Cover crop-driven shifts in soil microbial communities could modulate tomato early crop growth via plant-soil feedbacks

doi:10.21203/rs.3.rs-1005617/v1

Download PDF

Research Article

Cover crop-driven shifts in soil microbial communities could modulate tomato early crop growth via plant-soil feedbacks

https://doi.org/10.21203/rs.3.rs-1005617/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Sustainable agricultural practices such as crop diversification, cover crops and residue retention are increasingly applied to counteract detrimental effects of agriculture on natural resources. Since part of their effects occur via changes soil microbial communities, it is critical to understand how these respond to different practices. Our study analyzed five cover crop (cc) treatments (oat, rye, radish, rye-radish mixture and no-cc control) and two crop residue management strategies (retention/R+ or removal/R-) in an 8-year diverse horticultural crop rotation trial from ON, Canada. Cc effects were small but stronger than those of residue management. Radish-based cover crops tended to be the most beneficial for both microbial abundance and richness, yet detrimental for fungal evenness. Cc species, in particular radish, also shaped fungal and, to a lesser extent, prokaryotic community composition. Crop residues modulated cc effects on bacterial abundance and fungal evenness (i.e., more sensitive in R- than R+), as well as microbial taxa. Several microbial structure features, some affected by cc, were correlated with early tomato growth in the following spring (e.g., composition, taxa within Actinobacteria, Firmicutes and Ascomycota). Our study suggests that, whereas mid-term cc effects were small, they need to be better understood as they could be influencing crop productivity via plant-soil feedbacks.

Agroecology

General Microbiology

fungi

bacteria

sequencing

crop diversification

residue removal

Cover crops increase the number of plant species in crop rotation (i.e., diversification), extend the time under living plant cover (i.e., ‘perennialization’ of annual cropping systems), and act as green manure. These aspects aim to resemble some aspects of natural ecosystems such as their higher diversity, complexity and functionality, which make them less reliant on external inputs ¹. Thus, it is not surprising that cover crops can provide benefits such as increasing soil organic carbon and nutrients, improving soil structure, minimizing soil erosion and suppressing weeds ^2–4. Under certain conditions, cover crops may also increase crop yields and productivity ^5–7.

These changes in soil properties and plant growth occur mainly via changes in the soil biota ⁸. For instance, cover crops were shown to shift soil microbial community composition ^9–11, and increase microbial abundance and diversity ^3,12,13. They may also have functional repercussions by favoring or affecting specific microbial functions or life strategies ^9,14,15. Still, the nature of soil biological processes is complex and results are modulated by site-specific factors such as soil type, climate and management practices ^2,12,13. The influence of cover crop (cc) species composition on soil microbial communities is not yet fully clear, but different cc species have been found to favor specific microbial groups ^11,16,17. Yet, in other studies, cc species did not affect soil microbial biomass, composition or functional traits markedly ¹⁷, or their effect was surpassed by other cc management strategies ¹⁸. Interactions with other agricultural practices (e.g., tillage, residue management) could also modulate the response of soil properties and crop productivity to cc implementation ^5,17,19.

Previous studies in this 8-year horticultural crop rotation suggested that repeated utilization of some cover crops could improve soil health and crop growth ^6,19,20. The goal of this study was to better understand the microbial mechanisms underlying those changes by studying the structure of soil microbial communities. Besides a no-cc control, cc treatments consisted of oat, rye, radish, and a mixture of the last two species, which present contrasting root structure and nutrient requirements ²¹. While most microbiome studies in the literature evaluate prokaryotes and short-term responses to cc (i.e., single growing season), this is a medium-term experiment (i.e., 6 times in 8 years) that encompasses both prokaryotes and fungi. Microbial community structure was studied using quantitative PCR (qPCR) to assess abundance, and high-throughput sequencing to assess diversity and community composition. These effects were evaluated in interaction with the removal (R-) or retention (R+) of winter wheat crop residues, which sometimes must be removed for agricultural purposes ^19,20.

We hypothesized that both the inclusion of cover crops (i.e., cc vs no-cc) and residue retention would shift microbial community composition and increase microbial abundance via changes in the belowground environment (e.g., C and nutrient inputs, moisture, temperature fluctuations). In addition, cc species with different morphological and functional traits would promote different microbial groups, leading to changes in community composition and potentially increasing alpha diversity. For all variables, we expected to find some degree of interaction between cover crops and crop residue management, since they both act as sources of C and nutrients, and both can modify the soil physical environment. Finally, even though soil microbial communities are known to undergo seasonal fluctuations, we expected some degree of relationship between microbial structure in the fall and crop growth in the following spring (i.e., plant-soil feedbacks). In particular, specific taxa or functional groups present in the fall could have a role in soil functions or plant-microbe interactions the following crop season.

2.1- Response of soil prokaryotic communities to cc and residue treatments

Cover crops affected bacterial abundance only in R-, where no-cc was lower than radish and marginally lower than rye-radish (P=0.005 and P=0.051, respectively) (Fig. 1a). Intermediate values were found for cereal cover crops (oat and rye) (Fig. 1a). Prokaryotic alpha diversity was virtually unaffected by the cc and residue treatments, with the exception of no-cc soils having higher evenness in R+ than R- (Fig. 1b, Table S1). Still, Venn diagrams showed that rye-radish, followed by rye, was the cc treatment with the highest number of total and unique ASVs (Fig. S2)

No significant effects of cc or residue management on the phylogenetic composition of prokaryotic communities were detected with PERMANOVA (Table S2), but some cc effects were visualized via partial CAP (Fig. 2a). The largest compositional differences were observed between oat and rye-radish, followed by oat and no-cc (CAP1=31.6%, Fig. 2a). Oat also presented higher beta diversity than rye-radish (Fig. 2b). On the other hand, CAP2 (25.0%) was explained by the presence of radish, with rye-radish and radish separating from oat, no-cc and rye alone (Fig. 2a). Still, soil prokaryotic communities presented high unexplained variability (unconstrained inertia=74.18%) and high dispersion within cc treatments (Fig. 2b).

Cover crops also changed the relative abundance of some prokaryotic taxa. In terms of bacterial phyla, radish promoted Actinobacteriota, rye promoted RCP2-54, and cereals were detrimental to Firmicutes (Fig. 2c). Among archaea, Thermoplasmatota was lower in no-cc and Crenarchaeota was higher in radish. The response of some phyla was modulated by residue management (e.g., Crenarchaeota, Verrucomicrobiota, GAL15, Spirochaetota, Nanoarchaeaeota and Zixibacteria) (Fig. 2c). Even phyla with a more consistent response between residue managements (e.g., RCP2-54, Firmicutes) were generally more sensitive in R+ than R- (Fig. 2c). At the genus level, indicator species analysis identified 37 bacterial and 2 archaeal taxa associated with specific cover crops, although half of them were detected in either R- or R+ (Table S3). Overall, oat and radish had less indicator taxa while the opposite was observed for rye-radish (11 and 16 vs. 27, respectively) (Table S3). Besides, ALDEx detected a large number of taxa associated to cover crops, among which 12 were also indicator species (Table S4). One of these was an unidentified Nitrososphaeraceae archaeon with higher relative frequency (%) in radish soils (Fig. 3b). Besides this, and a slightly higher frequency of Actinobacteria and lower Acidobacteria in radish, treatments presented similar proportions of prokaryotic taxa (Fig. 3a, 3b).

