Probiotics & Antimicro. Prot. DOI 10.1007/s12602-017-9312-8
In Vitro Evaluation of the Probiotic and Safety Properties of Bacteriocinogenic and Non-Bacteriocinogenic Lactic Acid Bacteria from the Intestines of Nile Tilapia and Common Carp for Their Use as Probiotics in Aquaculture Pierre Marie Kaktcham 1,2 & Jules-Bocamdé Temgoua 1 & François Ngoufack Zambou 1 & Gloria Diaz-Ruiz 3 & Carmen Wacher 3 & María de Lourdes Pérez-Chabela 2
# Springer Science+Business Media, LLC 2017
Abstract In this study, seven bacteriocinogenic and nonbacteriocinogenic LAB strains previously isolated from the intestines of Nile tilapia and common carp and that showed potent antibacterial activity against host-derived and non-hostderived fish pathogens were assayed for their probiotic and safety properties so as to select promising candidates for in vivo application as probiotic in aquaculture. All the strains were investigated for acid and bile tolerances, transit tolerance in simulated gastrointestinal conditions, for cell surface
* Pierre Marie Kaktcham
[email protected];
[email protected];
[email protected] Jules-Bocamdé Temgoua
[email protected] François Ngoufack Zambou
[email protected];
[email protected] Gloria Diaz-Ruiz
[email protected];
[email protected] Carmen Wacher
[email protected];
[email protected]
characteristics including hydrophobicity, co-aggregation and auto-aggregation, and for bile salt hydrolase activity. Moreover, haemolytic, gelatinase and biogenic amineproducing abilities were investigated for safety assessment. The strains were found to be tolerant at low pH (two strains at pH 2.0 and all the strains at pH 3.0). All of them could also survive in the presence of bile salts (0.3% oxgall) and in simulated gastric and intestinal juices conditions. Besides, three of them were found to harbour the gtf gene involved in pH and bile salt survival. The strains also showed remarkable cell surface characteristics, and 57.14% exhibited the ability to deconjugate bile salts. When assayed for their safety properties, the strains prove to be free from haemolytic activity, gelatinase activity and they could neither produce biogenic amines nor harbour the hdc gene. They did not also show antibiotic resistance, thus confirming to be safe for application as probiotics. Among them, Lactobacillus brevis 1BT and Lactobacillus plantarum 1KMT exhibited the best probiotic potentials, making them the most promising candidates. Keywords Aquaculture . Nile tilapia and common carp . Bacteriocinogenic and non-bacteriocinogenic LAB . Fish pathogens . Probiotics . Safety
María de Lourdes Pérez-Chabela
[email protected] 1
Introduction Laboratory of Biochemistry, Food Science and Nutrition (LABPMAN), Department of Biochemistry, Faculty of Science, University of Dschang, Cameroon, P.O Box 67, Dschang, Cameroon
2
Departamento de Biotecnología, Universidad Autónoma Metropolitana-Iztapalapa (UAM-I), Av. San Rafael Atlixco 186, 09340 Mexico, Distrito Federal, Mexico
3
Departamento de Alimentos y Biotecnología, Facultad de Química, Universidad Nacional Autónoma de México (UNAM), Ciudad Universitaria Coyoacán, 04510 Mexico, Distrito Federal, Mexico
World aquaculture production of fish accounted for 44.1% of total production (capture fisheries and aquaculture) in 2014, up from 42.1% in 2012 and 31.1% in 2004, making aquaculture to continuously be the world’s fastest growing food-production sector [1]. The global importance of fish aquaculture is related to its contribution to the reduction of the supply–demand gap of fish products, thus addressing the world food security and
Probiotics & Antimicro. Prot.
