: find a starting point

Chemists rarely design a drug from nothing. Starting points, called hits and then leads, come from:

• Natural products — compounds from plants, fungi and microbes. Aspirin traces back to salicylic acid from willow bark; morphine comes from the opium poppy; many antibiotics and cancer drugs come from microorganisms.
• Natural ligands — modifying the body’s own messenger. Many adrenaline-like drugs started from adrenaline itself.
• High-throughput screening — robots testing hundreds of thousands of compounds against the target.
• Structure-based design — using X-ray crystallography or cryo-electron microscopy to see the binding site, then designing molecules to fit it on a computer. HIV protease inhibitors in the 1990s were an early major success of this approach.
• Fragment-based design — finding tiny molecules that bind weakly in different parts of the site, then linking or growing them.

Step 4: structure–activity relationships

With a lead in hand, chemists make many close relatives — changing one group at a time — and measure how each change affects activity. This builds a structure–activity relationship (SAR). Typical questions:

• Does replacing –H with –CH₃ fill an empty hydrophobic pocket and improve binding?
• Does adding –OH create a new hydrogen bond, or does it just make the drug more water-soluble?
• Can a fluorine atom block a site where the liver would otherwise oxidise the drug?
• Does swapping a carboxylic acid for a similar acidic group (a bioisostere) keep activity but improve absorption?

Fluorine deserves a special mention. It’s small, highly electronegative (3.98 on the Pauling scale; see fluorine) and forms a very strong bond to carbon. Adding fluorine can make a drug last longer in the body and change its acidity and fat solubility. A large fraction of modern drugs contain at least one fluorine atom.

Step 5: getting the drug to the target (ADME)

A molecule that binds perfectly in a test tube is useless if it never reaches the target. Pharmacologists summarise the journey as ADME:

• Absorption: can it cross the gut wall into the blood?
• Distribution: does it reach the right tissue (and, for brain drugs, cross the blood–brain barrier)?
• Metabolism: how quickly does the liver change it, often by oxidation with cytochrome P450 enzymes?
• Excretion: how is it removed, usually through the kidneys?

For oral drugs, chemists often use Lipinski’s “rule of five” (from 1997) as a quick guide. A compound is more likely to be poorly absorbed if it breaks more than one of these:

• molar mass over 500 g mol⁻¹
• more than 5 hydrogen-bond donors
• more than 10 hydrogen-bond acceptors
• log P (a measure of fat vs water solubility) over 5

The rules reflect a balance: a drug must be fat-soluble enough to cross membranes made of phospholipids, yet water-soluble enough to dissolve in blood. Acid–base chemistry matters too — many drugs are weak acids or bases whose charge depends on the pH of the stomach, intestine or blood.

Step 6: safety and trials

Promising compounds are tested for toxicity in cells and animals, then enter clinical trials in three phases: small safety studies in volunteers, then studies in patients to find the right dose, then large trials comparing the drug with existing treatments. Only a small fraction of compounds entering trials are eventually approved.

Common misconceptions

• “A drug is a key that fits one lock.” Most drugs bind several proteins to some extent; side effects often come from these “off-target” interactions.
• “Tighter binding is always better.” Potency matters, but absorption, metabolism and safety often decide whether a drug succeeds.
• “Natural means safer.” Many potent poisons are natural products; many drugs are natural products refined for safety.
• “Mirror-image forms behave identically.” In a chiral body, they often don’t.

Key takeaways

• Drugs usually act by binding a protein target — enzyme, receptor, channel or transporter.
• Binding uses hydrogen bonds, ionic, hydrophobic and van der Waals interactions; shape and chirality decide the fit.
• Chemists optimise leads using structure–activity relationships, often adding rings, fluorine or bioisosteres.
• A good drug must also have sound ADME; Lipinski’s rule of five is a useful guide for oral drugs.
• The whole process balances potency, selectivity and safety.

Next, see pharmacology basics for chemistry students and how antibiotics work.