2.2- Response of soil fungal communities to cc and residue treatments

The response of soil fungal communities to cc and residue treatments often contrasted that of prokaryotes. Fungal abundance was not affected by cc or residue management, although in R- it tended to be lower in no-cc and oat (Fig. 1c). Even though fungal richness was not sensitive to the applied treatments (Table S1), radish-based cover crops presented the highest number of total and unique ASVs (Fig. S2). Fungal evenness was affected by cover crops only in R-, where no-cc was higher than radish (P=0.011) and rye-radish (P=0.049), and oat higher than radish (P=0.011) (Fig. 1d). Also, no-cc presented higher fungal evenness in R- than R+ (Fig. 1d).

Fungal community composition was affected by cover crops (R²=0.22, P=0.042) but not by residue management (PERMANOVA, Table S2). CAP indicated that cc-based differences were mostly explained by the presence of radish (CAP1=23.0%), followed by rye (CAP2=22.0%) (Fig. 2d). CAP1 showed not only a cc gradient (oat<rye<no-cc<rye-radish<radish), but also a slight distinction between R+ and R- (Fig. 2d). Radish soils also presented lower beta diversity than no-cc, oat and rye (Fig. 2e). Despite these effects, fungal community composition also exhibited a large proportion of unexplained variability (unconstrained inertia=76.91%) and relatively high dispersion within cc treatments (Fig. 2e).

Both cc and residue treatments affected the relative abundance of fungal taxa, again with interactions between them. Oat and no-cc increased Mortierellomycota and Ascomycota, and, in R+, Basidiomycota (Fig. 2f). Oat also favored unidentified fungi while, in R-, both cereals increased Basidiomycota and reduced Kickxellomycota (Fig. 2f). In R+, Kickxellomycota was negatively affected by all cc species compared to no-cc (Fig. 2f). Olpidiomycota, one of the most sensitive phyla, was lower in no-cc, followed by both cereals in R- and only by oat in R+. Indicator species analysis found 20 genera associated with the different cc treatments, half of them in either one of the two residue treatments (Table S5). Similar to prokaryotes, oat presented the lowest number of indicator taxa (5 out of 20) (Table S5). Besides, ALDEx detected 24 fungal taxa associated to cover crops, 5 of which were also indicator species (Table S6). As opposed to prokaryotes, changes in the relative frequency of fungal taxa were clearer (e.g., higher Olpidiomycota and lower Ascomycota in radish, rye and rye-radish) (Fig. 3c). Radish, which had the lowest fungal evenness, presented a high frequency of an unidentified Olpidiaceae (~21%) and, especially in R-, Cercophora sp. (~10%) (Fig. 3d). This Olpidiaceae fungus, absent in other cc treatments, was an indicator species for radish (Fig. 3d, Table S5). Contrarily, cereals, but mostly rye, presented higher proportion of Olpidium brassicae (rye R+: 23.47%, rye R-: 7.58%) (Fig. 3d).

Retaining crop residues (R+) increased the relative abundance of Glomeromycota and Rozellomycota, as well as lower Zoopagomycota, with some variation between cc treatments (Fig. 2f). Mortierella was a highly abundant genus in most samples (14.6-29.4%), and, except for rye, it was more dominant in R+ than R- (Fig. 3d). In no-cc, other taxa were also more dominant in R+ than R-, consistently with the lower evenness: Rozellomycota (3.19% vs. 0.80%), Myrmecridium sp. (3.02% vs. 0.19%), Periconia macrospinosa (3.2% vs. 0.47%), Solicoccozyma terrea (2.61% vs. 1.65%), Acremonium furcatum (3.00% vs 1.97%), and unidentified Chaetomiaceae (0.62% vs. 0.17%) (Fig. 3c, 3d).

2.3- Relationship between soil microbial communities and plant growth

At the time of sampling (October 2015), cover crops presented different levels of biomass, as follows: rye<oat<rye-radish<radish (P<0.001) (Fig. 4a). Cc biomass was positively correlated with soil bacterial abundance (Spearman r (r_s)=0.50, P=0.001) and, to a lesser extent, with fungal abundance (r_s=0.30, P=0.037) (Fig. S3). In addition, higher cc biomass was associated with higher fungal ASV richness (r_s=0.39, P=0.014) and lower fungal evenness (r_s=-0.46, P=0.003) (Fig. S3). Correlations were also found between cc biomass and microbial community composition (PCoA axes), especially for the fraction of explained by cc treatments (CAP axes) (Fig. 4b-4e). In both cases, correlations were detected with the PCoA/CAP axes that distinguished radish-based from other cover crops (e.g., Fig. 2a, 2d). In the case of fungi, when excluding no-cc, a positive correlation was also found between cc biomass and CAP2 (r_s=0.51, P=0.003). Correlation values were always stronger when excluding no-cc, which we considered as having no aboveground biomass at sampling (Fig. 4b-4e).

We also looked for a relationship between soil microbial community data and early crop biomass of the following’s year tomato crop, after transplant (July 2016). Early crop biomass was higher with radish than no-cc (P=0.002), particularly in R+, with no clear differences among other cover crops (Fig. 4f). There was some degree of correlation between early crop biomass and cc biomass (r_s=0.30, P=0.06), especially when excluding no-cc (r_s=0.40, P=0.02). Early crop biomass was also positively correlated with fungal abundance (r_s=0.35, P=0.026) and ASV richness (r_s=0.47, P=0.002), and marginally with bacterial abundance (Pearson r=0.29, P=0.06) (Fig. S3). Microbial community composition, both unconstrained (PCoA axes) and constrained (CAP axes), was also associated with early tomato growth (Fig. 4g-4j). These correlations were found for the same axes as the ones correlated with cc biomass, as well as fungal CAP2 both including (r_s=0.33, P=0.036) and excluding no-cc (r_s=0.38, P=0.034). In all correlations with early crop biomass, no major changes were observed when including/excluding no-cc from the analysis (Fig. 4g-4j).

