|Main authors:||Giulia Bongiorno, Stefan Geisen, Else K. Bünemann, Lijbert Brussaard, Paul Mäder, Ron de Goede, Natacha Bodenhausen|
|iSQAPERiS editor:||Jane Brandt|
|Source document:||Bongiorno G., Bünemann E., de Goede R., Brussaard L., Mäder P. (2018) Screening of novel soil quality indicators. iSQAPER Project Deliverable 3.4, 66 pp|
Note: The following publication is based on the material contained in this section of iSQAPERiS
- Giulia Bongiorno, Natacha Bodenhausen, Else K. Bünemann, Lijbert Brussaard, Stefan Geisen, Paul Mäder, Casper W. Quist, Jean‐Claude Walser, Ron G. M. de Goede. 2019. Reduced tillage, but not organic matter input, increased nematode diversity and food web stability in European long‐term field experiments. Mol Ecol.; 28: 4987– 5005 https://doi.org/10.1111/mec.15270
In this article we report the preliminary results on the effect of soil management on the nematode communities studied with molecular methods. The statistical analysis is still in progress and will be used to detect relationships between the nematode communities and soil parameters and to identify specific nematode taxa which can be considered as indicators for certain management practices.
|2. Material and methods|
|3. Preliminary results|
|4. Preliminary considerations and future directions|
Nematodes are ubiquitous organisms which have an important role in soil functions such as nutrient cycling and pest and pathogen population regulation. They are present in all trophic levels, and can feed on microorganisms, insects, plants and nematodes themselves. In addition, they have different life strategies and for these reasons they can be divided in different functional groups (Bongers and Bongers, 1998; Neher, 2001; Yeates and Bongers, 1999). It has been shown that specific nematode groups or taxa respond differently to agricultural disturbances and management. For example fast-growing, bacterial-feeding nematodes are more associated with disturbed systems (Quist et al., 2016; Zhao and Neher, 2013). These characteristics make nematodes ideal soil quality indicators, principally for their linkage with soil functions and for their sensitivity to disturbances. Until recent years, the most popular method to study nematodes was with the traditional microscope approach. However, this method is time-consuming, requires specialists and is becoming more and more expensive. New molecular methods, which can assess nematode absolute abundances (qPCR) and community structure with relative abundances (sequencing) are promising tools that can help researchers, and ultimately land managers, to better understand the impact of soil management on soil functions. Molecular methods have the benefit of being high-throughput, fast and relatively cheap, especially compared to old microscopic and visual methods (Ahmed et al., 2016). Moreover, they can be used for targeted (with qPCR) or non-targeted (sequencing) studies and research questions.
The goal of our study was to assess if molecular methods can be used to detect nematode taxa that are specific to certain soil management and can, therefore, be used as indicators of soil functions and, ultimately, soil quality. Moreover, we wanted to associate these taxa to soil chemical, physical and biological parameters linked to soil functions. With this aim we assessed total nematode abundance with qPCR, nematode OTU richness, OTU diversity and nematode community structure with Illumina sequencing in ten European long term field experiments with different soil managements. We hypothesised that nematode abundance, richness and diversity will be increased by reduced tillage and high organic matter addition. We expected that in more disturbed agricultural systems (with conventional soil cultivation, addition of mineral fertilizer etc.) bacterial feeding, plant parasitic and fast growing nematodes will prevail over other nematode functional groups, i.e. fungivorous, carnivorous and omnivorous nematodes. We hypothesised that conventional tillage and low organic matter input will affect nematodes in such a way that nutrient cycling, decomposition, and population regulation are negatively affected.
2. Material and methods
Nematode extraction, DNA extraction and purification The 167 soil samples (from the 0-10 cm, 10-20 cm and 0-20 cm) stored at 3˚C were homogenised and nematodes were extracted from 100 g subsamples using an Oostenbrink elutriator (Oostenbrink, 1960). Ten samples for each LTE (a total al 100 samples) were analysed microscopically to count the number of nematodes expressed per 100 g of soil with the aim to correlate this with total nematode abundance measured with molecular method (qPCR). The nematode suspensions were subsequently concentrated and the dry concentrated suspension were lysed with proteinase K and β-mercapto-ethanol (Holterman et al., 2006). This lysis produced nematode DNA extracts that were then purified using a glass fibre column-based procedure (Ivanova et al., 2006; Vervoort et al., 2012). During the process of nematode DNA extraction we added an external control consisting in mammalian DNA which was subsequently used in the qPCR reactions to check the quality of the DNA extraction. All the purified DNA extracts were stored at -20°C until further use. Nematode extractions were done in the biological laboratory of the Soil Quality Department at Wageningen University, while nematode DNA extraction and purification was carried out in the molecular laboratory of the Nematology department at Wageningen University.
