Angiosperm root evolution: the relationship between root anatomy and mycorrhizal colonization among Angiosperm trees. Amber L. Horning, Oscar J. Valverde-Barrantes, Christopher B. Blackwood, Department of Biological Sciences, Kent Sate University, Kent, OH Discussion Introduction Sample Collection and Analysis •Our results suggest divergences in root anatomy and The association between plants and arbuscular • Samples of roots were collected at The Holden Arboretum Statistical Analysis chemistry along evolutionary pathways between mycorrhizal fungi (AMF) is possibly the most prevalent and Boone County Arboretum. Roots from two individuals of •To test whether the distribution of morphological and chemical traits in angiosperms. All variables except stele:root ratio showed a symbiotic relationship in nature. In this relationship, roots 34 species of woody Angiosperms were extracted using a soil angiosperms is structured phylogenetically we used an autocorrelative model significant phylogenetic signal (Fig. 2). Rosids showed rely on AMF for the acquisition of nutrients; in return the corer (10 cm diameter X 15 cm deep). (Moran Index) that tested if the distribution of traits across species is more similar thinner roots, less cortex area, higher lignin content and plant provides carbon and habitat to the fungi. among closely related species than expected by chance. lower N content than magnoliids, whereas asterids showed • Anatomically, plant dependence should be reflected in a • Soil samples were soaked in intermediate values between the other two groups. higher cortex area with respect to the vascular tissue, thus water over night and roots were •Since most root traits show some degree of phylogenetic structure, we employed a •Root diameter was thinner in rosid than in magnoliid , but providing a proportionally larger area for fungal habitat. carefully removed from soil. Root generalized least square (GLS) approach, using phylogenetic correlation structures not stele diameter (Fig 3A). This result indicates that rosids • Chemically, high lignin and systems representing the most as the error structure of the model to test the influence of stele:root ratio and maintain less cortex area on average per unit of root area low nitrogen content are usually distal portion of the roots were lignin:N ratio on AMF colonization. than magnoliids, suggesting less habitat for AMF fungi associated with secondary dissected and ~0.2 g first order roots used for chemical analysis. among rosid trees. growth in roots, which is Results: Phylogenetic Structure correlated with decreased • We measured N content using a Costech elemental analyzer. •Chemically, rosids showed a significantly negative

mycorrhizal colonization. Lignin content was estimated as the residual of a series of relationship between lignin and N levels, whereas the

Lignin_N Mycorrhiza Stele_Root relationship was not significant among other trees (Fig It has been hypothesized that flowering plants methanol, dichloromethane and sulfuric acid digestions Catalpa bignonioides o Ilex opaca o Diospyros virginiana o 3B). This result suggests a strong divergence in root (angiosperms) were initially highly dependent on their following Melillo et al. (1990). o Halesia tetraptera Cornus florida o Asterids Nyssa sylvatica chemistry between rosid trees and other angiosperms. AMF and became more independent as they evolved over Fraxinus americana o o Chionanthus virginicus Syringa reticulata time. However, there are relatively few studies describing Acer saccharum o Aesculus hyppocastanum o •In accordance with our hypotheses, both stele:root and o o Aesculus parvifolia how changes in root anatomy and chemistry occur across Estimation of AMF colonization Koelreuteria paniculata Gleditsia tricanthos o o lignin:N ratios were negatively associated with Styphnolobium japonicum o angiosperm lineages and how this is related to AMF Juglans nigra and Root Anatomy o Prunus sargentii o o o mycorrhizal colonization (Fig 3 C and D). In fact, both Rosids Ulmus americana colonization rates. In this study we evaluated trends in fine Populus deltoides Cercidiphyllum japonicum o factors were significant when included in a single model o Hamamelis virginiana root evolution in angiosperms and the possible consequences •Dry root systems were rehydrated and bleached Liquidambar styraciflua Platanus occidentalis (p= 0.04 and 0.01 for stele:root and lignin:N ratio Magnolia denudata o Angiosperms Magnolia kobus on mycorrhizal colonization using 34 woody species grown sequentially with a solution of 1% KOH and 5% Acetic o Magnolia acuminata o o respectively), suggesting that they function as additive Magnolia stellata Magnolia macrophylla in two common garden experiments. acid. o Magnolia grandifolia o mechanisms of AMF colonization control. Seed plants o o Magnolia virginiana Magnolia tripetala •Samples were then dyed with Trypan blue, dissected and o Liriodendron tulipifera Magnoliids Asimina triloba mounted on to slides. Sassafras albidum •In conclusion, our results support the idea of a decrease in Ginkgo biloba Gymnosperms •Root and stele diameter measurements were taken as well Thuja occidentalis mycorrhizal dependency during angiosperm evolution. Fig. 2. Phylogenetic relationship and trait value distribution for 36 woody species grown in common gardens in Hypothesis/ Purpose as percentage of colonization using a light microscope. Midwest USA. All values refer to first order roots. Associated p-values were 0.45;<0.001; and <0.001 for stele:root ratio; Decreases in root diameter seems to be a common feature lignin:N ratio; and mycorrhizal colonization respectively. Symbols indicated relative trait values for each species, black in more derived groups, although some asterids still squares represent values above the mean and white values below the mean. Brach lengths were standarized to have the Assuming that root systems in angiosperms have evolved to be same length. The gymnosperms Gingko biloba and Thuja occidentalis were used as root group for the tree. maintain thick roots. Additionally, rosids seem to have Gleditsia triacanthos Arbuscules Honey Locust more independent from AMF associations, we expected : Hyphae undergone further chemical changes, possibly associated 1) a decrease in cortex area as a proportion of the root cross Vesicles with early secondary growth that enhance limitations for section in more derived angiosperm lineages. Stele Results: Controls on AMF Colonization AMF colonization. 2) an increase in distal root lignin content, and decrease in Cortex A) C)