Finally, we explored which taxa, among those correlated with early crop growth, were also sensitive to cc treatments. For example, Actinobacteriota, Crenarchaeota and Firmicutes were positively correlated with early crop biomass, while the opposite relationship was found for Spirochaetota and Zixibacteria (Fig. 5). At the genus level, 36 prokaryotic taxa were both sensitive to cover crops and correlated with tomato growth (20 negatively and 16 positively) (Fig. S4, Table S7). More than half of these taxa belonged to 4 out of 16 phyla: Actinobacteria (8), Proteobacteria (4), Chloroflexi (4) and Acidobacteria (4) (Table S7). Besides, only 12 of them belonged to phyla also associated with early crop growth (Fig. 5). Notably, except for Actinocorallia, Actinobacteria showed a positive relationship with crop growth (e.g., Nocardioides, Iamia, Streptomyces, Gaiella and Rubrobacter) (Fig. 5, Table S7). Contrarily, all Acidobacteria and three out of 4 Chloroflexi, most of them uncultured, were negatively associated with early tomato growth (Fig. 5, Table S7). Within the archaeal phylum Crenarchaeota, members of the family Nitrososphaeraceae showed a positive correlation with crop growth, while the opposite was found for a member of Nitrosopumilaceae (Fig. 5, Table S7). In fungal communities, 2 phyla (Ascomycota and Mortierellomycota) and 4 genera within these phyla were negatively related to early crop biomass while also being affected by cover crops (Fig. 5, S4, Table S7). Three of these taxa belonged to the phylum Ascomycota (Fusarium, Collembolispora, Humicola) and the fourth one, Mortierella, to Mortierellomycota. Overall, these prokaryotic and fungal genera were relatively abundant (clr>0) and, as expected, those positively associated with crop growth were generally higher in radish or radish-based cover crops and vice versa (Fig. S4, Table S7).

Cover crops can shape the soil microbiome by providing C and nutrients via rhizodeposits and litter, by secreting signaling compounds in root exudates ²², or by modifying the soil abiotic environment. In this study, cover crops affected some aspects of soil microbial structure, consistently with other soil biological, chemical and physical properties ²⁰. These structural changes were driven by plant species identity more so than by cc inclusion or increased plant richness in the rotation. Generally, radish was the cc species exerting the strongest effects on microbial structure. Consistently, radish had higher aboveground biomass at sampling, it produces antifungal compounds that inhibit arbuscular mycorrhizal fungi (AMF) and other fungi ¹³, and it has a distinct root morphology and nutrient requirements compared to cereals ²¹. Still, microbial response was relatively minor considering cover crops had been applied for 7 successive years and were also present at sampling, suggesting neither short- or mid-term effects were strong. It is possible plant-driven effects were more evident in the root vicinity (e.g., rhizoplane and rhizosphere), without having extended to the bulk soil ²³, although microbes in these compartments could also be unaffected by plant species identity ²⁴. Cover crop effects may also be less defined in a diverse rotation like this one (i.e., horticultural and grain crops), where microbial communities are more adapted to fluctuations in plant species ²⁵. In fact, it is not uncommon that cc effects on soil microbial communities are surpassed by other practices such as fertilization, tillage or crop rotations ^9,10,15,24. Lastly, small cc effects could also be a consequence of timing if microbial response fluctuates with the season ^19,26, or as cover crops grow and their residues decompose ^14,18.

Increasing plant richness and the time under living plant cover via cover crops can have positive effects on microbial biomass and abundance ^{9,10,12,17,27,28}. In our study, changes in microbial abundance were small, only significant in bacteria, and dependent on the cc species ^17,29,30. The strongest driver of cc-driven changes in bacterial abundance seemed to be the overall supply of C via litter and rhizodeposits ^9,13, considering it behaved similarly to aboveground biomass ^16,27, and was only sensitive in R-. Input quality seems to be less relevant, considering the C:N ratio was relatively stable ⁶, although belowground inputs may have been more relevant ²⁸, especially considering fall cover crops generally present high root to shoot biomass ratio ⁴. We also predicted microbial diversity could be enhanced by the increased plant and resource diversity of cc treatments ³¹, although this was observed only as a trend in rye-radish soils. So far, clear positive effects of cc on microbial diversity are limited ¹⁰ or small ¹². Moreover, microbial communities subjected to a diverse crop rotation may be less sensitive to the addition of one or two extra species via cc ²⁵. Additionally, there might be other reasons why microbial diversity does not respond to cc or cc diversity ^9,26,32, such as high weed diversity or specific cc termination practices ^12,15.

Without affecting richness, radish reduced fungal evenness by disproportionally promoting some fungi over others (e.g., unidentified Olpidiaceae and Cercophora sp.). While Olpidiaceae fungi are potential brassica pathogens ³³, saprotrophs like Cercophora may have become enriched due to a preference towards radish inputs. This hypothesis is supported by changes in evenness (and Cercophora proportion) being less clear in R+, where wheat residues acted as an alternative C source. Substrate preference could be driven by differences in quality, but we do not have these data for belowground inputs. Still, higher quality could be expected on radish inputs, as it tended to have higher N content ⁶ and usually has higher P content ³⁴ than cereal inputs.

Soil microbial composition can be shaped by plant species identity and functional traits ^35,36, and similar effects were observed with cover crops ^{9–11, 26}. In this study, microbial composition shifts were small and mostly driven by presence-absence changes, especially for prokaryotes. The latter suggests they were probably driven by changes in rarer organisms ³⁷, as found by Cloutier et al. ³². The main differentiation of prokaryotic communities was driven by oat, mostly due to a loss of taxa/genomes, especially in R+, but also to a higher beta diversity (i.e., between oat plots). The latter may have been the most important factor, considering PERMANOVA, which is not always sensitive differences in dispersion, did not detect significant cc effects. Prokaryotic communities also showed changes between radish and rye that may be caused by differences in aboveground biomass or input quantity and quality, but also by N uptake and availability. In contrast, the main distinction in fungal community composition was driven by radish-based cover crops, possibly driven by the morphological and functional aspects discussed earlier. Rye-based cover crops also had a strong influence in fungal community composition, although it is not clear what is behind these effects. We hypothesize these could be related to the lower N uptake of rye and to the fact that it was the only cc that overwintered, hence subjecting soils to longer time under the influence of living roots.

Microbial shifts were also reflected in the relative abundance of specific phyla and genera. Some bacterial taxa were favoured (i.e., Pedosphaeraceae, Burkholderiales) or excluded (i.e., Desulfuromonadia, Zixibacteria, Sandaracinaceae, Chloroflexi) with radish, but their functional traits remain unknown for being uncultured and/or partially identified. The exception was Ammoniphilus, a bacterial genus adapted to high ammonium concentrations whose only source of C and energy is oxalate ³⁸. Even though brassicas generally have antifungal activity ¹³, radish promoted two potential pathogens of this plant family: Olpidiaceae and Leptosphaeria ³³. In cereal soils, especially rye, unidentified Olpidiaceae seemed to be replaced by O. brassicae, an obligate plant pathogen commonly found in brassicas, which can survive in bulk soil as spores ³³. This result seems contradictory but it is possible that the unidentified Olpidiaceae are in fact O. brassicae, while the O. brassicae in rye could either be a genotype adapted to a different host ³⁹ or O. virulentus, a related species more frequently found in non-brassicas ³³. Neither Olpidium nor Leptosphaeria, however, are likely pathogens of tomato, which may explain why radish was still the treatment with highest early crop growth. Other taxa that may also be behind the positive effect of radish on tomato growth. Finally, we noticed Glomeromycota fungi were not markedly affected by radish, despite brassicas not associating symbiotically with AMF and sometimes affecting their populations ¹³. Most probably, main crops in the rotation kept soil AMF populations stable.