nutritious food concerns [1, 2]. This global trend of aquaculture development gaining importance in total fish supply has remained uninterrupted, as one of the world’s greatest challenges is to feed the forecasted more than 9 billion people by 2050 in a context of climate change, economic and financial uncertainty, and growing competition for natural resources [1]. Among the 15 countries known as top farm-food fish producers, the single African country found is Egypt [1]. Although the introduction of aquaculture in many countries of the subSaharan Africa traced back to 1950s, the productions have remained very tiny to meet the population demand, while the population growth is still increasing [3, 4]. Consequently, huge amounts of money are being spent annually for fish importation in the majority of the countries in the sub-Saharan Africa (Cameroon, Nigeria, Malawi, etc.…), and the governments’ trend is the intensification of aquaculture [5]. However, outbreaks of bacterial, viral and fungal infections are a primary constraint for the development of aquaculture worldwide [6]. In fact, the aquatic environment serves as support to pathogenic bacteria, and consequently, fish diseases may arise, especially at larval and early fry stages, leading to massive economic losses in the aquaculture sector. Coupled with the disease concern mentioned above is the indiscriminate and abusive worldwide use of antibiotic in aquaculture, which has not only led to the emergence and development of antibiotic-resistant bacteria but also to the contamination of aquaculture livestock [7, 8]. In this respect, probiotics are promising alternatives to chemicals and antibiotics treatments in aquaculture. They constitute a good ecofriendly measure and owe high efficacy and low cost. Probiotics used in aquaculture include lactic acid bacteria, (Lactobacillus, Lactococcus, Leuconostoc, Enterococcus and Carnobacterium), Vibrio sp., Bacillus sp., Pseudomonas sp. and Saccharomyces cerevisiae species [9–13]. Several recent studies have shown that they are applied for disease control, enhancing immune response, substituting the use of antimicrobial compounds, supplying nutrients and enzymatic contributions and upgrading water quality [2, 12, 14–16]. In our previous investigation, LAB were isolated from the intestines of tilapia and common carp cultured in earthen ponds in the western highland regions of Cameroon. Seven of them (Lactococcus lactis subsp. lactis 1FT, 1FW and 3FT; Lactobacillus plantarum 1MTK, 4BC and 13BC and Lactobacillus brevis 1BT) presented strong direct and extracellular antimicrobial activity against host-derived and nonhost-derived fish pathogens (Escherichia coli, Listeria monocytogenes, Salmonella sp., Salmonella Typhimurium, Staphylococcus aureus, Staphylococcus sp., Pseudomonas aeruginosa, Vibrio sp.). The bacteriocins produced by Lactococcus lactis subsp. lactis 1FT, 1FW and 3FT strains also exhibited strong inhibition towards Vibrio sp., Staphylococcus sp. and P. aeruginosa [5]. Therefore, These
LAB strains could be promising candidates for use as probiotics under the tropical climate of the sub-Saharan African countries. But, before their in vivo application, they need to be selected based on their in vitro probiotic and safety properties. The aim of this study was to investigate the in vitro probiotic and safety properties of bacteriocinogenic and nonbacteriocinogenic LAB strains from the intestines of tilapia and common carp cultured in earthen ponds in the western highland regions of Cameroon. Their tolerances to low pH, in the presence of high bile salt concentrations and simulated gastric and intestinal juices conditions as well as their bile salt hydrolase activity, cell surface characteristics, haemolytic activity, gelatinase activity, antibiotic resistance and ability to produce biogenic amines were determined.
Materials and Methods Bacterial Strains and Growth Conditions The 7 LAB strains used in this study were previously isolated from tilapia and common carp intestines as well as pond water [5]. They included 3 Lactococcus lactis subsp. lactis strains (1FT, 1FW and 3FT), 3 Lactobacillus plantarum strains (1KMT, 4BC and 13BC) and 1 Lactobacillus brevis strain (1BT). The accession numbers for their 16S rRNA sequences are KU921689 (strain 1FT), KU921690 (strain 1FW), KU921691 (strain 3FT), KU921695 (strain 1KMT), KU921697 (strain 4BC), KU921699 (strain 13BC) and KU921688 (strain 1BT). Strains 1FT, 1FW and 3FT showed the ability to produce bacteriocins that were active against Vibrio sp., Staphylococcus sp. and P. aeruginosa. The fish pathogens, selected based on their sensitivity to the LAB strains mentioned above and used as the indicators for the coaggregation test included Vibrio sp. 1T1 isolated from the intestine of farmed tilapia in Foumban–Cameroon, Listeria monocytogenes CFQ-103, Salmonella enterica subsp. enterica serovare Typhimurium ATCC 14028, and Staphylococcus aureus ATCC 25923, from the culture collection of the Faculty of Chemistry, Universidad Nacional Autonoma de México (UNAM). All the LAB strains were routinely grown in de Man, Rogosa and Sharpe broth (MRS broth, Difco) at 30 °C for 15 h. The fish pathogens were grown at 37 °C in brain heart infusion broth (BHI, Difco) for 24 h. Acid and Bile Tolerances of the LAB Strains Phenotypic Tests Acid and bile tolerances were performed according to the protocol of Verdenelli et al. [17] with some modifications. Acid tolerance of the strains was determined in phosphate-buffered
Probiotics & Antimicro. Prot.
