Isolation, Identification, and Validation of Strains from Commercial Probiotics: Do We Get What We Expect?

TL;DR

This study checks if commercial probiotics actually contain the bacteria they claim, finding inconsistencies in viability and composition.

Contribution

The study provides strain-level validation of commercial probiotics, revealing discrepancies in product labeling and microbial quality.

Findings

Only four products met their labeled CFU counts, while two had no viable microorganisms.

Most isolates were L. reuteri and L. rhamnosus, with some containing additional species.

Strains showed variable functional properties and D-lactate production, but no transferable virulence markers were found.

Abstract

This study evaluated the viability, microbiological composition, functional traits, and safety of probiotic bacteria isolated from commercial products marketed as containing Limosilactobacillus reuteri. Viable cell counts, biochemical characterization, strain-level identification, functional properties, gastrointestinal tolerance, and safety attributes were assessed. Among the evaluated products, only four presented colony-forming units (CFU) counts consistent with label claims (products E, F, G, and H), while two showed no detectable viable microorganisms (products B and L). All isolates were Gram-positive, catalase-negative, and predominantly rod-shaped. rep-PCR analysis revealed strain homogeneity in most products, whereas others (products A and K) exhibited heterogeneous microbial compositions. Molecular identification based on 16S rRNA sequencing showed a predominance of Lmb. reuteri and Lacticaseibacillus rhamnosus, with some products containing additional species such as Lactiplantibacillus plantarum and Lactobacillus acidophilus. Functional assays demonstrated strain-dependent proteolytic and diacetyl-producing capacities, as well as variable tolerance to simulated gastrointestinal conditions. Most strains preferentially produced L-lactate, although some generated substantial amounts of D-lactate. All isolates were susceptible to antibiotics recommended by EFSA, except for intrinsic vancomycin resistance, and no transferable virulence markers, biogenic amine production, or Salmonella contamination were detected. Furthermore, virulence-related genes such as hdc, tdc, odc, hyl, cylA, and ace were not identified. Overall, the results highlight pronounced discrepancies between label claims and microbiological quality among commercial probiotic products and reinforce the importance of strain-level characterization to ensure safety, functional performance, and regulatory compliance.

Figures

Full text

1. Introduction

Since the visionary suggestions of Ilya Metchnikoff and his collaborator Stamen Grigorov signifying that microorganisms could be responsible for some of the health benefits of yogurt, the scientific community has extensively explored the role of beneficial microorganisms and highlighted their importance for human health and well-being. Following their work, the idea of probiotics was born, which are defined by the Food and Agriculture Organization of the United Nations (FAO) and the World Health Organization (WHO) as “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host” [1]. For almost a century, the benefits of microorganisms have remained under continuous discussion and constant revision, reflecting the relevance of the diverse applications of probiotics for health promotion. They have been extensively studied, applied in different health-promoting objectives, and explored by the nutritional and pharmaceutical sectors [2]. Strains belonging to Lacticaseibacillus rhamnosus, Limosilactobacillus reuteri, Lactobacillus acidophilus, and Bifidobacterium spp. are perhaps some of the most extensively studied as probiotics [2,3].

The probiotics industry represents a growing market worldwide. Products marketed as probiotics are sold in the form of foods, dietary supplements, pharmaceutical preparations, and medicines [4,5] and are available in different commercial presentations, including powders, granules, drinks/liquids, tablets, and gel capsules, among others. The question is whether marketing strategies consistently align with the scientific definition of the term “probiotic.” According to the science-based WHO recommendations, probiotics must be viable, meaning they should remain metabolically active and capable of replication [4,5]. A minimum effective dose of 1 × 10 ^9^ colony-forming units (CFU) per serving has been accepted as an adequate dose for a probiotic microorganism, as recommended by Health Canada and the Italian Ministry of Health [6]. Additionally, subsequent studies have reinforced that, to be effective, probiotics should be present at levels of 10^8^–10^9^ CFU/g [7]. When lower viable counts are recovered from probiotic products, their additional health benefits may be compromised, or they may become biologically unstable [4]. Furthermore, probiotic products should list microorganisms down to the strain level, since this information is essential for safety and valuable for healthcare professionals in properly recommending suitable probiotics to consumers; it constitutes clear evidence of the beneficial properties that need to be available [8,9].

It is important to emphasize that probiotic products contain living organisms whose specific metabolic activities may not always be suitable for every consumer. For a product to be marketed as a probiotic, the particular strain must be scientifically proven to provide the claimed benefits to consumers [10]. However, it is essential to clearly state any potential risks. Certain metabolites, such as biogenic amines, can present health hazards [11]; moreover, the production of D-lactic acid by some strains or species may cause acidosis in infants and young children [12]. While diacetyl is often considered a desirable aroma compound in dairy products, excessive production by certain probiotics can render it hazardous [13].

The health benefits of probiotics have been reported in numerous scientific studies and broadly disseminated through popular media, leading to a wide variety of supplements marketed as probiotics [14,15]. However, most of these products lack approved therapeutic claims and are sold as dietary supplements. Unlike therapeutic drugs, strict regulations are not consistently applied to these supplements. Unfortunately, the absence of standardized regulations enforced by governments or industries has led to mislabeling, leaving consumers without clear standards for probiotic product labeling [4,9]. The lack of appropriate controls and misuse in certain national markets has resulted in “probiotic” preparations with questionable properties and quality; in some cases, products have even contained other microorganisms, such as Staphylococcus species, including Staphylococcus aureus and Staphylococcus epidermidis, which have been detected in probiotic products in several studies [16,17,18,19].

