In Vitro Toxicity Studies of Bioactive Organosulfur Compounds from Allium spp. with Potential Application in the Agri-Food Industry: A Review

Organosulfur compounds (OSCs) are secondary metabolites produced by different Allium species which present important biological activities such as antimicrobial, antioxidant, anti-inflammatory antidiabetic, anticarcinogenic, antispasmodic, etc. In recent years, their use has been promoted in the agri-food industry as a substitute for synthetic preservatives, increasing potential accumulative exposure to consumers. Before their application in the food industry, it is necessary to pass a safety assessment as specified by the European Food Safety Authority (EFSA). This work reviews the scientific literature on OSCs regarding their in vitro toxicity evaluation following PRISMA guidelines for systematic reviews. Four electronic research databases were searched (Web of Science, Scopus, Science Database and PubMed) and a total of 43 works were selected according to predeterminate inclusion and exclusion criteria. Different data items and the risk of bias for each study were included. Currently, there are very few in vitro studies focused on investigating the potential toxicity of OSCs. Most research studies aimed to evaluate the cytotoxicity of OSCs to elucidate their antiproliferative effects focusing on their therapeutic aspects using cancer cell lines as the main experimental model. The results showed that diallyl disulfide (DADS) is the compound most studied, followed by diallyl trisulfide (DATS), diallyl sulfide (DAS), Allicin and Ajoene. Only 4 studies have been performed specifically to explore the safety of OSCs for agri-food applications, and genotoxicity studies are limited. More toxicity studies of OSCs are necessary to ensure consumers safety and should mainly be focused on the evaluation of genotoxicity and long-term toxicity effects.


Introduction
The name Allium is derived from the Greek word "aleo" defined as "to avoid" due to its strong odor [1]. The Allium genus includes about 600 to 700 species, with onion (A. cepa) and garlic (A. sativum) as the most well-known characteristic and edible species [2]. They were among the first domestic plants documented [3].
The main interest of these species lies in their organosulfur compounds (OSCs), which are secondary metabolites (e.g., ajoenes and thiosulphinates) with biological action and a distinctive smell of Allium species. When tissues are damaged, a cascade of enzymatic reactions by alliinase occurs, resulting in a degradation of initial compounds, and thus, sequentially, new highly reactive and biologically active compounds appear, giving rise to a great variety of byproducts [4]. These phytochemical compounds are biosynthesized for two main purposes: as a defense mechanism against biotic stresses, and as a mediator for pollination [1]. The Allium spp. are also well known for their multiple biological effects, Table 1. Organosulfur compounds present in Allium species, chemical structure and main properties.

Name
Chemical Structure Mode of Action Reference
Because of this all, to carry out all possible applications in the agri-food industry, it is necessary that OSCs pass a safety assessment as specified by the European Food Safety Authority (EFSA) [48][49][50][51]. This safety evaluation includes a wide array of tests both in vitro and in vivo. The "Guidance for submission for food additive evaluations" by EFSA [48] described the need of a toxicological evaluation in the following core areas: genotoxicity, toxicokinetics, toxicity comprising chronic, subchronic and carcinogenicity, developmental and reproductive toxicity. For the toxicological research, a tiered approach is followed, initially using fewer complex tests to obtain hazard data. These are then evaluated to determine if they are sufficient for risk assessment or, if not, for designing studies at higher tiers [48]. Therefore, in vitro studies are usually the initial step in the toxicological evaluation of any compound.
Thus, the objective of this work is to provide a systematic overview of the scientific literature of OSCs with antibacterial, antifungal, and other properties present in the Allium genus (Table 1) in regard to their in vitro toxicity evaluation. To achieve this purpose, the authors have followed the PRISMA 2020 guidelines for systematic reviews. This information could contribute to the safe use of these compounds in the agri-food sector.

Materials and Methods
This research was performed according to the Preferred Reporting Items for a Systematic Review and Meta-analysis (PRISMA) statement [52]. The question to be answered was: Do Allium compounds have toxic activity in vitro?

Protocol and Registration
The protocol for the present systematic review was not registered in any Systematic Review and Meta-analysis database.

