Dynamics of microbial diversity during the composting of agricultural straw

CHANG Hui-qing,ZHU Xiao-hui,WU Jie,GUO Da-yong,ZHANG Lian-he,FENG Yao

1 Henan University of Science and Technology,Luoyang 471023,P.R.China

2 Institute of Agricultural Resources and Regional Planning,Chinese Academy of Agricultural Sciences/Key Laboratory of Plant Nutrition and Fertilizer,Ministry of Agriculture and Rural Affairs,Beijing 100081,P.R.China

Abstract The dynamic changes in microbial diversity during the aerobic composting of agricultural crop straw with additives were evaluated using high-throughput sequencing at four phases of composting (mesophilic,thermophilic,cooling and maturation phases).In addition,the physicochemical parameters of the composting system were determined in this study.The fermentation time of the thermophilic period was prolonged with the addition of urea or urea combined with a microbial agent.The ratio of C/N and germination index variation indicated that the additives were favorable for composting,because the additives directly changed the physicochemical properties of the compost and had effects on the diversity and abundance of bacteria and fungi.The abundance of operational taxonomic units (OTUs),diversity index (Shannon) and richness index(Chao1) of fungi and bacteria were found to significantly increase when urea+microbial agents were added to straw in the thermophilic phase.The relative abundance of the predominant bacteria and fungi at the phylum and genus levels differed during different composting phases.The abundance of the phyla Firmicutes and Proteobacteria declined in the order of treatments SNW>SN>S (S is straw only compost;SN is straw+5 kg t–1 urea compost;and SNW is straw+5 kg t–1 urea+1 kg t–1 microbial agent compost) in the thermophilic phase.The abundance of the genera Staphylococcus,Bacillus and Thermobifida followed the same order in the mesophilic phase.Ascomycota accounted for more than 92% of the total fungal sequences.With the progression of the composting process,the abundance of Ascomycota decreased gradually.The abundance of Ascomycota followed the order of S>SN>SNW during the thermophilic phase.The abundance of Aspergillus accounted for 4–59% of the total abundance of fungi and increased during the first two sampling periods.Aspergillus abundance followed the order of SNW>SN>S.Additionally,principal component analysis (PCA) revealed that the community compositions in the straw and straw+urea treatments were similar,and that the bacterial communities in treatments S,SN and SNW in the mesophilic phase (at day 1) were different from those observed in three other phases (at days 5,11,and 19,respectively),while the fungal communities showed only slight variations in their structure in response to changes in the composting process.Canonical correlation analysis (CCA) and redundancy analysis (RDA) showed that total carbon(TC),NO3–-N (NN),electrical conductivity (EC) and pH were highly correlated with community composition.Therefore,this study highlights that the additives are beneficial to straw composting and result in good quality compost.

Keywords:crop straws,microbial diversity,community composition,dynamic characteristics,high-throughput sequencing

1.Introduction

In China,a large amount of crop straw is produced every year.Approximately 900 million tons of crop straw was produced in 2017,and only approximately 30% of this amount was used (Wanget al.2018).Much of the straw is disposed of by burning,being discarded or being used for mulching directly in the field.The straw causes many problems for farmers as well as the environment if not properly handled.Although the application of straw or strawderived biochar in the field is an alternative option for its cost-effective utilization (Yuanet al.2019;Zhaoet al.2020),the soil application of large doses of non-decomposted plant residues can lead to unfavorable effects on subsequent plant growth and crop yields due to the production of phytotoxic allelochemicals (Chunget al.2001;Inderjitet al.2004) and the spread of diseases and insect pests that originate from the crop straw.Burning or discarding the straw seriously harms the environment.Therefore,one of the best alternatives for managing this resource is its bioconversion to fertilizer through composting,because crop straw contains large amounts of N,P,K and micronutrient elements.Aerobic composting has been considered an attractive alternative for disposing of crop straw due to the reduction of environmental pollution,the lack of harmful effects of the final product and the effective utilization of this resource (Kleinet al.2011;Kanget al.2014;Liet al.2016).This process depends on bacterial,actinobacterial and fungal activities,which not only play an important role in the composting process (Novinscaket al.2009;Bonitoet al.2010;Adrianet al.2014;Zhang W Het al.2018),but also are responsible for the rapid succession of the community structure with the composting process.Studies have revealed that microbial abundance and diversity in the composting process depend on the type of straw and fermentation conditions including the temperature,oxygen concentration,moisture content and C/N ratio of the straw(Tkachuket al.2014;Covinoet al.2016).

