Lactic acid bacteria isolated from traditional Iranian butter with probiotic and cholesterol‐lowering properties: In vitro and in situ activity

Abstract Producing butter from yogurt is known as a traditional way practiced in Iran and elsewhere, particularly in rural areas. Lactic acid bacteria (LAB) with probiotic and cholesterol‐lowering properties were isolated from traditional butter collected in different regions of Iran. Then, isolates were identified and applied as adjunct culture in industrial butter production. Ten samples of traditional Iranian butter were collected from local farms. Fifty‐four isolates were considered LAB due to biochemical examinations. Molecular techniques then identified 10 strains showing high cholesterol reduction ability and tolerated bile and acid. The sequence analysis revealed that four isolates belonged to Enterococcus durans, four isolates to Lactobacillus, one isolate to Pediococcus, and one isolate to Neoscardovia. Lactobacillus brevis IBRC‐M 11044, Pediococcus pentosaceus IBRC‐M 11045, Neoscardovia arbecensis IBRC‐M 4391 4378, and Lactobacillus pentosus IBRC‐M 11043 were selected and applied as adjunct culture in producing four treatments of industrial butter. All examined strain treatments showed significant changes in cholesterol level of butter samples. Furthermore in all samples, the cholesterol content was significantly lower than control (p < .5). The highest level of cholesterol reduction was achieved in the butter sample prepared by Lactobacillus pentosus IBRC‐M 11045. Sensory analysis showed that the butter sample with Neoscardovia arbecensis IBRC‐M 4391 4378 was more acceptable than other butter samples. Due to our finding, it is valuable to incorporate these strains in butter production and introduce novel functional butter to market.

On the other hand, the approach based on the increased awareness of health benefits is attributed to probiotic products, particularly their ability to reduce serum cholesterol (Ishimwe et al., 2015).
Probiotic foods such as dairy products were classically defined as "foods containing live microorganisms believed to actively enhance health by microbiota balance improvement in gut (Tamime, 2008)." Today, a tremendous increase is shown in microbial species incorporated into probiotic dairy products (e.g., pasteurized milk, ice-cream, fermented milks, cheeses, and infant milk powder). However, fermented foods remain the primary source of probiotic organisms. Among fermented milk products, yogurt is considered the most crucial mean/medium to deliver probiotic organisms (Tamime, 2008).
Many attempts have been made to identify isolated bacterial strains from traditional dairy products in Iran for using as a starter (Edalatian et al., 2012;Ghiyamati Yazdi, 2012;Milani et al., 2012).
However, additional studies on strains' characterization isolated from traditional dairy products such as stability to acid and bile salts' conditions can select appropriate probiotic strains to increase health level and reduce cholesterol in butter. Kim et al., 2021, isolated lactic acid bacteria (LAB) from a traditional fermented Korean food, kimchi. LRCC5307 strain was isolated and showed a 74.5% decrease in cholesterol with 0.2% bile salts.
After producing butter with LRCC5307 strain as an adjunct culture, it showed 8.74 Log CFU (colony-forming units)/g viable cells, pH 5.43, and a 11% decrease in cholesterol (Kim et al., 2021). Albano et al. (2018), examined the lowering cholesterol ability in 58 probiotic strains and applied the selected strains in cheese preparation.
All strains were able to reduce cholesterol in cheese. Lactobacillus paracasei and Enterococcus lactis had the highest reduction (23%) (Albano et al., 2018). Ding et al. identified strains by screening LAB from local yogurt. The strains were able to reduce cholesterol more than 60% in a liquid culture medium. Lactobacillus plantarum LP3 had the highest cholesterol reduction (73.3%).
This study aimed to investigate the possibility for developing and introducing an adjunct starter with probiotic properties to produce functional butter with low cholesterol and acceptable physicochemical properties.

| Samples
Ten samples of traditional Iranian butter were obtained from local farms located in Ardebil, Semnan, and Gorgan. All samples were collected according to ISO 707 in 250 ml sterile bottles which were transported to the laboratory under refrigeration (4°C) within 36 h (Anon, 2008).

