Abstract
Microbial exopolysaccharides (EPS) are structurally diverse biopolymers with ecological and biotechnological relevance due to their physicochemical and functional properties, including emulsification, metal chelation, and contributions to biofilm formation. In this study, an EPS-enriched extracellular fraction (EPS-EF) produced by Vreelandella zhaodongensis BS253, isolated from the Brazilian Pantanal, was characterized using an integrated approach combining microscopic, physicochemical, molecular, genomic, and transcriptomic analyses. Morphological and structural profiling by microscopy revealed a highly porous, reticulated web-like supramolecular network. The EPS-EF exhibited cation-dependent colloidal stability, with turbidity decreasing by approximately 95% after divalent-cation chelation. The emulsification index after 24 h (E24) against kerosene reached approximately 67%, and the surface tension of water was reduced to approximately 45 mN/m, with an apparent surface-tension inflection (STI) near 10.7 mg/mL. X-ray diffraction (XRD) revealed a predominantly amorphous profile with discrete crystalline reflections. Thermogravimetric analysis (TGA) demonstrated thermal stability with a multi-step degradation profile and a maximum decomposition rate centered at 280 °C. CHNS/O and FTIR were consistent with a carbohydrate-associated, nitrogen-poor molecular profile containing carbonyl/carboxylate and glycosidic functionalities. UHPLC–ESI–QTOF–MS/MS resolved a reproducible set of oligomeric features across four chromatographic zones, without establishing monosaccharide composition, together with minor hydrophobic features tentatively associated with lipid-related compounds. One-dimensional 1H NMR provided complementary molecular-fingerprint information, but the primary structure remains unresolved. EPS-EF material analysis may also include contributions from co-extracted components because the analyzed material is an enriched extracellular fraction rather than a chemically homogeneous polysaccharide. Genome mining and targeted RNA-seq analysis of 47 EPS-associated genes identified candidate pathways for precursor metabolism, glycan assembly, and export. Collectively, these findings indicate that V. zhaodongensis BS253 produces an ionic-sensitive EPS-EF with emulsifying activity and support a working hypothesis for EPS-EF biosynthesis and export.
1. Introduction
Microbial exopolysaccharides (EPS) are extracellular biopolymers that form hydrated adaptive matrices at the interface between microbial cells and their environment. In bacteria exposed to osmotic stress, desiccation, nutrient limitation, and toxic compounds, EPS contribute to biofilm formation, surface adhesion, water retention, protection against environmental stressors, and cell–cell interactions [1,2,3,4]. Due to their structural diversity and functional versatility, microbial EPS have attracted increasing interest in biotechnology, particularly as biodegradable emulsifiers, viscosifying agents, metal chelators, and pharmaceutical biomaterials [5,6,7].
Among EPS-producing microorganisms, halophilic and halotolerant bacteria have emerged as particularly valuable sources of novel extracellular polymers [8]. Members of the family Halomonadaceae, including the recently established genus Vreelandella, are recognized as prolific producers of structurally diverse heteropolysaccharides with distinctive physicochemical properties [9]. These polymers often exhibit high anionic charge density, exceptional hydration capacity, metal-binding ability, and strong emulsifying performance, making them attractive candidates for environmental and industrial applications [10]. Despite this potential, the composition, biosynthetic pathways, and biological functions of EPS produced by Vreelandella and other recently described members of the Halomonadaceae remain poorly understood [11].
The structural characterization of bacterial EPS remains particularly challenging because these polymers often consist of heterogeneous units containing neutral sugars, uronic acids, amino sugars, and acetyl substituents, with lipid- or protein-associated components also contributing in some cases [12,13,14,15]. No single analytical technique is sufficient to resolve this complexity. Recent advances in structural identification techniques, such as high-resolution tandem mass spectrometry, proton nuclear magnetic resonance (1H NMR), and Fourier-transform infrared spectroscopy (FTIR); physicochemical characterization techniques, such as X-ray diffraction (XRD) and thermogravimetric analysis (TGA); and comparative genomics have considerably improved the ability to infer, respectively, EPS composition, physicochemical properties, and biosynthetic organization, especially when combined with transcriptomic analyses that reveal active metabolic pathways during polymer production [13]. In Gram-negative bacteria, EPS biosynthesis is generally coordinated through complex gene clusters involved in carbon uptake, nucleotide-sugar biosynthesis, glycosyltransferase-mediated assembly, membrane translocation, and extracellular export [14]. The coordinated activity of these pathways ultimately determines the monosaccharide composition, charge distribution, degree of acetylation, and physicochemical properties of the final polymer.
Previous physiological characterization of V. zhaodongensis BS253 by our group demonstrated its tolerance to salinity, pH, temperature, pressure, and elevated CO2 conditions, together with the production of extracellular material displaying pronounced emulsifying activity [15]. These observations suggested the production of an EPS with emulsifying properties that may contribute to environmental adaptation.
In the present study, we performed a comprehensive physicochemical, molecular, functional, genomic, and transcriptomic characterization of the exopolysaccharide enriched extracellular fraction (EPS-EF) produced by V. zhaodongensis BS253, a halophilic bacterium isolated from the Brazilian Pantanal. EPS-EF was investigated using microscopy, spectroscopy-based biochemical assays, emulsification assays, surface-tension assays, XRD, thermogravimetric analysis (TGA), elemental analysis (CHNS), FTIR, mass spectrometry (UHPLC–ESI–QTOF–MS/MS), and proton nuclear magnetic resonance (1H NMR), while whole-genome sequencing and RNA-Seq were used to define its main functional and chemical features and to identify candidate pathways associated with its biosynthesis and export. The combined evidence suggests the presence of a structurally heterogeneous extracellular matrix displaying prominent carbohydrate-associated molecular features together with minor hydrophobic features that may contribute to the observed emulsifying activity. However, the present data do not allow us to distinguish whether these hydrophobic species are physically associated with the bioactive extracellular matrix. Moreover, the working hypothesis model integrates nucleotide-sugar metabolism, glycosyltransferase activity, polymer export, and structural organization, which is consistent with the observed assay results. These findings expand the known chemical diversity of EPS produced by extremophilic and halophilic bacteria and provide a framework for linking biosynthetic potential, molecular composition, and environmental functionality in Vreelandella spp., with potential applications in emulsification, bioremediation, and environmental biotechnology.
2. Materials and Methods
2.1. Microorganism and Culture Conditions
Vreelandella zhaodongensis BS253 (IPR microorganism bank) was isolated from a saline–alkaline lagoon of the Brazilian Pantanal (GCF_051175405.1) and previously characterized as a strain producing extracellular material with emulsification activity. Strain maintenance and basal cultivation procedures followed the previously published protocol [15]. Briefly, V. zhaodongensis BS253 was maintained in saline medium (basal medium) and cultivated in mineral salts medium (MSM) for production assays. MSM consisted of, per liter, 5 g yeast extract, 5 g NaCl, 5.7 g KH2PO4, 5.25 g K2HPO4, 3 g NaNO3, 1 g KCl, 0.51 g MgSO4·7H2O, 0.21 g CaCl2·2H2O, and 0.20 g Na2S2O3. For production of the extracellular material used in the present structural and physicochemical characterization, V. zhaodongensis BS253 was cultivated in MSM under a 10% CO2-enriched atmosphere at 28 °C and 180 rpm for 72 h.
2.2. EPS-EF Extraction and Enrichment
After cultivation, the pH of the whole culture was adjusted to 9.0, and cells were separated from the culture broth by centrifugation (10,000 rpm for 10 min at 20 °C). The resulting cell pellet was washed three times with NaOH buffer adjusted to pH 9.0 to detach any cell-associated EPS. The initial culture supernatant and the alkaline wash fractions were combined, adjusted to pH 7.0, and filtered through 0.22 µm membranes to remove residual cells and debris. Cold absolute ethanol was added to the filtered aqueous phase at a 3:1 ethanol/water ratio, and the mixture was incubated at 4 °C for 24 h to precipitate EPS [6]. The precipitated fraction was recovered by centrifugation (10,000 rpm for 10 min), resuspended in ultrapure water, and dialyzed for 72 h using a 3 kDa cutoff membrane (Merck Millipore, Jaffrey, NH, USA) to remove salts and low-molecular-weight impurities. The dialyzed material was further concentrated using Amicon ultrafiltration units with a 3.7 kDa cutoff (Merck Millipore). For further enrichment, the concentrated sample was fractionated by size-exclusion chromatography using a Superdex®75 column (Cytiva, Marlborough, MA, USA) (1.6 cm × 60 cm; total column volume ~120 mL) equilibrated and eluted with ultrapure water at a constant flow rate of 2 mL/min. A sample volume of 2.0 mL (containing approximately 10 mg of concentrated crude EPS) was injected per run, with the enrichment completed over 3 consecutive chromatographic runs. Fractions of 2.0 mL were collected and screened for emulsification activity using the E24 assay described below. The active fractions were pooled and lyophilized, resulting in a final EPS-EF. All subsequent analyses were performed using this EPS-EF. The size-exclusion chromatography step was used solely for fractionation and activity-based selection and was not calibrated for molecular-weight determination; therefore, absolute molecular weight and polydispersity index were not determined. Because the fractionation strategy was designed to enrich extracellular polymeric material based on precipitation, dialysis, size exclusion, and emulsification activity rather than to obtain a chemically homogeneous polysaccharide, the resulting material is referred to throughout the manuscript as an EPS-enriched extracellular fraction.
2.3. Optical and Electron Microscopy
Microscopic analyses were performed using washed bacterial cells and EPS-EF of V. zhaodongensis BS253. For cellular observation, bacterial smears were prepared on glass slides, heat-fixed, and stained with Sudan Black B [16]. Excess dye was removed using 70% ethanol, followed by decolorization with xylene. The cells were then counterstained with 0.05% safranin. Stained preparations were examined under oil immersion using a 100× objective. For observation of the extracellular material, the EPS-EF was analyzed under 10× and 20× magnification.
For scanning electron microscopy (SEM), samples were fixed by immersion in 3 mL of 2.5% glutaraldehyde solution for 24 h at 4 °C [17]. After fixation, samples were mounted onto aluminum stubs using a 0.1 M sodium cacodylate buffer (pH 7.4). The samples were then dehydrated through a graded acetone series (30%, 50%, 70%, 90%, 95%, and 100%, with two final washes in 100% acetone for 15 min each). Samples were then subjected to critical point drying and sputter-coated with a gold/palladium alloy to ensure surface conductivity. The prepared specimens were examined using a Zeiss EVO MA10 SEM (Zeiss, Peabody, MA, USA), and the resulting micrographs were analyzed for morphological characterization.
For transmission electron microscopy (TEM), samples were diluted in 0.1 M sodium cacodylate buffer (pH 7.4) and fixed in 2.5% glutaraldehyde solution, post-fixed with 1% osmium tetroxide, and dehydrated through the same graded acetone series described above. The samples were then embedded in epoxy resin, ultrathin-sectioned, stained with 2% uranyl acetate and 0.3% lead citrate, and analyzed using a TECNAI FEI G20 TEM (FEI, Eindhoven, NB, Netherlands) for ultrastructural visualization [18].
2.4. Biofilm and EPS-EF Matrix-Disruption Spectrophotometric Analysis
Biofilm formation was evaluated using the crystal violet microplate assay [19,20]. The strain was cultivated statically for 72 h at 28 °C in 96-well polystyrene microplates containing 200 µL per well of basal or MSM. After incubation, wells were gently washed three times with sterile 0.9% NaCl to remove planktonic cells. The adhered biomass was treated for 45 min at 37 °C with 200 µL of one of the following solutions: 10 mM EDTA, 3.46 µM proteinase K, 10 mM EDTA plus 3.46 µM proteinase K, or untreated control. Wells were washed again, stained with 0.4% crystal violet for 10 min, dried, and the bound dye was solubilized with 200 µL of 96% ethanol. Absorbance was measured at 570 nm.