Specifically, this study aims to evaluate probiotic products marketed as containing Limosilactobacillus (formerly Lactobacillus) reuteri with respect to: (A) the viable count of probiotic microorganisms present; (B) the identification of microorganisms using culture-dependent methods; and (C) the safety and functional properties of isolated microbial strains. This study seeks to provide a comprehensive overview of microbiological quality and labeling accuracy of marketed products. Furthermore, the analysis reinforces the need for governments and industries to enforce appropriate standards for probiotic products in order to protect consumers, particularly children, from mislabeled preparations.

2. Materials and Methods

2.1. Commercial Acquisition of Limosilactobacillus reuteri

Ten different commercial probiotic products labeled as containing exclusively Lmb. reuteri were obtained from both physical retail stores (in Bulgaria and the Czech Republic) and online platforms (two products from the United States and six from Brazil) (Table 1). Preference was given to products from different companies and countries (Bulgaria, the Czech Republic, the USA, and Brazil) in order to obtain a more diverse collection of Lmb. reuteri strains for the subsequent stages of the present project. Products were stored according to the manufacturer’s recommendations, and all tested products were within their stated shelf life. For this study, each product was assigned a code designated by the letters A–H, K, and L to ensure confidentiality of the brands and manufacturers.

2.2. Sample Processing

Culture-based techniques, following general microbiological approaches, were applied to determine the number of viable microbial cells present in each of the tested probiotic products. Since the commercial products analyzed were available in different pharmaceutical forms, minor modifications to sample preparation were performed to ensure cell viability.

Initially (fully applicable for powdered and granulated samples), two capsules of each commercial probiotic product were suspended in 18 mL or one sachet in 9 mL of sterile saline solution (0.85% NaCl) and homogenized in a Stomacher (Smasher, AES Laboratories, Ames, IA, USA) for 2 min. Following homogenization using a vortex, all samples were serially diluted up to 10^−8^ in the same diluent.

For cultivation and enumeration of Lmb. reuteri, 0.1 mL of each previously prepared dilution was spread onto de Man Rogosa Sharpe (MRS) agar (BD Difco, Detroit, MI, USA) and allowed to dry for approximately 1 h. Since it was assumed that the commercial products studied contained the correct bacterial counts, only the three highest dilutions were plated in duplicates. Plates were incubated for 24–48 h at 37 °C under aerobic conditions. Based on colony counts, the CFU per capsule (or per sachet, or per recommended volume for serving) was determined for each product. The experiments were performed on at least 3 independent occasions, and the results are presented as the mean with calculated and presented standard deviations.

2.3. Characterization: Gram Staining, Gas Production, and Catalase Activity

Selected bacterial cultures were individually grown in MRS broth and subjected to purity verification in accordance with Good Microbiological Practices. Pure cultures were stored in the presence of sterile glycerol (25%, v/v) at −20 °C and subsequently tested for basic biochemical and physiological attributes according to Bergey’s Manual of Systematic Bacteriology [20]. The morphology of the selected colonies was described. Gram staining was performed, and bacterial cell morphology was examined under a light microscope. Carbon dioxide (CO_2_) production from glucose fermentation was evaluated using Durham tubes to distinguish between homofermentative and heterofermentative metabolism. Catalase activity was assessed following the recommendations of Fugaban et al. [21]. A 24–48 h single-colony culture grown on MRS agar for 24 h at 37 °C was transferred into 50 μL of H_2_O_2_, and gas release (O_2_) was observed as evidence of catalase activity produced by the bacterial culture.

2.4. Differentiation and Identification of the Microbial Cultures Studied

Microbial cultures were grown in MRS broth for 24 h at 37 °C under aerobic conditions, as previously described. DNA was isolated using the ZR Fungal/Bacterial DNA Kit (Zymo Research, Irvine, CA, USA), according to the manufacturer’s instructions. The extracted DNA was quantified and evaluated with a NanoDrop spectrophotometer (Spectrostarnano, Ortenberg, Germany) to estimate concentration and purity.

The obtained DNA was initially assessed for differentiation (rep-PCR) of the cultures studied by PCR [18], using the primer (GTG)5 in a Veriti 96 thermal cycler (Applied Biosystems, Thermo Fisher, Waltham, MA, USA) under the following conditions: denaturation at 95 °C for 5 min, 30 cycles of 95 °C for 30 s, annealing at 40 °C for 30 s, and extension at 65 °C for 8 min, followed by a final DNA extension at 65 °C for 16 min. The generated amplicons were separated by electrophoresis on 1.5% (w/v) agarose gels, stained with SYBR^®^ Safe DNA Gel Stain (Thermo Fisher), and visualized using a Bio-Rad gel documentation system (Bio-Rad, Hercules, CA, USA).

