Antimicrobial and Antibiofilm Activities of Probiotic Lactobacilli on Antibiotic-Resistant Proteus mirabilis

The emergence of biofilm-forming, multi-drug-resistant (MDR) Proteus mirabilis infections is a serious threat that necessitates non-antibiotic therapies. Antibiotic susceptibility and biofilm-forming activity of P. mirabilis isolates from urine samples were assessed by disc diffusion and crystal violet assays, respectively. Antimicrobial activities of probiotic Lactobacilli were evaluated by agar diffusion. Antibiofilm and anti-adherence activities were evaluated by crystal violet assays. While most P. mirabilis isolates were antibiotic-resistant to varying degrees, isolate P14 was MDR (resistant to ceftazidime, cefotaxime, amoxicillin-clavulanic acid, imipenem, ciprofloxacin, and amikacin) and formed strong biofilms. Cultures and cell-free supernatants of Lactobacillus casei and Lactobacillus reuteri exhibited antimicrobial and antibiofilm activities. The 1/16 concentration of untreated supernatants of L. casei and L. reuteri significantly reduced mature biofilm formation and adherence of P14 by 60% and 72%, respectively (for L. casei), and by 73% each (for L. reuteri). The 1/8 concentration of pH-adjusted supernatants of L. casei and L. reuteri significantly reduced mature biofilm formation and adherence of P14 by 39% and 75%, respectively (for L. casei), and by 73% each (for L. reuteri). Scanning electron microscopy (SEM) confirmed eradication of P14’s biofilm by L. casei. L. casei and L. reuteri could be utilized to combat Proteus-associated urinary tract infections.


Introduction
Proteus mirabilis is one of the most common pathogens associated with urinary tract infections (UTI) [1,2]. In addition, it can cause surgical wound infections, biliary tract infections, wound infections, and nosocomial infections [3]. It is part of the normal gut flora and can cause opportunistic infections in immunocompromised and elderly patients [3]. The emergence of antibiotic resistance in clinical isolates of P. mirabilis is a major healthcare issue [4,5]. P. mirabilis poses a major challenge in infection management as it produces AmpC β-lactamases and extended spectrum β-lactamases [6]. Furthermore, it can develop complex biofilms with accumulated layers of polysaccharides in which sessile cells are embedded, which adds to the severity of the infection [7]. The severity, chronicity, and dissemination of Proteus infections have been mainly attributed to its ability to form biofilms [7].
Probiotics are living microorganisms that belong to Lactobacillus and Bifidobacterium genera [8,9]. The intestinal probiotic Lactobacillus strains have been recognized for their antimicrobial activities

Detection of Biofilm Formation by P. mirabilis Isolates
The mature biofilm of P. mirabilis isolates was formed in flat-bottomed 96-well microtiter plates. The overnight culture of each isolate was adjusted to 0.5 McFarland (1.5 × 10 8 CFU/mL) using Muller Hinton broth. Then, 200 µL of the diluted cultures was distributed in each well and incubated at 37 • C for 48 h. The planktonic cells were removed and the attached cells were gently washed twice with sterile physiological saline. Then, 200 µL of methanol (99%) was added to each well and retained for 15 min to fix the sessile cells. The methanol was discarded, and the plate was left until complete dryness. To stain the adherent cells, 200 µL of 2% w/v crystal violet solution was added to each well and was left for 20 min. The wells were washed gently and left to dry. Glacial acetic acid (200 µL; 33% w/v) was added to release the bound dye and the absorbance was measured at OD 540 nm using ELISA microplate reader (MR-960, Perlong Medical Equipment Co. Ltd., Nanjing, China). Depending on the optical density (OD) generated by the bacterial biofilms, the tested P. mirabilis isolates were classified into non-biofilm producers, weak biofilm producers, moderate biofilm producers, and strong biofilm producers [33]. The cut-off OD of the negative control (ODc) was determined by adding the mean of the negative control to three standard deviations of it. The lack of biofilm formation is indicated by ODs ≤ ODc for the tested isolates. Weak biofilm production is indicated by ODc < OD ≤ (2 ODc). Moderate biofilm formation is indicated by (2 ODc) < O.D. ≤ (4 ODc). Strong biofilm formation is indicated by (4 ODc) < O.D. Tests were conducted in triplicates.

