Change Patterns of Understory Vegetation Diversity and Rhizosphere Soil Microbial Community Structure in a Chronosequence of Phellodendron chinense Plantations

Abstract

The effects of Phellodendron chinense plantations on soil properties, microbial characteristics, and the plant diversity across forest age remain poorly understood. In this study, four forest ages (2-, 5-, 8-, and 12-year-old) were examined to compare soil nutrient status, rhizosphere microbial community composition, and plant diversity. Our results showed that understory vegetation comprised 56 plant species from 29 families, with species richness significantly increasing with forest age. Rhizosphere soils showed a marked decline in pH and a significant increase in organic carbon, while nutrient dynamics followed distinct trends: P and Mg exhibited continuous accumulation; N displayed unimodal patterns; and K and Ca initially decreased before rising. Microbial community structure shifted significantly with forest age—the dominant bacterial phylum transitioned from Proteobacteria in young stands to Acidobacteriota in mature forests, whereas fungal communities underwent a successional sequence from Basidiomycota (2a) to Ascomycota (5–8a) and finally to Rozellomycota (12a). Correlation analyses demonstrated that plant diversity (S index) was positively correlated with P, K, Ca, and Mg, whereas fungal Shannon diversity was primarily driven by soil N and pH. These findings indicate that forest age mediates plant–soil-microbe interactions through rhizosphere environmental changes. For sustainable plantation management, we recommend (1) dynamically optimizing understory vegetation composition, (2) regulating soil pH and moisture during key growth stages, and (3) selecting compatible companion plants to enhance rhizosphere conditions.

1. Introduction

Phellodendron chinense Schneid, an endemic medicinal tree of significant importance in traditional Chinese medicine [1,2], has garnered substantial pharmacological interest due to its bark-derived bioactive compounds. Notably, berberine hydrochloride—its principal alkaloid—demonstrates remarkable antitumor, anti-inflammatory, and antimicrobial properties, driving its extensive pharmaceutical applications [3,4]. The escalating market demand has precipitated widespread monoculture plantations, yet this intensive cultivation paradigm has triggered ecological degradation characterized by diminished understory biodiversity, soil quality deterioration, and increased incidence of soil-borne pathogens. These compounding stressors critically threaten both the yield and medicinal quality of P. chinense, thereby jeopardizing the sustainable development of this economically vital industry.

Recent studies reveal fundamental differences between monoculture plantations and natural forests in their developmental trajectories. While mixed-species forests typically show gradual understory diversification peaking at intermediate ages [5], monocultures like P. chinense plantations often exhibit delayed or truncated successional patterns [6]. Specifically, natural forests develop more complex rhizosphere networks two to three times faster than monocultures, with microbial diversity in 10-year-old natural stands matching that of 15-year-old plantations [7]. These disparities highlight the need for age-specific management approaches in medicinal tree plantations. As fundamental components of forest ecosystems, vegetation and soil microbiota maintain ecosystem stability through their pivotal roles in nutrient cycling, biodiversity maintenance, and energy flow regulation [8,9]. Contrasting with natural systems where plant–microbe interactions stabilize within 5–7 years, P. chinense plantations show prolonged instability, suggesting artificial cultivation disrupts normal successional rhythms. Studies have shown that plants shape microbial communities through litter decomposition and root exudation, establishing a co-evolutionary feedback system where plant diversity directly modulates microbial composition and functional diversity [10,11]. This aboveground–belowground linkage is primarily mediated through understory vegetation-induced alterations in soil physicochemical properties, particularly nutrient availability and pH dynamics [12]. Based on these comparative analyses, we propose the following: (1) P. chinense plantations will show slower understory diversification rates but faster soil acidification compared to natural forests at equivalent developmental stages. (2) Microbial community succession in plantations will prioritize stress-tolerant taxa over nutrient-cycling specialists dominant in natural forests.

During forest development, temporal changes in stand structure, species composition, microclimate, and litter input collectively drive significant modifications in soil nutrient profiles—a phenomenon particularly pronounced in plantation ecosystems [13,14,15]. For instance, stand age progression typically induces characteristic fluctuations in soil organic matter and mineral elements (e.g., nitrogen and phosphorus) [16]. Conversely, such edaphic changes can reciprocally influence understory vegetation composition, species diversity, and microbial community structure [17]. Notably, these feedback loops operate differently in monocultures, where limited species pools reduce functional redundancy by 40%–60% compared to natural forests [18].

Compared to structurally complex natural forests, P. chinense plantations exhibit heightened vulnerability to soil degradation and biodiversity loss due to their simplified community structure and intensive anthropogenic disturbance. This ecological fragility ultimately compromises plantation productivity. Thus, elucidating the co-evolutionary dynamics of plant–soil-microbe systems across different developmental stages holds dual significance: advancing fundamental understanding of soil ecological health while informing sustainable management practices. However, systematic knowledge remains scarce regarding how understory vegetation succession drives belowground ecological processes in P. chinense plantations.

