Background

Olea europaea L. is a small tree belonging to the family Oleaceae, commonly known as olive in English and Zaitoon in Arabic. O. europaea produces edible fruits, the main source of olive oil; olive fruit and its oil have a long history of nutritional and medicinal values. Olive oil is the major dietary fat of the traditional healthy “Mediterranean diet”, which has been strongly associated with reduced prevalence of cardiovascular diseases and certain types of cancer [38]. The tree has been cultivated for centuries in tropical and warm temperate regions, particularly in the Mediterranean. Spain, Italy, and Tunisia produce over 50% of the global production of olives, with a production volume of approximately 12.5 million metric tons in 2020 [22]. The quality and chemical composition of olive fruit and oil are affected by several factors, including agricultural practices, cultivars, genetics, and seasonal and environmental factors [19]. Moreover, the storage and transporting conditions could significantly impact olive products' quality, taste, and health benefits. The health uses of olive fruit and leaves have been reported to include respiratory tract, urinary tract infections, and GIT diseases, while the oil is applied to the scalp to prevent hair loss and to fracture limbs [11, 28, 36]. Apart from folk use, the fruit and leaves have recently been reported for anti-oxidant effects with potential skin benefits [15].

The health benefits of olive oil, as well as its unique flavor and taste, are principally attributed to the occurrence of high amounts of monounsaturated fatty acids (MUFAs). These MUFAs are oleic acid (18:1) and palmitoleic acid (16:1) and functional bio-actives, including phenolics, tocopherols, phospholipids, and carotenoids with multiple biological properties [23]. The biological properties of olive oil and its components have been suggested due to its anti-oxidant, anti-cancerous, and antimicrobial activities and the modulation of gene functions [40]. Olive leaf and seed extracts have shown considerable activity against bacteria, fungi, and viruses in several in vitro and in vivo studies [12]. Virgin olive oil has been reported to reduce the count of inoculated Salmonella enteritidis and Listeria monocytogenes by approximately 1000 CFU/g in salad and mayonnaises, suggesting a strong protective effect against foodborne pathogens [37].

Similarly, olive oil has shown strong bactericidal activity against several Helicobacter pylori strains, including three antibiotic-resistant strains [42]. Olive oil affects cancer development through biological roles [40]. The constituents of olive oil, e.g., oleic acid and another hydrocarbon known as squalene, are essential in inhibiting cancerous cell growth [49]. The phenolic constituents, including oleuropein and hydroxytyrosol, have exhibited cancer cell inhibition effects in several studies and strong anti-oxidant activity [49]. The cytotoxic effect of olive oil constituents has been demonstrated against different cell lines, including breast, prostate, and colorectal cancerous cell lines [40]. The most popular tool to determine the amount of phytochemical profile of samples is gas chromatography–mass spectrometry (GC–MS). This analytical tool has been widely employed to study the fatty acid composition of olive oil for quality assessment and authentication purposes. Variations between olive oil from different origins have been reported based on the fatty acid content. In one study, the amount of oleic acid (18:1), a monounsaturated fatty acid, was found in different ratios in Syria, Greek, and commercial oils, with the highest amount present in Syrian oil [3].

Moreover, Syrian oil contained the highest amount of polyunsaturated fatty acid linolenic acid (18:3), followed by Greek oil. Overall, Syrian oil contained the highest content of unsaturated fatty acids compared to Greek and commercial oils. This shows an enormous variation of the phytochemical profile and its composition in O. europaea samples based on geographical origin.

The objective of the study is to evaluate the quality of an enormous number of olive samples (42) from different origins in terms of phytochemical profile using GCMS, as well as a comparative biological evaluation of the cytotoxicity and antimicrobial activity in correlation to the volatile chemical markers observed.

Material and methods

Olive (Olea europaea) fruit samples

The 42 seedless fresh-olive fruit (OF) samples from different geographical origins, as reported in our previous study, were collected during the year 2021 at local markets and properly coded; Egyptian (O1–O3), Greek (O–O9), Jordan (O10–O19), Moroccan (O20–O27), Palestinian (O28), Spanish (O29–O32), Syrian (O33–O39), and Turkish (O40–O42) samples [6].

