Soil Chemistry and Microbial Community Patterns Across Tricholoma matsutake Fairy-Ring Developmental Stages in Yeongju, South Korea

Abstract

This study investigates how shiro developmental stages, soil physicochemical properties, and seasonal variation shape fungal and bacterial communities associated with Tricholoma matsutake in a Pinus densiflora forest in Yeongju, South Korea. Seasonal soil samples from past, present, and future shiro zones were analyzed using ITS and 16S rRNA metabarcoding together with soil chemical measurements. Shiro-driven spatial heterogeneity, rather than seasonality, was the dominant factor structuring microbial communities. Fungal assemblages differed significantly among shiro stages, with exchangeable potassium (K⁺) emerging as the primary driver and a strong predictor of T. matsutake abundance. Elevated K⁺ in active shiro zones corresponded to reduced fungal diversity, suggesting competitive exclusion by dominant ectomycorrhizal taxa. In contrast, bacterial communities were shaped mainly by water-soluble iron (Fe), shifting from Acidobacteria-rich assemblages in past zones to Proteobacteria in active zones. The enrichment of siderophore-associated taxa suggests a potential role of Fe acquisition processes, broadly consistent with mechanisms proposed in the mycorrhiza helper bacteria hypothesis, though not directly tested here. Overall, T. matsutake development generates nutrient-specific biogeochemical gradients-K⁺ for fungi and Fe for bacteria-that reorganize soil microbial communities. These findings underscore tightly linked biotic-abiotic interactions in shiro ecology and highlight microbial and chemical features that may serve as indicators of shiro activity.

1. Introduction

Ectomycorrhizal (ECM) fungi are central to forest ecosystem functioning: they enhance host uptake of water and mineral nutrients and mediate coupled C-N-P cycles and belowground biogeochemistry across scales [1–3]. Within this guild Tricholoma matsutake (S. Ito & S. Imai) Singer is a culturally and economically important species in temperate East Asian conifer forests. T. matsutake forms dense mycelial aggregates (shiro) that are visually and physicochemically distinct from surrounding bulk soils, typically within mineral horizons at ca. 5–20 cm depth and often associated with sandy textures, low pH, and low water-holding capacity [4]. These distinctive edaphic conditions suggest a strong habitat filter on co-occurring soil biota and on T. matsutake productivity itself.

Amplicon-based next-generation sequencing (NGS) has revealed consistent patterns in shiro soils: fungal α-diversity is reduced and community composition is simplified relative to adjacent non-shiro soils, with T. matsutake dominance shaping the mycobiome across landscapes [5–7]. While fungal dominance is well-documented, the shiro also harbors distinctive bacterial assemblages that may influence or respond to fungal activity. Previous work has shown that Acidobacteria, Proteobacteria, and Actinobacteria often dominate T. matsutake-associated soils, with shifts in these groups reflecting nutrient availability, redox conditions, and metal dynamics [8–10]. Some of these bacteria may function as mycorrhizal helper bacteria (MHB) [10], hypothesized to support fungal growth by mobilizing recalcitrant nutrients or producing siderophores for iron acquisition. Yet, specific links between bacterial community succession and soil chemical gradients remain understudied compared to fungal dynamics, and integrated fungal-bacterial perspectives remain limited [11]. In parallel, comparative work across production regions indicates that soil chemistry varies with location and developmental stage for example, acidic pH (ranging approximately from 3.95 to 6.56) and heterogeneity in base cations and trace metals (e.g., Fe, Mn, Zn and Cu) are common features of T. matsutake habitats, and region-specific differences in K, Mn, and Ca have been noted [12,13]. Together, these studies point to an interplay between shiro development and soil physicochemistry.

However, most field designs remain snapshot-style, often restricted to the harvest season, limiting inference about seasonality and its interaction with shiro progression and soil chemistry [5–7]. As a result, we still lack integrated assessments that, on the same samples, jointly evaluate (i) shiro developmental stage, (ii) season, and (iii) soil physicochemistry, and relate these to NGS-derived community metrics and T. matsutake relative abundance.

