Anthocyanin and proanthocyanidin from Aronia melanocarpa (Michx.) Ell.: Purification, fractionation, and enzyme inhibition

Abstract Aronia melanocarpa (Michx.) Ell. is a rich source of anthocyanins and proanthocyanidins with confirmed health benefits. Individual cyanidin glucosides (cyanidin 3‐galactoside, cyanidin 3‐arabinoside, cyanidin 3‐xyloside, and cyanidin 3‐glucoside) of anthocyanins (calculated by individual cyanin glycoside fractions was 419.9 mg/100 g FW) were isolated by Sephadex LH‐20 column and different parts of proanthocyanidins with a different mean degree of polymerization (mDP) were fractionated by the solubility differences in different solvents. The composition of different mDP of proanthocyanidins was as follows: monomers (1.51%), oligomer (mDP of 4.2 ± 0.9, 20.57%), CPP‐50 (mDP of 78.9 ± 4.1, 22.17%), CPP‐60 (mDP of 66.1 ± 1.2, 27.94%), CPP‐70 (mDP of 36.8 ± 3.9, 36.8%), CPP‐75 (mDP of 25.2 ± 1.3, 6.14%), CPP‐L (mDP of 10.2 ± 2.6, 6.95%), and there were recycling loss of 0.34%. Cyanidin 3‐glucoside showed the strongest inhibition effects on α‐amylase and lipase and cyanidin 3‐arabinoside showed the strongest inhibition effect on α‐glucosidase, while cyanidin 3‐xyloside has no inhibition effect on the α‐amylase, and cyanidin 3‐galactoside, cyanidin 3‐arabinoside, and cyanidin 3‐xyloside have no inhibition effects on lipase. The inhibition effect of proanthocyanidins with different mDP to the enzymes all showed high negative correlations between the mDP and IC50 (half‐maximal inhibitory concentration). This study suggests that A. melanocarpa (Michx.) Ell. can have beneficial effects due to inhibition of the digestion enzyme.

Extraction is the crucial step for isolation, identification, and use of phenolic compounds. However, there is no general extraction method, and different methods will be used according to different raw materials. Pretreatment may be a good choice to release the polyphenol compounds, such as fermentation, enzyme lysis, grinding, and freeze-thawing. The most commonly used techniques for the isolation of phenolic compounds are solvent extraction (with or without assistance of microwave and ultrasound) and supercritical fluid extraction. This method has the advantages of being a simple process, low cost, and high purity. Different adsorbents were tested, such as Amberlite XAD7HP absorption and acidified ethanol elution (Galván D'Alessandro et al., 2013), Sep-Pak C-18 cartridges absorption, 1% acetic acid acidified water (Jing et al., 2008) or methanol (Sosnowska et al., 2018) elution for anthocyanin, and Sephadex LH-20 for the purification of anthocyanins (Bräunlich et al., 2013) and proanthocyanidins (Fan et al., 2007) in order to remove the concomitant substances such as sugars, amino acids, and proteins from anthocyanin and/or proanthocyanidin-containing extracts. For the fractionation of different mDP of proanthocyanidins, a rapid method based on liquid/liquid extraction and relative solubility of these compounds in different solvents (water, ethyl acetate, methanol, and chloroform) was established (Saucier et al., 2001).
The prevalence of diabetes and obesity is alarmingly increasing in the last few decades, leading to many serious public health concerns worldwide. α-Glucosidase, α-amylase, and lipase are the key enzymes that affect the digestion and absorption of major carbohydrates and lipids in diet (Türkan et al., 2020). Thus, inhibiting α-glucosidase, α-amylase, and pancreatic lipase would prevent the breakdown of carbohydrates and triglyceride and delay the absorption of glucose and fatty acids into the systemic circulation and adipocytes (Rajan et al., 2020). However, the conventional αglucosidase, α-amylase, and lipase inhibitor available in clinics are all chemicals (side effects often occurred), and identifying safe clinical alternatives from plants to inhibit these enzymes have been considered a significant advancement (Rajan et al., 2020).
Reports indicated that phenolic in Aronia can inactivate αamylase, α-glucosidase, and lipase through non-specific binding to enzymes (Bräunlich et al., 2013). However, crude polyphenol was F I G U R E 1 Flow chart of extraction, fractionation, characterization, and enzyme inhibition. studied and the individual effects of pure chemicals have not been considered, which will be important for the studying of the mechanism of the inhibition. Therefore, the individual cyanidin glycosides of anthocyanins and proanthocyanidins with different mDP were fractionated and the enzyme inhibition effects were studied ( Figure 1). This study will be useful to elucidate the inhibition effects of polyphenol from A. melanocarpa (Michx.) Ell. for obesity and diabetes and helpful for product development.

