Effects of Transformation of Inefficient Camellia oleifera Plantation on Soil Quality and Fungal Communities

Abstract

Camellia oleifera, a key economic forestry species in southern China, struggles with low productivity due to suboptimal planting management. Recently, transforming old or unadopted varieties of C. oleifera plantations has been recognized as a means to enhance economic benefits and production. However, the impact of these transformations on soil properties and fungal communities has received little attention. In this study, we targeted pre-renewal old C. oleifera and post-renewal young C. oleifera, Pinus massoniana, and Cunninghamia lanceolata. Through field sampling and soil physicochemical property analysis, we developed a soil quality evaluation system that effectively analyzes fungal community structures and identifies key arbuscular mycorrhizal fungi (AMF) species for soil health. We found that the soil quality evaluation system for this region comprises pH, TK, AK, NO₃⁻, PO₄⁻ BG, ACP, F.simpson, AMF.shannon, and AMF.ace, which collectively indicated significant improvements in soil quality following transformation. Notably, the nutritional characteristics of the dominant fungal communities underwent marked changes, with an increase in pathogenic fungi in young C. oleifera and an expansion of ectomycorrhizal fungi in P. massoniana forests. The AMF communities in all four types of forest exhibited aggregation, and Scutellospora and Diversispora emerged as key species in the AMF community of C. oleifera. Additionally, Mortierella and Trichoderma were found to enhance plant resistance to pathogenic fungi. This study demonstrates that forestland transformation positively impacts soil quality and fungal community structure in C. oleifera, which provides valuable insights for future soil management in the region, both in terms of soil quality evaluation and fungal conservation.

1. Introduction

Oil tea (Camellia oleifera Abel.), an evergreen plant unique to China, is one of the world’s four major woody oil-bearing plants, including oil palm, olive, and coconut [1]. As a crucial economic forestry crop, it thrives in southern China’s Hunan, Hubei, and Guizhou regions, spanning approximately 7.5 million acres [2]. Well-managed, high-yielding C. oleifera plantations can produce over 600 kg of oil per hectare. However, over 70% of these plantations yield less than 22.5 kg per hectare annually, significantly diminishing their economic value [3]. The primary causes of this underperformance include old tree age, poor variety quality, and continuous cropping, leading to soil acidification, compaction, and the overall deterioration of soil quality [4,5]. Moreover, the improper assessment of soil conditions and climate characteristics leads to unsuitable variety selection, thereby affecting the fruiting rate and oil yield. A thorough analysis of the factors limiting the efficiency of oil tea plantations, coupled with the adoption of suitable stand transformation techniques, can significantly improve yield and productivity [6].

To enhance the efficiency of low-yielding C. oleifera plantations, strategies like plantation renewal, variety improvement, and fertilization are primarily employed [7]. Such land conversion directly impacts C. oleifera soil properties (e.g., moisture content and soil bulk density marked increase after conversion) [8]; chemical properties (e.g., soil pH, organic matter, and total nitrogen significantly different) [9]; and enzyme activities related to carbon (C), nitrogen (N), and phosphorus (P) cycles [10]. On the other hand, diversity and changes in above-ground vegetation influence the accumulation and quality of soil litter, altering root interactions and driving shifts in the structure and function of underground microbes, especially for root-associated fungal communities [11]. For example, Liu et al. [10] identified significant shifts in both the alpha and beta diversities of fungi consequent to the conversion of forests into artificial oil tea plantations, resulting in alterations to fungal communities and their life strategies. Lan et al. [12] documented a decline in Basidiomycota alongside an increase in the dominance of Ascomycota during this transformation process. Notably, arbuscular mycorrhizal fungi (AMF), predominantly represented by Glomeromycota, constitute the most widespread symbiotic fungi [13]. The dependence of C. oleifera plantations on AMF is remarkable, with the inoculation of these fungi significantly enhancing plant biomass [1]. Furthermore, saprophytic fungi play a vital role in decomposing litter organic matter, thereby enhancing nutrient absorption for the host plants [14]. Consequently, a comprehensive understanding of the dynamics of fungal communities, particularly in the wake of plantation transformation, is essential for effective plantation management.

Soil quality forms the foundational material basis for the sustainability of agricultural and forestry production, playing a critical role in regulating energy flow, material cycling, and maintaining biodiversity [15]. Various methods have been employed to evaluate the soil quality of different types of forests, with the comprehensive index evaluation method being particularly prevalent [16]. This approach simplifies influencing factors into several primary quality elements and establishes evaluation criteria, integrating physical, chemical, and biological soil factors to calculate the soil quality index (SQI) [17]. In C. oleifera plantations, soil evaluations typically include factors such as soil density, sand content, organic matter, available phosphorus, potassium, iron, and acid phosphatase [18,19]. While studies on soil quality in these C. oleifera plantations have predominantly focused on physical and chemical properties, there has been limited focus on biological and microbiological indicators. Consequently, the methodologies and systems for soil evaluation require further research and refinement.