Representatives of each cluster were identified through 16S rRNA gene sequencing after amplification using universal primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′) [18]. PCR reactions were performed in a Veriti 96 thermal cycler (Applied Biosystems) with the following parameters: initial denaturation at 95 °C for 7 min,35 cycles of 95 °C for 1 min, annealing at 58 °C for 1.5 min, and extension at 72 °C for 2.5 min, followed by a final DNA extension at 72 °C for 5 min. Generated amplicons were separated by electrophoresis on 1% (w/v) agarose gels, stained with SYBR^®^ Safe DNA Gel Stain (Thermo Fisher), run at 100 V for 45 min, and visualized using a Molecular Imager^®^ Gel-Doc™ XR system (Bio-Rad).

Generated amplicons from the performed PCR targeting the 16S rRNA gene were sequenced at the Human Genome Research Center, Institute of Biomedical Sciences, University of São Paulo (ICB/USP), São Paulo, Brazil, through Sanger sequencing (BigDye Terminator v3.1 Cycle Sequencing Kit, Thermo Fisher). The sequences were processed using Sequencing Analysis Software v7.0 (Base Caller KB™). Generated sequences were analyzed using the Basic Local Alignment Search Tool (BLAST + 2.17.0, National Center for Biotechnology Information, Bethesda, MD, USA) for identification (http://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 15 October 2025).

2.5. Safety Tests

All safety tests were performed on at least 3 independent occasions.

2.5.1. Hemolytic Activity and Antibiotic Resistance of Selected Strains

The hemolytic activity of the isolates was determined according to Fugaban et al. [21]. An 18 h culture of each strain was transferred onto Blood Agar plates prepared with 5% (v/v) defibrinated sheep blood and incubated at 37 °C for 24 h. Hemolytic activity was assessed by observing clear zones of hydrolysis (β-hemolysis), greenish zones of partial hydrolysis (α-hemolysis), or absence of hydrolysis (γ-hemolysis) around the colonies. Laboratory culture collection strains were applied as controls.

Antibiotic resistance testing was performed by agar diffusion assay, following the guidelines established by the Clinical and Laboratory Standards Institute (CLSI). The selection of antibiotics was based on recommendations from the European Food Safety Authority (EFSA) and included ampicillin, vancomycin, gentamicin, kanamycin, streptomycin, erythromycin, clindamycin, tetracycline, and chloramphenicol.

2.5.2. Proteolytic Activity, Gelatinase, and Diacetyl Production

Proteolytic activity of the isolates was determined according to Fugaban et al. [21]. An 18 h culture of each strain was transferred onto MRS agar plates supplemented with 20% milk or 20% casein and incubated at 37 °C for 24 h. Proteolytic activity was assessed by observing clear zones surrounding the growing colonies. Strains from the laboratory culture collection were used as controls.

To evaluate gelatinase activity of the investigated strains, the approach proposed by Colombo et al. [22] was followed. Briefly, 1 µL aliquots of each culture were inoculated into 10 mL of MRS broth supplemented with 3% (w/v) gelatin (Difco). Tubes were incubated at 37 °C for 48 h, followed by refrigeration at 4 °C for 4 h. Evidence of gelatin hydrolysis was recorded when the culture medium remained in a liquid state. Positive and negative controls were applied.

For the assessment of diacetyl production, the previously cultured strains were inoculated into reconstituted skim milk (1%) (autoclaved at 110 °C for 15 min) and incubated for 24 h at 37 °C. Subsequently, 2.5 mL of α-naphthol in ethanol (0.5%) and 5 mL of creatine-KOH solution (0.3 g creatine in 40% KOH solution) were added to 5 mL of the milk culture. The suspensions were mixed for 30 s and incubated at room temperature for 30 min. The appearance of a pink color at the upper surface of the solution was considered as evidence of diacetyl production, as recommended by Moraes et al. [23].

2.5.3. Test for Biogenic Amine Production

The assay for the identification of biogenic amine synthesis by the studied strains was performed according to the methodology described by Bover-Cid and Holzapfel [24]. Initially, the strains were cultivated at 37 °C and subsequently subjected to a five-day induction period, with daily transfers into modified MRS broth supplemented with 0.1% of each biogenic amine precursor: histidine, ornithine, and tyrosine (Synth S.A.™, Volta Redonda, RJ, Brazil). After the induction period, plating was performed on solid MRS agar medium, also supplemented with the precursors (1%); pyridoxal-5-phosphate (0.005%), adjusted to pH 5.3; and bromocresol purple (0.006%). Plates were incubated at 37 °C for 48 h. Since amino acid decarboxylation led to pH changes in the medium, positive cases of biogenic amine formation were indicated by a color change of the medium to violet shades.

2.5.4. Presence or Absence of Virulence Genes

Bacterial samples were incubated in 10 mL of MRS broth for 24 h at 37 °C. Subsequently, genomic DNA was obtained and quantified as described before. The extracted DNA was screened for genes related to vancomycin resistance (vanA, vanB, vanD), genes involved in the biosynthesis of biogenic amines (odc, hdc, tdc), the collagen adhesin (ace), genes encoding hyaluronidase (hyl), and cytolysin A (cylA), following the methodologies described by Fugaban et al. [21] and Kim et al. [25]. PCR reactions were carried out in thermal cyclers (Applied Biosystems) according to recommendations from Fugaban et al. [21] and Kim et al. [25], and amplified products were separated by electrophoresis in agarose gels (1.0–2.0% w/v), depending on the expected fragment size, using SYBR^®^ Safe DNA Gel Stain (Thermo Scientific) as the dye. Electrophoresis was performed at 100 V for 45 min, and results were documented with a Molecular Imager^®^ GelDoc™ XR system (Bio-Rad).