| Chemicals and materials
All the chemicals used for analyses were of analytical grade and high- Co., Ltd. (September 18, 2018) and was freeze stored until use.

| Isolation of anthocyanins from A. melanocarpa (Michx.) Ell.
The pre-grounded slurry of A. melanocarpa (Michx.) Ell. (4 kg) was subjected to n-hexane (6:1, v/m) to remove lipophilic substances, after which was divided into two parts (part 1 for anthocyanins isolation and the other part for proanthocyanidins extraction) as pre-treated slurry.
Then, the slurry was extracted twice by ethanol solution (60% ethanol with 0.1% HCl, v/v) for 2 h at ambient temperature bubbled with nitrogen, and was centrifuged at 9090 g for 6 min and the supernatant was collected and combined. Then, the ethanol was removed from the supernatant and the retained solution was concentrated through rotatory evaporation under vacuum at 40°C to give the crude extract of anthocyanins (CEA). The CEA was subjected to a bed of Amberlite XAD-7HP (5 × 50 cm column) until the eluate has no more anthocyanins detected by pH differential method (Jing et al., 2008), followed by elution with methanol (0.1% HCl) to give the anthocyanin-enriched extract (AEE). The AEE was then purified by a Sephadex LH-20 column (5 × 150 cm) by a step gradient of 15% and 30% methanol (0.1% HCl v/v), the fractionation steps follow Bräunlich et al. (2013), and was detected with HPLC at 520 and 280 nm for purity. Then, the individual of the cyanidin glycosides (cyanidin 3-galactoside, cyanidin 3-arabinoside, cyanidin 3-xyloside, and cyanidin 3-glucoside) were obtained.

| Crude proanthocyanidins extract
The pre-treated slurry of A. melanocarpa (Michx.) Ell. was extracted two times successively with 7 L of acetone:water (70:30, v/v) bubbled with nitrogen with mechanical agitation for 6 h. All the slurries were centrifuged at 8000 rpm for 6 min and the supernatants were combined and evaporated at 40°C using a vacuum rotary evaporator to remove acetone. And the aqueous solution was freeze-dried to give the crude proanthocyanidins extract (CPE) powder.

| Isolation of crude oligomer fraction (COF) and crude polymerized proanthocyanidins (CPP) from CPE
The CPE powder was dissolved in redistilled water to 5 g/L followed by ethyl acetate extraction two times with 1:1 (v/v) of the organic and aqueous phase. The ethyl acetate extract was evaporated at 40°C using a vacuum rotary evaporator to dry for crude oligomer fraction from CPE (COF). And the CPP was retained in the aqueous followed by freeze drying for the powder of CPP.

| Removal of monomer from COF
The COF powder was dissolved to 150 mL by redistilled water.
Then solid-phase extraction (SPE) with column (ENVI 18,Supelco) was conducted by applying the dissolved COF on each column.
The monomers were then removed by eluting diethyl ether until no monomers (epicatechin) were detected in the eluent by HPLC (Saito et al., 2006). Then, the remained oligomers were eluted with methanol until no more oligomers were detected by Folin-Ciocalteu method (Tan et al., 2017). Then, the methanol was evaporated, the residue was dissolved in minimum water, and freeze-dried to obtain oligomers without monomers (OWM).

