A commercial arbuscular mycorrhizal inoculum increases root colonization across wheat cultivars but does not increase assimilation of mycorrhiza‐acquired nutrients

Societal Impact Statement Production and heavy application of chemical‐based fertilizers to maintain crop yields is unsustainable due to pollution from run‐off, high CO2 emissions, and diminishing yield returns. Access to fertilizers will be limited in the future due to rising energy costs and dwindling rock phosphate resources. A growing number of companies produce and sell arbuscular mycorrhizal fungal (AMF) inoculants, intended to help reduce fertilizer usage by facilitating crop nutrient uptake through arbuscular mycorrhizas. However, their success has been variable. Here, we present information about the efficacy of a commercially available AMF inoculant in increasing AMF root colonization and fungal contribution to plant nutrient uptake, which are critical considerations within the growing AMF inoculant industry. Summary Arable agriculture needs sustainable solutions to reduce reliance on large inputs of nutrient fertilizers while continuing to improve crop yields. By harnessing arbuscular mycorrhizal symbiosis, there is potential to improve crop nutrient assimilation and growth without additional inputs, although the efficacy of commercially available mycorrhizal inocula in agricultural systems remains controversial. Using stable and radioisotope tracing, carbon‐for‐nutrient exchange between arbuscular mycorrhizal fungi and three modern cultivars of wheat was quantified in a non‐sterile, agricultural soil, with or without the addition of a commercial mycorrhizal inoculant. While there was no effect of inoculum addition on above‐ground plant biomass, there was increased root colonization by arbuscular mycorrhizal fungi and changes in community structure. Inoculation increased phosphorus uptake across all wheat cultivars by up to 30%, although this increase was not directly attributable to mycorrhizal fungi. Carbon‐for‐nutrient exchange between symbionts varied substantially between the wheat cultivars. Plant tissue phosphorus increased in inoculated plants potentially because of changes induced by inoculation in microbial community composition and/or nutrient cycling within the rhizosphere. Our data contribute to the growing consensus that mycorrhizal inoculants could play a role in sustainable food production systems of the future.

The following Supporting Information is available for this article:  Methods S1 This file provides specific details on methods referred to in the main text, including biological materials and growth conditions, equations for calculation of phosphorus and carbon exchange between symbionts. Additional methods for measuring extraradical hyphal lengths, PCR conditions, restriction enzyme digests and Genemapper analysis.  P-orthophosphate which only the intact fungal hyphae could access in the static core treatment. In the rotated control treatment, the labelled core was rotated to sever hyphal connections and control for isotope diffusion outside of the core. 14 CO2 was released into sealed chamber to measure the fixation of C by the wheat plants and transfer to fungi within the static cores. The third core with an airtight seal (orange) was used to track plant-fixed carbon passed to the fungal partner through fungal respiration of 14 C.

Biological material and growth conditions
Wheat seedling were sterilised with chlorine gas, germinated on damp filter paper at 20°C for six days then transplanted into pots containing agricultural soil and sand in a 1:1 mix. Agricultural soil was collected from Leeds University Farm (Spen Common Lane, Tadcaster, North Yorkshire, LS24 9NU, England), sand was added to reduce compaction to aid water drainage. The soil, a slightly alkaline sandy clay loam with a pH of 7.5, was air-dried and passed through a 2 mm sieve.
Analysis of the soil characteristics showed soil organic C content represented ~2% of soil dry weight and soil solute concentrations of PO4, NO3, and NH4 were 0.08 mg L -1 , 6 mg L -1 , and 0.04 mg L -1 respectively (Holden et al., 2019). At the time of planting, the wheat seedlings were inoculated with a commercially available inoculum (PlantWorks, Kent, UK), which contained a mixture of infective AMF propagules including colonised root fragments, hyphae and spores (100 per gram) and an inert carrying substrate without fertiliser additions (1:1 pumice and zeolite). The recommended application rate for the inoculum is ~ 6 g per 1 L pot.
At the time of planting, mesh windowed cores were inserted into the pots perforated capillary tubes were fixed inside the cores using a waterproof silicon sealant (Aqua Mate, Ever Build, Dublin, Ireland) applied to the end of the tube before it was placed onto the mesh at the bottom of the core. The cores were filled with the agricultural soil/sand growing substrate contained in the rest of the pot. Greenhouse growing conditions were supplemented with LED lighting and electronic blinds to create a 16-hour photoperiod, light intensity: 350 mol.m -2 s -1 , average temperature: 23C, the plants were watered every three days with each being given 30 ml of 40% nitrate type Long Ashton solution weekly (Hewitt, 1966).

