Capturing and Selecting Senescence Variation in Wheat

Abstract

Senescence is a highly quantitative trait, but in wheat the genetics underpinning senescence regulation remain relatively unknown. To select senescence variation, and ultimately identify novel genetic regulators, accurate characterisation of senescence phenotypes is essential. When investigating senescence, phenotyping efforts often focus on, or are limited to, visual assessment of the flag leaves. However, senescence is a whole plant process, involving remobilisation and translocation of resources into the developing grain. Furthermore, the temporal progression of senescence poses challenges regarding trait quantification and description, whereupon the different models and approaches applied result in varying definitions of apparently similar metrics.To gain a holistic understanding of senescence we phenotyped flag leaf and peduncle senescence progression, alongside grain maturation. Reviewing the literature, we identified techniques commonly applied in quantification of senescence variation and developed simple methods to calculate descriptive and discriminatory metrics. To capture senescence dynamism, we developed the idea of calculating thermal time to different flag leaf senescence scores, for which between year Spearman’s rank correlations of r ≥ 0.59, P < 4.7 × 10−5(TT70), identify as an accurate phenotyping method. Following our experience of senescence trait genetic mapping, we recognised the need for singular metrics capable of discriminating senescence variation, identifying Thermal Time to Flag Leaf Senescence score of 70 (TT70) and Mean Peduncle senescence (MeanPed) scores as most informative. Moreover, grain maturity assessments confirmed a previous association between our staygreen traits and grain fill extension, illustrating trait functionality.Here we review different senescence phenotyping approaches and share our experiences of phenotyping two independent RIL populations segregating for staygreen traits. Together, we direct readers towards senescence phenotyping methods we found most effective, encouraging their use when investigating and discriminating senescence variation of differing genetic bases, and to aid trait selection and weighting in breeding and research programs alike.

Balancing grain yield trade-offs in ‘Miracle-Wheat’

“…S3A). The number of senesced peduncles per 10 peduncles was counted from the inner rows to determine peduncle senescence (%) (Chapman et al, 2021a). In this context, we found a gradient of yellowness in the peduncle across the RILs; however, in the current study, this was not differentiated, i.e.…”

“…Floret number was measured from the non-branching genotypes from two spikelets at the centre of the spike at harvest. Besides, derived traits such as grains per spikelet, grain filling duration (Chapman et al, 2021a), and harvest index was calculated as follows:…”

Unsupervised analysis of NIRS spectra to assess complex plant traits: leaf senescence as a use case

“…As most often the elementary traits are not easily visually observable, extensive works related to the use of some non-destructive and non-invasive methods, largely based on the use of visible and near infrared (VIS-NIR) imaging and spectroscopy, have been undertaken [3]. Thanks to supervised models based on linear or nonlinear signal treatment methods, the calibrations provided relevant predictions with documented accurateness and robustness which can be used as proxies of the key traits targeted [4,5]. The second strategy directly targets the complex trait to document some changes, or to detect some possible discrepancies along its time course; this global approach is frequently used in a wide range of disciplines such as transformation process monitoring, agronomy and phytopathology [6].…”

“…Based on spectra reflectance, numerous supervised models have been extensively used to generate some proxies of these traits to facilitate the collection of phenotype data sets. However these components have different time courses and they interact with the amount of resources available (nutrients and water in particular), demonstrating different senescence dynamics and an environmental dependent response [5]. These studies have shown that there is an intra specific variability in the onset and rate of senescence (for example see [16,19,20].…”

“…The onset of senescence is a complex trait that illustrates the transition of the leaf when grain N demand exceeds soil N uptake, generating translocation of N from stems and leaves and affecting carbon assimilation [21]. A delay in this onset or a lower rate of senescence will have a direct impact on wheat yield [5,[22][23][24].…”