Effect of Conservational and Conventional Tillage Systems on Functional Soil Archaea Diversity in Wheat-Pea Rotation Field

Soil borne archaea in agricultural systems is crucial for cycling of nutrient such as Nitrogen, Carbon and Sulphur. The objective of the study was to assess the effect of different tillage systems on functional archaea diversity in a 15 years cereal-legume rotation field using Illumina sequencing platform for archaea 16S rRNA gene. The treatments in the study included conventional tillage with stubble removed (T), no-till with stubble removed (NT), conventional tillage with stubble incorporated (TS) and no-till with stubble retained (NTS). The results showed that the dominant soil archaea phyla was Crenarchaeota (> 96%), followed by Euryarchaeota with a lower abundance of < 3% and then Parvarchaeota and other bacteria phyla made up < 1% across the treatments and depths. The treatment means were ranked as NT > NTS > TS > T for 16S rRNA number of OTUs, Shannon and Simpson indices calculated for the 0-10cm soil depth. Analysis of factor effect revealed that tillage but not stubble retention or their interaction significantly influenced (P < 0.01 and P < 0.05) 16S rRNA diversity. Non metric Multidimensional Scaling (NMDS) analysis clearly grouped the microbial communities according to depths. Linear Discriminant Analysis Effect Size (LEfSe) identified Crenarchaeota and Thaumarchaeota (to genus level) as significantly enriched clades in 0-10 cm depth of T while Euryarchaeota and Thermoplasmata were significantly enriched in TS. The conservational tillage systems (NT and NTS) promoted even distribution of archaea diversity while conventional tillage systems (T and TS) enriched the archaea communities identified in the study.


Introduction
Archaea constitutes a domain of microbes that exhibit diverse functional metabolic activity, habitat and typically adapted to chronic energy stress conditions than bacteria (Swan & Valentine, 2009).Archaea may obtain their source of energy from inorganic compounds such as ammonia or sulfur, use sunlight as a source of energy in non-oxygen-generating photosynthesis, or by autotrophic fixation of atmospheric CO 2 as a source of carbon (Pratscher et al., 2011;Zhalnina et al., 2012).However, their utilization of organic carbon source is still inadequately understood (Zhalnina et al., 2012).In non-extreme terrestrial environment such as agricultural soil, archaea play crucial functional roles in the cycling of nutrients such as carbon, nitrogen and sulphur (Jarrell et al., 2011) but also have the potential to contribute to atmospheric greenhouse gas emission (Offre et al., 2013).For instance, in the carbon cycle, methanogenesis and anaerobic methane oxidation are important intermediate steps that are performed exclusively by anaerobic Euryarchaeota and Methanomicrobia respectively (Offre et al., 2013).Also in the nitrogen cycle, Crenarchaeota (Thaumarchaeota and Candidatus Nitrososphaera) are reported to be involved in fixation of dinitrogen (N 2 ) gas, oxidation of ammonia to nitrite, and denitrification (Taylor et al., 2010).
The 16S rRNA marker gene has become the gold standard in molecular identification of archaea and bacteria diversity (Case et al., 2007).These small subunit ribosomal RNA genes are universally present in archaea and bacteria and contain both highly conserved fragments and variable regions that allow for the discrimination of different taxonomic levels (Vos et al., 2012).Illumina high-throughput sequencing platform has been employed to study microbial community composition and is reported to be accurate and rapid in the identification than traditional methods such as culture and serial dilution methods, Biolog system and biomarkers which often isolate only a few soil microbes (Klingler et al., 1992).Soil archaea are difficult to culture but molecular methods have been used to demonstrate the presence of archaeal 16S rRNA gene sequences in agricultural, grassland and forest soils (Navarrete et al., 2011).
In agricultural fields, tillage practices negatively affect soil chemical, physical and biological properties by exacerbating soil degradation which may consequently trigger changes in the soil microbial community structure and composition (Dorr de Quadros et al., 2012).It is widely accepted that no-till, residue retention, cover crops and crop rotations are conservation agriculture practices that protect soil, water, nutrients and increase microbial activity and biomass (Feng, 2003).However, varying research outcomes have been reported (Kaurin et al., 2015;Dong et al., 2017) making specific impacts of conservation agriculture practices on soil bacteria and archaea particularly complex to explain (Ng et al., 2012).
In the western Loess Plateau of China, conventional and traditional methods of soil preparation, application of Nitrogen fertilizers (Fan et al., 2005) and consistent crop residues removal from the fields are common farming practices (Lamptey et al., 2017) for the production of crops such winter and spring wheat, maize, soybean, potato among others (Nolan et al., 2008).These practices have worsened soil degradation processes, contributed to the decline of soil carbon and fertility (Zhang et al., 2016) and therefore, may impact microbial community structure.Conservation tillage techniques have been employed to improve soil physicochemical properties and greenhouse gas emissions for sustainable cultivation of spring wheat and field pea (Huang et al., 2013;Yeboah et al., 2016aYeboah et al., , 2016b;;Yeboah et al., 2017) but the rate of adoption is still relatively low in the region.The impact of tillage practices on soil archaea community structure has received low specialised research attention in the study area.Therefore, the study hypothesized that withdrawal of tillage and retention of stubble from previous crops in rotation systems will influence functional soil archaea diversity.We set out an objective to investigate the effect of different tillage systems on functional archaea communities in the 15 years crop rotation field using the 16S rRNA marker gene.