saline (PBS, 0.8% NaCl, 0.144% Na2HPO4, 0.024% KH2PO4 and 0.02% KCl, pH 7.4) adjusted to pH 2.0 and 3.0 by adding 6 M HCl. In brief, LAB cells were harvested from a 15-h-old culture (5000 rpm, 15 min, 4 °C), washed twice in sterile PBS, and re-suspended in PBS adjusted to pH 2.0 and 3.0 such as to obtain a cell suspension equivalent to 4 Mc Farland turbidity standard (~12 × 108 cfu/ml). The cell suspensions were incubated at 37 °C for 0, 2 and 5 h and tolerant LAB were considered as residual viable cells count following enumeration on MRS agar at the end of the incubation period. The surviving percentages were calculated with reference to the number of residual viable cells in control (t = 0 h). For evaluation of bile tolerance, a 15-h-old culture of each LAB isolate was used to inoculate (1% v/v) MRS broth containing 0.3% (w/v) bile salts (Sigma-Aldrich) and incubated at 37 °C for 0 and 5 h. The residual viable cells count as well as the surviving percentages were determined as described earlier. PCR Detection of Acid and Bile Resistance Genes Genetic screening was based on sets of genes involved in bile salt tolerance and pH survival. These genes are listed in Table 1 below. Direct colony PCR (DC PCR) was performed in a total reaction volume of 50 μl containing 25 μl of 2x PCR master mix (Tsingke, China), 2 μl (50 pmol) of primer (Table 1) and a speck of isolated bacterial colony. The amplification was performed as described by Stack et al. [18] and Vrancken et al. [19] and amplicons were visualised after electrophoresis on 2% agarose gel (run at 90 V for 90 min) and staining with ethidium bromide using Gel Doc XR + Imaging System (BioRad, Madrid, Spain).
prepared by dissolving 1000 U/ml pepsin (from porcine gastric mucosa, Sigma, USA) and 0.01 g/l lysozyme (from chicken egg white, Sigma-Aldrich, USA) in a sterile buffer solution containing NaCl (2.05 g/l), KH2PO4 (0.60 g/l), CaCl2 (0.11 g/l) and KCl (0.37 g/l) and adjusting the pH to 3.0 with sterile 1 M HCl. Simulated intestinal juice (SIJ), was formulated by adding 3 g/l bile salts (Sigma-Aldrich, USA) and 1000 U/ml trypsin (from bovine pancreas, Sigma-Aldrich, USA) in a buffer at pH 8.0 consisting of 50.81 g/l Na2HPO4, 8.5 g/l NaCl and 1.27 g/l KH2PO4. Afterwards, LAB cells from a 15-h-old culture were harvested (5000 rpm, 15 min, 4 °C), washed twice with sterile PBS, and re-suspended in the SGJ so as to obtain a cell suspension equivalent to 4 Mc Farland turbidity standard. The cell suspensions were incubated at 37 °C for 0, 1 and 2 h and the residual viable cells count as well as their percentages were determined as previously described. Then cells from the previous SGJ tolerance step were harvested, re-suspended in the SIJ and the residual viable cells count were enumerated as same as for SGJ tolerance assay. Their surviving percentages were calculated with reference to the number of residual viable cells from the simulated gastric juice experiment. Bile Salt Deconjugation The ability to deconjugate bile salts was tested by using the plate assay described by Dashkevicz and Feighner [21]. Briefly, 15-h-old culture of a LAB isolate was spotted (5 μl) onto MRS agar plates containing 0.5% (w/v) oxgall (Sigma) and incubated at 37 °C for 72 h. The plates were finally observed for an opaque halo surrounding colonies, as a positive sign of bile salt deconjugation. Determination of Cell Surface Characteristics
Tolerance to Simulated Gastric and Intestinal Juices Auto-Aggregation Transit tolerance in the fish GIT was assessed through an in vitro model simulating gastric intestinal juices, following the protocol described by Corcoran et al. [20] with slight modifications. Briefly, simulated gastric juice (SGJ) was
Table 1
Auto-aggregation assays were performed according to the methodology described by Sahoo et al. [2] with slight modifications as follows: LAB cells were harvested from a 15-h-old culture in
List of primers used to screen our LAB strains for pH and bile salt survival
General function
Gene
Predicted function
Primer Orientation Sequence (5′ to 3′)
pH and bile salt survival
gtf
Glucan synthase
clpL
ATPase
F R F R
ACACGCAGGGCGTTATTTTG GCCACCTTCAACGCTTCGTA GCTGCCTTYAAAACATCATC TGG AATACAATTTTGAARAACGC AGCTT
Annealing temperature (°C)
Expected Relevant amplicon reference(s) size (bp)
58.0
374
Stack et al., 2010
56.0
158
Vrancken et al., 2009
Probiotics & Antimicro. Prot.
