The fungal grid

Abstract

The observation that soil‐dwelling fungi seem to exchange information via electrical impulses has raised new interest about their interactions with plants and their ecological significance.

Fungi play a crucial role in the soil as degraders of organic material, as a carbon sink and a supplier of nutrients for many other species. Their ecological importance is reflected in their size: soil‐dwelling fungi form mycorrhizal networks in the ground that can span kilometres and interconnect with plant root systems and bacteria to create a diverse and enormous meta‐organism.

Despite their important role, soil fungi have remained largely hidden from human attention short of scientific inquiry. Recently, however, they entered the spotlight over their role in underground communication. The suggestion that fungi could have a neuron‐like method of transmitting and processing information has prompted not only coverage in the mainstream press but also heated debate among scientists on the extent of communication among fungi and possible interchanges with plant root systems. This heightened interest spills over into conservation biology and ecology given soil fungi's crucial role in forest ecosystems and carbon storage.

There is universal agreement over the importance of mycorrhizal networks both for fungi themselves and plants. Fungi generate hyphae, which are long, filamentous tendrils for expansion and a permanent underground reservoir to grow the sexual organs aboveground, some of which are highly appreciated for their taste. Fungi also transmit electrical signals to each other through these hyphae networks, and this has long been suspected to play an important role in determining spatial aspects of growth depending on the availability of nutrients (Harold et al, ¹⁹⁸⁵). A current question is how sophisticated this communication is and whether it can be considered an elementary form of language.

The suggestion that fungi could have a neuron‐like method of transmitting and processing information has prompted […] heated debate among scientists on the extent of communication among fungi and possible interchanges with plant root systems.

Some fungal species, notably the mycorrhizae, also evolved a symbiotic relationship with vascular plants, executed through specialized hyphae called arbuscules. These are found in the roots or phylum of vascular plants, providing the roots with nutrients and water from the soil in return for more complex carbon‐based compounds. Current questions now concern whether plants also exploit the fungal networks for communication among themselves, such as warning other species about the arrival of insects or pathogens, or even conducting chemical warfare to deter competitors.

Generation of electrical signals

On the first area of underground communication among fungi themselves, the scientific debate and the media interest was sparked by a paper arguing that the exchanges of information among fungi transmitted across their hyphae networks via electrical spikes amounted to a sophisticated language (Adamatzky, 2022). The study analysed species‐specific patterns of oscillations in extracellular electrical potential measured by differential electrodes inserted into a substrate colonized by the mycelium, or directly into sporocarps, the fruiting bodies above ground. The author analysed the spiking characteristics of four species to assess the variations in the patterns, in ghost fungi (Omphalotus nidiformis), Enoki fungi (Flammulina velutipes), split gill fungi (Schizophyllum commune) and caterpillar fungi (Cordyceps militaris). Their spike durations varied from one to 21 h and amplitudes in potential from 0.03 to 2.1 mV, reflecting differing dynamics and scales.

…the scientific debate and the media interest was sparked by a paper arguing that the exchanges of information among fungi transmitted across their hyphae networks via electrical spikes amounted to a sophisticated language.

While spikes of electrical potential are key attributes of neurons, many other organisms without nervous systems also produce them, including plants and slime moulds. Yet, according to the paper's author, Andrew Adamatzky, Director of the Unconventional Computing Laboratory at the University of the West of England, Bristol, there is a possibility that mycelium networks transform information through these spikes in a manner homologous to neurons.

Adamatzky suggested that these electrical action‐potential‐like spikes are by‐products of intercellular calcium waves that propagate within individual cells to coordinate multicellular responses. “The calcium waves are used by fungi firstly to keep their integrity, that is exchange of presence information between distant parts of the mycelium, analogous to wolves howling just to tell each other about their presence,” Adamatzky commented. “Secondly the waves ‘report’ to other parts of mycelium about discovered sources of attractants and repellents.” He also conceded a third option, namely that the waves were not saying anything at all. “Propagating mycelium tips are electrically charged and, therefore, when the charged tips pass a pair of differential electrodes a spike in the potential difference is recorded,” Adamatzky added. The implication is that the waves could be by‐products of that electrical activity, comparable to the way evolutionary spandrels are phenotypic traits that have no adaptive value themselves but ride on another trait that was positively selected for.

Yet, Adamatzky argues that the electrical signals might be associated with growth or spore production in response to environmental variables, or damage to a fungal cluster. The signals would propagate from the damaged body to intact ones on the network, to accelerate either growth or maturation of spores. “The sporulation could be induced by partial desiccation, and a number of fruit bodies could be larger in proximity of injury. This observation is in line with the finding that damaged mycelium responds with branching, and is similar to the sprouting response of a slime mould Physarum polycephalum to a dissection of its protoplasmic tubes,” Adamatzky explained.

