Chilling injury in fruit

Why does a banana kept in the fridge turn black outside while staying stubbornly hard inside?

Cold is supposed to be a preservative. It keeps milk, it keeps meat, it keeps apples for the better part of a year. Put a banana in the same fridge and within two days the skin is the colour of a bruise — yet cut it open and the flesh is pale, firm, and starchy, exactly as unready as it was going in.

That combination is the puzzle. Not "the banana spoiled faster", which would merely be surprising. Something in the fruit sped up and something else stopped, in the same fruit, at the same temperature. Any account that treats ripening and spoiling as one process running at one rate cannot produce that result.

So ask what the black is. It is not rot: chilled bananas are not microbially spoiled, and the blackening is diffuse over the whole peel rather than starting at a wound or a fungal spot. It is not freezing either — a domestic fridge sits near 4 °C and banana tissue does not freeze until roughly a degree or two below zero, so no ice forms.

What is left is the fruit's own chemistry. Plant cells keep phenolic compounds locked in the vacuole and the enzyme that oxidises them, polyphenol oxidase, outside it. Bring the two together in the presence of oxygen and you get quinones, which polymerise into brown-black pigments. That is the same reaction as a cut apple going brown, and it needs no programme and no permission — only that the wall between substrate and enzyme fails.

Which raises the question: what breaks the wall? Membranes are lipid bilayers, and a bilayer is fluid only above a temperature that depends on how unsaturated its fatty acids are. Below that point it stiffens into a gel-like state, and a membrane that is part gel and part fluid leaks at the boundaries. Tropical fruits are built from more saturated membrane lipids than temperate ones, so their transition sits high — for green bananas it is usually put around 13 °C, well above a fridge. Apples, built for a temperate winter, sit far below it. That single difference predicts which fruits suffer in a fridge and which are stored at 0 °C for months, and it predicts a threshold, not a slope: 15 °C does nothing, 8 °C does damage, and colder is worse only after the line is crossed.

Now the other half. Why does the flesh not ripen? Because ripening in a banana is not a reaction, it is a programme — starch broken down to sugar, pectins cut so the flesh softens, aroma compounds synthesised — and the programme is gated. It runs when the fruit makes ethylene, and ethylene makes more ethylene, so the whole thing goes off like a fuse once lit. Cold suppresses the enzymes that produce ethylene and the ripening enzymes downstream of it. The gate never opens.

So both effects come from the same cold, but they arrive by different routes. One process needed only a wall to fail, and cold failed it. The other needed a trigger to fire, and cold prevented it. Blackening is what happens when nothing is in charge.

Two refinements matter here.

The first is that "chilling injury" is a dose, not an event. Damage accumulates with time below the threshold, which is why a banana left an hour in the fridge is fine and one left three days is not, and why commercial cold chains for chilling-sensitive produce sometimes use intermittent warming — brief returns to a safe temperature — to let tissue recover before the dose accumulates. That practice is itself an argument for the mechanism: if cold merely slowed things, warm interruptions would only waste energy.

The second is a caution about how tidy I have made this. The membrane phase-transition account is the classical and still the most widely taught explanation, but the modern picture is broader: reactive oxygen species accumulate, ion balance is disturbed, and cold-responsive gene expression is involved, and researchers argue about how much of the damage is membrane physics versus downstream oxidative stress. What is not in dispute is the shape of the thing — a threshold temperature, cumulative dose, compartmentation lost, ripening blocked.

One prediction to hang it on, in case it is wrong: this account says the peel browning is enzymatic oxidation, so it should be suppressible by excluding oxygen and should not require any living ripening programme. It also says a chilling-tolerant fruit is one whose membrane lipids are more unsaturated — so raising the unsaturation of a sensitive plant's membranes should raise its cold tolerance. Both of those have held up; if either had failed, the account would have to go.

• Why some cultivars of the same species differ sharply in chilling tolerance.
• How controlled-atmosphere storage suppresses ripening without the membrane cost that cold imposes.

Every term the collection defines is gathered in the glossary.