In this study, our objective was to quantify the alterations in soil quality and fungal communities resulting from the renewal and transformation of C. oleifera. We evaluated the physicochemical properties, biological characteristics, and fungal communities of old C. oleifera plantations and three types of monoculture artificial forests. Our goals were to (1) analyze alterations in soil physicochemical properties following forest transformation; (2) assess the critical indicators impacting soil quality; and (3) elucidate the shifts in soil fungal communities, with a focus on the pivotal species of AMF.

2. Materials and Methods

2.1. Study Site and Sampling

The research site is located in the southeastern part of Tongren City, Yuping County, Zhu Jia Chang Town, in the eastern part of the Guizhou Province, China (27°28′–27°31′ N, 108°34′–109°09′ E) (Figure S1). The experimental site features a subtropical monsoon humid climate. The total annual sunshine duration is 1328.4 h, the average annual temperature is 17 °C, and the average precipitation ranges from 1100 to 1400 mm. The topography of the experimental site is dominated by low mountains and hills, with altitudes ranging from 600 to 800 m. The soil type is primarily Quaternary yellow soil developed from limestone slope deposits, classified as Luvisols in the World Reference Base for Soil Resources [20].

In Zhu Jia Chang Town, the low-efficiency C. oleifera plantations have been cultivated with the native variety known as “Han Lu Zi” for over 50 years. This plantation was planted at a density of 1050 trees per hectare. Strip felling was carried out, and all other trees, shrubs, and deep-rooted weeds within the plantation were completely removed. In 2010, the young C. oleifera (“Chang Lin” No. 54 and “Xiang Lin” No. 27) were interplanted at a density of 1155 trees per hectare. At the same time, Pinus massoniana (P. massoniana) and Cunninghamia lanceolata (C. lanceolata) plantations were planted at a density of 2 m × 3 m. During this period, 1.5 to 3 kg of organic fertilizer and 0.5 to 1 kg of calcium magnesium phosphate fertilizer were applied per pit annually in C. oleifera plantations.

In this study, three spatially interspersed sample plots (25 m × 25 m) of each aforementioned plantation type were randomly established, with three sampling points in each plot chosen diagonally for detailed investigation and sampling. In mid-March 2021, soil samples were collected at depths of 0–15 cm, 15–30 cm, and 30–45 cm at each sample point in the experimental plantations using a ring knife, yielding a total of 108 soil samples. After debris such as stones, branches, and fallen leaves were removed, the samples were homogeneously mixed and pooled by soil layer for each plot. The collected soil samples were divided into two portions, with one portion stored in a −80 °C ultra-low-temperature freezer for soil fungal DNA amplification and extraction, and the other portion air-dried at room temperature for the determination of soil physical and chemical properties.

2.2. Analysis of Soil Properties and Enzyme Activities

Soil bulk density (BD) and soil water content (SWC) were determined using the ring knife method, with a ring knife specification of 100 cm³ (diameter 7 cm, height 5.2 cm, and inner wall thickness 0.2 cm). The soil pH was determined with a 1:2.5 mixture (soil–water ratio) using an FE20K pH meter. After settling until the supernatant became clear, the pH value was measured using a Delta320 pH meter instrument (Melter-Toledo Instruments Co., Shanghai, China). The soil organic carbon (SOC) content was determined using the potassium dichromate oxidation method. The total nitrogen (TN) content was determined using the Kjeldahl method. The total phosphorus (TP) content was determined using the persulfate digestion + molybdenum antimony anti-colorimetric method. Ammonium nitrogen (NH₄⁺), nitrate nitrogen (NO₃⁻), and available phosphorus (PO₄⁻) were extracted with 1 M KCl and measured using a continuous flow analyzer (Skalar-analytic B.V, SAN++). The total potassium (TK), available potassium (AK), calcium (Ca), magnesium (Mg), iron (Fe), and manganese (Mn) contents were determined after digestion with nitric acid–hydrochloric acid–hydrofluoric acid, using an inductively coupled plasma atomic emission spectrometer (ICP-AES, Thermo Fisher Scientific Inc., Waltham, MA, USA). The enzyme activity was determined using the microplate method: referring to Deng et al.’s [21] method for enzyme assays, the activity of β-glucosidase (BG) related to carbon acquisition, N-acetylglucosaminidase (NAG) related to nitrogen acquisition, and acid phosphatase (ACP) related to phosphorus acquisition were measured.

