Extra virgin olive oil: A comprehensive review of efforts to ensure its authenticity, traceability, and safety

Abstract

The growing demand for extra virgin olive oil (EVOO), appreciated for its unique organoleptic properties and health benefits, has led to various fraudulent practices to maximize profits, including dilution with lower value edible oils. The adulterated oils would be of poor nutritional quality, more readily oxidized, and may contain unhealthy substances formed during processing. Nevertheless, the range of available techniques to detect fraud in EVOO production has been growing. Reliable markers of EVOO adulteration include fatty acids and minor components such as sterols, tocopherols, triterpene alcohols, phenolic compounds, phospholipids, volatile compounds, and pigments. Additionally, increasing consumer interest in high-quality EVOO has led to the development of robust scientific methods for its traceability.

This review focuses on (i) the usefulness of certain compounds as markers of EVOO adulteration; (ii) the potential health risks of consuming adulterated EVOO; and (iii) reliable methods for the geographical traceability of olive oil. In conclusion, fraudulent production practices need to be detected to preserve the beneficial health effects of EVOO and to avoid the potential risks associated with ingesting substandard oil. In this work, as EVOO certification and regulatory framework limitations have already been extensively reviewed, we focus our attention on biomarkers that guarantee both the authenticity and traceability of oil, and consequently its health properties. When it is unavailable to obtain a high-resolution mass spectrometry full fingerprint, stigmastadienes and the sterolic profile are proposed as reliable markers.

1 INTRODUCTION

Extra virgin olive oil (EVOO), considered to be the highest quality olive oil, is much appreciated for its organoleptic and nutritional properties. Moreover, its consumption has been gaining popularity beyond the Mediterranean area, where it is the main dietary source of fat (De Santis et al., 2019; Flori et al., 2019). EVOO is mostly composed of triglycerides (TAG) (98%), mainly monounsaturated fatty acids (MUFA) (80%) such as oleic acid (C18:1), which are responsible for its physicochemical properties (Carranco et al., 2018; Jimenez-Lopez et al., 2020). In the remaining unsaponifiable fraction (1–2%) stand different compounds which have known health benefits, such as phenolic compounds and triterpenic acids (Cárdeno et al., 2014; Carranco et al., 2018; Criado-Navarro et al., 2021; Gelmini et al., 2016). The nutritional profile of EVOO, particularly its richness in antioxidant and anti-inflammatory phenolic compounds together with its fatty acid (FA) profile, is associated with multiple health benefits, including the prevention of cardiovascular diseases, cancer, neurodegeneration, and diabetes (Castro-Barquero et al., 2020; Estruch et al., 2018; García-Gavilán et al., 2018; Guasch-Ferré et al., 2014; Lecerf, 2009; Tresserra-Rimbau et al., 2014).

Although all EVOO is produced following the same basic procedure, its composition and sensory properties vary depending on factors such as the olive variety and ripeness, environmental conditions and the technological methods used in processing, and the storage conditions (Boneza & Niemeyer, 2018; Correa Fuentes et al., 2017; López-Yerena et al., 2019, 2020, 2021; Polari et al., 2018). However, to be in EVOO category (instead of virgin olive oil or olive oil), certain criteria must be fulfilled: Production should be exclusively by mechanical procedures; free fatty acidity should not be over 0.8%; and the oil must not have sensory defects and include fruity attributes (Jimenez-Lopez et al., 2020). EVOO should also have a low peroxide index and specific spectrophotometric values (K₂₃₂ ≤ 2.5, K₂₇₀ ≤ 0.22, ΔK ≤ 0.01), which are indicators not only of quality, but also authenticity (Jimenez-Lopez et al., 2020; Ün & Salim, 2018).

Even though the olive oil sector is highly regulated by the European Union (EU) by Reg. (EEC) 2568/91 as amended (European Union, 2013b), and the International Olive Council (IOC), which also establishes methods for their analysis, it is known that there are still drawbacks in their analytical methods (Conte et al., 2020). Authentication of food and the detection of adulteration have been concerns throughout the evolution of food industries (Bansal et al., 2017; Perez et al., 2020; Stadler et al., 2016), and EVOO production is no exception (Carranco et al., 2018; González-Domínguez et al., 2019). Owing to the growing complexity of food production chains, more exhaustive measures and strategies are required to tackle fraudulent attempts by food producers to boost their profits (Creydt & Fischer, 2018).

In this review, we focus on the most recent approaches to detect adulteration of EVOO with lower value oils based on the profiling of FA and minor components such as phenolic compounds, volatile compounds, tocopherols, phospholipids, and pigments, proposing stigmastadienes and sterols as markers for detecting its adulteration. We also look at the potential health risks of consuming adulterated EVOO and describe reliable methods for the geographical traceability of olive oil. Although there are excellent reviews focused on olive oil authentication (Bajoub et al., 2018) and the weaknesses in the regulatory framework (Conte et al., 2020), our approach draws attention to global traceability organized by biomarkers, not by analytical techniques or type of adulteration and, most importantly, includes a section that highlights the possible risks of ingesting fraudulent EVOO.

2 FRAUD, AUTHENTICATION, AND TRACEABILITY OF EVOO

2.1 Fraudulent activities in EVOO production

Fraudulent activities in the food industries, a global issue of increasing concern, comprise a wide range of malpractices, including adulteration (dilution, substitution, and unapproved enhancement), mislabeling, and smuggling (Stadler et al., 2016). Economic gain is the main reason for committing food fraud, but its significance is not only a question of illegality, as it can lead to adverse health consequences for the consumer (Bansal et al., 2017). An extensive study analyzing the evolution of food fraud over 30 years (1980–2010) found that the type of adulterated food most frequently reported in academic articles is olive oil (Moore et al., 2012). As traditional methodologies for identifying such adulterations through the assessment of quality and purity have some limitations (Conte et al., 2020), new analytical techniques need to be developed (Barbieri et al., 2020; Creydt & Fischer, 2018; Damiani et al., 2020; Violino et al., 2021).

The most commonly reported frauds in EVOO production are: (i) mixing EVOO with lower quality oils that are cheaper to produce; (ii) mislabeling, for example, falsely claiming that refined olive oil is EVOO, or concealing the real place of origin; and (iii) price fraud, which consists of reducing the oil price to artificially lower the market value (García Martínez, 2020). Adulteration can be with refined olive oil (Garcia et al., 2013; Torrecilla et al., 2010), refined olive−pomace oil (Fragaki et al., 2005; Torrecilla et al., 2010), lampante olive oil (Fragaki et al., 2005), and olive pomace oil (Guimet et al., 2005; Yang & Irudayaraj, 2001). Other edible oils from other plants such as hazelnut (Arlorio et al., 2010; Chiavaro et al., 2008), soybean (da Silveira et al., 2017; Fasciotti & Netto, 2010; Jabeur et al., 2014; Mendes et al., 2015; Milanez et al., 2017; Poiana et al., 2015; Tiryaki & Ayvaz, 2017; Yang & Irudayaraj, 2001), corn (Jabeur et al., 2014), canola (Salivaras & McCurdy, 1992), and sunflower (Jabeur et al., 2014) have also been used.

Different EVOO components have been proposed as possible markers of fraudulent activities, including natural chlorophyll, diacylglycerols (DAG), FA, wax esters, phenolic compounds, and sterols (Garcia et al., 2013; Gómez-Coca et al., 2020).

2.2 Authentication

Consumer demand for authenticity and traceability of food products means that quality certification is of great interest for the food sector. In this context, a range of government legislation and regulations as well as international agency guidelines have been published (Conte et al., 2020; Melucci et al., 2016).

The quality of olive oil is governed by the IOC together with the European Community and the Codex Alimentarius Commission (Stadler et al., 2016), an organization that proposes several analytical methods, including gas chromatography (GC), high performance liquid chromatography (HPLC) or gravimetry, among others. European regulation derives from the IOC and Codex Alimentarius Commission regulations, as well as most of the national regulations, such as Brazilian one. However, there are some countries still lacking internal regulations for EVOO; in the case of the Untied States, the FDA still has no regulation defining this product.

