Deficiency of estrogen could change the balance between osteoclast and osteoblast which is the reason for increasing the risk of fracturing bone in menopause women and is defined as a kind of osteoporosis. The Women’s Health Initiative (WHI) showed that Hormone Replace Therapy (HRT) prevents the fracturing bone due to the osteoporosis that is in turn the result of estrogen deficiency, but it could increase the risk of breast cancer as well as cardiovascular diseases in women who take HRT for a long time [1
]. ERα is the major ER mediating the effects of estrogen on bone in females and in males; however, ERβ may play a role in estrogen action in the trabecular bone under certain circumstances as a substitute or enhancement of ERα action [2
]. The result of [3
] demonstrated that among the natural estrogens (estriol, estrone, 17α-estradiol, and 17β-estradiol), 17β-estradiol (E2) was the compound with the highest potency toward the three zebrafish estrogen receptors (zfERα, zfERβ1
) and among the phytoestrogens (genistein, ferutinin, and liquiritigenin), ferutinin with similar affinity for hERα and hERβ but different activity (full agonist on hERα and partial agonist on hERβ), behaved as full agonist on the three zfERs. Therefore, studies have been focused on phytoestrogens because of their estrogenic activities and lack of adverse side effects associated with estrogens [4
In recent years, several publications from Italy addressed the concept of ferutinin as an interesting phytoestrogen with an osteoinductive capability with application either for treatment of osteoporosis or in bone tissue engineering [4
]. The result also demonstrated the role of ferutinin in the impairment of female sexual function and its effect on promoting proliferation and differentiation in human dental pulp and amniotic fluid stem cells [10
]. The importance of ferutinin has been realized in bone metabolism since it is capable to prevent and treat osteoporosis including estrogen deficiency in ovariectomized rats [5
]. Ferutinin, through ERα, improves bone reconstruction, when orally administered in rats with a calvaria critical size bone defect, filled with a collagen type 1 and human amniotic fluid stem cells (hAFSCs) construct. This construct leads to an approximately 70% bone reconstruction showing that ferutinin could act as a healing promoting factor, on hAFSCs including osteogenic differentiation [4
]. In another study [11
], it was tried to explain a possible molecular mechanism for ferutinin-induced osteoblastic differentiation of hAFSCs, through ERα and GPR30, evaluating the role of the MEK/ERK and P13K/Akt signaling pathway, and indicated that ferutinin is able to stimulate both MEK/ERK and P13K/Akt signaling in undifferentiated hAFSCs, although with different timing of the phosphorylation pattern. Moreover, [11
] indicated that in the canonical osteoblastic differentiation model, both pathways were involved, but P13K/Akt is required to ferutinin stimulated osteoblastic differentiation through ERα.
Zebrafish is a well-developed model system for studying both embryonic development and human diseases. The primary advantages of zebrafish for drug discovery include small size, optical transparency, rapid development, large number of their embryos and larvae, as well as their high genetic, physiologic, and pharmacologic similarity to humans, particularly, high similarity with humans in terms of bone architecture, bone cells, matrix proteins, and molecular signaling defined as an ideal in vivo model for the systematic identification of bioactive natural products with therapeutic capability and suitable model for screening of the agents to prevent and treat osteoporosis [14
]. The ovariectomized rat is the most common animal model in the study of anti-osteoporosis medication which was used in the study of [5
] to find effect of ferutinin on osteoporosis. Zhang et al. [20
] used osteoporotic model induced by glucocorticoids as chemical drugs to show anti- osteoporosis effect of Chinese traditional drug after discussing about the disadvantages of using ovariectomised rat. Luo et al. [18
] and Vrieze et al. [19
] used chemical drugs such as prednisolone and dexamethasone to induce osteoporosis in zebrafish. Luo et al. [18
] recommended monitoring bone formation directly in transgenic zebrafish tg (sp7:egfp)
and inferred to fine and complicated procedure of staining a tiny model for directly observation of bone formation in the model of wild-type zebrafish larvae.
There are several studies which used zebrafish embryos to study the effects of estrogens on organ formation and function [21
] and to trace the effect of E2 on chondrogenesis and expression of related genes [24
]. Pashai Ahi et al. [26
] demonstrated E2 mediated differential expression of some genes which involved in craniofacial skeletal development (e.g., bmp2a/b
) as well as genes co expressed with esr1
(a ligand-activated receptor for estrogen). However, a significant amount of research has been conducted for the effect of E2 on cartilage development and related gene expression using zebrafish larvae [24
], but there is not any study about the effect of E2 on the ossification of zebrafish.
