Comparison of soil biological properties and soil bacterial diversity in sugarcane, soybean, mung bean and peanut intercropping systems

Aims Sugarcane intercropping with soybean (Glycine max (Linn.) Merr.), mung bean (Vignaradiata (Linn.) Wilczek) and peanut (Arachishypogaea Linn.) as well as a sugarcane monoculture were conducted to study the impacts of intercropping on soil biological characteristics and bacterial diversity. Methods The soil samples were collected from twelve random sites and mixed well at the experimental farm of the Guangxi South Sub-tropical Agricultural Science Research Institute, Longzhou, China. Traditional analysis methods and modern high-throughput sequencing technology was used to compare and analyze the soil enzyme activity, microbial biomass, soil cultivable microorganisms and other biological traits and bacterial diversity. Results The results showed that soil cultivable microorganisms, the activities of soil enzymes and microbial biomass carbon, nitrogen, and phosphorus were all signicantly improved by intercropping with soybean and mung bean. Additionally, soil bacterial diversity and richness in sugarcane elds were also signicantly enhanced by intercropping with soybean and mung bean. In addition, soil bacterial community structures in sugarcane elds can be altered by intercropping with different legumes. Proteobacteria, a high-nutrient-tolerant bacterial assemblage, became the dominant bacteria in the sugarcane-soybean and sugarcane-mung bean intercropped soils. Twenty four, 28, 26 and 27 dominant soil bacterial genera were found after the sugarcane-soybean, sugarcane-mung bean, sugarcane-peanut and sugarcane monoculture treatments, respectively. Conclusions Sugarcane-mung bean intercropping suggested to be the most promising system for regaining and improving soil fertility and soil heath and facilitate agriculture intensication of sugarcane.


Introduction
Sugarcane (Saccharum L.) is the primary source of sugar and is also utilized as a major biofuel and bioenergy crop worldwide (Tomes et al. 2011, Chandel et al. 2011. China is the third largest sugar producing country in the world after Brazil and India. In China, approximately 90% of sugarcane crops are planted in the southern and southwest regions, which are mainly in Guangxi, Guangdong, and Yunnan Provinces. Among these provinces, Guangxi Province is the top sugarcane and sugar producer and accounts for more than 65% of the total sugar production in China (Li 2004). In China, sugarcane production is largely con ned to hilly terrain under rainfed conditions that result in relatively low yields.
The problem is worsened by the long-term overuse of chemical fertilizers and pesticides to improve cane and sugar yields (Robinson et al 2011). For example, higher amounts of N fertilizer, as high as 600-800 kg N ha − 1 in some regions, are applied annually to sugarcane crops in China, while only 60-120 kg N ha − 1 are applied in Brazil (Li and Yang 2014). Long-term chemical fertilizer overuse negatively in uences soil microbial ecology and terrestrial and aquatic ecosystem function (Robertson and Vitousek 2009). Therefore, minimal chemical fertilizer inputs for maintaining healthy soil and high crop productivity are urgently needed for commercial sugarcane production in China.
Intercropping, which involves growing two or more crop species simultaneously in the same eld, is an ancient cropping system that is practiced all around the world (Solanki et al. 2016). Intercropping contributes to the ecofunctional and sustainable intensi cation of crop production (Raseduzzaman and Jensen 2017) and is considered an e cient way to achieve agriculture sustainability (Vandermeer 2011).
At present, intercropping is more common in developing countries than developed countries and is practiced mostly by small and subsistence farmers (Sileshi et al. 2012). Intercropping enables agricultural intensi cation, which delivers higher yields per unit area and increases resource use e ciency compared with monoculture crops (Hauggaard-Nielsen et al. 2008). In particular, the application rates of synthetic nitrogen fertilizer can be reduced by legume intercropping owing to its capacity for biological nitrogen xation. Moreover, intercropping promotes biodiversity in cropping systems and causes them to be more resilient when faced with environmental stresses, diseases and pests (Frison et al. 2011, Brooker et al. 2014). However, not all intercropping systems deliver yield bene ts or other positive outcomes. For example, some cereal-legume intercropping methods produce lower biomass and nitrate accumulations than monoculture crops , Li et al. 2001. Recently, some studies have compared intercropping to monocultures by focusing mainly on weed control, management factors, intercrop productivity, and resource use e ciency. (Weerarathne et al. 2017, Yu et al., 2015Pelzer et al. 2014). However, little is known about the effects of different intercrops on soil quality, particularly soil biology and related processes.
Soil quality depends on a large number of physical, chemical, biological, biochemical and microbiological parameters (Chaer et al. 2009). In particular, the latter two are the most sensitive indicators and respond rapidly to changes (Bastida et al. 2008). Soil enzyme activity is capable of re ecting ecosystem processes (Doran and Zeiss 2000). In addition to enzymatic activity, soil microbial biomass carbon (MBC), microbial biomass nitrogen (MBN) and microbial biomass phosphorus (MBP) are also used to monitor soil quality (Pandey et al. 2014). Soil microorganisms play an important role in soil biogeochemical processes, such as nitrogen, phosphorus and other element cycles (Urbanová et al. 2015). It is now recognized that soil microbial community composition and diversity determine soil health and crop productivity to a great extent (Mangan et al. 2010).
Therefore, in the present study, we investigated soil fertility and soil bacterial diversity under different sugarcane-legume intercropping systems, which are an important but overlooked aspect of the very promising crop diversi cation systems in China.

