Biology › Section H · Gene expression and its regulation › Topic H.1

The genetic code

Four nucleotides must spell twenty amino acids - and life settled on three-letter words. The central dogma traces the flow from DNA through mRNA to protein, and the triplet code that makes it work turns out to be degenerate, nearly universal, and beautifully armoured against typos.

The Central Dogma

Genes specify proteins through a two-step relay described by the Central Dogma: information flows from DNA to RNA to protein. In transcription, a gene is copied into messenger RNA (mRNA), a mobile replica of the instructions. In translation, the mRNA directs the synthesis of a protein. Two other RNA types assist: transfer RNA (tRNA) ferries amino acids to the ribosome, and ribosomal RNA (rRNA) builds the ribosome's catalytic core. Why the middleman? DNA is the cell's irreplaceable archive - keeping it intact and protected while disposable mRNA copies do the traveling is safer than reading the original over and over.

Three alphabets

The genetic code spans three alphabets. DNA writes with four letters (A, T, C, G); RNA copies use the same letters with uracil replacing thymine (A, U, C, G); and proteins are written in twenty amino acids. The twenty differ chemically - acidic or basic, polar or nonpolar, large or small - because each carries a distinctive side chain. Stringing different side chains in different orders is what gives proteins their enormous variety of shapes and functions, from the interleukins of the immune system to the globins that carry oxygen.

Why three-letter words

How many RNA letters spell one amino acid? One is obviously too few - four single nucleotides cannot name twenty amino acids. Doublets fail too: four letters taken two at a time give only 16 combinations, short of twenty. Triplets give 4 to the third power - 64 combinations - more than enough. Scientists theorized the code was triplet and degenerate (several codons naming the same amino acid), and experiments confirmed it. The code is also colinear: mRNA nucleotides 1-3 spell amino acid 1, nucleotides 4-6 spell amino acid 2, and so on down the line.

Crick and Brenner's frameshift test

Francis Crick and Sydney Brenner proved the triplet structure with a chemical mutagen, proflavin, that inserts or deletes single nucleotides. Insert one nucleotide into a viral gene and protein synthesis collapses; insert two and it collapses too. But insert three nucleotides, and a functional protein appears - carrying one extra amino acid, but intact from there on. The logic: if the code reads in groups of three, a shift of one or two letters scrambles every codon after the insertion (a frameshift), wrecking the whole message. A shift of exactly three restores the frame, and only one amino acid changes. Three letters per word, demonstrated by sabotage.

Start and stop signals

Of the 64 codons, 61 specify amino acids and three stop protein synthesis - the nonsense or stop codons - releasing the finished polypeptide from the machinery. One codon does double duty: AUG specifies methionine and also serves as the start codon that initiates translation. The AUG near the mRNA's 5' end sets the reading frame, establishing where the three-letter groups begin. Read from the wrong starting point and the same nucleotides spell gibberish - which is why the frame-setting codon matters as much as the message itself.

Degenerate and universal

The code is degenerate: most amino acids answer to several codons, like synonyms in English. CCU, CCC, CCA, and CCG all specify proline; CAU calls histidine. And it is universal: with minor variations in mitochondria and a few microorganisms, virtually every species on Earth reads the same codons the same way. Purify the mRNA for a horse's globin protein and put it in a tulip cell, and the tulip will make horse globin. One shared code among the roughly 10 to the 84 possible codes is powerful evidence that every living thing descends from a common ancestor.

Degeneracy as armour

Degeneracy is not wasted redundancy - it is mutation insurance. Codons for the same amino acid usually differ in only their third letter, so a random substitution there often changes nothing at all: a CCA becomes a CCG and proline stays proline. Even better, codons for chemically similar amino acids tend to resemble each other, so a mutation that does change the amino acid often swaps in one with similar behavior. The result: single-nucleotide substitutions frequently produce either no effect or a harmless tweak, rather than a wrecked protein. Random errors still occur - the code just mutes most of them before natural selection ever sees the damage.

