c8.7.9 – Leaf shape and size differentiation in white oaks: assessment of allometric relationships among three sympatric species and their hybrids

Viscosi V*, Antonecchia G, Fortini P, Gerber S, Lepais O, Loy A

Collocazione: c8.7.9 – Tipo Comunicazione: Presentazione orale
8° Congresso SISEF *
Sessione 7: “Biodiversità e Risorse genetiche forestali” *

Contatto: Vincenzo Viscosi (vincenzo.viscosi@unimol.it)

Abstract: Leaf allometry is considered to play a key role in plant evolution, growth, survival, ontogeny, ecology and taxonomy, representing an important component of morphological plasticity that facilitates the adaptation of plant organisms to heterogeneous environments and improves the resource acquisition. In this study a combined approach of genetic and morphometric data was used to study leaf shape and size variability of three sympatric and inter-fertile white oak species (Q. frainetto, Q. petraea and Q. pubescens) sampled in a mixed forest in central Italy (Monte Vairano, Molise). At least 30 mature individuals were sampled in nine sampling areas of 50 × 50 meters and for each tree 10 leaves were sampled. A total of 273 oak trees were sampled and genotyped at 12 nuclear microsatellite markers (EST-SSRs) using a PCR 12-plex protocol (Guichoux et al. Mol. Ecol. Res. 2011, 11: 591-611). Assignments of individuals to species or hybrid were performed on the estimate membership coefficient (Q) for each individual in each cluster: we assumed for Q≥0.900 individuals as pure genotype. Leaf shape was analyzed by a configuration of 11 landmarks digitalized on the right half of the leaf (Viscosi et al. Plant Biosyst. 2009, 143: 575-587), while the centroid size, which is mathematically independent of shape, was used as measure of leaf size. The centroid of each landmark configuration was translated to the origin; configurations were scaled to a common unit centroid size and optimally rotated by using the Generalized Procrustes Analysis; leaves of each individual tree were processed to create a mean leaf configuration. Canonical Variate Analysis, run on the weight matrix of the residuals from GPA, showed a clear distinction of Q. frainetto along CV1, while Q. petraea and Q. pubescens are partially distinguished along CV2: test of cross-validation indicated that the 91.0% of specimens were correctly classified. In addition, significant differences in leaf size were found among the three species by means of ANOVA. Multiple regression of Procruste’s coordinates onto centroid size showed a significant contribution of allometry to leaf shape variation in Q. frainetto and Q. petraea, but not for Q. pubescens. In particular, allometry explained the 9.16% and 8.81% of the total leaf shape variability for Q. frainetto and Q. petraea respectively; while for Q. pubescens allometry represented only the 2.18% of the total leaf variation. Allometric variation in leaf morphology was also tested for the hybrids: it explained a significant portion (25.79%) of the total variability. Small leaves characterized Q. petraea, Q. pubescens and their hybrids, while the great leaves belonged to Q. frainetto and its hybrids. MANCOVA on allometric trajectories among species and hybrids indicated no significant differences in slope among pure species, while significant differences in slope were observed between the pure species and the group of hybrids. Hence, for the three pure species, an extension of the MANCOVA was implemented in MorphoJ software to correct the dataset for the allometric component of shape by means of a pooled regression within species, accounting for the 9.55% of total leaf shape variability. The size-corrected data (residuals) were subjected to CVA among the species: it resulted that the three species were significantly discriminated along both CVs and the test of cross-validation indicated 91.5% of trees correctly classified. Concluding, a highly significant leaf shape differentiation among the three species was observed, while the leaf morphology of hybrids was meanly influenced by the genetic dominant species. These evidences indicate that the leaf morphology of hybrids could be informative about the dynamics of hybridization and introgression within mixed oak stands. The analysis of covariation between size and shape revealed that leaf morphology of the pure white oak species is significantly affected by allometric variability, which could be related to the environmental variability of the stands. On the other hand, the high portion of leaf allometric variation recorded for the group of hybrids was suspected to be mainly due to the influence of genetic inheritance of size and shape. Finally, a significant inter-specific variation in leaf shape and size was observed within an area characterized by the same macroclimatic conditions; while the similar leaf allometric patterns of species could be related to a micro-geographical variability of habitats. We suggest that allometric variability could be related to ecological strategies of different species to adapt to similar environmental conditions. Concluding, the possible effects of allometry may help us to interpret the significance of leaf oak variation among and within species in their natural environments, defining the important role of leaf size and shape in adaptive and evolutionary processes in white oaks.

Parole chiave: Allometry, Genetic Assignment, Leaf Shape, Leaf Size, Mancova, Quercus

Citazione: Viscosi V, Antonecchia G, Fortini P, Gerber S, Lepais O, Loy A (2011). Leaf shape and size differentiation in white oaks: assessment of allometric relationships among three sympatric species and their hybrids . 8° Congresso Nazionale SISEF, Rende (CS), 04 – 07 Ott 2011, Contributo no. #c8.7.9