Phytochemicals and Enzyme Inhibitory Capacities of the Methanolic Extracts from the Italian Apple Cultivar Mela Rosa dei Monti Sibillini

The phytochemical profile of the methanolic extracts (pulp and peel) obtained from two dehydration methods (drying and freeze-lyophilization) of the traditional Italian apple Mela Rosa dei Monti Sibillini, as well as their inhibitory properties against some biological enzymes (α-glucosidase, lipase, monoamine oxidase A, tyrosinase and acetylcholinesterase) were assessed in this study. HPLC-DAD-MS technique was used for the determination of polyphenolic and triterpenic compounds. The determination of the enzymes inhibitory effect was made through spectrophotometric techniques. The peel extracts were richer in bioactive compounds than the pulp. In this regard, the extracts from freeze-lyophilization displayed higher levels of flavan-3-ols, flavonol glycosides and dihydrochalcones. However, the extracts obtained from dried material displayed a stronger enzyme inhibition. Notably, the peel extracts showed a higher activity than the pulp ones, especially in terms of α-glucosidase whereby some samples exerted a similar enzymatic inhibition than acarbose (100% inhibition) at high concentrations (1 mg/mL). These results encourage thus further studies on this traditional Italian apple as a potential source of nutraceuticals helpful to prevent the insurgence of some pathologies.


Introduction
The high consumption of fruits and vegetables has been associated to a good health state. Indeed, it is reported that the low mortality rate due to fruit and vegetable consumption is due to the phytochemicals they contain [1]. Apple fruit (Malus domenica Borkh) have shown to be one of the most important dietary sources of polyphenols, whose consumption has been associated with human wellness [2]. It has also been reported that traditional and overlooked apples were proved to be higher sources of phytonutrients than the commercial ones [3]. Despite their particularity (shape, taste, nutritional values), these old apples are on the verge of extinction because of globalization. Hence, the characterization of old apple cultivars is important to enhance their value and increase their production [4,5].

Hydroxycinnamic Acids
In pulp extract analyses, as in the case of peel, the LEPu had a higher content of polyphenol compounds than the DEPu. The most abundant classes present in both DEPu and LEPu were flavan-3-ols and hydroxycinnamic acids. Epicatechin (494.5-1725.2 mg/kg for LEPu and 586.9-1410.9 mg/kg for DEPu), chlorogenic acid (167.  This analysis also showed variability between the samples studied. In the DEPe, a variation of 15.2% of the total polyphenol was noticed and was similar to that of the LEPe. Whereas in the LEPe, the triterpenes content varied highly in respect to that of the DEPe (83.7 and 56.7% respectively). In the pulp samples, the variability was quite similar for both total polyphenols (29.4 and 31.1% for DEPu and LEPu respectively) and triterpenes concentration (41.2 and 41.8% for DEPu and LEPu respectively).

α-Glucosidase Inhibition
From the results, both DEPu and LEPu at concentrations of 1 mg/mL inhibited the enzyme at a percentage less than 50% (Figure 1a). Similarly, almost all LEPe (exception with sample 5) and samples 1, 2, 3 and 4 of DEPe have shown an inhibitory effect less than 50%. While at the same concentration (1 mg/mL), DEPe from samples 5 and 6 inhibited α-GLU with a percentage higher than that of acarbose (90% and 80%, respectively) ( Figure 1b). DEPe and LEPe inhibited α-GLU in a concentration dependent manner (Figure 1c). IC 50 values calculated by nonlinear regression are reported in Table 3.

Lipase Inhibition
The extracts showed lipase inhibitory activity ( Figure 2c). Indeed, at concentration of 1 mg/mL lipase was inhibited at a percentage lower than 50% by all the samples even if the pulp extracts in this case seemed more active than the peel ones (Figure 2a,b). In contrast, at 10 mg/mL of DEPe, the inhibition obtained was 100%. IC 50 values were calculated by nonlinear regression (Table 3).

MAO-A Inhibition
The extracts were able to inhibit MAO-A in a dose-dependent manner with a similar profile of clorgyline ( Figure 3c). The percentages of inhibition were between 67% and 88% with all the samples studied at 1 mg/mL of extract concentration (Figure 3a,b). However, there were significant differences between the IC 50 of the extracts (DEPe: 533 µg/mL; LEPe: 473 µg/mL; DEPu: 1793 µg/mL and LEPu: 846 µg/mL) and the reference (0.2 µg/mL), calculated by nonlinear regression (Table 3).

