Difficulty: Intermediate | Prerequisites: Parts 1 to 3 of these notes (DNA structure, replication, RNA types, transcription)
This is the final piece of the central dogma: RNA → protein. By the early 1960s, scientists knew that DNA is transcribed into mRNA, but they still needed to crack the genetic code (which triplets of nucleotides specify which amino acids) and work out how the ribosome actually assembles a polypeptide. This section covers how the code was deciphered, the mechanics of translation (initiation, elongation, termination), and how the direction of protein synthesis was determined. Once you finish this, you have the complete picture: DNA → DNA (replication), DNA → RNA (transcription), RNA → protein (translation).
The genetic code is a triplet code (three bases = one codon) that is degenerate (multiple codons per amino acid) but not ambiguous (each codon specifies only one amino acid). Nirenberg, Matthaei, and Khorana cracked the code using synthetic polynucleotides. Translation starts at AUG, proceeds N-terminus to C-terminus, and terminates at one of three stop codons (UAG, UAA, UGA). Dintzis's pulse-labelling experiment on haemoglobin confirmed that polypeptides grow from N to C.
Codon
A triplet of three consecutive nucleotides in mRNA that specifies a particular amino acid (or a stop signal). In simple terms, codons are the "words" in the genetic language: a four-letter alphabet (A, U, G, C) read three letters at a time.
Degeneracy (of the genetic code)
The property that most amino acids are encoded by more than one codon. For example, leucine, serine, and arginine each have six codons. This is not the same as ambiguity; each codon still codes for only one amino acid.
Reading frame
The way a nucleotide sequence is partitioned into consecutive, non-overlapping triplets. Shifting the starting point by one or two bases changes every codon downstream. Think of it as where you start reading a sentence with no spaces: the wrong starting point garbles every word.
Frameshift mutation
An insertion or deletion of nucleotides (not in multiples of three) that shifts the reading frame, altering every codon downstream and typically destroying protein function.
Wobble position
The third position of a codon, where base pairing with the tRNA anticodon is less strict. Inosine (I) in the anticodon can pair with A, C, or U, which is why fewer tRNAs than codons are needed.
Shine-Dalgarno sequence
A purine-rich sequence (AGGAGGU) upstream of the AUG start codon in prokaryotic mRNA. It base-pairs with the 16S rRNA of the 30S ribosomal subunit to position the ribosome correctly for initiation.
Kozak sequence
The sequence context around the AUG start codon in eukaryotic mRNA, recognised by the 40S ribosomal subunit during scanning for the initiation site.
Release factors
Proteins that recognise stop codons (UAG, UAA, UGA) and trigger disassembly of the ribosome and release of the completed polypeptide. They contain a conserved GGQ motif essential for polypeptide release.
Collinearity
The principle that the linear order of nucleotides in a gene corresponds to the linear order of amino acids in its encoded protein. Demonstrated by Yanofsky and Brenner.
DNA has four bases, and proteins use 20 amino acids. How many bases per amino acid?
1 base at a time: 4^1 = 4 combinations (not enough)
2 bases at a time: 4^2 = 16 (still not enough)
3 bases at a time: 4^3 = 64 (more than enough)
George Gamow proposed the code must be at least triplet. With 64 codons for 20 amino acids + stop signals, the code is degenerate by necessity.
Introduced frameshift mutations into bacteriophage genes.
Inserting 1 or 2 bases disrupted the gene product entirely.
Inserting 3 bases restored function (the downstream reading frame is preserved, with only a small stretch of wrong amino acids).
Deleting bases followed the same pattern: multiples of three could be tolerated, other numbers could not.
This confirmed that the code is read in non-overlapping triplets from a fixed starting point.
Marshall Nirenberg and Heinrich Matthaei (1961) used a cell-free protein synthesis system with synthetic polynucleotides.
Key results:
Poly-U (UUUUUU...) → poly-phenylalanine. UUU = Phe was the first codon deciphered.
Poly-A → poly-lysine (AAA = Lys)
Poly-C → poly-proline (CCC = Pro)
Mixed polymers (e.g. poly-AC) produced proteins containing amino acids encoded by all eight possible triplet combinations of A and C (CCC, CCA, CAC, CAA, ACC, ACA, AAC, AAA), with proportions reflecting base ratios.
Nirenberg and Khorana then used defined-sequence polynucleotides (e.g. AGUAGU repeating) to assign specific codons more precisely.
By 1966, all 64 codons were assigned.
Degenerate but not ambiguous: most amino acids have multiple codons, but no codon encodes more than one amino acid.
