Maize roots can be colonized by free-living atmospheric nitrogen (N2)-fixing bacteria (diazotrophs). However, the agronomic potential of non-symbiotic N2-fixation in such an economically important species as maize, has still not been fully exploited. A preliminary approach to improve our understanding of the mechanisms controlling the establishment of such N2-fixing associations has been developed, using two maize inbred lines exhibiting different physiological characteristics. The bacterial-plant interaction has been characterized by means of a metabolomic approach. Two established model strains of Nif+ diazotrophic bacteria, Herbaspirillum seropedicae and Azospirillum brasilense and their Nif- couterparts defficient in nitrogenase activity, were used to evaluate the impact of the bacterial inoculation and of N2 fixation on the root and leaf metabolic profiles. The two N2-fixing bacteria have been used to inoculate two genetically distant maize lines (FV252 and FV2), already characterized for their contrasting physiological properties. Using a well-controlled gnotobiotic experimental system that allows inoculation of maize plants with the two diazotrophs in a N-free medium, we demonstrated that both maize lines were efficiently colonized by the two bacterial species. We also showed that in the early stages of plant development, both bacterial strains were able to reduce acetylene, suggesting that they contain functional nitrogenase activity and are able to efficiently fix atmospheric N2 (Fix+). The metabolomic approach allowed the identification of metabolites in the two maize lines that were representative of the N2 fixing plant-bacterial interaction, these included mannitol and to a lesser extend trehalose and isocitrate. Whilst other metabolites such as asparagine, although only exhibiting a small increase in maize roots following bacterial infection, were specific for the two Fix+ bacterial strains, in comparison to their Fix- counterparts. Moreover, a number of metabolites exhibited a maize-genotype specific pattern of accumulation, suggesting that the highly diverse maize genetic resources could be further exploited in terms of beneficial plant-bacterial interactions for optimizing maize growth, with reduced N fertilization inputs.