This article was originally published on the WinterTurf project blog.
By Katrina Freund-Saxhaug
As part of the WinterTurf project, a diverse panel of approximately 190 perennial ryegrass genotypes is being screened using metabolomic methods to understand how plant metabolites shift as temperature decreases in the fall (see earlier posts about using untargeted metabolomics to assess cold tolerance in perennial ryegrass). Prior to bringing the full panel through a traditional fall acclimation program, a central question emerged: can the experimental design be improved to more accurately reflect real-world conditions and better capture the timing of key metabolic shifts? These questions motivated reevaluation of two components:
- How should we design a controlled growth chamber program that mirrors field temperature dynamics?
- How should we select sampling time points that best capture biologically meaningful metabolic transitions during acclimation?
Previous work by former postdoctoral researcher Dr. Lindsey Hoffman Chappell showed a strong temperature-driven effect on metabolomic profiles, with cold-tolerant genotypes responding more quickly to temperature decreases. In PCA plots from this earlier work (Figure 1), both the tolerant (TOL2) and susceptible (SUS1) genotypes showed distinct metabolic shifts between 20 °C and 2 °C. However, the profiles differed between 2 °C and –2 °C: the tolerant genotype remained metabolically stable, while the susceptible genotype continued to shift. These observations prompted several questions: Does the susceptible genotype continue adjusting its metabolism during freezing? Are tolerant genotypes characterized by an earlier or more proactive metabolic response, suggesting preparation for freezing through upregulation of protective compounds or suppression of vulnerable pathways? Is cold susceptibility due to slower or insufficient metabolic adjustment as temperatures fall? Taken together, these findings indicated that a new acclimation program was needed, one that better reflects actual outdoor conditions and allows identification of when meaningful metabolic shifts occur. This would help ensure that sampling captures the most informative points during acclimation.
Developing a relevant growth chamber program for WinterTurf
Because the perennial ryegrass genotypes from the panel may move into our turfgrass breeding program, it was important that their acclimation conditions mimic field-relevant environments. Therefore the goal was to design a growth chamber program that reflects fall temperature patterns in Minnesota (September–December). With help from Jessica Till, a former research scientist in the University of Minnesota Department of Bioproducts and Biosystems Engineering, 20-year average surface temperature data were examined for Minnesota. Jessica provided daily and weekly averages from 2004–2023 using data from the North American Land Data Assimilation System (NLDAS), and a graph of the NLDAS 2-meter temperature data for the fall acclimation period revealed a characteristic pattern: an initial decline, a plateau near 17 °C, another decline, a plateau near 5 °C, a further decline, and a plateau near –8 °C. The “WinterTurf acclimation gradient” is a condensed version of this curve in which 3–4 days of real temperature change are condensed into 1 day in the growth chamber, resulting in a 37-day program (Figure 2).
In consultation with Douglas Brinkman, Researcher and Growth Chamber Manager at the University of Minnesota, the following program was established: hold temperature at 17 °C (days 1–4), ramp down to 5 °C, hold at 5 °C (days 15–20), ramp down to –8 °C, and hold at –8 °C (days 31–37). Following recommendations from Dom Petrella, photoperiod was set to 12 hours at the start (days 1–14), reduced to 10 hours after the first temperature ramp (days 15–30), and reduced to 8 hours after the second ramp (days 31–37) to roughly mimic fall daylength patterns. Photosynthetic photon flux density was maintained at 200 µmol m⁻² s⁻¹ throughout.
Daily sampling to capture shifts in metabolism
Cold acclimation in cool-season turfgrasses involves coordinated biochemical transitions that are not necessarily steady processes. When plants are first exposed to low temperatures, they initiate immediate metabolic adjustments - such as shifts in carbohydrate allocation and production of osmoprotectants - that can occur within hours to a few days. If sampling is restricted to the end of temperature plateau phases, which is common in turfgrass acclimation studies, transient but biologically informative events may not be captured. Many regulatory metabolites spike early and then decline, while others only briefly appear as intermediates of stress-response pathways. As a result, previous methods of sampling in acclimation studies can underestimate the timing, magnitude, and diversity of acclimation responses. Including a time-course approach not only improves detection of short-lived regulatory compounds but also strengthens the interpretation of downstream metabolomic patterns, leading to a more complete view of metabolic strategies associated with winter survival.
In this experiment, clonal tillers of SUS1 and TOL2 were established in cone-tainers in the greenhouse for two months, then moved into a GCW15 Environmental Growth Chamber. Tillers were maintained at 17 °C for 14 days before starting the acclimation program. Four replicates per genotype were sampled daily for 37 days. For each sampling event, leaves were removed just above the ligule on all tillers in a cone-tainer. The leaf tissue was chopped, homogenized, and approximately 200 mg of fresh tissue was extracted in 80% methanol (1 mL MeOH per 200 mg of leaf sample). Metabolomic profiles were obtained on a C18 column using LC-MS with a 15-minute gradient (mobile phase A: 0.1% formic acid in water; mobile phase B: 0.1% formic acid in acetonitrile; gradient: initial 2% B, 1 min 2% B, 15 min 98% B, 0.5 min 98% B, 0.5 min 2% B). Data conversion and processing were performed with MSConvertGUI4 and MZmine 2.53, and statistical analyses were conducted in R.
What did we find?
This experiment produced several interesting results that have informed how the full perennial ryegrass panel will be brought through the acclimation program, and these findings will be presented in the next part of this blog post. Although sampling every genotype daily throughout acclimation would provide the most detailed view of metabolic dynamics, practical limitations require trade-offs. Growth chamber capacity and time constraints mean that the number of genotypes being sampled during a given acclimation program must be balanced with the sampling intensity. The results of this experiment help identify the most informative time points to capture metabolic shifts, allowing the experiment to scale to the full panel while remaining feasible.