Effects of Pieris japonica (Ericaceae) dominance on cool temperate forest altered-understory environments and soil microbiomes in Southern Japan

Yuji Tokumoto; Ayumi Katayama

doi:10.1371/journal.pone.0296692

Abstract

The number of plants unpalatable to deer increases with increasing deer numbers. In the Kyushu Mountain area of Southern Japan, Pieris japonica (Ericaceae), an unpalatable shrub, has become the monodominant vegetation under evergreen conifer and deciduous broad-leaved tree stands. The monodominance of unpalatable plants in the understory has potential advantages and drawbacks; however, the effects of Pieris dominance are not well understood. To assess the effects of P. japonica dominances on forest environments and ecosystems, we investigated understory environments and soil microbiomes in Pieris-dominant sites. Under the deciduous broad-leaved trees, Pieris dominance leads to considerable Pieris leaf litter and humus weights and low soil bulk density and canopy openness. In the soil fungal community and fungal functional groups, the relative abundance of symbiotrophic fungi, particularly ectomycorrhizal fungi in Pieris-dominant sites were lower than in other-vegetation understory sites and saprotrophic fungi vice versa. Because few seedlings and saplings were found under Pieris shrubs, Pieris dominance in the understory might exclude other plant species. The results of this study will contribute to the Pieris population and forest management following deer overgrazing.

Citation: Tokumoto Y, Katayama A (2024) Effects of Pieris japonica (Ericaceae) dominance on cool temperate forest altered-understory environments and soil microbiomes in Southern Japan. PLoS ONE 19(1): e0296692. https://doi.org/10.1371/journal.pone.0296692

Editor: Jian Liu, Shandong University, CHINA

Received: August 30, 2023; Accepted: December 17, 2023; Published: January 11, 2024

Copyright: © 2024 Tokumoto, Katayama. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All the raw sequence data of the soil prokaryotic 16S rDNA, the soil eukaryotic 18S rRNA genes, and the root-associated fungal 18S rRNA genes were submitted in the Sequence Read Archive of DDBJ database under the accession number DRA016984, DRA016985, and DRA016986, respectively. All relevant environmental data are within the manuscript and its Supporting Information files.

Funding: Yuji Tokumoto, the Botanical Research Grant of ICHIMURA Foundation for New Technology (No. 2022-15 and 2023-20) Yuji Tokumoto, Leading Initiative for Excellent Young Researchers (LEADER) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (Grant Number JPMXS0320220123) Ayumi Katayama, Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Scientific Research (B) (Grant Number 22H03793). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Increasing deer populations have altered forest environments worldwide [1]. Deer overgrazing affects plant covers, which leads to changes in species composition, biodiversity, and environments. Deer overgrazing increases the populations of plants unpalatable to deer. This can lead to unpalatable plants overlaying forest understories, which are referred to as understory canopies [2, 3]. Unpalatable plants have specific physical traits, such as tough leaves and needles, and they contain poisonous chemicals, which prevent deer eating and/or digesting them easily [4]. Although some of these plants can be used for slope revegetation to prevent soil erosion and decrease plant species diversity [5], the effects of increasing unpalatable plants on ecosystems, both positive and negative, are not well understood. In addition, little is known about the management of overabundant unpalatable plants compared with the effects of deer overgrazing.

Ericaceae plants include several types of trees unpalatable to deer, and they form the monodominant understory layers in temperate and boreal forests in Europe and North America [2]. The effects of Ericaceae dominance have been confirmed by the increased thickness of the humus layers, and changes in soil chemical properties [6, 7]. Ericaceae plants produce polyphenol-rich litter that contain high levels of tannin, which influence the regeneration of overstory trees, soil organisms, and carbon (C) and nitrogen (N) cycles [6–8]. Ericaceae plants have mutualistic relationships with ericoid mycorrhizae (ErM), and the dominance of the Ericaceae plants can affect other symbiotic fungi, ectomycorrhiza (ECM), and arbuscular mycorrhizae (AM) fungi [7]. The growth of trees from other mycorrhiza types is inhibited and the ECM fungal root colonization is affected by the presence of Ericaceae plants and ErM [9, 10]. Tree species in forest upper layers often affect microbial community compositions, and several tree species-specific fungi have been confirmed in two main mycorrhiza types (i.e., ECM and AM) [11]. However, studies on ErM and ericoid plants are scarce [7, 12, 13], and little is known about Ericaceae plant dominance related to changes in soil microbial communities and plant–soil feedback.

One Ericaceae plant, Pieris japonica (Thunb.) D.Don ex G.Don is an evergreen shrub distributed broadly in forest understories and on the open ridges in nutrient-poor soils throughout the Japanese islands of Honshu, Shikoku, and Kyushu [14]. Pieris japonica has mutualistic relationships with ErM and AM [15], and is also known to be poisonous to mammals, as it contains asebotoxin and grayanotoxin in the plant organs [14]. In the past, few tree individuals were observed in the matured natural forest and they did not dominate understories. However, after sika deer (Cervus nippon) grazing following light environmental changes in the understories, the species population has gradually increased in forest understories and formed understory canopies because sika deer prefer not to eat this plant [14, 16–18]. It has been reported that Pieris domination leads to low soil water content and a high decomposition rate of the leaf litter in the short research term (i.e., 3–4 months) [19]; however, other environmental parameters, such as soil properties, microbiomes, and ecological functions, have not been studied. Since these phenomena caused by Ericaceae dominance can also occur in Pieris-dominant forests, the effects of expansion of this shrub should be investigated for the purpose of forest management after deer overgrazing.

To investigate Pieris dominance in the forest understory, we assessed differences in the understory environments, such as the light environment, chemical and physical soil properties, and microbial communities, between Pieris-dominant sites and those Pieris-nondominant sites.

Materials and methods

Study site

Our study was conducted at the Shiiba Research Forest of the Department of Agro-environmental Sciences, Kyushu University, Shiiba, Miyazaki Prefecture, Japan (32°23’N, 131°10’E; 1029–1079 m a.s.l.) located in the southern Kyushu Mountain area. The mean annual precipitation is 3,207.9 mm, and annual mean temperature is 10.8°C [20]. The monthly mean temperature ranges from −0.3°C in January to 24.2°C in August. The natural vegetation is a cool temperate forest comprising deciduous broad-leaved trees—Fagus crenata (Fagaceae), Quercus crispula (Fagaceae), and Carpinus spp. (Betulaceae), evergreen broad-leaved trees—Quercus salicina and Quercus acuta (Fagaceae), and conifers—Tsuga sieboldii (Pinaceae), Abies firma (Pinaceae), and Pinus densiflora (Pinaceae). Shika deer were frequently observed from 1976–1984 and the population density from 2005–2008 was approximately 20–50 individuals km⁻² [21]. Because of shika deer overgrazing, the understory vegetation, previously represented by dwarf bamboo (Sasa borealis), is now dominated by unpalatable shrubs, such as P. japonica (Ericaceae) and Illicium anisatum (Schisandraceae) [17]. We selected two study regions, Maruju (division No. 35) and Hirono (division Nos. 22, 23, and 24), where Pieris shrubs have been observed in the understory (S1 Fig, Table 1). The soil group of study site is andosole according to the classification system of the IUSS Working Group World Reference Base (2014). In the Maruju region, we selected three Pieris-dominant sites and three without Pieris (i.e., they have been understory grazed; however, they contain some grasses or mosses and no Pieris shrubs have been established) near the Pieris-dominant sites. In total, six sites were selected, and 2–3 plots (20 × 20 cm each) were selected for each site. In the Hirono region, we selected eight sites (four Pieris-dominant sites and four without Pieris) and selected 2–3 plots per site. For each plot, the types of canopy trees nearby were recorded in addition to detailed site information, including the canopy trees, tree genera, understory layers, and the numbers of plots per site, which are presented in Table 1. We categorized combinations of canopy tree types and understory trees as follows: C-Pieris; sites with a conifer canopy and Pieris in the understory, C-Other; conifers and other understory vegetation, such as grasses and mosses, DB-Pieris; deciduous broad-leaved trees and Pieris in the understory, and DB-Other; deciduous broad-leaved trees and other understory vegetation. This study was approved by the Committee for Forest Management, Kyushu University Forest, Kyushu University, Japan (#M2023-002).

