Beneficial factors for biomineralization by ureolytic bacterium Sporosarcina pasteurii

Background The ureolytic bacterium Sporosarcina pasteurii is well-known for its capability of microbially induced calcite precipitation (MICP), representing a great potential in constructional engineering and material applications. However, the molecular mechanism for its biomineralization remains unresolved, as few studies were carried out. Results The addition of urea into the culture medium provided an alkaline environment that is suitable for S. pasteurii. As compared to S. pasteurii cultivated without urea, S. pasteurii grown with urea showed faster growth and urease production, better shape, more negative surface charge and higher biomineralization ability. To survive the unfavorable growth environment due to the absence of urea, S. pasteurii up-regulated the expression of genes involved in urease production, ATPase synthesis and flagella, possibly occupying resources that can be deployed for MICP. As compared to non-mineralizing bacteria, S. pasteurii exhibited more negative cell surface charge for binding calcium ions and more robust cell structure as nucleation sites. During MICP process, the genes for ATPase synthesis in S. pasteurii was up-regulated while genes for urease production were unchanged. Interestingly, genes involved in flagella were down-regulated during MICP, which might lead to poor mobility of S. pasteurii. Meanwhile, genes in fatty acid degradation pathway were inhibited to maintain the intact cell structure found in calcite precipitation. Both weak mobility and intact cell structure are advantageous for S. pasteurii to serve as nucleation sites during MICP. Conclusions Four factors are demonstrated to benefit the super performance of S. pasteurii in MICP. First, the good correlation of biomass growth and urease production of S. pasteurii provides sufficient biomass and urease simultaneously for improved biomineralization. Second, the highly negative cell surface charge of S. pasteurii is good for binding calcium ions. Third, the robust cell structure and fourth, the weak mobility, are key for S. pasteurii to be nucleation sites during MICP.


Background
Microbially induced calcite precipitation (MICP) refers to the formation of calcium carbonate in the presence of the different metabolic activities of microorganisms, including urealysis, photosynthesis, denitrification, ammonification, sulfate reduction and methane oxidation [1,2]. It is widely used to repair cracks of constructions and roads, solidify sand and gravels, manufacture self-healing cement, develop bio-mineralization materials and so on [3]. Up to now, a large number of microorganisms have been discovered to perform MICP by ureolysis for various applications. For example, Shewanella has a positive effect on the compressive strength of mortar specimens [4]. A two-component self-healing bio-chemical agent was developed by mixing calcium lactate with spores of Bacillus alkalinitrilicus isolated from alkaline lake soil (Wadi Natrun, Egypt) to improve the durability of concrete structures [5]. Ureolytic bacterium Pararhodobacter sp. was used to solidify coral sand [6]. Among these Ma et al. Microb Cell Fact (2020) 19:12 well-known mineralizing bacteria, S. pasteurii was the most frequently used in MICP [7][8][9].
Under proper conditions, S. pasteurii can produce urease to decompose urea into carbonate and ammonia. The resulted carbonate reacts with calcium ions, forming the calcite (calcium carbonate) precipitation [10,11]. S. pasteurii-mediated MICP has been widely applied in constructional engineering and bio-materials. The culture conditions of S. pasteurii were optimized to endow it with the strong ability of mineralization for hardening a sponge bar [12]. Through morphological observations, it was revealed that S. pasteurii could form a high degree of calcite crystals for bridging, coating on sand particles and infilling pore spaces, which was used to solidify sandy soil columns [13]. S. pasteurii and bacterially produced γ-polyglutamate (PGA) solution were applied to produce nacre-inspired layered calcium carbonate-polyglutamate composite materials that exhibited high extensibility and stiffness [14]. However, there are application limitations for S. pasteurii-mediated MICP including its vulnerability to different application conditions, uncontrolled growth, undesired by-products, and inadequate engineering scale-up. Some previous researches have tried to solve these problems by optimization of growth and mineralization conditions for S. pasteurii. The concentrations of urea, calcium chloride and nickel were optimized to maximize the growth rate and the calcium carbonate precipitation ability of S. pasteurii using the response surface methodology [15]. A combination of meat extract and sodium acetate was a suitable replacement for yeast extract in the growth medium for S. pasteurii with enhanced calcium carbonate formation in concrete by markedly reducing the retardation of cement hydration [16]. Furthermore, as S. pasteurii could not grow and synthesize urease anaerobically, repeated supply of cells and/ or oxygenated medium might be necessary, especially when the ureolysis happens in areas far from the injection point [17]. In contrast to these extensive efforts on the practical applications of S. pasteurii-mediated MICP, few studies have been carried out to unravel the mechanism underlying biomineralization of S. pasteurii, which definitely would benefit its further practical applications.
In this study, we explored the potential processes involved in the biomineralization of S. pasteurii. The genome of S. pasteurii was sequenced and its cell surface charge was measured and compared to the non-mineralizing microorganism. The effect of urea was investigated on the biomass, the urease production, morphology, cell surface charge and the transcriptional level of S. pasteurii, and the pH of culture medium to demonstrate why S. pasteurii for MICP is generally grown with urea. Finally, the impact of MICP on the morphology and the transcription level of S. pasteurii was studied. This study sheds light on the molecular mechanism underlying the biomineralization of S. pasteurii, which will help to guide its application studies in the future.

