Differential Immobilization of Pb²⁺ and Cd²⁺ by Marine Bacillus velezensis Hao 2023: Mechanisms and Fermentation Optimization for Enhanced Exopolysaccharide Production

Abstract

Soil contamination with lead (Pb) and cadmium (Cd) poses a severe threat to agricultural safety. This study explored the marine bacterium Bacillus velezensis Hao 2023 for bioremediation potential and EPS yield enhancement. Soil filtrate tests under metal stress revealed high tolerance to Pb²⁺ (250 mg/L) and Cd²⁺ (2.5 mg/L), with distinct mechanisms: Cd²⁺ removal was strongly correlated with significant pH increase (up to 8.10), suggesting that immobilization likely occurred through precipitation, while Pb²⁺ was sequestered via EPS synthesis, achieving a yield of 1.62 g/L under stress. To decouple production from metal stress, fermentation was optimized using single-factor and response surface methodology. Key conditions (sucrose, ammonium sulfate, 45 g/L sea salt, 35 °C, pH 6.0, 8% inoculum, 150 rpm) achieved 1.081 g/L EPS under metal-free conditions. These results demonstrate strain Hao 2023’s metal-specific resistance and provide a scalable process for soil remediation agent development.

1. Introduction

Heavy metal pollution in farmland soils has increased concurrently with mining and smelting activities and the misuse of pesticides and chemical fertilizers [1]. The soil is a long-term sink for the group of potentially toxic elements often referred to as heavy metals, including zinc (Zn), copper (Cu), nickel (Ni), lead (Pb), chromium (Cr) and cadmium (Cd) [2]. Among them, cadmium (Cd) and lead (Pb) are potentially toxic elements that are non-biodegradable and highly persistent in soils [3]. Conventional physicochemical remediation methods (e.g., soil replacement, chemical leaching) are often limited by high cost, structural soil damage, and secondary pollution risks [4]. Thus, developing efficient, eco-friendly in situ bioremediation technologies has become an urgent priority for sustainable environmental management [5].

Among various bioremediation strategies, approaches utilizing microorganisms and their metabolites demonstrate unique advantages. Natural biosorbents, including microbial cells and their by-products, have emerged as promising solutions [6]. In particular, microbial extracellular polymeric substances (EPSs)—natural high-molecular-weight polymers secreted by bacteria, fungi, and microalgae into the extracellular environment—play a critical role in the passivation and immobilization of heavy metals. Synthesized by microorganisms under environmental stress, EPS is a complex mixture of macromolecules such as proteins, polysaccharides, nucleic acids, and humic acids [7]. Its anionic functional groups enable efficient electrostatic binding with cationic heavy metals, making EPS an effective biosorbent for soil remediation [8]. Numerous studies indicate that the adsorption capacity of EPS for heavy metal ions often exceeds that of the microbial cells themselves, underscoring its central role in heavy metal removal [9].

Beyond direct immobilization of heavy metals, EPS exerts multiple indirect beneficial effects in soil ecosystems [10]. For example, EPS promotes the formation and stabilization of soil aggregates via hydrogen bonding and ionic bridging, thereby improving soil pore structure, water retention, and nutrient-holding capacity [11]. As an important carbon source, it can stimulate rhizospheric microbial activity and enhance nutrient cycling [12]. Studies have also shown that exogenous EPS addition can effectively alleviate heavy metal toxicity in plants [13].

Under stress conditions, including heavy metal exposure, microbial EPS production is often enhanced, serving as a key adaptive response [14]. For instance, EPS from Pseudomonas aeruginosa OMCS-1 exhibited significantly higher removal efficiency for Cr, Pb, and Cd compared to EPS-removed cells [14], while marine diatom EPS showed optimized Pb²⁺ adsorption through parameter adjustment [15]. This stress-induced EPS synthesis underscores its role as a dynamic defense mechanism, broadening its application potential in heavy metal remediation [16].

However, translating this potential into practical applications faces a major challenge: the native yield of EPS from functional strains under standard conditions is often insufficient for large-scale use [17,18]. This is particularly relevant for microbes from extreme environments, such as marine systems, which may produce EPS with unique properties but require optimized cultivation [19]. Thus, systematic fermentation optimization is critical to enhance yield and bridge the lab-to-field gap [20]. For example, RSM-based optimization for Stenotrophomonas species significantly boosted EPS output by tuning parameters like temperature and pH [21].

To address this, our study focuses on Bacillus velezensis Hao 2023, a marine bacterium with high Pb and Cd tolerance and inducible EPS synthesis. Furthermore, the efficacy of adsorption processes, whether utilizing microbial cells or their metabolites, is governed by a complex interplay of physicochemical factors. These include the specific surface area and functional groups of the adsorbent, the nature and concentration of the heavy metal ions (adsorbate), solution pH, temperature, and contact time [22]. A comprehensive understanding of these parameters is crucial for optimizing bioremediation strategies and translating laboratory findings into field applications. For instance, the solution pH significantly influences the ionization state of functional groups on both the microbial cell wall and EPS, thereby affecting the electrostatic interactions with metal cations [22]. This study, while focusing on the inherent capabilities of Bacillus velezensis Hao 2023, systematically investigates the impact of key variables such as metal concentration and cultivation time on adsorption kinetics, thereby establishing a foundational framework for optimizing its application in heavy metal-contaminated soil remediation. Soil filtrate tests revealed distinct immobilization mechanisms: Cd²⁺ removal was strongly correlated with pH-induced precipitation, whereas Pb²⁺ was sequestered primarily via EPS. To overcome its low native yield, we employed single-factor and response surface methodology to optimize fermentation conditions, aiming to establish a scalable process for efficient bioremediation [23].

2. Materials and Methods

2.1. Bacterial Strain, Soil, and Culture Media

A marine bacterial strain was isolated from an abalone seedling collection plate in Rongcheng, Shandong Province, China. It was identified as Bacillus velezensis and deposited in the laboratory collection under the designation Bacillus velezensis Hao 2023 (NCBI accession number: CP163446). The strain is preserved in our laboratory at −80 °C in 25% (v/v) glycerol. Soil samples were collected from the greenhouse of Qilu University of Technology (Shandong Academy of Sciences), located in the Guyunhu Subdistrict, Changqing District, Jinan City, Shandong Province, China (approximately 36.65° N, 116.89° E). The greenhouse soil had low inherent heavy metal content, and blank controls were included to account for any background interference. For bacterial revival and seed culture preparation, Zobell 2216E liquid medium (a marine salt medium) was used, containing (per liter): peptone, 5 g (Liangshan Ketai Biological Products Co., Ltd., Jining, China); yeast extract, 1 g (Beijing Aoboxing Bio-Tech Co., Ltd., Beijing, China); marine salt, 35 g (Tianjin Yidali Commercial & Trading Co., Ltd., Tianjin, China); pH 7.6–7.8. The basal fermentation medium contained (per liter): glucose, 4.5 g (Tianjin Kemeng Chemical Industry & Trade Co., Ltd., Tianjin, China); ammonium sulfate [(NH₄)₂SO₄], 2.5 g (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China); marine salt, 35 g (Tianjin Yidali Commercial & Trading Co., Ltd., Tianjin, China); pH 8.0–9.0.

