You check your phone before deciding what to wear tomorrow. Plants do something similar: they carry an internal gadget called a Season-Meter that tracks temperatures over the preceding weeks or months to help determine when to flower. By comparing two wild perennial plant species over two years, the researchers found that each of the three genes in this Season-Meter operates over a different timescale, ranging from days to nearly five months, and that together they reliably forecast all four seasons of the year.

Tsukuba, Japan—Many challenges plants face do not occur only once; they recur on a predictable, roughly annual cycle, such as winter cold preceding spring or shortening daylight before autumn. Because these environmental signals are predictable, evolution has favored systems that allow plants to prepare in advance rather than react only to today's weather. Such preparation requires an internal memory. Since DNA itself cannot record a changing environment, plants instead preserve information about past temperatures through epigenetic changes in chromatin.

Rather than responding only to the temperature on a given day, plants appear to carry something like a built-in gadget, which the researchers have named the Season-Meter. It keeps a running tally of how temperatures have unfolded over the preceding weeks and months, allowing plants to forecast the season ahead and decide when to grow, flower, or wait. A single warm day in midwinter fools this internal Season-Meter into blooming early, whereas a single cold snap in spring holds back flowering that is already on schedule.

Although scientists have long known that plants remember past temperatures, no standard method has existed for measuring how far back this epigenetic memory extends or whether it operates similarly across plants and genes. To address this gap, the researchers studied two wild, long-lived species in the mustard family that grow naturally in Japan: wasabi, which grows in a mountain forest, and A. halleri, which grows along a mountain stream. They tracked the same three genes in both species over two full years.

By comparing each gene's activity with temperature records from the preceding weeks and months, the team estimated the span of past temperatures most strongly reflected in the activity of each gene. They found that the Season-Meter comprises three genes that integrate past temperatures over distinct timescales. One gene looked back only a few days, another a few weeks, and the third nearly five months, the longest interval measured in this system so far. Together, these overlapping memory windows allow the Season-Meter to track seasonal progression throughout the year. The same pattern appeared in both plant species despite their very different life cycles, suggesting a general strategy rather than a species-specific quirk.

The team next tested whether the Season-Meter could predict responses to climate. Using a model constructed solely from past temperature data, the researchers successfully predicted how the genes would behave in a separate year that had not been used to build the model. This result indicates that the seasonal forecast captured by the method is reproducible and not merely a coincidence arising from the data used to construct the model.

Because this approach requires only measurements of gene activity and temperature, rather than a detailed understanding of the underlying biological machinery, it could be extended to other genes, environmental signals such as rainfall or daylight, and species. As climates continue to shift, approaches like this could help researchers predict how wild and cultivated plants will respond to unfamiliar weather patterns.

###
This research was supported by the Japan Society for the Promotion of Science, (JSPS) KAKENHI (JP20K06699) and Scientific Research on Innovative Area (grant No. JP16H01459) to DB; JSPS KAKENHI (JP15K07289) to KY; JSPS KAKENHI (JP21H05659)to HN; and JSPS KAKENHI (JP21H04977), the Japan Science and Technology Agency (JST) CREST (JPMJCR15O1) and a Joint Usage program at the center for Ecological Research, KyotoUniversity to HK. Molecular biology equipment was available at Tsukuba-Plant Innovation. Research Centre (T-PIRC) at the University of Tsukuba.

Original Paper

Title of original paper:

Conserved gene- and network-level thermal memory intervals in two divergent perennial crucifers in nature

Journal:

PLOS ONE

DOI:

10.1371/journal.pone.0336733

Correspondence

Associate Professor Diana Mihaela Buzas
Tsukuba Plant Innovation Research Centre and Institute of Life and Environmental Sciences, University of Tsukuba

Related Link

Institute of Life and Environmental Sciences
Tsukuba-Plant Innovation Research Center（T-PIRC）