Results and discussion
Genomic properties of S. pasteurii BNCC 337394 The genomic information of S. pasteurii plays an essential role in its further molecular mechanism study on bio-mineralization. Recently, S. pasteurii NCTC4822 was sequenced by whole genome shotgun with the NCBI Accession Number: NZ_UGYZ01000000 (https ://www. ncbi.nlm.nih.gov/nucco re/NZ_UGYZ0 00000 00.1). Its chromosome is 3.3 Mb, containing 3063 coding genes [18]. Nevertheless, S. pasteurii BNCC 337394 used in this study has not been sequenced. Thus, we performed the de-novo whole genome sequencing for S. pasteurii strain BNCC 337394, showing that its genome is 3.25 Mb with GC content of 39.17% and contains 3105 coding genes (Additional file 1), similar to that of strain NCTC4822. The COG analysis of the genome of S. pasteurii showed there were 2106 genes with known protein function and 999 genes whose protein functions were unknown (Additional file 2). The ratio of the number of genes in each classification to the total number of functional genes was calculated. Among the genes of known function, the ratio of genes related to amino acid transport and metabolism (E) was the largest, followed by genes involved in inorganic ion transport and metabolism (P) and transcription (K) (Fig. 1). S. pasteurii produces urease to convert urea to CO 3 2− that reacts with Ca 2+ to form CaCO 3 , which is the key to biomineralization [19]. Urease is composed by amino acids, requiring the functions such as Amino acid transport and metabolism (E), and Transcription (K). More particularly, through the Inorganic ion transport and metabolism (P), the inorganic ions nickel, a key part of the active center of urease, could regulate the urease activity to improve the ability of urealysis [20]. Therefore, the high proportion of functions E, P, and K in S. pasteurii might be beneficial for the action of urease, helping implement its mineralization ability.
Additionally, seven genes involved in urease assembly and activation were found in the sequenced genome of S. pasteurii as a gene cluster (Additional file 3: Table S1). UreA, ureB and ureC were three subunits of the main structure of urease holoprotein (UreABC) 3 , while ureD, ureF and ureG were three essential urease accessory proteins forming the complex UreDFG required for the functional incorporation of the urease nickel metallocenter to mature urease. One (UreABC) 3 and three UreDFG form the (UreABC-UreDFG) 3 complex. Then ureE as a nickel donor takes nickel ions into the enzyme active center of the complex to dissociate the accessory proteins in order to activate urease [21,22]. As found by COG analysis, genes ureA, ureB and ureC belong to Amino acid transport and metabolism (E), while ureD, ureE and ureF are assigned to Post-translational modification, protein turnover and chaperones (O). Gene ureG is attributed to both Transcription (K) and Post-translational modification, protein turnover and chaperones (O). These ureaserelated genes can serve as potential targets by metabolic engineering to improve the efficiency of urease production in S. pasteurii for enhanced bio-mineralization in future study.

