Evolution of mycoheterotrophy in Polygalaceae: The case of Epirixanthes.

The mycoheterotrophic lifestyle has enabled some plant lineages to obtain carbon from their mycorrhizal symbionts. The mycoheterotrophic genus Epirixanthes (Polygalaceae) consists of six species from tropical Asia. Although it is probably closely related to the chlorophyllous genus Salomonia and linked to arbuscular mycorrhizal fungi, lack of DNA sequence data has thus far prevented these hypotheses from being tested. Therefore, the evolutionary history of Epirixanthes remains largely unknown. We reconstructed the phylogenetic relationships of Epirixanthes based on nuclear ITS and plastid matK data. Divergence times were inferred using a Bayesian relaxed clock approach, and we phylogenetically analyzed its mycorrhizal symbionts. We furthermore assigned these symbionts to operational taxonomic units, compared them with symbionts of other Polygalaceae, and measured their phylogenetic diversity. We found that Epirixanthes is placed in tribe Polygaleae as sister to Salomonia. Epirixanthes has a Miocene-Oligocene stem age and grows exclusively in symbiosis with fungi of Glomeraceae. Salomonia and some Polygala species are linked to both Glomeraceae and Acaulosporaceae, resulting in higher phylogenetic diversity values. The majority of the symbionts of Epirixanthes are not found in Salomonia or Polygala, although a few shared fungal taxa are found. Epirixanthes forms a relatively young mycoheterotrophic lineage. The Oligocene-Miocene origin suggests its evolution was influenced by the environmental dynamics in Southeast Asia during this time. Although comparison of fungi from Epirixanthes with those from Salomonia and Polygala suggests some specialization, many other mycoheterotrophic plants are linked to a more narrow set of Glomeraceae.

Mennes CB ，Moerland MS ，Rath M ，Smets EF ，Merckx VS ... -  《-》

Mycoheterotrophic Epirixanthes (Polygalaceae) has a typical angiosperm mitogenome but unorthodox plastid genomes.

Fully mycoheterotrophic plants derive carbon and other nutrients from root-associated fungi and have lost the ability to photosynthesize. While mycoheterotroph plastomes are often degraded compared with green plants, the effect of this unusual symbiosis on mitochondrial genome evolution is unknown. By providing the first complete organelle genome data from Polygalaceae, one of only three eudicot families that developed mycoheterotrophy, we explore how both organellar genomes evolved after loss of photosynthesis. We sequenced and assembled four complete plastid genomes and a mitochondrial genome from species of Polygalaceae, focusing on non-photosynthetic Epirixanthes. We compared these genomes with those of other mycoheterotroph and parasitic plant lineages, and assessed whether organelle genes in Epirixanthes experienced relaxed or intensified selection compared with autotrophic relatives. Plastomes of two species of Epirixanthes have become substantially degraded compared with that of autotrophic Polygala. Although the lack of photosynthesis is presumably homologous in the genus, the surveyed Epirixanthes species have marked differences in terms of plastome size, structural rearrangements, gene content and substitution rates. Remarkably, both apparently replaced a canonical plastid inverted repeat with large directly repeated sequences. The mitogenome of E. elongata incorporated a considerable number of fossilized plastid genes, by intracellular transfer from an ancestor with a less degraded plastome. Both plastid and mitochondrial genes in E. elongata have increased substitution rates, but the plastid genes of E. pallida do not. Despite this, both species have similar selection patterns operating on plastid housekeeping genes. Plastome evolution largely fits with patterns of gene degradation seen in other heterotrophic plants, but includes highly unusual directly duplicated regions. The causes of rate elevation in the sequenced Epirixanthes mitogenome and of rate differences in plastomes of related mycoheterotrophic species are not currently understood.

Petersen G ，Darby H ，Lam VKY ，Pedersen HÆ ，Merckx VSFT ，Zervas A ，Seberg O ，Graham SW ... -  《-》

Mycoheterotrophic interactions are not limited to a narrow phylogenetic range of arbuscular mycorrhizal fungi.

