HookThe anticancer drug that was found by accident
In 1965 the biophysicist Barnett Rosenberg was not looking for a drug at all. He wanted to know whether an electric field changed the way bacteria grow, so he ran a current through a colony of E. coli using what he thought were inert platinum electrodes. The bacteria stopped dividing — they grew into long filaments up to 300 times their normal length but could not split in two. The electric field turned out to be irrelevant. The platinum electrodes had quietly leaked a tiny amount of a platinum compound into the solution, and that compound was blocking cell division. Within a decade that molecule, cisplatin, was in hospitals curing testicular cancer at rates above 90%.
Everything in this section is the chemistry that story runs through. Cells divide by copying DNA; DNA is read to build proteins; proteins are chains of amino acids folded into precise shapes; and the most important of those shapes are the enzymes that catalyse life. Cisplatin works by jamming one link in that chain — it grabs hold of DNA so it can no longer be copied. To understand why a small square-planar platinum complex can stop a tumour, you first have to understand the molecules it sabotages. So this section climbs the ladder from a single amino acid up to the double helix, and then shows exactly where the drug bites.
ModelAmino acids as zwitterions — one molecule, two personalities
An amino acid carries two opposing groups on the same molecule: a basic amine, \(-\text{NH}_2\), and an acidic carboxyl, \(-\text{COOH}\). The ones that build proteins are 2-aminocarboxylic acids (α-amino acids) with the general formula \(\text{RCH(NH}_2)\text{COOH}\), where R is the side chain that makes each of the twenty amino acids different. Because an acid and a base sit on the same molecule, they react with each other internally: the \(-\text{COOH}\) donates its proton to the \(-\text{NH}_2\). The result is a zwitterion — a species with a positive \(-\text{NH}_3^+\) and a negative \(-\text{COO}^-\) but no overall charge.
That internal ionisation explains the physical properties AQA expects you to justify. Amino acids are crystalline solids with surprisingly high melting points for small molecules, because the zwitterions attract one another through strong ionic forces, and they dissolve readily in water because those charges hydrogen-bond to it. It also means an amino acid behaves differently depending on the pH of its surroundings. In a strongly acidic solution the extra \(\text{H}^+\) protonates the carboxylate, giving a positively charged cation, \(\text{RCH(NH}_3^+)\text{COOH}\). In a strongly alkaline solution the \(\text{OH}^-\) removes a proton from the \(-\text{NH}_3^+\), giving a negatively charged anion, \(\text{RCH(NH}_2)\text{COO}^-\).
The pH at which the molecule exists purely as the neutral zwitterion, with no net charge, is its isoelectric point. There is one more property to flag: every amino acid except glycine (where R is just H) has four different groups on its central carbon, so it is chiral and exists as two optical isomers — a fact that becomes the whole point when we reach enzymes.
A student dissolves alanine, \(\text{CH}_3\text{CH(NH}_2)\text{COOH}\), in three buffers — pH 1, pH 6 and pH 11 — and applies a voltage across each. Predict the structure and the direction of migration in each.
At pH 1 the solution is flooded with \(\text{H}^+\), so the carboxylate is protonated and the molecule is the cation \(\text{CH}_3\text{CH(NH}_3^+)\text{COOH}\), net charge +1. A positive ion is attracted to the negative electrode, so it migrates to the cathode.
At pH 6 (close to alanine's isoelectric point) it is the neutral zwitterion \(\text{CH}_3\text{CH(NH}_3^+)\text{COO}^-\), net charge 0, so it does not migrate to either electrode.
At pH 11 the \(\text{OH}^-\) deprotonates the \(-\text{NH}_3^+\), giving the anion \(\text{CH}_3\text{CH(NH}_2)\text{COO}^-\), net charge \(-1\), which migrates to the positive electrode, the anode. The rule to carry into the exam: low pH pushes an amino acid positive, high pH pushes it negative, and the isoelectric point is the crossover.
ModelProteins — a peptide bond, then three levels of folding
Amino acids join by condensation: the \(-\text{COOH}\) of one reacts with the \(-\text{NH}_2\) of the next, a molecule of water leaves, and an amide link, \(-\text{CONH}-\), forms. In a protein this amide is called a peptide bond, so a protein is a condensation polymer of amino acids — a polyamide, chemically the same class of link you met in nylon and in Kevlar. The reverse reaction, hydrolysis, breaks a protein back into its amino acids: refluxing it with 6 mol dm⁻³ hydrochloric acid for around 24 hours splits every peptide bond, and the mixture of amino acids can then be separated and identified by thin-layer chromatography.
AQA builds protein shape in three named layers, and the exam rewards keeping them straight. The primary structure is simply the sequence — the order in which the amino acids are joined along the chain. The secondary structure is the local folding held together by hydrogen bonds between the \(\text{C=O}\) and \(\text{N–H}\) groups of the peptide backbone; these regular patterns are the α-helix (a coil) and the β-pleated sheet.
The tertiary structure is the overall three-dimensional shape of the whole chain, and it is held by interactions between the R side chains, not the backbone. Four types are examinable: further hydrogen bonds, ionic interactions between charged R groups (for example a \(-\text{NH}_3^+\) attracting a \(-\text{COO}^-\)), disulfide bridges (strong covalent \(\text{S–S}\) bonds formed between two cysteine side chains), and hydrophobic interactions where non-polar R groups cluster away from water. This tertiary shape is not decoration — it is the whole reason a protein can do a job, because it creates the precise pocket an enzyme uses to grip its substrate.
MechanismEnzymes — why shape makes them fussy about mirror images
An enzyme is a protein that acts as a biological catalyst, and it works because its tertiary structure folds into an active site — a cavity with a shape and a pattern of charges that fits one particular molecule, the substrate. The substrate binds, the reaction happens with a lowered activation energy, and the product leaves. Because the active site is a specific 3-D shape, only a molecule with the exactly complementary shape can bind: this is why enzymes are described as stereospecific.