Crop residues constitute a source of aboveground C and nutrient inputs, but also modify soil temperature and moisture content, hence shaping the environment surrounding soil microbiota ⁴⁰. Consistent with previous results on biological, chemical and physical soil properties ²⁰, residue management effects in this study were smaller than cc effects, and negligible for most of the measured variables. This was not surprising considering that, throughout the 8-year trial, cover crops were applied 6 times but crop residues were removed only twice ¹⁹. Still, we expected residue effects to be clearer since this treatment was applied during the sampling year. Another reason behind the smaller effects of residue management may be that crop residues represent a more limited source of C and nutrients, while growing cover crops provide multiple sources of inputs (e.g., aboveground residues, belowground residue, rhizodeposits). This is supported by previous findings highlighting the importance of belowground inputs of cover crops as a C source for soil microorganisms ^28,41.

In spite of residues being incorporated via tillage ~2 months before our sampling, microbial abundance did not increase in R+. It is possible that the timing of our sampling missed a transient increase in microbial abundance ⁴². Alternatively, incorporating residues could have accelerated mineralization, with a large proportion of C being lost as CO₂ instead of used for microbial growth ⁴⁰. This explanation is supported by the higher C mineralization levels measured one month before sampling ²⁰. In terms of composition and taxa, crop residue effects were stronger on fungi than prokaryotes, possibly because fungi are more capable of breaking down plant cell wall polymers (e.g., lignin, cellulose) such as those found in straw. Surprisingly, residue retention modified the relative abundance of three phyla hosting few or no saprophytic taxa: it favored Glomeromycota (AMF, obligate symbionts) and Rozellomycota (mostly animal and protist parasites), while reducing Zoopagomycota (animal, protist and mycoparasite, sometimes with saprophytic capacity) ⁴³. These effects may have been indirect, via other soil microbes, standing cover crops and/or previous crops. Also, in the case of AMF, recent studies suggest they could remain viable after the host shoots are removed ⁴⁴ and even participate in organic matter decay ⁴⁵.

Fallow soils (no-cc) were the only ones to experience an increase in prokaryotic evenness and a decrease in fungal evenness when retaining crop residues (R+). Since prokaryotic evenness increased together with bacterial abundance, it seems a wide array of organisms were favored by R+ conditions, whether it was the overall higher C and nutrient availability after incorporating crop residues, or the increase in temperature that could follow tillage. Contrarily, the incorporated wheat straw favored only a few soil fungi over others. Some fungi may have been increased during the previous wheat crop and, in treatments where no other substrates were available (i.e., no-cc), they remained dominant in the retained residues. This could be the case of Fusarium sp., Periconia macrospinosa and Mortierella exigua, all of which were more dominant in no-cc R+ than R-. While the first two are potential plant pathogens or endophytes, Mortierella can live both in a saprotrophic or a root-associated lifestyle.

Despite the small effects of residue management, our results suggest that this treatment could modulate the response of soil microbial communities to cover crops and vice versa. Since residue management did not affect aboveground cc biomass, we believe such modulation did not occur via plant growth. On the contrary, it seemed to be explained by the fact that both cover crops and residue act as C and nutrient sources for the soil biota, as previously reported ⁴⁶. In our study, for example, detrimental effects of no-cc and oat on soil bacterial abundance were buffered when retaining crop residues (R+), probably by providing an additional C and nutrient source in treatments where it was limiting. A similar buffering effect was observed on fungal evenness, as cc effects were smaller in R+ than R-. Wheat residues reduced evenness in no-cc, as discussed above, but they also increased it in radish-based treatments, by acting as an alternative substrate that could feed a wider range of fungi in radish-based treatments. Residue management also modulated taxonomic changes, although not consistently for all taxa.

Early growth of tomato crops after transplant can provide a critical benefit for crop establishment. Here, early growth was highest under radish cover crops, independently of the residue management applied. This response could be driven by several abiotic and biotic changes induced by the presence of radish, such as its positive effects on spring soil mineral N (Table S8). Compared to the other cover crops, radish presented higher aboveground biomass which, due to its higher quality (N and P content), can be rapidly mineralized in the spring ^7,34. As we showed earlier, radish also induced some structural shifts in microbial communities, which may have been triggered by cc input quality ⁴⁷. Previously, we discussed how plants can modify the soil environment and microbiota. At the same time, soil organisms can modify the physical and chemical environment where plants live and interact with plants directly or indirectly. Hence, it is not surprising to find links between soil microbial communities and plant growth and productivity ^22,48.

In our study, correlation analyses detected associations between early crop growth and soil microbial composition and specific microbial taxa, as well as fungal abundance and richness, although only the first two were affected by cc treatments. Benefits of increased fungal biomass on plant growth could be observed if, for example, it enhanced organic matter cycling and aggregation. On the other hand, increased fungal diversity could also increase functional capacity, or even disease suppression. Changes in community composition may have shifted the soil and plant-associated microbiome of tomato crops ⁴⁹, which could affect their health and fitness (e.g., if beneficial microorganisms are promoted over pathogenic ones) ^47,50. These variables were not equally affected by cc species but, generally speaking, radish-based treatments were the ones both favoring fungal abundance, diversity and potentially beneficial taxa, while having lower pathogenic taxa.

Since soils encompass intricate interaction networks, it is difficult to identify individual microbial effects on plant growth. Still, three major mechanisms are known to play a role in plant-soil feedbacks: pathogenicity, plant-growth promotion (PGP) (including disease suppression), and indirect effects on nutrient availability ^48,50. Changes in the relative abundance of soil borne pathogens could be an important mechanism in this study, considering two of the Ascomycota genera negatively correlated with early crop growth are known plant pathogens: Fusarium (F. solani and Fusarium sp.) and Gibberella (G. intricans) ⁴³. While Fusarium sp. could have infected tomato crops directly, F. solani and G. intricans are not common tomato pathogens but may be compromising the growth of other crops in the rotation. Unexpectedly, Mortierella was also negatively correlated with crop growth despite bearing several PGP strains, some of them associated with Fusarium wilt disease suppression ⁵¹. Yet, the latter could explain why it is sometimes correlated with Fusarium ⁵². Contrarily, bacterial taxa with potential PGP and/or disease-suppressive traits positively correlated with early crop growth, including the phylum Firmicutes (Bacillus and Tumebacillus) ⁵³ and Actinobacteria (Iamia, Nocardioides, Streptomyces, Gaiella) ^50,54. In terms nutrient cycling regulation, two ammonia-oxidizers (AOA) from Nitrososphaeraceae positively correlated with crop growth, although the opposite was found for two other AOA taxa (Candidatus Nitrosotenuis, Nitrosopumilaceae). This result could reflect different ecological adaptations or affinity for ammonia in these AOA taxa ⁵⁵. In general, radish-based cover crops presented lower relative abundance of the potentially pathogenic fungi, as well as a higher relative abundance of potential PGP and disease suppressive bacteria, as previously reported ^13,15,56. However, this response was not always consistent among residue management treatments or with early crop growth.