Preparation of Cell-Free Supernatant from Lactobacillus Strains
In order to prepare the Lactobacillus culture, the MRS medium was inoculated with Lactobacillus strains with inoculum size 1% v/v and was incubated anaerobically at 37 • C for 48 h. The grown culture was centrifuged at 6000 rpm for 15 min to separate all cells and the supernatant was filtered through a 0.22 µm membrane filter. The cell-free supernatant was labeled as untreated supernatant (U) and stored at −20 • C. The Lactobacillus supernatant was highly acidic due to lactic acid production. To adjust the pH of the supernatant to pH 6.5-7.0, 1N NaOH was used and this fraction of the supernatant was labeled as treated supernatant (T) and stored at −20 • C [34].

Antimicrobial Activity of Lactobacillus Supernatants
The antimicrobial activity of Lactobacillus supernatants (treated or untreated) was assessed against the P. mirabilis isolates P2, P4, P14, P15, P23, P24 and P25 using microtiter plate assays. The antibacterial activity was assayed by the agar diffusion method [35]. The P. mirabilis isolates were incubated in Muller Hinton broth at 37 • C for 24 h. The culture inoculum was adjusted to 0.5 McFarland (1.5 × 10 8 CFU/mL) and was used to inoculate melted Muller Hinton agar at 50 • C. After medium solidification, wells were cut in agar using a cork borer. Wells were filled with 100 µL of the whole cell culture of Lactobacillus or cell-free supernatants (treated or untreated), and the plates were incubated aerobically at 37 • C for 24 h. Inhibition zones were measured to indicate the antimicrobial activity of the corresponding Lactobacillus.

Effect of Lactobacillus Supernatant on Mature Biofilms
The mature biofilms of P. mirabilis isolates P2, P4, P14, P15, P23, P24, and P25 were formed in flat-bottomed 96-well microtiter plates. The overnight culture of each isolate was adjusted to 0.5 McFarland using Muller Hinton broth. Then, 100 µL of the diluted Proteus cultures was added to each well and incubated at 37 • C for 48 h. The planktonic cells were removed and the attached cells were gently washed twice with sterile physiological saline. Untreated or treated supernatants of Lactobacillus were then added to the wells and the plates were incubated at 37 • C for 24 h. The biofilm of each Proteus isolate without Lactobacillus supernatants was used as a positive control. The influence of different concentrations of Lactobacillus supernatants (treated or untreated) on the mature biofilm was detected using crystal violet microtiter plate assays. The treated and the mature Proteus biofilms were washed, fixed using methanol, and stained with crystal violet as described before [36].

Anti-Adherence Effect of Lactobacillus Supernatants
In order to study the effect of Lactobacillus on biofilm formation, 100 µL of treated or untreated supernatants of Lactobacillus was mixed at low concentrations (1/8, 1/16 and 1/32) with 100 µL of the P. mirabilis isolate P14 (with inoculum size 1.5 × 10 8 CFU/mL). The microtiter plate was incubated for 48 h at 37 • C and the biofilm formation was detected using the crystal violet microtiter plate method [17]. Wells containing the Proteus isolate P14 in contact with different concentrations of Lactobacillus supernatant (At) were compared to wells containing Proteus culture without the Lactobacillus supernatant (control, Ac). The percent reduction in biofilm formation was calculated as follows = ((Ac − At)/Ac) ×100.

Scanning Electron Microscopy (SEM)
The overnight culture of P14 was diluted to 0.5 McFarland using tryptic soy broth (TSB). Similarly, cultures of the tested Lactobacillus strains were diluted in MRS and co-cultured with equal volumes of P14 (1:1) in sterile six well plates (Greiner Bio-One, Kremsmünster, Austria) at 37 • C for 24 h. The control wells containing P14 only with MRS and TSB media were also prepared. A clean sterile cover slide was added to each well. The slides were removed and washed gently with phosphate buffered saline (PBS) to remove the planktonic cells. The biofilm was fixed and prepared for examination by SEM (JSM-7600F, JEOL USA, INC., Peabody, MA, USA) as previously described [37].

