Detrimental effect of Hypericum perforatum on ovarian functions

Objective: Hypericum perforatum is widely used for depression and distress treatment as an over-the-counter plant at any age. This study investigated the safety of H. perforatum on ovarian function and infertility. Material and Methods: H. perforatum was given to rats in two different dosages (100 and 300 mg/kg/day) with drinking water for four weeks. Half of the treatment groups were sacrificed at the end of the four-week intervention, the remainder was sacrificed after an additional four-week waiting period to see if there was reversibility. At the end of the experiment, blood samples and both ovarian tissues were obtained under anesthesia with ketamine and xylazine (50 mg/kg and 5 mg/kg, respectively). Results: Although primordial follicle numbers were not affected with a dose of 100 mg/kg, they were significantly decreased (28.6%) when the dose was tripled. Primary follicle numbers stayed the same, but secondary and tertiary follicles numbers were significantly dose-dependently decreased, and remained significantly low four weeks after the intervention. Anti-mullerian hormone (AMH) levels were not significantly different between the groups. Conclusion: H. perforatum treatment did not change serum levels of AMH because the primary follicle number did not decrease. However, the other follicle counts decreased in a dose-dependent manner and full recovery was not regained after four weeks. The detrimental effect of H. perforatum on primordial follicles should be taken into consideration because any woman using H. perforatum could also experience ovarian failure.


Introduction
Although psychological distress by itself can cause psychosomatic infertility or decrease treatment success (1), the treatments of distress and depression with medicines and plants may also have a negative impact on ovarian function.
There is a growing interest in plants and their derivatives, with an assumption that they are a safe method of treatment.
Patients with infertility are increasingly using complementary and alternative medicines to support or replace medical fertility treatments (2). Extracts from the plant Hypericum perforatum (St John's wort) have become an appealing alternative therapy to prescription serotonin-modulating drugs and are widely available to women of childbearing age (3,4).
There is little research on H. perforatum and its use in pregnancy, and its effect on fertility has not been established, even in European Medicines Agency guidelines (EMA) (5). Therefore, we evaluated impact of H. perforatum on ovarian tissue by counting follicles and assessing anti-mullerian hormone (AMH) levels. DOI 86-23, revised 1985) and specific national laws were followed throughout the study. On the study day, the rats were randomly assigned to five groups of seven animals each.

Experimental design
Control group: The rats in this group were allowed free access to tap water.

Low-H. perforatum group:
The rats in this group were administered 100 mg/kg H. perforatum with drinking water for four weeks.

Low-H. perforatum waiting group:
The rats in this group were administered 100 mg/kg H. perforatum with drinking water for four weeks, and then the rats received no medication and were sacrificed 4 weeks later.
High-H. perforatum group: The rats in this group were administered 300 mg/kg H. perforatum with drinking water for four weeks.
High-H. perforatum group: The rats in this group were administered 300 mg/kg H. perforatum with drinking water for four weeks, and then the rats received no medication and were sacrificed 4 weeks later.
The rats were weighed every monday and the doses of H. perforatum were adjusted for every cage. At the end of the experiment, under the anesthesia with ketamine and xylazine (50 mg/kg and 5 mg/kg, respectively), blood samples were obtained by cardiac puncture, centrifuged at 3000 rpm/4 °C, and sera were stored at -80 °C for AMH measurement. In addition, both ovaries were harvested and kept in 10% formalin.

Follicle counting
After routine tissue processing, the obtained samples were sliced in 5-micrometer thickness and evaluated under an optical microscope (Zeiss Primo star, Ankara, Turkey) with hematoxylin-eosin (H-E) staining. Oocyte classification was based on previous studies (6,7). Secondary follicle (pre-antral follicle); follicles containing two or more layers of granulosa cells, without formed antrum follicles (Figure 1h, i).

Measurement of serum rat AMH
The serum level of rat AMH was determined using a commercial enzyme-linked immunosorbent assay (ELISA) kits (rat ELISA kit, Sunred, Baoshan District, Shanghai, China). The rat AMH ELISA kit sensitivity was 0.101 ng/mL with a coefficient of variation of <5%. Procedures were performed according to manufacturer's instructions.