Preparation of OF samples

Liquid–liquid extraction was performed where the samples (1 g of the methanolic extracts) were dissolved and partitioned in n-hexane solvent (20 mL). The LLE process was repeated 3 times, and the n-hexane solvent was collected and evaporated. The final extract was dissolved in 10 mL GCMS grade solvent for a final concentration of 1 mg/mL of the extract. The n-hexane extracted samples were filtered (0.2 μm), diluted (5 ppm), and subjected to GCMS. The GC–MS chemical study for qualitative and quantitative analysis of the phytochemical profile is performed and reported herein.

GC–MS profiling for volatile compounds

The GC separation was performed using a Shimadzu-2010 plus gas chromatograph equipped with a split/splitless injector and coupled with QP2010 Ultra MS detector. The column used for the application was a nonpolar Rxi-5MS capillary column (30 m × 0.25 mm, 1.00 μm; Restek Corporation); whereas, helium was used as a carrier gas (1.5 mL/min). For GCMS analysis, the oven was initially maintained at 50 °C for 1 min, ramped to 150 °C at a rate of 5 °C/min (held for 1 min), and finally ramped to 280 °C at a rate of 10 °C/min (held for 5 min). The temperatures for the ion source and MS transfer lines were maintained at 250 and 280 °C, respectively. Mass spectra (33–450 m/z) were acquired following a 6.5-min solvent delay. Shimadzu GCMS Solution® (V. 4.52) was used for data acquisition, processing, and GC–MS control. The area normalization method (%content) was used for semi-quantification [48]; whereas, compound identification was achieved by mass spectral searching within the NIST11 Library database.

Cell lines and culture conditions

Cancer cell lines, MCF7 (human breast adenocarcinoma), and HCT116 (human colorectal carcinoma) were obtained and sub-cultured in RPMI-1640 media (10% FBS) at 37 °C, 5% CO2, and 100% relative humidity.

Cytotoxicity assay

The MTT assay was performed using MCF7 and HCT116 cells to evaluate OF-samples' cytotoxicity. One concentration for each of the 42-OF-samples, i.e., 100 μg/mL (DMSO 0.1%; n = 3), was tested for comparative evaluation of the samples. The positive control used in the study was doxorubicin (5 µM) [1, 2].

Microorganisms and culture media

Bacterial strains: Staphylococcus aureus (ATCC-25923), methicillin-resistant Staphylococcus aureus (MRSA, ATCC-43300), Pseudomonas aeruginosa (ATCC-15442), and Escherichia coli (ATCC-35218); whereas, the culture media: Muller Hinton agar (Oxoid, CM-0337) and Muller Hinton broth (Oxoid, CM-0405) were used for agar diffusion (MIC) and broth dilution method (MBC), respectively.

Preparation of the standard inoculum

The tested organisms were grown on Muller Hinton agar medium (37 °C, 24 h.), and selected colonies were inoculated with the help of a sterile loop in Muller Hinton broth to form a homogenous suspension of bacterial strains. The suspension was standardized (0.5 McFarland turbidity) using calibrated Vitek Densichek Biomerieux Analyzer.

Agar diffusion assay

The assay was performed for individual bacterial strains suspended in Mueller Hinton broth in three directions of the agar plate using sterile cotton swabs (100 μL), as reported [21].

MIC and MBC determination

For MIC and MBC, 200 μL of the selected optimum extracts (O17 and O25) (100 µg/ml) were added to the first column of the 96-well microtiter plates, followed by the addition of 100 μL Mueller Hinton broth to the second column. Individual extract (100 μL) was transferred from the first well into the subsequent wells to produce twofold dilutions (50, 25, 12.5, 6.2, and 3.1 µg/ml). The adjusted McFarland bacterial suspension (10 μL of 0.5) in the Muller Hinton broth of either P. aeruginosa or E. coli was added to each well, and the experiment was repeated three times. The plates were incubated overnight (37 ºC) according to the clinical and laboratory standard institute, and the MIC and MBC were determined (CLSI M26− A, 1998).

Statistical analysis

The results were analyzed using SPSS (statistical package for social science students) software V 22.0 using principal component analysis and k-mean cluster analysis.