| Isolation and identification of isolates
Twenty-five grams of each sample was inoculated in 100 ml MRS broth (deMan, Rogosa, and Sharpe, HiMedia, Mumbai, India) at 37°C, M17 broth at 40°C, and MRS with L-Cysteine hydrochloride and L-Mupirocin at 37°C incubated, for 24-48 h in anaerobic and aerobic conditions until growth was observed. Then, decimal dilutions were directly prepared in a 0.1% ringer. Isolation and colony counts were made on the following media: MRS agar, M17 agar, M1712 agar (HiCrome Nickels and Leesment Medium), and MRS + L-Cysteine hydrochloride + L-Mupirocin. Finally, plates were anaerobically incubated, and aerobically at 30, 37, and 40°C. The incubation period varied between 24, 48, and 72 h depending on bacterial groups. Counting was conducted only for those plates involving 30-300 colonies and from plates corresponding to the highest dilution. Four to five different colonies (due to shape, size, and color) were randomly selected.
The selected colonies were regrown two or three times on the same media. A single colony from each plate was examined by Gram staining, catalase production, and microscopic morphology. Finally, only Gram-positive, catalase-negative isolates were considered, and longterm conservation of purified isolates was carried out in a mixture of MRS broth, M17 broth, and MRS+ L-Cysteine hydrochloride with sterile glycerol (30%) and stored at −70°C (Sagdic et al., 2002).

| Acid tolerance
Acid tolerance of isolates was studied by inoculating each microorganism on appropriate medium. Then, the pH of all media was adjusted to 3.0 with aliquot from each dilution by HCl, which was incubated at an appropriate temperature (30, 37, and 40°C) for 2 h. Serial dilutions were performed and grown on their related medium. The strains capable of growing to >10 7 CFU (colony-forming units)/ml after 24 h were considered acid-resistant strains (Liong & Shah, 2005).

| Bile tolerance
The selected isolates were examined for their ability to grow in appropriate medium supplemented with and without bile salts 0.3% (oxgall, Sigma-Aldrich, St. Louis, MO, USA), the latter was considered as control. The inoculated media were incubated at an appropriate temperature, and the growth was monitored by measuring the absorbance at 620 nm (A620) before and after 8 h of incubation. For quantifying growth inhibition of examined isolates by bile, the inhibition coefficient was calculated, using the following formula: Where T8 denotes control-optical density of culture broth without oxgall after 8 h incubation. T0 denotes control-optical density of culture broth without oxgall before incubation. T8 denotes treatment-optical density of broth involving oxgall after 8 h incubation, to treatment-optical density of broth involving oxgall before incubation. Isolates whose inhibitory coefficient is equal to or <0.5 are considered bile-tolerant isolates (Goderska & Czarnecki, 2007).

| Assimilation of cholesterol
Water-soluble cholesterol was filter-sterilized and added at a final concentration of 300 μg/ml to a MRS broth tube involving 0.3 g/100 ml oxgall. The tubes were inoculated with each selected strain (at 1 ml/100 ml) and incubated at 37°C for 2, 4, 9, and 24 h (bacterial turbidity was adjusted in phosphate-buffered saline (PBS) equivalent to 0.5 McFarland). After incubation, the inoculated medium was centrifuged at 10,000 g for 10 min. Measuring absorbance at 546 nm (A546) using a spectrophotometer, the cholesterollowering ability was determined . Using the following equation, the ability of selected strains to remove cholesterol from medium was calculated: where A = % removed cholesterol, B = Control absorbance at time 0 min, C = Examined strain absorbance.

| Identification of selected strains by the molecular method
Molecular techniques identified strains showing the high ability of cholesterol reduction and tolerance to bile and acid. The cell mass of freshly grown selected isolates was harvested, and their DNA was extracted using High yield DNA Purification Kit (CinnaGen Molecular Biology and Diagnostic, Iran).
Polymerase chain reactions (PCRs) were carried out in a thermocycler (Bio-Rad, USA) in a total volume of 90 μl containing 9 μl buffer, 4.5 μl of each primer 27f (5'-AGAGTTTGATCM TGGCTCAG-3') and 1492r ( Determining the PCR product sequence was done through Korean Bioneer Corporation Company. The sequence received from this company was edited and reread using ChromasPro software which was compared to the sequences stored in EzTaxon genomic database. Isolates with 97% or higher similarity in sequences were identified as the same species.