Download:

Table 1. Research site descriptions with region names, types of canopy trees, understory layers, and numbers of plots per site.

https://doi.org/10.1371/journal.pone.0296692.t001

Environmental properties of the forest understories

To investigate differences in the environmental properties of the Pieris-dominant and other understory sites, we analyzed the light environment and chemical and physical soil properties (Table 2). To analyze the light environment at each plot, fish-eye images of the plots were captured using a 360° camera (theta-V; Ricoh, Japan) in June, August, and October 2022, and canopy openness as a percentage was quantified using CanopOn2 software (http://takenaka-akio.org/etc/canopon2/). In June, plot soil hardness was measured four times using a soil hardness meter (Yamanaka, Fujiwara, Japan), and the measurements were averaged per plot. To quantify the litter and humus layers in each plot, we collected A₀ horizons using a 20 × 20 cm frame. The litter was air dried and sieved using 4-mm mesh. Particles >4 mm were categorized as litter, and particles ≤4 mm were categorized as humus. The litter and humus were oven dried at 72°C for 72 h and then weighed (± 0.1 g). After weighing the litter, the Pieris leaves were divided and weighed separately. The total litter, humus, and Pieris leaf weights were then converted into g m⁻². To quantify the soil bulk density (g cc⁻¹) and the soil water content (g 100 cc⁻¹) for each plot, soil layers were collected using a 100-cc syringes. The syringes were then oven dried at 108°C for 72 h and weighed (± 0.01 mg). The soil bulk density and soil water content were calculated using the before-and-after oven-dried weights. The rock (g 100 cc⁻¹) and plant root (g 100 cc⁻¹) contents of the syringes were divided and weighed (± 0.01 mg). To analyze the soil chemical properties, we collected topsoil under the A₀ horizon in plastic bags. The soil was then sieved using 2-mm mesh and air dried, To determine the soil pH (H₂O), the soil was extracted with water at a ratio of soil to deionized water of 1:2.5 (w/w) and the pH was measured using a soil pH tester (HI-981030; Hanna Instruments, MA, US). To determine the soil conductivity, the soil was extracted with water at a ratio of soil to deionized water of 1:5 (w/w) and the electrical conductivity (EC) was measured using a soil conductivity tester (HI-98331; Hanna Instruments). The percentage of C and N and the ratio of C to N (C/N) in the soils were analyzed using an N, C, and hydrogen analyzer with an oxygen circulating combustion system (Sumigraph NC-220F; Sumika Chemical Analysis Service, Japan) according to the manufacturer’s protocol. Total soil organic matter (SOM content; g m⁻²) was calculated by soil combustion at 550°C for 12 h using a maffle furnace (FUW210PB; Advantec, Japan). Six soil nutrient contents—nitrate nitrogen (NO₃⁻), ammonium nitrogen (NH₄⁺), plant-available phosphorus (P), exchangeable potassium (K⁺), exchangeable calcium (Ca²⁺), and exchangeable magnesium (Mg²⁺)—were measured using a simplified soil nutrient analyzer (EW-THA1J; Air Water Biodesign Inc., Japan) according to the manufacturer’s protocol, described as follows: The 1.0-g air dried soils were digested in the analyzer solution (EW-T201J; Air Water Biodesign Inc.) for 30 s and then filtered through a glass fiber filter (GA-55; Advantec). The filtered solutions were then placed in the analyzer using exclusive cartridges (EW-T102J; Air Water Biodesign Inc.) and the contents were estimated using absorbance detected by LED photo detectors. The machine analyzed the seven nutrient contents within the following ranges: NO₃⁻: 1–50 mg 100 g dry soil⁻¹, NH₄⁺: 1–50 mg 100 g dry soil⁻¹, P: 1–350 mg 100 g dry soil⁻¹, K⁺:1–200 mg 100 g dry soil⁻¹, Ca²⁺:1–1,000 mg 100 g dry soil⁻¹, and Mg²⁺: 1–120 mg 100 g dry soil⁻¹. However, the P and Ca²⁺ contents of the samples were less than the lower limits of the analyzer, therefore, we did not use the P and Ca²⁺ results.

Download:

Table 2. Environmental properties of the sampling points with the results of linear-mixed-model analyses of the effects of understory vegetation and canopy trees.

https://doi.org/10.1371/journal.pone.0296692.t002

Assessment of soil microbial communities and their functions

Soil for the microbial community analysis was collected at the same time as the soil for the chemical analysis. The soils were stored in a freezer at −20°C until analysis. DNA was extracted using a kit (NucleoSpin Soil; Macherey–Nagel, Germany) following the manufacturer’s protocol. Extracted DNA was amplified in the prokaryotic 16S rRNA V3–V4 regions using a primer set for 16S 341f and 805r, and the fungal ITS1 regions using a primer set (ITS1-F_KYO1 and ITS2_KYO2) [22] with an enzyme (KOD-FX Neo; Toyobo, Japan). The constructed libraries were sequenced on the illumina MiSeq System using paired-end 300 bp read lengths and the MiSeq Reagent Kit (ver. 3) (Illumina, CA, US) at the Bioengineering Lab. Co., Ltd. (Kanagawa, Japan). The sequence quality was checked using the FASTX-Toolkit (ver. 0.0.14), and the sequences under Q20 and shorter than 130 bases were discarded using sickle (ver. 1.33). Each paired read was united using FLASH (ver. 1.2.11). Chimera and noise were removed using the Qiime2 (ver. 2022.2) dada2 plugin, and the amplicon sequence variants (ASVs) were exported with the read count. Phylogenetic estimations of prokaryotes and fungi were conducted using a feature-classifier plugin with Greengene (ver. 13_8) and UNITE (ver. 8.3) references with a 97% threshold. Fungal functionality was estimated using FUNGuild [23].

Identification of Pieris root-associated fungi

To identify root-associated fungi in Pieris roots, we analyzed DNA extracted from the freshly cut roots of six Pieris saplings (three samples were collected from each of the M01 and H03 sites). The roots were washed with sterilized distilled water three times before being sampled. We extracted the DNA using a kit (NucleoSpin Plant II; Macherey–Nagel) following the manufacturer’s protocol. Fungal ITS1 amplification, library construction, MiSeq sequence analyses, raw data curation, and taxonomical and functional identifications of the read data were conducted using the same methods described in the previous section. Detailed sample information of the height, diameter at ground level (D₀), and raw data count of the sequences are shown in S1 Table.

Statistical analysis

To assess whether the environmental conditions differed due to the presence of Pieris shrubs and the types of canopy trees, we constructed linear mixed models (LMMs) for each environmental variable. The response variables were the measured environmental variables. The fixed effects were the understory layers (Pieris shrubs or others), the types of canopy trees in each plot (coniferous or deciduous broad-leaved trees), and the interaction term of for these two variables. The random effect was the region (Maruju or Hirono). The significance of the fixed effects was tested using Type II analysis of variance (ANOVA). For a post-hoc analysis, significant variables (p < 0.05) and marginally significant variables (p < 0.10) were tested using Tukey’s multiple comparison test at the p < 0.10 threshold.

To set the read count evenly among the soil microbial community data, we rarefied the read counts for each sample of prokaryotes and fungi. The lowest number of read counts for prokaryotes was 28,592, and 36,301 for fungi. We resampled the read based on the least number for all samples and set the data matrix for the following analyses.

To compare the soil microbial alpha diversity among the sites, we calculated four alpha diversity indices (i.e., the number of ASVs, Shannon, Simpson, and inverse Simpson). The scores were analyzed using the same LMM structures described above.

To assess whether the compositions of the microbial communities differed among the sites, we calculated the Bray–Curtis dissimilarity index for all combinations of samples. The effects of Pieris shrubs in the understory, the types of canopy trees, and the interaction term for the similarity indices were tested using permutational multivariate ANOVA (PERMANOVA) with 10,000 permutations. We plotted the samples based on the Bray–Curtis index using nonmetric multidimensional scaling (NMDS). To analyze the relationships between the microbial community and the eight significant environmental variables detected by Tukey’s test (the details are presented in the Results section), the environmental variables were tested using the envfit function, and variables with p-values <0.10 were drawn on the NMDS plot.

To assess whether microbial community taxa and fungal functionality were affected by significant environmental variables detected by the envfit functions, the relative read counts of each microbial taxon at the phylum level for both prokaryotes and fungi and fungal functionality estimated by FUNGuilds were analyzed using LMM. For the prokaryote taxa, we set the SOM content as a fixed effect because the variables represented prokaryotic community changes among the samples (the details are presented in the Results section). For the fungal taxa and two fungal functionality indices (i.e., trophic mode and guild), we set the Pieris leaf and litter weights as fixed effects (the details are presented in the Results section). In all LMM structures, the regions were set as random effects. All explanatory variables were normalized and tested using Type II ANOVA.

The statistical analyses were performed using R (ver. 4.1.2) [24] with the following packages: car [25], emmeans [26], lme4 [27], and vegan [28].

Results

Environmental properties

Canopy openness ranged from 28.9% (DB-Pieris in August) to 52.2% (DB-Other in October) throughout the research period (Table 2, S1 Appendix). The canopy trees’ leaf buds were opening in June, and defoliation started in October; therefore, the effects of the understory on canopy openness were significant (Table 2; LMM: understory, p < 0.05). In both months, the DB-Pieris plot had the lowest canopy openness percentages, and the DB-Other plot had the highest percentages. In August, the general trends were the same as for June and October; however, significant Pieris-dominant effects were not detected. Throughout the year, the interaction terms were significant (p < 0.05); however, the Pieris-dominant effects were likelier to be detected in the deciduous broad-leaved tree plots rather than in the coniferous tree plots.

Total litter weight was higher in Conifer plots (LMM: canopy, p = 0.023) but did not differ by understory (LMM: understory, p = 0.072). Humus and Pieris leaf weights were significantly increased according to the Pieris dominance (LMM: understory, p = 0.001, p < 0.001, respectively). The humus weight was about five times greater in the DB-Pieris plot compared to the DB-Other plots, while the humus weights of the C-Pieris and C-Other plots were not significantly different (Table 2). The DB-Pieris plots had significantly lower scores for soil bulk density and higher scores for SOM content compared with those of the DB-Other plots; however, the scores of the C-Pieris and C-Other plots were not significantly different (Table 2) (LMM: the interaction term, p < 0.05). For the other eight environmental variables (i.e., soil hardness, water content, pH, EC, soil C and N, NO₃⁻, and NH₄⁺), the understory, canopy, and interaction terms were significantly or marginally significant; however, the scores for the four plot types based on Tukey’s test were not significantly different. The five variables (i.e., rock, root mass, C/N, K, and Mg) were not significantly different for any terms. For the microbial community analyses, the eight environmental variables (i.e., canopy openness over three seasons, total litter weight, humus weight, Pieris leaf, soil bulk density, and SOM content) that showed significant differences according Tukey’s tests were used.