Results

GCMS profiling

The GCMS revealed the presence of 111 compounds from various chemical classes with a descending order (largest to smallest) of percentages/chemical class: esters (21.61–60.49%), alcohols (20.73–49.2%), hydrocarbons (3–38.88%), ketones (0.16–3.87%), acids (0.07–2.62%), and aldehydes (0.12–1.47%). Regarding the average (%) of 111 volatile compounds/chemical classes, the 42-OF samples showed a descending pattern of esters (44.62%), alcohols (38.06%), hydrocarbons (15.39%), ketones (0.75%), aldehydes (0.45%), others (0.41), acids (0.27%), and ethers (0.06%) as shown in Additional file 1: Table S1.

Among the 42-OF samples, O34 exhibited the highest (60.49%), whereas O23 revealed the lowest (21.61%) concentration for the ester compounds present in these samples. The major among these esters compounds was 13-methyl-pentadecanoic acid methyl ester, with an average occurrence of 11.9%. The representative chromatograms for OF-GCMS analysis are shown in Fig. 1 for sample Greek stuffed olive with pepper (O-09), Spanish stuffed olive pine green (O-11), Spain black olive (O-13), which shows the (tR) maximum up to 39 min for each sample. At the same time, the peak height is proportional to the concentration of a compound in the sample, which is given in more detail in Additional file 1: Table S1.

Fig. 1
figure 1

Representative GC–MS chromatograms for olive samples

Full size image

Cytotoxic assay

The 42-OF-samples exhibited cytotoxic activity within the range of 8–64% and 0.11–44% against MCF7 and HCT116 cell lines, respectively. Among the 42-OF-samples, the most significant inhibition was observed for O17 and O25 on MCF7 (64% and 52%) and HCT116 (25% and 34%), respectively (Table 1).

Table 1 Cytotoxic inhibition activity of the 42 extracts against two cell lines (MTT 72 h, % ± SD)
Full size table

Antimicrobial activity

The antimicrobial activity for 42-OF-samples revealed significant zones of inhibition (mm) against the tested strains: P. aeruginosa (0.00–17.00), E. coli (0.00–15.00), S. aureus (0.00–13), and resistant S. aureus, i.e., MRSA (0.00–12.00). The maximum zones of inhibition (mm ± SD) were observed for O8 and O39 as E. coli (12 ± 1.0) for O8 and (15 ± 1.1) for O39; P. aeruginosa (17 ± 1.1) for O8 and (13 ± 1.0) for O39. These two extracts did not exhibit any activity against the resistant S. aureus (MRSA), while significant activity was observed against the sensitive S. aureus (25,923) as: (11 ± 1.0) for O8 and (12 ± 1.1) for O39. Likewise, O42 showed a zone of inhibition against P. aeruginosa (14 ± 1.0), S. aureus, i.e., MRSA (12 ± 1.1), and S. aureus (13 ± 1.0). Except for O41 and 42, none of the OF samples revealed any activity against MRSA strain, as shown in Table 2. Further evaluation of the OF-samples with optimum activity, i.e., O39 and O8 revealed an identical MIC and MBC of 12.5 and 25 mg/ml against P. aeruginosa. In contrast, a MIC (50 and 25) and MBC (100 and 50) against E. coli were determined for O39 and O8, as shown in Table 3.

Table 2 Evaluation of the antimicrobial activity of the 42 olive variety extracts against four reference bacterial strains using agar diffusion assay
Full size table
Table 3 Assessment of MIC (µg/ml) and MBC (µg/ml) of the selected extracts against two bacterial strains
Full size table

Statistical analysis

SPSS V 22.0 was used to classify and analyze the data.

Principal component analysis (PCA)

To evaluate the variability and correlation of the components in the extracted data, PCA was applied where the Eigenvalue was applied to classify the data. The PCA for 42-OF samples suggested five components (PC1−PC5) with a cumulative %variance of 94.16 and an individual %variance for each component: PC1 (36.27), PC2 (31.61), PC3 (17.03), PC4 (6.50), and PC5 (2.72). The major variability of 36% was exhibited by PC1 loaded with OF-samples: O2, O5, O7, O8, O10, O13, O15, O17, O20, O24, O26, O29–32, O34, O36–39, and O40−42. This implicates the presence of major amounts of volatile compounds in these groups. The next major variability of 31.61% as represented by PC2 consisted of the OF-samples: O1, O3–4, O6, O9, O14, O16, O22, O25, O28, and O35. The samples with more potent cytotoxicity and antimicrobial activities, i.e., O25, O39 and O42 (PC1), O17 and O8 (PC2) are loaded in these two components where a distribution of an enormous number of volatile components has been observed. The remaining components with %variability and OF samples are briefly presented in Table 4.

