Nitrogen fixation and transcriptome of a new diazotrophic Geomonas from paddy soils

ABSTRACT Nitrogen gas (N2) fixation driven by diazotrophs is a crucial process for supplying nitrogen to paddy soil ecosystems. The genus Geomonas has been considered to be an important potential diazotroph in paddy soils, but direct experimental evidence of the nitrogen-fixing ability of Geomonas in pure culture is still lacking. Hence, we aimed to demonstrate this nitrogen-fixing capability and shed light on how this process was regulated in response to ammonium (NH4+) in Geomonas. In this study, we determined that a key nitrogenase gene (nifH) was present in 50 isolates from paddy soils. Members of Geomonas contained the minimum nitrogen fixation gene cluster (nifBHDKEN) based on genomic analysis, implying Geomonas species had the potential to fix nitrogen. Acetylene reduction assay (ARA), 15N2 isotope labeling, and total nitrogen accumulation assays validated that Geomonas was, indeed, able to fix nitrogen in pure culture. Under nitrogen-fixing conditions, the cell morphology of Geomonas changed from short rod-shaped (with NH4+) to long rod-shaped and flagella became longer and thicker. The expression of genes correlated to nitrogen fixation in the Geomonas transcriptome was quantified in response to NH4+. Expression of genes associated with nitrogenase, flavin-based electron bifurcation complexes (such as the FixAB system), NH4+ uptake, and transformation (e.g., glutamine and glutamate synthetases) were significantly upregulated under nitrogen-fixing conditions, suggesting these mechanisms might be involved in N2 fixation in Geomonas. These results were verified by RT-qPCR. Taken together, our results demonstrate that Geomonas species possess the ability to fix N2 and expand our understanding on the ecological significance and potential applications of Geomonas in paddy soil ecosystems. IMPORTANCE The ability of Geomonas species to fix nitrogen gas (N2) is an important metabolic feature for its application as a plant growth-promoting rhizobacterium. This research is of great importance as it provides the first comprehensive direct experimental evidence of nitrogen fixation by the genus Geomonas in pure culture. We isolated a number of Geomonas strains from paddy soils and determined that nifH was present in these strains. This study demonstrated that these Geomonas species harbored genes encoding nitrogenase, as do Geobacter and Anaeromyxobacter in the same class of Deltaproteobacteria. We demonstrated N2-dependent growth of Geomonas and determined regulation of gene expression associated with nitrogen fixation. The research establishes and advances our understanding of nitrogen fixation in Geomonas.

