Metabolite Profile and In Vitro Beneficial Effects of Black Garlic (Allium sativum L.) Polar Extract

Over the centuries, humans have traditionally used garlic (Allium sativum L.) as a food ingredient (spice) and remedy for many diseases. To confirm this, many extensive studies recognized the therapeutic effects of garlic bulbs. More recently, black garlic (BG), made by heat-ageing white garlic bulbs, has increased its popularity in cuisine and traditional medicine around the world, but there is still limited information on its composition and potential beneficial effects. In this study, the metabolite profile of methanol extract of BG (BGE) was determined by high-performance liquid chromatography coupled to tandem mass spectrometry in high-resolution mode. Results allowed to establish that BGE major components were sulfur derivatives, saccharides, peptides, organic acids, a phenylpropanoid derivative, saponins, and compounds typical of glycerophospholipid metabolism. Characterization of the BGE action in cancer cells revealed that antioxidant, metabolic, and hepatoprotective effects occur upon treatment as well as induction of maturation of acute myeloid leukemia cells. These results are interesting from the impact point of view of BG consumption as a functional food for potential prevention of metabolic and tumor diseases.


Introduction
Nowadays, plants are well known to produce a species-specific array of small molecules, i.e., secondary metabolites, which have the potential to contribute to the maintenance of health status as well as to prevent, delay, and reduce a number of chronic diseases and health concerns in humans [1,2]. In vitro, in silico, and in vivo studies have revealed the potential beneficial health effects of these phytochemicals and their capacity to improve human wellness and health in terms of prevention and treatment of aging-associated diseases and metabolic disorders, affecting more than 25% of the global adult population [3][4][5]. Among the natural plant products, common garlic (Allium sativum L.) is used worldwide as spice, medicinal plant, and vegetable having a key role in the traditional diets with a history of thousand years. It is an important species belonging to the genus Allium (with about 500 species), within the monocotyledonous family Amaryllidaceae (ex Alliaceae), mostly formed by bulb or corm-forming herbaceous plants [6]. Of note, the regular consumption of common garlic was shown to reduce arterial hypertension, blood levels of cholesterol, and coronary heart disease; further evidence suggest that garlic also has antibiotic and

Plant Material and Preparation of the Methanol Extract of Black Garlic
Fermented and aged black garlic (Allium sativum L.) bulbs (Garlicorea) used in this study were supplied by a South Korean company (Menari Corporation, Ltd.). In brief, the bulbs (100 g) were crushed with a pestle in a mortar and then extracted in n-hexane (2 × 1 L, 8 h each) and CHCl 3 (3 × 1 L, 8 h each). The sample was then carefully resuspended in methanol (solid:liquid = 1:3). The plant material was subjected to extraction with methanol three times for 1 h at room temperature in a shaker to obtain the methanol extract of black garlic (BGE) used in this study. The extracts were centrifuged (4000 g for 10 min), then the supernatants were collected and filtered by syringe filters (0.22 µm) prior to the LC-ESI/HRMS/MS analysis.
For the biological assays, after methanol evaporation, BGE was resuspended in dimethyl sulfoxide (DMSO) and used at the final concentrations indicated in each experiment.

LC-ESI/HRMS/MS Analysis
Qualitative analysis by liquid chromatography coupled to electrospray ionization tandem mass spectrometry in high-resolution mode (LC-ESI/HRMS/MS) was performed using a LC-ESI/HRMS system consisting of an Accela HPLC system, equipped with a quaternary Accela 600 pump and an Accela autosampler, coupled to a LTQ Orbitrap XL mass spectrometer (ThermoScientific, San Jose, CA, USA), operating in negative ionization mode. The Orbitrap mass analyzer was calibrated according to the manufacturer's directions by using a mixture of caffeine, methionine-arginine-phenylalanine-alanine-acetate (MRFA), and Ultramark 1621 in a solution of acetonitrile, methanol, and acetic acid. Data were collected and analyzed by using the software provided by the manufacturer. The separation was carried out by using a Kinetex EVO 5.0 µm RP C18 column (Phenomenex, 150 × 2.1 mm), at a flow rate of 0.2 mL/min and a mobile phase consisting of a combination of A (0.1% formic acid in water, v/v) and B (0.1% formic acid in acetonitrile, v/v). A linear gradient from 5 to 13% B in 8 min, from 13 to 38% B in 12 min, from 38 to 46% B in 8 min, held to 46% B for 10 min, from 46 to 53% B in 7 min, held to 53% B for 15 min, was used. The autosampler was set to inject 10 µL of extract (0.5 mg/mL).
In the negative ion mode, the following experimental conditions for the ESI source were adopted: sheath gas at 15 (arbitrary units), auxiliary gas at 5 (arbitrary units), source voltage at 3.5 kV, capillary temperature at 280 • C, capillary voltage at −48 V, and tube lens at −176.47 V.
The mass range was from 100 to 1400 m/z with a resolution of 30,000. In order to obtain structural information, data-dependent experiments were performed by acquiring tandem mass spectra of the first and the second most intense ions from the HRMS scan event. A normalized collision energy at 30%, a minimum signal threshold at 250, and an isolation width at 2.0 were used.

