
Molecular phylogenetics of Alternanthera (Gomphrenoideae, Amaranthaceae): resolving a complex taxonomic history caused by different interpretations of morphological characters in a lineage with C 4 and C 3 –C 4 intermediate species
Author(s) -
SÁNCHEZDEL PINO IVONNE,
MOTLEY TIMOTHY J.,
BORSCH THOMAS
Publication year - 2012
Publication title -
botanical journal of the linnean society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.872
H-Index - 68
eISSN - 1095-8339
pISSN - 0024-4074
DOI - 10.1111/j.1095-8339.2012.01248.x
Subject(s) - biology , clade , lineage (genetic) , phylogenetic tree , maximum parsimony , botany , molecular phylogenetics , phylogenetics , evolutionary biology , zoology , genetics , gene
Alternanthera (Amaranthaceae) is a diverse genus (80–200 species) largely restricted to the American Tropics. With Pedersenia and Tidestromia , it makes up the ‘Alternantheroid clade’ in Gomphrenoideae. Parsimony and Bayesian analyses of nucleotide sequences of nuclear (ITS) and plastid ( rpl16 , trnL‐F ) and morphological characters identify that the capitate stigma of Alternanthera is a synapomorpy within the Alternantheroids. Within Alternanthera , two major clades were resolved, both of which were marked by otherwise homoplasious characters of the gynoecium: Clade A [99% jackknife (JK); 1.0 posterior probability (PP)] with nine species and Clade B (60% JK; 0.98 PP) with 22 species. Four subclades (B1–B4), strongly supported statistically, were identified in Clade B. Previous subgeneric classifications of Alternanthera appear artificial in light of our new molecular phylogenetic analyses. Most major lineages are congruently resolved by nuclear and plastid data but some incongruence between the nrITS and plastid phylogenetic trees suggests hybridization may have played a role in the rampant speciation in Alternanthera . Whereas C 4 photosynthesis appears to have evolved in a single clade, the position of A. littoralis var. maritima (C 3 ) in this clade may be explained by hybrid speciation rather than a reversal from C 4 to C 3 . All C 3 –C 4 intermediates belong to a different clade that also contains C 3 species, but species limits, including the widely studied A. tenella , are unclear. The clade including A. tenella and A. halimifolia contains most of the species endemic to the Galápagos whereas A. nesiotes , also endemic to the islands, is nested among widespread American taxa. This suggests that the Galápagos radiation of Alternanthera may have arisen from at least two independent colonization events followed by a subsequent radiation in the former lineage. © 2012 The Linnean Society of London, Botanical Journal of the Linnean Society , 2012, 169 , 493–517.