(NH 4 + ) used by plants (1)(2)(3).Rice is the primary economic food crop for over half of the global population.To fulfill the food requirements for the growing global popula tion, chemical nitrogen fertilizers had to be applied to rice crops to achieve high yields (4).Nevertheless, approximately 56% of the applied nitrogen fertilizer is lost to land and water or converted into nitrous oxide through denitrification (5), which leads to various environmental issues including soil acidification, heightened greenhouse gas emis sions, and jeopardizing the sustainable development of agriculture and food security (6).BNF is friendly to the environment and nitrogen-fixing microorganisms mainly include symbiotic and free-living diazotrophs (7,8).Compared to symbiotic diazo trophs, free-living diazotrophs possess the advantage of lacking host-specificity and are extensively distributed in various environments (9,10).Recent studies have highlighted the significance of free-living diazotrophs in NH 4 + production in situ and their crucial role in nitrogen fixation in rice fields (11,12).Free-living diazotrophs are primarily composed of the phylum Proteobacteria (genus Azotobacter, Klebsiella, Pseudomonas, Geobacter, Anaeromyxobacter), as well as Cyanobacteria, Firmicutes (Clostridium and Bacillus), and Actinobacteria (2,13,14).These taxa have the ability to fix nitrogen and, thus, maintain the nitrogen requirements for plant growth when nitrogen sources become limited or absent.Currently, genome analysis has unveiled that a broad spectrum of bacteria harbor nitrogen fixation genes (nif), suggesting that the diversity of diazotrophs extends beyond what was previously recognized in terrestrial environments (15)(16)(17).Therefore, it is imperative to explore the potential significance of unknown diazotrophs and conduct further research to better comprehend their functional roles in supporting ecosystem health and productivity.
The genus Geomonas established by Xu et al. in 2019 displays a close genetic relationship with Geobacter, a member of the family Geobacteraceae of the class Deltaproteobacteria. Geomonas species have been shown to be strictly anaerobic and capable of iron reduction (16)(17)(18)(19)(20)(21).Until now, there have been 16 validly published species of Geomonas, primarily found in paddy soils (22).Subsequent research has inferred that Geomonas may function as a novel type of potential diazotroph similar to Geobacter in rice soil ecosystems (16,17,19).Previous studies indicated that the presence of nif gene transcripts from Deltaproteobacteria, particularly from the genus Geobacter and Anaeromyxobacter, was frequently detected and had a high relative abundance in paddy soils (12,(23)(24)(25)(26).Although the reports in this area have been limited, nitro gen-fixing activity has been demonstrated by culture-dependent methods in Geobacter (Geobacter metallireducens and Geobacter sulfurreducens) and Anaeromyxobacter spp.(2,13,(27)(28)(29)(30).As such, it is deduced that Geomonas may represent the dominant diazo trophs in the rice soil ecosystem.Consequently, it is crucial to explore the distribution of Geomonas in paddy soils and further study its nitrogen fixation properties at the culture-dependent level.
The primary objective of this study was to determine the nitrogen-fixing activity and gene regulatory changes in Geomonas under nitrogen-fixing conditions.We hypothesize that Geomonas is an important new diazotroph, playing a crucial role in fixing nitrogen in paddy soil.However, to date, there has been no comprehensive direct research on nitrogen fixation activity of Geomonas in pure culture.Therefore, demonstrating the diazotrophic activity in Geomonas is required for a comprehensive understanding of the microbial drivers of nitrogen cycling in soil environments.In this study, we con ducted genomic, transcriptomic, and culture-dependent analyses on Geomonas strains isolated from paddy soils to verify their diazotrophy.This research contributes to a better understanding of the genetic mechanisms of diazotrophy in Geomonas as well as providing a scientific foundation for exploring anaerobic microbial ecology in rice fields.

Identification of diazotrophic Geomonas from paddy soils
Fifty Geomonas strains were isolated from paddy soils by cultural methods and identified using 16S rRNA gene sequences.Seven of them have been identified as five novel species of genus Geomonas through polyphasic taxonomy and validly published (16,19).Based on 16S rRNA gene sequences analysis, other isolates were identified as six known species of the genus Geomonas, including Geomonas terrae, Geomonas ferrireducens, Geomonas paludis, Geomonas oryzae, Geomonas edaphica, and Geomonas bremensis (Fig. S1).
PCR results showed that all isolates displayed a nifH band with a fragment length of about 360 bp (Fig. S2), indicating that Geomonas potentially possessed the ability to fix nitrogen.To estimate the nitrogen-fixing ability, nitrogenase activity was measured in representative isolates by the acetylene reduction assay (ARA) method and showed that Geomonas displayed nitrogenase activity (Fig. S3).Among these isolates, a novel species, Geomonas nitrogeniifigens RF4, exhibited the highest nitrogenase activity and was used for further studies.

