Reassessment of phylogenetic relationships of some members of the Monotropoideae based on partial 28S ribosomal RNA gene sequencing

Abstract

2In 1992, 1994, and 1996, I published papers describing the evolution of the Monotropoideae and Ericaceae (Cullings and Bruns 1992; Cullings 1994; Cullings 1996) using 28S ribosomal RNA (rRNA) gene sequences (Fig. 1). The phylogenetic analysis presented in these manuscripts depicted a polyphyletic Monotropa, and an unexpectedly close relationship between Monotropa hypopithys and Pterospora andromedea. At the time, there were no other molecular treatments of the Monotropoideae for comparison, and morphological and biochemical characters were identified that supported all relationships. In 1997, Kathleen Kron gave a presentation and published an abstract (Kron and Johnson 1997) that depicted monotropoid relationships based on 18S rRNA gene and internal transcribed spacer (ITS) sequences. Kron and Johnson’s data differed from my own in two primary respects. First, the genus Monotropa, while still polyphyletic, was placed not with Pterospora andromedeaand Sarcodes sanguinea, but on the large clade containing Allotropa virgata, Hemitomes congestum, Pityopus californicus, and Monotropa uniflora. Second, Pityopus californicus, rather than being placed with Pleuricospora fimbriolataas in the 28S tree, was depicted as sharing a most recent common ancestor with H. congestum. Based on this information, I re-examined my phylogeny. I obtained four additional M. hypopithys individuals from a second population of this species in Yellowstone Park; both neighbor-joining and cladistic (parsimony) analysis of sequences from the additional four individuals from the second population of M. hypopithys depict a phylogeny in keeping with Dr. Kron’s combined 18S/ITS data (Fig. 1). This casts further doubt on my original data; however I have reason to believe the data were correct. The original two individuals used in Cullings and Bruns (1992) (also from Yellowstone) had been sequenced several times using DNA obtained from at least five independent polymerase chain reaction (PCR) amplifications using both terminal and internal, and univer