The greenhouse industry has become one of the most important components for the development of agriculture. Nowadays, the vegetable greenhouse industry has been greatly promoting the world's economy. The greenhouse industry is developing rapidly in China, which accounts for more than all global greenhouse cultivation areas (Chang et al., 2013; Wang et al., 2017a). After the increase in this production style, the supply of fresh and off-season vegetables on the market has improved (Chang et al., 2011). Greenhouse production should be developed consistently in future agricultural systems (Muller et al., 2017). Initially, sole cropping is used in the greenhouse production, but with increased market demand, intercropping or two times of repeated planting per year has been done in the greenhouse (Firouzi et al., 2017; Wang et al., 2014). As a result, the improvement of the cultivation environment conditions in the greenhouse has become the priority in growing crops, especially in protecting soil quality and fertility (Yeboah et al., 2018). Therefore, maintaining production and improving or balancing soil quality have become important in greenhouse production.
Melon (
Reduced TL systems can increase long-term soil productivity and health (Tillman et al., 2015). However, depending on the melon variety, growers can reduce yields compared with TL to avoid soil compaction and low fertility. Reduced TL systems can produce similar or higher yields than conventional TL systems in vegetable crops (Haramoto and Brainard, 2012; Tillman et al., 2015). However, UT is one form of conservation TL that has been widely adopted in agricultural production to improve soil sustainability (O’Rourke and Petersen, 2016). The UT treatment is widely applied in vegetable production, but striking a balance between retention of soil nutrients and quality, and increased production is still a contentious issue. Microbes are extremely abundant in soils and play important roles in soil productivity and biogeochemistry (Kumar et al., 2019). Microbial groups, such as beneficial bacteria and fungi, maintain soil functionalities and promote plant growth (Kandlikar et al., 2019; Touceda-González et al., 2015). Microbial diversity increases soil quality and fertility. Thus, increasing microbial diversity can improve greenhouse practices to maintain soil health and quality (Karlen et al., 2019; Xue et al., 2020). It is useful to analyse the microbial community structures and microbial abundance and diversity in UT and TL soils in the greenhouse. It can be considered as one of the important indicators for greenhouse soil fertility.
Despite the availability of various soils management practices (e.g. soil texture, nutrient and irrigation), the mechanism of improving soil quality under UT practice for facility production is still unavailable, causing difficulties and confusion in melon production. Given that UT and TL influence agriculture production and the soil environment, we attempted to explore the effects of these treatments on microbial abundance and diversity under facility condition. In this study, the pyrosequencing of the V4–V5 16S rRNA and fungi internal transcribed spacer (ITS) gene region were used to analyse microbial community structures in UT and TL soils. The melon yield and the correlation of soil microbial communities under UT treatment were also investigated under facility conditions.
Grafted melon seedlings were used in this study. The scion was a cultivated melon variety, ‘Cuimi’. This cultivar was suitable for greenhouse cultivation, and it could be available in the market in the near future. The rootstock was a pumpkin (
The greenhouse experiment was accomplished in a planting base in Hexin County (118.35° E and 31.74° N), Anhui Province, China. Soil samples were collected thrice from January 1, 2018, to August 31, 2019. A total of 18 rhizosphere soil samples were collected from 10 points by using a sterile blade at a depth of 0–15 cm and composited together. Each sample was then divided into two portions. One portion was stored at 4°C for biogeochemical analysis and the other one at −80°C for DNA analysis.
A protected field experiment with two operation practices (i.e. UT and TL treatments) was conducted in Hexin County. Soil pH ranged between 7.1 and 7.3, and the organic matter (OM) ranged from 20 to 22 g · kg−1 in two treatments. The complete basic soil properties are listed in Table 1. The soil was tilled with a depth of 20–35 cm in TL. Ridging ways and planting density were arranged the same way as UT. Grafted melon seedlings were transplanted at the beginning of January 2019 under greenhouse soils. Normal and regular management was followed to maintain melon seedling growth in UT and TL treatments. Before the fruit setting time, 20 kg · 667 m−2 fertiliser (N:P:K ratio of 3:1:3) was added to plant soils. The experiment was terminated at the end of July 2019. The melon matured within 80 days from planting. The average yield from 667 m2 was recorded immediately after harvest. Five fruits from each replicate were randomly selected to assess fruit weight (kg) and nutrient elements. Ten plants were used with three replicates.
Basic properties of soils in two treatments
Treatment | pH | OM | TN | TP | TK | AN | AP | AK |
---|---|---|---|---|---|---|---|---|
g · kg−1 | mg · kg−1 | |||||||
UT | 7.30 ± 0.05 | 20.70 ± 0.08 | 1.07 ± 0.02 | 1.91 ± 0.04 | 6.34 ± 0.09 | 155.23 ± 3.21 | 89.42 ± 2.12 | 686.05 ± 3.14 |
TI | 7.12 ± 0.03 | 22.01 ± 0.11 | 1.18 ± 0.05 | 1.69 ± 0.07 | 5.81 ± 0.02 | 163.51 ± 1.51 | 95.61 ± 1.11 | 767.00 ± 1.89 |
AK, available potassium; AN, available nitrogen; AP, available phosphorus; OM, organic matter; TI, tillage; TK, total potassium; TN, total nitrogen; TP, total phosphorus; UT, no-tillage.
