Leaf δ15N, δ13C and Their Associations with Soil Fungal Biodiversity, Ectomycorrhizal and Plant Pathogenic Abundance in Forest Ecosystems of China

Leaf δ 15 N and δ 13 C are important functional traits in biogeographic studies of forest ecosystems. However, little is known about their relationships with soil fungal biodiversity, ectomycorrhizal, and plant pathogen abundance at large scales. In this study, leaf and soil samples were collected from 33 forest reserves along a large range across China to explore the associations between leaf δ 15 N and δ 13 C and soil fungal biodiversity, ectomycorrhizal, and plant pathogen relative abundance using molecular and stable isotope techniques. We found large-scale biogeographic patterns for leaf δ 15 N, δ 13 C, soil fungal biodiversity, and ectomycorrhizal relative abundance. The soil-plant-microbial interaction may contribute to the variations in leaf δ 15 N, δ 13 C, and soil fungal communities across different types of forest ecosystems. Temperature and precipitation were the main factors affecting large-scale biogeographic patterns of latitude and longitude. Leaf δ 15 N was mainly affected by the relative abundance of ectomycorrhizal fungi and leaf δ 13 C was affected by the relative abundance of plant pathogens. Leaf δ 15 N and δ 13 C may be indicators reecting soil fungal communities in forest ecosystems.


Introduction
Soil microbes can exhibit biogeographic patterns in species diversity and distribution (Hanson et al. 2012), and soil fungi are crucial components of microbial communities in forest ecosystems, where they play fundamental ecological roles in soil formation, conservation, and regulating nutrient cycling (Zeilinger et al. 2015). Recent studies have reported distinct distribution patterns for different fungal taxa like ectomycorrhizal fungi (EMF) and arbuscular mycorrhizal fungi (AMF) across latitudes (Tedersoo et al. 2012; Davison et al. 2015). However, the mechanisms underlying these biogeographic patterns are di cult to distinguish in forest ecosystems because multiple community assembly processes may govern the biogeographic variation in soil fungal diversity (Ricklefs et al. 2004; Kraft et al. 2011;Bahram et al. 2018). In addition, biotic interactions within fungal communities may vary along large-scale geographic gradients, resulting in different strengths in the assembly of mechanisms across forest ecosystems (Hu et al. 2019). Yet, to date, few studies have focused on how these complex geographic processes together in soil fungal diversity in different forest types of China (Huang et al. 2019).
Therefore, a more comprehensive perspective is required to understand how and why their relative in uences vary across forest types (Powell et al. 2015).
How plant pathogens and EMF affect the biodiversity and construction of forests along large-scale gradients remains underreported. Recent studies suggest that plant pathogens play an important role in promoting plant species coexistence through negative plant-soil feedback at local scales in forests, which subsequently affects the biodiversity and construction of forest ecosystems (Bever et  focused on the variations in plant performance using potting experiments with the application of fungicide or inoculation with several fungal isolates (e.g., plant pathogens and EMF), which may overlook the relationships between different fungal taxa in natural conditions (Wang et al. 2019). Thus, a comprehensive study linking the spatial distributions of fungal biodiversity and functional taxa, e.g., EMF and plant pathogens, to plant functional indicators across large scales (i.e., across China) is urgently needed.
As important indicators for studying plant ecophysiology, stable isotopes provide fundamental insights into how plants interact with and respond to environmental factors, improving our understanding of how the plants adapt to their environment (Dawson et al. 2002). Leaf δ 15 N and δ 13 C were found more closely associated with belowground communities than soil isotopic values, where they could be used to predict the properties of belowground communities ). In particular, leaf δ 13 C re ects the balance between leaf conductance and photosynthetic rate, which can used to calculate intrinsic water use e ciency (iWUE) created by Farquhar et al. (1989). Based on this, leaf δ 13 C is widely used to ). However, we still lack the answers to the following questions: how do soil fungal biodiversity, ectomycorrhizal, and plant pathogen abundance vary in forest ecosystems across China? What affects these large-scale variations? Can leaf δ 15 N and δ 13 C be used as indicators to re ect soil fungal communities in forest ecosystems? To answer these questions, we collected leaf and soil samples from 33 forest reserves along a large range across China to explore the associations between soil fungal communities and leaf C, N stable isotopes using molecular and stable isotope techniques.

