Characterization of Lactobacillus fermentum PL9988 Isolated from Healthy Elderly Korean in a Longevity Village

Abstract

In this work, we wanted to develop a probiotic from famous longevity villages in Korea. We visited eight longevity villages in Korea to collect fecal samples from healthy adults who were aged above 80 years and had regular bowel movements, and isolated lactic-acid-producing bacteria from the samples. Isolated colonies that appeared on MRS agar containing bromophenol blue were identified by means of 16S rRNA sequencing, and 102 of the isolates were identified as lactic-acid-producing bacteria (18 species). Lactobacillus fermentum was the most frequently found species. Eight isolates were selected on the basis of their ability to inhibit the growth of six intestinal pathogens (Escherichia coli O157:H7, Salmonella enterica subsp. enterica Typhimurium, Salmonella enterica subsp. enterica Enteritidis, Enterococcus faecalis, Staphylococcus aureus, and Listeria monocytogenes) and their susceptibility to 15 antimicrobial agents. Among these eight isolates, four Lactobacillus fermentum isolates were found not to produce any harmful enzymes or metabolites. Among them, Lactobacillus fermentum isolate no. 24 showed the strongest binding to intestinal epithelial cells, the highest immune-enhancing activity, anti-inflammation activity, and anti-oxidation activity as well as the highest survival rates in the presence of artificial gastric juice and bile solution. This isolate, designated Lactobacillus fermentum PL9988, has all the characteristics for a good probiotic.

Introduction

The relationship between the microbiome and physical health as well as mental health is a hot issue at present [3, 33, 40]. The beneficial intestinal microbiome is very important for good human health, and controls and enhances the various intestinal activities such as the immune response in the intestine [36]. Several trials of microbiome transfer from healthy donors to recipients have produced successful results in both animal and human experiments [19, 38]. However, the total microbiome may contain harmful bacteria, which can cause a serious problem in the recipient [20]. This is why probiotics are a better choice for health, without health risk. Probiotics’ beneficial role for health and longevity have been proven for decades, and people are looking for new probiotics with additional functions such as anti-Helicobacter pyroli [33] or anti-obesity activities [25, 26] together with the well-known inhibitory activities on intestinal pathogens and immuneenhancing activities [34, 35].

The composition of the intestinal microbiome is known to be markedly influenced by food. For example, Lactobacillus acidophilus is frequently found in the gut of Western people, whereas this species is not commonly found in Eastern people [7, 27, 40]. The longevity of Bulgarian people has been attributed to their daily consumption of yogurt [28], which contains lactic-acid–producing bacteria (LAB) that inhibit the growth of harmful intestinal bacteria. Researchers have been trying to develop probiotics from their traditional foods and people [30]. In Korea, people enjoy various fermented foods daily that contain a large numbers of LAB. However, few LAB originated from Korean people or food have yet widely recognized compared with L. rhamnosus GG or L. delbrueckii subsp. bulgaricus.

One thing to be cautious about in developing a probiotic is antimicrobial resistance [31, 37]. In the past, when extrinsic and intrinsic resistances were not understood, people used to consider antimicrobial resistance as one of the good characteristics for probiotics. As antimicrobial resistance is becoming a problem worldwide, probiotics have been suspected as a reservoir for antimicrobial resistance to the human microbiome [5, 8, 12, 41, 42]. Since LAB are usually consumed in a large amount, extrinsic resistance in LAB, which can be transferred to a normal microbiome, would render a big health problem. Because only extrinsic resistance can be transferred to intestinal microflora, it is necessary to commercialize only those LAB with intrinsic resistance. Nowadays, many countries prohibit antimicrobial-resistant bacteria as probiotics, according to the resistance criteria set by ISO [13], EUCAST [9], and CLSI [45]. This is one of the reasons why strain number must be included on the label. If not, problems regarding intellectual property violation, safety, or claiming its beneficial functions as well as antimicrobial resistance will be provoked.

When the Korea Food and Drug Administration compared the intestinal microbiome compositions of Korean people aged over 40 years, people in the Korean longevity villages were found to have 2.4 times more Lactobacillus spp. and 5 times more Lactococcus spp. compared with city dwellers (unpublished data), which implies the close relationship between LAB and longevity. Based on this report, we decided to develop a probiotic candidate from healthy Korean senior citizens aged above 80 years and living in the well-known longevity villages.

