Soil nitrification process played a key role in alleviating continuous cropping limitation induced by fumigation

Continuous cropping causes enormous crop produce reduction, and soil fumigation is an effective approach to alleviate the limitation. Understanding the impacts of agriculture management on microbial community and its association with nutrient availability would provide strong supports for alleviating continuous cropping limitation. However, the mechanisms of fumigants in enhancing plant growth and alleviating continuous cropping barriers was not clear. In this study, fumigation treatments including chloropicrin (CP), dazomet (DZ), and untreated control (CK) were carried out at field scale, and rhizosphere bacterial community and plant phytochrome were analyzed. The results showed that fumigation had strong effects on rhizosphere bacterial community and soil properties. Fumigation treatment caused significantly reduction in rhizosphere bacterial diversity. The nitrifiers (Nitrospira and Nitrospirillum) and functional gene (ammonia oxidizing bacterial AOB amoA) were significantly inhibited by fumigation treatment, which caused significant reduction in nitrification potential (PNF). The inhibition of nitrifiers, AOB amoA gene and PNF led to significant reduction of soil NO3−-N, but increase of NH4+-N. Subsequently, plant photosynthesis was enhanced as a result of increasing leaf chlorophyll a content caused by fumigation treatment. Therefore, fumigation treatment would promote crop growth through increasing the photosynthetic pigment. The study indicated the key mechanisms fumigation promoting plant growth and alleviating cropping limitation were closely related to soil nitrifiers and nitrogen nutrients.

bacterial community and plant phytochrome were analyzed.
Results The results showed that fumigation had strong effects on rhizosphere bacterial community and soil properties. Fumigation treatment caused significantly reduction in rhizosphere bacterial diversity. The nitrifiers (Nitrospira and Nitrospirillum) and functional gene (ammonia oxidizing bacterial AOB amoA) were significantly inhibited by fumigation treatment, which caused significant reduction in nitrification potential (PNF). The inhibition of nitrifiers, AOB amoA gene and PNF led to significant reduction of soil NO 3 − -N, but increase of NH 4 + -N. Subsequently, plant photosynthesis was enhanced as a result of increasing leaf chlorophyll a content caused by fumigation treatment. Therefore, fumigation treatment would promote crop growth through increasing the photosynthetic pigment. Conclusion The study indicated the key mechanisms fumigation promoting plant growth and alleviating cropping limitation were closely related to soil nitrifiers and nitrogen nutrients.

Introduction
Continuous mono-cropping patterns after a few years of use could trigger continuous cropping obstacles which exert negative effects on the production and quality of crops. The negative effects of continuous cropping appeared after several years, including the decline in soil fertility, the deterioration of soil physicochemical properties, and the imbalance of soil nutrients (Adams et al. 2020). Monocrop had limited demands for soil nutrients, resulting in imbalance nutrient consumption and soil infertility. In particular, continuous cropping obstacles led to the retardation of plant growth, the reduction of plant pigment, and the low productivity of crops (de Faccio Carvalho et al. 2010). This was considered to be an increasing challenge for many important crops, such as potato, rice, cotton, and tobacco. It was reported that tuber yields had a significant decrease of 27%, 75% and 85% in the 2, 3 and 4-year continuous cropping soils, respectively compared to the 1-year cropping soils. (Martinez-Feria et al. 2016) found that continuous cultivation of wheat and other grains would result in the loss of about 20% in average yield. For instance, significant decreases in soil organic matter, available Nitrogen, Phosphorous, and Potassium were observed in Argentinean Pampas soil after 7 years of graincropping to a continuous cropping system (Aparicio and Costa 2007). With the prolongation of continuous cropping time, the soil pH had a significant decrease trend (She et al. 2017). (Gonzalez-Chavez et al. 2010) found that soil organic carbon and total nitrogen concentrations were significantly reduced in continuous cropping soil compared with non-continuous cropping soil.