Eligibility and Exclusion Criteria
International studies were considered. The eligibility criteria of the present systematic review were as follows: Inclusion criteria: (1) articles on Allium toxicity in vitro; (2) articles published prior to 21 September 2021; and (3) articles reporting comprehensive results and/or information on the field. Exclusion criteria: (1) unsystematic and narrative reviews; (2) articles published in a language other than English; (3) proceedings of conferences and dissertations; (4) books or book chapters; (5) editorial material; (6) articles dealing with Allium in vitro in which the test item is not a naturally occurring alliaceous compound (e.g., synthetically modified allicin) in the genus Allium or when the test has no toxicological relevance (e.g., protective effects of organosulfur compounds towards N-nitrosamineinduced DNA damage) as represented in the flowchart ( Figure 1).

Information Sources and Search Strategy
The electronic research databases Web of Science, Scopus, Science Database and PubMed were searched on 21 September 2021. The search identified articles published from inception to 21 September 2021 inclusive. The Boolean strings chosen were: ("Propyl thiosulfinate oxide" OR "propyl-propane-thiosulfonate" OR "propyl propane thiosulfinate" OR "propyl-propane-thiosulfinate" OR "organosulfur compound*" OR diallyl* OR allicin OR alliin OR ajoene OR "dipropyl disulphide" OR "dipropyl sulphide" OR propiin) AND (toxicity OR cytotoxicity OR genotoxicity) AND ("in vitro" OR "cell line"). The searches included works published in all languages. The Web of Science database option search was "theme" in all databases. The Scopus database options search were: "title, abstract and keywords". The Science Database option search was "all fields except full text (NOFT)" and the PubMed option search was "all fields".

Study Selection
Once the selection criteria have been established, a three-step process was performed to review all records according to the eligibility criteria: first was reading the title, second, the abstract, and third, the entire text of the publication. The works obtained by the four databases were crossed with the EndNote X9 (Bld 12062) software to identify possible duplicates and to classify the works according to the exclusion and inclusion criteria. Two authors (PA-M and AC-L) formed the review team to implement measures to reduce random mistakes and bias at all review phases and independently examined titles, abstracts and full texts of the articles for possible addition. Conflicts on whether a given reference should be incorporated or not were determined through discussion.

Information Sources and Search Strategy
The electronic research databases Web of Science, Scopus, Science Database and Pub-Med were searched on 21 September 2021. The search identified articles published from inception to 21 September 2021 inclusive. The Boolean strings chosen were: ("Propyl thiosulfinate oxide" OR "propyl-propane-thiosulfonate" OR "propyl propane thiosulfinate" OR "propyl-propane-thiosulfinate" OR "organosulfur compound*" OR diallyl* OR allicin OR alliin OR ajoene OR "dipropyl disulphide" OR "dipropyl sulphide" OR propiin) AND (toxicity OR cytotoxicity OR genotoxicity) AND ("in vitro" OR "cell line"). The searches included works published in all languages. The Web of Science database option search was "theme" in all databases. The Scopus database options search were: "title, abstract and keywords". The Science Database option search was "all fields except full text (NOFT)" and the PubMed option search was "all fields".

Study Selection
Once the selection criteria have been established, a three-step process was performed to review all records according to the eligibility criteria: first was reading the title, second, the abstract, and third, the entire text of the publication. The works obtained by the four databases were crossed with the EndNote X9 (Bld 12062) software to identify possible duplicates and to classify the works according to the exclusion and inclusion criteria. Two authors (PA-M and AC-L) formed the review team to implement measures to reduce random mistakes and bias at all review phases and independently examined titles, abstracts

Data Extraction and Data Items
The data items included for data extraction were: Assays performed, experimental model, concentration ranges and time exposure, and main results. This data extraction form is presented in Table 2.

Risk of Bias in Individual Studies
Bias can be judged to be a systematic mistake that can lead to an underestimation or overestimation of the true result [53]. The risk of bias for each incorporated work was evaluated using The Methods Guide for Comparative Effectiveness Reviews [54]. The characteristics of bias considered are shown in Table 3. Each risk of bias question scored 2 (fully reported), 1 (partially reported) or 0 (unclear/not reported). Scores for each risk of bias question were added together to give a total score between 0 and 10. A score of 0-4 was considered a high risk of bias; 5-7 was considered a moderate risk of bias; 8-10 was considered a low risk of bias. The risk of bias assessment was performed by the reviewer team. Differences were resolved by discussion between the three reviewers (AC-L, AC and AIP).

Results Construction and Statistical Analysis
Compiled information and findings in the studies were examined independently by the reviewer team in order to find conflicts in the extracted data, should they be present.

