Lesson 19: Medicinal Chemistry — How Drugs Are Designed

A medicine is a molecule with a job interview: it must bind its target strongly, reach that target through a hostile body, and leave the rest of the patient alone. Medicinal chemistry is the discipline of designing molecules that pass all three tests, balancing potency against solubility, stability and safety. This lesson follows its flagship example — aspirin — from synthesis to the dose arithmetic that keeps benefit ahead of harm.

How drugs work: targets and binding

Most drugs bind a protein receptor or enzyme active site (Lesson 16), held by the same weak interactions as substrates — hydrogen bonds, ionic attractions, hydrophobic packing — but binding far more tightly or blocking the site persistently. Aspirin goes further: it covalently acetylates a serine residue in cyclooxygenase (COX) enzymes, permanently disabling that enzyme molecule; the body must synthesise fresh COX, which is why one small dose suppresses clotting for days. Selectivity is the eternal problem: COX-1 protects the stomach lining while COX-2 drives inflammation and pain, so a drug hitting both trades relief against gastric side-effects — the trade-off that launched the (troubled) selective COX-2 inhibitor class. Chirality (Lesson 9) pervades pharmacology: binding sites are chiral, so enantiomers commonly differ in potency or effect, and regulators now require each to be characterised.

Dose arithmetic: ED50, LD50 and therapeutic index

Dose–response curves are quantified by the ED50 (dose effective in 50% of the population) and LD50 (dose lethal in 50% of test animals). Their ratio, the therapeutic index TI = LD50 ÷ ED50, measures the safety margin: penicillin's TI is enormous (hundreds), while digoxin's is about 2 — a drug dosed with blood tests and respect.

Worked example 1: therapeutic index and a patient dose

A drug has ED50 = 40 mg/kg and LD50 = 400 mg/kg. TI = 400 ÷ 40 = 10 — the effective dose must rise tenfold before half the test population dies; workable, but demanding respect in overdose. For a 70 kg adult at a prescribed 5 mg/kg, the dose is 5 × 70 = 350 mg. Note the units discipline: doses scale by body mass (mg/kg) because distribution volume scales with size — the same reason children's doses are not just "half a tablet" guesswork but mg/kg calculations pharmacists verify. Half-life completes the picture: a drug with t½ = 5 h falls to (½)³ = 12.5% of peak after 15 h (Lesson 13's mathematics), which is why dosing intervals are quoted in half-lives.

Aspirin: synthesis and analysis

Aspirin (acetylsalicylic acid) is made by esterifying salicylic acid with ethanoic anhydride, acid-catalysed: C₇H₆O₃ + C₄H₆O₃ → C₉H₈O₄ + C₂H₄O₂. Salicylic acid (Mr 138.12) limiting gives aspirin (Mr 180.16).

Worked example 2: yield and purity of an aspirin preparation

A student reacts 2.00 g of salicylic acid and obtains 1.85 g of crude aspirin. Moles of salicylic acid = 2.00 ÷ 138.12 = 0.01448 mol, so theoretical aspirin = 0.01448 × 180.16 = 2.61 g. Percent yield = 1.85 ÷ 2.61 × 100 = 70.9%. But the balance cannot tell aspirin from unreacted salicylic acid: purity is interrogated chemically. Iron(III) chloride gives an intense violet colour with the free phenol group of salicylic acid but not with aspirin's blocked phenol — a sensitive spot test for starting-material contamination. Melting point corroborates: pure aspirin melts sharply at 138–140 °C; impure samples melt lower and across a range, the analytical logic of Lesson 17 applied to identity. If the FeCl₃ test is positive, the 70.9% "yield" overstates the true aspirin yield — yield and purity are separate axes, and pharmaceutical quality control certifies both.

In the lab: titrating the product

Aspirin's content in tablets is assayed by back titration: excess NaOH hydrolyses the ester, and the leftover NaOH is titrated with standard HCl. The measurement uncertainty is two titrations deep, so errors compound: burette readings of ±0.05 mL on two titres of ~20 mL give roughly ±0.5% overall — provided the end points behave. They often do not: aspirin hydrolyses slowly, so a rushed analyst titrates unreacted NaOH before hydrolysis completes and over-reports leftover base (under-reporting aspirin). Warming and a standing time are prescribed not from fussiness but because the reaction's kinetics (Lesson 12) set the timetable. Tablet fillers add a further systematic wrinkle — some bind aspirin or absorb indicator — so pharmacopoeias validate the whole method against certified reference tablets: the analytical chemist's version of Lesson 17's calibration discipline.

Chemistry in the real world

Prodrugs exploit the body's chemistry: codeine is converted by a liver enzyme (CYP2D6) into morphine, its active form — and patients with low CYP2D6 activity get little relief, while ultra-rapid metabolisers have died from standard doses. Genetic variation in drug metabolism is now printed on drug labels: medicinal chemistry meeting Lesson 16's biochemistry in the clinic.