The effect of urea in culture medium on S. pasteurii
We investigated the effect of urea, a key component of the bio-mineralization, on the growth, intracellular urease activity, morphology, cell surface charge and mineralization ability of S. pasteurii grown in TSB medium. With urea, the optical density (OD) of S. pasteurii at 600 nm was increased sharply to OD 600 ≈ 5 within 12 h and then entered a steady period (Fig. 2a). Accordingly, its enzyme activity had a rapid increase to the maximal 6.8 IU/OD at 6 h with a slight decrease to 4.5 IU/OD before reaching the plateau (Fig. 2b). Without urea, the OD 600 of S. pasteurii was only 0.31 at 12 h, much lower than that with urea (OD 600 ≈ 5), and reached 3.6 at 38 h (Fig. 2a). Meanwhile, its enzyme activity did not increase observably until 10 h. The maximal enzyme activity was about 12 IU/ OD at 25 h, exceeding that (6.8 IU/OD) in the presence of urea at 6 h (Fig. 2b). Obviously, S. pasteurii cultivated with urea showed faster growth and urease production than that without urea, though its maximal urease activity was much lower. This finding was consistent with previous studies that revealed the urease synthesis was repressed in the presence of ammonia in Klebsiella aerogenes, Bacillus subtilis and Rhizobium meliloti [23][24][25]. TSB itself did not produce ammonium, while the addition of urea can generate a high level of ammonium ( Fig. 2c), inhibiting the production of the urease synthesis in S. pasteurii.
During the growth of S. pasteurii with urea, the generated urease quickly converted urea into ammonium (Fig. 2c), making the pH of culture medium increase rapidly from 7.4 to 9.4 in 5 h (Fig. 2c), accompanied by the same upward trend of ammonium ions. The resulted alkaline environment is suitable for the proliferation of S. pasteurii, an alkalophilic bacterium. By contrast, the medium without urea kept at pH 7.7 throughout the whole growth phase due to the unavailable ammonium (Fig. 2c), which was not suitable for the growth of S. pasteurii, resulting in slow proliferation. As observed by scanning electron microscopy (SEM), S. pasteurii cultivated with urea was health with intact cells (Fig. 2d), while S. pasteurii cultivated without urea became out of shape and slightly maldeveloped (Fig. 2d). Besides, S. pasteurii cultivated with urea displayed more negative charge (− 57 to − 63 mV) than that cultivated without urea (− 48 to − 53 mV) during the whole growth process (Fig. 2e).
The weight of precipitation produced by the bacterial suspension of S. pasteurii grown with and without urea was analyzed either with the same growth OD 600 (OD 600 ≈ 1) where the urease activity was ~ 5.6 IU/ OD with urea, and ~ 11.6 IU/OD without urea or at the same growth time (12 h). In both cases, the weight of precipitation produced by S. pasteurii cultivated with urea was larger than that produced by S. pasteurii without urea (Fig. 2f ). Even when there was no extra urea added for mineralization, S. pasteurii cultivated with urea still produced more precipitation than that cultivated without urea (Fig. 2f ). In the absence of urea culture, with extra urea, the precipitation produced at OD 600 ≈ 1 was about 100 mg, more than that at 12 h because the urease activity at OD 600 ≈ 1 was about 11.6 IU/OD, bigger than about 5 IU/OD at 12 h (Fig. 2b), which meant that high urease activity led to to the total number of functional genes. COG designations are described as follows: B, chromatin structure and dynamics; C, energy production and conversion; D, cell cycle control, cell division, chromosome partitioning; E, amino acid transport and metabolism; F, nucleotide transport and metabolism; G, carbohydrate transport and metabolism; H, coenzyme transport and metabolism; I, lipid transport and metabolism; J, translation, ribosomal structure and biogenesis; K, transcription; L, replication, recombination and repair; M, cell wall/ membrane/envelope biogenesis; N, cell motility; O, posttranslational modification, protein turnover, chaperones; P, inorganic ion transport and metabolism; Q, secondary metabolites biosynthesis, transport and catabolism; T, signal transduction mechanisms; U, intracellular trafficking, secretion, and vesicular transport; V, defense mechanisms; Z, cytoskeleton. This figure didn't contain the genes of unknown function (S) and general function prediction only (R) producing more CO 3 2− for CaCO 3 formation. While without extra urea, there was little precipitation comprised by bacterial cells at both OD 600 ≈ 1 and 12 h, which indicated that the mineralization process of S. pasteurii mainly depended on urea hydrolysis. The mineralization of S. pasteurii worked best when grown in the presence of urea for 12 h where both OD 600 and the intracellular urease activity of S. pasteurii reached a plateau, which was not achieved by other conditions tested here. The presence of urea enabled a good correlation of biomass growth and urease production of S. pasteurii, which is beneficial for its biomineralization. Taken together, S. pasteurii grown in the presence of urea could simultaneously proliferate and produce urease more rapidly with a healthier cell form and more negative cell surface charge in an alkaline environment as compared to that grown in the absence of urea, showing significantly enhanced mineralization ability. That S. pasteurii was grown slowly in TSB without urea contradicted the early finding that S. pasteurii exhibited no growth in ordinary neutral media without urea supplementation [26]. We have grown S. pasteurii in TSB + 2% urea for pre-culture. Pre-cultures were diluted 1:100 into 100 mL TSB in 500-mL conical flasks. Although no extra urea was supplied, there might be a trace amount of urea or its degraded product ammonium from preculture, enabling S. pasteurii to grow in TSB without urea. To prove this, we have washed the preculture with PBS three times to remove the residue urea before inoculation into 100 mL TSB. With the removal of residue urea, S. pasteurii was not grown out at all in TSB without urea (Fig. 2a) as found in the previous study [26], evidencing our speculation.

