Next Article in Journal
Effects of Transplantation and Microhabitat on Rhizosphere Microbial Communities during the Growth of American Ginseng
Previous Article in Journal
Effects of Bacillus subtilis HS5B5 on Maize Seed Germination and Seedling Growth under NaCl Stress Conditions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agronomy
Volume 13
Issue 7
10.3390/agronomy13071875
agronomy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

PacBio Sequencing Unravels Soil Bacterial Assembly Processes along a Gradient of Organic Fertilizer Application

by
Wenhui Wang
1,2,
Yuan Gao
1,
Na Li
1,
Hongmei Lu
1,
Ranxiang Lan
2 and
Xungang Gu
1,*
1
College of Resources and Environment, Anhui Agricultural University, Hefei 230036, China
2
School of Life Sciences, Anhui Agricultural University, Hefei 230036, China
*
Author to whom correspondence should be addressed.
Agronomy 2023, 13(7), 1875; https://doi.org/10.3390/agronomy13071875
Submission received: 25 May 2023 / Revised: 12 July 2023 / Accepted: 13 July 2023 / Published: 15 July 2023
Browse Figures
Review Reports Versions Notes

Abstract

:
The application of organic fertilizer is an important agricultural practice for improving soil health and the soil microflora, and the microbial community assembly process relating to this application is also closely associated with soil health. However, the effects of organic fertilizer intensification on the bacterial community assembly processes of farmland soil are often overlooked. In this study, bacterial community structure, ecological networks, and bacterial community assembly processes were evaluated using the investment soil-cultivation test and PacBio sequencing. The PCoA, Mantel test, and Procrustes analysis showed that overfertilization changed soil physicochemical properties and caused significant succession of soil bacterial communities (p < 0.05). The neutral community model indicated that the spread of bacteria in the low-fertilization group was greater than that in the high-fertilization group. Under conditions of overfertilization via organic fertilizer (organic matter ≥ 50% and N-P2O5-K2O ≥ 5%), the bacterial network topology and stability of nutrient-rich loess brown (H) soil were improved compared with those of red (R) soils, and the slope of the robustness analysis displayed a 10.9% decrease in H soil and a 37.2% decrease in R soil. The inference of community assembly mechanisms via phylogenetic-bin-based null model analysis (iCAMP) confirmed that with increasing fertilization, the relative importance of ecological drift gradually increased, and the importance of homogeneous selection was reduced (p < 0.01, permutational ANOVA). A total of 103 bins (in the selected top 200 bins) of the dominant process were different between the H and R soils. The results clarified that homogeneous selection and drift were the dominant processes driving the assembly of bacterial communities in different soil types along the gradient of organic fertilizer application and confirmed that excessive fertilization enhanced the relative importance of drift among the construction mechanisms. Changes in soil construction mechanisms due to overfertilization are related not only to soil type but also to different microbial lineages.

1. Introduction

Fertilization is crucial for the sustainable use of agricultural systems and a key factor in soil bacterial community assembly processes. Fertilization directly affects soil pH and soil nutrients, and it is the main factor causing changes in the soil microbiome. Previous studies have shown significant changes in the structure of bacterial and fungal communities under excessive fertilization, which also appear to be directly associated with a decrease in soil pH [1]. In contrast, our previous study showed that the overuse of organic fertilizer (1.5 times) did not significantly alter the community structure of myxobacteria in red soil [2]. This suggest that the effects of excessive fertilization on community processes differ among microbial lineages. However, the community assembly process of individual microbial lineages along fertilization gradients has often been overlooked.
With the development of third-generation sequencing technology, full-length 16S rRNA has returned to the public eye, playing a more important role in identification studies of microbial mechanisms. Second-generation sequencing yields less species-level information, and the disadvantage is increasingly prominent, especially in microbial ecology research, which is becoming increasingly accurate. Therefore, experimental verification is needed. There is no doubt that the full 16S gene provides better taxonomic resolution [3]. A large amount of experimental data at the microbial species level can be obtained quickly using the PacBio and Oxford Nanopore sequencing platforms [3].
Understanding the composition of bacterial communities at the species level and defining the relative importance of deterministic and stochastic processes in controlling community diversity and distribution are still important issues in microbial ecology [4]. There is a consensus among ecologists that deterministic and stochastic processes explain the assembly of microbial communities [5]. Under determinism, microbiomes are shaped by deterministic nonbiological factors (environmental factors such as pH, temperature, etc.) and biological factors (species interactions such as competition, predation, etc.) due to the different habitat preferences and adaptabilities of microorganisms [6]. Under stochasticity, random processes, such as birth, death, migration, species formation and finite diffusion, shape the structure of the microbiome [7]. The neutral community model (NCM) proposed by Sloan et al. [8] is particularly useful in quantifying the importance of neutral processes. NCMs have proven that occasionality and migration are important forces in the formation of microbial communities [9]. However, they do not explicitly represent deterministic processes and therefore do not fully describe the community assembly mechanism [8].
The Quantifying assembly Processes based on Entire-community Null model analysis (QPEN) method was developed to obtain quantitative information on community assembly processes [10]. This statistical method is a major advancement in microbial ecology and can generate quantitative information about community assembly deterministic and randomness processes. Ning et al. [4] reported a powerful framework (R package) for quantitatively inferring community assembly mechanisms via phylogenetic-bin-based null model analysis (iCAMP) based on the turn-over of individual bins across microbial communities. The iCAMP framework provides a more sophisticated approach for obtaining quantitative information on five community assembly processes, namely, heterogeneous selection (HeS), homogeneous selection (HoS), dispersal limitation (DL), homogenizing dispersal (HD), and drift (DR) [4]. The iCAMP framework focuses on phylogenetic bin-based assembly processes, exploring the role of different ecological processes at a more detailed biological level (species or gene type), and distinguishes different ecological processes in terms of their importance for different microbial lineages.
The application of organic fertilizers is a key anthropogenic factor driving changes in soil microbial diversity in farmland, but the effect of organic fertilizer intensification on the bacterial community construction mechanism is unclear. Commonly used organic fertilizers include composted animal manure, compost, sewage sludge, food processing wastes, and municipal biosolids [11]. Organic fertilizers often directly activate soil microbial biomass, gene abundance, taxonomic diversity, and respiratory activity [12]. The application of organic fertilizers has long-term effects on soil microbial communities. However, the duration of this effect largely depends on the dose of fertilizer applied, as well as the period for which the fertilizer is applied [12]. The main scientific questions of this study are as follows.(1) Are the changes in soil physicochemical properties caused by multigradient fertilization consider the main factors driving microbial community succession? (2) Along a fertilization gradient, are the ecological processes that dominate the assembly of bacterial communities in different soil types consistent, and how does the importance of these dominant ecological processes change with the increase in fertilization? (3) Are these quantified community assembly processes that dominate similar phylogenetic relationships of bacterial lineages along fertilization gradients consistent, and how do they relate to soil type?