To evaluate the organization of the EPS-EF, 10 mg of lyophilized sample was dissolved in 2.5 mL of ultrapure water. Aliquots of 100 µL were distributed into 96-well plates and incubated for 45 min under the same treatment conditions described above. Turbidity was measured at 600 nm before and after centrifugation (1000 rpm for 3 min) to compare total and soluble fractions. Protein-associated signals were assessed using the Bradford reagent and expressed as apparent protein-associated absorbance at 595 nm before and after centrifugation (20,000 rpm for 10 min). Congo Red binding was evaluated at 495 nm using a 20 µM solution, before and after centrifugation (20,000 rpm for 10 min). Appropriate blanks were used for all biofilm and EPS-EF conditions, including water, basal medium, MSM, crystal violet + water, crystal violet + basal medium, crystal violet + MSM, EPS-EF + water, EDTA + water, proteinase K + water, and EDTA + proteinase K + water. The relative treatment-associated signal percentage of the matrix was calculated by subtracting the optical density OD ratio of the treated sample to the untreated control from 1 and multiplying the result by 100: [1 − (EPS-EF biofilm + treatment/EPS-EF or biofilm)] × 100. All biofilm and EPS-EF disruption assays were performed in three independent biological experiments, with three technical replicates per condition within each experiment. Technical replicates were averaged before statistical analysis, and the independent biological experiment was considered the experimental unit. Data are presented as mean ± standard deviation (SD). For biofilm measurements, a two-way analysis of variance (ANOVA) was used to evaluate the effects of cultivation medium and treatment, including their interaction, followed by Tukey’s multiple-comparison test. For EPS-EF measurements, treatment and centrifugation stage were analyzed using a mixed-design two-way ANOVA, with treatment considered a between-experiment factor and centrifugation stage (before versus after) considered a within-experiment repeated factor. Šídák-adjusted comparisons were used for multiple comparisons where appropriate. Partial eta squared (ηp2) was calculated as an effect-size measure for factorial effects. Statistical assumptions were assessed before inferential analysis, and a p-value < 0.05 was considered statistically significant. All analyses were performed using GraphPad Prism version 9 (GraphPad Software, Boston, MA, USA).
2.5. Physicochemical Characterization of EPS-EF
Emulsification activity and surface-tension behavior of the EPS-EF were evaluated according to previously described methods [21]. EPS-EF samples were evaluated across a range of concentrations from 0.034 to 100.07 mg/mL. The emulsification index after 24 h (E24) was determined by mixing equal volumes of the aqueous sample solution and kerosene, followed by vigorous homogenization. Surface tension measurements were conducted using the pendant drop method with a Theta Flex tensiometer (Biolin Scientific, Manchester, UK). Three independent experimental measurements were obtained at each EPS-EF concentration. The individual surface-tension measurements were used for concentration–response analysis. Because the replicate measurements at different concentrations were experimentally independent, concentration was not treated as a repeated-measures factor. The concentration dependence of surface tension was characterized using a four-parameter logistic (4PL) nonlinear regression model: Y = Bottom + (Top − Bottom)/[1 + (X/EC50)Hill slope]. The model was fitted to the individual observations across the concentration range. The EC50, Hill slope, and corresponding 95% confidence intervals were estimated from the nonlinear regression. As a complementary categorical analysis, differences among concentrations were assessed using Welch’s one-way ANOVA because of unequal variances among concentration groups. The effect size for the concentration factor was summarized using eta squared (η2) and omega squared (ω2). A p-value < 0.05 was considered statistically significant. The results were plotted as surface tension versus concentration curves, and the apparent surface-tension inflection (STI) of EPS-EF was determined as the intersection point from the linear equations between the decreasing and steady stated observed in the fitted curve.
Thermogravimetric analysis (TGA) was performed using a PerkinElmer Pyris Diamond TGA/DTA system (PerkinElmer, Shelton, CT, USA) equipped with a gas station. Lyophilized EPS-EF samples (4–5 mg) were heated from 30 °C to 600 °C at a heating rate of 10 °C/min under a nitrogen flow of 40 mL/min [22].
X-ray diffraction (XRD) analysis was performed using a Bruker D8 Advance powder diffractometer (Bruker, Madison, WI, USA) with Ni-filtered CuKα radiation (λ = 1.54056 Å), operated at 35 kV and 30 mA. The instrument was calibrated using a silicon standard, and lyophilized EPS-EF samples were scanned over a 2θ range of 2–90° using a scintillation detector under slow-scan conditions. Interplanar d-spacing values were calculated from the diffraction angles using Bragg’s equation [22].
2.6. Compositional and Molecular Analysis of EPS-EF
The elemental composition of the EPS-EF was determined using a FlashSmart™ Elemental Analyzer (Thermo Scientific, Milan, MI, Italy) equipped with automated switching between combustion and pyrolysis modes. For carbon, hydrogen, nitrogen, and sulfur analysis, approximately 2 mg of lyophilized sample was mixed with vanadium pentoxide oxidizer, weighed into tin capsules, and subjected to dynamic flash combustion at 950 °C under oxygen and helium flow, generating N2, CO2, H2O, and SO2. For oxygen determination, samples were placed in silver capsules and pyrolyzed at 1060 °C under helium flow. The resulting gases were separated by gas chromatography and quantified using a thermal conductivity detector. Elemental percentages (CHNS/O) were calculated using the instrument’s data acquisition and calibration software 1.0 [13].
Fourier-transform infrared spectroscopy (FTIR) was performed using a PerkinElmer Spectrum 100 spectrometer (PerkinElmer, Shelton, CT, USA) equipped with a universal attenuated total reflectance accessory fitted with a diamond crystal. Lyophilized EPS-EF samples were placed directly onto the crystal surface, and spectra were recorded at room temperature over the range of 4000–400 cm−1, with a spectral resolution of 4 cm−1 and 16 accumulated scans. Data were acquired using Spectrum software 1.0 (PerkinElmer, Shelton, CT, USA) and presented in transmittance mode [23].
2.7. Mass Spectrometry of EPS-EF
Mass spectrometric analyses were performed using an ultra-high-performance liquid chromatography system coupled to a quadrupole time-of-flight mass spectrometer (UHPLC-QTOF-MS/MS) equipped with an electrospray ionization (ESI) source (Shimadzu LCMS 9030, Canby, OR, USA). EPS-EF samples were separated on a polar reversed-phase Luna® Omega Polar column (Phenomenex, Torrance, CA, USA) (100 Å, 150 × 3.0 mm, 3.0 μm) maintained at 35 °C, using an injection volume of 2 μL and a flow rate of 1.0 mL/min. The mobile phase consisted of (A) 5 mM ammonium formate containing 0.1% formic acid and (B) acetonitrile/5 mM ammonium formate (95:5) containing 0.1% formic acid. Gradient elution was performed from 0 to 100% B over 57 min (Table S1). Mass spectrometry data were acquired in negative electrospray ionization mode using Data-Dependent Acquisition (DDA). Full-scan MS spectra were acquired over an m/z range of 150–1000, and MS/MS spectra were acquired over an m/z range of 50–1000. The capillary voltage was set at −3.5 kV. Nitrogen was used as nebulizing and drying gas at flow rates of 3 and 10 L/min, respectively. The desolvation line temperature was maintained at 250 °C, the heat block at 400 °C, and the interface at 320 °C [12].
2.8. Nuclear Magnetic Resonance (NMR) of EPS-EF
Analyses for 1H NMR spectroscopy were performed using approximately 5.0 mg of EPS-EF dissolved in 0.8 mL of D2O containing 0.05% 3-(trimethylsilyl) propionic-2,2,3,3-d4 acid sodium salt (TSP) (99.9% D, Sigma-Aldrich). The spectra were recorded on a 400 MHz Bruker Avance spectrometer (Bruker, Billerica, MA, USA) equipped with a 5 mm PABBO BB/19F-1H/D Z-gradient probe at a controlled sample temperature of 50 °C. Chemical shifts (δ) were referenced internally against the TSP signal (δ H 0.00 ppm). A total of 32 scans were accumulated for data acquisition.
2.9. Genomic and Transcriptomic Analysis
Genome and transcriptomic data for V. zhaodongensis BS253 were retrieved from the annotated complete genome (RefSeq assembly GCF_051175405.1; chromosome NZ_CP195113.1, 3,765,793 bp, 53.35% G + C, 3335 chromosomal coding sequences, available under BioProject PRJNA1275882) and previously published datasets [24]. The previous RNA-seq experiment evaluated V. zhaodongensis BS253 cultivated in MSM under two conditions: basal MSM without additional inorganic carbon supplementation and MSM under 10% CO2-enriched atmosphere. Cultures had been incubated at 28 °C and 180 rpm, and samples were collected at 12 h and 18 h with three biological replicates per condition. Raw count matrices were imported into R (version 4.3.1) and processed using the DESeq2 package (v1.40.1). A targeted subset of 47 genes potentially involved in EPS biosynthesis was selected for downstream evaluation. To adjust for differences in sequencing depth and to stabilize variance across the dataset, read counts for selected genes were subjected to a variance stabilizing transformation (VST) using the experimental design. Differential expression of the emulsification-associated genes was evaluated using the Wald test in DESeq2, with pairwise contrasts performed to assess temporal shifts within the same condition (Control 18 h vs. Control 12 h; CO2 18 h vs. CO2 12 h) as well as condition-specific effects at the late exponential phase (CO2 18 h vs. Control 18 h). Transcripts were classified as differentially expressed genes (DEGs) if they exhibited a Benjamini–Hochberg adjusted p-value < 0.05 and an absolute log2 fold change >1. To identify expression sub-networks associated with EPS production, the VST-normalized counts of the targeted loci were analyzed using the WGCNA package (v1.72-1). A signed co-expression network was constructed using a soft-thresholding power of 6 and a minimum module size of 3 to accommodate the focused 47-gene subset. Module eigengenes (MEs) were calculated and correlated with the experimentally determined emulsification index (E24) of the 12 samples using Pearson correlation coefficients, alongside individual gene-trait Pearson correlations to identify specific loci highly correlated with emulsifying activity. For visualization, the VST-normalized expression data was transformed into row-wise Z-scores, ordered primarily by their assigned WGCNA module color, and secondarily in descending order based on their absolute Pearson correlation with the emulsification index. Samples were hierarchically clustered within their respective time-point groups. The final comprehensive heatmap—incorporating condition metadata, WGCNA module assignments, emulsification correlation annotations, and binary DEG status matrices—was generated using the ComplexHeatmap package (v2.16.0), with matrix scaling handled via tidyverse (v2.0.0) and circlize (v0.4.15), and rendering performed using the ragg package (v0.1.3).
2.10. Genomic Context, Transcription-Unit Prediction and Comparative Analysis
For the genomic-context analysis described in Section 2.9, coordinates, strand orientation and product descriptions were extracted from the GFF3 annotation, and the intergenic distance between each coding sequence and its immediate upstream neighbour was calculated, negative values denoting overlapping coding sequences. Gene symbols were assigned by orthology (eggNOG-mapper v2.1.12 reciprocal best hits) and NCBI CD-Search, not by experimental characterization. Transcription units were predicted by grouping consecutive co-directional coding sequences separated by no more than 50 bp; this threshold resolved the chromosome into 2128 transcription units, of which 676 were polycistronic (maximum 11 coding sequences per unit). Functional assignments of the initiation, chain-length-determining and export proteins, and of the nucleotide-sugar dehydrogenases, were confirmed by NCBI CD-Search against the Conserved Domain Database at an expect-value threshold of 0.01. The presence of Wzx-type flippases and Wzy-type polymerases was assessed by searching all chromosomal product descriptions for the corresponding protein families. Local G + C content of the candidate modules was compared with the distribution across 10-kb sliding windows of the chromosome (mean 53.35%, SD 1.93%).
Predicted transcription units were tested against the RNA-Seq data. Read counts for the twelve samples described in Section 2.9 were transformed with the variance-stabilising transformation of DESeq2, and genes with a mean normalised count below 10 were excluded. For every pair of adjacent chromosomal coding sequences the Pearson correlation of the transformed profiles was computed, and pairs were classified as belonging to the same predicted transcription unit, co-directional but separated by more than 50 bp, convergent, or divergent. Distributions were compared with one-sided Mann–Whitney tests and effect sizes reported as Cliff’s delta. Mean within-unit correlations were computed for each multi-gene transcription unit of the three EPS-related modules.
Orthologues of the candidate loci were sought in six related Halomonadaceae genomes retrieved from RefSeq: Vreelandella zhaodongensis NEAU-ST10-25 (GCF_013415115.1, type strain), V. aquamarina DSM 15723 (GCF_054110395.1), V. profundi MT13 (GCF_019722725.1), V. piezotolerans NBT06E8 (GCF_012427705.1), V. titanicae SOB56 (GCF_014697095.1) and Halomonas elongata DSM 2581 (GCF_000196875.2). Proteomes were compared with DIAMOND blastp (v2.2.6, expect-value threshold 1 × 10−10) and orthologues defined as reciprocal best hits against the BS253 proteome. Orthologue coordinates were mapped onto each genome annotation, and conservation of gene order was assessed as the number of intervening coding sequences between orthologues of adjacent BS253 genes. For display, each genome was centred and oriented on kpsE.
Transcription units reported here are computational predictions supported by co-expression; transcript boundaries were not mapped experimentally.
3. Results
3.1. Recovery and Morphology of the EPS-EF
Microscopic analyses were performed on both bacterial cultures and EPS-EF. Optical microscopy after Sudan Black B staining revealed hydrophobic or lipid-associated components in association with cellular and extracellular structures (Figure 1A). SEM images showed extracellular material distributed around and between bacterial cells (Figure 1B), whereas TEM revealed hair-like structures associated with the cell surface (Figure 1C,D). These microscopic observations indicate that the bacterial cells were surrounded by an extracellular matrix consistent with the presence of EPS. Extracellular material was observed in close association with the cell surface, suggesting the presence of a cell-associated matrix, potentially including capsular components.
