IJFS#1029_bozza Ital. J. Food Sci., vol. 30, 2018 - 414 PAPER EFFECT OF DIFFERENT STORAGE CONDITIONS ON THE SHELF LIFE OF NATURAL GREEN TABLE OLIVES S.J. LOMBARDI, V. MACCIOLA, M. IORIZZO and A. DE LEONARDIS* Department of Agricultural, Environmental and Food Sciences, University of Molise, Via De Sanctis, 86100 Campobasso, Italy E-mail address: antomac@unimol.it ABSTRACT The aim of this research is to study the effects of different storage conditions on Spanish (alkaline debittering) and natural (directly brined) green olives. Laboratory-processed olives were stored in 6% brine or in a vacuum bag without brine, at 6 or 20°C. After 18 months, natural olives showed higher microbial and olive oil stability than NaOH-treated olives. The lower pH (<4.80) and higher total phenol content (0.2 g/100 g wet pulp) influenced positively the long shelf life of natural olives. The packaging in 6% NaCl brine and in a vacuum bag stored at 20°C gave better performance, while growth of psychrophilic spoilage bacteria occurred at cold temperature. Keywords: microbial count, natural green olives, oil oxidation, phenols, sensory profile Ital. J. Food Sci., vol. 30, 2018 - 415 1. INTRODUCTION World consumption of table olives (TO) continues to increase, passing from 957 to 2,480 thousand tones in the period from 1990/91–2014/15 (IOOC, 2017). Major producer countries are located in the Mediterranean basin (Spain, Greece, Italy, Turkey, Tunisia, Morocco, Egypt, and Algeria). Green Spanish-style, natural black Greek-style and black ripe Californian are the most popular TO types; however, many other typologies are produced according to local traditions (ERCOLINI et al., 2006). In fact, TO types can vary depending on olive fruit kind (green or black olives), debittering method (chemical or biochemical), fermentation conditions (temperature, levels salt, and type of acid) and finally, canning (SANCHEZ et al., 2006). Debittering and fermentation are the main steps of the table olive process. The first is fundamental to remove natural bitterness, while the fermentation is important to prevent microbial spoilage. The typical strong bitter and pungent taste of green olives is due to the high presence of oleuropein, a secoiridoid containing elenolic acid glucoside linked to hydroxytyrosol (AMBRA et al., 2017; GHANBARI et al., 2012). In the chemical process, such as the Spanish-style method, olives are debittered in a few hours through a bath in a sodium hydroxide solution that hydrolyzes oleuropein quickly into less bitter compounds (CHAROENPRASERT and MITCHELL, 2012); subsequently, olives are water-washed repeatedly and then brined in 8-10% NaCl where a spontaneous lactic acid fermentation occurs in the following months (APONTE et al., 2010). Olive fermentation is caused by indigenous microbiota, including lactic acid bacteria (LAB) and/or yeasts, which transform the sugars in lactic acid with a consequential decrease in pH; in addition, substances inhibiting undesirable microorganisms are produced (BEVILACQUA et al., 2009). Moreover, several microbial strains are able to hydrolyse oleuropein contributing to olive debittering (IORIZZO et al., 2016; SERVILI et al., 2006). IOOC (2017) indicates the natural olives as “green olives, olives turning colour or black olives placed directly in brine in which they undergo complete or partial fermentation, preserved or not by the addition of acidifying agents”. Therefore, raw olives are placed in water or brine as long as required for the spontaneous loss of bitterness. In this case, oleuropeinolytic bacterial/yeast strains degrade oleuropein in synergy with specialized enzymes naturally occurring in the olives (RAMIREZ et al., 2017a; RAMIREZ et al., 2016). However, olive debittering is slow and several months are needed to obtain a satisfactory good-tasting product (SANCHEZ et al., 2006). In recent years, demand of organic and natural foods has become trendy (ROZIN et al., 2004). In this contest, the demand for natural TO risen since they have become popular and appreciated for the nutritional value and distinctive sensory characteristics. Several studies on natural TO have been carried out focusing on the content of bioactive components (CHAROENPRASERT and MITCHELL, 2012; SAKOUHI et al., 2008) and microbial characterization (BLEVE et al., 2015; PEREIRA et al., 2008). Moreover, also new systems to accelerate olive debittering with biotechnological approaches as an alternative to chemical treatment have been tried obtaining positive results by using enzymes (DE LEONARDIS et al., 2016), selected LAB (PERRICONE et al., 2013; HURTADO et al., 2010), batches in modified brine (RAMIREZ et al., 2017b) and vacuum impregnation (TAMER et al., 2013). In this study, several chemical, microbial and sensory parameters have been tested in green olives laboratory-processed through both the Spanish-style and natural methods and stored for a period of time 18 months, under different conditions. Specifically, the effects of storing in brine were compared with those in vacuum bags, without brine, in order to evaluate the possibility to reduce the salt, potentially harmful to health. Finally, considering that in the supermarkets the packaged olives are frequently placed in Ital. J. Food Sci., vol. 30, 2018 - 416 refrigerator, the effects of cold temperatures have been studied. The results could be useful to the improvement of natural olives, in olive packaging and conservation systems. 