Long-term Effects of Mixed Management on Arbuscular Mycorrhizal Fungal Community of Root and Soil in Juglans mandshurica Plantation


 Background: The establishment of mixed plantations is an effective way to improve soil fertility and increase forest productivity. Arbuscular mycorrhiza (AM) fungi are a kind of obligated symbiotic fungi, which can promote plants absorption of mineral nutrients and regulate intraspecific and interspecific competition. However, the effects of mixture on the community structure and abundance of AM fungi are still unclear. The Illumina MiSeq sequencing technique was used to determine the AM fungal community in the roots and soils of pure and mixed plantations (Juglans mandshurica × Larix gmelinii). The objective is to compare the response from root and rhizosphere soil AM fungal communities of Juglans mandshurica to long-term mixed plantation management. Results: Glomus and Paraglomus were the dominant genus in root samples, which accounted for more than 80% of the sequences. Compared with the pure plantation, the relative abundance of Glomus was higher in mixed plantation. Glomus, Diversispora and Paraglomus accounted for more than 85% of the sequences in soil samples. The relative abundance of Diversispora and Unclassified_c__Glomeromycetes were higher and lower in pure plantation, respectively. Samples of the Root_P (roots in pure plantation) had the highest number of unique OTUs (Operational Taxonomic Units), which were mainly composed of sequences belonging to unclassified_c__Glomeromycetes, Paraglomus, Glomus and Acaulospora. The amount of unique OTU detected in soil in pure and mixed plantation was relatively close. In the same types of sample (whether root or soil), the forest type did not have significant effect on AM fungal diversity, but the Sobs, Shannon, Chao1 and Ace indices of AM fungi in the roots were significantly higher than those in the soil. Conclusions: The mixed forest management had little effect on the AM fungal community in the root of Juglans mandshurica, but significantly changed the community composition of the soil AM fungi rather than diversity.