The consequence that AQA leans on is stereochemical. Most biological molecules are chiral, existing as two non-superimposable mirror images (optical isomers). An active site is itself chiral, so it typically binds only one of the two enantiomers — the other simply will not fit, in the same way a right hand does not fit a left glove. This is why living things make and use almost exclusively one enantiomer of each amino acid and sugar, and why the two mirror-image forms of a drug can behave completely differently in the body.
That last point is not abstract. Drug molecules are often designed to bind an enzyme's active site — either as the intended key or as a deliberate blocker that stops the natural substrate binding. If a drug is chiral, only one enantiomer may have the therapeutic effect while its mirror image is useless or harmful, which is exactly why the pharmaceutical industry spends heavily to make single-enantiomer drugs rather than 50:50 mixtures.
ModelDNA — a four-letter code held by hydrogen bonds
DNA (deoxyribonucleic acid) is a condensation polymer of nucleotides. Each nucleotide is built from three parts: a phosphate group, the sugar 2-deoxyribose, and one of four nitrogen-containing bases — adenine (A), thymine (T), cytosine (C) or guanine (G). Nucleotides condense together — the phosphate of one bonding to the sugar of the next, losing water each time — to build a sugar–phosphate backbone with the bases sticking out sideways.
What turns two strands into the famous double helix is complementary base pairing. The bases pair up through hydrogen bonds, but only in fixed partnerships: adenine always pairs with thymine through two hydrogen bonds, and cytosine always pairs with guanine through three. The pairing is dictated by which bases can line up their \(\text{N–H}\), \(\text{C=O}\) and \(\text{N}\) groups to hydrogen-bond cleanly; A cannot make a tidy set of hydrogen bonds with C or G, so it never pairs with them.
Because the pairing is fixed, the two strands are not identical — they are complementary. Wherever one strand reads A the other must read T; wherever one reads G the other must read C. This is the chemical basis of heredity: unzip the two strands and each acts as a template to rebuild its partner, copying the code exactly. It is also the feature cisplatin exploits, because anything that clamps the strands together stops that unzipping — and therefore stops the cell copying its DNA.
A sample of double-stranded DNA is analysed and found to contain 28% adenine (by number of bases). Deduce the percentage of each of the other three bases.
Start from the pairing rules. Adenine pairs only with thymine, so there must be as much thymine as adenine: %T = %A = 28%.
Together A and T account for \(28 + 28 = 56\%\) of the bases, so cytosine and guanine make up the remaining \(100 - 56 = 44\%\).
Cytosine pairs only with guanine, so those two are present in equal amounts: each is \(44 \div 2 = 22\%\). The answer is T 28%, C 22%, G 22%. The reasoning — that %A = %T and %C = %G in any double-stranded DNA — is a real experimental result (Chargaff's rule) and a favourite AQA way to test whether you have understood that base pairing is one-to-one and fixed.
CaseCisplatin — how a platinum complex jams the double helix
Cisplatin is a complex of platinum(II) with a square-planar shape, carrying two ammonia molecules and two chloride ions as ligands, \(\text{Pt(NH}_3)_2\text{Cl}_2\). The cis arrangement — the two chlorides next to each other rather than opposite — is essential; the trans isomer is far less effective as a drug, a neat payoff of the E–Z isomerism you learned earlier applied to a square-planar complex.
Inside the cell the two chloride ligands are lost and the platinum bonds instead to DNA. Specifically, a nitrogen atom on a guanine base donates a lone pair to the platinum, forming a coordinate (dative) bond, and because cisplatin has two available sites next to each other it binds two guanine bases at once. This cross-links the DNA and kinks it, so the strands can no longer separate and be copied. The cancer cell, unable to replicate its DNA, cannot divide, and it is triggered to die. Because cancer cells divide rapidly, they are hit hardest.
The honest complication — and the part AQA frames as a discussion of benefits and risks — is that cisplatin cannot tell a cancer cell from a healthy one. It binds the DNA of any dividing cell, so the worst side effects fall on the body's other fast-dividing tissues: bone marrow (suppressing the immune system) and hair follicles (causing hair loss). These harms are reduced by giving carefully controlled doses and by targeting the drug as precisely as possible to the tumour, so that the benefit — a very high cure rate for cancers such as testicular cancer — outweighs the damage. Society, not chemistry alone, decides that trade-off.
VocabularyKey terms the mark scheme pays for
TrapsMisconceptions that cost marks
ExamWhat examiners want
Draw the ionised structure that matches the pH the question gives. At low pH show the cation with –NH₃⁺ and –COOH; at high pH show the anion with –NH₂ and –COO⁻; at neutral or physiological pH show the zwitterion. A question that says \"in acidic solution\" is testing whether you protonate the carboxylate, and marks are lost by candidates who always draw the neutral form.
Name the bonds precisely when describing protein structure. Secondary structure is hydrogen bonding between C=O and N–H of the backbone; tertiary structure adds interactions between R groups — and if you are asked for the strongest, that is the covalent disulfide bridge between cysteine residues. Keep primary (sequence), secondary (local H-bonded coils and sheets) and tertiary (whole 3-D shape) clearly separated.
For enzymes and drugs, tie stereospecificity back to shape: the active site is a specific 3-D shape, so only a complementary molecule — often just one enantiomer — can bind. For cisplatin, a full-mark answer states that it bonds to DNA via a nitrogen on guanine, cross-links and prevents the strands separating so the cell cannot replicate, and then addresses the risk–benefit balance (side effects on healthy dividing cells versus a high cure rate). In base-ratio calculations, always start from %A = %T and %C = %G.