Genomics and enzymatic background of nitrogenase in Geomonas
The genome size of the genus Geomonas was ranged between 4.46-5.26Mb, and the number of protein-coding genes was 3,889-4,455.The genomic DNA G+C content was 60.3-62.6%(Table S1).The nifHDK nucleotide sequences retrieved from genomes in Geomonas showed the closest phylogenetic relationship to those in Geobacter and Anaeromyxobacter within the class Deltaproteobacteria (Fig. 1A).
Based on genome predictions, KEGG results showed that the genes nifHDKENX encoding functions associated with nitrogenase were found on sequenced genomes of Geomonas.nifHDK encoded the proteins of the nitrogenase complex, and one copy of nifDK and one to two copies of nifH were predicted in Geomonas (Fig. S4).Five genes adjacent to nifH (nifHDKENX) were conserved in the investigated Geomonas nif gene clusters (Fig. 1B).The nif gene cluster of Geomonas contained genes encoding regulatory proteins (draG, fdxN, and draT), nitrogenase complex structural genes (nifHDK), and biosynthesis of the FeMo cofactor (nifBENX).The nif gene cluster structure of Geomonas and Geobacter was different and even exhibited differences within the genus.Strains G. nitrogeniifigens RF4 and G. terrae Red111 T had a similar nif gene cluster that was different from G. paludis Red736 T .Moreover, the structure of the nif gene cluster of the genus Geomonas was more compact than that of other genera in class Deltaproteobacteria, namely, these genes were located closer to each other in Geomonas.
Residues in alignments of NifH sequences showed that [4Fe-4S] iron sulfur clusterligating cysteines (Cys97 and Cys132) and P-loop/MgATP binding motif were conserved in Geomonas as well as in other diazotrophs (Fig. 1C), suggesting that these pro teins might function as the dinitrogenase reductase.Similarly, nifDK sequences were conserved in these Geomonas species.The crucial residues of the FeMo cofactor binding site in NifD (Cys275 and His442) and in the P cluster in NifK (Cys70, Cys95, and Cys153) were conserved, as in other diazotrophs such as Anaeromyxobacter and Geobacter.

Nitrogen fixation and growth responses of Geomonas
G. nitrogeniifigens RF4 grew best with 20 mM acetate and 10 mM glucose simultaneously as electron donors in comparison to acetate as the sole electron donor (Fig. S5).Hence, we chose two electron donors to perform assays in the following studies.In the absence of NH 4 + addition, G. nitrogeniifigens RF4 grew well using N 2 as the sole nitrogen source, indicating RF4 was able to fix nitrogen.Moreover, no growth was observed with Ar in replacement of N 2 .However, through the visual observations presented in Fig. 2A, G. nitrogeniifigens RF4 exhibited better growth cultured with NH 4 + than without NH 4 + .The growth OD 600 of G. nitrogeniifigens RF4 was more than twofold higher with NH 4 + than with N 2 after 4 days (Fig. 2B).Morphological changes in cells can accompany different stresses such as high temperature (31) and oxidative stress (32).Compared to the non-nitrogen-fixing condition (with 1 mM NH 4 + ), we observed significant differences in G. nitrogeniifigens RF4 viability, optical density, growth rate, and cell morphology under nitrogen-fixing conditions (no NH 4 + added).For example, cells changed from being short rod-shaped (with NH 4 + ) to long rod-shaped under nitrogen-fixing conditions (Fig. 2CD).Flagella under nitrogen-fixing conditions became longer and thicker, increasing their ability to obtain energy for survival.
To further confirm nitrogen fixation ability, 15 N 2 isotope labeling was used.Isotope mass spectrometer detection showed that the ratio of 15 N/ 14 N approximated 50.0135% under the nitrogen-fixing conditions, but 15 N/ 14 N was only 0.366% in the presence of 1 mM NH 4 + (Fig. 3A).It was, indeed, confirmed that strain G. nitrogeniifigens RF4 was able to transform 15 N 2 to nutrients, thereby providing the necessary energy for its growth.To evaluate the nitrogen-fixing ability of G. nitrogeniifigens RF4, the total nitrogen concen tration (TNC) was measured.When using N 2 as the sole nitrogen source, we observed a significant 11.7-fold increase in TNC (increased to 6.4 mg L −1 ) after culturing for 4 days (Fig. 3B), indicating strain G. nitrogeniifigens RF4 was, indeed, able to fix N 2 and convert it into nutrient nitrogen.The nitrogen-fixing activity was estimated by ARA.In the absence of NH 4 + , ARA of G. nitrogeniifigens RF4 reached ~1.22 × 10 4 µmol C 2 H 4 g −1 protein h −1 (Fig. 3C).ARA was 4 µmol C 2 H 4 g −1 protein h −1 in the presence of 1 mM NH 4 + .The nitrogen-fixing activity of strain G. nitrogeniifigens RF4 decreased as the NH 4 + concentration in the medium increased and was completely inhibited by 0.8 mM NH 4 + (Fig. 3D).This was consistent with previous reports showing that NH 4 + inhibited nitrogenase activity of diazotrophs such as Geobacter (33) and Anaeromyxobacter (2).However, NH 4 + was not detected in the culture of G. nitrogeniifigens RF4 under nitrogen fixing conditions, indicating ammonia produced by nitrogen fixation process was almost completely used for the growth of Geomonas.