2. MATERIALS AND METHODS 2.1. Chemicals All reagents were of analytical or HPLC grade. Gallic acid, hydroxytyrosol and oleuropein were purchased from Sigma-Aldrich Co (St. Louis, MO, USA). All the media and the supplements for the microbiological analysis, unless otherwise stated, were from Oxoid, Basingstoke, Hampshire, UK. 2.2. Olive processing and storage conditions About 10 kg of green or yellow-green olive fruits were handpicked from ‘Cazzarella’ trees, a dual-purpose autochthonous cultivar grown in the Molise region (Italy). At the laboratory, the olive batch was split and processed as described synthetically in Table 1. Table 1. Olive sample typologies. 1. Raw olives: fresh untreated olives. 2. B1-NaOH / B1-Nat: olives brined in 6% NaCl solution (first brining) and stored in the dark under 20 °C constant temperature for 9 months. 3. B2-NaOH / B2-Nat: olives of point 2 brined in 6% NaCl fresh solution (second brining) and stored in the dark under 20 °C constant temperature for other 9 months. 4. C-NaOH / C-Nat: olives of point 2 packaged without brine in vacuum bags and stored in a refrigerator under 6 °C temperature (C = cold) for other 9 months. 5. T-NaOH / T-Nat: olives of point 2 packaged without brine in vacuum bags and stored in the dark under 20 °C constant temperature (T = tempered) for other 9 months. -NaOH: processed by Spanish method; -Nat: natural processed. In the preparation of the B1-NaOH sample, olives were placed into a 2% NaOH solution for 16 hours (ratio olive/lye 1:1.3, w/v); olives were washed repeatedly with water until neutral pH and brined into 6% NaCl solution (ratio olive/brine 1:1.3, w/v) in three different closed glass jars. In the case of the B1-Nat sample, the olives were immersed directly into 6% NaCl solution (ratio olive/brine 1:1.3, w/v); three weeks after (21 days), the brine was renewed with a fresh 6% NaCl solution. All jars were stored in the dark under 20°C constant temperature for 9 months. Thus, the olives were water-washed repeatedly, analyzed, repackaged and stored for another 9 months as described in Table 1 (B2-, C-, T-). 2.3. Analytical determinations Moisture and total fat were determined on the pulp separated manually from the stone and homogenized. Moisture was determined at 105°C on about 5 g of pulp; successively, oil content was measured in dry pulp by using Soxhlet extraction with 40-60°C petroleum ether. Lactic acid, pH and total phenol were determined on both olive pulp and brine. Ital. J. Food Sci., vol. 30, 2018 - 417 Brine was filtered through a 0.45 μm PVDF syringe-filter before assaying. As for the pulp, about 20 g of fresh pulp was dispersed in 60 mL of distilled water (1:3, w/v) and the flask was put into a sonicator bath for 20 minutes at room temperature by filtering the aqueous phase on paper. The pH measurement was performed through a pH-meter, while the lactic acid was measured by titration with 0.1 M NaOH up to pH 7.0. Free acidity and fatty acid composition were determined on the oil extracted at cold temperatures from 100 g of olive pulp by following the methods described in PASQUALONE et al. (2014). The free acidity (expressed as g oleic acid per 100 g pulp olive) was measured following the olive oil European Union Commission (1991), while the fatty acid composition was determined by gas-chromatographic analysis as described in DE LEONARDIS and MACCIOLA (2012). p-anisidine value was determined on the cold extracted oil according to AOCS (1998). Phenol extraction from the olive pulp, determination of total phenols and the HPLC analysis with UV detector were carried out in the same conditions described in DE LEONARDIS et al. (2016). Hydroxytyrosol (Hy), dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol (HyEDA) and oleuropein (OLE) were monitored and each of them was quantified through a hydroxytyrosol calibration curve derived from a plot of area counts versus concentration. Hy-compounds were calculated as the sum of Hy, HyEDA and OLE expressed as mg/100 g wet pulp olives as Hy equivalent. 