2.2. Determination of Heavy Metal (Cd, Pb) Tolerance in Strain Hao 2023

This study primarily assessed the tolerance of strain Hao 2023 to the heavy metals lead (Pb) and cadmium (Cd). To evaluate heavy metal tolerance, a seed medium was used as the base. Stress systems containing Cd²⁺ at concentrations of 0, 0.5, 1, 1.5, 2, and 2.5 mg/L, and Pb²⁺ at concentrations of 0, 50, 100, 150, 200, and 250 mg/L were prepared via the aseptic addition of respective metal stock solutions.

A mid-log-phase bacterial culture was centrifuged under aseptic conditions, and the supernatant was discarded. The cell pellet was resuspended in sterile phosphate-buffered saline (PBS) to an initial optical density (OD) of 1.0 at 600 nm, ensuring consistent inoculum biomass across all treatment groups. This bacterial suspension was then inoculated into the heavy metal-supplemented media at a ratio of 8% (v/v). The final working volume in each Erlenmeyer flask was 100 mL. Each treatment was performed in triplicate. Cultures were incubated in an orbital shaker at 37 °C with constant shaking at 180 rpm.

During cultivation, 200 μL aliquots of culture were sampled hourly and transferred into sterile 96-well UV plates. The dynamic change in optical density at 600 nm (OD₆₀₀) was monitored using a multifunctional microplate reader (Shanghai Metash Instruments Co., Ltd., Shanghai, China).

2.3. Determination of Cd²⁺ and Pb²⁺ Concentrations in Soil Filtrate and EPS Yield After Inoculation with Strain Hao 2023

Sieved, air-dried soil samples were mixed with ultrapure water at a 1:4 (m/v) ratio. The suspension was horizontally shaken at 25 °C and 150 rpm for 48 h. Subsequently, the mixture was aliquoted into 50 mL centrifuge tubes and centrifuged at 5000 rpm for 15 min at 4 °C. The supernatant was collected and sequentially filtered through a sterile filter membrane and a 0.45 µm membrane to obtain the final soil filtrate.

The fermentation medium was mixed with the soil filtrate at a 1:4 ratio. Graded concentrations of heavy metals were added to create stress systems: Cd²⁺ at 0, 0.5, 1, 1.5, 2, and 2.5 mg/L, and Pb²⁺ at 0, 50, 100, 150, 200, and 250 mg/L. A bacterial suspension (1 × 10⁸ CFU/mL, calibrated by serial dilution with physiological saline) was inoculated at 8% (v/v). An uninoculated control was included for each treatment, with all groups prepared in triplicate. Cultures were incubated in a shaker at 25 °C and 180 rpm for 96 h.

Samples were collected at predetermined time points (0, 24, 48, 72, and 96 h). For analysis, 5 mL of culture was centrifuged at 10,000 rpm for 10 min. The supernatant was filtered through a 0.45 µm membrane, and the Cd²⁺ and Pb²⁺ concentration was determined by ICP-MS. The pH of the fermentation broth was measured using a precision pH meter. Additionally, 1 mL of the sample was used to determine the polysaccharide (EPS) content.

The yield of extracellular polymeric substances (EPS) produced by the marine bacterium was determined using the phenol–sulfuric acid method. The detailed procedure was as follows: A standard stock solution (0.04 mg mL⁻¹) was prepared by accurately weighing 20 mg of glucose and bringing it to a final volume of 500 mL in a volumetric flask. A standard curve was established by aliquoting 0, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, and 1.8 mL of this stock solution into separate tubes. Deionized water was added to each tube to achieve a final volume of 2.0 mL for the reaction system. Subsequently, 1 mL of freshly prepared 6% (v/v) phenol solution and 5 mL of concentrated sulfuric acid (guaranteed reagent grade) were sequentially added to each tube. After standing for 10 min, the mixtures were vortexed for 30 s and then incubated at room temperature in the dark for 20 min to allow for color development. The absorbance of each standard was measured at 490 nm using a spectrophotometer to construct the calibration curve.

For EPS extraction from fermentation broth, the culture was first centrifuged at 5000 rpm for 20 min to remove bacterial cells. The collected supernatant was mixed with 95% ethanol (analytical grade) at a 1:4 (supernatant: ethanol) volume ratio. The mixture was vortexed thoroughly and subjected to ethanol precipitation at 4 °C for 12 h. The resulting precipitate was collected by centrifugation, redissolved in an appropriate volume of ultrapure water to obtain the test solution, and then diluted if necessary. A 1 mL aliquot of the sample solution was taken, and the colorimetric reaction was performed following the standard curve procedure described above. The polysaccharide content in the sample was finally calculated based on the standard curve.

2.4. Screening and Optimization of Culture Medium for Strain Hao 2023

Hao 2023 was first activated in seed medium (30 °C, 150 rpm) for 12 h. To screen optimal components, five carbon sources (glucose, maltose, lactose, sucrose, glycerol), five nitrogen sources (beef extract, peptone, yeast extract, urea, ammonium sulfate), and five marine salt concentrations (2.5%, 3%, 3.5%, 4%, 4.5%) were tested. Each component was used to equivalently replace the corresponding constituent in the basal fermentation medium. The initial cultivation conditions were: 500 mL Erlenmeyer flasks with a 20% working volume, 2% inoculation volume, 150 rpm, 30 °C, for 48 h. EPS yield was determined by the phenol–sulfuric acid method to identify the optimal carbon source, nitrogen source, and marine salt concentration.

Response Surface Methodology (RSM) was employed using Design-Expert software (version 10.0.1) to optimize the fermentation medium components. An experimental design was created with one low and one high level for each variable. All trials were performed in triplicate. The EPS content in the fermentation broth of strain Hao 2023 served as the response value for the optimization.

2.5. Optimization of Fermentation Conditions for Strain Hao 2023

2.5.1. Single-Factor Experiment Design

Using the optimized medium composition, batch cultures were established under the initial cultivation conditions. A series of single-factor experiments was sequentially conducted to examine the effects of five different parameters on EPS yield in the fermentation broth of Hao 2023. The tested factors and their levels were: fermentation temperature (20, 25, 30, 35, 40 °C), initial pH (5, 6, 7, 8, 9), inoculation volume (2%, 5%, 8%, 11%, 14%), cultivation time (12, 24, 36, 48, 60 h), and shaker speed (100, 150, 200, 250 rpm). The results were used to determine the approximate control ranges for subsequent fermentation condition optimization.

2.5.2. Plackett–Burman (PB) Experiment

Based on the results of the single-factor experiments, a Plackett–Burman (PB) experimental design was generated using Design-Expert software (version 10.0.1). Each tested variable was assigned a low (−1) and a high (+1) level. All trials were performed in triplicate. The EPS content in the fermentation broth of strain Hao 2023 served as the response value to screen for factors exerting a significant effect on EPS production. The specific design is presented in Supplementary Table S1. The significant factors identified from the PB experiment were then used, along with their corresponding concentration ranges, to design a subsequent Box–Behnken Design (BBD) experiment. Data from the BBD were analyzed and processed using Design-Expert software (version 10.0.1).

