Plant Genetics for Study of the Roles of Root Exudates and Microbes in the Soil

Abstract

Although plants can be grown in sterile soil in aseptic growth chambers, their natural lives involve an intense and intimate interaction with a vast number of microbes, especially those found in soils. The number of different bacterial species in a single gram of soil has been estimated to be anywhere from a few thousand to many millions, depending on the soil source and the method of analysis, with still-undescribed species making up a large share of the total. In addition to eubacteria and archaebacteria, many species of fungi, protists, and algae are also found in the soil, often in association with plant roots. The great majority of these soil microbes have not been studied to any significant degree, partly because conditions for their axenic culture have not been developed. For instance, only 26 of the approximately 52 identified major lineages, or phyla, within the domain Bacteria have cultured representatives. In fact, it is estimated that less than 1% of the bacterial species in the soil could be grown in culture with current approaches, and this number is certain to be much lower if one considers that most rare microbial components of the soil are completely unknown. Plants actively secrete very large quantities, and a great diversity, of organic compounds into the soil. Exudation of anywhere from 5 to 60% of total photoassimilate has been reported and found to be highly variable across environmental conditions (e.g., soil type, time of day, soil moisture, temperature) and plant genotype or growth stage. The roles of only a few of these compounds are known or guessed at. Citrate is secreted, sometimes in very large quantities, to help acidify the soil and thereby promote root growth, and this compound also helps bind aluminum in the soil, thereby decreasing its phytotoxic effects. Some plants have been shown to exude phenolic compounds that exhibit allelopathic effects like the sorghum exudate sorgoleone that is an inhibitor of broadleaf and grass weeds at concentrations as low as 10 mM in hydroponic assays. Many other compounds, such as amino acids and sugars, are believed to be secreted by plant roots in order to promote rhizosphere microbial growth, although the value to the plant of 1% of the rhizosphere microbes are not known in any system. Specific secreted phenolic compounds have been shown to be signal molecules that attract root colonization by useful microbes, nitrogen-fixing bacteria such as Rhizobium, and mycorrhizal fungi. The question remains, what do most of these soil microbes do? The active secretion of so much of the fixed carbon produced by a plant suggests that these microbes are very important to the plants, but this idea is challenged by the observation that plants can grow efficiently in sterile soil. Of course, plants that are grown with fertilizers in a controlled environment do not need symbiotic relationships that yield limiting growth substances, like the fixed nitrogen provided by rhizobia or the phosphate access provided by mycorrhizae. Perhaps, a more frequent value of rhizosphere microbial associations to a plant is exemplified in the “take-all” disease, where the Gaeumannomyces graminis var. tritici fungus that infects wheat roots is overcome in the soil by a beneficial bacterial competitor, a specific isolate of Pseudomonas fluorescens. Unlike sterile soil, potential microbial pathogens in field soil may exist in staggering numbers and variety, and only attraction of beneficial or neutral microbial competitors of these pathogens to the rhizosphere would provide comprehensive protection to host plants. In the absence of the ability to grow most soil microbes in pure culture, it is difficult to test their possible contributions to plant growth or plant disease. One cannot simply inoculate the soil with a single microbe and see its effects on a potential host plant if one cannot first grow that microbe. However, we have postulated that we can use our control over host plant genetics to accomplish the same goals of understanding the roles of microbes in the soil. If one can find mutations in plants, or segregating natural variation, which determines the presence/absence or abundance of specific rhizosphere microbes, then this demonstrates a specific relationship between the product of the mutated or varying plant gene(s) and the biology of the affected microbe. For instance, if one finds a natural variation for a low level of sorgoleone production, and sees that this causes the root to no longer be colonized by mycorrhizae, then this indicates that sorgoleone is involved in mycorrhizal colonization. We have been pursuing this approach to use plant host genetics to dissect plant–microbe interactions in the soil for the last 10 years. This research has proceeded very slowly because of the need to establish a foundation for the experiments, a very limited tool set, a challenging level of environmental variation in the experiments, a surprisingly low level of plant genetic variation for rhizosphere exudates (at least in Arabidopsis thaliana, see below), and the lack of funding for such research in the absence of compelling preliminary results. However, recent advances in DNA sequencing technology have offered the possibility that studies of plant genetic control of microbial interactions in the rhizosphere and root can be analyzed comprehensively. This chapter describes our initial results with the genetic and metagenomic analysis of these interactions.