Statistical Analysis
Statistical analysis was performed using one-way ANOVA in order to compare the effect of Lactobacillus, cell-free, and treated supernatants on the mature biofilm and on the adhesion of P. mirabilis isolates. A p value of <0.05 indicated statistical significance.

Antimicrobial Susceptibility and Biofilm Formation of P. mirabilis isolates
Activity of the tested antimicrobial agents against P. mirabilis isolates was evaluated according to CLSI standards [32]. As shown in Table 1, 86% of the tested P. mirabilis isolates were resistant to amoxicillin-clavulanic, 57% were resistant to cefotaxime, 57% were resistant to ceftazidime, 71% were resistant to ciprofloxacin, 57% were resistant to amikacin, 43% were resistant to imipenem, and 0% were resistant to cefoxitin. Furthermore, isolate P14 was resistant to cefotaxime, ceftazidime, amoxicillin-clavulanic acid, ciprofloxacin, and amikacin and intermediately resistant to imipenem (Table 1). In contrast, isolate P2 was sensitive to all the assessed antimicrobials (Table 1).
P. mirabilis isolates P2, P4, P14, P24 and P25 showed strong biofilm formation. Isolates P15 and P23 showed moderate biofilm formation. Isolate P14 showed the strongest biofilm formation and was resistant to the tested antimicrobials. Isolate P2 showed a strong biofilm formation, but was sensitive to all assessed antimicrobials.  3.2. Antimicrobial Activities of L. casei and L. reuteri against P. mirabilis Isolates The untreated supernatants of L. casei and L. reuteri had inhibitory effects on the tested P. mirabilis isolates ( Table 2). The treated (pH-adjusted) supernatants of L. reuteri were effective against almost all tested P. mirabilis isolates with inhibition zone diameters ranging from 12 to 16 mm. The treated (pH-adjusted) supernatants of L. casei was effective against four isolates (P2, P14, P15 and P24) with inhibition zone diameters ranging from 13 to 16 mm. Notably, the treated supernatants of L. casei and L. reuteri were effective against the multidrug resistant (MDR) Proteus isolate P14 with inhibition zone diameters ranging from 14 to 15 mm (Table 2).

Effect of Lactobacillus Supernatants on Mature Biofilm Formation of P. mirabilis Isolates
The effect of treated and untreated supernatants of L. casei and L. reuteri on mature biofilm formation of P. mirabilis isolates P2, P4, P14, P15, P23, P24, and P25 was studied ( Figure 1).

Anti-Adherence Effect of Lactobacillus Supernatants
The anti-adherence effect of diluted Lactobacillus supernatants (treated and untreated) on the P14 isolate was assessed. As in Figure 3, the untreated supernatants of L. casei and L. reuteri significantly reduced the adhesion of the P14 isolate (1/16 concentration) by 72% and 73%, respectively (p < 0.01) ( Figure 3A, C). The anti-adherence effect of the untreated supernatants on the P14 isolate decreased  The 1/16 and 1/32 concentrations of the untreated supernatants of L. reuteri caused a significant reduction in biofilm formation of the P14 isolate by 73% and 32%, respectively (p < 0.05) ( Figure 2C). The 1/8 concentration of the treated supernatants of L. reuteri significantly reduced biofilm formation of the P14 isolate by 73% (p < 0.05) ( Figure 2D).

Anti-Adherence Effect of Lactobacillus Supernatants
The anti-adherence effect of diluted Lactobacillus supernatants (treated and untreated) on the P14 isolate was assessed. As in Figure 3, the untreated supernatants of L. casei and L. reuteri significantly reduced the adhesion of the P14 isolate (1/16 concentration) by 72% and 73%, respectively (p < 0.01) ( Figure 3A,C). The anti-adherence effect of the untreated supernatants on the P14 isolate decreased by further diluting the supernatant.