MRS broth, washed twice with PBS, re-suspended in the same buffer and adjusted to OD 600nm = 0.607 ± 0.05. Bacterial cell suspensions were vortexed for 10 s and subsequently incubated at 37 °C for 0, 2, 4, 6 and 24 h. The auto-aggregation percentage was determined using the equation: Auto−aggregation% ¼ ð1−At=AoÞ 100: Where At represents the absorbance at any time (2, 4 or 6 h), and A0 the absorbance at time t = 0 h.
Co-aggregation For co-aggregation, the LAB strains were grown in MRS broth for 15 h at 30 °C whereas the fish pathogens (Vibrio sp. 1T1, L. monocytogenes CFQ-103, S. Typhimurium ATCC 14028, and S. aureus ATCC 25923) were grown in BHI broth for 24 h at 37 °C. Bacterial suspensions were prepared as described in the auto-aggregation test above. Equal volumes (2 ml) of LAB and fish pathogen suspensions were mixed by vortexing (10 s) and incubated at room temperature without agitation for 4 h. Control tubes contained 2 ml of the suspension of each bacterial species. The absorbance (OD600nm) of the mixtures and controls were monitored after incubation [2]. Percentage of co-aggregation was calculated according to the following formula: Co−aggregation% ¼ ½ðApro þ ApatÞ−2Amix=ðApro þ ApatÞ 100:
Where Apro and Apat refer to the OD600nm of the LAB cell suspension and fish pathogen cell suspension respectively in control tubes and Amix represents the absorbance of the mixed bacterial suspension tested after 4 h.
Safety Evaluation of the LAB Strains Antibiotic Susceptibility Assay Antibiotic sensitivity and resistance of the strains were analysed according to the standardised broth microdilution method [23] using LAB susceptibility test agar medium (LSM), consisting of 90% (v/v) Iso-Sensitest broth (Oxoid, England) and 10% (v/v) MRS broth, pH adjusted to 6.7. In brief, strains were grown on MRS agar and colonies were resuspended in LSM. The suspension was diluted in the same medium and then added to each well of the microdilution plate (such as to obtain approximately 3 × 104 CFU/well) and incubated at 30 °C under anaerobic conditions for 48 h. After incubation, growth within each well was determined visually by comparing with the positive control and the MICs (μg ml−1) were subsequently determined. To distinguish resistant from susceptible strains, the MIC values were compared to the microbiological cut-off values defined by the FEEDAP panel [24] as well as Danielsen and Wind [25]. The antibiotics tested were penicillin G, ampicillin, chloramphenicol, erythromycin, tetracycline and gentamicin (All from Oxoid, England). Prior to use, they were dissolved and diluted in their suitable solvents according to the protocol of the antimicrobial susceptibility testing manual [26]. Haemolytic Activity Haemolytic activity was investigated as described by Gerhardt et al. [27]. Incubated blood agar plates (48 h at 30 °C) were examined for signs of β-haemolysis (clear zones around colonies), α-haemolysis (green zones around colonies) or γhaemolysis (no clear zones around colonies). Gelatinase Activity
Hydrophobicity The cell surface hydrophobicity of each strain was assessed by measuring microbial adhesion to hydrocarbons (MATHC) using the procedure described by Ekmekci et al. [22] with slight modifications. Briefly, cells at the stationary phase were centrifuged (10,000xg, 5 min). The resulting pellet was washed twice with PBS, re-suspended in the same buffer and the OD600nm was measured (A0). One millilitre of xylene or toluene was then added to 5 ml of cell suspension, and mixed by vortexing for 2 min. Then, the water and xylene/toluene phases were separated by incubation for 1 h at 37 °C. The aqueous phase was removed with care, and the OD600nm was measured (A1). The percentage of the cell surface hydrophobicity (H%) was calculated using the formula: Hydrophobicity% ¼ ð1−A1=AoÞ 100:
