Volatile components and nutritional qualities of Viscum articulatum Burm.f. parasitic on ancient tea trees

Abstract Volatile flavor compounds (VFCs) and nutrients in Viscum articulatum Burm.f. parasitic on ancient tea trees (named TM) were studied in this research by headspace solid‐phase microextraction (HS‐SPME)/gas chromatography–mass spectrometry (GC–MS) and conventional methods. Sixty‐six volatile compounds belonging to different classes were identified by GC–MS. The ketones, alcohols, and aldehydes were the principal aroma groups in TM according to principle component analysis (PCA). The most abundant aroma components in TM included benzaldehyde (9.64%), geranylacetone (7.92%), epoxy‐β‐ionone (7.71%), β‐linalool (7.35%), methyl salicylate (6.96%), and hotrienol (6.14%), significantly higher than CKs (p < .05). The positive PC1 and PC2 in TM were correlated with benzaldehyde, hotrienol, methyl salicylate, and geranylacetone. The mistletoes could be differentiated from CKs due to the difference in aroma compounds. Clean and fresh, woody and nutty odor with minor floral scent was the characteristics of TM. Analysis of the nutritional components showed that contents of polyphenols and catechins in TM were at trace levels, significantly lower than CKs (p < .05). The total contents of polyphenols, amino acids, carbohydrates, and caffeine in TM were significantly lower from the total soluble solids (p < .05), indicating that there were still lots of compounds undetected in TM. The sensory test showed that the taste and aroma in TM can be accepted, which indicates TM could be developed into alternative tea drinks in the future.

genera in the Loranthaceae (Zhao et al., 2011), which contains special biological components, such as flavanols, flavonoids, alkaloids, organic acids, and other high molecular weight compounds (Ni, Li, Zhao, Shao, & Li, 2013). Experiments have been conducted to attest their potential health activities, such as antibacterial, antimutagenic, anti-arrhythmia effects, immunomodulating effects, eliminating rheumatalgia, curing knee-waist malaise and hepatitis (Bachhav, Bhutada, Patil, Baser, & Chaudhari, 2012;Lee, Kim, Kim, Park, & Lee, 2009;Yang, Xiang, Feng, & Liu, 2009). Therefore, TM becomes a popular drink among local residents, which has been developed into commercial products for sale by tea companies. When soaked, TM releases a special fresh aroma, which is significantly different from teas. However, there is little information in the literature about the aroma characteristics and nutrients of TM. It is also worth studying whether TM has similar nutritional ingredients to teas because of their parasitism.
The various aromas of teas such as fruity-sweet, floral-sweet, and fresh grass present an attractive odor organoleptic quality.
In this study, HS-SPME/GC-MS methods were used to analyze the aroma of TM, oak mistletoe (OM), and the control samples. The nutritional qualities including the polyphenols, catechins, caffeine, and soluble carbohydrates were analyzed by HPLC and spectrophotometry methods. The objectives of our research were to reveal the aroma profile of TM as well as its nutritional compounds in order to provide scientific foundation for its deeper development.

| Materials
One TM sample, one mistletoe hosted in oak trees (OM), and two control tea samples (CK1 and CK2) were provided by Tea Research Institute, Guangdong Academy of Agricultural Sciences in May, 2016. The TM sample was collected from old, former tea trees no longer in production. The control tea samples were collected from local tea garden. All the samples were dried, ground, sieved with a 40 mm (aperture) sieve, and stored in airtight glass containers at 4°C for further use. The details of samples were shown in Table 1.

| HS-SPME procedure
The HS-SPME method was used to extract the VFCs from experimental samples. The HS-SPME holder was fiber-coated with Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS) at a thickness of 50/30 μm (Supelco Co.). As indicated by the instruction manual, the fiber was activated for 5 min in the GC injection port to eliminate impurities before GC analysis.
The VFCs were extracted according to the method presented by Lv et al. (2012) with slight modification. Of 10.0 g of tea powder was weighed and placed in a 250 ml conical flask and immediately infused with 100 ml of boiling water. A magnetic heating stirrer was F I G U R E 1 Fresh mistletoe hosted in ancient tea trees used to establish equilibrium faster at the temperature of 50°C. The SPME fiber was exposed in the headspace of the conical flask for 80 min and then introduced into the GC injection port to allow desorption for 3.5 min.

