A urea agar base plate was prepared for isolation and identification of urease-producing bacteria, consisting of 1.0 g/L peptone, 5.0 g/L NaCl, 2.0 g/L KH₂PO₄, 0.2% phenol red, 2.0 g/L urea, and 0.1 g/L glucose. Luria-Bertani (LB) broth was used for presentation and pre-culture, comprising 10.0 g/L tryptone, 5.0 g/L yeast extract, and 5.0 g/L NaCl. The mineral salt medium (MSM) contained (NH₄)₂SO₄ at a concentration of 1.0 g/L, KH₂PO₄ at 7.0 g/L, K₂HPO₄ at 3.0 g/L, MgSO₄•7 H₂O at 3.0 g/L, and NaCl at 0.5 g/L (pH 7.0–7.2). For the calcium deposition tests, a mixture of 1,000 mL yeast powder (20.0 g/L), ammonium sulfate (10.0 g/L), and distilled water (pH 8.0) was used, following the NH4-YE culture medium advised by ATCC (Zhao et al., 2019). All culture media were adjusted to pH 7.0 and sterilized at 121°C for 20 min, except for urea and glucose that were filtered through a sterile 0.45 um membrane before being added to the urea agar base plate. All organic solvents and other reagents used in this study were of analytical grade.

The ureolytic bacteria were isolated using the dilution plate and direct streak techniques. In detail, 10 mL of sandstone oil was added to 90 mL of MSM and cultured for 7 d at 30°C with 180 rpm shaking to enrich the bacteria. Then, 1 mL of the gradient-diluted enrichment solution from the MSM culture was coated onto a urea agar base plate and incubated for 24–48 h at 30°C. Colonies that exhibited the ability to hydrolyze urea and change the color of the agar base plate from orange to pink were selected for purification. A pure culture, named strain NS-6, was identified as having high levels of urease activity (Van et al., 2010; Gupta et al., 2022; Leeprasert et al., 2022). A negative control was prepared using E. coli DH5α, a known non-ureolytic bacterium. Strain NS-6 was then identified through morphological and biochemical characteristics, as well as 16S rRNA sequencing. Gram staining, nitrate reduction testing, and determination of other enzyme activities such as oxidase and catalase follow Bergey’s Manual of Determinative Bacteriology (Liu X. G. et al., 2021). The 16S rRNA gene sequence of strain NS-6 was amplified, the amplification of DNA fragments was carried out using the primer pairs including forward primer (CAGAGTTTGATCCTGGCT) and reverse primer (AGGAGGTGATCCAGCCGCA). The PCR amplification program consisted of an initial pre-denaturation step at 94°C for 5 min, followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 60 s, The final extension was performed at 72°C for 10 min. and compared to other sequences in GenBank through the Basic Local Alignment Search Tool (BLAST) to construct a phylogenetic tree using the neighbor-joining method in MEGA 7 software (Yoo et al., 2021; Uyar and Avcı, 2023). The NS-6 strain was stored at −80°C in 30% glycerol (Xu et al., 2022).

Urease activity can be determined by measuring the change in conductivity per unit time. One unit of urease activity is defined as the amount of enzyme that hydrolyzes 1 μmol of urea per mL per minute, as determined by the method described by Yi et al. (2021) In brief, strain NS-6 was inoculated into NB-urea broth (0.3% nutrient broth and 2% urea, pH 8.0) at 30°C, and time-dependent urease activity was measured as follows. A mixture of 13.5 mL of 1.6 mmol/L urea solution and 1.5 mL of bacterial solution was prepared and incubated at 37°C for 5 min. During the cultivation process, the conductivity of the mixture (in mS/cm) was measured using a conductivity meter (DDS-307, Digital Conductivity Meter, China). Each treatment was repeated at least three times.