Transcriptomic analysis of Geomonas under nitrogen-fixing conditions
Transcriptomic analysis was performed to determine the gene expression differences between nitrogen-fixing and non-nitrogen-fixing conditions.The clean data obtained from the cDNA libraries constructed from each sample under nitrogen-fixing and non-nitrogen-fixing conditions reached at least 3.08 Gb, with Q30 > 95.25%.More than 97% of the total reads were mapped to genes of the reference strain (Table 1).Principal component analysis (PCA) showed that the same treatment clustered together, indicating that the sequence data were reasonable and could be used for further studies (Fig. S6).
Using soft DESeq2, we defined differential expression as a log 2 -fold change of |log 2 FC| ≥ 1.00 and P < 0.05.A total of 1,826 genes were significantly differentially expressed, including 950 downregulated genes and 876 upregulated genes under nitrogen-fixing conditions.Gene expression under nitrogen-fixing and non-nitrogen-fixing conditions exhibited significant differences (Fig. 4A).We grouped these differently expressed genes by metabolic function in the KEGG database: 1,247 annotated genes were obtained with 956 genes enriched into 167 pathways (Fig. S7).Among the six major classifications of KEGG, the number of genes involved in metabolism was the highest.Under the metabolism classification, carbohydrate, energy, cofactors, and vitamins and amino acid metabolism contained more genes.Under cellular process classification, cellular community and cell motility contained more genes, flagella assembly encoding genes were most up regulated under nitrogen-fixing conditions.These genes might play an important role in the nitrogen fixation process of strain G. nitrogeniifigens RF4.
As expected, expression of most genes associated with nitrogen fixation increased under nitrogen-fixing conditions.Nitrogenase enzyme encoding genes nifHDK (log 2 FC = 2.42, 1.60, 1.35, respectively) were upregulated, while nifEX critical to N 2 fixation exhibited upregulation but not significantly (P > 0.05).Nitrogen fixation-related genes glnG and hfq were upregulated (log 2 FC = 1.44, 2.37, respectively), which helped regulate and express genes for N 2 fixation at low-nitrogen level.The ammonia transporter encoded gene amtB was more highly expressed (log 2 FC = 1.39), glnAB and gltS (log 2 FC = 1.38, 1.98, 0.39) associated with NH 4 + uptake and transformation were upregulated under nitrogen fixation.The transcript level of glutamate dehydrogenase (gdhA) was downregulated in nitrogen-fixing cells but did not display a significant difference (P > 0.05).
To verify the transcriptome analysis, the genes glnB, glnG, hfq, fixA, and nifHD were selected for RT-qPCR analysis.These genes encoded functions needed for nitrogen fixation such as nitrogenase, nitrogen regulators, and an electron transfer flavoprotein.Through the melting curve analysis, the products amplified were all single products, indicating that the primers used are of high quality.As shown in Fig. 4B, the results of RT-qPCR were consistent with the transcriptome analysis, which proved the reliability of the RNA-seq data.Based on the above analysis, we proposed a hypothetical scenario of nitrogen fixation pathways and their regulation in Geomonas under nitrogen-fixing conditions (Fig. 5).Nitrogen fixation and NH 4 + assimilation by glutamine synthetase (GlnA) and glutamate synthase (GltS) were activated.Nitrogen-fixing genes such as nifHD were highly expressed.Cells need to obtain sufficient amounts of energy to generate electron donors such as ferredoxins/flavodoxins and ATP for nitrogen fixation.