2.4. Microbial counts Microbiological analysis was carried out on the olive pulp according to the method reported in IORIZZO et al. (2016) with modifications. For each sample, about 30 g of olives were homogenized in 0.9% NaCl (w/v) for 1 min in a Stomacher bag (Bag Mixer-400). Pulp homogenates were diluted serially and plated in triplicate by using the following growth media and incubation conditions: Plate Count Agar (PCA) at 30°C (72 h) and 15°C (5 days) for total mesophilic and psychrotrophic aerobic bacteria; Violet Red Bile Glucose Agar (VRBGA) at 37°C for 24 h for Enterobacteriaceae; MRS Agar, added with 0.17 g/L of cycloheximide (Sigma-Aldrich Co, St. Louis, MO, USA), at 30°C for 4 days in anaerobiosis (Anaerogen kit, Oxoid, Basingstoke, United Kingdom) for lactic acid bacteria (LAB); YPD agar (1% yeast extract, 2% peptone, 2% glucose and 2% agar) supplemented with 100 mg/L chloramphenicol at 28°C for 72 h for yeasts; Pseudomonas Agar Base (PAB), added with CFC selective supplement, at 30°C for 48 h for Pseudomonas spp. 2.5. Sensory analysis Sensory profile was evaluated according to Galán-Soldevilla and Pérez-Cacho (2012), with modifications. The panel group was composed of 15 untrained volunteers selected and led by one olive oil expert taster. Training and alignment of panel and sensory profile sheet was developed by using similar commercial products. Nine attributes (defects, fruit odor, salty, bitter, acid, firmness, fibrous, crunchy and overall opinion) were selected and evaluated through a seven-point hedonic scale (7: like extremely; 6: like very much; 5: like slightly; 4: neither like nor dislike; 3: dislike slightly; 2: dislike very much; 1 = dislike extremely). Value zero was given for absent attribute. During the panel training, panel leader held a discussion on the descriptors and a consensus lexicon was developed about the following nine attributes (GALÁN-SOLDEVILLA and PÉREZ-CACHO, 2012): a) defects (presence of abnormal negative odor/aroma); b) gustatory sensations (salty: typical taste produced by sodium chloride aqueous solutions; bitter: basic taste produced by diluted aqueous solutions of caffeine: acid: basic taste produced by aqueous solutions of substances like citric acid); d) fruity odor (odor/aroma characteristic of fresh olives); e) kinesthetic sensations (firmness: mechanical property of texture related to the strength Ital. J. Food Sci., vol. 30, 2018 - 418 required to attain a certain penetration of the olive; fibrous: geometrical property of texture related to the perception of strands oriented in the same direction; crunchy: mechanical property of texture related to the cohesion and strength necessary to break an olive with teeth); f) overall judgment (general grade of appreciation). 2.6. Elaboration data Each jar was analyzed independently in duplicate (six independent replicates for every sample), by calculating mean and standard deviation. Means obtained were compared through ANOVA (Duncan's multiple range post hoc tests at p<0.05) by using the SPSS 13.0 software (SPSS, Inc., Chicago, IL, USA). 3. RESULTS AND DISCUSSION 3.1. Basic olive pulp characteristics Processing method strongly affects the final characteristics of table olive (SANCHEZ et al., 2006) and the data reported in Table 2 confirm this evidence. Olive water content was initially 69.4%. Overall, water content in the processed olives was generally higher in the olives treated with NaOH than in the natural-processed (Nat). B1- NaOH and B2-NaOH samples, stored in the brine, kept the original moisture (69.6 and 72.6%, respectively), while the matching B1-Nat and B2-Nat samples showed a lower value (63.7 and 67.2%, respectively). However, the observed differences were not statistically significant. Conversely, significant moisture reduction was observed in the olives stored in the vacuum bags. Specifically, water content was 60.1, 59.0, 53.0 % in the C-Nat, T-NaOH and T-Nat samples, respectively. An increase of lactic acid and a decrease of pH expected as result of the microbial activity. All the NaOH-treated olives had about one-higher-point of pH than the natural olives. More specifically, pH values were 5.17, 5.39, 5.56 and 5.73 in B1-NaOH, B2-NaOH, C- NaOH and T-NaOH, respectively. These high pH values are potentially harmful being not fit to inhibit growth of spoilage microorganisms. Actually, brine acid characteristics of the NaOH treated samples (Table 2) were below the requirements for trade according to the IOOC (2004) that has a set the minimum brine pH value at 4.1; however, addition of acidity regulators is permitted. The pH values of Nat olive pulp were 4.62, 4.42, 4.80 and 4.75 in B1-Nat, B2-Nat, C-Nat and T-Nat, respectively. The pH value was obviously related to lactic acid, which was found in the Nat olives more-than-three-times higher than NaOH-treated olives. According to MARSILIO et al. (2005), lye treatment and subsequent water washing presumably caused sugar and nutrient loss from olive pulp giving rise to insufficient acidification in the NaOH-treated samples. In the set group of Nat olives, the maximum and minimum values of lactic acid were C-Nat (0.24%) and B2-Nat (0.11%), respectively. These values are in line with literature (DE LEONARDIS et al., 2016; APONTE et al., 2012). 