To investigate and quantify the formation of CaCO₃ by strain NS-6, 3.0 mL of stationary phase bacterial cells (1% v/v) were inoculated into 300 mL liquid NH4-YE medium supplemented with 2% urea and 25 mM CaCl₂. The mixture was incubated at 30°C and 200 rpm for 48 h. After incubation, 30 mL of the culture was sampled every 8 h. The samples were centrifuged at 5,000 rpm for 3 min to remove the supernatant, and the resulting precipitated CaCO₃ was collected. The collected precipitates were then filtered and washed with absolute ethanol and deionized water to eliminate any residual cells and culture components. Subsequently, the precipitates were dried for 24 h at 80°C in a vacuum drying oven and weighed. To further analyze the CaCO₃, it was washed with HCl solution, dried again, and weighed. The quantity of CaCO₃ was determined by calculating the mass difference between the two dryings. The dried CaCO₃ samples were preserved for future studies.

A field emission scanning electron microscope (SEM, FEI Quanta 450, America) was employed to observe the micro-morphology of the mineralized products using an accelerating voltage of 1.5 kV. Concurrently, the sample surface was coated with a conductive gold film to facilitate electron flow and prevent bright spots in the captured images caused by electron aggregation, as the sample is non-conductive. Fourier-transform infrared spectroscopy (Nicolet iS50 FT-IR, Thermo Fisher, United States) was utilized to characterize the functional groups present in the mineralized products. The measured wavenumber range was from 4,000 cm⁻¹ to 500 cm⁻¹ at 25°C (Mohan et al., 2018). Additionally, X-ray diffraction analysis (XRD, Smartlab 9kw, Japan) was performed to determine the characteristic X-ray diffraction peaks of the mineralized products. The phase composition and crystal form of the X-ray diffraction peaks were analyzed using Jade 6 software. The scanning range was set at 5–75°, with a scanning step size of 0.02°. Thermogravimetry and differential scanning calorimetry (TG-DSC, TA Q600, America) were employed to investigate the simultaneous thermal analysis of the precipitates under a nitrogen atmosphere, with a heating rate of 20°C/min within the temperature range of 30°C to 950°C (Yi et al., 2021). All measurements were performed in triplicate.

The response surface method (RSM) was utilized to optimize the critical variables and their interactions in the formation of CaCO₃ by strain NS-6. The independent variables, pH, urea concentration, and calcium ion concentration, were selected based on the results of the single-factor experiment (Supplementary Figure S1). Following the Box–Behnken central experimental design principle, a three-factor, three-level response surface analysis was conducted using a single-factor experimental design approach, with the optimal value as the central point and values above and below as the response surface experimental design levels (Wu et al., 2023). The experimental design, consisting of 17 treatments and 3 variables, was generated using Design-Expert V 10.0 software (Supplementary Tables S1, S2), with three replicates at the midpoint. By utilizing equation (1), a quadratic polynomial regression equation between the factor levels and response was derived.

In this study, Y_i was used to represent the predicted response, while X_i and X_j were variables. The constant b₀ was denoted as a constant term, b_i referred to a linear coefficient, b_ij indicated an interaction coefficient, and b_ii represented a quadratic coefficient. Analysis of variance (ANOVA) was conducted to assess the statistical significance. The experiments were performed in 50 mL centrifuge tubes with transparent polypropylene material and a pointed bottom. The quantity of CaCO₃ was determined using the gravimetric method. The calculation formula used was as follows:

In this context, “M₁” represents the amount of CaCO₃ generated by strain NS-6, while “M₂” stands for the quantity resulting from the abiotic treatments.