Background
Arbuscular mycorrhizal (AM) are widely spread and form symbiotic associations with about 80% of terrestrial plant species [1]. AM fungi can not only promote plant absorption of mineral nutrients (C, N, P) [2][3][4], provide resistance to environmental stress [5,6] and regulate intraspeci c and interspeci c competition [7], but also directly and indirectly improve soil structure and affect the circulation of matter and ow of energy in the ecosystem [8]. Some studies have found that host plants preferentially distribute carbohydrates to more bene cial symbionts while they provide photosynthetic products to AM fungi [9][10][11], and the preference/selection of different plants species leads to the difference in the growth rate of AM fungi, thus forming different AM community richness, composition and diversity [12].
Previous studies have indicated that AM fungal taxonomic groups differ in terms of the main propagule form for colonizing new roots and the allocation of biomass to the compartments of roots and soil [13].
To a certain extent, the composition and community structure of AM fungi depends on their propagation forms (spores, infected root segments, extensive extraradical mycelium), and the difference between the propagation of AM fungi in root (intraradical) and soil (extraradical) samples may be related to the location and density of AM fungal propagules [14]. In addition, host plant selection for AM fungi exhibited strong selection pressure for AM fungi in the root system [15]. Moreover, the response of different AM fungal groups to host plants or the rhizosphere microenvironment may also lead to host plant selection preferences for some AM fungal species [16]. There are differences in the distribution patterns of root carbon and the pathway to secondary metabolites or secretions, which may cause changes of soil environmental conditions [9,17]. A recent study found that AM fungal structures in root and soil have different responses to biotic and abiotic factors. The AM fungal community structure in roots is mainly affected by host plants and disturbances (grazing), while AM fungi in soil is greatly affected by environmental factors [18].
AM fungi could form a huge mycorrhizal network and connect individual plants within the community which facilitate the transport of nutrient resources between plants [19]. Therefore, neighboring plants could change the impact on host plants on AM fungi or form speci c AM fungal communities when multiple plants are mixed planting in terrestrial ecosystems [12,20]. Previous studies have found that the construction of AM fungal communities is in uenced by the identity of adjacent plants which mostly occur to greenhouses and invasive systems [21]. However, it is not clear how coexistence of multiple plants affects arbuscular mycorrhizal fungal communities in natural ecosystems, especially in forest ecosystems. Forests play an important role in the production of wood and fuel, controlling soil erosion and maintaining ecosystem functions [22]. Monoculture area accounts for 80% of the total planting area in China. The establishment and management of monoculture is relatively easier compared with mixed plantation, but it will reduce the ecological function of the forest, and long-term planting of monoculture will cause litter quality and soil fertility decline and other problems [23]. Mixed plantations building is an effective way to improve soil fertility and increase forest productivity [22,24]. Numerous studies have shown that rational mixed plantations can improve soil fertility [25], nutrient cycle [25], stand productivity [26,27], tree nutrition and resistance to pests and diseases [28]. However, the mixed effects vary from tree species [29,30]. Therefore, comparing the differences of AM fungal communities in different plantation types is conducive to a profound understanding of the stimulation mechanism of mixed plantations.
Larix gmelinii is a major afforestation and fast-growing species in north China [31]. There always had some problems, such as biodiversity loss, soil degradation occurs to rapid development of larch monoculture [32], which seriously affect the sustainable management. Juglans mandshurica (arbuscular mycorrhizal species) is one of the precious timber trees species in northeast China and has important economic value. It has been reported that the mixed management of Juglans mandshurica and Larix gmelinii can improve the soil fertility and stand productivity [33], however, until now, the synergistic mechanism of mixed management, especially the interaction between the host plant and soil AM fungal community structure is still obscure. For this purpose, we compared the AM fungal community composition, structure and diversity of root and soil of Juglans mandshurica of pure and mixed plantations in order to provide theoretical basis for the stimulation mechanism of temperate mixed plantations. We hypothesized that (1) mixed management signi cantly changed the AM fungal community structure and composition in the root and soil of Juglans mandshurica, and have higher AM fungal diversity and richness were higher in mixed plantation compare with pure plantation; (2) in all two forest types, the diversity and richness of AM fungi in the soil are higher than those in the roots.

Results
Soil properties and AM fungal colonization Soil pH and total phenol content was signi cantly higher in pure plantation than in the mixed plantation (Table 1; P < 0.05). The NH 4 + -N content was signi cantly higher in mixed plantation than in the pure plantation. Compared with mixed plantation, soil moisture, P, C/N ratio, N mic , CPh, WSPh and colonization mainly showed downward trends in pure plantations, although no signi cant differences were found (P > 0.05).  Fig. S1), which also indicated that the sequencing depth was adequate for assessing the diversity of AM fungal communities of all samples.

Am Fungal Community Composition
Across all root samples, nine AM fungal genera were detected. Glomus and Paraglomus were the dominant genus, which accounted for more than 80% of the sequences (Fig. 1A). Compared with the pure plantation, the relative abundance of Glomus in root sample was higher in mixed plantation (Fig. 1A). Across all soil samples, ten AM fungal genera were detected. Glomus, Diversispora and Paraglomus accounted for more than 85% of the sequences (Fig. 1B). The relative abundance of Diversispora and Unclassi ed_c__Glomeromycetes were higher and lower in pure plantation, respectively. At the OTU level, 178 AM fungal taxa were detected. OTU79 and OTU76 were the dominant OTU and had mean relative abundances of 12.15% and 6.54%, respectively (Supplementary Table S2, Fig. 2).

Am Fungal Community Structure And Diversity
The number of Sobs, Shannon, Chao1, Ace and Faith PD indices of root samples were higher in pure plantation ( Fig. 3A-3F; P > 0.05), while the Sobs, Shannon, Chao1 and Faith PD indices of soil samples were higher in mixed plantation (P > 0.05). The Simpson index of soil sample was signi cant higher in pure plantation than in mixed plantation (P < 0.05). In the same types of sample (whether root or soil), the forest type did not have signi cant effect on AM fungal diversity, but the Sobs, Shannon, Chao1 and Ace indices of AM fungi in the roots were signi cantly higher than those in the soil (P < 0.05).

