Nitrogen fixation by Paenibacillus polymyxa WLY78 is responsible for cucumber growth promotion

To study nitrogen contribution to cucumber derived from nitrogen fixation of Paenibacillus polymyxa WLY78. The nif gene cluster deletion mutant (ΔnifB-V) of Paenibacillus polymyxa WLY78 was constructed by a homologous recombination method. The effects of plant-growth promotion were investigated by greenhouse experiments. The nitrogen fixation contribution was estimated by 15N isotope dilution method (also being called the 15N natural abundance technique). Deletion of nif gene cluster of P. polymyxa WLY78 resulted in complete loss of nitrogenase activity. Greenhouse experiments showed that inoculation with P. polymyxa WLY78 could significantly enhance the lengths and dry weights of cucumber roots and shoots, but inoculation with ΔnifB-V mutant could not. 15N isotope dilution experiments showed that cucumber plants derive 25.93% nitrogen from nitrogen fixation performed by P. polymyxa WLY78, but the ΔnifB-V mutant nearly could not provide nitrogen for plant growth. This present study demonstrated that nitrogen fixation performed by P. polymyxa WLY78 contributes to plant growth.


Introduction
Nitrogen is one of the most important nutrients in plant growth, but plants can not directly use nitrogen in the atmosphere. Biological nitrogen fixation (BNF) is a process in which nitrogen-fixing microorganisms reduce nitrogen in the air to ammonia through nitrogenases. Biological nitrogen fixation is an important part of the natural nitrogen cycle (Dart 1986) and plays an important role in the sustainable development of agriculture (Raymond et al. 2004) . Nitrogen-fixing microorganisms include symbiotic nitrogen-fixing bacteria, free-living nitrogen-fixing bacteria and associative nitrogen-fixing bacteria (Xu et al. 2017). Associative nitrogen-fixing bacteria can colonize root surface cells, invade plant roots, and form close contact with plants, thereby promote plant growth (Baldani et al. 1997). Associative nitrogenfixing bacteria promote the absorption of nitrogen by non-legume plants (Geddes et al. 2015). Biological

Abstract
Aims To study nitrogen contribution to cucumber derived from nitrogen fixation of Paenibacillus polymyxa WLY78. Methods The nif gene cluster deletion mutant (ΔnifB-V) of Paenibacillus polymyxa WLY78 was constructed by a homologous recombination method. The effects of plant-growth promotion were investigated by greenhouse experiments. The nitrogen fixation contribution was estimated by 15 N isotope dilution method (also being called the 15 N natural abundance technique). Results Deletion of nif gene cluster of P. polymyxa WLY78 resulted in complete loss of nitrogenase activity. Greenhouse experiments showed that inoculation with P. polymyxa WLY78 could significantly enhance the lengths and dry weights of cucumber roots and shoots, but inoculation with ΔnifB-V mutant could not. 15 N isotope dilution experiments showed that cucumber plants derive 25.93% nitrogen from nitrogen fixation performed by P. polymyxa WLY78, nitrogen fixation not only reduce the use of nitrogen fertilizer, but also improve soil fertility and the absorption of nutrients by crops (Farrar et al. 2014).
It has been reported that the associative nitrogenfixing bacteria play an important role in promoting growth of non-legumes by fixing nitrogen and producing phytohormone (Chalk 1991). In earlier research, 15 N isotope and N balance studies had shown that several sugarcane varieties obtain over 60% of their nitrogen (>150 kg N ha -1 year -1 ) from biological nitrogen fixation performed by diazotrophs (Boddey et al. 1995). Diazotrophic bacteria present in the mucilage of aerial roots contributed 29-82% of the N nutrition of Sierra Mixe maize (Van Deynze et al. 2018). Inoculation with nitrogen-fixing Klebsiella pneumoniae 342 (Kp342) increased total N and N concentration in the wheat plant (Iniguez et al. 2004). Inoculation of the rhizobacteria including Azospirillum brasilense and Azospirillum lipoferum contributed up to 20-50% of the total nitrogen requirement of the oil palm seedlings through nitrogen fixation (Amir et al. 2003). Diazotrophic Paenibacillus beijingensis BJ-18 provided nitrogen for wheat, maize and cucumber plants and promoted plant growth, nitrogen uptake and metabolism . A recombinant nitrogen-fixing Pseudomonas protegens Pf-5 X940 was constructed by introducing the nif genes of Pseudomonas stutzeri A1501 via the X940 cosmid to the beneficial rhizobacterium Pseudomonas protegens Pf-5, and inoculation of Arabidopsis, alfalfa, tall fescue and maize with Pf-5 X940 increased the ammonium concentration in soil and plant productivity under nitrogen-deficient conditions (Fox et al. 2016;Setten et al. 2013). Inoculation with Azospirillum brasilense Ab-V5 cells enriched with exopolysaccharides and polyhydroxybutyrate enhances the productivity of maize under low N fertilizer input (Oliveira et al. 2017) Paenibacillus polymyxa WLY78 is a nitrogen-fixing bacterium containing a compact nif gene cluster consisting of 9 genes (nifBHDKENXhesAnifV) (Wang et al. 2013;Xie et al. 2014). In addition to nitrogen fixation, this bacterium has the ability of phosphate solubilization and indole-3-acetic acid (IAA) production (Xie et al. 2016). Also, this bacterium can produce fusaricidins that are a class of cyclic lipopeptide antibiotics to inhibit plant pathogenic fungi . These specific traits suggest that P. polymyxa WLY78 is a member of plant growth-promoting bacteria (PGPB) and has great potential as an inoculant in agriculture. However, the nitrogen contribution to plants derived from nitrogen fixation of P. polymyxa WLY78 is unclear. In this study, a nif gene cluster deletion mutant (ΔnifB-V) of P. polymyxa WLY78 was constructed. Comparisons of P. polymyxa wildtype and ΔnifB-V mutant in plant-growth promotion and nitrogen fixation contribution rate were investigated. Our study was to analyze the role of nif gene in N 2 fixation to stimulate plant growth.

