Safety evaluation of mutagenicity, genotoxicity, and cytotoxicity of Lactobacillus spp. isolates as probiotic candidates

Abstract Background Probiotics promote a healthy balance of gut bacteria and have many beneficial effects on human digestive physiology. Although, few side effects of probiotics have been reported. This study aimed to assess the safety of five probiotic candidate Lactobacillus strains isolated from healthy individuals by examining mutagenicity, genotoxicity, and oral toxic effects. Methods Five selected candidate probiotic (SCPs) strains were evaluated for genotoxicity (Ames test with Salmonella typhimurium), in vitro mammalian chromosome aberration test and an in vivo mouse micronucleus assay on peripheral blood of mice. To evaluate the oral dose toxicity, BALB/c mice models were treated repeatedly (2000, 1000, and 500 mg/kg body weight /day) for 28‐days. Results The Ames test performed for two S. typhimurium strains TA 98 and TA100 (both in the absence and in the presence of S‐9 metabolic activation system) did not show an increase in reverse mutation because of exposure to the SCPs in any of the doses (5.0, 2.5, 1.25, 0.625, and 0.3125 mg/plate). There was no genotoxicity in the SCPs treatment in the vitro chromosome aberration assay with Chinese hamster ovary cells (CHO‐K1). In addition, none of the tested strains increased the frequency of micronucleated reticulocytes in reticulocytes, the SCPs with the studied doses caused no substantial variation in the experimental groups compared to the negative control group (p > 0.05). SCPs were not acutely toxic when administered to male and female BALB/c mice by single gavage at (2000, 1000, and 500 mg/kg b.w/day) with no mortality or clinical signs, change in body weight or macroscopic abnormalities were observed in this dose range. Conclusion As a result, SCPs did not induce mutagenic potential in vitro with bacterial reverse mutation, clastogenicity, and in vivo tests in the ranges of concentrations evaluated in our study.


| INTRODUC TI ON
Probiotic bacteria are microorganisms that have a wide range of health benefits for the host, including protection against direct colonization of the gastrointestinal tract (GI), maintaining the equilibrium of intestinal microflora, controlling digestion, and stimulating the immune system. 1 Many lactic acid bacteria (LAB) strains are considered safe in most cases, given their long history of usage. 2 Among all LAB, Lactobacillus is considered the largest genus generally recognized as safe (GRAS) and induces several health benefits on the host as a probiotic. 3 In recent years, the use of live bacterial species such as Lactobacillus spp. due to safety benefits has received much attention. The GI is the location of too many bacterial species with critical functions in maintaining GI functionality, accounting for 70% of the human immune system functions. 4 Moreover, in the European Union, 36 Lactobacillus strains are headed with Qualified Presumption of Safety (QPS). 5 Several studies have shown that different strains of LAB have anti-tumor and anti-mutagenic properties, which are the chief characteristics of probiotics that could increase human lifespan. 6 Nowadays, studies show that new or specific species of probiotics are constantly being identified. The efficacy of new strains must be carefully evaluated before incorporating them into products, and a case-by-case evaluation must be performed to determine whether they share the safety status of traditional organisms with food grade and whether probiotics have the potential to enter the industry. [7][8][9] This is very important for humans because of the growing concern about the safety associated with probiotics whose extracts are being tested for their possible harmful effects on the host DNA. [10][11][12] Presently, at least three actual tests are used universally to check the genotoxicity of different compounds. Bacterial gene mutations, chromosomal aberration assays in mammalian cells, and micronucleus tests in rodents as the fast and trustworthy tests are usually employed for genotoxicity investigation. 13 Studies have shown that in each community, specific species could show better and more beneficial probiotic effects. 14 Therefore, in this study, five Lactobacillus strains as a selective candidate probiotics (SCPs) from the previous study 15 were assessed in terms of genetic toxicity, their potential mutagenic effects according to OECD guidelines (Organization for Economic Cooperation and Development) by performing laboratory evaluations and in vivo studies using mouse models to depict the safety of the strain in question.

| Ethics statement
The Ethics Committee approved all procedures and techniques used

| Bacterial strains and growth conditions
Five Lactobacillus strains (SCPs), including two strains of Inclusion criteria were; healthy humans without present or past underlying diseases, lack of any antibiotic therapies over 3 months prior to sampling, and no gastrointestinal (GI) disorders or consumed probiotics at the sampling time. Smokers, alcohol users, and people with underlying diseases were excluded from the study. Phenotypic methods identified the isolates as described previously 15 SCPs were used as a lyophilized powder with a concentration of 4 × 10 11 CFU/g. The isolates grew in deMan, Rogosa, and Sharpe medium (MRS; Sigma, Aldrich, UK) at 37°C for 18 h. After incubation, the broth was centrifuged at 4000 × g for 12 min at 4°C, the supernatants of SCPs were discarded, and the cell pellets were freezedried and stored at −20°C until use.