3.2. Phenolic content Many studies have evidenced a close relationship between processing method and content/composition of phenols in table olives (AMBRA et al., 2017; CHAROENPRASERT and MITCHELL, 2012; BLEKAS et al., 2002). In Table 3 phenolic compound evolution is given only for the Nat olives due to the phenol content in the NaOH-treated olives was negligible. Ital. J. Food Sci., vol. 30, 2018 - 419 Table 2. Analytical characteristics of the pulp olive samples and relative canned brine. Pulp olive Brine Moisture % pH Lactic acid % pH Lactic acid g/l Raw olives 69.4(4.3)a,b 5.67(0.06)a nd 7.02(0.06)a n.d. Stored in the first brine B1-NaOH 69.6(5.2)a,b 5.17(0.38)b,c,d 0.02(0.00)a 5.05(0.27)b 0.4(0.0)a B1-Nat 63.7(5.3)b 4.62(0.34)b,c 0.22(0.02)b 4.26(0.16)c 2.5(0.2)b Stored in the second brine B2-NaOH 72.6(4.2)a,b 5.39(0.41)b,c,d 0.03(0.00)a 5.02(0.25)b 0.5(0.0)a B2-Nat 67.2(4.1)a,b 4.42(0.26)b 0.11(0.01)c 4.10(0.22)c 2.5(0.2)b Stored in vacuum bags at 6°C C-NaOH 69.1(3.8)a,b 5.56(0.29)c,d 0.07(0.03)a C-Nat 60.1(3.3)b 4.80(0.31)b,c,d 0.24(0.02)b Stored in vacuum bags at 20°C T-NaOH 59.0(3.6)b 5.73(0.29)d 0.03(0.00)a T-Nat 53.0(3.2)b 4.75(0.26)b,c,d 0.19(0.03)b Values are means (standard deviation) of six independent replicates; letters on the column point out significant difference at p<0.05. -NaOH: lye processed; -Nat: natural processed; n.d. = not detected. Table 3. Total phenol and hydroxytyrosol (Hy) compounds obtained in pulp olives and related brine (natural olives). Pulp olives Brine Total phenols % GAE Hy-compounds mg/100g as HyE Total phenols mg/L GAE Raw olives 1.19(0.08)a 199(13)a B1-Nat 0.21(0.01)b 53(4)b 1,791(102)a B2-Nat 0.14(0.01)b 25(1)c 1,006(132)b C-Nat 0.20(0.01)b 47(3)b T-Nat 0.20(0.01)b 62(4)b Values are means (standard deviation) of six independent replicates and different letters on the column point out significant difference (p<0.05). Specifically, Nat olive samples showed total phenol values dropped from 1.19 (raw olives) to about 0.20 g/100g (Nat processed olives) as GAE (gallic acid equivalent), respectively. Although the reduction of the starting phenol content is a coveted result from an organoleptic point of view, presence of residual phenols may be positive from a nutritional point of view, since several phenols are recognized as antioxidant and healthy substances (D’ANTUONO et al., 2016; DE LEONARDIS et al., 2013; DE LEONARDIS et al., 2008; SANCHEZ et al., 2006). At the end of the first brining (B1-Nat), Hy-compound content was 53 mg per 100 g olives and insignificant changes were observed in C-Nat (47 mg/100 g) and T-Nat (62 mg/100g); conversely, Hy-compounds halved during the second brining (B2-Nat, 25 mg/100g). Ital. J. Food Sci., vol. 30, 2018 - 420 The European Union Commission (2012) recently approved a health claim for extra virgin olive oil stating that ‘daily intake of 5 mg of hydroxytyrosol and its derivatives is able to prevent low density lipoprotein (LDL) oxidation’. Therefore, Nat olives may be claimed as a source of phenols having health-promoting properties. However, the Hy-compound values reported in Table 3 were certainly underestimated because other hydroxytyrosol linked compounds, known to be present in olives (AMBRA et al., 2017; CHAROENPRASERT and MITCHELL, 2012), have been omitted in this study for purposes of simplification of the model. According some studies (RAMIREZ et al., 2017a; Ramirez et al., 2016; DE LEONARDIS et al., 2015), biological and enzymatic oleuropein depletion may occur through a rapid transformation in Hy-EDA and subsequent hydrolysis with gradual Hy release. A similar pathway occurred also in the Nat olives, as noted in the graph of Fig. 1. Figure 1. Changes in the table olives of the monitored phenolic compounds, expressed in mg (Hy equivalent) per 100 g dry olive pulp. Raw olive pulp contained 37, 89 and 158 mg/100 g in dry matter (d.m.) of Hy, Hy-EDA and OLE, respectively. After 21 days of brining (when the first brine was replaced, see Material and Methods), the olive phenol profile was changed in 73, 167 and 22 mg/100 g d.m. of Hy, Hy-EDA and OLE, respectively (data not shown). Hy-EDA is a recognized antimicrobial compound inhibiting the growth of lactic bacteria (MEDINA et al., 2007). After 9 of months brining (B1-Nat), OLE was completely depleted, while Hy-EDA was 45 mg/100 g d.m. Successively, Hy-EDA was undetectable in sample B2-Nat, C-Nat and T- Nat in which only Hy was found in the amounts of 37, 77 and 116 mg/100 g d.m., respectively. 3.3. Lipid fraction The characteristics of lipid fraction of the raw olives and samples stored for 18 months are given in Table 4. Ital. J. Food Sci., vol. 30, 2018 - 421 Table 4. Changes of lipid fraction of olives samples. Raw olives NaOH-processed olives Nat-processed olives B2-NaOH C-NaOH T-NaOH B2-Nat C-Nat T-NAT Total oil % w.m. 18.4 (1.1)a 19.5 (1.2)a 19.3 (1.2)a 17.5 (1.1)a 22.1 (1.3)b 22.2 (1.3)b 21.6 (1.3)a,b Total oil % d.m. 26.3 (1.6)a 26.8 (1.6)a 28.2 (1.7)a 29.3 (1.8)a 33.0 (2.2)b 36.5 (2.2)c 40.5 (2.5)d Free acidity % 0.5 (0.0)a 11.2 (0.7)b 15.3 (0.9)c 13.7 (0.8)d 2.8 (0.2)e 5.5 (0.3)f,a 3.4 (0.2)e p-Anisidine value 5.9(0.3)a 33.5(1.7)e 30.7(1.6)d 31.6(1.7)d,e 12.8(0.7)b 29.2(1.5)d 23.1(1.2)c Fatty acids % C16:0 16.5 (1.0)a 23.7 (1.4)b,c 24.4 (1.5)b 21.6 (1.3)c,d 17.0 (1.0)a 19.9 (1.2)d 18.2 (1.1)a,d C16:1 1.4 (0.1) 1.4 (0.1) 1.3 (0.1) 1.2 (0.0) 1.3 (0.1) 1.4 (0.2) 1.3 (0.1) C18:0 2.9 (0.2)a 4.2 (0.3)b 4.6 (0.3)b 4.5 (0.3)b 3.4 (0.2)c 3.9 (0.3)b,c 3.8 (0.2)c C18:1 65.5 (4.0)a 60.5 (3.7)a,b 