All chemicals were purchased from Sigma-Aldrich (St. Louis, MO, United States) unless otherwise stated. The strains of Leuconostoc carnosum, all originating from MAP sausage or cooked ham (Raimondi et al., 2018, 2019), were obtained from our laboratory collection. The strains were routinely cultured for 48 h at 30°C in static tubes of deMan, Rogosa, and Sharpe broth (Lactobacilli MRS Broth; BD Difco, Sparks, MD, United States), hereinafter referred to as MRS.

Cultures were incubated at 4, 8, 15, 23, 30, 37, and 42°C to determine the effect of temperature on growth kinetic and yield. Turbidity at 600 nm (OD₆₀₀) was used to evaluate growth. OD₆₀₀ data from the exponential tract of the growth curve were used to calculate the specific growth rate (μ_max). Optimum, minimal, and maximal growth temperature (Topt, Tmin, and Tmax) were deduced with the Sym’Previus software¹ fitting in a secondary growth model the μ_max data as a function of temperature (Rosso et al., 1993).

Tolerance of Leuconostoc carnosum strains to osmotic and oxidative stress was tested in MRS broth supplemented with increasing concentrations of NaCl (0, 20, 40, 60, 80, and 100 g/L) and H₂O₂ (0, 0.0625, 0.125, 0.25, 0.5, and 0.75 g/L), respectively (Montanari et al., 2018; Fessard and Remize, 2019). The cultures were seeded (5% v/v) with 24-h MRS cultures and incubated at 30°C for 48 h, then the OD₆₀₀ was measured.

The strains were tested for the fermentation of 49 carbohydrates using API 50 CH test strips (bioMerieux, Marcy-l’Etoile, France), according to the manufacturer’s instructions. Briefly, the bacterial biomass from the surface of MRS-agar plates was used to create a suspension with a turbidity equivalent to 2 McFarland in 10 ml of API 50 CHL Medium. The suspension was used to inoculate each tube of the strip. Anaerobiosis was obtained in the inoculated cupules by using sterile paraffin oil. The results were read after 48 h of incubation at 30°C.

A modified formulation of MRS, where glucose was replaced by 5 g/L arginine (hereinafter referred to as MRS-arg), was used to evaluate arginine utilization. MRS-arg cultures, firstly seeded (5% v/v) with 24-h MRS cultures, were incubated for 24 h and propagated six times in the same medium. Then, OD₆₀₀ was measured and compared with that achieved by control MRS cultures.

Citrate fermentation was evaluated in citrate medium (Milliere et al., 1989), containing the following: 4.5 g/L citric acid, 10 g/L proteose peptone (no. 3; BD Difco), 5 g/L yeast extract, 2 g/L K₂HPO₄, 5 g/L sodium acetate ⋅ 3H₂O, 1 g/L Tween 80, and 0.2 g/L MgSO₄⋅7H₂O. Cultures in citrate medium, firstly seeded (5% v/v) with 24-h MRS cultures, were incubated for 48 h and propagated six times in the same medium. Then, OD₆₀₀ was measured and compared with that achieved by control cultures containing glucose.

Proteolytic activity was assayed in MRS-agar plates supplemented with 30 g/L gelatin (BD Difco) (Smith and Goodner, 1958). Spots of 5 μl of grown cultures of Leuconostoc carnosum were deposited onto the surface of the plates. After 72 h of incubation at 30°C, the plates were placed at 4°C for 5 h, then the diameter of the hydrolysis halo around the spots was measured.

Lipolytic activity was assayed in a MRS-agar medium containing triglyceride emulsions. It was composed of 900 ml of MRS, supplemented with 20 ml of 1 g/L rhodamine solution and an emulsion of 50 ml ddH₂O, 30 ml of seed oil, and 150 μl of Tween 80. Rhodamine solution and the oil emulsion were sterilized separately (i.e., filtered at 0.22 μm and autoclaved, respectively) and mixed to the sterile MRS. Spots of 5 μl of grown cultures of Leuconostoc carnosum were deposited onto the surface of the plates. After 72 h of incubation at 30°C, the plates were placed under an UV lamp and the diameter of the hydrolysis halo around the spots was measured.

