Report

Host-to-Parasite Gene Transfer in Flowering Plants: Phylogenetic Evidence from Malpighiales

See allHide authors and affiliations

Science  30 Jul 2004:
Vol. 305, Issue 5684, pp. 676-678
DOI: 10.1126/science.1100671

Abstract

Horizontal gene transfer (HGT) between sexually unrelated species has recently been documented for higher plants, but mechanistic explanations for HGTs have remained speculative. We show that a parasitic relationship may facilitate HGT between flowering plants. The endophytic parasites Rafflesiaceae are placed in the diverse order Malpighiales. Our multigene phylogenetic analyses of Malpighiales show that mitochrodrial (matR) and nuclear loci (18S ribosomal DNA and PHYC) place Rafflesiaceae in Malpighiales, perhaps near Ochnaceae/Clusiaceae. Mitochondrial nad1B-C, however, groups them within Vitaceae, near their obligate host Tetrastigma. These discordant phylogenetic hypotheses strongly suggest that part of the mitochondrial genome in Rafflesiaceae was acquired via HGT from their hosts.

Malpighiales are one of the most diverse clades of flowering plants uncovered in recent phylogenetic analyses. The order comprises 27 families (1) previously assigned to 13 different orders (2), including more than 16,000 species spanning tremendous morphological and ecological diversity (3). Recent surprising additions to Malpighiales are the endophytic holoparasites Rafflesiaceae (4), which lack leaves, stems, and roots, and rely entirely on their host plants, species of Tetrastigma (Vitaceae), for their nutrition. Despite their extreme vegetative reduction, they are unmistakable in flower, producing the largest flowers in the world, which mimic rotting flesh—an enticement to the carrion flies that pollinate them (5).

Barkman et al. (4) used mitochondrial (mt) matR sequences to place Rafflesiaceae firmly with Malpighiales [100% bootstrap percentage (BP)]. Their use of a single mt gene was appropriate in a family that has resisted placement with standard genetic loci. To further examine this placement, we obtained sequences representing all families of Malpighiales, all genera of Rafflesiaceae, and numerous basal eudicots for four loci from the mt and nuclear genomes (6). Low-copy nuclear genes are an underused resource for resolving the placement of problematic taxa, and phytochrome C (PHYC), as used here, has been useful for revealing relationships within Malpighiales (7).

Our phylogenetic analyses are summarized in Fig. 1 (8). The tree created from the matR and nuclear loci firmly (100% BP) place Rafflesiaceae within Malpighiales. In contrast, the mt locus nad1B-C suggests that Rafflesiacecae are not members of Malpighiales but belong (100% BP) in Vitaceae near their host Tetrastigma. Each of these mutually exclusive hypotheses cannot be attributable to contamination (9), and each receives strong support from parsimony analyses and from alternative topology tests.

Fig. 1.

Two conflicting hypotheses about the phylogenetic placement of Rafflesiaceae. (A) The strict consensus of 136 angiosperms for combined mt matR and nuclear (PHYC and ribosomal 18S) data showing a well-supported (100% BP) Malpighiales clade (in blue), which includes all members of the order sensu APG II (1) plus Rafflesiaceae (in red; Rafflesia, Rhizanthes, and Sapria). (B) The strict consensus of 147 angiosperms for mt nad1B-C (the nad1 intron 2 and part of the adjacent exons b and c) showing a well-supported (100% BP) Malpighiales clade, which includes all members of the order except Rafflesiaceae. Rafflesiaceae (Rafflesia and Sapria) are strongly placed (100% BP) in the basal eudicot family Vitaceae (in yellow) near their host genus, Tetrastigma. The dashed line is the hypothesized host/parasite HGT.

Which of these conflicting hypotheses reflect the true species affinities of Rafflesiaceae? Vitaceae possess several synapomorphies that are rare among angiosperms, including sieve-tube plastids with starch and protein inclusions, pearl glands, stamens opposite the petals, and seeds with a cordlike raphe. If Rafflesiaceae were embedded in Vitaceae, as suggested by nad1B-C, we would expect species to possess at least some of these characters, but they do not (2, 3). A definitive malpighialean sister group for Rafflesiaceae is unclear, given our data. However, the closest relatives suggested in the combined analysis (10), Ochnaceae and Clusiaceae sensu lato, share tenuinucellate ovules (among mostly crassinucellate relatives) and staminal fusion with Rafflesiaceae (2, 3).

The position of Rafflesiaceae based on nad1B-C provides a new example of horizontal gene transfer. If nad1B-C were vertically transmitted, as we believe to be the case for the other loci, we would expect Rafflesiaceae to group with Malpighiales. Instead, phylogenetic evidence from nad1B-C suggests that part of the mt genome in Rafflesiaceae originated from their hosts, Tetrastigma (either stem or crown group members), and was horizontally transferred to these obligate parasites. A similar horizontal gene transfer (HGT) of nad1B-C was recently reported (11) in seed plants, involving a transfer from an asterid to Gnetum. And Bergthorsson et al.(12) have documented several instances of mt HGT between distantly related angiosperm groups.

The underlying mechanism for HGT between sexually unrelated plants, however, has been elusive. Various pathogens have been suggested as primary vector agents (11, 12). Our study documents a case in which there is no need to propose an intermediary vector for HGT. In these plants, the transfer appears to have been facilitated by the intimacy of the association between the host and the endophytic parasite, which lives its whole vegetative life as “an almost mycelial haustorial system,” “ramifying and anastomosing throughout the [tissues of the] host” (13). This pattern may be an important mechanism by which parasites assemble their genetic architecture, and additional cases of HGT should be sought among other endophytic parasites and their hosts. It will also not be surprising if reciprocal genetic transfers are found to have occurred, from parasite to host.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1100671/DC1

Materials and Methods

Figs. S1 to S5

References

References and Notes

View Abstract

Navigate This Article