Viscotoxin and lectin content in foliage and fruit of Viscum album L. on the main host trees of Hyrcanian forests

Mistletoe (Viscum album L.) is a hemiparasitic plant that absorbs water and nutrients from the host tree. Mistletoe contains two groups of cytotoxic, immunomodulatory and antitumor proteins, viscotoxins and lectins. This study evaluated the quantity and quality of viscotoxins and total lectins in the stems with leaves (foliage) and fruit of mistletoe on Parrotia persica and Carpinus betulus in September with immature green berries and in December with mature white berries. Viscum album L. plants were harvested from host species located in the Hyrcanian forests of Iran in 2019. The highest level of viscotoxins was detected in the December foliage of V. album settled on C. betulus (9.25 mg/g dry weight [DW]), and the highest content of lectins was found in the December foliage of V. album settled on P. persica (0.79 mg/g DW) and C. betulus (0.73 mg/g DW) respectively. The immature green berries of V. album from both host species contained much higher concentrations of viscotoxins and lectins than the mature white berries. Four isoforms of viscotoxins, viscotoxin A1, A2, A3 and B could be identified in all samples of both host species. Viscotoxin A3 was the predominant viscotoxin isoform followed by viscotoxin A1.

Hyrcanian forests of Iran with a main area of about 1.9 million hectares in northern Iran are located on the northern slopes of the Alborz Mountains and the southern shores of the Caspian Sea. These forests belong to the oldest and most valuable forests of the world protected as an UNESCO world heritage site since 2019 36,37 Two deciduous trees, Persian ironwood (Parrotia persica C.A. Meyer, Hamamelidaceae) and Hornbeam (Carpinus betulus L., Betulaceae) are the most frequently colonized host species of V. album in the Hyrcanian forests of Iran [38][39][40] . Parrotia persica is an endemic species in these forests 39 .
Previous studies have shown that the quantity and quality of mistletoe bioactive compounds depend on various parameters like ecological characteristics of host species, mistletoe life cycle, and also organ type, which consequently impress the therapeutic properties of mistletoe 10,[41][42][43] . For instance, total phenolic content and antioxidant activity between leaves and stems of V. album were different and also were influenced by different host trees and seasons 10 . Viscotoxin and lectin concentrations in leaves of V. album were also impressed by seasonal changes 44 . Similarly, the effect of host trees, harvesting time, and organ type on quantity of triterpene acids of V. album has been confirmed by other studies 42,45 To our knowledge, there is no information on viscotoxins and lectins content of V. album on P. persica and C. betulus grown in Hyrcanian forests of Iran. Therefore, this study aimed to evaluate the quantity and quality of viscotoxins and total lectins in stems with leaves (foliage) and fruit of V. album on these two Iranian host species in two fruit ripening stages of mistletoe, green berries in September and white berries in December.

Results
Total amount of viscotoxins and lectins in foliage and fruit of V. album on P. persica and C. betulus in September and December. The analysis of variance indicated that the interaction of host species and mistletoe organ type had a significant effect on viscotoxin and lectin contents at the 1 and 5% probability levels, respectively ( Table 1). As shown in Fig. 1a,b, viscotoxins were the prevailing proteins in all samples studied. The highest level of viscotoxins was detected in the December foliage of V. album settled on C. betulus (9.25 mg/g dry weight [DW]), and the lowest amount was measured in the December fruit of V. album from P. persica (0.55 mg/g DW). However, the maximum content of viscotoxins in V. album from P. persica was found in the September foliage (5.51 mg/g DW), which was almost half the maximum amount of viscotoxins in V. album from C. betulus (Fig. 1a).
It is noticeable that in V. album from both host species, the immature green berries in September contained a higher concentration of viscotoxins than the mature white berries in December. The viscotoxin content in the foliage of V. album from P. persica and C. betulus was almost four-and ninefold more concentrated than in fruits, respectively (Fig. 1a).
Regarding the lectins, as depicted in Fig. 1b, the highest concentration was found in the December foliage of V. album settled on P. persica (0.79 mg/g DW) and C. betulus (0.73 mg/g DW), respectively. The lowest amount of lectins was measured in the December fruit of V. album from P. persica (0.21 mg/g DW) (Fig. 1b). It is interesting that the green fruit of V. album on P. persica in September contained about 2-and 2.5-fold higher amounts of lectins (0.55 mg/g DW) than the foliage at the same time (0.3 mg/g DW), and the fruit in December (0.21 mg/g DW), respectively (Fig. 1b). Also, in V. album from C. betulus in September the lectins content of the green fruit was a little higher (0.56 mg/g DW) in comparison to the September foliage (0.43 mg/g DW) (Fig. 1b).
Concentrations and portions of viscotoxin isoforms in foliage and fruit of V. album grown on P. persica and C. betulus in September and December. Four isoforms of viscotoxin, viscotoxin A1, A2, A3 and B, could be measured in all samples of V. album on both host species. Viscotoxin A3 was the predominant isoform in all samples followed by A1, while A2 had the lowest percentage in all conditions ( Table 2).
The maximum percentage of A3 (93%) and A1 (25.37%) was detected in the December foliage and September fruit of mistletoe from C. betulus, respectively. The maximum percentage of isoforms B and A2 was detected in the December fruit (8.83%) and September fruit (3.68%) of mistletoe from P. persica, respectively ( Table 2). Table 1. Analysis of variance of viscotoxins and lectins concentrations in the foliage and fruit of V. album settled on P. persica and C. betulus in September and December. *,**Indicate a significant difference at 5% and 1% level, respectively. www.nature.com/scientificreports/ The analysis of variance showed that the concentrations (mg/g DW) of viscotoxin isoforms A1, A2, A3 and B were affected by the interaction of host species and organ type at a 1% probability level.

