Overwintering Conditions Reshape Flowering Time Genetics in Arabidopsis
Flowering time in Arabidopsis thaliana did not behave the same way outdoors as it did in the growth chamber. In a common garden that exposed plants to an overwintering environment, investigators found extensive genotype-by-environment interaction, or GxE, for a trait that is central to plant fitness, seasonal timing, and local adaptation. The result is biologically familiar but still consequential: the genes that matter under tightly controlled conditions are not necessarily the ones that matter when temperature, photoperiod, and other outdoor cues act together across a real season.
The study compared 479 accessions previously phenotyped in chamber conditions through the 1001 Genomes Project with the same material grown in an ecologically realistic overwintering setting. That design matters because flowering time is one of the clearest examples of a trait shaped by environment-sensitive genetic regulation. Plants do not simply “carry” a single flowering-time program; they integrate cues through multiple pathways, and those pathways can shift in relative importance when the environment changes. In this case, the outdoor setting exposed a broader and more variable genetic architecture than the chamber environment alone.
One sentence captures the central finding: the large-effect loci were not shared across growth conditions.
Distinct Loci Emerged Outside the Chamber
Multivariate genome-wide association analysis identified large-effect loci in each environment, but none overlapped between the overwintering common garden and the growth chamber conditions. For interpretation, that is a sharp reminder that a locus with a strong effect in one setting may disappear in another because its biological influence depends on temperature exposure, developmental timing, or the sequence of environmental cues the plant receives. Controlled environments remain useful, but they can compress the range of responses that matter in nature.
The study also found evidence consistent with several classic forms of GxE. Allelic differential sensitivity appeared when the same allele changed in effect size across environments. Conditional neutrality was present when a variant mattered in one setting but not the other. Antagonistic pleiotropy also contributed, meaning some alleles likely favored one environmental context while being disadvantageous in the other. Together, those patterns describe a trait that is not governed by a single stable genetic map but by an environment-dependent one.
This is exactly the kind of architecture expected for flowering time, which sits at the intersection of photoperiod, vernalization, temperature, and developmental state. In an overwintering context, the plant has to interpret cold exposure and spring transition cues rather than the simplified, steady-state signals of a chamber. A locus that influences timing under constant laboratory light and temperature may become irrelevant if a different pathway dominates under fluctuating outdoor conditions.
Partial Overlap, Not a Complete Rewrite
The picture was not one of total genetic turnover. Effect sizes for the most significant 1% of SNPs were moderately correlated between outdoor and growth chamber conditions, suggesting a partially shared genetic basis even as the major loci diverged. That combination is important because it implies both continuity and change: some genomic signal persists across environments, but the strongest signals are redistributed when the plants face realistic seasonal conditions.
In practical evolutionary terms, a moderate correlation in SNP effect sizes can reflect conserved regulatory architecture underneath a layer of environment-specific modulation. Shared effects may arise from core flowering pathways that remain relevant regardless of context, while the large-effect loci identified in one setting but not another point to conditional control points that only become visible under naturalistic cues. The genetic architecture is therefore not erased by the environment; it is rearranged.
For evolutionary biology, this matters because inferences about adaptation drawn from growth-chamber studies may miss the loci most relevant to survival and reproduction in the field. Selection does not operate in a vacuum, and flowering time is especially sensitive to the seasonal timing that determines whether a plant reproduces successfully before adverse conditions return. If the environment changes the ranking of loci, then estimates of which variants contribute most to adaptation can shift as well.
Why Ecological Realism Changes the Interpretation
The investigators framed their results around a core evolutionary problem: how genetic variation is expressed across environments. The overwintering garden revealed genetic effects across multiple flowering-time regulatory pathways that are not as apparent in standard chamber conditions. That suggests the outdoor setting captured interactions among cues that laboratory protocols often isolate or flatten.
The biological logic is straightforward. Flowering is not a single switch; it is a coordinated developmental decision controlled by several inputs that can reinforce or compete with each other. When those inputs vary seasonally, the phenotype becomes a moving target, and the underlying genetics can appear simpler or more complex depending on where and how the experiment is run. In this study, ecological realism increased the visibility of that complexity.
The authors are not claiming that chamber work is uninformative. They are showing that chamber-based inference has boundaries. Some of those boundaries are subtle, but here they were large enough to change which loci reached significance and how flowering-time architecture was partitioned across pathways. For a trait embedded in seasonal ecology, that is a meaningful shift in interpretation.
Limits of the Evidence and the Next Biological Test
Because this is a preprint, the findings have not yet undergone peer review. The study also centers on a particular overwintering common garden and a defined set of accessions, so the exact genetic architecture may differ under other climates, seasons, or field sites. Even so, the data make a strong case that flowering time should not be treated as genetically static across environments.
The next test is not whether environment matters; it clearly does. The sharper question is how much of the flowering-time map survives across settings, which loci are specific to seasonal exposure, and how these patterns generalize beyond this single overwintering site. Until then, the main lesson is concrete: the genetic basis of a complex trait can change when plants leave the chamber and enter the season.