Figure 1.
Microscopic characterization of extracellular material associated with V. zhaodongensis BS253 cells. (A) Light microscopy image showing bacterial cells surrounded by extracellular material. (B) Scanning electron microscopy (SEM) image showing cells embedded in an extracellular matrix-like structure. (C,D) Transmission electron microscopy (TEM) images showing dense extracellular aggregates and fibrillar, hair-like structures associated with the bacterial surface.
After enrichment and lyophilization (additional details of the enrichment workflow are provided in Figure S1), EPS-EF-associated emulsification activity was initially evaluated by optical microscopy under hydrated, dried, and kerosene-containing conditions (Figure S2). Droplet-like structures and organized emulsion-associated arrangements were observed in the kerosene-containing preparation. Moreover, the recovered EPS-EF exhibited a white, flocculent, fibrillar appearance (Figure 2A). SEM and TEM analyses showed flocculent aggregates and filamentous structures under the evaluated microscopic conditions (Figure 2B–E). Overall, these results indicate that the enrichment process was able to desorb EPS from the cell surface. No evident morphological signs of extensive cell lysis were observed under the evaluated conditions, as also shown in Figures S1E and S3.
Figure 2.
Microscopic characterization of EPS-EF produced by V. zhaodongensis BS253. (A,B) Optical microscopy images showing white, flocculent, and fibrous EPS-EF at low and higher magnification. (C) Scanning electron microscopy (SEM) image revealing the porous and filamentous organization of the EPS-EF. (D,E) Transmission electron microscopy (TEM) images showing electron-dense aggregates and fibrillar, hair-like structures of the EPS-EF.
3.2. Differential Responses of Adherent Biofilms and EPS-EF to Matrix-Disrupting Treatments
The effects of EDTA, proteinase K, and their combination were evaluated in both adherent biofilms and EPS-EF samples (Figure 3) and their statistical analysis are available in Data Availability Statement. The two materials exhibited distinct treatment-response profiles.
Figure 3.
Effects of matrix-disrupting treatments on biofilm and EPS-EF of V. zhaodongensis BS253. (A) Quantification of adhered biofilm mass via crystal violet staining (OD570) following treatment with EDTA, proteinase K, or a combination of both, in basal and MSM media. (B) Calculated relative reductions in biofilm-associated crystal violet signal following EDTA and proteinase K treatments. Functional and relative treatment-associated signal reduction assays of the EPS-EF include: (C,D) turbidity assessment (OD600) and the corresponding relative treatment-associated signal changes; (E,F) Bradford-reactive signal (OD595); and (G,H) supramolecular organization using the Congo Red binding (OD495) and the corresponding relative treatment-associated changes in dye binding. EPS-EF measurements were performed using untreated control, EDTA, proteinase K, and EDTA plus proteinase K treatments. Biofilm data were analyzed by two-way ANOVA with cultivation medium and treatment as factors, followed by Tukey’s multiple-comparison test. EPS-EF data were analyzed using a mixed-design two-way ANOVA with treatment as a between-experiment factor and centrifugation stage as a within-experiment factor, followed by Šídák-adjusted multiple comparisons where appropriate. Partial eta squared (ηp2) was used as the effect-size measure. Three independent biological experiments were performed, with three technical replicates per condition; technical replicates were averaged before statistical analysis. Data are presented as showing individual data points overlaid on the bar/column. Different lowercase letters indicate statistically significant differences at p < 0.05.
In adherent biofilms grown in basal medium, proteinase K and EDTA reduced biofilm-associated crystal violet signal by approximately 82% and 52%, respectively, compared to the untreated control (OD570 = 1.392). The combined treatment produced the greatest reduction, exceeding 95%. A similar response was observed in the MSM (control OD570 = 1.405), in which proteinase K and EDTA reduced the signal by approximately 95% and 27%, respectively, whereas the combined treatment again produced a reduction greater than 95% (Figure 3A). Based on these treatment responses, proteinase K treatment accounted for approximately 82% and 95% of the total treatment-associated crystal violet signal reduction observed in basal medium and MSM, respectively, whereas the relative reduction associated with EDTA treatment was approximately 52% and 26%. Two-way ANOVA identified a strong main effect of treatment on biofilm-associated crystal violet signal [F(3.16) = 164.37, p < 0.001, ηp2 = 0.969]. The main effect of cultivation medium was not statistically significant [F(1.16) = 0.53, p = 0.476, ηp2 = 0.032]. However, the medium × treatment interaction was significant [F(3.16) = 4.29, p = 0.021, ηp2 = 0.446], indicating that the magnitude of the response to the matrix-disrupting treatments differed between cultivation media. These values describe treatment-associated effects on crystal violet-retained biomass and should not be interpreted as quantitative estimates of matrix composition (Figure 3B).
The EPS-EF was further evaluated using turbidity, Bradford-reactive signal, and Congo Red binding assays to assess colloidal stability, the contribution of protein-associated material, and changes in dye-accessible supramolecular organization, respectively. To assess colloidal stability, turbidity (OD600) was measured before and after centrifugation. For all three measurements, centrifugation stage was analyzed as a within-experiment factor because the before and after measurements corresponded to the same experimental replicate.
For turbidity, the untreated EPS-EF showed a decrease from OD600 = 0.162 before centrifugation to 0.090 after centrifugation. EDTA treatment induced a near-complete collapse of the colloidal suspension, reducing baseline turbidity by approximately 95% (OD600 = 0.0083). Interestingly, proteinase K treatment had the opposite effect, increasing turbidity by nearly 37% (OD600 = 0.223), suggesting altered particle dispersion or aggregation. The combined treatment resulted in low turbidity values of approximately 0.055 before and after centrifugation (Figure 3C,D). The mixed-design ANOVA showed a strong main effect of treatment [F(3.8) = 255.43, p < 0.001, ηp2 = 0.990] and a significant main effect of centrifugation stage [F(1.8) = 7.54, p = 0.025, ηp2 = 0.485]. The treatment × centrifugation interaction was not statistically significant [F(3.8) = 2.28, p = 0.156, ηp2 = 0.461]. The strong treatment effect, together with the EDTA-associated decrease in turbidity, is consistent with ion-sensitive colloidal stabilization of the EPS-EF.
Bradford measurements were interpreted as assay-reactive protein-associated signals rather than direct protein quantification. Values were corrected using treatment-specific reagent blanks because proteinase K, EDTA, and their combination could affect the Bradford response. The untreated control exhibited OD595 values of 1.24 before centrifugation and 1.28 afterward. Proteinase K treatment substantially reduced the Bradford-reactive signal, yielding values of approximately 0.21 under both centrifugation conditions. In contrast, EDTA treatment slightly elevated the signal (OD595 = 1.50 before and 1.35 after centrifugation), while the combined treatment yielded intermediate values of approximately 0.59 (Figure 3E,F). The mixed-design ANOVA showed a strong treatment effect [F(3.8) = 290.92, p < 0.001, ηp2 = 0.991], no significant main effect of centrifugation stage [F(1.8) = 1.59, p = 0.243, ηp2 = 0.166], and a significant treatment × centrifugation interaction [F(3.8) = 5.23, p = 0.027, ηp2 = 0.662]. Thus, the effect of centrifugation on the Bradford-reactive signal depended on the matrix-disrupting treatment.
Congo Red binding was used to evaluate changes in dye-accessible supramolecular organization. The untreated fraction exhibited OD495 values of 0.970 before centrifugation and 0.524 afterward. EDTA reduced the signal to approximately 0.674 under both conditions, whereas proteinase K further reduced it to 0.456 before centrifugation and 0.361 afterward. The combined treatment nearly eliminated the detectable Congo Red signal, yielding OD495 values of approximately 0.025 (Figure 3G,H). Mixed-design ANOVA showed a strong treatment effect [F(3.8) = 377.20, p < 0.001, ηp2 = 0.993] and a significant centrifugation-stage effect [F(1.8) = 6.46, p = 0.035, ηp2 = 0.447]. The treatment × centrifugation interaction was not statistically significant at the predefined α = 0.05 threshold [F(3.8) = 4.05, p = 0.050, ηp2 = 0.603]. Overall, the Congo Red response indicates that dye-accessible supramolecular organization within the EPS-EF is sensitive to matrix-disrupting treatments and centrifugation.
3.3. Emulsification Activity and Surface-Tension Behavior
The EPS-EF was serially diluted in water (0.034 to 100.07 mg/mL) to evaluate its emulsifying capacity and surface activity. Increasing the concentration of EPS-EF reduced the surface tension of water from 72.8 to 45 mN/m, revealing an apparent surface-tension inflection (STI), or breakpoint, at approximately 10.7 mg/mL. Concurrently, the emulsification index (E24) against kerosene increased markedly, from 15% to 67%, with the emulsification response approaching its maximum at approximately 0.310 mg/mL (Figure 4). The surface-tension data showed a strong concentration effect (Raw Data at Zenodo). Because the three experimental measurements obtained at each concentration were independent rather than repeated observations of the same experimental units across concentrations, repeated-measures ANOVA and mixed-effects modeling with replicate as a random longitudinal factor were not used. Welch’s one-way ANOVA confirmed significant differences among concentrations [F(8, 7.25) = 1046.74, p < 0.001], with η2 = 0.997 and ω2 = 0.996. The nonlinear concentration–response analysis provided a complementary description of the continuous response. The 4PL model showed an excellent fit (R2 = 0.977), with an estimated Top of 73.840 mN/m, Bottom of 45.656 mN/m, EC50 of 1.488 mg/mL (95% CI: 1.103–1.872 mg/mL), and Hill slope of 1.148 (95% CI: 0.841–1.456).
Figure 4.
Concentration-dependent surface-tension and emulsification activity of the EPS-EF. Surface tension of EPS-EF concentration from 0 to 100.07 mg/mL was measured by the pendant-drop method. Individual observations from three independent biological replicates, with 139 or 140 technical replicates per condition are shown as individual data points overlaid on the fitted line. Technical replicates were averaged before statistical analysis. The fitted curve represents a four-parameter logistic concentration–response model fitted to the individual surface-tension measurements. The 4PL model yielded an EC50 of 1.488 mg/mL (95% CI: 1.103–1.872 mg/mL), with a Top of 73.840 mN/m, a Bottom of 45.656 mN/m, and a Hill slope of 1.148. The right inset shows the corresponding 24-h emulsification index (E24) against kerosene across the concentration range. The apparent surface-tension inflection (STI) at approximately 10.7 mg/mL was determined from the intersection of linear fits to the decreasing and plateau regions of the curve. Unlike the EC50, which represents the midpoint of the fitted 4PL response, the STI represents an operational transition between these regions. Welch’s one-way ANOVA was used as a complementary categorical analysis of surface-tension differences among concentrations, with η2 and ω2 reported as effect sizes. Repeated-measures ANOVA was not used because the replicate measurements at different concentrations were experimentally independent.
The E24 response therefore developed at substantially lower concentrations than those required to produce the largest changes in surface tension. This dissociation between emulsification efficiency and surface-tension reduction is consistent with a bioemulsification mechanism in which stabilization of the oil–water interface depends on interfacial film formation and/or supramolecular interactions, rather than on surface-tension reduction alone. The present results should nevertheless be interpreted specifically for kerosene and under the experimental conditions evaluated here.
3.4. Thermal Behavior and X-Ray Diffraction Profile
Thermogravimetric analysis was performed to evaluate the thermal behavior of the EPS-EF (Figure 5). The TGA profile showed four main temperature-associated degradation regions. The first mass-loss event, corresponding to approximately 5% of the initial mass, occurred between 30.0 and 62.8 °C and may be associated with the release of free and weakly bound water, as commonly observed in hydrated biopolymeric materials. A second mass-loss event of approximately 15% was observed between 62.8 and 190.0 °C and may reflect the release of more strongly associated water and the initial degradation of thermolabile functional groups or side-chain substituents. The most pronounced mass-loss region occurred between 190.0 and 311.1 °C, with an additional loss of approximately 17%. This event is consistent with thermal degradation of the carbohydrate-associated features and may involve cleavage of glycosidic linkages, decomposition of side groups, and progressive depolymerization of polysaccharide-associated components. However, this profile should be interpreted as the behavior of the EPS-EF as a composite extracellular material since contributions from associated proteins, lipids, salts, and other co-extracted components cannot be excluded. Above 311.1 °C, continued degradation and carbonization were observed. At the end of the analysis, the residual mass was approximately 26%. This relatively high residue may reflect the presence of ash-forming components, residual salts, co-purified inorganic material, thermally stable constituents, or carbonized residues. Although coordination of inorganic cations within the EPS-EF may also contribute, this interpretation cannot be established from TGA alone.