Search Results
The procedure of article selection is shown in the flow chart of Figure 1. The research database Web of Science returned 461 works; Scopus returned 190 works; Science Database returned 9 works and PubMed returned 287 works. The 947 works found by the four databases combined were crossed with the EndNote X9 (Bld 12062) software to identify possible duplicates. A total of 307 works were eliminated in this stage. After evaluating the abstract of each of the remaining articles, those that were associated with the subject of the study (43) were selected. According with the flow diagram (Figure 1), some articles were excluded since they had no toxicological importance, e.g., "Effect of allicin on promastigotes and intracellular amastigotes of Leishmania donovani and L. infantum". It should be noted that some of the studies could be introduced in more than one exclusion group but the final criterion was agreed by discussion of the review team. Finally, a total of 43 works were found to be eligible for the present systematic review following the full-text eligibility assessment.

Study Characteristics and Results of Individual Studies
The characteristics and main findings of the 43 selected articles in the present systematic review, such as Assays performed, experimental model, concentration ranges and time exposure and main results, are summarized in Table 2. Thus, the main tests carried out have focused on investigating cytotoxicity, cell apoptosis and ROS production against antioxidant assays. The MTT assay highlights how to determine the EC50 of these compounds in different cellular models. The flow cytometry is used for different determinations such as apoptosis, cell cycle, mitochondrial membrane potential and ROS. On the other hand, assays related to the mutagenicity and genotoxicity showed less interest. In this sense, Ames test, MN, comet assay and MLA have been performed. Regarding the experimental models, different cell lines have been used. Melanoma, lymphoma and gastrointestinal tract cells were the ones most often selected ( Figure 2). The 43 selected articles were published between 1990 and 2021. According to the corresponding authors addresses, the articles were published in 19 different countries: 8 in China, 6 in Korea, 4 in USA, 3 in Spain, 3 in Poland, 2 in India, 2 in Taiwan, 2 in Germany, 2 in UK, 2 in Venezuela, 1 in Brazil, 1 in Israel, 1 in Egypt, 1 in Japan, 1 in Iran, 1 in France, 1 in Croatia, 1 in Turkey and 1 in Italy. The 43 included articles were published in 33 different journals, Food and Chemical Toxicology being the journals with most publications.

In Vitro Studies of OSCs from Allium spp. Focused on Safety Evaluation for Agri-Food Application
Of the 43 studies selected, only 4 have been performed specifically to explore the safety of OSCs from Allium spp. for agri-food applications [16,17,55,56] (see Table 2). Mellado-García et al.  studied the genotoxicity potential of PTSO and PTS using in vitro test battery including Salmonella typhimurium reverse mutation assay (Ames Test), The cytokinesis-block micronucleus cytome assay (MN) and single cell gel electrophoresis assay (comet assay). Both compounds reported negative results by Ames Test and genotoxic effects were described in similar concentration for PTSO and PTS by MN assay. However, MLA and comet assays showed contradictory results between these compounds. No cytotoxicity or mutagenicity of DPS, DPDS, and mixtures were reported by Llana-Ruiz-Cabello et al. [16]. In relation to polymer films of Allium cepa L., no induction of MN was observed, although the HTP films showed signs of mutagenicity by the Ames Test [56].
addresses, the articles were published in 19 different countries: 8 in China, 6 in Korea, 4 in USA, 3 in Spain, 3 in Poland, 2 in India, 2 in Taiwan, 2 in Germany, 2 in UK, 2 in Venezuela, 1 in Brazil, 1 in Israel, 1 in Egypt, 1 in Japan, 1 in Iran, 1 in France, 1 in Croatia, 1 in Turkey and 1 in Italy. The 43 included articles were published in 33 different journals, Food and Chemical Toxicology being the journals with most publications.