Soil and leaf sampling
We surveyed 33 mountain forest reserves located across a broad range of latitudes (21.40 • ∼53.56 ºN) and longitudes (101.03º∼128.52 ºE) in China ( Figure 1 and Table S1). In each forest reserve, [9][10][11][12][13][14][15] sampling plots were randomly chosen along the same aspect of the mountain. Five topsoil samples (within a 5 cm depth) were collected from each plot (5 m apart) and stored in icy sterilized bags. Soils were stored at -20 ºC for molecular experiments. The leaves of the dominant trees species (such as Pinaceae,Betulaceae, and Fagaceae) in each plot were sampled at the same time. Each healthy adult trees was chosen 10 leaves were sampled from each tree was sampled 10 leaves. Low-quality bases with Phred quality scores <20 from the 3' end of the raw reads were removed and then the paired-end reads were merged using FLASH (version 1.2). MOTHUR (version 1.36) was used to demultiplex merged sequences to samples based on their unique barcodes. Operational Taxonomic Units (OTUs) were clustered at the 97% similarity threshold and chimeras were removed by using USEARCH (version 8.0). After removing singletons, a taxonomic name was assigned to each OTU using the Ribosomal Database Project (RDP) classi er with a con dence threshold of 0.

Biogeographic variations
About 420 EMF OTUs and only 15 AMF OTUs were found in the dataset, suggesting that the forests in China are mainly ectomycorrhizal systems. Soil fungal diversity (Shannon index) was found to signi cantly and parabolically vary with latitude and peaked at approximately 40 ºN, which is the boundary between the North and Northeast zones of China (Figure 2A). There was no signi cant relationship between the relative abundance of plant pathogens and biogeographic factors. However, the relative abundance of ectomycorrhizas increased signi cantly with latitude, except for Changbai Mountain, which showed a considerably higher relative abundance of ectomycorrhizas than the other forest reserves ( Figure 2B). Leaf δ 15 N increased signi cantly with latitude ( Figure 2C). The relationship between leaf δ 13 C and latitude was different for the South (tropical and subtropical) and North (temperate and boreal) zones of China (the boundary is at located at approximately 35 ºN). Leaf δ 13 C decreased and then increased signi cantly with latitude, and the lowest point was at approximately 30 ºN in the South, while leaf δ 13 C decreased signi cantly in North China ( Figure 2D).
There was no signi cant relationship between the fungal Shannon index and longitude in the South, but it decreased signi cantly in North China with increasing longitude ( Figure 3A). The relative abundance of ectomycorrhizas was decreased and then increased related to longitude signi cantly, and the lowest point was at approximately 110 ºE, which is the boundary between the West and East zones of China ( Figure 3B). There was no signi cant relationship between leaf δ 15 N and longitude ( Figure 3C). Leaf δ 13 C increased signi cantly with longitude in South China, while there was no signi cant relationship between Leaf δ 13 C and longitude in North China ( Figure 3D).
The fungal Shannon index was signi cantly and unimodally related to mean annual temperature (MAT), and peaked at about 7 o C ( Figure 4A). The relative abundance of ectomycorrhizas decreased signi cantly with MAT, except for Changbai Mountain, which had a considerably greater relative abundance of ectomycorrhizas than the other forest reserves ( Figure 4B). Leaf δ 15 N decreased signi cantly as MAT increased ( Figure 4C). Leaf δ 13 C was decreased and then increased related to MAT signi cantly, and the lowest point was at approximately 17.5 o C in South China. Leaf δ 13 C increased signi cantly as MAT increased in North China ( Figure 4D).
The fungal Shannon index decreased signi cantly as mean annual precipitation (MAP) increased ( Figure  5A). The relative abundance of ectomycorrhizas was decreased and then increased related to MAP signi cantly, and the lowest point was at approximately 1340 mm, except for Changbai Mountain, which had a considerably greater relative abundance of ectomycorrhizas than the other forest reserves ( Figure   5B). Leaf δ 15 N and δ 13 C decreased signi cantly as MAP increased (Figures 5C, D).
Relationships betweenleaf δ 15 N, δ 13 C and soil fungal community Leaf δ 15 N was signi cantly positively correlated with the soil fungal Shannon index ( Figure 6A).
Moreover, leaf δ 15 N was signi cantly negatively correlated with the relative abundance of plant pathogens, but signi cantly positively correlated with the relative abundance of ectomycorrhizas ( Figures   6B, 6C). There was no signi cant relationship between the fungal Shannon index and leaf δ 13 C ( Figure   7A). Leaf δ 13 C was signi cantly and unimodally correlated with the relative abundance of plant pathogens and valleyed at approximately 8.5% ( Figure 7B). Leaf δ 13 C was signi cantly positively correlated with the relative abundance of ectomycorrhizas ( Figure 7C). Ectomycorrhizal relative abundance was signi cantly negatively correlated with plant pathogen relative abundance ( Figure S1). Precipitation in China gradually decreases from Southeast to Northwest. In the wet south, moisture is not the main limiting factor; hence, we found no relationship between the fungal Shannon index and longitude. However, precipitation is an important factor in the dry north. High soil moisture can inhibit rhizomicrobial activity (Song et al. 2018), high fungal diversity can help plants to resist the adverse effects of drought (Fahey et al. 2020), thus causing the fungal Shannon index to decrease signi cantly as MAP increased (Preece et al. 2019). Furthermore, the fungal Shannon index decreased signi cantly with longitude from the dry west to the wet east in North China in the present study. Similarly, the relative abundance of ectomycorrhizas decreased as longitude increased, which may be due to the lack of soil moisture for microbial activity and biodiversity in low MAP zones (Song et al. 2018;Preece et al. 2019;Fahey et al. 2020). As the soil becomes more waterlogged in the east, trees rely more on mycorrhizal fungi to protect their roots against the adverse effects of excessive soil moisture (Erlandson et al. 2016), which explains why the relative abundance of EMF increased as longitude increased in high MAP zones in the present study. For these reasons, the relative abundance of EMF showed a unimodal variation from the dry west to the wet east of China.