Overall, in this 8-year horticultural crop rotation, cover crops had small effects on soil microbial community structure, despite their repeated application and being present at fall sampling. Still, these small effects evidenced belowground-aboveground interactions that could have productive implications. Even though these results are exploratory, the fact that there was some degree of relationship between cc-associated microbial communities and early crop growth in the following season suggests the existence of plant-soil feedbacks linking cover crops and cash crops ¹. Detecting such links in field trials constitutes a valuable data resource for future studies where these hypotheses are put to the test. Such studies should not disregard interacting effects with other agricultural practices, especially those influencing C and nutrient inputs.

4.1- Site description

The experimental site was located at the Ontario Crops Research Centre, Ridgetown, Ontario, Canada (42.44° N, 81.88° W, 201 m.a.s.l). The soil at the site is Orthic Humic Gleysol with a sandy loam texture (68% sand, 21% silt, 11% clay) ²⁰Γ. The trial (est. 2008) consisted in a diverse horticulture and grain crop rotation analyzing cc species compositions and main crop residue management ¹⁹. The experimental design was a split-plot randomized complete block design with four replicates, where cc were applied to whole plots and residue management to sub-plots of 6 m x 8 m ²⁰. Besides a no-cc control, the trial comprised four different cover crops: oat (Avena sativa L.), fall rye (Secale cereale L.), radish (Raphanus sativus L.) and an intercropped mixture of the last two species. Aboveground winter wheat (Triticum aestivum L.) residues were either removed or retained (R- and R+, respectively) at harvest (early August 2015). Following straw removal in R- treatments, a light tillage was applied to all treatments in order to prepare the seed bed for planting the cover crops (mid-August 2015). The following year (early May 2016), the entire trial was sprayed with glyphosate to terminate rye, the only cc that overwinters. Then, cover crops were incorporated in the soil before transplanting tomato (Solanum lycopersicum L.) (late May 2016). More details about the trial can be found in previous publications ^19,20.

4.2- Soil sampling and processing for microbial analyses

Bulk soil samples were collected on 19 October 2015, two months after planting cover crops, which were in vegetative stage. In each sub-plot, we collected 9 soil cores (0-10 cm depth, 2.5 cm diameter) along two perpendicular transects, excluding ~1 m from the plot borders. Soil samples, collected aseptically, were manually homogenized, stored at 4°C and processed within 24 h for DNA extraction. A sub-sample was used to measure gravimetric dry weight.

Soil DNA was extracted using a MO BIO PowerSoil DNA Isolation Kit (MOBIO Laboratories, Inc.) according to the manufacturer’s guidelines. DNA concentration and purity were determined using a Nanodrop 8000 (Thermo Fisher Scientific, Waltham, MA, USA) and gel electrophoresis.

4.3- Quantitative PCR (qPCR)

Bacterial and fungal abundance were estimated with qPCR using a Bio-Rad CFX detection system (Bio-Rad Laboratories, Inc.). Bacterial 16S rRNA was targeted with the 338F-518R ⁵⁷ and fungal 18S rRNA with the primers FR1-FF390 ⁵⁸. Each reaction (20 µL final volume) consisted of 400 nM reverse and forward primers, 50x diluted template (according to inhibition tests), molecular biology grade water, and 1x SsoFast™ EvaGreen® Supermix for bacteria or 1x BioRad iQ™SYBR® Supermix for fungi (Bio-Rad Laboratories, Inc.). Template concentration was optimized via inhibition tests using M13 template and primers. The thermal profile for bacteria was as follows: initial denaturation at 98°C for 2 min followed by 40 cycles of 98°C for 5 s and annealing at 55°C for 5s. The thermal profile for fungi comprised an initial denaturation at 95°C for 5 min, 40 cycles of 95°C for 15 s, annealing at 50°C for 30 s, and extension at 70°C for 45 s. For both amplicons, the analysis was followed by a melt curve analysis (41 cycles of 5 s at 65°C to 95°C) was carried out to verify specific target amplification. Standard curves for quantification were constructed using serial dilutions of plasmid DNA containing the cloned target gene (Clostridium thermocellum for bacterial 16S rRNA and environmental fungal genomic DNA). All qPCR assays included no template controls and were optimized to ensure reaction efficiencies of 95-110% and R² values between 0.99-1.00.

4.4- DNA sequencing and bioinformatics

High-throughput sequencing (2 x 250 bp) was performed by Génome Québec (McGill University, Montreal, QC) on an Illumina MiSeq platform. Amplicon libraries were prepared as recommended by the Earth Microbiome Project (https://earthmicrobiome.org). For prokaryotes, we used the primers 515F (5’-GTGCCAGCMGCCGCGGTAA-3’) and 806R (5’-GGACTACHVGGGTWTCTAAT-3’) targeting the V4 region of the 16S rRNA gene ⁵⁹. The fungal ITS region was amplified using ITS1f (5’-CTTGGTCATTTAGAGGAAGTAA-3’) paired with ITS2 (5’-GCTGCGTTCTTCATCGATGC-3’) ⁶⁰.

Bioinformatics were carried out in QIIME 2 2020.6 ⁶¹. Firstly, we used q2-dada2 denoise-paired for denoising, dereplication, chimera filtering and merging of demultiplexed paired-end reads ⁶². Because fungal ITS sequences present length polymorphism, they were previously trimmed using q2-itsxpress ⁶³, which finds and trims the ends of the ITS region. We obtained a total of 3,141,597 quality filtered and trimmed reads and 32,555 amplicon sequence variants (ASVs) for 16S rRNA, and 3,564,848 reads and 2,079 ASVs for ITS. For 16S rRNA data, we used q2-phylogeny align-to-tree-mafft-fasttree to align ASVs and construct a phylogenetic tree ^{64 65}. Phylogenetic analyses were not carried out on ITS sequences, as their higher variability makes them unreliable for this type of analysis. Taxonomy was assigned using classify‐sklearn naïve Bayes in q2‐feature‐classifier ⁶⁶ with taxonomic classifiers trained against the prokaryotic reference sequences from Silva v. 138 with 99% OTUs ⁶⁷ and the fungal reference sequences from UNITE v. 8.2 with dynamic use of clustering thresholds ⁶⁸. The database FungalTraits helped us inquire on the functional traits of sensitive fungal taxa ⁴³.

