Potential toxicity of Schisandra chinensis to water environment: acute toxicity tests with water crustaceans

Fruits of Schisandra chinensis, an East Asian liana plant, are currently more and more used to produce nutrient supplements that positively affect human health due to the content of various secondary metabolites. On the other hand, these substances because of their bioactivity can cause possible allelopathic or toxic effects concerning other organisms (algae, plants, animals). But the ecotoxicological properties of S. chinensis outside its area of origin have yet to be sufficiently verified. Two crustaceans, Daphnia magna and Thamnocephalus platyurus, were selected as model aquatic organisms to test the potential impact of S. chinensis active compounds on the aquatic environment. Crude water extract from S. chinensis fruits, simulating the natural leakage of active substances in water, was tested in treatments from 0.0045 to 45 mg/L (according to the content of schisandrin as the dominating lignan). Effective concentration (EC50) causing 50% lethal effect for D. magna was established to 0.0448 mg/L after 24 h and 0.0152 mg/L after 48 h. EC50 for T. platyurus reached 0.4572 mg/L after 24 h, i.e. more than ten times higher than for D. magna. This study showed that the potential environmentally relevant concentrations of S. chinensis bioactive compounds could represent a severe risk to aquatic ecosystems.


Introduction
Schisandra chinensis (known as magnolia vine or wuweizi) is a liana used as a natural herbal medicine to treat many human diseases.This plant grows wild and is cultivated in the Far East, like northeastern China, Japan and Korea (Panossian and Wikman 2008).A work by Li Shih-Chen, "Pên T'shao Kang Mu", on Chinese medicine published in 1596, contains the first written mention of the Schisandra species (Szopa et al. 2017).However, the seeds of Schisandra plants were also discovered in Europe, as confirmed (Teodoridis 2005) in a morphological and anatomical study of the holotype of S. moravica from Šafov and additionally seeds of the same species from the Cheb, Sokolov and Most Basins (Czech Republic) where the assignment of the seeds to the genus Schisandra MICH was confirmed.
As confirmed by many studies, S. chinensis is a promising plant with a positive effect on human health and is classified as an adaptogen.Hancke et al. (1999) describe the proven positive impact of S. chinensis on human health and focus on anti-hepatotoxic, antioxidant and anti-tumour effects and on improving physical performance and effects on the central nervous system.Ma et al. (2023) have shown in the study that S. chinensis has promising clinical efficacy in treating diabetes mellitus as one of the most common diseases in our current civilisation.S. chinensis and its compounds affect the endocrine system, e.g.cortisol and testosterone levels, and the metabolism of lipids and muscles (Leis et al. 2020).Another advantage of S. chinensis and its derivatives is that it can be a source of natural antioxidants (Wang and Wang 2013).
Plants from Schisandraceae family contain many pharmacologically usable substances including schisandrin, deoxyschisandrin, gomisin A, gomisin O and gammaschisandrin (Smejkal et al. 2010;Wang et al. 2018).Most of the bioactive compounds can be found in fruits and seeds, lower contents are in leaves and stem (Lee et al. 2022).The real content of schisandrin as the dominant compound and other lignans found in the fruits of magnolia vine is highly variable.It depends on many factors, such as the method and place of cultivation or the fruit ripening stage.According to the literature sources, the content of schisandrin can be from 2.1 mg/g (Kohda et al. 2012) to 5.0 mg/g of dry fruit weight (Slanina et al. 1997).
Growing demands for the production of food supplements containing bioactive substances from this plant promote its farming in larger areas outside its place of origin.It evokes a question if S. chinensis can influence the environment in such sites.Schisandrin and other active compounds from S. chinensis belong to lignans, secondary plant metabolites known as allelochemicals (Costas-Gil et al. 2018;Scavo et al. 2019a;Scavo et al. 2019b) which can be toxic not only to plants but even to other organisms in the ecosystem.Currently, the data about the allelopathic effect of schisandrin are scarce.
S. chinensis grows wild in East Asia, but in some areas, it is farmed.For example, the average size of the farmland for S. chinensis cultivation in South Korea was 0.5 ha per farm (Choi et al. 2015) and the scale of cultivation areas is still rising.In European area, S. chinensis is often planted only as an individual plant nearby garden ponds but because of the increasing inquiry on the fruits, it is considered its farming on larger areas.However, any study does not address the possible impact of S. chinensis farming on the surrounding environment.
Delicate water ecosystems can be affected by the effects of lignans leaking from fruits accidentally falling into the water body as proved on Lemna minor (Valickova et al. 2023).The effects of bioactive substances from Schisandra chinensis on aquatic animals are insufficiently researched and this lack of ecotoxicological data should be addressed.
An integral part of aquatic ecosystems and food webs is also invertebrate zooplankton species like Daphnia magna and Thamnocephalus platyurus.Because of their high sensitivity to environmental pollution, these species are included in standardised acute toxicity tests (Kerberova et al. 2022;Szklarek et al. 2022).
This study targets to assess the possible acute toxicity of aqueous extract from S. chinensis fruits to indicator zooplankton species Daphnia magna and Thamnocephalus platyurus to enhance the knowledge about the possible ecotoxicological risks of S. chinensis farming outside its areas of origin.