The C/N ratio,one of the indexes of compost maturity,must be less than 50:1 at the initiation stage to achieve an acceptable rate for the composting process (Tuomelaet al.2000).A C/N ratio of compost materials of (25–30):1 is generally considered suitable for microbial fermentation and decomposition of rice straw or other agricultural solid waste (Harutaet al.2002;Zhang Met al.2018).Crop straw generally exhibits a relatively high C/N ratio.In particular,paddy straw,one of the most recalcitrant materials among crop straws,is not easily degraded due to its high C/N ratio and silica content,but some vegetable or fresh crop straws present a lower C/N ratio (Liet al.2016;Maet al.2018).Therefore,different straws can be mixed to obtain a suitable C/N ratio and achieve rapid decomposition by microorganisms.In some provinces of China,due to the high rate of replanting,the time for disposal or reuse of straw is very short and it is therefore important for straw decomposition to be rapid.The composing rate of straw is related not only to its C/N ratio,but also to the bioavailability of the nutrients in the straw at the initial stage of composting.In the early stage of straw composting,the amount of nitrogen available to microorganisms is usually low.As urea can be directly metabolized by chemoautotrophic microorganisms as a singular and stable inorganic nitrogen source,it is commonly used as a nitrogen supplement during straw composting (Zhanget al.2005;Huanget al.2013;Wanget al.2018).Microbial agents are often used to accelerate the maturation of organic waste compost (Xuet al.2005;Jusohet al.2013).The effects of microbial agents on physical and chemical parameters during composting have been studied by Knerr and Tripepi (2012),Wanget al.(2018) and Zhuet al.(2018).

The process of composting primarily relies on microbial activities and interactions among the highly diverse microbial communities.The composition and succession of microbial communities are closely related to the maturity and quality of the compost (Tianet al.2013;Hsset al.2015).Bacteria play an important role in degrading proteins,lipids,cellulose,lignin and even toxic compounds during composting(Takakuet al.2006;Watanabeet al.2009;Partanenet al.2010;Karadaget al.2013).In addition,bacteria have a remarkable ability to adapt to harsh conditions,such as high temperature or low pH (Liet al.2013).Fungi are of vital importance in the decomposition of composting materials due to their filamentous growth,extracellular mode of obtaining nutrition,and secretion of extracellular enzymes;and they are essential during composting for the degradation of recalcitrant compounds and stable organic matter as well as the release of nutrient elements that are beneficial for plant growth and fertility (Abdel-Wareth 2017).However,the effects of urea and microbial agents on bacterial and fungal diversity during the composting of crop straws have not been well documented.Therefore,in this study,the diversity and dynamic variation of the bacterial and fungal community structure was investigated in composting straw originating from rice,corn and tomato crops with the addition of urea and microbial agents,using high-throughput sequencing technology.In addition,the relationships between the microbial communities and the physicochemical factors of composting were evaluated.This study may contribute to a better understanding of the microbial community involved in composting and promote the improvement of straw composting technology.

2.Materials and methods

2.1.Composting procedures

The crop straw used in this study was dried and crushed to a 2–3 cm particle size for storage.The straw carbon and nitrogen concentrations were determined using a Vario MAX Carbon and Nitrogen Analyzer (Elementar,Germany).The carbon and nitrogen mass fractions of the rice,corn and tomato straws were 40 and 0.8%,36 and 1.5%,and 32 and 1.6%,respectively,while the C/N ratios were 50:1,24:1 and 20:1,respectively.Corn,rice and tomato stalks were mixed at a mass ratio of 4:1:2.This straw ratio was conducive to the rapid fermentation of compost,and the ratio of carbon to nitrogen in the mixture was 27.8:1.

The compost site was set up at the Foshan Institute of Agricultural Sciences of Guangdong Province in China(23°32´16´´N,112°51´25´´E),where the average annual temperature and rainfall were 22.5°C and 1 681 mm,respectively.Three treatments were applied to the mixed straw as follows:(1) straw alone (S),(2) straw+5 kg t–1urea (SN),and (3) straw+5 kg t–1urea+1 kg t–1microbial agent (SNW).The urea contained 46% nitrogen and the microbial agent provided by Henan Yangshao Biochemical Engineering Co.,Ltd.(China) includedBacilluslicheniformis,Bacillusamylolytius,Saccharomycessp.,Aspergillussp.,protease and cellulase.Each treatment was repeated three times.The urea was dissolved in water and sprayed on the surface of composting materials that were then mixed thoroughly with the microbial agents.This procedure was repeated until all of the solutions of urea and microbial agents were added to the composting material.

The composting materials were stacked in reactors that were 0.7 m long,0.6 m wide and 0.5 m high.At the beginning,the moisture content of the materials in each treatment was controlled at 60%.When the temperature rose to 60°C,the composting pile in each reactor was turned every two days.Samples were collected at four stages of composting:day 1 (mesophilic phase),day 5 (thermophilic phase),day 11 (cooling phase),and day 19 (maturation phase).In order to guarantee a representative sample of the pile,samples were randomly collected from five different sites in three different layers (surface,middle and bottom)of the reactors,and mixed thoroughly.One part of the fresh sample was placed in a 50-mL sterile centrifuge tube and stored in a refrigerator at–80°C for subsequent microbial diversity analysis.The rest of the sample was divided into two parts,one of which was stored in the–20°C refrigerator and the other part was air-dried and sieved to a 0.12-mm particle size for the determination of physicochemical properties.

2.2.Physicochemical analysis

There are many maturity indicators for composting.In this study,temperature,pH,EC,total carbon (TC),total nitrogen(TN),C/N ratio (TC/TN),NH4+-N (AN),NO3–-N (NN) and the germination index (GI) were determined to indicate the maturity of the compost (Zhang Met al.2018).