Site Description
The study was carried out in the Rainfed Agricultural Experimental Station of the Gansu Agricultural University (35°28′N, 104°44′E, elevation 1971 m above sea level), Dingxi, Gansu Province, Northwest China.The soil type in the site has a sandy loam texture with ≥ 50% sand and is locally known as Huangmian (Chinese Soil Taxonomy Cooperative Research Group, 1995), which is equated to the Calcaric Cambisol by the FAO soil classification (1990).The soil has organic matter content of < 14.75 g/kg (< 1.48%) and an average pH of about 8.44.The annual temperature ranges between -22-38 o C in January and July respectively while the average long-term annual rainfall in the area is 390.7 mm per year.

Design of Experiment
This research was carried out in 2016 on a long-term field experiment which was initiated in 2001.Before then, the site had a long history of conventional tillage and continuous cropping of flax (Linum usitatissimum L.).The experiment is a two factorial design with two phases of rotation and four tillage treatments (Table 1) arranged in a randomised complete block design with three replicates.Spring wheat (cv. Dingxi No. 35) and Field pea (cv.Yannong) were sown in rotation with both phases represented in each year for the past 15 years and is also known as double sequence rotation (W→P→W and P→W→P sequence).The crop rotation started with spring wheat in the first phase, followed by field pea while the second phase started with field pea followed by spring wheat in that sequence each year for the past 15 years.In conventional tillage (T) plots, all the stubbles were removed before ploughing to a depth of 10-20 cm.Conventional tillage with stubble incorporated (TS) plots had all stubbles from the previous crop returned to those same plots after threshing and then incorporated into the soil during ploughing.In the No-till (NT) treatment plots, no ploughing was performed and all the stubbles were removed at harvest whiles in No-till with stubble retained (NTS) plots, all the stubbles from the previous crops were returned to the original plots and retained on the surface of the soil.Each year, spring wheat was cultivated in early March at a rate of 187.5 kg ha -1 with a row spacing of 20 cm and harvested in late July to early August.Field Pea is sown in early April at a rate of 180 kg ha -1 with a row spacing of 24 cm and harvested in early July each year.Low rates of Nitrogen and Phosphorus fertilizers are applied at sowing with the no-till seeder at a rate of 105 kg N/ha as urea (46% N), 45.9 kg P/ha as calcium superphosphate (6.1% P) for spring wheat, 20 kg N/ha and 45.9 kg P/ha for field pea.

Soil Sampling
Soil samples were collected before seeding the field Pea phase of the experiment in April for the 2016 planting season.At sampling, three soil cores samples were randomly collected from each plot at the depth of 0-10 cm and 10-30 cm.The three soil cores from each plot were pooled to form a composite sample which was further divided into subsamples.The subsamples were stored on dry ice, conveyed to the laboratory and stored at -80 o C for molecular analysis.

Soil DNA Extraction and Amplification
Genomic DNA extraction from soil samples was done by the use MoBio PowerSoil® DNA Isolation Kit (MoBio Laboratories, Solana Beach, CA, USA) following the manufacturer's instructions.PCR amplification of V3-V4 of 16S rRNA (480bp with barcode) was executed using the primers 338F 5'-ACTCCTACGGGAGGCAGCA-3' and 806R 5'-GGACTACHVGGGTWTCTAAT-3' to identify archaea genes.Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific, Wilmington, USA) was used to determine the concentration of genomic DNA.Gel electrophoresis condition of 1% agarose gel and 120 V was used to run the gel for 30 min and subsequent visualization of the DNA fragments.

DNA Sequencing and Data Analysis
The 16S rRNA (V3-V4) hypervariable region of archaea was sequenced using Illumina Hiseq sequencing platform to obtain Paired-ends sequenced.PANDAseq software (Masella et al., 2012) was used to assemble the paired-end reads.Chimera were removed with USEARCH v7.1 (Edgar, 2010) by Denovo method and unique sequences were clustered at 0.03 cutoff (97% similarity) into representative operational taxonomic units (OTUs) using the UPARSE software (Edgar, 2013).Using UCLUST, the representative sequences were compared with the archaea 16S rRNA database of known species to classify each OTU.Alpha and Beta diversity analysis were performed for both treatments and samples using QIIME software.Diversity indices such as observed species, Chao1estimator, Shannon index, and inverse Simpson index were determined.Beta diversity analysis was based on the weighted and unweighted UniFrac evolutionary relationship or distance between species and was used to generate Non Metric Multidimensional Scaling analysis (NMDS) graphs.R statistical package (Kruskal-Wallis test function) was used to analyze Linear Discriminant Analysis Effect Size (LEfSe) to identify significantly (P < 0.05) different biomarkers between treatments and the impact on the species or clades with Linear Discriminant Analysis (LDA) score > 3. Molecular analysis, sequencing reactions and sequence analysis were out sourced in a commercial laboratory (Genepioneer Biotechnologies, Nanjing, China).SPSS software's (version 19.0; SPSS, Chicago, IL, USA) general linear model function was used to compare means, factor effect and interaction of factors on microbial diversity indices and their significant differences.
Post Hoc analysis was tested by Tukey's HSD Test (P < 0.05).