Table 4 PCA analysis for GCMS, cytotoxicity and antimicrobial dataset
Full size table

The PCA for cytotoxicity and antimicrobial activity produced two components with a cumulative variability of 50.96% and individual variability of 28.75% (PC1) and 22.21% (PC2). The components loaded in PC1 consisted of antimicrobial activity, which showed more component variation of 28.75% compared to cytotoxicity activities loaded in PC2. The data suggest more variation in OF samples regarding antimicrobial activity due to lack of activity at most instances, particularly against the resistant strain of MRSA. On the contrary, the OF samples revealed a comparatively high activity for cytotoxicity assay, representing a low variation in components (Table 4 and Fig. 2).

Fig. 2
figure 2

Three-dimensional representation for cytotoxicity and antimicrobial assay components

Full size image

K-mean cluster analysis

The K-mean cluster analysis classifies the data into clusters based on the nearest mean among the data set. For GCMS, six clusters with high F-value and significance (P = 0.001) were observed as cluster 1 (2 samples), cluster 2 (1 sample), cluster 3 (1 sample), cluster 4 (15 samples), cluster 5 (1 sample), and cluster 6 (91 samples). The two samples in cluster 1 and one sample in cluster 5 represent the ester chemical class of volatile oils; one sample in cluster 2 represents alcohols, and one in cluster 3 represents hydrocarbons. These samples denote the presence of more amounts (%) of such chemical classes in the 42-OF samples. This may be appropriately explained by the high quantity of esters (44.62%) followed by alcohols (38.06%), and hydrocarbons (15.39%) among the chemical classes reported by GCMS in these 42-OF samples. The cluster with chemical class distribution is shown in Table 5 and Fig. 3.

Table 5 K-mean cluster analysis for GCMS, cytotoxicity and antimicrobial dataset
Full size table
Fig. 3
figure 3

Cluster distribution for the GCMS data of 42-OF-samples

Full size image

Regarding K-mean cluster analysis for cytotoxicity and antimicrobial activity, four clusters were proposed: cluster 1 (2 samples), cluster 2 (4 samples), cluster 3 (21 samples), and cluster 4 (15 samples). As denoted in Fig. 4, cluster 1 represents two extracts (OF-41 and 42) which are the only samples with activity against S. aureus (MRSA). Cluster 2 represents samples with activity against all cell lines and microbial strains except MRSA. These extracts are O8, O23, O25, and O39. The K-mean effectively distinguished the GCMS samples with more amounts of chemical compounds and the samples with more potency against the cell lines and microbial strain studied. The K-mean for cytotoxicity and antimicrobial activity is shown in Table 5 and Fig. 4.

Fig. 4
figure 4

Cluster distribution for cytotoxicity and antimicrobial data of 42-OF-samples

Full size image

The data produced with the current methodology were analyzed with the help of descriptive, PCA, and K-mean cluster analysis, which revealed a significant difference in terms of phytochemical profile as well as a correlation for the biological activities linked with a specific class and volatile chemicals of the phytochemical class in these olive samples. The statistical models fully support the significance and correlation of the data extracted based on the current methodology.

Discussion

Olive oil and fruits are considered healthy because of their nutritional and medicinal properties. Bioactive compounds like triterpenoids and phenolics contribute to their high quality. Olive is known to be produced by over a hundred cultivars with different colors, shapes, and sizes where the quality and oil content of olive may be compromised, prone to various environmental or geographical conditions [5, 33]. For instance, many studies have previously reported the variation in natural products such as black seeds and black pepper varieties [7, 8]. Current research collected 42 available olive fruit (OF) samples of different sizes, colors, and shapes. The selection for these samples was non-specific and generalized, i.e., all the samples available at all the stores, hypermarkets, suppliers, shops, etc., were searched and collected irrespective of any targeted numbers or geographical origin.