The temperature at the center of each reactor was measured daily with a digital thermometer during composting.pH and EC were measured in a 1:10 (w/v) water-soluble extract using a pH meter and an EC meter,respectively.TC and TN were determined by the potassium dichromate volumetric method and the Kjeldahl method (Bao 2008),respectively.The ratio of TC to TN was reported as the C/N ratio.AN was extracted with 2.0 mol L–1potassium chloride and analyzed by indophenol blue colorimetry.NN was extracted with a saturated calcium sulfate solution and analyzed by MgO distillation and FeSO4Zn reduction method (Bao 2008).The GI was determined using seeds of ChineseBrassicachinensis.The compost samples were added to deionized water at a ratio of 1:7.5 (mass concentration),followed by shaking at 25°C for 3 h.Then,5 mL of the extract was dripped into a culture dish containing filter paper.Twenty seeds were evenly distributed on the filter paper and cultured in the dark at 25°C for 48 h(Zucconiet al.1981).The germination index of the seeds was calculated according to the following formula:GI (%)=average germination rate of treatment×Average root length of treatment/(Average germination rate of control×average root length of control)×100,where the control was treated with 5 mL of distilled water instead of the extract.All analyses were carried out in triplicate.

2.3.DNA extracted and high-throughput sequencing

In order to clarify the change of the microbial community during the composting process,the bacterial 16S rDNA(V3–V4) region and the fungal ITS1 region were used as the DNA regions of conserved microbial diversity for PCR amplification.The universal primers used for bacteria were 338F:5´-ACTCCTACGGGAGGCAGCA-3´,and 806R:5´-GGACTACHVGGGTWTCTAAT-3´;and the universal primers used for fungi were ITS1F:5´-CTTGGTCATTTAGAGGAAGTAA-3´,and ITS2:5´-GCTGCGTTCTTCATCGATGC-3´.DNA was extracted based on the procedures described by Tianet al.(2013)and Awasthiet al.(2017).The PCR amplification products were purified,quantified and homogenized to generate a sequencing library.Finally,the Illumina HiSeq 2500 Platform was used for the high-throughput sequencing of qualified libraries.

2.4.Microbial data analysis and processing

The two-terminal sequence data obtained by Hiseq sequencing were merged into sequence tags,FLASH ver.1.2.7 Software was used for PE read stitching,Trimmomatic ver.0.33 Software was used for tag filtering,UCHIME ver.4.2 Software was used to remove chimeras and to apply other quality control filters,and the effective data were ultimately obtained (effective tags).Using UCLUST in QIIME Software(ver.1.8.0),the tags were clustered at a 97% similarity level to obtain operational taxonomic units (OTUs),and the OTUs were annotated based on the Silva (bacteria) and Unite(fungi) taxonomic databases.A columnar diagram of the community structure was drawn using R language tools.Mothur Software (ver.1.30) was used to evaluate the alpha diversity index of the samples.The relative abundance was the ratio of a certain microbial group’s abundance to the total microbial abundance in the same period.Principal component analysis (PCA) was used to analyze OTU similarity among the samples.Finally,canonical correlation analysis (CCA) or redundancy analysis (RDA) in the vegan package in R language was used to reflect the relationships between environmental factors and microflora.In CCA and RDA,environmental parameters are generally expressed with arrows.The length of an arrow link represents the degree of correlation between an environmental factor and the community distribution and species distribution.The longer the link,the greater the correlation,andvice versa.The angle between two arrows indicates the correlation between the two environmental parameters,where an acute angle indicates a positive correlation,and an obtuse angle indicates a negative correlation.The angle between environmental factors and samples is positively correlated with an acute angle and negatively correlated with an obtuse angle.

Seven parameters,including pH,EC,TC,TN,C/N,AN,NN,were selected for studying their correlations with microbial community properties.According to the results of dynamic component analysis (DCA),if the length of the first axis of a gradient is greater than 3.5,CCA is selected,whereas RDA is selected if this value is less than 3.5.The bacterial and fungal axis lengths in this study were 4.3 and 1.8,respectively.Therefore,CCA based on a single peak model was used for bacteria,and RDA based on a linear model was used for fungi.

2.5.Statistical analysis

Microsoft Excel 2017 Software was used for data processing and analysis.Two-factor analysis of variance (ANOVA)in SPSS 22.0 Software was used to test for significant differences and the Tukey HSD method was used to determine if individual means were significantly different.A significant difference was indicated byP<0.05,and an extremely significant difference was indicated byP<0.01.Data are expressed as the mean±the standard deviation(SD).

3.Results

3.1.Variation of parameters in the composting process

Fig.1 Changes in straw temperature during the composting process.S is straw only compost;SN is straw+5 kg t–1 urea compost;and SNW is straw+5 kg t–1 urea+1 kg t–1 microbial agent compost.Bars show standard deviations (n=3).

