Determination of nine bioactive phenolic components usually found in apple juice by simultaneous UPLC‐MS/MS

Abstract The functional food ingredients of apple juice can significantly change during processing, transportation, and storage, thus affecting the quality of the product. A simple and derivation‐free analytical method based on ultra‐high‐performance liquid chromatography–tandem mass spectrometry (UPLC‐MS/MS) was developed and optimized for the simultaneous determination of functional food ingredients in apple juice bought in the market. Cleanup steps and chromatographic conditions were optimized to remove interference and decrease the matrix effect. The nine target analytes were separated on an Acquity UPLC system equipped with a BEH C18 column and detected by electrospray ionization source (ESI) operating in positive subsection acquisition mode under multiple reaction monitoring (MRM) conditions. The results showed that p‐hydroxybenzoic acid, protocatechuate, caffeic acid, chlorogenic acid, epicatechin, phloridzin, hyperoside, procyanidin B2, and rutin could be sufficiently separated for content determination within 6 min. In the concentration range of 20 μg/L–50 mg/L, nine standard samples exhibited a good linear fit with correlation coefficients above .985.

immunopotentiators (Naveed et al., 2018). Evidence from studies in humans and experimental animals suggested that consumption of epicatechin (EC) and EC-rich foods may improve insulin sensitivity. Cremonini et al. (2016) investigated the ability of EC supplements to prevent insulin resistance in high-fat diet (HFD)-fed mice. The results indicate that eating EC-rich foods may be a viable dietary strategy to reduce obesity-related insulin resistance (Cremonini et al., 2016).
Phlorizin is a natural product and a dietary constituent found in a number of fruits that have already been approved as a food additive for long-term use with few side effects. Phlorizin was found to activate the Tyk2/STAT3 signaling pathway to induce thermogenesis in brown adipose tissue, with therapeutic potential for the treatment of obesity and comorbidities (Yuan et al., 2016). Proanthocyanidin B2 is a highly effective natural antioxidant that can remove free radicals and inhibit the occurrence of lipid peroxidation, helping maintain the homeostasis between free radicals and antioxidant enzymes in the body and prevent related diseases (Acosta-Estrada et al., 2014).
In addition, high concentrations of protocatechuate in apple peels have been shown to exert anti-inflammatory, antihyperglycemic, and antiapoptotic activities in animal studies using rats and mice (Yoswaris et al., 2015). Apples also contain hyperoside, which can alleviate high glucose-induced vascular inflammation (Tian et al., 2018). Caffeic acid can attenuate lipopolysaccharide-induced disease behavior and neuroinflammation in mice (Mallik et al., 2016).
Rutin is important for the aroma of certain apple cultivars (Damin et al., 2019) and p-hydroxybenzoic acid (PHBA) is a controversial class of chemicals used as cosmetic and pharmaceutical preservatives (Guo et al., 2014). However, some experimental evidence suggests that it is also an endocrine disruptor and that there is an association between human exposure to parabens and adverse health outcomes (Kang et al., 2002).
Fruit juice could be a great alternative to whole fruits (Wootton-Beard & Ryan, 2011). Compared with fruits, fruit juice has the advantages of convenient carrying, long shelf life, and few seasonal restrictions. Its fluidity makes it easier for the elderly and children to ingest. Juice demand is growing rapidly as consumers are looking for convenient fruit products with good taste and nutritional qualities (Teck-Chai et al., 2012).
Apple juice is one of the most popular fruit juices in the world because of its agreeable taste, easy processing leading to a low price, and widely known health benefits (Wodarska et al., 2016).
Currently, apple juice is consumed by millions of people every day (Tian et al., 2018). Polyphenolic compounds account for most of the antioxidant activity of apples and are distributed in the skin, pulp, and seeds. However, most of these apple polyphenols are removed during fruit juice production (Pei et al., 2017). Clarification is the step in which most antioxidants are lost from the final product, significantly reducing the polyphenol content because these compounds are mainly found in the pulp (Fahad et al., 2018). Phenolic compounds are sensitive to process conditions during juice extraction as well as further thermal and nonthermal processing (Kong et al., 2016). In order to determine whether the phenolic small molecules in apples are preserved in apple juice, the nine active small molecules reported above were selected to study their contents in various apple drinks (Matos et al., 2017).
Finding an efficient and reliable analytical method to simultaneously measure these nine compounds is a challenge. So far, several analytical methods have been reported for the determination of phenolic substances in food samples. These methods include flow injection methods, the Fast Blue BB method, GC-MS, and LC-MS methods. Costin et al. (2003) used the Fast Blue BB method to estimate total phenol/antioxidant levels in wine, and although the method provided the total phenolic content quickly, it could not quantify a specific compound (Costin et al., 2003). Medina (2011) quantified polyphenols or phenols by direct interaction with Fast Blue BB in an alkaline medium.
This method is economical but is also not specific and can only detect the total content of a class of substances (Medina, 2011). Zuo et al. (2002 used GC-MS to detect phenolic substances in food samples (Zuo et al., 2002). However, GC-based methods are not suitable for the determination of nonvolatile or thermally labile substances.
Some highly sensitive methods were also used for the analysis of functional food ingredients such as HPLC-FLD (Bi et al., 2018). In order to facilitate the analysis of phenolic compounds, it is necessary to avoid derivatization, as this takes a significant amount of time. As an important modern method, LC-MS is highly favored in the food industry due to its simplicity, speed, sensitivity, and accuracy (Bi et al., 2018). In this study, we applied the advantages of this method to establish a rapid protocol for the determination of nine representative phenolic active substances in apple juice.