A fungal “neural network”?

The findings that these electrical spikes occur in various patterns in response to destructive stimuli and seem to involve a complex exchange of information come after almost 50 years of painstaking investigation of the electrical activity of fungi. Spikes resembling neuronal action potentials were first reported as early as 1976 by making intracellular recordings from mycelium of the red bread mould Neurospora crassa (Slayman et al, ¹⁹⁷⁶). A key milestone followed a decade later in 1986, when McGillivray & Gow (1986) found that the direction of hyphal extension, as well as the frequency of branching and germination, could be affected by electric fields. Another decade later came observations of what looked like spontaneous action potential in the hypha of the oyster mushroom Pleurotus ostreatus and the honey mushroom Armillaria bulbosa, from intracellular recording using a reference electrode in an agar substrate (Olsson & Hansson, 1995).

It was then in 2018 that Adamatzky laid the foundation for the latest analyses with the finding that oyster fungi exhibit two types of spiking activity when they generate impulses of electrical activity, similar to neuronal action potentials (Adamatzky, 2018). In his 2022 paper, he argues that the trains of spikes that are transmitted in a mycelium network could be broken down in the same way as human languages, firstly as characters encoded in the signals, then size of the character lexicon, grammar, syntax and spelling. He claims that the distributions of lengths of spike trains, measured across a number of spikes, are similar to that of word lengths in human languages. “We found that size of fungal lexicon can be up to 50 words, although the core lexicon of most frequently used words does not exceed 15–20 words,” Adamatzky wrote.

These latest deductions have prompted some excitable and predictable responses in the public media about fungi being able to talk, just as happened after some comparable discoveries in plants. But Adamatzky's own view is much more nuanced. He argues that the spike patterns equate to a vocabulary, but one that is limited and more analogous to the communication of some domestic animals. “With regards to forming complex sentences, I think that is unlikely,” he commented. “Take for example, the language of domestic cats and dogs. Surely, owners understand vocal and body ‘words’. And the number of such ‘words’ is very limited, for example for a dog ‘want a poo/pee’, ‘hungry’, ‘love you’, ‘want to go for a walk etc.’ They do not form long sentences. Even people, unless it is their job to write, communicate with very short sentences and limited vocabularies of about 500 words.”

Tapping into the network

More controversial at a scientific level is the claim that plants can also tap into the electrical mycorrhizal networks to communicate with each other over wider distances—which has inspired the term “Wood Wide Web” (WWW). Suggestions include the idea that plants can communicate the same sort of information through mycorrhizal networks that they transmit above ground through volatile organic chemicals (VOCs). It is clearly established that plants use a variety of VOCs, including phenolics and tannins, to coordinate their chemical defences and regulatory networks (Mithöfer & Boland, 2016). Methyl jasmonate emitted by sagebrush (Artemisia tridentata) was the first VOC shown to trigger a response to nearby herbivores even when the plant had no direct contact with the predator yet. Moreover, VOCs can also be detected by neighbouring plants and other parts of the emitting plant to initiate a response to attack. The question then is whether trees can also convey such information below ground through their root systems and mycorrhizal networks. Among proponents of this theory is Suzanne Simard from the University of British Columbia, who is now pursuing these ideas in the Mother Tree Project (https://mothertreeproject.org/), but she did not respond to requests for comment.

More controversial at a scientific level is the claim that plants can also tap into the electrical mycorrhizal networks to communicate with each other over wider distances…

Among the critics is Dan Bebber, Co‐Director of Global Engagement for Biosciences at the University of Exeter, who referred to a recent paper showing that there is no evidence for this idea, only a positive citation bias (Karst et al, ²⁰²³). This paper refers to common mycorrhizal networks (CMNs) that connect the roots of multiple plants of the same or different species below ground and expresses concern that bias towards their positive effects in the scientific literature has stoked the popular idea of trees engaging in sophisticated interactions over the plant/fungal networks. The authors wrote that “The claim that mature trees preferentially send resources and defence signals to offspring through CMNs has no peer‐reviewed, published evidence.” Bebber conceded that the possibility of plants communicating through mycorrhizal networks could not be ruled out but clear evidence is lacking so far.

Indeed, if these networks have existed for a long time, it would be surprising if plants had not evolved some ways to exploit them, argues Frantisek Baluska at the Institute of Cellular and Molecular Biology at the University of Bonn in Germany. “This controversy is due to the lack of relevant studies on this important topic,” he said. “Moreover, it is almost impossible to perform relevant studies in the soil using the intact root‐fungal networks.” Baluska commented that most studies so far have focussed mainly on transport of nutrients and water/solutes. “However, a few new reports indicate that fungal hyphae are supporting action potentials between different plants,” he added. “This mycelium‐mediated electrophysiological plant‐plant communication supports the Wood Wide Web concept, based on the integrated root‐fungal networks, very strongly.”