2.5.3. Box–Behnken Design (BBD) Experiment

Based on the significant variables and their concentration ranges identified from the Plackett–Burman (PB) experiment, a Box–Behnken Design (BBD) was formulated. The factor levels and the experimental design are detailed in Supplementary Table S2. The experimental results were analyzed and processed using Design-Expert software (version 10.0.1).

2.6. Statistical Analysis

Data were processed using one-way analysis of variance (one-way ANOVA) followed by Tukey’s test (p < 0.05) for multiple comparisons of treatment means, and the results are expressed as mean ± standard deviation. All statistical analyses were performed with SPSS software (version 20.0; SPSS Inc., Chicago, IL, USA). Graphs were generated using Origin 8.5, and response surface plots were exported from Design-Expert software (version 10.0.1).

3. Results

3.1. Analysis of Strain Hao 2023 Tolerance to Cd²⁺ and Pb²⁺

The growth dynamics of strain Hao 2023 under varying Cd²⁺ concentrations clearly reflect its tolerance characteristics towards this heavy metal. As shown in Figure 1A, in the control group without Cd²⁺ stress, the initial OD600 of the cultivation system was approximately 0.08 after inoculation, and it increased to 0.587 at 12 h, indicating unimpaired normal growth. As the Cd²⁺ concentration increased, the strain exhibited a significant adaptive response. At 0.5 mg/L Cd²⁺, the early growth phase was similar to the control, but the OD₆₀₀ decreased significantly at 5 h before gradually recovering, indicating mid-phase inhibition followed by partial recovery. At 1 mg/L Cd²⁺, the growth curve was highly similar to the control, with an OD₆₀₀ of 0.570 at 12 h, nearly matching the control level and even slightly exceeding it at some time points, showing no significant inhibition. At 1.5 mg/L Cd²⁺, the initial OD₆₀₀ was significantly lower than the control, but growth resumed after 6 h, reaching an OD₆₀₀ of 0.576 at 12 h, suggesting considerable adaptive capacity. At 2 mg/L Cd²⁺, the OD₆₀₀ was 0.531 at 12 h, lower than the control but higher than the 0.5 mg/L group, and approached the control level in the later growth phase, demonstrating a degree of tolerance. At 2.5 mg/L Cd²⁺, growth was slightly inhibited (OD₆₀₀ decreased by approximately 8.6% compared to the control) but remained stable without complete cessation, indicating a relatively high tolerance threshold. Notably, Hao 2023 displayed growth recovery capability in the presence of Cd²⁺, suggesting the potential presence of Cd²⁺ detoxification or adsorption mechanisms. In summary, strain Hao 2023 exhibits significant tolerance to Cd²⁺, demonstrating its potential for application in the remediation of cadmium-contaminated soils.

The growth dynamics of strain Hao 2023 under varying Pb²⁺ concentrations clearly illustrate its tolerance to this heavy metal. As shown in Figure 1B, in the control group without Pb²⁺ stress, the growth curve showed a sustained increase, with a mean OD₆₀₀ of 0.5876 at 12 h, confirming normal, uninhibited growth. As the Pb²⁺ concentration increased, the strain demonstrated a notable adaptive response. At 50 mg/L Pb²⁺, early growth was comparable to the control, but the OD₆₀₀ decreased significantly at 8 h before a gradual recovery, indicating mid-phase inhibition followed by partial restoration. At 100 mg/L Pb²⁺, the OD₆₀₀ was 0.5097 at 5 h, and the growth curve was highly similar to the control, even slightly exceeding it at certain time points, showing no obvious inhibition. At 150 mg/L Pb²⁺, the OD₆₀₀ was 0.3647 at 5 h, significantly lower than the control and lower-concentration groups; 5 h also marked a turning point, after which growth gradually recovered to a pattern similar to the control. When the Pb²⁺ concentration increased to 200 mg/L, the initial OD₆₀₀ showed no significant decrease, and even reached 0.5669 at 9 h, exceeding the control level. Growth slowed after 10 h, demonstrating a degree of tolerance with only slight, non-significant inhibition in the later phase. At 250 mg/L Pb²⁺, growth was mildly inhibited (decreasing approximately 7.8% compared to the control) but did not completely cease. The extent of inhibition was even slightly less than that observed at 150 mg/L (which showed an ~8.3% decrease compared to the control), suggesting that high concentrations of Pb²⁺ may rapidly induce efflux systems and EPS secretion, thereby reducing intracellular toxic load, and indicating the presence of Pb²⁺ detoxification or adsorption mechanisms.

3.2. Determination of pH and EPS Concentration in Cadmium-Containing Soil Filtrate After Inoculation with Strain Hao 2023

Soil filtrate was used to simulate the field soil environment. The pH of the fermentation broth was measured at different time intervals over 0–96 h. As shown in Figure 2A, the pH in the soil filtrate without Cd²⁺ stress continuously decreased. In the treatment groups with Cd²⁺ concentrations of 0.5 and 1 mg/L, the pH was higher than that of the control at 24 h and 48 h but lower at 72 h and 96 h. The pH in all other treatment groups was elevated to varying degrees compared to the control at every measured time point. Notably, at a Cd²⁺ concentration of 2.5 mg/L, the pH remained alkaline throughout the incubation period, increasing from 6.96 to 8.10 within 24 h and peaking at 8.12 at 72 h. This suggests that the strain efficiently alkalized the environment, which was strongly correlated with Cd²⁺ immobilization and likely contributed to reducing ionic toxicity through precipitation.

As shown in Figure 2B, in experiments with soil filtrate under Cd²⁺ stress, the EPS content produced by all treatment groups increased over time, reaching peak levels at 96 h. The EPS yields at Cd²⁺ concentrations of 0, 0.5, 1, 1.5, 2, and 2.5 mg/L were 0.854, 1.51, 1.39, 1.78, 2.11, and 2.13 g/L, respectively. In the absence of Cd²⁺ stress (0 mg/L), EPS production increased steadily over time, consistent with normal microbial metabolic secretion of EPS. At Cd²⁺ concentrations of 0.5, 1, and 1.5 mg/L, EPS secretion was inhibited during the early experimental stage (24–48 h) compared to the control, indicating that Hao 2023 required time to adapt to the stress environment. In the later stage (72–96 h), EPS production increased significantly, peaking at 96 h, demonstrating that the strain responded to heavy metal stress by enhancing EPS secretion. At Cd²⁺ concentrations of 2 and 2.5 mg/L, the initial inhibitory effect was less pronounced compared to other treatment groups, and EPS secretion increased further in the later stage, highlighting the strain’s strong tolerance and remediation potential under high Cd²⁺ concentrations.

3.3. Determination of pH and EPS Concentration in Lead-Containing Soil Filtrate After Inoculation with Strain Hao 2023

Under Pb²⁺ stress, the pH modulation by strain Hao 2023 exhibited concentration dependence and temporal dynamics, with an overall declining trend after 48 h. The pH in the lead-free control group (0 mg/L) decreased continuously from an initial 6.08 to 4.52 at 96 h, indicating the acid-producing nature of the strain’s basal metabolism (Figure 3A). In the low-concentration lead groups (50–150 mg/L), a transient pH increase was observed at 24–48 h (e.g., reaching 7.05 at 48 h in the 50 mg/L group), likely resulting from the secretion of alkaline substances (e.g., NH₃/OH⁻) to counteract Pb²⁺ toxicity. However, the pH dropped to levels near the control after 72 h, suggesting a high metabolic cost associated with this alkalinization strategy. At 200 mg/L Pb²⁺, the pH trend was similar to the control before 48 h, after which it declined along with the other treatment groups. Notably, at 250 mg/L, the pH remained alkaline before 48 h, peaked at 7.86 at 48 h, and then decreased significantly. This may be attributed to competitive exchange between soluble Pb²⁺ and exchangeable H⁺ on soil particle surfaces, leading to a release of H⁺ and subsequent pH reduction.