2. Materials and Methods

2.1. Experimental Design and Soil Sampling

Two typical farmland soils in the middle and lower reaches of the Yangtze River were selected for follow-up experiments. The two experimental soils were collected in May 2020 from Anhui Province; both were collected from areas with a subtropical monsoon climate. The loess brown soil (H soil) was taken from the outskirts of Hefei City (30°48′ N, 114°36′ E) and had moderate nutrients. This sampling site had an average temperature, precipitation, and sunshine duration of 20.0 °C, 744 mm, and 792 h, respectively. The red soil (R soil) from a nutrient-poor tea garden was collected in Dongzhi County (29°63′ N, 116°85′ E), Chizhou City, with corresponding conditions of 20.0 °C, 744 mm, and 792 h. One hundred kilograms of soil was collected at a depth of 20 cm, air-dried, and sieved (1 cm × 1 cm) to remove plant and other debris. The physical and chemical properties of the experimental soil are listed in Table S1. The test organic fertilizer was provided by Anhui Huishang Agricultural Fu Co., Ltd. (Hefei, China), with organic matter ≥ 50% and N-P2O5-K2O ≥5%.

2.2. Indoor Soil Cultivation Test

An indoor soil cultivation test was implemented to study the effect of the fertilization gradient on bacterial assembly processes. With reference to the organic matter content of the experimental soil, a gradient comprising 5 different fertilization treatments was designed: CK, no fertilization; T1, 3.33 g/kg; T2, 6.67 g/kg; T3, 13.33 g/kg; and T4, 20 g/kg. T1 and T2 were defined as low-dose fertilization treatments, and T3 and T4 are high-dose fertilization. Each set of treatments also consisted of 6 basins of parallel experiments, and each basin was filled with 1 kg of soil. For the fertilization treatments, soil was mixed with organic fertilizer and added to the basins, and the water content was then adjusted with distilled water to 70% of the field water-holding capacity. The samples were placed in a thermostatic incubator at a constant temperature of 25 °C. The weighing method was used to replenish the moisture during the experiment, and culture took place until 90 days, when sampling occurred. For each treatment, 3 basins of parallel soil samples were randomly selected for soil microbial analysis. One part was stored at 4 °C for determination of its physical and chemical properties [13], and the other was stored at −80 °C for DNA extraction.

2.3. 16S rRNA Gene Sample Preparation, Sequencing, and Analysis

Soil DNA was extracted using the FastDNA®® SPIN Kit for Soil (MP Biomedicals, Santa Ana, CA, USA). Before extraction, soil samples were crushed three times for 20 s each using a FastPrepTM FP120 machine (MP Biomedicals) at level 4 speed. The full-length bacterial 16S ribosomal RNA gene was amplified via PCR (95 °C for 2 min, followed by 27 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 60 s and a final extension at 72 °C for 5 min) using the primers 27F (5′-AGRGTTYGATYMTGGCTCAG-3′) and 1492R (5′-RGYTACCTTGTTACGACTT-3′), where the barcode was an eight-base sequence unique to each sample.

2.4. Processing of Sequencing Data

SMRTbell libraries were prepared from the amplified DNA via blunt ligation in according with the manufacturer’s instructions (Pacific Biosciences, Menlo Park, CA, USA). All amplicon sequencing was performed using Shanghai Biozeron Biotechnology Co., Ltd. (Shanghai, China). PacBio raw reads were processed using SMRT Link Analysis software version 9.0 to obtain demultiplexed circular consensus sequence (CCS) reads. Raw reads were processed through SMRT Portal to filter sequences for length (>1400 bps) and quality. Operational taxonomic units (OTUs) were clustered with a 98.65% similarity cut-off using UPARSE (version 7.1 http://drive5.com/uparse/) (accessed on 7 June 2021), and chimeric sequences were identified and removed using UCHIME. The phylogenetic affiliation of each 16S rRNA gene sequence was analyzed by the UCLUST algorithm (http://www.drive5.com/usearch/manual/uclust_algo.html) (accessed on 9 June 2021) against the Silva (SSU138.1) 16S rRNA database using a confidence threshold of 70% [14]. These sequence data have been submitted to the GenBank database under accession number PRJNA899660.

2.5. Statistical Analyses

Principal coordinate analysis (PCoA) and one-way permutational analysis of variance (PERMANOVA) based on Bray–Curtis dissimilarities were performed using R (version 4.1.0, vegan package) to assess the statistically significant effects of treatments on bacterial communities [15]. Then, a partial Mantel test and a Procrustes analysis were performed to assess the correlations between bacterial community composition dissimilarity and soil physicochemical dissimilarity, respectively (R 4.1.0, vegan package). All statistical analyses were performed using the R stats package.
NCM was adopted to detect stochastic processes shaping bacterial community assembly under low-dose and high-dose fertilization [7]. The ecological width index was calculated using the “niche.width” function within the R package “spaa”. Bacterial species under low-dose and high-dose fertilization conditions were further classified as generalists or specialists based on their occurrence and by using the R package “EcolUtils” [16]. A bacterial species was defined as a generalist when the observed occurrence exceeded the upper 95% confidence interval, as a specialized species when the observed occurrence was below the lower limit of the 95% confidence interval of the zero distribution, and as a neutral species when the observed occurrence was within the 95% confidence interval of the zero distribution [16].
Co-occurrence network analysis was performed using the Molecular Ecological Network Analysis pipeline (http://ieg2.ou.edu:80/MENA/) (accessed on 25 July 2021). The network files were generated as “output for Cytoscape visualization” and visualized with the Gephi platform [17]. A robustness test was applied to measure network stability [18,19].
iCAMP is an emerging tool for quantifying microbial assembly processes, such as selection, dispersal, diversification, and drift [4]. Its main innovation is that the OTUs are first divided into different groups (‘bins’) based on their phylogenetic relationships [4]. Therefore, iCAMP is more accurate, sensitive, and specific than previous methods, such as tNST, NP, and QPEN [4]. In this study, iCAMP analysis was performed using a web-based pipeline (http://ieg3.rccc.ou.edu:8080) (accessed on 29 July 2021) built on the Galaxy platform (version 18.09) to quantify these microbial assembly processes. iCAMP analysis results were further visualized using the “ggplot2” R package. A phylogenetic tree of the top 200 bacterial bins was constructed using MEGA 10 (https://www.megasoftware.net/home) (accessed on 5 August 2021) and visualized using Interactive Tree of Life (iTOL, version 4.3.2) [20].