2.3. Molecular Analysis

The soil fungal community diversity and composition were sequenced using the Illumina high-throughput sequencing platform for soil microorganisms. For each cryopreserved sample, 0.25 g of soil was taken, and the total DNA was extracted using the Mobio Power Soil kit (MoBio Laboratories Inc., Carlsbad, CA, USA). The extracted genomic DNA was checked using 1% agarose gel electrophoresis.

Specific primers with barcodes were synthesized according to the specified sequencing regions. For fungi, universal primers ITS1F (5′- CTTGGTCATTAGAGAGAGTAA-3′) and ITS2 (5′- GCTGCGTTCTTCTCTCGATGGC-3′) and AMF-specific primers AMV4-5NF (5’-AAGCTCGTAGTTGAATTTCG-3′) and AMDGR (5’-CCCAACTATCCCTATTAATCAT-3’) were used to amplify the ITS and 18S-SSU regions of rRNA subunits, respectively. PCR was performed using TransGen AP221-02 (TransStart Fastpfu DNA Polymerase, Beijing, China), and the PCR machine used was ABI Gen Amp^® 9700 (ABI, Waltham, MA, USA). All samples were processed under standard experimental conditions, with three replicates for each sample. PCR products from the same sample were mixed and checked by 2% agarose gel electrophoresis. Gel-purified PCR products were recovered using the AxyPrepDNA Gel Recovery Kit (AXYGEN Company, Union City, CA, USA), followed by Tris-HCl elution and 2% agarose gel electrophoresis.

Based on the preliminary quantification results from gel electrophoresis, PCR products were quantified using the Quanti Fluor™-ST Blue Fluorescence Quantitation System (Promega Corporation, Madison, WI, USA). Subsequently, the PCR products were mixed in proportion to the sequencing requirements for each sample. According to the standard protocol of Majorbio Technic Group Laboratory (Shanghai, China), high-throughput sequencing (2 × 300 bp) was performed on the Illumina Miseq PE300 platform.

The raw sequence data obtained from the sequencing were filtered, clustered, and trimmed using QIIME (Version 9.1) in the platform, removing chimeras. Biological information filtering, quality control, and denoising were performed on both ends of the reads, and the clustering of data was conducted using FLASH (Version 1.2.11) and Usearch (Version 11). The sequences with identification thresholds above 97% were clustered into operational taxonomic units (OTUs). Finally, the obtained data were compared with the National Center for Biotechnology Information (NCBI) nucleotide sequence database (https://unite.ut.ee/) for the fungal classification of each OTU, and the microbial community structure was statistically analyzed at various taxonomic levels (phylum, class, order, family, genus, and species).

2.4. Soil Quality Evaluation

Using soil physical and chemical properties, enzyme activity, and fungal diversity as indicators, the soil quality index (SQI) was evaluated through Principal Component Analysis (PCA). Various soil attributes, including bulk density, nutrients, enzyme activity, and fungal diversity, were compiled into the dataset. PCA was then applied to all indicators, employing maximum variance rotation for a clearer interpretation of the uncorrelated factors [22]. Indicators with eigenvalues (λ) ≥ 1 and absolute factor loadings ≥ 0.5 were selected for inclusion in the first Minimum Dataset (MDS1) for further analysis [23]. To balance information preservation and minimize redundancy, norm values for each indicator were calculated, selecting those within the 10% range of MDS1’s maximum standard value.

A Pearson correlation analysis of the soil physical, chemical, and biological indicators was conducted to identify multicollinearity by examining the correlation coefficients among indicators. Redundant indicators within a component were identified through this analysis [24]. Variables correlated at ≥0.6 were further refined, retaining only the indicator with the highest loading, and excluding others to prevent data redundancy. Finally, the actual measured data of soil indicators were standardized uniformly using the membership function, and calculations for each membership value in MDS2 indicators (range 0–1) were performed (Formula (1)) [24].

The weight of a standardized indicator is equal to its commonality divided by the sum of the commonalities of the selected indicators.

$$f\left(x_i\right)=\left\{\begin{array}{cc}1.0& ,x_i\ge x_2\hfill \\ 0.1+\frac{0.9\left(x_i-x_1\right)}{x_2-x_1}& ,x_1<x_i<x_2\hfill \\ 0.1& ,x_i\le x_1\hfill \end{array}\right.$$

Or type reverse S:

$$f\left(x_i\right)=\left\{\begin{array}{cc}0.1& ,x_i\ge x_2\hfill \\ 0.1+\frac{0.9\left(x_i-x_1\right)}{x_2-x_1}& ,x_1<x_i<x_2\hfill \\ 1.0& ,x_i\le x_1\hfill \end{array}\right.$$

(1)

In the equation, f(x) represents a linear score between 0.1 and 1.0, x_i is the measured value of soil indicators, and x₁ and x₂ are the minimum and maximum values of the soil indicators, respectively.