57.1 (3.5)b 60.0 (3.6)a,b 64.5 (3.9)a 63.0 (3.8)a,b 64.3 (2.7)a,b C18:2 11.6 (0.7)a 4.2 (0.3)b 3.2 (0.2)c 5.1 (0.2)d 11.3 (0.5)a 8.2 (0.3)e 9.3 (0.3)f C18:3 0.6 (0.0)a 0.1 (0.0)b 0.1 (0.0)b 0.2 (0.0)b 0.6 (0.1)a 0.3 (0.1)a 0.4 (0.1)a C20:0 0.5 (0.0)a 0.7 (0.0)b 0.8 (0.0)b 0.7 (0.0)b 0.5 (0.0)a 0.6 (0.0)a 0.6 (0.0)a C20:1 0.2 (0.0) 0.3 (0.0) 0.3 (0.0) 0.3 (0.0) 0.3 (0.0) 0.2 (0.0) 0.2 (0.0) C22:0 0.1 (0.0)a 2.9 (0.7)b 4.8 (0.2)b 3.8 (0.6)b 0.3 (0.0)c 1.2 (0.5)c 0.7 (0.3)c C22:1 0.0a 1.5 (0.6)b 2.7 (0.2)b 1.9 (0.8)b 0.1 (0.0)a 0.7 (0.2)c 0.3 (0.3)c Values are means (standard deviation) of six independent replicates and different letters on the lines point out a significant difference (p<0.05). A different oil content was observed between NaOH- and Nat-processed olives. Apparently, total oil content increased in the Nat-olives, especially considering total oil on dry matter. TAMER et al. (2013) have found that alkali treatment caused oil loss from the olives. Moreover, the advanced fermentation status (with large sugar consumption) and the reduced lipolytic activity (with low fat consumption) could explain the apparent increase of oil in the Nat-olives (Table 4). Unfortunately, we did not have enough data to establish a possible link between oil increase and microbial counts and types (Table 5). According to what was observed in previous studies (PASQUALONE et al., 2014; LOPEZ et al., 2011), a higher hydrolytic and oxidative oil degradation was observed in the NaOH- vs Nat-olives. Specifically, a considerable increase of free acidity was evident in B2-NaOH, C-NaOH and T-NaOH with values of 11.2, 15.3 and 13.7% on oil extracted, respectively; conversely, free acidity was less than 5.5% for the Nat-olives. Formation of free fatty acids could be catalysed by the lipases contained in the fruits or synthesized by the environmental microflora. A positive effect of low temperature on the triacylglycerol hydrolysis was evident due to the high free acidity values found in C-NaOH (15.3%) and C-Nat (5.5%). Similarly, oil oxidation was higher in the cold temperature stored olives, as the p-anisidine values show (Table 4). As known, p-anisidine assay measures secondary lipid oxidation compounds; the highest p-anisidine values were obtained in the oils extracted from the B2-NaOH, C-NaOH, T-NaOH and C-Nat olives. A fatty acid composition of raw olives (Table 4) was in agreement with the literature (MALHEIRO et al., 2012). Oleic acid was the main fatty acid (65.5%), followed by palmitic acid (16.5%) and linoleic acid (11.6%). In order to make Table 4 more readable, the C17:0, Ital. J. Food Sci., vol. 30, 2018 - 422 C17:1 and C24:0 fatty acids (found in equal or less amount 0.1%) were deliberately omitted; however, these fatty acids did not change during the storage of the olives. Significant modifications of fatty acid composition occurred especially in the NaOH- olives, specifically, a reduction of linoleic and linolenic acid was registered in the B2- NaOH, C-NaOH, T-NaOH samples; in the same samples a decrease of oleic acid and an increase of palmitic and stearic acid were obtained. It is known that polyunsaturated fatty acids were more affected by oxidation and therefore decreased at a greater extent (CAPONIO et al., 2003; DE LEONARDIS and MACCIOLA, 2012). In the Nat-olives, especially in B2-Nat and T-Nat, the olive residual phenols have limited oil oxidation, as evidenced by the minor changes of p-anisidine and fatty acid profile. Unexpectedly, significant increase of the peaks corresponding to behenic (C22:0) and erucic (C22:1) fatty acids were found in the lipid fraction of stored olives. Generally, these long chain fatty acids are present in olive oil in amounts lower than 0.2%, as the data of raw olive oil shows (Table 4). In the NaOH treated olives, C22:0 and C22:1 were found in amounts higher than 2.9 and 1.5%, respectively. In particular, lipid fraction of the C-NaOH sample showed the highest content (C22:0 = 4.8%; C22:1 = 2.7%). Conversely, in the Nat-olives percentage of C22:0 and C22:1 was lower than NaOH olive oil counterpart, but higher than raw olive oil. We have no explanation for this and similar results are missing in literature. Perhaps, it is reasonable to hypothesise a microbial origin of C22:0 and C22:1 in the olives, by considering that few oleaginous microorganisms are able to synthesize long chain fatty acids (EL BIALY et al., 2011; RATLEDGE, 2004). 