Temperature,pH and ECThe temperature changes followed similar trends in all treatments,increasing to a maximum value and then decreasing toward the end of composting.In this study,the environmental temperature before composting was 34.5°C.As shown in Fig.1,the changes in temperature clearly indicated the composting phases (mesophilic,thermophilic,cooling and maturation phases).After one day of composting,the temperature of the compost had increased rapidly,and the temperatures under the S,SN and SNW treatments reached 44.4,46.7 and 52.8°C,respectively.On day 5 of composting,the temperatures of all treatments reached their highest levels,with the temperatures of the S,SN and SNW treatments being 68.7,71.7 and 77.6°C,respectively.On day 19,the pile temperatures were close to the ambient temperature.The addition of urea and microbial agents increased the composting temperatures.As composting progressed,the pH values of the S and SN treatments first decreased,and then increased,while the pH of different treatments showed no significant difference at any given sampling time (Table 1).Treatment,sampling time and treatment×sampling time have extremely significant effects on EC,which increased in all treatments as composting progressed,and the EC of the SNW treatment was significantly greater than those of the S and SN treatments at the same first three sampling dates,but there was no significant difference in EC between SN and SNW treatments at the maturation phase (Table 1).

TC,TN and C/NTwo factor ANOVA indicated that treatment and sampling time have extremely significant effects on TC,TN and C/N,but the treatment×sampling time showed no significant effects on these parameters(Table 1).Straw TC,TN and C/N decreased as composting progressed,especially on days 11 and 19.Straw TC concentrations decreased by about 49.0% in all treatments during the composting process,and TC concentrations showed no significant differences at any given sampling time.The total nitrogen concentrations of straw in the SN and SNW treatments were also significantly higher than that in the S treatment at any given sampling time.At each sampling date,total N concentrations were in the order SN>SNW>S.The TN concentrations in the straw in all treatments decreased by 17.0–20.0% during the composting process.The initial C/N ratios of straw in the S,SN,and SNW treatments were 27.5,21.9 and 22.2,respectively.Straw C/N ratios gradually decreased to 16.6,13.0 and 12.9 on day 19,when the C/N ratios of the SN and SNW treatments were significantly lower than that of the S treatment.

AN,NN and GIAN,NN and GI were extremely significantly affected by treatment,sampling time and treatment×sampling time.AN concentrations also declined as composting progressed,the declines being greater in the SN and SNW treatments (72.0%) than in the S treatment(55.0%).On days 1 and 5,straw AN concentrations in both SN and SNW treatments were significantly greater than in the S treatment.On day 19,the straw AN concentration in the SN treatment was significantly higher than in the other two treatments.Straw NN concentrations were the highest on day 1 and lowest on day 11,with differences between all treatments being significant on days 1 and 19.The straw GI was the lowest at the initial phase of composting(38.0–42.0%),and at that stage there was no significant difference between treatments (Table 1).The GI gradually increased with composting time and exceeded 80% in all treatments at the end of composting.At all sampling dates GI declined in the order of SNW>SN>S.At the cooling and maturation stages,the GI of the SNW treatment was significantly higher than those of the other two treatments.

3.2.Dynamics of microbial diversity

Microbial coverage and OTUsThe higher value of microbial coverage indicates the higher of the species in the sample will be detected,otherwise the lower species will be detected.Two factor ANOVA indicated that treatment,sampling time and treatment×sampling time have no significant effect on microbial coverage (Table 2).The coverage of bacterial 16S rDNA (V3–V4) and fungal ITS1 at the different sampling phases reached 99.7–99.9%,indicating that the prepared MiSep library covered over 99.7% of the bacterial and fungal groups in the compost samples and could reflect the true composition of bacteria and fungi in the compost samples.

Two-factor ANOVA indicated that treatment,sampling time and treatment×sampling time have extremely significant effects on microbial OTUs.The numbers of bacterial OTUs in the compost samples ranged from 261 to 345 (Table 3).On days 1 and 5,the addition of urea and microbial agents increased the numbers of OTUs in the samples.In these sampling times,the number of OTUs in the SNW treatment was significantly higher than in the other two treatments,and there were significantly more OTUs in the SN treatment than in the S treatment on day 5.In contrast,the OTU number in the treatment S was significantly higher than those in the other two treatments on days 11 and 19.

The numbers of fungal OTUs in the compost samples ranged from 118 to 186 (Table 3).On days 1 and 5,the addition of urea alone significantly increased the numbers of fungal OTUs and the addition of microbial agents further increased the numbers of fungal OTUs.On day 11,the numbers of fungal OTUs in the S and SN treatments exceeded that in the SNW treatment.The number of fungal OTUs in the S treatment was significantly lower than in the other two treatments on day 19.