In Europe, to achieve the category of EVOO, the oil must comply with a range of parameters set out in EU Regulation No. 2568/91 (European Community, 1991) and consequently modified in the EU Regulation 1348/2013 (European Union, 2013b), which are assessed by chemical and sensory tests using IOC methods and standards (Jimenez-Lopez et al., 2020). The quality criteria evaluated in olive oil are free acidity, the peroxide index, ultraviolet (UV) absorption (coefficients K₂₃₂, K270, and ΔK), fatty acid ethyl esters, and sensory characteristics (Conte et al., 2020; Jimenez-Lopez et al., 2020). These basic quality parameters, together with composition in fatty acids and sterols as well as profiles in phospholipids, tocopherols, phenolic molecules, volatile compounds and so on, can offer information very useful to check authenticity of EVOO and possible applied fraudulent activities to it (Azizian et al., 2015; Mikrou et al., 2020). Some examples of detection of EVOO adulterants are presented in Table 1 with the analytical techniques and the data processing methods used for their determination.

TABLE 1. Summary of the trending or novel analytical techniques, biomarker compounds, and data processing methods for the detection of EVOO adulterants

Methods	Determination	Data processing	Adulterants	References
GC-FID	FA composition	PCA	Geographical varieties	(Mikrou et al., 2020)
GC-FID	FA composition	Direct comparison	Corn, sunflower, soybean, and canola oils	(Aykas et al., 2020)
GC-MS	FA composition	CARS-PLS-LDA and MCTree	Corn, peanut, rapeseed, and sunflower oils	(Yang et al., 2013)
SPE-GC-FID	DAG and FFA	Direct comparison	Soft deodorized oil	(Gómez-Coca et al., 2020)
GC-FID	FA composition	Direct comparison	Refined soybean oil	(da Silveira et al., 2017)
MALDI-TOF MS	Complete lipidomic profile	UHC, PCA, and Pearson's correlation	Corn oil	(Di Girolamo et al., 2015)
FIA-HESI-HRMS	TAG profile	PLS-DA	High linoleic and high oleic vegetable oils	(Quintanilla-Casas et al., 2021)
FIA-MRMS	TAG, DAG and FFA	PCA and OPLS-DA	Geographical varieties	(Nikou et al., 2020)
FT-NIR	FA composition	PLS	Oils high in linoleic or oleic acid, palm olein, and refined olive oil	(Azizian et al., 2016)
FT-NIR	Volatile compounds and FA	PCA and SIMCA, PLS	Soybean, sunflower, corn, canola, hazelnut, high oleic acid safflower, peanut and refined olive oils, and palm olein	(Azizian et al., 2015)
FT-NIR	1,2-DAG, 1,3-DAG, and FFA	PLS	Oils high in linoleic or oleic acid, palm olein, and refined olive oil	(Azizian et al., 2018)
FT-Raman	FA composition	PLS	Geographic location, olive variety, harvest year, and PDO	(Sánchez-López et al., 2016)
FT-Raman	FA composition	iPLS and SiPLS	Waste cooking oil	(Li et al., 2018
MALDI-TOF-MS	Phospholipid	Direct compariso	Hazelnut oil	(Calvano et al., 2012
FL	Tocopherols, phenolic compounds, and chlorophyll	PLSR and ANN, PC	Corn, soybean, linseed, or sunflower oils	(Lia et al., 2018
FL	Tocopherols, phenolic compounds, and pigment	LDA-PARAFAC and UPLS-D	ROO, RPOO	(Merás et al., 2018
EEM-FL	Tocopherols, tocotrienols, phenolic compounds, and oxidation product	SVM and PLS	EVOO vs. VOO	(Omwange et al., 2021
HPLC-FL	Tocopherols (α/β-tocopherol ratio	Direct compariso	Sunflower, hazelnut and peanut oils	(Chen et al., 2011
Voltametric analysis	Phenolic compounds and tocopherol	PCA, PLS-DA, SIMCA, PLS	Sunflower, soybean, and corn oils	(Tsopelas et al., 2018)
e-tongue	Phenolic compounds and tocopherols	PLS-DA and PLSR	Sunflower, soybean, and corn oils	(Apetrei & Apetrei, 2014)
HPLC-DAD	Phenolic compounds	PLSR	Different cultivars, sunflower oil, and ROO	(Carranco et al., 2018)
GC-FID, HPLC-FL, HPLC-DAD-ESI/MS, UV	Volatile substances, polar phenolic substances, antioxidant activity, FA composition, and α-tocopherol	PCA	Low price vs. high price EVOO	(Fiorini et al., 2018)
LC-DAD/ESI–MS/MS, GC-FID	Phenolic compounds and sterol profile	PCA and UHC	RPOO	(Drira et al., 2020)
HPLC-DAD	Squalene and tyrosol	Scatter diagram of standard scores (z)	Sunflower and grapeseed oils	(Hayakawa et al., 2020)
SIFT-MS	Volatile compounds	SIMCA, PLSR	Corn, sunflower, high oleic sunflower, and olive oils	(Ozcan-Sinir, 2020)
GC-MS, GC-FID, GC-O-MS, ADEA	Volatile compounds and FA	PCA	RPOO	(Drira et al., 2021)
HS-SPME-GC–MS	Volatile compounds	PCA and SIMCA	Sunflower oil	(Van Durme & Vandamme, 2016)
vis-NIR	Volatile compounds and pigments	Artificial intelligence model	Corn, sunflower, rice, peanut, hazelnut, virgin wheat germ, and virgin cornstarch oils	(Violino et al., 2021)
GC-IMS, FGC E-nose	Volatile compounds	SIMCA	ROO	(Damiani et al., 2020)
UV–vis	Pigments, phenolic compounds, and tocopherols	PLS-JK, SW-MLR, GA-MLR	Soybean oil	(Milanez et al., 2017)
SBS-vis	Pigments	Linear and surface regressions	Rapeseed, soybean, peanut, and sunflower oils	(Shi et al., 2019)
FL spectroscopy	Pigments and oxidation compounds	PCA, PLSR	Sunflower oil	(Ali et al., 2018)
Frontface-FL and vis	Pigments, phenolic compounds, and tocopherols	PLSR	Corn, soybean, and sunflower oils	(Tan et al., 2018)
vis	Pigments, phenolic compounds, and tocopherols	ANN, MLP	ROO	(Aroca-Santos et al., 2016)
vis	Pigments, phenolic compounds, and tocopherols	ANN	ROO, PO, sunflower, and corn oils	(Aroca-Santos et al., 2015)
vis	Chlorophylls and carotenoids	CR, PCA and LDA	OO, PO	(Ferreiro-González et al., 2017)
MidIR, UV–vis, FL	Pigments, phenolic compounds, tocopherols, FA, and oxidation products	OPLS-DA and PLSR	Aged EVOOs	(Uncu & Ozen, 2019)
GC-FID	Campesterol and stigmasterol	Direct comparison	Corn, soybean, sunflower, and cotton oils	(Al-Ismail et al., 2010)
SPE-GC-FID	Uvaol, sterol and erythrodiol	Direct comparison	Pomace olive oil	(Mathison & Holstege, 2013)
HPLC-DRI, GC-FID	FA and sterol profile	LDA	Soybean, sunflower, and corn oils	(Jabeur et al., 2014)
TLC-GC-FID	Desmethylsterols	Direct comparison	Sunflower and soybean oils	(Youseff et al., 2014)
GC-FID	Desmethylsterols and triterpene dialcohols	Direct comparison	Canola, corn, peanut, safflower, soybean, and sunflower oils	(Srigley et al., 2016)
^1H NMR	Acyl groups, squalene, sterols, triterpene acids/esters, fatty alcohols, wax esters and phenols (lignans, tyrosol, hydroxytyrosol, oleocanthal, oleacein, oleokoronal, oleomissional, ligstrodials, and oleuropeindials)	Direct comparison	Different cultivars	(Ruiz-Aracama et al., 2017)
HPLC-DRI, GC-FID	FA and sterol profiles and ΔECN42	PCA	Cotton and sunflower oils	(Kesen, 2019)
GC-FID	Sterol profile	PCA, UHC	Different cultivars and EVOO from oleasters	(Manai-Djebali et al., 2021)
SPE-GC-FID	Free and esterified sterols and free and esterified triterpenic alcohols	Linear regression	Sunflower oil	(Valli et al., 2021)

