2.1 Preparation of Pomegranate Waste

Pomegranate (Punica granatum L., cv. Hicaz nar) peels were obtained from fruit processing industry in Turkey and stored at 4°C. The peels were lyophilised in a freeze dryer (VirTis Ultra Pilot Lyophilizer with a Wizard 2.0 control system, SP Industries, USA) by freezing at –30°C for 5 h and drying under 10 Pa pressure at 20°C for 24 h. The freeze-dried peels were ground to powder (No. 48 sieve) using a grinder and stored at 4°C until use.

2.2 Chemicals

Ellagic acid, punicalagins (A and B forms) and gallic acid standards were purchased from Fluka^TM (USA). Malic acid and citric acid standards were obtained from Merck KGaA (Darmstadt, Germany). Ethanol (99% v/v), methanol (99% v/v), acetone (99% v/v), and hydrochloric acid (HCl; 37% v/v) were used as solvents (Merck).

2.3 Preparation of Pomegranate Peel Extracts

5 g of pomegranate peel powder was mixed with 100 ml of different solvents and incubated at 50°C for 30 min or 2 h in an ultrasonic water bath (150 W, 40 KHz) (E1: ethanol for 30 min, E2: ethanol for 120 min, EA: ethanol with HCl, M: methanol, MA: methanol with HCl, A: acetone and W: distilled water, for 30 min, Table I). The mixtures were centrifuged at 4°C at 4000 rpm for 15 min as described by Zhang et al. (15) and Türkyýlmaz et al. (14) with slight modifications (Scheme I). The supernatants were filtered with Whatman^® Grade 1 paper and then evaporated in a rotary evaporator (Heidolph Instruments, Germany) at 50°C under a vacuum of 400 mbar. The extracts were stored at 4°C until used for further studies.

2.4 Qualitative Analysis of Phenolics and Sugars

Sugar and phenolic contents of PPEs were screened by thin layer chromatography (TLC). About 1 μl of each extract and standards were applied on TLC plate. For screening of phenolics, the plate was run into ethyl acetate:glacial acetic acid:formic acid:distilled water (100:11:11:5, v/v; adapted from Kumar et al. (16)). The plates were then sprayed with 5% w/v ferric chloride reagent (13). For sugars, the plate was run into acetonitrile:water (85:15, v/v) solvent system and then stained with α-naphthol (0.5% w/v) dissolved in ethanol solution acidified with H₂SO₄ (5% v/v), followed by heating at 110°C for 10 min (17). The colours of the spots were identified. An individual retention factor (R_f) value for each spot was measured and compared with standard reference sugars and phenolic compounds run in the same respective solvent systems.

2.5 Quantitative Analysis of Phenolics, Sugars and Organic Acids

Detection and quantification of phenolics, organic acids and sugars were carried out by high-performance liquid chromatography (HPLC) system. Each sample was centrifuged and then filtered through 0.22 μm membrane filter before HPLC analysis.

Ellagic acid and punicalagin separations were achieved at 30°C on a C18 column (150 mm × 4.6 mm, 5 μm, GL Sciences Inc, Japan). HPLC analysis was performed using Class VP, 20 AD series (Shimadzu Corporation, Japan) equipped with photodiode-array detector (PDA) and an autosampler. The mobile phase consisted of formic acid (1%) and acetonitrile with gradient mode elution (0–18 min, 15% v/v acetonitrile, 20 min 65% v/v acetonitrile, 25 min 5% v/v acetonitrile and 30 min 5%, v/v acetonitrile) at a flow rate of 1 ml min^–1. The injection volume was 10 μl. The quantitation wavelength was set at 255 nm (18). 1000 μg ml^–1 of ellagic acid, gallic acid and punicalagins were prepared by dissolving in 5 ml of HPLC grade methanol for standards. The solutions were stored at –20°C. The calibration curves were established from the standards of punicalagins and ellagic acid at concentrations between 0.005–0.02% and 0.25–1%, respectively.

