Microbial community analysis of apple rhizosphere around Bohai Gulf

Bohai Gulf is the main area for apple tree cultivation in China. Consecutive replanting significantly affects the yield and quality of apple trees in this area. Microecological imbalance in apple trees’ rhizospheres caused by variation in the soil microbial community is considered the primary cause of apple replant disease (ARD). This study analysed the microbial communities of the rhizospheres of perennial apple trees (PAT) and apple tree saplings under replanting (ATS) around Bohai Gulf using high-throughput sequencing. The results revealed increased populations of typical pathogenic fungi Verticillium and bacteria Xanthomonadaceae, and decreased populations of beneficial bacterial populations Pseudomonas and Bacillus with replanting, suggesting that competition between pathogens and beneficial microbes varies according to the ratio of pathogens to beneficial microbes in rhizosphere soil under the replanting system. Meanwhile, replanting was accompanied by an increase in the antagonistic bacteria Arthrobacter and fungus Chaetomium, suggesting that increased numbers of pathogens can lead to more instances of antagonism. Redundancy analysis (RDA) revealed site position and the main soil properties (pH, organic matter, available N, available K, available P, and moisture) affected the microbial community composition. It found clear differences in soil microbial communities and demonstrated a better understanding of the causes for ARD.

the apple tree's rhizosphere of Bohai Gulf were investigated through high-throughput sequencing. Redundancy analysis was carried out so as to detect the abundance of microbial groups and the interaction between microbial groups and soil environmental properties. Our research compared the microbial community's composition and structure by examining the presence of bacteria and fungi in perennial apple tree (PAT) and apple tree saplings under replanting (ATS) soils of Bohai Gulf to monitor variations after replanting. This was done to provide a basis for reasonable understanding of the rhizosphere microbial community's role in overcoming ARD.

Results
Estimators of the diversity and species richness of microbial communities. All of the raw sequence data obtained was assigned to each sample based on their barcode sequence. A total of 623,094 raw bacterial reads and 524,083 raw fungal reads were obtained from the rhizosphere of PAT and ATS soils. After trimming the barcodes, adapters, and primers and filtering the low-quality sequences, 480,262 bacterial sequences (ranging from 34,291 to 59,734 sequences per sample) and 468,134 fungal sequences (ranging from 38,785 to 59,051 sequences per sample) were used for further analysis.
Using a 3% dissimilarity cut-off for clustering, the reads were grouped into different OTUs. All chimera were removed during the process of clustering and the representative OTUs were obtained to classify the species. Rarefaction curves ( Supplementary Fig. S2) for the microbial community at distance levels of 0.03 had not reached an asymptote, which indicated that the sequence was not sufficient to represent the different bacterial and fungal communities; however, by combining the rarefaction curves with the Shannon diversity index (Fig. 1), the Shannon diversity curves approached a plateau. Therefore, the data were sufficient to allow an analysis of microbial communities.
Soil available K (AK) and available P (AP) decreased significantly in ATS soils compared with PAT soils overall (Table 1). Available N (AN) increased in samples A, B, and C from the Shandong area and decreased in samples D and E. Compared with PAT soils, pH did not differ significantly in ATS soils. No statistically significant difference was observed in organic matter (OM) and moisture between PAT soils and ATS soils. The indices of diversity and richness of bacteria and fungi are shown in Table 2. The depth of all sample sequences was 98% indicating adequate sequence coverage, thus meeting the criteria of our experiments. The ACE and Chao values are indicators of species richness. The Shannon value is an indicator of microbial diversity. The Qixia sample showed the highest microbial community diversity in PAT soil for both bacteria and fungi. The results above showed that soil Figure 1. Shannon-Wiener curves of bacterial (a) and fungal (b) depicting the effect of 3% dissimilarity on the number of OTUs identified in the 10 soil samples. The "A1", "B1", "C1", "D1" and "E1" refers to the five PAT soils in Qixia, Muping, Laizhou, Huludao and Changli. The "A2", "B2", "C2", "D2" and "E2" refers to the five ATS soils in Qixia, Muping, Laizhou, Huludao and Changli, respectively. biodiversity and soil properties differed between sampling locations, and replanting did not have a significant overall effect on microbial diversity and richness.