Quantitative PCR (qPCR) analysis of total nematode DNA
The purified extracts were used as templates in qPCR using a general primer to assess total nematode density and one for the external control mammalian DNA (used to check for losses of DNA during the extraction procedure and sampling handling). A standard curve was constructed by extracting DNA from a dilution series of nematodes. After the qPCR reaction, the Ct values obtained were converted into nematode densities using the linear relationship between the Ct values and the 10log (number of target nematodes) (Vervoort et al., 2012). We checked that the nature of the amplicon was correct, as estimated by the absolute values of the first mathematic derivative. Quantitative PCR analysis was carried out in the molecular laboratory of the Nematology department at Wageningen University.
18 SSU rDNA amplification and sequencing
The DNA of one part of the purified extracts was quantified with Nanodrop® and sent in dry ice to the company GenomeQuebec (Montreal, Canada) for 18 SSU rDNA amplification and sequencing on the Illumina platform. As a first step, a targeted PCR amplification with the tagged primers for 18S SSU rDNA was done (Table S17). The primers used to quantify the hypervariable eukaryotic V4 DNA region were: forward primer 3NDf (5'-GGCAAGTCTGGTGCCAG-3') and reverse primer 1132 mod (5'-TCCGTCAATTYCTTTAAGT-3') (Geisen et al., 2018). After this step, barcodes and Illumina adapter sequences were added to each sample (barcoding step) (Table S2). For each sample, the barcoding step was verified with gel electrophoresis. The DNA concentration was quantified with Quant-iTTM PicoGreen® dsDNA Assay kit (Life technologies) and for each sample, an equal amount of DNA was pooled for a sequencing library. After purification with AMPure beads (Beckman Coulter), the pooled DNA library was quantified using the Quant-iTTM PicoGreen® dsDNA Assay kit (Life technologies) and the Kapa Illumina GA with revised primers-SYBR Fast Universal kit (Kapa Biosystems). Average size fragment was determined using a LabChip GX (PerkinElmer) instrument. Sequencing was performed with MiSeq Reagent kit v3 (600 cycles) from Illumina. After sequencing, the sequences were demultiplexed by the company GenomeQuebec (Montreal, Canada) before the bioinformatic analysis.
DNA sequences were analyzed on the server from the Genetic Diversity Centre. After sequencing, the primer sites were trimmed and all the reads were trimmed to 280 nt with the Usearch software platform. Only the forward reads were used because the reads from the two primers did not merge. Subsequently, the reads were quality filtered with PRINSEQ-lite version 0.20.4 and error corrected with UNOISE III. Finally, the sequences were clustered into OTUs based on 97% similarity using UPARSE. The sequences were blasted to NCBI nt database. The ten top hits from each BLAST were collected to make a new database, which was used for taxonomic assignment with SINTAX .
Nematode richness and diversity (Shannon Index) were calculated for each site. Nematode richness was calculated as the sum of the species, and nematode diversity was calculated as the exponential of the Shannon Index:
where H is the Shannon diversity index, Pi is the fraction of the entire population made of species i, S is the number of species encountered, and Ʃ is the sum from species 1 to species S.
The difference in richness and diversity between the sites was tested with analysis of variance (ANOVA). The Tukey HSD post-hoc tests were used to assess significant differences between sites when the ANOVA indicated a statistically significant difference. All test results were considered statistically significant at p<0.05. Principle coordinate analysis (PCoA) was used to visualize the difference in community compositions between the different sites and canonical analysis of proximities (CAP) was performed to visualize the relationship with soil chemical, physical and biological parameters and nematodes communities. Bray-Curtis distances were calculated and used in the multivariate analysis. Spearman correlation was used to test the correlation between richness and diversity and the chemical, physical and biological indicators.
Two groups of LTEs were created and used for subsequent analysis on the effect of tillage and organic matter addition on nematode communities.
