Inhibitory Effects of the Spore-forming Bacillus spp. on the Expression Levels of eae, luxS, u, and ctxM Genes in E. coli Isolates

Background: Escherichia coli antibiotic resistance are one of the major health problems in many countries. Today, researchers are focusing on novel approaches for the treatment of E. coli infections, including the use of spore-forming probiotics for their high stability in the harsh gastrointestinal (GIT) environment. Results: The presence of resistance genes (u, eae, luxS, and ctxM ) was conrmed by PCR in four out of 40 E. coli isolated from 300 fecal samples of patients suffering from diarrhea. Following conrming the ability of broiler-derived Bacillus coagulans and Bacillus subtilis to survive under the simulant gastric environment, they were co-cultured with the four E. coli isolates harboring the studied resistance genes. Results of the expression of u, luxS, eae, and ctxM genes revealed that the expression levels of these genes decreased signicantly (p-value < 0.05) after co-culture of E. coli with B. subtilis ATCC 6633 and broiler-derived B. subtilis. However, the broiler-derived B. subtilis showed a greater effect on the expression levels of eae, luxS, and ctxM. B. coagulans MTCC 5856 and broiler-derived B. coagulans also led to signicant decreases in expression levels of all the studied genes (p-value < 0.05), while the standard strain had a greater impact on the expression of ctxM than the broiler-derived strains. Conclusion: The broiler-derived isolates had a greater capacity to decrease the expression of these genes than the standard strains, proposing their adoption for dietary supplementations. However, further studies are required to study the effects of these spore-forming bacteria on E. coli infections in vivo and their possible inuence on lactose tolerance, nutritional absorption, and reduced cholesterol levels.


Background
Escherichia coli normally colonizes the gastrointestinal tract of human infants and forms the normal intestinal micro ora over time. These commensal E. coli strains can lead to infections following compromising the host immune system. In fact, several E. coli clones have evolved by acquiring speci c virulence traits, allowing them to cause a wide array of infections, including the enteric disease, urinary tract infections (UTIs), and sepsis [1][2][3].
In fact, pathogenic E. coli strains harbor various virulence factors, including Ag43, LuxS, and intimin. The surface protein antigen 43, Ag43, is encoded by the u gene in 43 min on the E. coli chromosome with the ability of promoting bacterial adhesion and 3D bio lm structures [4]. LuxS has been suggested to be an interspecies signal with a crucial role in physiologic functions of bacteria by contributing to quorum sensing [5]. Finally, the intimin surface protein, encoded by the eae gene, contributes to bacterial attachment to the intestinal cells and induces attaching and effacing lesions [6].
The emergence of resistant E. coli strains and the subsequent failure in antibiotic therapy have become a worldwide health concern. Resistance in Gram-negative bacteria is attributed to various mechanisms, including altered target sites, enzymatic inactivation of antibiotics, and active e ux pumping [7].
However, one of the most frequent resistance mechanisms adopted by E. coli strains is the production of Extended-Spectrum Beta-Lactamase enzymes (ESBL). The CTX-M beta lactamase types are currently the most common forms of these enzymes leading to antimicrobial resistance [8].
Antimicrobial resistance has therefore necessitated the development of other therapeutic methods to combat E. coli infections. In this regard, one of the safest ways is the exploitation of probiotic bacteria. For many years, Lactobacilli and Bi dobacteria have been used to treat gastrointestinal tract (GIT) diseases [9]. However, these bacteria are sensitive to physiological conditions, such as pH of the stomach and bile salts. In addition, various conditions of production, storage, and transportation may affect their bioavailability [10]. To overcome these hurdles, therefore, using spore-forming Bacillus spp. as probiotics has recently come to notice. These bacteria have an innate ability to produce a wide number of enzymes and vitamins, and are highly tolerant to the harsh environment of GIT. Moreover, Bacillus spp. are ideal candidates for probiotics owing to their stability during food processing and storage [11][12].
Considering the merits of Bacillus spp. as probiotics, this study aimed to investigate the effects of B. coagulans and B. subtilis against the expression levels of eae, u, ctxM, and luxS genes to achieve a deeper insight into the mechanisms by which spore-forming Bacillus spp. can affect E. coli infections.