Soil physicochemical and microecological environment were considered to be key factors affecting continuous cropping obstacles, which were important to keep the soil quality. The decrease of soil quality deterioration was proved by low fertility efficiency (e.g. Nitrogen fertilizers) which was related to the change in soil microbial community (Culman et al. 2010). Our previous study showed that the soil microbial community composition, diversity and function were affected by continuous cropping systems (Tan et al. 2021). Soil microbial diversity could mediate many biochemical processes, including the cycling of plant nutrients, maintenance of soil structure and degradation of agrochemicals supporting agricultural production (Pascale et al. 2020). Long-term fertilization caused low nitrogen use efficiency (i.e. urea), which was the result of a decrease of microbial community diversity and enrichment of strong nitrifying bacteria (Sheng et al. 2013). Nitrogen is one of the essential nutrient elements for plant growth (Li et al. 2011). As the nitrogen is the essential component of the chlorophyl, nitrogen use efficiency is important for promoting plant growth. In the continuous cropping field, the nitrification process was enhanced as a result of enrichment of nitrifiers that exhibited strong nitrification ability. Therefore, it was important to mediate nitrification processes in long-term cropping field to keep soil fertility.
Fumigation treatment was a direct and effective method to alleviate continuous cropping limitation, which was widely used. Common and reliable fumigation treatments include Chloropicrin (CP) and Dazomet (DZ) (Duniway 2002). It has been reported that soil fumigants affected the microbial diversity and community composition, changed soil nutrient transformation, and had strong effects on soil fertility and the productivity of agricultural systems (Yan et al. 2013). Soil mineral nitrogen showed a significant increase in CP and DZ fumigated field (Yan et al. 2015), and the similar tendency was found for organic matter, urease and protease activities (Zhang et al. 2019a). It was also reported that fumigation caused a significant increase in the abundance of ammoniaoxidizing bacteria (AOA), but a significant decrease in the abundance of ammonia-oxidizing bacteria (AOB) (Li et al. 2017a). Because AOB dominated the nitrification process in nitrogen rich agriculture soil, and AOB is more sensitive to environmental changes than AOA, (Shahid and Prosser 2006), ammoniaoxidizing bacteria (AOB) inhibited by fumigation would lead to accumulation of NH 4 + -N, but decrease in NO 3 − -N (Zhang et al. 2019a). Due to its ecological importance in regulating soil microbial community and soil nutrient availability, fumigation was widely used in mitigating the continuous cropping limitation. The nitrogen was the important element for plant pigment, change in the nitrogen nutrient availability caused by fumigation would also have significant impact on the plant photosynthesis. Thus, fumigation may also impact the plant photosynthesis. However, whether the change was related to nitrogen uptake, photosynthesis, and growth of plant remained unclear. Moreover, the inherent relationship between microbial community, nitrogen cycling, and plant nitrogen uptake, and the effects in alleviating continuous cropping in fumigation treated field have never been stress enough.
Although previous studies on relieving continuous cropping obstacles by fumigation mainly focused on soil microecology and crop yield, the mechanism of relieving continuous cropping obstacles and promoting plant growth by fumigation has not been explored in depth.
In this study field scale experiments were designed, and two type of fumigants, CP and DZ were used to relieve continuous cropping obstacles. Rhizosphere soil were sampled, bacterial community, phytochrome, and nitrification were analyzed. The purpose of this study was (i) to explore the mechanisms of fumigation to reduce continuous cropping obstacles, and (ii) to evaluate the effect of fumigation on soil properties, rhizosphere bacterial community, potential nitrification rate and plant photosynthetic pigments, and their correlation with plant physiological characteristics.