Soil pH and OM content were detected following our previously reported methods (Zhang et al., 2019). The total N (TN) and alkaline N (AN) of the soil and the N content of the plant were determined following a reported method (Bremner, 1960). The total P (TP) in the melon plant was determined using the combustion and the molybdenum blue colorimetric methods (Sjösten and Blomqvist, 1997). Available P (AP) was extracted and detected following the procedures of a reported method (Li et al., 2004). Soil available K (AK) and plant total K were measured using flame atomic absorption spectrophotometry following a previous method (Zhang et al., 2019).
Approximately, 0.5 g of fresh rhizosphere soil was used to extract soil DNA in triplicate from each soil site by using the MoBio Power Soil DNA Isolation Kits (MoBio Laboratories, Carlsbad, CA, USA) following the manufacturer's instructions. The purity of the soil DNA extracts was determined using the NanoDrop™ 2000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). The obtained DNA was stored at −80°C for future analyses.
The amplification of 16S rRNA for sequencing was performed following a previously described method (Shen et al., 2013; Zhang et al., 2019). Briefly, the V4–V5 regions of the bacterial 16S rRNA genes were amplified using the primers F515 and R907. The primers ITS1 and ITS2 were used to amplify the ITS genes (Table 2). PCR was performed following a previous report (Zhang et al., 2019). The PCR products were purified, and sequencing was performed using the Illumina MiSeq platform from Biozeron Biotechnology Co., Ltd. (Shanghai, China).
Oligonucleotide primers for PCR
Microbes | Regions | Forward primer (5′-3′) | Reverse primer (5′-3′) |
---|---|---|---|
Bacteria | V4–V5 | GTGCCAGCMGCCGCGG | CCGTCAATTCMTTTRAGTTT |
Fungi | ITS1 | CTTGGTCATTTAGAGGAAGTAA | GCTGCGTTCTTCATCGATGC |
The operational taxonomic units (OTUs) with identities of 97% were identified using the Mothur software (
All data were analyzed using a one-way analysis of variance and considered statistically significant and highly significant at
In this study, two operation practices (i.e. UT and TL) were used. Under the UT and the TL treatments, the N, P and K were absorbed similarly by the melon plant, but the total amounts of absorbed nutrients were different. The differences in nutrient contents in leaves were significant. The most absorbed element was N. The N, P and K contents in the UT treatment were 2.71, 0.51, and 1.04 g · plant−1 (Figure 1A), respectively. The N, P and K contents in the TL treatment were 1.94, 0.35 and 0.89 g · plant−1, respectively. The N content was the highest, followed by K, and the P content was the lowest. The N content of the UT treatment was significantly higher than that of the TL treatment.
In the stem of the melon plant, the N, P and K contents in the UT treatment were 0.52, 0.09 and 0.65 g · plant−1 (Figure 1B), respectively. The N, P and K contents in the TL treatment were 0.53, 0.08 and 0.97 g · plant−1, respectively. K was the most absorbed, followed by N, and P was the least absorbed. The K content in the stems in the TL treatment was significantly higher than that in the UT treatment.
In melon fruit, the N, P and K contents in the UT treatment were 7.04, 1.37 and 6.45 g · plant−1, respectively. The N, P and K contents in the TL treatment were 3.42, 0.73 and 4.21 g · plant−1 (Figure 1C), respectively. The N, P and K contents in the UT treatment were significantly higher than those in the TL treatment. The total contents of the three elements in the UT treatment (20.41 g · plant−1) were significantly higher than those in the TL treatment (13.14 g · plant−1). The melon yields in the UT and TL treatments were 2813.34 and 2138.32 kg · 667 m−2, respectively (Figure 1D), indicating that the UT treatment increased melon yield.