Materials and methods
Bacterial strains, plasmids, media and growth conditions Bacterial strains and plasmids used in this study are summarized in Table 1. Paenibacillus polymyxa WLY78 was isolated from the rhizosphere of bamboo in Beijing (Wang et al. 2013). P. polymyxa  (Wang et al. 2013) was used for assay of nitrogenase activity (Wang et al. 2013).
Escherichia coli JM109 was used as routine cloning. Thermo-sensitive vector pRN5101 (Zhang et al. 2013) was used for gene disruption in P. polymyxa WLY78. When appropriate, antibiotics were added in the following concentrations: 100 μg/ml ampicillin and 5 μg/ml erythromycin for maintenance of plasmids.

Construction of ΔnifB-V mutant of P. polymyxa WLY78
The nitrogen-fixing gene cluster deletion mutant ΔnifB-V of P. polymyxa WLY78 was constructed by a homologous recombination method. For doing this, the upstream fragment and downstream fragment flanking nif gene cluster were amplified by PCR using Phanta®Max Super-Fidelity DNA Polymerase (Vazyme Biotech Co., Ltd., Nanjing, China), respectively. Primer 1 (5′ CGG CCA CGA TGC GTC CGG CGT AGA GGA TCC GCG TGG TG GAT GTG GA CG 3′) and Primer 2 (5′ AAC GCT TTT TCG GTT ATC ATT CCT TCA CAT CTA TTT TCGTC 3′) were used to amplify a 950 bp-long sequence located upstream of nifB. Primer3 (5′ GAA GGA ATG ATA ACC GAA AA AGC GTT CCC GTC 3′) and Primer 4 (5′ GAC TGC GCA AAA GAC ATA ATC GAT AAG CTT CCT GAT AAG GCA G ACA AGG CTC 3′) were used to amplify a 1107 bp-long sequence located downstream of nifV. The two fragments were then fused with BamH I/Hind III digested pRN5101 vector using Gibson assembly master mix (New England Biolabs), generating the recombinant plasmid pRDnifB-V. Then, the recombinant plasmid pRDnifB-V was transformed into P. polymyxa WLY78 as described by Wang et al. (2018), and the marker-free deletion mutant (the double-crossover transformant) ΔnifB-V mutant was selected from the initial erythromycin resistance (Em r ) transformants after several rounds of nonselective growth at 39°C and confirmed by PCR using the primer 5 (5′ GCA TAA ATT GTA CAC GTT GA 3′) and primer 6 (5′ AGG CTC ATA AAC ACC GTA TC 3′).
Growth and nitrogenase activity of the wild type and mutant strains To measure growth, P. polymyxa WLY78 and ΔnifB-V mutant were grown in 20 mL of LB media in 50 mL flasks shaken at 200 rpm at 30°C to logarithmic growth phase. The cultures were collected by centrifugation, washed three times with sterilized water and then resuspended in sufficient nitrogen medium (nitrogen-limited medium supplemented with 100 mM NH 4 Cl) to a start OD 600 of 0.02. Every 2 h, the growth of P. polymyxa WLY78 and ΔnifB-V mutant strains were determined by absorbance at 600 nm.
Acetylene reduction assays were performed as described previously to measure nitrogenase activity (Wang et al. 2013). P. polymyxa WLY78 and ΔnifB-V mutant strains were grown in 50 mL of LB medium overnight. The cultures were collected by centrifugation, washed three times with sterilized water, and then resuspended in a nitrogen-limited medium to a final OD 600 of 0.4. Then, 4 mL of the culture was transferred to a 26-mL test tube and the test tube was sealed with a rubber stopper. The headspace in the tube was then evacuated and replaced with argon gas. Then, approximately 2.2 mL of C 2 H 2 (10% of the headspace volume) was injected into the test tubes. After incubating the cultures at 30 °C for 8 h, a 100 μL gas sample was taken out and injected into gas chromatography to quantify ethylene (C 2 H 4 ) production. The nitrogenase activity was expressed in nmol C 2 H 4 /mg protein/h.