The effect of urea on the transcriptome of S. pasteurii
To understand how urea influences S. pasteurii at the transcriptional level, we performed the transcriptome analysis of S. pasteurii cultivated in TSB medium with or without urea at the same OD 600 ≈ 1, at which the corresponding intracellular urease activity was ~ 5.6 IU/ OD and ~ 11.6 IU/OD, respectively. The total reads of S. pasteurii cultivated with urea were mapped to the reference genome (as we sequenced above) with coverage of ~ 98.15-98.44%. Genes were differentially expressed when the average reads of the corresponding transcripts differed with log 2 (fold change) ≥ 1 and padj ≤ 0.05. By comparing S. pasteurii cultivated without urea (SP2) to with urea (SP1), there were a total of 1008 significant differentially expressed genes (DEGs) in the absence of urea, of which 421 were up-regulated and 587 down-regulated (Additional file 4). The DEGs were classified according to their gene functions (Fig. 3a). Except functions Translation, Ribosomal structure and biogenesis (J), Inorganic ion transport and metabolism (P) and Cell motility (N), most gene functions contained relatively more down-regulation genes, such as Energy production and conversion (C), Amino acid transport and metabolism (E) and Transcription (K).
Specifically, genes ureD, ureE, ureF and ureG that adjust the nickel active center and mature urease, and ureC for the urease structure assembly were notably increased, while ureA and ureB for the urease structure assembly stayed unchanged (Fig. 3a). Meanwhile, 41 out of 135 ribosomal genes were up-regulated DEGs (Additional file 3: Table S2). The up-regulation of genes involved in urease synthesis and ribosomal genes might be linked with the higher urease activity in S. pasteurii cultivated without urea as we observed (Fig. 2b).
The genes of F-type ATPase are all DEGs that were significantly up-regulated (Fig. 3b). F-type ATPase is divided into two regions F 1 (containing genes atpA, atpD, atpG, atpH, and atpC) and F 0 (containing genes atpB, atpF, and atpE). The F 1 region is responsible for ATP hydrolysis, while the F 0 region for proton transfer [27,28]. Similar to urease genes, all these genes were also clustered in the genome of S. pasteurii (Additional file 3: Table S1). The ATP generation in S. pasteurii was coupled with urea hydrolysis At pH 9.4 optimum for S. pasteurii growth, the extracellular proton concentration was lower than the intracellular one, leading to a reversed ∆pH (the transmembrane pH gradient) with the tendency for protons to move from inside to outside the cells. To combat this, S. pasteurii have developed two alternative ways to increase proton motive force and drive protons into the cells for ATP generation: 1) alkalinisation of the cytoplasm, leading to a reduced ∆pH 2) efflux of other cations than H + , giving rise to an increased ∆ψ (membrane potential). If ∆ψ is large enough, proton motive force becomes sufficient to drive the influx of protons into the cells against the concentration gradient for ATP generation. Urea diffuses into bacterial cells according to concentration gradient, where it is hydrolyzed by urease to ammonium and carbonate, leading to the alkalinisation of the cytoplasm and a decreased ∆pH. Ammonium was excreted outside the cells according to its concentration gradient, causing an increased ∆ψ. The increased ∆ψ is reversed through driving protons into the cells against the concentration gradient, resulting in ATP generation [29][30][31]. This coupling was probably broken in the absence of urea as indicated by the slow growth of S. pasteurii in the absence of urea no matter the pH was optimum or not (Fig. 2a), leading to the lack of ATP. To supply enough ATP, the efficiency of ATPase can be improved by up-regulating the expression of ATPase genes.
14 genes related to flagella were significantly up-regulated and only 1 gene down-regulated (Table 1). Among these genes, FliD and FliC are related to the flagella filament that acts as a helical screw to produce thrust for swimming motility. FliN, FliM and FliG are responsible for the C ring structure acting not only as a central part of the rotor for torque generation but also as a structural device to switch the direction of motor rotation. FliP, FliO, FliI and FliH constitute the type III protein export apparatus that is responsible for transporting axial component proteins from the cytoplasm to the distal end of the growing flagella structure to construct the axial structure beyond the cellular membranes. FliK is related to hook structure to control the kook length and FlgG is related to the distal rod. Both FliK and FlgG are involved in regulating the bending flexibility of the hook and rod structure [32,33]. Except FlgG, all genes were up-regulated in the absence of urea. The increased expression of these genes might allow for enhanced motility of S. pasteurii to approach the nutrients to survive the nitrogen starvation caused by the absence of urea from the culture medium [34]. However, it is north noting that FlgG was reduced. Why the expression of FlgG was downregulated specifically in the absence of urea and how it is going to affect the efficiency and function of flagella is unknown.
In summary, S. pasteurii cultivated in the absence of urea notably up-regulated the genes involved in urease synthesis, ATP synthesis and flagella to cope with the unfavorable culture conditions due to the absence of urea, which might compromise its biomineralization ability. The presence of urea provided S. pasteurii favorable culture conditions, saving extra energy to make it more prepared for biomineralization.

