3D reconstructions of infection thread networks in nodules

One long term goal of our research group is to understand how S. meliloti colonizes the root surface of its host, how it colonizes infection sites, and how it behaves while growing in infection structures (infection threads) inside of the root. We have engineered S. meliloti to produce a variety of fluorescent proteins belonging to the GFP family. Because these bacteria are constitutively fluorescent they can be easily tracked as they grow in soil and in association with roots.

We have learned that while initial infection sites are colonized by many S. meliloti cells, only a few of these cells (typically 5 to 7) go on to populate the resulting infection thread. In addition, we have found that often only descendents of the tipmost bacterium in the infection thread go on to populate the root nodule. Thus following an inoculation with two fluorescently marked strains of S. meliloti, mixed nodules containing both strains occur less frequently that expected based on the observed frequency of infection threads containing both strains. These results have implications for the ecology and evolution of these types of symbiotic bacteria. They indicate that there should be competition for access to the tip of the infection thread, assuming that populating the nodule is advantageous. In addition theoretical work, done by others, suggests that the frequency at which mixed nodules occur may impact the evolution of rhizobial/host symbioses. If nodules are usually populated by one strain of bacteria then host plants may directly select against only those bacterial strains that are not fulfilling their end of the mutualism by restricting oxygen and other resources to nodules not actively fixing nitrogen.

Recently we have begun investigations into how infection threads form networks in nodules. 3-D reconstructions of nodule infection thread networks from serial sections are being used to understand how such networks are organized, and how they change over time.