Date of Award

Summer 8-15-2015

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Evolution, Ecology & Population Biology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Studying the evolutionary developmental morphology of leaves using next-generation phylogenetics, a candidate gene approach and comparative developmental studies in the plant family Araceae is the overarching theme of the dissertation.

The plant family Araceae is an ancient lineage from the Early Cretaceous and belongs to the monocotyledons. Members of Araceae display striking variation in leaf development; such variation contradicts traditional models of monocot leaf development. Additionally, dissected leaves, which are rare in monocots, seem to have evolved independently multiple times in Araceae by various developmental mechanisms.

Despite extensive efforts to elucidate the evolutionary history of Araceae, phylogenetic ambiguity in the backbone of the tree has precluded answering questions about the early evolution of the family. To depict the sequence of morphological and developmental modifications to leaf ontogeny over time, it is essential to have a strongly supported hypothesis of the evolutionary relationships among species in the family.

To resolve the remaining questions in the deep phylogeny of Araceae a phylogenomic analysis was carried out using next-generation sequencing technology and reference-based assembly of chloroplast and mitochondrial genomes for 37 genera representing 42 of the 44 major clades in the family. Chloroplast sequences produced strongly supported phylogenies in contrast to mitochondrial sequences, which produced poorly supported trees although smaller clades were recovered. The plastid phylogeny obtained from this study is the first for Araceae with a strongly supported backbone and was used for subsequent studies of evolutionary developmental leaf morphology in the family.

Studies of the genetic basis of dissected leaf morphology via blastozone fractionation in plants outside monocots have almost always implicated the action of class I KNOX (KNOX1) genes with one exception - in peas a homolog of the floral meristem gene FLO/LFY is implicated. However, studies of dissected leaf development in monocots, and an examination of the developmental genetics for those monocots that putatively share the blastozone fractionation mechanism are lacking. Two genera in Araceae, Anthurium and Amorphophallus were studied and confirmed to produce lobes and leaflets through blastozone fractionation. To test whether KNOX1 genes are involved in leaf dissection in these genera, immunolocalizations using both a full-length and C-terminus anti-KN1 antibodies were performed on histological sections of developing dissected leaves. KNOX1 protein expression detected by the full-length anti-KN1 antibody and by the C-terminus anti-KN1 antibody was absent and present in developing dissected leaves, respectively. To resolve these conflicting results, an RT-PCR assay was designed to test for the presence of KNOX1 mRNA transcripts during leaf development in Anthurium. Results of the RT-PCR assay support the KNOX1 protein expression pattern seen in immunolocalizations using the C-terminus anti-KN1 antibody. This suggests that monocots share the same genetic mechanism for dissected leaf development with other angiosperms.

Historical models of leaf development posit that structural similarities between monocot and dicot leaves are the result of convergence, although this hypothesis has been contested. Araceae displays both dicot and monocot leaf characters. Previous researchers have remarked on the departure of leaf development in Araceae from traditional models of monocot leaf development. Araceae displays both dicot and monocot leaf characters. To test the hypothesis of a developmentally independent origin of dicot-like leaf characters in monocots, leaf primordium diversity was evaluated in 30 genera of Araceae, along with 36 taxa spanning the angiosperm phylogeny. Leaf primordia were scored for 14 developmental, morphological and anatomical leaf characters. Ancestral character state reconstruction was carried out using the phylogeny obtained from Chapter One, embedded in two contrasting phylogenetic hypotheses of angiosperm evolution. Taxa were plotted in morphospace constructed using the morphological matrix to test whether dicot and monocot leaves occupy similar or different parts of the morphospace. The results of ancestral character state reconstruction and morphospace plotting suggest that at the developmental morphological level, aroid and dicot leaves are homologous. However, at the molecular genetic level, a review of the literature suggests that statements of homology between monocot and dicot leaves must be tested within a framework of the hierarchically organized gene regulatory networks regulating leaf development.

The leaves of Araceae have historically been considered “odd” within monocots. However, the incredible morphological and developmental diversity of leaves in Araceae has provided a powerful study system with which to investigate the unifying aspects of leaf development across angiosperms.


English (en)

Chair and Committee

Allan Larson

Committee Members

Thomas B Croat, Justin C Fay, Elizabeth A Kellogg, Kenneth M Olsen, Peter H Raven, Barbara A Schaal


Permanent URL:

Included in

Biology Commons