Scanning Electron Micrographs
As in Figure 4A, there is a dense mass of biofilm-forming P. mirabilis P14 (as a positive control). Co-culture of L. casei with P. mirabilis P14 caused the complete elimination of biofilm formation by P14 ( Figure 4B). Co-culture of L. reuteri with P. mirabilis P14 showed scattered cells and a loose biofilm architecture but no dense aggregates ( Figure 4C). As in Figure 3, the treated supernatants of L. casei and L. reuteri significantly reduced biofilm formation of the P14 isolate (1/8 concentration) by 75% and 73%, respectively (p < 0.01) ( Figure 3B,D).

Scanning Electron Micrographs
As in Figure 4A, there is a dense mass of biofilm-forming P. mirabilis P14 (as a positive control). Co-culture of L. casei with P. mirabilis P14 caused the complete elimination of biofilm formation by P14 ( Figure 4B). Co-culture of L. reuteri with P. mirabilis P14 showed scattered cells and a loose biofilm architecture but no dense aggregates ( Figure 4C). 8 Figure 3. Adherence of the Proteus isolate P14 in the presence of supernatants of L. casei and L. reuteri (* p < 0.05). (A) Adherence of the Proteus isolate P14 in the presence of untreated supernatants of L. casei (* p < 0.05). (B) Adherence of the Proteus isolate P14 in the presence of treated supernatants of L. casei (* p < 0.05). (C) Adherence of the Proteus isolate P14 in the presence of untreated supernatants of L. reuteri (* p < 0.05). (D) Adherence of the Proteus isolate P14 in the presence of treated supernatants of L. reuteri (* p < 0.05).

Scanning Electron Micrographs
As in Figure 4A, there is a dense mass of biofilm-forming P. mirabilis P14 (as a positive control). Co-culture of L. casei with P. mirabilis P14 caused the complete elimination of biofilm formation by P14 ( Figure 4B). Co-culture of L. reuteri with P. mirabilis P14 showed scattered cells and a loose biofilm architecture but no dense aggregates ( Figure 4C).

