A Fungus Eats a Cyanobacterium: The Story of the Geosiphon pyriformis Endocyanosis

A FUNGUS EATS A CYANOBACTERIUM: THE STORY OF THE GEOSIPHON PYRIFORMIS ENDOCYANOSIS Manfired Kluge ManfredKluge(e-mail: Klugeabio.tu-darmstadt. de), Institut fOr Botanik derTechnischen Universitft Darmstadt, Schnittspahnstrasse 10, D-64287Darmstadt, Germany. INTRODUCTION This article reports on the Geosiphon pynformis Nostoc association, up to now the only known example of an endosymbiosis between a fungus and a cyanobacterium, Nostoc punctfIorme. Included is a general descnrption of the symbiotic system, the taxonomic position of the partners, the initiation and life history of the symbiosis, the structure of the consortium, the metabolic aspects of the sym biosis and the partner relationships. WHAT IS GEOSIPHON? Geosiphon pyriformis (Kiitz.) von Wettstein is a coenocytic soil fungus that lives in endocytobiotic association with a cyanobacterium, Nostoc puncti forme. At the tip of the hyphae the fungus fonns unicellular 'bladders' up to 2mm long and 5mm in diameter (P1. 1),where the cyanobacterial cells are located and where they differentiate and become physiologically active. Thus, Geosiphon represents an example of an endocyanosis. An arable site near Bibergemiind (SpessartMountains, Germany) is at present the only known site worldwide where Geosiphon is naturally abundant. Some particularly interesting features of Geosiphon pyriforrniswil be described briefly in this paper; formore details, see the reviews by Kluge et al. (1997) and SchufBler and Kluge (2001). The case of Geosiphon provides interesting information both about the fungus aril about Nostoc. They belong together and cannot be separated, but the following considerations will focus more on Nostoc as the partner in the Geosiphon symbiosis. TAXONOMIC POSITION OF THE PARTNERS Analysis of the SSU rRNA genes (Gehrig et al. 1996) showed unequivocally that Geosiphon be longs to the Glomales, a fungal order that includes the arbuscular mycorrhiza-forming fungi. It is therefore reasonable to propose that Geosiphon not only forms the endosymbiotic association with Nostoc but also interacts with plant roots to form arbuscular mycorrhiza (AM), though this remains to be proven. The endosymbiont of Geosiphon is Nostoc punctiforme. In iso-osmotic media, the cyanobac terium can be isolated from Geosiphon and culti vated easily without the fungus. There is evidence (D. Mollenhauer and R. Mollenhauer, pers. comm.) that the fungus is capable of incorporating various strains of Nostoc punctiforme, but not all of them lead to functionaly active bladders. Accord ing toD. Mollenhauer (pers. comm.), Nostocpuncti fonne strains derived from other symbiotic systems (e.g. Anthoceros, Blasia and Gunnera) can give rise to an active Geosiphon consortium. On the other hand, some strains of N. punct4formeare not incorp orated by the fungus. It isworth mentioning that Geosiphon is quoted occasionally in textbooks as an example of a lichen. However, in our opinion there is a fundamental difference between Geosiphon and true cyanobacterial lichens: in the Geosiphon system, Nostoc lives endocytobiotically in the fungal cell, whereas in lichens, Nostoc remains outside the fungal cell. INITIATION AND FURTHER DEVELOPMENT OF THE SYMBIOTIC CONSORTIUM In this paper only a short overview of this topic is given; further details can be found inMollenhauer et al. (1996), Kluge et al. (1997) and Schiiffler and Kluge (2001). Initially, the two future partners live independently of each other in the upper layer and on the surface of the soil before they interact to bring about the symbiotic consortium. For success ful interaction with the fungus, Nostoc has to be converted into the early stage of the immobile form, called a primordium. The late primordial stages and the motile trichomes (hormogonia) of Nostoc are not recognised and incorporated by the fungus. Upon contact with the appropriate develop mental stage of Nostoc, the portion of the cyto plasm at the tip of the fungal hypha bulges outwards, surrounding some cells of the cyanobac terial trichome and incorporating them into the fungal cell, presumably by endocytosis. Dunrng the BIOLOGY AND ENVIRONMENT: PROCEEDINGS OF THEROYAL IRISHACADEMY, VOL. 102B, No. 1, 11-14 (2002). 0 ROYAL IRISHACADEMY 11 BIOtLOG(Y AND )EFNVIRO(NMENT _ :Ajbstrate Hyphaee g i _ _ n~~~~~~~~Yug bladders PI aueadyugbadr fGoihnprfrnswthprso h uglmclu pedn ntesbtae The., Noto cell inid th bldesaentvsbei hspoorp poorp oreyo leWl) incorporation of a Nostoc trichome, its heterocysts are not incorporated into the fungus. Binding studies using fluorescence-labelled lectins specific for defined sugar residues have...