Bacterial community compositionThe relative abundance of the top 10 bacterial species at the phylum and genus levels in different treatments and sampling times of straw composting are shown in Fig.2.Firmicutes,Proteobacteria,Actinobacteria and Bacteroidetes were the four dominant phyla in all compost samples,accounting for92.0–100.0% of the total bacterial sequences (Fig.2-A).The abundance of Firmicutes during straw composting ranged from 23.1 to 62.5%.The abundance of Firmicutes was the highest on day 5 and the addition of urea and microbial agents increased the abundance of Firmicutes in this phase.On day 1,the abundance of Firmicutes in the SN and SNW treatments showed the greatest increases (49.3 and 94.8%,respectively),followed by the maturation phase,when there were increases of 15.5 and 25.4%,respectively.The abundance of Proteobacteria varied from 16.3 to 21.3% on day 5 and its abundance in the three treatments declined in the order of SNW>SN>S.In other sampling times,the abundance of Proteobacteria in the treatments followed the order of S>SN>SNW.The abundance of Actinobacteria in each treatment did not change markedly on days 1 and 5.However,on days 11 and 19,the Actinobacteria abundance increased by 52.7 and 31.1%,respectively,in the SN treatment and by 62.7 and 46.8%,respectively,in the SNW treatment.The abundance of Bacteroidetes ranged from 1.1 to 28.6% during the composting period,and the trend in its abundance of the three treatments followed the order of S>SN>SNW.The abundance of Bacteroidetes on days 1 and 19 was higher than at the other two sampling times.

Table 2 Effect of urea and microbial agents on bacterial and fungal coverage (%) in straw compost samples at four sampling times

Table 3 Effect of urea and microbial agents on operational taxonomic units (OTUs) of straw compost samples at four sampling times

The relative abundance of the top 10 bacterial genera differed among the treatments (Fig.2-B).The six genera with the highest abundance were successivelyBacillus,Thermobifida,Sphingobacterium,Staphylococcus,RhodospirillaceaeandAcinetobacter.Among these taxa,Sphingobacterium(19–26%),Staphylococcus(16–28%)andAcinetobacter(11–22%) exhibited the highest abundance on day 1.Thereafter,the abundance of these genera gradually decreased with the composting process.The abundance ofSphingobacteriumandAcinetobacterin the three treatments followed the order of S>SN>SNW during all composting phases,while the abundance ofStaphylococcusfollowed the order of SNW>SN>S.BacillusandThermobifidahad the lowest abundance on day 1.The abundance ofBacillusincreased to its highest level on day 5.The abundance ofThermobifidaincreased most rapidly on day 11,and its abundance in the treatments declined on day 1 in the order of SNW>SN>S.

Fig.2 Relative abundance of bacterial phyla (A) and bacterial genera (B).SNW is straw+5 kg t–1 urea+1 kg t–1 microbial agent compost;SN is straw+5 kg t–1 urea compost;S is straw only compost.The numbers 1,5,11 and 19 after SNW,SN and S indicate the sampling time during composting.

Fungal community compositionThree major fungal phyla,Ascomycota,Mucoromycota,and Basidiomycota,were observed in all samples (Fig.3).As shown in Fig.3-A,the relative abundance of the major phyla differed very little among the different treatments.The fungi in the S treatment at different phases mainly belonged to phylum Ascomycota,which accounted for more than 92% of the total fungal sequences.With the progress of the composting process,the abundance of Ascomycota decreased gradually.The abundance of Ascomycota in the different straw treatments followed the order of S>SN>SNW on day 5,and SNW>SN>S at the other three sampling times.On days 1 and 5,the abundance of Mucoromycota was in the order of SNW>SN>S,and the abundance was lower than on days 11 and 19.The abundance of Basidiomycota followed the order of S>SN>SNW on day 1,whereas its order was SNW>SN>S on day 5.

The relative abundance of the top 10 fungal genera differed among the treatments (Fig.3-B).Aspergillus,Candida,Thermomyces,Melanocarpus,Penicillium,andMycothermuswere the six most dominant genera in all compost samples.Among these genera,Aspergillusaccounted for 4.6–59.1% of the total abundance,and its abundance increased during the first three sampling periods.In the three treatments,Aspergillusabundance followed the order of SNW>SN>S.Candidahad the highest abundance(more than 54.0%) on day 1,but its abundance decreased thereafter,falling to 6.0% on day 19,when it followed the order of SNW>SN>S in the different treatments.On day 1,the relative abundance ofThermomyces,Melanocarpus,Penicillium,andMycothermuswas lower than the other sampling phases,and then increased as composting progressed.The abundance ofKodamaeaandMeyerozymadecreased on day 5 and then increased on day 11.However,when fermentation proceeded to the maturation phase,the relative abundance of these genera decreased,and followed the order of SNW>SN>S in the three treatments during the first three sampling periods.

Fig.3 Relative abundance of fungal phyla (A) and fungal genera (B).SNW is straw+5 kg t–1 urea+1 kg t–1 microbial agent compos;SN is straw+5 kg t–1 urea compost;S is straw only compost.The numbers 1,5,11 and 19 after SNW,SN and S indicate the sampling time during composting.

Diversity and richness of bacterial and fungal communitiesTreament,sampling time and treatment×sampling time have extremely significant effects on the diversity index (Shannon) and richness index (Chao1) of the bacterial and fungal communities in compost samples(Table 4).The Shannon and Chao1 indexes of bacteria in the last three composting phases were higher than those on day 1,and these two indexes were the highest on day 5.The application of urea plus microbial agents had a significant (P<0.05) effect on bacterial diversity on day 1,and on bacterial richness in the first two sampling phases.The Shannon and Chao1 indexes of fungi were lower than those of bacteria.In each of the first three sampling times,the order of the indexes was SNW>SN>S.The highest Shannon and Chao1 indexes for fungi in the SN and SNW treatments appeared on days 11 and 5,respectively.Compared with the S treatment,the application of urea and microbial agents together had a significant (P<0.05) effect on the diversity and richness of fungi in each of the sample days (except the samples of day 11).