2.6. Behavior of the Studied Strains in Simulated Stomach/Duodenum Passage (SSDP) Models

To evaluate survival under gastrointestinal tract (GIT) conditions, the strains were exposed to in vitro conditions of a simplified GIT model, adapted from Fugaban et al. [21], with certain modifications and optimizations. Bacterial cultures were grown in 10 mL of MRS broth for 24 h at 37 °C. Cells were then centrifuged (4000× g, 5 min, 20 °C), washed twice with sterile saline solution (pH 6.0; 0.85% NaCl, w/v), and resuspended in the original volume of saline, adjusted to pH 2.5. Suspensions were homogenized for approximately 1 min and incubated for 60 min at 37 °C to simulate gastric passage. Subsequently, the following were added to the treated bacterial suspensions: 4 mL of sterile bile salt solution (10% bovine bile, w/v, Difco); 17 mL of synthetic duodenal juice (pH 6.0; NaHCO_3_ 6.4 g/L, KCl 0.239 g/L, and NaCl 1.28 g/L). The mixture was incubated for 2 h at 37 °C to simulate passage through the small intestine. At each stage (initial control, post-gastric, post-intestinal), samples were collected, serially diluted in sterile saline solution, and plated on MRS agar (2% w/v). After incubation at 37 °C for 48 h, colony-forming units (CFU/mL) were counted and compared with the initial bacterial population to determine survival rates.

2.7. Lactic Acid Production

The amount of lactic acid produced by the studied strains was evaluated enzymatically using a specific kit for the quantification of D-lactate and L-lactate (Megazyme™, Bray, Co., Wicklow, Ireland). This analysis was performed in duplicate with the entire cell-free supernatant (CFS) from each strain cultured in MRS broth at 37 °C for 24 h. The assay was conducted according to the manufacturer’s instructions.

2.8. Test for the Identification of Salmonella spp.

The test for the identification of Salmonella was performed using the MC—Media Pad kit (Merck, Darmstadt, Germany), following the manufacturer’s instructions. The kit applies an immunochromatographic methodology, a lateral flow assay (“rapid test”) that allows qualitative detection of Salmonella spp. antigens in biological or environmental samples, particularly after microbiological enrichment. Initially, samples were enriched by first being incubated in a pre-enrichment medium, generally Buffered Peptone Water (BPW) broth, for approximately 18–24 h at 37 °C. Then, an aliquot was transferred into a selective enrichment broth, Tetrathionate broth, and incubated again at 42 °C for 24 h, favoring the selective growth of Salmonella spp. After enrichment, 100 µL was applied to the sample well of the rapid test device. The sample migrated by capillarity along the reagent strip, where monoclonal antibodies specific to Salmonella spp. antigens were immobilized. Results were visually interpreted after 10 min and interpretation was performed according to the manufacturer’s instructions.

2.9. Growth Curve Assay of Strains in the Presence of Different pH, NaCl, and Bile Salt Concentrations

Selected bacterial cultures were grown for 24 h in MRS medium (Difco) at 37 °C. After this period, cultures were washed two times with sterile saline and resuspended in the original volume of saline solution and inoculated into flat-bottom 96-well microtiter plates (Kasvi™, Pinhais, PR, Brazil) by adding 10 μL of culture to 200 μL of modified MRS medium. The medium was previously adjusted to different pH values (2.0, 4.0, 6.0, 8.0, 10.0, and 12.0) using 1 M HCl or NaOH solutions before autoclaving [19].

Additionally, medium variations were prepared with the addition of bile salts (Oxoid) at concentrations of 0.1% and 10.0% or sodium chloride (NaCl) at 0.5%, 1.0%, 2.0%, 3.0%, 5.0%, and 10.0%. Bacterial growth was monitored by absorbance readings at 595 nm, taken hourly with the Ultrospec 2000 spectrophotometer (Pharmacia Biotech™, Marlborough, MA, USA) [19]. Commercial MRS medium without additives served as the control. All tests were performed in duplicates.

3. Results and Discussion

3.1. Identification, Characterization, and Safety Tests of Probiotic Strains

Initially, bacterial growth derived from commercial products (all into at least 8 weeks before the end of the shelf-life period) dilutions of the probiotic products was analyzed qualitatively to exclude any product that did not exhibit colony development. Based on this preliminary evaluation, commercial products B and L were excluded, as no colony growth was observed on the Petri plates. Subsequently, the products that showed satisfactory results in the qualitative test (probiotics A, C, D, E, F, G, H, and K) were subjected to colony count analysis to obtain an approximate quantitative estimate. The results of the quantitative analysis are presented in Figure 1.

Based on the information provided on the original product labels, the amounts indicated in CFU for products A, B, C, D, and F were 5 billion CFU per capsule (5 × 10^9^, equivalent to 9.70 log CFU/capsule), whereas product E, G, H and L contained 100 million CFU per capsule (1 × 10^8^, equivalent to 8 log CFU/capsule or recommended volume in case of liquid commercial preparations). Finally, product K contained 600 million CFU per capsule (6 × 10^8^, approximately 8.78 log CFU/capsule).