Comparison of microbial communities from PAT and ATS soils. The relative abundance of different bacterial and fungal groups from the rhizosphere was compared (Fig. 2). For bacteria, the most abundant OTUs belonged to Bacillus, Xanthomonadaceae, and Nitrosomonadacea. By exploring the composition of the ten samples, Bacillus was found to be the dominant group in both PAT and ATS soils. Compared with PAT soils, Bacillus was present in relatively smaller quantities in ATS soils. Xanthomonadaceae had relatively greater population size in ATS soils, especially in Qixia. Nitrosomonadaceae was found in relatively higher amounts in ATS soils from all five sites, and the abundance of Arthrobacter and Pseudomonas in ATS samples was also relatively high from four sites with the exception of Changli. For fungi, the most abundant 10 OTUs accounted for 67.63% of the total number of sequences, belonging to Microascaceae, Geomyces, Chaetomium, Verticillium, and Filobasidiaceae. Microascaceae, Filobasidiaceae, and Verticillium were higher in ATS soils than in PAT soils overall, while Geomyces and Chaetomium were relatively less, especially in Qixia and Changli.

Relationship between microbial community and environmental variables. Redundancy analysis
showed that the relative abundance of bacteria was affected both by growing positions and soil properties (Fig. 3). The abundance of bacteria in D1 and D2 correlated with moisture, AN, and AP, while the abundance of bacteria in both E1 and E2 had no significant correlation with soil properties. These results suggest that the abundance of bacteria of two treatments in D and E had no obvious difference. However, the bacteria abundance in A1 correlated positively with pH and OM, while the abundance in A2 correlated negatively with pH and OM. Oppositely, the abundance in B1 correlated negatively with pH and OM while the abundance in B2 had no significant correlation with pH and OM. The abundance in C2 correlated with AK; the abundance in C1 presented no significant correlation with soil properties. Thus, replanting displayed a significant effect on microbial abundance, especially in areas of Shandong, and less of an effect on other areas around Bohai Gulf.
Redundancy analysis showed that the first and second components explained 78.88% of the total bacterial fungal variations. For bacteria, the first component of the redundancy analysis (RDA1), which explained 50.43%

Discussion
Soil biological activity can be measured through the changes in microbial quantity and species. It is considered that imbalance of the soil microbial community is an important contributor to replant disease. Microbes can affect the growth of crops by affecting the physical and chemical properties of soil and the transportation of nutrients needed for the cultivation of crops. In this study, Illumina MiSeq sequencing analysis of 16S rRNA and 18S rRNA gene sequences revealed that replanting altered rhizospheric microbe diversity and the community structure of apple trees in the Central Bohai area.
In this study, no significant differences among the OTUs were observed at the phylum level between ATS and PAT soils, both in bacteria and fungi. This result is in accordance with the findings by Xu et al. and Jumpponen and Jones, who also investigated soil fungal communities using deep amplicon sequencing. According to the diversity and richness index of the rhizosphere soil, no overall effect was observed in PAT soils compared to ATS soils.
Further study at the level of the family and genus was conducted. Comparisons of PAT and ATS soils revealed differences in the rhizosphere resident bacterial genus. Latz et al. suggested that plant diversity improved protection against soil-borne pathogens by fostering antagonistic bacterial communities, while plant monocultures had a negative impact on the abundance of antagonistic microbes. The present study found an abundance of fungal OTUs belonging to Verticillium, which causes wilt or root rot in many plants 17 . The proportion of Verticillium in ATS soils of B, C, D and E increased 2.89%, 6.73%, 1.31%, and 11.29%, respectively, compared with PAT soils. Its relative abundance significantly increased with continuous farming of apples, suggesting that this pathogen may be the causal agent of ARD in our sample fields. Similarly, when crops such as winter rapeseed and cotton were replanted, a significant increase in the abundance of Verticillium in soil and aggravated root wilt was observed 18 . Previous results have indicated that replanting contributes to the growth of harmful fungi Verticillium 9, 11, 19 , similar to the results shown in this study. Xanthomonadaceae has been reported as a cotton pathogen linked to decline . Redundancy analysis of abundant bacterial (a) and fungal (b) genus and soil properties for PAT and ATS samples from 5 sampling areas. The "A1", "B1", "C1", "D1" and "E1" refers to the five PAT soils in Qixia, Muping, Laizhou, Huludao and Changli. The "A2", "B2", "C2", "D2" and "E2" refers to the five ATS soils in Qixia, Muping, Laizhou, Huludao and Changli, respectively. in cotton growth and soil suppressiveness in replanted soils, also observed in ATS soils 9,20 . Bacillus showed a decline in ATS soils in this study. As reported, it is beneficial for the growth of black pepper, zea, Spartina maritima, and wheat 7 ; additionally, Bacillus spp. is known to be a plant-beneficial rhizobacteria with the potential to combat phytopathogens through the release of active secondary metabolites. In accordance with our previous study results, compared with PAT soils, the microbial community in the rhizosphere of ATS soils was more conducive to pathogenic microorganisms and less beneficial to microorganisms, which may be the root cause of poor growth of apple trees in ATS soils.
