On resistive spiking of fungi
OOn resistive spiking of fungi
Andrew Adamatzky a , Alessandro Chiolerio b,a , Georgios Sirakoulis c,a a Unconventional Computing Laboratory, UWE, Bristol, UK b Center for Sustainable Future Technologies, Istituto Italiano di Tecnologia, Torino, Italy c Department of Electrical and Computer Engineering Democritus University of Thrace, Xanthi, Greece
Abstract
We study long-term electrical resistance dynamics in mycelium and fruit bodies of oyster fungi
P.ostreatus . A nearly homogeneous sheet of mycelium on the surface of a growth substrate exhibitstrains of resistance spikes. The average width of spikes is c. 23 min and the average amplitude isc. 1 kOhm. The distance between neighbouring spikes in a train of spikes is c. 30 min. Typicallythere are 4-6 spikes in a train of spikes. Two types of resistance spikes trains are found in fruit bodies:low frequency and high amplitude (28 min spike width, 1.6 kΩ amplitude, 57 min distance betweenspikes) and high frequency and low amplitude (10 min width, 0.6 kΩ amplitude, 44 min distancebetween spikes). The findings could be applied in monitoring of physiological states of fungi andfuture development of living electronic devices and sensors.
Keywords: fungi, biomaterials, neuromorphic materials, soft robotics
1. Introduction
Electrical resistance of living substrates is used to identify their morphological and physiologicalstate [1, 2, 3, 4, 5]. Examples include determination of states of organs [6], detection of decaying woodin living trees [7, 8], estimation of roots vigour [9], study of freeze-thaw injuries of plants [10], as wellas classification of breast tissue [11]. The aim of this paper is two-fold.First aim is to study the dynamics of the fungal resistance during long-term (up to two daysof intermittent measurements). Whilst resistive properties of plants and mammals tissue have beenstudied extensively, results on electrical resistance of fungi are absent. This gap should be properly filledbecause the fungi is the largest, widely distributed and the oldest group of living organisms [12]. Fungi“possess almost all the senses used by humans” [13]: they can sense light, chemicals, gases, gravityand electric fields. Fungi show a pronounced response to changes in a substrate pH [14], demonstratemechanosensing [15] and sensing of toxic metals [16], CO [17], and chemical cues, especially stresshormones, from other species [18]. Thus mycelium networks can be used as large-scale distributedsensors. To prototype fungal sensing networks we should know their electrical features and resistanceis definitely one of these characteristics.Second aim is to assess whether fungi can be employed as electronic oscillators. The applicationdomain of the fungal electronic oscillators could be the field of unconventional computing [19], especiallyin the framework of organic electronics, living sensor and living computing wetware. Feasibility studieswith plants [20, 21], slime mould [22, 23, 24, 25] and fungi [26, 27] have shown that it is possible todevelop electrical analog computing circuits either based or with these living creatures. Biologicalmolecules such as suine microtubules have been shown to enable very fast oscillations, in the tents ofMHz range [28]. However, to have a full functional analog computer, we probably need an oscillator.As Horowitz and Hill reported — “A device without an oscillator either doesn’t do anything or expectsto be driven by something else (which probably contains an oscillator)” [29]. Thus, we envisage thatthe resistive spiking can be utilised to produce fungal electronic oscillators. Preprint submitted to Elsevier September 2, 2020 a r X i v : . [ c s . ET ] S e p a) (b) (c) (d)Figure 1: Experimental setup. (a) Electrodes are inserted in a hemp substrate nearly fully colonised by P. ostreatus .(b) Electrodes are inserted in stalk of a fruit body of
P. ostreatus . (c) Scheme of resistance measurement. (d) Schemeof measuring fungal electrical potential under DC applied.
2. Methods
Oyster fungi
P. ostreatus have been cultivated on hemp substrate in plastic containers in darknessand at ambient temperature 20-23 o C. We used the substrate after it was nearly fully colonised bymycelium, which was indicated by an almost everywhere white colour and white film of nearly homo-geneous mycelium, sometimes called ‘skin’ on surface of the substrate that was formed. The electricalresistance of the skin was measured as follows. We used iridium-coated stainless steel sub-dermal nee-dle electrodes (Spes Medica S.r.l., Italy), with twisted cables. The pairs of electrodes were inserted infungal skin, while the distance between electrodes was kept 1 cm (Fig. 1(a)). Twelve trials of measure-ments were undertaken with fungal skin. In six trials, we also undertook recordings of the fruit body’sresistance, where electrodes were inserted in stalks of the bodies (Fig. 1(b)). The resistance was mea-sured (Fig. 1(c)) and logged using Fluke 8846A precision multimeter, the test current being 1 ± µ A, once per 10 sec, 5 · samples per trial. When characterising trains of spikes, we measured spikeaverage width w , average amplitude a and average distance between spikes d . To check if there arepotential oscillations of voltage applied to the fungi, we applied direct current voltage with GW InstekGPS-1850D laboratory DC power supply and measured voltage with Fluke 8846A (Fig. 1(d)).