3. Results

3.1. Effects of Fertilizer Treatments on Soil Properties and the Soil Bacterial Community

The PCoA and PERMANOVA (Figure 1A, p < 0.05) results showed that the soil bacterial community similarity distance was mainly influenced by soil type. Additionally, the H and R soil results also showed that fertilization had a significant driving effect on bacterial community structure (Figure 1B,C, p < 0.05). Samples from the treatment without fertilization could be clearly distinguished from those from treatments with fertilization in terms of community composition (PERMANOVA, p < 0.05). The application of organic fertilizer had a significant effect on structural changes in the bacterial communities in both types of soil.
The Mantel test and Procrustes analysis were further used to verify whether there was a significant correlation between bacterial community structures and overall soil environmental factors. Mantel tests revealed that the application of organic fertilizer had a significant effect on structural changes in bacterial communities in both types (H and R) of soil (Figure 1D,E, Spearman, p < 0.05). Procrustes analysis showed that there was a significant correlation (M2 = 0.424, p = 0.001, 999 permutations) between the soil bacterial community and the soil physicochemical properties in the different samples (Figure 1F). Figure 1G,H display a p value of 0.057 for H soil and a p value of 0.001 for R soil, suggesting that the soil physical properties were significantly associated with the microbial community under the application of organic fertilizer. The similarity changes in soil physical and chemical characteristics along a gradient of fertilization treatments were significantly correlated with the similarity changes in bacterial communities.

3.2. The Potential Importance of Stochastic Processes in Bacterial Community Assembly

The NCM was used to determine the potential importance of stochastic processes in bacterial community assembly. The results showed that the relative contribution of stochastic processes increased gradually with increasing fertilization dose. At high fertilization doses, stochastic processes explained H soil at 77.3% (HH) and R soil at 78.7% (RH) of the community variance in H soil and R soil, respectively (Figure 2). However, at low fertilization doses, the explanatory rates were only 71.2% (HL) and 73.5% (RL), respectively (Figure 2). For H soils, the Nm value of bacterial species at low doses (Nm = 15,193) was higher than that of the high-dose group (Nm = 13,153), indicating that the spread of bacteria in the low-dose group was greater than that in the high-dose group.
In general, generalist and specialist species were widely present in the H and R soils, and the proportion of nonsignificant species was greater than 94% (Figure S1). For the H soil, the proportion of generalist species in the low-fertilization group (HL, 0.7%) was higher than that in the HH group (0.6%, Figure S1A,B). However, R soils had the opposite results, and the low-fertilization group (RL, 0.3%) was much lower than that in RH (0.6%, Figure S1C,D). However, the proportion of specialist species in the low-fertilization group (2.9%, HL; 5.5%, RL; Figure S1A,C) was much higher than that in the high-dose fertilization group (1.8%, HH; 2.8%, RH; Figure S1B,D) at the bacterial genus level.

3.3. Microbial Co-Occurrence Network and Its Stability

Figure 3A shows the co-occurrence network of H and R soils. Natural connectivity accurately depicts the nuances of network damage resistance and was calculated to determine the robustness of the microbial networks among the treatments. Figure 3B shows that the resistance of microbial networks related directly to soil type. The absolute value of the slope for the bacterial network of the nutrient-rich H soil was less than that of the R soil. A smaller slope value indicates more stability within networks. A higher fertilization dose also had a stabilizing effect on the bacterial network. It was confirmed that the slope displayed a 10.9% decrease in H soil and a 37.2% decrease in R soil (Figure 3B).

3.4. Soil Bacterial Community Assembly Processes along a Gradient of Organic Fertilizer Application Levels

According to iCAMP analysis, homogeneous selection (51.2% in H soil, 44.6% in R soil) and ecological drift (37.4% in H soil, 35.1% in R soil) were more important than other processes in bacterial community assembly (Figure 4A,B). The fertilization gradient significantly altered the relative importance of different processes. Overall, the gradient of increasing fertilization amounts gradually increased the relative importance of ecological drift and reduced the importance of homogeneous selection (Figure 4C–F and Table S2, p < 0.01, PERMANOVA). The relative importance of homogenizing dispersal and dispersal limitation was generally less than 10%. The effect of fertilization amount fluctuated, but homogenizing dispersal showed a clear downwards trend as fertilization increased, and dispersal limitation showed a clear increasing trend (Figure 4G,H).