4.5- Complementary data: plant growth and soil properties

Microbial data was related to both cc biomass and early crop biomass of the following year’s crop (tomato). Cc aboveground biomass was measured on 11 November 2015, before severe frost conditions ¹⁹. Within each sub-plot, aboveground biomass from living and dead plants was collected from two randomly located quadrants (0.25 m²) ²⁰. No weed biomass was sampled due to negligible amounts. Early crop biomass was sampled in the following spring (17 June 2016) by harvesting aboveground biomass of 6 tomato plants from the inside guard row per sub-plot ⁶. All plant biomass samples were dried at 60ºC to express data on a dry weight basis (kg ha⁻¹). For further detail regarding plant growth measurements refer to previous publications ^6,19,20.

Soil physicochemical data was obtained at different dates, according to availability. Soil nitrate, ammonium, SOC, total N and C:N were measured in April 2016 (before tillage) and texture in September 2016 (at tomato harvest). All soil measurements were carried out on composite samples of 6 cores (0-15 cm) obtained at each sub-plot as described in Chahal and Van Eerd ²⁰. Results are available in previous publications ^19,20 and Table S8.

4.6- Data analysis

Data analysis was carried out in QIIME 2 and R v. 3.6.3 ⁶⁹, and graphics were made using GraphPad Prism9 or ‘ggplot2’ in R. Alpha and beta diversity analyses (within- and between-sample, respectively) were carried out on rarefied ASV tables to avoid biases caused by differences in library size (see Fig. S1). Microbial richness was analyzed both as ASVs and genera counts, and with a phylogenetic index (Faith PD), while ASV and genera evenness were measured using Pielou’s index. These indices, as well as microbial abundance data (log copies g⁻¹ dry soil) from qPCR, were analyzed using ANOVA with linear mixed-effect models to account for the hierarchical nature of the experimental design. Tukey pairwise comparisons were carried out using package ‘multcomp’. Residuals were tested for normality and homogeneity of variance, and variance structures were applied if the latter was not satisfied.

Community composition and beta diversity were analyzed in the R package ‘vegan’ ⁷⁰ using presence-absence metrics, which resulted more sensitive to these longer term treatments. We used the phylogenetic distance metric unweighted UniFrac for prokaryotes ³⁷ and Jaccard distance for fungi. Permutational multivariate ANOVA (PERMANOVA) was used to test the effect of cc and residue management on community composition, previously assessing beta dispersion with betadisper. A PERMANOVA with restricted permutations was carried out with adonis2 following recommendations for split-plot designs by Anderson et al. ⁷¹. Because there was a high proportion of unexplained variability, canonical analysis of principal coordinates (CAP) allowed us to uncover and visualize treatment-driven patterns that were masked by other sources of variability (Anderson and Willis 2003). This analysis was run using the function capscale, removing block effects (partial CAP) to focus on treatment effects. For prokaryotes, residue management effects were also set as condition as they had a negligible influence.

Finally, taxonomic changes were explored using two methods: indicator species analysis in package ‘indicspecies’ ⁷² and ANOVA-like differential expression (ALDEx) in package ‘ALDEx2’ ⁷³. All analyses were carried out previously filtering highly rare taxa (less than 10 reads in total and/or present in less than 2 samples). Indicator species analysis was carried out using multipatt (multi-level pattern analysis) to analyze taxa distribution patterns among treatments. Changes in the relative abundance of taxa between cover crops were analyzed using aldex.kw and aldex.corr correlations with CAP axes, while aldex.t was used to test residue management effects. We also used aldex.corr to explore correlations between the relative abundance of taxa and early crop growth.

Acknowledgements

This work was supported by Food from Thought, a program funded by Canada First Research Excellence Fund (CFREF), and the Natural Sciences and Engineering Research Council (NSERC). Land-use fees for the field experiment were funded by the Ontario Agri-Food Innovation Alliance. Special thanks to Jonathan Gaiero, Eduardo Kovalski Mitter and Kamini Khosla for their help and advice.

Author contributions

MT and KD wrote the original manuscript. MT and DOA carried out bioinformatics and data analyses. JD and IC designed and carried out sampling. JD carried out sample processing and lab analyses for microbial data. IC performed sample processing and lab analyses for plant and soil physicochemical data. LVE conceived the field trial. LVE and KD provided funding. All authors contributed to editing and improving the original draft.

Competing interests

The author(s) declare no competing interests.

Data availability

Raw sequencing data have been deposited in the Sequence Read Archive (SRA) of the National Centre for Biotechnology Information (NCBI) under BioProject ID xxx.

Plant material

All plant material in the study was collected from long-term field experiments at the Ontario Crops Research Centre at Ridgetown Campus, University of Guelph. Plants were collected in accordance with University of Guelph and Ontario Ministry of Agriculture, Food and Rural Affairs guidelines.