In Vitro Studies of OSCs from Allium spp. Focused on Safety Evaluation for Agri-Food Application
Of the 43 studies selected, only 4 have been performed specifically to explore the safety of OSCs from Allium spp. for agri-food applications [16,17,55,56] (see Table 2). Mellado-García et al.  studied the genotoxicity potential of PTSO and PTS using in vitro test battery including Salmonella typhimurium reverse mutation assay (Ames Test), The   In both experiments, ↓ cell viability in a dose and time dependent manner. In the first minutes after exposure, GSH decreased and GSSG increased. The further course strongly depended on the dose.
The Glu pool showed an immediate increase and in a later stage decreased. The Asp pool showed the contrary pattern. [58]

DDS DAS
Clonal survival Mass growth rate. Anchorage-Independent Growth.
Control and differentiated HT29 cells 1-3 µg/mL DDS 24 h 100-300 µg/mL DAS 24 h after which the cells were incubated in fresh medium for 10-12 days. Cells were exposed to 24-h intervals for seven days.
Dq (concentration at which no cell killing occurs) 0.10 ± 0.03 µg/mL for DAS and was not found for DDS. D37 (dose required to reduce survival from 100% to 37%) are 2.93 ± 0.14 µg/mL for DAS and 164 ± 12 µg/mL for DDS. The cell growth was significantly inhibited by DAS and DADS and slightly inhibited by garlic extract. The OSCs compounds and garlic extract have apoptotic potential on lung cancer cells, and the mechanism was regulated through p53-dependent or p-53 independent related Bax/Bcl-2 dual pathway.  ↑ the expressions of p85a PI3K, phosphorylated Akt and phosphorylated GSK-3 in N18D3 cells pretreated with 25 µM (2 h) and subsequently exposed to 100 µM H 2 O 2 (30 min). Treatment with 100 µM reduced these biomarkers in N18D3 cells.
↑ the levels of free radicals and membrane lipid peroxidation a concentration-dependent manner.  The three garlic compounds induced cytotoxic effects via ROS production, increase in endoplasmic reticulum (ER) stress, decrease mitochondrial membrane potential, activation of stress kinases and cysteine proteases.       Viability inhibition in a concentration-dependent way. Apoptosis induction. Nuclear shrinkage/condensation and nuclear fragmentation. ROS production induction and decreased level of mitochondrial membrane potential. Activation of caspase 9 and 3. Increased protein levels of cytochrome c, caspase -9 and caspase-3.  No cytotoxicity or mutagenicity and no significant adverse effects were reported. ROS scavenger activity was observed for both compounds. Not mutagenic neither in the Ames test nor in MLA. Genotoxic effects were reported in the MN test at the highest concentration assayed (17.25 µM) without S9, and also its metabolites (+S9, from 20 µM). ↑ breaks damage on CaCO 2 cells at the highest concentration tested (280 µM) but it did not induce oxidative DNA damage.  ↑ viability in ethanol-exposed 3T3L1 cells treated with 200-500 µM for 24 h and 50-500 µM for 48 h. ↓ ROS production, reduces expression of pro-inflammatory cytokines, and enhance anti-inflammatory cytokine production in ethanol-exposed 3T3L1 cells treated with 50-100 µM for 24 or 48 h. 100 µM for 24h ↑ expression of M2 phenotype-specific genes in ethanol-exposed RAW 264.7 cells. [91]

Risk of Bias
Studies were considered to have a low, moderate or high risk of bias in terms of score out of 10. A moderate risk of bias was found in each of the 43 works chosen for the present systematic review. When reviewing the quality of selection, the studies show more limitations in "reproducibility" and "adequate statistical analysis" items. Full details are given in Table 3. Table 3. Risk of bias for the methodological quality of studies reporting the toxicological evaluation in vitro of OSCs from Allium spp. 0: not reported; 1: not appropriately or clearly evaluated; 2: appropriately evaluated. M: medium (5-7); L: low (8)(9)(10); H: high (0-4).

Reference
Clear

Limitations
The present systematic review was restricted by the databases used, the search conditions, and the recognized inclusion/exclusion principles chosen. However, the exploration strategy was quite comprehensive, so it is expected that relatively, only a few important studies could not be identified and considered. Only works reported in English were included, and this point could indicate bias in the source searching and selection pro-cess [96]. Finally, the lack of sufficient statistical information made impossible to combine the results of different studies into a meta-analysis section that it had to be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.