Cell Lines and Culture Conditions
Human hepatocellular carcinoma cell line HepG2 (ATCC HB-8065) from American Type Culture Collection (ATCC, Manassas, VA, USA) was cultured in RPMI-1640 medium with 10% heat-inactivated FBS, 100 U/mL penicillin, 100 µg/mL streptomycin, and 1% L-glutamine. The cell cultures were incubated at 37 • C in a 5% CO 2 and 95% air humidified atmosphere. Then, cells were treated using the different amounts of BGE, as reported for each experiment performed in duplicate.
Human histiocytic lymphoma cell line U937 (ATCC CRL-1593.2) was obtained from ATCC and human acute promyelocytic leukemia cell line NB4 was provided by Michel Lanotte (INSERM U-496, Centre G. Hayem Hôpital Saint-Louis, Paris, France). Cells were grown at 37 • C in 5% CO 2 atmosphere in RPMI-1640 medium (Gibco, NY, USA), supplemented with 10% heat-inactivated fetal bovine serum (FBS), 1% L-glutamine, 1% ampicillin or streptomycin, and 0.1% gentamicin. Cells were plated at 1000 cells/mL density and then treated using the different amounts of BGE, as reported for each experiment performed in triplicate. HepG2 and U937 cells were plated in 96-well plates at the seeding density of 5 × 10 3 cells/well in RPMI complete medium. After 4 h incubation at 37 • C, the cells were treated for 24 h with different concentrations of black garlic extract (BGE, 0-2.50 mg/mL). Then, cell viability was evaluated by adding 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) solution in PBS to a final concentration of 5 mg/mL, as previously described [19,20]. The culture medium was removed after 4 h of incubation at 37 • C, and the purple formazan crystals were solubilized by mixing for 5 min with 150 µL dimethyl sulfoxide (DMSO). Then, measurements of absorbance at 570 nm by Bio-Rad 550 microplate reader (Bio-Rad Laboratories, Milan, Italy) estimated cell viability inhibition, expressed as a percentage of the control. Each experiment was carried out in triplicate.

Cellular Oxidative Stress: Nitrite Levels in the Culture Medium
Nitric oxide (NO) is rapidly converted into the stable end products nitrite and nitrate. Briefly, nitrite was measured by the Griess reaction using 10 µL of medium mixed with an equal volume of Griess reagent (0.5% sulfanilamide, 2.5% H 3 PO 4 , and 0.05% naphthyl ethylene diamine in H 2 O) and then incubated for 10 min at room temperature. Measurements of absorbance at 550 nm were compared with a standard curve of sodium nitrite [21].

Cellular Oxidative Stress: Lipid Peroxidation in Cytosol
The cellular lipid peroxidation was evaluated determining the toxic reactive aldehydes by thiobarbituric acid reactive substances (TBARS) assay. In brief, the homogenates of U937 and HepG2 cells treated for 24 h without and with 1.25 and 2.50 mg/mL BGE, respectively, were incubated at 95 • C for 45 min with 0.5 mL of 20% acetic acid solution (pH 3.5) and 0.5 mL of 0.78% aqueous solution of thiobarbituric acid, as previously described [20,21]. After cooling and centrifugation at 4000 g for 5 min, TBARS were quantified in the supernatant fractions by absorbance at 532 nm. Each experiment was performed in duplicate.

Metabolic Parameters
HepG2 cells were grown in high glucose (30 mM, HG) concentration medium for 72 h and total cholesterol (CHL), triglycerides (TRG), and glucose were determined using standard chemical methods of clinical laboratory.

Oil Red O Staining
Oil Red O (ORO) staining of neutral lipids and lipid droplets was carried out as previously described [19]. Briefly, the cells, fixed by 10% formalin solution and incubated with 60% isopropanol, were stained for 10 min with ORO solution. Then, after washing the cell cultures, the dye was extracted by isopropanol and the absorbance was measured at 520 nm. Results were expressed as A 520 /cells number.