The drug screening using zebrafish larvae require a system to expose embryos to the compound in culture solution over a specific length of time. The test platform of the screening mainly relies on multi-well plate in which the compounds or culture solution require being renewed and the dead embryos should be removed [27
]. While the washing embryos are time-consuming and may injure the embryos, a large and growing body of literature has investigated the flow conditions in microfluidic chip to cultures them [27
]. For example, [28
] developed a modified 24-well plate with a flow-through system which requires a large volume of the test compound. A research team [29
] established a microfluidic chip that docks zebrafish embryos automatically and cultures them under flow conditions, but it does not allow for the timely removal of the dead embryos. Akagi et al. [30
] tried to overcome this problem and introduced the 3D multilayer microfluidic system for real-time developmental analysis of zebrafish embryos, but it was impossible to expose the fish eggs to multiple substances at different concentration. Li et al. [27
] introduced a microfluidic device to simultaneously evaluate the developmental toxicity of an anti-asthmatic agent on zebrafish embryos and larvae using real-time imaging.
Here, we investigated whether the effect of ferutinin on bone mineralization in vivo is similar to E2 by exposing zebrafish larvae to a range of E2 and ferutinin or not. In this context three sections for the experiments were designed: (i) investigating the effect of E2 on bone mineralization of zebrafish larvae as positive control; (ii) monitoring the developmental toxicity of ferutinin; and (iii) finding the effective concentration of ferutinin on bone mineralization and expression of target genes as main purpose of the study. The positive effect of ferutinin on bone mineralization of wild-type zebrafish would be helpful to increase studies in this area such as investigating the side effect of ferutinin and possibly to introduce it as a drug.
4. Materials and Methods
4.1. Zebrafish Maintain, Husbandry, and Embryo Care
All procedures involving zebrafish were performed in accordance with protocols approved by the University of Saskatchewan Committee on Animal Care and Supply and Animal Research Ethics Board (#200090108). Since the current study was the first in ferutinin-treated zebrafish studies, wild-type of zebrafish was used to optimize the method of treating for the next study on osteoporotic zebrafish. Adult wild-type zebrafish maintained in Dr. Brian F. Eames lab in an Aquatic Habitats Flow-Through System (Apopka, FL, USA) on a 14/10 day/night cycle to mimic natural conditions and were fed with alive brine shrimp and chironomids (Hikari, Hayward, CA, USA) at least once a day. The couples of wide type adult zebrafish (AB strain) mated in separated tanks, and the eggs were harvested the next day. Healthy eggs which were recognized under microscope, were transferred into 0.5× E2 (7.5 mM NaCl, 0.25 mM KCl, 0.5 mM MgSO4, 75 mM KH2PO4, 25 mM Na2HPO4, 0.5 mM CaCl2, 0.35 mM NaHCO3, 0.5 mg/L Methylene Blue, pH ~7.0) and were incubated at 28 °C. The dead embryos (opaque white rather than transparent) were removed and remaining embryos were rinsed once more on the day of collection and every 24 h thereafter. All embryos and larva were kept in an incubator at 28 °C when they are not being treated or cleaned.
4.2. Chemical Treatments
According to the purposes of current study four groups of chemicals were applied: Group 1 for investigating the effect of E2 on bone mineralization included 2, 4, 6, and 8 µM of E2; group 2 for comparing the effect of E2 and ferutinin on bone mineralization included 0.625, 1.25, 2.5, 5, and 10 µg/mL concentration of ferutinin (Sigma-Aldrich, Oakville, ON, Canada); 2 µM of E2 (Sigma-Aldrich, Oakville, ON, Canada) and 0.1% concentration of DMSO; group 3 for comparing the effect of E2 and ferutinin on target gene expression included0.625, 1.25 and 5 µg/mL concentration of Ferutinin; 10 µM of E2 and 0.1% concentration of DMSO and group 4 for real-time PCR included 5 µg/mL concentration of ferutinin; 10µM of E2 and 0.1% concentration of DMSO.
Firstly, E2 stock solution at 10mM was prepared by dissolving 17β-estradiol in 100% DMSO. Then, the stock solution was diluted in embryo medium (EM) to the final concentration of 2 µM (540 µg/L). Furthermore, ferutinin stock solution at 10 mg/mL was prepared by dissolving ferutinin in 100% DMSO. The stock solution was added to the embryo medium (EM) to the final concentration of 0.625, 1.25, 2.5, 5 and 10 μg/mL. The final DMSO concentration after diluting treatments in embryo medium was 0.1% to avoid malformation and positive/negative false results [32
] and negative control contained 0.1% DMSO. The treatment experiment was performed in 24-well plates, which was designed to add treatment at each time point.