| Sample source
In this experiment, 18 major brands of apple juice commonly found in the Chinese market were selected. Among them are seven kinds of pure apple juice (PAJ1-7), all of which are labeled as 100% pure apple juice, with the ingredient list indicating 100% apple, or containing only water and apple juice concentrate. Six medium-sized composite apple drinks (MAJ1-6) have a more complex ingredient list with more ingredients, but they contain apple juice concentrate, most of which are in the fourth or fifth place. Three apple-derived apple vinegar drinks (VAJ1-3) contain basically apple vinegar concentrate and apple juice concentrate in the ingredient list. In one apple-flavored carbonated beverage and one apple-flavored lactic acid bacteria beverage (OAJ1-2), the ingredient list contains apple juice concentrate, but is ranked lower. A representative sample consisting of at least three samples collected at different retail stores or large supermarkets for a period of 3 months (January to March 2021).

| Materials and standards
Reference standard for nine phenolic active small molecules, i.e., hyperoside, epicatechin, caffeic acid, phloridzin, protocatechuate, procyanidin B 2 , chlorogenic acid, rutin, PHBA sourced from Shanghai Biological Technology Co., Ltd. with the purity of HPLC grade 98%-99%. The structure of the compound is shown in Figure 1. Methanol and acetonitrile were of HPLC grade from American Tedia company and water for ultrapure water.
ACQUITY system ultra-high-performance liquid chromatographytandem quadrupole mass spectrometry (Waters, companies in the United States, including automatic dual gradient pump, injector, column temperature box, and Masslynx4.1 mass Spectrometry workstation).

| Preparation of the test solution
Preparation of apple extract solution: 2-mg apple extract dissolved in 1 mL of acetonitrile, mixed the solution at a centrifugal speed of 1280 g, 600-μL supernatant was taken and dried, 1 mL of acetonitrile and 200-μL methanol added in 600-μL supernatant.
Preparation of apple juice solution: 400-μL apple juice was taken in the pipette and added to 600-μL acetonitrile solution, mixed the solution at a centrifugal speed of 1600 rpm, 600-μL supernatant was taken and dried, 1 mL of acetonitrile and 200-μL methanol added in 600-μL supernatant.