Data presentation and statistics
AMH levels and ovarian count evaluation was performed using the Mann-Whitney U test. Data are presented as mean ± standard error mean, p values below 0.05 were considered significant. Follicle counts are also represented as percent changes.
Serum AMH level of the three groups were not significantly different from each other (p>0.05); the control group's AMH level was found as 4.39±0.47 ng/mL, the low-dose H. perforatum group was 4.63±0.63 ng/mL, and the high-dose H. perforatum group was 5.02±0.29 ng/mL.

Discussion
Many researchers have focused on the beneficial anti-depressive effects of H. perforatum, and despite there being few reports about its harmful effects, unfortunately, this does not signify the plant's safety, it shows a lack of rigorous research. This over-thecounter plant is available in many countries as an alternative treatment modality to anti-depressive medicines (5). The EMA determined in its guidelines the adult dosage is a total 900 mg/ day (three-times 300 mg/day) for the treatment of depression (5). Unfortunately, it has been reported that it can be abused among adolescents who believe it can make them feel good (8), in this case, its limited upper dose has not been established due to its non-prescription use. Depression can affect people of any age and disturb fertility, and any woman could be exposed to H. perforatum, assuming that it is safe, either before being pregnant and even while pregnant. Therefore, this study has investigated harmful effects of H. perforatum on ovarian tissue in two different doses and evaluated reversibility and if there was any alteration on ovarian reserve.
In this study, 100 mg/kg doses of H. perforatum did not inhibit the primordial stage, but when the dose was tripled, these parts  of the follicles were also affected. The primary follicle number remained the same in both doses; secondary and tertiary follicles numbers decreased in a dose-dependent manner and the ovarian cell count waiting did not recover after a four-week treatment cessation period. An earlier study showed that onehour pretreatment of H. perforatum (0.6 mg/mL) resulted in zero penetration of hamster zona-free oocytes and significant denaturation of sperm DNA to decrease the sperm viability, leading to concerns that its use may lead to decreased fertility (9). When H. perforatum was given as 180 mg/kg to mice two weeks before mating, the authors found no specific concerns about the pregnancy rates of the treated and non-treated groups; however, the study was focused on pup development and the number of live pups per litter were not found different between the groups (3).
There are no toxicity reports with regard to exposure to H. perforatum on direct ovarian tissues in the literature, but it is possible to find some deleterious effects on other organs. Gregoretti et al. (10) gavaged rats during gestation and for 21 days during lactation at doses 100 or 1000 mg/kg and renal and hepatic damage was identified in the pups at both doses. We did not want to use 1000 mg/kg doses of H. perforatum, which seem very high, our aim was just to mimic real life and worked with lower-moderate doses as 100 mg/kg and 300 mg/ kg, following Rayburn et al.
We also searched reports of HP on hormonal status in the literature. One study evaluated H. perforatum (900 mg/kg) interaction with oral contraceptive therapy in sixteen healthy women. As the CPY3A4 enzyme was induced, which is wellknown aspect of H. perforatum, the metabolism of these hormonal components was increased by approximately 25% and resulted in breakthrough bleeding, pre-ovulatory follicles and follicle exceeding (11). Thirty-six women aged 18-45 years with regular menstrual cycles who were diagnosed as having mild premenstrual syndrome were given H. perforatum (900 mg/day); in the follow-up, there were no differences in plasma follicle-stimulating hormone, luteinizing hormone, estradiol, progesterone, prolactin, and testosterone found compared with the non-treated group (4). Contrarily, hypo-prolactinemic activity has been reported in healthy male volunteers and plasma growth hormone levels increased due to dopaminergic action of H. perforatum (12). Plasma cortisol levels were significantly elevated in only four male volunteers out of twelve (12). Similarly, Di Carlo et al. (13) showed an inhibitory activity of H. perforatum (100 mg/kg) on prolactin production in male rats with 15 days' treatment and suggested its clinical reflection might be luteal inadequacy, probably observed due to following pharmacologically-induced low prolactin secretion. An animal study on the hypothalamic-pituitary-adrenal axis