Finally, for all indicators in MDS2, normalization, scoring, and weighted calculations are performed. The SQI formula is as follows (Formula (2)).

$$SQI={\displaystyle \sum }_{i=1}^n{W}_i\times N_i$$

(2)

SQI values range from 0 to 1, and W_i is the weight value of the i-th soil index, N_i is the membership degree of corresponding indicators, and n is the number of soil indicators.

2.5. Statistical Analyses

Based on 12 soil sampling points, the Levene test method was employed to examine whether the data satisfy the homogeneity of variance and normal distribution. If normal distribution assumptions were not met, square root or logarithmic transformations were applied to achieve normality. Subsequently, one-way analysis of variance (ANOVA) and Tukey’s Honestly Significant Difference (HSD) test were performed to analyze significant differences (p < 0.05) among different plantations and soil layers. PCA and Pearson correlation analysis utilized IBM SPSS software (v26, SPSS Inc., Chicago, IL, USA) for SQI calculations.

For sequencing data, α-diversity indices including Shannon–Wiener’s, Simpson, ACE, and Sob were computed using Mothur. Fungal community β-diversity was analyzed in R (v 4.3.1) software using packages such as “vegan”, “dplyr”, “ggcor”, and “phyloseq” to assess the impact of different types of forest and soil properties on fungal community composition. Permutational multivariate analysis of variance (PERMANOVA) was conducted to decompose the total variance, elucidating the explanatory power of different grouping factors or environmental factors in different plantations. PERMANOVA tests were employed to assess the statistical significance of the variables’ explanatory power. Non-metric multidimensional scaling (NMDS) ordination using Bray–Curtis distance was employed to analyze community β-diversity in R.

Correlations between soil properties in different plantations and the top five relative abundances of fungi and soil nutrients were assessed. DCA analysis was initially conducted using Species-sample to determine the size of the first axis of gradient lengths; results indicated a value greater than 4.0, leading to the choice of Canonical Correspondence Analysis (CCA). Venn diagrams were generated using the “Venn” package in R to illustrate the distribution of fungal OTUs and a visualization of the top 100 relative abundances of OTUs was created. The top 100 fungal OTUs and the top 20 ectomycorrhizal fungi in each forest stand were selected, and the “igraph” package was used to construct a matrix for network nodes and line segments based on correlations (r > 0.4 and p < 0.05) as filtering criteria. The visualization of the network graph and estimation of its natural connectivity were achieved using Gephi v. 10.0. After that, the network’s natural connectivity was estimated by “attacking” nodes in the static network to reveal the robustness of the network [25].

The top 20 species of ectomycorrhizal fungi at the OTU level in each forest were selected, and the “indicspecies” package was utilized to obtain indicator species for ectomycorrhizal fungi through 8000 random fitting tests. A dominance heatmap of ectomycorrhizal fungi was generated using the “pheatmap” package in R, based on original sequence data. Matching was performed individually with the (UNITE) database, selecting species with the highest theoretical matching scores to determine the host and origin.

3. Results

3.1. Soil Properties

Significant variations in pH, SOC, TK, Mn, Fe, Mg, and NO₃⁻ were observed across different plantations (Figure 1 and detailed in Table S1) (p < 0.05). After transformation, SOC in young plantations increased by 20.6%~52.8%. In the young C oleifera plantation, the contents of TK, Mn, and NO₃⁻ significantly increased by 36%, 45%, and 116%, respectively. A similar increasing trend in TK, Mg, and Fe levels was also observed in the conifer plantations.

After the forest transformation, there were no significant differences observed in the activities of the three enzymes (Figure 2). The activity of the phosphatase (ACP) and N-acetylglucosaminidase (NAG) enzyme was observed in the following order: P. massoniana > C. lanceolata > old C. oleifera > young C. oleifera. For the β-glucosidase (BG) enzyme activity, the old C. oleifera plantation exhibited the highest activity, with the P. massoniana plantation following closely behind.

3.2. Soil Quality Index

The PCA results indicate that the cumulative variance explained by the first seven principal components (with eigenvalues ≥ 1) is 77% (Table S2). In PC1, indicators TP, NH₄⁺, AMF.sobs, and AMF.ace were selected based on absolute loading values (≥0.5), and AMF.sobs and AMF.ace were chosen based on the maximum normalized range values (within 10%) (Table S3). Additionally, through correlation analysis (Table S4), the high-standard indicator AMF.ace was identified in PC1. Similarly, pH, total potassium, available potassium, nitrate nitrogen, phosphate, BG, ACP, F.simpson, and AMF.shannon were selected from PC2 to PCA7, respectively. In summary, pH, TK, AK, NO₃⁻, PO₄⁻, BG, ACP, F.simpson, AMF.shannon, and AMF.ace are the key indicators selected by MDS2. Based on the scores and weighted indicators (Table S5), the SQI values for the four afforested plantations ranged from 0.5 to 0.34 (Figure 3). Following the transformation, soil quality saw a notable improvement, with the P. massoniana plantation achieving the highest score and the C. lanceolata plantation registering the lowest among the three transformed plantations.