 

Materials and Methods

Isolation of LAB from Fecal Samples

Eight villages with more than 200 residents each were selected from the three Korean longevity counties, Suncheon, Jangsoo, and Kurea in Chulla Province. From 69 healthy adults who were aged above 80 years and had regular bowel movement, fecal samples were obtained and transferred to the laboratory on ice within 6 h of collection. As soon as the samples arrived in the laboratory, they were diluted with sterile saline and inoculated on de Man Rogosa and Sharpe medium (BD, Sparks, MD, USA) containing 0.002% bromophenol blue (MRS-BPB) [23]. Well-isolated colonies were streaked on MRS-BPB to obtain pure colonies. The presumed Lactobacillus spp. colonies that were gram-positive, catalase negative, and rod shaped were further identified by means of 16S rRNA sequencing by Bionics (Seoul, Korea). The generated sequences were compared with known sequences on the EzTaxon-e server (http://www.ezbiocloud.net/eztaxon) [21] using the BLAST algorithm at the National Center for Biotechnology Information Web server (http://www.ncbi.nlm.nih.gov).

Assay of Susceptibility to Antimicrobials

The susceptibilities of LAB to antimicrobials were assayed by means of the liquid dilution method according to the ISO guideline [13]. The minimum inhibitory concentrations (MICs) of the following 15 antimicrobials were determined: ampicillin, chloramphenicol, ciprofloxacin, clindamycin, erythromycin, gentamicin, kanamycin, linezolid, neomycin, rifampicin, streptomycin, a quinupristin and dalfopristin combination (Synercid), tetracycline, trimethoprim, and vancomycin. Cells were incubated in the presence of each antimicrobial agent at 37℃ under anaerobic condition for 48 h. L. paracasei ATCC 334 was tested as a quality control. The MIC assay was performed in duplicates.

Assay of Ability to Inhibit Pathogen Growth

Escherichia coli O157:H7 ATCC 43894, Salmonella Typhimurium CCARM (Culture Collection of Antimicrobial Resistant Microbes) 8001, Salmonella Enteritidis CCARM 8010, Enterococcus faecalis ATCC 29212, Staphylococcus aureus CCARM 0045, and Listeria monocytogenes ATCC 19113C3a were inoculated on Muller–Hinton solid medium (BD), and a mixture of the culture supernatant of each LAB isolate and 3% agar (1:1) was added to each well formed with a sterile Pasteur pipette. Growth inhibition zones were measured after overnight incubation at 37℃, in triplicates.

Safety Test

Published methods were used to detect various virulence factors. For detection of hemolysis, LAB was inoculated on sheep blood agar and the color change to black or the halo around the colony was observed after 24 h [15]. Gelatin liquefaction was observed by inoculating bacterial cells on MRS agar slants containing 12% gelatin. After 6 weeks of incubation at 35℃, the inoculated slant medium was relocated at 4℃ for 4 h. Gelatin liquefaction was positive if the slant medium was not solidified [14]. Indole was detected by adding Kovac’s reagent on colonies grown on SIM medium (BD) [16]. Urease was detected by inoculating LAB on urea agar, and color change was observed after 12 h [17]. Amine production from phenylalanine was detected by adding 10% ferric chloride on colonies on MRS agar containing 0.2% D,L-phenylalanine (Sigma) [18]. Yellow color production by β-glucosidase was observed after the same volumes of bacterial cells (Abs600 = 4) and 0.2% ρ-nitrophenyl-β-D-glucopyranoside (Sigma) were mixed and incubated for 16h [4]. Yellow color production by β-glucuronidase was detected by mixing the same amount of 0.2% ρ-nitrophenyl-β-D-glucuronide (Sigma) and bacterial cells (Abs600 = 4) and incubating for 16 h [29].