| Preparation of butter sample
Preparing butter was carried out in Pak dairy factory (Tehran, Iran).
The cream was separated from milk by a separator and standardized as 35% cream. It was pasteurized for 15 s at 85°C and divided into five sections. Then, the pasteurized cream was cooled to 37°C and inoculated with selected strains (5% v/v, 10 8 CFU/ml) at 37°C for 24 h. It was cooled to 14°C and churned. Butter samples (500 g) were packed and kept in refrigerator until the day of the experiment ( Figure 1).

| Cholesterol measurementand fatty acid compositions
The cholesterol and fatty acid compositions in butter samples (control and treated samples) were determined by gas chromatography (GC), as described by Fletouris et al. (1998). The control sample, which was not inoculated by any isolated strains, was prepared under the same conditions.

| Sensory analysis
Some sensory properties of butter samples (appearance, flavor, and consistency) were evaluated by 20 trained panelists (from 23 to 48 years old) using the Hedonic scale test in five levels (1, 2, 3, 4, and 5 by which 1 is very bad and 5 is very good) (Anon, 1997). The mean scores obtained for each sensory property were determined, and using Duncan's multiple comparison method, the differences among samples and their significance were calculated.

| Statistical analysis
To evaluate the effect of added isolated strain on the cholesterol reduction of butter, all experiments were triplicate. The average rating of three values was calculated for each sample (n = 3). One-way analysis of variance (ANOVA) was used to analyze the data following a general linear model in SPSS version 14.0 (SPSS Inc., Chicago, IL, USA). The significance level was set at p < .05 to compare among means and obtain the standard deviation (SD) using the Duncan test.
All graphs were drawn by Microsoft Excel software (version 2013). were similar to those reported in previous studies on butter (Sagdic et al., 2002).

| Chemical analyses
More moisture content in traditional butter is probably due to the difference in the processing equipment and the conditions used to prepare the butter. In traditional methods, butter granules are formed in fermented milk, which has a higher moisture content than the raw materials (cream) used to make industrial butter. Furthermore, due to high dry matter, particularly proteins in granules, the strong binding of proteins with water around butter granules prevents the outflow of water during kneading (Nielsen & Ullum, 1989). Sagdıc et al. (2004) reported moisture and fat contents of 15.20% and 82.90%, respectively, in traditional butter produced in Turkey, whose production method was similar to traditional Iranian butter.
The chemical properties of butters prepared with selected strains are mentioned in Table 2. Statistically, there was significant difference (p < .05) between moisture, fat, nonfat dry matter, pH, acid value-oleic, iodine index, and soap index among the samples.
According to the existing laboratory conditions for the preparation of butter samples, the moisture and fat contents of the samples did not match the standard of butter preparation. The results of chemical properties obtained from this study are consistent with the results of previous studies (Sagdic et al., 2002).

| Isolation and identification
Fifty-four isolates from 10 samples of traditional butter were identified (Figures 2 and 3). All strains were recorded as catalase-negative and Gram-positive cocci in pairs or long chains, bacilli in pairs or chains, and coccobacilli.

| Acid tolerance
The effect of acid on isolates is shown in Table 3. All strains showed the tolerance to pH 3.0 for 2 h, despite variations in the degree of viability.
Isolates with code nos. 4331, 4335, 4368, and 4373 were the most acid-tolerant strains, with more than 7 (log CFU/ml) after incubation for 2 h at pH 2.0, while isolates with code nos. 4326, 4344, 4352, and 4366 were the most acid-sensitive strains.
The ability to survive and grow at low pH environment is a key characteristic of probiotic bacteria. Acidic stress may inhibit bacterial growth by acidifying the cytoplasm, increasing energy consumption required for maintenance of intracellular pH, and inhibiting enzymatic reactions (Shabala et al., 2006). In many acid-tolerant fermentative bacteria, the intracellular pH decreases as the extracellular pH decreases during growth in order to maintain a constant pH gradient rather than a constant intracellular pH. Generating a large proton gradient is disadvantageous for fermentative LAB, because proton translocation consumes energy, and anaerobic organisms gain significantly less energy from sugar metabolism than what aerobes gain. Furthermore, a large proton gradient results in the accumulation of organic acid anions in the cytosol. There are several possible mechanisms by which a bacterium can regulate the intracellular pH, but the most important mechanism in fermentative bacteria appears to be the proton-translocating p-type adenosine triphosphatase (ATPase) (Siegumfeldt et al., 2000). Several proteins that are able to protect or repair macromolecules such as DNA are also effective in acid tolerance. It is believed that acidification in the cell reduces purine and pyrimidine in DNA (Almedia et al., 2015).

