Soil microbes and nutrient inputs influence root nodulation and tree performance in Alnus glutinosa dataset

The aim of this study is to understand the influence of soil biotic (microbial communities, characteristic of young and mature forests) and abiotic (available N and P) characteristics on F. alni nodulation and A. glutinosa tree performance. Seeds of black alder were obtained from the Dutch nature agency Staatsbosbeheer. The root nodules that were used to make the inoculum for all experiments in this study were collected from a single source: A. glutinosa trees that were growing adjacently in natural conditions (small forested area) in Leiden, Netherlands. Nodules were collected fresh each time on the same day that a new experiment was initiated. Only healthy, intact nodules from visibly robust root systems from mature A. glutinosa trees were selected for inoculum preparation. 2.1 Microbial community inocula To obtain microbial inocula, we set up an experiment where communities of bacteria, fungi and bacteria + fungi that originated from forest soils of two distinct developmental stages were cultured in artificial media. These forests were classified based on their planting periods as "young" (2010–2015) or "mature" (1880–1927). Six forests from each developmental stage were sampled in February 2022, resulting in 12 distinct forest sites, each serving as a biological replicate representative of its developmental stage (See SI, Table S1). All forests were located in Drenthe, Netherlands, featured sandy soils, and were initially planted with oak species (Quercus robur (L.) and Quercus petraea (Matt.) Liebl.). All six of the young forest plantations were planted on agricultural land and shared a similar crop cultivation history. Despite the initial planting with oaks, several other deciduous trees, including A. glutinosa, which are often found together with oak (Funk, 1990), were present in all forest stands at the time of sampling. In each forest stand, mineral soil was collected from multiple points at a depth of 15 cm, each point ranging at least 10 m apart. Soil samples from these points were combined to create a homogenized sample for each forest. Additionally, three soil cores (150 cm³ each, 12 cm deep) were collected from each forest to analyze soil properties using standard methods (see SI, Methods S1; Table S1). All experiments were conducted in the laboratories of the Institute of Biology at Leiden University. To create community cultures of bacteria, fungi and bacteria + fungi from each forest, we thoroughly mixed phosphate buffer (1 g KH₂PO₄ in 1 L DI H2O, pH 6.5) with forest soil (1:2, w/w) which was then filtered through two decreasing mesh sizes (first 1 mm and then 250 µm) to remove large debris and soil particles. The filtrate from each forest soil was centrifuged (4700 rpm, 10 ℃) to form a pellet which was resuspended in LB medium after decanting the supernatant. Each resuspension was stored in a 3:2 ratio of filtrate to glycerol (80%) and kept at -20 ℃ (Riis et al., 1998) until the time of preparing the cultures. To prepare the community cultures, 100 μl of frozen glycerol stocks for each forest was inoculated in tryptic soy broth (TSB) supplemented with the fungicide nystatin (0.1 mg/ml) to create bacterial cultures and another 100 μl in potato dextrose broth (PDB) supplemented with the antibiotics chloramphenicol (50 mg/ml) and ampicillin (100 mg/ml) to create fungal cultures. Cultures were prepared in duplicate to also create the treatment of bacteria + fungi and were incubated for a week (180 rpm, at 28°C) in an Innova S44i incubator shaker (Eppendorf, Nijmegen, Netherlands). After one week of incubation, fungal hyphal balls (formed in the fungal cultures) were fragmented and homogenized with sterile tweezers under a laminar flow cabinet. Each culture was centrifuged, and the microbial pellet was resuspended in 50 ml of KH₂PO₄ buffer. For bacteria + fungi-community cultures, 25 ml of bacterial and 25 ml of fungal cultures were combined to standardize microbial volume between treatments. To obtain the initial composition of each community culture, 1 ml of each suspension was centrifuged, and the microbial pellet was stored at -20°C for DNA analysis. The microbial suspensions were adjusted to 300 ml with KH₂PO₄ buffer and divided into two 150 ml aliquots — one for pots that would later also receive F. alni inoculum and one for pots without F. alni inoculum. In total, 72 inocula were made (6 young and 6 mature forests × 3 microbial treatments × F. alni 2 inoculation conditions). Next, from each of those suspensions, 2 ml was transferred to Eppendorf tubes, where four-week-old A. glutinosa seedlings (similar root and shoot length with at least two mature leaves) were left to incubate for 30 minutes at root level to facilitate seedling inoculation. For seedlings of the control treatment, 2 ml MiliQ H2O was used instead. These seedlings were previously surface sterilized and germinated in 0.5 MS medium (Kahrizi et al., 2018) to ensure seedlings were not colonized by other microbes (SI, Methods S3). Following the inoculum, the seedlings were transplanted to 1 L pots (11x11x12cm) filled