Experimental Design
To examine the influence of AMF species and microbial inoculation with a soil suspension on the bacterial community structure associated with their mycelia, we conducted a greenhouse experiment. We employed a microcosm setup consisting of two pots (6.5 × 6.5 × 8.5 cm each), each featuring a cut side and a mesh nylon membrane with an opening size of 44 µm. These pots were then attached to each other and sealed with silicone adhesive to create a unit with two compartments, allowing connectivity through two nylon membranes (Fig. 1). One pot was designated for plant growth (root compartment, RC), while the other facilitated hyphal proliferation (root-free compartment, RFC) and allowed microbial inoculation with a soil suspension. Both pots were interconnected through two 44 µm membrane layers (SEFAR Incorporation, Buffalo, NY, USA), allowing only the passage of hyphae while excluding roots.
We utilized a randomized block design with six replicates, incorporating two factors: (1) three distinct AMF species (M1: Rhizophagus irregularis isolate DAOM 197198, M2: Rhizophagus clarus isolate DAOM 234179, and M3: Rhizophagus cerebriforme isolate DAOM 227022, all cultivated in vitro under identical conditions) provided by Premier Tech, Rivières du Loup, QC, Canada; and (2) Consequently, there were six treatment combinations: two levels of microbial inoculation (B0 and B1). M1B0, M1B1, M2B0, M2B1, M3B0, and M3B1. Additionally, two control groups without AMF spore inoculation (M0) and with autoclaved distilled water (B0) were included to evaluate plant growth and root colonization, resulting in two combinations (M0B0 and M0B1). The experiment was arranged on a mesh table to prevent cross-contamination via water.
The microbial suspension was made from soil collected from the rhizosphere of the old maple forest at the Gault Nature Reserve in Mont Saint-Hilaire, QC, Canada [32, 33]. This soil was selected because it is known for hosting a diverse AMF community, thereby increasing the likelihood of fostering highly interactive microbial communities with the AMF species utilized in this experiment. A composite sample was created by combining four soil samples obtained from the forest, which were then sieved at 2 mm to remove large particles. The soil was mixed with sterilized distilled water, agitated for 30 minutes to achieve suspension, and left at room temperature for at least 2 hours. The suspension was subsequently filtered through sterilized sieves with opening sizes of 53 and 20 µm and collected in a sterilized Erlenmeyer flask (Fig. 1). The filtrate was confirmed to be free from AMF spores through binocular observation on a Petri dish. The suspension was used to inoculate the pot in the root-free compartment as described below.
The growth medium used for the substrate was composed of a mixture of sandy loam soil, sand, and Turface in a proportion of 1/1/1 (v/v/v) to enhance water drainage and facilitate AMF mycelia recovery at the harvest of the experiment. Sandy loam soil was collected from the organic farm at the L’Institut de Recherche et de Développement en Agroenvironnement (IRDA) in St-Bruno, QC, Canada (45°32’59.6” N, 73°21’08.0” W). This soil was collected from the top 15 cm layer and had a pH value of 6.01, and its characteristics were described in Renaut et al. (2019). After air-drying and sieving at 2 mm, it was combined with sand from Bomix, St-Léonard, QC, Canada, and Turface from Hydro Dionne, Montreal, QC, Canada. Each compartment of the microcosm was filled with approximately 250 mL of substrate.
Leek (Allium porrum L., cultivar Monstrous Carentan), a highly mycotrophic host plant, was used in the experiment. The seeds were purchased from McKenzie Seeds (Brandon, MB, Canada). These seeds underwent surface disinfection at room temperature under laminar flow conditions as follows: immersion in 70% ethanol for 30 s, immersion in 15% bleach for 5 min, and thorough rinsing with sterile distilled water. Subsequently, the seeds were soaked in sterile distilled water for 2 hours. Pre-germination took place on 10% water agar plates, which were then incubated at 28°C for one week. Following this, two germinated seeds were meticulously placed into each root compartment pot.
The germinated seeds were inoculated with 300 µL of a spore suspension in sterile distilled water containing approximately 200 spores of each AMF species separately, as per the experimental design. One month later, one plant per microcosm was retained, and a second inoculation was directly applied to the culture substrates. To mitigate microalgae contamination, a 1-cm layer of autoclaved Turface was placed atop the microcosms and replaced weekly. The leek plants were cultivated under a 16/8-hour photoperiod and a temperature of 21/18°C (day/night). Additional microcosm units were utilized to monitor plant colonization and the presence of mycelia in RFCs.