Microscopic Analysis
After treatment with BGE at 2.5 and 5 mg/mL, human leukemia cell lines U937 and NB4 were grown for 48 h on multi-well plates. Cell growth in three independent experiments, each performed in duplicate, was analyzed by bright-field light microscope observation. Images were captured with DMIRB inverted microscope (Leica) using N-Plan 10× objective (Leica) and a DFC 450C camera (Leica), and then analyzed with Application Suite software (Leica).
Compounds 3-4, 6 were identified as sugar derivatives according to their molecular formula, fragmentation pattern, and literature data. In particular, compounds 3 and 6 have been already reported in black garlic, the former as disaccharide molecule and the latter as hexose-phosphate characterized by product ions at m/z 79 and 97, originated by neutral loss of the hexose unit as whole or dehydrated form, and by product ions at m/z 199 and 169, formed by inner breakdown of the hexose moiety (  Legend: DGMG, di-galactosylmonoacylglycerol; l-PE, lyso-phosphatidylethanolamine; l-PC, lyso-phosphatidylcholine; l-PI lysophosphatidylinositol. According to literature reports and mass spectrometric data, compounds 1, could be tentatively identified as the organosulfur compounds, characteristic o species, S-allyl-cysteine (1), alliin (2), and allicin (5), belonging to the families of L-c sulfoxides, and thiosulfinates derivatives, respectively (Table 1) [13,26].
Compounds 3-4, 6 were identified as sugar derivatives according to their m formula, fragmentation pattern, and literature data. In particular, compounds have been already reported in black garlic, the former as disaccharide molecule latter as hexose-phosphate characterized by product ions at m/z 79 and 97, origin neutral loss of the hexose unit as whole or dehydrated form, and by product ion 199 and 169, formed by inner breakdown of the hexose moiety (Table 1) [26]. Com 4 displayed in LC-ESI/HRMS spectrum a [M-H] − pseudomolecular ion at m/z 3 along with the [(M-H) + HCOOH] − formic acid adduct ion, accounting for the m formula C12H20O10. The fragmentation pattern suggested the structure of 4 as bis tose 2′,1-2,1′-dianhydride, a disaccharide derivative already found in black garli 1) [26].
Compounds 7, 10, and 11 exhibited a nitrogenous molecular formula and a fr tation behavior ascribable to peptide derivatives. In particular, according to its ma trometric behavior and literature data, compound 10 could be assigned as the di γ-glutamylphenylalanine, already described in black garlic (Table 1) [26]. Intere compound 11 differed from 10 being 162 Da higher, as confirmed by the product H)-162] − at m/z 293 occurring in its LC-ESI/MS/MS spectrum. Moreover, as for com 10, compound 11 yielded product ions at m/z 164 and 128, corresponding to the p anine anion and to the dehydrated form of the glutamic acid anion, respectively, a to assume that 11 could be a γ-glutamylphenylalanine-O-hexose ester, never de before. The fragmentation pattern of 11 was also characterized by the presenc product ion at m/z 290, originated by a neutral loss of 162 Da, suggesting that the unit was likely esterified at the α-carboxyl group of the glutamic acid residue (T This observation guided the identification of compound 7 as L-pyroglutamic acid ose ester, according to its molecular formula and to the fragmentation pattern Compounds 7, 10, and 11 exhibited a nitrogenous molecular formula and a fragmentation behavior ascribable to peptide derivatives. In particular, according to its mass spectrometric behavior and literature data, compound 10 could be assigned as the dipeptide γ-glutamylphenylalanine, already described in black garlic (Table 1) [26]. Interestingly, compound 11 differed from 10 being 162 Da higher, as confirmed by the product ion [(M-H)-162] − at m/z 293 occurring in its LC-ESI/MS/MS spectrum. Moreover, as for compound 10, compound 11 yielded product ions at m/z 164 and 128, corresponding to the phenylalanine anion and to the dehydrated form of the glutamic acid anion, respectively, allowing to assume that 11 could be a γ-glutamylphenylalanine-O-hexose ester, never described before. The fragmentation pattern of 11 was also characterized by the presence of the product ion at m/z 290, originated by a neutral loss of 162 Da, suggesting that the hexose unit was likely esterified at the α-carboxyl group of the glutamic acid residue (Table 1). This observation guided the identification of compound 7 as L-pyroglutamic acid-O-hexose ester, according to its molecular formula and to the fragmentation pattern mainly characterized by the product ion at m/z 128, corresponding, as mentioned above, to the dehydrated form of the glutamic acid anion, i.e., the pyroglutamic form. The fragmentation pattern of 7 was further characterized by product ions at m/z 200 and 170, originated by inner fragmentation of the sugar unit (Table 1) [27]. Both compound 7 and 11 have been here described for the first time in black garlic.
Compound 12 showed a molecular formula and fragmentation pattern that allowed to suppose a phenylpropanoid diglycoside nature (Table 1). LC-ESI/MS/MS spectrum showed a main product ion at m/z 307 originated by neutral loss of 210 Da, that, according to the literature data, could be assumed as sinapyl alcohol, a phenolic compound already found in garlic [28]. Thereby, compound 12 could be tentatively supposed to be a syringin derivative, also described in A. macrostemon and A. victorialis, and precisely, a rhamnoside derivative, as supported by the presence of the product ion at m/z 163, likely corresponding to the dehydrated rhamnose anion (Table 1) [29].
According to literature data attesting the occurrence of saponins in Allium spp., compounds 15 and 16, exhibiting consecutive neutral losses of hexose units from the [M-H] − pseudomolecular ion in their LC-ESI/MS/MS spectra, could be referred as spirostanol saponins with an aglycone structure more likely corresponding to gitogenin or β-chlorogenin, already described in A. sativum (Table 1) [30][31][32][33]. Analogously, on the basis of its molecular formula, retention time, and fragmentation behavior, compound 17 could be tentatively assigned as tigogenin, a spirostanol aglycone already reported in A. sativum [30]. Finally, compound 13 could be tentatively assigned as porrigenin B 3-O-pentasaccharide, a spirostanone saponin found in A. ampeloprasum, but never reported so far in A. sativum [34].
The C 12 H 18 O 7 S molecular formula of compound 18 pointed to identify it as a sulfated compound that, in agreement with the fragmentation pattern and literature data, could be defined as hydroxyjasmonic acid sulfate, a stress indicator metabolite characterized by a LC-ESI/MS/MS spectrum showing a main [(M-H)-80] − product ion at m/z 225, due to neutral loss of a SO 3 molecule. Minor product ions at m/z 261 and 97, the first formed by neutral loss of a carbon dioxide molecule and indicating a carboxylic acid moiety in the parent ion, and the second corresponding to the sulfuric acid anion, were also detectable ( Table 1) [35,36].
The molecular formula of compound 14 pointed to suggest it as a C 18 trihydroxylated fatty acid, as supported by the tandem mass spectrum characterized by diagnostic product ions and neutral losses. These were generated by molecular rearrangements involving the head and the end of the acyl chain, respectively, so promptly suggesting the position on the fatty acyl chain of both hydroxyl groups and double bonds [37]. So, the observation of the product ion at m/z 171, corresponding to the shortened acyl chain having the carboxyl group as COO − and an aldehyde end-group originated from the rearrangement of one hydroxyl function and subsequent cleavage of the C9-C10 bond, allowed to locate the OH group at C-9 (Table 1). Analogously, the product ions at m/z 229 and 199, originated by neutral loss of 100 and 130 Da via the same CHOH→CHO rearrangement of the hydroxyl function involving the end of the acyl chain to form a hexanal and a hydroxylated heptanal, respectively, allowed to locate the hydroxyl groups at C-12 and C-13 (Table 1). Consequently, in agreement with the assigned molecular formula, a double bond between C10-C11 carbons has to be assigned (Table 1). To our knowledge, this is the first report of oxylipin (14) in black garlic.
Finally, according to literature data describing the occurrence of polar lipids in garlic, metabolites belonging to glycolipid (DGMG) and lyso-phospholipid (l-PE, l-PC, and l-PI) classes could be identified (Table 1) [13,38]. On the basis of the presence in LC-ESI/MS/MS spectra of typical product ions involving the characteristic head-group of lipid molecule, the identification of 5 l-PE, 3 DGMG, 5 l-PI, and 7 l-PC species could be obtained (Table 1) 30, 28 and 31, 36 and 38), for which the occurrence of both regioisomers could be asserted (Table 1).