with gamma-sterilized grassland soil for which soil analysis was also performed as in Georgopoulos et al., (2025) This soil was sandy (98% sand, 1.4% silt + clay) with very little gravel (0.6%), had moderate to low available N levels (NH4: 37.82 ± 1.61 mg/kg soil; NO3: 15.41 ± 0.57 mg/kg soil) and low orthophosphate (PO4: 3.86 ± 0.41 mg/mg soil) as well as slightly basic pH (7.37 ± 0.02) and low SOM (1.85 ± 0.04%). Immediately after transplantation, the corresponding 150 ml aliquot of the same microbial community culture used for seedling incubation was evenly applied to the soil surface around each seedling.. To study the effects of bacteria, fungi and bacteria + fungi communities from each developmental stage on F. alni nodulation, half of the pots of each community treatment were additionally inoculated with F. alni slurry, which was created fresh on the day of the experimental setup (see: SI, Methods S2). This was done right after inoculating with a community culture, by pipetting 1 ml of nodule homogenate (100 mg nodule tissue / 1 ml H2O) into a small indentation made right next to the roots of each seedling. Finally, six control pots with gamma sterilized soil received just the phosphate buffer without any inocula while another six controls received the phosphate buffer alongside the F. alni inoculum making up the total number of pots to 84. Plants grew for 12 weeks in a climate room with relative humidity 70%, light regime of 16h:8h (light:dark), air temperature of 20 ℃ (light) and 18 ℃ (dark), and were watered three times per week until saturation point of the soil (~30% of soil volume). During this time, their stem height, number of leaves and chlorophyll were recorded weekly. After 12 weeks, all plants were harvested, and soil subsamples were taken and oven-dried (40 ℃) for nutrient analysis. At the harvest, plants were unpotted and their roots were thoroughly rinsed with running tap water. Hereafter, root nodules were counted, cut from the roots with a razorblade and dried at 40°C to obtain biomass. Additionally, small root subsamples were stored at -20 ℃ for DNA analysis and larger subsamples were preserved in water at 4°C, for one day, to be used later for root morphological characterization via scanning. Root scanning was performed using an Epson Perfection V850 Pro scanner at 1200 dpi. Root length, volume and diameter, specific root length (SRL), and the percentage of fine roots (diameter < 0.3 mm) were measured using WinRHIZO software (Regent Instruments, Quebec, QC, Canada). The subsamples were then oven-dried for 96 hours (40°C). Following leaf removal, the stems were separated from the roots, and all components were oven-dried similarly to the root subsamples to obtain aboveground, belowground and nodule biomass. For all experiments in this study, nodule biomass was used instead of the number of nodules or nodule density to avoid redundant metrics (See SI, Fig. S1, Methods S4). Nodule dry weight directly reflects the biomass of nodules, which is closely related to their capacity to fix nitrogen (Aranjuelo et al., 2014). Since nitrogen-fixing nodules contribute more effectively to plant growth as their biomass increases (Fischinger and Schulze, 2010, Fischinger et al., 2010), dry weight offers a more direct measure of this biological function. Both the number of nodules and nodule density may not translate as directly to nitrogen-fixing capacity, as they don’t necessarily represent the size or quality of the nodules. To analyze the N-content of the leaves, a QIAGEN TissueLyser II Bead Mill (Hilden, Germany) was used to grind the oven-dried leaf samples at 370 rpm for 5 min. The percentage of leaf N was measured using the dry combustion method (Matejovic, 1997) with a Thermo Scientific FLASH 2000 CN analyzer (Milan, Italy). For the soil samples from the beginning and the end of the experiment, the soil NH4+-N and NO3--N concentrations were measured with a spectrophotometer using a standard 1M KCl extraction method (Kachurina et al., 2008), whereas PO43--P concentrations were determined employing a 0.01M CaCl2 extraction technique (Houba et al., 2008). The final concentrations were expressed in mg of NH4+-N, NO3--N, and PO43--P per kg of soil. 2.2 Microbial isolates Besides evaluating the effect of bacteria, fungal and bacterial + fungal communities on nodulation, we were also interested in the effects of individual microbial species from these communities, thus at the time of the harvest, we subsampled small root pieces from the treatments where fungi were added from young and mature forests. We decided to focus on these two treatments after assessing the results of the above-described mesocosms (see results), in an attempt to identify specific microbes that may contribute to the observed patterns. For isolation, roots were ground to a slurry in a mortar using 4 ml MQ H2O. Then, 50 μl from each root slurry was pipetted and spread onto potato dextrose agar (PDA) plates with bactericide (chloramphenicol), which were incubated in the dark at room temperature (21 ℃) for ~10 days. From the resulting growth, morphologically distinct colonies were isolated by streaking them in clean PDA