- The first group (Group 1) included LTEs in which we sampled the layers 0-10 cm and 10-20 cm: CH1, CH2, NL1, NL2, SL1, HU4 and ES4.
- The second group (Group 2) included LTEs where the layer 0-20 was sampled: CH3, PT1 and HU1.
In the first group we assessed the effect of tillage and organic matter addition, while in the second group we assessed only the effect of organic matter addition because all three LTEs had conventional tillage. Correlations between total nematode abundance assessed with microscope and total DNA abundance were visualised with Excel.
Total nematode DNA, nematode richness and diversity was calculated also for these two groups as described above and the effects of management practices were assessed using linear mixed effect models. Mixed models were used to take into account the possible correlations introduced by the multi-site field experiments and to generalize the effect of the management practices across the different LTEs (Bradford et al., 2013; Lucas and Weil, 2012). The tillage and/or the soil organic matter input and, if present, the layer, their two-way and possibly three-way interactions were used as fixed factors. Random effects for trials, blocks, main plots and subplots were introduced in the models, to represent the experimental designs of the different trials. The effect of the soil pedoclimatic zone was not included in the fixed part of the model because we were interested in the management effects across the pedoclimatic zone. Moreover, the number of LTEs was limited and they were biased towards heavy soils, which could strongly bias the analysis.
The effects of tillage and organic matter addition and their interaction on the nematode DNA quantity, nematode richness and diversity were addressed by performing an analysis of variance (function anova) on the fitted linear mixed effect model. The model assumptions of normality and homogeneity of variances of the residuals were checked both visually and with the Shapiro-Wilk and Levene’s tests. Total nematode abundance assessed with qPCR was squared root transformed and nematode richness and diversity were log-transformed in order to meet the assumption of normality. All test results were considered statistically significant at p<0.05. We tested and visualized (if present) the effect of tillage, organic matter and layer on the nematode communities with CAP. In this analysis, we put the LTE as a conditional factor in order to control for the effect of the pedo-climatic zone on the nematodes communities. We tested for management effects using ANOVA with 999 permutations. We performed the CAP for both layers together and for the two layers separately, in order to investigate the interaction between layer and tillage.
All statistical calculations were carried out using R version 3.3.2 (R Development Core Team, 2013). For the linear mixed effects model, the packages nlme, and emmeans were used, while for the correlation analysis the packages car and stats were used (Pinheiro et al., 2018).
3. Preliminary results
Long Term Experiments
The LTEs had different total nematode abundances per 100 g soil assessed with qPCR, and richness and diversity of the community composition (Figure S4). The LTEs CH3, NL2 and HU4 had the highest total nematode abundance assessed with qPCR. The LTEs CH3, SL1, ES4 and HU4 showed more OTU richness, while CH3, NL1 and ES4 showed greater OTU diversity. Moreover, the community composition was clearly clustering based on the LTEs and their soil parameters (Figure 6). The nematode community profiling yielded a total of 6,057,917 nematode sequences, with the range of number of sequence per samples being 10188 – 91624, and the median being 36179. We found 835 nematode operational taxonomic unit (OTUs) across all samples.
Effect of soil management on total nematode DNA
The mixed model performed on group 1 revealed that neither tillage nor organic matter additions had an effect on the total nematode DNA measured with qPCR (Table 12). However, we found that there was a higher quantity of nematode DNA in the first layer (0-10 cm) compared to the second layer (10-20 cm) but only in RT (Table 11). Regarding group 2, we found that the organic matter addition did not have an effect on total nematode DNA (Table 13). The correlation between total nematode abundance assessed with qPCR and other main soil chemical, physical, chemical and biological parameters did not show very strong and significant results (Table S18). Remarkably, some of the labile organic carbon fractions (POXC, HWEC and POM-C) resulted to be the parameters more strongly correlated with total nematode abundance, especially in the first layer (Table S19). The relationship between the number of nematodes manually counted with the microscope and the total number of nematodes measured with qPCR was very week (R2=0.26) (Figure S5).
