PerspectivePlant Science

Nodules and Hormones

See allHide authors and affiliations

Science  05 Jan 2007:
Vol. 315, Issue 5808, pp. 52-53
DOI: 10.1126/science.1137588

Since the advent of the “green revolution,” agriculture has been highly reliant on industrial nitrogen fertilizers to maximize crop production by enhancing the amount of nitrogen in soils. This comes at a heavy price. The industrial fixation of nitrogen accounts for ∼50% of fossil fuel usage in agriculture. The leaching of fertilizers into aquatic systems accounts for increasing plant and algal blooms, now a global problem. The rising cost of nitrogen fertilizers—driven by the rising cost of fossil fuels—and the need for improved sustainability are making alternatives to nitrogen fertilizers ever more attractive. Two reports in this issue, one by Murray et al. on page 101 (1) and the other by Tirichine et al. on page 104 (2), provide insights into a natural replacement for industrial nitrogen fertilizers.

A number of plant species have developed an intimate association with nitrogen-fixing rhizobial bacteria that provide plants with a reliable nitrogen source. Central to this plant-bacterial interaction is the formation of unique organs on plant roots called nodules, which accommodate the bacteria and provide a suitable environment for nitrogenase, the bacterial enzyme responsible for nitrogen fixation. Nodules are plant-derived structures that form following recognition of rhizobial bacteria by the plant. Murray et al. and Tirichine et al. reveal that activation of a plant hormone signaling pathway in the legume Lotus japonicus, most likely by rhizobial bacteria, is sufficient to activate nodule formation. This discovery could pave the way to transferring this symbiotic process into other plant species.

It has long been known that the plant hormone cytokinin is important for nodule development. One of the earliest indicators of this was the observation that transfer of cytokinin production allows normally nonsymbiotic bacteria to activate nodule formation in alfalfa (3). This suggested that increased amounts of cytokinin could induce nodule formation. It was also reported that expression of a gene engineered to be responsive to cytokinin is activated in nodule primordia (4), and suppression of a cytokinin receptor expression by RNA interference, reduces nodulation in the legume Medicago truncatula (5). Curiously, treatment of roots with exogenous cytokinin activates expression of the nodulation gene ENOD40 (6, 7), but fails to induce nodule formation. Thus, cytokinin is involved in nodulation, but whether this plant hormone alone is sufficient to activate nodule formation has not been clear.

Murray et al. and Tirichine et al. resolve this question by demonstrating that activation of a cytokinin receptor, lotus histidine kinase 1 (LHK1), is both necessary and sufficient for nodule organogenesis (see the figure). Both studies use a genetic approach to provide complementary evidence for the central role of this cell surface receptor in nodule development. Murray et al. demonstrate that loss-of-function mutations in the LHK1 gene of L. japonicus abolish nodule primordium formation, but do not affect bacterial invasion of the root. In these mutants, infection threads in root cells, which enable bacterial invasion, become highly elaborated and lose normal targeted growth toward the developing nodule primordia. Conversely, Tirichine et al. show that a gain-of-function mutation in LHK1 leads to spontaneous formation of nodules in the absence of rhizobial bacteria. These mutant plants show additional phenotypes indicative of a constitutive cytokinin response and also show hypersensitivity to exogenous cytokinin.

Nodule organogenesis.

Nod factor, which is released by rhizobial bacteria, is perceived by the plant root epidermis. This triggers a calcium-dependent signaling pathway and the production of the hormone cytokinin. Localized increase in cytokinin may activate cortical cell division, leading to formation of the nodule primordium. Rhizobial bacteria infect the nodule through infection threads that are initiated in root hair cells and invade the developing nodule.


Both studies reveal that LHK1 functions as a cytokinin receptor by expressing this protein in a heterologous system (bacteria or yeast). In these tests, the gain-of-function LHK1 mutant shows constitutive cytokinin-independent activity. Activation of LHK1 thus appears critical for nodule organogenesis, but bacterial entry through infection thread formation appears to be independent of cytokinin signaling.

During nodule formation, there must be coordinated development between the root epidermis and the root cortex. Cortical cell divisions that define the nodule meristem only occur in regions immediately below epidermal cells where infection threads are initiating. Most epidermal responses are activated by the secreted rhizobial signaling molecule Nod factor. This molecule activates a signaling pathway in root epidermal cells that leads to oscillations in the intracellular concentration of calcium. This calcium signal is perceived by a calcium- and calmodulin-dependent protein kinase (CCaMK), and it has recently been shown that, analogous to the cytokinin receptor, activation of CCaMK is sufficient to induce nodule formation in the absence of rhizobial bacteria (8, 9). Hence, the spontaneous formation of nodules can be induced by gain-of-function mutations in both the CCaMK and LHK1 genes. Tirichine et al. show that a plant carrying the gain-of-function mutations in both of these genes generates more nodules than plants with either mutation alone, indicating an additive effect. Hence, it is likely that two separate signaling pathways function in nodulation.

It is therefore very surprising that the LHK1 gain-of-function mutation depends on the gene NSP2 to activate spontaneous nodules. NSP2, a transcriptional regulator, is a component of the Nod factor signaling pathway (10, 11) and is also necessary for CCaMK-induced spontaneous nodulation (9). It appears that even though these two pathways function independently, they converge at NSP2, suggesting that this regulator has dual functions in both Nod factor and cytokinin signal transduction during nodulation.

Perception of Nod factor by the plant is one of the first steps during the interaction between root cells and rhizobial bacteria. Because Nod factor accumulates in cell walls (12), it is highly unlikely that it can traverse the epidermis to induce responses in root cortical cells. It is possible that the localized production of cytokinins follows Nod factor perception, acting as a mechanism to coordinate epidermal and cortical responses during nodule formation. In this model, activation of Nod factor signaling at the epidermis leads to increased localized production of cytokinins. Cytokinins are then transported (most likely by active transport from cell to cell, but the mechanism remains unclear) to cortical cells where they are perceived by LHK1 at the cell surface. This initiates cell division, leading to formation of the nodule primordium. This model predicts that rhizobial bacteria induce cytokinin production in legume roots or redirect cytokinin transport, and also predicts that CCaMK-induced spontaneous nodulation should require LHK1.

Cytokinins are also used by plants that do not form nodules in diverse developmental pathways, including the regulation of root and shoot branching. A key question is whether plants that form nodules have evolved a unique response to cytokinins, or whether the appropriate activation of LHK1 and its orthologs in other plant species is sufficient to induce nodule-like structures. Answering this question will provide insights into the ease of transfer of nodule organogenesis, a first step in transferring this symbiotic interaction into agriculturally important species.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.

Navigate This Article