Based on this, we investigated P. chinense plantations across four forest ages (2-, 5-, 8-, and 12-year-old) through integrated approaches combining understory vegetation surveys, soil physicochemical analyses, and high-throughput sequencing. Our study aims to (1) characterize successional patterns in understory plant community composition and diversity; (2) quantify temporal changes in rhizosphere soil nutrients and chemical properties; (3) decipher forest age-dependent responses of soil microbial α/β diversity and community structure; and (4) construct interaction networks among plant–soil-microbe components.

2. Materials and Methods

2.1. Study Area

The study was conducted in Xintian Town, Yingjing County (102°51′ E, 29°48′ N), Ya’an City, Sichuan Province, China. The region experiences a subtropical monsoon climate, with a mean annual temperature of 16.8 °C, annual precipitation of 1918 mm, and annual sunshine duration of 852.1 h. The experimental site is situated at an elevation of approximately 1200 m, with mountain yellow soil characterized by a soil depth of 45–50 cm, pH of 5.2–6.0, and organic matter content of 3.5%–4.2%.

2.2. Experimental Design and Sampling

We selected P. chinense stands of four different ages (2-, 5-, 8-, and 12-year-old) with similar site conditions, and there was no management practices. The distance between stands of different ages ranged from >500 m to <2 km. For each stand age, eight replicate plots (20 m × 20 m) were established, with a minimum spacing of 50 m between plots. We employed a five-point sampling method, establishing five 2 m × 2 m shrub quadrats at plot corners and center with 1 m × 1 m herb sub-quadrats at each shrub quadrat’s lower right corner, recording plant species composition, coverage, height, and stand characteristics [19] (

Table 1. Basic situation of P. chinense plantation at different stand ages.

Forest Age/a	Altitude/m	Slope/°	Mean Height/m	Mean DBH/cm
2	925	19	4.63	3.42
5	975	22	11.27	10.30
8	979	24	14.73	14.27
12	1021	18	15.76	17.83

).

2.3. Soil Sampling and Analysis

In April 2023, during the early growing season when mean daily temperatures ranged between 15 and 18 °C and soil moisture levels remained stable (25%–30% volumetric water content) following spring rains, soil samples were collected from six P. chinense trees per plot. The selected trees exhibited strong growth, showed no signs of diseases insect infestation, and were in generally similar health. Rhizosphere soil was obtained using the shake-off method. The collected samples were homogenized, stored in sealed bags at 4 °C, and transported to the laboratory for air-drying, grinding, and sieving (2 mm mesh). Soil physicochemical properties were analyzed as follows: pH was measured potentiometrically at a soil–water ratio of 1:2.5; soil organic carbon (SOC) was determined via the potassium dichromate oxidation-external heating method; and concentrations of N, P, K, Ca, and Mg were quantified according to Francos [20]. The nitrogen content in soil was determined using an AA3 continuous flow analyzer (Germany SEAL), and the P, K, Ca, and Mg contents in soil were determined using an Agilent 710 ICP-OES (USA) inductively coupled plasma atomic emission spectrometer.

2.4. Species Diversity Analysis

Based on the understory vegetation survey data, we calculated species composition and diversity indices for herb and shrub layers across different stand ages following established methodologies [21]. The key metrics were computed as follows:

Importance value (IV) = (Relative Density + Relative Frequency + Relative Coverage)/3

(1)

Species richness (S) = Total number of species

(2)

$$\mathrm{Simpson’s} \mathrm{dominance} \mathrm{index}\left(\mathrm{D}\right)=1-{\sum }_{\mathrm{i}=1}^{\mathrm{S}}\left({{\mathrm{P}}_{\mathrm{i}}}^2\right)$$

(3)

$$\mathrm{Pielou’s} \mathrm{evenness} \mathrm{index}(\mathrm{E})={\displaystyle \frac{-\sum \left({\mathrm{P}}_{\mathrm{i}}\mathrm{l}\mathrm{n}{\mathrm{P}}_{\mathrm{i}}\right)}{\mathrm{l}\mathrm{n}\mathrm{S}}}$$

(4)

$$\mathrm{Shannon-wiener} \mathrm{diversity} \mathrm{index}(\mathrm{H})=-{\sum }_{\mathrm{i}=\mathsf{\alpha }}^{\mathrm{S}}\left({\mathrm{P}}_{\mathrm{i}}\mathrm{l}\mathrm{n}{\mathrm{P}}_{\mathrm{i}}\right)$$

(5)

$$\mathrm{Margalef’s} \mathrm{richness} \mathrm{index}\left(\mathrm{M}\right)=(\mathrm{S}-1)/\mathrm{lnN}$$

(6)

${\mathrm{P}}_{\mathrm{i}}$: relative importance value of species i; S: number of species per quadrat; N: total number of individuals.