3.4. Sensory analysis Graphical presentations of the obtained sensory evaluation (average values) are shown in Fig. 2. The panel did not evaluate color; however, NaOH-olives were overall lightly yellow- green, while Nat-olives were dark gray-green. In general, sensory profile of NaOH-olives was very different from that of Nat-olives; these differences were amplified during the second storage. In general, ‘overall opinion’ was more positive for Nat- than NaOH- olive samples. The B1-NaOH obtained an ‘overall opinion’ of 4 points convincing us to continue the storage work for this sample despite its high pH (Table 2). However, at the end of study, B2-NaOH and T-NaOH showed perceptible sensory defects and loss of firmness, fibrous and crunchy, whereas C-NaOH was not evaluated due to its pronounced smell of rotten caused by an abnormal fermentation. In addition, salty was the prevailing, if not unique, taste in the tasted NaOH olives. Conversely, taste of Nat-olives was more complex and well-balanced between salty, bitter and acid tastes. Salty in the brined B1-Nat and B2- Nat was imperceptible, while bitter taste was perceived clearly in all Nat-olives. Effectively, bitter taste is a distinctive but pleasant flavor of the natural-style olives (LANZA and AMORUSO, 2016). The intensity of the bitter taste remained essentially unchanged between the B1-Nat, C-Nat and T-Nat samples, while it decreased in B2-Nat. Finally, a positive fruit odor was perceived clearly up to 18 storage months only in the Nat-olives. 3.5. Microbiological analysis Fermentation of both NaOH- and Nat-olives occurred spontaneously without adding any starter culture. The microbial count of raw olives highlighted presence of yeasts (log 7.3 CFU/g), LAB (log 2.0 CFU/g) and Enterobacteriaceae (log 2.0 CFU/g). Heperkan (2013) reports that microbiota of olives include principally yeasts and lactic acid bacteria (LAB), members of Enterobacteriaceae, Clostridium, Pseudomonas, Staphylococcus, and Ital. J. Food Sci., vol. 30, 2018 - 423 occasionally moulds. The processing method impacts microbial dynamics affecting greatly quality and shelf life of the TO (DE ANGELIS et al., 2015). Generally, indirectly brined green olives yeasts become the predominant population; however, a correct adding of LAB starter may improve lactic fermentation performance (CAMPUS et al., 2017; DE LEONARDIS et al., 2016; PERRICONE et al., 2013; CORSETTI et al., 2012). Figure 2. Sensory profile rings of the table olives. The mean of microbial count and standard deviation of the microorganisms searched in the olive pulp are given in Table 5. The most relevant result was absence of detectable microorganisms in B1-Nat and B2-Nat, apart from yeast cells in B1-Nat (7.1 log CFU/g); Ital. J. Food Sci., vol. 30, 2018 - 424 conversely, in all other samples LAB and yeasts coexisted until the end of study (Table 5). In addition, neither Enterobacteriaceae nor Pseudomonas spp. were found in B2-Nat, C-Nat e T-Nat. Therefore, low pH (Table 2) and high phenol level (Table 3) influenced the microbial profile of B1-Nat affecting positively the subsequent storage of the Nat-olives. Generally, the Enterobacteriaceae spp., eliminated during fermentation, are not detected at the end of the process (HEPERKAN, 2013). Nevertheless, Enterobacteriaceae cells were found in all NaOH-olives ranging from 3.3 to 4.8 log CFU/g (Table 5). Certainly, in these samples, high pH values and lack of phenols have favoured the development of Enterobacteriaceae and Pseudomonas spp., which have contributed to the spoilage of B2- NaOH, C-NaOH and T-NaOH samples. Moreover, in C-NaOH sample, Enterobacteriaceae and Pseudomonas spp. were not inhibited at cold temperature. Indeed, low temperature storage has penalized the populations of yeast and LAB favouring the growth of psychrophilic bacteria, which caused lipolysis, oil oxidation (Table 4), pectolytic action and, putrefaction (Fig. 2). Moreover, also for the Nat-olives, the conservation at cold temperature was proven to be less effective than that in brine and under temperate conditions. Table 5. Cell count (log CFU/g) of microorganisms determined on pulp olives. Media YPD MRS PCA (at 30°C) PCA (at 6°C) VRBGA PCFC NaOH processed olives B1-NaOH 6.20 (0.10)a 4.90 (0.10)a 6.10 (0.20)a 6.13 (0.12)a 3.50 (0.20)a 4.10 (0.20)a B2-NaOH 5.47 (0.15)c 4.00 (0.20)b 5.00 (0.10)b,c 4.90 (0.20)b 4.80 (0.20)b 0.00 (0.00) C-NaOH 6.13 (0.15)a 5.47 (0.15)c 6.10 (0.30)a 5.07 (0.15)b 3.30 (0.20)a 3.80 (0.20)b T-NaOH 5.90 (0.20)a 4.67 (0.25)a 5.70 (0.20)d 5.60 (0.20)d 4.03 (0.15)c 4.60 (0.30)c Nat processed olives B1-Nat 7.10 (0.20)b n.d. n.d. n.d. n.d. n.d. B2-Nat n.d. n.d. n.d. n.d. n.d. n.d. C-Nat 4.60 (0.10)d 4.83 (0.59)a 4.70 (0.10)c 4.00 (0.10)c n.d. n.d. T-Nat 5.17 (0.35)c 4.50 (0.40)a 5.20 (0.40)b n.d. n.d. n.d. Values are means (standard deviation) of three independent replicates; letters on the column point out significant difference at p<0.05. n.d. = not detected. 