Mariotte, P. et al. Plant–soil feedback: bridging natural and agricultural sciences. Trends Ecol. Evol, 33, 129–142 (2018).
Daryanto, S., Fu, B., Wang, L., Jacinthe, P. A. & Zhao, W. Quantitative synthesis on the ecosystem services of cover crops. Earth-Science Rev, 185, 357–373 (2018).
Shackelford, G. E., Kelsey, R. & Dicks, L. V. Effects of cover crops on multiple ecosystem services: Ten meta-analyses of data from arable farmland in California and the Mediterranean. Land use policy, 88, 104204 (2019).
McDaniel, M. D., Tiemann, L. K. & Grandy, A. S. Does agricultural crop diversity enhance soil microbial biomass and organic matter dynamics? A meta-analysis. Ecol. Appl, 24, 560–570 (2014).
Wittwer, R. A., Dorn, B., Jossi, W. & van der Heijden, M. G. A. A. Cover crops support ecological intensification of arable cropping systems. Sci. Rep, 7, 41911 (2017).
Chahal, I. & Van Eerd, L. L. Cover crops increase tomato productivity and reduce nitrogen losses in a temperate humid climate. Nutr. Cycl. Agroecosystems, 119, 195–211 (2021).
Belfry, K. D., Trueman, C., Vyn, R. J., Loewen, S. A. & Van Eerd, L. L. Winter cover crops on processing tomato yield, quality, pest pressure, nitrogen availability, and profit margins. PLoS One, 12, 1–17 (2017).
Wall, L. G. et al. Changes of paradigms in agriculture soil microbiology and new challenges in microbial ecology. Acta Oecol, 95, 68–73 (2019).
Schmidt, R., Gravuer, K., Bossange, A. V., Mitchell, J. & Scow, K. Long-term use of cover crops and no-till shift soil microbial community life strategies in agricultural soil. PLoS One, 13, 1–19 (2018).
Schmidt, R., Mitchell, J. & Scow, K. Cover cropping and no-till increase diversity and symbiotroph:saprotroph ratios of soil fungal communities. Soil Biol. Biochem, 129, 99–109 (2019).
Ali, A. et al. Hiseq base molecular characterization of soil microbial community, diversity structure, and predictive functional profiling in continuous cucumber planted soil affected by diverse cropping systems in an intensive greenhouse region of Northern China. Int. J. Mol. Sci, 20, 2619 (2019).
Kim, N., Zabaloy, M. C., Guan, K. & Villamil, M. B. Do cover crops benefit soil microbiome? A meta-analysis of current research. Soil Biol. Biochem, 142, 107701 (2020).
Vukicevich, E., Lowery, T., Bowen, P., Úrbez-Torres, J. R. & Hart, M. Cover crops to increase soil microbial diversity and mitigate decline in perennial agriculture. A review.Agron. Sustain. Dev.36, (2016).
Nevins, C. J., Nakatsu, C. & Armstrong, S. Characterization of microbial community response to cover crop residue decomposition. Soil Biol. Biochem, 127, 39–49 (2018).
Peralta, A. L., Sun, Y., McDaniel, M. D. & Lennon, J. T. Crop rotational diversity increases disease suppressive capacity of soil microbiomes. Ecosphere, 9, e02235 (2018).
Finney, D. M., Buyer, J. S. & Kaye, J. P. Living cover crops have immediate impacts on soil microbial community structure and function. J. Soil Water Conserv, 72, 361–373 (2017).
Calderón, F. J., Nielsen, D., Acosta-Martínez, V., Vigil, M. F. & Lyon, D. Cover Crop and Irrigation Effects on Soil Microbial Communities and Enzymes in Semiarid Agroecosystems of the Central Great Plains of North America. Pedosphere, 26, 192–205 (2016).
Romdhane, S. et al. Cover crop management practices rather than composition of cover crop mixtures affect bacterial communities in no-till agroecosystems. Front. Microbiol, 10, 1–11 (2019).
Chahal, I. & Van Eerd, L. L. Cover crop and crop residue removal effects on temporal dynamics of soil carbon and nitrogen in a temperate, humid climate. PLoS One, 15, e0235665 (2020).
Chahal, I. & Van Eerd, L. L. Evaluation of commercial soil health tests using a medium-term cover crop experiment in a humid, temperate climate. Plant Soil, 427, 351–367 (2018).
Zhao, M. et al. Intercropping affects genetic potential for inorganic nitrogen cycling by root-associated microorganisms in Medicago sativa and Dactylis glomerata. Appl. Soil Ecol, 119, 260–266 (2017).
Wardle, D. A. et al. Ecological linkages between aboveground and belowground biota. Science (80-.), 304, 1629–1633 (2004).
Xiong, C. et al. Host selection shapes crop microbiome assembly and network complexity. New Phytol, 229, 1091–1104 (2021).
Semenov, M. V., Krasnov, G. S., Semenov, V. M. & van Bruggen, A. H. C. Long-term fertilization rather than plant species shapes rhizosphere and bulk soil prokaryotic communities in agroecosystems. Appl. Soil Ecol, 154, 103641 (2020).
McDaniel, M. D., Grandy, A. S., Tiemann, L. K. & Weintraub, M. N. Eleven years of crop diversification alters decomposition dynamics of litter mixtures incubated with soil.Ecosphere7, (2016).
Fernandez-Gnecco, G. et al. Microbial community analysis of soils under different soybean cropping regimes in the Argentinean south-eastern Humid Pampas. FEMS Microbiol. Ecol, 97, 1–14 (2021).
Thapa, V. R., Ghimire, R., Acosta-Martínez, V., Marsalis, M. A. & Schipanski, M. E. Cover crop biomass and species composition affect soil microbial community structure and enzyme activities in semiarid cropping systems. Appl. Soil Ecol, 157, 103735 (2021).
Buyer, J. S., Teasdale, J. R., Roberts, D. P., Zasada, I. A. & Maul, J. E. Factors affecting soil microbial community structure in tomato cropping systems. Soil Biol. Biochem, 42, 831–841 (2010).
Drost, S. M., Rutgers, M., Wouterse, M., de Boer, W. & Bodelier, P. L. E. Decomposition of mixtures of cover crop residues increases microbial functional diversity., 361, 114060 (2020).
Di Simeone, R., Müller, G., Felgentreu, M., Glaser, B. & C. & Soil microbial biomass and community composition as affected by cover crop diversity in a short-term field experiment on a podzolized Stagnosol-Cambisol. J. Plant Nutr. Soil Sci, 183, 539–549 (2020).
Liu, L., Zhu, K., Wurzburger, N. & Zhang, J. Relationships between plant diversity and soil microbial diversity vary across taxonomic groups and spatial scales.Ecosphere11, (2020).
Cloutier, M. L. et al. Fungal community shifts in soils with varied cover crop treatments and edaphic properties. Sci. Rep, 10, 1–15 (2020).
Lay, C. Y., Hamel, C. & St-Arnaud, M. Taxonomy and pathogenicity of Olpidium brassicae and its allied species. Fungal Biol, 122, 837–846 (2018).
White, C. M. & Weil, R. R. Forage Radish Cover Crops Increase Soil Test Phosphorus Surrounding Radish Taproot Holes. Soil Sci. Soc. Am. J, 75, 121–130 (2011).
Leff, J. W. et al. Predicting the structure of soil communities from plant community taxonomy, phylogeny, and traits. ISME J, 12, 1794–1805 (2018).
Milcu, A. et al. Functionally and phylogenetically diverse plant communities key to soil biota., 94, 1878–1885 (2013).
Lozupone, C. A., Hamady, M., Kelley, S. T. & Knight, R. Quantitative and qualitative β diversity measures lead to different insights into factors that structure microbial communities. Appl. Environ. Microbiol, 73, 1576–1585 (2007).