Plant material and extraction
The ripe fruits of S. chinensis were harvested from a plant grown in the area of Vracov, Czech Republic.The whole fruits (including seeds) were freed from impurities, detached from the stalks and left to dry naturally in the air.The average portion of their dry mass was established at 20% of the fresh weight.

Acute toxicity tests
Daphtoxkit F® and Thamnotoxkit F® (Microbiotests, Belgium) were used to test the acute toxicity of SCE to freshwater crustaceans Daphnia magna and Thamnocephalus platyurus.According to producer's guides, resting stages of both organisms (D. magna ephippia and T. platyurus cysts included in the kits) were re-activated in a standard freshwater and cultivated on Petri dishes under continuous light for obtaining a population of adult animals (D. magna for 72 h, T. platyurus for 20-22 h).
The SCE was diluted with standard freshwater to the concentrations 0.0045 mg/L, 0.009 mg/L, 0.045 mg/L, 0.09 mg/L, 0.45 mg/L, 0.9 mg/L, 4.5 mg/L and 9 mg/L.Undiluted SCE (45 mg/L) was included in both tests too.Tested solutions and animals were transferred to test plates (4 wells per each treatment with 5 individuals per well for D. magna; 3 wells with 10 individuals per well for T. platyurus) and the cultivation continued in the darkness and controlled temperature (D. magna 20 ± 2 °C, T. platyurus 25 °C).According to producer's guides and ISO standards (ISO 6341: 2012;ISO 14380: 2011), the mobility of D. magna was evaluated after 24 and 48 h and the mortality of T. platyurus was assessed after 24 h.The tests were run in triplicates.

Statistics
The effect of SCE on both species was evaluated as the percentage of immobilised/dead individuals in each treatment according to the following equation: where N L is the number of living individuals and N T is the number of total individuals in each treatment.
The software STATISTICA (StatSoft Inc.®) was used to evaluate the obtained results statistically.The significance of the differences in the average values between the treatments was assessed by the non-parametric Kruskal-Wallis test and multiple comparison method (P < 0.05).The effective concentrations EC 50 /LC 50 for both species were evaluated using GraphPad Prism® software (Dotmatics) by a dose-response curve and non-linear regression analysis.

Acute toxicity to Daphnia magna
In accordance with the standards ISO 6341: 2012 and OECD Guidelines Test No. 202, daphnids were exposed to SCE in an acute toxicity test.The number of dead and immobilised individuals was assessed after 24 and 48 h (Fig. 1).The lowest tested treatment (0.0045 mg/L SCE) did not influence the vitality of daphnids either after 24 h or 48 h as compared to the untreated control.The number of surviving individuals decreased in 0.009 mg/L SCE treatment 24 h by 25% and after 48 h by 50%.Higher treatments of 0.045 and 0.09 mg/L SCE caused more than 50% mortality on both days.From 0.45 mg/L SCE, all treatments exhibited 100% mortality.Undiluted SCE 45 mg/L caused even the destruction of the daphnid's bodies.
The effective concentration (EC 50 ) for Daphnia magna was established at 0.0448 mg/L after 24 h and 0.0152 mg/L after 48 h (Table 1), showing high acute toxicity of SCE to this invertebrate.

Acute toxicity to Thamnocephalus platyurus
Following the ISO standard 14380: 2011, T. platyurus larvae were exposed to SCE in an acute 24 h toxicity test (Fig. 2).The lowest tested treatments (0.0045 and 0.009 mg/L SCE) did not influence the vitality of T. platyurus larvae after 24 h as compared to untreated control.Treatments 0.045, 0.09 and 0.45 mg/L SCE caused a decrease of surviving individuals' number by 3, 10 and 36%, respectively.In higher treatments (0.9 to 45 mg/L SCE), the mortality reached 100%.
The effective concentration (EC 50 ) for T. platyurus was established at 0.4572 mg/L after 24 h (Table 1).

