Introduction

The dietary imbalance, especially High-Sucrose Diet (HSD) intake, induces chronic metabolic disorders such as hyperlipidemia and hyperglycemia [1]. Based on the hypothesis that the development of the hyperglycemia is associated with the dysregulation in intestinal microbial ecology, many scholars have tried to employ potential therapeutic probiotics to restore gut dysbiosis preclinically [2-14] or clinically [15-19], leading to the alleviation of the hyperglycemia. The probiotic strains reported to exert the anti-hyperglycemic effects belonged to various kinds of bacterial species such as Lactiplantibacillus plantarum, Lacticaseibacillus casei, Lacticaseibacillus paracasei, Lacticaseibacillus rhamnosus, Limosilactobacillus reuteri, and Limosilactobacillus fermentum. However, most of those works lacked the description of the initial screening test for searching the target strain from varied probiotic libraries.

Due to the time and cost limitations, the rodents or human subjects are not available for a massive screening of probiotic libraries. Although being time and cost-efficient, in vitro cell lines are inappropriate for the probiotics due to the susceptibility to the probiotics-derived acidic compounds. Several studies made an effort to address this issue as applying the heat-killed probiotics to the in vitro cell lines [3,20]. However, the subsequent rodent test used the live probiotics [21], which could distort the initial screening results.

The fruit fly, Drosophila melanogaster, can be alternatively used for the massive screening of probiotic libraries due to the acid resistance, short life cycle, and high fecundity [22]. Importantly, HSD feeding to Drosophila effectively induced hyperglycemia in larva [23,24] or in adult flies [25]. In addition, the molecular mechanisms underlying hyperglycemia were shared between fly and rodent [26]. The HSD-fed Drosophila showed the decrease in mRNA expression of Insulin-Like Peptide 2 (Ilp2), the Drosophila ortholog of the insulin, in Insulin-Producing Cells (IPCs) located in the brain. Similarly, the HSD-fed mice exhibited the reduction in insulin secretion from pancreatic islets [26]. The insulin deficiency induced the shortage of glucose uptake into the peripheral cells, which was accompanied by cellular fasting responses represented by the elevation of the mRNA expressions of gluconeogenic genes such as phosphoenolpyruvate carboxykinase (Pepck) and glucose 6-phosphatase (G6Pase) [23-25,27]. In particular, Drosophila contains trehalose 6-phosphatase (Tps1) in addition to G6Pase for gluconeogenesis [28].

Human oral cavity is colonized with a variety of intrinsic probiotics. The aim of this study is to take advantage of HSDfed Drosophila to search out a new anti-hyperglycemic probiotic strain among 168 probiotic libraries derived from human Tongue Coating (TC) biospecimens donated from 120 healthy volunteers (100 adults and 20 younger children).

Materials and Methods

Isolation and identification of TC-originating probiotic strains

All procedures were conducted in accordance with relevant guidelines and regulations approved by the ethics committee of the Institutional Review Board of the Apple Tree Dental Hospital (approval number: ATDH-2021-0001) and the Korea National

Institute for Bioethics Policy (approval number: P01-202111-31002). The TC samples, collected from 100 adults and 20 younger children, were distributed by the Biobank of Apple Tree Dental Hospital, a part of the Korea Biobank Network (KBN). The TC samples were resuspended in 3 mL of phosphate-buffered saline (PBS) and stored at -80 ^°C until used. To isolate TC-derived probiotic strains, 1 mL of each TC sample was plated onto ten plates (100 mL per plate) of LactoBacillus Selective (LBS) agar (Kisan Bio Inc., Korea) and anaerobically incubated for 48 h at 37 ^°C. The colonized isolates were identified by 16S rRNA partial sequencing (Macrogen Inc., Korea).

Fly culture and food

The Drosophila melanogaster white-eyed w¹¹¹⁸ line was kept at a density of fifteen males and five females in each vial with 12 hour on/off light cycle at 25 ^°C. Three-day old flies fed ad libitum were used in all experiments. Flies were raised on food containing 2% (w/V) yeast and three different amounts of sucrose (w/V) (10%, 20%, and 40%), indicated with S10Y2, S20Y2, and S40Y2, respectively. The starvation diet contains only 1% sucrose.

Probiotics supplementation and hemolymph glucose measurements

Three-day old flies were transferred to vials containing the starvation diet and maintained for 12 h. Fasted flies were refed the experimental foods supplemented with probiotics at a concentration of 10¹⁰ Colony-Forming Units (CFU) per mL. After three days, males were anesthetized with posterior abdomen cropped, and vortexed in phosphate-buffered saline (PBS), allowing hemolymph to be eluted. After centrifuging at 9,000 g for 5 minutes at 4 ^°C, supernatants were used for determination of glucose levels by using the glucose detection kit (Sigma Aldrich Inc., USA). At least three biological replicates were measured on a microplate reader (SpectraMax iD3, Molecular Devices Inc., USA).