Figure 5.
TGA results for EPS-EF. Solid lines represent the weight decrease while the dashed lines show derivative thermogravimetry with temperature. (A) Thermal degradation between 0–62.80 °C. (B) Thermal degradation between 62.8–190.0 °C. (C) Thermal degradation 190.0–311.1 °C. (D) Thermal degradation above 311.1 °C. TGA assays were executed in independent biological triplicates (Raw Data at Zenodo).
X-ray diffraction analysis revealed a predominantly amorphous profile with discrete diffraction peaks (Figure 6). The main reflections were detected at 2θ values of 22.4°, 23.0°, 31.9°, 33.0°, 45.4°, 56.5°, 75.1°, and 84.1°, indicating the presence of localized ordered or crystalline domains within an otherwise largely amorphous material. These discrete reflections may arise from partially ordered regions of the EPS-EF. The partial ordering observed by XRD may be compatible with localized supramolecular organization within the EPS-EF, potentially involving interactions among polymer chains and ionic components; however, this ordering is not necessarily exclusive to the polysaccharide component, as it may also include ordered organic or inorganic co-associated constituents. Therefore, the present data do not allow cation-mediated junction zones to be demonstrated directly. Birefringence was also observed under polarized light microscopy (Figure S4), supporting the presence of optically anisotropic or partially ordered regions within the EPS-EF.
Figure 6.
X-ray diffraction profile of the EPS-EF. The diffractogram shows a predominantly amorphous background with discrete crystalline peaks (e.g., 23°, 31°, 45°, and 56°). XRD assays were executed in independent biological triplicates (Raw Data at Zenodo).
3.5. Elemental and Macromolecular Characterization of the EPS-EF
To elucidate the elemental composition of the EPS-EF, CHNS/O analysis and Fourier transform infrared (FTIR) spectroscopy were employed. Elemental profiling revealed that carbon is the most abundant element, comprising approximately 46% of the total mass, followed by oxygen (23%) and hydrogen (8%). Notably, nitrogen accounted for only about 3%, while sulfur was negligible (Figure 7A). The relatively low nitrogen content in EPS-EF is consistent with the carbohydrate-associated profile indicated by our earlier functional assays but does not exclude minor proteinaceous or amino-containing components. Because proteins typically contain approximately 16% nitrogen by mass, the measured nitrogen content of about 3% places an estimated upper bound of approximately 19% on the protein-equivalent material in the lyophilised EPS-EF. The true protein content is likely lower because amino sugars and other nitrogen-containing components may also contribute to the measured nitrogen pool.
Figure 7.
Elemental composition and FTIR profile of the EPS-EF. (A) CHNS/O elemental distribution profile demonstrating the relative mass percentage (%) of carbon (C), hydrogen (H), nitrogen (N), sulfur (S), and oxygen (O) within the EPS-EF. (B) Fourier transform infrared (FTIR) transmission spectrum (4000 to 500 cm−1). Key vibrational coordinates are annotated with their corresponding wavenumber frequencies (cm−1) and linked to specific functional groups, highlighting the structural signatures of putative carboxylate/carbonyl groups with acetyl modifications and glycosidic bonds. Both assays were executed in independent biological triplicates (Raw Data at Zenodo).
To further resolve the specific macromolecular bonds within the EPS-EF, FTIR spectroscopy was utilized (Figure 7B). The resulting spectrum exhibited strong carbohydrate-associated absorption bands, whereas characteristic peptide-associated amide bands were predictably weak or poorly defined. A broad band at 3271.59 cm−1 indicated extensive O–H stretching vibrations, typical of highly hydrated polymers, while bands at 2922.36 and 2855.13 cm−1 corresponded to aliphatic C–H stretching. Intense bands at 1633.33 and 1410.13 cm−1 were assigned to carbonyl- and carboxylate-associated vibrations, indicating the presence of carboxylate/carbonyl-containing functionalities. Together with the detection of putative uronic-acid-like molecular features among the UHPLC–MS/MS fragments, these observations are consistent with acidic carbohydrate-associated components. Finally, the highly resolved, low-frequency fingerprint region (1300–800 cm−1) displayed a complex array of peaks (1242.65, 1148.29, 1064.48, 979.19, and 856.30 cm−1) assigned to C–O–C and C–O stretching, consistent with carbohydrate-associated C–O and C–O–C functionalities. The band at 1242.65 cm−1 falls within the region in which S=O stretching of sulfate esters also absorbs. Because sulfur was negligible in the CHNS/O elemental analysis, this band is assigned to C–O–C and C–O stretching, and significant sulfation of the EPS-EF can be excluded.
3.6. UHPLC–ESI–QTOF–MS/MS Analysis
To map the structural heterogeneity and determine the compositional subunits of the EPS-EF, high-resolution UHPLC–ESI–QTOF–MS/MS analysis was performed in negative ionization mode (Figure 8). The principal macromolecular and oligomeric signals systematically emerged between a retention time (Rt) window of 3.3 and 31.2 min. Signals eluting outside this specific interval matched baseline solvent configurations or well-defined background noise and were resolved independently. Based on localized retention behavior, structural mass defects, and collision-induced dissociation (CID) product-ion dynamics, the chromatogram was partitioned into four distinct zones (Figure 8, Zones A–D).
Figure 8.
Comparative UHPLC–ESI–QTOF–MS/MS base peak chromatogram of the EPS-EF. The micro-structural elution profile of the EPS-EF (orange trace) is overlaid against the analytical solvent blank (black trace) in negative electrospray ionization (ESI-) mode. Dashed geometric boundaries designate the four distinct analytical elution regimes (A–D) mapped across the functional mobile phase gradient on a polar reversed-phase stationary phase: (A) early hydrophilic oligomeric window (3.3–6.0 min); (B) dominant, high-density acidic glycan core (6.0–9.5 min); (C) intermediate oligosaccharide transitional zone (9.8–24.0 min); and (D) late-eluting region (24.5–31.2) containing minor hydrophobic features tentatively compatible with lipid-associated or isoprenoid-carrier-related components.
Zone A (3.3–6.0 min) was mainly composed of relatively low-to-medium molecular weight species (m/z 150 to 755), representing early-eluting hydrophilic oligomers. Zone B (6.0–9.5 min) represents an intense main glycan-rich core of the EPS-EF. Here, higher-mass ions became more abundant (e.g., m/z 997.13, 996.14, 980.15, and others). To determine the building blocks of this core, low-molecular-weight reporter ions were tracked across both zones. The simultaneous presence of these fragments is consistent with chemically related uronic acid-containing species. To elucidate the specific connectivity of these complex carbohydrates, tandem mass spectrometry (MS/MS) was evaluated for a representative peak at 9.45 min, corresponding to the deprotonated ion at m/z 924.2110 [M–H]− (Figure 9). Based on accurate mass and fragmentation behavior, this ion was interpreted as a putative oligosaccharide-like compositional feature compatible with acidic and N-acetylated carbohydrate components. The full-scan spectrum at this retention time (Figure 9A) contained further ions at m/z 657.36, 587.28, 541.26, 474.22 and 444.21. Because these were recorded in MS1 rather than as product ions of a mass-selected precursor, they are reported as co-eluting species and are not assigned as fragments of m/z 924.2110. Tandem MS of the m/z 924.2110 precursor (Figure 9B) may be related to dehydration, decarboxylation, and acetamide-loss events. Together, these spectral data suggest that the 9.45 min peak is putatively an acidic oligosaccharide-like compositional motif. Several low-mass ions were detected across the majority of retention times, including in the solvent blank, and are therefore interpreted as background features rather than as glycan-specific reporter ions. The mass spectrometric data are accordingly presented as a molecular fingerprint of the EPS-EF, and no monosaccharide-level compositional assignment is proposed from them.
Figure 9.
Mass spectrometric analysis of proposed representative oligosaccharide-like molecular feature detected at RT 9.45 min. (A) Mass spectrum showing the precursor ion at m/z 924.2110 and co-eluted and possible associated ions. (B) Tandem mass spectra showing putative oligosaccharides productions used for fragmentation-based compositional annotation.
Finally, to characterize the later-eluting, less polar fractions of the extract, fragmentation patterns in Zones C and D were analyzed. Zone C (9.8–24.0 min) displayed a prolonged, low-abundance elution profile, containing related oligosaccharide-like species with similar fragmentation patterns. However, Zone D (24.5–31.2 min) exhibited a distinct high-mass-defect ion pattern (>0.35), shifting away from pure glycans. Tandem MS/MS of precursor ions m/z 987.4770 and 895.4005 yielded hypothetical structural pathways tentatively compatible with lipid-associated or isoprenoid-carrier-related components. However, their precise identities and relationship with the EPS-EF remain unresolved (all spectrometric data are shown in Figures S5–S8, and Table S2).
3.7. 1H NMR Profile of the EPS-EF
To further validate the structural motifs and functional group environments within the EPS-EF, proton nuclear magnetic resonance (1H NMR) spectroscopy was performed (Figure 10). The resulting spectrum displayed a complex profile with broad, overlapping resonances typical of microbial heteropolysaccharides. An intense signal centered at approximately 4.71 ppm was assigned primarily to residual HDO and may overlap with anomeric proton signals from carbohydrate residues, for instance H1 protons of hexose (Hex), hexuronic acid (HexA), and their N-acetylated derivatives (HexNAc/HexNAcA). Adjacent to this, the broad cluster of signals observed between 3.50 and 4.40 ppm fell within the carbinolic region, representing the H2–H6 overlapping ring protons consistent with oxygenated ring protons of carbohydrate residues.
Figure 10.
Proton nuclear magnetic resonance (1H NMR) spectrum of the EPS-EF. Key structural domains are annotated, including the anomeric region (H1 protons at ~4.7 ppm), the carbinolic region (H2–H6 ring protons at 3.50–4.40 ppm), acetyl, amide, and amine-associated modifications (1.9–3.0 ppm), and the aliphatic resonance region (1.0–1.5 ppm).
Furthermore, resonances in the 1.9–3.0 ppm region were compatible with acetylated substituents and other overlapping aliphatic environments, compatible with acetylated and other aliphatic substituents identified in our MS/MS analysis. Signals between 1.0 and 1.5 ppm were consistent with methyl resonances potentially arising from 6-deoxy sugars, terminal methyl groups of lipidic or acyl substituents, or other aliphatic components. Notably, the spectrum lacked prominent resonances in the aromatic region (6.0–8.0 ppm); however, this observation does not exclude minor proteinaceous or other aromatic components. Overall, this 1H NMR profile is highly consistent with a heterogeneous, acetylated, carbohydrate-associated matrix, with MS and FTIR data successfully complementing the structural overlaps inherent to this one-dimensional analysis.
3.8. Genomic and Transcriptomic Analysis of EPS-Related Genes
To elucidate the genetic basis underlying the biosynthesis of the EPS-EF, whole-genome and transcriptomic datasets of V. zhaodongensis BS253 were evaluated (Figure 11). Genome mining successfully identified a set of candidate loci within chromosomal gene clusters potentially associated with EPS precursor generation, assembly, modification, and export (Figure 11A).
Figure 11.
Targeted co-expression analysis of the predefined 47-gene EPS-associated set potentially associated with EPS biosynthesis and export in V. zhaodongensis BS253. (A) Complete genome map showing the positions of genes implicated in EPS production. (B) RNA-Seq heatmap of the highlighted genes. Genes were classified into modules according to their relationship with emulsification activity and their expression variation. (C) Working hypothesis for the candidate pathways associated with EPS-EF biosynthesis and export.
Specifically, genetic candidates associated with sugar uptake (MFS transporters and fruA), uronic acid metabolism (ugd), and N-acetylated sugar formation (glmU, wecB, wecC) were detected. The presence of these nucleotide-sugar pathways is consistent with the acidic and acetylated structural motifs identified in our FTIR, MS/MS, and 1H NMR analyses. Furthermore, genes linked to lipid-carrier metabolism (uppS, uppP), polymerization initiation (wcaJ), and ABC transporter-mediated export (kpsM, kpsT, alongside polysaccharide co-polymerase and outer membrane export proteins) were identified, which increase the chance of lipid derivatives be involved in the EPS-EF constitution.
Targeted expression analysis was performed using RNA-Seq transcriptomic profiles at 12 and 18 h under basal and CO2-enriched MSM conditions (Figure 11B). Expression data revealed condition- and time-dependent upregulation at 18 h for multiple genes governing nucleotide-sugar precursor biosynthesis, glycosyltransferase activity, and capsule formation. Notably, within the targeted 47-gene EPS-associated dataset, the co-expression module showing the strongest positive correlation with the measured emulsification index (turquoise module; r = +0.85, p = 0.002 after Bonferroni correction for the five modules tested) included key glycosyltransferases, capsule-associated proteins, ABC transporters, and nucleotide-sugar metabolism enzymes.