Here, we address this gap by systematically sampling soils across shiro stages, covering past, present, and future zones, and seasons. We quantified soil chemistry (pH, organic matter, available N and P, exchangeable cations, and related edaphic variables) and characterized both fungal and bacterial communities using ITS and 16S rRNA gene metabarcoding, respectively [14]. We test three broad expectations: (i) shiro developmental stage will explain substantial variation in soil microbial communities; (ii) soil physicochemical properties, particularly specific nutrient gradients such as K and Fe, will correlate differently with shifts in fungal and bacterial assemblages; and (iii) seasonal effects, if present, will be secondary compared with zone level differences.

2. Materials and methods

2.1. Study site and soil sampling

Soil sampling was conducted in Yeongju, South Korea, within a T. matsutake production forest dominated by Pinus densiflora (Siebold & Zucc). To examine seasonal variation, soil sampling was conducted across four time points corresponding to Spring (April), Summer (June), Autumn (September), and Winter (December). Shiro developmental zones were classified into past, present, and future zones based on the visual presence of the characteristic grayish white hyphal aggregation (Supplementary Figure 1). The past zone refers to areas where shiro activity has diminished, often showing weak or degraded hyphal traces; the present zone corresponds to the active mycelial front where dense white hyphae are concentrated; and the future zone indicates the outward expanding margin where hyphae are beginning to extend but have not yet formed a clearly visible, continuous shiro. After removing the litter layer, soils were collected at a depth of approximately 10 cm using sterilized tools. Control soils were collected 3–6 m outside of shiro boundaries in areas lacking visible mycelial formation. All soil samples were transferred into sterile polyvinyl bags, homogenized thoroughly, and transported on ice. Subsamples for physicochemical analysis were kept chilled, whereas soils intended for DNA extraction were stored at −20 °C. To capture within zone variability, three replicate soil cores were collected for each shiro zone in every season, homogenized within zones, and pooled to generate one composite sample per zone-season combination. In total, 23 fungal (ITS) samples and 33 bacterial (16S) samples were analyzed in this study, reflecting the number of soil samples successfully obtained for each marker.

For chemical profiling, air-dried and sieved (2-mm mesh) soil samples were subjected to detailed analysis. We determined soil pH and quantified key nutrient concentrations, including total nitrogen (Total N), total phosphorus (Total P), and inorganic nitrogen species (NH₄⁺–N and NO₃⁻–N). Additionally, soil organic matter and available iron (Fe) levels were measured to assess the specific micro-environmental conditions of the shiro zones.

2.2. DNA extraction and library preparation

Genomic DNA was extracted from 0.5 to 1 g of homogenized soil using the DNeasy PowerSoil Kit (Qiagen, Hilden, Germany), following the manufacturer’s protocol. DNA concentration and purity were quantified using the VICTOR Nivo^™ system (PerkinElmer, USA) with PicoGreen assays. Amplicon libraries were prepared according to Illumina guidelines. For bacterial communities, the V3-V4 region was amplified using primers 341 F (5′—CCTACGGGNGGCWGCAG—3′) and 805 R (5′—GACTACHVGGGTATCTAATCC—3′) [15] with standard Illumina adapters. For the fungal community, the ITS region was amplified using the primers ITS3 (5′—GCATCGATGAAGAACGCAGC—3′) and ITS4 (5′—TCCTCCGCTTATTGATATGC—3′) [16]. PCR reactions were performed using Herculase II Fusion DNA Polymerase (Agilent Technologies, USA). The thermal cycling conditions were as follows: 25 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 30 s, with a final extension at 72 °C for 5 min. The PCR products were purified using AMPure XP beads. Subsequently, 10 μL of the purified product was used as a template for the second PCR (10 cycles) with Nextera XT index primers to attach dual indices and sequencing adapters. Final libraries were purified, quantified using PicoGreen reagents, and fragment size distributions were confirmed on the TapeStation D1000 ScreenTape system (Agilent Technologies). Libraries were normalized, pooled, quantified by qPCR (KAPA Library Quantification Kit), and sequenced on the Illumina MiSeq platform. The sequencing was performed by Phyzen (Seongnam-si, Korea).