The majority of achlorophyllous mycoheterotrophic plant species associate with arbuscular mycorrhizal fungi (AMF). Previous studies have shown that some species are highly specialized towards narrow lineages of AMF and have suggested that only particular lineages of these fungi are targeted by mycoheterotrophic plants. To test this hypothesis, we analyzed all available partial SSU sequences of AMF associated with mycoheterotrophic plants including data from 13 additional specimens from French Guiana, Gabon and Australia. Sequences were assigned to 'virtual taxa' (VT) according to the MaarjAM database. We found that 20% of all known Glomeromycota VT are involved in mycoheterotrophic interactions and the majority of associations involve Glomeraceae (Glomus Group A) fungi. While some mycoheterotrophic plant species have been found growing with only a single VT, many species are able to associate with a wide range of AMF. We calculated significant phylogenetic clustering of Glomeromycota VT involved in mycoheterotrophic interactions, suggesting that associations between mycoheterotrophic plants and AMF are influenced by the phylogenetic relationships of the fungi. Our results demonstrate that many lineages of AMF are prone to exploitation by mycoheterotrophic plants. However, mycoheterotrophs from different plant lineages and different geographical regions tend to be dependent on lineages of AMF that are phylogenetically related.

Merckx VS ，Janssens SB ，Hynson NA ，Specht CD ，Bruns TD ，Smets EF ... -  《-》

Morpho-anatomical and molecular characterization of the mycorrhizas of European Polygala species.

The mycorrhizas of 12 species of Polygala (Polygalaceae), including herbs, subshrubs and one shrub, collected from Germany, Mallorca (Spain) and Malta, were investigated by morpho-anatomical and molecular methods. Aseptate hyphae, arbuscules and vesicles indicate an arbuscular mycorrhiza in all species examined. Hyphal spread in Polygala is predominantly, but not exclusively, intracellular and comprises three characteristic stages of colonization: (i) intracellular, linear hyphal growth in a cascading manner after penetration towards the penultimate parenchyma layer (layer 2), (ii) initially linear hyphal growth in the cells of layer 2 from where hyphal branches repeatedly penetrate the anatomically distinct innermost parenchyma layer (layer 1), forming arbuscule-like structures therein which are subject to degeneration, (iii) more branches from the linear hyphae in layer 2 develop, but coil and make contact to the layer outside layer 2 (layer 3) in which arbuscule-like structures similar to those in layer 1 form and degenerate. This general colonization pattern differs in details between the species, and critical comparisons, in particular between the woody P. myrtifolia, the herbaceous Polygala spp. and the mycoheterotrophic Epirixanthes spp. (Polygalaceae) suggest an evolutionary shift of mycorrhizal features within the family towards an optimization of plant benefit through the fungus. Based on the molecular marker 18S rDNA mycorrhizal fungi detected in roots of Polygala spp. are largely restricted to five clades of Glomeraceae 1 (Glomus Group A). This result rejects the hypothesis of a strict symbiotic specificity in Polygalaceae but may stimulate a discussion on functionally compatible groups of fungi.

Rath M ，Weber HC ，Imhof S 《-》

Specific arbuscular mycorrhizal fungi associated with non-photosynthetic Petrosavia sakuraii (Petrosaviaceae).

Mycorrhizal fungi in roots of the achlorophyllous Petrosavia sakuraii (Petrosaviaceae) were identified by molecular methods. Habitats examined were plantations of the Japanese cypress Chamaecyparis obtusa in Honshu, an evergreen broad-leaved forest in Amami Island in Japan and a mixed deciduous and evergreen forest in China. Aseptate hyphal coils were observed in root cortical cells of P. sakuraii, suggesting Paris-type arbuscular mycorrhiza (AM). Furthermore, hyphal coils that had degenerated to amorphous clumps were found in various layers of the root cortex. Despite extensive sampling of P. sakuraii from various sites in Japan and China, most of the obtained AM fungal sequences of the nuclear small subunit ribosomal RNA gene were nearly identical and phylogenetic analysis revealed that they formed a single clade in the Glomus group A lineage. This suggests that the symbiotic relationship is highly specific. AM fungi of P. sakuraii were phylogenetically different from those previously detected in the roots of some mycoheterotrophic plants. In a habitat in C. obtusa plantation, approximately half of the AM fungi detected in roots of C. obtusa surrounding P. sakuraii belonged to the same clade as that of P. sakuraii. This indicates that particular AM fungi are selected by P. sakuraii from diverse indigenous AM fungi. The same AM fungi can colonize both plant species, and photosynthates of C. obtusa may be supplied to P. sakuraii through a shared AM fungal mycelial network. Although C. obtusa plantations are widely distributed throughout Japan, P. petrosavia is a rare plant species, probably because of its high specificity towards particular AM fungi.

Yamato M ，Yagame T ，Shimomura N ，Iwase K ，Takahashi H ，Ogura-Tsujita Y ，Yukawa T ... -  《-》