Impact of bacterial culture medium on composition and characteristics of Burkholderia pseudomallei extracellular polymeric substances

Culture media

The culture media for biofilm formation in this study included brain heart infusion (BHI) (Product No. M210-500G, Hi-media, India), an enriched medium with high carbon and nitrogen content (1.00% (w/v) proteose peptone, 1.25% (w/v) dehydrated calf brain infusion, 0.50% (w/v) dehydrated beef heart infusion, 0.20% (w/v) glucose, 0.50% (w/v) NaCl, 0.25% (w/v) Na₂HPO₄), Luria-Bertani (LB) (BDH), a common medium for bacterial cultivation (1.00% (w/v) tryptic soy broth, 0.50% (w/v) yeast extract and 1.00% (w/v) NaCl) and modified Vogel–Bonner medium, MVBM, a chemically defined medium with high carbon and low nitrogen availability (0.02% (w/v) MgSO₄⋅7H₂O (Merck), 0.20% (w/v) C₆H₈O₇ (Kemaus), 0.35% (w/v) NaNH₄HPO₄⋅4H₂O (Sigma, Singapore), 1.00% (w/v) K₂HPO₄ (Merck), 0.036% (w/v) CaCl₂⋅2H₂O (BDH), 2.00% (w/v) glucose (Kemaus)) (Kunyanee et al., 2016; Lam et al., 1980; Taweechaisupapong et al., 2005).

All chemicals used in this study were of analytical reagent grade or equivalent.

Bacterial strain and biofilm cultivation

An environmental B. pseudomallei ST-39 isolate (Wang-Ngarm, Chareonsudjai & Chareonsudjai, 2014) was used throughout this study. The bacterium from frozen stocks was streaked on Ashdown’s agar plates and incubated at 37 °C for 48 h. A single B. pseudomallei colony was grown in five mL of LB broth with agitation at 200 rpm (Innova 42R; New Brunswick Scientific, Edison, NJ, USA) at 37 °C overnight. The bacterial suspension was subsequently diluted to an OD₅₅₀ ≈ 0.1 (UV-160 A, Shimadzu, Japan). One mL of diluted culture was inoculated into 99 mL of LB broth, resulting in a final concentration of 1% (v/v) inoculum. This culture was then further incubated to obtain an OD₅₅₀ ≈ 1.0 (corresponding to ≈1 ×10⁸ CFU/mL) as biofilm inoculum (Pakkulnan, Sirichoat & Chareonsudjai, 2024). One mL of the biofilm inoculum was added to 24 mL of fresh BHI, LB, or MVBM in 50 mL conical centrifuge tubes (Corning, Corning, NY, USA), prepared in quadruplicate for each media. Biofilm cultures were further incubated at 30 °C under static condition with vented lids for 2, 4, or 6 days, based on previously established growth and biofilm formation previously demonstrated (Kamjumphol et al., 2015; Wang-Ngarm, Chareonsudjai & Chareonsudjai, 2014). Biofilms cultures were harvested by centrifugation at 10,000× g for 15 min at 4 ^∘C. Following removal of the supernatant, the pellets were washed three times with 25 mL phosphate-buffered saline (PBS). The pellets were then stored at −20 °C until further used. On the day of the experiment, the pellets were thawed at room temperature for 10 min and resuspended in 10 mL PBS. Subsequent analyses were performed to determine the total biofilm dry weight, extract EPS, and quantify total carbohydrate and protein concentrations.

Carbon to nitrogen (C/N) ratio in B. pseudomallei culture media

Supernatants from all three B. pseudomallei culture media on days 2, 4 and 6 were collected by centrifugation at 10,000× g for 15 min at 4 ^∘C. Total organic carbon (TOC) left in the supernatant was determined followed the previous described (Walkley & Black, 1934) as originally described. The TOC was performed using the wet oxidation method. A 50 mL of the supernatant was transferred into a 250 mL Erlenmeyer flask and mixed with 10 mL of 0.1 M K₂Cr₂O₇ solution. Subsequently, 20 mL of concentrated H₂SO₄ was then added, and the mixture was gently swirled for 1 min before being allowed to cool to room temperature. After cooling, 100 mL of distilled water, 10 mL o-phosphoric acid, and 0.2 g NaF were added, followed by the addition of 3–4 drops of diphenylamine indicator, which developed to violet-blue color (Nelson & Sommers, 1982). The mixture was then titrated with ferrous ammonium sulfate (FAS), prepared by dissolving 196.1 g Fe(NH₄)₂(SO₄)₂⋅6H₂O and 20 mL concentrated H₂SO₄ in distilled water to a final volume of 1 L. The titration endpoint was identified by a color change from violet-blue to greenish blue.

