Olive oil authentication carries out routinely through chemical analysis by monitoring several components such as sterols, phenols, fatty acids, alcohols etc (1-6). However, several difficulties have been encountered in distinguishing pure olive oil from its admixture with other vegetable oils especially with hazelnut. Hazelnut oil provides characteristics strongly similar to olive oil such as similar fatty acids, triglyceride, sterol and tocopherol components. These components do not differ sufficiently to readily distinguish olive oil from hazelnut oil even before mixing. On the other hand, the determination of stigmastadiene produced by the dehydration of sterols during the bleaching of oils can enable the detection of refined oils in unrefined olive oil (IOC method) but it is not always successfully assessed. Recently DNA-based techniques introduced to overcome these barriers and to promise satisfactory performance of the results in terms of precision and sensitivity (7). Furthermore, the deposition of sequences of olive genome on NCBI database and the application of molecular markers in this discipline offered more benefits although olive oil usually provides very low yields of DNA and has variable degrees of degradation which may limit the applicability of molecular markers (8, 9). The first documented research on olive oil DNA isolation refers to Muzzalupo and Perri (10). However, other progress using SCAR (11), AFLP (9), RAPD, ISSR, and SSR (12), SSRs (13, 14), carried out since they are considered the most interested molecular markers in olive oil varietal identification. Rotondi et al. (14) performed a comparison between genetic results, chemical and sensory properties of monovarietal olive oils and demonstrated a very good correspondence between the clustering obtained by SSR analysis and the clustering based on selected fatty acids composition (15).
As plant cells have one nuclear and two cytoplasmic (chloroplast and mitochondrial) genomes. The nuclear genome undergoes recombination during sexual reproduction, the other two do not, and therefore chloroplast and mitochondrial DNAs are more useful for taxonomic studies. Chloroplast DNA (cpDNA) has been extensively used in phylogenetic reconstructions and a number of potentially useful regions are easily amplified using universal primers (16) even more it has used for cultivar identification in olive oil (17). It is supposed that PCR analysis on cpDNA/mtDNA with further molecular analysis such as SNP detection will provide trustable results in olive oil authentication.
The chemometrics techniques play an important role in the study of edible fats and oils, especially for the authentication study (18). Among them, FTIR received more attention as assesses the relationship between the concentration of analyte and its spectra (19) by fast acquisition of a great numbers of spectral data (20). Furthermore it is considered as a valid tool in the study of edible oils and fats could serve as a “fingerprint technique”. This analytical approach presents good sensitivity and a great simplicity in sample preparation and data elaboration and considers as a valid tool to authenticate extra virgin olive oils (21). Some attempts in using FTIR to distinguish olive oils from different geographical origin (22, 23) and different genetic varieties (24) have been proposed, as well (25).
The present research has been carried out with the aim of offering powerful molecular/chemical tools suitable to prove olive oil authenticity and to prevent its adulteration. SBE and FTIR techniques were compared when olive oil has been admixture with different ratios of hazelnut oil.
3. Materials and Methods
3.1. Sample Preparation and DNA Extraction
Cold pressed unfiltered virgin olive oil and hazelnut oil were prepared at Core facilities lab, NIGEB, Tehran, Iran and stored at 4°C until worked out. Commercial kit Qiagen QIAamp DNA stool (cod. 51504) was used for DNA extraction which DNA was extracted from 250 µg oil samples (13). This kit is based on resin tablets that absorb PCR inhibitors and silica-gel columns that allow separation of nucleic acids. To test the quality of DNA extracted, two microsatellite markers of the literature (DCA17 and DCA9) were tested on DNA isolated from olive oil by means of PCR (PTC-200 machine, MJ research, USA) amplification and agarose gel electrophoresis separation (Figure 1) (13). From the young leaves of the same olive cultivar which olive oil was extracted, DNA was extracted by commercial DNA extraction kit (DNeasy Plant Mini Kit (cod. 69104)) to compare the reliability of the results when compared with DNA extracted from oil.
3.2. Analysis of Chloroplast DNA
3.2.1. Design of Universal and Specific Primers
Upon a survey on NCBI GenBank, rbcL (ribulose-bisphosphate carboxylase large unit) gene was nominated for primer design. Alignment did on the sequence of rbcL that used as a template for the design of flanking primers with the software PRIMER3 (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi/) for two species of hazelnut (Corylus avellana L. access number: 33113306) and olive (Olea europaea L. access number: 2598012). The main concept to select this region was being well-conserved and high polymorphic, which conduct to design two primer pairs of common and specific amplification, rbcL1 and rbcL2, 143 bp and 118 bp, respectively (Figure 2).
It is suggested to provide more DNA copy number of the target zone before performing Single Base Extension (SBE) analysis. SBE analysis requires the use of a single primer that extends one of the strands. Therefore two one-handed primers were designed on rbcL sequence and it was decided to carry out the analysis on both strands using separately either forward or reverse primer. In Figure 4, the two designed SBE primers are shown. Forward primer is in dark box and it exactly ends before a G, so the addition of a G fluorescent dye-labeled terminator means the presence of hazelnut in the oil sample. In case of Rev-primer (clear box), the addition of a dye-labelled C, should confirm the presence of hazelnut in the sample. The primer pairs’ information is addressed in Table 1.