Field site description and experimental designs
Field experiments were carried out in the 2016-17 and 2017-18 crop seasons at the experimental farm of the Guangxi South Sub-tropical Agricultural Science Research Institute, Longzhou (106°47′34″E and 22°19′42″N). Experiments were conducted using a randomized block design with three replications to study the performances of a sugarcane monoculture and sugarcane intercropping treatments with soybean (Glycine max (Linn.) Merr., mung bean (Vignaradiata (Linn.) Wilczek, or peanut (Arachishypogaea Linn. with a 2:2 design (two rows of soybean, mung bean or peanut planted between each sugarcane row). The sugarcane monoculture was used as the control.
Soil sampling and soil biological properties analysis Soil samples were collected in July 2018 from 12 plots that represented all the treatments in the intercropping experiments. From each plot, soil samples (0-30 cm layer) were collected from 12 random sites and mixed well. These soil samples were collected in sterile plastic bags and placed on ice in an ice box. The samples were immediately transferred to the laboratory, where they were sieved through a 2-mm stainless steel sieve, and then stored in a refrigerator at 4°C for immediate analysis or were stored at -80°C for later use. Meanwhile, portions of the soil samples were air dried for soil chemical analyses. The sample soils had an average pH of 6.2, while the organic matter, total nitrogen, available phosphorus and potassium contents were 23.3 g kg − 1 , 1.77 g kg − 1 , 12.4 mg kg − 1 and 66.1 mg kg − 1 , respectively.

Soil microbial numbers
Microbial numbers were determined using the agar plate dilution method modi ed with cycloheximide (100 µgL − 1 ) as described by Martin (1950). Rose Bengal-streptomycin agar medium and starch casein medium were used to determine the fungi and actinomycetes numbers in fresh soil samples as described by Miyashita (1997). The pH levels of the media were adjusted to 6.8 with HCl or NaOH. Microbial counts were determined for 5 replicates.

Soil microbial biomass
The soil microbial biomass N (MBN) and soil microbial biomass C (MBC) contents were determined using the chloroform fumigation-extraction method as described by Brookes et al. (1985) and Vance et al. (1987). The contents of soil microbial biomass P (MBP) contents were determined by the phosphorus molybdenum blue colorimetric method (Powlson et al. 1987).
Soil enzyme activities β-Glucosidase (EC.3.2.1.21) assays were based on ρ-nitrophenol (pNP) release after cleavage of a synthetic substrate (Deng and Tabatabai 1994). In brief, the color of the released ρ-nitrophenol was measured at 400 nm using a spectrophotometer (UV-1700, Shimadzu, Japan). A standard curve was plotted using 0-80 µg mL − 1 concentrations of ρ-nitrophenol. Enzyme activities are expressed as nmol pNP released per g dry soil per minute (nmol pNP g − 1 min − 1 ). Acid phosphatase activity in soils was estimated by measuring the amount of ρNP released after incubating the samples with ρ-nitrophenyl-phosphate (Alef et al. 1995). In a reaction tube, 0.25 mL of toluene, 4.0 mL of modi ed universal buffer (5x MUB, pH 6.0, which was made by dissolving 12.1 g of Tris, 11.6 g of maleic acid, 14.0 g of citric acid and 6.3 g of boric acid in 500 mL of 1 M NaOH to make a volume of 1 L), and 1.0 mL ρ-nitrophenylphosphate (15 mmol L − 1 ) were added to 1.0 g of soil sample and incubated at 37°C for 1 h. The reaction was terminated by adding 1.0 mL of 0.5 mol CaCl 2 and 4.0 mL of 0.5 mol NaOH to the mixture prior to ltration. The absorbance of the released ρNP was measured at 400 nm using a spectrophotometer (UV-1700, Shimadzu, Japan), and the phosphatase activity is expressed in mg ρ-NP g − 1 h − 1 .
Aminopeptidase activity was measured using the method described by Pansombat et al. (1997) with 0.002 M N-benzoyl-Lxycarbonylglycyl L-phenylalanine (ZGP). The absorbance at a wavelength of 570 nm was measured using a spectrophotometer (UV-1700, Shimadzu, Japan). All analyses were conducted with 5 replicates.