The TOC concentration was calculated using the following equation: ${\displaystyle \text{TOC}\left(\text{\%}\right)=\frac{\left(V_{\mathrm{b}}-V_{\mathrm{s}}\right)\times N\times 0.003\times 100}{\text{sample volume (mL)}}}$ where V_b is the volume of FeSO₄ used for the blank (mL), V_s is the volume of FeSO₄ used for the sample (mL), and N is the normality of K₂Cr₂O₇ solution.

Total Kjeldahl nitrogen (TKN) was quantified according to the method previously described by Kirk (1950). A 200 mL of culture supernatant was digested with 50 mL of digestion reagent (0.77 M K₂SO₄, 0.029 M CuSO₄, 2.46 M H₂SO₄) using a Kjeldahl Digester (Model 177550, Gerhardt) until the solution became clear. Following digestion, the mixture was cooled to room temperature, after which 200 mL deionized water, 0.5 mL phenolphthalein indicator, and 50 mL of mixed NaOH–Na₂S₂O₃ solution (0.1 M each) were added. Nitrogen was subsequently distilled using a Gerhardt Distillation System (Model 7630, Gerhardt) under the following conditions: 0.5 min reaction time, 15 min distillation time, and 50% steam intensity. A total of 150 mL distillate was collected and titrated with 0.02 N H₂SO₄ to the endpoint, indicated by a color change from green to light purple. A blank, prepared using deionized water, was processed in parallel for correction.

The TKN concentration was calculated according to the following equation: ${\displaystyle \text{TKN}\left(\text{\%}\right)=\frac{\left(\mathrm{A}-\mathrm{B}\right)\times 280\times 100}{\text{sample volume (mL)}}}$ where A is the volume (mL) of H₂SO₄ used for titration of the sample, and B is the volume (mL) used for the blank.

C/N ratio was calculated using the following formula: ${\displaystyle \text{C/N ratio}=\frac{\text{Total carbon content (g)}}{\text{Total Kjeldahl nitrogen content (g)}}.}$

Measurement of B. pseudomallei biofilm biomass

Burkholderia pseudomallei biofilm biomass was determined as the total dry weight of B. pseudomallei using the volatile suspended solids (VSS) method (American Public Health et al., 2005). Briefly, glass fiber filter papers (GF/C, 1.1 µm pore size, four cm diameter; Whatman) were pre-dried at 105 °C for 1 h, cooled in a desiccator, and stored until use. For biomass collection, filters were placed on a Büchner funnel, pre-wetted with distilled water, and used to filter cell pellets that had been resuspended in 10 mL distilled water under vacuum. The retained biomass was subsequently rinsed with 10 mL of distilled water and subjected to vacuum filtration for additional 3 min. Filters containing the biomass were then dried at 105 °C for 1 h, cooled in a desiccator, folded, placed in pre-weighed crucibles, and weighed to obtain the initial dry mass (A). The crucibles were then ignited at 550 °C for 20 min until the content turned to gray ash, cooled in a desiccator, and reweighed to determine the post-ignition mass (B). The organic dry biomass was calculated as the difference between these two weights, with measurement performed using an analytical balance (Model MSE225S-100-DI; Sartorius, Germany).

Biofilm biomass was calculated using the following formula: ${\displaystyle \text{Total dry weight of biofilm (mg/mL)}=\frac{\left[\text{weight before ashing (mg) (A)}-\text{weight after ashing (mg) (B)}\right]\times 10^6}{\text{Volume (mL)}}.}$

EPS extraction

EPSs are classified into two main types based on their physical association with microbial cells: cell-bound EPS and soluble EPS. In this study, only cell-bound EPS was extracted from biofilm pellets using the cation-exchange resin (CER) method (Frølund et al., 1996; Jahn & Nielsen, 1995; Oliva et al., 2025; Oliva et al., 2024). Briefly, the frozen pellets were suspended in 10 mL PBS containing 1.6 g of CER Dowex (Dowex Marathon C sodium, strongly acidic, 20–50 mesh; Sigma, Singapore), then mixed thoroughly using a vortex mixer. The mixture was then shaken at 400 rpm for 1 h, after which the supernatant was collected. This was followed by centrifugation at 10,000× g for 15 min at 4 ^∘C. The supernatant was transferred again to a fresh tube and centrifuged at 20,000× g for 20 min at 4 ^∘C. EPS purification by monomer removal was performed via dialysis (MWCO 3,500 Da; T1 5015-46; Cellu.Sep) in deionized water (EPS:DI, 1:10) at 4 ^∘C for 24 h. The retained EPS in the dialysis membrane was stored at −20 ^∘C for subsequent analyses.