2.3. Bioinformatic processing and statistical analysis

To correct for differences in sequencing depth across samples, the dataset was normalized using rarefaction. We determined the optimal rarefaction depth based on the minimum sequencing depth observed to retain all samples for downstream analysis. Consequently, the ITS and 16S rRNA gene datasets were rarefied to 19,768 and 13,402 reads per sample, respectively. Alpha rarefaction curves for observed features confirmed that these depths were sufficient to capture the representative microbial diversity, as the curves reached a plateau (Supplementary Figure 2 and 3). Quality-filtered paired-end reads were processed in QIIME2 (v2024.2.0) [17] using the DADA2 pipeline for denoising, dereplication, chimera removal, and amplicon sequence variant (ASV) inference. Taxonomic identification of fungal (ITS2) and bacterial (16S rRNA) ASVs was performed using BLAST+ against the UNITE (v10.0) and SILVA (v138.2) reference databases with ≥80% identity and query coverage thresholds. Alpha diversity indices (Observed ASVs, Chao1, ACE, Shannon, Simpson, and Good’s coverage) and rarefaction curves were computed in QIIME2. Beta diversity was assessed using Bray-Curtis dissimilarity and visualized via Principal Coordinate Analysis (PCoA). Further statistical analyses were conducted in R (v4.4.3). Differences in community composition across shiro developmental stages and seasons were tested using PERMANOVA (function “adonis2” in the vegan package) [18]. Similarity Percentage (SIMPER) analysis (function “simper” in the vegan package) was used to identify bacterial taxa contributing to between-group dissimilarities. For correlation analyses between soil physicochemical variables, T. matsutake abundance, and dominant bacterial taxa, centered log-ratio (CLR) transformed relative abundances were used to reduce compositional bias. For univariate comparisons, data normality and homogeneity were tested prior to analysis. Parametric comparisons were conducted using pairwise t-tests with Benjamini-Hochberg (BH) corrections, while non-parametric data were analyzed using Kruskal-Wallis tests followed by Dunn’s post hoc tests. Spearman’s rank correlation and linear regression were used to evaluate relationships between soil chemistry and fungal or bacterial abundance patterns. Statistical visualizations, including scatterplots and regression models with R² and p values, were generated using ggplot2 [19]. Statistical significance was set at p < 0.05.

3. Results

3.1. Soil chemical properties across fairy-ring developmental zones

Soil chemistry showed distinct variation across the three developmental zones of the Tricholoma matsutake mycelium. Concentrations of water-soluble iron (Fe), available phosphorus (P) and electrical conductivity (EC) were significantly higher in the shiro (Present) zone than the inner (Past) zone (p < 0.05, Figure 1). While exchangeable potassium (K⁺) showed an increasing trend in the shiro (Present) zone, the most pronounced significant alterations driven by T. matsutake activity were observed in Fe, P, and EC gradients (Figure 1). In contrast, organic matter, total N, and pH showed no significant differences among zones or seasons.

Figure 1.

Variations in soil chemical properties across developmental zones of Tricholoma matsutake mycelia. Box plots display the median (center line), interquartile range (box), and range (whiskers). Significant differences (p < 0.05) between the Present (active) and Past (inner) zones were observed for water-soluble iron (Fe), available phosphorus (P), and electrical conductivity (EC), as indicated by asterisks.