AChE Inhibition
The extracts had a lower inhibitory effect against AChE than the positive control, galantamine (Figure 4a-c). At 1 mg/mL, the DEPu seems to be more active (inhibition less than 52%) than the DEPe and LEPe (inhibition less than 27%). On the other hand, LEPu showed no activity against the enzyme. Furthermore, the values of IC50 obtained from extracts (DEPe: 1889 μg/mL; LEPe: 2261μg/mL and DEPu: 3963μg/mL) were significantly different from those achieved by galantamine (Table 3).

AChE Inhibition
The extracts had a lower inhibitory effect against AChE than the positive control, galantamine (Figure 4a-c). At 1 mg/mL, the DEPu seems to be more active (inhibition less than 52%) than the DEPe and LEPe (inhibition less than 27%). On the other hand, LEPu showed no activity against the enzyme. Furthermore, the values of IC 50 obtained from extracts (DEPe: 1889 µg/mL; LEPe: 2261 µg/mL and DEPu: 3963 µg/mL) were significantly different from those achieved by galantamine (Table 3).

TYR Inhibition
As can be noted from the following figures (Figure 5 a,b), the extracts were not able to inhibit significantly TYR at concentration of 1 mg/mL (inhibition less than 50%), and increasing the concentration to 10 mg/mL, the percentage remains almost the same (between 49% and 60%). IC50 values were calculated by nonlinear regression (Table 3). One-way analysis of variance (ANOVA) followed by Dunnet's test for multiple comparisons: ** p < 0.01; *** p < 0.001 against galantamine. DEPe: dried peel methanolic extract; LEPe: lyophilized peel methanolic extract; DEPu: dried pulp methanolic extract; LEPu: lyophilized pulp methanolic extract.

TYR Inhibition
As can be noted from the following figures (Figure 5a,b), the extracts were not able to inhibit significantly TYR at concentration of 1 mg/mL (inhibition less than 50%), and increasing the concentration to 10 mg/mL, the percentage remains almost the same (between 49% and 60%). IC 50 values were calculated by nonlinear regression (Table 3). Pharmaceuticals 2020, 13,

Correlation Analysis between Phytochemical Composition and Bioactivity
In order to establish a relationship between phytochemical content and the enzymatic inhibitory capacity of the extracts, correlation analyses were performed for each type of extract. Results showed a strong negative correlation between the content of triterpenes and MAO-A inhibition (Pearson r Value = −0.8241) in LEPe, without a good correlation between polyphenols and the MAO-A bioassay (Pearson r Value = −0.2331). A strong positive correlation was found in the DEPu sample between the content of phenolic compounds and tyrosinase inhibition (Pearson r Value = 0.7924). Additionally, a very strong positive correlation in the analysis was found for the triterpene content versus lipase inhibition (Pearson r Value = 0.9568) in the same sample (Table 4). One-way analysis of variance (ANOVA) followed by Dunnet's test for multiple comparisons: * p < 0.05; ** p < 0.01; *** p < 0.001 against kojic acid. DEPe: dried peel methanolic extract; LEPe: lyophilized peel methanolic extract; DEPu: dried pulp methanolic extract; LEPu: lyophilized pulp methanolic extract.

Correlation Analysis between Phytochemical Composition and Bioactivity
In order to establish a relationship between phytochemical content and the enzymatic inhibitory capacity of the extracts, correlation analyses were performed for each type of extract. Results showed a strong negative correlation between the content of triterpenes and MAO-A inhibition (Pearson r Value = −0.8241) in LEPe, without a good correlation between polyphenols and the MAO-A bioassay (Pearson r Value = −0.2331). A strong positive correlation was found in the DEPu sample between the content of phenolic compounds and tyrosinase inhibition (Pearson r Value = 0.7924). Additionally, a very strong positive correlation in the analysis was found for the triterpene content versus lipase inhibition (Pearson r Value = 0.9568) in the same sample (Table 4).