The antibiotic susceptibility of Leuconostoc carnosum was assayed according to the EUCAST protocol (Matuschek et al., 2014). The 24-h MRS cultures were diluted in sterile saline to create suspensions with a turbidity equivalent to 0.5 McFarland and spread onto MRS-agar plates, on the surface of which the antibiotic paper disks were applied. The following antibiotics were assayed: amoxicillin/clavulanic acid, AMC (30 μg); ampicillin, AMP (10 μg); azithromycin, AZM (15 μg); bacitracin, BAC (10 IU); cefixime, CFM (5 μg); ciprofloxacin, CIP (5 μg); clarithromycin, CLR (15 μg); kanamycin, KAN (30 μg); neomycin, NEO (30 μg); tetracycline, TET (5 μg); vancomycin, VAN (5 μg). The inhibition halo was measured after 48 h of incubation at 30°C. Strains were assessed as sensitive (≥20 mm), intermediate (15–19 mm), and resistant (≤14 mm), as previously described (Szutowska and Gwiazdowska, 2021). Escherichia coli ATCC 2592 was used as reference strain in the disk diffusion test.

Grown cultures of Leuconostoc carnosum strains and the corresponding supernatants were assayed for the ability to inhibit Listeria monocytogenes in a plate assay. The supernatants were recovered by centrifugation (5,000 × g for 10 min) and split in aliquots, treated with one of the following: (1) no treatment; (2) the pH was adjusted to 6.0 with 1 M NaOH; (3) addition of 0.1 mg/ml catalase, incubation at 25°C for 30 min, then at 80°C for 10 min; (4) addition of 0.2 mg/ml proteinase K, incubation at 37°C for 1 h, then at 65°C for 10 min. The supernatant aliquots were filter-sterilized at 0.22 μm. A lawn of L. monocytogenes was streaked on surface of Brain Heart Infusion (BHI) agar plates (BD Difco), then 20 μl of sample (i.e., L. carnosum culture or supernatant aliquots) was spotted on the plates. The diameter of the zone of inhibition was measured after 16 h of incubation at 37°C.

For cross-inhibition test, Leuconostoc carnosum strains were inoculated at approximately 10⁷ CFU/ml in agarized MRS plates and challenged, by the agar well diffusion method, against 50 μl of filter-sterilized supernatant of each L. carnosum MRS culture grown at 30°C for 48 h. The diameter of the inhibition halos was measured after 48 h of incubation at 30°C.

Exopolysaccharide (EPS) production was deduced from the mucoid aspect of Leuconostoc carnosum colonies that originated onto MRS-agar plates supplemented with 40 g/L sucrose, after 14 days of incubation at 4°C or 48 h at 25, 30, and 37°C (Fessard et al., 2016).

Biofilm formation was quantified with crystal violet in a microtiter assay (Landeta et al., 2013). Leuconostoc carnosum was inoculated in MRS broth supplemented with 0.1 g/L Tween 80 and incubated for 24 h at 30°C. Grown cultures were 10-fold diluted in the same medium and poured in a 96-well microtiter plate, 150 μl each well. After 14 days of incubation at 4°C or 48 h at 30°C, the supernatants were removed and the biofilms adhering the wells were washed twice with PBS and stained with 125 μl of 1% Crystal Violet dye for 15 min. Stained biofilms were rinsed six times with PBS and extracted with ethanol/acetone solution (8:2 v/v) for 15 min. The absorbance of the extract was measured at 570 nm and normalized by the OD₆₀₀ of the culture. Strains with values >1 were regarded as positive to biofilm formation (Amaretti et al., 2020).

Simulated gastric juice consisted of 3.2 g/L pepsin, 0.084 M HCl, and 0.03 M NaCl, with the pH adjusted to 1.4 (Vecchione et al., 2018). Simulated intestinal fluid consisted of 8.5 g/L NaCl, 3.0 g/L bile salts (Oxgall, BD Difco), and 1.0 g/L pancreatin, with the pH adjusted to 7.4 (Castro-Rodríguez et al., 2015). Both the gastric and the intestinal fluid were filter sterilized at 0.22 μM.

The biomass of the strains was recovered by centrifugation (13,000 × g at 4°C for 5 min) from 1 ml of a 24-h MRS culture and resuspended in sterile saline. The suspension was diluted to 10⁶ CFU/ml in the gastric or the intestinal juice and incubated at 37°C for 1 or 2 h, respectively. Viable counts were determined with MRS-agar plates. To test the resistance to a sequential treatment with simulated gastric and intestinal fluids, the biomass was recovered by centrifugation (13,000 × g at 4°C for 5 min) after the gastric incubation, washed in PBS, then resuspended and incubated in simulated intestinal fluid.