| Fractionation of the polymer with different mDP from CPP
CPP powder was prepared by freeze drying the aqueous solution, followed by dissolving it with methanol. The same volume of chloroform (1:1, v/v) was added to the CPP-methanol solution. Then, the precipitate was harvested by filtration and recovered by methanol washing. Then, this methanol solution was evaporated and the residue was dissolved in water and freeze-dried as a fraction of CPP fractionated by 50% chloroform (CPP-50). The retained filtrate was repeatedly precipitated by adding more chloroform successively to the volume ratio of 1.5:1, 2.33:1, and 3:1 (v/v, chloroform:methanol), and the corresponding precipitate was harvested by filtration, washed in methanol, dissolved in water, and freeze-dried to make the fraction of CPP-60, CPP-70, and CPP-75. And the last filtrate was also evaporated, redissolved in water, and freeze-dried to make CPP-L fraction. All these samples were kept at −80°C until further studies (Saucier et al., 2001;Sosnowska et al., 2018).
proanthocyanidins (10 mg) were dissolved in 1.0 mL of 95% ethanol to prepare a 10 mg/mL proanthocyanidins solution. The thiolysis method was carried out according to Gao et al. (2018).

| Determination of total phenolic contents
The total phenolic content was measured by the Folin-Ciocalteu

| Determination of total proanthocyanidins
Proanthocyanidins were depolymerized into anthocyanidins by use of an n-BuOH-HCl-ferric ammonium sulfate mixture and HPLC was used to determine the content as stated in Skupien and Oszmianski (2007). (−) Epicatechin was used as a calibration standard and results were expressed as mg (−)epicatechin equivalents (EE) per 100 g of FW (mg of EE/100 g FW).

| HPLC and mass spectra determination of the cyanidin glycosides, proanthocyanidins, and thiolytic products
An Agilent 1260 Infinity HPLC (Agilent Technologies, Santa Clara, CA, USA) was used for the purity check of the cyanidin glycosides, (−)epicatechin, catechin, and thiolysis products by an Eclipse XDB-C8 (4.6 × 150 mm, 5 μm) column (Agilent Technologies, Santa Clara, CA, USA) according to the method of Bräunlich et al. (2013) with a diode array detector (DAD). Anthocyanin standard stock solutions were prepared in methanol containing 0.1% formic acid and were used to identify the compounds. Wavelengths of (−)epicatechin at 280 nm and anthocyanin glycosides at 520 nm were determined. The mass spectra of thiolytic products were acquired in positive mode using electrospray ionization on an Agilent 6220 ESI-TOF mass spectrometer. The mDP was calculated by the following formula (Ci et al., 2018): 2.9 | Enzyme inhibition essays 2.9.1 | α-Amylase inhibition studies α-Amylase inhibitory activity was determined using the method from Tan et al. (2017), with a slight modification as Table 1 indicates. One hundred microliters of different concentrations of the fractionated compounds and 100 μL of the enzyme solution (5 mg/L) were mixed in a centrifuge tube at ambient temperature for 10 min. After adding 200 μL, 0.5% soluble starch with phosphate solution as a buffer (25 mmol/L, pH = 6.8), the tube was incubated at 37°C for 5 min. Then, 1 mL of DNS color reagent solution (96 mM 3,5-dinitrosalicylic acid and 5.31 M sodium potassium tartrate in 2 M NaOH) was added into the tube. The tube was placed into a boiling water bath for 5 min to inactivate the enzyme. Then, the solution was diluted by adding 3 mL of distilled water. Then, 200 μL of the mixture was taken and added to a 96-well plate and the absorption at 540 nm was determined. To eliminate the background absorbance produced by the fractionated compounds, an appropriate extract control without enzyme was included. α-Amylase inhibitory activity was measured at five different concentrations, and a logarithmic regression curve was (1) mDP = 1 + area of catechin and epicatechin derivatives area of catechin and epicatechin (1 mol/L) was added to the reaction mixture to terminate the reaction followed by determining the absorbance at 405 nm by a microtiter plate reader. The percentage of inhibition was calculated using Equation 3. α-Glucosidase inhibitory activity was measured at five different concentrations, and a logarithmic regression curve was established to calculate IC 50 values (Gong et al., 2020;Tan et al., 2017).