3.2.2. Single Base Extension Approach
Two simple analyses of the rbcL polymorphism were designed, both based on the SBE: one makes use of the ABI PRISM automatic sequencer and the SNaPshot kit (APPLERA); the other one makes use of the VICTOR machine (Perkin Elmer, USA) and the modified TDI-FP kit (Template-directed Dye-terminator Incorporation with Fluorescence Polarization detection technique) (26).
188.8.131.52. Acycloprime Single Nucleotide Polymorphism Detection
The Acycloprime TM II SNP Detection kit used to determine the base present at a specific location in an amplified DNA by a modification of TDI-FP. This approach is more robust and accurate even in the presence of few copies of DNA (27).
A preliminary PCR with a primer pair designed to amplify the zone including SBE was carried out (rbcL3 primer pairs, Figure 3). Then a clean-up pre-SBE step operates to degrade and remove extra dNTPs, inorganic pyrophosphate (PPi) and the rest of primers from the amplification steps to prevent interference with primer extension, by means of enzymatic purification, pyrophosphatase (PPase). By incubating at 80°C for 15 min the enzyme deactivates. Acycloprime reaction by SNP primer to extend the primer only with one base during the thermal cycles between 10 to 30 cycles performs. In the TDI-FP, the combination of R110 and Tamra terminators was used because their spectral wavelengths do not overlap. As the final step, FP values read by using VICTOR machine and the allele-calling software, calculates the results (SNPscorer software cat no: ASP001 from PerkinElmer).
184.108.40.206. ABI PRISM SNaPshot
ABI PRISM SNaPshot is based on the di-deoxy single base extension of an unlabeled oligonucleotide primer that has performed by ABI PRISM SNaPshot Multiplex Kit (Applied Biosystem).
DNA amplification was carried with the aim of amplifying a region including SBE (rbcL3 primer) followed by an enzymatic purification (SAP, Shrimp Alkaline Phosphatase and Exo I, Exonuclease I) to obtain purified template. The second purification or post-SBE clean-up (SAP and ExoI) carried out to remove unincorporated ddNTPs. Then the samples run on ABI PRISM 3700 DNA Analyzer instrument while it has been already added Hi-Di formamide to each of them and samples were denatured.
3.3. Spectrum Wavelength Evaluation by Fourier-Transform InfraRed Spectroscopy
FTIR spectra were recorded on FT-IR spectrophotometer (BRUKER, Germany) using KBr discs that were approximately 5 mm in diameter and approximately 1 mm in thickness. IR spectra were recorded in the 4000–700 cm−1 range at scanning speed of 2 mm.s-1 with a resolution of 4 cm−1 at room temperature (25 °C) and relative humidity of 30%.
Oil samples (2 µL) were coated on the KBr discs to form thin liquid films for infrared spectrometry analysis. The sample measurements were replicated 5 times with 4 scans each for a total of 20 spectra, and then the average chart was taken as a last sample spectrum. The background air spectrum, water vapor and CO2 interference were subtracted from these spectra.
4.1. Sample Preparation and DNA Extraction
Total DNA was extracted from both olive plant and oil successfully and after performing PCR reaction the amplicons were observed on agarose gel electrophoresis stained with “Syber green” since it is more sensitive and the bands appear more intense than the gels stained with ethidium bromide (Figure 1). In microsatellite studies, we should expect generally two alleles for each microsatellite amplification (in this case, as we hypothesize the used olive cultivar is Carolea cv. and DCA 17 should provide 117 bp and 143 bp alleles and in contrary164 bp and 199 bp for DCA9). Recording missing allele could be the result of either the preferential amplification of one of the two alleles in oil-derived DNA templates, or to the excess of degradation of the DNA template of the missed allele, that limited the production of a sufficient number of copies of that allele to be detected.
4.2. Chloroplast DNA Analysis
The PCR analyses carried out with the designed primers (Figure 2) in single species-derived oils and mixture of olive: hazelnut oils in the ratio of 50:50, 70:30, and 30:70 v:v). As expected, that specific primer designed for hazelnut did not anneal the olive DNA, and that primers were enough selective to identify the contamination of olive oil by means of hazelnut oil (Figure 3).
Attempt were made to change the ratio between the two oils by reducing the contribution of hazelnut oil to < 10%, until we decided to move towards more sensitive methods of analysis such those based on single base extension.
4.3. SBE Analysis
Future SNP technology would allow direct revealing of adulteration of olive oil with oils of other species. It could be possible also to design SNP-based experiments to detect individual olive varieties when a sufficient amount of intra-specific SNPs is documented in olive. Figure 4 shows the SBE specific primers.
