Spices Volatilomic Fingerprinting—A Comprehensive Approach to Explore Its Authentication and Bioactive Properties

Volatile organic metabolites (VOMs) present in different spices can provide distinct analytical biosignatures related to organoleptic properties and health benefits. This study aimed to establish the volatilomic fingerprint of six of the most consumed spices all over the world (saffron (Crocus sativus L.), cinnamon (Cinnamomum verum), cumin (Cuminum cyminum L.), black pepper, (Piper nigrum L.), sweet paprika (Capsicum annuum L.), and curry (a mix of different herbs and spices)). Based on headspace solid phase microextraction (HS-SPME) followed by gas chromatography-mass spectrometry (GC-MS) analysis, this is a powerful strategy to explore and establish the spice’s volatile pattern and unravel the potential health benefits related to the most important VOMs identified in each spice. This comprehensive knowledge will help in the definition of their authenticity, while simultaneously protecting against potential frauds and adulterations. A total of 162 VOMs were identified. Semi-quantitative assessments revealed that terpenoids and sesquiterpenoids amounted to the major volatile class in the investigated spices, except for cinnamon, where carbonyl compounds are the major group. Most of the studied spices comprised key characteristics of aroma and health bioactive compounds, e.g., dihydrojuneol in saffron, cinnamaldehyde in cinnamon, cuminaldehyde in cumin and curry, and caryophyllene in black pepper. The principal component analysis (PCA) and partial least-squares discriminant analysis (PLS-DA) successfully discriminated the investigated spices, being α-cubebene, 3-methyl butanal, β-patchoulene and β-selinene, the most important VOMs (highest VIP’s) that contributed to its discrimination. Moreover, some VOMs have a high influence on the spice’s bioactive potential, helping to prevent certain diseases including cancer, inflammatory-related diseases, diabetes, and cardiovascular diseases.


Introduction
Since antiquity, spices have been used worldwide for a variety of purposes. From natural flavoring and preservative agents in beverages, pharmaceuticals, and foods, to ingredients in cosmetics and perfumes, to their use as preventive agents with potential benefits in human health [1][2][3].
Its composition includes the presence of natural phytochemicals such as polyphenols, volatile organic metabolites (VOMs), vitamins and sulfur-containing compounds, among others. These natural phytochemicals, together with the complementary actions they can exert on human health and their bioactive potential (antioxidant, antimicrobial, antiviral, anti-inflammatory, antidiabetic, anti-obesity, anti-pyretic, antihypertensive, and antidepressant effects, cytochrome and other enzyme inducers, reducing induction and advancement of cancer cell development, and cardio and neuroprotective effects), helps to of different polyphenols, ascorbic acid, and carotenoids, which play a key role in preventing certain diseases including cardiovascular diseases and some kinds of cancer [16,17]. In addition, it possesses other bioactive compounds (capsaicin, curcumin, tocopherol, lutein, carotene, capsanthin, and quercetin) with potential antioxidant and anti-inflammatory effects [13,16,17].
Curry is a mixture of different herbs and spices, including chili, basil, fennel, celery, saffron, cinnamon, cardamom, dried onion, coriander, cloves, cumin, turmeric, fenugreek, ginger, mustard, nutmeg, pepper, cayenne or tamarind. Currently, curry is used in cuisines all over the world, with its origins being in the Asian continent. Its composition can vary, which is why curry can contain several bioactive compounds. However, turmeric is usually found in large proportions in the composition of curry, where curcumin is the main bioactive compound. In addition to conferring the typical yellow color, curcumin is responsible for the various pharmacological effects that help to fight against some cardiovascular diseases and cancer [13,15].
The extremely high market prices of most spices (for instance, saffron can reach 50 M €/Kg), increases the adulteration susceptibility of the original product. In addition, due to its nutritional value, high demand, and associated potential health benefits, a comprehensive insight into the characteristic secondary metabolites responsible for its authenticity and quality is of paramount importance as a powerful strategy for the prevention and detection of potentially fraudulent activities. In this context, gas chromatography-mass spectrometry (GC-MS) is considered the gold standard instrumental technique for VOMs analysis in a wide range of samples [18]. However, the previous sample preparation is fundamental in concentrating the VOMs and removing interferences, particularly from complex samples, as food matrices [19]. Several conventional analytical extraction techniques, such as solid-liquid extraction (hydro-distillation), supercritical fluid extraction (SFE), Soxhlet and others, can be employed for the extraction and pre-concentration of the VOMs from food matrices. However, these techniques present some drawbacks since they require significant amounts of organic solvents and samples, they are time-consuming and require exhaustive concentration steps, which can generate artefacts leading to an inaccurate elucidation of the volatiles [20]. In this context, headspace solid-phase microextraction (HS-SPME) is a well-established technique that has been successfully applied in the characterization of the volatile profile of new fruits and vegetables [21,22]. This technique requires relatively short extraction times and minimal sample handling, combining extraction and concentration in a single step. In addition, it does not require the use of potentially harmful organic solvents, making it an environmentally and analyst-friendly extraction technique. Moreover, its combination with GC coupled to a mass spectrometer with a quadrupole inert mass selective detector (qMS) retrieves high sensitivity, reproducibility, and robustness. In addition, a few studies applied multivariate statistical data analysis to chromatographic data sets as a powerful and fast strategy to obtain volatile fingerprints and discriminate between the different food matrices [21][22][23].
Therefore, the main purpose of this work was to establish the comprehensive volatilomic profile of five of the most consumed spices worldwide, namely saffron, cinnamon, cumin, black pepper and sweet paprika, in addition to curry, using HS-SPME/GC-qMS to examine the dominant VOMs and evaluate the diversity and similarity in the VOMs patterns. The definition of the volatilome of each spice will allow for a better understanding of its ability in authenticity determination and quality control, and therefore the detection of potential adulterations which affect the genuine volatile profile. The combination of the chromatographic data set with multivariate statistical data analysis (MSDA) was also used to extract useful information concerning the spice's quality and authenticity. This approach provides insights into understanding the chemistry behind the bioactive and flavor properties of these spices, consequently providing consumers with quality and safety guarantees.

