Why Do Refrigerated Tomatoes Lose Aroma? Changes Inside the Fruit

What happens inside a tomato when refrigeration weakens its flavor? A 2016 study exposed red-ripe tomatoes to low temperatures and examined their aroma compounds, gene activity, and chemical marks attached to DNA together.

After about a week at 5°C, aroma compounds decreased, and under some conditions they did not recover fully even after the fruit was returned to room temperature. Behind this finding were changes in the machinery that keeps replenishing aroma after harvest.

Following refrigeration’s effects on flavor down to the molecular level

Tomato flavor involves volatile aroma compounds sensed by the nose, alongside the sweetness of sugars and the sourness of acids. The study’s central question was how genes that produce these compounds, and the machinery that regulates fruit ripening, change when low temperatures reduce aroma.

The team used the heirloom variety Ailsa Craig and the more recent breeding line FLA 8059, selecting fully red-ripe fruit without damage or other defects. They first established when changes appeared by storing fruit at 5°C for one, three, or seven days, then returning it to 20°C for one or three days.

For the detailed chemical and genetic analyses, they compared fruit at harvest, fruit stored at 5°C for eight days, and fruit stored at 5°C for seven days followed by one day at 20°C. Volatiles were measured by collecting the amount released from chopped fruit over one hour. The main chemical analyses used three pooled samples, each containing six fruit.

One to three days gave different results from about a week

In the initial experiment, volatile compounds did not decrease significantly relative to harvest after one or three days of refrigeration. After seven days, however, they decreased in both varieties and did not return to harvest levels even after three days at 20°C.

In the detailed Ailsa Craig analysis, total volatile emissions from samples stored at 5°C for eight days were approximately 65% lower than at harvest. Samples returned to room temperature for one day after seven days of refrigeration showed some recovery but remained below harvest levels. This 65% describes a difference in measured volatile emissions, not an equivalent percentage decrease in perceived aroma intensity.

The researchers also assessed eating quality. Seventy-six consumers evaluated Ailsa Craig fruit refrigerated for seven days and then kept at room temperature for one day, alongside fruit of the same ripeness freshly harvested on a different day. Overall liking was lower for the refrigerated fruit. The control was freshly harvested fruit, not fruit stored for the same eight days at 20°C.

Meanwhile, glucose, fructose, malic acid, and citric acid did not differ significantly. The findings suggest that changes in aroma compounds, rather than large losses of sugars or acids, contributed to the decline in flavor.

Not every aroma compound decreases in the same way

Twelve compounds showed consistent refrigeration effects across both varieties and two growing seasons. Those that decreased included the fatty-acid-derived compounds hexanal, (Z)-3-hexenal, and (E)-2-hexenal, as well as the amino-acid-related compound 3-methyl-1-butanol.

By contrast, the carotenoid-derived compounds 6-methyl-5-hepten-2-one and geranial increased. Tomato aroma therefore does not change simply as though its overall volume had been turned down: the mixture of compounds changes too.

For the compounds that increased, activity of the corresponding biosynthetic-enzyme genes did not increase. The authors proposed that chilling stress might promote nonenzymatic oxidation and breakdown of carotenoids. This was offered as a hypothesis to explain the results.

Changes in aroma-producing genes and the regulation of ripening

Gene activity was tracked by measuring the amounts of RNA transcribed from those genes. RNA abundance changed across a wide range of genes during refrigeration, and many returned toward harvest levels after the fruit was brought back to room temperature. They did not all recover at the same speed, however.

For example, RNA levels for the enzyme genes LoxC, HPL, and ADH2, which help produce fatty-acid-derived aroma compounds, decreased during refrigeration and partly recovered at room temperature. Some genes involved in amino-acid-related aromas and esters also remained downregulated.

RNA levels for RIN, NOR, and CNR, which coordinate the regulation of many ripening-related genes, also decreased. Production of ethylene, a plant hormone regulating ripening, fell and partly recovered at room temperature. However, RNA for ethylene-synthesis enzymes did not uniformly decrease, suggesting that regulation after the RNA stage was also involved.

Because aroma compounds escape from fruit, they need to be replenished after harvest. The authors explain that low-temperature suppression of this production contributes to aroma loss. An important contribution of the study was to understand refrigeration’s effects as changes in the fruit’s metabolism.

Temporary changes in chemical marks on DNA

The team also examined DNA methylation. This does not rewrite the DNA sequence itself; it involves attaching small chemical marks called methyl groups to DNA. Such changes can be associated with how genes are read.

After refrigeration, average methylation increased in promoter regions that regulate gene expression. After one day back at 20°C, it returned close to harvest levels. For some genes, including RIN and genes involved in aroma synthesis, increased methylation coincided with decreased RNA abundance.

However, the study did not establish whether the methylation changes caused the RNA decrease or the reverse. Rapid recovery of methylation is also a different outcome from sufficient recovery of the aroma as a whole. Measuring compounds, RNA, and DNA marks together showed that the response to refrigeration spans several levels.

Different conditions from the study finding no difference after four days

Our article on a 2020 study (in Japanese) describes a comparison of ripe tomatoes stored at 7°C or 20°C for four days after three days of storage intended to simulate distribution. Under those conditions, trained assessors found no significant storage-related differences in perceived aroma or taste.

The 2016 study also used fully ripe fruit, so ripeness alone cannot explain the different findings. Temperature, duration, variety, simulated distribution treatments, and controls differed. So did the assessments: consumer liking versus attribute intensity judged by a trained panel. The 2020 volatile analysis addressed relative composition and also examined the suitability of an electronic tongue.

Read together, the studies show why it matters to ask at what temperature, for how many days, which tomatoes were stored, and what they were compared with. Even within the 2016 experiment, short refrigeration and about a week of refrigeration produced different results.

Summary

Storing red-ripe tomatoes at 5°C for about a week reduced aroma compounds and liking, with incomplete recovery after return to room temperature under some conditions. While sugars and acids changed little, substantial changes occurred in aroma-producing genes, ripening regulation, and DNA methylation.

The study highlighted the importance of continued production in sustaining aroma after refrigeration. Different compounds showed different increases, decreases, and recovery patterns. Considering storage temperature and duration helps explain more specifically why flavor changes.

The study received support from the US NSF, Chinese research programs, Zhejiang University, and other sources. The authors declared no conflicts of interest.

References

Zhang B, et al. (2016). Chilling-induced tomato flavor loss is associated with altered volatile synthesis and transient changes in DNA methylation. PNAS, 113(44), 12580–12585. Original paper.

Kanski L, Naumann M, Pawelzik E. (2020). Flavor-Related Quality Attributes of Ripe Tomatoes Are Not Significantly Affected Under Two Common Household Conditions. Frontiers in Plant Science, 11, 472. Original comparison paper.