After plating chemicals, clutches of zebrafish embryos from several pairs of adult fish were divided and transferred into 24-well plate. Treating with small molecules was done with adding 3–4 larva to each well and exposed to 2 mL EM containing treatments at each time point. In the other words, they exposed to treatments based on aims of the study from 6 hpf, 1 dpf, 2 dpf, and 3 dpf.
4.3. Sample Selection and Screening
Coding of each treatment was performed according to the time point of adding treatment and after raising embryos of each clutch in EM to target age, 3–4 larvae were added to each well. Then, plates including larvae of each clutched were incubated at 28 °C which is a more physiologically relevant temperature for zebrafish and increased the potency of estradiol approximately 10-fold compared to incubation at 37 °C [3
The screening was performed to investigate changes in expression of the target genes and phenotype of target bones after treating larvae at each time point according to the timelines (Figure 7
). For example, approximately 10 larvae of each clutch were treated at 1 dpf and remained in 24-well plate at 28 °C incubator for staining at 6 dpf. The rest of them were treated at 2 dpf and 3 dpf, respectively. Then, harvesting was performed in two steps; the 30 larvae of each treatment samples were anesthetized with 0.4% tricaine (MS-222, Sigma) and were placed into microtube at 4 dpf for RNA isolation [22
] and rest of them were collected at 6 dpf for staining [18
In order to investigate the effect of E2 on bone mineralization of zebrafish larvae, three biological replicates of approximately six larvae were collected at 6 dpf for a total of 20 larvae at each exposure time (6 hpf, 1 dpf, and 2 dpf) and treatments (0, 2, and 8 μM of E2). Moreover, an extra treatment with 8 μM of E2 at 3 dp was performed to find the effect of exposure time on bone mineralization. The ferutinin part including six treatments (0.625, 1.25, 2.5, and 5 μg/mL of ferutinin, E2, and DMSO) at three exposure time (1 dpf, 2 dpf, and 3 dpf) for analysis the mortality rates and four treatments (0.625 and 1.25 μg/mL of ferutinin, E2 and DMSO) at three exposure time (1 dpf, 2 dpf, and 3 dpf) for the molecular and histology section.
4.4. Calculation the LC50 of Ferutinin
The LC50 is the concentration of the compound which causes mortality in 50% of the treated-test subjects over a specific period of time [35
]. The mortality of the larvae which were exposed to ferutinin (0.625, 1.25, 2.5, and 5 μg/mL) at three exposure time (1 dpf, 2 dpf, and 3 dpf) was calculated by counting the number of dead larvae of each clutches in each treatment until 6 dpf. Treated- larvae were observed daily, and dead larva were removed but were noted in final calculation. Several wells of the 24-well plates were assumed as quarantine and sick larvae or which showed malformation were transferred to quarantine with similar treatment. The linear logarithm of the exposure time and the mortality percentage until 6 dpf were plotted to reach the toxicity curves of each concentration of ferutinin. According to the mortality percentage of exposed larvae, the LC50 of the concentrations of ferutinin at each treatment group were calculated using the LC50 calculator [31
4.5. RNA Isolation and cDNA Synthesis
Total RNA was isolated from embryos using the RNeasy mini kit (including the RNase-free DNase set) as described by the manufacturer (Qiagen, Mississauga, ON, Canada). It started with pooling around 30 larvae of each group of treatment from different clutches which were treated with ferutinin (0.625, 1.25 and 5 µg/mL) at different time points (1 dpf, 2 dpf, 3 dpf), DMSO at 1 dpf, 2 dpf, and 3 dpf, and 10 µM of E2 at 3 dpf in TRI Reagent (Sigma) and homogenized with a disposable tissue grinder pestle with matching 1.5 mL microtube (Kimble Kontes, Waltham, Massachusetts, USA). The quantity of the resulting RNA samples was assessed using a NanoDrop -Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). cDNAs were prepared from 0.1 μg of total RNA using a RevertAid H Minus First Strand cDNA Synthesis Kit as described by manufacturer (Thermo scientific, Lithuania, EU). The 20 µL reaction volume was diluted 40-fold prior to PCR amplification.