From comparison with the results presented in Figure 1, it can be observed that only the samples from products E, F, G, and H corresponded to the values declared on their labels. Products A, C, D, and K showed the presence of viable bacterial cells; however, the counts did not match those reported on the commercial product labels. Furthermore, to provide beneficial effects, probiotics must be consumed at specific levels of 10^8^–10^9^ CFU/g [7]. Considering the results, it is possible that products A, C, D, and K may not deliver the expected beneficial effects due to the low microbial count.

Subsequently, the colonies were analyzed through biochemical tests, all of which showed Gram-positive results (purple staining) and negative catalase activity (absence of catalase enzymatic activity). Regarding morphology, short rod-shaped cells predominated (Table 2).

Additionally, in the CO_2_ gas production test, as an indicator of heterofermentative metabolism, all strains isolated from products C, D, E, G, and H were classified as CO_2_ producers, while strains from product F did not produce gas. Conversely, some strains from products A and K produced gas, whereas others did not (Table 2). In the analysis of commercial products purported to contain the Lmb. reuteri strain, a heterofermentative species [20], CO_2_ production was utilized as a qualitative test to exclude homofermentative isolates.

In the identification of cultures were included the initials of principal investigators, sequence numbers of the isolates, and code names of the commercial probiotics (see Table 1). The differences observed can be explained by the presence of distinct strains within these commercial products. Strain identification was performed using two biomolecular assays: rep-PCR and 16S rRNA analysis. The images obtained from the rep-PCR tests, demonstrating the differences among strains isolated from the same products, are presented in Figure 2.

From the rep-PCR analysis, it was possible to determine whether isolates 1–10 from each probiotic corresponded to the same strain in most cases, confirming that commercial products should represent the presence of a single probiotic strain. In probiotics C, D, E, F, G, and H, identical banding patterns were observed for all isolates (1–10), indicating that the strains within each product were homogeneous but distinct between products. Thus, the bacteria present in the samples from each probiotic (C–H) were similar among isolates (1–10) but not identical across products (with G and H showing high similarity). In contrast, analysis of probiotics A and K revealed heterogeneity among isolates, as distinct banding profiles were observed in the agarose gel.

After evaluating similarity and divergence, representative isolates from each group representing products A and K (IS01A, IS05A, IS06A, IS07A, IS12A, IS15A, IS01K, IS02K, and IS06K) were selected for further experiments. Identification by 16S rRNA sequencing is summarized in Table 2. Sixteen isolates from eight commercial probiotics that showed growth were selected: IS01E, IS01F, IS01C, IS01D, IS01A, IS05A, IS06A, IS07A, IS12A, IS15A, IS01G, IS01H, IS01K, IS02K, IS06K, and IS09K. Based on repPCR analysis, isolates IS01H and IS01G were identical; therefore, these were not submitted for sequencing and were excluded from subsequent analyses.

The sequencing results demonstrated the presence of targeted strains among products marketed as probiotics C, D, E, G, and H, all identified as Lbs. reuteri. Strains isolated from product F were identified as Lacticaseibacillus rhamnosus. Similarly, strains isolated from product K included Lbs. reuteri (IS02K, IS06K, IS09K) and Lbs. rhamnosus. In product A, three strains of Lmb. reuteri (IS05A, IS06A, IS15A), two strains of Lpb. plantarum (IS01A, IS07A), and one of Lab. acidophilus (IS12A) were identified (Table 2). Probiotic products must include not just certain species but also the specific strains that are linked to the health benefits described on their labels. The effectiveness of these products at the strain level depends on how well the microbes are identified and how accurately their claimed benefits are supported, which is especially important for meeting consumer expectations and regulatory standards. For example, Jovanović et al. [26] carefully assessed a defined probiotic strain (Lmb. reuteri DSM 17938) used in a food matrix, stressing that thorough functional testing is essential beyond what is indicated on product packaging.

The ability of the strains to lyse red blood cells (hemolytic activity) and their proteolytic activity (casein or milk degradation) were evaluated (Table 2). All strains tested were γ-hemolytic and some displayed negative results for milk or casein protein hydrolysis. Proteolytic activity may be considered either beneficial or virulence-associated, depending on the application of the strain. In fermented food products, proteolytic activity can be beneficial, contributing to flavor and texture development. However, when applied as probiotics, this activity may have undesirable consequences for consumers, as it can interact with intestinal proteins [27]. In the present study, Lbs. rhamnosus IS01F, Lpb. plantarum IS01A and IS07A, and Lmb. reuteri IS05A demonstrated proteolytic activity when tested in skim milk (Table 2).

Regarding the ability to produce enzyme gelatinase, all strains tested were negative, except for Lmb. reuteri IS06A (Table 2). Gelatinase production has been suggested as a potential virulence factor, aiding in proteolysis and cell lysis [28].

Diacetyl production is considered a beneficial property for starter and probiotic cultures, as it contributes to the characteristic buttery aroma and exhibits antimicrobial properties. In the early 20th century, diacetyl and its derivatives were applied in treatments for Mycobacterium tuberculosis infections [29]. In this study, the tested strains showed variable diacetyl production: Lbs. rhamnosus IS01F and IS01K; Lpb. plantarum IS01A and IS07A; Lab. acidophilus IS12A; and Lmb. reuteri IS15A are diacetyl producers (Table 2).