Discussion
It should be noted that the relative abundance of EMF samples from Changbai Mountain was much higher than that from other forest reserves, which is probably due to the special geological history of Changbai Mountain. Changbai Mountain is a dormant volcano; the last eruption was in 1702 (Yuan and Sun 1990). After the eruption, substantial amounts of nutrients were left in the soil, which caused the mycorrhizal fungi to proliferate. Furthermore, the combination of positive density dependence and advantages of nutrition utilization may have led to a clustering of EMF seedlings around adult trees and a lack of tree saplings with AMF around conspeci c adult AMF trees ( Leaf δ 13 C re ects the response and adaptation mechanisms of plants to speci c environments, and may be used as an alternative indicator of the long-term water use e ciency of plants (Cernusak et al. 2013; Gautam and Lee 2016; Acosta-Rangel et al. 2018). In the present study, leaf δ 13 C decreased signi cantly with latitude, except at 30º∼37 ºN. This decreasing trend is similar to most studies, which is explained as the phenotypic acclimation of plants to climate (Diefendorf et al. 2010;Sun et al. 2016;Li et al. 2017). The decreasing trend may due to the special ability of plants to adapt to cool and dry environments that occur from 30º to 37 ºN (Du et al. 2015). MAP is the strongest predictor of leaf δ 13 C among global climate variables (Diefendorf et al. 2010). Water de cits might reduce either stomatal conductance or stomatal density, leading to improved water use e ciency and positive leaf δ 13 Kadowaki et al. 2018). Therefore, in our study, the negatively correlation between leaf δ 15 N and plant pathogen relative abundance may be explained by the higher relative abundance of EMF, which enables the plant-microbial system to be more e cient at transferring 15 N-enriched N (Kranabetter et al. 2015). Further, as the transfer e ciency of the root-microbial system is enhanced, niche availability increases, thereby increasing biodiversity in the belowground ecosystem (Bardgett and van der Putten 2014). This may explain why leaf δ 15 N was signi cantly positively correlated with soil fungal Shannon index in the present study.
Leaf δ 13 C mainly re ects the long-term water use e ciency of plants (Cernusak et  Uroz et al. 2016), therefore leaf δ 13 C was signi cantly positively correlated with the relative abundance of EMF. Similarly, plant pathogens are harmful to plants; hence, leaf δ 13 C decreased as the relative abundance of plant pathogens increased. Notably, leaf δ 13 C increased with the relative abundance of plant pathogens when both the relative abundance of plant pathogens and EMF were more than 10%. This is probably due to enhanced adaptability and immunity of the plants and the ability of EMF to protect the plants against pathogens, which may allow the ecosystem to bear more plant pathogens.

Conclusions
In summary, we reported large-scale biogeographic patterns in soil fungal diversity, ectomycorrhizal, and plant pathogen abundance, and subsequently linked their diversity distributions to large-scale leaf δ 15 N and δ 13 C patterns. The large-scale biogeographic patterns of EMF relative abundance suggest that the soil-plant-microbial interactions contribute to the variations in soil fungal diversity, leaf δ 15 N, and δ 13 C across different forest ecosystems in China. Climatic factors, i.e., MAT and MAP, may be affecting these large-scale biogeographic variations. Leaf δ 15 N was mainly affected by the relative abundance of EMF while leaf δ 13 C was mainly affected by the relative abundance of plant pathogens. Leaf δ 15 N and δ 13 C may be indicators for re ecting the soil fungal communities in forest ecosystems; thus, they could be used to predict the component and function of belowground fungal communities.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.  Map showing sampling sites within 33 forest reserves in China. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.