In experiments with soil filtrate under Pb²⁺ stress, the EPS content produced by all treatment groups increased over time, revealing a core detoxification strategy of strain Hao 2023 (Figure 3B). The EPS yields for Pb²⁺ concentrations of 0, 50, 100, and 150 mg/L all peaked at 72 h, reaching 0.94, 0.98, 1.15, and 1.49 g/L, respectively. Overall, EPS production in the control group increased slowly, consistent with its basal metabolic pattern. Compared to the control, EPS secretion was inhibited in the 100 mg/L group at 0–24 h, and in the 150 mg/L and 200 mg/L groups at 24–48 h (being 50.15% and 32.29% lower than the control, respectively), indicating that Pb²⁺ initially suppressed extracellular polymeric substance synthesis. Notably, EPS production at 200 mg/L and 250 mg/L peaked later at 72 h, reaching 1.39 g/L and 1.62 g/L, respectively. This confirms that the strain employs a delayed EPS production surge to chelate Pb²⁺. Furthermore, the high-concentration lead groups (200–250 mg/L) showed no initial inhibition period in EPS secretion, instead displaying a consistent linear increase. This suggests that under high-stress conditions, the strain rapidly activates EPS synthesis pathways. The subsequent stabilization of the pH in the acidic range from 48 to 96 h may further enhance the affinity of EPS carboxyl groups for Pb²⁺, facilitating its immobilization and contributing to tolerance. Additionally, the later EPS peak (96 h) compared to the Cd²⁺ experiment (72 h) indicates that Pb²⁺ causes a more prolonged disruption to the strain’s metabolic network, requiring a longer time to mount a remediation response.

3.4. Effect of Strain Hao 2023 on Cd²⁺ and Pb²⁺ Concentrations in Soil Filtrate

The experimental results are shown in Figure 4A. The overall trend across all treatment groups was a continuous decrease in solution-phase Cd²⁺ concentration over time. In the treatment with no added Cd²⁺, the background Cd²⁺ concentration in the soil filtrate significantly decreased, dropping to 0.025 mg/L by 96 h. This suggests that Hao 2023 may have adsorbed or immobilized the native soil Cd²⁺, or reduced its availability by altering soil pH or redox conditions. When 0.5 mg/L of Cd²⁺ was added, the concentration showed a sharp initial drop within 48 h, followed by a partial rebound after 72 h. This indicates rapid adsorption of Cd²⁺ by the strain in the early stage, with potential subsequent redissolution due to the release of metabolic byproducts or microbial cell lysis. At Cd²⁺ levels of 1 mg/L and 1.5 mg/L, the concentration did not decrease significantly after 48 h and showed notable fluctuations. For concentrations of 2 mg/L and 2.5 mg/L, the Cd²⁺ levels decreased in the initial phase (0–48 h), but the immobilization efficiency declined in the later phase (72–96 h).

In the experiment with soil filtrate under Pb²⁺ stress, the Pb²⁺ concentration in all treatment groups exhibited a declining trend over time (Figure 4B). Without exogenous Pb²⁺ addition, the native background Pb²⁺ concentration in the filtrate decreased significantly, reaching 0.4837 mg/L at 96 h. In the 50 mg/L Pb²⁺ treatment, the concentration decreased within 48 h but partially rebounded at 72 h. In the 100 mg/L treatment, the Pb²⁺ concentration dropped notably to 2.3887 mg/L by 72 h, followed by a non-significant increasing tendency. This pattern suggests rapid initial adsorption of Pb²⁺ by the strain, with potential re-release later due to decreased activity or cell death. For Pb²⁺ additions of 150 mg/L, 200 mg/L, and 250 mg/L, the concentrations fluctuated noticeably within the first 48 h, ultimately reaching their lowest soluble levels of 2.121 mg/L, 1.514 mg/L, and 1.083 mg/L, respectively, at 96 h. It is noteworthy that EPS accumulation in the 200 mg/L and 250 mg/L treatment groups also peaked at 72 h. This suggests that EPS likely played a key role in adsorbing Pb²⁺ through mechanisms such as ion exchange, complexation, and precipitation, leading to the significant decrease in Pb²⁺ concentration observed after 72 h.

3.5. Screening and Optimization of Medium Components for EPS Production by Strain Hao 2023

3.5.1. Glucose Standard Curve (Phenol–Sulfuric Acid Method)

The standard curve (Figure 5A) was plotted with glucose mass as the x-axis and absorbance at 490 nm as the y-axis, as illustrated in the figure.

3.5.2. Effect of Different Carbon Sources on EPS Production by Strain Hao 2023

Analysis revealed that the EPS yield of Bacillus velezensis Hao 2023 varied with different carbon sources, in the following descending order: sucrose > lactose > glucose > maltose > glycerol (Figure 5B). Although strain Hao 2023 can utilize various carbon sources, EPS yield was highest (0.4386 g/L) when sucrose was used. Therefore, sucrose was selected as the optimal carbon source. Experiments were then conducted with sucrose concentrations of 30, 35, 40, 45, and 50 g/L. The results (Figure 5C) showed the highest EPS yield at 35 g/L. This optimal concentration is likely because lower sucrose levels are depleted prematurely, limiting later-stage polysaccharide synthesis, while excessively high concentrations lead to vigorous microbial growth, rapidly depleting both the carbon source and dissolved oxygen, which also ultimately reduces polysaccharide yield.

3.5.3. Effect of Different Nitrogen Sources on EPS Production by Strain Hao 2023

This experiment investigated the effect of five organic and inorganic nitrogen sources—yeast extract, beef extract, peptone, urea, and ammonium sulfate—on EPS yield. Preliminary fermentation results indicated that strain Hao 2023 has a broad range of nitrogen source utilization. EPS yields were significantly higher when peptone or ammonium sulfate was used compared to the other three sources (Figure 5D). Peptone contains various amino acids and peptides, which can directly provide nitrogen for EPS synthesis and serve as precursors, explaining the higher EPS yield. Ammonium sulfate, an inorganic nitrogen source, also supported high EPS production. The marine-derived strain Hao 2023 may possess an efficient inorganic nitrogen assimilation mechanism, rapidly converting it into organic nitrogen compounds for growth and metabolism, thereby providing sufficient nitrogen for EPS synthesis and resulting in faster growth and higher EPS yield. Considering polysaccharide yield, ammonium sulfate at a concentration of 2 g/L was selected as the sole nitrogen source for subsequent studies (Figure 5E).