| GC-MS analysis
The extracted VFCs from samples were analyzed on a GC-MS (TRACE DSQ, Thermo Finnigan). GC conditions were as follows: injector temperature, 230°C; injector mode, splitless mode; column, HP-5MS, 30 m, 0.25 mm i.d., 0.25 µm film thickness. The GC oven temperature was programmed to hold at 50°C for 1 min, then increase to 180°C at a rate of 2°C/min for 1 min; and finally to 230°C at a rate of 10°C/min, which was maintained for another 1 min.
Helium (percentage purity > 99.999%) was used as the carrier gas at a constant flow velocity of 1 ml/min. MS spectrometer conditions were as follows: interface temperature, 280°C; ion source temperature, 220°C; ionization mode, EI; electron energy, 70 eV; mass scan range, 40-400 AMU. The National Institute of Standards and Technology (NIST) library (98 L), MS, retention time compared with those reported in the literature were used to identify aroma compounds.
The relative proportions of the constituents were obtained by FID peak area normalization and were quantified by the following equation: Relative content % = single constituent area/total area × 100% The internal standard was not added in this research. Because the internal standard changed the equilibrium system of tea soup and would also be absorbed by the teas, it was feasible to use the method of relative peak area to analyze and compare the aroma components of the different samples in the same experimental conditions.

| The total soluble solids
The total soluble solids were determined according to the method of Yin, Xu, Yuan, Luo, and Qian (2009), with minor modifications.
The ground samples (1.0 g) were extracted in 100 ml of boiled distilled water for 45 min. The extracted infusion (50 ml) was placed in a weighed evaporation dish to evaporate to dryness on a heated water bath. Then, the evaporation dish was dried at 120°C for 2 hr until the weight of the dishes with the residues reached to a constant content. The contents of soluble solids were calculated according to the weight difference of the evaporation dishes before and after evaporation.

| Analysis of tea polyphenols
Tea polyphenols were determined with spectrophotometric method using FeSO 4 , potassium sodium tartrate and phosphate of pH7.5 as a buffer (Xu et al., 2018). Of 1.0 g samples was extracted with 100 ml of where (A1 − A2) was the difference between two absorbance, m was the dry weight of samples, g, V was the total volume of infusion, ml.

| Analysis of free amino acids
The content of amino acids was determined according to the ninhydrin dyeing method at 570 nm using glutamic acid as the standard

Sample name Details Tenderness
Tea mistletoe (TM) Hosted in ancient tea trees One stem with one to two branches Oak mistletoe (OM) Hosted in ancient oak trees One stem with one to two branches Steamed green tea (CK1) Fresh tea leaf from local tea trees, dried One bud with two leaves Pureh raw tea (CK2) Fresh tea leaf from local tea garden, made into Pu-erh raw tea One bud with two leaves TA B L E 1 Sample details in this research for the amino acids (Yao, Guo, Lu, & Jiang, 2005). Of 1.0 ml of tea infusion with 0.5 ml of ninhydrin solution (2%) and 0.5 ml of phosphate buffer (pH = 8.0) was heated in water bath for 15 min before cooling and fixing to 25 ml with water. The absorbance (A) was determined at 570 nm with a spectrophotometer. The amino acids content (%) = (C × V 1 )/(10 × V 2 × M × W), where C was the quality of glutamic acid calculated from the standard curve based on the absorbance value (A), mg; V 1 was the total volume of the infusion, ml; V 2 was the volume of test solution, ml; M was sample mass, g; W was the dry matter content, %. M × m was the dry weight of the sample, g.