Experimental design and soil samples
All field experiments were carried out at Huayuan Agricultural Science Park, located in Xiangxi China (109°27′5″E and 28°24′57″N). The site was belonged to subtropical monsoon moist climate region, and had mean annual rainfall of 1031 mm, and mean annual temperature of 17 °C. According to IUSS Working Group WRB (2022) the parent material of the soil was the Metamorphic rock, Slate (MF3), and the surface layer soil was the strongly weathered, slightly moist, weak acidic, reddish sandy loam (SL). The field was divided into 18 equal plots (three treatments × six replicates). Each plot had an area of about 120 m 2 with 8 ridges. According to the crop density, 168 seedlings (8 rows with 21 seedlings in each row) were planted in each plot. The agricultural management practices and fertilization regimes were similar before the experiment were carried out. The fertilization regime was: 50 kg/ha special basal fertilizer, 20 kg/ha special top dressing, 15 kg/ha bio-organic fertilizer, and 5 kg/ha hole-applied fertilizer. The ratio of N/P/K (N, P 2 O 5 , and K 2 O) applied was 1:1.2:2.43. Ploughing and ridging were carried out before the field was divided into equal plots. Fumigation treatments were performed right before transplanting of tobacco seedlings in April, 2019. Three fumigation treatments included chloropicrin (CP), dazomet (DZ), and untreated control (CK). The fumigant CP (99.5% purity) and DZ (98.5% purity) were obtained from Zhejiang Haizheng Chemical Co. Ltd., China. Fumigation treatments were carried out before filed ploughing (30-day before transplanting) and the application amount of CP and DZ were 50 mg kg −1 and 80 mg kg −1 in each plot according to previous report (Yan et al. 2015). The field was ploughed to mix the soil with fumigants. Rhizosphere soil sampling, plant sampling and plant physiological investigation were carried out in Aug, 2019, when plants were at mature stage. Entire plant was pull out to sample the rhizosphere soil. After shaking off the bulk soil, the rhizosphere soil was carefully collected using root brush. In each plot, five rhizosphere soil samples were merge into one composite sample. In each plot, five pulled out plants were used for tissues sampling. Plant leave tissues were sampled using scissors by cutting the longest leaf of each plant. All samples were placed on ice before transport to the laboratory, and were stored in −80 °C freezer before DNA or phytochrome were extracted.

Phytochromes and chlorophylls analysis
Sampled leaves were ice frozen and transported into laboratory. A volume of 25 ml of 90% acetone solution was used to extract phytochrome from 0.2 g of liquid nitrogen ground plant leaf material. For extracting the phytochrome, leaf material in acetone solution was ultrasonically crushed for 20 min in ice bath, and then filtered with 0.45 μm filtering membrane. The filtrate was used to measure phytochrome using high performance liquid chromatography (HPLC). The column used for HPLC was Waters Nova-Pak-C18 (3.9 × 150 mm, 4 μm). Analyzing condition was set as follows: 30 °C, 0.5 ml/min, 5 min for equilibrium, 90% acetone solution (V/V) as Mobile Phase A and 80% acetonitrile water (V/V) solution as Mobile Phase B.

3
Vol:. (1234567890) Soil physicochemical parameters Soil samples were sieved and sent to Nanjing Institute of Geography & Limnology, Chinese Academy of Science for measuring pH, organic matter, total nitrogen, ammonia nitrogen, nitrate nitrogen, available potassium and available phosphorate. Soil pH was measured when the soil: water ratio was 1:5. The concentration of mineral nitrogen in each sample was determined using a continuous flow automated analyzer (Futura Continuous Flow Analytical System, Alliance Instruments, France). Soil samples were extracted for mineral nitrogen (NH 4 + -N and NO 3 − -N) extraction using 2 M KCl and measured with a continuous flow automated analyzer. Soil potential nitrification rate (PNF) was determined through laboratory soil incubation experiment. Soil samples with 1% (NH 4 ) 2 SO 4 were incubated at 25 °C for 7 days . Nitrate-N was determined before and after incubation to calculate the PNF.

DNA extraction, sequencing and qPCR
Soil samples were stored at −80 °C pending DNA extraction. Soil 16S rRNA was extracted from 0.25 g of each soil sample using a MoBio Powersoil DNA Isolation Kit (MoBio Laboratories, United States) according to the manufacturer's protocol. The V3-V4 region of 16S rDNA was amplified using primer pair 341F (5'-CCT ACG GGNGGC WGC AG-3′) and 805R (5'-GAC TAC HVGGG TAT CTA ATC C-3′) with 12 bp barcode sequences. PCR amplification was performed on BioRad S1000 (Bio-Rad Laboratory, CA, USA). The PCR reaction system included 25 μl 2x Premix Taq (Takara Biotechnology, Dalian Co. Ltd., China), 1 μl each primer (10 μM) and 3 μl DNA (20 ng/μl) template in a volume of 50 μl. The PCR progress was set as follows: 5 min at 94 °C for initialization; 30 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 30 s extension at 72 °C; followed by 10 min final elongation at 72 °C.The quality and concentration of extracted DNA were determined using gel electrophoresis (1% agarose) and a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, United States). Mixed PCR products was purified with E.Z.N.A. Gel Extraction Kit (Omega, USA).