3.5.4. Effect of Marine Salt Concentration on EPS Production by Strain Hao 2023

This study examined the influence of different marine salt concentrations (25, 30, 35, 40, 45, and 50 g/L) on EPS production by strain Hao 2023 (Figure 5F). At 25 g/L, EPS yield was relatively low (approximately 0.1887 g/L). As the concentration increased to 35 g/L, EPS yield rose significantly to about 0.457 g/L. A further increase to 45 g/L resulted in the peak EPS yield of approximately 0.4915 g/L. However, at 50 g/L, the yield slightly decreased but remained relatively high at around 0.45 g/L. Error bars representing standard deviation are shown for each data point, indicating some data dispersion, but the overall trend is clear. In summary, EPS production by strain Hao 2023 is dependent on marine salt concentration, with an optimal concentration of approximately 45 g/L for maximizing yield.

3.5.5. Response Surface Optimization of Basic Medium Components

Based on a Box–Behnken central composite design, a response surface experiment was conducted. Three key medium components affecting EPS yield—sucrose, ammonium sulfate, and marine salt—were selected for a three-factor, three-level response surface analysis. The experimental design and results are presented in Supplementary Table S3, and the corresponding analysis of variance is shown in Supplementary Table S4.

Data analysis using Design-Expert 10.0.3 software yielded the following second-order polynomial regression equation:

Y = 0.5395 + 0.0505A − 0.0026B + 0.0367C − 0.0504AB + 0.0460AC + 0.0061BC − 0.1850A² − 0.2069B² − 0.1962C²

The model regression was significant (p < 0.0001), and the lack of fit was not significant (p = 0.2077 > 0.05). The model’s coefficient of determination (R²) was 0.9852, and the adjusted R² (Adj R²) was 0.9662, indicating a high degree of fit and good predictive capability for the EPS yield of strain Hao 2023 based on the fermentation medium formulation. According to the F-values, the order of influence of the three factors on EPS yield was: A (sucrose) > C (marine salt) > B (ammonium sulfate). The linear and interaction terms were not significant, while the quadratic terms had an extremely significant effect (p < 0.0001).

Using the optimization function in Design-Expert 10.0.1, the interaction response surfaces and contour plots (Figure 6) were analyzed. The optimal conditions were determined as follows: sucrose 3.577%, ammonium sulfate 0.199%, and marine salt 4.556%. Under these conditions, the model predicted an EPS yield of 0.545 g/L. Considering practical feasibility, the conditions were adjusted to sucrose 3.58%, ammonium sulfate 0.2%, and marine salt 4.56%. Three replicate verification experiments under these conditions yielded an average EPS production of 0.5392 g/L, which deviates from the predicted value by only 1.06%. This confirms the validity and feasibility of the response surface methodology for optimizing EPS production.

Figure 6. Response surface plots of the interaction of sucrose, ammonium sulphate and sea salt on EPS yield. (A) Response surface plots for Y = ∫(A, B); (B) Response surface plots for Y = ∫(A, C); (C) Response surface plots for Y = ∫(B, C).

Figure 6. Response surface plots of the interaction of sucrose, ammonium sulphate and sea salt on EPS yield. (A) Response surface plots for Y = ∫(A, B); (B) Response surface plots for Y = ∫(A, C); (C) Response surface plots for Y = ∫(B, C).

3.6. Optimization of Fermentation Conditions for EPS Production by Strain Hao 2023

3.6.1. Single-Factor Experiments

In studies on microbial extracellular polysaccharide (EPS) synthesis, single-factor experiments serve as a fundamental approach for investigating key influencing factors. As shown in Figure 7A, the effect of temperature on EPS yield was evaluated. Within the range of 20 °C to 35 °C, EPS production initially decreased and then increased, reaching a peak at 35 °C. This indicates that the optimal temperature for EPS synthesis by this microorganism falls within this range, likely because temperature regulates the biosynthetic pathway by affecting relevant enzyme activities. Therefore, 35 °C was selected as the most suitable fermentation temperature.

Figure 7B illustrates the influence of pH. As pH increased from 5 to 9, EPS production first increased and then decreased, with the highest yield observed at pH 6. This suggests that a slightly acidic environment is more conducive to EPS synthesis, which may be related to intracellular acid–base balance and enzyme stability in microorganisms.

The effect of inoculum size on EPS yield is presented in Figure 7C. EPS production showed a positive correlation with inoculum size up to 8%, but beyond this threshold, a further increase in inoculum led to a significant decline in EPS yield. This may be attributed to excessive microbial proliferation depleting carbon sources prematurely, thereby impairing EPS synthesis. Consequently, an inoculum size of 8% was chosen for subsequent experiments.

Figure 7D reflects the impact of cultivation time. Within the range of 12 to 48 h, EPS production initially increased and then decreased, peaking at 36 h. This indicates the existence of an optimal cultivation period, which aligns with the microbial growth cycle and the patterns of metabolite accumulation.

The effect of shaker speed is shown in Figure 7E. As the speed increased from 100 rpm to 150 rpm, EPS production increased, but further increases from 150 rpm to 250 rpm resulted in a decline. This suggests that moderate shaking optimizes mass transfer and promotes EPS synthesis, while excessively high speeds may introduce adverse effects such as shear stress. The highest EPS yield was achieved at 150 rpm, and this speed was adopted for subsequent experiments. Together, these single-factor experiments provide a critical foundation for further multi-factor optimization and mechanistic investigations.

3.6.2. Plackett–Burman Experiment

Using polysaccharide content as the evaluation index, a Plackett–Burman (PB) experimental design was created with Design-Expert software (version 10.0.1) software. The results and analysis are presented in

Table 1. Analysis of variance (ANOVA) of factors in the Plackett–Burman design.

Source	Coefficient Estimate	Stdized Effect	Contribution/%	Sum of Squares	Mean Square	F	p	Important Ranking
A—Temperature	0.0763	0.1527	30.8046	0.0699	0.0699	38.72	0.0008	1
B—pH	0.0426	0.0852	9.5973	0.0217	0.0218	12.06	0.0133	3
C—Shaker speed	0.0579	0.1157	17.6913	0.0402	0.0402	22.24	0.0033	2
D—Inoculation volume	−0.0213	−0.0425	2.3901	0.0054	0.0054	3.00	0.1338	5
E—Incubation time	−0.0223	−0.0447	2.6396	0.0060	0.0060	3.32	0.1184	4
Model		0	0	0.1433	0.0287	15.87	0.0021

. The model exhibited an R² of 0.9297 and an adjusted R² (R²adj) of 0.8711, with a significance level of p < 0.01, indicating that the model is reasonably reliable. As shown in Supplementary Table S4, increasing the levels of the variables ‘inoculum size’ and ‘inoculation time’ showed negative effects, which were not statistically significant. In contrast, increasing the levels of the three variables ‘temperature’, ‘pH’, and ‘shaker speed’ demonstrated positive effects, with p-values all less than 0.05. Consequently, the three most significant factors—temperature, pH, and shaker speed—were selected as the influential factors for the subsequent Box–Behnken response surface methodology experiments.

3.6.3. Results of the Box–Behnken Experiment

Based on the results of the single-factor and Plackett–Burman experiments, temperature (A), pH (B), and shaker speed (C) were selected as the independent variables. Extracellular polysaccharide (EPS) yield was used as the response variable. A Box–Behnken response surface methodology experimental design was employed to analyze the effects of these factors. The design and results of the Box–Behnken experiments are presented in

Table 2. Box–Behnken experimental design and results.