To determine the dry weight, functional groups, and morphological characteristics of the EPS, the EPS suspension was first concentrated using a rotary evaporator (Maxivac Beta, LaboGene) and stored at −70 ^∘C for 24 h. It was then freeze-dried (MiniVac, LaboGene) to obtain a powdered sample, which was subsequently stored at −20 ^∘C.

Measurement of B. pseudomallei EPS biomass

The biomass of extracted EPS was determined from 100 mL of culture and extrapolated to a per-liter yield. The EPS yield was expressed as the percentage of extracted B. pseudomallei EPS in biofilm grown in each culture medium. ${\displaystyle \text{Percentage of EPS}=\left(\text{Dry weight of EPS in biofilm (g/L)/Dry weight of biofilm (g/L)}\right)\times 100.}$

Biochemical composition and characterization of EPS

Determining functional groups of EPS using FTIR

The functional groups of EPS were determined using FTIR (TENSOR 27, Bruker, Germany), following the method previously described by Bramhachari (Bramhachari et al., 2007) using potassium bromide (KBr) as a baseline control. EPS powder was finely ground with KBr to a final concentration of 0.01% (w/w) in a mortar. The mixed sample was placed in a mold and was compressed with a hydraulic press at 15,000 psi for 1–2 min. The sample was analyzed with an FTIR, and the spectra were interpreted to identify functional groups and chemical bonds, particularly OH-stretch compounds and N-compounds, using public databases, such as PubChem (https://pubchem.ncbi.nlm.nih.gov/), WebSpectra (https://webspectra.chem.ucla.edu/irtable.html), and ChemAnalytical (http://www.chemanalytical.com/services/ft-ir-spectra/).

Characterization of EPS using SEM

Structural characterization of EPS followed a previously described method (Schrand et al., 2008). Briefly, the extracted EPS samples were desiccated at −20 ^∘C for 24 h. Subsequently, the dried samples were mounted on stubs using carbon adhesive tape and then sputter-coated with a gold layer approximately 10–20 nm thick prior to imaging with a scanning electron microscope (SEC, SNE-4500M).

Statistical analysis

Data was obtained from three independent experiments, each with three replicates (n = 9). Statistical analysis was performed using GraphPad Prism version 10 (GraphPad Software, Boston, MA, USA). Comparisons of the total dry weight of biofilms and EPS, carbon-to-nitrogen ratios, carbohydrate and protein concentrations, were performed using one-way ANOVAs followed by Tukey post-hoc test for comparison between pairs. The levels required for statistical significance were *p < 0.05, ^∗∗p < 0.01 and ^∗∗∗ p < 0.001. Bar graphs in the figures display the mean values with error bars representing the standard deviation (SD). Linear regression analysis was performed to analyze correlation of C/N ratio of media on the biofilm and EPS dry weight.

Effect of culture medium on biofilm biomass, EPS dry weight and the EPS yield in B. pseudomallei biofilm

The dry weight (the total dry mass) of B. pseudomallei biofilms grown in BHI, LB and MVBM increased throughout the course of observations (Table 1). The dry weight of biofilm was highest in cultures grown in MVBM, followed by those cultivated in BHI and LB. On day 6, the biofilm dry weight in MVBM was significantly greater than that observed in both BHI and LB cultures (p < 0.001). In contrast, biofilms formed in LB exhibited the lowest biofilm biomass among the three media.

The dry weight of EPS extracted from B. pseudomallei biofilms increased over the experimental period in all media (Table 1). On day 4, the EPS dry weight monitored in BHI medium was significantly higher than those of the other two media (p < 0.001). The dry weight of EPS in both LB and MVBM media increased significantly from day 2 to day 6 (p < 0.01).