Assay of Binding to Intestinal Epithelial Cells

Binding of the isolates to intestinal epithelial cells (Caco-2 cells; Korea Cell Line Bank, Seoul, Korea) was assayed by the method reported by Baccigalupi et al. [2] with slight modification. Caco-2 cells were cultured in Dulbecco’s modified minimal essential medium (Gibco, Grand Island, NY, USA) containing 20% inactivated fetal bovine serum (Gibco). After confluence, cells were washed three times with 10 mM phosphate-buffered saline (pH 7.0) and divided into new medium in a 30 mm cell culture plate. LAB grown in MRS (1 × 108 CFU/ml) were washed three times with 10 mM phosphate-buffered saline (PBS) with centrifugation and then added to each well. After incubation for 1 h at 37℃ in a CO2 incubator, the cells were washed three times with 10 mM PBS to remove unattached LAB. The washed cells were fixed in 4% fixing solution (100 ml of 35% formaldehyde, 16g Na2HPO4, 4 g NaH2PO4·H2O, distilled water to 1 L), and then bacterial cells were stained with Gram stain and observed under a light microscope. To count the number of bacterial cells attached to the intestinal epithelial cells, Caco-2 cells and bacterial cells were detached from the plate with 0.1% Triton X-100. After serial dilution, an aliquot (100 μl) was spread on MRS agar. After incubation for 2 days at 37℃, the CFU was counted. Experiments were conducted in triplicates, and standard deviations were obtained using the Sigma Plot 12.5 (Systat Software Inc., USA).

Assay of Resistance to Artificial Gastric Juice and Artificial Bile Solution

Resistance to artificial gastric juice was assayed by the method of Cho et al. [6] with slight modification. LAB grown overnight in MRS broth was collected by centrifugation and washed three times with sterile saline with centrifugation. Artificial gastric juice (1,000 U/ml pepsin, pH 3.0) was added to cells in sufficient volume to equal the original volume of the MRS broth and the mixture was incubated at 35℃. After 90 min, an aliquot was removed, serially diluted, and inoculated on MRS agar. The CFU was counted after overnight incubation at 37℃.

LAB treated with the artificial gastric juice as described above were collected by centrifugation. Then artificial bile solution (0.3% porcine bile extract, 1,000 U/ml trypsin, pH 7.0) was added to the cells in sufficient volume to equal the original volume of the MRS broth. The mixture was incubated at 35℃ for 90 min, and then an aliquot of the mixture was removed, serially diluted, and inoculated on MRS agar. The CFU was counted after overnight incubation at 37℃.

Assay of Immune-Enhancing Activity

The immune-enhancing activity of the LAB was assayed by means of a previously reported method [1] with slight modification. Macrophage cells (Raw 264.7, mouse monocytes, KCLB) were cultured in RPMI1640 medium (Gibco) containing 10% fetal bovine serum (Gibco) and 1% antimicrobial-antimycotic agents (Gibco) in a 96-well plate (1.0 × 105 cells/well), and then the cells were stimulated with 1 μg/ml lipopolysaccharide. Cells were suspended in RPMI1640 medium and added to each well (1.0 × 108 CFU/well). After incubation at 37℃ in a CO2 incubator for 24 h, the culture supernatant was collected and transferred to a tube. After the cells were completely removed by centrifugation, the culture supernatant was stored at –70℃ until use. The concentrations of TNF-α, IL-1β, and IL-6 in the cell culture supernatant were assayed by means of a cytokine sandwich ELISA kit (Biosource, Camarillo, CA, USA), in triplicate, according to the manufacturer’s protocol. The absorbance at 450 nm was measured in a microplate reader (SPECTRAmax 250 Microplate Spectrophotometer, Molecular Devices, Sunnyvale, CA, USA) using SOFTmaxPRO ELISA (Molecular Devices).

Assay of Resistance to Superoxide Anions

The resistance of the LAB to superoxide anions induced by paraquat (1,1’-dimethyl-4,4;-bipyridinium dichloride) was determined by means of the diffusion assay method [22]. Overnight-cultured LAB cells were collected by centrifugation and dispersed in saline at a concentration of 107 CFU/ml. After 0.1 ml of the dispersed cells was spread on an MRS agar plate, a paper disk containing 10 μl of paraquat solution at 10 mM or 100 mM in sterile saline was placed in the center of the plate. After overnight culture at 37℃, the diameter of the growth inhibition zone was measured with a ruler. L. fermentum PL9005 [35] was used as a positive control.

 

Results and Discussion

Recently, several trials of transfer of microbiomes have produced successful results in animals and humans [19, 20]. Since microbiomes may contain pathogens and antimicrobialresistant bacteria, the wide use of microbiome therapy needs to be done carefully. Compared with the total microbiome transfer, LABs have been generally regarded as safe (GRAS) and have a long history of use as probiotics. However, LABs need to be checked for their safety, such as antimicrobial resistance and production of harmful metabolites or enzyme before their use as probiotics. In this study, we wanted to develop a well-characterized and safe probiotic candidate from Koreans.