| The preparation of standard solutions
The PHBA, protocatechunate, caffeic acid, chlorogenic acid, epicatechin, phloridzin, hyperoside, procyanidin B 2 , and rutin standard were weighed in a certain mass using a high-precision analytical balance. The hydroxybenzoic acid standard was prepared as 20, 10, 5, F I G U R E 1 Nine phenolic active molecular compound structures.

| Chromatographic and mass spectrometric conditions
Chromatographic column: WATERS ACQUITY UPLC BEH C18   Table 1. MS spectra were studied in both positive and negative ion modes. In the negative ion mode, the response value was higher, as was the sensitivity for some of the compounds, and the mass spectrum obtained in the negative ion mode was clearer, so that the separation of each ion peak was more obvious and easier to recognize. Therefore, the negative MS ion mode was chosen for further study.

| Optimization of UPLC-MS/MS conditions
The precursor-product ion pairs for MRM detection were generated by the Intellistart protocol (automatic tuning and calibration of the Waters Xevo TQD), which was embedded in the MassLynx software. Otherwise, the signal of each compound has to be manually optimized by altering the cone voltage and collision energy.

| Results of separation
Under the above conditions, the nine apple phenolics exhibited a good separation degree within 6 min at the same run time. If only for separation, the time can be shorter (up to 3 min in this study).
However, in order to better maintain reproducibility and protect the chromatographic column from the accumulation of impurities in actual biological samples. We should better extend the retention time to 6 min. A sample comprising 50 μg/mL of an authentic reference standard was used to optimize single extraction, and the total ion chromatograms of the mixture are shown in Figure 2, respectively.

| Linear range
Using the above method for the preparation of standard solutions, the integral of the peak area was plotted as the ordinate Y, the concentration as the abscissa X. The weight coefficient was 1/X, and the standard curve was obtained by fitting to a linear equation (Table 2).
The linear range of compound quantification was obtained by diluting each stock solution with a mixed solution of methanol and water to obtain seven different concentrations. The correlation coefficients were between .98574 and .99797.

| Determination of functional food ingredients
The 18 batches of samples prepared using the above method were measured and the contents of each component in the samples were calculated, as shown in Table 3.
Protocatechuate was only detected in one sample, while chlorogenic acid was detected in 11, phloridzin in 13, and hyperoside in three of the 18 samples. The contents of PHBA, caffeic acid, epicatechin, proanthocyanidin B2, and rutin were below the detection limit. All apple vinegar samples contained chlorogenic acid at concentrations of 0.03-0.05 mg/L.
The detection and analysis method established in this article has significantly higher sensitivity compared to other research and analysis methods such as HPLC (Alu'datt et al., 2017), GC-MS (Zuo et al., 2002), and The Fast Blue BB method (Medina, 2011).
Moreover, this characteristic of being able to simultaneously determine nine phenolic compounds in this article is significantly more efficient. Some studies have shown that most polyphenols in apple juice are not heat resistant, and heat treatment may lead to degradation or hydrolysis (Tian et al., 2018). Therefore, different processing techniques for apple juice may lead to the loss of phenolic substances in apple juice to varying degrees. In addition, different apple varieties selected for commercial fruit juices also have different effects on the determination of phenolic compounds. Therefore, further research involving more factors is interesting.
TA B L E 2 Calibration curve, correlation coefficient of nine compounds. The compounds were detected but their distribution was not specific, and the content was not necessarily higher than in other types of beverages. Therefore, these findings cannot be used as a basis for judging that certain products contain more phenolic substances than other types of apple juice. The vast majority of consumers prefer juices that are more cloudy, higher in phenolic compounds, and considered more natural (Wodarska et al., 2016).

Compounds Regression equation
But the type of juice is not the main factor determining the content of phenolics. The loss of phenolic substances in marketed apple juice compared to the whole fruit before processing is very significant.

ACK N OWLED G M ENTS
The authors gratefully acknowledge the National Natural Science

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that there is no conflict of interest and that they do not have any possible conflicts of interest.

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 from the corresponding author upon reasonable request.