• Abbreviations: GC-FID: gas chromatography flame ionization detector; FA: fatty acid; PCA: principal component analysis; GC-MS: gas chromatography mass spectrometry; CARS-PLS-LDA: competitive adaptive reweighted sampling partial least square linear discriminant analysis; MCTree: Monte Carlo tree; SPE: solid phase extraction; DAG: diacylglycerides; FFA: free fatty acids; MALDI-TOF-MS: matrix-assisted laser desorption/ionization mass spectrometry; UHC: unsupervised hierarchical cluster; FIA-HESI-HRMS: flow injection analysis-heated electrospray ionization high resolution mass spectrometry; TAG: triacylglycerides; PLS-DA: partial least square discriminant analysis; FIA-MRMS: flow injection analysis magnetic resonance mass spectrometry; OPLS-DA: orthogonal projection to latent structures discriminant analysis; FT-NIR: Fourier transform near infrared spectroscopy; PLS: partial least square; SIMCA: soft independent modeling of class analogy; 1,2-DAG: diacylglycerides in position 1 and 2; 1,3-DAG: diacylglycerides in position 1 and 3; PDO: protected designation of origin; iPLS: interval partial least square; SiPLS: synergy interval partial least square; FL: fluorescence spectroscopy; PLSR: partial least squares regression; ANN: artificial neural network; ROO: refined olive oil; RPOO: refined pomace olive oil; LDA-PARAFAC: linear discriminant analysis parallel factor analysis; UPLS-DA: unfolded partial least squares discriminant analysis; EEM-FL: excitation emission matrix fluorescence spectroscopy; EVOO: extra virgin olive oil; VOO: virgin olive oil; SVM: support-vector machine model; HPLC-FL: high-performance liquid chromatography-fluorescence spectroscopy; e-tongue: electronic tongue; HPLC-DAD: high-performance liquid chromatography diode array detector; HPLC-DAD-ESI/MS: high-performance liquid chromatography diode array detector electrospray ionization mass spectrometry; UV: Ultraviolet spectroscopy; LC DAD/ESI–MS/MS: liquid chromatography diode array detector electrospray ionization mass spectrometry in tandem; SIFT-MS: selected ion flow tube mass spectrometry; GC-O-MS: gas chromatography olfactometry mass spectrometry; ADEA: aroma dilution extraction assay; HS-SPME-GC-MS: headspace solid phase microextraction gas chromatography mass spectrometry; vis: visible spectroscopy; GC-IMS: gas chromatography ionic mobility spectrometry; FGC E-nose: flash gas chromatography electronic nose; PLS-JK: partial least squares Jack-Knife; SW-MLR stepwise multiple linear regression; GA-MLR: genetic algorithm multiple linear regression; SBS- vis: stimulated Brillouin scattering visible spectroscopy; MLP: multilayer perception model; PO: pomace olive oil; CR: multivariate curve resolution method; LDA: linear discriminant analysis; MidIR: Mid infra-red spectroscopy; HPLC-DRI: high-performance liquid chromatography differential refractometer detector; TLC-GC-FID: thin layer chromatography gas chromatography flame ionization detector; 1H NMR: proton nuclear magnetic resonance.

Authenticity of food or beverages can be established by an exact match between the characteristics, properties, and composition of the product and the description on the label. Although the EU closely monitors the olive oil market, the risk of EVOO adulteration remains high due to the economic gains at stake and the increasingly sophisticated techniques employed (Bansal et al., 2017).

Among the EU regulations established to standardize the production and commercialization of EVOO in Europe (Melucci et al., 2016) (the largest producer, exporter, and consumer of olive oil in the world), Regulation EU No. 1019/02 defines its labeling and packaging. Certified labeling is a way of providing the consumer with assurance of food authenticity. The information on EVOO labels also includes its classification as light, medium, or intense, according to its sensory attributes.

2.2.1 Fatty acids and glycerides derivatives

FAs are present in the oil as triacylglycerols. Its main component is oleic acid, a MUFA that accounts for 65–80% of the FA content (ranges obtained from Konuskan et al., 2019; Orsavova et al., 2015; Rotondi et al., 2021; Vávra et al., 2021; Wang et al., 2019; Yahay et al., 2021). Oleic acid is reported to have beneficial cardiovascular effects and, as MUFA, with its single double bond, it is more stable than polyunsaturated fatty acids (PUFA) (Yubero-Serrano et al., 2019). Other FAs, such as linoleic are much less abundant in EVOO than in other vegetable oils (Jimenez-Lopez et al., 2020). The FA profile of EVOO thus distinguishes it from other edible oils (such as sunflower and peanut oil), as shown in Figure 1, and its analysis can provide useful information for the detection of adulteration. However, other edible oils such as rapeseed oil also contain a high proportion of MUFAs. Several studies have compared the FA profiles of EVOO and different oils commonly employed as adulterants, such as sunflower, corn, peanut, coconut, or rapeseed (Aykas et al., 2020; Mikrou et al., 2020; Yang et al., 2013), using GC coupled with a flame ionization detector (FID) or mass spectrometry (MS), the traditional method to evaluate the lipid fraction of EVOO. This methodology, however, requires a pretreatment of the sample to obtain FA methyl esters. Deodorized olive oil in EVOO was successfully detected by the determination of DAGs and free FAs (Gómez-Coca et al., 2020). A comparative study analyzing TAG by electrospray ionization MS (ESI-MS) and FA by GC-FID concluded that the spectroscopic technique was faster and more efficient than the chromatographic method. Applying this novel methodology, olein lineo 10-heptadecenoic was established as a lipid marker for soybean oil added to EVOO (da Silveira et al., 2017).

FIGURE 1

Main FA components of vegetable oils, ranges in % obtained from Alves et al., 2019; Chen et al., 2020; Feizabadi et al., 2021; Holey et al., 2021; Konuskan et al., 2019; Król et al., 2021; Rotondi et al., 2021; Vávra, Hájek, & Kocián, 2021b; Wang et al., 2019; Yahay et al., 2021

In recent years, spectroscopic methods have emerged as useful and efficient techniques for the determination of FA; not requiring a derivatization step, they are less time-consuming than chromatographic methods. Thus, a simple technique able to characterize EVOO and detect adulterations with corn oil was developed by Di Girolamo et al. (2015) based on matrix-assisted laser desorption/ionization MS (MALDI-TOF MS) with unsupervised hierarchical clustering (UHC), principal component analysis (PCA), and Pearson's correlation analyses. Authentication models for the detection of high linoleic and high oleic vegetable oils in EVOO were developed based on flow injection analysis–heated electrospray ionization–high-resolution MS (FIA-HESI-HRMS) combined with partial least squares (PLS)–discriminant analysis (DA), which provided TAG profiles of olive oil samples (Quintanilla-Casas et al., 2021).

The lipid profile of EVOO was also successfully characterized using an untargeted metabolic approach based on flow injection analysis–magnetic resonance MS (FIA-MRMS) and chemometrics (Nikou et al., 2020). Indeed, it provided better projection and prediction models than LC-Orbitrap MS, with the additional advantage that it allows simultaneous monitoring of both lipophilic compounds and polyphenols. Moreover, Fourier transform-near infrared (FT-NIR) spectroscopy coupled to chemometric tools such as PLS analysis was able to detect adulterants (i.e., edible oils high in 16:0 [palm oil], 18:1n-9 [palm olein, hazelnut, canola, or high OLA sunflower and safflower oils], 18:2n-6 [soybean, sunflower or corn oil], and ROO) in EVOO (Azizian et al., 2015, 2016, 2018). Another vibrational technique, FT-Raman, combined with chemometrics (PCA and PLS) accurately classified EVOO samples according to the harvest year, olive variety, geographical origin, and protected designation of origin (PDO), and also detected adulteration with sunflower or waste cooking oil, due to the different unsaturation degree of the FAs (Li et al., 2018; Sánchez-López et al., 2016).

In summary, analytical techniques such as NIR or Raman spectroscopy are likely to grow in importance and replace traditional chromatographic methods for the detection of adulterants based on FA determination. Overall, they provide efficient results with the advantage of a simpler sample treatment step compared to chromatography, which reduces the time of analysis.

2.2.2 Phospholipids

Olives and olive oil contain a diversity of phospholipids as minor components (Alves et al., 2005). The profile of phospholipids, whose concentration in EVOO is lower than in other vegetable oils (Antonelli et al., 2020), can provide a distinct “fingerprint” for traceability and authenticity studies (Alves et al., 2018; Gallina Toschi et al., 2013).