4.6. RT-PCR and Real-Time PCR
Gene- specific primers used for RT-PCR and real-time PCR which was previously reported by [26
] listed in table. The candidate target genes included estrogen receptors (esrra
), and potential skeletogenic targets of estrogen pathway (bmp2a/b
) which were upregulated in the larvae which were exposed to E2 at 8 hpf with 2 and 5 μM of E2 and the media were refreshed daily until the target time (3–7 dpf) of harvesting samples [26
]. Moreover, [26
] validated ppi2
, and tbp
are three suitable reference genes to accurately quantify the small differences in gene expression in developing heads of zebrafish larvae across the E2 treatment groups, but when we tested them, just rpl8
showed constant expression and we selected it as a reference gene for real-time PCR. RT-PCR reaction conditions were performed using T100™ Thermal Cycler (Bio-Rad, Mississauga, ON, Canada). The PCR products of the larvae which were treated with ferutinin (0.625, 1.25 and 5 µg/mL) at three time points (1 dpf, 2 dpf, 3 dpf), DMSO at different time points (1 dpf, 2 dpf, 3 dpf), and E2 10 µM at 3 dpf were visualized using a 2% agarose gel electrophoresis. For RT-qPCR an ABI 7500 real-time PCR System (Applied Biosystems, ON, Canada) was employed using Power SYBR Green PCR MasterMix (Thermo Fisher Scientific, Foster City, CA, USA) following the manufacturer’s instructions. The Cycle threshold values (CT) was automatically determined by StepOne Software v2.1 (Bio-Rad, Mississauga, ON, Canada). The group of treatment including larvae which were treated with 5 µg/mL of ferutinin, 0.1% of DMSO (negative control to calculate ΔΔCt
) and 2−ΔΔCt
) and 10 µM of E2 at 3 dpf were assayed in duplicate which had two replicates of cDNA of esr1
, bmp2 a/b
, no template control (NTC) for checking the contamination in primer/probe mix or formation of primer dimer and no reverse transcriptase control (NRC) which was prepared to confirm the absence of genomic DNA contamination in RNA samples. The abundance of target and reference genes within each sample was evaluated using relative standard cure method. The data of each amplified genes were averaged and normalized to rpl8
(reference gene to calculate ΔCt
4.7. Staining and Scoring
Alizarin red staining was used to evaluate bone mineralized matrix deposition which is an important indicator of bone formation [18
]. However, alizarin red can attach to calcium salt, staining the cartilage using alcian blue was necessary in current study to recognize and trace bone mineralization without fluorescent. Zebrafish larvae were collected and anesthetized at 6 dpf (when the cranial bone did not develop completely) to be fixed for 1hour in a 2% paraformaldehyde solution. The staining was performed based on two-color acid-free cartilage and bone staining protocol for zebrafish larvae [36
]. After staining the larvae at 6 dpf, images of dorsal aspect head bone of zebrafish were taken using a DFC310 FX camera (Leica, Wetzlar, Germany) of M205 FA stereomicroscope (Leica, Wetzlar, Germany).
Regarding cartilage and bone developments of the head skeleton in 10 dpf zebrafish [37
], an atlas of zebrafish craniofacial development at a cellular resolution [38
], and the schematic images of zebrafish bone development related to gene expression [39
] the bone elements especially hyomandibular and ceratohyal were recognized. Pashai Ahi [39
] specified the signaling pathways which were related to morphological changes of the cartilage elements, for example developmental changes in ceratohyal, palatoquadrate, and Meckel is related to estrogen pathway and the changes in hyomandibular is related to BMP pathway. The quantity of hyomandibular and ceratohyal mineralization area was determined according to the scoring system [40
] of the target bones area which was stained with alizarin red staining. The applied scoring system including four scores: it was 0 when the target area was just blue without bone mineralization, 1 for bone mineralization of one part of target cartilage, 2 for red stained two parts of cartilage but not completed, and 3 was for completely mineralized bone (Figure 8
4.8. Data Analysis and Statistics
The mineralization of the target bones (hyomandibular and ceratohyal) according to scoring system was assessed for chemical concentration in special time point which was tested in triplicate (number of larvae in each well) in at least three independent experiments (number of tested clutches). The data was sorted according to the coding system using Excel. Then the ceratohyal and hyomandibular mineralization was analyzed using IBM SPSS Statistics 19 (IBM Corp., Armonk, NY, USA) as the mean ± SEM. The comparison of the means were statistically compared by one-way ANOVA (using coded treatment as independent factors and scored bones as dependent factor) for the analysis of variance and followed by a Tukey’s post-hoc test (when warranted) for multiple comparison of the means. The bone mineralization of the treated larvae were analyzed by two-way ANOVA (using the concentration and exposure time as independent factors and scored bones as dependent factor) followed by a Tukey’s post-hoc test to confirm the result of one-way ANOVA. Statistical significance for all groups of treatments was set at p < 0.05. The graphs were modeled by the Microsoft Excel 2010 (Microsoft Corp., Redmond, WA, USA).