Engineering Root Exudation of Lotus toward the Production of Two Novel Carbon Compounds Leads to the Selection of Distinct Microbial Populations in the Rhizosphere P.M. Oger 1,2,4 , H. Mansouri 1 , X. Nesme 3 and Y. Dessaux 1 (1) Institut des Sciences du Ve ´ge ´tal, CNRS, Gif-sur-Yvette F-91198, France (2) Laboratoire de Sciences de la Terre, Ecole Normale Supe ´rieure, Lyon F-69364, France (3) Laboratoire d’Ecologie Microbienne, Universite ´ Lyon I-Claude Bernard, Villeurbanne F-69622, France (4) Present address: Laboratoire de Sciences de la Terre, Ecole Normale Supe ´rieure, 46, Alle ´e d’Italie, F-69364 Lyon cedex 07, France Received: 16 April 2003 / Accepted: 7 July 2003 / Online publication: 31 October 2003 Abstract The culture of opine-producing transgenic Lotus plants induces the increase in the rhizosphere of bacterial communities that are able to utilize these molecules as sole carbon source. We used transgenic Lotus plants producing two opines, namely mannopine and nopaline, to characterize the microbial communities directly in- fluenced by the modification of root exudation. We showed that opine-utilizers represent a large community in the rhizosphere of opine-producing transgenic Lotus. This community is composed of at least 12 different bacterial species, one third of which are able to utilize the opine mannopine and two thirds the opine nopaline. Opine utilizers are diverse, belonging to the Gram-posi- tive and -negative bacteria. We described two novel mannopine-utilizing species, Rhizobium and Duganella spp., and five novel nopaline-utilizing species, Duganella, Afipia, Phyllobacterium, Arthrobacter, and Bosea spp. Although opine utilizers mostly belong to the a-Proteo- bacteria, Rhizobiaceae family, there is little overlap be- tween the populations able to utilize each of the two opines produced by the plants. Noticeably, in the rhizo- sphere of transgenic Lotus, only the opine mannopine favors the growth of Agrobacterium tumefaciens, the bacterium from which opines have been characterized. The diversity of opine utilizers from the rhizosphere of Lotus plants is greater than that observed from any other environment. Therefore, transgenic plants with engi- neered exudation constitute an excellent tool to isolate and characterize specific microbial populations. Introduction The presence in the soil of plant roots and root exudates has a strong structuring effect on soil microbial com- munities. The extent of this plant influence on the soil compartment defines what Hiltner termed the ‘‘rhizo- sphere’’ a century ago [15, 18]. The critical importance of this plant/soil/microbiota interface in terms of plant physiology, plant defense, or plant nutrition is now largely recognized. Metabolic resources, such as organic acids, amino acids, sugars, and vitamins, are key deter- minants in the colonization of the root system by the soil microbiota. The quality and quantity of these carbon and nitrogen compounds released by rhizodeposition greatly varies from one plant species to another, or from one cultivar to another [1, 17, 18, 31, 40]. These variations are mirrored by differences in microbial communities of the rhizosphere [37]. The quality of root exudates also varies according to the growth stage of the plant and its phys- iological status [14], including that of the root system [11]. These variations also impact the structure of the bacterial equilibrium at the local level [2, 3, 30]. Lastly, substantial differences in root colonization can be related to pathogen infections, whether the pathogen indirectly interferes with root exudation via catabolism of exudates or directly induces the production of novel substrates [20]. As early as 1978, Petit et al. suggested to take ad- vantage of the close relationship existing between a plant and its associated microflora to engineer plant root ex- udation [27]. This should provide the microorganism of interest with a selective advantage that may help establish it in the rhizosphere, a strategy later termed ‘‘biased rhizosphere’’ or ‘‘artificial symbiosis’’ [23, 26, 35]. Most of the early data on artificial symbiosis were obtained Correspondence to: P.M. Oger; E-mail: poger@ens-lyon.fr 96 DOI: 10.1007/s00248-003-2012-9 d Volume 47, 96–103 (2004) d Ó Springer-Verlag New York, Inc. 2003