Sugars (glucose and sucrose) were determined by using NH₂ column (250 mm × 4.6 mm, 5 μm, GL Sciences Inc). The column temperature was 25°C. The eluted samples were detected by refractive index detector (RID). The mobile phase consisted of acetonitrile (60%) and ultra-pure water (40%). Flow rate was 1 ml min^–1 and injection volume was 20 μl (19). The standards of glucose and sucrose (5 mg ml^–1, 10 mg ml^–1, 20 mg ml^–1, 40 mg ml^–1 and 50 mg ml^–1) were used for calibration curves.

Organic acids (citric and malic acids) were determined with an ultraviolet-visible (UV-vis) detector (Shimadzu SPD-10A VP, Shimadzu Corporation). All organic acid analyses were carried out with a Kromasil^® C18 HPLC column (5 μm, 4.6 mm × 156 mm). The mobile phase was prepared by using 0.005 N sulfuric acid. Injection volume was 50 μl and the column temperature was 25°C. Flow rate was 0.3 ml min^–1. The data were recorded at 210 nm (20). Citric and malic acid standards (0.1–2% w/v) were used to prepare calibration curves.

2.6 Antimicrobial Analysis

The antimicrobial efficacies of PPEs were evaluated against E. coli (ATCC^® 25922), E. faecalis (ATCC^® 29212), B. subtilis (ATCC^® 6633), B. cereus (ATCC^® 11778), Pseudomonas aeruginosa (ATCC^® 27853), Streptococcus uberis (ATCC^® 700407), E. aerogenes (ATCC^® 13048) and Candida albicans (ATCC^® 10231) by using agar well diffusion method (Scheme I). Each microbial culture was incubated in Mueller Hinton Broth (MHB, Merck) overnight. The OD₆₀₀ values of cultures were adjusted to 0.1 and then 100 μl of each microbial culture was spread on petri dishes containing 17 ml of Mueller Hinton Agar (MHA, Merck). The media were punched with 7 mm diameter wells and these wells were filled with 40 μl of each extract. The plates were then incubated for 24 h at 37°C. After incubation, inhibition zones for microorganisms for each extract were measured in millimetres. Each extract was tested three times.

2.7 pH Measurements

The pH levels of extracts were measured by a pH meter (HI-2211 Bench Top pH & mV Meter, Hanna Instruments Ltd, UK).

2.8 Statistical Analysis

Values are expressed as mean ± standard deviation. Data were analysed by student t-test. The level of statistical significance was accepted as p<0.05.

3.1 Phenolics

Punicalagins (punicalagin A and B) and ellagic acid were identified in all extracts by TLC. However, gallic acid was not determined in all extracts according to both TLC and HPLC analyses.

The highest levels of punicalagins and ellagic acid were detected in the EA extract as 13.86% and 17.19% respectively, whereas the lowest levels were obtained with W extract (p<0.05) (Figure 1) by HPLC analyses. This is consistent with previous results which reported that the solvents could affect the phenolic contents of plant extracts (21, 22). The difference in phenolic contents of extracts depends on the solvent polarity that affects solubility of selected groups found in antimicrobial bioactive compounds (8). Water, ethanol and methanol are polar solvents while acetone is an intermediate polar solvent that can dissolve both polar compounds including phenolics and nonpolar compounds. In addition, the extracts obtained by mixture of solvents (combination of acid and ethanol or methanol) could contain more radical scavenger than the pure solvents (23) by changing polarity that affects antimicrobial activities. It was also observed that ethanol alone or ethanol-acid combination could be more effective than other solvents to obtain high levels of phenolic compounds. The antimicrobial activities of phenolics (ellagic acid and punicalagins) were shown previously (24, 25). These polyphenols found in PPEs can work as antimicrobial agents by forming a complex with the bacterial cell to cause death or by inhibiting protein activities. The position and the number of hydroxyl groups on the phenolic components may also increase this inhibitory effect on the microorganisms (26, 27).

3.2 Sugars

The highest total sugar contents (glucose and sucrose) were obtained with MA (5.95%), E2 (4.98%) and EA (4.93%) extracts (p<0.05) (Table II). MA and EA extracts exhibited the highest antimicrobial activities (Figure 2). Sugars might help the antimicrobial efficacies of these extracts due to the osmotic effects of carbohydrates on microorganisms. However, there was no clear consistency for the sugar contents and antimicrobial activities of other extracts.