Effects of Black Garlic Methanol Extract on Cancer Cell Viability and Oxidative Stress
The effects of black garlic methanol extract (BGE) on cell viability and oxidative stress were evaluated in vitro using human hepatocellular carcinoma (HepG2) and human leukemic (U937) cell lines. Both HepG2 and U937 cells were initially exposed to increasing BGE doses up to 2.5 mg/mL, then cell viability was assessed after 24 h by MTT assay. BGE induced dose-dependent cell viability inhibition with IC 50  were evaluated in vitro using human hepatocellular carcinoma (HepG2) and hum kemic (U937) cell lines. Both HepG2 and U937 cells were initially exposed to inc BGE doses up to 2.5 mg/mL, then cell viability was assessed after 24 h by MTT ass induced dose-dependent cell viability inhibition with IC50 = 0.8 mg/mL and 2.0 mg HepG2 and U937 cells, respectively (Figure 2, upper panel). The effects on cellular oxidative stress upon BGE treatments (1.25 and 2.50 m were determined by evaluating free nitric oxide (NO) concentration as nitrites in ture medium and lipid peroxidation as TBARS in cytosol of both HepG2 and U9 (Figure 2, central and bottom panels). After 24 h of 2.50 mg/mL BGE treatment, U9 showed a significant increase (p = 0.0038) of free NO release of about 49% compare untreated U937, while there was no effect on HepG2 cells compared to the un HepG2 (Figure 2, central panel). In rats, BGE also demonstrated vasodilating effec artery segments by NO induction and increased the NO release in aorta segmen The effects on cellular oxidative stress upon BGE treatments (1.25 and 2.50 mg/mL) were determined by evaluating free nitric oxide (NO) concentration as nitrites in cell culture medium and lipid peroxidation as TBARS in cytosol of both HepG2 and U937 cells (Figure 2, central and bottom panels). After 24 h of 2.50 mg/mL BGE treatment, U937 cells showed a significant increase (p = 0.0038) of free NO release of about 49% compared to the untreated U937, while there was no effect on HepG2 cells compared to the untreated HepG2 (Figure 2, central panel). In rats, BGE also demonstrated vasodilating effect in tail artery segments by NO induction and increased the NO release in aorta segments [39]. NO generated by the NO synthase (NOS) enzymes plays the role of intercellular messenger that regulates different key functions, such as gene expression, differentiation, cell cycle, and apoptosis [40][41][42]. The inhibition effect of BGE on U937 cells could be due to its ability of activating the production of NO by NOS enzymes and, indirectly, of scavenging superoxide anion radicals (O 2 − ) that induce the transformation of NO into peroxynitrite (ONOO − ), which consequently cannot play the function as a second messenger.
On the other side, the lipid peroxidation evaluated by TBARS assay in cytosol resulted in a significant increase (p = 0.0089) in the BGE-treated (1.25 and 2.50 mg/mL) HepG2 cells of about 25% compared to untreated cells, and to a lesser extent in U937 cells (Figure 2, bottom panel). In support of these results, previous studies demonstrated that the products of cellular lipid peroxidation act as intracellular signals able to inhibit cell proliferation and to induce differentiation [43]. Up to now, white garlic extracts have been amply demonstrated to induce ROS-dependent cytotoxicity on cancer cell lines, while only a limited number of studies showed that BGE can cause ROS-dependent cell death in cell lines [9,44]. The cytotoxic activity of saponins was shown in different cancer cell lines, such as Hep-G2, HT1080 (fibrosarcoma), HeLa (cervical cancer), HL-60 (promyelocytic leukemia), and MDA-MB-453 (breast cancer) [45]. Our data demonstrated that the BGE treatment of HepG2 cells induced antiproliferative effect and a concomitant increase of lipid peroxidation. During the fatty acid β-oxidation, the physiological collateral productions of superoxide anion radicals (O 2 − ) and hydrogen peroxide (H 2 O 2 ) trigger the self-propagating chain reactions of non-enzymatic lipid peroxidation in cells. In rat liver, Ha et al. (2015) [46] reported that the supplementation with BGE of a high fat diet led to increased expression of carnitine palmitoyltransferase-1 (CPT-1), causing TG break down to supply free fatty acid as an energy source. We think that the antiproliferative effect of BGE is most likely due to its pro-oxidant actions related to lipid peroxidation induced by an increase of fatty acid β-oxidation.