3.5. Assembly Mechanisms across Phylogenetic Groups

A total of 535 bins were generated according to iCAMP’s default parameter settings, where the minimal bin size requirement was set to 24. The relative importance of different ecological processes in individual bins (lineages) was calculated. More details on the 523 bins are listed in Table S3. The top (dominant) 200 bins were selected based on their relative abundance, which accounted for 37.4% of bins and 69.3% of all bin relative abundances. The top taxa of the 200 bins belonged to 13 phyla, including 66 Proteobacteria, 54 Acidobacteriota, 19 Actinobacteriota, 19 Gemmatimonadota, 10 Bacteroidota, 8 Planctomycetota, 6 Chloroflexi, and 5 Myxococcota (Figure 5A). The relative abundance of these bins was between 0.17 and 1.37%. The most abundant bin (Bin 281, Proteobacteria, average 1.37%) was governed by homogeneous selection (Tables S4 and S5). Along the fertilization gradient, the process of community assembly that dominated in these two types of soil changed greatly (Figure 5A–C).
The colored columns of different sizes represent the interpretation rate of different processes, but the interpretation rate of the five processes changed significantly. In summary, the community assembly process of these 200 bins was rarely consistent within the two types of soils, and even the process that dominated the bins changed significantly. The results in Figure 5B and Table S5 show that 103 (51.5%) of the 200 bins in the H soil were dominated by HoS, with 63 (31.5%) by DR and 34 (17%) by HD. There were 113 (56.5%) cases of HoS, 56 (28%) of DR, 29 (14.5%) of HD, and 2 (1%) of DL for R soils (Figure 5C). A total of 103 bins showed dominant processes that differed between the H and R soils. These 103 bins are listed in Figure 5D, including 30 Acidobacteriota bins (55.6% of the total number of Acidobacteriota bins), 33 Proteobacteria bins (50%), 11 Actobacteriota bins (57.9%), and 11 Gemmatimonadota bins (57.9%). These results demonstrated that HoS, HD, and DR were the most important processes in H and R soils, but these bacterial community assembly processes were not consistent in the two types of soil. Half of the dominant processes in the top 200 bins were different. In general, there are complex and varied assembly mechanisms for different bacterial bins under different fertilization doses in different soil types. This bin-level mechanistic information is richer and more accurate than information obtained with other methods.

4. Discussion

4.1. Soil Microbes Detected Using PacBio Full-Length 16S rRNA Sequencing

For insect gut microbiomes, PacBio full-length 16S rRNA sequencing was more effective than the Illumina (V4 zone) platform at the genus level [21]. The use of full-length 16S rRNA greatly reduced the proportion of unclassified species, and unclassified genera were not detected in Tabanus nigrovittatus samples [21]. However, for samples of complex habitats, PacBio sequencing still yields some unclassified genera, such as with soil and animal gut samples [22]. Our study revealed 70 unclassified genera among 962 total genera (30%) and 612 unclassified species among 1928 total species. We believe that these results are associated with the selected 16S rRNA database and thresholds. PacBio full-length 16S rRNA sequencing can more accurately reduce the composition of the microbial community in the sample [23] and lays the foundation for the next step of community assembly analysis. Full-length 16S rRNA gene amplicons might be used to construct phylogenetic trees with highly supported topologies that are more suitable for iCAMP and the phyloscore analytical framework [24].

4.2. Fertilization Dose Mediates Soil Properties to Drive Soil Microbial Community Processes

The application of organic fertilizer increases the biomass and activity of soil microorganisms, and studies have shown that organic fertilizers have a stronger impact on bacterial structure than the application of chemical fertilizers [25,26]. Many studies have demonstrated that manure fertilization alters soil microbial communities mainly due to organic component inputs [26]. Although the organic fertilizer used in this study was not sterilized before the experiments, the majority of organic fertilizer microorganisms do not survive in soil [12]. After 2 weeks of application of organic fertilizer, the above microbial characteristics showed a downwards trend [12]. The vast majority of microorganisms associated with organic fertilizers, especially Gram-negative bacteria, die almost immediately in the soil environment, and only a few genera can survive in soil for several months. Fungal abundance and diversity begin to decline after the 9th week of application of organic fertilizers [12]. Variations in soil organic component input and pH can directly or indirectly modify microbial habitats [27]. This study revealed that soil bacterial community similarity was mainly influenced by soil type, followed by the application of organic fertilizers. Our results suggested that the habitat created by soil type and fertilization played an important role in bacterial community assembly processes. This was further supported by the results of the Mantel test and Procrustes analysis (Figure 1).
Our study revealed that the bacterial community in red soils is primarily affected by the chemical and organic components of manure fertilizer. This was further supported by the physiochemical property results. The variation in environmental conditions along latitudinal gradients shapes soil bacterial β diversity by mediating the strength of heterogeneous selection [28]. Our study showed that fertilization directly changes soil environmental conditions, triggering soil bacterial community assembly processes. Consistent with the findings of this previous work, the application of biofertilizer triggered microbial assembly in microaggregates [29].

4.3. High-Dose Fertilization Increases Stochastic Processes

The NCM results support the prominent role of stochastic processes in shaping bacterial community assembly, and high-dose fertilization increases stochastic processes. Additionally, fertilization at high doses increases the migration rate of H-soil microorganisms, and nutrient-rich H soils have better community immigration than R soils. A higher proportion of specialist species (Figure S1) was detected under low fertilization. Specialist species have a narrower fundamental niche and are at a disadvantage under resource competition [30].
High-dose fertilization directly increases soil nutrients by creating specific niches and exerting selection forces to screen microfloral taxa [31]. Studies have shown that increased sources of aggregation enhance the interactions and development of shared niches [30]. Our results suggest that high-dose fertilization causes the soil to have narrower niches, and similar results suggest that rhizosphere soils have narrower niches than bulk soil. We suspect that high-dose fertilization increases the migration rate of soil microbes but further weakens the competitive advantage of specialist species to shape the microbial community structure.
The increase in fertilization mentioned earlier causes changes in bacterial interactions via the increase in fertilization-specific niches. According to network analysis, fertilization dose not only affects the topology of the bacterial network nodes, connections, modules, etc., but also affects the stability (robustness) of the network. While the functionality and performance of complex networks depend on their stability, Wu et al. [19] mentioned that the resilience of networks stems from the redundancy of alternative paths. High-dose fertilization in this study increased the structural robustness of the H and R soil networks. This may relate to the interaction between microorganisms affected by soil nutrients. Studies have shown that competition for nutrients is more intense among microbial communities under poor nutrient conditions, while competition among microbes reinforces the robustness of networks [18,32]. This is also consistent with lower-nutrient R soils having better natural connectivity than H soils (Figure 3).

