Plant–fungal associations

Cue for the branching connection

Article metrics

Plant roots release potent molecules that activate symbiotic fungi and initiate a harmonious relationship. It turns out that the same compounds are detected by parasitic weeds for less benign purposes.

Of all the fungi, it is a group known as the Glomeromycota that arguably has the greatest claim to public notice. The reason is that its members form symbiotic associations known as arbuscular mycorrhiza with plant roots, and without this symbiosis our planet's flora would have a very different constitution. The latest episode in understanding how plants communicate with their fungal partners appears on page 824 of this issue1. There, Akiyama et al. describe the characterization of a factor that is released by the plant as a component of the symbiotic dialogue.

It seems likely that arbuscular mycorrhiza had a key role in shaping life on Earth by helping the first primitive plants to colonize land2,3. Plant biomass constitutes the base of food chains, and was the prerequisite for the evolution of land-based animals and ultimately for our own existence. Early land plants did not have elaborate root systems for extracting nutrients from the soil, and therefore probably relied on fungal symbioses to explore the substrate and deliver those nutrients. In return, the fungi obtained carbohydrates produced by photosynthesis. Because arbuscular mycorrhiza are so beneficial, all major plant lineages have retained the association — more than 80% of higher plant species form a symbiotic relationship with these fungi, which are found in most terrestrial ecosystems4.

Probably as a consequence of millions of years of coevolution, arbuscular mycorrhizal fungi rely so much on the interaction with plants that they cannot be grown in isolation. A spore buried in the ground can only ramify through the branching threads known as hyphae if it encounters a host root. The spores will germinate on agar surfaces in Petri dishes. In the absence of a host root, however, the germ tubes become inactive and their cytoplasm retracts into the spore, until another attempt is made to find a suitable host root.

But once the fungal hyphae sense the presence of a potential host, they branch and proliferate profusely; this probably increases the chance of contact with the root and subsequently infecting it. The branching phenomenon was first described more than 30 years ago5, and since then it has always seemed likely that the agent involved was a ‘branching factor’ released by roots. Indeed, root exudates have been shown to contain such a factor6. But its chemical structure has escaped identification, mostly because the concentration in root exudates is very low and the factor seemed to be rather unstable.

That changes with the report of Akiyama et al.1, which describes the chemical structure of branching factor and constitutes a leap forward in our understanding of plant–fungal symbioses. Akiyama and colleagues came up with a procedure for concentrating the factor, in which water containing root exudates of a plant called Lotus japonicus was pumped through a cartridge containing active charcoal. Branching factor was enriched on the charcoal surface, and ultimately enough of it could be isolated to allow determination of its chemical structure. Another essential element in the authors' success was the establishment of a sensitive bioassay that allowed the activity of branching factor towards fungi to be monitored during the purification procedure. The molecules identified are members of a group of sesquiterpene lactones, known as the strigolactones, that are active towards fungi at very low concentrations. Their relative instability seems to be a virtue, because a stable molecule would accumulate in the soil and so not convey positional information.

In another symbiotic relationship — that between legume roots and nitrogen-fixing, root-nodule bacteria — chemical communication also occurs before infection. But it seems that fungal and bacterial root symbionts detect different plant signalling molecules. Plant flavonoids activate the bacterial production of symbiotic signalling molecules known as nodulation factors7, but strigolactones are structurally and biosynthetically unrelated to flavonoids.

Arbuscular mycorrhizal fungi are exceptionally promiscuous with respect to the host plants they colonize; for example, my group has used a single strain of one species, Glomus intraradices, to successfully infect a wide range of plants, including liverworts, grasses (rice, wheat) and dicotyledonous plants such as tomato and pea. This is consistent with the broad distribution of strigolactones in the plant kingdom, and isolates from the two main groups of flowering plants, the dicotyledons and monocotyledons, do not significantly differ in their biological activity1.

A twist in the story comes from the close relatedness between branching factor and molecules detected by parasitic weeds8. For example, the pest Striga (witchweed) causes devastating losses in crops, and is a major obstacle to food production in Africa. The roots of Striga and related plants can attack the roots of their victims and rob them of water and nutrients9. It seems that these parasites rely on the very same molecules for seed germination that facilitate symbiosis of roots with arbuscular mycorrhizal fungi. Interestingly, roots increase strigolactone production when phosphate availability is low, possibly to attract the fungi, and these are the very conditions under which Striga infestations occur10.

To understand molecular communication between plant and fungus, one of the most pressing questions is now the chemical structure of the fungal signals that induce symbiosis-related gene expression in plant roots. Some observations suggest that the fungal signal is produced only by hyphae that have undergone the branching response — that is, after perception of branching factor11. Akiyama and colleagues' results open up the possibility of inducing fungal responses in the absence of a host root, which might also help in identifying the fungal signal.

Despite the global importance of arbuscular mycorrhizal fungi and their potential in agriculture, research into their molecular genetics is still in its infancy. What deters geneticists who work with model systems, such as yeast, is that these fungi contain several nuclei within a single spore and the nuclei are probably genetically diverse12. Together with a complete absence of sex in all arbuscular mycorrhizal fungi that have been investigated, this hinders classical genetic approaches. A programme to sequence the entire 15-megabase genome of G. intraradices is under way — I hope the results will prompt more researchers to tackle the molecular biology of these fungi that are so essential to plant life on our planet.


  1. 1

    Akiyama, K., Matsuzaki, K. -i. & Hayashi, H. Nature 435, 824–827 (2005).

  2. 2

    Remy, W., Taylor, T. N., Hass, H. & Kerp, H. Proc. Natl Acad. Sci. USA 91, 11841–11843 (1994).

  3. 3

    Redecker, D., Kodner, R. & Graham, L. E. Science 289, 1920–1921 (2000).

  4. 4

    Brundrett, M. New Phytol. 154, 275–304 (2002).

  5. 5

    Mosse, B. Annu. Rev. Phytopathol. 11, 171–196 (1973).

  6. 6

    Buee, M. et al. Mol. Plant Microbe Interact. 13, 693–698 (2000).

  7. 7

    Lerouge, P. et al. Nature 344, 781–784 (1990).

  8. 8

    Cook, C. E. et al. Science 154, 1189–1190 (1966).

  9. 9

    Yoder, J. I. Curr. Opin. Plant Biol. 4, 359–365 (2001).

  10. 10

    Yoneyama, K., Takeuchi, Y. & Yokota, T. Physiol. Planta 112, 25–30 (2001).

  11. 11

    Kosuta, S. et al. Plant Physiol. 131, 952–962 (2003).

  12. 12

    Hijri, M. & Sanders, I. R. Nature 433, 160–163 (2005).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Parniske, M. Cue for the branching connection. Nature 435, 750–751 (2005) doi:10.1038/435750a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.