When the dogma bends

The Central Dogma is the standard flow, not a law of nature. HIV stores its genome in RNA - and on infecting a host cell, the viral enzyme reverse transcriptase copies that RNA into DNA, which then follows the usual path to mRNA and protein. Influenza skips DNA entirely: its RNA genome is copied by an RNA-dependent RNA polymerase, going RNA to RNA to protein. And mitochondria, along with a few microorganisms, use slightly modified codon assignments. The dogma describes the cell's default; viruses and organelles show where the arrows can be redrawn.

Worked example: from DNA to peptide

A DNA template strand reads 3' GCT GTC AAA TTC GAT 5'. Transcribe it: RNA polymerase builds the antiparallel complement, replacing thymine with uracil - mRNA 5' CGA CAG GUU AAG CUA 3'. Now translate, codon by codon: CGA is arginine, CAG is glutamine, GUU is valine, AAG is lysine, and CUA is leucine. The peptide: Arg-Gln-Val-Lys-Leu. Notice the bookkeeping: transcription is one-for-one (each nucleotide read becomes one nucleotide of mRNA), while translation is three-for-one (each codon becomes one amino acid) - and the frame starts from the 5' end, so any shift would scramble every codon after it.

Worked example: predicting frameshift damage

Suppose proflavin inserts one nucleotide after the first codon of that same mRNA. The reading frame shifts for everything downstream: what was CAG GUU AAG CUA becomes a wholly different string of codons, and the protein is almost certainly nonfunctional - or stops prematurely if a new codon happens to be a stop. Now insert three nucleotides at the same spot instead. One extra amino acid appears at that position, but the frame is preserved and the rest of the chain reads normally. This is Crick and Brenner's result as a prediction: damage scales with the shift modulo three. A genetic counselor would put it this way - single deletions or insertions are catastrophic; multiples of three are comparatively gentle.

Common mistakes to avoid

First, calling the code ambiguous - degenerate is not ambiguous: many codons per amino acid, but never one codon per several amino acids. Second, forgetting uracil - the mRNA alphabet swaps T for U. Third, treating AUG as only a start signal - it also names methionine. Fourth, miscounting the codon budget: 64 total, 61 for amino acids, 3 stops. Fifth, confusing transcription with translation - copying DNA to RNA versus decoding RNA to protein. Sixth, assuming frameshifts merely change one amino acid - they scramble every codon after the shift. Seventh, overstating the dogma - reverse transcriptase and RNA viruses legitimately reverse the arrows.

Key takeaways

The Central Dogma traces information from DNA through mRNA to protein via transcription and translation, with tRNA and rRNA supporting the second step. Three alphabets meet: DNA's ATCG, RNA's AUCG, and 20 amino acids whose side chains drive protein variety. Doublets (16) cannot cover 20 amino acids, but triplets (64) can - Crick and Brenner proved the triplet reading frame with one-, two-, and three-nucleotide insertions. AUG starts translation and names methionine; three stop codons end it. The code is degenerate (synonymous codons buffer point mutations) and universal - shared by nearly all life, evidence of common ancestry - with known exceptions: HIV's reverse transcriptase, influenza's RNA-dependent RNA polymerase, and mitochondrial codon variations.

Flashcards

Flashcard

Central Dogma

Information flows DNA to RNA to protein

Transcription makes mRNA from a gene; translation turns that mRNA into a polypeptide chain.

Test your understanding

Question 1 of 5

Why can a two-nucleotide code not specify all 20 amino acids?

Question 2 of 5

In the standard genetic code, what are the TWO roles of the codon AUG?

Question 3 of 5

In Crick and Brenner's proflavin experiment, inserting THREE nucleotides into a gene produced a functional protein. Why?

Question 4 of 5

How does degeneracy of the genetic code protect cells against mutations?

Question 5 of 5

HIV stores its genome in RNA. Which enzyme lets it insert its genes into the DNA-based flow of the host cell?

← Previous topicNext topic →