All experimental protocols involving mice were approved by the Animal Care and Use Ethics Committee of the University of Padova under license from the Italian Ministry of Health, and they were in compliance with the national and European guidelines for handling and use of experimental animals.

The biomass of three strains (Leuconostoc carnosum WC 0318, L. carnosum WC 0319, and L. carnosum WC 0321) grown for 48 h in MRS broth was harvested by centrifugation, suspended in PBS containing 15% v/v glycerol to obtain single-dose aliquots of 5 × 10⁸ CFU in 150 μl, then stored at −80°C. Aliquots were thawed one by one and used in the animal trial.

Six-week-old Balb/c male mice were obtained from Envigo Laboratories (Oderzo, Italy) and housed in groups of two per individually ventilated cage, at 22 ± 2°C under a 12-h light–dark cycle. Chow food and water were provided ad libitum. Mice were allowed to acclimate to the laboratory for 1 week before entering experimentation. Three groups of six mice were established, each group being assigned a Leuconostoc carnosum strain. Each mouse received a daily dose of L. carnosum biomass via gastric gavage for 10 days. A control group consisted of six mice daily receiving 150 μl of PBS containing 15% v/v glycerol. Fecal pellets were collected at the end of the treatment, then immediately stored at −80°C in pre-weighted tubes containing 50% glycerol. Viable L. carnosum in mice feces were enumerated with pour plate method within MRS-agar supplemented with 2 μg/ml vancomycin, 25 μg/ml trimethoprim, and 12.5 μg/ml cycloheximide. The plates were incubated at 30°C for 72 h.

The mice were sacrificed after 10 days of daily supplementation with Leuconostoc carnosum WC 0318, L. carnosum WC 0319, L. carnosum WC 0321, or PBS. The whole ileum was collected, washed in cold PBS, and the Peyer’s patches were isolated under a dissecting microscope. Mucosa was scraped off from the underlying muscle layer of distal ∼5 cm of ileum and immediately frozen in liquid nitrogen and stored at −80°C.

Peyer’s patches were placed in DMEM containing dithiothreitol for 30 min to remove the mucus and then incubated for 90 min at 37°C in Hanks’ balanced salt solution without calcium but added with EDTA to remove epithelial cells. Peyer’s patches were extensively washed with Hanks’ balanced salt solution without calcium and then digested with collagenase D (2 mg/ml) and DNase (10 μg/ml) for 5 min at 37°C. Mononuclear cells were incubated for 30 min in ice-cold PBS containing 2% BSA or murine serum (blocking buffer) to block unspecific bindings. Cells were washed twice by centrifugation in PBS (1,600 rpm, 6 min) and incubated with the appropriate antibodies (Supplementary Table 1). Finally, cells were washed with blocking buffer and analyzed using a FACSCalibur (Becton Dickinson, San Jose, CA, United States).

Cells were first selected on a forward scatter and side scatter dot plot. CD11c double-positive cells were recorded in 10,000 events. For intrakine analysis, CD4 or CD8 positive lymphocytes were selected on FL-3/SSC dot plot and CD25/Foxp3 or IL17 positive cells were recorded in 50,000 events.

Frozen mucosal specimens were weighed and placed in ice-cold PBS buffer (1:10 w/vol ratio) supplemented with protease inhibitors (0.1 mM PMSF, 1 μM leupeptin, 150 nM aprotinin) and homogenized for 30 s. Cellular debris was removed by centrifugation (10,000 × g for 10 min at 4°C). Levels of interleukin (IL)-1β and IL-10 were determined in the cleared supernatants by ELISA, using commercially available kits (eBioscience, Prodotti Gianni, Milano, Italy). The assays were conducted following the manufacturer’s recommended protocols.

Comparisons were performed with Student’s t-test or one-way ANOVA followed by Tukey post hoc test. Statistical analysis was done using SPSS Statistics 21 (IBM Corp., Armonk, NY, United States).