The percentage of EPS relative to total biofilm biomass was analyzed (Fig. 1). Notably, the LB medium consistently produced the highest EPS percentage across all time points throughout the incubation period, in comparison to both BHI and MVBM media (p < 0.001). The MVBM medium demonstrated the lowest EPS percentage within the biofilm matrix (p < 0.001).

The carbon and nitrogen (C/N ratio) of the nutrients directly influence bacterial metabolism, growth strategy, and biofilm development. Therefore, the C and N contents were measured in freshly prepared BHI, LB and MVBM media as well as in spent media collected after biofilm cultivation on days 2, 4 and 6 (Table S1). The corresponding C/N ratios were calculated (Fig. 2). The C/N ratios of fresh BHI, LB and MVBM were 8.5, 12 and 98, respectively. The C/N ratios in MVBM following 2, 4, and 6 days of incubation were 190, 359 and 417, respectively, the highest in any of the culture media (p < 0.001) according to the low N content. BHI and LB exhibited relatively low C/N ratios of 11–22 and 11–16, respectively. These findings suggest that MVBM contains excess carbon and limited nitrogen, resulting in the observed extremely high C/N ratio. No significant difference was observed between the C/N ratios of BHI and LB (p > 0.05).

Linear regression analysis was used to investigate the correlation of the C/N ratio of BHI, LB and MVBM media with the dry weight and percentage of EPS in the biofilm matrix of B. pseudomallei measurement on days 2, 4 and 6. As the C/N ratio changes over time, we found a moderate positive correlation between biofilm dry weight and C/N ratio of MVBM medium (R² = 0.5570) (Fig. 3A) and a strong positive correlation between percentage of EPS in biofilm matrix and the C/N ratio of MVBM medium (R² = 0.7131) (Fig. 3B), indicating the correlation of C/N ratio of MVBM medium to dry weight biofilm and percentage of EPS in B. pseudomallei biofilm matrix but not in the case of BHI and LB medium (data not shown).

Effect of culture medium on biochemical composition of B. pseudomallei EPS

The biochemical composition of the EPS extracted from B. pseudomallei biofilms varies according to the cultivation medium. The total carbohydrate content in EPS showed a progressive increase in all media, with a substantial elevation observed by day 6 (Fig. 4A). In contrast, nitrogen content, representing protein quantity, was highest in biofilms grown in BHI medium, followed by those in LB and MVBM media (Fig. 4B). These results suggest a critical influence of nutrient composition on EPS profiles, with BHI medium preferentially supporting protein synthesis, while all tested media facilitated time-dependent carbohydrate accumulation.

FTIR spectroscopy analysis of EPS

FTIR spectroscopy analysis was employed to characterize the functional groups present in the EPS extracted from 6-day-old B. pseudomallei biofilm grown in BHI (Fig. 5A), LB (Fig. 5B), and MVBM (Fig. 5C). The EPS consisted primarily of carbohydrates, proteins and lipids, with both similarities and subtle differences among the media. Notably, EPS produced in LB and MVBM media was rich in carbohydrate-associated functional groups, whereas EPS from BHI medium exhibited comparable levels of carbohydrate and protein functional groups. The characteristic peaks associated with functional groups observed including broad O–H and/or N–H stretching vibrations around 3,400–3,200 cm⁻¹ and prominent amide C=O stretch (1,700–1,600 cm⁻¹) bands corresponding to polysaccharide (O–H stretching) and protein components (N–H and C=O stretching), respectively. Notably, the EPS extracted from bacteria grown in BHI and LB medium exhibited the highest peak intensities, particularly in the amide (C=O stretch (1,700–1,600 cm⁻¹)). These findings indicated that the different C and N components in each culture medium impacted on the biochemical composition of bacterial EPS by day 6. However, on days 2 and 4, the functional group compositions of EPS were similar in LB, MVBM, and BHI (data not shown). These observations were consistent with the results obtained from the quantitative analysis of total carbohydrate and protein concentrations in EPS. Additionally, lipid-associated functional groups were also detected across all media.

Scanning electron microscopic structure of the B. pseudomallei EPS

SEM images revealed the EPS architecture extracted from the biofilms of B. pseudomallei cultivated in BHI, LB and MVBM media for 2, 4 and 6 days (Fig. 6). On day 2, EPS from all three media was smooth in morphology but had a more granular appearance on days 4 and 6. EPS from BHI cultivation appeared as tightly packed clusters while EPS from LB and MVBM cultures demonstrated large and compact aggregates with irregular and coarse surface morphology.