How often does the ribosome get it wrong?
From the last few guides you know the machinery: a ribosome clamps the mRNA, the right tRNA arrives, a peptide bond forms, and the message advances one codon at a time. Run that loop and a worry should nag at you. A bacterium adds roughly 15-20 amino acids per second; a mid-sized protein is hundreds of residues long; a cell is making thousands of proteins at once. At that blistering pace, how does the right tRNA ever find the right codon often enough? You might expect a blizzard of mistakes. The remarkable fact is that the ribosome misreads only about one codon in 1,000 to 10,000 — accurate enough that most proteins come out perfect.
There is a deep puzzle hiding in that number. A correct codon-anticodon match and a near-miss that is wrong in just one position barely differ in their binding energy — sometimes by only a hair. If the ribosome simply waited to see which tRNA stuck more tightly, that tiny energy gap could never deliver one-in-ten-thousand accuracy. Pure thermodynamics is not selective enough. So the cell does something cleverer than passive selection: it spends energy to amplify a small difference into a large one. Accuracy, it turns out, is not free — it is bought.
Two checkpoints and the energy you pay for them
The first checkpoint comes before the ribosome ever sees the tRNA. Recall from the tRNA guide that an [[molbio-aminoacyl-trna-synthetase|aminoacyl-tRNA synthetase]] is the enzyme that loads each tRNA with its correct amino acid — the true guardian of the code. Many of these enzymes carry a separate editing pocket: if they accidentally attach the wrong, similarly-shaped amino acid, that mischarged product is shaped to slip into the editing site and gets hydrolyzed off before it ever leaves. This is genuine proofreading — make the bond, then check it, then cut it if it is wrong — and it is why the amino acid riding each tRNA is almost always the right one.
The second checkpoint is at the ribosome itself, when an incoming tRNA tests its anticodon against the codon in the A site. This is where the cell spends GTP to buy accuracy, in two timed stages. First, the ribosome inspects the geometry of the pairing — a correct three-letter match snaps the codon and anticodon into a precise shape that the ribosome physically senses, while a near-miss sits slightly crooked and is more likely to fall off. Only a good match triggers the elongation factor to burn its GTP. Then comes a deliberate pause — a brief delay before the amino acid is committed — giving a wrong tRNA a second chance to dissociate before any bond is made. Inspect, commit, then pause-and-recheck: two filters in series multiply their selectivity.
Same job, two different machines: bacteria vs us
Every cell translates, but the machine differs in detail across the great [[prokaryote-eukaryote-divide|prokaryote-eukaryote divide]] you met back in the foundations rung. A bacterial ribosome is smaller — biologists label it 70S, built from 30S and 50S subunits — while ours is bigger, the 80S made of 40S and 60S. (Those S numbers are sedimentation rates, not simple sizes, which is why 30 + 50 'equals' 70.) Initiation differs even more sharply. In bacteria the small subunit lands directly on the message, guided by a short marker just upstream of the start codon — the [[shine-dalgarno-sequence|Shine-Dalgarno sequence]] — so a bacterium can even begin translating one stretch of an mRNA while the rest is still being transcribed. In eukaryotes there is no Shine-Dalgarno; the small subunit instead binds the mRNA's 5' cap and scans along until it finds the first AUG.
BACTERIA (70S) EUKARYOTES (80S)
30S + 50S 40S + 60S
Shine-Dalgarno marks 5' cap; small subunit
the start codon scans to first AUG
transcription & translation transcription in nucleus,
coupled (same place/time) translation in cytoplasm
---> the differences are SMALL but real, and that
narrow gap is the target many antibiotics aim atOne more difference is geographic and matters for the next rung. A bacterium has no nucleus, so transcription and translation happen in the same space at the same time — ribosomes can grab a message the instant it is born. A eukaryote splits the two: the mRNA is made and processed in the nucleus, then exported to the cytoplasm to be translated. That separation gives eukaryotes room for the editing steps you saw in the RNA rung (capping, splicing, the poly-A tail) and, as we will see, extra layers of control over which messages get read and when.
Why an antibiotic can hit a germ but spare you
Here is where those small differences become life-saving. A great many antibiotics work by jamming the bacterial ribosome — and because the bacterial machine differs from ours, a drug can be shaped to wedge into the 70S ribosome while barely touching our 80S one. That gap is exactly the source of selective toxicity: the holy grail of a drug is to poison the pathogen and leave the patient alone, and the prokaryote-eukaryote divide in the ribosome hands us a built-in target. The ribosome is, in fact, one of the single most-drugged structures in all of medicine.
The drugs hit different steps of the cycle you learned last guide. Tetracyclines block the A site so an incoming tRNA cannot dock — decoding stalls. Aminoglycosides (like streptomycin) bind the small subunit and make it misread, so the wrong tRNAs are accepted — they sabotage the very accuracy this guide is about. Macrolides (like erythromycin) plug the tunnel the growing chain exits through, so the peptide jams. Chloramphenicol blocks peptide-bond formation outright. Each one is a precise spanner thrown into one stage of bacterial translation.
Translation as a control knob — and the handoff to folding
It is tempting to picture translation as a dumb final step that simply executes whatever mRNA arrives. It is not. Whether and how fast a message is translated is itself a layer of control, distinct from deciding whether the gene is transcribed at all. A cell can stockpile an mRNA and translate it only on a signal; it can dial a single message up or down; it can shut translation down globally under stress to stop wasting energy. The amount of a protein in a cell is set not just by how much mRNA exists, but by how busily each message is read.
The mechanisms are varied and clever. In bacteria, a folded patch of the mRNA itself — a riboswitch — can hide the Shine-Dalgarno sequence until a small molecule binds and frees it, so the message reads its own surroundings. In our cells, small RNAs (the microRNAs you met in the RNA rung) pair with a message to block its translation or trigger its decay, and a single mRNA is usually read by many ribosomes at once — a polysome — so loading more or fewer ribosomes per message tunes output. None of this rewrites the genetic code; it governs how vigorously the same code is read.
Step back and the whole arc of this rung is complete: a coded message, an adapter that reads it, a machine that builds the chain, three stages of assembly, and the energy spent to keep it accurate. But the moment the ribosome releases a finished chain, the story is not over — that chain is a limp string of amino acids that must collapse into a precise three-dimensional shape before it can do anything. Folding often begins while the chain is still emerging from the ribosome's exit tunnel, sliding down its [[protein-folding-funnel|folding funnel]] toward its working form. That handoff — from the ribosome that wrote the protein to the machinery that shapes, ships, and quality-checks it — is exactly where the next rung begins.