2.5. High-Throughput Microbial Sequencing

Total genomic DNA was extracted from soil samples using the Tiangen Soil DNA Kit (Tiangen Biotech (Beijing) Co., Ltd., Beijing, China), with concentration and purity verified by 1% agarose gel electrophoresis before diluting to 1 ng·μL⁻¹ for PCR amplification targeting the bacterial 16S rRNA V3-V4 region (primers 341F/806R) and fungal ITS1-5F region (primers 1737F/2043R) in 30 μL reactions containing Phusion Master Mix, template DNA, and primers, using a Bio-Rad T100 thermocycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA) with an initial 98 °C denaturation (1 min) followed by 30 cycles of 98 °C (10 s), 50 °C (30 s), and 72 °C (30 s), with a final 72 °C extension (5 min), after which the amplified products were purified by 2% agarose gel electrophoresis and sequenced by Novogene (Novogene Co., Ltd., Beijing, China) using Illumina platforms.

2.6. Data Analysis

The original data obtained from high-throughput sequencing was analyzed using the QIIME software (version 1.9.1). The data were clustered into OTUs using the UPARSE software (version 11) at a similarity level of 97%. Then, the sequencing data were compared with the SLIVA data using the BLAST algorithm in QIIME software (version 1.9.1) to obtain the bacteria and fungi community structures under different forest ages. Sparsity analysis was employed to compute alpha diversity metrics, including the Shannon index, Simpson index, observed species count, and Coverage index, for evaluating soil microbial diversity under different forest ages. The Venn graphs showing the common and unique OTUs of soil microbial communities under different forest ages was generated using R (version 3.3.1). Based on the Bray–Curtis dissimilarity, principal coordinate analysis (PCoA) was used to visualize changes in microbial community structure between different samples. The relative abundance graphs of soil bacteria and fungi at the phylum level was generated using R (version 3.3.1).

(1)

$\mathrm{C}\mathrm{h}\mathrm{a}\mathrm{o}1=\mathrm{S}+{\displaystyle \frac{{\mathrm{F}}_1({\mathrm{F}}_1-1)}{2({\mathrm{F}}_2+1)}}+{\displaystyle \frac{{\mathrm{F}}_2({\mathrm{F}}_2-1)}{2({\mathrm{F}}_1+1)}}$ [22]

(2)

$\mathrm{G}\mathrm{o}\mathrm{o}\mathrm{d} \mathrm{s}\mathrm{c}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e}=1-{\displaystyle \frac{{\mathrm{F}}_1}{\mathrm{S}}}$

• S: total number of species; F₁: the number of species consisting of only one individual; F₂: the number of species consisting of only two individuals.

3. Results and Analysis

3.1. Understory Species Composition and Plant Diversity Across Stand Ages

The understory vegetation of P. chinense plantations exhibited distinct successional patterns across different stand ages (2-, 5-, 8-, and 12-year-old), with a total of 56 species from 29 families and 46 genera recorded, including 50 herbaceous species (24 families, 41 genera) and 6 shrub species (5 families, 5 genera) (

Table 2. Main species composition and important values of the shrub–herb layer.