Acid-resistance test of probiotics

Resistance to low pH was evaluated as described by Ye et al. [28] with minor modifications. Briefly, probiotic strains were anaerobically cultured in de Man Rogosa Sharp (MRS, SigmaAldrich Inc., USA) liquid for 48 h at 37 ^°C. The cultures were inoculated into PBS adjusted to pH 2.5 or pH 6.5. After 6 h, the cultures were ten-fold diluted four times, thereafter each 100 μL of 1^st ~ 4^th dilutions were cultured on MRS agar. The x number of CFU was counted in y times diluted plate which presented the number of CFU between 30 and 200. The survival rate was calculated as the ratio of CFU of pH 2.5 to that of pH 6.5. The experiments were performed in triplicate and mean values were calculated.

Quantitative RT-PCR

Total mRNA was extracted using the RNeasy mini kit (Qiagen Inc., USA). The cDNA was generated with 0.5 μg total mRNA using the PrimeScript^TM RT-PCR kit (Takara Bio Inc., Japan) according to the manufacturer’s instructions. The generated cDNA was used for real-time RT-PCR using Exicycler^™ 96 Real-Time PCR systems (Bioneer Inc., Korea), SYBR Green PCR master mix (Thermo Fisher Scientific, USA), and primers for Ilp2 (5-CT-

GAGTATGGTGTGCGAGGA-3 and 5-CAGCCAGGGAATTGAGTACAC-3), CCHa2 (5-TCGTTATCTGCACCGTGGTC-3 and 5-CCCTTTTTCGCTTGGCTCT-3), Pepck (5-AGAAGAAGTACATCACTGCCGCCT-3 and 5-TCCCTGCGAGTCAAACTTCATCCA-3), Tps1 (5-TCCGATGAGATCCTACAGGGTATG-3 and 5-CGCCATGTTCCACCAGCAGATTG-3), and Rp49 (5-CGGATCGATATGCTAAGCTGT-3 and 5-GCGCTTGTTCGATC-CGTA-3). The reaction conditions consist of the first denaturation step for 3 min at 95 ^°C followed by 40 cycles of a denaturation step for 10 s at 95 ^°C and an annealing/extension step for 20 s at 60 ^°C.

Antibiotic susceptibility test

Minimal Inhibitory Concentration (MIC) values for probiotics were determined in LSM medium (90% of Iso-Sensitest broth (KisanBio Inc., Korea) and 10% MRS broth) according to the ISO 10932:2010 broth microdilution procedure [30]. LSM medium was supplemented with serial dilutions of antibiotic compounds including GEN (0.5–256 mg/L), KAN (2–1,024 mg/L), STR (0.5–256 mg/L), TET (2–1,024 mg/L), ERY (0.016–8 mg/L), CLIN (0.032–16 mg/L), CHL (0.125–64 mg/L), and AMP (0.032–16 mg/L). Overnight-cultures of probiotics were inoculated into the LSM medium containing antibiotics and anaerobically incubated at 37 ^°C for 2 days. MIC was determined as the lowest concentration of antimicrobial compounds at which the growth of the probiotics was inhibited. The growth was measured at 600 nm absorbance using a SpectraMax iD3 microplate reader (Molecular Devices, USA) and MICs were compared to the cut-off values recommended by the European Food Safety Authority (EFSA) [31].

Hemolysis test

Hemolysis was observed by anaerobic incubation of the probiotics in Tryptic Soy Agar (TSA) (Fisher Scientific  Inc., USA) supplemented with 5% sheep blood (KisanBio Inc., Korea) at 37 ^°C for 2 days [32]. The loss of blood color around colonies indicates hemolysis.

Statistical analysis

All data were expressed as mean ± Standard Deviation (SD). Statistical analysis was performed through the analysis of variance (ANOVA) followed by the Tuckey’s multiple comparison.

Statistical significance was attributed to p values < 0.05 (*), 0.01 (**), and 0.001 (***). The software GraphPad Prism v5 (GraphPad Inc., USA) was used for the analysis.

Results

Isolation of probiotic strains from the human TC biospecimens

The human TC specimens distributed by the Korea Oral Biobank Network (KOBN) were originated from 100 adults (aged from 19 to 80) and 20 younger children (aged under 9). As the results of cultivating each TC specimen on LBS agar plates, a total of 168 colonies were isolated, which belonged to 9 different genera and 14 different species according to the reclassified nomenclature (Table 1) [33]. Lacticaseibacillus rhamnosus (n = 52) showed the highest incidence, followed by Limosilactobacillus fermentum (n = 32), Lactiplantibacillus plantarum (n = 15), Lactobacillus gasseri (n = 14), Latilactobacillus curvatus (n = 14), Limosilactobacillus vaginalis (n = 13), Latilactobacillus sakei (n = 8), Lacticaseibacillus paracasei (n = 7), Ligilactobacillus salivarius (n = 6), Limosilactobacillus mucosae (n = 2), Lentilactobacillus sunkii (n = 2), Levilactobacillus brevis (n = 1), Levilactobacillus graminis (n = 1), and Liquorilactobacillus nagelii (n = 1). Among them, Lcb. rhamnosus, Llb. fermentum, Lpb. plantarum, Llb. sakei, and Lcb. paracasei are species found in both adults and younger children.