Sorces of variation DF
The maximum concentration of viscotoxin A1 (0.69 mg/g DW) and A2 (0.09 mg/g DW) was detected in the September foliage of V. album L. from P. persica (Fig. 2a,b). The highest concentration of A3 (8.60 mg/g DW) and B (0.13 mg/g DW) was detected in the December and September foliage of mistletoe samples from C. betulus, respectively (Fig. 2c,d).

Discussion
Mistletoe is a hemiparasitic plant that depends on nutrient transfer from the host. Therefore host species might affect phytochemical characteristics and biological activities of mistletoe 10 . It has been shown that also harvest time and type of organ produce variations in the chemical composition of mistletoe 29 . In this study two groups of the most important bioactive compounds, viscotoxins and lectins, have been analyzed in V. album grown on P. persica and C. betulus for the first time. Total viscotoxin, its isoforms and mistletoe lectin were quantified in the foliage and fruit of V. album in the phenological stages with green berries in September and white berries in December.
Our findings showed significant differences in the quantity and quality of viscotoxins and total lectins depending on host tree, season, and organ type of mistletoe. The highest level of viscotoxins and lectins in V. album from P. persica was achieved in September and December respectively and in case of samples from C. betulus, both compounds were in the highest concentrations in December. Seasonal and host dependent variations of different bioactive compounds in mistletoe have already been found in V. album (different compounds 41     www.nature.com/scientificreports/ Our results regarding the effect of mistletoe organ type on viscotoxin and lectin accumulation showed that the mistletoe foliage contained higher amounts of these compounds in comparison with the fruit. Such a preferential accumulation of compounds in the leaves of V. album has been shown before also for other compounds. Soursouri et al. 45 reported that the highest amount of triterpene acids was detected in the foliage, and Stefanucci et al. 43 measured the highest levels of total phenolics, flavonoids, and antioxidant effect of V. album L. in leaf extract compared with fruit and seed. The highest concentrations of ML in this study extracted from the last three generations of V. album foliage in December (0.73-0.8 mg/g DW) were within the range of ML concentrations related to the first (1.8 mg/g DW) and second (0.4 mg/g DW) year of growth of V. album subsp. album leaves reported by Urech et al. 44 .
The concentrations of total viscotoxins detected here (4.20-9.25 mg/g DW in the foliage of V. album) corresponded almost to the concentrations evaluated in V. album subsp. album on different deciduous host trees by Schaller et al. 48 and Holandino et al. 49  www.nature.com/scientificreports/ The isoforms viscotoxin A1, A2, A3, and B known to be present in V. album could be assessed in all samples of V. album measured here. However, the composition of the viscotoxin isoforms in the mistletoe samples from the two Iranian deciduous host trees P. persica and C. betulus did not correspond to the composition in the European V. album subsp. album. The extremely high portion of about 87% and 93% viscotoxin A3 and the low concentrations of viscotoxin A2 comply with the viscotoxin pattern of V. album subsp. abietis growing exclusively on fir (Abies ssp.) 44 . The European V. album subsp. album growing on deciduous trees contains rather well-balanced portions of the four viscotoxin isoforms 44 so that viscotoxin A2 and A3 are present in almost equivalent portions in this subspecies. Previous studies on V. album subsp. album from different host species have shown that the maximum percentage of each viscotoxin isoform in this subspecies was A1: 16%, A2: 35%, A3: 41.2% and B: 14% 44 meanwhile in V. album subsp. abietis these percentages changed to A1: 6.5%, A2: not detected, A3: 74.5% and B: 5.4% 44 being close to the portions measured in our samples of C. betulus and P. persica. This correspondence leads us to the hypothesis that the Iranian V. album on the two host species might represent a specific genetic variant of V. album. Impact of ecological factors on the composition of these small proteins is very unlikely in view of the constant viscotoxin patterns detected in V. album subsp. album on 7 different host species representing a big range of ecological conditions 29,48 .
Zuber and Widmer 50 concluded in their study on genetic and geographic differentiation of V. album, that highly differentiated populations and possibly new taxa exist at the range limit of this species. Iranian mistletoe belongs to one of the easternmost populations of V. album.
Our study detected significant variations in quantity and quality of viscotoxins and total lectins of V. album under the effect of the host species and mistletoe organ type at two stages of fruit development in September with green berries and in December with white berries. It has to be expected that the pharmacological activities of mistletoe extracts such as cytotoxicity and induction of apoptosis in cancer cells vary accordingly. It has been shown that the four isoforms, viscotoxin A1, A2, A3, and B, have cytotoxic effects on cancer cell lines. Viscotoxin A3 proved to be the isoform with the highest specific activity 49,51,52 . Therefore it is expected that the Iranian V. album on the two host species measured with high concentrations of viscotoxin A3 exert high viscotoxin dependent cytotoxic effects exceeding that of the European V. album subsp. album. It will be crucial for the pharmaceutical use of V. album to take into consideration the results presented here.