3.2. Fungal community composition and diversity

The fungal community composition showed clear distinct patterns across the different zones of the mycelial ring. In this study, dominant taxa were defined as those with a relative abundance greater than 1%. PCoA based on Bray-Curtis dissimilarity demonstrated clear separation into two major groups, between samples dominated by T. matsutake (relative abundance > 40%; Past and Present) and those where it was absent or less dominant (Control and Future) (Figure 2). PERMANOVA results confirmed that the fungal community composition differed significantly among the three shiro zones (p = 0.001) but showed no significant seasonal effect (Table 1). When soil variables were fitted to the ordination, exchangeable K was the only factor significantly explaining the fungal community pattern (p = 0.025, Table 1). Univariate PERMANOVA analyses assessing the effects of individual soil properties on fungal beta diversity are shown in the Supplementary Table S1.

Figure 2.

Principal coordinates analysis (PCoA) of fungal community composition. Principal coordinates analysis (PCoA) based on Bray-Curtis distances shows the fungal community composition across shiro stages (colors) and seasons (shapes). The separation along the first axis (PCoA1, 19%) distinguishes Tricholoma matsutake-dominant zones (Past and Present) from non-dominant areas (Future and Control).

Table 1.

Permutational multivariate analysis of variance (PERMANOVA) of fungal community composition based on Bray-Curtis dissimilarities.

	Term	Df	SS	R2	F value	p value
Categorical Factors	Shiro stage	3	2.440	0.361	3.722	0.001***
	Season	3	0.705	0.104	1.076	0.367
	Residual	16	3.496	0.517
	Total	22	6.755	1.000
Soil property	Exchangable K+	1	0.560	0.277	1.918	0.025*
	Residual	5	1.460	0.723
	Total	6	2.019	1.000

The model evaluates the effects of shiro stages, seasons, and significant soil environmental drivers. Available potassium (K⁺) was identified as the sole environmental factor significantly influencing fungal community structure (p = 0.025). Significance levels based on p-values are indicated as follows: p ≤ 0.001 (***); 0.01 ≥ p > 0.001 (**); and 0.05 ≥ p > 0.01 (*). Df: degrees of freedom, SS: sum of squares.

To identify the fungal taxa responsible for community differentiation across shiro developmental stages, a SIMPER analysis was performed. The analysis revealed that T. matsutake (16.1%) was the primary contributor to the dissimilarity, exhibiting a markedly higher relative abundance in the Present zone. In contrast, several Genus taxa, including Phialocephala sp. (11.5%), Thelephoraceae sp. (6.1%), and Fungi sp. (8.8%), showed higher abundance in the Past zone. Additional contributors such as Umbelopsis sp. (9.0%) and Helotiales sp. (6.4%) also displayed zone-specific enrichment patterns, collectively indicating a distinct shift in fungal community structure between Past and Present zones (Figure 3). Furthermore, genus-level relative abundance bar plots of fungal communities across shiro developmental stages and control soils are provided in Supplementary Figure 4. The Past and Present zones were dominated by T. matsutake, whereas this genus was detected at very low relative abundance or was nearly absent in the Future zone and control soils. These profiles illustrate clear differences in fungal community composition across shiro stages and non-shiro soils.

Figure 3.

SIMPER analysis of fungal community differences (Past vs Present). Top fungal taxa contributing to community dissimilarity between Past and Present zones. Blue indicates higher abundance in Past, and red indicates higher abundance in Present.

Alpha-diversity analysis of the fungal community with shiro stages was performed the Observed ASVs, ACE, Chao1, Shannon, and Simpson indexes. Although alpha-diversity indices (ACE, Chao1, Shannon, and Simpson) showed no significant seasonal variation (p > 0.05, Figure 4A), they differed significantly according to T. matsutake mycelial development (Figure 4B). Specifically, the future zone exhibited higher richness (ACE, Chao1, and Observed ASVs) and diversity (Shannon and Simpson indices) compared to the past and present zones (p < 0.05), showing levels similar to the Control. These results suggest the reduction in fungal richness and diversity within the shiro (Past and Present zones) attributable to the dominance of T. matsutake.

Figure 4.