Ecological significance of forest fungi

What is beyond doubt is that mycorrhizal networks are of huge ecological significance. Plants, especially in forests, are heavily dependent on them, such that disruptions to one impact the other with knock‐on effects across the whole ecosystem. Ecological concerns and interest in the role of mycorrhizal networks in mitigating climate change helped spawn the Society for Protection of Underground Networks (SPUN), launched in November 2021, that involves researchers from the Netherlands, Canada, USA, France, Germany and the UK.

SPUN is collaborating with the GlobalFungi database and the Crowther Lab at ETH Zurich to build and train machine‐learning models for identifying, measuring and eventually predicting mycorrhizal richness across global terrestrial ecosystems. “Our latest spatial models reveal mycorrhizal biodiversity hotspots in a number of unexpected locations, including tropical conifer forests in Northern Mexico and ancient grasslands in Asia and South America,” said Toby Kiers, University Research Chair at Vrije Universiteit Amsterdam and Director of SPUN. “We are finalizing our spatial analysis to identify the top 10 mycorrhizal hotspots and assess their existing threats and protective status. As part of this work, we estimate that under‐sampled ecoregions cumulatively cover over 50% of Earth's terrestrial area, meaning that we are largely missing mycorrhizal information from more than half the world's land area.”

This has led to expeditions to Ecuador, Lesotho, Patagonia in Chile, Palmyra Atoll in the Pacific and the Alps and Apennines mountains in Italy, where SPUN researchers have worked with local collaborators to assemble a global dataset. “We are strategically adding new samples to this dataset to fill gaps from under‐explored regions, which is important for creating an accurate spatial model of mycorrhizal diversity, even when extrapolating to locations with fewer data points,” said Kiers. This will result in global maps useful for all types of managed ecosystems, including forestry and agriculture, she added.

Even if plants did not communicate directly over these networks, they are the major contributor of carbon to it, hence the great interest from a climate change perspective.

This work is spanning multiple scales from laboratory to ecosystem and embraces the flow of chemicals within them. “In the lab, we are studying the structure and flows inside arbuscular mycorrhizal networks at a micron scale, tracking nutrients and carbon to understand how fungi move resources under different climate perturbations,” Kiers explained. “To better leverage fungi in agriculture, forestry and climate scenarios, we need to understand the growth and trade strategies of the fungi themselves.”

Even if plants did not communicate directly over these networks, they are the major contributor of carbon to it, hence the great interest from a climate change perspective. “Mycorrhizal fungi have been shaped by natural selection for hundreds of millions of years. They have evolved strategies that result in the movement of massive amounts of nutrients and carbon,” Kiers said. “We know that plants that feed carbon to underground networks store an estimated eight times more carbon compared to ecosystems with non‐mycorrhizal vegetation. By studying the fungal strategies themselves, we can understand what the fungi do with the carbon once it is below ground.”

Indicators of forest health

While not a direct objective, the activities around SPUN could help to elucidate the extent to which plants do exchange information with and over mycorrhizal networks, or just integrate with it for the exchange of nutrients and carbon. There are though other possible societal benefits of work on underground mycelial networks, relating to the findings on electrical activity. As Adamatzky noted, fungi have been shown to be hypersensitive detectors for many environmental cues. “Fungi sense light, chemicals, gases, gravity and electric fields,” he explained. “Fungi show a pronounced response to changes in a substrate pH, demonstrate mechanosensing, they sense toxic metals, CO2, direction of fluid flow. Fungi exhibit thigmotactic and thigmomorphogenetic responses. Fungi are also capable of sensing chemical cues, especially stress hormones, from other species, thus they might be used as reporters of the health and well‐being of other inhabitants of the forest.”

Fungi are also capable of sensing chemical cues, especially stress hormones, from other species, thus they might be used as reporters of the health and well‐being of other inhabitants of the forest.

Even if the grand idea of Wood Wide Web with a sophisticated communication network involving plants and fungi remains elusive, further research and related projects such as SPUN can yield more knowledge about the enormous and diverse ecosystems below ground and the relationship between mycorrhizal networks and plants. This knowledge will be valuable for understanding how climate change affects forests worldwide and how we could use and strengthen their carbon storage capabilities to mitigate the effects of increasing CO₂ levels in the atmosphere (Fig 1).

Figure 1. Agaricus fungus.

Visible interface of a Wood Wide Web? Fruit body of the fungus Agaricus sp. in an Atlantic forest in Northeastern Bahia, Brazil. © Alex Popovkin, Bahia, Brazil/Wikimedia.

EMBO reports (2023) 24: e57255