DISCUSSION
Geomonas has mainly been isolated from paddy soil.Despite it having been reported that Geomonas possesses the nif gene cluster (16,17,19), whether it is able to fix nitrogen in pure culture has remained unclear.In this study, we provide evidence to support the hypothesis that Geomonas is capable of N 2 fixation.
Our data demonstrated that Geomonas species were abundant inhabitants of paddy soil.nifH was detected in all isolates, and the nif gene cluster was found on all sequenced genomes.The nif gene cluster structure in Geomonas was highly conserved, consisting of nifHDKENBX, draG, fdxN, and draT (Fig. 1B).Despite its close relation to Geobacter, Geomonas was shown to possess a distinct nif gene cluster.Compared to facultative anaerobic or strict anaerobic diazotrophs like Anaeromyxobacter, Azotobacter, Pseudomo nas, Pelobacter (33)(34)(35), Geomonas exhibited a remarkably compact structure of a nif cluster and also contained the minimum required gene set for nitrogenase, nifHDKENBX.Furthermore, the essential residues of the active site of nitrogenase were also conserved (Fig. 1C).Phylogeny of the nitrogenase nifHDK nucleotide sequences revealed clear differences between Geomonas and Geobacter as well as Anaeromyxobacter (2).Thus, it is reasonable to suggest that the proteins encoded by the nif gene cluster in Geomonas form an active nitrogenase complex.However, the regulatory mechanism of nitrogen fixation in Geomonas requires further studies.
Exocellular electron transfer is a crucial process and the availability of an electron donor limits the biomass and growth rate of strict anaerobes (36,37).Optimal electron donors were able to effectively increase the reducing reaction rate driven by anaerobes.Melo et al. (38) discovered that combing H 2 and pyruvate resulted in faster reduction of 2,4-dinitroanisole (DNAN) compared to using only one electron donor.Similarly, Santos et al. (39) found that lactate was the most effective electron donor for sulfate removal in wastewater treatment with high sulfate levels.Previous reports indicated that acetate was preferred as the electron donor for Geomonas growth (17), but the resulting biomass was low (Fig. S5).The addition of carbon-rich substrates has been shown to stimulate growth and activity, increasing the demand for nitrogen (40,41), glucose supplementa tion was able to promote a reduced rate of nitrogen fixation in the environment (42,43).Our results indicated that combining acetate and glucose as electron donors significantly increased Geomonas growth.
The cell morphology has been shown to undergo changes to adapt to environmental stress.For example, Rhodococcus changed the cell shape in response to temperature variations (31), cyanobacterium displayed elongated cells under limiting phosphorus but sufficient nitrogen levels (44), and the flagella of Ralstonia eutropha changed according to nutrient supply (45).However, no reports have shown the morphological comparison of cells of anaerobes grown under nitrogen-fixing versus non-nitrogen-fixing conditions.In this study, we observed the cell morphology of Geomonas was changing including cell shape and flagella when cultured under nitrogen-fixing condition.
Although it has been reported that some Geomonas species contained the nif cluster and were able to grow under nitrogen-fixing conditions without ammonia (17), there have been no studies characterizing nitrogenase activity of Geomonas in vitro.Our studies clearly demonstrated that Geomonas was able to fix and assimilate N 2 using nitrogenase.In addition, the nitrogenase activity of Geomonas was also suppressed by NH 4 + , a response that had been observed in other diazotrophs, including Anaeromyxo bacter (2,13,46).Previous studies had displayed physiological features of Geomonas, such as iron reduction (16,17), which was shown to play a crucial role in the rice soil ecosystem.The addition of ferric compounds has been predicted to enhance the nitrogen-fixing activity in paddy soil (47,48).Our study revealed that nitrogen fixation was an important feature of Geomonas, suggesting its importance in rice soil ecosystems.