TA B L E 4 Coefficient of inhibition in medium with oxgall
Optical density in T0 Optical density in T8

Coefficient of inhibition Control
The broth containing oxgall Control The broth containing oxgall

| Bile tolerance
Tolerance to bile allows LAB to survive in small intestine. By analyzing coefficients of growth inhibition ( Bile is a green-yellow alkaline liquid, secreted by the liver cells, which is involved in fat digestion. Bile is a solution of bile acids (weak organic acids) and their salts (containing amino acid, taurine and glycine), cholesterol, phospholipids, and bile pigments. Bile causes folding and/or denaturation of cell wall proteins, DNA and RNA degradation, pH reduction, osmotic stress, and oxidation, as well as bacterial cell wall damage. As a result, the ability to survive, grow, and reproduce in such a situation is used as a benchmark when assessing probiotic strains' potential (Lv et al., 2017).
Despite the proteomics studies and the expression of several genes reported for the resistance of probiotic strains to bile, the mechanism of tolerating probiotics to bile salts is still not completely clear. The active release of acids or bile salts to the outside, bile salts' hydrolysis, and changes in the composition of the bacterium cell membrane and the cell wall are believed to be the most common mechanisms for bile resistance in both Lactobacillus and Bifidobacterium (Kumar et al., 2006;Lv et al., 2017;Pfeiler & Klaenhammer, 2009).
Bacterial bile salt hydrolase (BSH), which controls bile salts' deconjugation reaction, is believed to take an active part in abating the bile salts' toxicity. The amino acid released by deconjugation can be further utilized as carbon and nitrogen sources to benefit bacterial sustenance and survival. BSH is an intracellular, nonallosteric enzyme which is nonsensitive to oxygen; the optimum pH is 5-6, and the activity of BSH is associated with biomass density (Niamah et al., 2017).
Besides, bacterial cells' colony shape and morphology play an essential role in the resistance to biliary salts in probiotics. Suskovic et al. (2000) reported that the Lactobacillus acidophilus M92 colonies are smooth and rough, and the rough type is more sensitive to biliary salts. They suggested that this difference was not due to a genetic mutation in the bacteria due to environment's phototypic response.
Smooth colonies have a compact structure with short chains, while rough colonies are longer and more vulnerable to biliary salts.
Increasing the production of exopolysaccharides in Bifidobacterium affected by biliary salts has a protective role for the cell wall (Mozzi et al., 2016). Also, in proteomics studies by Burns et al. (2010), they showed resistance to bile salts in Lactobacillus delbrueckii, which is directly related to the ability to code enzymes involved in the production of exopolysaccharides.  Liong and Shah (2005) suggested that the Lactobacillus cell wall absorbed the environmental cholesterol by studying the fatty acid profile in Lactobacilli, especially palmitic acid, stearic acid, total saturated and unsaturated fatty acids with and without cholesterol.