Serial Number	A: Temperature/°C	B: pH	C: Shaker Speed/rpm	Y: EPS/g·L−1
1	−1	−1	0	0.617
2	1	−1	0	0.688
3	−1	1	0	0.687
4	1	1	0	0.742
5	−1	0	−1	0.556
6	1	0	−1	0.735
7	−1	0	1	0.715
8	1	0	1	0.767
9	0	−1	−1	0.567
10	0	1	−1	0.646
11	0	−1	1	0.708
12	0	1	1	0.656
13	0	0	0	1.087
14	0	0	0	1.058
15	0	0	0	1.079
16	0	0	0	1.118
17	0	0	0	1.061

.

Analysis of the data from Table 2 produced the analysis of variance (ANOVA) table for the Box–Behnken experiment (

Table 3. Analysis of variance (ANOVA) table for the Box–Behnken design.

Source	Sum of Squares	Df	Mean Square	F-Value	p-Value
Model	0.6317	9	0.0702	97.13	<0.0001	significant
A—Temperature	0.0159	1	0.0159	22.06	0.0022	**
B—pH	0.0029	1	0.0029	4	0.0856	ns
C—Shaker speed	0.0146	1	0.0146	20.18	0.0028	*
AB	0.0001	1	0.0001	0.0887	0.7744	ns
AC	0.004	1	0.004	5.6	0.0498	*
BC	0.0043	1	0.0043	5.93	0.045	*
A2	0.1275	1	0.1275	176.41	<0.0001	***
B2	0.2095	1	0.2095	289.96	<0.0001	***
C2	0.1916	1	0.1916	265.18	<0.0001	***
Residual	0.0051	7	0.0007	-	-	-
Lack of Fit	0.0028	3	0.0009	1.59	0.3247	not significant
Pure Error	0.0023	4	0.0006	-
Cor Total	0.6368	16	-

Note: *, **, and * indicate statistical significance at the p < 0.05, p < 0.01, and p < 0.001 levels, respectively.

). Multiple regression fitting of the data from Table 3 yielded the following second-order polynomial regression equation:

Y = 1.08 + 0.0446A + 0.0190B + 0.0427C − 0.0040AB − 0.0318AC − 0.0327BC − 0.1740A² − 0.2231B² − 0.2133C²

As shown in Table 3 the model was highly significant (p < 0.0001), and the lack of fit was not significant (p = 0.3247 > 0.05). This indicates a high degree of model fit and suggests no other major unconsidered factors significantly influenced the response. The coefficient of determination R² was 0.9921, and the adjusted R² (Adj R²) was 0.9818, confirming the model’s excellent goodness of fit. Furthermore, the full model ANOVA revealed that the quadratic terms A², B², and C² all had p-values less than 0.0001, indicating they significantly affect the response value. This demonstrates that temperature, pH, and shaker speed are important factors influencing EPS yield.

Analysis of the response surface plots revealed a fermentation condition combination corresponding to the peak EPS yield (Figure 8). Software analysis indicated that the optimal fermentation parameters were a temperature of 35.598 °C, a shaker speed of 154.391 rpm, and a pH of 6.035. Under these conditions, the predicted theoretical EPS yield reached 1.085 g/L. Considering practical experimental feasibility, the parameters were adjusted to a temperature of 36 °C, a shaker speed of 155 rpm, and a pH of 6.0. Verification experiments conducted in triplicate under these adjusted conditions yielded an actual EPS production of 1.081 g/L in the fermentation broth. This value closely matched the predicted yield, confirming that the model effectively predicts EPS content under the given fermentation conditions.

Figure 8. Response surface plots of the effect of temperature, pH and shaker speed on EPS yield. (A) Response surface plots of Y = ∫(A, B); (B) Response surface plots of Y = ∫(A, C); (C) Response surface plots of Y = ∫(B, C).

Figure 8. Response surface plots of the effect of temperature, pH and shaker speed on EPS yield. (A) Response surface plots of Y = ∫(A, B); (B) Response surface plots of Y = ∫(A, C); (C) Response surface plots of Y = ∫(B, C).

4. Discussion

4.1. Differential Tolerance Strategies and Mechanisms of Strain Hao 2023 Towards Pb²⁺ and Cd²⁺

This study systematically assessed the tolerance of strain Hao 2023 by monitoring its growth dynamics under varying heavy metal stresses. The results demonstrated significant tolerance to high concentrations of both Cd²⁺ (2.5 mg/L) and Pb²⁺ (250 mg/L). However, the coping strategies employed by the strain differed markedly, reflecting an adaptive response to the distinct toxic mechanisms of these heavy metals.

The growth curves revealed a typical “inhibition-recovery” stress response pattern, which is consistent with the adaptive responses observed in many metal-tolerant bacteria [24]. Initial growth inhibition under stress is primarily attributed to the direct damage caused by heavy metal ions to cellular enzymatic activity and membrane integrity [5,25]. However, with prolonged cultivation, significant growth recovery was observed under medium to high stress concentrations, where the final biomass approached or even exceeded that of the control group. This indicates the strain’s capacity to rapidly initiate effective detoxification mechanisms, a process governed by the complex interplay of key factors in adsorption processes, including the physicochemical properties of the bacterial adsorbent, dynamic changes in solution pH, and the inherent characteristics of the metal ions [22]. For instance, the distinct nature of heavy metal ions (e.g., ionic radius and electronegativity differences between Cd²⁺ and Pb²⁺) significantly affects their affinity to functional groups on the cell surface [22], which may explain the differential stress response patterns observed. This rapid adaptability may involve the synergistic action of multiple resistance systems, such as the potential upregulation of heavy metal efflux pump gene expression alongside the observed induction of EPS synthesis [5,25,26]. The significant increase in EPS production induced by high heavy metal concentrations serves as direct evidence for the strain’s active “external immobilization” strategy, where the biosorption efficiency is ultimately determined by the comprehensive effects of these controlling factors.

Notably, the strain’s detoxification strategies for Cd²⁺ and Pb²⁺ showed significant differentiation. Under Cd²⁺ stress, the strain primarily elevated the environmental pH (up to 8.10), which strongly correlated with Cd²⁺ removal efficiency. Although this study did not directly characterize the solid precipitates formed, the efficient Cd²⁺ removal coupled with a significant pH increase strongly suggests immobilization via microbially induced precipitation (e.g., as Cd(OH)₂ or CdCO₃), a mechanism consistent with studies on microbial enhancement of metal stabilization in amended environments [27].

This “alkalization-precipitation-dominated” immobilization mode is consistent with findings from our team’s previous research on the marine bacterium Pseudoalteromonas agarivorans Hao 2018 [28] and parallels observations reported for other marine bacteria (e.g., certain Bacillus and Halomonas strains) in response to cadmium stress [29,30]. This collectively suggests that alkalizing the microenvironment through metabolic activity may represent a convergent adaptive strategy among phylogenetically diverse marine microorganisms to mitigate cadmium toxicity.

