Biology

Tomato study suggests fruits can warn mother plants of pest attacks

View 2 Images
The research indicates that tomato fruits can send electrical warning signals "upstream" to the main plant
Barabasa/Depositphotos
The research indicates that tomato fruits can send electrical warning signals "upstream" to the main plant
Barabasa/Depositphotos
One of the tomato plants that was monitored in the study
Gabriela Niemeyer Reissig

Because fruits just dangle from the main plant before ultimately falling off, one might think that they are unable to communicate with that plant. According to new research, however, tomato fruits are able to transmit alerts to their mother plant when attacked by caterpillars.

Given the fact that sap typically only runs from the plant to the fruit – not back and forth between the two – it has long been unclear whether or not fruits can transmit information to the plant.

For the new study, which set out to address that question, scientists at Brazil's Federal University of Pelotas started by placing tomato plants in a Faraday cage. Doing so blocked any external electromagnetic fields. Electrodes were attached to the ends of the plants' branches, at the points where they connected to the fruits.

Utilizing those electrodes, the researchers measured electrical responses within the branches before, during and after a 24-hour period in which the fruits were attacked by Helicoverpa armigera caterpillars. Machine-learning-based algorithms were used to detect patterns within the recoded signals.

One of the tomato plants that was monitored in the study
Gabriela Niemeyer Reissig

It was found that there was a "clear difference" in the patterns that were detected before and after the caterpillar attacks. Additionally, defensive biochemical responses were observed throughout the plants' bodies, suggesting that they had been triggered by signals sent by the fruits.

"Since fruits are part of the plant, made of the same tissues of the leaves and stems, why couldn't they communicate with the plant, informing it about what they are experiencing, just like regular leaves do?" asks Dr. Gabriela Niemeyer Reissig, first author of a paper on the study. "What we found is that fruits can share important information such as caterpillar attacks – which is a serious issue for a plant – with the rest of the plant, and that can probably prepare other parts of the plant for the same attack."

The scientists now plan on investigating whether other fruiting plants behave in the same manner as tomatoes, and whether their fruits respond to threats other than insects. It is hoped that the team's findings could ultimately lead to earlier detection of pest infestations in crops, along with improvements in fruit quality, shelf life and pest resistance.

The paper was recently published in the journal Frontiers in Sustainable Food Systems.

Source: Frontiers

  • Facebook
  • Twitter
  • Flipboard
  • LinkedIn
3 comments
christopher
"Machine-learning-based algorithms" - keep in mind that using such algorithms, you can always find any match for anything you're looking for - but that doesn't mean it's an actual result. Unless it's repeated a few times, and the double-blinded ML gets the answer right *again*, all you've done is found an arbitrary small sequence in an infinite run of random numbers. i.e. nothing.
Karmudjun
NIce article Ben.

I'm always fascinated by repeatable research investigations utilizing ML to find all the possible patterns in discrete investigations and debating the correlations. There is a supposition in scientific circles that the experiments MUST be repeatable with the same findings/results, and that the null hypothesis (the match for anything you are looking for - that renders the experiment a failure such as 'there is no communication difference between the pre-caterpillar attack tomato plant electrical signalling and post-caterpillar attack tomato plant signalling') which indicates a failed experiment. So this basic experiment appears to be a success and will be duplicated with other plants to determine if there are random patterns or similar patterns. Since the null hypothesis (what you look for to determine arbitrary randomness) wasn't the outcome, then the actual result IS an actual result. Fascinating how scientific theory works in practice - even with Machine Learning.
Worzel
An apple tree may have thousands of apples on it, but it can detects those with worms in, and jettison them, so it doesn't waste energy on them. Acacias in Africa, can also communicate with other acacias by releasing pheromones, which warn that they are being overgrazed. Then they, and those acacias down wind will produce a poison, that is capable of killing large herbivores, which then reduces the overgrazing.
Trees have been around for many hundred millions of years, they've had plenty of time to develop lots of survival tricks. They also have a symbiotic relationship with mycelium, that allows them to communicate with other trees. I'm wondering if the signals being transmitted back from the fruit, in this experiment, are actually carried by mycelium.