To understand the nitrogen fixation process in Geomonas, we examined gene expression by transcriptome analysis under nitrogen-fixing conditions.The amt gene, encoding an ammonium transporter, was activated at low nitrogen levels.When NH 4 + concentrations are sufficient (>1 mM), ammonium can diffuse through the membrane as NH (49).Thus, we found amt expression was upregulated under nitrogen-fixing conditions, consistent with findings in Geobacter (29).The hfq gene was shown to regulate nitrogen fixation in other diazotroph (50) and was upregulated in Geomonas under nitrogen-fixing conditions.Glutamine synthetase encoded by glnA and glutamate synthase encoded by gltS are needed to synthesize the respective amino acids during anabolism providing a source for growth (51)(52)(53).We observed an upregulation of these genes in this study.glnB encoding the P-II family nitrogen regulator was highly expressed under nitrogen-fixing conditions (29).gdhA encoding glutamate dehydrogenase was predicted to be repressed during nitrogen fixation (54) similar to previous findings in Geobacter (27).However, gdhA was not significantly downregulated in our study.
Flavin-based electron bifurcation is an important energy coupling mechanism in the anaerobic microbial metabolism (55,56), which involves the splitting of a hydride electron pair by flavoproteins into two separate electrons with different reduction potentials (56,57).Genes encoding the flavin-based electron bifurcating enzyme complexes NfnAB and EtfAB have been observed in Geobacter (13,29).In this study, we observed high expression of fixAB encoding flavin-based electron bifurcation complexes similar to EtfAB in Geobacter, suggesting FixAB system was involved in nitrogen fixation in Geomonas.The Fix system has also been identified in the diazotrophs A. vinelandii and Rhodopseudomonas palustris, which facilitates the reduction of ferredoxins or flavodoxins for N 2 fixation (58-60).Jing et al. (13) reported that EtfAB and NfnAB in Geobacter sulfurreducens were upregulated in microbial electrolysis cells under nitrogen-fixing conditions, showing these complexes might reduce ferredoxins to drive N 2 fixation.Ortiz-Medina et al. (29) found that Geobacter exhibited higher nitrogen fixation activity when gene expression of EtfAB was upregulated when exposed to a potential of −0.15 V.However, no previous study has investigated electron bifurcation in Geomonas.Our results suggest that nitrogen limitation promotes electron bifurcation providing substantial energy for N 2 fixation.
Although strategies for generating reduced ferredoxins/flavodoxins for nitrogen fixation in Geomonas have not been studied, understanding these mechanisms is crucial.Ferredoxins (encoded by fer) are involved in respiration, nitrogen fixation, and carbon dioxide fixation in strict anaerobes (55).During nitrogen fixation, ferredoxin has been shown to act as an electron donor for nitrogenase (60,61).We observed high expression of the fer gene in Geomonas under nitrogen-fixing conditions.In Rhodospirillum rubrum, pyruvate-ferredoxin oxidoreductases (PFOR) have been found to contribute modestly to nitrogen fixation (62).Although many diazotrophs possessed PFORs on their genomes, only a few were shown to support nitrogenase activity.We found that PFOR in Geo monas was upregulated under nitrogen-fixing conditions.However, further research is needed to confirm the contribution of PFOR to nitrogen fixation in Geomonas.Addition ally, fliC-encoded flagellin, the major constituent of bacterial flagella, was significantly upregulated under nitrogen-fixing conditions.In contrast, flagella-related genes were downregulated in Geobacter during nitrogen fixation.Similar to Geobacter, pilA was not listed under the "electron transfer activity" category (29,63) but might indirectly contribute to nitrogen fixation.
In conclusion, we demonstrated that Geomonas inhabited paddy soils and were the first to characterize the nitrogen-fixing properties of Geomonas at the cultural level.This research provides valuable insights into growth conditions, nitrogen fixation properties, and gene expression patterns associated with nitrogen fixation in Geomonas.These findings highlight the significance of Geomonas as a diazotrophic organism and its crucial role in nitrogen fixation in rice fields, thereby providing a new strategy for future reduction in the use of nitrogen fertilizer.Overall, our study contributes to a comprehen sive understanding of Geomonas and its potential applications as a new pathway for reducing the dependency on nitrogen fertilizers in sustainable agriculture.