Phospholipid analysis using an ionic liquid as a matrix and extraction solvent and MALDI-TOF-MS detected the presence of hazelnut oil in EVOO (still detectable at a 1% contamination level) due to a significant increase in phospholipid signals (Calvano et al., 2012). In another study based on phospholipids analysis, MALDI-TOF MS coupled to UHC and PCA was used to characterize the oil type and was able to reveal very low levels of corn oil in EVOO (as low as 0.5%) (Di Girolamo et al., 2015).

2.2.3 Tocopherols

Tocopherols (vitamin E compounds) are found in seed oils in four different forms: α-, β-, γ-, and δ-tocopherols (Ergönül & Köseoğlu, 2014) but in EVOO, δ-tocopherol has not been detected, only α-, β-, and γ-tocopherols have been described, with α-tocopherol representing more than 95% of the total tocopherol content (Beltrán et al., 2010). Tocopherol content and composition depend on several agronomic factors, including the cultivar, fruit ripeness, and agroclimatic conditions (Beltrán et al., 2010). Some of the efforts developed to combat EVOO fraud have focused on the tocopherol profile as a potential marker able to detect adulteration with high selectivity, sensitivity, and accuracy.

The concentration of tocopherols in EVOO changes when it is adulterated with other vegetable oils, including lower quality olive oils (Omwange et al., 2021). Depending on the adulterating oil, the concentration can increase, when adulterated with sunflower oil for example (Lia et al., 2018), or decrease, when adulterated with olive oil (Merás et al., 2018). Accordingly, autofluorescence excitation-emission profiles combined with multiway classification allowed approximately 15% of olive oil and 3% of olive pomace oil to be detected in EVOO (Merás et al., 2018). Front-face fluorescence spectroscopy coupled with UV-induced fluorescence imaging differentiated between pure EVOO and adulterated oils based on a specific region of excitation emission matrices (between 300 and 600 nm) corresponding to tocopherols, tocotrienols, phenolic compounds, oxidation products, and vitamin E (Omwange et al., 2021). The dilution of Maltese EVOO with several vegetable oils (corn, soybean, linseed, and sunflower) was identified by fluorescence spectrometry combined with PCA, PLSR, and an artificial neural network (Lia et al., 2018). In another study, the authenticity of EVOO was assessed using the α/β tocopherol ratio and the presence of δ-tocopherol to detect the fraudulent addition of oils from other sources (Chen et al., 2011).

Tocopherols have antioxidant properties and can therefore be detected by means of electroanalytic methods (Apetrei & Apetrei, 2014; Tsopelas et al., 2018). Voltametric fingerprinting of EVOO can reveal changes in the concentration of electroactive compounds such as tocopherols. A voltametric e-tongue successfully detected adulterations of olive oil with less than 10% of sunflower, soybean, and corn oils (Apetrei & Apetrei, 2014). In another study, voltametric fingerprinting combined with PLS-DA provided a clear discrimination between olive oils (extra virgin and regular) and olive pomace/ seed oils (Tsopelas et al., 2018). This assumption is supported by the much lower tocopherol content of olive oil compared to seed oils (Kamal-Eldin, 2006).

In summary, the advantage of determining the tocopherol profile is that it requires only minimal sample preparation, such as oil dilution (Chen et al., 2011; Lia et al., 2018), obtaining methanolic extracts (Tsopelas et al., 2018), or directly analyzing the EVOO (Apetrei & Apetrei, 2014; Merás et al., 2018), before subsequent analysis by fluorescence and/or voltammetry with chemometrics. One of the least used approaches for the detection of EVOO adulteration, tocopherol fingerprinting likely deserves wider application.

2.2.4 Phenolic compounds

The main phenolic compounds in EVOO are secoiridoids, which are accompanied by phenolic acids, lignans, flavonoids, and phenolic alcohols (Lozano-Castellón et al., 2020). As secoiridoids are characteristic of the Oleacea family (Jensen, 2002), a reduction in their content could indicate a possible fraudulent blend, either adding EVOO from another cultivar (i.e., Arbequina is less rich in phenolic compounds than Picual) or mixing EVOO with different types of oil. Most techniques for EVOO phenolic analysis are based on liquid–liquid extraction and various determination methods, the most common being liquid chromatography coupled to different detectors, such as a diode array detector (International Olive Council, 2017b), MS (Alessandri et al., 2014), or MS in tandem (Lozano-Castellón et al., 2021). Other methods are nuclear magnetic resonance (Olmo-Cunillera et al., 2020) or GC-MS (Chiou et al., 2007), which need an extra step for volatilizing the target compounds.

The squalene and tyrosol concentration allowed pure EVOO to be differentiated from EVOO blended with seed oils; pure EVOO presents higher concentration of both compounds, that when plotting the concentration of tyrosol against the concentration of squalene, the blended samples are located near the origin of the plot (0,0) and it is possible to identify them (Hayakawa et al., 2020). Using an e-tongue, it was possible to discriminate between pure EVOO and EVOO containing sunflower, soybean, and corn oils, based on the specific electrical properties of phenolic compounds and their lower concentration in the adulterated samples (Apetrei & Apetrei, 2014). Using a partial least squares regression (PLSR) model generated from the HPLC-UV spectrum of the phenolic extract, monovarietal Arbequina EVOO was accurately distinguished from the same variety mixed with Picual EVOO, refined olive oil, or sunflower oil (Carranco et al., 2018). In a recent study, a PLSR model generated from fluorescence spectra of the phenolic profile was used to discriminate between pure EVOO and EVOO blended with olive oil (Omwange et al., 2021). Finally, EVOO adulterated with refined olive pomace oil was identified based on the phenolic and sterol content, as these compounds are more hydrophilic in the pomace after the first oil extraction than in EVOO (Drira et al., 2020).

The EVOO phenolic profile can be affected by numerous factors, such as the olive cultivar, agronomic techniques (López-Yerena et al., 2019), the extraction procedure (López-Yerena et al., 2021), and storage (Castillo-Luna et al., 2021). As refined vegetable oils contain much lower content of phenolic compounds (Orozco-Solano et al., 2011), their usage to adulterate EVOO will result in a lower phenolic concentration, but not the addition of a new phenolic compound that could potentially serve as a marker. Hence, this type of adulteration is difficult to assess based on phenolics alone, as their concentration can also be reduced depending on the variety, by an extraction parameter, or improper storage, so they should be targeted along with other compounds (Nikou et al., 2020).

2.2.5 Volatile compounds

EVOO is highly appreciated by consumers, mostly for its pleasant aroma and characteristic flavor. Aroma depends on the volatile fraction, which differs according to the olive variety, environmental growing conditions, and technological factors during processing operations (Cecchi et al., 2021). Typical flavors and off-flavor compounds that affect the volatile fraction of EVOO are produced by different mechanisms. Positive odors are mainly generated from the oxidation of linoleic and linolenic acids by enzymes of the lipoxygenase pathway, which are released when the fruit is crushed and have a major impact during malaxation. Conversely, the main defective or off-flavors are due to sugar fermentations, amino acid conversion, enzymatic activities of molds (musty), and anaerobic microorganisms (muddy). They are also the result of auto- and photo-oxidation of FA during EVOO storage, which produces linear acids, alcohols, esters, and ketones (Cecchi et al., 2021; Clodoveo et al., 2014).

Oxidation is the principal cause of the deterioration of olive oil quality. Fatty acids are the fraction most vulnerable to oxidation, their degradation leading to the production of carbonyl compounds and subsequent development of unpleasant flavors and oxidative rancidity (Gargouri et al., 2015; Sanmartin et al., 2018; Silva et al., 2015). Autoxidation can occur even in the absence of light by a free radical mechanism in which the absorption of oxygen results in the formation of hydroperoxides. These labile compounds evolve to produce a complex mixture of volatile compounds such as aldehydes, ketones, hydrocarbons, alcohols, and esters, which negatively affect the flavor of olive oil (Frankel, 2014). In fact, the “extra virgin” or even “virgin” designation is granted only if lipid oxidation products such as hydroperoxides do not exceed a stipulated limit and/or produce rancid flavors (Hrncirik & Fritsche, 2005). Positive odors and/or volatile oxidation compounds can also be used as markers of EVOO adulteration (Zhou et al., 2021).