Discussion
Biologically active substances contained in plants, fruits or seeds represent a wide range of chemicals which can be helpful for humans but toxic for other organisms in the environment.One group of these secondary metabolites is lignans, compounds with allelopathic potential.Schisandrin and other lignans in S. chinensis fruits are highly beneficial for human health.Still, they can represent a severe risk to other organisms, as already proved for the aquatic plant Lemna minor et al. 2023).Currently, the ecotoxicological data on the effects of plant lignans in a water environment on other organisms, including zooplankton, are scarce, and schisandrin still needs to be evaluated.But as Mayorga et al. (2010) and others stated, using complex whole plant extracts in ecotoxicological studies can be more suitable and significant than studying fractions because the individual constituents can lose their specific bioactivity when isolated from the crude extract.
Crustaceans represent an essential part of the zooplankton.D. magna and T. platyurus belong to the common species known for their sensitivity to pollution in the aquatic environment and are, therefore, often used as model organisms (Alvarenga et al. 2016;Szklarek et al. 2022).Therefore, this paper aimed to simulate the conditions when the extraction of compounds contained in S. chinensis fruits occurs accidentally in a water body and zooplankton species get in contact with it.
Both tested species exhibited considerable sensitivity when exposed to S. chinensis fruit extract.As shown in Table 1, after 24 h, the EC 50 value for D. magna (0.0448 mg/L) was ten times lower than for T. platyurus (0.4572 mg/L), and after 48 h, the standard test period for D. magna, the EC 50 value decreased even to 0.0152 mg/L.The exact mechanism of SCE toxicity to the crustaceans is currently unknown.
According to the literature sources (Slanina et al. 1997;Kohda et al. 2012), the content of schisandrin in S. chinensis fruits can reach 0.2 to 0.5% of their dry weight, i.e. one S. chinensis berry (average dry weight approx.30 mg) can contain 0.06 to 0.15 mg of schisandrin.Hypothetically, when considering this schisandrin content, for reaching the schisandrin concentration in a water body close to the EC 50 value for D. magna, only dozens of Schisandra berries could be sufficient to contaminate hundreds of litres of water.
The susceptibility of these species to various chemicals, including natural compounds, is different.In this case, T. platyurus seems to be less sensitive to SCE as compared to D. magna, similar to the results published on other organic pollution, like olive oil mill wastewater (Paixao et al. 1999), various organic matters like compost, digestate or sludge eluents (Alvarenga et al. 2016) or urban wastewater contaminated by artificial sweeteners (Kerberova et al. 2022).Opposingly, in the case of inorganic de-icing salts, D. magna was less sensitive than T. platyurus (Szklarek et al. 2022).Some examples of their differences in sensitivity to other contaminants are given in Table 2.
The effective concentration (EC 50 ) of Trapa japonica leaves extract to D. magna was from 4 to 22 g wet mass per litre (Ishimota et al. 2019), and in contrast to this, EC 50 for SCE to D. magna is 0.0152 mg/L (Table 1) corresponding to 15 to 37.5 mg fresh weigh per litre of S. chinensis berries.Similarly, the extract from Hedychium coronarium rhizomes exhibited a 50% mortality rate to Daphnia similis in concentration approx.1.3 g fresh weight per litre (Costa et al. 2021).Both these results show that S. chinensis fruit extract is more toxic for daphnids than extracts from other plants.
In the case of T. platyurus, Mayorga et al. ( 2010) stated higher sensitivity of this species as compared to another crustacean Artemia salina when exposed to an extract of different Guatemalan plants.The lethal concentration (LC 50 ) range was from 10 to 500 mg/L.Cangiano et al. (2002) described the toxic effect of diterpenes from Ruppia maritima and Potamogeton natans on green algae and a few zooplankton species, including D. magna and T. platyurus.In this case, T. platyurus was more sensitive than D. magna, and the LC 50 value for the most toxic compound reached 0.84 μM, a lower value than in SCE 0.4572 mg/L (corresponding to 1 μM schisandrin).Ji et al. (2014) described that only the minority of S. chinensis bioactive compounds (unmetabolised forms) were It can be hypothesised that the waste water contaminated by schisandrin used by humans as food supplements would be a negligible source of contamination, and the primary source of contamination would be S. chinensis farming in large due to fruit or other plant parts fallout.

Conclusion
Although S. chinensis provides many beneficial effects for human health, it can harm aquatic ecosystems.Crustaceans D. magna and T. platyurus exhibited sensitivity to waterextractable bioactive compounds from S. chinensis fruits already in potentially environmentally relevant concentration.
Reaction of other zooplankton species to SCE is unpredictable; nevertheless, the accidental contamination of water by S. chinensis bioactive compounds can disturb fragile food webs.

Fig. 1
Fig. 1 Acute toxicity test of SCE on Daphnia magna after 24 and 48 h.Data points represent the mean over twelve replicates, and standard deviations are indicated by error bars.Different letters mark significant differences between treatments within the time point (Kruskal-Wallis test, P < 0.05)

Fig. 2
Fig. 2 Acute toxicity test of SCE on Thamnocephalus platyurus after 24 h.Data points represent the mean over nine replicates, and standard deviations are indicated by error bars.Different letters mark significant differences between treatments (Kruskal-Wallis test, P < 0.05)

Table 1
Evaluation of SCE acute toxicity in tests with freshwater crustaceans

Table 2
Kim et al. (2014)C 50 / LC 50 ) of selected contaminants to freshwater crustaceans D. magna and T. platyurus.EC 50 values are expressed as means excreted via bile, urine and faeces in rats fed with S. chinensis extract.Similarly,Kim et al. (2014)stated only negligible excretion of Schisandra lignans via urine after oral and intravenous administration of S. chinensis extracts to rats.