Discussion
Bacterial resistance is one of the major public health problems. It includes the emergence of MDR pathogens, such as resistant isolates of Proteus mirabilis, which have been associated with urinary tract infections and nosocomial infections worldwide [4,38]. P. mirabilis isolates have been reported to be resistant to penicillins [39]. They also show a high incidence of resistance to cephalosporins, carbapenems, and quinolones [4,40]. In the current study, a high percentage of the isolates was resistant or intermediately resistant to amoxicillin-clavulanic acid and cefotaxime. There was also significant resistance to ceftazidime, ciprofloxacin, and amikacin (Table 1). It is noteworthy that the isolate P14 was resistant to three different groups of antibiotics with various mechanisms of action (Table 1). Cefotaxime-resistant P. mirabilis has been previously described [41]. Kwiecińska-Piróg et al. reported that P. mirabilis showed resistance against ceftazidime and ciprofloxacin [42]. Amikacin resistance has been shown to be associated with extended spectrum beta-lactamase (ESBL(-producing isolates of P. mirabilis, which were resistant to amikacin (85.1%) [43].
Urinary tract infection with P. mirabilis is associated with biofilm formation and accumulation of the polysaccharide matrix [44]. This is followed by urease production, increase in the pH, attraction of calcium and magnesium ions, and development of crystals [44]. The deposition of crystals within the biofilm can cause catheter blockage and urinary retention. This is because P. mirabilis possesses various virulence factors (lipopolysaccharide, quorum sensing autoinducers, pili, adhesin, and other proteins) that enhance adhesion and crystalline biofilm formation on the abiotic surfaces of urinary catheters [45]. It has been reported that 48% of the isolated proteus species were biofilm producers [42]. The Proteus-biofilm assembly is dangerous, as it interferes with microbial penetration, increases antimicrobial resistance, and renders therapeutic treatments ineffective, which encourages the development and chronicity of infections [46].
Moreover, pathogenicity of P. mirabilis is enhanced by the complex architecture of its biofilm, which is characterized by a high ability for adapting to different environmental conditions, biocides, and antimicrobials [47,48]. Isolates in this study were able to form strong or moderate biofilms. This was true even for the antibiotic-sensitive isolate P2, which was observed to have a strong biofilm-forming ability (Table 1). Typically, the biofilm-producing isolates are more resistant to antibiotics than the non-biofilm producing isolates [49]. The reason is that the development of complex biofilm structures can convey protection to the internal cells from antimicrobials [50]. It can also support the persistence of P. mirabilis in the host cells [50].
Given the above, alternative treatments are required for the management of P. mirabilis infections and the disruption of its biofilm architecture. Probiotics as Lactobacilli, have been used for the treatment of burn infections and have been shown to interfere with activity of Pseudomonas aeruginosa [51,52]. Lactobacilli have also been used for the management of recurrent urinary tract infections caused by E. coli [53]. The antimicrobial activity of L. casei and L. reuteri against various enteropathogenic infections have been previously reported [10]. Their antimicrobial activities against uropathogens have also been reported [20]. L. reuteri has been shown to be effective against Escherichia coli and Listeria monocytogenes [54]. Thus, it was important to examine the antimicrobial and antibiofilm activities of Lactobacillus on P. mirabilis clinical isolates. We examined the cultures and the cell-free supernatants of L. casei, and L. reuteri against P. mirabilis isolates.
In our study, the treated supernatants of L. casei and L. reuteri retained antimicrobial activities against the sensitive P2 isolate and the other resistant Proteus mirabilis isolates (Table 2). Notably, treated supernatants of L. casei and L. reuteri retained significant antibiofilm and anti-adherence activities against P14 at 1/8 concentration (Figures 2 and 3). In addition, L. casei and L. reuteri had significant inhibitory effects on the typical biofilm formed by the MDR isolate P14, as detected by SEM (Figure 4). In another study, the untreated supernatants of fecal Lactobacilli caused an 85 to 95% reduction in biofilm formation of Vibrio cholerae isolates [27]. Similarly, the pH-adjusted supernatants of fecal Lactobacilli significantly reduced biofilm formation of Vibrio cholerae by 50-75% [27]. Additionally, Koohestani and colleagues showed that the untreated supernatants of L. casei significantly reduced biofilm formation of S. aureus [55]. Probiotic lactobacilli have been shown to negatively impact growth and biofilm formation of Streptococcus mutans [34], pathogens in the oral cavity [56], and Salmonella enterica serovar Typhimurium [28]. Moreover, probiotic lactobacilli inhibited cancer cells of the human colonic carcinoma cell line HT-29 [23]. Lastly, Lactobacillus reuteri DPC16 supernatants have been shown to exhibit activity against Escherichia coli, S. aureus, Salmonella derby, and Listeria monocytogenes. This is consistent with results in our study in which the untreated supernatants of L. casei reduced biofilm formation of P. mirabilis isolates by 45-67%.
It is important to mention that the use of probiotic Lactobacillus cell cultures and supernatants was proposed as an alternative to traditional antibiotic therapy against P. mirabilis. In other words, it was proposed as a way to avoid the adverse effects of antibiotic therapy and to circumvent the impact of antibiotic resistance developed by the microbe. Indeed, we saw a significant inhibitory effect of Lactobacillus cell cultures and supernatants on P. mirabilis biofilm formation and adherence. Future studies could examine the effect of Lactobacillus supernatants on the susceptibility of P. mirabilis to the tested antimicrobials. In this case, the focus would be on the possible synergistic effects of combining probiotic Lactobacillus supernatants with anti-Proteus agents.

Conclusions
The pathogenicity and virulence of Proteus infections includes a biofilm-forming ability that enables serious urinary tract infection. Isolates in this study showed multidrug resistance (MDR) against more than one antimicrobial agent, highlighting the importance of the development of alternative and adjuvant treatments for the efficient management of P. mirabilis infections. Using cell cultures, cell-free supernatants, and pH-adjusted supernatants, we have shown that L. casei DSM 20011 and L. reuteri DSM 20016 exhibit antimicrobial, anti-adherence, and antibiofilm activities against MDR P. mirabilis. Thus, L. casei and L. reuteri could be utilized to combat Proteus-associated urinary tract infections. Funding: This research received no external funding.

Conflicts of Interest:
The authors confirm that there are no conflict of interest.