Preparation of soil, seeds and bacterial suspension
The soil (0-20 cm deep topsoil) was taken from the Shangzhuang Experimental Station of China Agricultural University. The soil was low nitrogen (7.8 mg kg −1 ) sandy soil. After the soil was air-dried and crushed, the debris were removed with a 2 mm sieve to reduce heterogeneity, and then packed into plastic pots with a diameter of 20 cm and a height of 14 cm.
Each pot was filled with 2 kg of soil to grow cucumbers. No trace elements were applied during plant growth.
Cucumber seeds ("Zhongnong 8" of Beijing Shengfeng Garden Agricultural Technology Co., Ltd.) were first disinfected with 10% sodium hypochlorite for 10 minutes, then were rinsed with sterile water three times, and finally were distributed in a sterile petri dish with damp filter paper at room temperature (25°C) for 3-5 days in the dark.
The bacterial suspension of P. polymyxa WLY78 and ΔnifB-V used in inoculation was prepared as follows. P. polymyxa WLY78 and ΔnifB-V were inoculated into LB liquid medium, cultured at 30°C and 180 rpm to logarithmic growth phase, and then harvested by centrifugation and finally suspended with physiological saline (0.89% w/v NaCl in deionized water). The cell concentration was set to 10 8 cells mL -1 .

Greenhouse pot experiment
The research was conducted in the greenhouse of China Agricultural University using greenhouse potting. The experimental design was arranged by random factors, with three inoculation treatments (two bacterial inoculation and one mock inoculation) and two nitrogen level treatments. Each treatment was repeated three times, for a total of 18 pots of cucumber plants. Nitrogen level treatment included high nitrogen and low nitrogen levels. Nitrogen fertilizer was applied in the form of 15 N labeled (NH 4 ) 2 SO 4 (10.16% 15 N atom, Shanghai Research Institute of Chemical Industry, China). The high nitrogen level was 250 mg N kg -1 soil, and the low nitrogen level was 83 mg N kg -1 soil. Nitrogen fertilizer was applied in three times, one-third each time, and the first time was applied as a base fertilizer, followed by 7 and 14 days after transplantation.
The inoculation included three treatments: inoculation of P. polymyxa WLY78 (WT), ΔnifB-V, and equal amount of physiological saline (as a control). The germinated cucumber seeds with robust and consistent growth were picked and immersed in the bacterial suspension for 20 minutes. The seeds were immersed in physiological saline for 20 minutes as a control group, and then transplanted into plastic pots. Four seeds were planted in each pot, and three repetitions were set for each treatment. In the first and second weeks after planting, the 80 mL of the bacterial suspensions of P. polymyxa WLY78 and ΔnifB-V were applied to pot containing inoculated cucumbers, respectively, and 80 mL of physiological saline was applied to pot containing non-inoculated cucumbers. Pots were placed in the greenhouse under optimum conditions (15 h light/9 h dark cycle, 25-30/15-20°C day/night temperature and 40% day/60% night humidity). The seedlings were regularly watered every 3 days until the relative humidity of the soil reached 40%.

Plant sample collection
On 30th day of cucumber planting, the plants were collected by destructive sampling. The whole seedling was first uprooted, and rinsed with deionized water to remove the soil attached to the root system, then the root and shoot samples were separated, and the length of the root and shoot were measured. The root and shoot samples were killed in an oven at 105°C for 30 minutes, and then dried at 65°C until constant weight for dry weight analysis. Then the dried samples were immediately frozen in liquid nitrogen for subsequent analysis.