Black Garlic Methanol Extract Reduces Glucose, Triglycerides and Cholesterol Accumulation in Liver Cells
Hyperglycemia is known to exert injuries in liver, i.e., dysregulation of glucose, lipid, and triglyceride metabolism [47,48]. Previous studies reported that the in vivo administration of black garlic in rats led to decreased liver weights by reducing fat accumulation and decreased TBARS levels in liver, heart, and plasma [49]. HepG2 cells are an in vitro model system suitable to determine the effects of black garlic on lipid metabolism in human hepatocytes. To assess the functional role of BGE, we have grown HepG2 cells in high glucose (30 mM, HG) concentration medium for 72 h, in vitro miming the hyperglycemic effect on lipid metabolism [21,47]. Interestingly, 1.0 mg/mL BGE was significantly active in reducing glucose, cholesterol (CHO) and triglyceride (TG) cell uptake of 2.4, 3.8 and 4.9 folds, respectively, in HG-HepG2 cells compared to untreated cells (Table 2). Moreover, BGE reduced lipid peroxidation of about 1.5-fold compared to untreated HG-HepG2 cells, as evaluated by TBARS assay (Table 2).  To quantitatively assess lipid accumulation in HG-HepG2 cells, we performed ORO colorimetric assay [21]. ORO assays on BGE-treated HG-HepG2 cells evidenced about 45% and 80% reduction of the lipid droplets at 0.5 and 1.0 mg/mL, respectively, compared to untreated cells (p < 0.001). Malignant cells express high levels of glucose transporters in connection with an enhanced rate of glycolysis [50]. Some studies have reported beneficial effects of BG preventing some of the cardiometabolic alterations associated with metabolic syndrome, both in humans and in experimental animals [8,9]. In our study, BGE has been shown to contain the steroidal sapogenin Tigogenin. It has been extensively reported that steroidal sapogenins from various plants have anti-obesity effects [51]. Saponins consist of an amphiphilic structure with a hydrophobic steroid or a triterpene nucleus that could form complexes with cholesterol in the lipid bilayer membrane. The formation of the saponin/cholesterol complex may be the cause of the decrease in glucose transport. Therefore, the length of treatment with BGE of HG-HepG2 cells (72 h) can decrease the absorption of glucose and consequently cause a reduction in the accumulation of neutral lipids.
Non-alcoholic fatty liver disease (NAFLD) is a chronic damage due to a dysregulated movement of lipids from adipose tissue, muscle, and gut to liver [21]. The clinical importance of NAFLD is associated with the shift towards non-alcoholic steatohepatitis (NASH), which is in turn able to evolve in liver cirrhosis and cancer, through a series of steps related to oxidative stress and excessive lipid accumulation in hepatocytes [21]. Although there is no treatment for NAFLD and NASH, natural remedies such as black garlic, able to reduce oxidative stress and lower lipid accumulation, could help to control these disorders.