High-quality genomic DNA from strain NS-6 was extracted using the Bacterial Genome Extraction kit (Tiangen, Beijing), following the manufacturer’s instructions. Subsequently, the obtained DNA was sent to Shanghai Meiji Bioinformatics Technology Co., Ltd. for whole-genome sequencing, with the aim of understanding the genetic-level molecular mechanisms underlying CaCO₃ formation. Before being assembled into a contig using the hierarchical genome assembly technique (HGAP) version 2.2 of Canu, all clean reads underwent filtration and quality control based on the method described by Koren et al. (2017). CDS prediction, tRNA prediction, and rRNA prediction were performed using Glimmer, tRNA-scan-SE, and Barrnap, respectively, as described by Delcher et al. (2007). The predicted CDSs were annotated using sequence alignment techniques against the NR, Swiss-Prot, Pfam, GO, COG, and KEGG databases. Quick alignment of each set of query proteins with the databases allowed retrieval of gene annotations for the top matches (e-value <10⁻⁵). To analyze the genome of strain NS-6, the IslandViewer4 online website (Bertelli et al., 2022) was utilized, employing the IslandPick, IslandPath-DIMOB, and SIGI-HMM methods.

Numerous bacterial strains exhibiting morphological variations were observed to undergo a color change from orange to pink when inoculated on a urea agar base plate. Among these strains, a particular strain, designated NS-6, displayed a deep pink color due to its high urease activity. This strain was subsequently screened and identified. Following incubation on a urea agar base plate at 30°C for 48 h, individual colonies of the NS-6 strain exhibited a spherical shape with conical protrusions and ragged edges. Microscopic analysis revealed that the cell size was approximately 2.0 μm in diameter and rod-shaped (Figures 1A,B). The NS-6 strain exhibited positive activity in oxidase, catalase, nitrate reduction, and sugar fermentation tests shown in Table 1. Phylogenetic analysis further indicated that strain NS-6 shared a 99.8% homology with Neobacillus mesonae GCF-001636315.1 (Figure 1C). Additionally, Figure 1D demonstrated the changes in urease activity and CaCO₃ formation over time for strain NS-6, which increased steadily and reached maximum values of 7.9 mmol/L and 184 mg (4.60 mg/mL) at 32 h of cultivation. Moreover, the deposition of CaCO₃ was found to be quantitatively enhanced under conditions involving a calcium chloride concentration of 0.4–0.6 mmol/L, urea concentration of 0.8–1.0 mmol/L, temperature of 30°C, and pH range of 7–9, as depicted in Supplementary Figure S1. Those findings aligned with the optimal temperature and pH values reported in the earlier research works for the other urease-producing bacteria, and strain NS-6 exhibited significantly high urease activity at 32 h of cultivation compared to previously reported urease-producing bacteria (Qian et al., 2021; Song et al., 2022; Diez-Marulanda and Brandao, 2023).

The formation of precipitated CaCO₃ by strain NS-6 was investigated using NH4-YE culture medium, along with a negative control. The results showed that the CaCO₃ precipitation by NS-6 exhibited irregular spherical compact large particles with a dense filling (Figure 2A). EDS analysis revealed the presence of Na, Cl, and other elements in addition to C, Ca, and O in the CaCO₃ precipitate, indicating the presence of CaCO₃ crystals (Figure 2B). XRD spectra confirmed the results of FESEM imaging, showing that the CaCO₃ precipitation with the addition of strain NS-6 mainly consisted of calcite and vaterite (Figure 2C). Simultaneously, an analysis of calcite and spherulite content using XRD was conducted, and calculated their respective yield proportions by measuring the peak areas and masses of the respective minerals in the samples and underwent rigorous statistical analysis to ensure precision. The results revealed that calcite accounted for 31.7% in the sample, while spherulite constituted of 68.3%. Furthermore, the FT-IR spectra of the precipitated CaCO₃ showed characteristic diffraction peaks at 710 cm⁻¹ and 872 cm⁻¹, which corresponded to the characteristic diffraction peaks of calcite. The characteristic diffraction peaks at 1,070 cm⁻¹ and 868 cm⁻¹ mainly corresponded to the spherulite characteristic diffraction peaks. Notably, the characteristic diffraction peak at 1650 cm⁻¹ indicated the presence of a symmetric stretching of the carboxyl group (-COO-) in the protein secreted by strain NS-6 compared to the pure water system forming CaCO₃ (Figure 3A). Thermogravimetric analysis revealed that the thermogravimetric curve of CaCO₃ formed in the pure water system had a weight loss phase, with a total weight loss of 43.41% in the temperature range of 500–780°C, mainly due to the decomposition of CaCO₃. On the other hand, the CaCO₃ precipitated by strain NS-6 showed two weight loss stages on the TG curve (Figure 3B). The first weight loss phase occurred between 50 and 500°C, with a weight loss of 6.51%, attributed to the combustion and decomposition of bacterially secreted proteins adsorbed on the precipitated CaCO₃. The second weight loss phase of 36.90% occurred in the range of 500–780°C, which was attributed to the decomposition of CaCO₃.