GCMS analysis was performed for the extracts where several volatile chemical classes were reported, consisting of acids, alcohols, esters, ethers, hydrocarbons, aldehydes, and ketones. The predominant among these chemical classes was the ester chemical class (44.62%), followed by alcohols (38.06%) and hydrocarbons (15.39%). The predominant volatile compounds in these chemical classes were: 13-methyl-pentadecanoic acid methyl ester (esters), 2, 4-bis (1, 1-dimethyl ethyl) -phenol (alcohols), o-cymene (hydrocarbons), 2-pentadecanone, 6, 10, 14-trimethyl- (ketones), 3-hydroxy- dodecanoic acid (acids), and 4-methyl- benzaldehyde (aldehydes). Although the study of quality variation was reported previously for olive fruit, based on the comparative evaluation of fatty acids [3], this is the first report to evaluate 42-OF samples from different geographical origins based on the phytochemical profile and biological properties of the samples.

As olive oil consumption has been linked to the prevention of cancer [26, 31], these 42-OF-samples were tested for cytotoxic potential against various cell lines. An intra-comparative evaluation was performed for the 42-extracts, using MCF7 and HCT116 cell lines, where the two samples, O17 (Jordan black olive) and O25 (Jordan green balad olive) exhibited the maximum activity with an inhibition of 52% for O17 (Jordan black olive) and 64% for O25 (Jordan green balad olive). The findings in our study corroborate previous reports [9, 25]. The OF extracts in this study exhibited feeble antiproliferative effects against one cell line, showing strong inhibition against the other type. This may be due to the varied genetic context of each cell line, the uncertain response of cell lines against the tested compounds, as well as the use of non-standardized crude extracts against these cell lines by different researchers, which may affect the outcomes of the study [10, 32]. The antiproliferative effects, though, have been attributed to many compounds in most OF samples; 3, 4- (dihydroxyphenyl) -ethanol elenolic acid di-aldehyde and 3, 4- (dihydroxyphenyl) -ethanol elenolic acid ester secoiridoid aglycone has been reported as the most potent cytotoxic compounds at many instances [10]. This again supports the same phytochemical marker of olive to be used for a comparative quality evaluation and standardization of the olive samples. The GCMS analysis herein for the 42 OF samples showed the presence of 17-pentatriacontene in high concentration in O17 (Jordan black olive), whereas 10-octadecenal, trichloroacetic acid, hexadecyl ester, and citronellol epoxide were found in maximum amounts in O25 (Jordan green balad olive) compared to other cultivars. This may be a valuable biomarker for further mechanistic studies on OF.

It is of utmost importance to mention here that a consistent trend of halogenated and sulphonated compounds (chlorinated/fluorinated/sulphurous) have been observed in the phytochemical profile of most of the samples, as reported previously in an enormous amount of literature [4, 14, 20, 27, 34, 35, 39, 41, 43,44,45,46,47] [52]. Though most of these chemical compounds have been reported in these studies, they are not related to the chemistry of natural products. The authors suggest that such volatile components originate from environmental contamination during cultivation or instrumental and solvent contamination during extraction and GCMS analysis. Therefore, such compounds should be carefully evaluated when interpreting the results and selecting chemical markers for the quality control of medicinal plants.

The 42-OF-samples were studied for their antimicrobial potential against various bacterial strains. Similar to cytotoxicity, the general antimicrobial screening showed three OF samples: O8 (Greek olive, Kalamata medium black), O39 (Syrian black, selptin olive), and O42 (Palestine green olive) with significant inhibition of the tested microorganisms. The MICs and MBCs were calculated for the most potent extracts of O8 (Greek olive, Kalamata medium black) and O39 (Syrian black selptin olive), which showed a similar MIC of 12.5 μg/ml against P. aeruginosa. In contrast, MICs of 25 (O8) and 50 (O39) μg/ml with MBCs of 50 (O8) and 100 (O39) were observed against E. coli. A previous study reported a MIC of 0.625–1.25 mg/mL for olive oil against Salmonella typhimurium and Staphylococcus aureus. The other samples presented a lack of inhibitory effect, possibly due to dilute concentrations compared to Guo et al. work [18, 29]. The GCMS profile for these samples (O8 and O39) suggested the presence of high amounts of 2-hexyl-1-decanol, 2-butyl-1-octanol, 3, 7, 11-trimethyl-1-dodecanol, n-nonadecanol, (Z) -7-hexadecenal, (Z) -9-octadecenoic acid, hexyl ester, octacosyl trifluoroacetate, nonadecyl pentafluoropropionate, octatriacontyl trifluoroacetate, and octacosyl heptafluorobutyrate, compared to other cultivars.