Volatomic Fingerprinting from Spices
Due to high demand, nutritional value, and the sui generis typical aroma highly appreciated by consumers, a volatilomic-based analytical approach was used to establish the volatilomic fingerprint of each spice as a useful strategy to assess and define its quality and authenticity. Figure S1 (Supplementary Material) shows the typical chromatograms of the investigated spices, obtained by HS-SPME DVB/CAR/PDMS /GC-qMS. Their chromatographic profiles show that the concentration range of the volatile compounds differs enormously for the various spices. For instance, black pepper presents a very complex chromatogram with most volatiles at high concentrations (based on peak area intensity), whereas the volatile compounds identified in sweet paprika shows relatively low concentrations ( Figure S1).
In total, 162 VOMs were identified, most of them in black pepper (69), while in other spices the number of identified VOMs ranged from 49 (cumin) to 59 (cinnamon). The identified VOMs belong to several chemical groups including sesquiterpenoids (66), terpenoids (48), carbonyl compounds (14), and hydrocarbons (12). At a lower extent were identified eight esters, six volatile phenols, three furanic compounds, two higher alcohols, two nitrogen compounds and one sulfur compound ( Figure 1A). Figure 1B shows the distribution of the total volatile fraction for each investigated spice. The sui generis character of spices and other foods is influenced by the composition, the presence of key aroma compounds, and concentrations which can determine the kind of aroma perceived [24]. The major VOMs chemical class in the current study ( Figure 1A) is the class of terpenoids, which are widely produced in the plant kingdom [25], followed by sesquiterpenoids, and to a lower extent by carbonyl compounds.
The predominant VOMs present in the top 10 investigated spices include α-curcumene and dihydrojuneol in saffron; cinnamaldehyde and copaene in cinnamon; cuminaldehyde and γ-terpinene in cumin; caryophyllene and 3-carene in black pepper; p-cymene and γ-terpinene in sweet paprika; and cuminaldehyde and linalool in curry.
The detailed list of all VOMs identified in the analyzed spices and their respective obtained experimental data, including retention times, KIs, molecular formula (MF), chemical families, and relative peak area, is shown in Table 1. Of the 162 VOMs identified, only eight are common to all the spices analyzed-α-and β-pinene, limonene, p-cymene, copaene, caryophyllene, β-bisabolene, and cuminaldehyde. On the other hand, some VOMs, described in Table S1, were identified only in a certain type of spice (Table S1).

