Ontogenetic changes in the phenotypic integration and modularity of leaf functional traits

Abstract

Changes in resource availability, functional demands, hormonal regulation and developmental constraints can promote differences in the expression of leaf traits during plant development and foster changes in the targets of natural selection. As a consequence, the pattern and magnitude of covariation among traits, and therefore their phenotypic integration and modularity are equally expected to change throughout ontogeny. However, these changes have not been described yet. We measured leaf economic, defensive and morphological traits in plants of Turnera velutina and estimated the magnitude and pattern of foliar integration and modularity for juvenile and reproductive individuals. In addition, we assessed the relationship between plant biomass and foliar integration within and among ontogenetic stages. Both the pattern and magnitude of foliar integration changed across plant ontogeny. Foliar integration was lower in juvenile than in reproductive plants, and the pattern of phenotypic integration and modularity was different between ontogenetic stages, whereas leaves from juvenile plants showed two functional modules related to plant defence and leaf economy, traits from reproductive plants had greater interconnectivity and hence lower modularity. The relationship between plant biomass and foliar integration was negative within each ontogenetic stage but positive between ontogenetic stages, suggesting that processes intrinsic to plant development influenced the magnitude of foliar integration to a greater extent than plant size. Our findings indicate that plants can change the patterns of covariation among leaf traits during their development. However, a lower foliar integration in juvenile plants could allow for greater lability to explore a multi‐trait phenotypic space, canalisation of leaf attributes along ontogeny should promote greater phenotypic integration, constraining the number of multi‐trait combinations that plants can express. Hence, we suggest that ontogenetic changes in foliar integration allow plants to deal with changing selective dynamics and physiological priorities along their development. A http://onlinelibrary.wiley.com/doi/10.1111/1365-2435.12971/suppinfo is available for this article.

Natural selection acting on integrated phenotypes: covariance among functional leaf traits increases plant fitness

“…LMA was estimated as the dry mass (recorded after oven drying the leaves at 60°C for 72 h) divided by leaf area, which was estimated with a portable scanner (CI-202 CID Inc., Cama, WA, USA). Chl content per unit leaf area (lg cm -2 ) was calculated as the mean value of five measurements (on both sides of the leaf avoiding the middle vein plus a measure in the leaf tip) taken with a CCM-200 Chl content meter (Opti-Sciences, Tyngsboro, MA, USA), using the equation Chl content (lg cm -2 ) = À0.093 + 1.36 9 √(ICC units) (R 2 = 0.79, P < 0.001) (Dami an et al, 2018). ICC is the index of chlorophyll content, which comes from the readings obtained with the CCM-200 chlorophyll content meter (Dami an et al, 2018).…”

“…Chl content per unit leaf area (lg cm -2 ) was calculated as the mean value of five measurements (on both sides of the leaf avoiding the middle vein plus a measure in the leaf tip) taken with a CCM-200 Chl content meter (Opti-Sciences, Tyngsboro, MA, USA), using the equation Chl content (lg cm -2 ) = À0.093 + 1.36 9 √(ICC units) (R 2 = 0.79, P < 0.001) (Dami an et al, 2018). ICC is the index of chlorophyll content, which comes from the readings obtained with the CCM-200 chlorophyll content meter (Dami an et al, 2018). Dry material was ground in a TissueLyser II (Qiagen) and the carbon (d 13 C) isotopes and total nitrogen and carbon content in leaf samples were analyzed by mass spectrometry (Thermo Scientific Delta V Advantage IRMS with EA-2000Flash Elemental analyzer), the d 13 C (&) values are relative to the Pee Dee Belemnite (VPDB).…”

“…PCAs were conducted on R v.3.3.3 using the PHENIX package (R Core Team, 2014; Torices & Muñoz-Pajares, 2015) to calculate INT for each maternal family. Because variation in the correlation structure of traits can result from differences in resource acquisition among individuals of different size (Magwene, 2001;Torices & Muñoz-Pajares, 2015) and/or age (Dami an et al, 2018), INT estimations were based on partial correlation matrices after controlling for plant size, using the length of the main stem recorded at each sampling date. INT values were corrected as INT = (Var (k i ) À (number of traits À 1)/ number of individuals per genetic family) and transformed to percentage of the total maximum integration.…”

Edaphic history over seedling characters predicts integration and plasticity of integration across geologically variable populations of Arabidopsis thaliana

“…Within these broad functional modules, suites of correlated traits may represent submodules under partially independent developmental or genetic control, or with partially independent functions. Such substructuring may facilitate ecological responses to multivariate environments, facilitate functional changes through development, and ultimately influence evolutionary change in overall plant form (e.g., Bissell and Diggle, 2008; Damián et al, 2017). For example, in response to ecologically variable nutrient environments, individuals from populations where access to nutrients such as phosphorus is historically limited may respond to a low‐phosphorus environment with both longer roots and narrower roots to enhance resource acquisition (Hill et al, 2006), whereas populations historically from areas rich in resources may respond by simultaneous reduction in root length and root diameter commensurate with overall resource reduction.…”

“…From an ontogenetic perspective, a recent investigation of leaf physiological, defensive, and morphological traits demonstrated that foliar integration was lower earlier in ontogeny than in reproductively mature plants (Damián et al, 2017). Additionally, only a single functional foliar module was detected in reproductively mature plants, and two submodules in juvenile plants.…”