Shared And Unique Otus
Venn diagram analysis of OTUs at 97% sequence similarity showed that all samples shared 22 OTUs, which accounted for 7.86% of the total OTUs observed ( Fig. 5). At the genus level, these shared OTUs were mainly composed of sequences belonging to Glomus, Paraglomus and Diversispora (Table S4). Samples from the Root_P had the highest number of unique OTUs, which were mainly composed of sequences belonging to unclassi ed_c__Glomeromycetes, Paraglomus, Glomus and Acaulospora (Table  S5). The amount of unique OTU detected in soil in pure and mixed plantation was relatively close. At the genus levels, the Soil_P had signi cantly higher relative abundances of Diversispora than other treatments ( Fig. 6, P < 0.05).

Relationships Between Am Fungal Communities And Soil Characteristics
The RDA analysis and Mantel test were conducted to identify the key drivers of AM fungal community structure. In the RDA plots of both AM fungal community structure, soil NO 3 − -N, C mic and pH appeared to be the most important soil characteristics in controlling the root AM fungal community structure, which represented major variations among microbial communities (Fig. 7A). Mantel test demonstrated that root AM fungal community structure was signi cantly correlated to NO 3 − N (R 2 = 0.906, P = 0.029) and C mic (R 2 = 0.881, P = 0.044) ( Table 2). The soil complex phenol (CPh), C/N, pH, NO 3 − -N and NH 4 + -N had longer arrows than the others (Fig. 7B). It indicates that these variables have a greater impact on the AM fungal community. Of all the soil characteristics tested, C/N (R 2 = 0.864, P = 0.032) and CPh (R 2 = 0.994, P = 0.019) were correlated with community composition signi cantly (Table 2).

Discussion
Response of AM fungal diversity, composition and community structure on mixed management The core species composition of AM fungal communities was very similar and conservative in pure and mixed plantations (Glomus mainly, Fig. 1), which is consistent with the result of Senés-Guerrero and Schüßler [34]. They found that there is a conservative core species AM fungal community structure both at different stages of plant development and under different environmental conditions in the Andean ecosystem. Plant community diversity and AM fungal community interact to improve the diversity of AMF community [15]. Van der Heijden, et al. [35] believed that the change of host plants caused by the infection of a single mycorrhizal species was an important factor determining the species composition and diversity of plant community. However, some researchers have reported that there is a negative correlation between plant diversity and AM fungal communities [36]. In this study, the mixture reduced AM fungal community diversity and OTU number in root of Manchurian walnut (Fig. 3, P > 0.05). The lower diversity of AM fungi might have been attributable to a lower "carrying capacity" of fewer walnut roots than in the pure plantation. In mixture, larch roots possessed a greater plasticity in traits related to resource uptake than walnut roots [37]. In addition, larch root exudates could alleviate autotoxic effect caused by juglone, which secreted by Manchurian walnut after mixture. Our previous studies have found that larch root exudates promoted the variation of soil microbial communities of Manchurian walnut and improved soil invertase and urease activities [38]. Salahuddin, et al. [37] indicated that the ratio of root tip tissue of Manchurian walnut signi cantly increased after mixed with larch, but the mycorrhizal infection rate was inhibited. Achatz, et al. [39] found that mycorrhizal mycelia could promote the migration and transport of juglone secreted by plant roots and mycorrhizal could enhance interspeci c interaction thought the study on the species of the genus Juglans (Juglandaceae). AM fungi has certain host speci city [40]. Mummey and Rillig [20] found that spotted knapweed as an invasive plant could signi cantly change the AM fungal community in the invasive site.
AM fungal community composition and abundance were obviously different between pure and mixed plantations ( Fig. 2 and Fig. 6), and there was more OTUs in the soil of mixed plantation (Fig. 5). The reasons may be as follows: (1) Compared with pure plantation, mixed plantation has more advantages in terms of litter quantity and quality, soil nutrients and stand structure, etc. The change of soil microenvironment leads to the variation of AM fungal community; (2) larch affects the root system of Manchurian walnut after mixture, which reshapes the mycorrhizal network. [41] found that abiotic factors (soil moisture content, nitrate and soil enzymes, etc.) had a greater impact on the soil AM fungal community when Robinia pseudoacacia mixed with Platycladus orientalis. The growth of some plants may change soil quality or other abiotic characteristics, which may lead to changes in rhizosphere AM fungal communities [42].