The addition of other edible oils dilutes both the aroma and color of EVOO (Violino et al., 2021). However, to date, the technological determination of volatile organic compounds (VOCs) is not required to authenticate EVOO, even though they form an intrinsic part of the quality of the product and the perceived intensity of positive sensory attributes (Violino et al., 2021). There is clearly a need for reliable and inexpensive methods that can rapidly assess the VOC profile of EVOO on an industrial scale.

Over the last decade, numerous studies have targeted VOCs to identify EVOO adulteration with lower value oils such as corn, soybean, sunflower, high oleic sunflower, olive, soft-refined olive, and refined olive oils (Azizian et al., 2015; Damiani et al., 2020; Drira et al., 2021; Ozcan-Sinir, 2020; Van Durme & Vandamme, 2016; Violino et al., 2021; Zhou et al., 2021). The techniques applied include GC-MS, GC-FID, GC–olfactometry‑MS (GC-O-MS), GC ion mobility spectrometry (GC-IMS), flash GC electronic nose (FGC E-nose), thermogravimetric-GC/MS (TGA-GC/MS), selected ion flow tube MS (SIFT-MS), FT-NIR, and vis–NIR (Table 2). Among the strategies to isolate VOCs for analysis by GC-MS are purge and trap extraction (Drira et al., 2021), nonthermal plasma treatments (Van Durme & Vandamme, 2016) and profiling of the headspace composition (Damiani et al., 2020). On the other hand, for analysis by SIFT-MS, a temperature-controlled water bath at 30°C for 30 min was used to release volatile compounds until an equilibrium was reached at the headspace (Ozcan-Sinir, 2020). The application of PCA, PLS1, SIMCA, PLSR, and artificial intelligence to spectral data successfully predicted and determined adulterated samples.

TABLE 2. Identification of EVOO adulteration by assessing VOCs

Samples	Analysis	Findings	References
EVOO and adulterated EVOO (1, 2.5, 5, 10, and 20% of corn oil, sunflower oil, high oleic sunflower oil, and olive oil)	- SIFT-MS, SIMCA algorithm, PLSR model. - The excellence of the final model was calculated based on the number of latent variables, loading vectors, SECV, the coefficient of determination (R-value), SEP, and outlier diagnostics.	1-Octanol, 1-penten-3-one, 2-phenylethanol, dodecane, anisole, ethyl nonanoate, isobutanoic acid, ocimene, phenol, and toluene were the compounds that most successfully classified adulteration.	(Ozcan-Sinir, 2020)
EVOO adulterated with RPOO (1, 2, 3, 5, 6, 10, and 20%)	- GC-MS, GC-FID, GC-O-MS, and AEDA	Thirty-four relevant aroma compounds and twenty-one key odorants were quantified in EVOO, RPOO, and EVOO adulterated with 1−20% of RPOO.	(Drira et al., 2021)
EVOO and EVOO adulterated with sunflower oil (1%)	- NTP, PCA, and SIMCA	NTP treatments of 60 min (Ar/O2 0.1%) resulted in the formation of a unique set of secondary volatile lipid oxidation products enabling classification of adulterated oil samples.	(Van Durme & Vandamme, 2016)
EVOO and EVOO adulterated with RPOO	- FT-NIR, PLS1, and FT-NIR	FT-NIR cannot distinguish between naturally occurring volatile carbonyl-type compounds in edible oils and those derived from subsequent oxidation.	(Azizian et al., 2015)
EVOO, olive oil, pure seed oils and olive oil samples adulterated with 7 different seed oils in different ratios	- vis–NIR and artificial intelligence model	Analyzing the data produced by the instruments using artificial intelligence methods accurately distinguished between EVOO adulterated with sophisticated techniques and pure EVOO	(Violino et al., 2021)
EVOO blended with soft-refined olive oils	- GC-IMS, FGC E-nose, and SIMCA	Volatile fraction analysis might be the right strategy to overcome the lack of clear and specific process-related markers formed in soft-refinement processes.	(Damiani et al., 2020)

• Abbreviations: FGC E-nose: flash gas chromatography electronic nose; FT-NIR: Fourier transform near infrared; GC-IMS: gas-chromatography ion mobility spectrometry; NTP: non-thermal plasma; PCA: principal component analysis; PLS1: partial least squares; GC-O-MS: gas chromatography–olfactometry‑mass spectrometry; PLSR; partial least squares regression; SECV: standard error of cross-validation; RPOO: refined pomace olive oil; SEP; standard error of prediction; SIFT-MS: selected ion flow tube mass spectrometry; SIMCA: soft independent modeling of class analogy; VOCs: volatile organic compounds.

In summary, a strong reduction of pentanal and hexanal and its derivative compounds (which contribute to the green odor notes) has been observed in adulterated EVOO (Violino et al., 2021). The higher amounts of PUFA in oils other than EVOO (e.g., sunflower oil) result in a higher rate of volatile oxidation than observed for some oxidation markers (Van Durme & Vandamme, 2016). Although the composition of the volatile fraction of EVOO is not a legal requirement to guarantee its authenticity, this approach could be introduced as a useful tool to support EVOO quality.

2.2.6 Pigments

The color of EVOO is attributed to lipophilic chlorophyll and carotenoid pigments present in the olive fruit (Uncu & Ozen, 2020). EVOO contains a rich variety of chlorophyll derivatives (chlorophyll a and b, pheophytin a and b, and other minor derivatives) and carotenoids (β-carotene, lutein, violaxanthin, neoxanthin, and other xanthophylls) (Lazzerini et al., 2017; Uncu & Ozen, 2020). The presence of pigments depends on the olive variety, the stage of fruit ripeness, environmental growing conditions, the extraction process, and storage conditions (Giuffrida et al., 2007). Pigment profiling of EVOO has been applied as an indication of quality and/or authenticity. Using this approach, corn, rapeseed, soybean, peanut, sunflower, refined olive oil, olive pomace oil, and/or old olive oils have been detected as adulterants in EVOO (Ali et al., 2018; Aroca-Santos et al., 2015; Aroca-Santos et al., 2016; Ferreiro-González et al., 2017; Merás et al., 2018; Milanez et al., 2017; Shi et al., 2019; Tan et al., 2018; Uncu & Ozen, 2019).

A clear reduction in chlorophylls and carotenoids in adulterated EVOO has been evaluated by UV–vis, fluorescence spectroscopy and/or FT-IR + UV–vis, which are quick and reliable methods (Ali et al., 2018; Aroca-Santos et al., 2015; Ferreiro-González et al., 2017; Merás et al., 2018; Milanez et al., 2017; Uncu & Ozen, 2019). In this field, stimulated Brillouin scattering combined with UV–vis–NIR (Shi et al., 2019) and front-face fluorescence and visible spectroscopy (Tan et al., 2018) have also been proposed for the authentication of EVOO and the detection of adulteration with other vegetable oils.

Regarding chemometric analysis, PLS, PLSR, partial least squares-Jack-Knife algorithms (PLS-JK), successive projections algorithm-multiple linear regression, stepwise multiple linear regression (SW-MLR), and/or genetic algorithm-multiple linear regression (Ali et al., 2018; Milanez et al., 2017; Tan et al., 2018) have been used. In addition, artificial neural networks (Aroca-Santos et al., 2015), orthogonal projection to latent structures-DA (Uncu & Ozen, 2019), linear-DA (LDA) (Ferreiro-González et al., 2017), parallel factor analysis, and discriminant unfolded PLS (Merás et al., 2018) models have been used to identify and quantify adulteration in EVOO samples.

In summary, EVOO adulteration can be determined by comparing visible absorption spectra. In the visible spectrum of EVOO, certain absorption bands in the range of 430−480 nm and 670 nm stand out owing to the presence of various carotenoid and chlorophyll pigments (Ferreiro-González et al., 2017). Adulteration of EVOO with other seed and/or vegetable oils can be differentiated spectroscopically by the intensity of carotenoids and chlorophylls absorption peaks, as shown in Figure 2.

FIGURE 2

Comparison of pigment profiles in EVOO and seed and vegetable oils, in red pure EVOO, in green EVOO adulterated with refined oils, in blue or grey refined seed and vegetable oils

2.2.7 Sterol, triterpene dialcohol, and stigmastadiene composition

Sterols and triterpene dialcohols are among the parameters used to officially establish the purity of olive oil. Their composition permits discrimination between olive and other oils and between olive pomace oils and nonsolvent-extracted olive oils such as EVOO (International Olive Council, 2021). The addition of other oils will increase the concentration of those compounds and allow then the discrimination between pure EVOO and adulterated EVOOs (Al-Ismail et al., 2010). The official method involves several steps: Compound separation from the saponifiable fraction; partial purification by chromatography; and then derivatization and analysis by GC-FID. As this is time- and reagent-consuming, shorter and simpler alternative procedures have been developed (Gorassini et al., 2019; Mathison & Holstege, 2013; Valli et al., 2021).