Bacterial and fungal community structure analysisPCA based on the weighted UniFrac algorithm revealed that the bacterial and fungal communities varied among the treatments and the composting phases (Fig.4).For the bacterial community structure,the cumulative percentage variance of the samples explained by the first principal component axis (PC1) was 50.0%,and PC1 and the second principal component axis (PC2) together explained 65.8%of the variance of the samples (Fig.4-A).The samples on day 1 were separated from the samples of the other three sampling phases along the PC1 axis,and the bacterial community structures were similar in the latter three sampling phases.There was no significant separation between the treatments along the PC2 axis.On days 1 and 5,the distances between the S and SN treatments were shorter than the distances between the S and SNW treatments.The results indicate that the bacterial community structures in treatments with the addition of microbial agents differed from those without addition of microbial agents,but the effect of microbial agent addition was small on day 11.

The results of the PCA for fungi are shown in Fig.4-B.The cumulative percentage of the variance in the fungal communities in the samples explained by the PC1 was 69.7%,and the PC1 and PC2 axes together explained 82.6% of the variance in the fungal communities in the samples.The fungal community structures showed great differences in different sampling phases along both PC1 and PC2 axes.Moreover,the fungal community structures in S and SN treatments were similar and both differed from SNW treatment,indicating that the addition of microbial agents influenced the fungal community structure.

RDA and CCAThe correlations between microbial community and physicochemical parameters of different samples varied during the composting process.The correlations between the main environmental parameters and the bacterial community are shown in Fig.5-A.The first two ordination axes explained 33.3 and 9.7% of the variation in the bacterial community.The projection of the environmental parameters on the first axis showed that TC,NN,EC and pH were ranked as the first correlated with bacterial community composition;there were also positive correlations between TC and NN,pH and EC.The influence of the C/N and AN parameters on bacterial community structure was ranked as the second,and that of TN was the weakest.The RDA showed the relationships between environmental parameters and the fungal community(Fig.5-B).The first two ordination axes only explained 17.1 and 9.7% of the fungal community variation,respectively.TC,NN,EC,and pH were ranked as the first correlated with fungal community composition,and there were also positive correlations between TC and NN,EC and pH.Thecorrelation between the C/N,NN and fungal community composition was ranked second,while TN showed the weakest correlation with fungal community composition.

Table 4 Effects of urea and microbial agents on diversity indexes of straw compost samples at four sampling dates

Fig.4 Principal component analysis (PCA) of bacteria (A) and fungi (B).S is straw only compost;SN is straw+5 kg t–1 urea compost;and SNW is straw+5 kg t–1 urea+1 kg t–1 microbial agent compost.The numbers 1,5,11 and 19 after S,SN and SNW are the sampling time during composting.

Fig.5 Canonical correlation analysis (CCA) of bacteria (A) and redundancy analysis (RDA) of fungi (B).S is straw only compost;SN is straw+5 kg t–1 urea compost;and SNW is straw+5 kg t–1 urea+1 kg t–1 microbial agent compost.The numbers 1,5,11 and 19 after S,SN and SNW are the sampling time during composting.

4.Discussion

4.1.The effects of additives on the physicochemical parameters of straw composting

Successful composting depends on a number of environmental factors that both directly and indirectly influence the activities of microorganisms (Chroniet al.2009).During composting,the addition of additives affects the physicochemical parameters of straw composting,thus affecting the structure and diversity of microorganisms in the pile as well as the straw compost quality.pH can indirectly affect the microbial community structure by changing the physicochemical properties of compost (Maet al.2016).The pH range in the present study was narrow,varying from 7.2 to 7.9,and treatment and sampling time had significant and extremely significant effects on pH values.In general,the pH shows a decrease in the initial stage of composting.This decrease may be due to the accumulation of organic acids caused by microbial metabolism during composting (López-Gonzálezet al.2015).With the progression of composting,the pH increases due to the production of ammonia by ammonification and the decomposition of organic acids (Liuet al.2011).After the second turning,the pH may gradually decrease because of the volatilization of ammonia and consumption of ammonium (Rashadet al.2010).In the present study,pH decreased in the S and SN treatments on day 5,but in the SNW treatment pH showed a continuous increase during composting.pH ranges from 7.0–8.0 in decomposed compost (Montserrat and August 2008) and the pH of the final compost product in this experiment was within this range.