Fungal alpha diversity indices across seasons and shiro stages. Alpha diversity indices were analyzed using richness estimators (ACE, Chao1, Good’s coverage, and observed ASVs) and diversity indices (Simpson and Shannon). (A) No significant differences were observed among seasons, (B) whereas shiro stages showed significant variation in diversity. Different lowercase letters above the boxplots indicate statistically significant differences among shiro stages based on Kruskal—Wallis tests followed by post hoc multiple comparisons (p < 0.05), groups sharing the same letter are not significantly different.

The correlation heatmap among alpha diversity indices and soil properties showed that diversity metrics exhibited a strong negative correlation specifically with exchangeable K, while other soil variables showed no clear associations (Figure 5). Linear regression analysis revealed a strong positive relationship between the relative abundance of T. matsutake and exchangeable K⁺ concentration (adjusted R² = 0.815, p < 0.001, slope = 1.797, Figure 6).

Figure 5.

Spearman rank correlation heatmap between fungal alpha diversity indices and soil physicochemical properties. Colors indicate the strength and direction of the correlation: blue represents a positive correlation, and red represents a negative correlation. Exchangeable potassium (K⁺) shows a strong negative correlation with both richness indices. The asterisk indicates a significant difference (p < 0.05).

Figure 6.

Relationship between Tricholoma matsutake relative abundance and exchangeable K⁺ concentration. Relative abundance of T. matsutake increased above 0.3 cmol⁺/kg exchangeable K⁺ (adjusted R² = 0.815, p < 0.001, slope = 1.797).

3.3. Bacterial community structure and environment

In contrast to the fungal community, 16S rRNA gene-based analysis of bacterial communities showed that community structure was not influenced by season but differed significantly among shiro stages. PCoA based on Bray-Curtis distance revealed distinct clustering by shiro stage, with the Past zone clearly separated from the Present and Future zones (Figure 7). PERMANOVA confirmed that shiro stage had a significant effect on bacterial community composition (Table 2, p = 0.01). Among soil variables, water-soluble iron (Fe) was the only significant factor explaining bacterial community variation (Table 2, p = 0.005). Univariate PERMANOVA analyses assessing the effects of individual soil properties on bacterial beta diversity are shown in the Supplementary Table S2.

Figure 7.

Principal coordinates analysis (PCoA) of bacterial community composition based on Bray-Curtis dissimilarities. The ordination visualizes the structural differentiation of bacterial communities across shiro stages (colors) and seasons (shapes). Distinct clustering is observed for the Past zone compared to the Present and Future zones.

Table 2.

Permutational multivariate analysis of variance (PERMANOVA) of bacterial community composition based on Bray-Curtis dissimilarities.

	Term	Df	SS	R2	F value	p value
Categorical Factors	Shiro stage	3	1.632	0.295	1.261	0.010**
	Season	1	0.429	0.077	0.993	0.456
	Residual	8	3.452	0.623
	Total	12	5.540	1.000
Soil property	Water-soluble Fe	1	0.650	0.117	1.463	0.005**
	Residual	11	4.890	0.883
	Total	12	5.540	1.000

The analysis identifies the influence of shiro stages and environmental variables on bacterial assemblages. Water-soluble iron (Fe) was the only significant soil driver explaining community variation (p = 0.005). Significance levels based on p-values are indicated as follows: p ≤ 0.001 (***); 0.01 ≥ p > 0.001 (**); and 0.05 ≥ p > 0.01 (*). Df: degrees of freedom, SS: sum of squares.

Alpha-diversity indices, including observed ASVs, Chao1, Shannon, and Simpson, showed no significant differences across seasons or shiro stages. This suggests that bacterial diversity and richness remained relatively stable despite compositional shifts in community composition observed in beta-diversity analysis (Figure 8).

Figure 8.