3. Results
The resistance of fungal mycelium exhibits very slow, 1 . · –3 · sec, disordered changes of theresistance values with trains of spikes, of increased resistance, emerging. An example of the long-termrecording is shown in Fig. 2(a) and a train of spikes in Fig. 2(b). In over 16 trials we inferred thefollowing parameters of spikes: w = 1380 sec (median 1190 sec, σ = 77), a = 1 , σ = 674), d = 1830 sec (median 175 sec, σ = 87). Spike width versus amplitude distribution is shownin Fig. 2(c).Two trials of the resistance recording from substrate, colonised by fungi, have shown outstandingphenomena (although these have not been explicitly included in the above analysis).More specifically, in one trial (not included in the analyses above) we observed high frequency( w = 86 sec, median 86 sec, σ = 7 and d = 283 sec, median 250 sec, σ = 166) and high amplitude( a = 11 , σ = 3 , · from the start of the ascentto the end of the descent). On tops of these variations there were trains of 5-7 spikes. An example isshown in Fig. 2(e). Average widths w of these spike is 1 ,
094 sec ( σ = 475), a = 158Ω ( σ = 49) and d = 1 ,
135 sec ( σ = 442).In fruit bodies, we typically recorded two types of spikes: low frequency ( w = 1 ,
690 sec, σ = 32 and d = 3 ,
450 sec, σ = 161) and high amplitude ( a = 1 , σ = 116) and high frequency ( w = 580 sec,2 e s i s t a n ce , O h m × × × Time, ×
10 sec0 1000 2000 3000 4000 5000 6000 7000 × × × × × (a) R e s i s t a n ce , O h m × × × × Time, ×
10 sec400 500 600 700 800 900 1000 1100 (b) A m p lit ud e , O h m -1 sec0 50 100 150 200 250 300 350 400 (c) R e s i s t a n ce , O h m × × × × × Time, ×
10 sec8400 8450 8500 8550 8600 8650 8700 (d) R e s i s t a n ce , O h m × × × × × × Time, ×
10 sec5500 6000 6500 7000 7500 8000 (e) R e s i s t a n ce , O h m × × × × × Time, ×
10 sec2800 3000 3200 3400 3600 3800 (f) * * * s *** P o t e n ti a l , V ×
10 sec2000 2500 3000 3500 4000 (g)Figure 2: (a) Slow variations of resistance with trains of spikes, zoomed in the inserts, are usually observed in long termrecordings. (b) Example of a train of 5 spikes. (c) Spike width w versus amplitude a distribution. Line is a linear fit a = 6 w + 195, R = 0 .
49. (d) Examples of high amplitude and high frequency spikes. (e) Example of a spike train ontop of a very slow variation of resistance. (f) Example of resistance recorded on fruit bodies. (g) Oscillation of electricalpotential under 10 V DC applied, where spikes analysed are marked by ‘*’. = 16 and d = 2 ,
630 sec, σ = 188) and low amplitude ( a = 611 Ohm, σ = 266). Figure 2(f) shows atypical train of four high frequency spikes followed by a train of low frequency spikes.To assess feasibility of the living fungal oscillator, we conducted a series of scoping experimentsby applying direct voltage to the fungal substrate and measuring output voltage. An example of theelectrical potential of a substrate colonised by fungi under 10 V applied is shown in Fig. 2(g). Voltagespikes are clearly observed. Spikes with amplitude above 1 mV, marked by ‘*’, except the spike markedby ‘s’ have been analysed. We can see two trains of three spikes each. Average width of the spikes is1,050 sec ( σ = 9 .
2, median 1,090 sec), average amplitude 2.5 mV ( σ = 0 .
68, median 2.2 mV), whileaverage distance between spikes is 2,318 sec ( σ = 25 .
6, median 2370 sec).
4. Discussion
We demonstrated that oyster fungi
P. ostreatus undergo oscillations of resistance with trains ofresistive spiking emerging. Spikes amplitude vary from 1 k Ω to 1 . k Ω and width of spikes from 23 min upto 28 min. A distance between spikes in a train varies from 30 min to nearly 60 min. The oscillations ofresistance have so low frequency that could not be explained using conventional electronics framework(e.g. charging of a mycelium during probing) and resistance sampling was with very low frequency (onceper 10 sec). Thus the only feasible explanation, we see is the translocation of water and metabolitestaking place in the mycelium. This translocation is periodic, and more likely guided by calcium waves.Increase in a liquid in the mycelium loci leads to reduced resistance. When the translocated mass ofmetabolites leaves the area, the resistance increases. Rates of the translocation, measured by injectingfluoresecent dye in hyphae, reported in [30] are 2–6 cm/hour for small specimen and 9–15 cm/hourfor large specimen. The distance between electrodes in our experiments was 1 cm, thus the aboverate can be translated to the following width of resistive spikes — 10–30 min and 4–7 min. Thefirst estimates matches in scale with resistive spikes measured in our experiments. The widths ofresistive spikes are twice the widths of electrical potential spikes observed by us previously in fruitbodies of
P. ostreatus [31]. All the above indicate that the resistive spiking observed is not an artefactbut manifestation of physiological processes in fungal mycelium and fruit bodies. Therefore one of theapplication domains of the proposed methodological setup and delivered results could be in monitoringphysiological states of fungi: the physiological states might reflect states of ecosystems inhabited byfungi [32]. In experiments with fungal oscillator we have found that at some stages the fungal skinexhibits oscillations of the electrical potential. A width of a voltage spike is c. 18 min, which is slightlyless than an average width of resistive spikes, and an average amplitude c. 2.5 mV (at 10 V DC voltageapplied). The amplitude is not as high as would be expected from our previous experience with slimemould electronic oscillators [33]. This may be due to the fact that in the experiments with slimemould, the electrodes were connected by a single protoplasmic tube, so its resistance was crucial, whilein the fungal skin the current can also propagate along remnants of the wet hemp substrate. A verylow frequency of fungal electronic oscillators does not preclude us from considering inclusion of theoscillators in fully living or hybrid analog circuits embedded into fungal architectures [34] and futurespecialised circuits and processors made from living and functionalised with nanoparticles fungi.
Acknowledgement
This project has received funding from the European Union’s Horizon 2020 research and innovationprogramme FET OPEN “Challenging current thinking” under grant agreement No 858132.