Regarding antimicrobial susceptibility, all strains were susceptible to the antibiotics recommended by the EFSA, except for vancomycin, which is expected, since lactobacilli may possess intrinsic resistance to this antibiotic (Table 3). However, an atypical behavior was observed in Lmb. reuteri IS01D, which was susceptible to vancomycin, as well as in IS02K and IS06K (12 mm and 11 mm, respectively). These findings may be associated with specific genetic variations in these strains (Table 3). With respect to kanamycin, several strains displayed resistance, including Lpb. plantarum IS01A and Lmb. reuteri IS05A, IS01G, IS01H, IS01K, IS02K, IS06K, and IS09K (Table 3).

Consistent with these findings, Dongre et al. [30] noted that antibiotic resistance in beneficial microorganisms may exert either positive or negative effects depending on its nature. If resistance is considered an extrinsic property, it may be horizontally transferred to other microorganisms, including pathogens. Conversely, if it is intrinsic, limited antibiotic resistance in probiotic strains may be acceptable, as such strains can be co-administered with specific antibiotics and exhibit synergistic effects for the treatment of human and veterinary-associated infections [30]. Therefore, further investigations are required to confirm the non-transferability of antibiotic resistance genes among these strains.

Detailed analysis of strains IS01G, IS01H, IS01K, IS02K, IS06K, and IS09K revealed interesting patterns. For ampicillin, all strains were sensitive, although IS01G (19 mm) and IS01H (16 mm) exhibited smaller inhibition zones compared to the others (Table 3). Regarding erythromycin, IS01K showed excellent susceptibility (42 mm), while IS02K, IS06K, and IS09K were resistant (≤18 mm), possibly due to mutations in rRNA. Clindamycin results also varied: IS02K, IS06K, and IS09K exhibited complete resistance (0 mm inhibition zones), which may be associated with specific genes or ribosomal alterations (Table 3).

These results have important implications for the safety and applicability of these strains. The multiple resistance observed in IS02K, IS06K, and IS09K (particularly to tetracycline, erythromycin, and clindamycin) highlights the need to investigate the potential mobility of these resistance genes through PCR assays targeting specific genes and conjugation experiments. Conversely, strains such as IS01K, which exhibited susceptibility to most of the antibiotics tested, represent promising candidates for probiotic applications.

The assay for biogenic amine production was also performed. All analyzed strains tested negative for the production of biogenic amines. This finding is particularly relevant, as it indicates the absence of potentially toxic compounds such as histamine, tyramine, and putrescine in these bacterial strains. The lack of these amines is a critical factor for the microbiological safety of probiotics, especially considering their potential adverse effects in sensitive consumers. These negative results are consistent with the genomic analyses that will be presented later. However, it is important to note that biogenic amine production can be influenced by various environmental and physiological factors. As highlighted by Campedelli et al. [31], incubation temperature can significantly affect the expression of genes involved in the biosynthesis of these compounds. Furthermore, the availability of amino acid precursors in the growth medium and the presence of alternative metabolic pathways, such as transamination or amine oxidase activity, may also contribute to the formation of these nitrogenous derivatives under specific conditions.

The results obtained have important implications for the safety and applicability of these probiotic strains, establishing these strains as promising candidates for use in probiotic and fermented food products. This safety profile is particularly relevant for products intended for sensitive populations, such as individuals with food intolerances or specific metabolic conditions.

However, complementary studies under different experimental conditions and food matrices are recommended to definitively confirm these observations and ensure safety across all potential applications. These findings significantly contribute to the process of characterization and selection of safe probiotic strains, aligned with the strict requirements of international regulatory agencies.

For the identification of virulence-related genes, strains IS01A, IS05A, IS06A, IS07A, IS12A, IS15A, IS01C, IS01D, IS01E, IS01F, IS01G, IS01K, IS02K, and IS06K were analyzed. The results are presented in Table 4. It was observed that the strains studied did not possess genes encoding specific decarboxylase enzymes, such as histidine decarboxylase (hdc), tyrosine decarboxylase (tdc), and ornithine decarboxylase (odc) in their genomes. As demonstrated by de Fugaban et al. [21], the absence of these genes is directly related to the inability to synthesize the corresponding biogenic amines. This consistency between phenotypic and genotypic data reinforces the reliability of the results obtained in the previous biogenic amine production assays. Furthermore, genes encoding hyaluronidase (hyl) and cytolysin A (cylA) were not identified. The ace gene was also not detected in any of the analyzed samples, which represents an additional safety feature for the studied probiotics, since ace is associated with virulence, aiding bacteria such as Enterococcus faecalis in adhering to human tissues and causing infections (e.g., endocarditis, urinary tract infections, etc.). Therefore, its absence indicates that these strains likely lack pathogenic colonization capacity, an essential requirement for probiotics intended for human consumption.

Genes related to vancomycin resistance (vanA, vanB, vanD) were also evaluated (Table 3). All samples tested negative for vanB and vanD; however, the strains showed the presence of the vanA gene, which warrants further attention. A deeper investigation is required to confirm whether the generated amplicon, corresponding to the expected target size, represents the correct base sequence, an analysis that must be performed through sequencing of the obtained amplicons. If sequencing confirms the presence of the vanA gene in the DNA of the studied strains, an expression study at the RNA level will be necessary to determine whether the gene is actively expressed.