Comparison of AM fungal communities in root and soil samples
Our study found that the AM fungal community in the root system of Manchurian walnut was signi cantly different from that in rhizosphere soil, which is consistent with previous results on temperate steppe [43], temperate farmland [44] and Mediterranean shrub community [45]. Due to plants will have different degrees of speci c selection on AM fungi according to their own nutritional requirements in different growth and development stages, which will inevitably lead to differences in the relative abundance, species and quantity of AM fungi in soil and roots. In addition, there are great differences between the environmental conditions of AM fungi in plant roots and soil. The living environment of AM fungi in roots is mainly regulated by the physiological activities of individual plants, while AM fungi in soil are mainly affected by external environmental conditions.
Many scholars believe that the AM fungal community in soil represents a species pool, which plants can freely recruit some species, that is, the AM fungal richness in soil is the highest [46][47][48][49]. However, some scholars have found that the abundance of AM fungi in roots and soil is the same [50][51][52] or the AM fungi in the root system is more abundant than the soil [53,54]. In this study, the amount of OTU in roots was higher than that in soil samples (whether in pure or mixed plantations), and a large proportion of AM fungi were detected in roots. There are many potential methodological and biological explanations for differences in AM fungal community, especially the small amount of OTU detected in soil. Firstly, the biomass of AM fungi in soil is an order of magnitude lower than that of roots [55]. Therefore, the concentration of soil AM fungi DNA is relatively low. The AM fungi in root samples have higher sequencing depth than soil samples, and low DNA template quantity may lead to the underestimation of species richness.
Secondly, there are differences in the distribution of AM fungi from different taxa to the root (intraradical) and soil (extraradical) samples [56]. Compared with the internal structure, Glomeraceae spend less on external structure [57]. In this study, the Glomus of the roots showed a higher abundance than the soil, which was consistent with previous research results. Some studies have found that Gigasporaceae and Acaulosporaceae could produce a large number of external hyphae compared with internal structures. In this study, Diversisporaceae and Gigasporaceae were mainly found in soil, with few sequence numbers in the root system which is consistent with the early ndings that Diversisporaceae and Gigasporaceae are poor root colonizers [43,54,57]. Paraglomeraceae was mainly found in soil samples, with only few in the roots, which is consistent with previous studies [43,54].

Conclusions
Our study showed that long-term (almost 30 years) mixed management had little effect on the AM fungal community in the root of Juglans mandshurica, but signi cantly changed the community composition of AM fungi in soils rather than diversity. Samples from the root in pure plantation had the highest number of unique OTUs. The core species composition of AM fungal communities was very similar and conservative in pure and mixed plantations. In the future, the combination of root traits and mycorrhizal symbiosis should be considered to comprehensively evaluate the mechanism of nutrient absorption and utilization of mixed tree species, in order to lay a foundation for sustainable management of plantations.