4.4. Homogeneous Selection and Ecological Drift Mediate the Assembly Processes of Microbial Communities

The mechanisms (selective or neutral processes) of microbial community assembly were evaluated under different treatments. To this end, sequencing data and/or phylogenetic assessment of the assembly (construction) process of the target community under environmental stress over time and/or space were applied. Two algorithms commonly used in the evaluation of community assembly processes are the normalized stochasticity ratio (NST) method developed in 2019 by Ning et al. [33] and the beta nearest-taxon index (βNTI) method developed by Stegen et al. [34]. The V4 hypervariable region of the bacterial 16S rRNA gene was used to assess the relationship between βNTI values of red and black soils and soil dilution and pH levels of soil suspensions [35]. In this study, overall visualization of the bacterial community structure assembly patterns based on species richness under different fertilization treatments in two soils was carried out.
Soil type and fertilization dose were the two most important variables influencing bacterial community assembly in this study. iCAMP analysis showed that the two community assembly mechanisms of H soil and R soil were mainly dominated by homogeneous selection and ecological drift, while increasing the fertilization dose significantly reduced the relative importance of homogeneous selection and increased the relative importance of ecological drift. These results are also consistent with the findings of previous research, where homogeneous selection and ecological drift were found to be important assembly processes; for example, homogeneous selection was the dominant assembly process in a glacier-fed stream microbiome [24] and membrane biofilm [36], and ecological drift was dominant in a grassland microbial community [4].
Along a gradient of fertilization conditions, we further compared the assembly mechanisms of the microflora in H and R soils across different phylogenetic groups. Our analytical framework showed that the top 200 bins of both soils were dominated by HoS, DR, and HD processes. This was consistent with our expectations, and the processes dominating similar bins in the assembly process were not consistent. Proteobacteria, for example, was dominated by alternating HoS, DR, and HD processes, although the abundance of these bins did not vary much in order of magnitude. We hypothesize that this relates to their different responses to the fertilization gradient. Interestingly, the dominant process of 103 bins, covering 14 bacterial phyla, was transformed in H and R soils. This shows that soil type has an important influence on community assembly along fertilization gradients, influencing a wide spectrum of bacterial phyla.
Previous studies also reported a similar trend in which the dominant process of similar bins under warming conditions was inconsistent [4]. We hypothesize that this result relates to the fact that more closely related microorganisms are more similar in terms of resource competition and niche width. Interestingly, the dominant process of 103 bins, covering 14 bacterial phyla, was transformed in H and R soils. This shows that soil type has an important influence on the dominant bacterial community assembly process, which involves a wide spectrum of bacterial phyla. iCAMP can be used as a standalone tool to identify phylogenetic clades driving community assembly patterns. Compared with previous studies focusing on the community assembly processes of abundant and rare bacteria [5,37], our data showed that phylogenetic relationships and abundance were not decisive for community assembly. In addition, phyloscore analysis is a novel analytical framework used to identify phylogenetic clades under homogeneous ecological selection [24].

5. Conclusions

In conclusion, the high-dose application of organic fertilizer caused significant succession of the bacterial community in two different soil types. This succession is also closely associated with the physical and chemical changes in soil caused by fertilization. The NCM and iCAMP analysis demonstrated that an increase in the amount of fertilizer applied changed the mechanism of bacterial community assembly, in turn increasing the potential importance of stochastic processes. With an increase in the fertilization dose, the importance of drift (DR) processes increased significantly, and the importance of homogeneous selection (HoS) decreased significantly. The presented results provide new insights for understanding the process of bacterial community assembly driven by the application of high doses of organic fertilizers and for studying the mechanism of community assembly of the same microbial lineages in different soils. Further research is required to verify the contribution of the high-dose application of organic fertilizer on the community construction process of rhizosphere microbial communities, root endophytes, and functional genes, especially on a larger geographical scale.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/agronomy13071875/s1, Figure S1: Relative contributions of generalist and specialist species in soil bacterial communities with high- and low-dose fertilization; Table S1: Physical and chemical properties of the experimental soils; Table S2: Comparison of iCAMP processes between groups; Table S3: Each bin’s contribution was analyzed according to iCAMP; Table S4: The relative abundance of each bin and its top taxon; Table S5: The dominant process for each bin.

Author Contributions

Data curation, W.W. and N.L.; writing—original draft, W.W.; methodology, Y.G.; formal analysis, Y.G., N.L. and H.L.; visualization, W.W. and R.L.; project administration, X.G.; funding acquisition, X.G. All authors have read and agreed to the published version of the manuscript.