Results
Isolation of E. coli from fecal samples and the presence of the studied genes Of 300 fecal samples obtained from patients with diarrhea, 40 were positive for E. coli. The bacterial strains were identi ed and con rmed by cultivation of samples on MacConkey agar and EMB agar, in addition to performing complementary biochemical tests, including MR/VP, consumption of citrate, production of urease and lysine decarboxylase, and production of ornithine. The presence of u, luxS, eae, and ctxM genes in clinical E. coli isolates was con rmed through designing speci c primers and PCR assay. According to the results of ampli cation assay, the prevalence rates of luxS, u, ctxM, and eae genes were 35% (n=14), 62.5% (n=25), 37.5% (n=15), and 17.5% (n=7), respectively. Among these, four E. coli isolates carried all the u, luxS, ctxM, and eae genes.
Isolation of spore-forming Bacillus spp. from the intestinal content of broilers Spore-forming bacteria were selected by heat or ethanol treatment of the intestinal content of 10 broilers.
Treated samples were subsequently identi ed by the API CHB test. Collected colonies consisted of those identi ed as B. subtilis and B. coagulans. Isolates showing positive catalase test were differentiated from the anaerobic spore-forming Clostridium spp. Moreover, isolates showing no hemolysis on 5% sheep blood agar were considered as Bacillus spp. B. subtilis ATCC 6633 and B. coagulans MTCC 5856 were used as controls. Molecular analysis was used to con rm the characterization of the isolates. According to the results of biochemical and microbiological tests and PCR, one B. subtilis and one B. coagulans were isolated from the intestinal content of broilers.
Acid and bile tolerance of Bacillus cells The survival rates of B. subtilis and B. coagulans cells were evaluated after exposure to both simulated gastric conditions and bile salts. According to results of the CFU measurements, both B. coagulans and B. subtilis were resistant to 1 mg/mL of pepsin, 1 mg/mL of trypsin, 0.2% bile salts, and a pH of 2.
Expression levels of u, luxS, eae, and ctxM genes in E. coli after co-culture with B. coagulans and B. subtilis isolates In order to study the effects of B. coagulans and B. subtilis as the studied probiotics on the expression levels of genes involved in bacterial attachment, bio lm formation, signaling, and antimicrobial resistance, total RNA of four isolates harboring all u, luxS, eae, and ctxM genes was extracted after the co-culture assay at the logarithmic growth phase.
Results of the expression of u, luxS, eae, and ctxM genes revealed that the expression levels of these genes decreased signi cantly (p-value < 0.05) after co-culture of E. coli with B. subtilis ATCC 6633 and broiler-derived B. subtilis. However, the broiler-derived B. subtilis showed a greater effect on the expression levels of eae, luxS, and ctxM. B. coagulans MTCC 5856 and broiler-derived B. coagulans also led to signi cant decreases in expression levels of all the studied genes (p-value < 0.05), while the standard strain had a greater impact on the expression of ctxM than the broiler-derived strains. A comparison on the results of the expression levels of the studied genes indicated that B. subtilis had a more substantial effect on the reduced expression levels of eae, ctxM, and u than B. coagulans .