Sequencing libraries were generated using NEB-Next® Ultra™ II DNA Library Prep Kit for Illu-mina® (New England Biolabs, MA, USA) following manufacturer's instructions. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Fisher Scientific, MA, USA). At last, the library was sequenced on an Illumina No-va6000 platform and 250 bp paired-end reads were generated (Guangdong Magigene Biotechnology Co.,Ltd. Guangzhou, China).
Real-time qPCR was performed to analyze abundance of AOA and AOB amoA gene. The AOB amoA gene was amplified using primer pair amoA-1F (5′-GGGG TTT CTA CTG GTG GT-3′), and amoA-2R (CCCC TCK GSA AAG CCT TCTTC) (Rotthauwe et al. 1997), and the AOA amoA gene was amplified using primer pair arch-amoA-23F (ATG GTC TGG CTW AGACG) and arch-amoA-616R (GCC ATC CAT CTG TAT GTCCA) (Sanhan and Muyzer 2008). Real time PCR was performed in a volume of 10 μl mixture with triplicates. The qPCR mixture contained 2.5 ng purified DNA template, 150 nM of each primer, 5 μl 2X SYBR® Premix Ex TaqTM (Takara Japan). Two steps method was used for amoA gene quantifying, with desaturate at 95 °C for 30 ss, then 5 s at 95 °C and 30 s at 64 °C through 40 cycles. After amplification, melting curves were performed (95 °C for 15 s then temperature gradient from 55 °C to 95 °C in 81 steps) to check product size and homogeneity. Three technical replicates were performed for each sample. Sequencing data are available in the NCBI Sequence Read Archive database, following the Biosample accession ID of SAMN17278885 to SAMN17278902.

Data analyses and statistical analyses
Fastq format, library spliced raw data was obtained from Guangdong Magigene Biotechnology Co.,Ltd. for further analysis. Raw data was processed on the Galaxy pipeline developed by the Institute of Environmental Genomics, University of Oklahoma (http:// zhoul ab5. rccc. ou. edu: 8080/ root) to generate OTU table and representive sequences, as described in our previous studies (Gu et al. 2019;Meng et al. 2019). Briefly, the left and right reads were assembled with 10-200 bp overlapping using Flash (Version 1.0) (Mago et al. 2011). Low quality sequences (QC score < 20 and length < 250 bp) were trimmed using Btrim (Version 1.0) (Yong 2011), and sequences containing 'N' were also removed and only sequences with 400-440 bp in length was kept. Finally, chimeras were removed and sequences with 97% identity were assigned to the same operational taxonomic unit (OTU) using UPARSE (version v7.01001_ i86linux64) (Edgar 2013). Singletons that had no similar sequences were removed. Sequence number for each sample ranged from 33,931 to 45,860, therefore, we rarefied all samples to 33,931 by randomly choosing sequences. All downstream analyses were carried out using the rarefied OTU table.
Bacterial community diversity indexes calculation and beta diversity analyses were performed using R v4.1.1 statistical platform with 'vegan' v2.5-7 package (https:// github. com/ vegan devs/ vegan). LDA effect size (LEfSe) analysis was carried out to determine significantly differenced taxa between treatments. One-way analysis of variance (ANOVA) was used to analyze the effects of the hexaconazole on biochemical parameters and gene abundance using Minitab 16.0 statistical software. Significant differences (P < 0.05) between treatments were analyzed with LSD's multiple range tests.