The combined genomic and transcriptomic data were used to formulate the working hypothesis shown in Figure 11C. The working hypothesis integrates sugar uptake and activation, nucleotide-sugar precursor biosynthesis, WcaJ-associated chain initiation, glycosyltransferase-mediated assembly onto lipid-linked intermediates, ABC transporter-associated translocation across the inner membrane, and final extracellular export via PCP/OPX-associated envelope components.
The candidate loci are not dispersed across the chromosome. Genomic-context analysis placed them in three contiguous modules contained within a single 166.4-kb segment of the 3.77-Mb chromosome, spanning positions 1,057,779 to 1,224,193 (Figure S9, Table S3). The gene symbols and locus tags are summarized in Table S4.
The first module (15.1 kb; positions 1,057,779–1,072,888) comprises seven co-directional coding sequences in which every intergenic distance is 4 bp or less and three pairs of adjacent genes overlap; it was therefore predicted to form a single transcription unit. Conserved-domain analysis identified its first two genes as kpsC and kpsS (both E = 0), the diagnostic components of group 2, ABC-transporter-dependent capsule assembly. The module also encodes rfaH, the transcriptional antiterminator that governs capsule and O-antigen operons in Gram-negative bacteria.
The second module (59.7 kb; positions 1,082,793–1,142,454) begins 9.9 kb downstream and comprises 41 coding sequences, every one of which is transcribed from the same strand. Its proximal boundary is a galU-pgm pair whose coding sequences overlap by 3 bp, supplying UDP-glucose; the module then combines the remaining nucleotide-sugar genes—the UDP-glucose 6-dehydrogenase ugd, a second nucleotide-sugar dehydrogenase, wecB, wecC, rfbA-rfbD, glmM and two copies of glmS—with multiple glycosyltransferases and, at its distal end, a polysaccharide-export module. Within that export module, kpsE, encoding a polysaccharide co-polymerase of the PCP-3 family (E = 1.2 × 10−124), lies 4 bp upstream of an ABC transporter permease of the ABC-2 family and was predicted to form a transcription unit with it; the corresponding ABC ATP-binding subunit overlaps the upstream rfbB by 3 bp; and an outer-membrane component of the OPX (Wza/KpsD) family lies 156 bp further downstream on the same strand. The presence of rfbA-rfbD, which direct dTDP-L-rhamnose biosynthesis, is consistent with the methyl resonances observed at 1.0–1.5 ppm in the 1H NMR spectrum, which were attributed in part to 6-deoxy sugars.
The third module (17.8 kb; positions 1,206,361–1,224,193) lies 63.9 kb beyond the second and contains a seven-gene predicted transcription unit of 8.3 kb combining three glycosyltransferases with a second, independent ABC exporter of the Wzm/Wzt type, the ATP-binding subunit of which carries a C-terminal carbohydrate-binding domain. The module also contains wcaJ (E = 3 × 10−151), within a four-gene predicted transcription unit that additionally encodes a type I secretion apparatus, together with genes encoding mannose-1-phosphate guanylyltransferase/mannose-6-phosphate isomerase and phosphomannomutase, both transcribed from the opposite strand.
A genome-wide search recovered no Wzx-type flippase and no Wzy-type polymerase: the only O-antigen ligase encoded by the chromosome is WaaL, which acts in ligation of the O-antigen to the lipid A-core, and the five MATE-family transporters annotated in the genome are multidrug efflux proteins located outside the three modules. The G + C content of the three modules (55.18%, 52.42% and 51.24%) lies within the range of 10-kb windows across the chromosome (mean 53.35%, SD 1.93%), so these data provide no evidence of recent horizontal acquisition.
To test whether the predicted transcription units behave as co-expressed units, the expression profiles of adjacent chromosomal genes were compared across the twelve RNA-Seq samples (Figure S10). Gene pairs assigned to the same predicted transcription unit were substantially more correlated (median Pearson r = 0.69, n = 1189) than co-directional pairs separated by more than 50 bp (median r = 0.43, n = 1114; p = 4.0 × 10−26, one-sided Mann–Whitney, Cliff’s delta = 0.25), convergent pairs (median r = 0.26, n = 499; p = 1.6 × 10−37) or divergent pairs (median r = 0.21, n = 495; p = 2.8 × 10−47). Within the EPS-related modules, the export and initiation units were among the most internally coherent: mean within-unit correlation was 0.84 for kpsE with the ABC permease, 0.88 for the OPX-family component with its downstream gene, 0.91 for ugd-rfbC, 0.75 across the seven-gene wzt-wzm unit and 0.82 across the four-gene wcaJ unit. Two predicted units were exceptions: the seven-gene capsule module, despite its tight gene spacing, was only weakly co-expressed (mean r = 0.34), and one four-gene unit of the second module was not co-expressed (mean r = −0.11). Genomic context therefore predicts transcriptional co-regulation for most, but not all, of these loci.
To place this architecture in a phylogenetic context, the candidate loci were compared with six related Halomonadaceae genomes: the V. zhaodongensis type strain NEAU-ST10-25 (GCF_013415115.1), V. aquamarina DSM 15723, V. profundi MT13, V. piezotolerans NBT06E8, V. titanicae SOB56 and Halomonas elongata DSM 2581 (Figure S11, Table S5). Orthologues were identified by reciprocal best hits, which recovered between 2203 and 2864 orthologous pairs per genome. The export machinery is conserved throughout: kpsC, kpsS, kpsE, the ABC permease, the ABC ATP-binding subunit and the OPX-family component were recovered in all six relatives, with amino-acid identities of 56–99%, 63–99%, 65–90%, 70–91%, 74–86% and 66–90% respectively. Gene order is conserved as well. In every genome examined the six genes lie on a single replicon within a span of 49.4–82.7 kb; kpsC and kpsS are immediate neighbours in all seven genomes, as are the ABC permease and the OPX component, and kpsE is adjacent to the ABC permease in six of the seven, being separated by one intervening gene only in H. elongata. The glycan-specific genes, by contrast, are variable: the complete dTDP-L-rhamnose set rfbA-rfbD was recovered in three of the six relatives (the V. zhaodongensis type strain, V. aquamarina and V. titanicae), only rfbA and rfbB in H. elongata, neither in V. piezotolerans, and a single weak rfbB hit (26% identity) in V. profundi; wecB and wecC were recovered only in V. piezotolerans; waaL is absent from V. titanicae and H. elongata; and the Wzm/Wzt-type exporter has a close counterpart only in V. titanicae (87–89% identity), the reciprocal hits recovered in H. elongata (27–43%) and V. aquamarina (32%) being too divergent to assign with confidence. The EPS locus of V. zhaodongensis BS253 therefore combines an export cassette that is conserved across the six related genomes examined with a variable, strain-specific glycan-synthesis region, an architecture widely reported for bacterial surface-polysaccharide loci.
Predicted transcription units, intergenic distances, conserved-domain assignments and within-unit correlations for all coding sequences of the three modules are given in Table S3; these transcription units were inferred from genomic context and co-expression and were not mapped experimentally.
4. Discussion
The multi-platform analytical framework applied in this study indicates that V. zhaodongensis BS253 produces a structurally complex, ionic extracellular matrix with a predominantly carbohydrate-associated profile. This integrated interpretation is particularly relevant because microbial EPS are highly diverse in composition and architecture, particularly among halophilic and extremophilic bacteria [9].
Microscopy revealed a hybrid architecture potentially comprising both capsular and released extracellular polymeric substances (EPS): Sudan Black B staining localized hydrophobic/lipid-associated material in and around cells, while SEM and TEM showed an extracellular matrix distributed between cells and fibrillar, hair-like structures at the cell surface. In Gram-negative halophilic bacteria, EPS is typically composed of carbohydrate polymers and may include capsular polysaccharides (CPS), loosely bound EPS, released polymers (slime EPS), and envelope-associated glycoconjugates [16,25,26,27]. The concurrent presence of surface-anchored and released polymer is consistent with an adaptive strategy documented among halophilic bacteria, in which capsular and dispersed EPS jointly support biofilm morphogenesis, osmotic balance, and nutrient capture in fluctuating saline environments [1,16,28,29]. Alkaline extraction desorbed the extracellular matrix with no evident morphological signs of extensive cell lysis (Figures S1 and S3), supporting pH-modulated destabilization as an effective route to recover intact polyanionic networks, most plausibly by altering the protonation state of anchoring carboxyl and phosphate groups [30]. The EPS-EF displayed a white, flocculent, and fibrillar appearance consistent with other descriptions of microbial EPS-like polymers in the literature [13].
The matrix-disruption assays in Figure 3 revealed results in the adherent biofilm and the EPS-EF that merit careful interpretation. In the adherent biofilm, the very large treatment effect size (ηp2 = 0.969) supports a substantial alteration of crystal-violet-retained biomass following matrix-disrupting treatments. Proteinase K produced the larger reduction in the crystal violet-associated signal (~82% in basal medium, ~95% in MSM), whereas EDTA had a smaller effect, and the combined treatment produced an enhanced effect (>95%). This differential response is consistent with a contribution of protein-sensitive structures to biofilm-associated biomass, which is commonly observed in microbial adhesion [31,32].
The EPS-EF exhibited a distinct combination of treatment and centrifugation responses. The strong treatment effects observed for turbidity, Bradford-reactive signal, and Congo Red binding are consistent with substantial changes in the physicochemical organization of the extracellular material after EDTA and/or proteinase K exposure. In the EPS-EF, the EDTA-sensitive collapse of the colloidal suspension (turbidity −95%) is consistent with a role for divalent-cation-dependent interactions in matrix stabilization observed in other microorganisms [33]. For Bradford, the significant interaction suggests that centrifugation affected the assay-reactive signal differently depending on the treatment. Moreover, the disruption of EPS complexes by chelation is likely susceptible to carbohydrate-mediated optical interference [34]. These data suggest the interpretation of the EPS-EF as a heterogeneous extracellular fraction in which ionic interactions and protein-associated components can contribute to supramolecular organization. However, the assay does not establish whether these components are physically entrapped, non-covalently associated, or covalently linked to the extracellular matrix, and it is not proof of polysaccharide purity. Therefore, Bradford-reactive signal should not be directly interpreted as an increase in protein content. Whereas turbidity and Bradford data indicate a carbohydrate-associated matrix with a possible protein contribution, the Congo Red assay assesses whether the organization of the structure is affected by the treatments. Congo Red can bind glucan/amyloid and protein conformations that contribute to supramolecular organization and colloidal behavior [32]. The significant centrifugation effect for turbidity and Congo Red binding further indicates that part of the matrix response depends on the transition from the initial suspension to the resulting supernatant. In contrast, the absence of a significant interaction for turbidity indicates that the overall effect of centrifugation did not differ sufficiently among treatments at the present sample size. All these findings are complemented by qualitative chemical profiles previously published by our group indicating carbohydrate-associated and amino-containing components, as well as anionic functional groups, while neutral lipid-associated signals were limited [15].
Functionally, the EPS-EF is best described as a bioemulsifier rather than a classical biosurfactant, a distinction commonly observed across Halomonas and other halophilic species [35,36,37]. The concentration statistical analysis of surface tension provides a more appropriate quantitative framework for Figure 4 than a repeated-measures model. The independent experimental structure across concentrations precludes treating concentration measurements as repeated observations of the same experimental units. Welch’s ANOVA demonstrated a strong concentration effect despite heterogeneity of variance, while the 4PL model captured the continuous concentration-dependent response with a high goodness of fit. In Figure 4, the EPS-EF reduced the surface tension of water from ~72 to ~45 mN/m with an apparent surface-tension inflection (STI) near 10.7 mg/mL, yet the emulsification index against kerosene approached its maximum (~67%) at concentrations as low as 0.310 mg/mL—roughly 35-fold below the apparent STI. This decoupling of emulsification from surface-tension reduction is consistent with interfacial-film formation, hydration capacity, steric stabilization, and polymeric network organization of oil droplets rather than monolayer-driven tension reduction [35,38,39]. Therefore, we interpret the apparent STI at ~10.7 mg/mL cautiously because, for some polymers, a true micellar CMC is not necessarily the correct physical model, and the inflection may instead reflect a concentration-dependent change in interfacial packing or chain overlap. However, because emulsification was evaluated against kerosene only, these results should not be generalized to all hydrophobic substrates without additional comparative testing.