4. CONCLUSIONS After 18 months of storage, natural green olives showed good nutritional features (hydroxytyrosol, unchanged fatty acid profile), organoleptic identity, microbial safety, low oil hydrolysis and oxidation. Conversely, after 9 months of storage, pH values below the requirements for the trade were obtained in the NaOH-treated olives. Natural olives preserved high total phenols (0.2 g/100 g wet pulp) and a significant level of the Hy- compounds determined in this study. It is reasonable to suppose that residual phenols have influenced positively polyunsaturated fatty acid preservation and, together with the low pH level, have inhibited the growth of Enterobacteriaceae and Pseudomonas spp. Therefore, the slow olive debittering of natural olives was counterbalanced by a prolonged shelf life. The packaging in 6% NaCl renewed brine or in vacuum bag, under a storing Ital. J. Food Sci., vol. 30, 2018 - 425 temperature of 20°C, gave the best results, while conservation at cold temperature proved to favor the growth of psychrophilic spoilage bacteria. ACKNOWLEDGEMENTS The authors would like to thank Margherita Di Cristofaro for the valuable contribution in editing the manuscript. REFERENCES Ambra R., Natella F., Bello C., Lucchetti S., Forte V. and Pastore G. (2017). Phenolics fate in table olives (Olea europaea L. cv. Nocellara del Belice) debittered using the Spanish and Castelvetrano methods. Food Research International 100, 369- 376. American Oil Chemists' Society (AOCS) (1998). Official methods and recommended practices of the American Oil Chemists' Society. D. Firestone (Ed.), AOCS press. Aponte M., Blaiotta G., La Croce F., Mazzaglia A., Farina V., Settanni L. and Moschetti G. 2012. Use of selected autochthonous lactic acid bacteria for Spanish-style table olive fermentation. Food Microbiology 30(1):8-16. Aponte M., Ventorino V., Blaiotta G., Volpe G., Farina V., Avellone G., Lanza C.M. and Moschetti G. 2010. Study of green Sicilian table olive fermentations through microbiological, chemical and sensory analyses. Food Microbiology 27(1):62- 170. Bevilacqua A., Perricone M., Cannarsi M., Corbo M.R. and Sinigaglia M. 2009. Technological and spoiling characteristics of the yeast microflora isolated from Bella di Cerignola table olives. International Food Science Technology 44(11):2198- 2207. Blekas G., Vassilakis C., Harizanis C., Tsimidou M. and Boskou D.G. 2002. Biophenols in table olives. Journal of Agricultural Food Chemistry 50(13):3688-3692. Bleve G., Tufariello M., Durante M., Grieco F., Ramires F.A., Mita G., Tasioula-Margari M. and Logrieco A.F. 2015. Physico-chemical characterization of natural fermentation process of Conservolea and Kalamata table olives and developement of a protocol for the pre-selection of fermentation starters. Food Microbiology 46:368-382. Campus M., Cauli E., Scano E., Piras F., Comunian R., Paba A., Daga E., Di Salvo R., Sedda P., Angioni A. and Zurru R. 2017. Towards controlled fermentation of table olives: lab starter driven process in an automatic pilot processing plant. Food and Bioprocess Technology 10(6):1063-1073. Caponio F., Pasqualone A., and Gomes T. 2003. Changes in the fatty acid composition of vegetable oils in model doughs submitted to conventional or microwave heating. International Journal of Food Science & Technology, 38: 481-486. Charoenprasert S. and Mitchell A. 2012. Factors influencing phenolic compounds in table olives (Olea europaea). Journal Agricultural Food Chemistry 60(29):7081-7095. Corsetti A., Perpetuini G., Schirone M., Tofalo R. and Suzzi G. 2012. Application of starter culture to table olive fermentation: an overview on the experimental studies. Frontiers Microbiology 248:1-6. D’Antuono I., Garbetta A., Ciasca B., Linsalata V., Minervini F., Lattanzio V.M., Logrieco A.F. and Cardinali A. 2016. Biophenols from table olive cv Bella di Cerignola: chemical characterization, bioaccessibility, and intestinal absorption. Journal of Agricultural and Food Chemistry 64(28):5671-5678. De Angelis M., Campanella D., Cosmai L., Summo C., Rizzello C.G. and Caponio F. 2015. Microbiota and metabolome of un-started and started Greek-type fermentation of Bella di Cerignola table olives. Food Microbiology 52:18-30. De Leonardis A. and Macciola V. 2012. Heat-oxidation stability of palm oil blended with extra virgin olive oil. Food Chemistry 135(3):1769-1776. De Leonardis A., Angelico R., Macciola V. and Ceglie A. 2013. Effects of polyphenol enzymatic-oxidation on the oxidative stability of virgin olive oil. Food Research International 54(2):2001-2007. De Leonardis A., Macciola V., Cuomo F. and Lopez F. 2015. Evidence of oleuropein degradation by olive leaf protein extract. Food Chemistry 175:568-574. De Leonardis A., Pizzella L. and Macciola V. 2008. Evaluation of chlorogenic acid and its metabolites as potential antioxidants for fish oils. European Journal of Lipid Science Technology 110(10):941-948. Ital. J. Food Sci., vol. 30, 2018 - 426 De Leonardis A., Testa B., Macciola V., Lombardi S.J. and Iorizzo M. 2016. Exploring enzyme and microbial technology for the preparation of green table olives. European Food Research Technology 242(3):363-370. El Bialy H., Gomaa O.M. and Azab K.S. 2011. Conversion of oil waste to valuable fatty acids using oleaginous yeast. World Journal Microbiology Biotechnology 27(12):2791-2798. Ercolini D., Villani F., Aponte M. and Mauriello G. 2006. Fluorescence in situ hybridization detection of Lactobacillus plantarum group on olives to be used in natural fermentations. International Journal Food Microbiology 112(3):291-296. European Union Commission. Commission Regulation (EU) No 432/2012 establishing a list of permitted health claims made on foods, other than those referring to the reduction of disease risk and to childrens development and health. Official