Response to Waterlogging Stress in Wild and Domesticated Accessions of Timothy (Phleum pratense) and Its Relatives P. alpinum and P. nodosum

Abstract

Timothy (Phleum pratense) is a cool-season perennial forage grass widely grown for silage and hay production in northern regions. Climate change scenarios predict an increase in extreme weather events with fluctuating periods of high rainfall, requiring new varieties adapted to waterlogging (WL). Wild accessions could serve as germplasm for breeding, and we evaluated the responses of 11 wild and 8 domesticated accessions of timothy, P. nodosum and P. alpinum from different locations in northern Europe. Young plants at tillering stage were exposed to WL for 21 days in a greenhouse, and responses in growth allocation and root anatomy were studied. All accessions produced adventitious roots and changed allocation of growth between shoot and root as a response to WL, but the magnitude of these responses varied among species and among accessions. P. pratense responded less in these traits in response to WL than the other two species. The ability to form aerenchyma in the root cortex in response to WL was found for all species and also varied among species and among accessions, with the highest induction in P. pratense. Interestingly, some accessions were able to maintain and even increase root growth, producing more leaves and tillers, while others showed a reduction in the root system. Shoot dry weight (SDW) was not significantly affected by WL, but some accessions showed different and significant responses in the rate of production of leaves and tillers. Overall correlations between SDW and aerenchyma and between SDW and adventitious root formation were found. This study identified two wild timothy accessions and one wild P. nodosum accession based on shoot and root system growth, aerenchyma formation and having a root anatomy considered to be favorable for WL tolerance. These accessions are interesting genetic resources and candidates for development of climate-resilient timothy varieties.

Keywords: accessions; aerenchyma; anatomy; forage grass; perennial; root.

Figure 1

Root traits in P. pratense, P. nodosum and P. alpinum at 21 days in waterlogging treatment. (A) Root dry weight (RDW). (B) Percent root weight of total plant dry weight (%RDW). (C) Ratio of dry weight of the upper 5 cm of the root system and the dry weight of the total root (5RDW:RDW). The root anatomy traits: (D) root cross section area (RA). (E) Ratio of the cortex area to root cross section area (CEA:RA). (F) Ratio of aerenchyma to root cross section area (AA:RA). (G) Ratio of aerenchyma to cortex area (AA:CEA). (H) Ratio of the cortex to stele (CEA:SA). (I) Ratio of the stele area to root cross section area (SA:RA). Values are means and error bars are standard error (±SE). Significant differences between treatments within each species are represented by * = p < 0.05, ** = p < 0.01 and *** = p < 0.001 according to ANOVA. Differences among species in NWL are indicated by different letters when significantly different according to ANOVA and Tukey’s method, p < 0.05.

Figure 2

PCA based on the shoot dry weight, tiller and leaf production and growth rates of accessions of P. pratense and P. nodosum. P. alpinum was excluded due to no development of generative tillers. Data points represent the difference between WL and NWL. The loadings are shown for the variables of the shoot dry weight (SDW), tiller number (TN), leaf number (LN), the ratios of the tiller number to shoot dry weight (TN:SDW), the leaf number to shoot dry weight (LN:SDW), the leaf number per tiller (LN:TN) and the growth rates of the tiller number (RGR-TN) and leaf number (RGR-LN). The ellipse corresponds to the 90% confidence interval.

Figure 3

Effect of waterlogging on root traits of accessions. (A) Absolute difference between WL and NWL in percent root weight of total plant dry weight (%RDW). (B) Proportional difference in total root dry weight (RDW). (C) Absolute difference in ratio of the dry weight of the upper 5 cm of the root and the dry weight of the total root (5RDW:RDW) in wild (W) and domesticated (D) accessions of P. pratense, P. alpinum and P. nodosum. Values are means and error bars are ±SE. Significant differences between WL and NWL according to t-test are shown with *. Mean values that do not share the same letter are significantly different among accessions within each species according to ANOVA and Tukey’s method, p < 0.05.

Figure 4

Root systems of representative plants showing different responses in non-waterlogging conditions (NWL, plants to the left side) and in waterlogging conditions (WL, plants to the right side). (A) P. pratense, accession D4, (B) P. pratense, accession W6, (C) P. nodosum, accession W9, (D) P. nodosum, accession W7 and (E) P. alpinum, accession W10. Bar is 5 cm.

Figure 5

Effect of waterlogging on root anatomy in wild (W) and domesticated (D) accessions of P. pratense, P. nodosum and P. alpinum. (A) Proportional difference between WL and NWL in the root cross section area (RA). (B) Absolute difference in the ratio of the cortex area to root cross section area (CEA:RA) (C) Absolute difference in the ratio of the aerenchyma to cortex area (AA:CEA). (D) Absolute difference in the cortex to stele area (CEA:SA). (E) Absolute difference in the ratio of the stele to root cross section area (SA:RA). Values are means and error bars are ±SE. Significant differences between WL and NWL according to t-test are shown with *. Mean values that do not share the same letter are significantly different among accessions within each species according to ANOVA and Tukey’s method, p < 0.05.

Figure 6

Distribution of accessions of P. pratense, P. nodosum and P. alpinum based on differences between WL and NWL in the shoot and root traits and the root anatomy traits. (A) PCA where data points represent an accession. For measured traits (SDW, ΔTN, ΔLN and RA), proportional differences between WL and NWL were used, and for the remaining calculated traits (RGR-TN, RGR-LN, %RDW, 5RDW:RDW, AA:CAE, CAE:SA), the absolute differences were used. (B) HCA based on the same traits as for the PCA. Abbreviations can be found in Table 2.

Figure 7

Drawing of the studied shoot and root traits and their abbreviations; shoot dry weight (SDW), number of tillers (TN) and number of leaves (LN) per plant, root dry weight (RDW), dry weight of the upper 5 cm part of the root system (5RDW), area of the root cross section (RA), stele area (SA), cortex area including epidermis (CEA) and aerenchyma area (AA). Measured tissue areas in the root anatomy are indicated in black.