Results
A range of 11,093-20,601 clean tags were obtained per sample for archaea and bacterial 16S rRNA genes through sequence optimization and quality filtering.At 0.03 similarity cutoff (97%), a subsample of 11,093 16S rRNA clean tags gave 51 OTUs out of which 14 OTUs constituted 100% of the core microbiome in treatment samples.The OTUs were classified into 8 phyla, 8 classes, 8 orders, 5 families, 2 genus and other unidentified taxa at each taxonomic level.

LEfSe Analysis of Significantly Enriched Archaea Communities in Treatments
LEfSe analysis was used to identify significantly (p < 0.05) abundant and enriched microbial communities with LDA score > 3 in treatments and was presented from domain to genus levels (Figures 2a and 2b).Enriched microbial communities were only recorded in the two treatments of conventional tillage (T and TS).In the conventional tillage with stubble removed (T), Crenarchaeota (phylum to genus); Thaumarchaeota, Nitrososphaerales, Nitrososphaeraceae, Candidatus Nitrososphaera were enriched in the 0-10 cm soil layer.However, conventional tillage with stubble incorporated (TS) during ploughing enriched members of the phylum Euryarchaeota and class Thermoplasmata in the 10-30 cm layer.Crenarchaeota and Thaumarchaeota (previously a class of Crenarchaeota but now a novel phylum) were significantly enriched from phylum to genus in the 0-10 cm soil layer conventional tillage with stubble removed (T) treatment probably because of the chronic stress imposed by the annual ploughing of this field.The clade Thaumarchaeota predominantly comprise of ammonia oxidizing members (Zhalnina et al., 2012) that carry the amoA genes and oxidize ammonia using ammonia monooxygenase (AMO) enzyme during nitrogen cycling (Taylor et al., 2010;Shen et al., 2013).The other enrichment was detected in the conventional tillage with stubble incorporated (TS) plots at the 10-30 cm soil layer and the phylum Euryarchaeota and its class Thermoplasmata were the significantly enriched clades.Euryarchaeota constitute a diverse group consisting of extreme halophiles, thermophilic heterotrophs and anaerobic methanogens (Bapteste et al., 2005) that are involved in carbon cycling in their environment.Thermoplasmata is a recently discovered methanogenic class that can reduce methanol with hydrogen (Dridi et al., 2012) and may also use methylamines as substrate for methanogenesis (Poulsen, et al., 2013).
In the present study, the no-tillage systems (NT and NTS) recorded the highest archaea diversity respectively and our findings are congruent with Dorr de Quadros et al. (2012).Also, Dong et al. (2017) have recently reported that NT practices markedly enhanced the abundance of Thaumarchaeota (previously a class of Crenarchaeota but now a novel phylum) in the topsoil after 22 years of no-tillage managements in Northern China.On the contrary, archaea were found not to be affected by the different forms of tillage practices in the experiment conducted by Kaurin et al. (2015) who compared minimum tillage to conventional mouldboard ploughing.In no-tilled soil, physical disturbance is absent hence there is less soil aeration than in tilled soils and this may promote the growth of anaerobic microbes (Dorr de Quadros et al., 2012).Withdrawal of tillage in NT gave significantly high Shannon species diversity index and Inverse Simpson evenness index; however, only numerically high means were obtained for Chao1 richness estimator and number of observed species.Irrespective of the tillage practice, the archaea community was evenly distributed but the species diversity was relatively low across the treatments.In the study, treatment influenced clustering of microbial communities by NMDS was not obvious but clear depth wise distribution was observed.This may be explained by complex interaction of factors and mechanism that influence microbial activity and function in agro-ecosystem (Detheridge et al., 2016).

Conclusion
The dominant archaea phyla identified across the treatments and depths in the study site were Crenarchaeota, Euryarchaeota and Parvarchaeota, with Crenarchaeota making up > 96% of the total abundance.The treatments ranked as NT > NTS > TS > T for the various diversity indices calculated for 16S rRNA genes in the 0-10 cm topsoil.Factor effect analysis also revealed that tillage significantly influenced archaea diversity in this study.Conventional tillage with stubble removed (T) significantly enhanced the enrichment of Crenarchaeota and Thaumarchaeota in the 0-10 cm soil layer while Euryarchaeota and Thermoplasmata were stimulated in the 10-30 cm layer of tillage with stubble incorporated (TS).No-till promoted even distribution of archaea diversity but T and TS enriched the archaea clades detected in the study.The study results provide practical implications and benefits for the adoption of conservation tillage practices on soil microbes which are major drivers in the below ground plant-soil ecosystem.

Table 1 .
Treatment codes and names