| Bacterial reverse mutation test
Ames test was performed following the protocols of the OECD guideline No. 471 (1997) and Maron and Ames (1983). 9,10 Ames test was employed to evaluate the potency of five Lactobacillus strains in inducing genetic mutation in S. typhimurium strains TA98 and TA100 (National Collection of Industrial Microorganisms, Iran).
S. typhimurium TA 98, TA100 and S-9 (a metabolic activator of rat liver), were kept in a deep freezer at −80°C until utilized. For the preparation of metabolic activation, male rats (body weight~300g) were treated with 40 mg/kg sodium phenobarbitone (in 0.9% w/v saline) on day one and 60 mg/kg on days 2, 3, and 4. Then, five days later, animals were sacrificed and the livers were collected, homogenized in 0.18 M KCl. The homogenate was centrifuged at 12,000 g for 15 min. The supernatant was aliquoted (2 ml portions) and stored at −80°C. The S9 mix was prepared according to the recipe recommended by Maron and Ames, Mortelmans, and Zeiger. 3,16,17 It consists of 10 mM of MgCl, 36 mM of KCl, 6 mM of glucose-6phosphate, 4 mM of NADPH, and 3 mM of NADH, and 0.2 M sodium phosphate (pH 7.4) was added by 2 ml of S9.
Genotypes of S. typhimurium strains TA98 and TA100 were confirmed by examining histidine requirement, rfa mutation, uvrB mutation, and ampicillin resistance (Table 1). SCPs were used as lyophilized powders at the concentration of 4 × 10 11 CFU/g and were dissolved in phosphate-buffered saline (PBS) in five concentrations

| Chromosome aberration test
The principal media were bought from the Cell Bank of Pasteur Sigma-Aldrich) at two concentrations of (0.1 and 0.05 μg/ml) and 80-μMcyclophosphamide monohydrate (CPP; Sigma-Aldrich) were used as the positive controls.
In brief, 100 μl of a 10 (μg/ml) colcemid solution (Karyo MAX Colcemid Solution, Gibco, Life Technologies) was added into the culture and incubated for 3 h. Cells were harvested by trypsinization and centrifugation (4000 RPM for 10 min), swollen by freshly prepared potassium chloride at 0.075 M, and fixed with ice-cold freshly

| Micronucleus assay in mice
This assay was performed in accordance with the OECD guideline No. 474 for the testing of chemicals. 12  In this test, mice were anesthetized with isoflurane, and peripheral blood was sampled from the orbital sinus 48 and 72 h after receiving the first dose, and positive control was sampled only 48 h after the first dose.
Additionally, 36 h after the last day of administration, 3-4 μl of peripheral blood was collected from the mouse tail, placed on a slide to dry in the air, and fixed with methanol, then were incubated at room temperature for 1-2 h before examination with a light microscope. The

| Statistics
The collected data in this research were reported as mean ±SD.
Data related to Ames test, body weight, and frequency of RETs and MN% were analyzed using SPSS software (IBM, NY, USA) by employing one-way ANOVA and Duncan's multiple range assays. The significance of the differences between the groups was evaluated by p < 0.05.

| Bacterial reverse mutation test
Genotypes of S. typhimurium strains TA98 and TA100 were confirmed at first step by examining histidine requirement, rfa mutation uvrB mutation, and ampicillin resistance genotype (Table 1). Strains  (Table 3). Therefore, these five strains of SCPs were considered to have no mutagenic activity in histidine auxotrophy of S. typhimurium strains TA98 and TA100. All data were calculated as mean ±SD in three independent experiments (Table 3).

| Chromosome aberration test
The primary toxicity experiment of SCPs with five concentrations

| Micronucleus assay in mice
There were no signs of toxicity after consuming low (500 mg/kg  In the positive control group treated with mitomycin C, the frequency of RETs among total erythrocytes was 16 + 1.2 and 9 + 1.2 after 48 and 72 h, respectively, which were significantly lower than those in the negative control during the same period (

| Oral toxicity assay
During the study period, after 28 days of repeated oral administration of SCPs, no clinical signs of abnormality, ophthalmological abnormalities, or mortality were detected in the treatment groups (500, 1000, and 2000 mg/kg body weight). Though bodyweight gradually increased during the study period in the treatment groups; however, this amount was not significant (data not shown). Water and food intake in the groups treated with three doses of SCPs were not significantly different from the control group (Table 8). Statistical analysis was performed and did not observe a significant difference in the mean between male or female mice fed with control or three doses of SCPs mixtures.

| DISCUSS ION
Over the past few decades, the use of probiotic microorganisms in various foods, especially dairy products, have significantly increased. Lactobacillus and Bifidobacterium are probiotic bacteria that both make up the natural flora of the human gut and have many beneficial effects on human health. 23 In recent years, new probiotic species, especially native probiotic species, have become commercially available as effective supplements and foods used in industry and medicine. Therefore, the efficacy and safety of new probiotics must be appropriately tested. 24 The primary assay revealed no pronounced toxicity in experimental strains at concentrations up to 5 mg/plate, this concentration was set as the highest dose, and the Ames trial was performed with its serial dilutions. One of the essential non-clinical safety examinations is the Ames test, which detects mutagenic effects of food, chemicals, and plant extracts with high predictive results. 25   The study of micronucleus counting is an indirect method for investigating the possible genotoxic damage at the chromosomal level. 36 The results of our study showed that there were no genotoxic reactions in empirical animals at various maximum doses, which is consistent with the findings of various equivalent in vivo genotoxic studies on probiotics. 32,34,35 Furthermore, consistent with other studies, 32,36-38 28-day administration of the tested Lactobacillus strains in three doses does not cause any oral toxicity or mortality in animals.

| CON CLUS ION
Using OECD guidelines protocols (No. 471,473,474) for testing chemicals, this study showed that native Lactobacillus strains isolated from feces had no mutagenic or genotoxic effects at all doses tested using bacterial reverse mutation, chromosome aberration, micronucleus in mice, and oral toxicity assays. In this study, we assessed the Ames test using S. typhimurium strains TA98, TA100, and to confirm the results in the future, three more S. typhimurium

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.