Plant-soil feedback (PSF) and plant-phyllosphere feedback (PPF) critically influence plant performance and community assembly, yet their interactions remain poorly understood for subtropical tree species. To address this knowledge gap, we conducted a greenhouse experiment using eight subtropical tree species (four arbuscular mycorrhizal/AM and four ectomycorrhizal/EcM trees). All seedlings were exposed to two soil inoculum sources (conspecific vs. heterospecific soil) and two phyllosphere inoculum treatments (control vs. phyllosphere inoculation), aiming to elucidate PSF-PPF patterns and identify the underlying mechanisms shaping plant performance. Our findings revealed that AM trees exhibited a dramatic shift from weakly to strongly positive PSF following phyllosphere inoculation, whereas EcM trees maintained consistently positive PSF regardless of the treatment. Similarly, PPF strength in AM trees transitioned from positive to negative with soil inoculum switching from conspecific to heterospecific sources, while EcM trees showed consistently negative PPF across soil sources. PSF and PPF were positively correlated in AM but not EcM trees. Co-occurrence nwork analysis showed predominantly positive interactions (99%) within fungal communities and demonstrated that phyllosphere inoculation enhanced network complexity in AM systems but reduced connectivity in EcM systems. Mechanistically, the growth of EcM trees was primarily driven by plant functional traits, whereas that of AM trees depended on synergistic interactions between plant functional traits, soil abiotic properties, and soil fungal community profiles. Our findings highlight mycorrhizal type as a critical regulator of PSF-PPF relationships, offering novel mechanistic insights into subtropical tree community assembly by integrating above- and belowground microbial interactions.
Figure 1. The plant-soil feedback (PSF) index (a) and plant-phyllosphere feedback (PPF) index (b) calculated using plant total biomass, along with their relationships (c, d) in arbuscular mycorrhizal (AM) and ectomycorrhizal (EcM) trees under different treatments.
Figure 2. Shannon indices of soil fungal communities (A) and relative abundance of fungal functional guilds (B) associated with arbuscular mycorrhizal (AM) and ectomycorrhizal (EcM) tree species under different treatments: two phyllosphere inoculation regimes (control vs. phyllosphere inoculum) and two soil inoculation sources (conspecific soil [con-soil] vs. heterospecific soil [hetero-soil]).
Figure 3. Co-occurrence networks of soil fungal communities associated with arbuscular mycorrhizal (AM) (A) and ectomycorrhizal (EcM) tree species (B) under different treatments: two phyllosphere inoculation regimes (control vs. phyllosphere inoculum) and two soil inoculation sources (conspecific soil [con-soil] vs. heterospecific soil [hetero-soil]).
Figure 4. Zi-Pi plots showing the distribution of keystone genera based on their topological roles in soil fungal community networks associated with arbuscular mycorrhizal (AM) (A) and ectomycorrhizal (EcM) tree species (B) under different treatments: two phyllosphere inoculation regimes (control vs. phyllosphere inoculum) and two soil inoculation sources (conspecific soil [con-soil] vs. heterospecific soil [hetero-soil]).
Figure 5. Relationships between fungal functional guilds and both plant traits and soil abiotic properties (a, c) across all treatment combinations, and relationships between keystone fungal taxa identified under each treatment combination and both plant traits and soil abiotic properties (b, d) in arbuscular mycorrhizal (AM) (A) and ectomycorrhizal (EcM) tree species (B).
Figure 6. Piecewise structural equation model (SEM) on direct and indirect effects of multiple variables on total biomass of arbuscular mycorrhizal (AM) (a) and ectomycorrhizal (EcM) tree species (b) under different treatments: two phyllosphere/leaf inoculation regimes (control [N] vs. phyllosphere inoculum [Y]) and two soil inoculation sources (conspecific soil [C] vs. heterospecific soil [H]).
Literature:
Yanhong Wang, Yumei Pan, Qiqian Wu, Aiping Wu, Sirong Zhang, Xinping Wang, Yan Li, Naili Zhang*. 2026. Mycorrhizal type mediates coupled plant-soil and plant-phyllosphere feedbacks: divergent mechanisms driving subtropical tree growth. Geoderma. 471:117879.https://www.sciencedirect.com/science/article/pii/S0016706126002077.