Eight villages were selected for this study. Each village had at least more than 200 residents and higher percentages of seniors aged over 80 years. A total of 102 fecal samples were collected from seniors aged over 80 years with regular bowl movement. Every colony with a different morphology on MRS-BPB agar plate was collected. Several different colonies were collected from some people, while some had only one or no colony. A total 102 LAB were isolated from 69 people and these were identified as 18 LAB species by means of 16S rRNA sequencing (Table 1). L. fermentum was the most prevalent Lactobacillus species in healthy seniors in the longevity villages. More than one quarter of senior residents (19 people among 69, 27.5%) had L. fermentum. L. fermentum has been reported both in Western people and Eastern people and various foods [7,40]. This species is known to be safe and there has been no report of infection or disease due to it. Several strains of L. fermentum are on the market: L. fermentum KLD, L. fermentum RC-14, L. fermentum KC5b, L. fermentum CECT5716, PCC® L. fermentum, L. fermentum VRI-003, L. fermentum ACA-DC 179, L. fermentum CECT5716, L. fermentum ME-3, and L. fermentum CECT5716.

Table 1.aEvery colony was isolated from the same sample when they had a different morphology.

As the first step to check for safety, the MIC of the LAB was assayed. It is very important to use antimicrobialsusceptible LAB as a probiotic [37]. In the old days, people used to consider antimicrobial-resistant LAB as good probiotics that can survive during antimicrobial treatment. People used to select LAB that appeared on media containing antimicrobials. The MIC assay showed that only eight isolates (11.6%) among 69 LAB were susceptible to gentamicin, erythromycin, and clindamycin according to CLSI criteria and also susceptible to clindamycin, chloramphenicol, and ampicillin according to EUCAST criteria (Table 2). The high percentages of antimicrobialresistant LAB in these villages were not expected. This showed the importance of the MIC assay for a probiotic candidate.

Table 2.a16S rRNA sequencing revealed >99% homology. bAMP, ampicillin; CHL, chloramphenicol; CIP, ciprofloxacin; CLI, clindamycin; ERY, erythromycin; GEN, gentamicin; KAN, kanamycin; LIN, linezolid; NEO, neomycin; RIF, rifampicin; STR, streptomycin; SYN, Synercid (quinupristin + dalfopristin); TET, tetracycline; TRI, trimethoprim; VAN, vancomycin.

Among eight antimicrobial-susceptible LAB, four isolates were L. fermentum. Three L. fermentum isolates (isolate nos. 24, 66, and 76) inhibited the growth of five pathogens, and one L. fermentum isolate (no. 79) inhibited all six pathogens tested in this study (Table 3). When safety tests were performed, none of the antimicrobial-susceptible L. fermentum isolates produced any harmful enzymes or metabolites (data not shown).

Table 3.Diameter of growth-inhibition zone of intestinal pathogens by selected LAB isolates.

These four L. fermentum isolates showed other beneficial characteristics as a probiotic. Isolate nos. 24 and 66 showed good resistance to acid and bile (Table 4) and isolate no. 24 showed the best binding activity to Caco-2 cells (Figs. 1 and 2). Growth of all four L. fermentum isolates was not inhibited by superoxide anions produced by paraquat (even at 100 mM), suggesting the strong anti-oxidation activity, whereas the growth of the positive control, L. fermentum PL9005 [35, 36], was inhibited by the presence of 100 mM paraquat (Fig. 3). Based on the various results above, isolate no. 24 was chosen as the best candidate for a probiotic.

Table 4.alog10 CFU is the unit for the values; bincubation time.

Fig. 1.Micrographs of intestinal epithelial cells (Caco-2 cells) incubated with L. fermentum isolates: (A) cells without LAB; cells with (B) L. fermentum no. 24, (C) L. fermentum no. 66, (D) L. fermentum no. 76, and (E) L. fermentum no. 79. Each lactic acid bacterium was incubated with Caco-2 cells for 1 h, and then the plates were washed to remove unattached LAB. After fixing in 4% fixing solution, the bacterial cells were stained with Gram stain and observed under a light microscope, as described in Materials and Methods.