Soil properties and plant growth
The results showed that soil properties were significantly affected by fumigation treatment (Table 1). Soil pH increased significantly (p < 0.05) after fumigations, whereas fumigation treatment did not lead to significant change in the contents of total nitrogen and organic matter. The contents of ammonia nitrogen, availability potassium and phosphorate increased significantly, whereas the nitrate nitrogen decreased significantly after fumigation treatment. The soil properties, except for availability potassium, did not show significant difference between the two fumigation treatments. The increase in plant available nutrients by fumigation treatment would promote the plant growth (Table 2). Compared to the control soil, potential nitrification rate decreased 4.82 and 9.05 times in CP and DZ treated soils, respectively.
The plant height, length and surface area of the longest leaf, and number of leaves increased significantly in fumigation treated fields ( Table 2). As shown in Fig. 1, the chlorophyll a had the highest contents, and its contents increased significantly following both two fumigation treatments, and no significant difference between the two fumigation treatments was observed. The violaxanthin was Table 1 Soil chemical parameters and potential nitrification rates after fumigation with chloropicrin (CP), dazomet (DZ), and untreated control (CK) Results are means and standard deviation of 6 replicates. Different letters following the results indicate the differences are significant at p < 0.05 level. PNR Potential nitrification rate

Microbial community diversity and composition
Diversity and composition of soil bacterial communities were investigated by using Illumina sequencing of the 16S rRNA gene amplicons. There were 721,855 high-quality sequences that could be assigned to 3692 bacterial OTUs with 97% sequence identity. We found that fumigation with continuous cropping soil resulted in soil bacterial diversity decreases and significant changes in some keystone taxa abundances, whereas the difference between two type of fumigants was not significant. NMDS analysis and PCoA analysis showed CP and DZ samples clearly separated with CK, suggesting they had different bacterial community structure (Fig. 2a-b). Dissimilarity analysis by ADNOIS further confirmed that the differences between treatments were significant ( Fig. 2c-d). However, there was no significant difference in bacterial community structure between CP and DZ. The bacterial community structure differed obviously at the phylum level (Fig. 3). Actinobacteria, Proteobacteria, Firmicutes and Candidatus Saccharibacteria were relatively more abundant in DZ and CP group compared with the CK group. The relative abundance of Gemmatimonadetes and Verrucomicrobia was significantly higher in CP group than others, while Bacteroidetes was significantly higher in DZ group. The relative abundance of Acidobacteria, Chloroflexi, Nitrospirae and Armatimonadetes was significantly higher in untreated CK group, compared to the fumigation treated DZ and CP groups. We carried out the LefSe analyses to reveal significantly changed bacterial taxa after fumigation treatments (Fig. 3). The results showed that CP treatment enriched Micromonospora, Saccharibacteria and Rhizomicrobium, and DZ treatment enriched Achromobacter, Ralstonia, Micromonospora, Stenotrophomonas and Pseudomonas.
Abundance of nitrifiers and amoA gene Typical nitrifiers were selected from the OTU table and their relative abundance were analyzed (Fig. 4). The results showed that the main nitrifiers, including Nitrospira and Nitrospirillum decreased significantly in fumigation treated rhizosphere. Although Nitrolancea showed different trend with other nitrifiers, its relative abundance was extremely low. Furthermore, the total relative abundance of selected nitrifiers decreased significantly in fumigation treatments (33% of the CK in DZ, and 28% of the CK in CP).
To obtain the absolute abundance of nitrifiers in soils, we carried out qPCR analysis for bacterial (AOB) and archaeal (AOA) amoA genes. The amplification efficiency was 98.45% and 96.48% for AOA and AOB amoA gene, respectively. AOA amoA copy numbers (ranging from 6.41 × 10 8 to 9.48 × 10 8 copies per g soil) were higher than the AOB in soil. However, the AOA amoA copy number did not show significant difference among treatments, whereas, the AOB amoA copy number decreased significantly in fumigation treated field.