Common mistakes

• "LD50 measures safety in humans directly." LD50 comes from animal studies; it anchors comparisons, and human dosing is set from clinical trials with wide margins.
• Yield reported without purity. A 95% yield of 80%-pure material is a 76% yield of drug (the Lesson 15 multiplication trap in reverse).
• Assuming both enantiomers are equivalent. Chiral targets distinguish them; always ask which enantiomer the data describes.
• Half-life used as "time until gone." Decay is exponential (12.5% remains after three half-lives); "gone" is a detection-limit statement, not a time.

Exam Experience

1. The therapeutic index of a drug is defined as: A) ED50 ÷ LD50 B) LD50 ÷ ED50 C) the maximum safe dose D) the half-life in the body Answer: B. TI = LD50 ÷ ED50; larger means a wider safety margin. A inverts it (and would shrink as drugs get safer — a definitional red flag).

2. A drug has a half-life of 4 hours. What fraction of a dose remains in the body after 12 hours? A) 1/3 B) 1/4 C) 1/8 D) 1/16 Answer: C. 12 ÷ 4 = 3 half-lives; (½)³ = 1/8.

3. In the aspirin synthesis, the role of concentrated sulfuric acid is: A) a reactant providing the acetyl group B) a catalyst for the esterification C) the solvent for crystallisation D) an oxidising agent Answer: B. The acetyl group comes from ethanoic anhydride; H₂SO₄ catalyses the esterification (and the reaction is warmed to a workable rate, Lesson 12 logic).

Constructed response (5 marks). A new analgesic has ED50 = 12 mg/kg and LD50 = 180 mg/kg. (a) Calculate its therapeutic index. (b) Calculate the dose for a 60 kg patient at the ED50 level, in mg. (c) A second drug has TI = 3. Explain, in terms of the dose–response margin, why the second drug requires stricter dose control in the clinic even if its ED50 is lower (i.e., it is more potent). Scoring guide: (a) 1 mark: 180 ÷ 12 = 15. (b) 1 mark: 12 × 60 = 720 mg. (c) 1 mark: TI 3 means the lethal dose is only 3× the effective dose (a narrow margin); 1 mark: small errors in dose, body mass or patient sensitivity can bridge a 3× gap but not a 15× gap; 1 mark: potency (low ED50) is about how *much* is needed, not safety — a potent drug can still have a narrow margin. Accept equivalent margin-based reasoning; reject "TI 3 is more toxic because ED50 is lower" (confuses potency with margin).

Pharmacokinetics: the journey is half the design

A potent molecule that never reaches its target is a failed drug, so medicinal chemists design for the journey (summarised as ADME: absorption, distribution, metabolism, excretion). Lipinski's rules encode the survival statistics of oral drugs: Mr ≤ 500, no more than 5 hydrogen-bond donors, no more than 10 acceptors, and a fat/water partition (log P) below 5 — molecules outside these bounds usually absorb poorly from the gut. Aspirin (Mr 180, log P ≈ 1.2) passes everything and is absorbed partly *because* the stomach's acidity keeps it un-ionised and membrane-permeable — weak-acid chemistry from Lesson 6 deciding a drug's fate. (Ionised species cross lipid membranes badly; a weak acid is uncharged in acid, charged in blood, and effectively trapped where it is absorbed — ion trapping, used deliberately in overdose treatments that alkalinise urine to trap aspirin for excretion.)

Worked example 3: keeping a drug in its window

A drug has a half-life of 6 hours and a therapeutic window between 2 mg/L and 8 mg/L in blood. A dose produces a peak of 8 mg/L. After one half-life the level is 4 mg/L (in window); after two, 2 mg/L (the window's floor). The dosing interval is therefore about 12 hours — dose again just as the level leaves the window. Double the dose to stretch the interval and the peak becomes 16 mg/L, above the safety ceiling: exponential decay (Lesson 13) makes dosing intervals logarithmically stubborn, which is why "take twice as much, half as often" is dangerous advice for narrow-index drugs (Lesson 19's TI) and harmless for wide ones. Sustained-release formulations exist to flatten exactly this sawtooth, releasing drug at the rate the body clears it.

Clinical trials: where the numbers are earned

Before any drug of this lesson reaches a pharmacy it survives a statistical obstacle course: Phase I (tens of healthy volunteers; safety and dose), Phase II (hundreds of patients; does it work, at what dose), Phase III (thousands; randomised, double-blind comparison against standard treatment or placebo). Randomisation and blinding exist because chemistry alone cannot exclude hope: placebo groups routinely improve by 20–30% on subjective measures, so an unblinded trial measures belief as much as binding. Roughly nine of ten candidates entering Phase I never reach approval, and the survivors carry their ED50, LD50 and trial statistics into the regulator's dossier — the figures of Worked example 1 are the distilled summary of that decade-long process. When a headline announces a "miracle drug" from a Phase II result, the trained reader of this lesson asks for the phase, the blinding and the margin.