Bacterial alpha diversity indices across seasons and shiro stages. Alpha diversity indices were analyzed using richness estimators (ACE, Chao1, Good’s coverage, and observed ASVs) and diversity indices (Simpson and Shannon). (A) Seasonal variation in alpha diversity. (B) Alpha diversity across mycelial ring development zones. No significant seasonal or shiro stage differences were detected in any alpha-diversity indices.

Genus-level relative abundance bar plots of bacterial communities across shiro developmental stages and control soils are presented in Supplementary Figure 5. The Past zone was characterized by a high relative abundance of Edaphobacter sp., while the Present, Future, and Control soils showed different bacterial compositional profiles. The bar plots highlight stage-dependent variation in dominant bacterial taxa across shiro development.

3.4. Specific bacterial taxa and interaction with T. matsutake

To identify the specific bacterial taxa driving this community differentiation associated with iron availability, a SIMPER analysis was performed. The analysis identified Edaphobacter sp. (15.1%) and Alloacidobacterium sp. (9.6%) as the top contributors to the dissimilarity, with both taxa exhibiting higher relative abundance in the Past zone. In contrast, taxa such as Afipia sp. (6.5%), Chthoniobacter sp. (6.3%), and Pseudacidobacterium sp. (5.6%) were significantly enriched in the Present zone (Figure 9).

Figure 9.

SIMPER analysis of bacterial community differences (Past vs Present). Top bacterial taxa contributing to community dissimilarity between Past and Present zones. Blue indicates higher abundance in Past, and red indicates higher abundance in Present.

4. Discussion

The distinct spatial differentiation of microbial communities across the T. matsutake shiro stages indicates that shiro development profoundly alters the surrounding soil environment [5,7]. These differences indicate that T. matsutake activity alters the surrounding soil environment, potentially enriching specific nutrients within the active mycelial area (Figure 1) [4,20]. Chemical gradients, including water-soluble iron (Fe), available P, EC, and exchangeable K suggest that T. matsutake mycelia actively modify soil nutrient dynamics through organic acid secretion, mineral weathering, and selective nutrient uptake [13,21]. Similar micro-scale weathering and cation mobilization at the hypha-mineral interface has been reported for ectomycorrhizal fungi, demonstrating that fungal-driven mineral transformations can occur even when bulk-soil nutrient concentrations show minimal variation [22,23].

The strong relationship between exchangeable K and T. matsutake abundance corroborates the hypothesis that K availability is an important determinant of fungal proliferation, as observed in other soil-microbial systems [20,24]. Consistent with this, our result (Figure 6) indicates that T. matsutake responds sensitively to potassium availability, potentially utilizing or altering K dynamics within its shiro environment [20,24]. Although bulk exchangeable K⁺ did not differ significantly among zones, the Past–Present–Future pattern showed a subtle but directional increase toward the active shiro. Such small but consistent gradients are not always detected by univariate mean comparisons, yet can still structure fungal communities. This interpretation is supported by our PERMANOVA, where exchangeable K⁺ was the only significant environmental predictor of fungal composition, and by the strong positive relationship between K⁺ concentration and T. ­matsutake abundance. These findings suggest that even modest increases in K⁺ availability—potentially amplified within the hyphosphere—can influence fungal competition and contribute to T. matsutake dominance, rather than implying that absolute K⁺ quantity is unimportant.

The observed reduction in fungal diversity in K-rich, shiro zones is consistent with competitive exclusion, wherein dense T. matsutake hyphae monopolize nutrients and physical space, limiting the establishment of saprotrophic and non-symbiotic fungi [5,8,25]. This supports the tradeoff hypothesis between host specialization and community-level diversity, as the ecological strategies favoring T. matsutake success simultaneously reduce fungal coexistence.