Jaccard’s similarity scores among the strains were calculated with the software Past ver. 4.05 (Hammer et al., 2001) considering their metabolic profile, antibiotic susceptibility, and other physiological properties (Supplementary Table 2) and were used to compute an UPGMA (unweighted pair group method with arithmetic mean) cladogram.

The growth kinetics of 12 Leuconostoc carnosum strains was studied in MRS broth. All the strains grew in the temperature ranging from 4 to 37°C, with difficulty at 42°C. The optimum temperature was predicted in the range of 30.6–34.2°C (median = 32.5°C), while minimum and maximum ones lay in the range of 0.3–3.0°C (median = 1.4°C) and 42.6–44.8 (median = 44.0°C), respectively, (Figure 1). The maximum specific growth rate (μ_max) of all the strains was the highest at 30°C (p < 0.05), ranging from 0.23 to 0.34 h^–1 with a median of 0.30 h^–1 (Figure 2A). For all the strains, the highest biomass yields were achieved at 23°C (p < 0.05), laying in the OD₆₀₀ range 1.30–3.39, with a median value of 2.56 (Figure 2B). The medium generally presented a pH between 4 and 5 at the end of the growth phase, regardless of the growth temperature (data not shown).

To evaluate the adaptation to oxidative and osmotic stress, Leuconostoc carnosum strains were cultured in MRS broth with increasing concentrations of H₂O₂ and NaCl. H₂O₂ restricted the biomass yield by at least 1 magnitude (p < 0.05) at the concentration of 0.0625 g/L, while higher concentrations completely inhibited growth (Supplementary Figure 1). Differences among the strains were registered. In presence of 0.0625 g/L H₂O₂, L. carnosum WC319 did not grow, most strains achieved an OD₆₀₀ < 0.2, while L. carnosum WC0329 performed better, yielding an OD₆₀₀ of 0.36.

In presence of NaCl, Leuconostoc carnosum grew with decreasing yields up to 60 g/L (Figure 3). Marked differences were observed among the strains (Supplementary Figure 2). In presence of 60 g/L NaCl, L. carnosum WC0327 and L. carnosum WC0318 presented the lowest and the highest growth yield, with OD₆₀₀ values of 0.03 and 0.38, respectively. Only L. carnosum WC0328 grew to some extent also with 80 g/L NaCl, even though with an OD₆₀₀ < 0.1

Substrate utilization by the 12 Leuconostoc carnosum strains was assessed with API 50 CH strips and specific growth experiments in modified MRS media (Table 1 and Supplementary Figure 3). All the strains were positive for the utilization of glucose, ribose, sucrose, and esculin. Fructose, mannose, methyl-α-glucopyranoside, N-acetyl-glucosamine, trehalose, and turanose were used by at least 10 strains, while cellobiose, maltose, gentibiose, and gluconate were used by 5–7 strains. On the other hand, all the strains were unable to use 37 substrates, among which glycerol, erythritol, D- and L-arabinose, D- and L-xylose, galactose, L-sorbose, L-rhamnose, inositol, mannitol, sorbitol, lactose, melibiose, inulin, melezitose, raffinose, amidon, glycogen, xylitol, tagatose, fucose, citrate, and arginine. L. carnosum WC0329 exhibited the smallest range of growth substrates (6, i.e., glucose, fructose, ribose, sucrose, gluconate, and esculin) and L. carnosum WC0323 the largest one (14, i.e., glucose, fructose, ribose, mannose, methyl-α-glucopyranoside, N-acetyl-glucosamine, cellobiose, maltose, esculin, sucrose, trehalose, gentibiose, turanose, and gluconate).

Proteolitic and lipolytic activity were also assessed, showing that all the strains were negative to gelatin and triglyceride hydrolysis.

The 12 Leuconostoc carnosum strains were screened for susceptibility to 11 antibiotics (Figure 4). All the strains were resistant to cefixime and vancomycin and most of them were resistant or gave an intermediate response to ciprofloxacin and kanamycin. On the other hand, the strains were all susceptible to amoxicillin/clavulanic acid, azithromycin, clarithromycin, and tetracycline and were susceptible or gave an intermediate response to neomycin. Most strains were susceptible to ampicillin and bacitracin, except for a few resistant or intermediate strains. L. carnosum WC0329 was resistant to ampicillin, while L. carnosum WC0319 and L. carnosum WC0322 were resistant to bacitracin.