In Vivo Studies Excluded
Several in vivo studies (n = 26, Figure 1), despite being excluded by the criteria of this systematic review, have been analyzed by the authors. As in in vitro studies, many of them deal with the anticancer properties of natural organosulfur compounds, mainly assayed in mice. In this sense, Sundaram et al. [97] studied the growth inhibitory properties of DADS against colon cancer, Chu et al. [98] studied the compound S-Allylcysteine against prostate cancer and Nishikawa et al. [99] studied the inhibitory properties of ajoene against skin cancer. Other in vivo studies focused on measuring biological markers, such as catalase and monooxygenase activity, or their protective properties against toxic substances in animals treated with alliaceous compounds. In this sense, Zhang et al. [100] studied the protective effect of allicin against acrylamide. Only a few in vivo studies have focused on evaluating the toxicity of alliaceous compounds. Thus, acute studies [20] or subchronic toxicity studies of isolated substances (such as PTSO) [101] or extracts from plants of the Allium genus [102] were found. Among the in vivo toxicological studies, the genotoxicity tests (MN and comet) in rats are highlighted [103,104]. In general, no significant signs of toxicity neither genotoxic effect were observed in the subchronically studies or genotoxicity endpoints.

Discussion
The beneficial effects of OSCs compounds have been reviewed by different authors. In addition, their phytochemical profile has been well described [5,12]. However, as far as we know, the safety evaluation and toxicity effects of these compounds have not been reported.
The number of scientific publications dealing with the in vitro toxicity of OSCs that meet the criteria established in this review amounts to 43. In vitro studies play an important role in the toxicity evaluation of compounds. They can give valuable hints about mechanisms of toxicity, providing rapid and cost-effective screening, and allow one to reduce the use of live animal models in research.
Among the OSCs investigated, most studies have focused on DADS, DATS and DAS, followed by DATS, DAS, Allicin and Ajoene, whereas for others, the existing reports were limited (i.e., DPS, DPDS, PTSO, PTS) (see Figure 2). Moreover, most of the toxicity studies of OSCs are reported from 1990s and early 2000s, and only 14 of them have been published after 2015, so there are few current toxicological studies focusing on the toxicological effects of these compounds.
Most of these studies aimed to evaluate the cytotoxicity of OSCs on different cellular models and tried to elucidate the mechanisms involved. Indeed, the vast majority of them explored the antiproliferative effects of OSCs to justify their potential as chemoprotectants against carcinogenesis (see Table 2). Several studies that reported the anti-cancer effect of black garlic on the cancer cell line showed inhibition of tumor activity by regulating metabolism [12]. Furthermore, DADS has been proposed as a therapeutic strategy for oxidative stress-injury in neurodegenerative diseases [65] and DAS has proved to be effective in reducing ethanol induced injury of cells (Kema et al., 2018). Apart from therapeutic aspects, there are a limited number of papers dealing with genotoxicity [16,17,55,56,60,62,95]. Finally, only 4 out of 43 studies have been performed specifically to explore the safety of OSCs for further agrifood applications [16,17,55,56] Regarding results obtained for specific OSCs, DADS in relation to genotoxicity aspects has been reported to induce chromosome aberrations and sister chromatid exchanges in a Chinese hamster's ovary cell line (CHO) [60]. Additionally, it increased the mutagenicity of 2-amino-1-methyl-6-phenylimidazol [4,5-b] pyridine when the Ames test was performed with S9 fraction from rats exposed to DADS. Filomeni et al., [64] and Kim et al. [65] explored the effects of DADS on two different cellular models of the nervous system and obtained different results. Whereas the first one suggested a pivotal role for oxidative stress in DADS-induced apoptosis on SH-SY5Y cells and pointed out a potential use as antiproliferative agent in cancer therapy, the second one observed opposite results on N18D3 cells, depending on the concentration used, with a protective effect at low concentrations. Recently, important toxic effects have been reported of this compound associated to high doses [105]. For these reasons, more DADS toxicity studies are necessary to guarantee its safe use as an anticancer agent.
Two other studies [70,75] investigated its effects on a breast cancer cell line (MCF-7) and obtained similar results, inhibition of cell proliferation and apoptosis induction, with Lei et al. [70] providing mechanistic clues (inhibition of ERK and activation of SAPK/JNK and p38 pathways). Apoptosis was also observed in other different cell types where the antiproliferative effects of DADS were investigated such as p53-wild type H460 and p53null type H1299 non-small-cell lung cancer cells [63], in human glioblastoma cells [68], human leukemia cells [72], human esophageal carcinoma cells [84,85] or primary effusion lymphoma cells [87]. All these reports support the potential use of DADS as chemotherapeutic agent.