Study area and sample design
The study site was located in Maoershan Forest Research Station (127°30′-127°34′E, 45°21′-45°25′N) of Northeast Forestry University, Heilongjiang province, China. This area is characterized by a continental monsoon climate of a windy spring, a warm and humid summer, and a dry and cold winter. Mean annual temperature is 2.8℃, with the minimum temperature in January (-40.9℃) and maximum temperature in July (34.2℃). The frost-free period uctuates between 120 and 140 days. Annual precipitation ranges from 600 to 800 mm. Soils are Hap-Boric Luvisols [58] with high organic matter content and welldeveloped horizons, and are well drained.
In  Table  S1. Voucher specimen of J. mandshurica and L. gmelinii were not deposited in this study since they are the most common trees in Northeastern China.

Sample Collection
In April 2016, three random sampling plots (20 m × 30 m 0.06 ha) were respectively selected for pure and mixed plantation described above, which were identi ed to serve as replicates. Rhizosphere soil and plant root samples were collected in July 2016. In each of these experimental plots, root samples were collected in 0-10 soil layers from nine individuals of Juglans mandshurica and mixed as a composite sample of each plot. The rhizosphere soils were sampled adjacent to the roots and brushed off from the plant root systems. The soil and root samples were packed in an ice box and transported to laboratory. Soil samples were sieved (1 mm mesh) to remove roots and debris, and subsamples were stored at -80℃ for DNA extraction. The rst three root orders of the roots of Juglans mandshurica are infected by mycorrhiza [59]. Root samples were washed using distilled water to remove soil particles and were stored at -80℃ for DNA extraction. for annealing at 55 °C, and 45 s for elongation at 72 °C, and a nal extension at 72 °C for 10 min. The procedure of the second round PCR reaction was same as the rst round, except for the cycle number, which was 35 [60].

Illumina Miseq Sequencing
The resulted PCR products were extracted from a 2% agarose gel and further puri ed using the AxyPrep

Processing Of Sequencing Data
Raw fastq les were demultiplexed, quality ltered by Trimmomatic and merged by FLASH with the following criteria: (i) The reads were truncated at any site receiving an average quality score < 20 over a 50 bp sliding window. (ii) Primers were exactly matched allowing two nucleotide mismatching and reads containing ambiguous bases were removed. (iii) Sequences whose overlap longer than 10 bp were merged according to their overlap sequence.
Operational taxonomic units (OTUs) were clustered with 97% similarity cutoff using UPARSE(version 7.1 http://drive5.com/uparse/) and chimeric sequences were identi ed and removed using UCHIME. The taxonomy of each 18S rRNA gene sequence was analyzed by blast against the MaarjAM database using con dence threshold of 97%.

Soil Physicochemical Analyses
Total carbon (C) and total nitrogen (N) were measured by a Macro Elemental Analyzer (vario MACRO, Elementar Co., Germany), and total phosphorus (P) was determined colometrically with a UV spectrophotometer (TU-1901, Puxi Ltd., Beijing, China) after wet digestion with HClO 4 -H 2 SO 4 . Soil pH was measured using a pH meter (MT-5000, Shanghai). Soil nitrate-N (NO 3 − -N) and ammonium-N (NH 4 + -N) were extracted in 2 M KCl and measured using a continuous-ow ion auto-analyzer (Scalar SANplus segmented ow analyzer, The Netherlands). Soil total phenol content was measured by the ultraviolet spectrophotometer method [61]. The Folin reagent colorimetric method was used to determine soil watersoluble phenol and complex phenol content [61]. Soil microbial biomass carbon (MBC) and nitrogen (MBN) was measured using a chloroform fumigation extraction method [62]. The mycorrhizal colonization rate of ne roots was determined according to method of Guo [59].

Statistical analysis
For the Illumina MiSeq sequencing data, the Alpha diversity indices (OTU number of observed, Chao1, ACE, Faith's PD, Shannon and Simpson diversity indices) were generated using QIIME [63]. For beta diversity analysis, Bray-Curtis distances were calculated and Principal coordinates analysis (PCoA) was conducted to visualize the community similarity using by 'vegan' package in 'R' (Version 3.       Relative abundances of main AM fungal genera in root (A) and soil (B) samples.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.