Materials and methods
Plant material. The collection of plant samples in this study was done according to legislation and formal permission of Iran Natural Resources and Watershed Management organization. Viscum album L. plants were harvested in 2019 from two host species, P. persica C.A. Meyer and C. betulus L., located in the kelerd forest of the Hyrcanian forests in Mazandaran province of Iran (Table 3) in September (mistletoe green berry stage) and December (mistletoe white berry stage). The altitude of this region is between 700 and 1100 m above sea level, and the average annual rainfall is 668 mm.
For sampling, three individual trees of each host species which were similar in diameter, height and morphology were selected and three mistletoe plants were marked on each of these trees in order to perform harvesting in September and December on the same mistletoe plants. In both harvesting times five branches of each mistletoe plant including the last three generations of organs (foliage and fruit) were collected (Fig. 3), placed in liquid nitrogen for a few seconds and finally stored at -80 °C until extraction. In the following the designation "foliage" is used for the mistletoe branch including leaves and stems without fruit.
Preparation of extracts. The mistletoe foliage and fruit were separately freeze-dried for 72 h. Then the dried samples of foliage or fruits were powdered in liquid nitrogen using mortar and pestle.
Viscotoxins extraction. For this extraction, 200 mg of the plant powder was mixed with 5 ml of 2% acetic acid and left overnight at 4 °C in 3 replicates. The resulting mixture was then homogenized by an Ultra-turrax homogenizer. The samples were centrifuged at 4000 × g for 30 min and the supernatant was separated. The extraction process was repeated by adding new solvent to the pellet. The two supernatants were pooled and the volume adjusted to 10 ml. These viscotoxin extracts were stored at 4 °C until HPLC analysis 44 .
Lectins extraction. Regarding lectins extraction, 200 mg of the plant powder was suspended in 5 ml of phosphate buffered saline (PBS) and left overnight at 4 °C in 3 replicates. After homogenizing by an Ultra-turrax homogenizer the mixture was centrifuged at 4000 × g for 30 min, and the supernatant was separated. This procedure was repeated with the pellet and 5 ml PBS and once again by adding 5 ml of PBS containing 0.2 M galactose. The total volume of the combined extracts was adjusted to 15 ml and stored at − 20 °C until chemical evaluation 44 . Statistical analysis. The statistical analysis was performed using SAS 9 software. The residual normality of data was tested using the Shapiro-Wilk test. All the experiments were analyzed by two-way ANOVA (analysis of variance), and significant differences between means were compared by the least significant difference test (LSD) at the 0.05 probability level.

Conclusion
In this study, the quantity and quality of viscotoxins and also total lectins of the foliage and fruit of V. album L. plants growing on P. persica and C. betulus in Hyrcanian forests of Iran were evaluated in September in the phenological state of green berries and in December with white berries for the first time. Our findings indicate significant variations in the concentrations of two groups of pharmacologically important antitumor compounds, viscotoxins and lectins, depending on host species, mistletoe organ type, and harvest time. On the whole, viscotoxins were the prevailing proteins in all studied samples. Among the two mistletoe organs tested, the foliage extract contained the highest amount of both compounds. The foliage of V. album on C. betulus in December had the highest level of viscotoxins. The maximum amount of lectins was found in the December foliage of V. www.nature.com/scientificreports/ album settled on P. persica and C. betulus. In the mistletoes from both host species, the green berries in September showed higher contents of viscotoxins and lectins than the white berries in December. Regarding viscotoxin composition, 4 isoforms of viscotoxin, A1, A2, A3, and B, were detected in all samples of both host species. A3 was by far the most predominant viscotoxin isoform followed by A1 in all samples, whereas A2 represented the lowest concentration. This extraordinary pattern of viscotoxins in the Iranian V. album on the two deciduous trees contrasts with the patterns known for V. album subsp. album growing on deciduous trees in Europe. This points to possible new genetic variants of V. album in Iran which can have more potent cytotoxic activity than European V. album subsp. album.

Data availability
Data sharing is not applicable to this article as all new created data is already contained within this article.