In contrast, under Pb²⁺ stress, the strain tended to maintain a slightly acidic environment (pH ≈ 6.0), with its core detoxification strategy shifting to substantial EPS synthesis (yield up to 1.62 g/L). Functional groups such as carboxyl and hydroxyl groups abundant in EPS molecules can effectively immobilize Pb²⁺ via complexation and surface precipitation [31,32]. It is noteworthy that the immobilization efficiency of EPS for heavy metal ions is significantly regulated by environmental pH. Studies have indicated that the adsorption capacity of EPS for Pb²⁺ generally increases with rising pH, often demonstrating superior immobilization efficiency under alkaline compared to acidic conditions [33]. This phenomenon primarily stems from the pH-dependent dissociation state of anionic functional groups (e.g., carboxyl and phosphate groups) within EPS molecules. Under acidic conditions, a high concentration of H⁺ competes with Pb²⁺ for binding sites, reducing the number of available adsorption sites. A strongly acidic environment may even disrupt the structure of some functional groups in EPS, further impairing its binding capacity [34]. Conversely, under alkaline conditions, OH⁻ can combine with heavy metal ions to form precipitates. Additionally, the hydrolysis speciation of heavy metal ions in solution is pH-dependent, collectively influencing the adsorption process and mechanism of EPS [35,36,37]. These findings suggest that in practical soil remediation applications, adjusting soil pH to a slightly alkaline range may help fully realize the immobilization potential of EPS for heavy metals, thereby enhancing remediation efficiency. Furthermore, the “rebound” phenomenon observed in the growth curves under Pb²⁺ stress (e.g., the late-stage OD value of the 250 mg/L group surpassing the control) might be related to high-concentration ions acting as potent inducing signals, triggering a more rapid and comprehensive activation of defense mechanisms, including EPS synthesis [38,39].

In summary, the detoxification mechanisms of strain Hao 2023 toward Cd²⁺ and Pb²⁺ exhibit clear divergence, with the former primarily relying on environmental alkalization and the latter closely associated with substantial synthesis of EPS. This differentiated tolerance strategy reflects a specific adaptive response of microorganisms to the distinct toxicities of different heavy metals. While the current study has preliminarily elucidated these mechanisms in liquid culture, their practical remediation potential in complex soil environments still requires further validation through subsequent pot experiments.

4.2. Synergistic Immobilization via pH Dynamics and EPS Secretion

In the soil filtrate system, strain Hao 2023’s modulation of the culture environment pH revealed another key strategy for adapting to heavy metal stress. Under Cd²⁺ stress, especially in the high-concentration group (2.5 mg/L), the pH remained alkaline and increased continuously. This phenomenon may be explained by considering that solution pH is recognized as an important factor influencing adsorption efficiency, as it affects the ionization state of functional groups and the speciation of metal ions [22]. The resulting alkaline environment appears conducive to the formation of carbonate or hydroxide precipitates of Cd²⁺, which is consistent with reported adsorption characteristics where pH influences the precipitation tendency of different metal ions [22]. This precipitation-based immobilization mechanism likely contributes to the effective Cd²⁺ removal observed in our experiments [40]. These findings align with the observation that the strain maintained relatively high growth vitality under high cadmium concentrations, suggesting that environmental alkalization combined with precipitation formation may represent one of the core mechanisms for Cd²⁺ immobilization by this marine bacterium.

In contrast, under Pb²⁺ stress, the pH generally decreased to the acidic range after a transient increase. This may be related to the chemical behavior of Pb²⁺: Pb²⁺ competes with soil particles for H⁺ adsorption, temporarily increasing solution pH. However, the strain’s own basal metabolism (e.g., organic acid secretion) ultimately dominated the acid–base balance of the system. It is noteworthy that despite the acidic environment, the adsorption efficiency of EPS for Pb²⁺ remained significant in the later stages (72–96 h). This suggests the potential affinity of EPS functional groups (particularly carboxyl groups as supported by literature [41,42] for Pb²⁺ over a wide pH range. However, the specific binding mechanisms require further verification by spectroscopic techniques such as FTIR or NMR. The delayed peak in EPS production at high Pb²⁺ concentrations (200 and 250 mg/L), reaching 1.39 g/L and 1.62 g/L at 72 h, underscores the strain’s adaptive response to severe mental stress. This delay may reflect the time required to upregulate efflux pumps or cellular repair mechanisms before activating EPS synthesis, as reported in studies on microbial heavy metal resistance [14,43]. The high EPS yields further demonstrate the strain’s capacity to leverage polymeric substances for Pb²⁺ immobilization under prolonged stress, aligning with its role in enhancing metal adsorption efficiency [44,45].

4.3. Immobilization Potential and Stability of Hao 2023 for Cd²⁺ and Pb²⁺ in Soil Filtrate

The remediation potential of strain Hao 2023 was evidenced by the consistent decrease in both background and spiked Cd²⁺ and Pb²⁺ concentrations in soil filtrate throughout the cultivation period, demonstrating its capacity to immobilize heavy metals within complex soil matrices. However, the concentration rebound observed during later stages in some treatments (e.g., Cd²⁺ at 0.5 mg/L) suggests a dynamic adsorption–desorption equilibrium in the bio-immobilization process. This dynamic behavior may be understood through the lens of key operational parameters in adsorption processes, where contact time and adsorbent dosage significantly influence the attainment of equilibrium states [22]. The rapid initial decrease likely results from active adsorption onto living cell surfaces and complexation by extracellular polymeric substances (EPS). Subsequent partial metal re-solubilization could be attributed to cellular lysis or the release of weakly bound metals due to fluctuating environmental conditions (e.g., pH variations) [46,47]. Consequently, maintaining robust microbial activity and optimal environmental conditions emerges as crucial for ensuring stable long-term remediation efficacy in practical applications. This observed adsorption–desorption dynamic, previously documented for other biosorbents in complex systems [48], underscores a significant practical consideration for achieving reliable long-term immobilization.

Particularly noteworthy is the strong temporal correlation between the minimum soluble Pb²⁺ content and the peak EPS yield (both occurring at 72–96 h) in the high-concentration Pb²⁺ treatments (200, 250 mg/L). This provides compelling evidence that EPS secretion is the dominant mechanism for Pb²⁺ immobilization by Hao 2023. The underlying mechanisms may include ion exchange, surface complexation, and serving as nucleation sites promoting the formation of insoluble Pb²⁺ minerals [43,49]. The sequestration of heavy metals via EPS, facilitated by its abundant functional groups (e.g., carboxyl, hydroxyl), represents a common and effective strategy for microbial resistance to toxicity [50].

4.4. Medium Optimization and Resource Characteristics for EPS Production by Strain Hao 2023

Systematic medium optimization elucidated the key nutritional requirements and metabolic traits underlying high EPS yield in strain Hao 2023. The strain demonstrated a distinct preference for sucrose and ammonium sulfate, likely attributable to an efficient sucrose transport and metabolism system [51,52] coupled with an inorganic nitrogen assimilation pathway capable of directly incorporating ammonium into EPS precursors (e.g., aminosugars) [53]. These characteristics offer a potential cost advantage in utilizing low-cost inorganic nitrogen sources. The observed nutritional preferences are consistent with the fundamental principle that the chemical composition and surface properties of adsorbent materials significantly influence their adsorption performance, as the specific carbon and nitrogen sources directly determine the structural characteristics and functional groups of the produced EPS [22].