Bioactive Potential of VOMs Identified in the Investigated Spices
It is well known that the spices investigated in this work have long been used to add flavor to the foodstuffs and to enhance the quality of the food. In addition, many of them also act as excellent preservatives which increase the shelf life of foods by delaying the spoilage process [26]. Such spices are also a rich source of biologically active compounds, which have antioxidant, antimicrobial, anti-inflammatory, antidiabetic, cytotoxic, antiproliferative, anticancer and even anti-HIV properties, among others [4,13,32]. These properties are beneficial to human health and help to fight several ailments of the human body.
The VOMs are part of such biologically active compounds and have a high influence on the spice's bioactive potential. That is why many of the VOMs identified in the spices analyzed have been previously reported in other food samples, with a significant number of biological activities such as those mentioned above [33,34]. Therefore, these VOMs could help prevent and treat diseases like cancer [35], inflammatory diseases [36], diabetes [37], and cardiovascular diseases [38], among others. Table 2 shows the potential bioactive effects of some VOMs identified in the spice samples under investigation.
Among these VOMs, terpenoids are one of the most dominant chemical families found in fruits and vegetables [39], which agrees with the results obtained in this work. Terpenoids, biosynthesized through isopentyl diphosphate, and the methylerythritol phosphate and mevalonate pathways are the major contributors to the total volatile composition of cumin, curry and sweet paprika ( Figure 1A). In turn, sesquiterpenoids are the major contributors to the total volatile composition of saffron and black pepper and are abundantly present in the total volatile fraction of cinnamon, where carbonyl compounds are more abundant ( Figure 1A). Therefore, terpenoids and sesquiterpenoids are the ones that contributed most to the characterization of the total volatile profile of the spices investigated. Monoterpenes have several beneficial biological effects reported in several works [36,41]. Terpenoids, such as α-pinene (#15), sabinene (#23), (+)-3-carene (#25), β-myrcene (#27), α-phellandrene (#28), α-terpinene (#30), limonene (#31), β-phellandrene (#32), p-cymene (#42), linalool (#68), α-terpineol (#107), cuminaldehyde (#130) and thymol (#156), possess high potential antioxidant, antimicrobial, antibacterial, anti-inflammatory, antidiabetic, neuroprotective and immunostimulant properties [10,36,38,41,42]. However, the most remarkable bioactivity of terpenoids is related to their anticancer potential, acting at different stages of tumor development and in different mechanisms of action (inhibition, regulation of intracellular signaling pathways) [42,43]. Limonene was the third and fourth major VOM, in terms of relative peak area, found in the black pepper and sweet paprika samples, respectively. The chemopreventive and chemotherapeutic properties of limonene against human cancers were widely demonstrated by Paduch et al. [44] and Kris-Etherton et al. [45]. Cuminaldehyde is an oxidized aldehyde monoterpene found as a major VOM in the cumin and curry spices analyzed. The relative peak area obtained for this compound in the cumin sample was very remarkable (9694.8), being more than twelve times greater than that determined in the curry sample (Table 1). Cuminaldehyde was also identified in the sweet paprika sample, although in lower amounts. Therefore, the results obtained about cuminaldehyde agree with those reported by Ebada [11], with cuminaldehyde being the main bioactive compound of cumin and presenting many of the bioactive properties cited previously [10]. On the other hand, several sesquiterpenoids also stood out in the volatile composition of some of the spices investigated. For example, dihydrojuneol (#161) and α-curcumene (#129) were the most abundant VOMs, in terms of relative peak areas, in the saffron sample, and caryophyllene isomer (#87), in black pepper (Table 1). Even though in smaller relative peak areas, dihydrojuneol was also found in curry, α-curcumene was identified in the cinnamon and curry samples, and caryophyllene isomer was also present in the volatile profile of saffron and sweet paprika (Table 1). Even though they were not the most abundant VOMs in the cinnamon sample, the sesquiterpenoids copaene (#62), δ-cadinene (#126), and αmuurolene (#118) were identified in notable quantities, representing over 20% of the total volatile profile of this spice (Table 1). Allyl isothiocyanate (#49), belonging to the chemical family of organosulfur compounds, was identified in the analyzed curry sample. These compounds are highly reactive phytochemical metabolites and are usually present in the composition of cruciferous vegetables [46]. In addition, they also have interesting properties and bioactivities, mainly antibacterial, antiproliferative, cytotoxic and anticancer effects.
On the other hand, carbonyl compounds are present in all the analyzed spices, in greater or lesser amounts, but standing out above all in the cinnamon sample, where the contribution of carbonyl compounds to the total volatile profile was higher than 63%. Cinnamaldehyde (#150) was the most prominent carbonyl compound identified in cinnamon and, although in smaller amounts, was also identified in cumin. In the case of the cinnamon sample, the amount of cinnamaldehyde was very remarkable (6996.9 relative peak area), being 83 times the amount of cinnamaldehyde present in the cumin sample (Table 1). This result obtained in our work agrees with that stated by Singh et al. [4] in their research, where an extensive review of the phytochemical and pharmacological properties (antimicrobial, antioxidant, anti-inflammatory, anticancer, antidiabetic, anti-HIV, among others) of cinnamon is carried out, concluding that cinnamaldehyde is a main and characteristic VOM of cinnamon, and is also responsible for its typical aroma.
The obtained p values by one-way ANOVA with Fisher post-hoc test (p < 0.001) indicated that 94 of 162 VOMs identified were significantly different among the investigated spices. Moreover, HCA was also performed using the 15 most significant VOMs identified in the spice samples obtained by ANOVA. The resulting dendrogram associated with the heat map was performed by Euclidean distance through Ward's clustering method (Figure 4), providing intuitive visualization of the data set, which along with the statistical analyses carried out previously, allows better identification of the inherent clustering patterns between each spice. Furthermore, these chemometric analyses are an excellent tool for authentication and quality control of the investigated spices.