Soil microbial communities and functionality

The amplicon sequencing of the prokaryotic 16S rRNA V3–V4 regions confirmed 34 phyla in two domains (archaea and bacteria) (Fig 1A, S2 Table). About 90% of the total read counts comprised by the top five phyla: Proteobacteria (average 39.5%; range 37.8%−40.6%), Acidobacteria (24.2%; 23.0%−24.8%), Actinobacteria (12.7%; 11.5%−13.7%), Planctomycetes (7.9%; 7.5%−8.3%), and Chloroflexi (4.3%; 3.1%−5.9%) (Fig 1A, S2 Table). The alpha diversity indices of the understory and canopy terms were significant (Table 3, LMM: p < 0.05), except for the number of ASVs (LMM: understory, p = 0.471). DB-Pieris had the highest average scores for all indices; however, the scores were not significantly different in the plots of deciduous broad-leaved trees, based on Tukey’s test. In the fungal community, 14 phyla were confirmed by the sequencing (Fig 1B). The top five phyla, Basidiomycota (average 48.7%; range 44.9%−51.5%), Ascomycota (27.9%; 22.9%−33.2%), and Mortierellomycota (8.2%; 7.0%−8.7%), constituted about 85% of the total read counts (Fig 1B, S3 Table). In general, the alpha diversity indices of the plots were not significantly different (Table 3). Two alpha diversity indices (i.e., numbers of ASVs and Shannon) were marginally significant according to the interaction term (LMM: p = 0.051) and the understory term (LMM: p = 0.056), respectively; however, the scores were not significantly different according to the understory and canopy tree types based on Tukey’s test. The other two indices were not significant for any of the fixed effect terms (LMM: p > 0.10).

Download:

Fig 1.

Proportions of the relative abundance of each microbial taxon: a) prokaryotes and b) fungi. In b), root-associated fungi in Pieris root samples are shown on the right side.

https://doi.org/10.1371/journal.pone.0296692.g001

Download:

Table 3. Prokaryotic and fungal alpha diversities for the four study site types with the results of the linear-mixed-model analyses.

https://doi.org/10.1371/journal.pone.0296692.t003

According to the NMDS plot based on the Bray–Curtis indices and the PERMANOVA test, prokaryotic communities differed by canopy trees (PERMANOVA: canopy, F_1,32 = 2.160, p = 0.006), while the understory and interaction terms were not significant (PERMANOVA: understory, F_1,32 = 1.109, p = 0.268; interaction, F_1,32 = 1.228, p = 0.169). The deciduous broad-leaved tree samples plotted on the right side and those of the coniferous trees plotted on the left side (Fig 2A). However, the Pieris-dominant site samples were plotted with other vegetation samples. In the envfit analysis, three environmental variables were significant: SOM content, litter weight, and soil bulk density (p < 0.01, Table 4) with the same or opposite arrow directions (Fig 2A). Considering the results of the LMM for the environmental variables (Table 2) and the correlation coefficients among the environmental variables (S4 Table), these three variables roughly divided the samples into two groups: DB-Other (i.e., low SOM content and litter weight and high soil bulk density) and C-Pieris, C-Other, and DB-Pieris (i.e., high SOM content and litter weight and low soil bulk density). Since the R² values of the SOM content were highest among the three environmental variables (Table 4), we constructed an LMM to assess which prokaryotic taxon changed with SOM content. SOM content positively affected the relative abundances of Actinobacteria, Verrucomicrobia, Chlamydiae, and the other nine phyla, and it negatively affected the relative abundance of Chloroflexi (S2 Table, S2 Fig).

Download:

Fig 2.

Nonmetrical dimension scaling plot of two microbiomes based on Bray–Curtis indices and significant environmental variables detected via the envfit function (see Table 3, p < 0.10) shown as arrows: a) prokaryotes and b) fungi. The green dots indicate the samples from conifer trees, and the red dots indicate samples from deciduous broad-leaved trees. The solid dots indicate samples from understories dominated by Pieris, and the open dots indicate samples from under other understory vegetation.

https://doi.org/10.1371/journal.pone.0296692.g002

Download:

Table 4. Effects of the investigated environmental variables on prokaryotic and fungal communities analyzed using the envfit function for the nonmetric multidimensional scaling (NMDS) analysis.

https://doi.org/10.1371/journal.pone.0296692.t004

In the fungal communities, (Fig 2B), the understory vegetation and canopy tree type terms were significant (PERMANOVA: understory, F_1,32 = 1.596, p < 0.001; canopy, F_1,32 = 1.367, p = 0.008). Since the interaction term was not significant (PERMANOVA: F_1,32 = 1.036, p = 0.320), fungal community changes due to Pieris dominance were the same in coniferous and deciduous broad-leaved tree stands. In the NMDS plots, Pieris-dominant plots plotted on the top left side, and the other understory vegetation plotted on the opposite side (Fig 2B). Of the analyzed environmental variables, six environmental variables (i.e., Pieris leaf, humus weight, litter weight, canopy openness in October and June, and soil bulk density, in the order of the R² values) significantly affected the fungal communities, and SOM content marginally affected (p = 0.099) (Table 4). Considering the directions of the environmental variable arrows shown in the NMDS plot and the correlation coefficients among the environmental variables (S4 Table), the effects of the environmental variables on the fungal communities were divided into two directions: Pieris-dominant effects (i.e., high Pieris leaf and humus weights and low canopy openness) and specific effects in DB-Other (i.e., low litter weight and SOM content and high soil bulk density). According to the R² values from the envfit results, the former effect was represented by Pieris leaf (R² = 0.594), and the latter was represented by litter weight (R² = 0.320). We then assessed the effects of Pieris leaf and litter weight on the relative abundance of fungal phylum and the functionalities using an LMM. Two fungal phyla (i.e., Calcarisporiellomycota and Rozellomycota) were positively affected by Pieris leaf, and Mortierellomycota was positively affected by litter weight (S3 Table). For fungal functionality, the trophic mode was generally affected by Pieris leaf (Fig 3, S5 Table). The relative abundance of symbiotroph decreased (p < 0.001, Fig 3A) and that of saprotroph increased with Pieris leaf (p < 0.001, Fig 3B). Pathotrophs were not affected by either Pieris leaf or litter weight (Pieris leaf, p = 0.13, Fig 3B; litter weight, p = 0.98). Litter weight negatively affected the relative abundance of saprotroph (p < 0.05, Fig 3D). At the guild level, the relative abundance of ECM fungi decreased with Pieris leaf (p < 0.001, S6 Table, S3 Fig). However, other symbiotic fungi, such as endophyte, ErM, and AM, were not affected by Pieris leaf (p > 0.10, S6 Table, S3 Fig).

Download:

Fig 3.

Relationships between significant environmental variables and the relative abundance of fungal trophic modes: a) weight of Pieris leaf and the relative abundance of symbiotrophs, b) saprotrophs, and c) pathotrophs, and d) total litter weight and saprotrophs. The explanatory variables were standardized. The solid line indicates the significant effect of the explanatory variable, and the dotted line indicates that no significant effects were detected (see also S6 Table).

https://doi.org/10.1371/journal.pone.0296692.g003

Pieris root-associated fungi

Root-associated fungi of P. japonica have been confirmed by microscope observations [15]; however, molecular identification and host ranges have not been determined. Nine fungal phyla were confirmed in the six Pieris root samples by the amplicon sequencing (Fig 1B). The top three phyla, Ascomycota (average 48.5%; range 20.0%−78.3%), Basidiomycota (28.8%; 6.4%−63.0%), and Mortierellomycota (11.0%; 0.6%−45.3%), comprised about 90% of the total read counts (Fig 1B, S3 Table). At the order level, 48 orders were confirmed, and known ErM taxa [12], such as Helotiales (average 22.9%), Sebacinales (average 11.0%), Agaricales (average 5.84%), and Hypocreales (average 1.56%) were confirmed in all samples (S7 Table). Chaetothyriales (average 3.42%) and Trechisporales (average 0.11%), including ErM and the dark septate endophyte, an important endophyte found in Ericaceae plant roots, were found in some samples. ECM taxa—Cantharellales (average 6.50%), Thelephorales (average 2.75%), Russulales (average 1.66%), Atheliales (average 0.29%), Auriculariales (average 0.25%), and Boletales (average 0.12%)—and an AM taxon, Glomeromycota—Glomerales (average 0.57%) and Diversisporales (average 0.002%)—were also confirmed from some samples.

The functionalities of root-associated fungi were estimated. In the trophic mode, saprotroph (average 14.1%), symbiotroph (average 13.0%), and saprotroph–symbiotroph (average 12.5%) were frequently confirmed (S5 Table). Pathotroph was only confirmed by 2.21%, on average. At the guild level, undefined saprotrophs (average 12.0%) and multifunctional saprotrophs, such as endophyte-litter saprotroph-soil saprotroph-undefined saprotroph (average 11.0%), were confirmed as frequent guilds (S6 Table).

