doi:10.1016/j.ympev.2008.03.028 
Copyright © 2008 Elsevier Inc. All rights reserved.
Phylogeny of the basal angiosperm genus Pseuduvaria (Annonaceae) inferred from five chloroplast DNA regions, with interpretation of morphological character evolution
Yvonne C.F. Sua, Gavin J.D. Smithb and Richard M.K. Saundersa, 
, 
aDivision of Ecology & Biodiversity, School of Biological Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong, China
bDepartment of Microbiology, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong, China
Received 31 October 2007;
revised 5 March 2008;
accepted 11 March 2008.
Available online 23 March 2008.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Phylogenetic relationships within the magnoliid basal angiosperm genus Pseuduvaria (Annonaceae) are investigated using chloroplast DNA sequences from five regions: psbA-trnH spacer, trnL-F, matK, rbcL, and atpB-rbcL spacer. Over 4000 nucleotides from 51 species (of the total 53) were sequenced. The five cpDNA datasets were analyzed separately and in combination using maximum parsimony (MP), maximum likelihood (ML), and Bayesian methods. The phylogenetic trees constructed using all three phylogenetic methods, based on the combined data, strongly support the monophyly of Pseuduvaria following the inclusion of Craibella phuyensis. The trees generated using MP were less well resolved, but relationships are similar to those obtained using the other methods. ML and Bayesian analyses recovered trees with short branch lengths, showing five main clades. This study highlights the evolutionary changes in seven selected morphological characters (floral sex, stamen and carpel numbers, inner petal color, presence of inner petal glands, flowering peduncle length, and monocarp size). Although floral unisexuality is ancestral within the genus, several evolutionary lineages reveal reversal to bisexuality. Other phylogenetic transitions include the evolution of sapromyophily, and fruit-bat frugivory and seed dispersal, thus allowing a wide range of adaptations for species survival.
Keywords: Annonaceae; Pseuduvaria; Chloroplast DNA; Phylogeny; Magnoliid; Morphological character evolution
Fig. 1. Flowers and fruits of selected Pseuduvaria species. (A) Flowers of P. rugosa. (B) Flowers of P. reticulata. (C) Flower of P. setosa. (D) Flowers of P. clemensiae. (E) Fruits of P. megalopus. Photographs: A–C, Pindar Sidisunthorn and Simon Gardner; D and E, Wayne Takeuchi, reprinted from Su and Saunders (2006).
Fig. 2. Strict consensus of 2600 equally most parsimonious trees of Pseuduvaria resulting from maximum parsimony (MP) analysis using combined data of psbA-trnH spacer, matK, rbcL, trnL-F, and atpB-rbcL spacer (4473 aligned nucleotide positions of which 212 are parsimony informative; tree length = 937 steps; CI = 0.828; RI = 0.759). Numbers above the branches are MP bootstrap values over 50%.
Fig. 3. Phylogeny of the genus Pseuduvaria. (A) Maximum likelihood (ML) and (B) Bayesian trees generated under a homogeneous evolutionary model (GTR + I + G) using combined data of psbA-trnH spacer, trnL-F, matK, rbcL, and atpB-rbcL spacer. The Bayesian tree is a consensus of 76,000 trees following removal of 24,000 trees as burn-in. Numbers above the branches are ML bootstrap values and posterior probabilities on the ML and Bayesian trees, respectively. Geographical distribution: AMER = America; AUS = Australia; BOR = Borneo; CH = China; MYAN = Myanmar; MP = Peninsular Malaysia; NG = New Guinea; PHI = Philippines; SL = Sri Lanka; SUL = Sulawesi; SUM = Sumatra; THAI = Thailand; VIET = Vietnam; and W MALESIA = West Malesia.
Fig. 4. Bayesian tree generated under a partitioned model using combined data of psbA-trnH spacer, trnL-F, matK, rbcL, and atpB-rbcL spacer. The tree is a consensus of 90,000 trees following removal of 10,000 trees as burn-in. Numbers above the branches are posterior probabilities.
Fig. 5. Phylogram of the maximum likelihood (ML) topology, showing the number of substitutions per site for each branch.
Fig. 6. Evolution of selected characters mapped on one of the most parsimonious trees obtained in the combined five-region analysis. (A) Floral sex. (B) Stamen number. Drawings by Ngai Yuen Yi, reproduced from Su and Saunders (2006).
Fig. 7. Evolution of selected characters mapped on one of the most parsimonious trees obtained in the combined five-region analysis. (A) Carpel number. (B) Inner petal color.
Fig. 8. Evolution of selected characters mapped on one of the most parsimonious trees obtained in the combined five-region analysis. (A) Inner petal gland. (B) Flowering peduncle length. Drawings by Ngai Yuen Yi, reproduced from Su and Saunders (2006).
Fig. 9. Evolution of monocarp size mapped on one of the most parsimonious trees obtained in the combined five-region analysis. Drawings by Ngai Yuen Yi, reproduced from Su and Saunders (2006).
Table 1.
Information regarding species sampling, localities, collection dates, and GenBank accession numbers for sequences used in the analyses
Note: Asterisk indicates sequence obtained from GenBank; 1leaf samples collected in silica gel. Pseuduvaria nomenclature follows that of Su and Saunders (2006). Herbarium acronyms follow Holmgren et al. (2008).
Table 2.
List of primers used for amplification and sequencing of five DNA regions
Note: Numbers in primer names correspond to their locations relative to Pseuduvaria sequences. Symbols: F = forward, R = reverse.
Table 3.
P-values from incongruence length difference (ILD) test
Significant incongruence at P < 0.01.
Table 4.
Matrix of morphological character states used for MacClade reconstructions
Morphological characters: (1) Floral sex: 0 = bisexual; 1 = unisexual; 2 = staminate and bisexual flowers. (2) Stamen number per flower: 0 = fewer than 24; 1 = 26–80; 2 = 85–155 [character inapplicable for solely bisexual species]. (3) Carpel number per flower: 0 = solitary; 1 = 2–18; 3 = more than 22 [character inapplicable for solely bisexual species]. (4) Inner petal color: 0 = cream or light yellow; 1 = dark purple or red. (5) Inner petal gland: 0 = absent; 1 = discrete with one or paired; 2 = diffuse. (6) Flowering peduncle length: 0 = inconspicuous, or up to 7 mm long; 1 = 10–280 mm long. (7) Monocarp size: (0) <20 mm long; (1) >21 mm long.
Table 5.
Tree statistics of the psbA-trnH spacer, trnL-F, matK, rbcL, atpB-rbcL spacer, and combined datasets from maximum parsimony (MP) analysis
a Inclusive of outgroups.
Table 6.
Best-fitting models and parameter values for separate (psbA-trnH spacer, trnL-F, matK, rbcL, atpB-rbcL spacer) and combined datasets in this study
a Substitution models: GTR, General-time-reversible (
Yang, 1994); HKY, Hasegawa–Kishino–Yano (
Hasegawa et al., 1985). Ti/Tv, transition–transversion ratio; I, proportion of invariable sites; G, gamma distribution.
Abstract
Purchase PDF (2743 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (0)
Purchase PDF (154 K)