3.3. Diversity, Composition, and Fungal Traits of the Fungal Communities

After the sequence quality filtering, a total of 2,142,839 effective fungal sequences remained across 36 samples, with an average length of 248 bp. The average number of sequence reads per sample was 59,523 ± 8569. The RDP classifier Bayesian algorithm, based on the UNITE8.0/ITS_fungi database with a homogeneity threshold of 97%, resulted in the clustering of 3133 OTUs. The diversity indices of fungal community composition are presented in Table S6. Neither the tree species, the soil layer, nor their interaction had a significant impact on fungal community composition. At the 97% similarity level, the Venn diagram analysis of soil fungal community OTUs between different plantations is illustrated in (Figure S2). The shared OTUs among the four plantations amounted to 18, constituting 1.6% of the total OTUs.

After transformation, the prevalence of pathogenic fungi in young C. oleifera plantations surged by 20.7 times, with 80.4% of this increase concentrated in the 0–15 cm soil layer (Figure 4a). In P. massoniana plantations, ectomycorrhizal fungi increased by 5 times. Meanwhile, including those associated with soil, wood, litter, and saprotrophs, ectomycorrhizal fungi increased by 64% in young C. oleifera plantations, declined by 15.7% in P. massonina plantations, and saw a modest increase of 3% in C. lanceolata forests.

Within the top 100 species by relative abundance, a total of 12 phyla were identified, and Ascomycota, Basidiomycota, and Mortierellomycota emerged as the top three groups of fungal communities (Figure 4b). Notably, Ascomycota saw the most significant increases in C. lanceolata and young C. oleifera plantations, with rises of 128.9% and 47%, respectively. Basidiomycota experienced the most substantial decrease in the young C. lanceolata plantation, with a reduction of 87%, whereas Mortierellomycota surged by 84% in young C. oleifera plantations. In the young C. oleifera plantations, Ascomycota was predominantly found in the 0–15 cm soil layer, accounting for 53% of the total relative abundance.

3.4. Analysis of Composition and Indicator Species of AMF

Based on the amplification results using specific primers for AMF, a total of 839,966 effective sequences of AMF DNA remained across 36 samples. The average length of these sequences was 217 bp, with an average sequence read count of 23,332.39 ± 2748.51 per sample. The RDP classifier Bayesian algorithm, based on the MaarjAM database with a homogeneity threshold of 97%, resulted in the clustering of 515 OTUs.

The alpha diversity of the AMF community, as shown in (Table S7), indicated that the richness indices Ace (F = 3.198, p < 0.05), Chao1 (F = 3.574, p < 0.05), and Sobs (F = 3.188, p < 0.05) were significantly influenced by different tree species (F = 3.198, p < 0.05). Specifically, the Ace richness index (F = 3.570, p < 0.05) and the Chao1 richness index (F = 3.768, p < 0.05) were simultaneously affected by the soil layer. The interaction between tree species and soil layer did not have a significant impact on the diversity and richness of the AMF community in the samples. The Venn diagram analysis of soil AMF community OTUs between different plantations is illustrated in Figure S3. The shared OTUs among the four plantations amounted to 1, constituting 0.4% of the total OTUs.

At the genus level, there was a clear distinction in the AMF communities among the four plantations (Figure 5). In the old C. oleifera plantations, Dominikia was the dominant genus, comprising 94% of the total fungal community. In contrast, Scutellospora is the predominant genus in both the young C. oleifera and C. lanceolata plantations, constituting 72% and 93% of the total fungal community, respectively. The P. massoniana plantations are dominated by Rhizophagus, which accounts for 78% of the total fungal community.

At the species level, in the old C. oleifera, Dominikia OTU3 emerged as the dominant species within the 30–45 cm soil layer of the old C. oleifera plantations (Figure 6). Conversely, in the young C. oleifera, Scutellospora and Diversispora were primarily found in the 30–45 cm soil layer, while being notably absent in the 0–15 cm layer. Indicate Species Analysis revealed that Diversispora OTU488 and Diversispora OTU483 are indicative species in the young C. oleifera plantation.