S. pasteurii displays high negative surface charge
In MICP, cationic calcium ions need to bind to the negatively charged surface of biomineralization microorganism, suggesting the microbial surface charge might play a role in the biomineralization capability of S. pasteurii [35]. Therefore, we measured the zeta potential of S. pasteurii, and non-mineralizing bacteria Escherichia coli, Staphylococcus aureus and Bacillus subtilis, which were cultivated in TSB with 20 g/L urea (Fig. 4). The potential of S. pasteurii was − 67 mV, while the other three nonmineralizing bacteria E. coli, S. aureus and B. subtilis were − 28, − 26 and − 40.8 mV, respectively. Obviously, S. pasteurii exhibited much more negative surface charge than non-mineralizing bacteria E. coli, S. aureus and B. subtilis. When treated with 0.2 mol/L or 2 mol/L CaCl 2 , a noticeable upward trend of the zeta potentials was found for all the tested bacteria with the largest increase observed for S. pasteurii. As we know, bacterial surface has various components contributing to negative charge such as phosphate groups, carbonate groups and sulfate groups. The metal ions like calcium, magnesium and copper can bind to them. That S. pasteurii has much more negative surface charge than other non-mineralizing bacteria indicates that there might be more negative functional groups on its surface for binding more metal ions like calcium ions in the same ionic environment. Even without urea, the negative charge of S. pasteurii was more than that of non-mineralizing bacteria E. coli, S. aureus and B. subtilis (Fig. 2e), indicating that the superb negative charge of S. pasteurii is constitutive which might be due to the components of the cell membrane. This can greatly benefit its mineralization process. It has been reported that the zeta potential of S. pasteurii ATCC 6453 and BNCC 337394 was − 19.51 to − 23.10 mV and − 32.5 ± 1.3 to −35.6 mV respectively, higher than the zeta potential of S. pasteurii BNCC 337394 in this study [35,36]. These distinctive zeta potential values might be attributed to the different strains, different culture mediums and/or different measurement conditions used in each study.