| Lipase inhibition assay
The lipase inhibitory activity was determined according to the method described by Tan et al. (2017) with modifications (Table 3).

| Statistics
All measurements were performed in triplicate and the results were presented as mean ± standard deviation (SD). The data and figures were processed with Origin 9.0 and Adobe Photoshop CC 2018. And SPSS 26.0 was used to evaluate whether the data of mDP and IC 50 of different enzymes conform to normal distribution.

| Extraction, fractionation, and characterization of anthocyanins
Two-thousand-gram A. melanocarpa (Michx.) Ell. was used to extract and purify the crude anthocyanins and its cyanidin glycosides.
The results are shown in Table 5 and Figure 2. For crude extract of anthocyanins (CEA), there were not only anthocyanins but also water-soluble compounds such as protein, sugar, and vitamin.
Thus, more compounds were obtained at this stage. Followed by the absorbance of anthocyanins using Amberlite XAD7HP, most of the water-soluble compounds have been washed off and methanol was used to get the high-purity anthocyanin complex. There were mainly four cyanidin glycosides found in A. melanocarpa (Michx.) Ell. such as cyanidin 3-galactoside, cyanidin 3-arabinoside, cyanidin 3-xyloside, and cyanidin 3-glucoside, which were purified and showed the same migration time compared with the standard individual anthocyanins (Figure 2). Their chromatographic spectrum was in agreement with previous studies (Bräunlich et al., 2013).
Also, there still was an 11.08% loss of AEE which was ascribed to TA B L E 4 Content of total polyphenol, proanthocyanidins, and anthocyanin of A. melanocarpa (Michx.) Ell.

F I G U R E 2 HPLC profile of different individual anthocyanins.
Profile a-d are the solutions fractionated by a Sephadex LH-20 column. (a) for cyanidin 3-xyloside, (b) for cyanidin 3-arabinoside, (c) for cyanidin 3-glucoside, and (d) for cyanidin 3-galactoside, (e) is the standard individual anthocyanins as the reference compounds to identify the different fractions, I for cyanidin 3-galactoside, II for cyanidin 3-glucoside, III for cyanidin 3-arabinoside, and IV for cyanidin 3-xyloside.
the impurity substances in AEE, and elute and recycle loss during the fractionation process. Cyanidin 3-galactoside and cyanidin 3-arabinoside are the predominant representatives of 82.57%, although this was lower than Oszmiański and Wojdylo (2005) reported with a cumulative content of >90% in the berries (Oszmiański & Wojdylo, 2005).
The total anthocyanins content calculated by individual cyanin glycoside fractions was 419.9 mg/100 g FW, which was lower than that determined by pH differentiation method as shown in Table 5.
This error may have come from the recycle loss of the fractionation method, and impurities were calculated as anthocyanins during pH differentiation determination.  (Figure 3 and Table 6). The chain extension units and chain-terminating units in proanthocyanidins were predominant as (−) epicatechin trace catechin (Figure 3), and the small peak 1 may be catechin. The mDP value has been calculated according to the thiolysis of proanthocyanidins as indicated in Table 7.