Soil samples collection, Geomonas isolation, and identification
Paddy soil samples were collected from a rice field in Fujian Province, China (26.1076 ° N, 119.3014 ° E).No fertilizer had been applied to these paddy soils for at least 3 years.This soil was also used in the previous research (16).Geomonas strains were isolated and cultured as described before by Liu et al. (16).The single colonies obtained were purified and identified by their 16S rRNA gene sequence.Pairwise 16S rRNA gene sequence similarities between the isolates and the comparison to related type strains were calculated using the EzBioCloud platform (64).For phylogenetic analysis, 16S rRNA gene sequences of closely related type strains were downloaded from the EzBioCloud platform.Phylogenetic trees were constructed based on the neighbor-joining (65) method with the Kimura two-parameter model (66) and 1000 bootstrap replications (67) using MEGA version X (68).

DNA extraction and genomic sequencing
For genome sequencing, genomic DNA was extracted from cells of Geomonas iso lates cultured in R2A with 40 mM disodium fumarate using TIANamp Bacterial DNA kit (TIANGEN, China) according to the manufacturer's instructions.Genomes were sequenced using Nanopore PromethION platform and Illumina NovaSeq PE150 at the Novogene Bioinformatics Technology Company, Ltd. (Beijing, China).The Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used to predict gene functions (69,70).
The nucleotide sequences of nifHDK of Geomonas and related strains were retrieved from the genomes sequenced in this study and the NCBI database.Phylogeny was inferred based on the neighbor-joining (65) method with 1000 bootstrap tests (67) using MEGA version X (68).Structural schematic diagram of the nitrogenase was drawn online using ChiPlot (https://www.chiplot.online/gene_cluster.html) to compare the differences in nitrogenase structure between Geomonas and related diazotrophic species.

Growth and cell morphology
Geomonas nitrogeniifigens RF4 was cultured in MFM broth.To optimize the growth conditions for G. nitrogeniifigens RF4, electron donors at different concentrations were tested in the modified MFM broth.The modified MFM (L −1 ) was composed of 2.0 g NaHCO 3 , 0.2 g MgSO 4 •7H 2 O, 0.3 g KH 2 PO 4 , 0.16 g CaCl 2 •2H 2 O, 1.0 mL vitamin stock solution (L −1 , 0.02 g biotin, 0.05 g folic acid, 0.1 g pyridoxine-HCl, 0.05 g thiamine-HCl, 0.05 g nicotinic acid, 0.05 g aminobenzoic acid, 0.05 g Ca-pantothenate, 0.01 mg vitamin B12, and 0.05 g lipoic acid), and 1.0 mL mineral stock solution [L −1 , 1.5 g nitrilotriacetic acid, 3.0 g MgSO 4 , 0.1 g FeSO   ]], and supplemented with 20 mM sodium acetate and 40 mM disodium fumarate.The pH was adjusted to 6.5.To observe the changes in cell morphology of Geomonas under nitrogen-fixing and non-nitrogen-fixing conditions, strain RF4 was cultured in modified MFM with 1 mM NH 4 Cl or without NH 4 Cl but with N 2 as the sole nitrogen source at 30°C for 3 days without shaking.The cultured cells were observed by transmission electron microscope (Hitachi HT7700, Japan) using the negative staining method with 1% phosphotungstic acid.

Assays of N 2 fixation and nitrogenase activity
Under nitrogen-fixing conditions, the optimal cell growth conditions of strain RF4 were determined by growth curves (OD 600 ).Two milliliters of strain RF4 culture in log phase in modified R2A (R2A + 20 mM disodium fumarate) was centrifuged for 10 min at 4,500 × g.To estimate nitrogen-fixing ability and nitrogenase activity, the cells were washed three times with oxygen-and nitrogen-free sterile MFM broth and then transferred into MFM with and without ammonia and incubated for 3 days without shaking, respectively.The electron acceptor under both treatments was fumarate, and acetate and glucose were both provided as electron donors.As a control experiment, the same system was set up using pure Ar instead of N 2 /CO 2 gas.The strain was anaerobically incubated in serum bottles sealed with butyl rubber plug and aluminum crimp under N 2 /CO 2 (80:20 [vol/ vol]) atmosphere at 30 ℃ without shaking.Three replicates were set for each experiment in this study.
Nitrogen fixation activity was measured by the ARA based on acetylene (C 2 H 2 ) reduction into ethylene (C 2 H 4 ) by nitrogenase (71).Briefly, strain RF4 was precultured in the modified MFM medium obtained from the analysis described above for 4 days.The experimental conditions were identical except the head space in the bottle was Ar/C 2 H 2 (90:10 [vol/vol]) gas.The negative control was set up using pure Ar gas.C 2 H 4 production was measured by gas chromatography on a fused-silica column (Porapak; Hychrom) (16).The inhibitory effect of NH 4 + on nitrogen fixation activity was deter mined in N-free MM with 0-1 mM NH 4 Cl added (interval of 0.2).The effect of different concentrations of ammonium on nitrogen fixation activity was estimated based on ARA as described above.The nitrogen-fixing rate was estimated by calculating the amount of total nitrogen during the nitrogen fixation process.The total nitrogen was extracted using total nitrogen detection kit (HACH) according to manufacturer's instruction and measured on a UV Spectrophotometer (DR3900, HACH).
To further confirm whether strain RF4 facilitated nitrogen fixation, an isotope labeling enrichment assay was conducted as described previously (13).Briefly, 2 mL cell culture of strain RF4 in log phase cultured in modified R2A was centrifuged for 10 min at 4,500 × g.The cells were washed three times with oxygen-free sterile MFM broth (without N) and then transferred into 15 N 2 serum bottles and incubated for 3 days without shaking.The serum bottle contained 20 mL sterile N-free MFM (without NH 4 + ) and MFM (with 1 mM NH 4 + ), respectively.The headspace of a serum bottle containing N-free MFM was first full of Ar, and then 10% vol was replaced by 15 N 2 .The cells were harvested and freeze-dried for further 15 N isotope analysis.The δ 15 N value ( 15 N [m/z = 29]/ 14 N [m/z = 28]) of bacterial cells was determined by an isotope ratio mass spectrometer (EA-IRMS, Thermo Scientific MAT253 Plus gas bench isotope mass spectrometer).