Discussion
Two methods were used in this study for the dehydration of the fresh fruit: drying at 45 ± 5 • C and freeze-lyophilization. HPLC-DAD analysis revealed a higher content of phytochemicals in the LEPe and LEPu than in the DEPe and DEPu that can be due to the neutralization of the degradative enzymes by liquid nitrogen (−195.8 • C) used during the crushing of the fresh samples [2]. The peel extracts showed the highest content in polyphenols than the pulp ones because of the accumulation of these compounds in the peel according to the ecological role of this part such as protection against ultraviolet radiations, attraction for fruit dispersion and defense against pathogens [20,21]. Indeed, it is also known that triterpenes are concentrated on the surface of fruit peel [22]. This confirms the high concentrations detected in our samples. The variability of the total polyphenols and triterpenes content in extracts obtained from dried and freeze-lyophilized materials showed that the dehydration method used can influence the phytochemical content of a sample and is thus an important parameter to consider during the preparation of samples.
An increasing interest in the utilization of natural products as candidates for drug discovery, coadjuvant or alternative to drugs (food supplements, nutraceuticals) in the treatment of different pathologies has been demonstrated in the last years [23]. Thus, the present study was conducted in order to evaluate the nutraceutical or pharmaceutical potential of an old Italian apple variety. In particular, it was assayed the possible inhibitory effect of the apple extracts towards the enzymes α-glucosidase (α-GLU), lipase, monoamine oxidase A (MAO-A), tyrosinase (TYR) and acetylcholinesterase (AChE). These enzymes are related to pathologies such as diabetes, obesity, neurodegenerative disorders and melanogenesis. The phytochemical composition, in terms of concentration of phenolics and triterpenes [24], is strictly related to the inhibition of these enzymes.
In our investigation, the extracts from dried peels inhibited better α-GLU than the ones from freeze-lyophilized peels, especially the samples 5 and 6 that showed a greater activity than the reference compound although without significant differences. This may be the result of the formation of some bioactive compounds during apple drying as reported by Birtic et al. [25].
As known α-GLU and lipase are digestive tract enzymes, involved in the metabolism of carbohydrates and fats, respectively. The inhibition of α-GLU can thus have an impact on diabetes treatment due to the reduction of intestinal absorption and decrease of post-lunch insulin values, maintaining the glycemic variations under control [26]. It is reported that the preventive effects of polyphenols against diseases such as diabetes and obesity may be the results of the modulation of receptors and enzymes such as α-GLU and lipase [10]. Some flavan-3-ols (e.g., catechin) showed an inhibitory effect against the enzymes α-GLU and lipase [27]. However, according to our results, we did not find a positive correlation for polyphenols and α-GLU inhibition.
It is reported that procyanidins inhibit the gastrointestinal lipase, thus decreasing the plasma triglycerides [28]. Triterpenes such as ursolic acid are known to significantly inhibit the pancreatic lipase [29]. In fact, a very strong positive correlation was found in this study between triterpenes and lipase inhibition in the DEPu samples, which reveals that these compounds are responsible for this activity at least in that extract. Quercetin and other flavonoids contributed significantly to the inhibition of the MAO-A, as well their antioxidant activity is related to the central protective action [30]. The polyphenols contained in the apple are able to cross the brain-blood barrier and for this reason showed antidepressant activity [31]. It has previously been reported that the flavonol quercetin and some polyphenol-rich extracts are able to inhibit the enzymes anticholinesterase (AChE) and butyrylcholinesterase (BChE), demonstrating thereby neuroprotective effect against pathologies such as Alzheimer's disease [9]. It is noteworthy to mention that in spite of the use of the pulp for food and nutritive purposes, the peel contains a higher proportion of phytochemicals exerting better bioactive potential. There are no significant differences in triterpenes composition between the crude methanolic extracts prepared from dry material and the ones prepared from the freeze-lyophilized material, although the latter showed higher concentrations of polyphenols. Thus, the complex phytochemical composition and the synergism of each phytochemical in the extracts can be the result of the activity seen in the study [32]. The high phytochemical content of the peel extracts might justify their effectiveness with respect to pulp samples and their use for pharmaceutical applications.
The maximum activities achieved by the different extracts were at high concentrations. Although these doses are not physiological, these are in vitro studies, where the ability of the extracts to interact with the different enzymes is confirmed. In order to corroborate this action at a physiological level, subsequent studies should be considered.