Funding

This work was financially supported by the Anhui Provincial Natural Science Foundation, China (2108085QC89), the Natural Science Research Project of Anhui Educational Committee, China (2022AH050870), and the Anhui Postdoctoral Science Foundation, China (2020B410).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, Xungang Gu.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Li, B.-B.; Roley, S.S.; Duncan, D.S.; Guo, J.; Quensen, J.F.; Yu, H.-Q.; Tiedje, J.M. Long-term excess nitrogen fertilizer increases sensitivity of soil microbial community to seasonal change revealed by ecological network and metagenome analyses. Soil Biol. Biochem. 2021, 160, 108349. [Google Scholar] [CrossRef]
  2. Wang, W.; Wang, N.; Dang, K.; Dai, W.; Guan, L.; Wang, B.; Gao, J.; Cui, Z.; Dong, Y.; Wang, H. Long-term nitrogen application decreases the abundance and copy number of predatory myxobacteria and alters the myxobacterial community structure in the soil. Sci. Total Environ. 2020, 708, 135114. [Google Scholar] [CrossRef] [PubMed]
  3. Johnson, J.S.; Spakowicz, D.J.; Hong, B.-Y.; Petersen, L.M.; Demkowicz, P.; Chen, L.; Leopold, S.R.; Hanson, B.M.; Agresta, H.O.; Gerstein, M.; et al. Evaluation of 16S rRNA gene sequencing for species and strain-level microbiome analysis. Nat. Commun. 2019, 10, 5029. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  4. Ning, D.; Yuan, M.; Wu, L.; Zhang, Y.; Guo, X.; Zhou, X.; Yang, Y.; Arkin, A.P.; Firestone, M.K.; Zhou, J. A quantitative framework reveals ecological drivers of grassland microbial community assembly in response to warming. Nat. Commun. 2020, 11, 4717. [Google Scholar] [CrossRef] [PubMed]
  5. Mo, Y.; Peng, F.; Gao, X.; Xiao, P.; Logares, R.; Jeppesen, E.; Ren, K.; Xue, Y.; Yang, J. Low shifts in salinity determined assembly processes and network stability of microeukaryotic plankton communities in a subtropical urban reservoir. Microbiome 2021, 9, 128. [Google Scholar] [CrossRef]
  6. Fargione, J.; Brown, C.S.; Tilman, D. Community assembly and invasion: An experimental test of neutral versus niche processes. Proc. Natl. Acad. Sci. USA 2003, 100, 8916. [Google Scholar] [CrossRef]
  7. Chen, W.; Ren, K.; Isabwe, A.; Chen, H.; Liu, M.; Yang, J. Stochastic processes shape microeukaryotic community assembly in a subtropical river across wet and dry seasons. Microbiome 2019, 7, 138. [Google Scholar] [CrossRef] [Green Version]
  8. Sloan, W.T.; Lunn, M.; Woodcock, S.; Head, I.M.; Nee, S.; Curtis, T.P. Quantifying the roles of immigration and chance in shaping prokaryote community structure. Environ. Microbiol. 2006, 8, 732–740. [Google Scholar] [CrossRef]
  9. Gad, M.; Hou, L.; Li, J.; Wu, Y.; Rashid, A.; Chen, N.; Hu, A. Distinct mechanisms underlying the assembly of microeukaryotic generalists and specialists in an anthropogenically impacted river. Sci. Total Environ. 2020, 748, 141434. [Google Scholar] [CrossRef]
  10. Stegen, J.C.; Lin, X.; Fredrickson, J.K.; Chen, X.; Kennedy, D.W.; Murray, C.J.; Rockhold, M.L.; Konopka, A. Quantifying community assembly processes and identifying features that impose them. ISME J. 2013, 7, 2069–2079. [Google Scholar] [CrossRef]
  11. Khan, M.N.; Mobin, M.; Abbas, Z.K.; Alamri, S.A. Fertilizers and Their Contaminants in Soils, Surface and Groundwater. In Encyclopedia of the Anthropocene; Dellasala, D.A., Goldstein, M.I., Eds.; Elsevier: Oxford, UK, 2018; pp. 225–240. [Google Scholar]
  12. Semenov, M.V.; Krasnov, G.S.; Semenov, V.M.; Ksenofontova, N.; Zinyakova, N.B.; van Bruggen, A.H.C. Does fresh farmyard manure introduce surviving microbes into soil or activate soil-borne microbiota? J. Environ. Manag. 2021, 294, 113018. [Google Scholar] [CrossRef] [PubMed]
  13. Wang, W.; Wang, H.; Feng, Y.; Wang, L.; Xiao, X.; Xi, Y.; Luo, X.; Sun, R.; Ye, X.; Huang, Y.; et al. Consistent responses of the microbial community structure to organic farming along the middle and lower reaches of the Yangtze River. Sci. Rep. 2016, 6, 35046. Available online: https://www.nature.com/articles/srep35046 (accessed on 12 June 2020). [CrossRef] [PubMed] [Green Version]
  14. Yang, S.; Xu, W.; Tan, C.; Li, M.; Li, D.; Zhang, C.; Feng, L.; Chen, Q.; Jiang, J.; Li, Y.; et al. Heat Stress Weakens the Skin Barrier Function in Sturgeon by Decreasing Mucus Secretion and Disrupting the Mucosal Microbiota. Front. Microbiol. 2022, 13, 860079. [Google Scholar] [CrossRef]
  15. Anderson, M.J.; Walsh, D.C.I. PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: What null hypothesis are you testing? Ecol. Monogr. 2013, 83, 557–574. [Google Scholar] [CrossRef]
  16. Wu, W.; Logares, R.; Huang, B.; Hsieh, C.-h. Abundant and rare picoeukaryotic sub-communities present contrasting patterns in the epipelagic waters of marginal seas in the northwestern Pacific Ocean. Environ. Microbiol. 2017, 19, 287–300. [Google Scholar] [CrossRef] [PubMed]
  17. Bastian, M.; Heymann, S.; Jacomy, M. Gephi: An open source software for exploring and manipulating networks. In Proceedings of the Third International AAAI Conference on Weblogs and Social Media, San Jose, CA, USA, 17–20 May 2009; Volume 8, pp. 361–362. [Google Scholar]
  18. Shi, Y.; Zhang, K.; Li, Q.; Liu, X.; He, J.-S.; Chu, H. Interannual climate variability and altered precipitation influence the soil microbial community structure in a Tibetan Plateau grassland. Sci. Total Environ. 2020, 714, 136794. [Google Scholar] [CrossRef] [PubMed]
  19. Wu, J.; Mauricio, B.; Tan, Y.-J.; Deng, H.-Z. Natural Connectivity of Complex Networks. Chin. Phys. Lett. 2010, 27, 078902. [Google Scholar] [CrossRef]
  20. Letunic, I.; Bork, P. Interactive tree of life (iTOL) v3: An online tool for the display and annotation of phylogenetic and other trees. Nucleic Acids Res. 2016, 44, W242–W245. [Google Scholar] [CrossRef]
  21. Lefoulon, E.; Truchon, A.; Clark, T.; Long, C.; Frey, D.; Slatko Barton, E.; Auchtung Jennifer, M. Greenhead (Tabanus nigrovittatus) Wolbachia and Its Microbiome: A Preliminary Study. Microbiol. Spectr. 2021, 9, e00517–e00521. [Google Scholar] [CrossRef]
  22. Emu, Q.; Guan, H.; Zhu, J.; Zhang, L.; Fan, J.; Ji, Y.; Lin, Y.; Li, C.; Dan, X.; Aguo, Y.; et al. Grazing and Supplementation of Dietary Yeast Probiotics Shape the Gut Microbiota and Improve the Immunity of Black Fattening Goats (Capra hircus). Front. Microbiol. 2021, 12, 666837. [Google Scholar] [CrossRef]