Only Leuconostoc carnosum WC0321 and L. carnosum WC0323 inhibited the growth of L. monocytogenes to some extent (Figure 4). For both the strains, inhibition of Listeria occurred with the culture, the supernatant, the neutralized supernatant, and the supernatant treated with catalase, while the supernatant treated with proteinase K did not exert any inhibition (data not shown).

Growth inhibition by Leuconostoc carnosum WC0321 or, to a lesser extent, by L. carnosum WC0323 was observed also in the other L. carnosum strains, with the exception of L. carnosum WC0328 that was immune (Figure 4). L. carnosum WC0324 and L. carnosum WC0325 were the most sensitive, particularly to the supernatant of L. carnosum WC0321.

EPS production was assayed at 4, 25, 30, and 37°C (Figure 4). The ability to produce EPS was observed in six strains (i.e., Leuconostoc WC0319, L. carnosum WC0321, L. carnosum WC0322, L. carnosum WC0323, L. carnosum WC0324, and L. carnosum WC0325) that presented mucoid colonies when cultured at 4 and 25°C. EPS production tended to reduce with the increase of temperature. A decrease in mucosity was observed in the colonies of L. carnosum WC0325 at 25°C and in those of L. carnosum WC0321 and L. carnosum WC0323 at 30°C. No strains presented mucoid colonies at 37°C.

Leuconostoc carnosum WC0321 and L. carnosum WC0323 produced considerable biofilm both at 4 and 30°C. A slight biofilm formation was observed only at 4 and not at 30°C for L. carnosum WC0324, L. carnosum WC0326, L. carnosum WC0327, and L. carnosum WC0329 (Supplementary Figure 4).

The treatment with simulated gastric juice was lethal for all the strains (Figure 4). On the other hand, all the strains survived to some extent to the treatment with simulated intestinal juice, presenting a residual viability in the range of 16–100% (Figure 4 and Supplementary Table 3). Coherently, no strains survived the sequential incubation, first in the gastric and then in the intestinal juice.

The hierarchical clustering of the 12 Leuconostoc carnosum strains recognized three clades according to Jaccard’s similarity of the phenotypic traits (Figure 5). One clade encompassed L. carnosum WC0319, L. carnosum WC0324, L. carnosum WC0325, and L. carnosum WC0329, which were all unable to use the disaccharides cellobiose, maltose, and gentiobiose and were negative to bacteriocin production and biofilm formation, at least at 30°C. The second clade encompassed L. carnosum WC0321, L. carnosum WC0322, and L. carnosum WC0323, which were the most sensitive to kanamycin and neomycin, and produced EPS, with L. carnosum WC0321 and L. carnosum WC0323 that also exerted anti-Listeria activity and were the main producers of biofilm. The third clade, encompassing L. carnosum WC0318, L. carnosum WC0320, L. carnosum WC0326, L. carnosum WC0327, and L. carnosum WC0328, differed from the first one for the wider spectrum of substrates, particularly the disaccharides.

The strains Leuconostoc carnosum WC0318, L. carnosum WC0319, and L. carnosum WC0321, each belonging to a different clade, were selected to investigate their potential in colonizing the intestine of mice. They all failed to colonize the intestine of Balb/c mice after 10 days of daily administration at the dose of 5 × 10⁸ cells. For the three mice groups, each receiving a single strain, the viable counts in the fecal samples onto MRS plates did not yield any colony (data not shown).

To evaluate the impact of Leuconostoc carnosum on the immune system, the phenotype of dendritic cells (DC) and lymphocytes in Peyer’s patches was analyzed. Following 10 days of supplementation, L. carnosum WC0321 enhanced all the measured markers of DC maturation (namely CD-80, MHC-II, TLR2, and TLR4), while L. carnosum WC0318 and L. carnosum WC0319 had a limited effect (Figure 6). L. carnosum effects were restricted to DC, since no effects on polarization of CD3⁺ lymphocytes toward regulatory (CD4⁺ CD25⁺ FoxP3⁺) or inflammatory (CD3⁺ CD8⁺ IL17⁺) phenotype were observed.

To further assess the impact of Leuconostoc carnosum on mucosal immune system balance, the expression level of IL-1β and IL-10 was measured after 10 days of supplementation. No significant effects on mucosal cytokine levels were observed (p > 0.05), as reported in Figure 7.