3.3 Organic Acids

The organic acid contents of PPEs determined by HPLC are presented in Table II. MA extract had the highest malic acid (3.19%) content whereas the highest citric acid level (1.13%) was detected with EA extract (p<0.05) (Table II). Organic acids could affect integrity of the cell membrane, activities of enzymes or biosynthesis of macromolecules and cellular homeostasis (28, 29).

3.4 pH Values

The lowest pH level (0.26) was measured with MA extract while the highest pH (3.55) value was determined with W extract as expected (Table II). The highest inhibition zones for all tested microorganisms were determined with MA and EA extracts with the low level of pH (almost zero) that might also affect microbial growth.

3.5 Antimicrobial Activities of Pomegranate Peel Extracts

The antimicrobial effects of different extracts obtained by using different solvents were evaluated against food associated microorganisms. The antimicrobial activities were assessed by the presence or absence of inhibition zones and zone diameters. The results are given in Figure 2. The data of the study showed that MA and EA extracts of pomegranate peels had the highest antibacterial activities against all tested microorganisms (p<0.05) (Figure 2(b)). The inhibition zone diameters were found to be 17–36 mm with MA extract (including the diameter of the wells). EA extract, the second most efficient extract, resulted in 17–32 mm inhibition zones (Figure 2(a)). The extracts E1, E2, M and A showed similar antimicrobial effects (Figure 2). Increasing extraction time from 30 min to 120 min for E2 extract did not enhance (p>0.05) antimicrobial efficacies almost in most cases. The lowest inhibitions for all microorganisms were detected with W extract in the range of 0–15 mm (p<0.05) (Figure 2(a)).

In general, the antimicrobial effects of extracts could be attributed to their phenolics (30) and organic acid contents (31, 32). Therefore, according to results obtained in this work, the highest levels of organic acids (MA and EA) and phenolics (EA) could contribute antimicrobial activities of these extracts.

In previous works, Gram positive bacteria were more sensitive to plant extracts than Gram negative ones (33, 34) consistent with the results obtained from this study. It was found that B. subtilis, B. cereus, E. faecalis and S. uberis (Gram positive) strains were more sensitive than E. coli and E. aerogenes (Gram negative) strains (Figure 2). The cell walls of Gram positive bacteria were shown as more sensitive to antimicrobial compounds compared with Gram negative bacteria (35, 36) due to the lipopolysaccharide layer and periplasmic space present in Gram negative bacterial cell walls. However, Hama et al. (37) found that pomegranate juice had antibacterial activity on both Gram positive (S. aureus) and Gram negative bacteria including E. coli and P. aeruginosa. The multi-layered peptidoglycan was shown as the main factor for antimicrobial resistance (37). According to results obtained in this work, the antibacterial activities of extracts were similar for P. aeruginosa (Gram negative) and S. uberis (Gram positive) strains. The reason for this is not known, but might be related with some specific properties of these microorganisms or extraction methods. Therefore, no clear correlation was found between the cell wall structures and the antibacterial activities of the extracts. Previously, methanol extracts of peel showed the greatest activities on different bacteria depending on the pomegranate variety tested (38, 39). It was indicated that ethanol extracts of pomegranate had hydrolysable tannins including punicalagins, ellagic acid and gallic acid (30). In this work, punicalagins and ellagic acids were found in all extracts but gallic acid was not determined. That may be linked to the variety of the pomegranate or extraction techniques.

In the present study, the inhibition zones for C. albicans were between 8–25 mm in diameter. The highest inhibition zone was obtained with MA extract (25 mm). It was reported that the methanolic PPEs inhibited C. albicans with the inhibition zones of 6–6.5 mm (12) which were much lower than found in this study. Punicalagins were shown as antifungal components of ethanol extract of pomegranate peels. C. albicans treated with punicalagins extracted from pomegranate peels exhibited morphological alterations in cell structure and abnormal budding (40). The aqueous extract of pomegranate peels also showed inhibitory activity on C. albicans (41) which was consistent with the results obtained with this study.