3.2. Behavior of Selected Strains in Simulated GIT Models

Figure 3 illustrates the ability of the evaluated 14 strains to withstand adverse GIT conditions. The experiment simulated three main stages: t_0_ (before exposure to gastric conditions), t_1_ (after passage through the stomach, under acidic pH), and t_2_ (after passage through the duodenum, where bile salts and digestive enzymes are active). Bacterial viability was expressed as log CFU/mL, allowing the identification of which strains maintained stability and which underwent degradation throughout the process.

Among the analyzed strains, some stood out for their high resistance. This was the case for Lpb. plantarum IS01A and IS07A, which maintained elevated viability levels. However, even though these strains exhibited high counts after exposure to stomach and duodenal conditions, a decrease in the number of surviving bacteria was observed during passage. This behavior suggests that these bacteria possess efficient mechanisms to cope with hostile environments, such as the production of protective substances or the ability to adjust their metabolism in response to stress.

Other strains, such as Lbs. rhamnosus IS01F and IS01K, although belonging to the same species, exhibited different resistance patterns. Strain IS01F showed a slight reduction in counts after passage through the stomach and a pronounced decline after exposure to duodenal conditions. Strain IS01K showed no significant change during gastric passage but displayed a greater decrease in viable counts after the duodenal phase.

Conversely, some samples showed marked sensitivity to simulated GIT conditions. Strains Lmb. reuteri IS01C, IS01D, and IS01G, for instance, experienced a drastic reduction in viability after duodenal exposure, indicating low tolerance to basic pH and to substances (enzymes and bile) released in this region. Interestingly, other strains of the same species, such as Lmb. reuteri IS05A, IS15A, and IS02K, exhibited higher resistance, showing less pronounced reductions and greater stability in the duodenal phase, although with generally lower bacterial counts. Among them, IS15A showed better survival than IS05A and IS12A. This variation among strains of the same species reinforces the idea that individual characteristics, such as genetic or physiological differences, can significantly influence bacterial survival.

The heterogeneity observed among strains of the same genus and species is one of the most interesting aspects of this study. For example, while Lmb. reuteri IS05A and IS15A demonstrated moderate resistance, other strains such as IS01C, IS01D, and IS01G were highly sensitive. Similarly, Lab. acidophilus IS12A showed a significant decline in viability in the stomach, followed by an additional reduction after duodenal exposure, possibly due to specific metabolic adaptations. These differences emphasize the importance of evaluating strains individually, even when belonging to the same taxonomic classification, since small variations can lead to markedly different behaviors.

In conclusion, this study provides valuable insights into the survival of different bacterial strains under GIT conditions, highlighting both the resilience of some lineages and the vulnerability of others. The results reinforce that probiotic selection should be based on strain-specific data to ensure their efficacy and safety for human health.

3.3. Quantification of D- and L- Lactic Acid

The results of D- and L-lactate isomer production analysis are shown in Figure 4. Approximately half of the strains exhibited higher production of the L-lactate form (metabolically favorable for humans), with Lbs. rhamnosus IS01F (12.669 g/L) and IS01K (10.666 g/L) showing the highest levels of production. In contrast, the D-lactate form was produced in smaller amounts, except for strain IS01A (10.029 g/L), which displayed higher production of this isomer. Strains IS15A and IS06K produced both isomers in nearly equivalent amounts (Figure 4).

From a physiological standpoint, these results are relevant because humans preferentially metabolize L-lactate via the enzyme L-lactate dehydrogenase. Although D-lactate is tolerated in small quantities, excessive accumulation in the body may lead to metabolic acidosis. This raises concern, as some strains that express high levels of D-lactic acid were isolated from probiotic products recommended for infants and children.

The predominance of L-lactate production by most strains (particularly IS01F, IS01G, and IS01K) is a positive aspect, indicating that these lineages are metabolically suitable for use in probiotic products intended for human consumption. However, other strains evaluated in this study may lead to metabolic acidosis. It is a fact that D-lactic acid can be linked to acidosis, particularly in young children. Therefore, strains that mainly produce D-lactic acid should not be marketed as recommended for babies and small children.

3.4. Absence of Salmonella spp.

From a regulatory standpoint, the presence of Salmonella in probiotic products would be considered a serious non-compliance issue, violating the standards established by both ANVISA (a Brazilian Health Regulatory Agency, RDC No. 331/2019) and international agencies such as the FDA and EFSA. It is noteworthy that the negative results obtained are consistent with the Good Manufacturing Practices (GMP) required for this type of product. The absence of Salmonella reflects adequate quality control throughout all stages of production. However, it is important to emphasize that microbiological safety must be continuously monitored, as contamination can occur at any point along the production chain.

The results obtained from the detection tests for Salmonella spp. in all probiotic products analyzed (A to L), using the immunochromatographic method (MC–Media Pad kit, Merck), were consistently negative. This finding is of fundamental importance for ensuring the microbiological safety of these products, especially considering their use for human and animal consumption. The applied protocol proved to be reliable, as evidenced by the consistent appearance of the control line in all tests performed.

The absence of Salmonella in these products is particularly relevant considering the serious consequences that its presence could entail. If this pathogen were present, it would represent a significant public health risk, especially for vulnerable populations such as immunocompromised individuals, the elderly, and children, precisely the groups that most frequently consume probiotics. Salmonella spp. is responsible for major foodborne diseases, including acute gastroenteritis, typhoid fever, and systemic infections, with the potential to cause large-scale outbreaks [32].