Quantifying 33 P-and 15 N-for-C exchange between wheat and fungi.
Below-ground respiration was sampled every 2 hrs and above-ground gas throughout the labelling period by injecting gas samples into gas-evacuated scintillation vials containing 10 ml of the C trapping chemical CarbonTrap (Meridan Biotechnologies) and mixed with 10 ml scintillation chemical CarbonCount (Meridan Biotechnologies). Sample activity was monitored by scintillation counting (Packard Tri-carb 3100TR, Isotech, Chesterfield, UK). At the end of the 16h photoperiod, 4 ml 2M KOH was injected into vials within the sealed systems to trap remaining 14 CO2 gas.

Mycorrhizal-acquired 33 P
To quantify the amount of 33 P transferred from fungus to plant, samples (30-50 mg plant or 40-100 mg soil) were digested in 1 ml concentrated sulphuric acid at 365°C for 15 minutes. Upon cooling, 100 µl hydrogen peroxide (Acros Organics, Geel, Belgium) was added to the samples and reheated to 365°C for 1 minute and repeated until the solution cleared. The cleared sample was diluted up to 10 ml with distilled water. 33 P activity was quantified via Packard Tri-carb 3100TR (Isotech, Chesterfield, UK) using 2 ml of digest solution with 10 ml Emulsify-safe (Perkin-Elmer). 33 P transfer between symbionts was corrected for radioactive decay and measured using  Cameron et al. 2007).

Transfer of carbon from plant to fungus
Between 10 and 40 mg of plant tissue or soil was placed in a Combusto-cone (Perkin Elmer) and the CO2 released through oxidation was trapped in 10 ml CarbonTrap (Meridan Biotechnologies) and mixed with 10 ml CarbonCount (Meridan Biotechnologies). Radioactivity present in the sample measured using the following equations.  (Cameron et al., 2006)).  Cameron et al. 2008)).

Hyphal extractions and measuring extraradical hyphal lengths
Hyphal lengths were measured by a grid-line-intersect method over 50 fields of view (Tennant, 1975). A 1-3 g sample of soil was dispersed in 1000 mL of water on a stirring plate and hyphae were extracted from the spinning liquid via a syringe. This liquid was then dispensed into the Millipore filtration apparatus attached to a vacuum pump. The fluid was drawn through a filter paper disc which caught the hyphae, which was subsequently dyed with a trypan blue stain.

PCR and T-RFLP
The PCR was carried out in a 20 μl reaction using 2 μl of DNA template, 10 μl of Qiagen Mastermix (Qiagen, Hilden, Germany) and 0.5 μM of each primer, made up to the final volume with PCR water. Thermal cycling consisted of an initial DNA denaturation step of 3 min at 94°C followed by 35 cycles each of 30 s at 94°C, 40 s at 59°C and 60 s at 72°C with a final extension step of 10 min at 72°C, on a 96-well thermal cycler. All PCR plates included a negative control to ensure that no DNA contamination was present. Gel electrophoresis was used to verify the success of PCR amplification. 1 μl of loading buffer (Bioline, London, UK) and 4 μl of PCR product was run on a 1.5% agarose gel with 0.001 % (v/v) SYBR® Safe DNA stain (Invitrogen, Carlsbad, USA).
PCR products were run along-side a 100 bp ladder (Bioline, London, UK).
For the HpyCHIV/MboII/Sau96I digest, the optimal reaction set-up contained 1 unit of each enzyme, 1 μl of CutSmart buffer (All enzymes had 100% activity in CutSmart buffer) and 3 μl of PCR product, made up to final reaction volume of 10 μl with water. Digests were incubated for 60 minutes at 37°C then denatured at 65°C for 20 minutes. All digests were run alongside uncut samples and negative controls. Digests were diluted 1:10 with water to prepare the samples for capillary DNA analysis. 1 μl of the diluted digest was added to 9 μl of formamide containing 1% GeneScanTM LIZ 1200 size standard (Applied Biosystems, UK) and heated at 94°C for 3 minutes before immediate cooling on ice. Genotyping was carried out on an ABI 3730 PRISM® capillary DNA analyser (Applied Biosystems, UK).
T-RFLP data was analysed using Genemapper software v. 5 (Applied Biosystems, UK) with a background threshold of 50 fluorescent units, and a bin width of 5 bp. Peaks were analysed in the range of 50-850 bp. The relative abundance of each peak was calculated to the percentage of total sample fluorescence. Peaks containing <1% of the total sample fluorescence were discarded, and artefacts were detected by identifying peaks which frequently occurred in both the cut and uncut samples, after their removal the proportion of total sample fluorescence accounted for by the peaks was recalculated.