Link between nitrifiers, soil nitrogen, chlorophylls and plant growth Pearson correlation analyses (Fig. 5) showed that all parameters could be assigned to two groups with high positive correlations, and two groups showed significant negative correlation with each other. The first group consist of bacterial community diversity indices, relative abundance of nitrifiers, AOB amoA gene abundance and soil PNF. The second group consist of leaf growth characters, part of soil properties and content of phytochromes and chlorophylls. To be specific, AOB amoA gene abundance and soil nitrate N content showed significant (p < 0.05) positive correlation with soil PNF, whereas, ammonia N showed significant (p < 0.05) negative correlation with PNF. Soil PNF correlated negatively with plant leaf surface area and Chlorophyll a content, whereas, soil ammonia content correlated positively with plant leaf surface area and Chlorophyll a content. Soil PNF correlated significantly and positively with abundance of Nitrospira, but the correlation between PNF and Nitrosospira was insignificant. In addition, the bacterial community diversity (Shannon and Chao1 indices) showed significant negative correlation with plant growth (leaf surface area). AOA amoA gene showed few correlations with all parameters.
The PLSPM model (Fig. 6) had a goodness of fit of 0.709, indicating the model was reliable. The PLSPM model showed that AOB had strong and significant effect on both soil PNF and nitrogen content, whereas, the direct path from PNF to plant chlorophyll content was relative weak than the effect from PNF to nitrogen. Chlorophyll content could strongly affect plant growth.

Discussion
Continuous-cropping obstacle has significantly negative impact on soil productivity (Ashworth et al. 2020). Our study indicated that fumigation treatment could alleviate the problems caused by continuous cropping, as the plant growth in fumigation (Dazomet and Chloropicrin) treated field was obviously promoted. Occurrence and alleviation of continuous-cropping obstacle were closely related to soil microorganisms (Tan et al. 2021). Therefore, the bacterial community played a crucial role in alleviation of the continuous cropping obstacle by fumigation treatment.
Soil property change is responsible for the plant growth promotion following fumigation treatment Considering that the plant height, leaf length, and leaf surface area were significantly increased following fumigation treatment, the plant growth was obviously promoted by fumigation treatments. The chlorophyll a synthesis enhancement in fumigation treated fields further confirmed that fumigation could effectively promote crop production. The change in plant phytochrome content was a result of the significant reduction in soil nitrogen nutrients, as the nitrogen was the important element for plant pigment. Phytochrome content was essential for crop yield and physiology, because the phytochrome determined the rate of photosynthesis (Boccalandro et al. 2003). The phytochrome also regulated the immunity of crop (Moreno and Ballare 2014). In this study, we used HPLC to investigated phytochrome contents in crop leaves. (Wang et al. 2014) also detected the increase of leaf chlorophyll content in fields treated by fumigants and PGPR (plant growth promoting rhizobacteria). Increasing of phytochrome would lead to strong photosynthetic rate, and therefore, was responsible for the plant growth promotion. Chlorophyll is the material carrier for plant photosynthesis, and its content directly affects photosynthesis, and the growth of crops. The content of chlorophyll pigment increased with the increase of nitrogen application, and nitrogen application could effectively prolong the functional period of leaves. Generally, the higher the chlorophyll content of plant leaves, the stronger the photosynthetic performance of the leaves and the higher the accumulation and metabolism of carbohydrate in the plant (Xue et al. 2013).
In this study, fumigation led to significant change in soil available phosphate and potassium, ammonia and nitrate nitrogen, which should be responsible for the plant growth promotion (Huang et al. 2020). It is well recognized that soil properties were the major factors affecting plant growth (Strom et al. 2020). For example, potassium deficiency reduced plant growth and transpirational water flow (Coffey et al. 2018). In particular, crop production largely depended on nitrogen availability (Hartemink and Bourke 2000). Longterm continuous cropping had been proven to cause low efficiency of nitrogen fertilizers (Sheng et al. 2013). In continuous cropping field nitrogen mainly existed as nitrate format, which are available to crops. However, the nitrate nitrogen tends to escape from soil through denitrification and leaching. In contrast, the ammonia N, that with positive charge, often has lower mobility than nitrate N in field, was not leached from the surface soil (Shaw 2009), and could offer processive nitrogen for crops. Soil total nitrogen did not show significant difference between treatments. However, the NH 4 + -N in CP and DZ-fumigation treated soil increased significantly, compared to the unfumigated soils. About 1.75 and 1.63 times higher with NH 4 + -N content in CP and DZ-fumigated soil than the untreated control, respectively. On the contrary, the soil NO 3 − -N content in all fumigant treatments was significantly lower than the untreated control (Fumigated soils exhibited a substantial decrease in NO 3 − -N). Thus, the difference in nitrogen speciation should be responsible for the difference in plant growth. To be specific, fumigation treatment increased soil ammonia nitrogen, but decreased soil nitrate nitrogen. It is not to say that nitrate nitrogen is less efficient to crops than the ammonia nitrogen, but instead, upland crops benefitted from the ammonia rich environment (Thorup-Kristensen 2001). However, the ammonia nitrogen could continuously oxidate to nitrate nitrogen, and supply sufficient nitrate N to crops, while the nitrate nitrogen is easier to loss from soil than ammonia nitrogen (Posmanik et al. 2014). In consequence, high nitrate N often leads to low efficiency of nitrogen nutrients.