Black Garlic Mehanol Extract Induces Differentiation in Hematological Cancer Cell Lines
Previous research showed that the aqueous extract of BG (10, 50, and 100 mg/mL) can induce antiproliferative effect and apoptosis in the human gastric cancer cells SGC- To quantitatively assess lipid accumulation in HG-HepG2 cells, we performed ORO colorimetric assay [21]. ORO assays on BGE-treated HG-HepG2 cells evidenced about 45% and 80% reduction of the lipid droplets at 0.5 and 1.0 mg/mL, respectively, compared to untreated cells (p < 0.001). Malignant cells express high levels of glucose transporters in connection with an enhanced rate of glycolysis [50]. Some studies have reported beneficial effects of BG preventing some of the cardiometabolic alterations associated with metabolic syndrome, both in humans and in experimental animals [8,9]. In our study, BGE has been shown to contain the steroidal sapogenin Tigogenin. It has been extensively reported that steroidal sapogenins from various plants have anti-obesity effects [51]. Saponins consist of an amphiphilic structure with a hydrophobic steroid or a triterpene nucleus that could form complexes with cholesterol in the lipid bilayer membrane. The formation of the saponin/cholesterol complex may be the cause of the decrease in glucose transport. Therefore, the length of treatment with BGE of HG-HepG2 cells (72 h) can decrease the absorption of glucose and consequently cause a reduction in the accumulation of neutral lipids.
Non-alcoholic fatty liver disease (NAFLD) is a chronic damage due to a dysregulated movement of lipids from adipose tissue, muscle, and gut to liver [21]. The clinical importance of NAFLD is associated with the shift towards non-alcoholic steatohepatitis (NASH), which is in turn able to evolve in liver cirrhosis and cancer, through a series of steps related to oxidative stress and excessive lipid accumulation in hepatocytes [21]. Although there is no treatment for NAFLD and NASH, natural remedies such as black garlic, able to reduce oxidative stress and lower lipid accumulation, could help to control these disorders.