Table 12. Results of the mixed linear model for the trials of group 1 (CH1, CH2, NL1, NL2, SL1, HU4 and ES4) where we assessed the effect of tillage (CT vs RT) and organic matter addition (LOW vs HIGH) on total nematode DNA abundance per 100g of soil, OTU richness (total number of OTUs), and OTU diversity (Shannon diversity index-1) (number of observation=132). In the upper part of the table are reported the estimated means from the model and confidence intervals in parenthesis (95%). Different letters following mean and se have to be read per columns and per layer and they show treatments which are significantly different (p<0.05) according to Tukey post-hoc test. In the lower part of the table F statistics and p-values for the main factors and their interactions are reported.
Table 13. Results of the mixed linear model for the trials of group 2 (CH3, PT1 and HU1) where we assessed the effect of organic matter additions (LOW vs HIGH) on total nematode DNA abundance per 100g of soil, OTU richness (total number of OTUs), and OTU diversity (Shannon diversity index-1) (number of observation=35). The upper part of the table shows the estimated means from the model and confidence intervals in parenthesis (95%). In the lower part of the table F statistics and p-values for the main factors and their interactions are reported.
Effect of management on nematode communities assessed with Illumina sequencing richness and diversity
We found higher nematode richness and diversity in RT compared to CT across the trials of group 1 (p<0.05) (Table 12). In this analysis, we also found that OTU richness and diversity were affected by soil layer, with higher values of these indexes in the upper than in the lower layer (Table 12). Moreover, OTU diversity resulted to be lower in the high organic matter input plots (p<0.05). However, in the analysis of group 2 where only the effect of the organic matter input was analysed, we did not find a difference between low and high organic matter input (Table 13).
Richness and diversity calculated in each sample (n=167) were not strongly correlated with chemical, physical, biological and biochemical indicators described in the material and methods section (Table S20). The same was found if only the samples from the first layer were used for the correlation (Table S21).
Nematode community composition
We used constrained analysis of principal coordinates (CAP) to analyse if the management had an effect on the soil nematode communities. Partial CAP of group 1 - constrained by tillage, organic matter input and layer - revealed that tillage, organic matter input, layer and the interaction between tillage and layer had a significant effect on the nematode communities (p<0.05) (Figure 7).
In order to clarify the interaction between tillage and layer, we ran the partial CAP analysis on the two layers separately. The CAP revealed that tillage and fertilization had a significant effect on the nematode community in the first layer (Figure 8-A) as well as in the second layer (Figure 8-B).
β-diversity can be driven by true biological differences, differences in group dispersion, or both (Anderson, 2001). For this reason we tested for differences in group dispersion for nematode communities in group 1 and group 2. The dispersion tests were not significant, suggesting that differences between managements were driven primarily by true biological differences and not by an artifact of the differences of the within-group dispersion (Table S22). The partial CAP of group 2 revealed that organic matter addition did not influence the nematode community (Figure 9), even if the P value was close to significant.
4. Preliminary considerations and future directions
As expected from previous analysis, we found that the effect of the pedo-climatic characteristic on the nematode communities was stronger than the effect of the management. For this reason, we tried to eliminate the influence of the pedoclimatic zones using mixed linear models and constrained analysis of principal coordinates (CAP) adding the LTE in the random part of the model.
Preliminary results show that tillage has an effect on nematode communities in group 1. We found more richness and diversity in RT compared to CT, but the effect of tillage on the nematode community assemblage was even stronger. Moreover, we found a strong effect of the layer on nematode abundance assessed with qPCR, richness and diversity of nematodes, with higher values of these parameters in the upper (0-10 cm) than in the lower (10-20 cm) layer. Microbial and biotic parameters are more abundant and active in the upper soil layers compared to the deeper soil layers most probably because of the higher abundance of oxygen, nutrients and water which can be used by the organisms (Wall et al., 2013). Moreover, plant roots often concentrate in the first soil layer, influencing the distribution of soil organisms. Also the nematode community assemblage assessed by CAP was affected by the layer, but only in RT. This result does not come unexpected, since in the RT the two layers are not mixed with the action of ploughing as in CT. Also for other parameters (i.e. labile carbon fractions) we found a stratification effect in RT which was not present in CT. We also found an effect of organic matter additions on nematode community assemblage, but only in the first layer of group 1.
The next step of the analysis will be to identify which nematode taxa (also in terms of feeding behaviour) are specific for specific agricultural management with the indicspecies package in R. Moreover, we will determine the relation between these nematodes indicators and chemical, physical and biological soil parameters.
Note: For full references to papers quoted in this article see