Zaitsev, G. M. et al. New aerobic ammonium-dependent obligately oxalotrophic bacteria: description of Ammoniphilus oxalaticus gen. nov., sp. nov. and Ammoniphilus oxalivorans gen. nov., sp. nov. Int. J. Syst. Bacteriol, 48, 151–163 (1998).
Hartwright, L. M., Hunter, P. J. & Walsh, J. A. A comparison of Olpidium isolates from a range of host plants using internal transcribed spacer sequence analysis and host range studies. Fungal Biol, 114, 26–33 (2010).
Turmel, M. S., Speratti, A., Baudron, F., Verhulst, N. & Govaerts, B. Crop residue management and soil health: A systems analysis. Agric. Syst, 134, 6–16 (2015).
Austin, E. E., Wickings, K., McDaniel, M. D., Robertson, G. P. & Grandy, A. S. Cover crop root contributions to soil carbon in a no-till corn bioenergy cropping system. GCB Bioenergy, 9, 1252–1263 (2017).
Bai, Z., Liang, C., Bodé, S., Huygens, D. & Boeckx, P. Phospholipid 13C stable isotopic probing during decomposition of wheat residues. Appl. Soil Ecol, 98, 65–74 (2016).
Põlme, S. et al. FungalTraits: a user-friendly traits database of fungi and fungus-like stramenopiles. Fungal Divers, 105, 1–16 (2020).
Pepe, A., Giovannetti, M. & Sbrana, C. Lifespan and functionality of mycorrhizal fungal mycelium are uncoupled from host plant lifespan. Sci. Reports 2018 81, 8, 1–10 (2018).
Frey, S. D. Mycorrhizal Fungi as Mediators of Soil Organic Matter Dynamics. Annu. Rev. Ecol. Evol. Syst, 50, 237–259 (2019).
Barel, J. M. et al. Winter cover crop legacy effects on litter decomposition act through litter quality and microbial community changes. J. Appl. Ecol, 56, 132–143 (2019).
Liu, X. et al. Decomposing cover crops modify root-associated microbiome composition and disease tolerance of cash crop seedlings. Soil Biol. Biochem, 160, 108343 (2021).
Saleem, M., Hu, J. & Jousset, A. More Than the Sum of Its Parts: Microbiome Biodiversity as a Driver of Plant Growth and Soil Health. Annu. Rev. Ecol. Evol. Syst, 50, 145–168 (2019).
Maul, J. E. et al. Microbial community structure and abundance in the rhizosphere and bulk soil of a tomato cropping system that includes cover crops. Appl. Soil Ecol, 77, 42–50 (2014).
Wei, Z. et al. Initial soil microbiome composition and functioning predetermine future plant health. Sci. Adv, 5, 1–12 (2019).
Ozimek, E. & Hanaka, A. Mortierella Species as the Plant Growth-Promoting Fungi Present in the Agricultural Soils. Agriculture, 11, 7 (2020).
Li, F. et al. Mortierella elongata ’s roles in organic agriculture and crop growth promotion in a mineral soil. L. Degrad. Dev, 29, 1642–1651 (2018).
Sansinenea, E. & Type="Italic">Bacillus spp.: As Plant Growth-Promoting Bacteria. in Secondary Metabolites of Plant Growth Promoting Rhizomicroorganisms: Discovery and Applications225–237(Springer, Singapore, 2019). doi:10.1007/978-981-13-5862-3_11.
Palaniyandi, S. A., Yang, S. H., Zhang, L. & Suh, J. W. Effects of actinobacteria on plant disease suppression and growth promotion. Appl. Microbiol. Biotechnol. 2013 9722 97, 9621–9636 (2013).
Jung, M. Y. et al. Ammonia-oxidizing archaea possess a wide range of cellular ammonia affinities. ISME J. 2021, 1–12 https://doi.org/10.1038/s41396-021-01064-z (2021).
Larkin, R. P., Griffin, T. S. & Honeycutt, C. W. Rotation and cover crop effects on soilborne potato diseases, tuber yield, and soil microbial communities. Plant Dis, 94, 1491–1502 (2010).
Fierer, N. & Jackson, J. Assessment of soil microbial community structure by use of taxon-specific quantitative PCR assays. Appl. Environ. Microbiol, 71, 4117 (2005).
Vainio, E. J. & Hantula, J. Direct analysis of wood-inhabiting fungi using denaturing gradient gel electrophoresis of amplified ribosomal DNA. Mycol. Res, 104, 927–936 (2000).
Caporaso, J. G. et al. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl. Acad. Sci. 108, 4516–4522(2011).
White, T. J., Bruns, T., Lee, S. & Taylor, J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. in PCR protocols: a guide to methods and applications315–322(Academic Press, 1990).
Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol, 37, 852–857 (2019).
Callahan, B. J. et al. DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods, 13, 581–583 (2016).
Rivers, A. R. et al. Software to rapidly trim internally transcribed spacer sequences with quality scores for marker gene analysis. F1000Research, 7, 1418 (2018).
Katoh, K., Misawa, K., Kuma, K. & Miyata, T. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res, 30, 3059–3066 (2002).
Price, M. N., Dehal, P. S. & Arkin, A. P. FastTree 2 – Approximately Maximum-Likelihood Trees for Large Alignments. PLoS One, 5, e9490 (2010).
Bokulich, N. A. et al. Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2’s q2-feature-classifier plugin. Microbiome, 6, 90 (2018).
Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res, 41, D590–D596 (2013).
Abarenkov, K. et al. UNITE QIIME release for Fungi(2020) doi:10.15156/BIO/786385.
R Core Team. R: A Language and Environment for Statistical Computing. (2020).
Oksanen, J. et al. vegan: Community Ecology Package(2020).
Anderson, M. J., Gorley, R. N. & Clarke, K. R. PERMANOVA+ for PRIMER: Guide to Software and Statistical Methods (PRIMER-E, 2008).
De Cáceres, M. & Legendre, P. Associations between species and groups of sites: indices and statistical inference., 90, 3566–3574 (2009).
Fernandes, A. D. et al. Unifying the analysis of high-throughput sequencing datasets: characterizing RNA-seq, 16S rRNA gene sequencing and selective growth experiments by compositional data analysis. Microbiome, 2, 15 (2014).

No competing interests reported.

supplementarymaterial.pdf

Download PDF

Editorial decision: Major revision
29 Dec, 2021
Reviews received at journal
19 Dec, 2021
Reviews received at journal
29 Nov, 2021
Reviewers agreed at journal
10 Nov, 2021
Reviewers agreed at journal
08 Nov, 2021
Reviewers invited by journal
03 Nov, 2021
Editor assigned by journal
03 Nov, 2021
Editor invited by journal
28 Oct, 2021
Submission checks completed at journal
28 Oct, 2021
First submitted to journal
21 Oct, 2021

You are reading this latest preprint version

Cover crop-driven shifts in soil microbial communities could modulate tomato early crop growth via plant-soil feedbacks

Status:

Version 1

Abstract

Figures

1- Introduction

2- Results

2.1- Response of soil prokaryotic communities to cc and residue treatments

2.2- Response of soil fungal communities to cc and residue treatments

2.3- Relationship between soil microbial communities and plant growth

3- Discussion

4- Methods

4.1- Site description

4.2- Soil sampling and processing for microbial analyses

4.3- Quantitative PCR (qPCR)

4.4- DNA sequencing and bioinformatics

4.5- Complementary data: plant growth and soil properties

4.6- Data analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1