S. pasteurii displayed intact cell structure under the biomineralization conditions
It is reported that S. pasteurii serves as the nucleation sites in MICP for CaCO 3 precipitation [37][38][39], requiring that the structure of S. pasteurii must be robust without lysis and maintain highly intact under the conditions of MICP. To prove this speculation, we measured the morphology of S. pasteurii, B. subtilis and E. coli after MICP by SEM (Fig. 5). S. pasteurii grown in TSB medium with urea had perfectly intact cell forms. After being incubated with both CaCl 2 and urea, the shape of S. pasteurii was complete and embedded inside lumpy substances (Fig. 5). The lumpy substances contained 11.26% calcium, 26.86% carbon, and 61.67% oxygen (Additional file 3: Table S3). The molar ratio of the three elemental components was close to 1:1:3, demonstrating that these substances were mainly calcium carbonate (CaCO 3 ). Obviously, S. pasteurii maintained a highly intact cell structure even under mineralization condition. The non-mineralizing B. subtilis is supposed to share similar bacterial cell wall as S. pasteurii, for both of them are Gram-positive bacteria. However, B. subtilis was less healthy in TSB medium with urea than S. pasteurii. In the presence of CaCl 2 and urea, few malformed B. subtilis were observed among the massive aggregates with an amorphous loose porous structure, indicating that most of B. subtilis cells had lysed. Gram-negative bacterium E. coli, another nonmineralizing bacterium, also behaved similarly to B. subtilis under the MICP conditions. Those amorphous loose porous matters that appear in association with E. coli and B. subtilis were probably CaCl 2 and urea salts in addition to bacterial remains. According to the experimental conditions, only CaCl 2 and urea were added to the bacterial cells. For SEM, the samples were dehydrated and dried, resulting in the precipitation of CaCl 2 and urea. Collectively, we found that S. pasteurii were grown healthier in slightly alkaline medium than non-mineralizing bacteria E. coli and B. subtilis. Under mineralization condition, S. pasteurii was still in great shape, even being surrounded by the compact calcium carbonate precipitation. By contrast, B. subtilis and E. coli became damaged and lysed inside amorphous loose porous aggregate (Fig. 5). Seemly, the robust structure of S. pasteurii under the mineralization conditions play an important role in its superior biomineralization ability in addition to its highly negative surface charge as we found above.