| Extraction, fractionation, and characterization of proanthocyanidins
The crude proanthocyanidins extract (CPE) was fractionated into six parts with different mDP. The mDP of the CPE was 49.3 ± 3.8, almost as reported by Skupien and Oszmianski (2007). After fractionation of the CPE, the proanthocyanidins were nicely fractionated with different mDP from 4.2 to 78.9, which will be ready for the studying of the enzyme inhibition with different mDP of proanthocyanidins as indicated in Table 7. The sum of OWM and CPP contents were the total proanthocyanidins, which was a little bit high than reports and also has reported an exceptionally high mDP > 100 (Skupien & Oszmianski, 2007), differences may be caused by the different processing, storage, cultivation, determination methods, etc. Labarbe et al. (1999) also found that the proanthocyanidins from the seed and skin of grape also can be fractionated with different mDP, ranging increasingly from 4.7 to 17.4 in seed (8.1 for total extract) and from 9.3 to 73.8 in skin (34.9 for total extract). The monomer content of proanthocyanidins was the highest compared with several reports with content of 5.89 mg ECE/100 g FW (Dudonné et al., 2015), 5.17 mg/100 g FW (Wu et al., 2004) there were recycling loss of 0.34%. Wu et al. (2004) reported that the proanthocyanidins composition in Aronia is as follows: monomers (0.78%), dimers (1.88%), trimers (1.55%), 4-6-mers (6.07%), 7-10-mers (7.96%), and >10-mers (81.72%) (Wu et al., 2004). This   , 2022). Also, their pharmaceutical effects are interesting because they act beneficially on the circulatory system and are efficient free radical scavengers (Rigaud et al., 1993). Several studies found that proanthocyanidins with high mDP are more potent antioxidants than simple phenolics, and the increasing mDP may enhance the antioxidant power and lipase inhibition of proanthocyanidins (condensed tannins) (Sosnowska et al., 2016), which will influence its organoleptic or pharmacological properties. However, the high mDP feature should limit their absorption through the gut barrier; oligomers larger than trimers are unlikely to be absorbed in the small intestine in their native forms (Denev et al., 2012).

Peak number RT (min) [M-H] − (m/z) Compounds
These studies established a rapid method to fractionate different mDP proanthocyanidins by different solvents only, which avoid the poor resolution and irreversible adsorption during chromatographic separations of the highly polymerized fraction (Labarbe et al., 1999).

| Inhibition of the enzyme
The effects of crude extractions of anthocyanins and proanthocyanidins, the individual cyanin glycoside and proanthocyanidins with different mDP on the α-amylase, α-glucosidase and lipase are studied to identify the fractions of anthocyanins and proanthocyanidins that are responsible for the anti-diabetes and obesity functions in A. melanocarpa (Michx.) Ell., The results were showed in Figure 4 and Table 7.
Among the individual cyanidin glycoside and crude anthocyanins extracts, cyanidin 3-glucoside showed the strongest inhibition effects on α-amylase and lipase and cyanidin 3-arabinoside showed the strongest inhibition effect on α-glucosidase, while cyanidin 3-xyloside has no effect for the inhibition of α-amylase.
And cyanidin 3-galactoside, cyanidin 3-arabinoside, and cyanidin 3-xyloside have no inhibition effects for the lipase. Cyanidin 3-arabinoside showed the strongest inhibition effect on the three enzymes of the anthocyanins studied. These results showed the same tendency as indicated in Table 8. Considering the data obtained from our investigations ( Figure 4) and the reports in Table 8, cyanidin glycosides may be one of the reasons for the diabetes treatment effect of A. melanocarpa (Michx.) Ell. (Chrubasik et al., 2010), and studies suggested that intake of cyanidin and thus reducing substrate-binding affinity (Wei et al., 2017). All the proanthocyanidins fractions with different mDP showed strongest inhibition effects than anthocyanins individuals on α-amylase. And CPP-50 showed stronger α-glucosidase and lipase inhibition than that of all the anthocyanins. This may be ascribed to the most effective protein-precipitating effect of large molecules of proanthocyanidins (Hofmann et al., 2006), which is of particular importance, as bioavailability is not needed for proanthocyanidins of any size to exert activity by inhibiting the digestion of lipids or carbohydrates in the gastrointestinal lumen (Neilson et al., 2016).

| CON CLUS IONS
A. melanocarpa (Michx.) Ell. has been transplanted to China for about 12 years, and it has only been allowed to be sold as food in China for not more than 4 years. Therefore, the functional components and functions of the fruit planted in China still need further study. The results showed that the content, composition, and enzyme inhibi- writing -original draft (equal); writing -review and editing (equal).

Many thanks to Mr. Shifu Chen and Liaoning Fukangyuan Black
Chokeberry Technology Co., Ltd. for their support with Aronia melanocarpa (Michx.) Ell.

FU N D I N G I N FO R M ATI O N
This research was funded by the National Key R&D Program of the Ministry of Science and Technology of China (2018YFA0902200).

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.