There was rapid progression from days 1 to 5 in the three treatments in this study.The temperature in the SNW treatment reached 77.6°C,and remained above 50°C for more than 13 days.The results showed that the addition of urea and microbial inoculation can promote temperature increases in the compost.It is generally believed that if the temperature of compost is maintained at 55°C for more than 3 days (or above 50°C for 5–7 days),the pathogens contained in the compost can be killed,and the requirements of the compost hygiene index and compost maturity can be met (Liuet al.2014).Therefore,the three treatments performed in this study were able to achieve acceptable composting hygiene,since a composting temperature of more than 50°C was maintained for more than 7 days in all treatments in this experiment.However,the time required to reach a temperature of 50°C in the treatments with urea and urea+microbial agents was shorter than in the straw only treatment,and the duration of high temperature (>50°C)was longer.The positive effect of microbial agents on hightemperature composting of straw has been confirmed in many studies (Aloket al.2009).The addition of urea as a supplementary nitrogen source also played a positive role in the initiation and promotion of the composting process,which is consistent with Zhanget al.(2005).

The use of carbon and nitrogen by the microorganisms for energy production and growth led to changes in the C/N ratio during composting (Karadaget al.2013).The C/N ratio is an important parameter for assessing the maturity of the compost (Wanget al.2013) and can affect the composition of the microbial community during the composting process(Liet al.2017).In this study,application of urea and microbial agents resulted in the loss of more carbon than in the straw only treatment during the composting process.These results are in accordance with previous studies(Tumuhairweet al.2009;Makanet al.2013).Compost is generally considered to be mature (Garciaet al.1992) when the ratio of C/N is 10–20 and the final (C/N)/initial (C/N)ratio reaches 0.5–0.6.On day 19,the C/N ratio met these requirements,and the results showed that the addition of urea and microbial agents to the compost was beneficial to decreasing the C/N ratio and accelerating the composting process.

The GI is generally one of the most accurate biological indicators for evaluating compost maturity (Wanget al.2013;Zhang Met al.2018).It is generally believed that GI values greater than 80.0% indicate a loss of compost toxicity and maturity of compost (Paredeset al.2005).In the initial stage of composting,a low GI value might be due to the production of fatty acids or a high NH4+-N content (Sundberget al.2004).In this study the GI gradually increased during the composting process.The GI exceeded 80% after 19 days,which indicated that the final compost product was not toxic to seed germination.The addition of urea and microbial agents significantly increased the GI on day 19.

Straw TC concentrations declined by 49.7,52.0 and 53.5% in the S,SN and SNW treatments,respectively.Although the addition of urea and microbial agents caused no significant difference (P<0.05) in the compost,the present results were similar to those reported by Benitoet al.(2003).According to Tumuhairweet al.(2009),a reduction in TC content is due to pronounced microbial activity.The decreases in the TN concentrations observed at the end of the composting process were in accordance with the results obtained by Tognettiet al.(2007).It has been shown that the number and species of microorganisms existing in compost is directly related to the NH4+-N level in the compost (Wanget al.2011).In this study,urea was added to the composting system and NH4+-N produced in the compost could be used as a direct substrate for the NH3oxidation reaction,resulting in a greater correlation between the NH4+-N concentration and the microbial communities,when compared with TN and C/N.The concentration of NH4+-N decreased during the composting process,which might have resulted from the conversion of NH4+-N to NH3and its subsequent volatilization (Rashadet al.2010;Wanget al.2013).In addition,increases in microbial nitrification and immobilization may cause decreases in NH4+-N concentrations during composting (Raj and Antil 2011).In the present study,the low NO3–-N concentration observed may have been due to the inhibition of the activity of nitrifying microorganisms by the high temperature,resulting in the slow formation and accumulation of NO3–-N (Tiquiaet al.2002;Bernalet al.2009).Generally,a high temperature and a high concentration of NH4+-N will strongly suppress the growth of autotrophic nitrifying bacteria and weaken their nitrification ability (Nakhshinievet al.2014).The NO3–-N concentrations on day 19 were higher than those on day 11,which could be attributed to microbial nitrification(Jianget al.2015).In addition,the mineralization of some portion of organic nitrogen may have contributed to the accumulation of NO3–-N during composting.As shown in Table 1,NO3–-N concentrations increased on day 19.Tiquia and Tam (2002) reported that nitrification rarely occurs at temperatures above 40°C,which indicates that the nitrification mostly occured on days 11 and 19.

4.2.Compositional changes in bacterial and fungal communities during composting

The essence of straw composting is in the biochemical metabolic processes of microbial decomposition and the transformation of organic matter.The metabolic capacity and structure of the microbial community are the key factors contributing to the effects of aerobic composting.With the progression of composting,the abundance of the microbial community and the composition of the dominant microbial community undergo changes.The proportions of dominant microorganisms differ in different composting systems.

Richness is a representation of the number of species present in a sample,which,for bacteria,is expressed using the Chao1 index in this study.According to this index,the thermophilic phase was the most important for maturation,because 85% of the dominant bacterial species were abundant during composting at high temperature (Antuneset al.2016).These dominant bacterial species contribute to the decomposition of the different components of straw.In addition,the thermophilic phase is also where the removal of the harmful pathogenic bacteria which exist in the straw occurs.The Chao1 indexes of the treatments with added urea and microbial agents were the highest on day 5 (the thermophilic phase),and were higher than the straw only treatment.It may be that the addition of N and microbial agents improved the bacterial species richness.Diversity is commonly measured using the Shannon index (Liet al.2012).The higher the Shannon index,the more diverse the community.As shown in Table 4,the Shannon index for bacteria showed the highest diversity on day 5,and the indexes in the urea and urea plus microbial agents treatments were higher than that of the straw only treatment.For fungi,the highest Shannon index was observed on day 11.

