Selection of Lactobacillus plantarum strains for their use as starter cultures in Algerian olive fermentations

The aim of this research was to evaluate some technological traits of L. plantarum strains previously isolated from fermented olives. For this purpose, 11 strains were tested for their in vitro antibiotic susceptibility, resistance to low pH values, acidifying activity, proteolytic activity, haemolytic activity, lactic acid and exopolysaccharide production and resistance to freeze-drying .Collectively, the strains were susceptible to most of the antibiotics tested and showed survival at pH 2. Most strains showed high (1.035 ± 0.29 to 0.912 ± 0.21 mmol/l ± sd of lactic acid) or medium (0.556 ± 0.29 to 0.692 ± 0.18 mmol/l±sd) acidification activity with good proteolytic activity (1.49 ± 0.25 to 5.25 ± 0.11 mg L –1 tyrosine). None of the strains produced exopolysaccharides or haemolysis in sheep’s blood.


INTRODUCTION
Olives are the fruit of the olive tree (Olea europaea), which is cultivated in all the countries of the Mediterranean region, especially in the central and southern regions of Spain and Italy, Greece, Turkey, Tunisia, Algeria and Morocco.In western Algeria (Sig), the green olives are directly brined and preserved by natural fermentation.Starter cultures are not employed during fermentation.This process is still performed by simply allowing the fruit to ferment spontaneously.No means whatsoever are taken to control fermentation, which in most cases is incomplete or affected by microbial spoilage.For these reasons great attention must be paid to the quality of fermenting olives.
Antibiotic resistance has been reported for lactic acid bacteria (Salminen et al., 1998).This is of interest because of its possible use to reconstitute the intestinal microflora of patients suffering from antibiotic-associated colitis.On the other hand, the transmission of antibiotic resistant genes to unrelated pathogenic or potentially pathogenic bacteria in the gut is a major health concern (Danielsen and Wind, 2003).
In a previous study (Kacem et al. 2004), a total of 11 L. plantarum strains were isolated from the fermented green olives produced in western Algeria (region of Sig).Of these, L. plantarum OL15 and OL9 strains produced bacteriocins towards Gram-positive and negative bacteria (Kacem et al., 2005(Kacem et al., , 2006)).
In this work, we aimed to evaluate the resistance of these strains to a range of antibiotics, and to screen them for some criteria (carbohydrate fermentation, lactic acid and exopolysaccharide production, tolerance to acidic pH values, proteolytic and haemolytic activities and resistance to freeze-drying ) in order to select starter culture strains.

Testing for resistance to antibiotics
Bacterial antibiotic resistance was determined on solid MRS medium by the use of 11 different antibiotic discs (bioMérieux, Marcy-l'Etoile, France) (Table 3).The results (average of 3 reading) were expressed as sensitive (S) or resistant (R) thanks to the standard disc diffusion method (National Committee for Clinical Laboratory Standards, 1999).Two strains with known antibiotic resistances (Staphylococcus aureus ATCC 25923 and Enterococcus faecalis ATCC 29212) were used as the control strains (Kacem and Kaid-Harche, 2008).

Fermentation of carbohydrates
The ability of the different L. plantarum strains to ferment different carbohydrates was determined using API 50 CHL test kits (bio Mérieux SA).The API test strips were prepared as recommended by the kit supplier and scored after incubation for 24 and 48 hours at 37 or 30°C.The interpretation of the fermentation profiles was facilitated by the use of the computer-aided database ''API-LAB plus'' (API bioMe´rieux).

Lactic acid production
Production of lactic acid was measured by the precipitate method according to Pryce (1969) and expressed in mmol/l.Results are the average of three replicates.