Aldehydes might be potential chemicals responsible for antimicrobial effects [16, 30]. The organic acids might also be responsible for antimicrobial effects, though the amount might not be high, while the esters might not be the active ingredients responsible for antimicrobial effects [17, 50, 51]. Phenolic compounds might be another category responsible for antimicrobial effects. However, our GC–MS results did not report these compounds [13, 24].

The statistical analysis using GCMS-PCA proved the presence of high amounts of esters, alcohols, and hydrocarbons with significant quantities of ethers, aldehydes, acids, ketones, and aldehydes. PCA revealed more variation in antimicrobial data for biological activities than cytotoxicity, where the OF samples presented significant antiproliferative effects. The antimicrobial activity was sparse, and only three extracts showed reasonable activity. For the activity against MRSA, only O41 (Syrian green olive) and O42 (Palestine green olive) showed considerable activity, whereas the rest of the samples were ineffective to retard the growth of MRSA. The K-mean cluster analysis of GCMS data and biological activities scrutinized the most effective four extracts of O8 (Greek olive, Kalamata medium black), O23 (Jordan green olive), O25 (Jordan green balad olive), and O39 (Syrian black selptin olive). The MIC and MBC evaluation for the two most effective extracts of O8 (Greek olive, Kalamata medium black) and O39 (Syrian black selptin olive) revealed a low IC50 value against the tested organisms.

The outcomes of this research study with statistical calculations were the variation in the quality of food and herbs, as prone to various factors of environmental and geographical variations, including watering, salinity, humidity, temperature, altitude, processing, packaging, and storage, etc.[7, 8]. Hence, a proper evaluation using instrumental metabolomics analysis is of utmost importance to determine the quality of food and herbal products.

Conclusion

The GCMS analysis for 42-OF samples showed the presence of one hundred and eleven compounds which belong to the chemical classes of acids, alcohols, esters, ethers, aldehydes, hydrocarbons, and ketones. Among the 42-OF samples, O17 (Jordan black olive) and O25 (Jordan green balad olive) exhibited the most potent activity against the cancer human cell lines, whereas O8 (Greek olive Kalamata medium black) and O39 (Syrian black selptin olive) had the highest levels of inhibition against the tested microorganisms. These samples showed the presence of a unique pattern of volatile chemical compounds with a predominant amount of esters which may be considered markers and valuable sources for further mechanistic studies of anticancer and antimicrobial natural products. Olive fruit may be a potential source to prevent microbial growth in food products using Olea-based food preservatives.

Novelty and potential impact of the study

This is the first study comparing the quality variation of 42 olive samples from different geographical locations worldwide. The study aimed to evaluate the food quality of olive used in the Saudi Arabian market, as an enormous proportion of the local inhabitants are using olive and their oil for various food, nutraceutical, health, and therapeutic purposes. It is critical to know the quality of the olive in terms of its phytochemical and phytochemical profile as most of the health and therapeutic properties of olive are imparted due to the presence of these novel and active phytochemicals. This study evaluated the quality variation of olive samples not merely in terms of their phytochemical profile using GCMS; the samples were also subjected to various comparative biological evaluations (cytotoxic and antimicrobial). The purpose was to standardize the quality of olive samples as per WHO guidelines, where the chemical, physical, and biological evaluations of the samples are of utmost necessity. This unique quality variation study, with the help of GCMS and biological evaluation, may be useful for the food and nutraceutical manufacturers for the quality control of marketed products, as well as for the end-user to be aware of the olive product with good quality in terms of active components and health effects. This study is equally important for the researchers in further studies that evaluate the impact of geographical origins and other factors on the quality of olive oil.

Additionally, the study investigated the correlation between phytochemical profile and biological activities wherein specific classes with unique volatile chemicals were responsible for most of the olive activities. Notably, such volatile substances may act as potential markers for researchers to select the most potent olive products and explore their mechanistic properties for food and nutraceutical manufacturers for quality control and standardization purposes of olive products.