plates until isolates were pure. These were then stored in 80% glycerol at -80oC for later inoculation, and also in Cetyltrimethylammonium bromide (CTAB;1:1 ratio) for identification. Although the initial aim was to isolate fungi, some of the isolates were unexpectedly bacteria (see identification below). In total, 14 strains were isolated, of which six (three bacteria and three fungi) were selected from the roots of trees that had received fungal communities from young forests, and eight (two bacteria and six fungi) from those that received fungal communities from mature forests (see, SI, Fig. S2). To revive the inocula for each strain, 5 μl from the glycerol stocks was added to 2 ml of PDB, and incubated (100 rpm, 21 ℃). After one week, 10 μl of each culture was transferred to an Erlenmeyer flask containing PDB to generate biomass. These cultures were then processed as described in the previous experiment. As before, 2 ml of each culture was transferred to Eppendorf tubes (12 tubes per isolate), where four-week-old A. glutinosa seedlings of similar root and shoot length and bearing at least two mature leaves—were incubated at the root level for 30 minutes to facilitate inoculation. Following incubation, the seedlings were transplanted into 1 L pots filled with gamma-sterilized grassland soil. Immediately after planting, 1 ml of the corresponding isolate culture was applied directly to the soil near the seedling roots.. Out of 12 seedlings inoculated with each isolate, half were randomly selected to also receive F. alni inoculum, while the remaining seedlings did not receive F. alni (n = 6 replicates). In addition, six pots containing gamma-sterilized soil were set up as controls. The seedlings in these pots did not receive inocula. Another six pots received only the F. alni inoculum, totaling the number of pots to 180. The seedlings were left to grow for 12 weeks, in the same conditions as described before and then harvested as described above. 2.3 Microbial cultures and root-associated microbial communities Community culture DNA and bacterial isolate DNA were extracted using the DNeasy PowerSoil Pro kit (Qiagen Inc., Hilden, Germany), while DNA from root samples was isolated with the DNeasy Plant Pro kit (Qiagen Inc., Hilden, Germany), following the manufacturer's protocols. For the community cultures, 2ml of each culture was used for DNA extraction while for the roots, ~100 mg of tissue was used. Fungal isolate DNA was extracted using a CTAB-based method (Schenk et al., 2023). For bacterial analysis, the primers 515F (GTG YCA GCM GCC GCG GTA A) and 926R (GGC CGY CAA TTY MTT TRA GTT T) were used to target the V4 region of the 16S rRNA gene (Quince et al., 2011; Parada et al., 2016). For fungal analysis, the rRNA ITS2 region was amplified using the primers gITS7ngs (GTG ART CAT CRA RTY TTT G) and ITS4ngsUni (CCT SCS CTT ANT DAT ATG C) (Tedersoo and Lindahl, 2016). Prior to sequencing, PCR checks were performed for all reactions. Library preparation was performed at NovoGene UK and sequenced with 2x250 paired-end chemistry in Illumina NovaSeq6000 device. Isolated microbes were sequenced using Sanger sequencing and were identified based on the top five blast results (SI, Table S2). Isolate 6 gave different results for the forward and reverse sequences and was determined visually as Pseudomonas sp. by comparing its morphology to other isolated and identified colonies. 2.4 Soil nutrient manipulation An experiment was established to investigate the hypothesis that increasing nitrogen will inhibit F. alni nodulation while increasing phosphorus will increase F. alni nodulation. Seedlings were prepared as described above and transferred to 1 L pots inside the climate room under the same conditions and using the same soil as before. A gradient of increasing N was created where pots were fertilized with 7.5, 10, and 20 mM NH₄NO₃ solution (Sigma Aldrich). For the P treatment gradient, pots were fertilized with KH₂PO₄ solution (Sigma Aldrich) at concentrations of 1, 3 and 6 mM. For both nutrients, 15 mL of each respective solution was applied to the pots twice a week for a total of 12 weeks (SI, Table S3). Each nutrient concentration was applied to 20 pots. Of these, half of the pots received F. alni inoculum, while the remaining pots were not inoculated, serving as fertilized but non-inoculated controls (n = 10 replicates). Additionally, 20 unfertilized control pots were prepared, with half the pots receiving only F. alni inoculum and the remaining pots left untreated and unfertilized. Lastly, an additional treatment was established using the highest concentrations of NH₄NO₃ (20 mM) and KH₂PO₄ (6 mM) to assess whether phosphorus could mitigate the inhibitory effect of high nitrogen concentrations on F. alni nodulation through lowering the N/P ratio. For this treatment, pots were treated similarly, with 15 mL of the high N+P solution applied twice a week for a total of 12 weeks (SI, Table S3). The plants were left to grow for 12 weeks, during which time their stem height, number of leaves and chlorophyll were recorded weekly. After 12 weeks, all plants were harvested and treated similarly to the other experiments.