Sampling and Preparation of Apple Extracts
For this study, 8 apples (8 samples) (Table 5) of the cultivar Mela Rosa dei Monti Sibillini cultivated at an altitude between 250 and 500 m a.s.l. and harvested in October-November 2018 were collected from different farmers of the Montedinove, Montottone and Monterinaldo municipalities in the Marche region, Central Italy. The peel of the fruits (at least 5 fruits for each population of apples) was separated from the pulp. A portion (pulp and peel) was dried at 45 ± 5 • C for at least 18 h using a Biosec De Luxe B12 dryer (Albrigi Luigi, Verona, Italy) while the other part was crushed in a mortar with liquid nitrogen and lyophilized until the material was well dehydrated (Buchi, Cornaredo, Italy). The dehydrated material was then powdered using 2 mm-size particles using an IKA-WERK MFC DCFH 48 (Staufen, Germany).
All the dehydrated materials (dried and freeze-lyophilized) were submitted to the same bioactive compounds extraction with methanol as solvent [33,34] using an ultrasound bath (Ultrasonic Falc, Trviglio, BG, Italy). After a sonication for 45 min at room temperature of the sample with methanol (ratio 1 g of dehydrated material with 5 mL of methanol) and filtration, the residue was recovered and submitted to a further sonication for 20 min (ratio 1 g of dehydrated material with 4 mL of methanol). The filtrates of both sonication were gathered and concentrated under vacuum at 35 • C for at least 1 h using rotavapor.

Enzyme Inhibitory Activities
In all the enzyme inhibition assays, the concentration of 1 mg/mL was first used in order to compare the different samples extracts and the reference inhibitors. After this screening, one of the samples with the highest activity was selected to perform the dose response curves.

α-Glucosidase Inhibition Assay
α-Glucosidase inhibition assay was performed following the previous method [35]. Each 96-well microplate contained: 50 µL apple extract or the reference at different concentration and 100 µL of α-GLU (1U/mL) in phosphate buffer (pH 6.9). After 10 min, 50 µL of PNPG (50 µL) were added and incubated for 20 min at 37 • C. The absorbance was read at 405 nm. The percentage inhibition of α-glucosidase was calculated using the following formula: Acarbose was used as a positive control and IC 50 values were determined.

Lipase Inhibition Assay
A lipase inhibition assay was performed according to a previous method [35]. The extract or orlistat (reference inhibitor) at different concentrations were mixed with 40 µL of lipase type II (2.5 mg/mL in Tris-Buffer, pH 7.0). After 10 min of incubation at room temperature, 20 µL of pNPB (10 mM) were added. A 96-Well microplate was incubated for 15 min at 37 • C and the absorbance was read a 405 nm. The percentage inhibition on lipase was calculated using Equation (1).

Monoamine oxidase (MAO-A) Inhibition Assay
MAO-A inhibition assay was performed according to a previous protocol [12]. In each well the following reagents were added: 50 µL of apple extract or reference inhibitor (clorgyline) at different concentrations, 50 µL of chromogenic solution (0.8 mM vanillic acid, 417 mM 4-aminoantipyrine and 4 U/mL horseradish peroxidase in potassium phosphate buffer, pH = 7.6.), 50 µL of samples or reference inhibitor at different concentrations, 100 µL of tyramine (3 mM) and 50 µL of MAO-A (8 U/mL). Blanks and control wells were made. The absorbance was read at 490 nm every 5 min during 30 min. Clorgyline was used as reference inhibitor. The MAO-A inhibitory activity was calculated by the Equation (1).

Statistical Analysis
The experiments were carried out in three replicates on different days. Analysis was performed using GraphPad Prism v.6. The results are expressed as mean ± standard error (±SEM). The differences between the different extracts were analyzed using one-way analysis of variance (ANOVA) followed by Dunnet's test for multiple comparisons with a confidence interval of 95%. p values ≤ 0.05 were considered as significant differences. Pearson correlation analyses were performed in all samples (DEPu, LEPu, DEPe, LEPe) between polyphenol or triterpene content and bioactivity in terms of % of inhibition of the enzymes (α-GLU, lipase, MAO-A, AChE, tyrosinase) at 1 mg mL −1 . Correlation coefficients (r) and statistical significances were calculated using GraphPad Prism v.6.

Conclusions
This study investigated the phytochemical composition as well as the enzyme inhibitory properties of the extracts of an overlooked traditional apple, the Mela Rosa dei Monti Sibillini. Its polar extracts, especially those obtained from the peel have demonstrated a rich content of bioactive compounds such as flavan-3-ols, flavonols, dihydrocalchones, and triterpenes. The freeze-lyophilization dehydration method was more effective in maintaining the phenolic constituents than the drying method. Some extracts have demonstrated inhibitory properties against α-GLU, lipase and MAO-A.
The extracts obtained from the peel dried material exert better bioactivities than those obtained from lyophilized material. These results thereby demonstrated that this variety is a potential source of bioactive compounds for the production of pharmaceuticals or nutraceuticals to be used for the prevention and co-treatment of pathologies such as diabetes, obesity, Alzheimer's disease, depression, and hypermelanosis.