Discussion

Differences between understory environments according to Pieris dominance for different canopy trees

The understory environments differed according to the types of understory and canopy trees. Under deciduous broad-leaved trees, Pieris dominance led to high Pieris leaf litter and humus weights and low soil bulk density and canopy openness. These changes have also been observed in Ericaceae monodominant areas [e.g., 2, 29–31]. DB-Pieris had the lowest canopy openness percentages compared with other plots in the same season, whereas those of DB-Other seasonally fluctuated probably responding to the canopy trees’ leaf phenology, that is, leaf buds opening (June), leaves matured (August), and defoliation started (October). Pieris-dominances in understory might make the light environments less variable throughout the year. Due to the formation of the understory canopy, the light quantity and quality are often affected. Under the canopy of Rhododendron maximum (Ericaceae), photosynthetically active radiation decreases compared with that in nonrhododendron areas [32]. Messier et al. [29] reported the decreased red:far-red ratio in the understory canopy of Gaultheria shallon (Ericaceae), which can affect the seed germination rate. Along with the formation of the understory canopy, the light environment on the forest floor would become less variable [32, 33], and thus, Pieris dominance in the forest understory may change both the quantity and quality of the light environment. Thick layers of humus formed by Pieris dominance also impact the forest ecosystems and often physically prevent the establishment of seedlings, which inhibits forest regeneration [30, 34]. Ericaceae humus and roots contain many secondary compounds, such as phenolic acid and tannin, and extracellular enzymes, and the decomposition rates of the humus are slow [7, 35, 36]. These changes in chemical contents and the decomposition rate affect seed germination and growth and N cycles derived from the secondary compounds of Ericaceae plants [6–8, 37]. Moreover, chemical inhibition is often dependent on combinations of Ericaceae species and non-Ericaceae plants [38, 39]. A few individual seedlings and saplings of other-trees were observed in the Pieris-dominant forest [3, Tokumoto, Y. unpublished data]. Although an allelopathic test of the Pieris leaf on lettuce seeds showed no significant effect on seed germination [40], the effects of Pieris root, humus, and leaf litter have not been tested. Therefore, some chemical inhibition of Pieris species may exist; however, the inhibition process is not well understood. Regarding the nutrient cycles, our soil chemical analyses on soil N and C and other nutrient ions showed that nonsignificant changes were inflicted by Pieris dominance (Table 2), and the impact of Pieris dominance on nutrient cycles may be less than that previously reported for other Ericaceae plants [37]. Because the other variables, such as soil nitrogen complexes and enzyme activities, have not been measured in this study, further studies on the physical and chemical factors that inhibit plant regeneration under Pieris populations and the effects on the nutrient cycles in forest ecosystems are required to better understand the effects of Pieris dominance.

In this study, most of the effects of Pieris dominance were observed in deciduous broad-leaved tree stands rather than in evergreen coniferous stands, except for the amount of Pieris leaf. In particular, DB-Other had the highest soil bulk density scores and the lowest litter weights, which is similar to the results for soil-eroded regions in the same study regions (soil bulk density: 0.29−0.40 g/cm³) [41]. After understory layers have been grazed by deer, litter and soils are frequently removed, and high soil bulk density and low chemical contents, such as SOM, C, and N, are often observed in severely eroded regions [41–43]. At our research sites, surface soils were almost exposed under deciduous broad-leaved trees; therefore, the impact of direct rainfall on the soil surfaces of deciduous broad-leaved tree stands will be greater than that in evergreen coniferous stands. This implies that Pieris cover has prevented soil erosion in DB-Pieris; and thus, Pieris cover in the understory might play a role in preventing litter and soil runoff in deciduous broad-leaved tree stands and maintaining the soil’s physical properties and chemical contents after deer grazing. However, no significant effect of Pieris on soil properties was observed in the coniferous stands, which suggests that the effect of Pieris cover differs among forest types. Our study region has experienced severe deer overgrazing, and the original understory vegetation, especially dwarf bamboo, has disappeared from many areas. After the disappearance of understory vegetation, the invasion of thickets of Pieris may remediate forest soil environments.

Differences between soil microbial communities according to Pieris dominance for different canopy trees

The types of understory and the canopy trees affected the microbial community compositions, diversity indices, and functionalities. Pieris affected the fungal community compositions but did not affect the prokaryotic community compositions. The effects of ErM plants and fungi on vegetation cover, SOM, and C and N cycles have been assessed [7]; however, research on soil microbial communities is limited. In this study, the effects of Pieris dominance determined by the environmental variables on the microbial communities and Pieris root-associated fungi are discussed below.

For the soil fungal community, the community changes were associated with the environmental changes in two directions: Pieris-dominant effects (i.e., high Pieris leaf and humus weights and low canopy openness) and specific effects in the DB-Other plots (i.e., low litter weight and SOM content and high soil bulk density) (Fig 2B, Table 3). For the Pieris-dominant effects, the relative abundance of symbiotroph, especially ECM, decreased, and those of saprotroph increased with the increase of Pieris leaf, indicating that Pieris dominance led to symbiotroph-poor and saprotroph-enriched fungal communities in the soils (Fig 3A, 3B, S5 and S6 Tables). In Pieris root, we confirmed the association of not only ErM, AM, and endophyte but also ECM (Fig 3A, S3 Fig., S6 and S7 Tables), although previous microscopic observations of Pieris roots have only confirmed associations with ErM and AM [15]. Multiple types of mycorrhizae fungi have been found in the roots of other ericaceous plants using molecular identification (e.g., ECM, ErM, and AM in Pyrola japonica [44]; ErM and AM in Vaccinum oldhamii [45]). Our molecular identification of root-associated fungi suggests that P. japonica may potentially have broad mutualistic relationships with three mycorrhizae. However, in the soil fungal communities, the proportions of symbiotroph, particularly ECM fungi were negatively affected by Pieris dominance, while other fungi, such as ErM, AM, and endophyte, were not significantly affected (S5 Table, S3 Fig). Some studies showed that the ErM fungi commonly live in non-Ericaceae plant roots and these plants may maintain the ErM fungi populations [46]; therefore, Pieris dominance might not affect the relative abundance of ErM in soils. In contrast, the result of the decrease of proportions of ECM fungi was inconsistent with the previous study [47] that showed the proportions of ECM fungi increased to the amount of leaf litter inputs. The negative proportional changes of ECM fungi associated with the Pieris leaf could be discussed with the interactions between the Ericaceae plants or ErM. Previous studies have reported that the infection rate of ECM in plants, which have the symbiotic relationships with ECM, was decreased by Ericaceae plants in pot experiments, and that Ericaceae and ErM might suppress ECM fungal infection of plant roots [9, 10]. Although some studies have reported that ErM plant cover is positively related to the abundance of ECM plants at plot levels [48], the other study also showed that the directions and intensities of ErM effects have differed among combinations of the ErM and ECM plant species [10]. In our study regions, Pieris shrubs or ErM fungi infected in Pieris may suppress the ECM in soils, however, the underlying mechanisms on how the proportions of ECM decreased were unknown. Further studies are needed to understand these mechanisms and the interactions between ECM and Pieris shrubs or ErM.

Pieris-dominant effects were also confirmed by changes in the proportions of saprotrophic fungi; their increased proportions were associated with Pieris dominance (Fig 3B, S4 Fig). Pieris dominance in the understory leads to more undecomposed organic matter, such as humus layers and Pieris leaf, and less canopy openness (Table 1). This positive proportional change of saprotrophic fungi might relate to such environmental changes and the result was consistent with that of the previous study, which tested the soil microbial community changes according to the leaf litter input [47]. According to the other previous study, ErM fungi also show high saprotrophic capability and have been predicted to compete with free-living saprotrophs [7]. In our results, the proportional changes in ErM were not confirmed, and the number of other saprotrophic fungi may have increased in response to increased organic matter in the Pieris-dominant areas without the distinctive competition with ErM. In contrast, litter weight negatively affected the proportions of saprotroph (Fig 3D), which indicated that DB-Other, where the litter weight was the lowest among the sites, had relatively higher proportions of saprotrophs. Compared with the evergreen coniferous forest, the topsoil of the deciduous broad-leaved tree forests has high proportions of saprotrophs [49]. In DB-Other, deciduous broad-leaved trees, such as Quercus and Carpinus spp. are distributed, and the higher proportions of saprotroph in the plots may be due to the forest type. This suggests that the formation of Pieris cover may have changed the original diverse fungal communities, including saprotrophs, in different forest types into the similar soil fungal communities. Changes in soil fungal compositions sometimes affect the percentages of ECM fungal root colonization and decrease other plants’ growth [9, 10]. Because few seedlings were observed under the Pieris canopy, suppression effects of the changed fungal community due to Pieris cover could be expected. Together with the effects of environmental changes caused by Pieris dominance, the effects of soil microbial community changes induced by Pieris dominance on the plant performances should also be assessed. Moreover, previous studies have reported that ErM consistently inhabits after Ericaceae plants are cut back [6, 50]. This indicates that Pieris-dominant effects can remain even after Pieris shrubs are cut back. Further studies on long-term plant performance, including germination and growth, should be conducted to understand plant–microbe interactions under Pieris understory canopies.