Analysis of soil microbial diversity
Microbial community genomic DNA was extracted from samples using the E.Z.N.A.® soil DNA Kit (Omega Bio-tek, Norcross, GA, U.S.) according to the manufacturer's instructions. The DNA extract was checked on a 1% agarose gel, and the DNA concentrations and purity were determined with a NanoDrop 2000 UV-vis spectrophotometer (Thermo Scienti c, Wilmington, USA). PCR ampli cation and sequencing of the total DNA extracted from the rhizosphere soil samples were performed by Shanghai Majorbio Biopharm Technology Co., Ltd, (Shanghai, China) while PCR ampli cation was performed using an ABI GeneAmp 9700 instrument (ABI, USA), and the PCR products were recovered using 2% agar-gel electrophoresis. The products were puri ed by using an AxyPrep DNA Gel Extraction Kit (Axygen, USA) and quanti ed using a Quantus Fluorometer (Promega, USA). The puri ed amplicons were pooled in equimolar quantities and were paired-end sequenced (2×300) on the Illumina MiSeq platform (Illumina, San Diego, USA) according to the standard protocols of the Majorbio Bio-Pharm Technology Co. Ltd.
(Shanghai, China). Raw reads were deposited in the NCBI Sequence Read Archive (SRA) database (Accession Number: SRP284471).

Statistical analyses
The experimental data were analyzed using Excel 2019 and IBM SPSS Statistics 21, and the results are shown as means with their standard deviations (mean ± SD). Online data analysis was conducted by using the free online platform of the Majorbio Cloud Platform (www.majorbio.com) of the Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China).

Soil enzyme activities
The activities of soil β-glucosidase in the treatments using sugarcane-soybean and sugarcane-mung bean intercropping were signi cantly higher than those in the monoculture and sugarcane/peanut intercropping treatment (Table 1). No signi cant difference in soil β-glucosidase activity was observed between sugarcane-peanut intercropping and the monoculture. The highest β-glucosidase activity was found in the sugarcane-mung bean system, which was signi cantly greater than that in sugarcanesoybean treatment (Table 1). Acid phosphatase activity showed nearly the same trend as that of βglucosidase except that there were no signi cant differences between the sugarcane-soybean and sugarcane-mung bean intercropping treatments. Aminopeptidase activity was signi cantly different among all treatments, with the sugarcane-peanut system showing slightly lower activity than that of the monoculture (Table 1). Table 1 Soil enzyme activities (nmol g − 1 min − 1 at 30°C) in the sugarcane monoculture and different sugarcanelegume intercropping systems Notes: All data are presented as means ± SD (standard deviation). Different letters in the same column indicate signi cant differences among treatments at P < 0.05

