Rare microbial taxa as the major drivers of ecosystem multifunctionality in long-term fertilized soils

Highlights

• Inorganic fertilization decreased soil multifunctionality.

• Organic fertilization increased microbial diversity and multifunctionality.

• Rare microbial taxa had an over-proportional role in multifunctionality.

Abstract

Soil microbial communities play an essential role in driving multiple functions (i.e., multifunctionality) that are central to the global biogeochemical cycles. Long-term fertilization has been reported to reduce the soil microbial diversity, however, the impact of fertilization on multifunctionality and its relationship with soil microbial diversity remains poorly understood. We used amplicon sequencing and high-throughput quantitative-PCR array to characterize the microbial community compositions and 70 functional genes in a long-term experimental field station with multiple inorganic and organic fertilization treatments. Compared with inorganic fertilization, the application of organic fertilizer improved the soil multifunctionality, which positively correlated with the both bacterial and fungal diversity. Random Forest regression analysis indicated that rare microbial taxa (e.g. Cyanobacteria and Glomeromycota) rather than the dominant taxa (e.g. Proteobacteria and Ascomycota) were the major drivers of multifunctionality, suggesting that rare taxa had an over-proportional role in biological processes. Therefore, preserving the diversity of soil microbial communities especially the rare microbial taxa could be crucial to the sustainable provision of ecosystem functions in the future.

Introduction

Soil microbes represent the most abundant and diverse organisms on Earth (Locey and Lennon, 2016). It is estimated that 1 cm³ of soil contains 0.4–2 billion prokaryotic microbes, tens of thousands of taxa and up to ~200 m fungal hyphae, which play key roles in maintaining multiple ecosystem functions simultaneously (i.e. ecosystem multifunctionality) that are critical to the biogeochemical nutrient cycling, primary production, litter decomposition and climate regulation (Bardgett and van der Putten, 2014; Wagg et al., 2014; Bender et al., 2016). Recent studies provide evidence that global environmental drivers, such as land use changes, nitrogen deposition, and climate change, can severely impact multifunctionality in terrestrial ecosystems through manipulating the belowground soil biodiversity (Garcia-Pichel et al., 2013; Maestre et al., 2015; Delgado-Baquerizo et al., 2016, 2017a; Luo et al., 2018). Common agricultural practices, such as soil tillage, fertilization, pesticide application, and monoculture, can have adverse effects on the maintenance of soil microbial diversity and interactions (de Vries et al., 2012; Tsiafouli et al., 2015), with unknown consequences for soil multifunctionality. Given that farming intensity is projected to constantly increase on a global scale (Bender et al., 2016) to feed a growing human population (Ort et al., 2015), it is imperative to understand the consequence of agricultural practices on belowground biodiversity and multifunctionality.

It is estimated that ~23% of the world soil faces degradation and the area of degraded land increases at an annual rate of 5–10 million ha, which may affect the food security for approximate 1.5 billion people globally (Stavi and Lal, 2015). Fertilization as an important agricultural practice accelerates the rate of land degradation, as long-term inorganic fertilization may result in soil acidification (Guo et al., 2010). Before the innovation of industrial ammonia synthesis, the biological nitrogen fixation has sustained life on Earth for thousands of years. Modern agricultural practices are based predominantly on industrially produced mineral fertilizers, and have directly caused several environmental problems, such as surface and ground water eutrophication through excessive discharge of nutrients including Nitrogen and Phosphorus into water (Smith and Schindler, 2009), and global warming through conversion of ammonium to nitrogen oxides (Foley et al., 2005). A growing body of evidence indicated that intensive fertilization also indirectly influences a wide range of crucial ecosystem functions via altering the diversity of soil microorganisms (Hartmann et al., 2015; Ling et al., 2016), but we know little about how fertilization will impact the ecosystem multifunctionality. To the best of our knowledge, few studies have explicitly addressed the impact of fertilization on the ecosystem multifunctionality and the relationships between soil biodiversity and ecosystem functioning (Luo et al., 2018). Such knowledge is essential to the development of management frameworks to protect soil biodiversity involved in multifunctionality and reduce impacts of intensive fertilization on terrestrial ecosystems.

Herein, we hypothesized that the positive relationship between ecosystem multifunctionality (especially the functional traits related to nutrient element cycles) and microbial diversity is maintained in the agroecological system, as it has been widely demonstrated in the natural ecosystem (Mori et al., 2016; Delgado-Baquerizo et al., 2017a). We characterized the bacterial and fungal communities in soil samples collected from a long-term fertilization experimental field, using amplicon sequencing of bacterial 16 S rRNA genes and fungal internal transcribed spacer 2 (ITS2) region, respectively. Given that the relationships between biodiversity and multifunctionality were reported to be dependent on the identity and number of measured functions (Meyer et al., 2018), we measured multiple ecosystem functions from functional gene level to enzyme level and specific biological processes: (i) We used quantitative microbial element cycling (QMEC) for high-throughput quantitative assessment of 70 functional genes related to Carbon (C), Nitrogen (N), Phosphorus (P), and Sulphur (S) biogeochemical cycling (Zheng et al., 2018); (ii) We measured four enzyme activities including β-glucosidase, N-acetyl-β-glucosaminidase, urease and phosphatase; (iii) We also determined soil basal respiration, potential ammonia oxidation and denitrification enzyme activity.