Physicochemical profiling is consistent with a highly hydrated EPS-EF displaying a predominantly amorphous diffraction profile with discrete reflections of unresolved origin. Thermogravimetric analysis showed an early mass loss (~5–15%) attributable to free and coordinated water, followed by a mass loss between ~190 and 311 °C, consistent with the high hydration capacity, thermolabile polysaccharide side chains, volatilization of acetyl groups, and glycosidic linkages typical of halotolerant EPS [40,41]. The residual mass (~26%) may reflect ash-forming components, residual salts, inorganic material, thermally stable constituents, or carbonized residues found in microbial EPS [42]. Although coordinated inorganic cations may contribute, this possibility cannot be established from TGA alone. The major degradation event is compatible with thermal degradation of carbohydrate-associated organic components, although contributions from associated proteins, lipids, salts, and other co-extracted components cannot be excluded. X-ray diffraction showed a predominantly amorphous profile with discrete reflections. Similar reflections have also been observed in structurally distinct bacterial EPS because of monosaccharide composition, linkage patterns, and ionic interactions [1,5,9,13,16,22,25,26,27,40,43,44,45,46]. Together with the EDTA sensitivity and TGA ash residues, these observations are consistent with—though they do not by themselves prove—the formation of ordered junction zones in which cations can bridge negatively charged chains [9,47,48,49]. Supporting these findings, the EPS-EF displayed noticeable birefringence under polarized light, indicating the presence of some ordered regions (Figure S4).
Elemental and spectroscopic analyses converge on a carbohydrate-associated profile. CHNS/O analysis showed a carbon-rich, nitrogen-poor composition (C ~46%, N ~3%, S negligible). This profile is consistent with EPS structures containing hexoses, uronic acids, and other monosaccharides with low nitrogen content, suggesting a limited contribution from nitrogen-containing components, such as amino sugars or proteinaceous material [50,51,52,53]. FTIR displayed carbohydrate-associated absorptions (O–H at ~3271 cm−1; C–O/C–O–C in the 1300–800 cm−1 fingerprint region) together with carboxylate/carbonyl bands (~1633 and ~1410 cm−1) and weak or poorly defined amide bands. This profile is consistent with hydrated, gel-forming extracellular polymeric networks built on polyhydroxyl ring systems in carbohydrates, with C–H and C–H2 stretching vibrations from aliphatic methyl groups, together with carbonyl (C=O), carboxyl (O–C=O), and peripheral C–N or C=C functional groups typically arising from aldehydes, hexoses, and uronic acids, which may be structurally organized through interactions between the polysaccharides and ions present in different EPS from halophilic microorganisms [9,13,16,22,27,40,54,55]. In the same studies, bands below 1300 cm−1 were related to C–O–C glycosidic bonds, with additional contributions from carbonyl and C–H groups of uronic esters, O-acetyl esters, potentially N–H vibrations from proteins, and S=O groups. However, it is important to emphasize that these interpretations are not proof of polysaccharide purity but rather evidence of physicochemical and molecular characteristics.
Mass spectrometry provided a molecular fingerprint of the EPS-EF rather than a compositional assignment. After subtraction of the solvent blank, the glycan-rich elution window retained a set of oligomeric features, of which the ion at m/z 924.2110 [M–H]− (9.45 min) was taken as representative and subjected to tandem MS. The fragmentation behaviour included recurrent neutral losses corresponding to H2O (dehydration), CO2 (decarboxylation) and acetamide (CH3CONH2), which is compatible with acidic and N-acetylated carbohydrate components [12,56,57,58]. We describe this species as a putative acidic, N-acetylated oligosaccharide-like motif rather than a resolved repeating unit: the accurate mass and fragmentation pattern do not establish monosaccharide identity, sequence, linkage, anomeric configuration or stereochemistry, and no molar ratios can be derived from them. Several low-mass ions were present at the majority of retention times and in the solvent blank, and were therefore excluded as background rather than treated as diagnostic reporter ions. Negative-ion collision-induced dissociation can also produce different mass defects according to the oxygen content of the precursor: low mass defects (<0.35) are characteristic of carbohydrate-derived A/X ions, whereas high mass defects (>0.35) are more characteristic of aliphatic species [12]. On this basis the late-eluting, high-mass-defect ions are compatible with membrane-associated lipids and isoprenoid carriers; because only two precursors underpin these assignments, we regard them as provisional signatures rather than confirmed structures [12,59]. These results describe a chemically heterogeneous extracellular fraction that includes co-eluting membrane-associated components, and the annotations should be regarded as putative rather than as monosaccharide identifications.
The 1H NMR spectrum was consistent with this picture: a residual HDO and anomeric region near 4.7 ppm, a dense carbinolic envelope (3.5–4.4 ppm) from ring protons, and resonances at 1.9–3.0 ppm assignable to acetyl/N-acetyl groups—consistent with acetylated/N-acetylated components suggested by the MS/MS data. No prominent aromatic resonances were observed in the region 6.0–8.0 ppm; however, this does not exclude minor aromatic or protein-associated components. These signals are attributed to the potential anomeric hydrogens of hexoses and other substituted sugars, uronic acids, protons attached to oxygenated carbons within pyranose rings, and acetylated and amino sugar residues, or methyl groups associated with deoxy sugars, aliphatic substituents, or acetylated polysaccharide structures commonly observed in bacterial exopolysaccharides [13,22,40,60,61,62]. The region between 1.0 and 1.5 ppm could be related to the methyl protons of 6-deoxy sugars (such as rhamnose, fucose, and quinovose), terminal methyl groups of lipidic/acyl substituents, or other aliphatic components. We emphasize, however, that 1D 1H NMR alone cannot establish monosaccharide identity, ring configuration, or glycosidic linkage; the NMR data are therefore best regarded as providing complementary molecular-fingerprint information rather than defining monosaccharide composition or structure.
Genomic and transcriptomic data provide a plausible molecular basis for these chemical observations. Genome mining identified candidate loci for sugar uptake (MFS transporters and fruA), uronic-acid and nucleotide-sugar metabolism (ugd), N-acetylated sugar formation (glmU, wecB, wecC), lipid-carrier metabolism (uppS, uppP), chain initiation (wcaJ), and export (kps ABC-transporter components), consistent with the possible acidic, acetylated glycan and the lipid-carrier ions detected by MS. RNA-seq showed condition- and time-dependent upregulation (notably at 18 h) of nucleotide-sugar, glycosyltransferase, and capsule-associated genes, and WGCNA identified a co-expression module positively correlated with the measured emulsification index. This correlation is consistent with coordinated transcriptional regulation of the candidate pathways associated with matrix biosynthesis during the production phase, indicating that EPS production is coordinated at multiple metabolic levels, from carbon assimilation to extracellular matrix assembly, as commonly observed in bacteria [5,10,33,63,64,65].
One aspect of the proposed model requires explicit comment. In the original analysis we noted the co-occurrence of a WcaJ-type initiating enzyme, characteristic of Wzx/Wzy-dependent assembly, with kps-type ABC-transporter export, and considered whether two mechanistically distinct pathways might operate in parallel. Genomic-context analysis now allows the question to be framed more precisely. No Wzx-type flippase and no Wzy-type polymerase is encoded in the genome, and the single O-antigen ligase present is WaaL, which functions in lipid A-core ligation rather than in extracellular polysaccharide polymerisation. V. zhaodongensis BS253 therefore does not encode a complete Wzx/Wzy assembly pathway. What it encodes instead are two independent ABC-transporter-dependent export systems: a kps-type system, in which KpsC and KpsS, a KpsE-type polysaccharide co-polymerase, an ABC permease-ATPase pair and an OPX-family outer-membrane component are distributed across two neighbouring modules of the same chromosomal segment; and a Wzm/Wzt-type exporter embedded within a predicted glycosyltransferase transcription unit 63.9 kb further along the chromosome. Both export units are tightly co-expressed across the RNA-Seq samples, which supports their transcriptional coherence but does not identify their substrates. wcaJ is present, but it lies 73.9 kb beyond the distal end of the kps-containing module and in a separate predicted transcription unit, an arrangement that raises the possibility that wcaJ participates in a glycoconjugate pathway distinct from the EPS-EF; however, the present data do not establish its substrate. Which system exports the EPS-EF characterised here cannot be resolved by genomic context or co-expression alone and will require directed mutagenesis of the ABC permease-ATPase pairs and of the OPX component, followed by compositional analysis of the resulting extracellular fractions. Figure 11C is therefore presented as a working hypothesis rather than as a validated biosynthetic model.
Comparison with related Halomonadaceae places these observations in context and provides the precedent for the gene combination reported here. The kps-type export machinery is conserved across the family: orthologues of kpsC, kpsS, kpsE, the ABC permease-ATPase pair and the OPX component were recovered in all six relatives examined, including H. elongata, and their gene order is preserved, with kpsC-kpsS and the permease-OPX pair adjacent in every genome. The glycan-synthesis genes that lie between these blocks are, in contrast, variable between species. This pattern—a conserved export cassette combined with a variable biosynthetic region—is characteristic of surface-polysaccharide loci and is consistent with the structural diversity reported for Halomonadaceae exopolysaccharides. It also indicates that the combination of a kps-type exporter with the nucleotide-sugar repertoire described here is not unique to V. zhaodongensis BS253, although the particular complement of dTDP-L-rhamnose, N-acetylated-sugar and uronic-acid genes assembled in this strain differs from that of each relative examined.
Placed in the context of previously characterized halophile EPS, the EPS-EF shares some features—features compatible with acidic carbohydrate components, acetylated groups, strong hydration, and ion-sensitive organization—with acidic exopolysaccharides described from Halomonas and related Halomonadaceae [9]. The combined data support the production of an EPS-enriched extracellular fraction with strong emulsifying activity and a putative carbohydrate-rich composition. Molecular and genomic evidence further supports the involvement of nucleotide-sugar metabolism, glycan assembly, lipid-linked intermediates, and polysaccharide export. Although the primary chemical structure remains unresolved, the present findings establish a functional and molecular framework for understanding extracellular polymer production in this halophilic bacterium.
Finally, the limitations of this study should be stated plainly. The present study was designed to characterize the physicochemical, functional, molecular, and genetic features of the EPS-EF rather than to resolve its complete primary structure. Quantitative purity indicators for the lyophilised EPS-EF—total carbohydrate by the phenol–sulfuric method, residual protein by BCA or amino-acid analysis, and nucleic-acid contamination by A260—were not determined, nor was an additional polishing step such as anion-exchange chromatography on DEAE-Sepharose or a second size-exclusion separation applied to obtain a carbohydrate-dominant sub-fraction. The carbohydrate, protein and nucleic-acid mass fractions of the material therefore remain undefined, and this limits the compositional inferences that can be drawn from the FTIR, MS/MS and NMR data. Emulsifying activity was assessed against a single hydrophobic phase (kerosene) and surface tension against a single aqueous subphase (ultrapure water); the substrate range, emulsion stability beyond 24 h, the influence of pH, salinity and temperature on E24 and interfacial tension, and the dose–response relationship including an EC50 were not determined, so it cannot be established whether the EPS-EF is a generalist or a substrate-selective bioemulsifier. Spectral deconvolution of the 1500–1800 cm−1 envelope, FTIR of proteinase K-digested material, and comparative spectra of model uronic-acid polymers were likewise not performed. The conclusions from FTIR, 1D 1H NMR, and MS/MS help to elucidate the EPS-EF composition, but monosaccharide composition analysis (e.g., GC-MS of alditol acetates or HPAEC-PAD), linkage analysis, absolute configuration determination, molecular-weight distribution, and multidimensional 2D/13C NMR will be required in future studies to establish the primary structure. These analyses were beyond the scope of the present functional and microbiological characterization. Moreover, the absolute molecular weight and polydispersity of the EPS-EF were not determined in the present study because the size-exclusion procedure was used for fractionation and activity-based selection rather than calibrated molecular-weight determination. Accordingly, the m/z values detected by UHPLC–QTOF–MS/MS are interpreted as those of molecular ions and oligomeric/fragment species rather than as the molecular weight of the intact extracellular polymeric fraction. The proposed ion-coordination model awaits direct cation identification, and the biosynthetic model awaits functional genetic validation. Addressing these points in future work—together with an assessment of the EPS-EF emulsion stability, rheology, and metal-binding capacity under application-relevant conditions—would substantiate the biotechnological potential suggested here for bioremediation, emulsification, and biomaterial development, and would allow the primary structure of this Vreelandella-derived EPS to be defined with confidence.
5. Conclusions
This study characterized an EPS-enriched extracellular fraction (EPS-EF) produced by Vreelandella zhaodongensis BS253 using an integrated physicochemical, molecular, functional, genomic, and transcriptomic approach. The EPS-EF exhibited a carbohydrate-rich, ion-sensitive extracellular matrix with emulsifying activity against kerosene and a concentration-dependent reduction in surface tension. Complementary elemental composition, FTIR, UHPLC–ESI–QTOF–MS/MS, 1H NMR, microscopy, XRD, and TGA analyses supported the presence of a heterogeneous extracellular material containing carbohydrate-associated features, while also indicating contributions from other co-extracted components. Genomic and transcriptomic analyses identified candidate pathways potentially involved in nucleotide-sugar metabolism, glycan assembly, lipid-linked intermediates, and extracellular export, which are summarized as a working hypothesis rather than a validated biosynthetic model. Importantly, the present study does not establish the complete primary structure, molecular-weight distribution, monosaccharide composition, linkage pattern, or stereochemistry of the EPS-EF. Further purification and orthogonal structural analyses will therefore be required to resolve these parameters and define the molecular basis of its emulsifying properties.
