Antagonistic Activity of Bacteriocin-producing Lactobacillus Against Candida spp

In order to screen antifungal activity of lactobacilli that produce bacteriocin against the yeast Candida, ninety-two food and clinical samples were collected. Also, several commercial brands of Lactobacillus probiotics and ready-made isolates of Lactobacillus were used. The isolated lactobacilli were subjected to microscopic, macroscopic, and biochemical tests. Moreover, molecular identification was performed for the best producer isolate. For Candida spp. isolation, seventytwo samples were collected from different clinical sources; in addition, nine of ready-made isolates were also used. All the isolated Candida spp. were subjected to microscopic and macroscopic examinations which were confirmed by VITEK® 2 YST card automated system. Detection of bacteriocin production from Lactobacillus was investigated by primary and secondary screening techniques and the results showed that the agar wells diffusion method was the best. The most efficient isolate to produce bacteriocin was L. plantarum WZD3, in which antifungal activity of bacteriocin was at the level of 80 and 40 AU/ml against the most sensitive Candida isolates; C. albicans CA and C. albicans CB, respectively. After partial purification of bacteriocin by n-butanol extraction method, bacteriocin activity was increased to 320 AU/ml against both yeast isolates. In conclusion, L. plantarum WZD3 or its bacteriocin could be a good candidate as antifungal agent to treat Candida infections and more studies are required to evaluate its acivity against other types of medical fungi.


Introduction
One of the most difficult diseases to control in human beings are fungal diseases ]1], with mortality rate of 45% ]2]. The mortality rates are still high in spite of the availability of many effective antifungal medicines ] 3]. Fungal species that belong to Cryptococcus, Candida, Aspergillus, Histoplasma, and Pneumocystis are responsible for more than 90% of all reported fungal-related deaths ]4]. Candida spp. is one of the most common causes of invasive candidiasis and the most important one is C. albicans [5]. Development of resistant fungi and therapy failures following longterm use of antifungal drugs have been increased in immunocompromised patients ]6]. Therefore, finding an alternative of some useful compounds, such as bacteriocin, for the better control and treatment of microbial infections has been suggested ]7, 8,9]. The fungicidal effect of plantaricins (Pln E/F and J/K) against C. albicans was investigated by Sharma and Srivastava [10[. Also, the investigation of Santos and co-authors [11[ showed that bacteriocin produced from Lactobacillus plantarum has an anti-inflammatory effect against C. albicans. While, Graham et al.[12[ demonstrated that bacteriocin EntV from Enterococcus faecalis has potential as a novel antifungal agent against C. albicans.
Bacteriocins of lactic acid bacteria (LAB) are simple to produce, constant at low pH, non-toxic to humans, and susceptible to proteases [13]. In general, the antimicrobial activity of LAB species is well known [14]. Various investigators demonstrated anticandidal effects of different Lactobacillus species [15][16][17][18][19] and other clinical isolates of Lactobacillus. The genus Lactobacillus that comprises 261 species [20[ is part of the normal human microbiota that colonizes the mouth, gastrointestinal (GI) tract, and female genitourinary tract [21]. Lactobacillus comprises a high number of species which are generally recognized as safe ]22]. Members of this genus contribute to the health status of the gastrointestinal tract and vagina. They also produce a wide variety of bacteriocins that demonstrate a wider range of activities towards bacteria or fungi ]23[. This study aims to detect bacteriocin production by Lactobacillus spp. isolates and evaluate its antimicrobial effects against Candida spp., with the final aim of applying it therapeutically against this pathogenic yeast.

Materials and Methods
Lactobacillus Ready-made Isolates Ten isolates of Lactobacillus spp. were obtained from the Department of Biotechnology, College of swabs were immersed in 9 ml of de Man, Rogosa and Sharpe (MRS) broth medium tubes and transferred to the laboratory in ice box. For the isolation from food samples, each tube containing 9 ml of MRS broth was inoculated with 1 ml of liquid sample or 1 g of solid sample and mixed gently to get a uniform sample, separately. All the inoculated tubes were incubated anaerobically at 37°C for 48 hours. After that, serial decimal dilutions were made for each culture by using physiological saline solution and then they were streaked on MRS agar medium supplemented with 1% calcium carbonate. The plates were incubated anaerobically at 37°C for 48 hours. After incubation, large white colonies with intensity were selected, transferred, and purified by the streaking method on MRS agar and SL agar plates. The plates were incubated under the same incubation conditions above before being subjected to identification tests [24,25].