Discussion
E. coli strains are considered as harmless commensal bacteria; however, several strains have gained the ability to cause infections in human hosts. These pathogenic variants have obtained a variety of virulence factors that confer environmental adaptations and pathogenicity [13][14]. Moreover, these pathogenic variants have adopted several mechanisms to resist antibiotic therapy with the production of beta-lactamase enzymes being the most common strategy [15]. Therefore, nding novel therapeutic approaches to ght E. coli infections seems to be necessary. In this regard, using the probiotic properties of various bacteria has become an interesting subject in recent years [16].
Lactic acid-producing bacteria are commonly known as probiotics. In recent years, however, researchers have focused on developing drugs based on more resistant bacterial species due to the antagonistic environment of the human gastrointestinal tract [17]. Spore-forming Bacillus spp. are one of these bacteria with high levels of tolerance to harsh conditions. Moreover, these bacteria are highly stable during the food manufacturing and storage processes, making them suitable candidates for health improvement strategies. In fact, approved spore-forming Bacillus spp. are currently being used in dry probiotic products due to their ability of survival in conditions of high or low temperatures, aridity, and high oxygen levels [18]. This ability is ascribed to the possession of an endospore with a cortex replete with peptidoglycan and protein-containing materials [19]. According to several studies, the bene ts of using Bacillus spp. for their probiotic properties include the improvement of immunity systems and prevention of GIT disorders, including diarrhea and irritable bowel syndromes [20]. However, limited studies have focused on the effects these probiotic bacteria may have on the virulence factors of gastrointestinal pathogenic bacteria including E. coli. Here, the effects of the standard strains and broilerderived B. coagulans and B. subtilis were investigated on the expression of various virulence genes in E. coli, including u (coding for Ag43 protein involved in bio lm formation), eae (coding for a protein involved in intimate attachment of E. coli to the intestinal epithelial cells), and luxS (coding for a protein involved in bacterial quorum sensing). In addition, the potential effects of these strains were investigated on the expression levels of ctxM, a gene commonly associated with antimicrobial resistance in E. coli. To this end, 40 E. coli out of 300 fecal samples of patients with diarrhea were isolated using cultivation and biochemical methods. According to results of PCR, the prevalence rates of u, eae, luxS, and ctxM genes in these isolates were 62.5% (n=25), 17.5% (n=7), 35% (n=14), and 37.5% (n=15), respectively. Moreover, ampli cation results showed the presence of all the studied genes ( u, eae, luxS, and ctxM) in four E. coli isolates, indicating their importance in bacterial pathogenicity. In addition to the standard strains, one B. subtilis and one B. coagulans were isolated from the intestinal contents of 10 broilers to examine the effects of spore-forming Bacillus spp. and determine their tolerance to the simulant GIT environment. PCR assay was utilized for the molecular con rmation of B. subtilis and B. coagulans isolates. According to the results, the studied B. subtilis and B. coagulans were tolerant to low pH [2], 0.2% bile salts, and gastric enzymes, including pepsin and trypsin, suggesting their suitability for probiotic use. Then, the effects of B. coagulans and B. subtilis on the expression levels of the studied genes were studied using the real-time PCR (RT-PCR) method. According to the results, the expression levels of all the studied genes decreased signi cantly after co-culture of E. coli isolates harboring these genes with each of B. subtilis ATCC 6633 and B. coagulans MTCC 5856, as well as broiler-derived B. subtilis and B. coagulans strains. These results reveal the important roles of B. subtilis and B. coagulans isolates in reducing the expression of genes involved in the virulence and antimicrobial resistance in E. coli isolates. Moreover, since broiler-derived isolates showed similar to or even better effects than the standard strains, gut microbiota of broilers can be exploited as suitable sources of these probiotic bacteria in case of unavailability of commercial strains.
Furthermore, the results indicated that B. subtilis had a greater in uence on the expression levels of eae, ctxM, and u, while B. coagulans showed a greater in uence on the expression levels of luxS. These results suggest that B. subtilis mostly affects the expression of genes involved in attachment, bio lm formation, and antibiotic resistance, whereas B. coagulans has a greater potential to reduce the expression of genes involved in bacterial quorum sensing, proposing the different capabilities of these probiotic bacteria and different pathways they may adopt to combat bacterial pathogens.
Several studies have shown the probiotic properties of spore-forming Bacillus spp. on E. coli infections. Guo et al. collected a total of 124 intestinal samples from broilers and isolated six spore-forming Bacillus spp. with the ability to inhibit E. coli K88 and E. coli K99. Of these isolated Bacillus spp., B. subtilis MA139 showed a great tolerance to pH=2 and 0.3% bile salts, with the highest activity against E. coli strains by co-culture method [21].
In another study by Kim et al., dietary supplementation of B. subtilis DSM 25841 reduced the E. coli F18 infection signi cantly. Overall, they showed a positive in uence of this probiotic microorganism on the promotion of health after infecting pigs with pathogenic E. coli [22]. Also, Lin et al. revealed the effect of B. coagulans on the intestinal microbiota of broilers. In their study, probiotic supplementation with 0.02% and 0.04% B. coagulans led to a signi cant increase in Lactobacillus counts and a signi cant reduction in E. coli counts of duodenum and cecum [23]. The effects of these spore-forming Bacillus spp. on the expression of virulence genes and beta-lactamase genes in E. coli were not found in the literature. However, Medellin-Pena et al. showed decreased expression of autoinducer-2 and several genes associated with virulence in E. coli O157:H7 (EHEC) following exposure to L. acidophilus. They suggested the role of L. acidophilus as an inhibitor of quorum sensing in EHEC O157 strains [24].