Transcriptome analysis of S. pasteurii before and after mineralization
To understand how MICP influences S. pasteurii at the transcriptional level, transcriptome analysis was performed by comparing the transcriptional level of S. pasteurii after mineralization (SP3) with that before mineralization (SP1). The total reads of S. pasteurii were mapped to the reference genome (as we sequenced above) with coverage of ~ 80.42-89.47%. A total of 1316 DEGs of which 590 were up-regulated and 726 downregulated were found after mineralization and classified (Additional file 4). Functions Cell cycle control, cell division, chromosome partitioning (D), Translation, ribosomal structure and biogenesis (J), Cell wall/membrane/ envelope biogenesis (M) and Posttranslational modification, protein turnover, chaperones (O) contained obviously more up-regulated genes (Fig. 6a). The number of the up-regulated genes in Carbohydrate transport and metabolism (G) was similar to that of the down-regulated genes. Except those functions, the other functions contained more down-regulated genes.
During MICP, the genes related to ATPase of S. pasteurii were significantly up-regulated (Fig. 6b), whereas the genes of urease showed relatively unchanged expressions (Fig. 6b). We found that pH of S. pasteurii solution decreased from 9.31 to 8.22 during the mineralization process, which was not favorable for the coupling of ATP synthesis with urea hydrolysis, and the growth of S. pasteurii. Similar to the case without urea, S. pasteurii might increase the expression of genes encoding ATPase to keep supplying enough energy in unfavorable environment. In contrast, urease produced in sufficient quantities before mineralization might be stable enough that S. pasteurii did not need to enhance the expression of the related genes during MICP. It is demonstrated that the mineralization conditions promote the ATP synthesis of S. pasteurii but have no significant influence on urease production and activity.
Besides, it was interesting to observe that the gene expressions of flagellar proteins included in function cell motility (N) were almost inhibited ( Table 2). These proteins are components of bacterial flagellum, which are involved, in bacterial motility and chemotaxis as mentioned above [32,33,40]. The decreased expression of the corresponding proteins in the bacterial flagellum of S. pasteurii might result in the inhibition of bacterial motility under mineralization conditions, which is beneficial for the biomineralization, giving that S. pasteurii cells serve as the nucleation sites during bacterial calcium precipitation. As compared to that without urea, the mRNA level of genes related to bacterial flagellum in S. pasteurii cultivated with urea was significantly reduced (Table 1), which was further decreased after mineralization (Table 2). Hence, we conjecture that the activity of bacterial flagellum of S. pasteurii might be closely related to the urea concentration in the environment. But more studies are required to confirm this hypothesis in the future.

Conclusions
In this study, the genome of S. pasteurii BNCC 337394 was sequenced. Urea in cultural medium was rapidly converted to ammonium, increasing pH of the culture medium from 7.4 to 9.4 and creating an alkaline environment that is suitable for the proliferation of S. pasteurii. As a result, the growth and urease production in S. pasteurii were accelerated notably at the same time, providing sufficient biomass and urease for biomineralization. Moreover, S. pasteurii cultivated with urea displayed healthier cell surface for nucleation sites and more negative cell surface charge for calcium ion binding than that without urea. All these led to improved biomineralization of S. pasteurii grown on urea, as compared to that of S. pasteurii grown without urea. To survive the unfavorable environment caused by the absence of urea, S. pasteurii increased the expression of genes involved in urease production, ATPase synthesis and flagella, which might occupy the resources for biomineralization. It was first reported that S. pasteurii displays superb negative charge, much more negative than that of non-mineralizing bacteria E. coli, S. aureus and B. subtilis. During MICP, S. pasteurii increased the mRNA levels of ATPase synthesis genes, while no impact was observed on the expression of urease genes. Genes related to flagella were down-regulated during MICP, which might cause weak mobility of S. pasteurii. More interestingly, the expression of genes responsible for fatty acid degradation was remarkably inhibited, suggesting that fatty acid degradation pathway might be shut down. This finding was consistent with the intact cell structure of S. pasteurii during MICP that was not seen for non-mineralizing bacteria.

Bacterial strains and growth conditions
Deionized water was obtained from a Milli-Q synthesis system (Billerica, MA, USA

Optical density measurement
The growth of S. pasteurii cultivated in TSB medium with 20 g/L urea and no urea respectively, was monitored in standard quartz cells by measuring the optical density at OD 600 nm with a UV-2600 UV-Vis spectrophotometer (SHIMADZU, Japan). A standard quartz cell contains 1 mL solution. When their OD 600 was > 1, the test samples were diluted with sterile water with proper dilution factors for OD measurement to make sure OD 600 is below 1. And the actual values of OD 600 were obtained by multiplying the dilution factors.

Zeta potential measurement
Sporosarcina pasteurii, E. coli, and B. subtilis, respectively were grown in TSB containing 20 g/L urea. When OD 600 was about 1, the cell cultures were collected and centrifuged at 4200×g for 5 min. After discarding the supernatant, the cell pellets were washed with PBS three times to remove medium residuals. The bacterial aggregate was resuspended in 1 mL PBS, 1 mL calcium chloride solution (0.2 mol/L) and 1 mL calcium chloride solution (2 mol/L) respectively to obtain three kinds of bacterial suspensions (OD 600 ≈ 1), and then was transferred to Malvern polystyrene U-shaped cell individually. Zeta potentials of these samples were measured with a Zetasizer Nano ZS instrument (Malvern Instruments, United Kingdom). The values presented are the average of three test results, and the standard deviation was considered as the error range.