nitrogen content, in more derived angiosperm species. -1.8 100 Asterid Magnolid

3) a negative relationship between stele:root and lignin:N ratios -2.0

Rosid 90

with AMF colonization resulting in lower colonization in -2.2 Future Research

80

more derived angiosperm lineages. -2.4

• We will measure AMF colonization using qPCR techniques 70

Examples of Morphological -2.6 log Stele diameter (mm) logStelediameter

Mycorrhizal colonizationMycorrhizal (%) and try to define more accurately the total fungal biomass -2.8 Variation in First Order Roots AvgDiameter F=30.5, p<0.001 60 Stele:Root F=14.3, p<0.001 Clade F=6.03, p<0.01 Clade F=6.17, p<0.01 supported per unit of root biomass. Stele:Root*Clade F=1.98, p= NS Expected Results -3.0 AvgDiameter*Clade F=2.31, p= NS 50 • We will sequence fungal DNA extracted from root samples to -1.0 -0.5 0.0 0.10 0.15 0.20 0.25 Magnolia acuminata log Root diameter (mm) Stele:Root test to what extent root morphology influences AMF Magnoliid community composition. (Thick, heavily colonized roots) B) D)

100 .

1.2 90

1.0 Acknowledgements 80 0.8

Catalpa bignonoides 0.6

N content (%) content N This study was funded by the Department of Biological

Asterid 70 Mycorrhizal colonizationMycorrhizal (%) 0.4 Sciences of Kent State University through the Art and Margaret

Lignin F=12.72, p<0.001 60 Lignin:N F=12.2, p<0.04 Clade F=4.67, p<0.05 Clade F=1.38, p=NS 0.2 Herrick Aquatic Ecology Research Facility Student Research

Lignin*Clade F=1.91, p= NS Lignin:N*Clade F=1.13, p= NS 50

10 20 30 40 50 Grant. A special thanks to Nate Beccue and Charlotte Hewins -1.4 -1.2 -1.0 -0.8 Lignin:N Ulmus americana Lignin content (%) from Holden Arboretum and Kristopher Stone and Josh Selm Rosid Fig. 3. Relationship between root traits and with mycorrhizal colonization for first order roots of 36 woody from Boone Arboretum for their field assistance and lab Figure 1. Expected distribution of root diameter among three (Thin and less extensively colonized) species grown in common gardens in Midwest, USA. A) Relationship between first order stele and root diameter; B) relationship between lignin (measured as non-acid soluble carbon compounds) and N content in collaboration for this study. main clades of angiosperms based on dependency of first order roots; C) effect of stele:root ratio and D) lignin:N ratio on mycorrhizal colonization for first order mycorrhizal association. roots.