Figure 5 shows the clustering of Acycloprime data obtained by plotting the TAMRA polarization values (T incorporation) against R110 ones (G incorporation) after 30 thermal cycles; hazelnut, olive, and the blend of hazelnut: olive (3: 97 and 8: 92) and the negative control (the absence of DNA template).
As expected, for hazelnut samples, the values for R110 were high (lower right) and the values for TAMRA were low, reflecting incorporation of R110 but not TAMRA. The Tamra/T scores present either in the upper right or in the upper left. Negative controls, represented by PCR without DNA template, pure olive samples (the lack of base G, hazelnut DNA absent) and failed PCR reactions are in the lower left cluster.
In Figure 6 the results of ABI PRISM SNaPshot analysis is reported. The presence of extra peak involved with the extension of “G” in pure hazelnut and contaminated olive oil samples with different percent of hazelnut oil (3 and 8%) is considerable.
4.4. Spectrum wavelength evaluation by FTIR
Pure olive and hazelnut oils and their admixture at ratios 75: 25 and vice versa were analyzed by means of FTIR. Figure 7 exhibits the absorbance of the samples, at frequency region of 4000 – 700 cm -1.
Some differences in the absorbance of the spectra of pure olive oil and hazelnut-adulterated olive oil have revealed. The higher absorbance at frequency regions of 3007 cm -1 (attributed the C-H stretching), 1373 cm -1 (-CH3 bending), 1237, 1120, 1098, and 1032 cm -1 (-C-O stretching), 1160 cm -1 (-C-O stretching, -CH2 bending), 965 (trans–CH=CH-bending out of plane), as well as 722 (cis-CH=CH-bending out of plane) were observed for pure olive oil. However, abnormalities were recorded at frequency regions of 2922 and 2853 cm -1 (symmetrical and asymmetrical stretching of –CH2).
The assignment of such functional groups has reported by Liang et al. (29) in adulterated walnut oil. Rohman and Che Man (30) performed FTIR analysis in discriminating virgin olive oil from other edible oils. They got the same results but the only controversy was at frequency region 2853 with higher absorbance for olive oil. This minor discordance between results is supposed to be a function of higher/lower olive oil in admixture samples. De La Mata et al. (31) reported that FTIR uses as a semi-quantification analysis can be performed suitably for screening purposes at contribution lower than 50% (w/w). Such differences at higher wavelengths are the result of thermal treatments and/or oil degradation as described by Navarra et al. (32).
Hazelnut oil is sometimes used to adulterate olive oil because of its lower cost in the market and the ease it escapes analytical methods. DNA-based techniques can help to reveal this type of adulteration. Here, chloroplast sequences are introduced as the most suitable one for the differential analysis of olive and hazelnut. Gene-coding chloroplast DNA (cpDNA) sequences, that are usually conserved within a given species but differ from one species to another considers a good choice. This is because the mutations occurs at lower rate in cpDNA than in nuclear DNA and, secondly, because coding sequences are less variable than non-coding sequences due to the selection pressure.
We used for such an analysis the rbcL, ribulose-bisphosphate carboxylase Large unit, that is a gene conserved at low taxonomic levels (no differences between individuals of the same species) and is polymorphic at higher taxonomic levels (differences occurring in the DNA sequence between species and at higher hierarchical levels).
To investigate olive oil adulterated by hazelnut oil it was requested a new primer pairs not only capable to give a good DNA amplification from both olive and hazelnut oils but also a selective amplification to differentiate the presence of DNA of both species.
We designed a primer capable to amplify a fragment of the same size (143 bp) from both admixture of hazelnut and olive species, and a second primer pair capable to provide amplicons of 118 bp only from hazelnut species, at ratios of 1:1, 3:1, and 1:3. However, for lesser hazelnut oil contribution a very sensitive method of analysis, based on single base extension has proposed. Two parallel methods Acycloprime SNP detection or TDI-FP (Template-directed Dye-terminator Incorporation _ Fluorescence polarization) and ABI PRISM SNaPshot were considered eligible. Furthermore, both methods provided consistent and reliable results and appear promising.
When FTIR was used to identify hazelnut-adulterated olive oil as a rapid and non-destructive powerful analytical tool in olive oil adulteration detection (21, 31, 33), appropriate results in accordance with molecular analysis (using primers rbcLz1 and rbcLz2) were achieved. However, the intensities of the FTIR spectral bands are proportional to concentration. Furthermore, for less than 10% of hazelnut contribution in admixture samples, FTIR method provided inconsistent results and does not advise for such adulteration recognition.
The authors would like to appreciate the helpful assistance of Dr. Guisi Zaina in the application of the single base extension technique.
Leila Akbari: FTIR analysis, Zohreh Rabiei: SBE analysis, Sattar Tahmasebi Enferadi: elaboration of FTIR results, Sakineh Vanaii: FTIR analysis.
There is no conflict of interest.
This work was partially supported by grant No. 409 from National Institute of Genetic Engineering and Biotechnology (NIGEB), Tehran, Iran.