In this study, a Box–Behnken design based on response surface methodology (RSM) was employed to identify the key factors and their interactions that influence the induction of CaCO₃ precipitation by strain NS-6. The results revealed that the pH (A), urea concentration (B), and calcium ion concentration (C) were the critical factors responsible for inducing CaCO₃ precipitation, with a range of 130 mg to 196 mg. The corresponding precipitation rate ranged from 3.25 mg/mL to 4.90 mg/mL (shown in Supplementary Table S2). To model the quantity of CaCO₃ precipitation induced by strain NS-6, a quadratic binomial regression equation (3) was fitted.

The ANOVA analysis of the quadratic polynomial model was presented in Table 2. The results show that the fitted terms were not statistically significant (p = 0.8192 > 0.05). However, the model had a high R² value of 0.9984 and R²_Adj = 0.9964, indicating that the quadratic model accurately represents the relationship between the response and variables. Moreover, the results suggest that factors like pH (A) and calcium ion concentration (C), as well as the interaction term A × C and squared terms A², B², and C², had significant effects on the formation rate of precipitation (p < 0.05) (Karimifard and Alavi, 2019). To further enhance our understanding of the results, a three-dimensional response surface was generated (Figure 4), which illustrates the impact of pH (A), urea concentration (B), and calcium ion concentration (C) on the quantity of CaCO₃ precipitation while keeping other independent variables constant. From the 3D plot, it can be concluded that the variation in the surface due to pH (A) × calcium ion concentration (C) is greater than that due to pH (A) × urea concentration (B) and urea concentration (B) × calcium ion concentration (C) within the selected range of factors. The model predicts that a maximum amount of 193.8 mg (4.845 mg/mL) of CaCO₃ precipitation can be achieved at pH 8.0, urea concentration of 0.9 mmol/L, and calcium ion concentration of 0.5 mmol/L. The reliability of the polynomial model equation is assessed through the R² value, and its statistical significance is evaluated using the F-value. The value of ps of the model coefficients test the significance of the linear and squared effects of the influencing factors and their interaction effects.

Aiming to better understand the genetic characteristics of strain NS-6, a genome-wide analysis was performed to reveal functional genes involved in the mineralization process. The genome of strain NS-6 consisted of a single chromosome spanning 5,736,360 base pairs (bp) with an average GC content of 40.32 mol% (Figure 5). Additionally, the genome contained 41 rRNAs and 110 tRNAs, and 13 genomic islands (Gls) were identified in strain NS-6 (Table 3 and Supplementary Figure S2). In terms of coding genes, there were a total of 2,676, 4,518, and 2,914 genes annotated for KEGG, COG, and GO databases, respectively (Figure 6). Notably, the genome analysis of strain NS-6 revealed the presence of urease-producing genes, namely ureA, ureB, and ureC, which encoded the γ subunit, β subunit, and α subunit of urease respectively, as well as genes encoded the urease accessory proteins UreD, UreE, UreF, and UreG (Table 4). This indicates that strain NS-6 is a typical urease-producing bacterium with a trimeric structure composed of two identical monomers, representative of urease enzymes, which was in agreement with earlier research works (Kappaun et al., 2018; Moro et al., 2022). These data through comprehensive whole-genome sequencing and functional annotation pave the way for future rational design of synthetic precipitator strains optimized for specific applications. The genome sequence of strain NS-6 has been submitted to the GeneBank database under accession number CP128196.1.