The composting of different materials usually requires different mesophilic and thermophilic microorganisms that play key roles during composting (Farah Nadiaet al.2015;Wanget al.2016).The relative diversity of the bacterial community at the phylum level remained similar between different treatments and sampling periods in the tested composting system;however,the relative abundance of the phyla,especially that of the predominant phyla,differed with different treatments and sampling stages.Firmicutes,Proteobacteria,Actinobacteria,and Bacteroidetes have been reported as the most abundant phyla during agricultural solid waste co-composting processes (Zhang Met al.2018).According to Fig.2-A,the dominant bacterial phyla in this study were also Firmicutes,Proteobacteria,Actinobacteria,and Bacteroidetes,and their abundance exceeded 92%.Actinobacteria showed a marked increase during the composting process,possibly because Actinobacteria have an extraordinary ability to decompose organic matter(Venturaet al.2007;Xiaoet al.2011).The abundance of Firmicutes decreased in all treatments.Many members of Firmicutes are anaerobic degraders.The regular turning of a compost pile increases air flow,maintaining aerobic conditions within the pile,and aerobic composting can lead to a decrease in anaerobes.A previous study showed that significant differences in the distribution of Acidobacteria were primarily caused by differences in N and pH values.The relative abundance of Acidobacteria decreased with an increase in the amount of NH4+-N (Liuet al.2017).In another study it was found that the abundance of Acidobacteria increased during the composting process due to a decrease in pH (Zhanget al.2014).The increase in the proportion of Firmicutes could be explained by the ability of these bacteria to survive and participate in a variety of metabolic processes at high temperatures (>55°C) (Tianet al.2013).The dominant genera ofBacillus,Thermobifida,Sphingobacterium,Staphylococcus,RhodospirillaceaeandAcinetobacter were similar in abundance in the three treatments at all stages,but the abundance of the genera was different.The addition of urea and microbial agents influenced the abundance of genera during the composting process.It has been reported thatBacillusandThermobifidaspecies are associated with the degradation of lignocellulose in aerobic compost (Peterset al.2000;Szeklyet al.2009).The highest abundance ofBacilluswas found on day 5.This genus harbors the main species of cellulose and lignin degraders (Kausaret al.2011;Eiet al.2016).

Ascomycetes was the most abundant fungal phylum during the process of straw composting.Ascomycetes can secrete a variety of cellulose-and hemicellulose-degrading enzymes,and efficiently utilize nutrients in compost(Singhet al.2003;Salar and Aneja 2007).Zhang W Het al.(2018) found that Ascomycetes gradually became the most dominant group during the static composting of cow dung and rice straw,and the relative abundance of this phylum remained above 75%.In the present study,the addition of urea and microbial agents increased the abundance of Ascomycetes on days 11 and 19.At the genus level,Aspergillus,Candida,Penicillium,andRhizomucor species were found to be the most abundant fungi with developed mycelia.During composting,the mechanical interpenetration of filamentous fungal mycelia can cause the destruction of material to a certain extent and further promote the decomposition of various substances (Guet al.2017).At the same time,fungi secrete extracellular enzymes to accelerate the hydrolysis of organic compounds.Under the action of physical chemistry,the decomposition efficiency of refractory organic compounds is improved(Wanget al.1994).This study showed thatAspergillusbecome more abundant after the thermophilic phase (day 5)of composting,and the abundant presence of these fungi would promote the production of extracellular enzymes such as hydrolytic cellulase,ligninase,protease and lipase and thereby promote the ripening fermentation of the straw(Zhuet al.2013).Thermomyces produce thermostable lignocellulase to degrade lignocellulose (Natviget al.2015),and the abundance of this genus increased during the maturation stage.

5.Conclusion

The application of urea and microbial agents promoted the composting process in this study,promoting the degradation of crop straw.Microbial abundance and diversity indexes differed at different composting phases.The maximum Shannon and Chao1 indexes for bacteria appeared on day 5,but the maximum Shannon indexes for fungi appeared on day 11.Additionally,the application of urea and microbial agents increased the richness and diversity of bacteria and fungi on days 5 and 11,respectively.The addition of urea and microbial agents to the compost changed the relative abundance of bacteria and fungi at the phylum and genus levels during the composting process,although the dominant bacterial and fungal communities were similar.RDA and CCA revealed that several environmental parameters,such as pH,AN,TC and EC,played a predominant role in the distribution of the bacterial and fungal communities during the composting process.Therefore,the addition of urea and microbial agents can change the physical,chemical and biological characteristics of the compost,and these agents can therefore be considered as accelerators for improving the efficiency of crop straw composting.

Acknowledgements

The authors are thankful for foundation support from the National Key R&D Program of China (2017YFD0801304)and the National Natural Science Foundation of China(31972943).

Declaration of competing interest

The authors declare that they have no conflict of interest.