| Statistical analysis
Significance test was carried out using SPSS software (version 16.0; SPSS, Inc.), where one-way analysis of variance (ANOVA) and the Tukey test were used (a p-value of .05 was considered significant).
PCA was carried out using Excel 2019 supported by XLSTAT software to determine the relationships between samples and volatile aroma compounds. All the experiments were repeated three times under identical conditions. Data were expressed as the average of three replications.
Aroma classes and contents were shown in Figure 2. It was showed that ketones were the most abundant aroma class in TM, and was significantly different from other aroma classes (p < .05).
The alcohols in TM were significantly lower than CKs, in which the alcohols were the most abundant aroma classes, accounting for more than 60% of total aroma. However, the ketones, esters, and aldehydes content in the CKs were significantly lower than TM (p < .05). Figure 3 showed the PCA results of the aroma classes in different samples. The first two principal components (PCs) accounted for 97.88% of the variation in the data. The PC1 dimension, which explained 77.09% of the variance, showed that samples were mainly characterized by ketones, alcohols, alkenes, and phenolic compounds, etc. The PC2 dimension explained an additional 20.79% of the variance, characterized by ketones, alcohols, and aldehydes.
The positive PC1 and negative PC2 dimension was largely correlated with ketones and aldehydes, which were main aroma classes in mistletoes. The positive PC1 and PC2 were correlated with the alcohols, the dominant aroma class in CKs. It could be deduced that the mistleotes were differentiated from CKs due to their different aroma classes and material bases. Basically, the result of PCA agreed with that of Figure 3.
Epoxy-β-ionone, a degradation product of β-carotene (Ravichandran, 2002), contributed to the floral and sweet scents in TM. β-Ionone had violet and woody odors and a very low odor threshold for human detection (Schuh & Schieberle, 2006) and was usually found in black tea with high amounts (Rawat et al., 2007), and was an important contribution to tea aroma.  . β-Linalool was one of the terpene alcohols in teas aroma, endowing wood, sweet, and floral notes to tea (Rawat et al., 2007). The high concentration and low threshold level made it a great contribution to the total tea flavor (Ravichandran, 2004). Our results were consistent with the previous studies in that β-linalool was the principal alcohol in green tea (Kato & Shibamoto, 2001;Tontul et al., 2013) and black tea . The phenethyl alcohol content in OM (13.08%) was significantly greater than other samples, which imparted sweet and floral aromas. Hotrienol, with hyacinthine odor, rarely reported in tea aroma, was detected in all samples with high content. Aldehydes were detected from mistletoes and included benzaldehyde, (E)-2-decenal Z-2-decenal, β-cyclocitral, safranal, benzyl acetaldehyde, etc. Aldehydes content in mistletoes was significantly higher than CKs (p < .05). Benzaldehyde was an important active odor in almonds and nuts, contributing significantly to the almond odor (Lasekan & Abbas, 2010;Xiao et al., 2014). Benzaldehyde was rarely found in as high of an amount in teas as in mistletoes, which was a key difference from teas. Aldehydes such as 2-heptenal and pentanal mainly came from oxidative degradation of amino acids and fatty acids (Rizzo, 2014), and they were important VFCs due to their low odor threshold (Selli & Cayhan, 2009).
γ-Nonanolide was detected as one of the esters only in the mistletoes and was rarely found in teas, which imparted strong coconut flavor with 0.0097 mg/m 3 aroma threshold in water (Van Gemert, 2011). Our results suggested that γ-nonanolide was one of the major differences between TM and CKs in aroma compounds. Methyl jasmonate, releasing jasmine-like odor and often found in teas (Mu et al., 2018), was also identified with a low aroma intensity in this study.
Five kinds of hydrocarbons were identified with GC-MS, constituting 6.58% and 4.65% of the total aroma in TM and OM, respectively. Researches had found that few odors came from hydrocarbons (Du et al., 2014;Lv et al., 2012) and thus they had little effect on tea aroma. Alkenes, such as sabinene, β-caryophyllene, β-cedrene, δ-cadinene, and L-calamenene, were found to have an important contribution to the woody or herbal flavor in tea (Du et al., 2014;Lv et al., 2012). However, only β-Cedrene was detected in mistletoes in a low content.
Two kinds of acids, hexadecanoic acid and octanoic acid, were detected in this study. The result was different from the findings of Du et al., (2014), who did not detect hexadecanoic acid in Pu-erh tea by SPME and SDE methods. This discrepancy might be attributed to different experimental conditions and samples.
Two phenolic compounds, 2, 6-di-tert-butyl-4-methyl phenol and 4-vinylphenol, did not account for a significant proportion in the total VFCs in this study. Indole was detected in this study, which was an important compound in the formation of aroma in oolong teas (Zeng et al., 2016). The elemicin was only found in OM, providing spice and flower odors.

| Principal components analysis (PCA)
Before PCA, data in Table 3 were processed as follows. The hydrocarbons and aromas whose content was <1.0% were removed from the list; the remaining aroma substances were subjected to PCA ( Figure 4) For the negative PC2 dimension, it was characterized by geraniol, E-nerolidol, decanal and cis-hexanoic acid, 3-hexenyl ester. Figure 4 showed that TM could be differentiated from CKs due to their aroma compounds.
The major aroma components were significantly different in contents between CKs and TM ( Figure 5). It was evident that benzaldehyde, β-linalool, methyl salicylate, geranylaceone, and epoxy-β-ionone were important aroma components in TM. In OM, benzaldehyde, phenethyl alcohol, and β-linalool were principal aroma compounds. β-Linalool and geraniol were most abundant aroma compounds in CK1 and CK2 because of their large leaf origins.

| Aggregation of aroma compounds with different odor notes
According to the odor of different components (Table 2), compounds with flower, fruit, mint, wood, nut, grass, and fat odor were summarized in Figure 6. It was evident that the CKs contained more floral odor compounds than TM and OM. The nut flavor compounds in TM and OM were significantly higher than that in the CKs, mainly due to the benzaldehyde. It was worth noting that content of different aroma compounds with different odor note only represented the quantity in samples but not the real aroma smelt by human noses.
The threshold and content of each aroma compound both decided the aroma characteristics.
It was known that SPME/GC-MS method had drawbacks in identification of odor intensity, and the main contributors of fla-