Various studies have focused on the analysis of sterols and triterpene dialcohols to classify different EVOOs and detect their adulteration with other oils, including lower quality olive oil or olive pomace oil. Based on the percentages of campesterol and stigmasterol, adulteration with corn, soybean, sunflower, and cotton oils was identified (Al-Ismail et al., 2010), whereas uvaol and erythrodiol revealed the presence of olive pomace oil (Mathison & Holstege, 2013). Δ7-stigmastenol and campesterol proved to be effective markers of EVOO adulteration with sunflower and corn oils, respectively (Jabeur et al., 2014). Another study confirmed this usage of Δ7-stigmastenol as well as campesterol as a marker of adulteration with soybean oil (Youseff et al., 2014). The combination of several parameters, namely total sterol content, desmethylsterol composition, and triterpene dialcohols (erythrodiol and uvaol), was successfully used to identify EVOO adulteration with canola, corn, peanut, safflower, soybean, and sunflower oils, but this strategy failed to detect hazelnut oil (Srigley et al., 2016). The mean values of each sterol allowed EVOOs to be differentiated according to the olive variety and oleaster cultivar, including hybrids (Manai-Djebali et al., 2021). Sterol concentration could differ significantly from different varieties, in the case of Δ5-avenasterol for example, it was reported to be between 2.2% and 15.2% of all sterolic fraction (Manai-Djebali et al., 2018). Finally, a new method that analyses free and esterified sterols and free and esterified triterpenic alcohols was able to detect the adulteration of EVOO with only 2% sunflower oil (Valli et al., 2021).

Studies investigating ways of detecting EVOO adulteration have also targeted sterols together with other compounds. Using ¹H NMR spectroscopy with lipid signal suppression, Ruiz-Aracama et al. (2017) classified EVOOs according to various signals, including those of sterols. A PCA model based on sterol and FA profiles clearly separated EVOO containing cotton and sunflower oils (Kesen, 2019).

Thus, the sterol profile and fatty alcohols have demonstrated to be effective markers for the detection of EVOO adulteration. The official method for their determination has been improved upon, with the development of quicker and more ecological alternatives that require fewer sample preparation steps and consume less reactant to achieve similar results (Mathison & Holstege, 2013; Tena et al., 2015).

On the other hand, detection of stigmastadienes has been proved to be a highly sensitive method to notice addition of refined oils to EVOO. Virgin olive oils (VOOs) obtained by cold pressing do not contain enough quantity of these molecules to be measured (less than 0.01 mg/kg³), and EVOOs are defined by Regulation to contain less than 0.15 mg/kg³ of stigmastadiene (Uncu & Ozen, 2020). However, during the refining process, the high temperatures to which the oils are exposed lead to the formation of 3,5-stigmastadiene by the dehydration of sterols, particularly of beta-sitosterol, in measurable amounts ranging from 0.3 to 0.9 mg/kg³ (Gordon & Firman, 2001). Concretely, the existence of stigmastadienes indicates the use of bleaching clay or high-temperature applications executed during the deodorizing of the refining process. Thus, the detection of these steroidal hydrocarbons in VOOs is legally accepted to reveal adulteration of the product by the presence of refined vegetable oils (olive pomace, soybean, sunflower, palm, etc.) (International Olive Council, 2017a; Schneider, 2016).

Regarding the stigmastadienes detection, Crews et al. (2014) described a test which not only has been shown to be more rapid and easier to apply than other already standardized alternatives, but also is be a method that can detect other sterenes in oils. This method presents a high sensitivity to low levels of stigmastadienes. And even though it does not indicate the exact concentration of refined oil existing, mixtures of less than 5% refined oils can be revealed.

2.2.8 Oil fingerprint applied to the authentication

In recent years, chromatographic and related techniques with spectroscopic detection or coupled to MS, and combined with chemometrics, have proven to have an exceptional capacity to discourse complex food authentication issues by fingerprinting approaches (Medina et al., 2019). In this context, flow injection analysis coupled to high-resolution MS (FIA–HRMS), using a fingerprinting strategy and combined with PCA, PLS-DA, and SIMCA was used to distinguish olive oil from other vegetable oils, as well as to perform an evaluation of its quality category (Campmajó et al., 2022). After external validation, remarkable classification accuracies were reached. Moreover, putative identification of the most common ions was performed by HRMS, allowing excellent discrimination of olive oil in front of the other vegetable oil samples using PCA. As mentioned before, a similar approach based on TAG profile was satisfactorily used to detect high linoleic and high oleic vegetable oils in EVOO (Quintanilla-Casas et al., 2021).

On the other hand, using chromatographic fingerprints (HPLC coupled to charged aerosol detector and high-temperature GC coupled to FID) coupled to multivariate techniques was applied to authenticate the geographical origin of EVOO without identifying or quantifying the chemical compounds (Vera et al., 2019). The results were best when fingerprints from the data from the two techniques were combined applying low-level data fusion and PLS-DA was employed as the classification procedure. Similarly, Quintanilla-Casas et al. (2020) established fingerprinting as a more efficient approach than profiling sesquiterpene hydrocarbons to classify EVOO according to its origin.

2.2.9 Proposed markers for EVOO adulteration detection

During EVOO adulteration, FA profile and minor compounds profile are changed. Phenolic compounds and volatile compounds decrease, while waxes such as campesterol increase. Hence, monitoring several compounds could be useful to detect adulteration. This is easily achieved by recording EVOO fingerprint with HRMS. Nevertheless, this approach requires expensive equipment that is not always available. Another way of taking fingerprints could be monitoring the absorbance or emission spectra in the UV–vis range or using the FT-NIR spectroscopy. However, only active compounds in the NIR or the UV–vis will be detected; hence, less information will be obtained compared to the HRMS analysis and it might not be enough for the detection of adulterants.

EVOO is commonly adulterated with refined oils, such as rapeseed or hazelnut, or with solvent extracted oils, such as pomace olive oil. During the deodorization that takes place in the refinement process, the oil is heated at 180°C or even higher temperatures. This step is present in both chemical and physical refinements (Varona et al., 2021). Thus, analyzing markers of heating, such as stigmastadienes or glycyl esters (GE) (Gordon & Firman, 2001; Kamikata et al., 2019), could be a feasible approach to detect EVOO adulterations with refined oils (Crews et al., 2014). On the other hand, for detecting adulteration with solvent extracted oils, analyzing the sterolic profile could be a useful approach, as the concentration of those compounds will increase (Mathison & Holstege, 2013).

In conclusion, when HRMS is not available, analyzing both stigmastadienes and sterols could be a useful and cheap method to detect EVOO adulteration. As those compounds are analyzed for olive oil categorization (European Community, 1991), no extra equipment is needed for carrying out those determinations.

2.3 Health issues derived from EVOO fraud

The health benefits associated with EVOO consumption have been extensively demonstrated, particularly its protective effect against cardiovascular (CV) diseases (Covas et al., 2006; Estruch et al., 2018; Guasch-Ferré et al., 2014). This positive impact has been verified by numerous intervention studies focusing on specific markers of CV disease, which show that EVOO intake can prevent or reduce the inflammatory processes associated with chronic-degenerative conditions such as CV-cerebral diseases and cancer (Casas et al., 2017), and benefits plasma lipid levels and lipid oxidative damage (Covas et al., 2006). EVOO supplementation also leads to an improvement of postprandial glucose and lipid profiles, an antiatherosclerosis mechanism that can reduce the risk of developing diabetes (Carnevale et al., 2014). Diabetes prevention could be attributed to the antioxidant property of EVOO (Bullo et al., 2011), since oxidative stress appears to be involved in β-cell dysfunction and downregulates insulin secretion. Aside from investigating whether EVOO had any effect on the lipid profile, the Carnevale research group examined whether postprandial glycemic control occurred using an oxidative stress-mediated mechanism, demonstrating that a Mediterranean-type meal supplemented with EVOO is associated with reduced postprandial oxidative stress generated by NOX2 (Violi et al., 2015). Thus, the consumption of EVOO represents a simple but effective nutritional approach to modulating the deleterious effect of different CV risk factors on the vascular system, chiefly oxidative stress, inflammation, postprandial hyperglycemia, and hyperlipidemia (Nocella et al., 2018).