3.5. Growth Curve in the Presence of pH, NaCl, and Bile Salt Concentrations

Finally, the growth patterns of the selected strains with different pH levels (Figure 5) and different NaCl (Figure 6) and bile salt (Figure 7) concentrations were evaluated. pH is a crucial factor for bacterial development and growth, allowing direct observation of how acidic or basic environments influence microbial viability. Optimal pH levels for the growth of each strain were identified. At pH 2.0, an extremely acidic condition that simulates the gastric environment, most strains showed reduced growth, suggesting higher sensitivity to acidity. Despite this, microorganisms IS01G, IS01K, IS02K, and IS06K exhibited greater growth than the others under these conditions (Figure 5).

At pH 4.0, most strains demonstrated progressive adaptation to the acidic environment and a significant recovery in growth. Although this moderate acidity still poses a challenge, it is less hostile than pH 2.0. The strains displayed evident growth, with curves indicating a continuous multiplication rate over time.

At pH 6.0, the growth curves exhibited a pronounced and uniform upward trend. Strains IS12A and IS06K showed less growth compared to the others. This pattern suggests that, despite the generally good performance, subtle physiological differences exist among the strains, with some being more adaptable to this pH, indicating favorable growth under near-neutral conditions (Figure 5).

The medium adjusted to pH 8.0, which is slightly alkaline, proved to be the most favorable condition for all strains, representing the optimal point for many probiotic bacteria. This pH simulates the environment of the small intestine, where many probiotics exert their beneficial effects. The homogeneity of growth patterns among strains at this pH indicates that most possess a physiological profile compatible with the intestinal environment, an important attribute from both technological and therapeutic perspectives. Considering that the large intestine can reach pH levels between 7.5 and 8.0, maintaining viability within this range is a positive indicator of the functional performance of these probiotics throughout the gastrointestinal tract.

At pH 10.0, the medium becomes distinctly alkaline, and most strains showed increased growth under these conditions. At pH 12.0, most strains exhibited little or no growth, with linear absorbance curves over time, which showed moderate growth under this simulated condition (Figure 5). This result was expected, as highly alkaline environments cause protein denaturation, osmotic imbalance, and cellular membrane disruption, impairing bacterial survival. Nevertheless, the absence of growth at pH 12.0 does not compromise the probiotic potential of these strains, since such conditions are not found in normal physiological states of the digestive tract. Rather, these findings reinforce that microbial viability limits are respected, and that formulation development should focus on ensuring survival within more relevant pH ranges, such as 4.0–10.0.

Under the most extreme condition tested, 10.0% NaCl, despite the clearly elevated osmotic stress, some strains still exhibited notable growth, such as IS01G, IS01K, IS02K, and IS06K. In contrast, the remaining strains displayed low or no growth, revealing sensitivity to hypersalinity. These results indicate that, although this concentration exceeds the physiological conditions of the human gastrointestinal tract, the ability to grow under high salinity may represent an industrial advantage, conferring greater stability to probiotic products exposed to adverse formulation, storage, or transport conditions.

Subsequently, the growth of the strains at different bile salt concentrations was evaluated (Figure 7). The concentration of 0.1% represents a mild physiological level of bile salts, similar to that found in portions of the small intestine following moderate bile release. Strains such as IS01C, IS01D, IS07A, and IS15A stood out with higher absorbance values, indicating good adaptation and viability in this environment. Conversely, IS01E, IS01F, IS01A, IS05A, and IS06A showed more discreet growth, while the remaining strains exhibited low or no growth.

Under the higher concentration of 10.0% bile salts, all strains demonstrated excellent growth, as evidenced by well-developed curves. Overall, the homogeneity of the results indicates that most of the studied strains possess a basic tolerance to bile salts, an essential criterion for their probiotic functionality, since ingested microorganisms must survive the detergent action of bile in the intestine to exert their beneficial effects (Figure 7).

4. Conclusions

Probiotic formulations intended for commercial distribution must adhere strictly to the regulatory requirements established by national competent authorities and conform to internationally accepted scientific and safety standards. Compromising on product quality, strain authenticity, or consumer safety is not permissible under any circumstances. Accordingly, rigorous pre-market evaluation, including evidence-based characterization of microbial strains, demonstration of safety, and substantiation of health claims, is essential prior to commercialization. Furthermore, robust quality-assurance systems encompassing controlled manufacturing processes, validated storage conditions, and monitored distribution chains are critical to ensuring product consistency and maintaining the integrity of probiotic interventions.

This study demonstrates substantial variability in the microbiological quality and compliance of commercial probiotic products marketed as containing Lmb. reuteri. While some formulations met label claims regarding viable cell counts and target species, others exhibited reduced viability, heterogeneous microbial composition, or absence of cultivable microorganisms. Moreover, it cannot be stated that retail prices can serve as an indicator of quality, as tested products were chosen from relatively low to high cost. Strain-level analyses revealed pronounced differences in functional traits, including gastrointestinal tolerance, diacetyl-producing activities, and lactate isomer profiles.

From a safety standpoint, the absence of hemolytic activity, biogenic amine production, Salmonella contamination, and major virulence-associated genes supports the suitability of the evaluated strains for probiotic applications. Overall, these findings underscore the need for stricter quality control, mandatory strain-level identification, and harmonized regulatory standards to ensure the safety, functionality, and reliability of marketed probiotic products.