Conclusion
The results of the current study obviously indicate the capacity of broiler-derived B. coagulans and B. subtilis to signi cantly reduce the expression of genes involved in bacterial attachment, bio lm formation, quorum sensing, and antibiotic resistance in E. coli isolates. Standard strains of these sporeforming Bacillus spp. also showed the same effects. However, B. coagulans strains had a greater potential to reduce the expression of genes involved in quorum sensing while B. subtilis showed a comparatively greater ability to reduce the expression of genes involved in attachment, bio lm formation, and antibiotic resistance, suggesting the different mechanisms adopted by probiotics to combat E. coli infections. Moreover, the broiler-derived isolates showed a greater capacity to reduce the expression of these genes than the standard strains, proposing their large-scale usage in dietary supplementations. However, further studies are required to clarify the effects of these spore-forming bacteria on E. coli infections in vivo and their possible in uences on lactose tolerance, nutritional absorption, and reduced cholesterol levels.

E. coli isolation and culture conditions
A total of 300 fecal samples were collected from Imam Khomeini hospital, Tehran, Iran, from August 2018 to January 2019. Clinical samples were rst plated onto 5% sheep blood agar and MacConkey Agar (Biolife Laboratories, Milano, Italy). Samples were then con rmed as E. coli based on their morphology, Gram-staining, and routine biochemical tests, including MR/VP, utilization of citrate, presence of lysine decarboxylase and urease enzymes, and the production of ornithine. A single colony was obtained from each sample and maintained in TSB medium (Invitrogen, Paisley, Scotland) with 30% sterile glycerol at -80 °C for future experiments.
Primer design and polymerase chain reaction (PCR) assay To con rm the presence or absence of eae, u, luxS, and ctx-M genes, speci c primers were designed using the Primer-BLAST software (Table 1). Total genomic DNA of E. coli isolates was extracted using a DNA Extraction kit (Roche, Mannheim, Germany) according to the protocol recommended by the manufacturer. To investigate the presence of eae, u, luxS, and ctx-M genes, the PCR assay was performed in a volume of 25 μl in a DNA thermal cycler (Bio-Rad, USA) according to the following reaction conditions: initial denaturation at 94 °C for 5 min, 33 × denaturation at 94 °C for 30 s, annealing at 60 °C for 30 s, extension at 72 °C for 30 s, and nal extension at 72 °C for 5 min. No template control (NTC) was used as a negative control. Finally, amplicons were observed following gel electrophoresis, and sent for sequencing after puri cation. Isolation of spore-forming probiotics from gastrointestinal tracts of broilers A total of 10 broilers aged 6-12 months were chosen that did not take antibiotics or probiotics during their lifetime. After slaughter of birds in sterile conditions, intestinal contents were collected and diluted 1:1 (wt:vol) in buffered peptone-water (Oxoid) and resuspended by vigorous vortexing until obtaining an evenly distributed suspension. Then, aerobic spore-forming isolates were selected by heat (80 °C) and ethanol treatments. Ethanol treatment was performed by diluting the primary suspension (1:1) in ethanol ( nal concentration of 50% vol/vol) and incubation at room temperature for 1 h. Aliquots (0.1 ml) were cultured on nutrient agar plates and incubated at 37°C for 24-48 h. Colonies were picked randomly and puri ed by re-streaking on Luria-Bertani agar plates. The laboratory strain B. subtilis ATCC 6633 was used as a control throughout the experiments. Isolates were identi ed using the API 50 CHB strips according to the manufacturer's protocols (bioMérieux), and catalase and hemolysis tests were carried out to con rm the identi ed isolates. Finally, the identi ed B. subtilis and B. coagulans were selected for further analysis. Moreover, to con rm the production of spores, B. subtilis and B. coagulans isolates were grown on Difco sporulation medium (DSM) for 24-48 h. Then, cultures were puri ed as described by Henriques et al. [29] and stored in Difco heart-infusion broth (HIB) with 30% glycerol at -80°C for future use.

Molecular detection of spore-forming probiotics
Total genomic DNA of the isolated spore-forming probiotics was extracted using the pepGOLD Bacterial DNA kit (Roche, Mannheim, Germany) according to the manufacturer's protocol. For molecular identi cation of the isolated spore-forming probiotic bacteria, 16srRNA gene was investigated using the speci c primers designed by the Primer-BLAST software (Table 1). PCR assay was performed in a DNA thermal cycler (Bio-Rad, USA) in a volume of 25 μl according to the following reaction conditions: initial denaturation step at 94 °C for 5 min, 30 × denaturation at 94 °C for 30 s, annealing at 61 °C for 30 s, extension at 72 °C for 30 s, and nal extension at 72 °C for 7 min. No template control (NTC) was used as a negative control. After observing PCR products following gel electrophoresis, amplicons were nally sent for sequencing after puri cation.