Results reported for DATS are similar to those discussed for DADS, as DATS have also shown to induce cytotoxicity, ROS production or apoptosis (also evidenced by changes in the expression of related genes and proteins) in different cell types such as NIH-3T3, MCF-7, PC-3, AGS, U87M6, PNT-1A, MDA-MB468, MDA-MD231, A172, U343 and T98G. Only Das et al. [68] and Shigemi et al. [87] evaluated both DADS and DATS, and compared the results obtained. Both of them observed similar results: DATS was more potent than DADS and DAS for induction of cell death with involvement of mitochondria and ROS production.
With respect to DAS, it induced genotoxic effects similarly to DADS [60,62], but it was less cytotoxic to CHO cells and it increased the activation of a higher number of mutagens. Again, several studies in a variety of cell lines showed its antiproliferative effect [59,63,68,71,77,87]). It has been reported that garlic compounds (DAS, DADS, DATS) do not require a p53-dependent pathway for mediation of apoptosis [68]. Moreover, its potential to reduce the tissue injury caused by ethanol was also demonstrated [92]. In order to compare the effects produced by each of these compounds, differences have been found, mainly due to the diverse experimental models and conditions used (concentration, time of exposure, biomarkers, etc.).
Studies evaluating allicin mainly reported cytotoxicity [57], and apoptosis [66,83,86,90] as adverse outcomes, mediated by different key events such as altered genes and protein expression or cell cycle changes. Scharfenberg et al. [57] were the only ones that studied not only allicin but also its decomposition product ajoene and observed that allicin was less toxic than ajoene in three different cell lines. This compound, ajoene, was also investigated by different authors [58,67,73] that observed cytotoxicity (with different cell lines showing different sensitivity) and apoptosis.
Regarding cytotoxicity assays, there are only a few cell lines listed in these studies that are included in the guidelines by OECD guidance for toxicological evaluation of chemicals. Most of them are cancer cell lines, and the effects reported in these findings correspond to a therapeutic anticancer effect and not to a cytotoxic evaluation. In this sense, more studies focused on the toxic effect of OSCs isolated are necessary.
Moreover, those OSCs with the fewer number of studies available in the scientific literature were also those mainly focused on safety issues in relation to their use in the agri-food sector. Thus, neither mutagenicity by the Ames test nor cytotoxicity in the human intestinal carcinoma Caco-2 cell line was observed for DPS and DPDS [16]. Additionally, a complete battery of genotoxicity tests were performed for PTSO [17] and PTS [55].
Genotoxicity assessment plays a key role in the safety evaluation required by EFSA guidelines for the submission of dossiers of different substances, such as food and feed additives, etc. [48,50], with the basic battery performed with in vitro tests. However, results of this review showed that only few assays have been carried out and some studies do not include the basic battery of tests required by the EFSA. In this regard and taking into account that specific OSCs that have been mainly investigated for their chemotherapeutic potential such as DADS, DATS, DAS, etc., show also interesting activities for their use in the food industry (antimicrobial, antioxidant or antifungal activities, among others, see Table 1), the thorough study of their genotoxicity would be worthy of research. Moreover, advanced in vitro models (i.e., 3D) could provide new data to support in vitro-in vivo data extrapolation for OSCs in general, and the testing of relevant concentrations used in the agri-food sector would allow to consider both efficacy and safety aspects. Thus, in vitro assays on their own can still provide valuable information to contribute to the commercial use of OSCs.

Conclusions
In general, there are very few in vitro studies focused on investigating the potential toxicity of OSCs. Most research studies aimed at evaluating only the cytotoxicity of OSCs on different cellular models to elucidate antiproliferative effects of these compounds and justify their potential as chemoprotective agent against carcinogenesis. This makes it difficult to assess the safety of the use of these compounds for a correct risk assessment. In addition, it limits the preliminary information needed to proceed with an in vivo toxicity assessment. Therefore, other cellular models such non-cancer cell lines should be included to ensure a correct in vitro toxicity evaluation of these compounds. Specifically, considering that genotoxicity assessment plays a key role in the safety evaluation required by EFSA; more genotoxicity studies of OSCs are necessary to guarantee consumer safety before their use as a potential natural additive in the food industry. Funding: This research was funded by the Consejería de Economía y Conocimiento of Junta de Andalucía (AT17_5323_USE and P18-TP-2147) and Antonio Cascajosa Lira thanks the Spanish Ministerio de Universidades for the funding FPU grant (FPU2019/01247).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study is contained within the article.