Black Garlic Mehanol Extract Induces Differentiation in Hematological Cancer Cell Lines
Previous research showed that the aqueous extract of BG (10, 50, and 100 mg/mL) can induce antiproliferative effect and apoptosis in the human gastric cancer cells SGC-7901 [44]. Dose-dependent apoptosis was detected in the treated cells by using concentrations 10-to 20-fold higher than those used in our study. The same authors further demonstrated the anticancer ability in a tumor-bearing mouse model. Likewise, the ethanol extract of BG (20,50, and 100 mg/mL) inhibited growth of HT29 colon cancer cells through apoptosis and cell cycle arrest [52].
Then, we characterized the biological effects of the methanol extract of BG by evaluating its potential anticancer activity in myeloid cell lines. In our study, the treatment of U937 and NB4 cells with 2.5 and 5.0 mg/mL BGE for 48 h induced cell proliferation inhibition, and we also found differences in antiproliferative efficacy of BGE between the two cell lines (Figure 4). Our results showed a dose-dependent antiproliferative effect after treatment for 48 h with 2.5 and 5.0 mg/mL BGE concentrations, as clearly evidenced by microscopic analysis (Figure 4). 7901 [44]. Dose-dependent apoptosis was detected in the treated cells by using con tions 10-to 20-fold higher than those used in our study. The same authors further strated the anticancer ability in a tumor-bearing mouse model. Likewise, the etha tract of BG (20,50, and 100 mg/mL) inhibited growth of HT29 colon cancer cells t apoptosis and cell cycle arrest [52].
Then, we characterized the biological effects of the methanol extract of BG by ating its potential anticancer activity in myeloid cell lines. In our study, the treatm U937 and NB4 cells with 2.5 and 5.0 mg/mL BGE for 48 h induced cell proliferatio bition, and we also found differences in antiproliferative efficacy of BGE between cell lines (Figure 4). Our results showed a dose-dependent antiproliferative effe treatment for 48 h with 2.5 and 5.0 mg/mL BGE concentrations, as clearly eviden microscopic analysis (Figure 4). In our conditions, the cell line U937 was more responsive than NB4, so we sel for further investigations on BGE effects. A greater effect was observed after BG ment at the concentration of 5.0 mg/mL after 48 h, as shown by the extremely r number of cells. Consequently, to allow for the appropriate experimental observ the subsequent assays were performed at 2.5 mg/mL BGE for 48 h on U937 cells.
In the same model used in our study, U937 cells, Park et al. (2014) [53] show the hexane extract of BG (2.5, 5, 7.5, and 10 μg/mL) inhibited cell growth. This eff mediated, in a concentration-and time-dependent manner, by signaling cascades death receptor-mediated extrinsic and mitochondria-dependent intrinsic pathw apoptosis (caspase-8, caspase-9, and caspase-3), and by cleavage of poly (ADP-ribo ymerase (PARP), a substrate of caspase-3 considered as an apoptosis biomarker experiments were performed using concentrations of the hexane extract of BG w to 1000-fold lower doses than those used in our study. Then, we intended to ve presence of apoptotic effects of BGE at 2.5 mg/mL on U937 cells. The results show BGE does not seem to activate the caspase-dependent apoptosis pathway at 24 treatment, as evidenced by the absence of cleavage of caspase-9, -8, -3, and PARP 5). In our conditions, the cell line U937 was more responsive than NB4, so we selected it for further investigations on BGE effects. A greater effect was observed after BGE treatment at the concentration of 5.0 mg/mL after 48 h, as shown by the extremely reduced number of cells. Consequently, to allow for the appropriate experimental observations, the subsequent assays were performed at 2.5 mg/mL BGE for 48 h on U937 cells.
In the same model used in our study, U937 cells, Park et al. (2014) [53] showed that the hexane extract of BG (2.5, 5, 7.5, and 10 µg/mL) inhibited cell growth. This effect was mediated, in a concentration-and time-dependent manner, by signaling cascades of both death receptor-mediated extrinsic and mitochondria-dependent intrinsic pathways of apoptosis (caspase-8, caspase-9, and caspase-3), and by cleavage of poly (ADP-ribose) polymerase (PARP), a substrate of caspase-3 considered as an apoptosis biomarker. These experiments were performed using concentrations of the hexane extract of BG with 500to 1000-fold lower doses than those used in our study. Then, we intended to verify the presence of apoptotic effects of BGE at 2.5 mg/mL on U937 cells. The results showed that BGE does not seem to activate the caspase-dependent apoptosis pathway at 24 h post-treatment, as evidenced by the absence of cleavage of caspase-9, -8, -3, and PARP ( Figure 5). All together, our results and the literature data reported above [44,52,53] evi differences from 10-to 1000-fold in the biologically active doses, also depending the BGE is prepared (aqueous, ethanol, methanol, and hexane extract). This fact explain why restoration of the apoptotic way was not detected with our methano extract. In turn, this aspect is likely due to the different combinations of the bioact ondary metabolites present in each extract by which the chemical composition is mined. Moreover, properties could vary from different garlic cultivars due to agro genetic, and environmental factors. More recently, to confirm our results, Toleda dina et al. (2019) [54] showed BGE to induce a decrease in leukemia cells growth, b was unable to induce, consistently with the concentrations used our study, internu mal DNA fragmentation, a marker of the genomic integrity associated with the act of the apoptotic program.
Free NO is an inhibitor of the proliferation of myeloblastic leukemia cells and ducer of monocytic differentiation [40][41][42]. Therefore, the antiproliferative effec BGE was also investigated to verify the activation of cell differentiation pathways. aim, the action of BGE was studied on U937 cell expression of CD11c and CD14, m of granulocyte and monocyte differentiation, respectively. BGE induced matura lated pathways with consistent monocyte and granulocyte maturation of U93 treated with 1.25 and 2.5 mg/mL BGE for 24 h (Figure 6). All together, our results and the literature data reported above [44,52,53] evidenced differences from 10-to 1000-fold in the biologically active doses, also depending on how the BGE is prepared (aqueous, ethanol, methanol, and hexane extract). This fact would explain why restoration of the apoptotic way was not detected with our methanol-based extract. In turn, this aspect is likely due to the different combinations of the bioactive secondary metabolites present in each extract by which the chemical composition is determined. Moreover, properties could vary from different garlic cultivars due to agronomic, genetic, and environmental factors. More recently, to confirm our results, Toledano Medina et al. (2019) [54] showed BGE to induce a decrease in leukemia cells growth, but BGE was unable to induce, consistently with the concentrations used our study, internucleosomal DNA fragmentation, a marker of the genomic integrity associated with the activation of the apoptotic program.
Free NO is an inhibitor of the proliferation of myeloblastic leukemia cells and an inducer of monocytic differentiation [40][41][42]. Therefore, the antiproliferative effect of the BGE was also investigated to verify the activation of cell differentiation pathways. To this aim, the action of BGE was studied on U937 cell expression of CD11c and CD14, markers of granulocyte and monocyte differentiation, respectively. BGE induced maturation-related pathways with consistent monocyte and granulocyte maturation of U937 cells treated with 1.25 and 2.5 mg/mL BGE for 24 h ( Figure 6). Furthermore, the significant increases of free NO and TBARS in the culture medium and cytosol of U937 cells, respectively, observed after treatment with 2.5 mg/mL BGE for 24 h (Figure 2, central and bottom panels), correlated consistently with the expression of CD11c and CD14 markers ( Figure 6). The differentiative activity observed in U937 cells after treatment with BGE also correlates with microscopic analysis (Figure 4), and with viability assay (Figure 2, upper pane), which together indicate an evident cell proliferation inhibition. Although further studies are needed, these important effects strongly suggest that the induction of differentiation can contribute to the BGE-mediated anticancer action. ducer of monocytic differentiation [40][41][42]. Therefore, the antiproliferative effect of the BGE was also investigated to verify the activation of cell differentiation pathways. To this aim, the action of BGE was studied on U937 cell expression of CD11c and CD14, markers of granulocyte and monocyte differentiation, respectively. BGE induced maturation-related pathways with consistent monocyte and granulocyte maturation of U937 cells treated with 1.25 and 2.5 mg/mL BGE for 24 h (Figure 6).