Bacterial community was strongly affected by fumigation treatment
Our study showed that the fumigation treatments led to significant changes in the diversity and composition of the bacterial community. This is because the fumigation chemicals are toxic, microorganisms (e.g. Nitrospira and Nitrospirillum genus) that lack special strategies to live in extreme conditions would be suppressed by fumigation treatments (Zhang et al. 2019b). The sharp reduction in bacterial community richness (Chao1 and Observed OTU number) confirmed the toxicity of fumigants to some bacterial taxa. Other studies also reported the fumigant caused a significant shift in the predominant soil microbial populations (Huang et al. 2020;Liu et al. 2015). It is reported that DZ fumigation significantly reduced the soil microbial biomass, in which the total number of bacteria was reduced by 50%, and the effects lasted for 15 days (Parthipan et al. 1995). It was also indicated that the numbers of fungi, bacteria and actinomycetes were reduced by 58.8%, 15.3%, and 8.5%, respectively, after treatment with DZ in the continuous apple cropping soil (Wang et al. 2014). On the contrary, the number of aerobic gram-negative bacteria in the soil increased by 10 times after CP fumigation for 10 days. The biomass increased significantly after CP fumigation treatment, especially Pseudomonas which accounted for more than 70% of aerobic bacteria, but less than 10% in unfumigated soil (E. and Ridge 1976), and Pseudomonas is the main microorganism that degrades chloropicrin (Castro et al. 1983).
Proteobacteria, Firmicutes, Acidobacteria, Bacteroidetes, and Actinobacteria were the most abundant bacterial phyla across all samples, confirming what was already observed in agricultural soils (Mbuthia et al. 2015). The relative abundance of Proteobacteria and Actinobacteria was increased following fumigation treatment, which was consistent with several other studies (Bowles et al. 2014). It is because both the phyla Proteobacteria and Actinobacteria consist with a large group of microbe taxa that possess various strategies to survive in stressed conditions (e.g. heavy metal stress, drought stress or salt stress) (Li et al. 2017b). Proteobacteria and Actinobacteria are the most common phyla in soil environment, and play predominant roles in the soil nutrient (e.g. carbon, nitrogen and sulfur) cycling (Rong et al. 2021). Under the stress of fumigant, the internal balance of indigenous microbial community was changed, and different dominant microbial taxa appeared. According to the most abundant genera identified in each sample, the abundance of some initially dominant genera decreased or even disappeared after fumigation with DZ or CP. These genera may be susceptible to DZ or CP -fumigation. Following this, new predominant genera emerged, which are likely more tolerant to DZ or CP-fumigation. The significant change in soil physi-chemical properties (i.e. available P and K, and particularly, the speciation of Nitrogen) following fumigation treatment was due to the significant change in soil microbial community diversity, richness and composition. For example, the change in soil nitrogen speciation was a result of significant change in soil nitrifiers. Once the soil nitrifiers were inhibited by fumigants, soil nitrification potential rate reduced and therefore, the contents of soil ammonia and nitrate nitrogen would be significant changed. Soil functional microorganisms, especially bacteria, play vital roles in plant nutrient cycling, and therefore, in soil health and crop growth promotion (Crouzet et al. 2016).