Chemical and Reagents
Ultrapure water was obtained from a Milli-Q ® system (Millipore, Bedford, MA, USA). Internal standard (IS) 3-octanol and sodium chloride (NaCl, 99.5%) were obtained from Sigma-Aldrich (Madrid, Spain), whereas the GC carrier gas, helium of purity 5.0 was obtained from Air Liquide, Portugal. The SPME fiber coated with divinylbenzene/carboxen /polydimethylsiloxane (DVB/CAR/PDMS) (50/30 µm), SPME holder for manual sampling and glass vials were purchased from Supelco (Bellefonte, PA, USA). The alkane series, C8 to C20, with a concentration of 40 mg/L in n-hexane used to determine the kovats index (KI) was supplied from Fluka (Buchs, Switzerland).

Spice Samples
Five types of spices were selected for analysis i.e., saffron (Crocus sativus L.), cinnamon (Cinnamomum verum), cumin (Cuminum cyminum L.), black pepper (Piper nigrum L.) and sweet paprika (Capsicum annuum L.) and curry, a mixture of herbs and spices. A total of 18 samples, 3 packages per spice type, were collected in retail outlets in Funchal (Madeira Island, Portugal). There was no information available about how these spices were processed. The spices were purchased already powdered, packed in plastic bags, and stored in the dark at room temperature until analysis by HS-SPME/GC-qMS.

HS-SPME Procedure
A time-effective and solventless (HS-SPME) method was applied for VOMs extraction. With slight modifications, the HS-SPME procedure was performed based on the conditions described by Figueira et al. [22]. For headspace sampling, 1 g of spice powder, 0.3 g of NaCl (to promote the "salting-out" effect by decreasing the solubility of volatile metabolites in the water-based phase) and 6 mL of ultra-pure Milli-Q water were placed into a 20 mL amber headspace glass vial containing a magnetic stirring microbar. Before sealing the vial with a PTFE-faced silicone septum, 5 µL of 3-octanol (102 µg/mL) used as an internal standard was added. Then, the vial was placed in a thermostatic bath at 45 ± 1 • C under constant magnetic stirring (450 rpm). HS-SPME extractions were carried out by exposing the SPME fiber (DVB/CAR/PDMS) to the sample's headspace for 50 min. Finally, the fiber was withdrawn into the holder needle, removed from the vial, and the VOMs extracted by SPME were thermally desorbed by the direct insertion of the fiber into a GC injector at 250 • C for 6 min, in splitless mode. Experiments were carried out in triplicate (n = 3) for all samples. The SPME fiber was thermally conditioned according to the manufacturer's instructions before use, and a daily conditioning for 10 min was carried out before the first extraction to ensure the absence of carryover.