In contrast to fungal communities, bacterial assemblages showed a stronger association with water-soluble iron (Fe), highlighting the importance of redox-driven processes in shaping bacterial community structure within shiro soils [9,25,26]. The enrichment of siderophore-producing taxa such as Bradyrhizobium and Burkholderia in the Present zone suggests that these bacteria may not only tolerate Fe limitation but actively engage in Fe mobilization [27,28]. Previous work has demonstrated that root- and hypha-associated bacteria can contribute substantially to mineral weathering and Fe solubilization, thereby increasing the availability of micronutrients essential for fungal metabolism [29]. These findings align with the mycorrhiza helper bacteria (MHB) hypothesis [10], which posits that cooperative bacterial–fungal interactions may enhance ectomycorrhizal establishment under nutrient-limited conditions.

Importantly, several studies provide direct experimental support for bacterial facilitation of T. matsutake growth. Oh and Lim [30] showed that specific bacterial isolates increased T. matsutake mycelial growth in co-culture assays, while Oh et al. demonstrated that shiro-associated bacteria enhanced nutrient mobilization and supported T. matsutake physiological activity. These results confirm that certain bacterial partners can function as MHB for T. matsutake. In our study, putative MHB-like taxa including Bradyrhizobium- and Burkholderia-affiliated groups were enriched in the Present shiro zone (Supplementary Figure 5). This is consistent with the notion that bacterial–fungal feedbacks may contribute to the maintenance, nutrient acquisition, and continued development of T. matsutake shiro [31].

Microbial succession patterns further support this interpretation. Acidobacteria, which prefer oligotrophic and acidic environments, were dominant in the Past zone, whereas copiotrophic Proteobacteria expanded in the Present zone where T. matsutake metabolism likely increases nutrient turnover [9,25]. Proteobacterial taxa such as Bradyrhizobium and Afipia possess functional traits including nitrogen fixation and siderophore production [13,24], both of which may enhance nutrient availability for matsutake. Similar shifts in bacterial composition associated with T. matsutake activity have been observed in other forest systems [8], collectively indicating that matsutake-driven resource dynamics selectively promote or suppress specific bacterial taxa (Figure 8).

Notably, neither fungal nor bacterial communities exhibited significant seasonal turnover (Figure 2 and 7) and alpha diversity of both groups remained stable across seasons. This finding contrasts with the pronounced seasonal shifts frequently reported in temperate forest soils [32,33]. A plausible explanation is that the pine–shiro complex generates a buffered microenvironment that mitigates seasonal fluctuations in temperature, moisture, and nutrient availability, thereby stabilizing microbial diversity within shiro soils [32,33]. Dense T. matsutake hyphae and a stable mineral soil matrix may dampen climatic variation within the immediate hyphosphere [2,34,35]. Additionally, our sampling design focused on a single forest type and a consistent soil horizon, reducing habitat heterogeneity relative to broader landscape-level studies and potentially limiting detectable seasonal effects.

Bacterial alpha diversity did not vary across shiro zones or seasons, whereas beta diversity differed significantly with shiro stage (Figure 7 and 8). This pattern suggests that shiro development reorganizes bacterial community composition through species turnover rather than changes in overall richness (Supplementary Figure 5). In other words, taxa are being replaced rather than added or lost, indicating that environmental filtering and niche differentiation along the shiro gradient drive bacterial compositional shifts without impacting total diversity [36].

Taken together, our results show that T. matsutake shiro development is associated with steep biogeochemical gradients that distinctly restructure both fungal and bacterial communities. Fungal assemblages primarily responded to exchangeable K⁺ dynamics, while bacterial assemblages were strongly shaped by Fe availability and siderophore-mediated interactions. This coordinated response underscores the tight coupling between biotic interactions and soil chemical processes in ectomycorrhizal systems [7,20,24,37]. Understanding these relationships provides critical insight into the ecological mechanisms underlying T. matsutake distribution and the maintenance or decline of microbial diversity within shiro environments.

Funding Statement

This work was supported by the National Institute of Forest Science under a grant (Forest Science Research project number FG0603-2021-01-2025).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All data presented in the study are openly available in the National Center for Biotechnology Information (NCBI) database under Sequence Read Archive (SRA) at accession number PRJNA1370007.