RNA extraction and transcriptome sequencing
Strain RF4 was cultured in MFM medium with NH 4 + or without NH 4 + for 3 days.The cultures were collected by centrifugation at 10,000 × g for 2 min.The cells were frozen in liquid nitrogen and stored at −80°C for subsequent RNA extraction.RNA was extracted using TRIzol Reagent kit according to the manufacturer's instructions (Invitrogen).RNA concentration and purification were detected by NanoDrop 2000 (Thermo Scientific).Transcriptome sequencing was performed by Majorbio Bio-pharm Technology Co., Ltd (Shanghai, China).All statistical analysis was performed using the free online platform of Majorbio Cloud Platform (https://www.majorbio.com).

RT-qPCR
Fluorescence real-time quantitative PCR (RT-qPCR) was performed to evaluate the reliability of RNA-Seq transcriptome results.By comparing the expression of genes encoding functions involved in the nitrogen fixation process, a total of six upregula ted genes were selected for RT-qPCR analysis (Table S2).RNA was isolated using the same process as in the transcriptome analysis.Double-stranded cDNA was synthesized using a SuperScript double-stranded cDNA synthesis kit (Invitrogen, CA).RT-qPCR was performed using 2× Taq Plus Master Mix by following the manufacturer's protocol.For genes glnB, glnG, hfq, fixA, and nifHD, rpoB was used as internal reference for quantitative analysis of relative expression.The thermal cycling for all genes was performed as follows: preheat at 95°C for 300 s followed by 35°C cycles of 95°C for 30 s, 52°C for 30 s, and 72°C for 60 s.Three biological replicates were used in this study.

FIG 1
FIG 1 Phylogeny and nitrogenase gene regions in Geomonas based on genome analysis.(A) Neighbor-joining phylogenetic tree of the aligned amino acid sequences of NifHDK retrieved from genomes.(B) Comparison of the nif gene cluster in Geomonas and its relatives.Note: regions a to f contain 4, 5, 12, 1, 2, 8 genes, respectively.Homologous genes are shown in (Continued on next page)

FIG 3
FIG 3 Nitrogen fixing ability and inhibiting effect of NH 4 + on nitrogenase activity of strain Geomonas nitrogeniifigens RF4.(A) Elemental analysis-iso topic ratio mass spectrometry of 15 N 14 N and 14 N 14 N; (B) total nitrogen accumulation through nitrogen fixation; (C) nitrogenase activity estimated by the ARA method; (D) influence of NH 4 + on nitrogenase activity.

FIG 4 FIG 5
FIG 4 Differential expression levels (A) and selected differentially expressed genes confirmed by RT-qPCR (B) of strain Geomonas nitrogeniifigens RF4 when comparing growth with and without NH 4 + .