The most critical finding was the decisive role of marine salt concentration on EPS synthesis, with peak yield at 45 g/L. This concentration, substantially higher than typical for terrestrial bacteria yet consistent with its marine origin, confirms the strain’s halophilic nature [54]. This aligns with the observed enhancement of secondary metabolism (including EPS production) in many marine microorganisms at salinities approximating their native habitat [55]. Such salinity dependence likely functions through activating stress-response pathways linked to osmoregulation [56]. Finally, response surface optimization indicated interactive effects among sucrose, ammonium sulfate, and marine salt, forming a core system regulating EPS synthesis. These findings establish a basis for designing a cost-effective fermentation process leveraging the strain’s marine adaptations [20].

4.5. Fermentation Kinetics and Mechanism of Condition Optimization for EPS Production by Strain Hao 2023

Fermentation condition optimization revealed how strain Hao 2023 balances EPS synthesis with cellular growth. The strain showed optimal growth and EPS production at 35 °C, a temperature typical of many mesophilic Bacillus species [45]. In contrast, maximum EPS yield occurred under slightly acidic conditions (pH 6.0), which diverges from the neutral to mildly alkaline pH preference reported for many Bacillus strains during growth and metabolism [57]. This distinct pH optimum may indicate unique physiological adaptations in Hao 2023—possibly reflecting an acidic pH preference of its extracellular glycosyltransferases or more favorable intracellular conditions for accumulating EPS precursors such as UDP-glucose. Environmental pH is known to influence secondary metabolite flux by modulating membrane potential and substrate transport efficiency [58].

The optimized parameter combination collectively defines a cultivation window conducive to high-yield EPS production. An 8% inoculum size shortens the lag phase while avoiding premature nutrient depletion and metabolic inhibition often associated with high initial cell densities [43]. A shaking speed of 150 rpm supplies sufficient dissolved oxygen while minimizing shear-induced damage to high-molecular-weight EPS, which is a common concern in the microbial synthesis of biopolymers [59,60].

The peak EPS yield at 36 h (late exponential phase) clearly indicates a growth-associated production pattern. This aligns with the synthesis kinetics of many bacterial EPS as secondary metabolites, often linked to coping with nutrient stress or preparing for entry into the stationary phase [61]. However, for processes aimed at maximizing productivity, identifying and potentially decoupling the growth and production phases is a key strategy for further yield enhancement [20].

4.6. Application Potential of the Optimized Process and Future Perspectives

Through systematic optimization of medium components and fermentation conditions, EPS production by strain Hao 2023 was significantly enhanced. The success of this optimization not only validates the efficacy of statistical experimental design methodologies in microbial fermentation but also underscores the resource potential of Hao 2023 as a novel marine EPS producer. RSM has been successfully applied to achieve high-cell-density fermentation of Saccharomyces cerevisiae Y1402 for enhanced enzyme and flavor compound yields [62]. Similarly, RSM was employed to optimize the fermentation of rapeseed meal by Bacillus subtilis to improve protease activity and peptide content [63], and to increase polyphenol levels during the fermentation of Acanthopanax senticosus leaves by a probiotic Bacillus mixture [64]. The ability of Hao 2023 to utilize low-cost ammonium sulfate and a defined marine salt concentration for EPS production aligns with the goal of developing cost-effective bioprocesses [65]. Marine bacteria, inhabiting extreme environments characterized by low temperatures and high pressure, possess superior capabilities to produce unique bioactive compounds, including EPSs, with structural and functional properties that may offer advantages over EPSs derived from traditional terrestrial microorganisms [65,66,67,68].

While optimization at the shake-flask level serves as a crucial first step, the transition towards industrial application necessitates fermentation at the pilot-scale in bioreactors, which presents several key engineering challenges. In fermenters, research on antifungal substance production by Streptomyces yanglinensis 3–10 highlights the significant role of dissolved oxygen (DO) and pH in optimizing yield [69]. Similarly, the oxygen transfer rate (OTR) is a critical factor influencing polyhydroxybutyrate (PHB) production and the oxidative stress response in the deep-tank cultivation of Rhizobium phaseoli [70]. During industrial scale-up, a major challenge lies in precisely controlling these parameters to prevent microbial stress responses triggered by environmental fluctuations, which could adversely affect both the yield and quality of EPS. To maintain high and stable EPS production, such control strategies should be integrated with feeding protocols [71]. Furthermore, this study focused primarily on yield improvement. The fine structure and molecular weight distribution of the EPS, as well as the structure–activity relationships between these characteristics and its heavy metal adsorption capacity or plant growth-promoting activity, remain largely unexplored. Understanding how EPS structure correlates with its function—such as its ability to adsorb heavy metals or promote plant growth—is essential [72]. EPS from various microbial sources has demonstrated excellent heavy metal adsorption capacity, which is closely linked to its unique chemical composition. For instance, EPS produced by marine-derived Pseudomonas sp. or Bacillus sp. often contains abundant carboxyl (-COOH) and hydroxyl (-OH) groups. These functional groups can dissociate in aqueous solutions, forming negatively charged sites that effectively adsorb positively charged heavy metal ions such as Cd²⁺, Pb²⁺, and Cu²⁺ [65]. In the future, upon obtaining sufficient quantities of EPS, analyzing its structural properties and further validating its remediation efficiency within soil–plant systems will be critical steps for translating its “potential function” into “practical application” [56,73].

5. Future Perspectives

Based on the findings of this study, future work should focus on the following aspects. The immediate next step involves systematically evaluating the remediation efficacy and ecological safety of strain Hao 2023, applied as live cells or as EPS-rich fermentation products, in pot experiments using cadmium/lead-contaminated soil to better simulate real field conditions. Following this, the development of a composite microbial inoculant by combining this strain with plant-beneficial bacteria possessing specific growth-promoting traits (e.g., IAA production, phosphate solubilization) warrants exploration. This strategy aims to synergize the heavy metal immobilization capacity of Hao 2023 with the plant health-promoting functions of the partner bacteria, thereby addressing the functional limitations inherent to single-strain applications in complex agricultural settings. Ultimately, to advance toward practical deployment, well-designed field trials in typical cadmium/lead-contaminated farmland are essential to confirm the stability, persistence, and comprehensive ecological benefits of such microbial remediation technologies under actual farming conditions.

6. Conclusions

In summary, this study elucidated the differential tolerance strategies of the marine bacterium Bacillus velezensis Hao 2023 towards Pb²⁺ and Cd²⁺ under controlled laboratory conditions. The strain likely immobilizes Cd²⁺ primarily through an alkalization-mediated mechanism, whereas Pb²⁺ sequestration is strongly associated with substantial EPS biosynthesis. Furthermore, optimization of the medium and fermentation parameters significantly enhanced EPS yield. These findings provide new insights into the metal-specific resistance mechanisms of marine bacteria and establish a foundational fermentation process for developing remediation strategies based on this strain. This work indicates the potential of strain Hao 2023 for heavy metal bioremediation; however, its practical remediation efficacy requires further rigorous validation through subsequent studies (e.g., pot and field experiments) within more complex soil–plant systems.