  23. Singer, E.; Bushnell, B.; Coleman-Derr, D.; Bowman, B.; Bowers, R.M.; Levy, A.; Gies, E.A.; Cheng, J.-F.; Copeland, A.; Klenk, H.-P.; et al. High-resolution phylogenetic microbial community profiling. ISME J. 2016, 10, 2020–2032. [Google Scholar] [CrossRef]
  24. Fodelianakis, S.; Washburne, A.D.; Bourquin, M.; Pramateftaki, P.; Kohler, T.J.; Styllas, M.; Tolosano, M.; De Staercke, V.; Schön, M.; Busi, S.B.; et al. Microdiversity characterizes prevalent phylogenetic clades in the glacier-fed stream microbiome. ISME J. 2021, 16, 666–675. [Google Scholar] [CrossRef] [PubMed]
  25. Wang, H.; Xu, J.; Liu, X.; Zhang, D.; Li, L.; Li, W.; Sheng, L. Effects of long-term application of organic fertilizer on improving organic matter content and retarding acidity in red soil from China. Soil Tillage Res. 2019, 195, 104382. [Google Scholar] [CrossRef]
  26. Yang, Y.; Li, G.; Min, K.; Liu, T.; Li, C.; Xu, J.; Hu, F.; Li, H. The potential role of fertilizer-derived exogenous bacteria on soil bacterial community assemblage and network formation. Chemosphere 2022, 287, 132338. [Google Scholar] [CrossRef]
  27. Suleiman, A.K.A.; Gonzatto, R.; Aita, C.; Lupatini, M.; Jacques, R.J.S.; Kuramae, E.E.; Antoniolli, Z.I.; Roesch, L.F.W. Temporal variability of soil microbial communities after application of dicyandiamide-treated swine slurry and mineral fertilizers. Soil Biol. Biochem. 2016, 97, 71–82. [Google Scholar] [CrossRef]
  28. Zhang, X.; Liu, S.; Wang, J.; Huang, Y.; Freedman, Z.; Fu, S.; Liu, K.; Wang, H.; Li, X.; Yao, M.; et al. Local community assembly mechanisms shape soil bacterial β diversity patterns along a latitudinal gradient. Nat. Commun. 2020, 11, 5428. [Google Scholar] [CrossRef] [PubMed]
  29. Dong, M.; Zhao, M.; Shen, Z.; Deng, X.; Ou, Y.; Tao, C.; Liu, H.; Li, R.; Shen, Q. Biofertilizer application triggered microbial assembly in microaggregates associated with tomato bacterial wilt suppression. Biol. Fertil. Soils 2020, 56, 551–563. [Google Scholar] [CrossRef]
  30. Zhang, J.; Zhang, B.; Liu, Y.; Guo, Y.; Shi, P.; Wei, G. Distinct large-scale biogeographic patterns of fungal communities in bulk soil and soybean rhizosphere in China. Sci. Total Environ. 2018, 644, 791–800. [Google Scholar] [CrossRef]
  31. Mendes, L.W.; Kuramae, E.E.; Navarrete, A.A.; van Veen, J.A.; Tsai, S.M. Taxonomical and functional microbial community selection in soybean rhizosphere. ISME J. 2014, 8, 1577–1587. [Google Scholar] [CrossRef] [Green Version]
  32. Coyte Katharine, Z.; Schluter, J.; Foster Kevin, R. The ecology of the microbiome: Networks, competition, and stability. Science 2015, 350, 663–666. [Google Scholar] [CrossRef]
  33. Ning, D.; Deng, Y.; Tiedje, J.M.; Zhou, J. A general framework for quantitatively assessing ecological stochasticity. Proc. Natl. Acad. Sci. USA 2019, 116, 16892. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  34. Stegen, J.C.; Lin, X.; Konopka, A.E.; Fredrickson, J.K. Stochastic and deterministic assembly processes in subsurface microbial communities. ISME J. 2012, 6, 1653–1664. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Xun, W.; Li, W.; Xiong, W.; Ren, Y.; Liu, Y.; Miao, Y.; Xu, Z.; Zhang, N.; Shen, Q.; Zhang, R. Diversity-triggered deterministic bacterial assembly constrains community functions. Nat. Commun. 2019, 10, 3833. [Google Scholar] [CrossRef] [Green Version]
  36. Matar, G.K.; Ali, M.; Bagchi, S.; Nunes, S.; Liu, W.-T.; Saikaly, P.E. Relative Importance of Stochastic Assembly Process of Membrane Biofilm Increased as Biofilm Aged. Front. Microbiol. 2021, 12, 708531. [Google Scholar] [CrossRef]
  37. Wang, J.; Li, M.; Li, J. Soil pH and moisture govern the assembly processes of abundant and rare bacterial communities in a dryland montane forest. Environ. Microbiol. Rep. 2021, 13, 862–870. [Google Scholar] [CrossRef] [PubMed]
Agronomy 13 01875 g001 550
Figure 1. Compositional differences in bacteria of different fertilization samples and the correlation between soil bacterial communities and the soil physicochemical properties of different soil types. Principal coordinates analysis (PCoA) plot based on bacterial species-level abundances using a Bray–Curtis distance metric. (A) All soil samples, (B) loess brown soil (H soil) samples, and (C) red soil (R soil) samples. The Mantel test ((D): loess brown soil; (E): red soil, Spearman) and Procrustes analysis ((F): all soils; (G): loess brown soil; (H): red soil) of the correlation based on the Bray–Curtis results for OTU abundances and Euclidean results for soil physicochemical properties. Permutational multivariate analysis of variance, PERMANOVA.
Figure 1. Compositional differences in bacteria of different fertilization samples and the correlation between soil bacterial communities and the soil physicochemical properties of different soil types. Principal coordinates analysis (PCoA) plot based on bacterial species-level abundances using a Bray–Curtis distance metric. (A) All soil samples, (B) loess brown soil (H soil) samples, and (C) red soil (R soil) samples. The Mantel test ((D): loess brown soil; (E): red soil, Spearman) and Procrustes analysis ((F): all soils; (G): loess brown soil; (H): red soil) of the correlation based on the Bray–Curtis results for OTU abundances and Euclidean results for soil physicochemical properties. Permutational multivariate analysis of variance, PERMANOVA.
Agronomy 13 01875 g001
Agronomy 13 01875 g002 550
Figure 2. Neutral model applied to assess the effects of random dispersal and ecological drift on the assembly of HL, HH, RL, and RH soils. R2 indicates the goodness of fit to the neutral model. Nm indicates the metacommunity size times immigration. The solid blue lines indicate the best fit to the neutral model, and dashed blue lines represent 95% confidence intervals around the model prediction. H soil: loess brown soil; R soil: red soil; HH: H soil with a high fertilizer dose; HL: H soil with a low fertilizer dose; RH: R soil with a high fertilizer dose; and RL: R soil with a low fertilizer dose.