Moreover, EVOO intake is inversely associated with the prevalence of cancer (Psaltopoulou et al., 2011), as reported in a meta-analysis of 19 observational studies involving approximately 35,000 individuals and carried out within 10 years. More recently, a randomized trial found that women who adhered to a Mediterranean diet supplemented with EVOO had a 62% lower incidence of invasive breast cancer than a control group advised to restrict dietary fats (Toledo et al., 2015). Other health properties of EVOO are related to neuroprotection. Pitt et al. (2009) reported that low doses of oleocanthal, one of the main secoiridoids of EVOO, structurally altered amyloid-β oligomers, which play a role in the development of Alzheimer's disease. Moreover, Batarseh and Kaddoumi (2018) reported that high-oleocanthal EVOO reduced the amyloid-β load and related toxicity in a mouse model of Alzheimer's disease. Additionally, Li et al. (2009) found that oleocanthal inhibited tau fibrillization. Other phenolic compounds present in EVOO, hydroxytyrosol and oleuropein, also showed neuroprotective effects in the prevention of Parkinson's disease, as they can interfere in the pathophysiology of the disease by reducing the accumulation of neurotoxins, oxidative stress, and impair autophagy (Achour et al., 2016; Goldstein et al., 2016). Finally, EVOO has shown gut-modulating activity, by increasing the growth of beneficial bacteria and diminishing pathogenic bacteria (Millman et al., 2021).

The health-enhancing properties of olive oil are lost when its composition is altered by industrial practices focused on maximizing profits as is shown in Figure 3. Adulteration with other vegetable oils can result in a product of poor nutritional quality that oxidizes more readily and instead of offering health benefits, it can even be harmful. Consumption of rancid oil, due to the prooxidant substances present can induce oxidative stress and dysfunction of the vascular endothelium, which plays an essential role in the pathophysiology of several diseases (Carnevale et al., 2014; Casas et al., 2017; Nocella et al., 2018). Additionally, the sustained consumption of trans or saturated fats, which include many refined oils (Astrup et al., 2020; Liu & Lu, 2018), is also a health risk. When coupled with a sedentary lifestyle and other bad habits, it increases the likelihood of suffering from arteriosclerosis, myocardial infarctions, or embolisms (Carnevale et al., 2014; Casas et al., 2016, 2017).

FIGURE 3

General description of health benefits of extra virgin olive oil and health risk derived from the consumption of adulterated extra virgin olive oil

During the refining process of olive oil (the deodorization step), the content of 3-monochloropropane-1,2-diol (3-MCPD), 2-monochloropropane-1,3-diol esters (2-MCPDE), and GE was found to increase (Hung et al., 2017; Kamikata et al., 2019). As the processing of EVOO does not require the use of high temperatures, it is not expected to contain these substances at quantifiable levels. Therefore, 3-MCPDE, 2-MCPDE, and GE have been proposed as complementary indicators of EVOO adulteration (Kamikata et al., 2019; Weesepoel et al., 2021). The mixing of EVOO with refined oils can pose a threat to consumer health if it results in an increase in these compounds. GE and 3-MCPD are classified, respectively, as “probable human carcinogen” (category 2A) and “possible human carcinogen” (category 2B) by the International Agency for Research on Cancer (IARC, 2000). The European Food Safety Authority (EFSA) CONTAM Panel established a tolerable daily intake of 0.8 μg/kg body weight per day for 3-MCPD (EFSA Panel on Contaminants in the Food Chain [CONTAM], 2016).

The presence of polycyclic aromatic hydrocarbons, besides being related with EVOO fraud, is also a substance with potential health risk. These genotoxic and carcinogenic compounds (IARC, 2006; EFSA, 2008; World Health Organization, 2005) are formed during incomplete combustion or pyrolysis of organic matter, and their presence in oils is attributed to the drying process to which seeds, grains, and olive pomace are subjected during processing (Tfouni et al., 2017).

Finally, Arlorio et al. (2010) demonstrated that the adulteration of EVOO with hazelnut oil introduces a potential risk for consumers with hazelnut allergies, after analyzing the allergen and protein content of EVOO being deliberately spiked with raw cold-pressed hazelnut oil or solvent-extracted hazelnut oil. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) analysis confirmed the presence of hazelnut proteins in solvent-extracted hazelnut oil, which were still detectable at a 1% contamination level in solvent-extracted hazelnut oil-spiked EVOO, and therefore showing the potential health risk for sensitized people.

3 TRACEABILITY

Product traceability refers to the ability to monitor a product through all the stages of the production chain from its origin to the final destination. This process allows the product to be registered and identified, thereby guaranteeing its quality and protecting the consumer (Melucci et al., 2016). Due to the growing demand for high-quality EVOO, its characteristics and composition need to be specified, including the geographic origin and variety of the olives (Melucci et al., 2016; Violino et al., 2019).

Monovarietal olive oils are distinguished by a recognizable taste associated with a particular cultivar. In contrast, coupage olive oils are produced from a blend of olive varieties, with the main objective of obtaining an exotic and singular aroma and flavor (Campestre et al., 2017).

The EU Regulation 2081/92, which was established to protect and promote quality food products, specifies that EVOO must be labeled according to its origin. Accordingly, oils can be registered within different schemes, as shown in Table 3: PDO, protected geographical indication (PGI), and traditional specialty guaranteed (TSG) (Violino et al., 2019).

TABLE 3. General regimen for food and other agricultural products based on Regulation 510/2006

General regimen	Origin	Characteristics	Restriction
Protected designation of origin (PDO)	Region, specific place, or country	Quality essentially or exclusively due to a particular geographical area.	Produced, processed, and prepared in a given geographical area.
Protected geographical indication (PGI)	Region, specific place, or country	Slightly less strict; food reputation of a product from a given region is sufficient.	One of the stages of production, processing or preparation takes place in the area.

• Source: Data take from (European Union, 2006).

To qualify for a PDO, the olive oil must comply with specific requirements regarding the geographic origin, cultivar, organoleptic characteristics, production methods, and agronomic practice (Giménez et al., 2010). Certification and denomination require EVOO traceability to be established, focusing on the region of the olive tree (geographic traceability) and the cultivar (botanical traceability).

3.1 Geographic and botanical traceability

Olive oil from different geographic sources can vary considerably in its components, reflecting variable factors such as the growth environment, climate, soil, and water quality. The Commission Implementing Regulation 2013 EU No. 1335/13 stipulates that the location of the olive harvest must be stated on EVOO labels (European Union, 2013a), as well as olive oil labels, for example, “blend of olive oils (not) of EU origin” or “blend of olive oils of EU origin and not of EU origin” (Melucci et al., 2016).

Different strategies have been developed to characterize EVOO from several origins to detect fraudulent practices. Instead of analyzing one or a group of marker compounds, researchers have focused in analyzing whole spectra of techniques such as infrared or mass spectroscopy. These techniques usually require few or non-pretreatments steps; however, chemometrics is needed to process all the data.

Although few studies on NIR and FTIR spectroscopies as non-invasive techniques for food authentication have been published to date, the results indicate they are robust and reliable methods (Galtier et al., 2007; Garrido-Varo et al., 2017; Tapp et al., 2003). For example, NIR spectroscopy categorized 125 olive oils into five geographically close designations of origin in France (Galtier et al., 2007). This study demonstrated that the origin of olive oil can be traced by analyzing NIR spectra, which is related to the FA and TAG profiles, without the need for time-consuming physical and chemical procedures to analyze those compounds. In the same line, two models (spinning and static sample presentation) were developed to predict and classify olive oil quality parameters (Garrido-Varo et al., 2017). In another study, FTIR spectroscopy in combination with multivariate analysis was also carried out to distinguish EVOOs from different geographic origins (Tapp et al., 2003).