Similar to the fungal changes associated with litter weight, the prokaryotic community was also affected by SOM content, which was significantly positively correlated with litter weight and negatively correlated with soil bulk density (Fig 2A, Table 3). Proportions of the relative abundances of Actinobacteria and Verrucomicrobia changed positively with SOM and vice versa for Chloroflexi (S2 Fig). DB-Other had the lowest SOM content, and thus, the prokaryotic flora in DB-Other would have become less organic matter-dependent communities compared to the other three (i.e., DB-Pieris, C-Pieris, and C-Other). The above-mentioned phyla often respond to organic matter and soil depth [51, 52], and Chloroflexi is an oligotrophic bacterium that inhabits poor-nutrient soils [53]. In the previous study around our study regions, the severe soil erosion after deer overgrazing caused the soil ecosystems to change into organic matter-poor environments and oligotrophic bacterium-enriched prokaryotic communities [54]. As mentioned, the DB-Other soil environments closely resembled those of soil-eroded areas in terms of the litter weight, bulk density, and SOM content. In the absence of Pieris shrubs, soil prokaryotic flora might resemble those of deeper and poor-nutrient soils, especially in deciduous broad-leaved tree stands. A previous study has compared the extracellular enzyme activities between sites with or without Ericaceae removal, and the presence of Ericaceae decrease C-acquisition enzymes, such as β-glucosidase and β-xylosidase and soil C and N availabilities [55]. Although the soil chemical parameters and the soil prokaryote community in our study were not directly affected by the Pieris-dominances (Tables 2 and 4), a further study on the microbial activities, including enzyme activities is important to investigate in details the effects of the Pieris-dominances on the ecosystem.

Conclusions

This study focused on the understory environments and soil microbiomes of Pieris-dominant sites with two different types of upper tree stands. Due to the Pieris dominance in the understory, canopy openness decreased by the formed understory canopy and humus layers were thickened, especially in the deciduous broad-leaved tree stands. The soil chemical parameters, which were often changed by other Ericaceae plant dominance in the understory layer, were not different between Pieris-dominant and nondominant sites. However, soil fungal communities were affected by the Pieris dominance, and the proportions of symbiotrophs, especially ECM fungi, decreased, and those of pathotrophs increased, although prokaryote communities were not affected by the Pieris dominance. These changes in the environmental variables and soil microbial communities might affect forest regeneration and, indeed, few seedlings were observed under the Pieris populations. The results should provide fundamental knowledge of the effects of Pieris monodominant in the forest understories and will encourage future studies on the forest ecosystems and mountain protections in changing environments after deer overgrazing.

Supporting information

S1 Fig. Location of the research regions (Maruju, M; Hirono, H).

Pieris dominated sites were shown with odd numbers (M01, M03, M05, H01, H03, H05, and H07). Sites with other vegetation (sites with even numbers) were located near the sites with odd numbers.

https://doi.org/10.1371/journal.pone.0296692.s001

(TIFF)

S2 Fig.

Relationships between SOM content and the relative abundance of prokaryotic phyla: a) Actinobacteria, b) Verrucomicrobia, c) Chlamydiae, and d) Chloroflexi. The explanatory variables were standardized, and the results are shown as a solid line (also see S2 Table).

https://doi.org/10.1371/journal.pone.0296692.s002

(TIFF)

S3 Fig.

Relationships between the weight of Pieris leaves and the relative abundance of fungal symbiotrophic guilds: a) ectomycorrhizal fungi, b) endophyte, c) ericoid mycorrhizal fungi, and d) arbuscular mycorrhizal fungi. The explanatory variables were standardized. The solid line indicates the significant effect of the explanatory variable, and the dotted line indicates no significant effects (also see S6 Table).

https://doi.org/10.1371/journal.pone.0296692.s003

(TIFF)

S4 Fig. Relationships between the weight of Pieris leaves and the relative abundance of undefined saprotroph.

The explanatory variables were standardized, and the results are shown as a solid line (also see S6 Table).

https://doi.org/10.1371/journal.pone.0296692.s004

(TIFF)

S1 Table. Pieris sapling information for assessing root-associated fungi.

https://doi.org/10.1371/journal.pone.0296692.s005

(XLSX)

S2 Table. Results of the linear-mixed-model analyses of the relative abundance of prokaryotic phyla with SOM content.

https://doi.org/10.1371/journal.pone.0296692.s006

(XLSX)

S3 Table. Results of the linear-mixed-model analyses of the relative abundance of fungal phyla with two environmental variables and the read count (%) of root-associated fungi in the Pieris root samples.

https://doi.org/10.1371/journal.pone.0296692.s007

(XLSX)

S4 Table. Results of Spearman’s correlation coefficient test for the environmental variables.

The upper and lower parts of the triangular matrix are the correlation coefficient and p-value, respectively.

https://doi.org/10.1371/journal.pone.0296692.s008

(XLSX)

S5 Table. Results of the linear-mixed-model analyses of the relative abundance of fungal trophic mode with two environmental variables and the read count (%) of root-associated fungi in the Pieris root samples.

https://doi.org/10.1371/journal.pone.0296692.s009

(XLSX)

S6 Table. Results of the linear-mixed-model analyses of the relative abundance of fungal guild with two environmental variables and the read count (%) of root-associated fungi in the Pieris root samples.

https://doi.org/10.1371/journal.pone.0296692.s010

(XLSX)

S7 Table. Relative read count of the order-level Pieris root-associated fungi of six Pieris root samples.

https://doi.org/10.1371/journal.pone.0296692.s011

(XLSX)

S1 Appendix. Environmental data measured in this study.

https://doi.org/10.1371/journal.pone.0296692.s012

(XLSX)

Acknowledgments

We are grateful to the Shiiba Research Forest of the Department of Agro–environmental Sciences, Kyushu University, for their kind support during the research.