In the 30~45 cm soil layer of the old C. oleifera, the most abundant Dominikia OTU3, and the indicative species OTU488 in the young C. oleifera plantation are highly similar to the AMF LC508248 and LS446163 found in Marchantia foliacea, achieving a 100% matching degree (

Table 1. List of arbuscular mycorrhizal fungal OTUs approximately matched with NCBI Genbank.

OTU No.	Closest BLAST Match in Genbank	Distribution	Host	Fungal Origin
OTU487	MH205886 (100%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU115	LR795569 (100%)	Unspecified	Ammophila arenaria	Rhizophagus irregularis
OTU3	LC508248 (100%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU188	LS997552 (100%)	Unspecified	not indicated	Rhizophagus intraradices
OTU5	MH205808 (100%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU460	KF489071 (92.61%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU455	LC274508 (94.6%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU450	MK551290 (100%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU413	KF489071.1 (97.2%)	Brazil; Sete Lagoas	Zea mays	uncultured Glomeraceae
OTU412	MK84137 (98.59%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU411	LT798728 (100%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU410	MN597014 (100%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU488	LS446163 (100%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU486	MG829319 (100%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU483	MG429401 (98.17%)	Brazil; Barra do Pirai, Rio de Janeiro State	not indicated	Dentiscutata reticulata
OTU438	MH205823 (96.67%)	New Zealand; Ross	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU43	LT836938 (100%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU406	AB747013 (100%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU35	LR795582 (98.17%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea
OTU189	MG829514 (98.17%)	New Zealand; Otira valley	Marchantia	Glomus mycorrhizal symbiont of Marchantia foliacea

). Additionally, the young C. oleifera plantations feature a relatively high abundance of AMF species primarily aligned with Dentiscutata from the Gigasporaceae family.

3.5. Co-Occurrence Network of Fungi and AMF in Different Plantations

Following the transformation, the young C oleifera and P. massoniana plantations experienced increases of 47% and 41% in the number of nodes, respectively, while the edges surged by 446% and 251% (Figure 7). The number of nodes and edges in the C. lanceolata plantation did not show a significant increase.

Each forest exhibited a notable clustering of AMF, with Scutellospora and Diversispora identified as the core components. In the young C. oleifera plantations, the plant pathogenic fungus Boeremia, central to the ecosystem, exhibited antagonistic effects against saprotrophic fungi such as Geminibasidium, Sagenomellas, and Pleotrichocladium. Additionally, Boeremia and the plant pathogenic fungus Botrytis both showed antagonism towards saprotrophs Entoloma, Sagenomellas, and mycoparasite Trichoderma. Key litter saprotrophic fungi, like Articulospora, also displayed antagonism towards the plant pathogen Coniosporium, wood saprotroph Paraphaeosphaeria, and soil saprotrophic fungus Calcarisporiella. Cladosporium, another litter saprotroph, antagonizes Geminibasidium and Sagenomellas but correlates positively with the plant pathogenic fungi Phaeosphaeriopsis and Alternaria. In the P. massoniana plantations, ectomycorrhizal fungi such as Lactarius, Russula, and Amphinema were not only abundant but also closely connected with other fungal nutritional types. Conversely, the ectomycorrhizal fungus Inocybe exhibited weaker connections. In the C. lanceolata plantations, the saprotrophic fungus Entoloma, despite its low abundance, significantly correlated with other nutritional types in the community.

The linear fitting curve for the young C. oleifera plantations, with a slope of 11.5, exhibits the highest absolute value, indicating its fluctuation amplitude surpasses that of other plantations (Figure 7e). This trend signifies a more pronounced decrease in the connectivity of the community network, implying reduced stability in the fungal community composition.

3.6. Relationship between Soil Properties and Fungal Communities

PERMANOVA indicated significant differences in fungal communities among different plantations (R² = 0.267, p < 0.001). The fungal composition in P. massoniana plantations correlated with soil Fe and Mg content (p ≤ 0.01), whereas in old C. oleifera plantations, it is associated with Mn content (p ≤ 0.01) (Figure 8a). Dominant fungal genera exhibit strong correlations with soil Fe, Mg, Mn, pH, and SOC (Figure 8b). For instance, Retiboletus increased with higher Fe and Mg content, while Ilyonectria, Archaeorhizomyces, and Leohumicola decreased in abundance with rising soil pH.