Acidifying activity
Acidification was measured by the change in pH (∆pH) over time according to the methods of Lombardi et al. (2002) and Ayad et al. (2004) .Fifty millilitres of MRS (Oxoid) were inoculated with 2% culture (in order to standardize the assay, the inoculums were approximately 10 6 cfu/ml) and incubated at 37°C.The pH was measured at 0, 2, 4 and 6 h using a pH-meter (Micro pH 2002, Crison, Barcelona, Spain).The acidification values were expressed as pH decrease, calculated as the difference between the value immediately after inoculation and values at 0, 2, 4 and 6 h (∆pH = = pH at time -pH zero time) .The cultures were considered fast, medium or slow acidifying when a ∆pH of 0.4 U (pH units) was achieved after 3, 3-5 and > 5 h, respectively (Kacem and Kaid-Harche, 2008).

Tolerance to acidic pH values
Strains were grown in MRS broth (Oxoid) at 30°C overnight, then subcultured into fresh MRS broth and incubated for another 24 h.The cultures were centrifuged at 5000 ϫ g for 10 min.at 4°C.The pellets were washed in sterile phosphate-buffered saline (PBS) (Oxoid), pH 7 and resuspended in PBS.Each strain was diluted 1/100 in PBS at pH 1, 2 and 3. Incubation times were 2, 4 and 6 h.Bacteria were then transferred to MRS broth (Oxoid) and incubated at 37°C overnight (El-Naggar, 2004).Counts of surviving cells were determined as described by Kacem and Kaid-Harche (2008).Bacterial growth was expressed in colony forming units per milliliter (cfu/ml) and the survival percentage (% ± sd) of strains to different pH values was then calculated.The experiment was repeated twice and each reading represents the means of three observations.

Proteolytic activity
The proteolytic activity of strains was determined in a skimmed-milk medium (reconstituted skimmed milk powder 10% w/v) using the tyrosine method (Hull, 1947), in accordance with the International Dairy Federation (IDF) standard 149A (1997).Milk was inoculated at 0.2% with each strain precultured in MRS broth (Oxoid) at 37°C for 18 h, to obtain approximately 10 6 cfu / ml and then incubated at 30°C for 24 h.The IDF method is based on the reaction of the amino acids tyrosine and tryptophan released from the milk substrate at 72 h with a phenol reagent, yielding a blue color which is measured at 650 nm.The results were calculated from a calibration curve obtained from dilutions of tyrosine in distilled water and expressed as mg tyrosine l -1 of milk (Kacem and Kaid-Harche, 2008).

Exopolysaccharide production
Exopolysaccharide production was evaluated as reported by Kacem and Kaid-Harche (2008).Overnight cultures were streaked on the surface of plates containing ruthenium red milk (10% w/v, skim milk powder, 1% w/v, sucrose and 0.08 g/l ruthenium red, 1.5% w/v agar) (Oxoid).After incubation at 37°C for 24 h, non-ropy strains gave red colonies due to the staining of the bacterial cell wall, while ropy strains appeared as white colonies.

Influence of freeze-drying
Resistance of strains to freeze-drying was determined according to the Font de Valdez et al. (1985) method modified by Kacem and Kaid-Harche, 2008).Strains were grown in MRS broth (Oxoid) for 18 h at 37°C, centrifuged (10000 g at 4°C for 10 min), and then suspended to the initial volume in a cryoprotective solution containing 100 g/l non-fat milk solids, 100 g/l sucrose, and 4 g/l ascorbic acid.The mixture was frozen at -80°C for 24 h and freeze-drying was then carried out at -50°C for 48 h under 0.01 mbar vacuums (Model Lyph Lock 4.5, Labconco Corporation, Kansas City, MO, USA).Freeze-dried cultures were immediately rehydrated to their initial volume in MRS broth (Oxoid) for 10 min at 37°C, and homogenized by high speed vortexing.Cell survival was then determined by the pour plate method.

