Polyphenol Oxidase: Why Fresh Fruit Gelato Turns Brown
Table of contents
You blend a perfect pear, and twenty minutes later the purée has gone the colour of weak tea. Nothing burnt, nothing spoiled. A copper enzyme that was sitting quietly inside the intact fruit met the oxygen you introduced with the blade, and it started working immediately.
The clock starts at the first cut, not at the freezer.
What the enzyme actually does
Polyphenol oxidase, usually shortened to PPO, is a copper-containing enzyme present in the tissue of most fruit. In an undamaged fruit it is physically separated from its substrates: the enzyme sits in the plastids, the phenolic compounds sit in the vacuole. Cut, crush or blend the fruit and you destroy that separation.
From there the sequence is short and fast. PPO oxidises colourless phenolic compounds into ortho-quinones, using molecular oxygen as the acceptor. The quinones are highly reactive and do the rest of the work without any enzyme at all: they polymerise with each other and with amino groups into brown, high-molecular-weight melanin pigments. The enzyme is only the trigger. The colour itself is a spontaneous chemical cascade.
Three things must be present for it to run. The enzyme, a phenolic substrate, and oxygen. Chlorogenic acid is the dominant substrate in apple, pear and peach; catechins matter in several stone fruits; banana browns through dopamine. Remove any one of the three and the reaction stops, which is why there are exactly four useful levers and not forty.
Why the freezer does not save you
The instinct in a gelato lab is that cold fixes everything. It does not fix this.
Cold slows the enzyme. It does not destroy it. PPO retains measurable activity at refrigeration temperature and survives freezing entirely intact, because freezing is a physical process and denaturation is a chemical one. A raw purée frozen at minus 18 degrees Celsius comes back with its PPO fully functional the moment it thaws, and in a sorbet there is always an unfrozen serum phase where a slow reaction can keep creeping.
This is why fruit sorbet made from raw purée can look right on day one and read visibly duller in the display case a week later. The reaction never stopped, it just ran at a speed you did not notice.
The four levers that actually work
Heat. PPO is a protein, and heat denatures it permanently. Blanching is the classic industrial answer, and a normal pasteurisation hold at 85 degrees Celsius is comfortably enough to inactivate it. This is the only lever that removes the problem rather than delaying it. The cost is flavour: a cooked pear tastes like a cooked pear.
Acid. PPO activity peaks in the region of pH 6 to 7 and falls away sharply below pH 4. Pushing a purée under pH 3.5 with lemon juice or citric acid slows the enzyme to near-uselessness, and citric acid does a second job at the same time by chelating the copper the enzyme needs at its active site. Malic acid works on the pH side but is not the chelator citric is. If you are going to work this lever, buy a pH meter, because titratable acidity is not the same measurement and only one of them tells you where the enzyme sits.
Oxygen. No oxygen, no reaction. In practice this means filling the blender jug rather than whipping air into a half-empty one, working fast, covering the purée, and, if the equipment exists, using a vacuum pasteuriser. It also means being honest that high overrun is a browning risk on a raw fruit base.
Reducing agents. Ascorbic acid works mainly on the products rather than the enzyme. It does bind the active site as a competitive inhibitor at low concentration, but what carries the effect at working doses is that it reduces the quinones back to colourless phenols, so it buys time proportional to how much of it you added, and browning resumes the moment it is used up. In fresh-cut produce the usual working range is roughly 0.1 to 0.5 percent of fruit weight. It is a useful stopgap for a raw purée that has to sit, not a permanent fix. Sulphites are far more effective and heavily regulated as declarable allergens, which puts them out of reach for most artisanal labs.
Quick reference. PPO needs enzyme, phenol and oxygen together. Heat is the only lever that removes the enzyme; acid below pH 3.5 slows it; excluding oxygen starves it; ascorbic acid only buys time. Freezing does none of these things.
Figure 1. One pathway, four places to break it, and only one of them is permanent.
Which fruits are actually at risk
PPO activity varies enormously between species and even between cultivars of the same species, so the useful question is never "does fruit brown" but "does this fruit brown".
| Fruit | Browning risk | Main substrate | First move |
|---|---|---|---|
| Apple, pear, quince | high | chlorogenic acid | acidify hard, or cook |
| Banana | high | dopamine | acidify and work fast |
| Peach, apricot, nectarine | moderate to high | chlorogenic acid, catechins | pasteurise the purée |
| Avocado, fig | high | mixed phenolics | acidify, exclude air |
| Berries | low in practice | present but masked | protect pigment, not from PPO |
| Citrus | very low | low PPO activity | no action needed |
| Melon, pineapple | low to moderate | mixed | work fast, keep cold |
For anything in the top rows, the simplest professional answer is often to skip the problem entirely: a commercially frozen purée has already been heat treated, and choosing purées well removes an entire class of colour failure from the day. Freeze-dried fruit powder sidesteps it too.
A working protocol for raw fruit sorbet
- Weigh the acid before you cut anything. Have the lemon juice or citric solution in the bowl waiting.
- Blend directly into the acid, jug as full as it will go, short bursts rather than a long vortex.
- Check pH. Under 3.5 for a high-risk fruit. Adjust with acid, then rebalance the sugar, because acid changes perceived sweetness and sorbet balance has to hold.
- Decide, deliberately, whether this fruit gets heat. If flavour allows it, pasteurise and stop worrying.
- Cover the purée and get it cold. Minimum headspace in the container.
- Do not age a raw fruit purée any longer than the recipe needs. Unlike a dairy base, where aging earns its keep, resting time here only costs colour.
- Churn and freeze quickly, and put a real shelf life on the raw-fruit flavours rather than the same one you use for white base flavours.
When browning is not the enzyme
Not all beige is PPO, and treating the wrong cause wastes a week.
If the base browned during or after heating, and it contains dairy or protein, you are looking at the Maillard reaction instead: that one needs a reducing sugar and an amino group and heat, and it does not care about oxygen. Swapping dextrose for sucrose changes it, acid does not. The distinction is worth internalising because which sugars brown is a recipe decision, while PPO is a handling decision.
If the colour drift is a fruit sorbet fading rather than darkening, especially a red or purple one, that is pigment degradation, not browning, and the fix lives in pH and light exposure. If you want to argue about any of this with numbers rather than impressions, colour tolerance in delta E is how the industry does it.
Related Concepts
- Fruit purée selection is the decision that removes most of this problem before it starts
- Citric acid in sorbet covers the lever you will reach for most often
- Browning in gelato is the non-enzymatic sibling of this reaction
- Anthocyanin colour stability handles the fading problem rather than the darkening one
- pH meters for sorbet is the tool that turns all of this from guesswork into a number