It is worth noting that some beneficial microbes, such as Arthrobacter and Chaetomium, presented an increasing trend in ATS soils in Muping, Laizhou, and Huludao. Arthrobacter act as antagonistic bacteria against Sclerotinia sclerotiorum 21 , which can cause sunflower sclerotinia rot 22 . Arthrobacter also show a significant negative correlation with tobacco bacterial wilt disease 23 . The proportion of Arthrobacter bacteria has had significantly increased in ATS soils of the Bohai Gulf. Chaetomium can be used for controlling disease in other plants 24 . By acting as a biocontrol fungi against plant pathogenic bacteria, it also has the ability to decompose palm-oil mill fibres 25 . Nitrosomonadaceae and Pseudomonas act as beneficial bacteria and also present an increasing trend in ATS soils. A previous study found that the addition of Nitrosomonadaceae could reduce nitrogen loss, as well as the time required to stabilise the nitrogen profile 26 . Additionally, Nitrosomonadaceae is useful for biotechnological processes such as bioremediation of toxic chemicals in the soil 27 . High-throughput sequencing in this study confirmed an increase in the abundance of Pseudomonas. Root-associated Pseudomonas species are well known to produce metabolites that can limit the growth of other microorganisms 28 . There are reports of both beneficial and harmful species of Pseudomonas 29-32 , however, the majority of previous studies have found that Pseudomonas spp. have the potential to control specialised pathogens and suppress soil-borne diseases 30,31 . The increase of these beneficial microbes in ATS soils may suggest that populations are more competitive than other species, allowing them to prevail in ATS soils. A reasonable explanation of the above results is that beneficial microbes existing in soil may act as antagonists inhibiting the harmful microbes. Therefore, the larger population of pathogens, e.g. Verticillium in ATS soils, may result in more intense competition among some antagonistic microbes. Thus, promoting beneficial microbes acting as antagonists may warrant further investigation.
Above all, competition, antagonism, and hyperparasitism are generally the main interactive mechanisms between pathogenic fungi and antagonists. Some beneficial microorganisms act as potential antagonists of plant pathogens present in the soil. Many previous findings have demonstrated that when attacked by a root pathogen, plants can alter their specific rhizosphere microbial populations for protection 33 . Thus, our results suggest that the reduction of apple growth in ATS soils may arise from the decline in soil populations of beneficial microorganisms and an increase of pathogens in ATS soils.
Soil microorganisms not only have mutual effects on each other but also interact with the surrounding environment. Mounting evidence indicates that root exudates initiate and modulate the dialogue between the plant and soil microbes, composed of both pathogenic and beneficial microbes 34 . With the replanting of apple trees, the characteristics of rhizosphere soil changes, causing shifts in the characteristics of root exudates due to apple physiological changes. This may systemically affect interactions between plants and microbes by decreasing the abundance of rhizosphere microorganisms' taxa. In accordance with our previous study results, the soil AP and AN decreased significantly, that may have been caused by inhibition of the phosphate-solubilizing and potassium-solubilizing microbes' activities. In this study, further understanding of the mechanism of microbial activity was described by redundancy analysis, revealing a significant correlation between soil properties and the abundance of some microbes. The apple tree's growth position was associated with the abundance of soil microbes. Because the abundance of Acidobacteriaceae correlated significantly with soil moisture and AP, it was inferred that Acidobacteriaceae probably can dissolve phosphorus in the soil. The abundance of Anaerolineaceae significantly correlated with pH and soil OM. It was inferred that Anaerolineaceae may require particular pH levels; moreover, previous studies have found that anaerobic bacteria could promote the decomposition of OM in soil. For fungi, redundancy analysis indicated that the abundance of Verticillium had a positive correlation with OM. In future studies, it may be possible to control the content of the pathogenic Verticillium by manipulating the content of OM. The present study determined that different biocontrol microbes have different resistance mechanisms, and that some are affected by a single mechanism while others are affected by a variety of mechanisms. These findings suggest that with continued heavy apple cultivation, soil microbial communities may be transformed from a biological system suitable for apple tree growth to a biological system that promotes ARD.

