2. Materials and Methods
2.1 Study site
The study was carried on three sampling locations located in Southeast Georgia, including 1) Acacia (A) Farm (Latitude 32°34.784N, Longitude 82°32.313W), 2) Honeydew (H) Farm (32°32.354N, 81°50.053W), and 3) Strickland (S) Farm” (32°19.231N, 81°41.554W). At each of the location, three sites with different vegetation types and disturbance intensity levels. The first site, site one (1) is representative of a conventional tomato (Lycopersicon esculentum ) crop field, defined as highly disturbed agricultural crop production (C) soil. The second site is a transitional region, where a secondary successional event is occurring (dominated byAndropogon spp.) and defined as moderate disturbed transitional environment (T). The third site is representative of a recently undisturbed forest habitat, containing both deciduous hardwood (Quercus spp.). and coniferous pine forests (Pinus spp.) and defined as less-disturbed native forests (F). At a location, sampling sites were no more than 100 meters apart. Physical and chemical soil properties including soil organic (%), NO3-, P, Mg, Ca, soil pH, and cation exchange-capacity, were determined by Waters Agricultural Laboratories, Inc. Camilla, GA, and soil ammonium was determined at Georgia Southern University by extracting soils with 2M KCl and followed with the salicylate method (Nelson 1983).
2.2 Sample collection and DNA extraction
Soil samples were collected from there locations (A, H, and S) under three sites with different vegetation types and disturbance intensity levels (C, T, and F). Three replicates of the samples were collected from each of the three vegetation types of three different locations, thus a total of 27 samples were obtained. At each site, one approximately 6 x 50-meter plot was randomly selected, and a grid with 30-centimeter intervals was established on each plot. We collected soil samples from total of 30 generated coordinates and used a stratified random sampling regime. Total of thirty soil cores (3.4 cm diameter; 10 cm deep) from the rhizosphere of each site were collected with a core sampler and hand mixed in a single sterile plastic bag. We mixed ten soils cores into a single sample, thus each site yielding 3 samples. A grand total of 27 samples were collected with three sites at each of the tree locations. The DNA was extracted using PowerSoil DNA Isolation Kit from each sample (Mo Bio Laboratories, Inc., Carlsbad, CA).
2.3 Total bacterial DNA quantification
A NanoDrop spectrophotometer, ND-1000 (NanoDrop Technologies, Wilmington, DE) was used to quantify total DNA of each sample. Soil bacterial DNA was quantified by Real-time quantitative-PCR (Q-PCR) as the indicator of relative soil bacterial abundance (Fierer et al. 2005). The Q-PCR was performed in a the QuantStudio™ 6 Flex Real-Time PCR System (Life Technologies, Carlsbad, CA, USA) using the conditions as described previously (Wu et al. 2015). In brief, 16S rRNA gene was amplified with the primers 27F (5’-AGAGTTTGATCMTGGCTCAG-3’) and 355R (5’-GCTGCCTCCCGTAGGAGT-3’) for bacterial quantification. We extracted DNA from the pure culture of Micrococcus lutus (Item # 155160 from Carolina Biological Supply Company, Burlington, NC, USA) as the standard DNA. The standards with serial diluted DNA concentration extracted from pure culture was quantified using NanoDrop spectrophotometer, ND-1000 (NanoDrop Technologies, Wilmington, DE).
2.4. Illumina sequencing
We used the NanoDrop ND-1000 UV-Vis Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) to quantify the above extracted DNA. Primers 27F (5’-
AGAGTTTGATCMTGGCTCAG-3’) and 355R (5’- GCTGCCTCCCGTAGGAGT-3’) were used to amplify the V1–V2 hyper variable region of the 16S rRNA bacterial gene. DNA sequencing was conducted at University of Georgia Genomics and Bioinformatics Cores (Athens, Georgia, USA), using the MiSeq platform (Illumina, Inc., USA). QIIME software was selected for the purpose of integrating the original FASTQ format sequencing data (Caporaso et al. 2010). The USEARCH tool (version 7.0; http://drive5.com/usearch/) was used to vet and remove chimeric sequences. The operational taxonomic unit (OTU) partition threshold was identified at a 97% sequence similarity of classification results, which was subsequently used for the calculation of bacterial community diversity and relative abundance. To obtain species classification data corresponding to each OTU, the 16S Metagenomics from Illumina Sequence Hub (Illumina, Inc., San Diego, CA USA) was applied to analyze DNA from amplicon sequencing of prokaryotic 16S small subunit rRNA genes.