Gelatinase activity was carried out according to Harrigan and Mc Cance [28]. A 15-h-old culture was streaked onto nutrient gelatin agar (Oxoid). The plates were incubated anaerobically for 48 h at 30 °C, subsequently flooded with a saturated ammonium sulphate solution and observed for clear zones surrounding colonies. Determination of the Ability to Produce Biogenic Amines –
Phenotypic test
The capacity to produce biogenic amines (histamine, tyramine and putrescine) was assessed using the decarboxylase agar as described by Bover-Cid and Hopzapfel [29]. The precursor amino acids were histidine, tyrosine and ornithine respectively, purchased from Sigma. LAB strains strain were spotted onto the decarboxylase plates and
−
− +
− −
+
− −
−
− −
+
− −
8.64 ± 0.06
8.89 ± 0.36
Lb. plantarum 4BC
Lb. plantarum 13BC
t0 viable count at 0 h (control), t2: viable count at 2 h. t5 viable count at 5 h. + presence, − absence
0 ± 0.00 (0%)
9.64 ± 0.08 Lb. plantarum 1KMT
0 ± 0.00 (0%)
9.07 ± 0.18 Lb. brevis 1 BT
The values in brackets represent the mean survival percentages. The values presenting different superscript letters in the same column differ significantly (p ˂ 0.05)
7.72 ± 0.22
7.61 ± 0.11 6.22 ± 0.11 (65.30%)a
5.80 ± 0.29 (61.48%)a
a a
9.53 ± 0.28
8.60 ± 0.11 (90.19%)
a
8.06 ± 0.08 (85.49%)d 0 ± 0.00 (0%)a 0 ± 0.00 (0%)a
9.43 ± 0.13
7.49 ± 0.02
7.64 ± 0.23 8.91 ± 0.06 (97.10%)
8.78 ± 0.58 (94.81%)c 8.98 ± 0.04 (96.98%)b
9.17 ± 0.01 (99.95%) 9.18 ± 0.61
4.10 ± 0.14 (42.53%)c
8.40 ± 0.41 9.11 ± 0.16 L. lactis subsp. lactis 3FT
5.92 ± 0.11 (65.30%)
4.04 ± 0.05 (44.53%)
5.83 ± 0.24 (60.48%)c
9.26 ± 0.09
c c b b
6.53 ± 0.23 (73.25%)
5.69 ± 0.44 (60.27%)a 8.45 ± 0.21 (89.51%)a
8.57 ± 0.02 (96.05%) 8.92 ± 0.11
0 ± 0.00 (0%)a
9.34 ± 0.18 L. lactis subsp. lactis 1FW
0 ± 0.00 (0%)
0 ± 0.00 (0%)
0 ± 0.00 (0%)a
9.44 ± 0.04
6.61 ± 0.06
4.03 ± 0.25 a (55.20%) 5.71 ± 0.29 b (88.90%) 4.08 ± 0.11 c (58.57%) 8.23 ± 0.48 d (107.72%) 8.43 ± 0.04 e (112.55%) 8.44 ± 0.05 f (109.33%) 8.50 ± 0.07 e,f (111.69%) 7.30 ± 0.49 5.95 ± 0.07 (63.70%) 8.56 ± 0.01 (91.68%) 9.34 ± 0.09
a a
9.54 ± 0.14 L. lactis subsp. lactis 1FT
0 ± 0.00 (0%)
0 ± 0.00 (0%)
a
t2
t5
t2
The results of the phenotypic and genotypic assessment of bile salt and acid tolerances of the strains are summarised in Table 2. Only Lb. plantarum 1KMT and Lb. brevis 1BT could survive at pH 2.0 for 2 and 5 h, whereas all the strains survived to pH 3.0 conditions. Among all the strains tested, Lb. brevis 1BT strain was found to have the significantly highest acid tolerance, followed by Lb. plantarum 1KMT. However, concerning the bacteriocin-producing strains, L. lactis. subsp. lactis 1FW showed the significantly highest acid tolerance.
Strains
Acid and Bile Tolerances
Phenotypic and genotypic traits related to pH and bile salt tolerances of the LAB strains
Results
Table 2
All experiments were performed as independent triplicates. The results of the phenotypic characteristics related to probiotic potential (pH and bile tolerances, SGJ and SIJ tolerances, auto-aggregation, co-aggregation and hydrophobicity) were analysed by one-way ANOVA. When the means of more than two groups were different, they were compared two by two using Student’s t test. The differences were considered significant for p values