Microcosms received weekly watering with half-strength modified Long Ashton nutrient solution with a phosphorus concentration of 22 ppm [34] in the plant compartments to maintain soil moisture at the holding capacity. Irrigation with water occurred two to three times weekly in both compartments (RC and RFC). Two months post-second AMF inoculation, RFC substrate enrichment with a natural soil microbiome was conducted by applying 50 mL of a microbial suspension, as per the experimental design, labeled as treatment B1. For non-inoculated treatments (B0), 50 mL of sterile distilled water was added. This 50 ml volume was determined based water-holding capacity which was measured as 70 mL. Subsequently, the irrigation regime was adjusted for the remainder of the experiment, with a half-strength nutrient solution added weekly to the RFC compartments. Thripex Neoseiulus cucumeris (Koppert, Leamington, ON, Canada), a biocontrol agent, was employed for thrip control, and no chemical pesticides were utilized in the experiment.
Parameters of plant growth, chlorophyll content, and mycorrhizal colonization
Before harvesting the plants, the photosynthetic capacity was determined by measuring the chlorophyll concentration in µg per cm2 of the third or fourth fully expanded leaf from the top using an atLEAF chlorophyll meter (FT Green LLC, Washington, USA). Additionally, the stem diameter of the plants was measured using an electronic caliper 5 mm above the sprouting point. The microcosm pots were then carefully separated, and the plants were removed from the substrate. The shoots were detached from the roots and placed in paper bags. The roots were rinsed with sterile distilled water and weighed. A portion of the fine roots was preserved in 8 mL of 50% ethanol for mycorrhizal colonization assessment. The bags containing shoots and the remaining roots were dried at 80°C for 72 hours in a hot-air oven. The remaining roots were weighed to determine the total root dry biomass.
Mycorrhizal colonization was assessed using the ink and vinegar method (Vierheilig et al., 1998). Briefly, the roots were washed with tap water and cut into 1-cm-long segments. These segments were cleared in 10% KOH for 3–4 min at 90°C, rinsed with distilled water, and then treated with 1% acetic acid for 4 minutes to remove KOH residues. The cleared roots were stained with a 5% ink–vinegar solution at 90°C for 3 minutes and rinsed with distilled water. They were then stored in lactoglycerol solution until microscopic examination using a Zeiss Axio Imager 2 (Zeiss, Montreal, QC, Canada). The stained root segments were placed on microscope slides using fine-point forceps, and the mycorrhizal colonization ratio was determined using the intersection method [34].
Soil sampling and AMF mycelia collection
To mitigate the influence of surface and root exudates on the AMF hyphae-associated microbial community, the top layer of the RFC substrate (comprising Turface and the upper 1 cm of substrate) was removed along with a 1 cm width from the membrane side. Approximately 10 mL of substrate was then gathered in a 15 mL Falcon tube. The remaining substrate was placed in a sterile 1 L glass jar, covered with cold sterile distilled water, vigorously agitated, and allowed to settle for approximately 15–20 minutes to aggregate AMF mycelia. Subsequently, the solution was filtered through a sterile sieve with a mesh size of 40 µm. The aggregated mycelia (Figure S2A) were delicately collected from the sieve using fine forceps and transferred into a sterile plate containing approximately 1 mL of cold sterile physiological water. This process was repeated four times to maximize the collection of mycelia (Figure S2B). The collected mycelia were then rinsed with cold sterile physiological water to remove any attached debris in a 50 mL Falcon tube (Figure S2C) and then transferred into microcentrifuge tubes, which were gently squeezed to remove excess liquid, and their weights were recorded. Following sampling, all tubes were flash-frozen in liquid nitrogen and kept on dry ice in a cooler box before being stored at − 80°C. The absence of mycelia in the non-mycorrhizal RFCs (M0) was confirmed by the lack of root colonization by AMF.
DNA extraction, PCR amplification, and amplicon sequencing of bacterial 16S rRNA gene
Prior to the DNA extraction process, the soil samples underwent lyophilization for 3 days and were sieved at 1 mm to facilitate the separation of substrate particles. DNA was extracted from 250 mg of homogenized soil from each sample using the DNeasy PowerSoil Pro Kit (Qiagen, Toronto, ON, Canada) following the manufacturer’s instructions. For DNA extraction from the mycelia, the DNeasy Plant Mini Kit (Qiagen, Toronto, ON, Canada) was utilized. Initially, approximately 50 mg of frozen samples were ground in liquid nitrogen in a 1.5 mL tube containing sterile white sand to disrupt the material. Subsequent steps were carried out according to the manufacturer’s guidelines. In the final step of the extraction process, the DNA was eluted in 50 µL of elution buffer for the soil and in 20 µL for the mycelia and then stored at − 20°C.