The distribution of the clustered bar showed that the bacterial species (A) was significantly higher than the fungal species (B). The distribution characteristics of the microbial communities in UT and TL treatments were similar. The main bacterial phyla included
The CCA correlation showed a significant negative correlation between the pH value and
The correlation of the N, P and K contents absorbed by the leaves, stems and fruits of the melon plants with the soil microbial communities was further studied. The TN, TP, and TK contents of the leaves and fruits were significantly positively correlated with the bacterial genus
Concerning fungal genus communities, the TN, TP and TK contents in melon leaves and fruits were negatively correlated with
The TL and the UT treatments are important for improving soil structures, which are beneficial for plant growth and physical soil properties (Pöhlitz et al., 2018). TL treatment affects corn grain nutrient composition and yield (Houx et al., 2016). In the present work, the two tillage styles showed different effects on the nutrient uptake in melon plants under facility conditions. Our data showed that in the UT treatment, additional N was absorbed in the leaves and fruit (Figures 1A and 1B), and the N content was kept constant in the melon stem under the TL treatment. Previous findings have discovered that the autotrophic nitrification rate in the topsoil layer of UT is significantly higher than that of conventional TL (Liu et al., 2017), which provides a reasonable explanation on the absorption of additional N by the melon plant. TL and P fertilization showed no significant effects on root P contents (Li et al., 2017). By contrast, the P contents in leaves and fruits in the UT treatment in our study were higher than the values in TL treatment especially the P content in fruits (*
Here, additional K in the fruits under UT treatment was found, whereas high K content was observed in melon stem under TL treatment. K can affect fruit quality (Lester et al., 2010). Following previous studies, our data showed that the K accumulation in melon plants had different amounts under the two treatments. However, in a previous study, TL has produced no-effect on seed K concentrations (Farmaha et al., 2011). Soil management and rotations can affect the P and K in soil and increase fertiliser use efficiency (Rosolem and Calonego, 2013). In the present study, under UT treatment, the proper addition of nutrients can increase the N, P, and K contents in leaves and fruits. The total amount of the three elements in the UT treatment (20.41 g · plant−1) was significantly higher than that in the TL treatment (13.14 g · plant−1), thereby suggesting that nutrient uptake can also be improved by adding fertiliser in later management without TL. Melon yield demonstrated that the UT treatment produced more fruits than the TL treatment, thereby indicating that UT treatment can increase melon yields. Similar results have been reported that UT can ensure equivalent or even higher yields than conventional TL (Ruisi et al., 2016). Hence, the results of the present study indicated that the UT treatment exhibited the same positive effect on the yield of the melon crop compared with TL treatment.
Soil microbial correlation with melon plant and nutrients
Microbial groups | Nutrients | Soil microbial genus | Correlation | Significant level | Plant part | |
---|---|---|---|---|---|---|
Bacteria | TN, TP and TK | Pirellula | 0.771 | 0.038 | * | Leaf and Fruit |
TN | Dongia | −0.869 | 0.010 | * | Stem | |
TN | Gemmatimonas, Hamadaea | −0.927 | 0.002 | ** | Stem | |
TP | Dongia | −0.811 | 0.024 | * | Stem | |
TP | Nocardioides, Gemmatimonas | −0.753 | 0.044 | * | Stem | |
TP | Planctomycetes | 0.927 | 0.002 | ** | Stem | |
TP | Armatimonadetes | −0.898 | 0.005 | ** | Stem | |
TP | Patescibacteria | −0.898 | 0.005 | ** | Stem | |
Fungi | TN, TP and TK | Chaetomium | −0.771 | 0.038 | * | Leaf and Fruit |
TN, TP and TK | Dendrostilbella | 0.771 | 0.038 | * | Leaf and Fruit | |
TN | AcrophialophoraCephaliophora | 0.782 | 0.034 | * | Stem | |
TN | Polyschema | −0.898 | 0.005 | ** | Stem | |
TN | Cercospora | −0.788 | 0.032 | * | Stem | |
TN | Scedosporium | 0.779 | 0.035 | * | Stem | |
TN | Verticillium | −0.927 | 0.002 | ** | Stem | |
TP | Fusarium | −0.753 | 0.044 | * | Stem | |
TP | Cercospora | −0.857 | 0.013 | * | Stem | |
TK | Dendrostilbella | −0.771 | 0.038 | * | Stem | |
TK | Ascobolus | −0.885 | 0.007 | ** | Stem |
TN, total nitrogen; TP, total phosphorus; TK, total potassium.
Bacterial and fungal diversities changed differently under UT and TL operations. The OTUs in bacteria were higher than those in fungi, and the functional communities in the UT treatment were higher than those in the TL treatment. Results revealed that the predominant bacterial phyla were
In terms of fungi, the predominant fungal phylum was
A significant negative correlation was observed between the pH value and
Microbial communities are important for soil quality, and plant species and soil type shape the structure and function of the microbial communities in the plant rhizosphere (Berg and Smalla, 2009; Yao et al., 2003). The correlation of the N, P, and K contents that were absorbed by melon plant leaves, stems and fruits with the soil microbial communities was further analysed. However, although the nutrients in different melon parts were correlated to microbial groups, these bacterial or fungal phyla did not decide the nutrient uptake in melon plants. By contrast, the nutrient uptake in melon leaves, stems and fruits may collect those microbial groups. The efficient use of plant nutrient can be enhanced by beneficial bacteria or fungi (Adesemoye et al., 2008). Rhizosphere interactions between microorganisms and plants can also stimulate the P acquisition in plant roots (Marschner et al., 2011). Here, TN, TP, and TK were significantly positively correlated with
The UT and the TL treatments had minimal influence on soil chemical properties and microbial community diversity. The melon yield under the UT treatment was higher than that under the TL treatment.
The microbial communities may be correlated with nutrient uptake in melon leaves, stems and fruits. The further investigation focused on the mechanisms underlying the beneficial stimulations of these microbial groups and improvement of the utilisation of these microbial groups are useful in sustainable greenhouse production.
This study provided additional insights into the response of soil fertility and microbial structures to UT and TL treatments under greenhouse soils, which may help manage soil quality for the protected field production of melon.