Nearly universal: the same code is used from bacteria to humans, with minor exceptions:
UGA serves as both a stop codon and selenocysteine in some contexts.
Mitochondria have several variations (e.g. UGA = Trp instead of stop).
Some yeasts have ambiguous decoding of CUG (Ser, Ala, or Leu), possibly as a viral defence mechanism.
Second-position encodes chemical properties: the middle base of a codon correlates with the physicochemical nature of the amino acid (hydrophobic, polar, etc.).
A minimum of 31 tRNAs are needed (not 61) because wobble pairing at the third codon position allows one tRNA to read multiple codons.
Inosine (I) in the anticodon's wobble position can pair with A, C, or U.
Actual numbers: 30 to 50 tRNAs in bacteria, 50 to 100 in plants, at least 613 nuclear tRNA genes in humans (plus hundreds of pseudogenes and 22 mitochondrial tRNA genes).
Prokaryotes:
Start codons: AUG (83%), GUG (14%), UUG (3%)
The 16S rRNA of the 30S subunit binds the Shine-Dalgarno sequence (AGGAGGU), a purine-rich region upstream of AUG
Initiator tRNA carries N-formylmethionine (fMet), which is later removed by a deformylase or peptidase
Can have polycistronic mRNA (multiple genes on one transcript)
Eukaryotes:
Start codon: AUG only
The 40S subunit binds the 5' cap and scans to the first AUG in a favourable Kozak context
Initiator tRNA carries methionine (not formylated)
Many initiation factors are involved in both systems.
Stop codons: UAG (amber), UAA (ochre), UGA (opal)
These are not recognised by any tRNA. Instead, release factors bind the stop codon in the A site.
Release factors contain a conserved GGQ (glycine-glycine-glutamine) motif essential for catalysing peptide release.
The ribosome disassembles: mRNA, tRNA, and ribosomal subunits separate.
Multiple ribosomes can translate the same mRNA simultaneously, forming a polyribosome (polysome). This increases the rate of protein production from a single transcript.
Howard Dintzis used reticulocytes (immature red blood cells that synthesise haemoglobin from mRNA with no new transcription).
After protein synthesis was already underway, he added radiolabelled leucine for a brief pulse, shorter than the time needed to complete one full globin chain.
He then isolated only completed globin chains (removed incomplete fragments by gel electrophoresis).
Digested the completed chains with trypsin and measured radioactivity in each peptide fragment.
Result: radioactivity was lowest in peptides from the N-terminal region (synthesised before the label was added) and highest at the C-terminus (synthesised during the pulse).
Conclusion: polypeptide chains grow from N-terminus to C-terminus at approximately 2 amino acids per second.
Charles Yanofsky and Sydney Brenner studied tryptophan synthetase in E. coli.
Showed that successive nucleotides along a DNA chain correspond to successive amino acids along the polypeptide.
This confirmed the gene and its protein product are collinear.
A single E. coli cell contains 10,000 to 100,000 ribosomes.
Protein synthesis rate: approximately 20 amino acids per second per ribosome (~170,000 amino acids used per second across all ribosomes).
Codons: 4^3 = 64 total; 61 sense codons + 3 stop codons
Start: AUG (universally); also GUG (14%) and UUG (3%) in prokaryotes
Stop: UAG (amber), UAA (ochre), UGA (opal)
Minimum tRNAs needed: 31 (due to wobble)
Protein synthesis rate: ~2 amino acids/second (haemoglobin); ~20 amino acids/second (E. coli)
Haemoglobin: 146 amino acids in β-globin chain
Synthetase error rate reminder: ~1/10,000
Understanding the genetic code is essential for interpreting any DNA or protein sequence. In clinical genetics, identifying whether a mutation creates a missense (wrong amino acid), nonsense (premature stop), or frameshift (garbled downstream) change is the first step in predicting its effect on a patient. Codon degeneracy also matters in biotechnology: when designing synthetic genes for expression in a different organism, codon optimisation (choosing the codons preferred by that host's tRNA pool) can dramatically improve protein yield.
Students sometimes say the genetic code is "ambiguous" when they mean "degenerate." Degenerate means multiple codons per amino acid. Ambiguous would mean one codon encoding multiple amino acids, which is not the case (with very rare exceptions in certain yeasts).
The start codon AUG also encodes methionine internally. It does not always signal "start." It is the Shine-Dalgarno or Kozak context that identifies the initiator AUG.
UGA is often memorised only as a stop codon, but it also encodes selenocysteine in specific contexts (the 21st amino acid). Exam questions sometimes test this.