Table 1: The list of probiotic strains isolated from adults and younger children.

Induction of hyperglycemia in Drosophila

To find out the sucrose concentration to induce hyperglycemia in Drosophila, adult flies were fed three kinds of diets differing in sucrose concentrations (10%, 20%, and 40%, respectively). As the results, the hemolymph glucose levels increased in proportion to the sucrose concentrations, leading to peak glucose level at 40% sucrose (Figure 1).

Figure 1: The graphs showing hemolymph glucose levels measured in Drosophila adult flies fed 2% yeast and three different concentrations of sucrose as indicated. Three times (five males per time) were replicated and subjected to ANOVA followed by the Tuckey’s multiple comparison. Statistical significance was attributed to p values < 0.01 (**) and 0.001 (***).

The massive screening of probiotic libraries using Drosophila

The massive screening was conducted to examine the anti-hyperglycemic effects of 168 libraries using Drosophila adult flies fed 40% sucrose and 2% yeast diet (triplicate of five males for each probiotic strain). As the results, three strains belonging to Lpb. plantarum species (DM043, DM049, and DM083) were determined to be the most effective in suppressing hyperglycemia as indicated with gray bars in Figure 2.

Figure 2: The graphs showing hemolymph glucose levels of Drosophila fed 40% sucrose and 2% yeast (S40Y2, black bars) and S40Y2fed Drosophila supplemented with 168 probiotic libraries from DM001 to DM177 with exemption of nine. Gray bars indicated the probiotic strains exerting the strongest effects on anti-hyperglycemia.

Acid-tolerance test

The acid-tolerance test determined that Lpb. plantarum DM083 was significantly more acid-tolerant than DM043 or DM049 (Figure 3A), implying that the Lpb. plantarum DM083 is more likely to remain alive after passing through the stomach compared to DM043 or DM049. The significance of being alive in relation to the anti-hyperglycemic effect of Lpb. plantarum DM083 was reflected by the result that the heat-killed Lpb. plantarum DM083 failed to restore hyperglycemia (Figure 3B).

Figure 3: (A) The graphs illustrating survival rates (%) of Lpb. plantarum DM043, DM049, and DM083 which were calculated as the ratio of the number of CFU incubated in pH2.5 to that incubated in pH6.5. (B) The graphs demonstrating hemolymph glucose levels of S40Y2-fed Drosophila supplemented with live or heat-killed Lpb. plantarum DM083. Three times were replicated and subjected to ANOVA followed by the Tuckey’s multiple comparison. Statistical significance was attributed to p values < 0.01 (**) and 0.001 (***).

The mechanisms underlying the anti-hyperglycemic effect of probiotics

To investigate the molecular mechanisms under which Lpb. plantarum DM083 can alleviate the hyperglycemia, mRNA levels of genes related to the insulin and glucose metabolisms were measured in Drosophila adult flies feeding 10% sucrose (normal diet, ND), 40% sucrose (high-sucrose diet, HSD), and HSD + 10¹⁰ CFU/mL Lpb. plantarum DM083 (HSD+DM083). The mRNA levels of CCHamide-2 (CCHa2), the insulin secretagogue, measured in the intestine tissues were lowered in HSD compared to ND, but significantly upregulated in HSD+DM083 (Figure 4A). The mRNA levels of Ilp2, the Drosophila ortholog of insulin, measured in the brain tissues were also reduced in HSD compared to ND, but significantly increased in HSD+DM083 (Figure 4B). The mRNA levels of both Pepck and Trehalose6-phosphatase (Tps1), the gluconeogenic genes, measured in abdomen tissues were higher in HSD compared to ND, but significantly downregulated in HSD+DM083 (Figure 4C and D). Taken together, Lpb. plantarum DM083 intervention successfully restored HSD-induced dysregulations in insulin and glucose metabolisms.

Figure 4: The graphs showing mRNA levels of CCHa2 (A), Ilp2 (B), Pepck (C), and Tps1 (D) in Drosophila fed ND, HSD, and HSD+DM083. Three times were replicated and subjected to ANOVA followed by the Tuckey’s multiple comparison. Statistical significance was attributed to p values < 0.05 (*) and 0.001 (***).

Safety evaluations of Lpb. plantarum DM083

The antibiotic susceptibility of the Lpb. plantarum DM083 was evaluated by MIC values according to EFSA criteria for Lpb.

plantarum, resulting that Lpb. plantarum DM083 was susceptible to all the tested antibiotics as indicated in Table 2. Additionally, no hemolytic activity of the Lpb. plantarum DM083 was observed when cultured on TSA supplemented with sheep blood for 2 days at 37 ^°C (data not shown).

Table 2: Antibiotics susceptibility test of Lpb. plantarum DM083.