Probiotic characterization of isolated bacterial strains
Resistance of vegetative B. subtilis and B. coagulans to bile salts and simulated gastric conditions was determined using overnight LB cultures of B. subtilis and B. coagulans isolates. To this end, he tolerance of Bacillus spores to bile salts and simulated gastric conditions was tested as previously explained by Duc et al. [30] with some modi cations. Brie y, about 10 8 -10 9 spores per ml were suspended in an isotonic buffer containing 0.76% H 2 PO 4 , 0.6% [NH 4 ] 2 SO 4 , 0.1% trisodium citrate, and 1.24% K 2 HPO 4 , also known as Bott and Wilson salts (pH 6.7) as well as 0.2% bile salts consisting of 50% sodium deoxycholate and 50% sodium cholate or in 0.85% NaCl (pH 2), containing 1 m pepsin (Sigma) and incubated at 37°C with shaking. Aliquots were collected immediately and following 30 min-1 h for assessing acid tolerance and 1-3 h for assessing bile tolerance. After direct plating of adequate dilutions in isotonic buffer onto LB plates CFU was determined following incubation at 37°C for 24 h. Control samples were suspended only in the isotonic buffer or 0.85% NaCl.
Similar assay was used to determine the resistance of vegetative cells to bile salts and simulated gastric conditions by resuspending overnight LB cultures of Bacillus isolates in fresh LB, LB supplemented with 0.2% bile salts, or acidi ed LB (pH 2) with HCl supplemented with 1 mg/ml pepsin.
Bacterial co-culture assay Co-culture of the two Bacillus spp. strains with E.coli isolates harboring all the studied genes (luxS, u, ctxM, and eae) were performed to determine changes in expression levels of the studied virulence genes in E. coli strains. Brie y, overnight cultures of B. subtilis and B. coagulans were centrifuged, the supernatant was collected, and after ltering, the two strains were inoculated individually in the tubes containing 5 ml of nutrient broth. Then, overnight cultures of E. coli isolates were also inoculated in each tube and once these cultures were set up, the tubes were incubated at 37 ºC under microaerophilic conditions. Each strain was also cultured alone as a control. To determine the effects of B. subtilis and B. coagulans on the expression levels of the studied virulence genes in E. coli, samples were withdrawn at the logarithmic growth phase (OD = 0.08-0.1). Experiments were carried out three times independently.
RT-qPCR analysis of transcript levels of the studied virulence genes in E. coli RT-PCR experiment was carried out to investigate the expression of the studied virulence genes in E. coli after the co-culture assay . Brie y, after the logarithmic growth phase, 1 ml of tube content was collected for RNA extraction using commercially available kits (QIAGEN RNeasy Mini kit). Samples were treated with Turbo DNase (Ambion, Grand Island, NY, USA) to eliminate remaining genomic DNA whose absence was con rmed using PCR and running samples on a 1% agarose gel. The quality of total RNA was assessed using the NanoDrop 1000 (Thermo Fisher Scienti c, Waltham, MA, USA). cDNA was synthetized using random hexamers (Applied Biosystems, CA, USA) and SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) according to the recommended protocols.
Finally, quantitative RT-PCR was performed in a Rotor-Gene thermal cycler (Corbett 6000, Australia) using the SYBR Green method (AccuPower Green Star qPCR Master Mix, Bioneer, Korea). Thermal cycling consisted of an initial cycle of 95 °C for 10 min and 40 cycles of 95, 58 , and 72 °C for 12 s, 25 s, and 30 s, respectively. 16s rRNA was used as the internal reference gene. After con rming the absence of primer dimers, qRT-PCR results were analyzed by the 2 -(ΔΔC(t)) method [31]. A P-value less than 0.05 was considered statistically signi cant.

Statistical analysis
In order to evaluate the signi cant probiotic effects of properties of B. subtilis and B. coagulans on the expression of u, eae, luxS, and ctxM genes in E. coli isolates, one sample t-test was performed using SPSS v. 24 at a signi ce level of p-value < 0.05.