Conclusions
The great interest in functional foods with strong biomedical activities has stimulated the study of BG for its potential beneficial effects on human health. This work showed the metabolite profile of BGE and its potential benefits on liver cells and on leukemic cell lines. To the best of our knowledge, this is the first study in which the BG crude methanol extract was shown to be able to induce cell differentiation. These aspects are particularly relevant in view of the potential health effects obtained from the consumption of this food. On the other hand, the enrichment in the biologically active compounds by heat aging could have likely increased these benefits and helped to promote the functionalization of garlic. Likely, the biological properties showed by BGE are due to the presence in the extract of pharmacologically important compounds, such as the organosulfur compounds S-allylcysteine, alliin and allicin, organic acids, and saponins. However, the exact mechanism of action of all ingredients and their long-term effects are not yet fully understood, so further studies are needed to elucidate these important aspects. Moreover, studies on the human health impact by assessing bioavailability and metabolism have to be carried out. The results of this study will be helpful for the application of BG as a preventive anticancer agent as well as in the prevention and treatment of liver diseases.
Author Contributions: All the authors contributed to the study through the study design, performed experiments, analyzed data, and writing the paper. P.B., P.S., S.P., L.A., A.M.M. and D.R. designed the research; P.B., P.S., S.L., A.N. and D.R. performed the experimental work; P.B., P.S., A.N., S.P., L.D.M. and D.R. analyzed the data and wrote the manuscript. All authors have read and agreed to the published version of the manuscript.