Visualization of the extracted DNA was performed using 1% agarose gel electrophoresis stained with GelRed (1/10000) and the GelDoc System (BioRad, Montreal, QC, Canada). Quantification of the DNA was conducted using a Qubit 2.0 Fluorometer and the Qubit double-stranded DNA (dsDNA) HS Assay Kit (ThermoFisher, St-Laurent, QC, Canada).
The V3–V4 region of the bacterial 16S rRNA gene underwent amplification using 341F/806R barcoded primers, as previously described [33, 35]. The primer pair consisted of CS1_341F ACACTGACGACATGGTTCTACACCTACGGGNGGCWGCAG-3’ and CS2_806R 5′-TACGGTAGCAGAGACTTGGTCTGACTACHVGGGTATCTAATCC-3′) (Farnelid et al., 2019). PCR reactions were conducted in a 25 µL volume mixture comprising 10 µL of Platinum Direct PCR Universal Master Mix (ThermoFisher, St-Laurent, QC, Canada), 0.25 µM of each primer, 4 µL of Platinum GC Enhancer, 8 µL of water, and 1 µL of template DNA. Negative controls containing only water were included in each PCR run. The PCR reactions were carried out using an Eppendorf Mastercycler Pro thermocycler (Eppendorf, Mississauga, ON, Canada) with the following cycling conditions: initial activation at 94°C for 2 min, followed by 35 cycles of denaturation at 94°C for 15 s, annealing at 60°C for 30 s, and extension at 68°C for 20 s, with a final extension at 68°C for 1 min and held at 10°C. Amplification success was confirmed by analyzing the PCR products on a 1% agarose gel. Subsequently, the amplicons were subjected to sequencing on Illumina MiSeq using the paired-end 300 libraries at the Centre d'Expertise et de Services of Genome Quebec (Montreal, QC, Canada).
Sequence processing and bioinformatic analyses
Bioinformatic analyses were conducted within QIIME2 environment version 2021.4.0 [36]. DADA2 pipeline v1.18.0 [37] was also employed to process the 16S rRNA gene sequences. Initially, Cutadapt 3.4 was utilized to remove the primer sequences from the 16S rRNA gene amplicons, setting parameters such as “minimum-length = 50”, “times = 2”, “overlap = 6”, and “p-error-rate = 0.1”. Subsequently, sequences with less than 220 bp were filtered out using the command “--p-trunc-len”, as the base quality tended to degrade below that threshold in our dataset. Following this, the amplicon sequence variant (ASV) table was generated, and chimeras were removed. The ASVs were then taxonomically classified using the naive Bayesian classifier method with the SILVA and RDP databases, and the identities of specific ASVs were manually confirmed using BLASTn against the NCBI (nr/nt) database.
Statistical analyses
Statistical analyses of plant-related parameters, AMF biomass, and mycorrhizal colonization data were conducted using JMP Pro 17 software (SAS Institute Inc., Cary, NC, USA). The normality of the data was assessed using Shapiro–Wilk tests, and variance homogeneity was examined using Levene’s test. Prior to statistical analyses, the data on the shoot fresh biomass, chlorophyll concentration, mycorrhizal colonization, and AMF biomass were Box–Cox transformed to normalize the distributions. An analysis of variance (ANOVA) was then performed to evaluate the effects of AMF species, microbial addition, and their interactions. AMF inoculation and microbial addition were considered fixed effects, and the random effect was attributed to the block. Tukey’s honestly significant difference (HSD) test was used to compare the mean values that differed at P > 0.05.
Microbial analysis was carried out using the vegan v2.6-4 [38], agricolae v1.3-7 [39], and indicspecies v1.7-14 [40] packages in R (version 4.3.2, https://www.r-project.org). The rarecurve function of the vegan package was employed to standardize the dataset by randomly subsampling read data from each sample to the lowest number of reads encountered for a sample. The alpha diversity was assessed by calculating the Shannon and Simpson indices using the vegan package, and the Kruskal–Wallis test was employed to compare alpha diversity between the two biotopes. The significance of AMF identity on alpha diversity was tested using ANOVA, with post hoc tests conducted using the agricolae package. The beta diversity among the samples was computed using the Bray–Curtis distance on bacterial ASVs and visualized with principal coordinate analysis (PCoA) using the vegan package. The community dispersion of bacteria in each biotope was calculated using the “betadisperse” function of the vegan package. The effect of AMF identity and microbial addition on the bacterial community structure was evaluated using PERMANOVA with the Adonis function of the vegan package, employing Hellinger-transformed data and 999 permutations. Indicator species analysis was performed using the indicspecies package to identify the taxa significantly associated with treatments. The term “core bacteriota” was assigned to the set of bacterial taxa present in 90% of hyphae or hyphosphere soil samples associated with the AMF host species [41].