Quantification of biological nitrogen fixation (BNF) contribution
The plants un-inoculated, inoculated with P. polymyxa WLY78 (WT) and ΔnifB-V mutant were collected on 30th day of cucumber planting as described above. The dried plant samples were ground, sieved with a 1 mm sieve and placed in a bag. The δ 15 N values were assayed by using DELTA V Advantage isotope ratio mass spectrometer (Thermo Fisher Scientific, Inc., United States) in Institute of Botany, the Chinese Academy of Science, China.
The 15 N natural abundance technique (also being called the 15 N isotope dilution method) was used to quantitatively determine the contribution of BNF (Boddey et al. 2001;Bremer and van Kessel 1990). This method is based on difference of the relative abundances of the stable isotope 15 N in the atmosphere and soil, with 15 N abundance in the soil being higher than in the air. Atmospheric N 2 shows a natural abundance of 0.3663 atom% 15 N. Natural 15 N abundance was calculated as follows: δ 15 N (‰) = 1000 × (atom% 15 N sample -0.3663)/0.3663. The percentage of nitrogen derived from BNF (%Ndfa) was calculated using the following formula.
In the formula, "δ 15 N" is stable nitrogen isotopes, "ref" is the value from non-N-fixing reference plants, "fixing plant" is plant inoculated with nitrogen-fixing bacteria, and "B" is the N abundance in the air, assumed to be 0.0‰.

Statistical analysis
Graphs were prepared using GraphPad Prism software v. 8.0 (GraphPad Software Inc., San Diego, CA, USA). Statistical analysis was performed using SPSS software version 20 (SPSS Inc., Chicago, IL, United States). Two-way analysis of variance (ANOVA) was employed to check the significant differences between treatments. Means of different treatments were compared using the least significant difference (LSD) at the 0.05 or 0.01 level of probability.

Results
The nif gene cluster deletion mutant (ΔnifB-V) of P. polymyxa WLY78 resulted in complete loss of nitrogenase activity P. polymyxa WLY78 contains a compact nif gene cluster consisting of 9 genes (nifB nifH nifD nifK nifE nifN nifX hesA nifV) located within a 10. 5 kb region. The nif gene cluster deletion mutant (ΔnifB-V) was constructed by recombination as described in Fig. 1a. The disruption of nif gene cluster was confirmed by PCR (Fig. 1b). The ΔnifB-V and the wild-type strains exhibited similar growth phenotypes on nitrogen-limited medium with ammonium as nitrogen sources (Fig. 1c). However, the ΔnifB-V mutant did not have nitrogenase activity (Fig. 1d), indicating that the nif gene cluster is essentially required for nitrogen fixation under nitrogen limitation.

Effects of P. polymyxa WLY78 and ΔnifB-V mutant on the growth of cucumber
Cucumber samples were collected on the 30th day after plantation, and the length and dry weight of cucumber shoots and roots were measured to evaluate the effects of inoculation with P. polymyxa WLY78 and ΔnifB-V mutant on plant growth under low and high nitrogen conditions. The non-inoculated cucumber served as a control. Compared to the noninoculated control group, cucumber inoculated with P. polymyxa WLY78 under low nitrogen conditions showed increase of 68.18% in shoots dry weight, of 59.15% in root dry weight, of 37.61% in shoot length and of 38.52% in root length, but they showed a little increase in lengths and weights under high nitrogen conditions ( Fig. 2a-d). Compared with the non-inoculated control group, the cucumber inoculated with the ΔnifB-V mutant showed a little increase in the dry weight and length of the shoots and roots under both low and high nitrogen conditions (Fig. 2a-d). Fig. 3a, c is an experimental diagram of the greenhouse cultivation. The data indicated that the diazotrophic P. polymyxa WLY78 can effectively promote plant growth under low nitrogen conditions and disruption of nif genes encoding nitrogenase results in almost loss of the ability of promoting plant growth (Fig. 3b,  d).