Scanning electron microscope (SEM) characterization
Sporosarcina pasteurii, B. subtilis and E. coli were cultivated in TSB medium with 20 g/L urea until OD 600 was about 1. One milliliter bacterial suspension was collected and treated with 0.5 mL calcium chloride solution (1 mol/L), or calcium chloride solution (1 mol/L) and urea solution (1 mol/L) for 30 min, respectively. The untreated bacterial suspensions were set as controls. The samples were centrifuged at 4200×g for 5 min to obtain precipitation. After washing with sterile water three times, the precipitation was fixed in 2.5% glutaraldehyde solution for 4 h and then washed again to remove the residual glutaraldehyde. The samples were dehydrated with gradient dehydration by ethanol (30%, 50%, 70%, 85%, 95% and 100%) for 20 min for each concentration.
Then the diluted samples were loaded on the clean silicon wafer and left to dry. The images of the samples were obtained using a Zeiss Ultra Plus-43-12 FESEM (Zeiss, Germany). The energy dispersive spectrometer (EDS) results were evaluated with the electron probe configured in the FESEM for elemental analysis.

Intracellular urease activity and mineralization precipitation
The intracellular urease activity was determined by measuring the amount of ammonia released from urea. Firstly, the urease of S. pasteurii was extracted by Bacterial Protein Extraction Kit (BBI Life Sciences Corporation). Then 0.1 mol/L urea was added into the same volume of extracted urease solution for 30 min at 37 °C. The amount of ammonia released from urea was measured using phenol-hypochlorite assay method [41]. One unit of urease is defined as the amount of enzyme hydrolyzing urea to produce one micromole ammonia per minute. The specific urease activity (IU/OD) was obtained by dividing the urease activity (IU) by the bacterial biomass (OD 600 ). Two milliliter bacterial suspension of S. pasteurii under different conditions as indicated in the text was collected, and mixed with 1 mL urea (1 mol/L) and 1 mL CaCl 2 (1 mol/L) for 15 min to produce the precipitation. Precipitation mass was obtained by centrifugation (8000 rpm, 5 min) and then dried at 50 °C for 8 h before being weighed.

Whole genome sequencing
Sequencing was performed at Genewiz Bioinformatics Technology Co., Ltd (Suzhou, China). Genomic DNA was sheared, and then 20 Kb double-stranded DNA fragments was selected. DNA fragments were end repaired and ligated with universal hairpin adapters. Subsequently, the SMRTbell library was prepared and sequenced in PacBio Single Molecule Real-Time (SMRT) instrument. Then the PacBio reads were assembled using SmrtLink. The prodigal (Prokaryotic Dynamic Programming Genefinding Algorithm) software has been used for finding coding genes in bacteria. The coding genes were annotated with National Center for Biotechnology Information (NCBI) nr database by Diamond. The proteins encoded by genes were classified on a phylogenetic classification by the database of COG (Clusters of Orthologous Groups). The corresponding fastq files of S. pasteurii was deposited into NCBI SRA database with the NCBI Accession Number: SRR8727707.

Transcriptomics analysis
The total RNA of samples were extracted with the TRIzol Reagent (Invitrogen, USA)/RNeasy Mini Kit (Qiagen, Germany). Total RNA of each sample was quantified and qualified by Agilent 2100 Bioanalyzer (Agilent Technologies, USA), NanoDrop (Thermo Fisher Scientific Inc., USA) and 1% agarose gel. Illumina RNA sequencing was carried out by Genewiz following their standard analysis method. Gene FPKMs were computed by summing the FPKMs of the transcripts in each gene group with Htseq (V 0.6.1). FPKM represents "expected number of fragments per kilobase of transcript sequence per millions base pairs sequenced", which is calculated on the basis of the length of the fragments and the reads count mapped to each fragment. Genes with the absolute value of log 2 (fold change) ≥ 1 and the corrected p-values less than 0.05 were assigned as significantly differentially expressed genes (DEGs).