Protein synthesis direction (N → C) is not the same as mRNA reading direction (5' → 3'), though they proceed in parallel. Students sometimes conflate the two as if they are the same thing; they are related but describe different molecules.
⚠️ Be able to calculate the number of possible codons for a given number of bases per codon (4^n).
⚠️ Know the three stop codons and their names (amber, ochre, opal).
⚠️ Understand how Brenner and Crick's frameshift experiment proved a triplet code.
⚠️ Know the difference between Shine-Dalgarno (prokaryotic) and Kozak (eukaryotic) sequences for translation initiation.
⚠️ Dintzis's experiment and the logic of pulse-labelling to determine synthesis direction is a common exam question.
⚠️ Understand wobble pairing and why 31 tRNAs suffice for 61 sense codons.
True or False: The genetic code is both degenerate and ambiguous.
Fill in the blank: Brenner and Crick showed that inserting ______ bases into a gene restores the reading frame.
True or False: In prokaryotes, the Kozak sequence positions the ribosome at the start codon.
Fill in the blank: The three stop codons are UAG, UAA, and ______.
True or False: Dintzis showed that protein synthesis proceeds from the C-terminus to the N-terminus.
Answers: 1. False (degenerate, but not ambiguous). 2. Three. 3. False (that is the Shine-Dalgarno sequence in prokaryotes; the Kozak sequence is eukaryotic). 4. UGA. 5. False (N-terminus to C-terminus).
Q: How did Nirenberg and Matthaei determine that UUU codes for phenylalanine?
A: They added synthetic poly-U (UUUUUU...) to a cell-free protein synthesis system containing ribosomes, tRNAs, amino acids, and the necessary cofactors. The only protein produced was poly-phenylalanine. Since the only codon present was UUU, they concluded UUU encodes phenylalanine.
Q: Explain why inserting one or two nucleotides into a gene is more damaging than inserting three.
A: The genetic code is read in non-overlapping triplets from a fixed start point. Inserting one or two nucleotides shifts every downstream codon by one or two positions, producing entirely wrong amino acids for the rest of the protein (a frameshift). Inserting three nucleotides adds one extra amino acid but restores the correct reading frame for all subsequent codons, so only a small region of the protein is affected.
Q: What is the Shine-Dalgarno sequence, and how does it function in translation initiation?
A: The Shine-Dalgarno sequence is a purine-rich region (consensus AGGAGGU) located upstream of the AUG start codon in prokaryotic mRNA. It base-pairs with a complementary sequence on the 16S rRNA of the 30S ribosomal subunit, positioning the ribosome so that the AUG falls in the correct place for initiation. This is why prokaryotes can have polycistronic mRNA: each coding sequence has its own Shine-Dalgarno sequence.
Q: In Dintzis's experiment, why was radioactivity highest at the C-terminus of completed globin chains?
A: Radiolabelled amino acids were added partway through synthesis. Chains that were nearly finished at the time of labelling incorporated the label only in the last portion synthesised. Since protein grows from N to C, the C-terminal end is the last part made, so it picked up the most label. The N-terminal end had been synthesised before the label was added and therefore contained less radioactivity.
Q: Why are only 31 tRNAs needed to read 61 sense codons?
A: Wobble base pairing at the third position of the codon allows a single tRNA anticodon to recognise more than one codon. For example, inosine (I) in the anticodon's wobble position can pair with A, C, or U, so one tRNA can read up to three codons. This reduces the minimum number of tRNAs required from 61 to 31.
This completes the full central dogma (DNA → RNA → protein) that runs through the entire course. The genetic code connects to protein structure and function (later chapters on protein folding, enzyme kinetics). Frameshift mutations reappear in medical genetics (e.g. Tay-Sachs disease). The sickle cell example from Part 3 is a direct application of codon-level change producing an amino acid substitution. Translation initiation mechanisms (Shine-Dalgarno vs. Kozak) become important when studying gene expression differences between prokaryotes and eukaryotes, and in biotechnology when designing expression vectors.
genetic code, triplet code, codon, degeneracy, reading frame, frameshift mutation, Nirenberg Matthaei experiment, Khorana, Brenner Crick frameshift, start codon AUG, stop codons UAG UAA UGA, amber ochre opal, Shine-Dalgarno sequence, Kozak sequence, translation initiation, translation termination, release factors, polyribosome, Dintzis experiment, protein synthesis direction N to C, wobble base pairing, inosine, collinearity, tryptophan synthetase, selenocysteine, chapter 2 molecular biology