Soil microbial biomass
As shown in Table 2, the soil microbial biomass carbon (MBC), nitrogen (MBN) and phosphorus (MBP) contents were highest in the sugarcane-mung bean intercropping treatments. All of these three parameters were signi cantly higher in the sugarcane-soybean and sugarcane-mung bean intercropping systems than in the sugarcane-peanut and monoculture treatments. The soil microbial biomass C content in the sugarcane-peanut treatment was signi cantly lower than that in the monoculture, but the opposite trend was observed for MBP ( Table 2). The soil microbial biomass N contents in the sugarcanepeanut and monoculture treatments remained similar to those of MBN. The relative numbers of cultivable bacteria, fungi and actinomycetes in the soils of the sugarcanesoybean, sugarcane-mung bean and sugarcane-peanut treatments and the sugarcane monoculture followed a somewhat similar pattern, as the MBC, MBN and MBP contents (Table 3). In particular, the sugarcane-soybean and sugarcane-mung-bean systems were superior to the other two treatments.
All of these results show that the intercropping treatments not only changed the proportions of dominant soil bacteria but also altered the compositions and structures of soil bacterial communities (Table 5). A: sugarcane intercropped with soybean, B: sugarcane intercropped with mung bean, C: sugarcane intercropped with peanut, and CK: sugarcane monoculture.
The total numbers of bacteria in the sugarcane-soybean, sugarcane-mung bean, sugarcane-peanut and monoculture soil samples were 505, 492, 403 and 464, respectively ( Fig. 3-1A, B, C, and CK). Additionally, the numbers of unique bacteria in these treatments (in the same order) were 28, 22, 0 and 2, respectively ( Fig. 3-1A, B, C, CK). The total numbers of bacteria at the operational taxonomic units (OTUs) level in the sugarcane-soybean sugarcane-mung bean, sugarcane-peanut (C) and sugarcane monoculture treatments were 2,609, 2,553, 1,910 and 2,358, respectively ( Fig. 3-2A, B, C, CK). The unique bacteria numbers in these respective treatments were 214, 118, 23 and 49 ( Fig. 3-2A, B, C, CK). These results indicate that the soil bacterial community structures of sugarcane soils can be signi cantly altered by legume intercropping, particularly with soybean and mung bean.

Discussion
Previous studies have shown that intercropping is more bene cial to soil nutrient enrichment, plant nutrient acquisition and productivity than the corresponding monocultures. For example, researchers in China and sub-Saharan Africa have observed better crop yields and resource use even under adverse production conditions by using cereal-legume intercropping than monocultures (Kumar et al. 1998 legumes are known to contribute plant available nitrogen and boost yields and soil health, a mechanistic understanding of the intercropping-dependent improvement of soil health and crop yield remains poorly understood. As mentioned before, low soil fertility is a critical limiting factor for cane yields and quality in China (Zeng et al. 2015). Soil acidi cation and all accompanying soil biotic and abiotic constraints for sugarcane crop improvement are widespread in many sugar-producing regions in China. In this context, legume intercropping with sugarcane appears to be a very desirable crop management strategy for reversing soil degradation and improving soil fertility and crop productivity, as has been observed for other broadacre crops (Zhang and Li 2003, Mucheru-Muna et al. 2010. Collectively, the results obtained in this study also support this assertion. For instance, soil microbial biomass is an important indicator of soil quality, soil fertility and crop productivity (Powlson et al. 1987). The greater the microbial biomass in soil, the greater the capacity of the soil to provide plant nutrients by mineralization of organic nutrients (Dwivedi and Soni 2011). Soil microbial biomass carbon promotes the formation of new humus and increases the soil total carbon content (Doran et al., 1996). Similarly, soil microbial biomass nitrogen re ects the availability of soil nitrogen to crops and plays an important role in soil nitrogen turnover and supply (Doran et al. 1996). Additionally, turnover of soil microbial biomass phosphorus, although it is not directly available to plants, releases inorganic phosphorus, which is very important for plant growth (Khan and Joergensen 2009). In this study, soil microbial biomass carbon, nitrogen and phosphorous were shown to be signi cantly higher in the sugarcane-mung bean and sugarcane-soybean intercropping treatments than in sugarcane-peanut treatments and sugarcane monoculture. These results suggest that soil microbial biomass in sugarcane elds can be improved considerably by intercropping with soybean, mung bean or, with a much reduced effect, peanut.
Soil enzymes are produced by microorganisms, other soil organisms and plant roots, and they have key biochemical functions, such as decomposing organic matter in the soil system (Ellert et al. 1997), and thereby release nutrients that are readily available for crop uptake. Soil enzymes also play an important role in facilitating microbial processes in the soil that stabilize soil structure, balance soil microbial ecology and drive nutrient cycling (Dick et al. 1994). In our study, the activities of soil β-glucosidase, aminopeptidase and acid phosphatase in sugarcane-soybean and sugarcane-mung bean intercropping systems were signi cantly higher than in sugarcane monocultures. This nding parallels the soil microbial biomass levels observed in our study, which indicate a signi cant contribution to nutrient cycling that is facilitated by increased levels of soil micro ora in legume-intercropped sugarcane crops.
Intercropping with soybean or mung bean thus signi cantly accelerates soil carbon, nitrogen and phosphorus cycles in sugarcane elds and promotes soil fertility and healthy soil ecology, which in turn, result in better crop performance Soil microbial populations and their compositions are closely related to soil quality, which make them an ideal indicator of soil health (Brookes 1995, Zhang et al. 2014. In our study, we found that the soil microbial population in sugarcane monocultures can be signi cantly improved by intercropping with soybean and mung bean but not as effectively with peanut. High-throughput sequencing of microbial populations revealed that the dominant bacteria at the phylum level represented eight phyla in sugarcane monocultures, e.g. Actinobacteria, Proteobacteria, Chloro exi, Acidobacteria, Firmicutes, Planctomycetes, Bacteroidetes, and Gemmatimonadetes groups. In addition to these eight bacterial phyla in sugarcane monoculture elds, sugarcane-soybean soils were additionally enriched with Nitrospirae, while Saccharibacteria formed another dominant phylum in sugarcane-mung bean intercropped soils. Additionally, it is noteworthy that Saccharibacteria was also enriched in sugarcane-peanut intercropped elds. In addition to the new added dominant phyla, the orders of the dominant bacteria, based on their abundance levels, were also changed in intercropped soils. For instance, Proteobacteria was the third dominant phylum in sugarcane-peanut treatments but was rst and second most abundant in sugarcanesoybean and sugarcane-mung bean systems, respectively. Generally, Proteobacteria are considered to be copiotrophic microorganisms, which thrive under conditions of high nutrient availability (Chen et al. 2016). These results suggest that legume intercropping promotes nutrient-tolerant soil bacterial community structures in sugarcane elds and thus positively impacts soil fertility. As the microbial community structures in rhizospheres can be changed by one or synergistically by both plant species , Song et al. 2007a, Song et al. 2007b), we found signi cant variations in bacterial community compositions and abundances in intercropped soils compared to monoculture treatments.
Collectively, our data suggest that intercropping systems, such as sugarcane-soybean and sugarcanemung bean, are bene cial for improving soil biology and ecology, soil structure and soil fertility and lead to superior sustainable sugarcane agriculture. It is very likely that sugarcane-legume intercropping will also help meet the increasing demand for agricultural intensi cation and diversi cation without compromising the environmental obligations and economic outcomes of sugarcane agriculture.