4. Discussion

4.1. Impact of Conversion on Soil Properties and Soil Quality

The growth and development of plants, as well as litter composition, are intimately linked to soil property effectiveness. This study identified significant alterations in soil properties, particularly in soil pH, SOC, and TK, between the old C. oleifera plantations and the newly established young plantations following transfer. For an upward trend of soil pH in the young C. oleifera, P. massoniana, and C. lanceolata plantations, the observed shift may stem from transformation-induced disturbances that altered the original soil structure, leading to a reduction in plant species diversity and a decrease in organic acid secretion by roots and ectomycorrhizal fungi [26]. Furthermore, the removal of more acidic surface soil layers and the application of fertilizers contributed to an increase in soil pH by enhancing the concentration of exchangeable cations [27,28]. The newly planted forests displayed higher SOC content, indicating a rapid growth phase characterized by an active organic carbon input and dynamic symbiotic interactions between roots and microorganisms. The content of Mn, as a heavy metal element, showed significant variation across different plantations. C. oleifera demonstrates a pronounced capacity for Mn accumulation and tolerance, varying in absorption efficiency during different growth stages [29]. A comparison of soil Fe content among C. oleifera, P. massoniana, and C. lanceolata highlighted significant differences. C. oleifera showed lower soil Fe levels, as the Fe absorbed during growth phases was translocated from the roots and concentrated in the fruits of C. oleifera [30]. Plants primarily absorb NO₃⁻ and NH₄⁺; post-transformation, young C. oleifera retained more nitrate nitrogen, indicating a predominance of NH₄⁺ as the nitrogen source [31]. The transformation of the C. oleifera plantation did not significantly impact the activities of ACP, NAG, and BG enzymes, possibly due to an unchanged fungal biomass [32].

Soil quality assessment results indicated an improvement in soil quality following transformation, with AMF richness being the most significant contributor to SQI scores, followed by AMF diversity. For plants like C. oleifera that rely on AMF, the richness and diversity of AMF are vital for improving growth indicators [33]. AMF proliferation is influenced by environmental factors such as pH, NO₃⁻, and AK. Both available potassium and TK play critical roles in the management of oil tea, with potassium deficiency manifesting as the yellowing and eventual death of C. oleifera leaves. The application of AK fertilizers is key to enhancing carbohydrate synthesis, increasing stress resistance, and improving the content of crude fats and palmitic acid in oil crops [34]. The ACP enzyme facilitates the decomposition of organic phosphorus compounds, supplying vital phosphorus for plant growth. Enhancing soil microorganisms that decompose organic phosphates accelerates litter decomposition [35]. Consequently, SQI analysis offers new management approaches for C. oleifera plantations, suggesting that the increased use of microbial fertilizers and the judicious application of potassium fertilizer can enhance soil quality.

4.2. Changes in Fungal Community Diversity and Structure

Following the transformation of forests, the diversity and richness of fungal communities in four distinct types of forest exhibited no significant changes, aligning with Kerfahi’s [36] observations on the conversion of tropical rainforest to rubber plantations. This stability may stem from the fungal communities’ vulnerability to human intervention, where frequent disturbances diminish soil fungal diversity [36,37]. Notably, the richness of AMF communities underwent substantial changes. The disruption of native fungal communities in the old C. oleifera plantations led to a shift in AMF types, notably increasing the presence of ‘ruderal’ species [38].

After conversion, a marked difference was noted in the β-diversity of the fungal communities. This variance is likely attributable to the introduction of new species that modified soil properties, thereby inducing synergistic shifts in the structure of the fungal communities and becoming instrumental in the formation of dominant groups [39,40,41]. Different fungal functional types exhibited varied responses to the transformation. In the young C. oleifera plantations, an uptick in pathogenic fungus abundance was observed, possibly due to increased litter and root exudates attracting pathogenic insects, thereby heightening the risk of plant infections [42]. Network co-occurrence mapping in these forests revealed associations between the saprophytic fungus Cladosporium and the pathogenic fungi Phaeosphaeriopsi and Altemaria [43]. Studies indicate that as litter abundance escalates, Cladosporium may progressively become detrimental to plants [44]. Despite diverse infection mechanisms and nutrient acquisition strategies, all pathogenic fungi are identifiable by plant defense systems. Studies suggest that plants recruit specific microbial communities to combat pathogens, either by directly antagonizing the pathogens or modulating the host’s immune response [45,46,47]. Pathogens like Boeremia, in conjunction with saprophytic fungi, create antagonistic effects, with genera such as Geminibasidium boosting plant disease resistance [48]. Botrytis, by counteracting Trichoderma, serves as a biocontrol agent. Simultaneously, Chaetomium enhances plant disease resistance through interactions with other fungi [43,49]. The Articulospora, associated with AMF and soil saprophytic fungi, possesses a genetic type broadly distributed in aquatic and terrestrial environments [50].