Coefficient of inhibition Control
The broth containing oxgall

| Molecular identification of selected strains
The PCR was performed to amplify 16S rRNA (ribosomal RNA) region after DNA extraction of selected strains. After transferring the PCR product onto the gel, they had a length from 1400 bp to 1500 bp.
Due to sharp bands' formation in the electrophoresis gel, the extracted DNA's quantity and quality were confirmed and amplicons were sent to Bioneer Corporation for sequencing. The sequence received from this company was edited and reread using ChromasPro software and compared with the sequences recorded in the EzTaxon genomic database, and its similarity to different registered strains was studied (  (Table 6).
The similarity among isolated cocci strains was examined as shown in Table 7  Enterococcus's presence in fermented products has led to aroma development is still debatable. Some researchers believe that large amounts of Enterococcus produce disintegration and undesirable effects in some dairy products (Erdogrul & Erbilir, 2006). On the other hand, many reports point to this strain's desired effect on the aroma production and quality of dairy products (Graham et al., 2019).
Considering delbrueckii subsp. lactis (Shehata et al., 2016), and Lactobacillus pentosus (Iranmanesh et al., 2014).  Table 8 indicates the analysis of fatty acid and cholesterol levels in samples. The total saturated fatty acids in all samples were less than the control, while the total unsaturated fatty acids were higher than the control. In other words, using selected strains as an ad-  (Fox & McSweeney, 2006). While Ekinci et al. (2008) investigated the effect of cream fermentation with different probiotic bacteria as well as the addition of sunflower, soybean, and hazelnut oils on fatty acids of cream and they concluded that short-chain fatty acids such as butyric, caproic, and capric acids were significantly affected by culture media incorporation in sour cream and they increased compared to the control group where no fermentation took place. They also reported that the rate of change varied depending on the type of culture used and it was higher for sour cream with L. acidophilus than the others. However, unsaturated long-chain fatty acids were significantly affected by the type of oil added to them (Ekinci et al., 2008).

| The analysis results of fatty acid and cholesterol levels in the samples
Loric acid, myristic acid, and palmitic acid are known as atherogenic factors, and their presence in high level in food has a high risk of cardiovascular diseases (Kim et al., 2006). to 55% in the liquid culture medium. All strains could reduce cholesterol in cheese; however, Lactobacillus paracasei and Enterococcus lactis had the highest reduction (23%). There was also no negative effect on the sensory properties of cheese. It was recommended to use these strains to produce functional fermented dairy products (Albano et al., 2018).
The effect of low pH on the increase of cholesterol uptake by LAB has been discussed in studies performed by Rasic et al., 1992;. Reported that Lactobacillus fermentum strain KC5b, isolated from the human gut, was regarded as a candidate probiotic. It maintained viability for 2 h at pH 2 and grew in a medium with 4000 mg of bile acids per liter.
This strain was also able to remove a maximum of 14.8 mg of cholesterol per g (dry weight) of cells from the culture medium. Also, Aloğlu and Öner (2006) confirmed that the cholesterol uptake by Lactobacillus caesium at pH 4.6 and 3.96 was 4.05% and 25.35%, respectively.

| Sensory evaluation
As is evident in Table 9, comparing the average points about appearance, flavor, and consistency shows a significant difference among treatment samples and control (p < .05). Butter sample prepared with Neoscardovia arbecensis IBRC-M 4391 4378 gained higher scores for appearance, flavor, and consistency. Oxidation is the main cause of a reduction in the quality and shelf life of highfat dairy products. Peroxide is generated by the reaction of oxygen with unsaturated fatty acids, leading to the decomposition of fatty acids and carbonyl production, with unpleasant flavor and aroma in butter (Senel et al., 2011). In this study, undesirable flavor and aroma, such as bitterness and, rancidity, have not been detected by panelists.

| Conclusion
In this study, LAB with probiotic and cholesterol-lowering properties were isolated from traditional Iranian butters and identified to the species level. Due to our findings, it can be concluded that four strains mentioned above are considered suitable candidates as probiotic strains with a high ability to reduce cholesterol in butter. However,

Pediococcus pentosaceus IBRC-M 11045 and Lactobacillus pentosus
IBRC-M 11043 have a higher potential for cholesterol reduction among these strains. By examining the safety of these highly cholesterollowering strains (production of toxic metabolites, antibiotic resistance)

TA B L E 8 (Continued)
and their technological characteristics, we can recommend these strains to be used commercially in functional butter production.

ACK N OWLED G M ENTS
We would like to express our appreciation to Dr. Tajabadi Ebrahimi and Dr. Mohebat Mohebi for their help in running the project.

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

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared. TA B L E 9 Sensory evaluation of butters