Journal of European Communities, L. 50, 2012. European Union Commission. Regulation EEC/2568/91 on the characteristics of olive and olive pomace oils and their analytical methods. Official Journal of European Communities, L. 248, 1991. Galán-Soldevilla H. and Pérez-Cacho P.R. 2012. Panel training programme for the Protected Designation of Origin “Aceituna Aloreña de Malaga. Grasas y Aceites 63(1):109-117. Ghanbari R., Anwar F., Alkharfy K.M., Gilani A.H. and Saari N. 2012. Valuable nutrients and functional bioactives in different parts of olive (Olea europaea L.). International Journal Molecular Sciences 13(3):3291-3340. Heperkan D. 2013. Microbiota of table olive fermentations and criteria of selection for their use as starters. Frontiers Microbiology 4(143). Hurtado A., Reguant C., Bordons A. and Rozès N. 2010. Evaluation of a single and combined inoculation of a Lactobacillus pentosus starter for processing cv. Arbequina natural green olives. Food Microbiology 27(6):731-740. International Olive Oil Council (IOOC) 2004. Trade standard applying to table olives COI/OT/NC no. 1, December. International Olive Oil Council (IOOC) Internet: http://www.internationaloliveoil.org/estaticos/view/132-world-table- olive-figures (accessed April 4, 2017). Iorizzo M., Lombardi S.J., Macciola V., Testa B., Lustrato G., Lopez F. and De Leonardis A. 2016. Technological potential of Lactobacillus strains isolated from fermented green olives: In vitro studies with emphasis on oleuropein-degrading capability. The Scientific World Journal 2016, 1917592. Lanza B. and Amoruso F. 2016. Sensory analysis of natural table olives: Relationship between appearance of defect and gustatory-kinaesthetic sensation changes. LWT Food Science Technology 68:365-372. López A., Cortés A. and Garrido A. 2011. Chemometrics characterization of the fats released during the conditioning processes of table olives. Food Chemistry 26(4):1620-1628. Malheiro R., Casal S., Sousa A., de Pinho P.G., Peres A.M., Dias L.G., Bento A. and Pereira J.A. 2012. Effect of cultivar on sensory characteristics, chemical composition, and nutritional value of stoned green table olives. Food and Bioprocess Technology 5(5):1733-1742. Marsilio V., Seghetti L., Iannucci E., Russi F., Lanza B. and Felicioni M. 2005. Use of a lactic acid bacteria starter culture during green olive (Olea europaea L cv Ascolana tenera) processing. Journal Science Food Agriculture 85(7):1084-1090. Medina E., Brenes M., Romero C., García A. and de Castro A. 2007. Main antimicrobial compounds in table olives. Journal of Agricultural Food Chemistry 55:9817-9823. Pasqualone A., Nasti R., Montemurro C. and Gomes T. 2014. Effect of natural-style processing on the oxidative and hydrolytic degradation of the lipid fraction of table olives. Food Control 37:99-103. Pereira A.P., Pereira J.A., Bento A. and Estevinho M.L. 2008. Microbiological characterization of table olives commercialized in Portugal in respect to safety aspects. Food Chemistry Toxicology 46(8):2895-2902. Perricone M., Corbo M.R., Sinigaglia M. and Bevilacqua A. 2013. Use of starter cultures in olives: a not-correct use could cause a delay of performances. Food Nutrition Science 4(7):721-726. Ramírez E., Brenes M., de Castro A., Romero C. and Medina E. 2017a. Oleuropein hydrolysis by lactic acid bacteria in natural green olives. LWT-Food Science Technology 78:165-171. Ramírez E., Brenes M., García P., Medina E. and Romero C. 2016. Oleuropein hydrolysis in natural green olives: importance of the endogenous enzymes. Food Chemistry 206: 204-209. Ital. J. Food Sci., vol. 30, 2018 - 427 Ramírez E., Medina E., García P., Brenes M. and Romero C. 2017b. Optimization of the natural debittering of table olives. LWT-Food Science Technology 77:308-313. Ratledge C. 2004. Fatty acid biosynthesis in microorganisms being used for single cell oil production. Biochimie 86(11):807-815. Rozin P., Spranca M., Krieger Z., Neuhaus R., Surillo D., Swerdlin A. and Wood K. 2004. Preference for natural: instrumental and ideational/moral motivations, and the contrast between foods and medicines. Appetite, 43(2):147-154. Sakouhi F., Harrabi S., Absalon C., Sbei K., Boukhchina S. and Kallel H. 2008. α-Tocopherol and fatty acids contents of some Tunisian table olives (Olea europea L.): changes in their composition during ripening and processing. Food Chemistry 108(3):833-839. Sánchez A.H., García P. and Rejano L. 2006. Trends in table olive production. Elaboration of table olives. Grasas y Aceites 57:86-94. Servili M., Settanni L., Veneziani G., Esposto S., Massitti O., Taticchi A., Urbani S., Montedoro G.F. and Corsetti A. 2006. The use of Lactobacillus pentosus 1MO to shorten the debittering process time of black table olives (cv. Itrana and Leccino): a pilot-scale application. Journal Agricultural Food Chemistry 54(11):3869-3875. Tamer C.E., İncedayı B., Yıldız B. and Çopur Ö.U. 2013. The use of vacuum impregnation for debittering green olives. Food and Bioprocess Technology 6(12):3604-3612. Paper Received October 20, 2017 Accepted February 6, 2018