Conclusion
In conclusion, our research revealed that the ARD closely associates with the variation in micro-ecology in the rhizosphere's ecosystem. High-throughput sequencing of PAT and ATS soils revealed that the bacterial and fungal community composition did not differ significantly at the phylum level. However, further analysis showed that the apple replanting system caused changes in the structure of the microbial community, ultimately resulting in the deterioration of microbial structures in rhizosphere soil, with fewer beneficial microorganisms and more pathogenic microbes. This imbalance in the rhizosphere's microecological system leads to serious replanting problems in a continuous cropping system. Meanwhile, the antagonistic microbes play an important role in shaping the rhizosphere's microbial community, as a greater population of pathogens in ATS soils may result in more intense competition by some antagonistic microbes which open several new options related to these newly discovered involvements in ARD outside of the relationship between microbial communities and environmental variables discussed in our research. This finding provides a clue to open a new avenue for modulating the root microbiome to enhance crop production and sustainability. . Soil samples were randomly collected using the five-point sampling method. From each point, 1 kg of soil was collected within the root zone at a depth of 10-30 cm. Soils of the same type collected in the same plot were thoroughly mixed as one sample, placed in a sterilised bag, immediately placed on ice, and transported to the laboratory, where it was stored at −20 °C until physicochemical parameters measurement and DNA extraction. The physicochemical parameters of the soil were measured as follows. The oven dry-weight method allowed the estimation of soil moisture content. Soil pH was determined using a glass electrode meter (Sartorius PB-10, Sartorius scientific Instruments (Beijing) Co., Ltd.) in a suspension of 1 g of soil in 5 mL of distilled water. AP was extracted using sodium bicarbonate and then measured by the molybdenum blue method. AK was determined by flame photometry 32,35 . AN was determined by potassium persulfate oxidation. OM content was determined as described by Walkey and Black 36, 37 . DNA extraction and PCR amplification. Microbial DNA was extracted from 0.5 g soil samples using the Power Soil DNA kit (MoBio Laboratories, Solana Beach, CA), following to the manufacturer's protocols 38 . The V3-V4 regions of the bacterial 16S rRNA gene were amplified by polymerase chain reaction (PCR) (95 °C for 2 min, followed by 25 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s, with a final extension at 72 °C for 10 min) using the primers 39 338 F 5′-barcode-ACTCCTACGGGAGGCAGCA-3′ and 806 R 5′-barcode-GGACTACHVGGGTWTCTAAT-3′ 40 . The barcode is an eight-base sequence unique to each sample. The fungal 18S rRNA gene was amplified using PCR (98 °C for 2 min, followed by 35 cycles of 98 °C for 15 s, 56 °C for 30 s, and 72 °C for 40 s, with a final extension at 72 °C for 10 min) using the primers SSU0817F 5′-barcode-TTAGCATGGAATAATRRAATAGGA-3′ and SSU1196R 5′-barcode-TCTGGACCTGGTGAGTTTCC -3′ 41  Processing and analysing of sequencing data. Raw FASTQ files were de-multiplexed and quality-filtered using QIIME(version 1.17) with the following criteria: (i) The 300 bp reads were truncated at any site with an average quality score of <20 bp over a 10 bp sliding window, and truncated reads shorter than 50 bp were discarded; (ii) exact barcode matching, two nucleotide mismatch in primer matching, and reads containing ambiguous characters were removed; and (iii) only overlapping sequences longer than 10 bp were assembled according to their overlapped sequence. Reads that could not be assembled were discarded. Operational taxonomic units (OTUs) with 97% similarity cutoff were clustered using UPARSE (version 7.1), and chimeric sequences were identified and removed using UCHIME.
Rarefaction analysis based on Mothur v.1.21.1 42 was conducted to reveal diversity indices, including the Chao, abundance-based coverage estimator (ACE), and Shannon diversity indices. The beta diversity analysis was performed using UniFrac to compare the results of the principal component analysis (PCA) using the community ecology package, R-forge. Venn diagrams were implemented and Mantel test, redundancy analysis, and Heatmap figures were performed in Vegan packages in R.
After removing primers, all sequences were aligned with the Silva-ARB database. After chimeras were removed, all sequences were clustered into OTUs at a 97% identity threshold using the unsupervised Bayesian clustering algorithm CROP 43 , and all OTUs were included in downstream analyses. Shannon-Wiener curves of all the samples at a dissimilarity of 0.03 were generated. Rarefaction analysis suggested that the sequences sampled were sufficient to capture the total richness at a genetic distance of 0.03. To estimate the final ranking, we calculated the ACE and the estimated asymptotic microbial taxon richness Chao. After an identical number of reads were subsampled from each sample, microbial diversity was compared between PAT and ATS plants based on the calculated Shannon index. A graphical representation of the relative abundance of microbial diversity was created using Origin (version 8.0).
To ensure the data met the assumptions of normality, soil AN, soil AP, soil AK, soil moisture, and soil OM and pH were tested. Finally, redundancy analysis was used to identify the effect of environmental factors on bacterial and fungal community composition, based on the relative abundance of the detected phyla in each sample, with CANOCO 5.0 software.