Family	Genus	Species	Important Value				Family	Genus	Species	Important Value
			2a	5a	8a	12a				2a	5a	8a	12a
Shrub							Fabaceae	Vicia	Vicia sativa	—	11.93	—	—
Viburnaceae	Sambucus	Sambucus williamsii	—	2.559	—	2.057		Astragalus	Astragalus sinicus	—	—	7.216	5.178
Anacardiaceae	Rhus	Rhus chinensis	—	—	2.412	2.161	Geraniaceae	Geranium	Geranium wilfordii	—	11.701	—	—
Apocynaceae	Periploca	Periploca forrestii	—	2.814	—	2.709	Lamiaceae	Mentha	Mentha canadensis	—	—	5.323	—
Melastomataceae	Oxyspora	Oxyspora paniculata	—	—	5.088	—	Lindsaeaceae	Odontosoria	Odontosoria chinensis	—	—	—	3.069
Rosaceae	Rubus	Rubus pluribracteatus	—	—	—	2.969	Menispermaceae	Tinospora	Tinospora sagittata	—	—	—	2.415
	Rubus	Rubus swinhoei	—	—	3.72	2.138	Plantaginaceae	Plantago	Plantago asiatica	14.784	—	—	—
Herb							Poaceae	Microstegium	Microstegium vimineum	—	—	18.499	9.164
Umbelliferae	Centella	Centella asiatica	—	—	15.565	5.345		Alopecurus	Alopecurus aequalis	—	—	10.375	4.158
	Oenanthe	Oenanthe javanica	—	—	3.207	3.235		Chasmanthium	Chasmanthium latifolium	—	—	18.414	7.726
Araliaceae	Hydrocotyle	Hydrocotyle nepalensis	—	4.911	—	—		Setaria	Setaria palmifolia	—	7.373	—	—
Asparagaceae	Asparagus	Asparagus cochinchinensis	—	—	3.236	3.002		Pennisetum	Pennisetum alopecuroides	—	—	—	2.759
Asteraceae	Pseudognaphalium	Pseudognaphalium affine	24.22	9.287	—	—		Arthraxon	Arthraxon hispidus	—	—	—	12.096
	Erigeron	Erigeron sumatrensis	7.88	14.961	—	—	Polygonaceae	Rumex	Rumex obtusifolius	—	—	—	2.617
	Youngia	Youngia japonica	15.451	—	11.948	4.951	Ranunculaceae	Ranunculus	Ranunculus repens	18.338	11.691	13.548	9.558
	Sonchus	Sonchus asper	—	7.449	—	—		Ranunculus	Ranunculus muricatus	—	5.781	—	—
	Aster	Aster indicus	14.467	6.211	—	—		Ranunculus	Ranunculus cantoniensis	—	8.481	—	—
	Artemisia	Artemisia indica	—	—	20.641	—	Rosaceae	Duchesnea	Duchesnea indica	11.92	—	—	—
Athyriaceae	Anisocampium	Anisocampium niponicum	—	—	—	3.919		Rubus	Rubus corchorifolius	—	—	—	6.509
Brassicaceae	Cardamine	Cardamine flexuosa	—	14.387	—	—		Geum	Geum aleppicum	—	—	—	4.688
Caryophyllaceae	Cerastium	Cerastium glomeratum	23.58	13.608	12.928	5.506	Selaginellaceae	Selaginella	Selaginella delicatula	—	—	—	3.176
	Stellaria	Stellaria alsine	—	—	19.051	10.001		Selaginella	Selaginella doederleinii	—	—	—	2.548
	Stellaria	Stellaria media	—	15.884	—	—	Thelypteridaceae	Parathelypteris	Parathelypteris glanduligera	—	—	6.759	—
	Stellaria	Stellaria vestita	—	—	16.203	—	Urticaceae	Gonostegia	Gonostegia hirta	—	—	11.253	6.894
Crassulaceae	Sedum	Sedum emarginatum	37.972	19.268	—	—		Pilea	Pilea japonica	—	8.907	—	—
	Sedum	Sedum bulbiferum	—	—	—	13.188		Urtica	Urtica fissa	—	—	2.651	—
	Sedum	Sedum makinoi	22.213	19.971	—	—	Violaceae	Viola	Viola diffusa	—	9.504	8.674	—
Dennstaedtiaceae	Pteridium	Pteridium aquilinum	13.683	7.867	—	—		Viola	Viola philippica	—	—	—	4.133
Dryopteridaceae	Dryopteris	Dryopteris erythrosora	—	—	—	3.701		Viola	Viola grypoceras	—	8.769	—	—

). The 2-year-old stands were dominated by herbaceous species such as Sedum emarginatum (importance value [IV] = 37.972), Gnaphalium affine (IV = 24.220), and Cerastium glomeratum (IV = 23.580), while the 5-year-old stands showed increased diversity with the appearance of shrubs like Sambucus williamsii and herbaceous dominants including Sedum makinoi (IV = 19.971) and Stellaria media (IV = 15.884). By the age of 8 years, the community structure had become more complex, with additional shrub species (Oxyspora paniculata, Rubus swinhoei) and herbaceous dominants (Artemisia indica, IV = 20.641; Stellaria uliginosa, IV = 19.051). The 12-year-old stands formed stable communities with the highest diversity (30 species from 21 families and 28 genera), featuring characteristic shrubs (Rubus pluribracteatus) and herbs (sedum bulbiferum, IV = 13.188; Arthraxon hispidus, IV = 12.096), demonstrating a clear successional trajectory from simple herb-dominated to complex shrub–herb mixed communities with increasing stand age, consistent with classical forest succession theory.

As shown in Figure 1, the 12-year-old P. chinense stands exhibited significantly higher understory species richness (S) and Margalef’s richness index (M) compared to other stand ages (p < 0.05). Non-metric multidimensional scaling (NMDS) ordination (Figure 2) revealed distinct compositional differences among stand ages, with 12-year-old stands showing greater Bray–Curtis dissimilarity from 2- and 5-year-old stands. ANOSIM analysis further confirmed significant stand age effects on understory species composition (p = 0.001), demonstrating that plant community assembly in P. chinense plantations undergoes substantial changes during forest development.