References

1. Côté S, Rooney TP, Tremblay JP, Dussault C, Waller DM. Ecological impacts of deer overabundance. Annual Review of Ecology, Evolution, and Systematics. 2004; 35: 113–147.
- View Article
- Google Scholar
2. Royo AA, Carson WP. On the formation of dense understory layers in forests worldwide: consequences and implications for forest dynamics, biodiversity, and succession. Canadian Journal of Forest Research. 2006; 36: 1345–1362.
- View Article
- Google Scholar
3. Tsujino R, Matsui K. Forest regeneration inhibition in a mixed broadleaf-conifer forest under sika deer pressure. Journal of Forest Research. 2022; 27(3): 230–235.
- View Article
- Google Scholar
4. Hashimoto Y, Fujiki D. List of food plants and unpalatable plants of sika deer (Cervus nippon) in Japan. Humans and Nature. 2014; 25: 133–160. (in Japanese).
- View Article
- Google Scholar
5. Ishida H, Hattori T, Kodate S, Kuroda A, Sawada Y, Matsumura T, et al. The Hypolepis punctata community established under severe feeding pressure by sika deer and its application to revegetation. Japanese Journal of Conservation Ecology. 2008; 13: 137–150 (in Japanese).
- View Article
- Google Scholar
6. Mallik AU. Conifer regeneration problems in boreal and temperate forests with ericaceous understory: role of disturbance, seedbed limitation, and keytsone species change. Critical Reviews in Plant Sciences. 2003; 22(3–4): 341–366.
- View Article
- Google Scholar
7. Ward EB., Duguid MC, Kuebbing SE, Lendemer JC, Bradford MA. The functional role of ericoid mycorrhizal plants and fungi on carbon and nitrogen dynamics in forests. New Phytologist. 2022; 235: 1701–1718.
- View Article
- Google Scholar
8. Vandermast DB, Van Lear DH. Riparian vegetation in the southern Appalachian Mountains (USA) following chestnut blight. Forest Ecology and Management. 2002; 155: 97–106.
- View Article
- Google Scholar
9. Walker J, Miller O Jr, Lei T, Semones S, Nilsen E, Clinton BD. Suppression of ectomycorrhizae on canopy tree seedlings in Rhododendron maximum L. (Ericaceae) thickets in the southern Appalachians. Mycorrhiza. 1999; 9: 49–56.
- View Article
- Google Scholar
10. Kohout P, Sýkorová Z, Bahram M, Hadincová V, Albrechtová J, Tedersoo L, et al. Ericaceous dwarf shrubs affect ectomycorrhizal fungal community of the invasive Pinus strobus and native Pinus sylvestris in a pot experiment. Mycorrhiza. 2011; 21: 403–412.
- View Article
- Google Scholar
11. Urbanová M, Šnajdr J, Baldrian P. Composition of fungal and bacterial communities in forest litter and soil is largely determined by dominant trees. Soil Biology & Biochemistry. 2015; 84: 53–64.
- View Article
- Google Scholar
12. Leopold DR. Ericoid fungal diversity: Challenges and opportunities for mycorrhizal research. Fungal Ecology. 2016; 24: 114–123.
- View Article
- Google Scholar
13. Vohnik M. Ericoid mycorrhizal symbiosis: theoretical background and methods for its comprehensive investigation. Mycorrhiza. 2020; 30: 671–695.
- View Article
- Google Scholar
14. Kurokochi H, Chunlan L. The populations and distribution of Pieris japonica, a poisonous tree protected from herbivore browsing pressure, increase slowly but steadily. Journal of Forest Research. 2016; 21: 125–130.
- View Article
- Google Scholar
15. Fujiyoshi M, Nakatsubo T, Seo K, Horikoshi T. The distribution of mycorrhizal types in a Pinus densiflora forest. Natural Environmental Science Research. 1999; 12: 11–16.
- View Article
- Google Scholar
16. Kisanuki H, Nakasu M, Nakai A, Yurugi Y. Predicting the population dynamics of three understory broad-leaved evergreen species under the influence of Sika deer in primary and secondary forests of mid-western Japan. Journal of Forest Research. 2008; 13: 52–58.
- View Article
- Google Scholar
17. Enoki T, Hishi T, Tashiro N. Changes in the effects of neighboring trees on tree growth and mortality in a temperate mixed forest for 30 years. Bulletin of Kyushu University Forest. 2018; 99: 1–7.
- View Article
- Google Scholar
18. Takahashi K, Ikeyama Y, Okuhara I. Stand dynamics and competition in a mixed forest at the northern distribution limit of evergreen hardwood species. Ecology and Evolution. 2018; 8: 11199–11212.
- View Article
- Google Scholar
19. Kawakami E, Katayama A, Hishi T. Effect of the understory shrub Pieris japonica on litter decomposition. Bulletin of Kyushu University Forest. 2020; 101: 1–6 (in Japanese).
- View Article
- Google Scholar
20. Nakamura T, Cho K, Yamauchi K, Ogata T, Kaji K, Sasaki H et al. The performance of different fencing on growth, survival rate of plantation tree and forest floor vegetation. Bulletin of Kyushu University Forest. 2022; 103: 29–39 (in Japanese).
- View Article
- Google Scholar
21. Murata I, Inoue S, Yabe T, Kabemura Y, Kaji K, Kubota K, et al. Sika deer density and vegetation changes for 37 years in Shiiba Research Forest. Bulletin of Kyushu University Forest. 2009; 90: 13–24 (in Japanese).
- View Article
- Google Scholar
22. Toju H, Tanabe AS, Yamamoto S, Sato H. High-coverage ITS primers for the DNA-based identification of ascomycetes and basidiomycetes in environmental samples. PLoS ONE. 2012; 7(7): e40863.
- View Article
- Google Scholar
23. Nguyen NH, Song Z, Bates ST, Branco S, Tedersoo L, Menke J, et al. FUNGuild: an open annotation tool for parsing fungal community datasets by ecological guild. Fungal Ecology. 2016; 20: 241–248.
- View Article
- Google Scholar
24. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. (accessed 23 June 2022) 2022.
25. Fox J, Weisberg S. An R Companion to Applied Regression, Third edition. Sage, Thousand Oaks CA. 2019.
26. Lenth R. _emmeans: Estimated Marginal Means, aka Least-Squares Means_. R package version 1.8.5, <https://CRAN.R-project.org/package=emmeans>. 2023.
- View Article
- Google Scholar
27. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software. 2015; 67(1): 1–48.
- View Article
- Google Scholar
28. Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, et al. _vegan: Community Ecology Package_. R package version 2.6–4, <https://CRAN.R-project.org/package=vegan>. 2022.
- View Article
- Google Scholar
29. Messier C, Honer TW, Kimmins JP. Photosynthetic photon flux density, red:far-red ratio, and minimum light requirement for survival of Gaultheria shallon in western red cedar-western hemlock stands in coastal British Columbia. Canadian Journal of Forest Research. 1989; 19(11): 1470–1477.
- View Article
- Google Scholar
30. Maubon M, Ponge JF, Andre J. Dynamics of Vaccinium myrtillus patches in mountain spruce forest. Journal of Vegetation Science. 1995; 6: 343–348.
- View Article
- Google Scholar
31. Frak E, Ponge JF. The influence of altitude on the distribution of subterranean organs and humus components in Vaccinium myrtillus carpets. Journal of Vegetation Science. 2002; 13: 17–26.
- View Article
- Google Scholar
32. Clinton BD. Light, temperature, and soil moisture responses to elevation, evergreen understory, and small canopy gaps in the southern Appalachians. Forest Ecology and Management. 2003; 186(1–3): 243–255.
- View Article
- Google Scholar
33. Nilsen ET, Clinton BD, Lei TT, Miller OK, Semones SW, Walker JF. Does Rhododendron maximum L. (Ericaceae) reduce the availability of resources above and belowground for canopy tree seedlings? The American Midland Naturalist. 2001; 145(2): 325–343.
- View Article
- Google Scholar
34. Mallik AU. Conversion of temperate forests into heaths: role of ecosystem disturbance and ericaceous plants. Environmental Management. 1995; 19(5): 675–684.
- View Article
- Google Scholar
35. Read DJ. Mycorrhizas in ecosystem. Experientia. 1991; 47: 376–391.
- View Article
- Google Scholar
36. Wurzburger N, Hendrick RL. Plant litter chemistry and mycorrhizal roots promote a nitrogen feedback in a temperate forest. Journal of Ecology. 2009; 97: 528–536.
- View Article
- Google Scholar
37. Wurzburger N, Hendrick RL. Rhododendron thickets alter N cycling and soil extracellular enzyme activities in southern Appalachian hardwood forests. Pedobiologia. 2007; 50(6): 563–576.
- View Article
- Google Scholar
38. Adamczyk B, Ahvenainen A, Sietiö OM, Kanerva S, Kieloaho AJ, Smolander A, et al. The contribution of ericoid plants to soil nitrogen chemistry and organic matter decomposition in boreal forest soil. Soil Biology & Biochemistry. 2016; 103: 394–404.
- View Article
- Google Scholar
39. Jones GL, Scullion J, Worgan H, Gwynn-Joneset D. Litter of the invasive shrub Rhododendron ponticum (Ericaceae) modifies the decomposition rate of native UK woodland litter. Ecological Indicators. 2019; 107: 105597.
- View Article
- Google Scholar
40. Appiah KS, Li Z, Zeng RS, Luo S, Oikawa Y, Fujii Y. Determination of allelopathic potentials in plant species in Sino-Japanese floristic region by sandwich method and dish pack method. International Journal of Basic and Applied Sciences. 2015; 4(4): 381–394.
- View Article
- Google Scholar
41. Katayama A, Oyamada M, Abe H, Uemori K, Hishi T. Soil erosion decreases soil microbial respiration in Japanese beech forests with understory vegetation lost by deer. Journal of Forest Research. 2023; 28: 428–435.
- View Article
- Google Scholar
42. Furusawa H, Miyanishi H, Kaneko S, Hino T. Movement of soil and litter on the floor of a temperate mixed forest with an impoverished understory grazed by deer (Cervus nippon centralis Temminck). Journal of the Japanese Forestry Society. 2003; 85: 318–325.
- View Article
- Google Scholar
43. Obalum S, Chibuike G, Peth S, Ouyang Y. Soil organic matter as sole indicator of soil degradation. Environmental Monitoring and Assessment. 2017; 189: 176.
- View Article
- Google Scholar
44. Matsuda Y, Shimizu S, Mori M, Ito SI, Selosse MA. Seasonal and environmental changes of mycorrhizal associations and heterotrophy levels in mixotrophic Pyrola japonica (Ericaceae) growing under different light environments. American Journal of Botany. 2012; 99: 1177–1188.
- View Article
- Google Scholar
45. Baba T, Hirose D, Sasaki N, Watanabe N, Kobayashi N, Kurashige Y, et al. Mycorrhizal formation and diversity of endophytic fungi in hair roots of Vaccinium oldhamii Miq. in Japan. Microbes and Environment. 2016; 31(2): 186–189.
- View Article
- Google Scholar
46. Chambers SM, Curlevski NJA, Cairney JWG. Ericoid mycorrhizal fungi are common root inhabitants of non-Ericaceae plants in a south-eastern Australian sclerophyll forest. FEMS Microbiology Ecology. 2008; 65(2): 263–270.
- View Article
- Google Scholar
47. Marañón-Jiménez S, Radujković D, Verbruggen E, Grau O, Cuntz M, Peñuelas J, et al. Shifts in the Abundances of saprotrophic and ectomycorrhizal fungi with altered leaf litter inputs. Frontiers in Plant Science. 2021; 12:682142.
- View Article
- Google Scholar
48. Ward EB, Duguid MC, Kuebbing SE, Lendemer JC, Warren II, RJ, Bradford MA. Ericoid mycorrhizal shrubs alter the relationship between tree mycorrhizal dominance and soil carbon and nitrogen. Journal of Ecology. 2021; 109: 3524–3540.
- View Article
- Google Scholar
49. Prada-Salcedo LD, Prada-Salcedo JP, Heintz-Buschart A, Buscot F, Goldmann K. Effects of tree composition and soil depth on structure and functionality of belowground microbial communities in temperate European forests. Frontiers in Microbiology. 2022; 13:920618.
- View Article
- Google Scholar
50. Bergero R, Girlanda M, Bello F, Luppi A, Perotto S. Soil persistence and biodiversity of ericoid mycorrhizal fungi in the absence of the host plant in a Mediterranean ecosystem. Mycorrhiza. 2003; 13(2): 69–75.
- View Article
- Google Scholar
51. Zhan W, Zhou Y, Jeppesen E, Wang L, Tan H, Zhang J. Linking heterotrophic bacterioplankton community composition to the optical dynamics of dissolved organic matter in a large eutrophic Chinese lake. Science of the Total Environment. 2019; 679: 136–147.
- View Article
- Google Scholar
52. Chen J, Du Y, Zhu W, Pang X, Wang Z. Effects of organic materials on soil bacterial community structure in long-term continuous cropping of tomato in greenhouse. Open Life Sciences. 2022; 17: 381–392.
- View Article
- Google Scholar
53. Lan J, Wang S, Wang J, Qi X, Long Q, Huang M. The shift of soil bacterial community after afforestation influence soil organic carbon and aggregate stability in Karst Region. Frontiers in Microbiology. 2022; 13: 901126.
- View Article
- Google Scholar
54. Chen FC, Katayama A, Oyamada M, Tsuyama T, Kijidani Y, Tokumoto Y. Effects of soil environmental changes accompanying soil erosion on the soil prokaryotes and fungi of cool temperate forests in Southern Japan. Journal of Forest Research, 2023;
- View Article
- Google Scholar
55. Osburn ED, Elliottt KJ, Knoepp JD, Miniat CF, Barrett JE. Soil microbial response to Rhododendron understory removal in southern Appalachian forests: Effects on extracellular enzymes. Soil Biology and Biochemistry. 2018; 127: 50–59.
- View Article
- Google Scholar