Figure 2. Neutral model applied to assess the effects of random dispersal and ecological drift on the assembly of HL, HH, RL, and RH soils. R2 indicates the goodness of fit to the neutral model. Nm indicates the metacommunity size times immigration. The solid blue lines indicate the best fit to the neutral model, and dashed blue lines represent 95% confidence intervals around the model prediction. H soil: loess brown soil; R soil: red soil; HH: H soil with a high fertilizer dose; HL: H soil with a low fertilizer dose; RH: R soil with a high fertilizer dose; and RL: R soil with a low fertilizer dose.
Agronomy 13 01875 g002
Agronomy 13 01875 g003 550
Figure 3. Co-occurrence networks and robustness analysis for microbial communities between H (loess brown) and R (red) soils. (A) The nodes indicate individual OTUs; the size of each node is positively correlated with the node degree, and the modules are randomly colored. (B) Robustness analysis was performed to show the relationships between microbial natural connectivity and the proportion of removed nodes. The horizontal coordinates are the node removal scale, and the ordinates represent the natural connectivity of the network. HH: H soil with a high fertilizer dose; HL: H soil with a low fertilizer dose; RH: R soil with a high fertilizer dose; RL: R soil with a low fertilizer dose.
Figure 3. Co-occurrence networks and robustness analysis for microbial communities between H (loess brown) and R (red) soils. (A) The nodes indicate individual OTUs; the size of each node is positively correlated with the node degree, and the modules are randomly colored. (B) Robustness analysis was performed to show the relationships between microbial natural connectivity and the proportion of removed nodes. The horizontal coordinates are the node removal scale, and the ordinates represent the natural connectivity of the network. HH: H soil with a high fertilizer dose; HL: H soil with a low fertilizer dose; RH: R soil with a high fertilizer dose; RL: R soil with a low fertilizer dose.
Agronomy 13 01875 g003
Agronomy 13 01875 g004 550
Figure 4. Relative importance of different ecological processes along a fertilization gradient. (A): Different ecological processes in H (loess brown) soil. (B): Different ecological processes in R (red) soil. Processes under different fertilization doses in H soil (C) and R soil (D). Changes in the relative importance of homogeneous selection (E), drift (F), homogenizing dispersal (G), and dispersal limitation (H) under fertilization and control conditions. Mean values and standard deviations of importance were used to plot data as bars, with the significance markers *** p < 0.01, and * p < 0.1 (n = 3). L, M, S, and N represent large (|d| > 0.8), medium (0.5 < |d| ≤ 0.8), small (0.2 < |d| ≤ 0.5), and negligible (|d| ≤ 0.2) effect sizes of fertilization, respectively, based on Cohen’s d. More details are provided in the Source Data file (Table S2).
Figure 4. Relative importance of different ecological processes along a fertilization gradient. (A): Different ecological processes in H (loess brown) soil. (B): Different ecological processes in R (red) soil. Processes under different fertilization doses in H soil (C) and R soil (D). Changes in the relative importance of homogeneous selection (E), drift (F), homogenizing dispersal (G), and dispersal limitation (H) under fertilization and control conditions. Mean values and standard deviations of importance were used to plot data as bars, with the significance markers *** p < 0.01, and * p < 0.1 (n = 3). L, M, S, and N represent large (|d| > 0.8), medium (0.5 < |d| ≤ 0.8), small (0.2 < |d| ≤ 0.5), and negligible (|d| ≤ 0.2) effect sizes of fertilization, respectively, based on Cohen’s d. More details are provided in the Source Data file (Table S2).
Agronomy 13 01875 g004
Agronomy 13 01875 g005 550
Figure 5. Assembly mechanisms across different phylogenetic groups. (A) A phylogenetic tree of the top 200 bins is displayed in the center, and the top taxa of the 200 bins are colored. Relative abundance of each bin (first annulus). Relative importance of different ecological processes in each bin (stacked bars in the fourth and fifth annuli). Light cyan bars, heterogeneous selection (HeS); light yellow, homogeneous selection (HoS); lilac, dispersal limitation (DL); orange, homogenizing dispersal (HD); light blue, drift (DR). Dominant process in the top 200 bins of H and R soils (second and third annuli). (B) Dominant process in the top 200 bins of H soil (second annulus). (C) Dominant process in the top 200 bins of R soil (third annulus). (D) The consistency of dominant processes for the top 200 bins dominant process in different soils. H soil: loess brown soil; R soil: red soil.
Figure 5. Assembly mechanisms across different phylogenetic groups. (A) A phylogenetic tree of the top 200 bins is displayed in the center, and the top taxa of the 200 bins are colored. Relative abundance of each bin (first annulus). Relative importance of different ecological processes in each bin (stacked bars in the fourth and fifth annuli). Light cyan bars, heterogeneous selection (HeS); light yellow, homogeneous selection (HoS); lilac, dispersal limitation (DL); orange, homogenizing dispersal (HD); light blue, drift (DR). Dominant process in the top 200 bins of H and R soils (second and third annuli). (B) Dominant process in the top 200 bins of H soil (second annulus). (C) Dominant process in the top 200 bins of R soil (third annulus). (D) The consistency of dominant processes for the top 200 bins dominant process in different soils. H soil: loess brown soil; R soil: red soil.
Agronomy 13 01875 g005
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, W.; Gao, Y.; Li, N.; Lu, H.; Lan, R.; Gu, X. PacBio Sequencing Unravels Soil Bacterial Assembly Processes along a Gradient of Organic Fertilizer Application. Agronomy 2023, 13, 1875. https://doi.org/10.3390/agronomy13071875

AMA Style

Wang W, Gao Y, Li N, Lu H, Lan R, Gu X. PacBio Sequencing Unravels Soil Bacterial Assembly Processes along a Gradient of Organic Fertilizer Application. Agronomy. 2023; 13(7):1875. https://doi.org/10.3390/agronomy13071875

Chicago/Turabian Style

Wang, Wenhui, Yuan Gao, Na Li, Hongmei Lu, Ranxiang Lan, and Xungang Gu. 2023. "PacBio Sequencing Unravels Soil Bacterial Assembly Processes along a Gradient of Organic Fertilizer Application" Agronomy 13, no. 7: 1875. https://doi.org/10.3390/agronomy13071875

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop