Exopolysaccharide (EPS) Produced by Leuconostoc mesenteroides SJC113: Characterization of Functional and Technological Properties and Application in Fat-Free Cheese

Abstract

A Leuconostoc mesenteroides strain (SJC113) isolated from cheese curd was found to produce large amounts of a mucoid exopolysaccharide (EPS). An analysis revealed the glucan nature of the EPS with 84.5% (1→6)-linked α-d-glucose units and 5.6% (1,3→6)-linked α-d-glucose units as branching points. The EPS showed 52% dextranase resistance and a yield of 7.4 ± 0.9 g/L from MRS medium supplemented with 10% sucrose within 48 h. Ln. mesenteroides SJC113 was also characterized and tested for the production of EPS as a fat substitute in fresh cheese. Strain SJC113 showed high tolerance to a wide range of NaCl concentrations (2, 5 and 10%), high β-galactosidase activity (2368 ± 24 Miller units), cholesterol-reducing ability (14.8 ± 4.1%), free radical scavenging activity (11.7 ± 0.7%) and hydroxyl scavenging activity (15.7 ± 0.4%). The strain had no virulence genes and was sensitive to clinically important antibiotics such as ampicillin, tetracycline and chloramphenicol. Ln. mesenteroides SJC113 produced highly viscous EPS during storage at 8 °C in skim milk with 5% sucrose. Therefore, these conditions were used for EPS production in skim milk before incorporation into fresh cheese. Four types of fresh cheese were produced: full-fat cheese (FF) made from pasteurized whole milk, non-fat cheese (NF) made from pasteurized skim milk, non-fat cheese made from skim milk fermented with Ln. mesenteroides without added sugar (NFLn0) and non-fat cheese made from skim milk fermented with Ln. mesenteroides with 5% sucrose (NFLn5). While the NF cheeses had the highest viscosity and hardness, the NFLn5 cheeses showed lower firmness and viscosity, higher water-holding capacity and lower weight loss during storage. Overall, the NFLn5 cheeses had similar rheological properties to full-fat cheeses with a low degree of syneresis. It was thus shown that the glucan-type EPS produced by Ln. mesenteroides SJC113 can successfully replace fat without altering the texture of fresh cheese.

1. Introduction

Lactic acid bacteria (LAB) are an extensive group of food-grade bacteria that comprise many genera and are widely distributed in dairy products during the fermentation process [1]. Due to their probiotic and technological properties, they have attracted widespread interest in research [2]. One of these properties is the ability to produce exopolysaccharides (EPS) with different structures and properties. Microbial EPS generally exist in two forms, depending on their location. EPS produced by LAB are often distinguished as ropy or slime EPS [3,4].

EPS can be synthesized extracellularly or excreted into the extracellular environment by the producing microorganisms [5]. They can appear as capsular exopolysaccharides associated with the cell surface or as exopolysaccharides secreted into the environment as free polymers, with a filamentous appearance or in viscous secretions. Although both homopolymeric and heteropolymeric EPS can be produced by LAB, the interesting technological properties belong to homopolymeric EPS [6]. They can be divided into four groups—α-glucans, β-glucans, fructans and polygalactan. The α-glucan group contains dextrans, which consist of linear (1→6)-linked α-d-glucose units that can be branched with (1→2), (1→3) or (1→4)-linked α-d-glucose units. These dextrans have been used in various industrial and commercial applications, for example, to thicken and stabilize fermented dairy products and to improve the production yield and rheological properties of low-fat cheeses [6,7,8,9].

According to Park et al., the predominant LAB group of EPS producers are Leuconostoc species. In the early 1950s, it was discovered that Leuconostoc mesenteroides strains produce high molecular weight dextran when sucrose is added to the culture medium [10]. Indeed, Leuconostoc spp. have been shown to produce EPS with different structures and properties that influence the sensory and rheological properties of food [11,12].

In the food industry, the in-situ production of EPS has attracted particular interest due to their functional properties as thickening, stabilizing, emulsifying, texturizing and gelling agents [6]. These EPS improve the texture, stability and sensory properties, such as creaminess, of fermented dairy products, which include yogurt, fermented beverages, cheese and desserts [13]. In addition to the technological benefits, certain EPS exhibit physiological effects that are beneficial to the consumer. These health benefits include the stable growth of bacteria during gastrointestinal stress, microbial immunomodulation and antimicrobial properties against pathogens in the digestive tract [5,14,15,16,17,18,19].

In recent years, consumers have shown an increasing interest in reducing the fat content of dairy products [20,21]. One of the main characteristics of reduced-fat cheese is a lower moisture content, resulting in a more compact and rubbery texture [22,23]. It is necessary to compensate for the deficiencies caused by fat removal and to prepare low-fat cheese that has characteristics comparable to full-fat cheese to improve consumer acceptance. To compensate for the reduced fat content in the matrix, the moisture content of the cheese can be increased using various techniques, e.g., by adding ingredients that increase the water-holding capacity [24,25]. Several studies have shown that the water-holding capacity of low-fat cheese can be increased by polysaccharide hydrocolloids, whether they are produced in situ by LAB or used as additives from microbial, algal or plant sources [24]. Polysaccharide-based fat replacers have been shown to alter the microstructural properties of low-fat cheese and mimic some of the textural and sensory properties of fat molecules [26]. For example, the addition of beta-glucan from the mushroom Pleurotus ostreatus has been shown to improve the texture and organoleptic properties of low-fat white-brined cheese [27]. However, some studies suggest that EPS produced in situ have advantages in increasing moisture retention and thus improving the textural properties of fat-reduced cheese [28,29,30]. In this context, the addition of an EPS-producing culture to low-fat Kasar cheese resulted in a less compact protein matrix and a sponge-like structure [31]. Similarly, the addition of a ropy EPS producer in the manufacture of a low-fat cheddar cheese resulted in a cheese with the same texture and melting properties as the full-fat type [26]. Moreover, the use of LAB cultures producing EPS in situ during fermentation could be a suitable alternative for products whose addition of polysaccharides requires labeling as food additives, which is not appreciated by consumers [14].

Recently, a strain of Leuconostoc mesenteroides (SJC113) was isolated from an artisanal cheese with a protected designation of origin (São Jorge cheese from the Azores, Portugal) and was found to produce large amounts of EPS from sucrose. In the present study, the EPS produced by this strain was characterized, and the functional and technological properties of this strain were evaluated. In addition, the application of in-situ-produced EPS as a fat substitute in fresh cheese was investigated.

2. Materials and Methods

2.1. Materials

The following materials were used: MRS broth and agar (Biokar Diagnostics, Allonne, France), skim milk (VWR Chemicals, Leuven, Belgium), PCA (Biokar Diagnostics, France), sucrose (Sigma, Steinheim, Germany), glucose (Sigma, Saint-Quentin-Fallavier, France), lactose (Scharlau, Barcelona, Spain), fructose (Sigma, Burlington, MA, USA), TCA (PanReac AppliChem ITW reagents, Darmstadt, Germany), DNAse type I (Amresco, Solon, OH, USA), pronase E (Amresco, USA), Chaetomium erraticum (Sigma, Søborg, Denmark), pancreatin (Sigma), DPPH (Sigma, Germany), brilliant green (Sigma, India), ascorbic acid (Riedel-de Haën, Seelze, Germany), BHT (Sigma, Spain), cholesterol (Sigma-Aldrich, USA), NaCl (Merck, Darmstadt, Germany), methanol (Fisher Chemical, Loughborough, UK), PBS (Sigma, USA), Fe₂SO₄ (Riedel-de Haën, Berlin, Germany), H₂O₂ (Panreac Química SAU, Barcelona, Spain), dialysis membrane (10 kDa, Spectrum Laboratories Inc., Piscataway, NJ, USA), CaCl₂ (Merck, Germany, Darmstadt) and rennet (Lusocoalho, Portugal).

2.2. Bacteria

The strain Ln. mesenteroides SJC113 was isolated from the curd during cheese-making of São Jorge cheese (The Azores, Portugal). The 16S rDNA sequences of the strain were deposited in the GenBank under accession number MT742947. Stock cultures were kept at −80 °C in 30% (v/v) glycerol and propagated twice in MRS broth with 1% (v/v) of inoculum at 30 °C for 24 h.

2.3. Production and Purification of EPS

The bacterial culture (1%) was inoculated into 50 mL MRS broth and incubated overnight at 30 °C. The culture was then transferred to 1 L of modified MRS broth, in which the glucose was replaced with 10% (w/v) sucrose and incubated at 30 °C for 48 h. After fermentation, the EPS was extracted according to Domingos-Lopes et al. [32] with some modifications. Briefly, the medium was centrifuged at 9000× g for 5 min at 4 °C to remove the bacterial cells. The EPS was precipitated by adding 2 volumes of chilled ethanol at 4 °C for 48 h. After centrifugation at 10,000× g for 20 min at 4 °C, the precipitate was dissolved in Milli-Q water and dialyzed against Milli-Q water at 4 °C for 3 days. The water was changed twice daily, the crude EPS was obtained by freeze-drying (Scanvac Coolsafe 55-4 Pro, Labogene, Allerød, Denmark) and the dialysate sample was stored at −20 °C for further purification. The extraction procedure was repeated four times to determine the EPS yield, which was quantified using the phenol-sulfuric acid method [33]. In the second purification (to analyze EPS composition and linkage), the crude EPS (5 mg/mL) was dissolved in 50 mM Tris–HCl, 10 mM MgSO₄.7H₂O (pH 7.5) and was incubated with DNAse type-I (2.5 μg/mL) for 6 h at 37 °C and Pronase E (50 μg/mL in 50 mM Tris–HCl, 2% EDTA, pH 7.5) for 18h at 37 °C, as described by Domingos-Lopes et al. [32].

2.4. Characterization of EPS

2.4.1. Monosaccharide Composition and Linkage Analysis

The neutral sugar composition and linkage analysis were determined by gas chromatography (GC) and alditol acetates, as described by Nunes et al. [34]. Briefly, EPS hydrolysis was performed with 2M trifluoroacetic acid (TFA) for 1 h at 120 °C. The monosaccharides were reduced by NaBH₄ and were acetylated with acetic anhydride and 1-methylimidazole. To separate the alditol acetate derivatives, dichloromethane was used, and after the separation, they were analyzed by GC with an FID detector.

Determination of uronic acids was performed using the colorimetric method described by Blumenkrantz and Asboe-Hansen [35] with pre-hydrolysis of samples, as described by Coimbra et al. [36].

The linkage analysis of the monosaccharides was performed by methylation, as reported by Ciucanu and Kerek [37] with modifications [36]. The partially methylated alditol acetates were separated and analyzed by gas chromatography quadrupole mass spectrometry (GCqMS) (GC-2010 Plus, Shimadzu, Kyoto, Japan) using a non-polar column HT5 (30 m length, 0.25 mm internal diameter and 0.10 μm stationary phase, Trajan, Melbourne, Australia), as described by Concórdio-Reis et al. [38].

2.4.2. Dextranase Resistance

The susceptibility of the EPS to dextranase was performed using 1,6-α-d-glucan 6-glucanohydrolase from Chaetomium erraticum (D0443, Sigma), as described by Domingos-Lopes et al. [32].

2.4.3. Protein Content

Protein content was determined using an elemental analyzer Truspec (630–200-200) with a thermal conductivity detector, as described by Bastos et al. [39]. The conversion from % N to protein content was performed considering the Kjeldahl factor of 6.25.

2.4.4. EPS Production in Milk Whey and Skim Milk

The production of EPS was evaluated in milk whey and skim milk with different concentrations of sucrose (5%, 10% and 20%). The bacterial culture (1%) was inoculated into 50 mL of MRS broth and incubated overnight at 30 °C. The medium was centrifuged at 3500× g for 30 min at 4 °C, and the pellet was dissolved in either milk whey or skim milk containing 5%, 10% or 20% sucrose and incubated at 30 °C for 48 h. After fermentation, the EPS was extracted as follows: 25 mL of the fermented sample was diluted with MiliQ water (1:1). The casein fraction was precipitated by adding 4 mL of 0.2 M TCA and centrifuging at 3500× g for 30 min at 4 °C. The supernatant was collected, and the pH value was adjusted to 6.8 with 4 M NaOH. Then, the solution was boiled for 30 min and centrifuged at 3500× g for 30 min at 4 °C to remove the whey proteins. The volume of the supernatant was measured, and an equal volume of cold ethanol was added to further precipitate the carbohydrates overnight at 4 °C. The sample was centrifuged at 3500× g for 30 min at 4 °C. The pellet was resuspended in 10 mL MiliQ water and sonicated for 1 h at room temperature. The sample was vortexed to a uniform solution and dialyzed against MiliQ water for 7 days at 4 °C, changing the water twice daily. The dialysate sample was then stored at −20 °C. The EPS concentration in the suspension was quantified using the phenol-sulfuric acid method [33].

2.5. Functional and Technological Properties of Ln. mesenteroides SJC113

2.5.1. β-Galactosidase Activity

The enzymatic activity of the strain SJC113 was performed according to Miller [40] with some modifications [41]. The strain was grown in MRS broth overnight at 30 °C, and the OD was measured at 600 nm. Freshly prepared (80 µL) permeabilization solution (100 mM Na₂HPO₄, 20 mM KCl, 2 mM MgSO₄, 0.8 mg/mL hexadecyltrimethylammonium bromide—CTAB, 0.4 mg/mL sodium deoxycholate and 5.4 μL/mL β-mercaptoethanol added immediately before use) was added to 20 µL of bacterial culture and 600 µL of substrate solution (1 mg/mL o-nitrophenyl-β-d-galactoside—ONPG in phosphate buffer pH 7.0, and 2.7 μL/mL β-mercaptoethanol added just before use). The microtubes were incubated at 30 °C for approximately 30 min or until sufficient color developed. The reaction was stopped by adding 700 µL of stop solution (1 M Na₂CO₃) and centrifuged at 16,100× g (model 5415D; Eppendorf AG, Eppendorf, Hamburg, Germany) for 5 min. The tubes were carefully removed from the centrifuge, and the absorbance of the top layer was measured at 420 nm. β-galactosidase activity was calculated as follows:

$$\mathrm{Galactosidase\; activity\; (Miller\; units)}=1000\times {\displaystyle \frac{\mathrm{Abs}420}{\left(\mathrm{Abs}600\times 0.2\times \mathrm{reaction\; time}\right)}}$$

2.5.2. Cholesterol-Reducing Ability

The in vitro cholesterol-lowering ability of the strain SJC113 was evaluated, as described by Domingos-Lopes [42]. MRS broth was supplemented with 0.5 mg/mL (w/v) of cholesterol, inoculated with log phase grown bacterial culture and incubated at 30 °C for 24 h. The percentage of cholesterol reduction was calculated by measuring the cholesterol at the end of the incubation (cholesterol final) using the enzymatic colorimetric method (CHOD-PAP cholesterol kit; NS Biotec, Cairo, Egypt) and calculated as follows:

$$\mathrm{Cholestero\; reducing\; ability}\left(\%\right)=100\times {\displaystyle \frac{\mathrm{Cholesterol\; initial}-\mathrm{Cholesterol\; final}}{\mathrm{Cholesterol\; initial}}}$$

2.5.3. DPPH Scavenging Activity

The free radical scavenging activity or 2,2-Diphenyl-1-picrylhydrazyl (DPPH) was performed according to Aarti and Khusro [2], with some modifications. Overnight grown cells were freshly centrifuged and resuspended in dipotassium hydrogen phosphate buffer. The cells were added to a 1 mL DPPH solution (0.2 mM), and the mixture was incubated for 30 min at room temperature in the dark. During the assay, cells at the same concentration with methanol were used as blank, and the DPPH solution was used as control. Ascorbic acid was used as reference. All mixtures were read at 517 nm, and DPPH scavenging activity was calculated as follows:

$$\mathrm{DPPH\; scavenging\; activity}\left(\%\right)=\left[1-\left({\displaystyle \frac{{\mathrm{A}}_{\mathrm{sample}}-{\mathrm{A}}_{\mathrm{blank}}}{{\mathrm{A}}_{\mathrm{control}}}}\right)\right]\times 100$$

2.5.4. Hydroxyl Radical Scavenging Assay

The methodology previously described by He et al. [43] was used, with some modifications to estimate the hydroxyl scavenging activity of the strain. The reaction mixture consisted of brilliant green (0.5 mM; 1 mL), FeSO₄ (0.5 mM; 2 mL) and H₂O₂ (1.5 mL). The bacterial cells were previously centrifuged and inoculated into the reaction mixture. This was incubated for 15 min at room temperature, and the absorbance was read at 624 nm. Butylated hydroxytoluene was used as a standard.

$$\mathrm{Hydroxyl\; radical\; scavenging\; activity}\left(\%\right)=\left[{\displaystyle \frac{{\mathrm{A}}_1-{\mathrm{A}}_0}{\mathrm{A}-{\mathrm{A}}_0}}\right]\times 100$$

A₁ is the absorbance of a solution with the sample present, A₀ is the absorbance of a solution where the sample was not present and A is the absorbance without the sample and the Fenton reaction system.

2.5.5. Tolerance to Sodium Chloride (NaCl)

Overnight culture of LAB was inoculated into an MRS broth supplemented with either 2%, 5% or 10% (w/v) NaCl, as described by Aarti and Khusro [2]. Incubation was done at 30 °C for 48 h. After incubation, the survival potential was estimated by a plate counting method and expressed in log CFU/mL.

2.6. In Vitro Safety Evaluation of Ln. mesenteroides SJC113

2.6.1. Antibiotic Susceptibility

Antibiotic susceptibility was determined using the disc diffusion method according to the Clinical and Laboratory Standards Institute [44]. Antibiotic discs (Oxoid, Basingstoke, UK) were used to determine the susceptibility of the strains to 9 antibiotics: penicillins—ampicillin (2 μg per disc) and oxacillin (1 μg per disc); aminoglycosides—kanamycin (30 μg per disc), gentamycin (30 μg per disc) and streptomycin (300 μg per disc); glycopeptidesvancomycin (30 μg per disc); tetracyclines—tetracycline (30 μg per disc); macrolides—erythromycin (15 μg per disc) and amphenicols—chloramphenicol (30 μg per disc). The analyses were performed in duplicate. The discs were placed on the surface of inoculated Mueller-Hinton (AES, Liévin, France) agar plates colonized with LAB strains previously grown in MRS broth for 24–48 h at 30 °C. After the 24 h incubation at 30 °C, the diameter of the inhibition zones around the discs was measured using a digital caliper (Absolute Digimatic Caliper; Mitutoyo, Neuss, Germany). The strain was categorized as sensitive (S) or resistant (R) based on the diameter of the inhibition zones. Antibiotic susceptibility and resistance were interpreted as no growth and growth, respectively, as there are no CLSI guidelines for interpretation criteria for this genus.

2.6.2. Virulence Genes

The strain was tested for the presence of gelE (gelatinase), hyl (hyaluronidase), asa1 (aggregation substance), esp (enterococcal surface protein), cylA (cytolysin), efaA (endocarditis antigen), ace (adhesion of collagen), vanA and vanB (vancomycin resistance), hdc1 and hdc2 (histidine decarboxylase), tdc (tyrosine decarboxylase) and odc (ornithine decarboxylase) according to Ribeiro et al. [45]. The amplified fragments were separated by electrophoresis with 0.8–2% (w/v) agarose gel in 0.5 TAE buffer, and the gels were stained with SybrGreen (15 µg/mL).

2.7. Application of Ln. mesenteroides SJC113 in Fat-Free Fresh Cheese

2.7.1. Effect of Temperature on EPS Production in Skim Milk

The LAB were previously cultured in MRS broth for 48 h. They were then added to skim milk (10% w/v) and fermented overnight (approx. 16 h). Subsequently, this mixture was added to freshly prepared skim milk containing 0% sucrose, 5% sucrose or 10% sucrose to study the acidification and EPS production of Ln. mesenteroides at refrigeration temperatures (4 °C and 8 °C). The total number of colony-forming units (CFU/mL) was evaluated by plating appropriate dilutions on De Man–Rogosa–Sharpe agar (MRS), as described in Section 2.7.8. Viscosity (V-72 measuring system, speed: 12 rpm, time: 90 s) was determined using the RM100 Plus viscometer (Lamy Rheology Instruments, Lyon, France). These measurements were carried out during fermentation (0–120 h) in skim milk at 4 °C and 8 °C.

2.7.2. Cheese Making

Before starting cheese production, 1% of an overnight culture of Ln. mesenteroides was inoculated into skim milk media and left at 30 °C for 16 h. This inoculum was then used to ferment skim milk with 0% or 5% sucrose at 8 °C for 5 days (to promote EPS production). This fermented milk was later incorporated into skim milk (1:4) to produce fresh cheese. Fresh cow’s milk from the Azores University farm (Chegalvorada, Angra do Heroísmo, Portugal) was used for the production of fresh cheese according to Coelho et al. [46]. Whole milk (4.64% fat, 3.71% protein) was used for the production of full-fat cheese. For the production of fat-free cheese (no-fat cheese), the milk was skimmed. Four different types of fresh cheese were produced: full-fat cheese (FF), no-fat cheese (NF) made with skim milk, no-fat cheese with Ln. mesenteroides fermented with skim milk (NFLn0) and no-fat cheese with Ln. mesenteroides fermented in skim milk with 5% sucrose (NFLn5). The cheeses were stored at 4 °C until they were analyzed on day 1 and day 7.

2.7.3. Cheese Composition

The moisture content was determined using the oven-drying method [47]. The ash content was determined according to the AOAC method 945.46 [48]. Fat content was analyzed using the Van Gulik method [49], and total protein was quantified using the Kjeldahl method [50].

The pH values were measured directly with a potentiometer (HannaFoodcare HI99161, Hanna Instruments, Amorim, Portugal) on the 1st and 7th days of storage by inserting the electrode into the cheese.

2.7.4. Weight Loss

To determine the weight loss, the weight of each cheese was measured during the storage period using a precision scale (AE200, Metler Toledo, Columbus, OH, USA) according to the following calculation:

$$\mathrm{Weight\; loss}\left(\%\right)={\displaystyle \frac{{\mathrm{W}}_{\mathrm{i}}-{\mathrm{W}}_{\mathrm{f}}}{{\mathrm{W}}_{\mathrm{i}}}}\times 100$$

where W_i is the initial weight (day 0) and W_f is the weight measured at different times during storage (day 1 and day 7).

2.7.5. Water-Holding Capacity

The water retention capacity of fresh cheese was measured according to Linares et al. [51] using the following calculation:

$$\mathrm{WHC}\left(\%\right)=100\times {\displaystyle \frac{{\mathrm{C}}_{\mathrm{W}}-{\mathrm{D}}_{\mathrm{W}}}{{\mathrm{C}}_{\mathrm{W}}}}$$

where C_W is the weight of the cheese and D_W is the weight of the decanted whey.

2.7.6. Rheological Parameters

The viscosity was measured with a Viscometer RM100 Plus (LAMY Rheology Instru-ments). The device was equipped with an RV-7 spindle (dimensions—Ø3.20; reference—111007). The viscosity was measured for 60 s at a speed of 30 rpm.

Texture properties of fresh cheese samples were measured in a texture analyzer TMS-PRO (Food Technology Corporation, Sterling, VA, USA) equipped with a cylindrical probe (35 mm diameter) with a flat surface, a penetration depth of 15 mm and a probe speed of 1 mm/s. All analyses were performed in four independent experiments.

2.7.7. Determination of EPS Content in Fresh Cheese

The EPS of fresh cheese (25 g cheese sample) were extracted according to the method described by Kearney et al. [52]. The EPS content of the extracted suspension was estimated using the phenol-sulfuric acid method and expressed as glucose equivalent. All analyses were performed in triplicate.

2.7.8. Microbial Analysis

Microbial growth was assessed on the 1st and 7th day of storage at 4 °C. Microbial enumeration was performed by plating appropriate dilutions on De Man–Rogosa–Sharpe agar (MRS), a selective medium for enumeration of LAB (including Leuconostoc sp.), and plate count agar (PCA) a non-selective medium for enumeration of total aerobic microorganisms. Plates were incubated for 48–72 h at 30 and 37 °C for MRS and PCA, respectively. Results were expressed as log CFU/g. All analyses were performed in triplicate.

2.8. Statistical Analysis

The effect of sucrose on EPS production in whey and skim milk was tested using a factorial analysis of variance (ANOVA) with the treatments (milk whey and skim milk) and sucrose concentration (5%, 10% and 20%) as factors. The factorial ANOVA design was also used to investigate the effects of fermentation of skim milk with 0%, 5% and 10% sucrose at different temperatures (4 °C and 8 °C) by Ln. mesenteroides SJC113. Post-hoc multiple comparisons were performed using the Tukey test. An analysis of variance (ANOVA) was performed to evaluate the differences between the cheese formulations for the different parameters evaluated. For each value, the means ± standard error of the means (SEM) were reported. In cases where ANOVA revealed statistically significant differences (p < 0.05), Tukey’s test for multiple comparisons was applied. All statistical analyses were performed using the SPSS program (version 28; IBM SPSS Statistics, Armonk, NY, USA).

3. Results and Discussion

3.1. EPS Yield and Characterization

The estimated EPS production of Ln. mesenteroides SJC113 was 7.4 ± 0.9 g/L when cultured (1%) in MRS broth containing 10% (w/v) sucrose for 48 h at 30 °C. Several studies reported similar yields when using different Leuconostoc strains [32,53,54], although the yield of EPS may vary with the fermentation conditions, purification and quantification methods [55,56].

The carbohydrate composition and glycosidic linkages of EPS are presented in

Table 1. Total carbohydrate and protein concentration, dextranase resistance (%), monosaccharide composition (mole percentage) and relative abundances (mole percentage) of glycosidic linkages of purified EPS produced by Ln. mesenteroides SJC113.

Carbohydrate (mg/g)	352.2 ± 47.9
Protein (mg/g) Dextranase resistance (%)	415.2 ± 10.4 52.1 ± 2.3
Monosaccharide	Mol%
Mannose (Man)	2
Glucose (Glc)	92
Uronic acids (UA)	5
Glycosidic linkage	Mol%
t-Manp	1.3
2-Manp	0.3
2,6-Manp	0.7
3,6-Manp	0.7
Total Man	3
t-Glcp	6.3
4-Glcp	0.7
6-Glcp	84.5
3,6-Glcp	5.6
Total Glc	97

. The Ln. mesenteroides SJC113 produced a mucoid EPS that can be classified as glucan since glucose was the main monosaccharide detected (92 mol%). Moreover, taking into account the high percentage (84.5%) of (1→6) α-glycosidic linkages, the EPS is mainly a dextran with a low percentage of α-3,6 branched glycosidic linkages (Table 1). This result was further confirmed by the dextranase resistance since glucans presenting 85% of α-1,6 glycosidic linkages were shown to produce approx. 50% of resistance to dextranase [57].

EPS may be characterized by its susceptibility to dextranase. Dextranase resistance was shown to range from 4.3 to 37.4% for classical dextrans (>92% of α-1,6 glycosidic linkages), from 37.9 to 82% for dextrans containing α-1,2 linkages and from 64.0 to 97.8% for glucans with high percentage of α-1.3 linkages [57]. A novel water-soluble polysaccharide extracted from Leuconostoc citreum strains showed dextranase resistance ranging from 9.32 ± 0.23 to 88.96 ± 1.06 [58]. Also, higher dextranase resistance (82–95% resistance) expressed by Ln. mesenteroides strains were also reported [59].

Several Leuconostoc spp. were described to produce dextran from sucrose with high levels of (1→6)-linked α-d-glucopyranosyl units and low levels of (1→3)-linked α-d-glucopyranosyl units [9,57]. According to the study of London et al. [56], the percentages of α-1.6 linkages in EPS produced by Leuconostoc vary from 50% to 97.9%.

3.2. EPS Production in Milk Whey and Skim Milk

Ln. mesenteroides was cultured in two different types of media—skim milk and sweet whey (derived from the manufacture of fresh cheese), as shown in Figure 1, with three different concentrations of sucrose (5, 10 and 20%). The production of EPS in sweet whey with 5% sucrose was twice as high as the production of EPS in skim milk with 5% sucrose, namely 0.96 ± 0.11 g/L and 0.44 ± 0.14 g/L, respectively. However, this trend changed when 10% and 20% sucrose were added to skim milk and sweet whey, respectively. EPS production was 1.34 ± 0.19 g/L in sweet whey supplemented with 10% sucrose and 3.36 ± 0.15 g/L in sweet whey supplemented with 20% sucrose. In contrast, EPS production in skim milk supplemented with 10% sucrose doubled to 3.13 ± 0.04 g/L, and in skim milk supplemented with 20% sucrose, the yield was 6.81 ± 0.88 g/L. According to other authors, the highest EPS yields were produced by L. rhamnosus RW-9595 M, which was able to produce 2.7 g/L EPS in whey medium. However, the authors assume that due to the high variability in the chemical composition of whey, EPS could be produced with inconsistent biological and technological properties [60,61].

3.3. Functional and Technological Properties of Ln. mesenteroides SJC113

Strain SJC113 showed a high tolerance to NaCl, as it was able to grow in the range of 2% to 10% NaCl. The maximum growth (7.35 ± 0.08 log CFU/mL) was attained at 2% NaCl was), but the strain also showed high growth (6.33 ± 0.04 log CFU/mL) at 10% NaCl. Similar high tolerances to NaCl of LAB were reported by several studies [62,63,64]. Tolerance to high osmotic concentrations of NaCl could be an important requirement of bacteria to be used in food as commercial strains [62,64].

Ln. mesenteroides SJC113 presented high β-galactosidase activity (2368.4 ± 24.4 Miller units,

Table 2. β-galactosidase activity, cholesterol-reducing ability (%), radical scavenging activity—DPPH (%) and hydroxyl scavenging activity (% and BHT equivalent) of Ln. mesenteroides. Results are presented as mean ± SEM.

β-galactosidase activity (Miller units)	2368.4 ± 24.4
Cholesterol-reducing ability (%)	14.8 ± 4.1
Radical scavenging activity—DPPH (%)	11.7 ± 0.7
Hydroxyl scavenging activity (%)	15.7 ± 0.4
Hydroxyl scavenging activity (mg/mL BHT Eq.)	1.08 ± 0.04

). The β-galactosidases belong to a group of enzymes involved in the breakdown of lactose into glucose and galactose, and their activity in dairy products has beneficial effects in lactose-intolerant individuals [65,66]. In addition, the β-galactosidase activity expressed by LAB presents technological applications such as improving the technological and sensorial characteristics of dairy foods by overcoming the low solubility of lactose and providing greater sweetening power [67]. In recent years, several studies have been conducted to determine the production of β-galactosidase by LAB, reporting a wide range of activities. While some authors reported similar high values of β-galactosidase activity [68], others reported much lower β-galactosidase activities [69,70]. For example, Shukla et al. [70] reported reduced β-galactosidase activity (94.24 Miller units) for a Ln. mesenteroides strain.

Ln. mesenteroides SJC113 showed an ability to reduce cholesterol in vitro by approx. 15% (Table 2). Although there is not a high reduction in cholesterol, as different LAB strains were shown in other studies [42,71], this is a useful trait that can be used in the reduction of cholesterol in foods [72].

The antioxidant activity of LAB can be an important characteristic for the preservation of food. Low antioxidant activity was shown by the Ln. mesenteroides SJC113 using both a radical scavenging activity and hydroxyl scavenging activity (Table 2). The strain inhibited 11.7 ± 0.7% of DPPH radical and presented 15.7 ± 0.4% of hydroxyl scavenging activity, corresponding to 1.08 ± 0.04 mg/m BHT equivalents. These results are consistent with other authors that reported antioxidant activities of intact bacterial cells ranging from 4.5% to 19.7% and 8.8% to 30.6% of DPPH and hydroxyl radical scavenging activity, respectively [73]. In similar studies, LAB strains isolated from traditional artisanal milk cheese showed DPPH scavenging activity ranging from 2.79% to 39.3% [74]. However, higher DPPH and hydroxyl scavenging activities were reported by other authors (32.9–63.8% and 46–51.5% for DPPH and hydroxyl scavenging activities, respectively), presumably due to differences in protocol assays [75].

3.4. Antibiotic Susceptibility and Virulence Genes

The safe use of Ln. mesenteroides SJC113 was tested by screening for antibiotic resistance to clinically important antibiotics and the presence of virulence genes (

Table 3. Resistance/susceptibility to antibiotics and occurrence of virulence genes in Ln. mesenteroides SJC113.

Antibiotics 1		Virulence Genes 2
Ampicillin	S	gelE	-	vanA	-
Oxacillin	S	hyl	-	vanB	-
Kanamycin	S	asa1	-	hdc1	-
Gentamicin	S	esp	-	hdc2	-
Streptomycin	S	cylA	-	tdc	-
Vancomycin	S	efaA	-	odc	-
Tetracycline	S	ace	-
Erythromycin	S
Chloramphenicol	S

¹ S—sensitive (inhibition-zone >8 mm). ²—absence of virulence gene.

). Remarkably, the strain was sensitive to all antibiotics tested. These results were based on the formation of inhibition zones (>8 mm), as there are no CLSI guidelines for interpretation criteria for Leuconostoc sp. However, since genes involved in antibiotic resistance might be present but not expressed, these negative phenotypic results should be confirmed by a molecular biology approach. Regarding the presence of virulence factors, the strain did not harbor any of the tested virulence genes, including the resistance genes for vancomycin vanA and vanB. The presence of some virulence genes, including the genes for resistance to vancomycin (vanA), was observed in several studies using LAB as probiotics [76]. In contrast, other authors reported that LAB used as probiotics were negative for virulence genes [77].

3.5. EPS Production in Skim Milk

To determine the best conditions under which Ln. mesenteroides SJC113 can produce a highly viscous EPS, fermentation was tested in skim milk at 4 °C and 8 °C at different concentrations of sucrose (0%, 5% and 10%) for up to 5 days. The results for bacterial viability, pH values and apparent viscosity are shown in Figure 2.

Colony counts of Ln. mesenteroides SJC113 increased from ∼6 log (CFU/mL) to ∼10 log (CFU/mL) after 48 h at 8 °C and remained stable for up to 5 days. As expected, growth was lower at 4 °C and reached ∼9.5 log (CFU/mL) after 4 days. Bacterial growth was similar for milk with 5% or 10% of sucrose and slight lower for skim milk with no sugar added. As for the change in pH values, a slight decrease was observed during fermentation, with a maximum of 0.28 units for the highest level of sucrose (10%), indicating that this strain is a very low acidifier. This is important for this type of cheese, as it should have a low acidity since no acidification step is performed during cheese production [78].

The highest viscosity (3405 ± 212 mPa.s) was achieved after a five-day fermentation of skim milk with 5% and 10% sucrose at 8 °C (Figure 2). These results indicate that this strain is capable of producing large amounts of EPS at refrigerated temperatures (especially at 8 °C), as shown by the increase in milk viscosity. The incubation of skim milk at these chilling temperatures is important to limit the growth of undesirable bacteria that survive the pasteurization process, such as psychrotrophic Gram-positive spore-forming bacteria [79]. Some authors reported that the maximum production of dextran occurs in a logarithmic phase since dextran is a biopolymer associated with bacterial growth [80]. However, in the present study, a delay between bacterial growth and EPS production, which starts after the logarithmic phase, was observed at refrigerated temperatures (Figure 2A,B).

It is also noteworthy that EPS production was highest when the strain was fermenting skim milk with 5% sucrose at 8 °C. This result could be due to the decreasing pH values (higher at this sucrose concentration), which may influence EPS production. Indeed, several authors reported that pH = 7 is the best pH value for the production of glucan-type EPS by Ln. mesenteroides and other LAB [55,81]. Therefore, the highest pH drop observed during the fermentation of skim milk with 10% sucrose at 8 °C could cause slower EPS production under these conditions.

3.6. Application of Ln. mesenteroides SJC113 in Fat-Free Fresh Cheese

Considering the higher EPS production after the fermentation of skim milk with 5% sucrose at 8 °C after 5 days, these conditions were used in the next experiments to produce a fat-free fresh cheese. The aim was to replace the fat with the highly viscous EPS produced after fermentation with Ln. mesenteroides SJC113 to obtain the same physicochemical properties as a full-fat fresh cheese.

The physicochemical analysis of the different types of cheese (full-fat cheese and fat-free cheese) was carried out on day 7 of storage at 4 °C and is shown in

Table 4. Composition of fresh cheeses: full-fat cheese (FF), no-fat cheese (NF), no-fat cheese with 0% sucrose skim milk fermented by Ln. mesenteroides SJC113 (NFLn0) and no-fat cheese with 5% sucrose skim milk fermented by Ln. mesenteroides (NFLn5). Values are means ± SEM of three replicates.

Parameters	FF	NF	NFLn0	NFLn5
Moisture (%)	78.3  ±  1.3 a	85.8  ±  0.2 b	85.2  ±  0.6 b	87.0  ±  0.4 b
Crude fat (%)	11.4  ±  0.2	ND 1	ND 1	ND 1
Crude protein (%)	7.5  ±  0.8 ab	9.7  ±  0.4 bc	10.3  ±  0.6 c	6.4  ±  0.1 a
Crude ash (%)	1.98  ±  0.06 b	2.23  ±  0.03 b	2.26  ±  0.10 b	1.43  ±  0.08 a
Carbohydrates (%)	0.86  ±  0.44 a	2.24  ±  0.29 b	2.25  ±  0.14 b	5.18  ±  0.26 c

Different letters as superscripts in the same line indicate a significant difference (p < 0.05) between cheese treatments. ¹—Not detected.

. The gross composition of the standard full-fat cheese is similar to the values determined by other authors [78,82]. As expected, the cheeses made with skim milk (NF, NFLn0 and NFLn5) had a higher (p < 0.05) moisture content than the full-fat cheeses (FF). The lack of fat in the fat-free cheeses is compensated for by water, protein and carbohydrates (lactose) in the experimental control cheeses (NF and NFLn0). However, the cheese made with the 5% sucrose skim milk fermented by Ln. mesenteroides had a lower protein content (6.4%) compared to the fat-free control cheeses (p < 0.05). The application of fermented milk with 5% sucrose on cheese-making increased the carbohydrate content as expected, although some of these carbohydrates were presumably EPS. When compared to standard full-fat cheese, this experimental cheese has a protein content that is not significantly different (p > 0.05) and a slightly lower ash content (p < 0.05), as the fat was replaced by water and carbohydrates (partly as EPS).

The viability of Ln. mesenteroides SJC113 in fresh cheese and the in-situ production of EPS were evaluated on day 1 and day 7 of storage at 4 °C and are shown in

Table 5. Number of viable microorganisms (Log CFU/mL) in MRS and PCA, and EPS content on day 1 and day 7 of storage in no-fat cheese made with Ln. mesenteroides SJC113 fermented skim milk without sucrose (NFLn0) and no-fat cheese made with Ln. mesenteroides fermented skim milk with 5% sucrose (NFLn5). Values are mean ± SEM of three replicates.

Parameters	Day 1		Day 7
	NFLn0	NFLn5	NFLn0	NFLn5
MRS Log (CFU/mL)	6.09 ± 0.24 a	7.77 ± 0.15 b	6.53 ± 0.03 a	7.96 ± 0.01 b
PCA Log (CFU/mL)	6.06 ± 0.23 a	7.74 ± 0.13 b	6.50 ± 003 a	7.88 ± 0.02 b
EPS (mg/g)	ND 1	0.44 ± 0.02 A	ND 1	1.76 ± 0.28 B

Different lower- and upper-case letters in the same line indicate a significant difference (p < 0.05) between the cheese treatments and days, respectively. ¹—Not detected.

.

The number of viable cells was similar in MRS and PCA, indicating that only Ln. mesenteroides was present in the cheeses until day 7. The initial colony count of Ln. mesenteroides was slightly higher in cheese made from fermented skim milk with 5% sucrose and remained stable in both treatments (NFLn0 and NFLn5) until day 7. EPS was not detected in the cheese made from fermented skim milk without sucrose, as the strain requires sucrose for EPS synthesis. However, in fresh cheese made with the addition of fermented skim milk with 5% sucrose, EPS was already detected on day 1, and its concentration increased significantly (p < 0.05) to a concentration of 1.76 mg/g during storage at 4 °C (Table 5). These results indicate that at 4 °C, the strain actively converts the remaining sucrose (which was not consumed during the fermentation period) into EPS.

The presence of the bacterial strain was responsible for the slight decrease in pH values observed in cheeses with fermented milk (NFLn0 and NFLn5), especially in NFLn5 (

Table 6. Physicochemical characterization of fresh cheeses: full-fat cheese (FF), no-fat cheese (NF), no-fat cheese with 0% sucrose skim milk fermented by Ln. mesenteroides SJC113 (NFLn0) and no-fat cheese with 5% sucrose skim milk fermented by Ln. mesenteroides (NFLn5). Values of pH, water-holding capacity (WHC), weight loss, apparent viscosity and hardness are means ± SEM of four replicates.

Cheeses	Day 1				Day 7
	FF	NF	NFLn0	NFLn5	FF	NF	NFLn0	NFLn5
pH	6.58 ± 0.03 a	6.55 ± 0.02 ab	6.48 ± 0.01 bc	6.46 ± 0.02 c	6.57 ± 0.03 a	6.57 ± 0.03 a	6.45 ± 0.02 ab	6.36 ± 0.04 b
WHC (%)	57.78 ± 5.94 a	39.59 ± 3.34 b	41.58 ± 3.89 ab	57.48 ± 0.76 a	60.63 ± 1.07 a	42.35 ± 3.32 bc	39.95 ± 4.91 c	55.41 ± 1.13 ab
Weight loss (%)	2.85 ± 0.33 a	2.91 ± 0.17 a	1.75 ± 0.39 ab	0.71 ± 0.04 b	10.80 ± 0.62 ab	12.75 ± 0.48 a	7.60 ± 1.28 b	3.75 ± 0.61 c
Viscosity (Pa.s)	15.06 ± 2.22 ab	18.14 ± 2.58 b	14.89 ± 3.05 ab	5.67 ± 1.60 a	13.42 ± 1.75 b	19.64 ± 2.39 b	12.38 ± 1.39 b	3.33 ± 1.16 a
Hardness (N)	0.24 ± 0.03 a	0.52 ± 0.06 b	0.26 ± 0.02 a	0.13 ± 0.04 a	0.39 ± 0.08 b	0.62 ± 0.03 c	0.36 ± 0.03 b	0.11 ± 0.03 a

Different letters as superscripts in the same line indicate a significant difference (p < 0.05) in each day (day 1 or day 7) between cheese treatments.

). A decrease in pH value was also observed in these cheeses on day 7, possibly caused by the fermentation of residual lactose and the resulting formation of lactic acid [29]. Although the difference in pH compared to the control cheeses (FF) was significant (p < 0.05), this represents a decrease of 0.21 units of pH, being not high enough (> 6.3) to affect the organoleptic properties of this cheese, as reported in previous studies [46].

The WHC evaluates the ability of proteins to bind water molecules when exposed to centrifugal force and is an important parameter for foods with high moisture content such as yogurt and fresh cheese [78]. In the present study, cheeses made with skim milk had lower WHC values, with the exception of NFLn5 cheese (Table 6). For this cheese, the presence of EPS improves the WHC value to similar values as full-fat cheeses. The high WHC of fresh cheeses indicates that the proteins that make up the gel structure of the curd may prevent excessive whey loss. This can be observed in cheeses made from whole milk, as the interaction of the fat with the casein leads to high water binding [24]. In contrast, low-fat cheeses lose an excessive amount of whey over time, resulting in a more compact and drier structure [30].

The extent of syneresis in fresh cheese can also be assessed as the weight loss of the cheese during storage and is an indicator of the quality and shelf life of cheese [78]. The greatest weight loss was observed in fat-free cheese (12.75 ± 0.48%), followed by full-fat cheese (10.80 ± 0.62%), which was considered normal for this type of cheese [78,83]. In contrast, the addition of fermented milk with Ln. mesenteroides SJC113 decreased cheese syneresis during the 7-day storage period, resulting in minimal weight loss (3.75 ± 0.61%).

With respect to texture, apparent viscosity and hardness were measured on days 1 and 7 and are shown in Table 6. The presence of EPS reduced (p < 0.05) the apparent viscosity and hardness of NFLn5 cheese. In contrast, the fat-free (NF) cheese exhibited higher viscosity and hardness. The removal of fat from cheese without the addition of fat replacers usually results in a compact microstructure of the protein matrix, which is drier and harder compared to its high-fat counterpart [24]. Therefore, the use of in-situ-produced EPS could compensate for the changes in texture caused by the reduction in fat content during cheese production. Despite the high moisture content, the texture of the cheese was maintained, and the lack of fat was successfully compensated for by the EPS produced by the strain Ln. mesenteroides SJC113. Similar have been reported for low-fat cheddar cheese [26] and Kasar cheese [31]. However, to our knowledge, no studies have been conducted on fresh cheeses in which milk fat was completely replaced by in-situ-produced EPS.

4. Conclusions

Ln. mesenteroides SJC113 was shown to produce a glucan-type EPS from sucrose after fermentation of milk at different temperatures, including refrigeration. This strain showed important technological properties such as high tolerance to NaCl and high activity of β-galactosidase, while its safety was confirmed, as it did not exhibit virulence genes and was sensitive to clinically important antibiotics.

The results also show that EPS produced by fermenting skim milk with 5% sucrose at 8 °C for 5 days can improve the rheological properties of fat-free fresh cheese by softening it and reducing syneresis. As far as we know, this is the first report on the use of in-situ-produced EPS as a fat substitute in fat-free fresh cheese. The use of fermented skim milk with EPS for the production of fat-free fresh cheese showed great potential and could meet the growing consumer demand for low-fat and healthier dairy products.