[ref1] 1. Côté S, Rooney TP, Tremblay JP, Dussault C, Waller DM. Ecological impacts of deer overabundance. Annual Review of Ecology, Evolution, and Systematics. 2004; 35: 113–147.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Royo AA, Carson WP. On the formation of dense understory layers in forests worldwide: consequences and implications for forest dynamics, biodiversity, and succession. Canadian Journal of Forest Research. 2006; 36: 1345–1362.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref3] 3. Tsujino R, Matsui K. Forest regeneration inhibition in a mixed broadleaf-conifer forest under sika deer pressure. Journal of Forest Research. 2022; 27(3): 230–235.
View Article
Google Scholar

[8] View Article

[9] Google Scholar

[ref4] 4. Hashimoto Y, Fujiki D. List of food plants and unpalatable plants of sika deer (Cervus nippon) in Japan. Humans and Nature. 2014; 25: 133–160. (in Japanese).
View Article
Google Scholar

[11] View Article

[12] Google Scholar

[ref5] 5. Ishida H, Hattori T, Kodate S, Kuroda A, Sawada Y, Matsumura T, et al. The Hypolepis punctata community established under severe feeding pressure by sika deer and its application to revegetation. Japanese Journal of Conservation Ecology. 2008; 13: 137–150 (in Japanese).
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref6] 6. Mallik AU. Conifer regeneration problems in boreal and temperate forests with ericaceous understory: role of disturbance, seedbed limitation, and keytsone species change. Critical Reviews in Plant Sciences. 2003; 22(3–4): 341–366.
View Article
Google Scholar

[17] View Article

[18] Google Scholar

[ref7] 7. Ward EB., Duguid MC, Kuebbing SE, Lendemer JC, Bradford MA. The functional role of ericoid mycorrhizal plants and fungi on carbon and nitrogen dynamics in forests. New Phytologist. 2022; 235: 1701–1718.
View Article
Google Scholar

[20] View Article

[21] Google Scholar

[ref8] 8. Vandermast DB, Van Lear DH. Riparian vegetation in the southern Appalachian Mountains (USA) following chestnut blight. Forest Ecology and Management. 2002; 155: 97–106.
View Article
Google Scholar

[23] View Article

[24] Google Scholar

[ref9] 9. Walker J, Miller O Jr, Lei T, Semones S, Nilsen E, Clinton BD. Suppression of ectomycorrhizae on canopy tree seedlings in Rhododendron maximum L. (Ericaceae) thickets in the southern Appalachians. Mycorrhiza. 1999; 9: 49–56.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref10] 10. Kohout P, Sýkorová Z, Bahram M, Hadincová V, Albrechtová J, Tedersoo L, et al. Ericaceous dwarf shrubs affect ectomycorrhizal fungal community of the invasive Pinus strobus and native Pinus sylvestris in a pot experiment. Mycorrhiza. 2011; 21: 403–412.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref11] 11. Urbanová M, Šnajdr J, Baldrian P. Composition of fungal and bacterial communities in forest litter and soil is largely determined by dominant trees. Soil Biology & Biochemistry. 2015; 84: 53–64.
View Article
Google Scholar

[32] View Article

[33] Google Scholar

[ref12] 12. Leopold DR. Ericoid fungal diversity: Challenges and opportunities for mycorrhizal research. Fungal Ecology. 2016; 24: 114–123.
View Article
Google Scholar

[35] View Article

[36] Google Scholar

[ref13] 13. Vohnik M. Ericoid mycorrhizal symbiosis: theoretical background and methods for its comprehensive investigation. Mycorrhiza. 2020; 30: 671–695.
View Article
Google Scholar

[38] View Article

[39] Google Scholar

[ref14] 14. Kurokochi H, Chunlan L. The populations and distribution of Pieris japonica, a poisonous tree protected from herbivore browsing pressure, increase slowly but steadily. Journal of Forest Research. 2016; 21: 125–130.
View Article
Google Scholar

[41] View Article

[42] Google Scholar

[ref15] 15. Fujiyoshi M, Nakatsubo T, Seo K, Horikoshi T. The distribution of mycorrhizal types in a Pinus densiflora forest. Natural Environmental Science Research. 1999; 12: 11–16.
View Article
Google Scholar

[44] View Article

[45] Google Scholar

[ref16] 16. Kisanuki H, Nakasu M, Nakai A, Yurugi Y. Predicting the population dynamics of three understory broad-leaved evergreen species under the influence of Sika deer in primary and secondary forests of mid-western Japan. Journal of Forest Research. 2008; 13: 52–58.
View Article
Google Scholar

[47] View Article

[48] Google Scholar

[ref17] 17. Enoki T, Hishi T, Tashiro N. Changes in the effects of neighboring trees on tree growth and mortality in a temperate mixed forest for 30 years. Bulletin of Kyushu University Forest. 2018; 99: 1–7.
View Article
Google Scholar

[50] View Article

[51] Google Scholar

[ref18] 18. Takahashi K, Ikeyama Y, Okuhara I. Stand dynamics and competition in a mixed forest at the northern distribution limit of evergreen hardwood species. Ecology and Evolution. 2018; 8: 11199–11212.
View Article
Google Scholar

[53] View Article

[54] Google Scholar

[ref19] 19. Kawakami E, Katayama A, Hishi T. Effect of the understory shrub Pieris japonica on litter decomposition. Bulletin of Kyushu University Forest. 2020; 101: 1–6 (in Japanese).
View Article
Google Scholar

[56] View Article

[57] Google Scholar

[ref20] 20. Nakamura T, Cho K, Yamauchi K, Ogata T, Kaji K, Sasaki H et al. The performance of different fencing on growth, survival rate of plantation tree and forest floor vegetation. Bulletin of Kyushu University Forest. 2022; 103: 29–39 (in Japanese).
View Article
Google Scholar

[59] View Article

[60] Google Scholar

[ref21] 21. Murata I, Inoue S, Yabe T, Kabemura Y, Kaji K, Kubota K, et al. Sika deer density and vegetation changes for 37 years in Shiiba Research Forest. Bulletin of Kyushu University Forest. 2009; 90: 13–24 (in Japanese).
View Article
Google Scholar

[62] View Article

[63] Google Scholar

[ref22] 22. Toju H, Tanabe AS, Yamamoto S, Sato H. High-coverage ITS primers for the DNA-based identification of ascomycetes and basidiomycetes in environmental samples. PLoS ONE. 2012; 7(7): e40863.
View Article
Google Scholar

[65] View Article

[66] Google Scholar

[ref23] 23. Nguyen NH, Song Z, Bates ST, Branco S, Tedersoo L, Menke J, et al. FUNGuild: an open annotation tool for parsing fungal community datasets by ecological guild. Fungal Ecology. 2016; 20: 241–248.
View Article
Google Scholar

[68] View Article

[69] Google Scholar

[ref24] 24. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. (accessed 23 June 2022) 2022.

[ref25] 25. Fox J, Weisberg S. An R Companion to Applied Regression, Third edition. Sage, Thousand Oaks CA. 2019.