Compared to other plantations, the young C. oleifera plantations exhibited the highest abundance of AMF and saprophytic biomass, with saprophytic fungi playing a dominant role and significantly interacting with arbuscular mycorrhizal fungi. Cao et al. [14] found that AMF and saprophytic fungi mutually enhance each other’s functions, thereby accelerating litter mineralization. Following the transformation, soil quality in the young C. oleifera plantations improved; however, its enzyme activity was observed to be the lowest. This phenomenon could be attributed to the abundance of soil nutrients, which may reduce the activity of rhizosphere enzymes, a consequence of the interactions between AMF and saprophytic fungi. Given that both fungus types are filamentous microbes competing for the same niche, competition and antagonism can occur [51].

Soil quality assessments, along with NMDS and CCA analyses, reveal that the core communities of all four tree species are predominantly influenced by the Mn, Fe, Mg, SOC, and pH levels. Recent research has emphasized the significance of Mn in litter decomposition, specifically its role as a key component in the production of lignin-degrading enzymes by saprophytic fungi [52]. Additionally, Fe is an essential factor for microbial activities due to its ability to coordinate and activate molecular oxygen [53]. In a study by Yang et al. [54], the application of varying levels of Mg fertilizer to tea gardens revealed a notable impact on fungal community composition and network complexity. The diversity, complexity, and key taxa of these fungal communities directly regulate SOC storage [55]. Furthermore, fungi exhibit strong adaptability across different pH levels, allowing them to occupy dominant positions within their ecological niches. This adaptability underscores the crucial role of fungi in maintaining ecosystem function and stability [56].

4.3. Implication for Conservation of Key Fungal Species

Effective soil management is a cornerstone of successful agricultural systems, and the strategic use of fungal functions is essential for sustainable development [57,58,59]. Fungi are pivotal not only in decomposing organic matter and providing nutrients to plants but also in suppressing pathogenic microorganisms and safeguarding plants [60]. The extent of fungal involvement directly affects soil quality. Consequently, fungal biodiversity assessment should go beyond mere biodiversity indices to include fungal population structure analysis, elucidating their role in influencing soil quality and plant health [57]. Selecting appropriate cultivation practices to augment fungal biodiversity is crucial for preventing root diseases and preserving soil quality and health.

This study highlights the significant role of saprophytic fungi, particularly the Mortierella, in plant growth and environmental adaptation. As a saprophyte, Mortierella predominated in young C. oleifera plantations. Its notable characteristics include abundance in healthy soils, resilience under adverse conditions, the improvement of host phosphorus and iron uptake, synthesis of plant hormones and ACC (1-aminocyclopropane-1-carboxylic acid), and protection of crops from pathogens [61,62]. For instance, in the acidic soils of Mexico, over 400 apple trees were devoid of Mortierella pathogenicity [63]. Ozimek et al.’s [64] research underscores Mortierella’s potential to boost nutrient absorption, improve crop protection, and reduce fertilizer and pesticide usage. Leveraging fungal community functions can elevate soil fertility. Shifting from synthetic inorganic fertilizers to bio-fertilizers containing microorganisms can enhance plant mineral nutrient uptake. Network co-occurrence mapping shows the widespread presence of Trichoderma in all plantations, playing a crucial role in both aged and young C. oleifera plantations. Woo et al. [65] identified that Trichoderma activates plant enzymes responsible for disease resistance, effectively serving as a biocontrol agent. Concurrently, Cai et al. [66] highlighted Trichoderma’s contribution to litter decomposition and the enhancement of host nutrition, thereby supporting sustainable plant growth.

Regarding AMF distribution, Scutellospora and Diversispora are most abundant in the young C. oleifera plantations, with Diversispora identified as an indicator species. Karthikeyan et al. [67] studied the effects of six AMF on tea plant growth, noting each type’s growth-promoting capabilities, particularly Scutellospora calipers as the most efficient. Dong et al.’s [68] research on inoculating Diversispora showed increased tea plant stem diameter, height, leaf area, and flower count, significantly enhancing root system morphology and promoting root hair growth.

5. Conclusions

This study demonstrates that transforming land from old-aged C. oleifera to young plantations mitigates soil acidification and increases SOC, TK, NO₃⁻ content, and AMF abundance, thereby enhancing soil quality. The transformation markedly changes the nutritional dynamics within fungal communities, enhancing connectivity among fungal members. Notably, in young C. oleifera plantations, this includes an increase in pathogenic fungi, which leads to a significant aggregation of AMF and a subsequent decrease in the stability of the community. Concurrently, the AMF species Diversispora exerts a crucial function during the early growth stages of C. oleifera plantations, suggesting that the inoculating AMF from Diversispora can enhance root length and biomass accumulation. Moreover, the widespread usage of Mortierella and Trichoderma supports achieving biocontrol objectives, potentially reducing the reliance on pesticides and insecticides. In conclusion, the transformation has ameliorated soil quality and enriched the fungal community structure.