Nitrogen fixation in cucumber provided by P. polymyxa
To estimate the contribution of BNF, the natural abundance 15 N measurement technique was used to analyze the inoculated and un-inoculated cucumber plants grown in soil containing 15 N-labeled (NH 4 ) 2 SO 4 as N fertilizer in greenhouse conditions. As shown in  containing both 14 N and 15 N which were absorbed from soil containing 15 N-labeled (NH 4 ) 2 SO 4 . If biological nitrogen fixation occurs, atmospheric 14 N will be converted to NH 4 + and thus δ 15 N value will become smaller. Therefore, the δ 15 N value of plants without nitrogen-fixing bacteria is higher than that of plants with nitrogen-fixing bacteria.
The percentage of nitrogen derived from BNF (%Ndfa) was estimated from the δ 15 N value as described in Materials and Methods. The percentage of nitrogen derived from BNF (%Ndfa) in the P. polymyxa WLY78 inoculated cucumber under low N conditions was calculated as follows: (7699.37 -5705.87) / 7699.37 × 100 =25.9%. According to this calculation, the plants inoculated with P. polymyxa WLY78 under high nitrogen condition obtained N from BNF is 2.8%. The nitrogen derived from BNF in the plants inoculated the ΔnifB-V strain under low N and high N conditions is 1.5% and 0.2%, respectively. These results indicate that the cucumber plant has incorporated the nitrogen provided by BNF of P. polymyxa WLY78 under low N conditions and BNF was inhibited by high concentration of available nitrogen in the environment. The ΔnifB-V strain-inoculated cucumbers only derived a very low amount of nitrogen from BNF, consistent with the ΔnifB-V strain had no nitrogenase. The data also indicate that P. polymyxa WLY78 can be used to provide nitrogen nutrition to plants and reduce the use of nitrogen fertilizers.
The results came from three biological replicates, the error represents SD, lowercase letters a and b indicate that there is a significant difference between the groups (P <0.05), while the same letter indicates that there is no significant difference.

Discussion
To estimate the contribution of BNF, the cucumbers inoculated with wild -type P. polymyxa WLY78 and with a ΔnifB-V mutant and the cucumbers without inoculation (reference control) were grown in soil containing 15 N-labeled (NH 4 ) 2 SO 4 as N fertilizer in greenhouse conditions. The contribution of BNF was determined by using the natural abundance 15 N measurement technique (Boddey et al. 2001;Bremer and van Kessel 1990 (Padda et al. 2017). Nitrogen fixation is a highly energy-consuming process in which reduction of 1 mol of N 2 requires 16 mol of ATP in vitro (Seefeldt et al. 2009;Hu and Ribbe 2016). Thus, nitrogen fixation is strictly regulated according to ammonium concentration. Most of the N-fixing bacteria fix nitrogen only in absence of ammonium or in presence of low concentration (1-5 mM) of ammonium. Thus, P. polymyxa WLY78 and the ΔnifB-V mutant showed the similar growth rates in the presence of 100 mM NH 4 + (Fig 1c), since nitrogen-fixing ability of P. polymyxa WLY78 is inhibited by high concentration of ammonium.
The contribution of nitrogen-fixing bacteria on plant growth via BNF has been reported in Klebsiella pneumoniae 342, Pseudomonas stutzeri A1501 and Acetobacter diazotrophicus PAl5 (Iniguez et al. 2004;Ke et al. 2019;Sevilla et al. 2001). A difference between our study with other's is that the ΔnifB-V mutant of P. polymyxa WLY78 has a deletion of a compact nif gene cluster comprising 9 genes (nifBHDKENXhesAnifV) and the other's mutant (such as the nifHmutant or nifKmutant or nifDmutant) has a deletion of a single nif gene.
The effects of P. polymyxa WLY78 and ΔnifB-V mutant on cucumber growth under different nitrogen concentrations were studied through greenhouse cultivation experiments. Compared to the un-inoculated control cucumber plants and the inoculated cucumber plants with the ΔnifB-V mutant, inoculated cucumbers with P. polymyxa WLY78 were significantly increased in the dry weights and lengths of cucumber shoots and roots under low nitrogen conditions. These results have revealed that the nitrogen fixation of P. polymyxa WLY78 plays an important role in promoting plant growth. Phosphate solubilization and IAA production of P. polymyxa WLY78 may exhibit a minor role in promoting plant growth. Similarly, inoculation with Pseudomonas stutzeri A1501 strain can increase the root and shoot dry weight of maize, but this effect is not found in the maize inoculated with nifHmutant (Ke et al. 2019). In N 2 -deficient conditions, Kallar grass inoculated with Azoarcus sp. BH72 grew better and accumulated more nitrogen than plants inoculated with the nifKmutant strain (Hurek et al. 2002.). Inoculation with K. pneumoniae 342 resulted in increased dry weight, chlorophyll content, total N, and N concentration of wheat in comparison with the uninoculated and nifHmutantinoculated controls (Iniguez et al. 2004).