2.5 Quantitative PCR (qPCR) and Denaturing Gradient Gel Electrophoresis (DGGE) for N functional genes
Quantitative PCR amplification of selected nitrification and denitrification genes (AOB amo A and nir K genes) with corresponding primers (Wu et al. 2020) was performed using the QuantStudio™ 6 Flex Real-Time PCR System (Life Technologies, Carlsbad, CA, USA). The fluorescent dye SYBR-Green I which binds to double-stranded DNA was applied to quantify the relative abundance of nitrification AOB amo A and denitrification nir K genes. Each of the PCR mixtures contained 12.5 µL of 2× GoTaq® Colorless Master Mix (Promega, USA), 0.5 µL of 10 µM forward and reverse primer, 1 µL BSA, 2 µL SYBR® of 1×, 6 µL of nuclease-free water, and 2 µL of DNA template, which make the recommended 25 µL protocol. To control for mechanical and technical errors, the Q-PCR of each sample was run in triplicate, and the mean of all three DNA quantities was used for statistical analysis.
The GC clamp was added to the corresponding primer position for DGGE analysis. Amplified gene products (AOB amo A and nir K) were run on 8% (w/v) acrylamide with a linear chemical gradient ranging from 40%–70% using a DGGEK-1001 Cipher DGGE Kit (C.B.S. Scientific, Del Mar, CA, USA) and obtained DGGE band patterns were exported for further nonparametric multivariate analyses of soil microbial and N functional gene communities as described previously (Wu et al. 2020).
2.6 Link soil microbial community structure with environmental factors
We used PRIMER-E including PERMANOVA+ statistical software (PRIMER-E, Plymouth, UK) to generate soil bacterial and N functional gene similarity matrices of each sample from three different locations and three different vegetation types with different disturbance intensity levels. Bio-Env function and Principal Coordinates Analysis (PCO) with Spearman’s correlations of variables with the PCO axes of PRIMER-E were used to correlate the environmental factors with the similarity matrix (Wu et al. 2015, Wu et al. 2020).
2.7 Statistical analysis of operational taxonomic unit (OTUs) to test diversity and similarity
Univariate analysis of operational taxonomic unit (OTUs) by different fragment lengths or 97% sequence similarity was used to characterize diversity indices. Richness (S) was expressed as the total number of different OTUs identified. Diversity was calculated using the Shannon–Weiner (Weaver) index using the equation of Diversity (H): H’ = - ∑ (pi) (loge pi), where pi is the proportion of an individual OTU relative to the sum of OTUs detected in a sample. Evenness was calculated by Pielou’s evenness index using the equation of J′=H′/Log (S). Univariate analyses of richness, richness and Shannon diversity indices were performed with JMP® Pro 12.1.0 (SAS Institute Inc., USA). Cluster analysis was used to compare soil bacterial communities and N functional genes from there locations (A, H, and S) under three sites with different vegetation types and disturbance intensity levels (C, T, and F). The analysis of similarities (ANOSIM) procedure was applied to statistically discriminate the soil bacterial communities and N functional genes. The bacterial 16S rDNA OTUs obtained from Illumina sequencing from there locations under three sites with different vegetation types and disturbance intensity levels were determined using the similarity percentages (SIMPER) procedure. PRIMER-E including PERMANOVA+ statistical software (Primer-E, Plymouth Marine Laboratory, UK) was applied for all nonparametric multivariate analysis procedures, including calculation of Bray–Curtis similarity matrices, cluster analysis, SIMPER analysis, and ANOSIM.