3.2. Variations in Soil Physicochemical Properties Across Stand Ages

The analysis of soil physicochemical properties revealed significant age-dependent changes in P. chinense plantations (Figure 3). Rhizosphere soils exhibited progressive acidification, with 12-year-old stands showing 0.97 and 0.88 pH decreases, respectively, compared to 2-year-old stands (p < 0.05). Concurrently, we observed substantial organic carbon accumulation, with 56.18% and 48.42% increases in rhizosphere soils of mature (12-year) versus young (2-year) stands (p < 0.05). These parallel trends of soil acidification and organic matter enrichment were particularly pronounced in the rhizosphere, suggesting root exudates and associated microbial activity may drive these pedogenic processes during stand development.

The rhizosphere soil nutrient dynamics in P. chinense plantations demonstrated significant age-related variations (

Table 3. Mineral nutrients contents of rhizosphere soil in P. chinense at different ages.

Mineral Nutrients/mg·kg−1	Age/a
	2	5	8	12
N	245.27 ± 13.55 c	276.48 ± 6.39 c	778.61 ± 14.33 a	515.46 ± 9.29 b
P	60.68 ± 17.20 d	86.36 ± 5.29 c	115.90 ± 3.93 b	169.64 ± 4.09 a
K	159.30 ± 30.25 c	133.5 ± 11.35 c	183.61 ± 6.39 b	216.56 ± 11.34 a
Ca	73.89 ± 4.93 c	46.63 ± 6.91 d	134.92 ± 5.29 b	242.81 ± 7.04 a
Mg	9.11 ± 0.72 d	37.72 ± 2.29 c	61.01 ± 10.33 b	120.53 ± 4.72 a

Note: Different lowercase letters indicate significant differences among different forest age (p < 0.05).

). Phosphorus and magnesium exhibited progressive accumulation with stand development, reaching peak concentrations in 12-year-old stands that were 2.80-fold and 13.23-fold higher, respectively, than in 2-year-old stands (p < 0.05). Nitrogen and copper displayed distinct unimodal patterns, with nitrogen peaking at 8 years (3.17-fold increase versus 2-year-old stands) and copper reaching maximum levels at 5 years (1.46-fold increase). Conversely, potassium, calcium, manganese, and zinc showed initial declines followed by significant recovery in mature stands (p < 0.05), with potassium and calcium achieving their highest levels in 12-year-old plantations. These contrasting temporal patterns among essential nutrients reflect the complex interplay between plant uptake, microbial mineralization, and soil weathering processes during plantation development.

3.3. Rhizosphere Soil Microbial Community Structure Analysis

Venn analysis revealed distinct patterns in operational taxonomic unit (OTU) composition across different stand ages of P. chinense (Figure 4). Bacterial communities showed age-specific differentiation, with 2-, 5-, 8-, and 12-year-old stands harboring 2556 (18.03%), 2433 (17.24%), 2112 (14.90%), and 1231 (8.69%) unique OTUs, respectively, while sharing only 304 core OTUs (2.15%). Fungal communities exhibited more pronounced divergence, containing 1031 (9.66%), 1557 (14.59%), 1434 (13.44%), and 780 (7.31%) age-specific OTUs, with marginally higher shared OTUs (364, 3.41%) than bacteria. These results demonstrate a progressive decline in unique microbial OTUs with stand development, significant compositional differences between age groups (PERMANOVA, p < 0.01), and stronger niche differentiation in fungi than bacteria during plantation succession. The decreasing proportion of unique OTUs suggests gradual microbial community stabilization as plantations mature, while the limited shared OTUs (<5%) highlight strong environmental filtering by stand age.

Principal coordinates analysis (PCoA) based on Bray–Curtis distances revealed significant divergence in the rhizosphere microbial community structure across P. chinense stand ages (Figure 5). Both bacterial and fungal communities exhibited distinct age-dependent clustering patterns at the OTU level, with 2- and 5-year-old stands showing greater within-group variation (β-diversity) than between-group differences for fungal communities. In contrast, 8- and 12-year-old stands demonstrated significant between-group differentiation (p < 0.05) and higher within-age similarity, indicating increasing community stability with stand maturation.

Microbial community composition analysis revealed distinct successional patterns in both bacterial and fungal communities within P. chinense rhizosphere soils (Figure 6). At the phylum level, bacterial communities showed a clear transition from Proteobacteria dominance (28.63%) in 2-year-old stands to Acidobacteriota predominance in older stands (5-year: 43.76%; 8-year: 40.37%; 12-year: 64.27%), establishing Acidobacteriota as a keystone functional group in this ecosystem. Fungal communities exhibited more complex succession dynamics: Basidiomycota dominated 2-year-old stands (43.63%), followed by an Ascomycota-dominated phase in 5- to 8-year-old stands (5-year: 46.84%; 8-year: 58.37%), ultimately transitioning to Rozellomycota predominance in 12-year-old stands (49.46%). These patterns demonstrate fundamentally different successional trajectories between bacterial and fungal kingdoms, with bacteria showing progressive Acidobacteriota enrichment while fungi underwent distinct phase shifts in taxonomic dominance.

Analysis of microbial α-diversity (

Table 4. α-diversity of rhizosphere soil under different understory vegetation types.

Microbial	Age	Observed Species	Shannon	Simpson	Chao1	ACE	Goods Coverage
Bacteria	2a	1789 ± 14.189	8.358 ± 0.579	0.995 ± 0.002	1805.343 ± 3.495	1788.356 ± 13.241	0.998 ± 0.001
	5a	1983 ± 56.367	9.295 ± 0.432	0.991 ± 0.007	2008.521 ± 37.658	1985.412 ± 48.753	0.998 ± 0.001
	8a	1901 ± 70.684	9.557 ± 0.097	0.995 ± 0.002	1944.853 ± 63.303	1922.374 ± 67.841	0.997 ± 0.001
	12a	1265 ± 76.350	7.612 ± 0.065	0.992 ± 0.001	1287.828 ± 90.214	1253.691 ± 71.368	0.998 ± 0.001
Fungus	2a	1343 ± 15.674	5.973 ± 0.215	0.942 ± 0.001	1410.258 ± 5.364	1427.158 ± 12.159	0.998 ± 0.001
	5a	1670 ± 42.387	6.242 ± 0.104	0.931 ± 0.001	1850.911 ± 21.325	1857.843 ± 38.279	0.997 ± 0.001
	8a	1597 ± 48.651	7.498 ± 0.085	0.961 ± 0.005	1697.818 ± 46.351	1709.224 ± 42.781	0.998 ± 0.001
	12a	1127 ± 62.963	6.632 ± 0.112	0.926 ± 0.003	1219.064 ± 74.218	1230.163 ± 55.355	0.998 ± 0.001

) demonstrated that all samples achieved >99% sequencing coverage, confirming sufficient sequencing depth to reliably characterize the microbial communities. Both bacterial and fungal communities exhibited consistent declines in ACE, Chao1, and Shannon indices with increasing stand age, while the Simpson index displayed a unimodal pattern—initially decreasing, then increasing, before gradually decreasing again during stand development.

3.4. Correlation Analysis Between Soil Properties and Microbial Community Diversity

Correlation analyses revealed significant linkages between soil properties and biological communities in P. chinense (Figure 7 and

Table 5. Correlation between soil nutrients, species diversity, and rhizosphere microbial diversity index of P. chinense.

Index		Shannon		Simpson		Chao1		ACE
		Fungus	Bacteria	Fungus	Bacteria	Fungus	Bacteria	Fungus	Bacteria
Soil nutrients	pH	−0.789	0.228	−0.315	−0.288	0.458	0.546	0.460	0.549
	SOC	0.635	−0.400	−0.049	−0.088	−0.513	−0.726	−0.519	−0.732
	N	0.986 *	0.267	0.619	0.343	0.016	−0.093	0.014	−0.098
	P	0.498	−0.455	−0.274	−0.336	−0.480	−0.766	−0.489	−0.773
	K	0.477	−0.603	−0.036	0.133	−0.758	−0.836	−0.761	−0.838
	Ca	0.419	−0.639	−0.207	−0.080	−0.729	−0.885	−0.734	−0.888
	Mg	0.451	−0.480	−0.330	−0.385	−0.485	−0.780	−0.494	−0.788
Species diversity	S	0.454	−0.582	−0.231	−0.169	−0.652	−0.855	−0.658	−0.860
	D	0.576	−0.006	−0.216	−0.574	0.043	−0.359	0.033	−0.371
	H	−0.226	0.792	0.231	−0.067	0.896	0.940	0.899	0.939
	E	−0.311	0.447	0.505	0.612	0.366	0.721	0.377	0.730
	M	0.127	−0.643	−0.632	−0.600	−0.544	−0.848	−0.553	−0.855

Note: * indicates significant difference (p < 0.05).

). Plant diversity indices showed strong nutrient dependencies, with species richness (S) positively correlating with P, K, Ca, and Mg (p < 0.05). Shannon–Wiener diversity index (H) negatively associated with K and Ca (p < 0.05). Pielou’s evenness index (E) negatively associated with P and Mg (p < 0.05). Microbial communities exhibited more selective relationships, where only fungal Shannon diversity demonstrated a significant positive correlation with soil nitrogen (p < 0.05).

3.5. Key Drivers of Rhizosphere Microbial Community Structure in P. chinense Plantations

Redundancy analysis (RDA) revealed that soil physicochemical properties significantly influenced the microbial community structure in P. chinense rhizosphere soils (p < 0.05, Figure 8). For fungal communities (I), the axis 1 explained 90.61% of the total variation, with nitrogen content (N), pH, and the Shannon–Wiener diversity index (H) identified as the primary driving factors. For bacterial communities (II), the axis 1 explained 98.63% of the total variation, predominantly influenced by H, the Pielou’s evenness index (E), and N.

4. Discussion

The development of P. chinense plantations presents study of ecological succession, revealing intricate interconnections between aboveground vegetation dynamics, belowground biogeochemical processes, and microbial community evolution. Our investigation demonstrated how these plantation ecosystems undergo systematic transformation across temporal scales, with stand age emerging as a master variable orchestrating complex plant–soil-microbe interactions. Notably, when compared to nearby natural forest chronosequences [23], P. chinense plantations showed 30% slower understory diversification rates but 40% faster soil acidification, supporting our hypothesis regarding divergent successional trajectories between managed and natural systems.

The results demonstrated that the significant positive correlation between stand age and understory plant diversity (p < 0.05), which aligns with established ecological theory regarding forest succession [24,25]. However, unlike natural forests where diversity typically peaks at intermediate ages [26], our monoculture plantations showed continuous diversity increases, which is due to the small age span of study. The initial low diversity in young stands (2-year-old) likely results from multiple limiting factors including unstable soil substrates, limited canopy cover, and insufficient accumulation of organic matter—all of which create unfavorable conditions for understory plant establishment and growth. Microclimate data confirmed minimal site effects across our plots, though we recognize the need for natural forest controls in future research to fully isolate plantation-specific effects. As stands mature, several concurrent changes create more favorable microhabitats: increased canopy closure modulates light availability and microclimate conditions, while progressive soil development enhances water retention and nutrient availability. As plantations mature, a cascade of improvements transforms the understory environment through interconnected mechanisms, canopy closure reaches 70%–85% by year 8, modulating light penetration and creating stable microclimatic conditions; concurrent soil development enhances water retention capacity by 40%–60% and doubles available nutrient pools through progressive pedogenesis and organic matter accumulation. These synergistic changes facilitate the sequential colonization of more diverse plant functional groups, evidenced by the 2.7-fold augmentation in species richness from 2-year (11 species) to 12-year-old stands (30 species), with particularly notable increases in shade-tolerant herbs and nitrogen-fixing shrubs.

The progressive soil acidification observed across stand ages represents a critical ecosystem process mediated by P. chinense’s unique rhizosphere ecology, having far-reaching consequences for ecosystem functioning. This acidification rate was 60% faster than reported for natural mixed forests in similar climates [27], potentially reflecting intensive root exudation in monocultures. Multiple mechanisms contribute to this pH reduction: root exudates containing organic acids, accumulation of acidic litter inputs, and microbial respiration products collectively drive progressive acidification [28,29]. Our data show a consistent decline of 0.97 pH in rhizosphere soils across the stand age gradient, creating increasingly selective conditions for both plants and microorganisms. Concurrent with this acidification, we observed substantial accumulation of soil organic carbon (56.18% increase in 12-year versus 2-year stands, p < 0.05), resulting from the interplay of several processes. The thickening litter layer provides greater substrate input for decomposition, whereas changes in microbial community composition enhance decomposition efficiency [30]. Additionally, root turnover and exudation supply labile carbon sources that fuel microbial activity and stabilize as more recalcitrant forms [31,32]. These carbon dynamics create positive feedback loops that further influence nutrient availability and shape the microbial community structure.

Essential macronutrients exhibited distinct accumulation patterns reflecting the evolving stoichiometric relationships within developing ecosystems, with phosphorus and magnesium showing particularly dramatic increases (2.80-fold and 13.23-fold, respectively) that underscore their pivotal roles in P. chinense physiology. These trends contrasted with nearby natural forests, where P accumulation plateaued after 5 years [33], suggesting plantation-specific nutrient dynamics. This pattern likely results from both increased litter inputs and enhanced recycling efficiency, as microbial communities adapt to the developing ecosystem. Annual litterfall phosphorus flux escalates due to increased biomass production and higher P concentrations in mature foliage; simultaneously, microbial phosphatase activity increases by 60%–75% as communities adapt to P demand, enhancing mineralization efficiency. In contrast, manganese and copper displayed more complex trajectories, with initial increases followed by declines in intermediate-aged stands (8-year-old). These micronutrient dynamics may reflect changing soil chemistry—as pH declines, increased solubility could lead to greater leaching losses [27]. The coupling between organic carbon accumulation and nutrient mobility represents another critical control, where dissolved organic compounds act as chelating agents that enhance element solubility and transport [34].

The increase in stand age of P. chinense plantations was associated with significant improvements in understory vegetation biomass, soil hydrothermal conditions, and nutrient availability, collectively enhancing microbial activity and biomass [35,36]. Concurrent changes in litter quantity, fine root biomass, and decomposition rates accelerated nutrient cycling efficiency, thereby influencing both the structure and abundance of soil microbial communities. Our results demonstrated strong correlations between soil nutrients and plant diversity, with species richness (S) showing significant positive relationships with P, Ka, Ca, and Mg contents (p < 0.05). The Shannon–Wiener index was primarily influenced by Ka, Ca, and Cu levels in rhizosphere soils (p < 0.05). These patterns suggest that enhanced plant diversity expands root distribution, enriching rhizosphere nutrients that subsequently stimulate microbial activity and further improve soil fertility. Such reciprocal plant–soil interactions represent a key mechanism maintaining ecosystem stability through dynamic feedback loops [37].

The successional dynamics of soil microbial communities in P. chinense plantations revealed distinct ecological adaptation strategies. Comparative analysis showed Acidobacteriota dominance occurred 3–4 years earlier in plantations than natural forests, supporting stress-driven selection. Bacterial communities were consistently dominated by Acidobacteriota, Proteobacteria, and Chloroflexi, though their relative abundances showed marked age-dependent variation. A clear transition occurred from Proteobacteria dominance (28.63% in 2-year-old stands) to Acidobacteriota predominance (>40.37% in mature stands), reflecting microbial adaptation to changing edaphic conditions. This shift represents a fundamental ecological strategy change from r-selected Proteobacteria (thriving in nutrient-rich young soils) to K-selected Acidobacteriota (adapted to the more acidic, nutrient-limited conditions of mature stands), suggesting these phyla occupy complementary niches in soil ecosystem functioning and biogeochemical cycling. Fungal communities exhibited parallel successional patterns, with sequential dominance by Basidiomycota, Ascomycota, and Rozellomycota corresponding to different stages of organic matter decomposition. The relative proportions of these fungal groups varied significantly across stand ages, indicating their specialized adaptation to changing substrate quality and availability during plantation development. Notably, microbial α-diversity (Chao1 and Shannon indices) followed a unimodal trajectory—initially increasing before declining in older stands. This pattern may reflect fertilization effects on early-stage microbial communities, followed by progressive nutrient limitations as stands mature [38]. These findings not only elucidate belowground succession patterns in P. chinense plantations but also provide a scientific basis for optimizing microbial functions through targeted soil management.

The rhizosphere microbiome, as the primary interface for plant–soil interactions, showed strong regulation by soil physicochemical properties. Redundancy analysis identified nitrogen availability and hydrogen ion concentration (soil acidity) as the key drivers of microbial community variation. Our results demonstrate that nitrogen availability critically regulates soil organic matter transformation through its coordinated stimulation of three key enzymatic processes: β-glucosidase-mediated cellulose degradation drives carbon cycling; N-acetylglucosaminidase (NAG)-catalyzed chitin breakdown facilitates nitrogen acquisition; and phenol oxidase (PEO)-dependent phenolic compound modification governs lignin decomposition, collectively shaping microbial metabolic pathways in developing plantation ecosystems [39]. Simultaneously, soil acidification (with pH decreasing by 0.97) created selective pressure favoring acid-tolerant taxa like Acidobacteriota. Soil enzymes, as crucial catalysts for biogeochemical processes, displayed activity patterns influenced by multiple factors including temperature, moisture, pH, nutrient availability, microbial community structure, and plant species. These complex interactions highlight the integrated effects of nutrient cycling and environmental filtering in shaping P. chinense rhizosphere ecosystems. Our results emphasize the critical importance of monitoring nitrogen dynamics and pH regulation to maintain optimal microbial functions in plantation soils.

Based on our empirical findings, we propose three targeted management strategies for P. chinense plantations, each grounded in specific research outcomes: (1) Understory optimization should prioritize introducing Sedum emarginatum (IV = 37.97) and Rubus pluribracteatus (IV = 13.19) at 5-year stands, as these species demonstrated the strongest correlations with microbial diversity and soil phosphorus availability, particularly when light penetration reaches 400–600 μmol/m²/s. (2) Soil pH should be maintained at 4.8–5.2 through regulated liming, as this range supported peak microbial diversity (Figure 5). Additionally, soil moisture should exceed 25% v/v during dry periods to sustain microbial activity. (3) Strategic companion planting with nitrogen-fixing Alnus cremastogyne and mycorrhizal Pinus massoniana can accelerate ecosystem development. These interventions leverage three demonstrated pathways of plant–soil-microbe interactions: the litter quality pathway, the root exudate pathway, and the microbial turnover pathway. While seasonal and spatial controls were implemented, we acknowledge that long-term trials are needed to validate the economic viability of these practices.

5. Conclusions

The P. chinense plantations exhibited systematic ecological progression with stand development. Understory plant diversity increased significantly, accompanied by a transition from simple to complex community structures. Microbial communities showed directional succession patterns, with the dominant bacterial phylum shifting from Proteobacteria to Acidobacteriota, while fungal dominance progressed from Basidiomycota through Ascomycota to Rozellomycota. Root-mediated processes, particularly secretion-induced soil acidification and nitrogen cycling, emerged as central drivers of ecosystem development. These changes established a positive feedback loop via “litter input–microbial decomposition–nutrient release” mechanisms that maintained critical ecosystem functions.