Fig. 2.Assay of binding of L. fermentum isolates to intestinal epithelial cells (Caco-2 cells). Each LAB was incubated with Caco-2 cells for 1 h, and then the plates were washed to remove unattached LAB. Cells and LAB were detached from the plate and an aliquot of serial dilution was spread onto MRS agar, as described in Materials and Methods. After incubation for 2 days at 37℃, the number of CFU was counted.

Fig. 3.Growth of (A) L. fermentum PL9005 (positive control), (B) L. fermentum no. 24, (C) L. fermentum no. 66, (D) L. fermentum no. 76, and (E) L. fermentum no. 79 in the presence of superoxide anions generated with 10 mM paraquat (left disk) or 100 mM paraquat (right disk) and in the absence of superoxide anions (top disk).

Additional beneficial activities of isolate no. 24 were also observed. LABs with a gram-positive bacterial cell wall (capsular polysaccharides, pepetidoglycans, lipoteichoic acid) have immune-enhancing activity by inducing cytokines [11]. On the contrary, too high concentration of cytokines induced by lipopolysaccharide (LPS) will cause tissue damage and severe septic shock [47]. LPS is one of the constituents of cell walls in gram-negative bacteria and an important factor for septic shock and gram-negative sepsis. LPS activates mononuclear phagocytes and induces the production of cytokines, including TNF-α, IL-1β, and IL-6 [1]. Humans are much more sensitive to LPS than any other animals. In most experiments, macrophage cells are treated with 1 μg/ml LPS, which is equivalent to 1 mg/kg in humans, which can cause severe septic shock [44]. When Raw 264.7 cells were treated with LPS, they produced a large amount of TNF-α. When Raw 264.7 cells were treated with both LPS and LAB, the LAB inhibited the signaling pathways for TNF-α production or bind to LPS, preventing its interaction with the Raw 264.7 cell and resulting in a decrease in TNF-α production [24]. This is how LAB can suppress inflammation [10]. As shown in Fig. 4, isolate no. 24 increased the production of TNF-α, IL-6, and IL-1β, showing the immune-enhancing activity, and decreased the lipopolysaccharide-stimulated production of TNF-α, IL-6, and IL-1β in Raw 264.7 cells (Fig. 4), showing the antiinflammatory activity [11]. This suggested that isolate no. 24 can lessen ulcer colitis and intestinal bowel disease [10].

Fig. 4.Immune-enhancing activity of L. fermentum no. 24. Raw 264.7 macrophages were incubated for 24 h with lipopolysaccharide (1 μg/ml) and 1.0 × 108CFU of L. fermentum no. 24. The amounts of TNF-α, IL-6, and IL-1β were measured in triplicate by means of a cytokine sandwich ELISA assay, as described in Materials and Method. Standard deviations were obtained using the Sigma plot 12.5 (Systat Software Inc. USA). Control, cells without any treatment; LPS control, cells treated with LPS; LAB control, cells treated with L. fermentum no. 24; LPS+LAB, cells treated with LPS and L. fermentum no. 24.

Conclusively, L. fermentum isolate no. 24, which was isolated from a healthy 86-year-old woman living in Yongdong village of Sunchang county, one of the most famous longevity villages in Korea, was characterized as the best probiotic candidate and designated as L. fermentum PL9988. L. fermentum PL9988 has all the good characteristics of a probiotic as suggested by many researchers [22, 43]: 1, origin from healthy people; 2, anti-oxidation activity; 3, inhibitory activity of pathogens; 4, immune-enhancing activity; 5, resistance to low pH and bile acid; 6, safety without harmful metabolites and enzymes; 7, adhesion to intestinal epithelium; and 8, anti-inflammation activity. In addition, L. fermentum PL9988 is susceptible to antimicrobials. This strain was deposited in the Korea Agriculture Culture Collection (KACC, Korea) with the number KACC 91834P and it has been assigned Korean as patent application no. 2013-0086153. The 16S rRNA sequence was deposited in GenBank (acc. no. KF649831).

Since the human microbiome differs among the human population depending on age, cultural traditions, and geography [46], we hope that the L. fermentum PL9988 developed from a Korean will be widely used as a probiotic to exert better beneficial effects in the Korean population.