- American Society of Plant Biologists
Rhizobium-legume symbiosis is a fascinating phenomenon of fundamental importance to natural and agricultural ecosystems. Under conditions of low soil nitrogen, nitrogen-fixing bacteria of the family Rhizobiaceae infect the roots of legumes, induce the formation of root nodules, which house and feed the bacterial symbiont and provide the specialized conditions (such as low oxygen) necessary for nitrogen fixation, and process the fixed nitrogen into amino acids that are used by the host plant. In agricultural systems worldwide, legumes provide an amount of fixed nitrogen roughly equivalent to that produced by the chemical fertilizer industry, and intercropping with legumes is used in many parts of the world to reduce the need for expensive chemical nitrogen inputs.
Nodule development can be divided into three overlapping stages of pre-infection, nodule initiation, and differentiation. In the pre-infection stage, specific flavonoids released by legume roots serve as chemoattractants for the rhizobial symbiont and also activate expression of rhizobial nod genes. nod gene expression produces the Nod factors (certain lipochitin oligosaccharides) that are perceived by a receptor in the legume host, triggering a sequence of events, including curling of root hairs around the invading rhizobia, the entry of the rhizobia into the plant through infection threads, and the induction of cell division in the root cortex that marks formation of the nodule primordium. In addition to their role as chemoattractants for rhizobia in the pre-infection stage, flavonoids produced by the host plant have long been suspected to play a direct role in nodule formation (Hirsch, 1992).
Certain flavonoids have been found to act as auxin transport inhibitors (Stenlid, 1976; Jacobs and Rubery, 1988; Mathesius et al., 1998; Brown et al., 2001). Hirsch (1992) proposed that rhizobial Nod factors induce the synthesis of flavonoids in their hosts that interfere with auxin transport, thereby causing altered hormonal concentrations at the site of infection, leading to cell divisions that produce the nodule primordium. This has been supported by numerous studies. For example, Hirsch et al. (1989) showed that auxin transport inhibitors, such as N-(1-naphthyl)phthalamic acid (NPA) and 2,3,5-triiodobenzoic acid, induced the formation of nodule-like structures on alfalfa roots that resembled Rhizobium-induced nodules in that they contained transcripts for early nodulin genes. Mathesius et al. (1998) found that auxin transport inhibition likely precedes root nodule formation in white clover roots. These authors showed that nodulating bacteria induced a rapid, transient, and local downregulation in expression of an auxin-responsive reporter gene, followed by upregulation of reporter gene expression at the site of nodule formation. de Billy et al. (2001) showed that genes in Medicago truncatula related to the auxin import carrier gene At AUX1, termed Mt LAX, are expressed predominantly in regions of root tips and nodule primordia where vascular tissue arises (central regions for lateral roots and peripheral regions for nodules), and they suggested that auxin is required at two common stages of lateral root and nodule development: development of the primordia and differentiation of the vasculature. Several studies have also shown that genes for enzymes involved in the early stages of flavonoid synthesis are expressed at nodule and lateral root initiation sites. Taken together, these data support the hypothesis that Nod factors bring about an immediate and transient inhibition of polar auxin transport in a highly localized fashion, possibly through the action of specific flavonoids, resulting in a change in auxin concentrations and subsequent stimulation of cell divisions at the site of nodule initiation (Hirsch, 1992; de Billy et al., 2001). de Billy et al. (2001) further suggested that the expression of auxin transporters (encoded by LAX genes) in the region of the nodule primordium subsequently directs auxin transport to the developing nodule, after which normal root polar auxin transport resumes.
In this issue of The Plant Cell, Wasson et al. (pages 1617–1629) provide long-awaited genetic evidence that flavonoids play a key role in the initiation of nodule primordia by inhibiting auxin transport in M. truncatula roots. The authors reduced flavonoid biosynthesis in hairy root cultures of M. truncatula by silencing genes encoding CHALCONE SYNTHASE (CHS) using RNA interference. CHS catalyzes the first committed step of the flavonoid pathway, the synthesis of naringenin chalcone, from which the diverse flavonoid end products are derived (Stafford, 1990). The authors targeted the entire flavonoid pathway because it is not known which flavonoids in legumes are responsible for auxin transport inhibition. They show that silencing of CHS in M. truncatula induced flavonoid deficiency, inhibited nodulation and disabled the inhibition of auxin transport by rhizobia.
Wasson et al. used HPLC, thin layer chromatography, and fluorescence microscopy to examine flavonoid content and showed that decreases in CHS mRNA in RNA interference–transformed roots correlated with a reduction in flavonoids. Silencing the flavonoid pathway had no effect on any visible aspect of root development (i.e., width, length, growth, and appearance), but the CHS-silenced hairy roots were unable to form nodules when inoculated with nodule-forming Sinorhizobium meliloti. Next, the authors supplemented the growth medium with the exogenous flavonoids naringenin and liquiritigenin, which are precursors for the majority of flavonoids and isoflavonoids, respectively, to test if the inability to form nodules was due to the absence of flavonoids or if it were an unrelated secondary effect of silencing CHS. They found the accumulation of flavonoids and nodule formation were restored in the flavonoid-supplemented CHS-silenced root cultures (see figure).
Flavonoids Are Required for Nodule Formation in M. truncatula Hairy Roots.
Autofluorescence (blue) indicates abundant intracellular flavonoids in control hairy roots (top), whereas in CHS-silenced hairy roots (middle), autofluorescence was limited to cell wall fluorescence, indicating the absence of flavonoids, which correlated with an inability to produce nodules when inoculated with S. meliloti (data not shown). CHS-silenced hairy roots supplemented with flavonoid precursors regained capacity for nodule development and accumulated flavonoids (bottom). Top and middle panels show root sections; bottom panel shows section from a nodule. Infection threads of green fluorescent protein–labeled rhizobia can be seen at the arrowhead in the bottom panel. Bars = 100 μM (top and middle panels) and 500 μM (bottom panel).
Wasson et al. also examined auxin transport in CHS-silenced compared with control hairy root cultures. They found that flavonoid-deficient CHS-silenced hairy roots showed significantly higher rates of auxin transport compared with controls before inoculation with rhizobia. Auxin transport, as measured with a tritium-labeled indole-3-acetic acid assay, was significantly reduced in control hairy roots below the point of inoculation by either the auxin transport inhibitor NPA or nodule-forming rhizobia. By contrast, inoculation of CHS-silenced hairy roots with rhizobia did not significantly reduce auxin transport, whereas NPA treatment did. These results showed that the ability of nodule-forming rhizobia to inhibit auxin transport was flavonoid dependent.
Future studies will need to identify the specific flavonoids that function as auxin transport inhibitors during nodule development. It will also be of interest to investigate the role of flavonoids in Lotus, which (like soybean) produces determinate nodules, in contrast with the indeterminate nodules of M. truncatula. Other legumes that have indeterminate nodules include Medicago sativa (alfalfa), clover, pea, and vetch. In addition to lacking a nodule meristem, determinate nodules are distinct from indeterminate nodules in that certain isoflavones, rather than flavones, are the major inducers of nod gene expression. In addition, the organization of cells and cell division patterns differ between the two nodule types.