[ref26] 26. Lenth R. _emmeans: Estimated Marginal Means, aka Least-Squares Means_. R package version 1.8.5, <https://CRAN.R-project.org/package=emmeans>. 2023.
View Article
Google Scholar

[73] View Article

[74] Google Scholar

[ref27] 27. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software. 2015; 67(1): 1–48.
View Article
Google Scholar

[76] View Article

[77] Google Scholar

[ref28] 28. Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, et al. _vegan: Community Ecology Package_. R package version 2.6–4, <https://CRAN.R-project.org/package=vegan>. 2022.
View Article
Google Scholar

[79] View Article

[80] Google Scholar

[ref29] 29. Messier C, Honer TW, Kimmins JP. Photosynthetic photon flux density, red:far-red ratio, and minimum light requirement for survival of Gaultheria shallon in western red cedar-western hemlock stands in coastal British Columbia. Canadian Journal of Forest Research. 1989; 19(11): 1470–1477.
View Article
Google Scholar

[82] View Article

[83] Google Scholar

[ref30] 30. Maubon M, Ponge JF, Andre J. Dynamics of Vaccinium myrtillus patches in mountain spruce forest. Journal of Vegetation Science. 1995; 6: 343–348.
View Article
Google Scholar

[85] View Article

[86] Google Scholar

[ref31] 31. Frak E, Ponge JF. The influence of altitude on the distribution of subterranean organs and humus components in Vaccinium myrtillus carpets. Journal of Vegetation Science. 2002; 13: 17–26.
View Article
Google Scholar

[88] View Article

[89] Google Scholar

[ref32] 32. Clinton BD. Light, temperature, and soil moisture responses to elevation, evergreen understory, and small canopy gaps in the southern Appalachians. Forest Ecology and Management. 2003; 186(1–3): 243–255.
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref33] 33. Nilsen ET, Clinton BD, Lei TT, Miller OK, Semones SW, Walker JF. Does Rhododendron maximum L. (Ericaceae) reduce the availability of resources above and belowground for canopy tree seedlings? The American Midland Naturalist. 2001; 145(2): 325–343.
View Article
Google Scholar

[94] View Article

[95] Google Scholar

[ref34] 34. Mallik AU. Conversion of temperate forests into heaths: role of ecosystem disturbance and ericaceous plants. Environmental Management. 1995; 19(5): 675–684.
View Article
Google Scholar

[97] View Article

[98] Google Scholar

[ref35] 35. Read DJ. Mycorrhizas in ecosystem. Experientia. 1991; 47: 376–391.
View Article
Google Scholar

[100] View Article

[101] Google Scholar

[ref36] 36. Wurzburger N, Hendrick RL. Plant litter chemistry and mycorrhizal roots promote a nitrogen feedback in a temperate forest. Journal of Ecology. 2009; 97: 528–536.
View Article
Google Scholar

[103] View Article

[104] Google Scholar

[ref37] 37. Wurzburger N, Hendrick RL. Rhododendron thickets alter N cycling and soil extracellular enzyme activities in southern Appalachian hardwood forests. Pedobiologia. 2007; 50(6): 563–576.
View Article
Google Scholar

[106] View Article

[107] Google Scholar

[ref38] 38. Adamczyk B, Ahvenainen A, Sietiö OM, Kanerva S, Kieloaho AJ, Smolander A, et al. The contribution of ericoid plants to soil nitrogen chemistry and organic matter decomposition in boreal forest soil. Soil Biology & Biochemistry. 2016; 103: 394–404.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref39] 39. Jones GL, Scullion J, Worgan H, Gwynn-Joneset D. Litter of the invasive shrub Rhododendron ponticum (Ericaceae) modifies the decomposition rate of native UK woodland litter. Ecological Indicators. 2019; 107: 105597.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref40] 40. Appiah KS, Li Z, Zeng RS, Luo S, Oikawa Y, Fujii Y. Determination of allelopathic potentials in plant species in Sino-Japanese floristic region by sandwich method and dish pack method. International Journal of Basic and Applied Sciences. 2015; 4(4): 381–394.
View Article
Google Scholar

[115] View Article

[116] Google Scholar

[ref41] 41. Katayama A, Oyamada M, Abe H, Uemori K, Hishi T. Soil erosion decreases soil microbial respiration in Japanese beech forests with understory vegetation lost by deer. Journal of Forest Research. 2023; 28: 428–435.
View Article
Google Scholar

[118] View Article

[119] Google Scholar

[ref42] 42. Furusawa H, Miyanishi H, Kaneko S, Hino T. Movement of soil and litter on the floor of a temperate mixed forest with an impoverished understory grazed by deer (Cervus nippon centralis Temminck). Journal of the Japanese Forestry Society. 2003; 85: 318–325.
View Article
Google Scholar

[121] View Article

[122] Google Scholar

[ref43] 43. Obalum S, Chibuike G, Peth S, Ouyang Y. Soil organic matter as sole indicator of soil degradation. Environmental Monitoring and Assessment. 2017; 189: 176.
View Article
Google Scholar

[124] View Article

[125] Google Scholar

[ref44] 44. Matsuda Y, Shimizu S, Mori M, Ito SI, Selosse MA. Seasonal and environmental changes of mycorrhizal associations and heterotrophy levels in mixotrophic Pyrola japonica (Ericaceae) growing under different light environments. American Journal of Botany. 2012; 99: 1177–1188.
View Article
Google Scholar

[127] View Article

[128] Google Scholar

[ref45] 45. Baba T, Hirose D, Sasaki N, Watanabe N, Kobayashi N, Kurashige Y, et al. Mycorrhizal formation and diversity of endophytic fungi in hair roots of Vaccinium oldhamii Miq. in Japan. Microbes and Environment. 2016; 31(2): 186–189.
View Article
Google Scholar

[130] View Article

[131] Google Scholar

[ref46] 46. Chambers SM, Curlevski NJA, Cairney JWG. Ericoid mycorrhizal fungi are common root inhabitants of non-Ericaceae plants in a south-eastern Australian sclerophyll forest. FEMS Microbiology Ecology. 2008; 65(2): 263–270.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref47] 47. Marañón-Jiménez S, Radujković D, Verbruggen E, Grau O, Cuntz M, Peñuelas J, et al. Shifts in the Abundances of saprotrophic and ectomycorrhizal fungi with altered leaf litter inputs. Frontiers in Plant Science. 2021; 12:682142.
View Article
Google Scholar

[136] View Article

[137] Google Scholar

[ref48] 48. Ward EB, Duguid MC, Kuebbing SE, Lendemer JC, Warren II, RJ, Bradford MA. Ericoid mycorrhizal shrubs alter the relationship between tree mycorrhizal dominance and soil carbon and nitrogen. Journal of Ecology. 2021; 109: 3524–3540.
View Article
Google Scholar

[139] View Article

[140] Google Scholar

[ref49] 49. Prada-Salcedo LD, Prada-Salcedo JP, Heintz-Buschart A, Buscot F, Goldmann K. Effects of tree composition and soil depth on structure and functionality of belowground microbial communities in temperate European forests. Frontiers in Microbiology. 2022; 13:920618.
View Article
Google Scholar

[142] View Article

[143] Google Scholar

[ref50] 50. Bergero R, Girlanda M, Bello F, Luppi A, Perotto S. Soil persistence and biodiversity of ericoid mycorrhizal fungi in the absence of the host plant in a Mediterranean ecosystem. Mycorrhiza. 2003; 13(2): 69–75.
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref51] 51. Zhan W, Zhou Y, Jeppesen E, Wang L, Tan H, Zhang J. Linking heterotrophic bacterioplankton community composition to the optical dynamics of dissolved organic matter in a large eutrophic Chinese lake. Science of the Total Environment. 2019; 679: 136–147.
View Article
Google Scholar

[148] View Article

[149] Google Scholar

[ref52] 52. Chen J, Du Y, Zhu W, Pang X, Wang Z. Effects of organic materials on soil bacterial community structure in long-term continuous cropping of tomato in greenhouse. Open Life Sciences. 2022; 17: 381–392.
View Article
Google Scholar

[151] View Article

[152] Google Scholar

[ref53] 53. Lan J, Wang S, Wang J, Qi X, Long Q, Huang M. The shift of soil bacterial community after afforestation influence soil organic carbon and aggregate stability in Karst Region. Frontiers in Microbiology. 2022; 13: 901126.
View Article
Google Scholar

[154] View Article

[155] Google Scholar

[ref54] 54. Chen FC, Katayama A, Oyamada M, Tsuyama T, Kijidani Y, Tokumoto Y. Effects of soil environmental changes accompanying soil erosion on the soil prokaryotes and fungi of cool temperate forests in Southern Japan. Journal of Forest Research, 2023;
View Article
Google Scholar

[157] View Article

[158] Google Scholar

[ref55] 55. Osburn ED, Elliottt KJ, Knoepp JD, Miniat CF, Barrett JE. Soil microbial response to Rhododendron understory removal in southern Appalachian forests: Effects on extracellular enzymes. Soil Biology and Biochemistry. 2018; 127: 50–59.
View Article
Google Scholar

[160] View Article

[161] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Study site

Environmental properties of the forest understories

Assessment of soil microbial communities and their functions

Identification of Pieris root-associated fungi

Statistical analysis

Results

Environmental properties

Soil microbial communities and functionality

Pieris root-associated fungi

Discussion

Differences between understory environments according to Pieris dominance for different canopy trees

Differences between soil microbial communities according to Pieris dominance for different canopy trees

Conclusions

Supporting information

S1 Fig. Location of the research regions (Maruju, M; Hirono, H).

S2 Fig.

S3 Fig.

S4 Fig. Relationships between the weight of Pieris leaves and the relative abundance of undefined saprotroph.

S1 Table. Pieris sapling information for assessing root-associated fungi.

S2 Table. Results of the linear-mixed-model analyses of the relative abundance of prokaryotic phyla with SOM content.

S3 Table. Results of the linear-mixed-model analyses of the relative abundance of fungal phyla with two environmental variables and the read count (%) of root-associated fungi in the Pieris root samples.

S4 Table. Results of Spearman’s correlation coefficient test for the environmental variables.

S5 Table. Results of the linear-mixed-model analyses of the relative abundance of fungal trophic mode with two environmental variables and the read count (%) of root-associated fungi in the Pieris root samples.

S6 Table. Results of the linear-mixed-model analyses of the relative abundance of fungal guild with two environmental variables and the read count (%) of root-associated fungi in the Pieris root samples.

S7 Table. Relative read count of the order-level Pieris root-associated fungi of six Pieris root samples.

S1 Appendix. Environmental data measured in this study.

Acknowledgments

References