Identification of Lactobacillus Isolates
The initial identification of the achieved pure cultures was based on Gram staining, the ability to grow on a selective MRS and SL agar [26], the ability to grow on nutrient agar, and biochemical tests, including catalase, oxidase, and gelatinase tests [27]. Molecular identification was performed by sequencing of 16S ribosomal RNA (16S-rRNA) for Lactobacillus isolates. Genomic DNA was extracted using ABIOpure™ Total DNA purification kit (ABIOpure, USA). The extracted DNA was estimated using Quantus (Promega, USA) with the Quantiflour dye. The primers (27F; AGAGTTTGATCCTGGCTCAG and 1492R; TACGGTTACCTTGTTACGACTT) [28] were supplied by Macrogen Company (Korea) in a lyophilized form. PCR products were sent for Sanger sequencing, using ABI3730XL automated DNA sequencer, by Macrogen Corporation (Korea) and the results were analyzed by using genious software.

Isolation and Identification of Candida
In this study, 72 samples were collected from different sources (skin, mouth and vagina) of patients (children and women) from Al-Alwiya Hospital for Childbirth and the Central Children's Teaching Hospital in Baghdad. The taken swabs were cultured on Sabouraud Dextrose agar (SDA), then incubated aerobically at 37°C for 24-48 hours [29[. All isolates were identified macroscopically and microscopically ]30[, and the identification was confirmed by VITEK® 2 YST card automated system (BioMérieux ,France).

Detection of Bacteriocin Production from Lactobacillus Primary Screening A. Agar-Plug Diffusion Method
A volume of 0.1 ml of 1.5×10 8 CFU/ml (according to McFarland tube No. 0.5) of fresh Lactobacillus broth cultures was spread on MRS agar plates and incubated anaerobically at 37°C for 24-48 hours. After incubation, sterile cork borer (6 mm) was used to cut plugs of cultured MRS agar plates. A volume of 0.1 ml of 10 6 cell/ml [31] of the activated yeast suspension was spread on SDA agar plates. After that, the plugs of cultured MRS agar were placed on the cultured SDA agar plates and incubated aerobically at 37°C for 18-24 hours. The formed inhibition zone around each plug was measured and recorded. Uninoculated MRS agar was used as a negative control [32].

B. Agar-Wells Diffusion Method
Lactobacilli (1.5×10 8 CFU/ml) were inoculated in MRS broth under anaerobic conditions at 37 °C for 24-48 hours. The cultures were centrifuged at 6000 rpm for 15 minutes. A volume of 0.1 ml of 10 6 cell/ml of the activated yeast suspension was transferred and spread on SDA agar. Wells cut into the pour plates by using a 6 mm sterile cork borer were filled with 100 μl of the cell-free supernatant (CFS). The plates were kept at room temperature for 2 hours and then incubated at 37°C for 18-24 hours. Finally, the inhibition zones formed around the wells were measured in mm and compared with that of control which contained MRS broth only ]16[.

Secondary Screening By Filter Paper Disc Method and Agar-Wells Diffusion Method
Lactobacilli (1.5×10 8 CFU/ml) were inoculated in MRS broth and incubated anaerobically at 37°C for 24-48 hours. Then, the CFS was obtained by centrifuging at 6000 rpm for 15 minutes. The CFS was neutralized to pH 6.5 with 1N NaOH and sterilized by filtration through 0.45 μm membranes. A volume of 0.1 ml of 10 6 cell/ml of the activated yeast suspension was transferred and spread on SDA agar. In the filter paper disc method, sterile filter paper discs, measuring 5 mm diameter and saturated with 100 μl of CFS, were placed on the seeded SDA agar plates ]33[. While in the agar-wells diffusion method, wells cut into the pour plates with 6 mm sterile cork borer were filled with 100 μl of the CFS ]16[. All the plates were left in laboratory temperature for 2 hours and then incubated aerobically at 37°C for 18-24 hours. The inhibition zones formed around the paper discs or wells were measured in mm and recorded.
According to the results of secondary screening, the most efficient isolate in the bacteriocin production of Lactobacillus spp. and the most sensitive isolates of Candida spp. were selected and used in the subsequent experiments.