Conclusion
In this study, a eld experiment was carried out to elucidate the effects of intercropping sugarcane with different legumes on soil biological properties, soil bacterial diversities and community structures. The conclusions are as follows: (1) The biological indicators of soil fertility in sugarcane elds, such as the activities of soil cultivable microorganisms (e.g., bacteria, fungi and actinomycetes), soil enzymes (e.g., β-Glucosidase, acid phosphatase, and aminopeptidase) and microbial biomass carbon, nitrogen, phosphorus, were all signi cantly improved by intercropping sugarcane with soybean and mung bean.
(2) Soil bacterial diversity and richness in sugarcane elds were also signi cantly enhanced by sugarcane intercropping with soybean and mung bean.
(3) By intercropping, new bacterial phyla, such as Nitrospirae or Saccharibacteria, became the dominant groups in intercropped soils.
(4) Proteobacteria, which thrive under conditions of high nutrient availability, became the most and second-most dominant bacterial group in sugarcane-soybean and sugarcane-mung bean systems.
(5) Sugarcane-mung bean intercropping showed the greatest effects for improving soil fertility and soil health among the cropping systems studied in this work. Figure 1 Compositions of soil bacterial communities at the phylum level among the sugarcane monoculture and sugarcane-legume intercrops. A: sugarcane intercropped with soybean, B: sugarcane intercropped with mung bean, C: sugarcane intercropped with peanut, and CK: sugarcane monoculture.

Figure 2
Compositions of soil bacterial communities at the genus level among the sugarcane monoculture and sugarcane-legume intercrops. A: sugarcane intercropped with soybean, B: sugarcane intercropped with mung bean, C: sugarcane intercropped with peanut, and CK: sugarcane monoculture.