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/bacteria5040064/s1, Figure S1: Extraction and enrichment of EPS-EF; Figure S2: Microscopic observation of emulsification; Figure S3: Transmission electron microscopy (TEM) images after culture were treated with pH 9; Figure S4: Polarized microscopy of EPS-EF; Figure S5: V. zhaodongensis BS253 EPS-EF fragments between 3.3–6 min in negative mode [M-H]-; Figure S6: V. zhaodongensis BS253 EPS-EF fragments between 6–9.5 min in negative mode [M-H]-; Figure S7: V. zhaodongensis BS253 EPS-EF fragments between 9.8–24 min in negative mode [M-H]-; Figure S8: V. zhaodongensis BS253 EPS-EF fragments between 24.5–31.2 min in negative mode [M-H]-; Figure S9: Genomic context of the EPS-related candidate loci in V. zhaodongensis BS253; Figure S10: Predicted transcription units of V. zhaodongensis BS253 are co-expressed; Figure S11: Conservation of the polysaccharide-export locus across related Halomonadaceae; Table S1: Chromatography gradient used for UHPLC analysis; Table S2: Ions detected at each retention time (RT) after blank subtraction; Table S3. Coordinates, strand, intergenic distances, predicted transcription units, conserved-domain assignments and with-in-unit expression correlations for the coding sequences of the three EPS-related chromosomal modules of V. zhaodongensis BS253 (RefSeq GCF_051175405.1); Table S4: Correspondence between the gene symbols used in this study and the locus tags, product descriptions and con-served-domain assignments of RefSeq assembly GCF_051175405.1; Table S5: Orthologues of the EPS-related candidate loci of V. zhaodongensis BS253 in six related Halomonadaceae genomes, with amino-acid identities from reciprocal best DIAMOND hits, and the number of intervening coding sequences between orthologues of adjacent BS253 genes.
Author Contributions
H.A.d.B.: conceptualization, methodology, formal analysis, investigation, data curation, writing—original draft preparation, visualization. A.P.N.: investigation, visualization, writing—review and editing. D.U.H.d.B., E.V.A., L.S.D. and R.S.S.: investigation. F.M.d.S., C.L.M. and J.P.T.Z.: project administration. C.C.S.: writing—review and editing and supervision. All authors have read and agreed to the published version of the manuscript.
Funding
This research was funded by Petronas Petróleo Brasil Ltd.a (ANP number: 22140-8) and by the Brazilian National Agency of Petroleum, Natural Gas and Biofuels through the R&D levy regulation.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
The original contributions presented in the study are included in the article and Supplementary Materials, further inquiries can be directed to the corresponding author/s. Additional raw data and its statistical calculus in the study are publicly available in Zenodo at https://doi.org/10.5281/zenodo.22967314. Genomic additional data are publicly available in the NCBI BioProject under accession number PRJNA1275882 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1275882, accessed in 28 September 2026). The genome annotation underlying the genomic-context analysis is publicly available in the NCBI Assembly database under accession GCF_051175405.1 (chromosome NZ_CP195113.1); the comparative analysis used the publicly available assemblies GCF_013415115.1, GCF_054110395.1, GCF_019722725.1, GCF_012427705.1, GCF_014697095.1 and GCF_000196875.2.
Acknowledgments
The authors would like to acknowledge the support of ANP (Brazil’s National Oil, Natural Gas and Biofuels Agency) through the R&D levy regulation. During the preparation of this manuscript/study, the author(s) used Chat GPT 4 for the purposes of grammar correction, sentence structure, spelling, punctuation, formatting, and improvement of English language clarity and readability. The authors have reviewed and edited the output and take full responsibility for the content of this publication.
Conflicts of Interest
The authors declare that this study received funding from Petronas Petróleo Brasil Ltd.a. The funder was not involved in the study design, data collection, analysis or interpretation of data, the writing of this article, or the decision to submit it for publication.
References
- Casillo, A.; Lanzetta, R.; Parrilli, M.; Corsaro, M.M. Exopolysaccharides from Marine and Marine Extremophilic Bacteria: Structures, Properties, Ecological Roles and Applications. Mar. Drugs 2018, 16, 69. [Google Scholar] [CrossRef] [Scilit]
- Wang, X.; Liu, M.; Yu, C.; Li, J.; Zhou, X. Biofilm Formation: Mechanistic Insights and Therapeutic Targets. Mol. Biomed. 2023, 4, 49. [Google Scholar] [CrossRef] [Scilit]
- Kaur, N.; Dey, P. Bacterial Exopolysaccharides as Emerging Bioactive Macromolecules: From Fundamentals to Applications. Res. Microbiol. 2023, 174, 104024. [Google Scholar] [CrossRef] [Scilit]
- Joe, J.; Merlin, T.; Nair, I.C. Ecological and Economic Importance of Exopolysaccharides Produced by Microorganisms: A Review. J. Pure Appl. Microbiol. 2026, 20, 1021–1040. [Google Scholar] [CrossRef] [Scilit]
- Mouro, C.; Gomes, A.P.; Gouveia, I.C. Microbial Exopolysaccharides: Structure, Diversity, Applications, and Future Frontiers in Sustainable Functional Materials. Polysaccharides 2024, 5, 241–287. [Google Scholar] [CrossRef] [Scilit]
- Nguyen, H.-T.; Pham, T.-T.; Nguyen, P.-T.; Le-Buanec, H.; Rabetafika, H.N.; Razafindralambo, H.L. Advances in Microbial Exopolysaccharides: Present and Future Applications. Biomolecules 2024, 14, 1162. [Google Scholar] [CrossRef] [Scilit]
- Oliveira, J.M.; Radhouani, H.; Reis, R.L. Polysaccharides of Microbial Origin, 1st ed.; Oliveira, J.M., Radhouani, H., Reis, R.L., Eds.; Springer International Publishing: Cham, Switzerland, 2022; Volume 1, ISBN 978-3-030-42214-1. [Google Scholar]
- Upadhyaya, C.; Patel, H.; Patel, I.; Upadhyaya, T. Extremophilic Exopolysaccharides: Bioprocess and Novel Applications in 21st Century. Fermentation 2025, 11, 16. [Google Scholar] [CrossRef] [Scilit]
- Sabroso, E.; Martínez, J.M.; Sánchez-León, E.; Rodríguez, N.; Amils, R.; Abrusci, C. Production and Characterisation of an Exopolysaccharide by Vreelandella titanicae Zn11_249 Isolated from Salar de Uyuni (Bolivia). Polymers 2025, 17, 2362. [Google Scholar] [CrossRef] [Scilit]
- Gan, L.; Huang, X.; He, Z.; He, T. Exopolysaccharide Production by Salt-Tolerant Bacteria: Recent Advances, Current Challenges, and Future Prospects. Int. J. Biol. Macromol. 2024, 264, 130731. [Google Scholar] [CrossRef] [Scilit]
- de la Haba, R.R.; Arahal, D.R.; Sánchez-Porro, C.; Chuvochina, M.; Wittouck, S.; Hugenholtz, P.; Ventosa, A. A Long-Awaited Taxogenomic Investigation of the Family Halomonadaceae. Front. Microbiol. 2023, 14, 1293707. [Google Scholar] [CrossRef] [Scilit]
- Harvey, D.J. Negative Ion Mass Spectrometry for the Analysis of N-Linked Glycans. Mass Spectrom. Rev. 2020, 39, 586–679. [Google Scholar] [CrossRef] [Scilit]
- Halder, U.; Mazumder, K.; Kumar, K.J.; Bandopadhyay, R. Structural Insight into a Glucomannan-Type Extracellular Polysaccharide Produced by a Marine Bacillus altitudinis SORB11 from Southern Ocean. Sci. Rep. 2022, 12, 16322. [Google Scholar] [CrossRef] [Scilit]
- Kuklewicz, J.; Zimmer, J. Molecular Insights into Capsular Polysaccharide Secretion. Nature 2024, 628, 901–909. [Google Scholar] [CrossRef] [Scilit]
- Napp, A.P.; de Brito, H.A.; de Brito, D.U.H.; Abati, E.V.; dos Santos, F.M.; Melo, C.L.; Zielinski, J.P.T.; Staats, C.C. Inorganic Carbon Modulates Emulsification Activity and Transcriptional Responses in Vreelandella zhaodongensis BS253. Molecules 2026, 31, 2182. [Google Scholar] [CrossRef] [Scilit]
- Athmika; Ghate, S.D.; Arun, A.B.; Rao, S.S.; Kumar, S.T.A.; Kandiyil, M.K.; Saptami, K.; Rekha, P.D. Genome Analysis of a Halophilic Bacterium Halomonas malpeensis YU-PRIM-29T Reveals Its Exopolysaccharide and Pigment Producing Capabilities. Sci. Rep. 2021, 11, 1749. [Google Scholar] [CrossRef] [Scilit]
- Kąkol, M.; Tagliasacchi, E.; Borkowski, A.; Słowakiewicz, M. Influence of Different Sample Preparation Techniques on Imaging Viruses and Virus-like Particles by Scanning Electron and Scanning Transmission Electron Microscopes. Front. Microbiol. 2023, 14, 1279720. [Google Scholar] [CrossRef] [Scilit]
- Pirolli, N.H.; Jay, S.M. Analysis of Bacterial Extracellular Vesicles by Immunogold Transmission Electron Microscopy. In Bacterial Extracellular Vesicles: Methods and Protocols; Springer US: New York, NY, USA, 2024; pp. 15–23. [Google Scholar]
- Lu, L.; Zhao, Y.; Li, M.; Wang, X.; Zhu, J.; Liao, L.; Wang, J. Contemporary Strategies and Approaches for Characterizing Composition and Enhancing Biofilm Penetration Targeting Bacterial Extracellular Polymeric Substances. J. Pharm. Anal. 2024, 14, 100906. [Google Scholar] [CrossRef] [Scilit]
- Han, A.; Lee, S.-Y. An Overview of Various Methods for In Vitro Biofilm Formation: A Review. Food Sci. Biotechnol. 2023, 32, 1617–1629. [Google Scholar] [CrossRef] [Scilit]
- Brito, H.A.; Napp, A.P.; Pereira, E.; Bach, E.; Borowski, J.V.B.; Passaglia, L.M.P.; Melo, V.M.M.; Moreira, R.; Foster, E.J.; Lopes, F.C.; et al. Enhanced Low-Cost Lipopeptide Biosurfactant Production by Bacillus velezensis from Residual Glycerin. Bioprocess Biosyst. Eng. 2024, 47, 1555–1570. [Google Scholar] [CrossRef] [Scilit]
- Biswas, J.; Ganguly, J.; Paul, A.K. Partial Characterization of an Extracellular Polysaccharide Produced by the Moderately Halophilic Bacterium Halomonas xianhensis SUR308. Biofouling 2015, 31, 735–744. [Google Scholar] [CrossRef] [Scilit]
- Jia, F.; Yang, Q.; Liu, X.; Li, X.; Li, B.; Zhang, L.; Peng, Y. Stratification of Extracellular Polymeric Substances (EPS) for Aggregated Anammox Microorganisms. Environ. Sci. Technol. 2017, 51, 3260–3268. [Google Scholar] [CrossRef] [Scilit]
- Lautert-Dutra, W.; Napp, A.P.; Back Sivinski, E.A.; Staats, C.C.; dos Santos, F.M.; Melo, C.L. Complete Genome Sequence and Metabolic Features of Vreelandella zhaodongensis BS253: A New Isolate from Hypersaline Lakes from Brazilian Pantanal. bioRxiv 2026. [Google Scholar] [CrossRef] [Scilit]