The traceability of the geographic origin of EVOOs by PCA and DA of the headspace volatile profile as a fingerprint has also been reported (Melucci et al., 2016). In this preliminary study, an FGC E-nose accurately determined the geographic origin of EVOO (100% Italian versus non-100% Italian), and the results indicate it is suitable for the rapid screening of commercial EVOO to verify if the information declared on the label is correct. Similarly, Quintanilla-Casas et al. (2020) analyzed EVOO samples from seven countries by headspace solid phase microextraction-GC-MS (HS-SPME-GC-MS) combined with chemometrics. Sesquiterpene hydrocarbons proved to be accurate geographic markers, and fingerprinting was established as a more efficient approach than profiling to classify EVOO according to its origin. Further experiments of the same group showed that the sesquiterpene hydrocarbon fingerprints permitted the proper characterization of origin of EVOOs from and not from theEU (Quintanilla-Casas et al., 2022).

Apart from geographical traceability, botanical traceability is important to assess EVOO trueness. Botanical traceability of EVOO is the ability to identify the olive variety, whose characteristics depend on the type of soil, growing conditions, and climate adaptability (Montealegre et al., 2010). Although PDO verification is useful for EVOO classification and certification, its accomplishment is not always straightforward. Olive cultivars used to be distinguished by morphological and pomological traits, an approach limited by the influence of external uncontrolled factors (Sanz-Cortés et al., 2003). More efficient methods involve compositional analysis and genetic markers (Montealegre et al., 2010). The most relevant compositional markers for establishing the botanical traceability of olive oils are summarized in Table 4. As the chemical composition of EVOO is strongly affected by environmental conditions, fruit ripening, and the extraction technology, the botanical origin cannot be identified by a single compositional marker (i.e., FA, phenolic compounds, volatile compounds, pigments, etc.), and instead several parameters are analyzed together using chemometric tools.

TABLE 4. Relevant compositional markers for establishing the botanical traceability of olive oils

Compositional marker	Relevant chemical marker analysed	Olive/VOO variety example	References
FA	C18:0, C18:1	Cornicabra	(Aranda et al., 2004)
	C16:0, C17:1	Arbequina
	C17:1 and C18:0	Hojiblanca
	C16:0, C18:0, C18:1	Picual
TAG	OOO and SLO + POO	Manzanilla Cacereña	(Diaz et al., 2005)
	SOO, LOO, and PLO	Non manzanilla Cacereña
Sterols	Stigmasterol	Cobrançosa	(Alves et al., 2005)
	β-sitoesterol/ Δ^5-avenasterol	Madural
	Campesterol	Verdeal
Phenolic compounds	p-HPEA-EDA and ligstroside aglycon	Cornicabra	(Gómez-Alonso et al., 2002)
	1-Acetoxypinoresinol + trans-cinnamic acid and 3,4-DHPEA-AC	Arbequina
	3,4-DHPEA-AC	Hojiblanca
	ligstroside aglycon	Picual
Volatile compounds	(E)-Hex-2-enal	Bidh Hman, Rekhami, Jarboui 1, Regregui	(Haddada et al., 2007)
	Ethanol, 2-Methylpentane, (E)-Hex-2-enol, and Hexanol	Jarboui 2
	(Z)-pent-2-enol and two isomers of 3,4-diethylhexa-1,5-diene	Ain Jarboua, Chétoui 1 and 2, Neb Jmel
Pigments	Lutein/β-carotene ratio = 0.27	Cerasuola	(Giuffrida et al., 2007)
	Lutein/β-carotene ratio = 0.4	Nocellara
	Lutein/β-carotene ratio = 0.17	Biancolilla
Hydrocarbons	C29:0 to C34:0	Cacereña	(Bueno et al., 2005)
	C13:1	Carrasqueña
	C24:0 to C29:0, C13:1	Corniche
	C13:1	Picual

• Abbreviations: FA: Fatty Acid; TAG: Triacylglycerol; L, O, P, and S: linoleoyl, oleoyl, palmitoyl, and stearoyl fat acid radicals, respectively; VOO: virgin olive oil; 3,4-DHPEA-AC: hydroxytyrosol acetate.

Aranda et al. (2004) classified four Spanish cultivars (Cornicabra, Arbequina, Hojiblanca, and Picual) according to their distinct FA and triglyceride profiles and according to their distinct FA in the 2-position in the TAG profiles. For analyzing the FA in the 2-position, several sample preparation steps are needed, while the FA acid profile is quicker to obtain. The differences in the total FA profile permitted a successful discrimination with easier analysis methodology than the FA in the 2-position. Sterol composition together with chemometrics allowed the discrimination between three Portuguese cultivars (Cobrançosa, Madural, and Verdeal) (Alves et al., 2005). However, sterol composition alone was not enough to distinguish EVOOs from Manzanilla Cacereña cultivar from EVOOs from other cultivars (Diaz et al., 2005).

Different varieties will present different enzymatic activities, which result in different volatile and phenolic profiles; those have also been used to discriminate between varieties (Gómez-Alonso et al., 2002; Haddada et al., 2007). In addition, tocopherols, pigments, and hydrocarbons are also affected by the olive enzymes, which depend on the cultivar; hence, those markers have been used successfully as well as discriminants between different olive cultivars (Baccouri et al., 2007; Bueno et al., 2005; Giuffrida et al., 2007).

In spite of all that, the environmental and technical factors have such a high impact on EVOO composition (Olmo-Cunillera et al., 2021) that the best approach for assessing EVOO botanical origin are the genetic markers. Although olive oils have a low protein content, peptide separation by capillary electrophoresis has proved to be an effective method for the differentiation of monovarietal olive oils. However, UV detection of the protein profiles, as well as protein isolation and solubilization, needs to be performed first (Monasterio et al., 2013; Montealegre et al., 2010). This approach has been further developed by coupling capillary electrophoresis with MS, which constitutes a reliable and rapid method to assess EVOO authenticity and quality (Monasterio et al., 2013; Sánchez-Hernández et al., 2011).

4 CONCLUSIONS

In conclusion, analyzing the whole EVOO fingerprint seems to be the best approach to detect EVOO adulteration; however, expensive equipment (HRMS) is needed. Then, both stigmastadienes and sterolic profile are proposed as markers for detecting EVOO blended with refined oils and solvent extracted oils.

Nowadays, international consumer demand for EVOO is growing, and fraudulent practices to maximize profits by reducing oil quality are also on the rise. These deceitful practices not only reduce the quality of the oil and consequently its beneficial effects on health, but also may pose a significant risk due to the ingestion of toxic, carcinogenic or allergic substances. In the fight against fraud, the EU, IOC, and Codex Alimentarius have established several standardized regulations that EVOO producers need to comply with. These endeavors are supported by scientific research focused on developing and improving strategies and novel analytical technologies that can identify possible adulterations, such as mixing EVOO with refined olive oil or other edible oils.

In addition, analytical methods such as FTIR or GC-MS can also authenticate the product by determining either the geographic or botanical origin of the oil, and tracing all the stages in the production chain.

ACKNOWLEDGMENTS

This research was funded by CICYT [AGL2016- 75329-R]; it is part of the project [PID2020-114022RB-I00], funded by MCIN/ AEI/10.13039/501100011033; CIBEROBN from the Instituto de Salud Carlos III, ISCIII from the Ministerio de Ciencia, Innovación y Universidades, (AEI/FEDER, UE) and Generalitat de Catalunya [2017SGR 196]. Julián Lozano-Castellón and Inés Domínguez-López thank the Ministry of Science Innovation and Universities for the FPI and FPU contracts, respectively [BES-2017-080017] and [FPU20/02478]. Anna Vallverdú-Queralt thanks the Ministry of Science Innovation and Universities for the Ramon y Cajal contract [RYC-2016-19355].

AUTHOR CONTRIBUTIONS

Julián Lozano-Castellón: Data curation; investigation; visualization; writing – original draft; writing – review & editing. Anallely López-Yerena: Data curation; investigation; visualization; writing – original draft; writing – review & editing. Inés Domínguez-López: Data curation; investigation; visualization; writing – original draft; writing – review & editing. Aina Siscart-Serra: Formal analysis; investigation; writing – original draft. Nathalia Fraga: Formal analysis; investigation; writing – original draft. Samantha Sámano: Formal analysis; investigation; writing – original draft. Maria Pérez: Conceptualization; formal analysis; investigation; methodology; project administration; supervision; writing – original draft; writing – review & editing.

CONFLICT OF INTEREST

Rosa María Lamuela-Raventós declares to receive lecture fees from Cerveceros de España and lecture fees and travel support from Adventia. The other authors declare no conflict of interest.