RESULTS AND DISCUSSION
In this study, a total of 11 Lactobacillus plantarum strains were screened for their performance regarding several growth characteristics.This characterization could be used as selection criteria for novel starter cultures in olive fermentations.
The results of carbohydrate fermentation by the 11 strains of lactobacilli tested are shown in Table 1.All strains fermented D and L-arabinose, galactose, D-glucose, inositol, salicin, lactose, cellobiose and inulin.Additionally, Table 2 shows the best two final identifications for each strain on API gallery.Bacteria were identified as L. plantarum which confirms our previous results (Kacem et al., 2004).
Table 3  resistance was observed.This is not in accordance with various reports indicating that lactic acid bacteria are normally resistant to the principal antibiotics, such as penicillin G, ampicillin, vancomycin, cloramphenicol or ciprofloxacin (Halami et al., 2000 andCoppola et al., 2005).In a similar study conducted by Herreros et al. (2005) most of the tested L. plantarum strains from different sources were resistant to the antibiotics used.In addition, our results show that 4 strains were tetracycline resistant which is in accordance with other reported studies (Danielsen, 2002;  Herreros et al. 2005).It is well known that vancomycin is an antibiotic belonging to glycopeptide antibiotics and inhibits peptidoglycan synthesis which is an important structural component of the bacterial cell wall.Therefore, Gram-positive bacteria, including lactic acid bacteria are especially vulnerable to vancomycin treatment (Reynolds, 1989).In our case, all strains tested were sensitive to vancomycin.This result not confirm the finding of Salminen et al. (1998), who reported that vancomycin resistance is an intrinsic property of lactobacilli.
Generally, antibiotic sensivity or resistance has been studied in strains of lactobacilli isolated from dairy products, animals or humans (Torriani et al., 1988 andSalminen et al. 1998).All these works have reported that several isolates showed multiple resistance to most antibiotics tested.However, our strains of L. plantarum tested for sensivity or resistance to antibiotics were isolated from vegetable material (fermented green olives) (Kacem et al., 2004) where this type of treatment is not practiced.This is probably one of the reasons, among others, that can explain our findings.
As shown in Table 4, the values of proteolytic activity of the 11 strains studied ranged from 1.49 ± 0.25 and 5.25 ± 0.11 mg l -1 tyrosine at 72 h.Our results are high in comparison with those reported by other authors for lactobacilli isolated from milk (Schmidt et al., 1994 andBadis et al., 2004).
Lactic acid production is reported in Table 4.These results are not in accordance with those reported by Ayad et al. (2004) who indicated that most strains of L. plantarum isolated from different sources show a slow acidification rate.
Resistance to freeze-drying is described in Table 4.The results indicate that the strains readily grew in MRS agar after freeze-drying and show high survival percentages (up to 60 ± sd %) (Table 4).This characteristic is critical for the storage of frozen stock cultures used in industrial production.Finally, none of the strains produced exopolysaccharides or haemolysin in sheep's blood.
Table 5 shows the results of acid tolerance (survival percentage of L. plantarum strains at various pH values).All tested strains survived incubation periods of 2 h to 6 h at pH 2.0 and pH 3.0 with a decrease in survival percentage when exposure time progresses for the strains.Generally, L. plantarum OL12, OL15, OL16 and OL33 strains survived acidic conditions better than the rest of strains.At pH 2.0, L. plantarum OL15 strain showed the highest survival percentage (65 ± 0.3%, 53 ± 2% and 28 ± 1.4%) after 2, 4 and 6 h incubation periods, respectively.No growth occurred after incubation at pH 1 for 2 h.Similar results were reported by Draser et al. (1969); Dunne et al. (2001) and El-Naggar (2004).
In addition, among these interesting strains, it has been reported in previous studies (Kacem et  al., 2005, 2006; Kacem, 2007) that L. plantarum OL15 and OL9 strains produced bacteriocins (proteinaceous antimicrobial substances compound) with inhibitory activity against olive spoilage and phytopathogenic bacteria organisms, including, Pseudomonas genera that can cause severe damage in olive groves and occasionally on the fruits (Holtsmark and al., 2008).This suggests that these strains are favorable for use as starter cultures for olives production.