Bacteriocin Activity Assay
To quantify the bacteriocin activity, CFS or crude bacteriocin was serially diluted two-fold with physiological saline solution. These dilutions were used to examine the antifungal activity of bacteriocin against the indicator yeast by agar well diffusion assay (as previously described). Bacteriocin activity was expressed as AU/ml and defined as the reciprocal of the highest dilution showing a distinct inhibition zone of the indicator yeast. AU was calculated as: (1000 / 100) × D, where 1000: constant, 100: volume of supernatant in a well (μl), and D : the dilution factor ]34[. Partial Purification of Bacteriocin MRS broth was inoculated with the bacterial isolate (D) and incubated at 37°C for 48 hours. Cells were harvested by centrifugation at 6000 rpm for 15 minutes. CFS was heated at 80°C for 10 minutes, then cooled and centrifuged at 6000 rpm for 15 minutes ]35[. The supernatant was mixed thoroughly with n-butanol at a ratio of 1:1.The mixture was centrifuged at 4000 rpm for 10 minutes to achieve phase separation. The organic phase was evaporated at 65°C by rotary evaporator, then the sediment was re-suspended in 1.0 mM sodium phosphate buffer (pH 6) and referred to as partially purified bacteriocin (PPB) ]36 [ . The antimicrobial activity of bacteriocin was determined, as previously described, by the agar well diffusion method.

Results and Discussion Identification of Bacterial Isolates
The results showed that 22 of the bacterial isolates appeared as having white colored, soft, large or small, convex, creamy, smooth, and circular colonies with entire margins and surrounded by clear zones after being cultured on MRS agar containing calcium carbonate, as a result of dissolving it by the acid produced by the bacteria. Lactobacillus showed white colored, convex, large, slightly mucoid, smooth, and round forms with entire margins on SL agar, whereas no colonies on nutrient agar were developed. Under microscope, they were purple rods, singly, in pairs, or in short chains, and non-spore forming. They were catalase, oxidase, and gelatinase negative ]26, 37 [ . Moreover, the most efficient isolates (D and L1) in the bacteriocin production were selected for molecular identification to confirm the diagnosis. According to16S rRNA analysis, the two isolates were identified as L. plantarum (Table  -1). *Identity values were determined using the basic local alignment search tool (BLAST) of the GenBank.

Identification of Yeast Isolates
Under microscopic examination, the shape of yeast cells was spherical to oval, with the presence of budding, and they were much larger than bacterial cells. On SDA, yeasts appeared as raised, glossy, smooth, glabrous yeast-like, with distinctly smelling, white to cream colored, and circular to oval colonies ]38, 39]. Moreover, the identification of these isolates was confirmed by Vitek2 YST card automated system. The results showed that 24 isolates were identified as Candida spp., 12 isolates as C. albicans (10 from mouth and 2 from vagina), 6 isolates as C. famata (one from vagina and 5 from mouth), C. dubliniensis and C. parapsilosis from vagina, and C. guilliermondii, C. tropicalis, C. ciferrii and C. catenula from mouth.

Detection of Bacteriocin Production from Lactobacillus
All isolates which were confirmed through the identification process, as well as the commercial brands of Lactobacillus probiotics, were subjected to primary and secondary screening techniques in order to select the most efficient isolate as bacteriocin producer.

Primary Screening
The results showed that the agar-plug diffusion method did not demonstrate any antifungal activity, while the agar well diffusion method demonstrated this activity. Among 38 isolates of Lactobacillus, only 20 isolates recorded inhibition zones against 17 isolates of Candida (Table-2). While 18 isolates of Lactobacillus did not show any antifungal activity, since no inhibition zones were recoded. Thus, the selection of isolates for the secondary screening was relied on the results of this method. The ability of Lactobacillus isolates to inhibit yeast growth is clearly strain and culture condition-dependent, with antimicrobial substances such as bacteriocins and mechanisms being involved in the inhibition process in vitro, as reported by ] 40,41].