Fingerprinting of Spices by GC-MS Analysis
The fingerprint analysis of the investigated spices was performed using an Agilent Technologies 6890N (Palo Alto, CA, USA) gas chromatography system. This was equipped with a SUPELCOWAX ® 10 fused silica capillary column (60 m × 0.25 mm i.d. × 0.25 µm film thickness) supplied by Supelco (Bellefonte, PA, USA), with helium (Helium N60, Air Liquid, Portugal) as a carrier gas at a flow rate of 1 mL/min (column-head pressure: 13 psi). The injector temperature was fixed at 250 • C and a splitless injector equipped with an insert of 0.75 mm i.d. was attached. The oven temperature program was run as follows: initial temperature 40 • C for 1 min, 2.5 • C/min ramp until 220 • C and then held isothermally at 220 • C for 10 min, for a total GC run time of 83 min. MS detection was performed in full scan in an Agilent 5975 quadrupole inert mass selective detector (Santa Clara, CA, USA), the ion energy used for the electron impact (EI) was 70 eV and the source temperature was 230 • C. The electron multiplier was set to the autotune procedure. The mass acquisition range, made in full scan mode, was 30-300 m/z. VOMs were identified based on their mass spectra compared with those in the National Institute of Standards and Technology (NIST) MS 05 spectral database (Gaithersburg, MD, USA) with a matching probability > 85%, and determining the Kovat's retention indices (RI) of each identified VOM according to the van den Dool and Kratz [47] equation: RI x = 100n + 100(t x − t n )(t n + 1 − t n ), where t n and t n + 1 are the retention times of the reference n-alkane hydrocarbons eluting immediately after and before the compound "x", and t x is the retention time of compound 'x'. A C 8 -C 20 saturated n-alkane solution was used to determine the RI, and the values were compared, when available, to values reported in the literature for similar columns [48][49][50][51] and databases available online (the Pherobase and Flavornet). Each sample was analyzed in triplicate.

Data Treatment and Multivariate Statistical Analysis
MSDA was performed using the MetaboAnalyst 5.0 web-based tool [52]. The raw GC-qMS data were pre-processed to remove the VOMs with missing values and then normalized (data transformation by cubic root and data scaling by autoscaling). The data matrix is subjected to a one-way analysis of variance (ANOVA) followed by Fisher's test for post hoc multiple comparisons of means from the six spice varieties data at p-value < 0.001 to identify significant differences. The principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) were used to provide insights into the separations among the spices under study, and to detect the VOMs that may represent differences among the sample sets. Importantly, PLS-DA can identify VOMs sets that best discriminate among the different spices analyzed by reducing the size of the data matrix through eliminating redundant variables. The VOMs with variable importance in the projection (VIP) score ≥ 1.45 and differentially expressed in the univariate analysis were potential candidates for characterizing spice varieties. A hierarchical cluster analysis (HCA) was carried out using the 15 most significant VOMs identified in the spice samples obtained by ANOVA. This was generated through Ward's algorithm and Euclidean distance analysis, aiming to identify clustering patterns for the characterization of the spices analyzed. Significance was established at p < 0.05.

Conclusions
The volatilomic composition of spices obtained by volatiles extraction through HS-SPME DVB/CAR/PDMS followed by GC-qMS analysis revealed a useful strategy for the quality and authenticity purposes of spices. For each investigated spice, the most dominant chemical classes (terpenoids, sesquiterpenoids and carbonyl compounds) and major VOMs (dihydrojuneol, #161, in saffron; cinnamaldehyde, #150, in cinnamon; cuminaldehyde, #130, in cumin and curry; caryophyllene, #87, in black pepper; and p-cymene, #42, in sweet paprika), were determined and established. In addition, the VOMs common to all spices under study, and the VOMs identified only in one spice type were also determined. Even though all of the spices share the plant-related VOMs groups, the combination of both chromatography techniques and chemometrics analysis on the volatile analysis of the spices, resulted in a deep comprehensive characterization of the spices. This can be useful for the authentication analysis, helping with the detection of frauds and adulterations. In addition, some VOMs are part of biologically active compounds and have a high influence on the spice's bioactive potential. This helps in the prevention of certain diseases including cancer, inflammatory-related diseases, diabetes, and cardiovascular diseases.