Fumigants alleviated continuous cropping obstacles through their strong effects on soil nitrifiers and nitrification Drastic change in the forms of inorganic nitrogen was observed following fumigation treatment, therefore, attentions were paid to the nitrification process and microorganisms and genes driving the process. Nitrification process consists of two steps: one is aerobic ammonia oxidation of NH 4 + to NO 2 − , and the other is nitrite oxidation of NO 2 − to NO 3 − (Ji et al. 2012). Results in this study were consistent with substantial number of studies showing that soil fumigation slowed down the microbial nitrification and led to the accumulation of ammonium nitrogen (Fang et al. 2019), therefore, one can hypothesize that soil nitrification process played a key role in alleviating continuous cropping limitation induced by fumigation. Typical nitrification conversion of NH 3 to NO 2 − is mainly dominated by AOA and AOB (He et al. 2007). Our results showed that AOA amoA gene abundance was not significant increased with the application of fumigation, and the AOA amoA gene abundance did not show significant correlation with the potential nitrification rate. The results suggested that the AOA was not responsible for the nitrogen speciation and nitrification change in fumigation treated field. This is due to the fields being nitrogen-rich agricultural land, and the high ammonia concentrations in the soil might have provided an unfavorable condition for the AOA populations. In contrast to the AOA, the AOB was responsible for the significant change in nitrogen speciation and nitrification rate following fumigation treatment. The AOB amoA gene abundance decreased 15%-18% in fumigation treatment field compared to the control (Fig. 4). It also well explained that there was significant positive correlation between AOB amoA gene and nitrification rate. It is because the AOB populations can adapt and flourish under conditions of high nutrient availability (Shahid and Prosser 2006). This is consistent with many findings that nitrification was driven by bacteria but not archaea in nitrogen-rich soils (Di et al. 2009).
In the case that AOB amoA gene abundance was sharply reduced by fumigation treatment, one could infer that the AOB might be sensitive to fumigation treatment. There was a distinct shift in AOB community composition among different fumigation treatments, with all AOB sequences affiliating to Nitrosospira or Nitropira-like species. A large number of investigations have confirmed that the representative of Nitrosporium is a major ammonia-oxidizing bacteria (AOB). For example, Wang et al. (2014) showed that fumigation with 1,3-propene dichloride significantly reduced the abundance of Nitromonas, while fumigation with mianlong increased the abundance of Nitrospira in the early stage. The results showed that there was hardly any difference between the two fumigation treatments, indicating that the inhibition effects may not be type specific. Also, many studies showed the nitrification was sensitive to fumigation treatment, regardless the type of fumigants, soil type, plant species or experimental scale (Yan et al. 2015), the inherent mechanisms underlying the fumigation in alleviating continuous cropping barriers would be a general relevance. According to the report that ammonia concentration contributes to the definition of distinct ecological niches of AOA and AOB in soil (Verhamme et al. 2011). The PLSPM model suggested the important role of AOB in regulating the nitrogen nutrient cycling, plant chlorophylls synthesis, and plant growth, which is in agreement with our hypothesis. Therefore, one of the key mechanisms that fumigation alleviated continuous cropping obstacles could be closely related to soil nitrifiers, nitrification process and nitrogen nutrients.

Conclusions
In summary, this field-scale study showed fumigation had strong effects on soil microbial community and soil properties, particularly the nitrogen speciation. Microbial populations, e.g. Nitrospira and Nitrospirillum, that were sensitive to fumigants were suppressed. Soil nitrifiers, such as AOB, were drastically inhibited by fumigation treatment. The inhibition of AOB led to lower PNF, resulting in high efficiency of nitrogen nutrients to crops. High photosynthesis, as a result of increase of leaf chlorophyll content was caused by high nitrogen efficiency, would promote crop growth to alleviate continuous cropping obstacles. Future work at laboratory scale using pure culture and pot cultivation is needed, and functional analysis using omic tools and isotope technology will offer more solid, precise and deep understanding on the active mechanisms of fumigants.