- García, A.; Fernández-Sandoval, M.T.; Morales-Guzmán, D.; Martínez-Morales, F.; Trejo-Hernández, M.R. Advances in Exopolysaccharide Production from Marine Bacteria. J. Chem. Technol. Biotechnol. 2022, 97, 2694–2705. [Google Scholar] [CrossRef] [Scilit]
- Arias, S.; del Moral, A.; Ferrer, M.R.; Tallon, R.; Quesada, E.; Béjar, V. Mauran, an Exopolysaccharide Produced by the Halophilic Bacterium Halomonas maura, with a Novel Composition and Interesting Properties for Biotechnology. Extremophiles 2003, 7, 319–326. [Google Scholar] [CrossRef] [Scilit]
- Mukherjee, P.; Mitra, A.; Roy, M. Halomonas rhizobacteria of Avicennia Marina of Indian Sundarbans Promote Rice Growth Under Saline and Heavy Metal Stresses Through Exopolysaccharide Production. Front. Microbiol. 2019, 10, 1207. [Google Scholar] [CrossRef] [Scilit]
- Cuthbertson, L.; Mainprize, I.L.; Naismith, J.H.; Whitfield, C. Pivotal Roles of the Outer Membrane Polysaccharide Export and Polysaccharide Copolymerase Protein Families in Export of Extracellular Polysaccharides in Gram-Negative Bacteria. Microbiol. Mol. Biol. Rev. 2009, 73, 155–177. [Google Scholar] [CrossRef] [Scilit]
- Amjres, H.; Béjar, V.; Quesada, E.; Carranza, D.; Abrini, J.; Sinquin, C.; Ratiskol, J.; Colliec-Jouault, S.; Llamas, I. Characterization of Haloglycan, an Exopolysaccharide Produced by Halomonas stenophila HK30. Int. J. Biol. Macromol. 2015, 72, 117–124. [Google Scholar] [CrossRef] [Scilit]
- Cui, Y.; Zhang, S.; Zhao, P.; Cui, J.; Song, S.; Qu, Y.; Zhang, H.; Ma, D. Assessment System and Optimization of the Thermal Extraction Methods for Extracellular Polymeric Substances (EPS) from Microcystis. Microorganisms 2026, 14, 116. [Google Scholar] [CrossRef] [Scilit]
- Hu, X.; Huang, Y.-Y.; Wang, Y.; Wang, X.; Hamblin, M.R. Antimicrobial Photodynamic Therapy to Control Clinically Relevant Biofilm Infections. Front. Microbiol. 2018, 9, 1299. [Google Scholar] [CrossRef] [Scilit]
- Zhao, A.; Sun, J.; Liu, Y. Understanding Bacterial Biofilms: From Definition to Treatment Strategies. Front. Cell. Infect. Microbiol. 2023, 13, 1137947. [Google Scholar] [CrossRef] [Scilit]
- Zang, J.; Lin, T.; Shi, Y.; Lin, Y.; Xu, C.; Ma, K.; Zhang, C.; Rui, X.; Gan, D.; Li, W. Recent Advances in Exopolysaccharide–Protein Interactions in Fermented Dairy- and Plant-Based Yogurts: Mechanisms, Influencing Factors, Health Benefits, Analytical Techniques, and Future Directions. Compr. Rev. Food Sci. Food Saf. 2025, 24, e70219. [Google Scholar] [CrossRef] [Scilit]
- Noble, J.E.; Bailey, M.J.A. Chapter 8 Quantitation of Protein. Methods Enzymol. 2009, 463, 73–95. [Google Scholar] [CrossRef] [Scilit]
- Uzoigwe, C.; Burgess, J.G.; Ennis, C.J.; Rahman, P.K.S.M. Bioemulsifiers Are Not Biosurfactants and Require Different Screening Approaches. Front. Microbiol. 2015, 6, 47–52. [Google Scholar] [CrossRef] [Scilit]
- Dhasayan, A.; Kiran, G.S.; Selvin, J. Production and Characterisation of Glycolipid Biosurfactant by Halomonas Sp. MB-30 for Potential Application in Enhanced Oil Recovery. Appl. Biochem. Biotechnol. 2014, 174, 2571–2584. [Google Scholar] [CrossRef] [Scilit]
- Uad, I.; Silva-Castro, G.A.; Abrusci, C.; Catalina, F.; González-López, J.; Manzanera, M.; Calvo, C. Production Index: A New Index to Evaluate EPSs as Surfactants and Bioemulsifiers Applied to Halomonas variabilis Strain W10 for Hydrocarbon Bioremediation. Ecotoxicol. Environ. Saf. 2019, 175, 66–73. [Google Scholar] [CrossRef] [Scilit]
- Nikolova, C.; Morris, G.; Ellis, D.; Bowler, B.; Jones, M.; Mulloy, B.; Gutierrez, T. Characterization of the Surface-Active Exopolysaccharide Produced by Halomonas sp. TGOS-10: Understanding Its Role in the Formation of Marine Oil Snow. PLoS ONE 2024, 19, e0299235. [Google Scholar] [CrossRef] [Scilit]
- Berry, J.D.; Neeson, M.J.; Dagastine, R.R.; Chan, D.Y.C.; Tabor, R.F. Measurement of Surface and Interfacial Tension Using Pendant Drop Tensiometry. J. Colloid. Interface Sci. 2015, 454, 226–237. [Google Scholar] [CrossRef] [Scilit]
- Joulak, I.; Finore, I.; Poli, A.; Abid, Y.; Bkhairia, I.; Nicolaus, B.; Di Donato, P.; Dal Poggetto, G.; Gharsallaoui, A.; Attia, H.; et al. Hetero-Exopolysaccharide from the Extremely Halophilic Halomonas smyrnensis K2: Production, Characterization and Functional Properties in Vitro. 3 Biotech 2020, 10, 395. [Google Scholar] [CrossRef] [Scilit]
- Kavitake, D.; Delattre, C.; Devi, P.B.; Pierre, G.; Michaud, P.; Shetty, P.H.; Andhare, P. Physical and Functional Characterization of Succinoglycan Exopolysaccharide Produced by Rhizobium radiobacter CAS from Curd Sample. Int. J. Biol. Macromol. 2019, 134, 1013–1021. [Google Scholar] [CrossRef] [Scilit]
- Vall, M.; Hultberg, J.; Strømme, M.; Cheung, O. Inorganic Carbonate Composites as Potential High Temperature CO2 Sorbents with Enhanced Cycle Stability. RSC Adv. 2019, 9, 20273–20280. [Google Scholar] [CrossRef] [Scilit]
- Can, H.K.; Kavlak, S.; Gurbuz, F.; Odabaşı, M. Insights into the Viscoelastic Peculiarities of Cyanobacterial Extracellular Polymeric Substance (EPS). J. Polym. Environ. 2022, 30, 3055–3062. [Google Scholar] [CrossRef] [Scilit]
- Li, L.; Fang, Y.; Vreeker, R.; Appelqvist, I.; Mendes, E. Reexamining the Egg-Box Model in Calcium−Alginate Gels with X-Ray Diffraction. Biomacromolecules 2007, 8, 464–468. [Google Scholar] [CrossRef] [Scilit]
- Singh, R.P.; Shukla, M.K.; Mishra, A.; Kumari, P.; Reddy, C.R.K.; Jha, B. Isolation and Characterization of Exopolysaccharides from Seaweed Associated Bacteria Bacillus licheniformis. Carbohydr. Polym. 2011, 84, 1019–1026. [Google Scholar] [CrossRef] [Scilit]
- Asker, M.S.; El Sayed, O.H.; Mahmoud, M.G.; Yahya, S.M.; Mohamed, S.S.; Selim, M.S.; El Awady, M.S.; Abdelnasser, S.M.; Abo Elsoud, M.M. Production of Exopolysaccharides from Novel Marine Bacteria and Anticancer Activity against Hepatocellular Carcinoma Cells (HepG2). Bull. Natl. Res. Cent. 2018, 42, 30. [Google Scholar] [CrossRef] [Scilit]
- Leśniewski, B.; Kotula, M.; Kubiak, A.; Pajewska-Szmyt, M. Thermostability of Selected Biological Materials. Lett. Appl. NanoBioScience 2022, 12, 88. [Google Scholar] [CrossRef] [Scilit]
- Benli, B. Effect of Borax Addition on the Structural Modification of Bentonite in Biodegradable Alginate-based Biocomposites. J. Appl. Polym. Sci. 2013, 128, 4172–4180. [Google Scholar] [CrossRef] [Scilit]
- Nörnberg, A.B.; Gehrke, V.R.; Mota, H.P.; Camargo, E.R.; Fajardo, A.R. Alginate-Cellulose Biopolymeric Beads as Efficient Vehicles for Encapsulation and Slow-Release of Herbicide. Colloids Surf. A Physicochem. Eng. Asp. 2019, 583, 123970. [Google Scholar] [CrossRef] [Scilit]
- Kiran, N.S.; Yashaswini, C.; Singh, S.; Prajapati, B.G. Revisiting Microbial Exopolysaccharides: A Biocompatible and Sustainable Polymeric Material for Multifaceted Biomedical Applications. 3 Biotech 2024, 14, 95. [Google Scholar] [CrossRef] [Scilit]
- Yang, R.; Liu, L.; Gao, D.; Zhao, D. Purification, Structural Characterization, and Bioactive Properties of Exopolysaccharides from Saccharomyces cerevisiae HD-01. Front. Bioeng. Biotechnol. 2024, 12, 1455708. [Google Scholar] [CrossRef] [Scilit]
- Ning, Y.; Cao, H.; Zhao, S.; Gao, D.; Zhao, D. Structure and Properties of Exopolysaccharide Produced by Gluconobacter frateurii and Its Potential Applications. Polymers 2024, 16, 1004. [Google Scholar] [CrossRef] [Scilit]
- Oliva, R.L.; Khadka, U.B.; Camenzind, T.; Dyckmans, J.; Joergensen, R.G. Constituent of Extracellular Polymeric Substances (EPS) Produced by a Range of Soil Bacteria and Fungi. BMC Microbiol. 2025, 25, 298. [Google Scholar] [CrossRef] [Scilit]
- Deshmukh, S.; Kanekar, P.; Bhadekar, R. Production and characterization of exopolysaccharide from marine moderately halophilic bacterium Halomonas smyrnensis SVD III. Int. J. Pharm. Pharm. Sci. 2017, 9, 146. [Google Scholar] [CrossRef] [Scilit]
- Maajid, H.S.; Nurliyani, N.; Widodo, W. Exopolysaccharide Production in Fermented Milk Using Lactobacillus casei Strains AP and AG. AIMS Microbiol. 2022, 8, 138–152. [Google Scholar] [CrossRef] [Scilit]
- Grabarics, M.; Lettow, M.; Kirschbaum, C.; Greis, K.; Manz, C.; Pagel, K. Mass Spectrometry-Based Techniques to Elucidate the Sugar Code. Chem. Rev. 2022, 122, 7840–7908. [Google Scholar] [CrossRef] [Scilit]
- Dadovska, L.; Paskova, V.; Novak, P.; Hrabak, J. Mass Spectrometric Profiling of Microbial Polysaccharides Using Laser Desorption/Ionization—Time-of-Flight (LDI-TOF) and Liquid Chromatography/Mass Spectrometry (LC/MS): A Novel Method for Structural Fingerprinting and Derivatization. Front. Cell. Infect. Microbiol. 2025, 15, 1658802. [Google Scholar] [CrossRef] [Scilit]
- Helms, A.; Brodbelt, J.S. Mass Spectrometry Strategies for O-Glycoproteomics. Cells 2024, 13, 394. [Google Scholar] [CrossRef] [Scilit]
- Jin, C.; Harvey, D.J.; Struwe, W.B.; Karlsson, N.G. Separation of Isomeric O- Glycans by Ion Mobility and Liquid Chromatography–Mass Spectrometry. Anal. Chem. 2019, 91, 10604–10613. [Google Scholar] [CrossRef] [Scilit]
- Meena, S.V.K.M.; Tripathi, A.D.; Ts, R.L. Optimization and Characterization of Alginic Acid Synthesized from a Novel Strain of Pseudomonas stutzeri. Biotechnol. Rep. 2020, 27, e00517. [Google Scholar] [CrossRef] [Scilit]
- McMurry, J. Organic Chemistry; OpenStax: Houston, TX, USA, 2023; ISBN 9781951693985. [Google Scholar]
- Yamauchi, K.; Kuroki, S.; Fujii, K.; Ando, I. The Amide Proton NMR Chemical Shift and Hydrogen-Bonded Structure of Peptides and Polypeptides in the Solid State as Studied by High-Frequency Solid-State NMR. Chem. Phys. Lett. 2000, 324, 435–439. [Google Scholar] [CrossRef] [Scilit]
- Netrusov, A.I.; Liyaskina, E.V.; Kurgaeva, I.V.; Liyaskina, A.U.; Yang, G.; Revin, V.V. Exopolysaccharides Producing Bacteria: A Review. Microorganisms 2023, 11, 1541. [Google Scholar] [CrossRef] [Scilit]
- Wang, W.; Ju, Y.; Liu, N.; Shi, S.; Hao, L. Structural Characteristics of Microbial Exopolysaccharides in Association with Their Biological Activities: A Review. Chem. Biol. Technol. Agric. 2023, 10, 137. [Google Scholar] [CrossRef] [Scilit]
- Wünsche, J.; Schmid, J. Acetobacteraceae as Exopolysaccharide Producers: Current State of Knowledge and Further Perspectives. Front. Bioeng. Biotechnol. 2023, 11, 1166618. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.