| The total soluble solids
To investigate the differences in nutritional qualities between mistletoes and teas, the quality index was determined in this research.
The contents of nutritional components were found different between mistletoes and tea samples ( Table 3). The total soluble solids were 35.12% in TM, significant lower than the CKs. It was because that TM did not experience rolling process like tea leaves, and its tissues and cells were not broken, which led to less releasing of the soluble solids. The total soluble solids represented the thickness of the infusion, and the higher soluble solids would lead to the mellow and smooth mouth taste. In sensory evaluation, the taste of TM was not as mellow and smooth as CK2 (Table 4).
The total contents of polyphenols, amino acids, carbohydrates, and caffeine in TM were significantly lower than the total soluble solids, which meant that there were still lots of components undetected and needed to be analyzed further. Previous researches showed that mistletoes contained more functional flavonoids, sesquiterpene lactones, lectins, steroids, organic acids (small molecules), peptides, proteins, polysaccharides (polymers), and alkaloids (Harikrishnana, Balasundaramb, & Heo, 2011;Ni et al., 2013;Samuelsson, 1958). Therefore, mistletoes could be used as

| Tea polyphenols, catechins, and caffeine
Polyphenols are important factors in tea qualities (Wu et al., 2019;Xu et al., 2018), which contribute astringency to the tea taste. The catechins are the family of polyphenols, mainly include EC, EGC, ECG, EGCG, C, GC, CG, and GCG and have been used as a large group of natural antioxidants in many aspects (Singh, Shankar, & Srivastava, 2011;Wang, Dong, Men, Tong, & Zhou, 2013). Tea catechins can be categorized into two groups: nonesterified and esterified catechins (Theppakorn, 2016). Nonesterified catechins include C, EC, GC, and EGC; whereas the esterified catechins contain EGCG, ECG, GCG, and CG (El-Shahawi, Hamza, BahaffiA, Al-Sibaai, & Abduljabbar, 2012;Theppakorn, 2016). Table 3 showed that the contents of polyphenols and catechins in TM were at trace levels, significantly different from CKs (p < .05). The CKs were green teas in nature, and therefore, they were rich in polyphenols and catechins (Wei et al., 2011). Nonesterified catechins were significantly different from the esterified catechins in TM. However, there was an opposite tendency in CKs. EGCG was the principal catechin with a high content in CKs, which was in agreement with the reports of Khan and Mukhtar (2007) and Zuo, Chen, and Deng (2002). The results also showed that the content of caffeine in TM   (Harikrishnana et al., 2011), which agreed with our results.

| Amino acids and soluble carbohydrates
Our results showed that the nutritional compositions in mistletoes were quite different from tea samples, indicating that the mistletoes did not absorb large quantities of chemicals from tea trees as a source of nutrition despite they were parasitic on tea trees.
Therefore, the TM should have its own material circulation and metabolism, which needs to be further studied in the future.
The theaflavins, the index of redness in black tea soup, were not determined in this research because they were not important quality parameters in mistletoe, green teas, and Pu-erh teas (Okinda Owuor et al., 2006).

| Sensory evaluation
Sensory evaluation was carried out by five experienced tea experts from our research institute. The mistletoes were evaluated according to sensory test method of green teas because of their unfermented nature. Sensory scores of the mistletoes and CKs ranged from 84.80 to 90.40, according to the outer shape, the color, aroma, and taste of the infusion, the residues (Table 4). Fresh, lasting, woody, and nutty aroma was the characteristic of TM. TM had its own sensory attributes, significantly different from teas. Despite the lower scores, the panelists believed that the two mistletoes were acceptable both in aroma and taste, which could be further developed into alternative tea drinks.

| CON CLUS IONS
Sixty-six volatiles were identified in research samples, including alcohols, aldehydes, ketones, esters, hydrocarbons, acids, phenolic compounds, and alkenes. According to PCA, ketones, alcohols, and aldehydes were the principal aroma classes in TM. Benzaldehyde (9.64%), epoxy-β-ionone (7.71%), geranylacetone (7.92%), β-linalool (7.35%), and methyl salicylate (6.96%) were the most abundant and significant components in TM, which contributed to a fresh, lasting woody, and nutty aroma with minor floral scent. The mistletoes could be differentiated from CKs due to the difference of the aroma compounds according to PCA.
Analysis of nutritional components showed that a large portion of compositions in teas such as the polyphenols, catechins, caffeine, free amino acids were relatively low in TM, indicating there were still ingredients undetected in the water extract. This research also testified that the TM did not absorb large quantities of chemicals from tea tree as a source of nutrition despite it being parasitic on tea trees. Sensory test showed that the TM and OM had acceptable aroma and taste, which could be developed into alternative tea TA B L E 4 Sensory quality and scores of the mistletoes and CKs drinks in the future. This study sheds light on the volatile and nutritional qualities in the mistletoes and will provide a scientific foundation for their further utilization. Research Institute, and Chinese Academy of Agricultural Sciences are also acknowledged.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing. Written informed consent was obtained from all study participants.