Secondary Screening
In the secondary screening, only three isolates of L. plantarum (D and L1) and Lactobacillus sp.(L2) had antifungal activity against only 11 yeast isolates; 7 isolates of C. albicans (C1,C3, C4,C8, C40, CA and CB ), C. famata (C15), C. guilliermondii (C17), C. parapsilosis (C30), and C. dubliniensis (C33) . The isolate D was the most active one. Also, the agar wells diffusion method was more efficient than the other method (Tables-3 and 4). Thus, depending on the results of the secondary screening, the most efficient Lactobacillus isolate (D) was selected as a bacteriocin producer and called L. plantarum WZD3. Also, the most sensitive Candida isolates, namely C. albicans CA and C. albicans CB, were selected as indicator yeasts due to heir higher sensitivity toward Lactobacillus isolates, where the diameters of the inhibition zones of L. plantarum WZD3 against C. albicans CA and C. albicans CB were 20 and 30 mm, respectively. The activity of bacteriocin against C. albicans CA was at the value of 80 AU/ml, whereas that against C. albicans CB was 40 AU/ml (Figure-1).   C4  C8  C15  C16  C17  C21  C30  C33  C39  C40  C51  C53  C27s  CA  CB   B  12  ----------------D  15 18 14 15 20  14.  5  14  -14  ------2  2 24    C3  C4 C8 C15 C17  C30  C33  C40  CA  CB  D  10  18  16  15  15  18  16  15  10  20  30  L1  10  --------15  29  L2 12 In antimicrobial activity researches, the agar plug method is considered as a practical and suitable technique and the filter paper disc method is regarded as appropriate and easy to use. Nevertheless, in the present investigation, it was observed that agar wells diffusion method was more appropriate and more effective than the filter paper disc method and the agar plug method. It is mainly based on forming a zone of inhibition around the well where the indicator has not grown, and the size of the zone is based on how much the antimicrobial compound is effective. Therefore, this method was used in this study to define the antimicrobial activity because of its sensitivity to determine the antifungal activity in next steps, as reported previously ] 42,43].

Partial Purification of Bacteriocin
Crude bacteriocin extract was heated before starting purification to denaturize proteases and any heat-labil proteins, as investigated by Powell et al. [35]. The heating step did not affect bacteriocin activity, as reported earlier [44]. Bacteriocin partially purified by extraction with n-butanol was named plantaricin WZD3, according to the producer isolate, L. plantarum WZD3. By using this method for purification, the antifungal activity of bacteriocin reached to 320 AU/ml against both of C. albicans CA and C. albicans CB, compared with the antifungal activity results of crude bacteriocin extract, which was only 80 AU/ml against C. albicans CA and 40 AU/ml against C. albicans CB, respectively (Figure -2). Butanol extraction exhibited complete recovery of bacteriocin activity, suggesting that at least part of the bacteriocin molecule has a hydrophobic character and shares this property with other bacteriocins ] 45,46]. Researchers such as ten Brink et al. [47] showed that butanol extraction yields almost pure (≥ 90%) bacteriocin, and that this simple procedure could be adopted for partial purification of bacteriocins. Extraction of bacteriocins using n-butanol in a 1:1 ratio was reported for several plantaricins [36,44,48].

Conclusion
In the present study, Lactobacillus isolates exhibited variations in their antagonistic activity and bacteriocin production against Candida spp. It is clear that the capability of bacterial isolates to inhibit the growth of yeast in vitro is related to the isolation and culturing conditions, as well as, the used method for detection of antifungal activity and also mechanisms which involved in the inhibition process as mentioned by Siro [49]. L. plantarum WZD3 and its bacteriocin could be used as antifungal agents against Candida infections instead of antibiotics, but more investigations are actually needed to evaluate the antifungal activities of lactobacilli and their bacteriocins.