2024 Volume 71 Issue 4 Pages 317-333
Soybean is a source of protein, fibers, and phytochemical isoflavones which are considered to have numerous health benefits for children and adulthood. On the other hand, isoflavones are widely known as phytoestrogens that exert their action via the estrogen signaling pathway. With this regard, isoflavones are also considered as endocrine-disrupting chemicals. Endogenous estrogen plays a crucial role in brain development through binding to estrogen receptors (ERs) or G protein-coupled estrogen receptors 1 (GPER1) and regulates morphogenesis, migration, functional maturation, and intracellular metabolism of neurons and glial cells. Soy isoflavones can also bind to ERs, GPER1, and, furthermore, other receptors to modulate their action. Therefore, soy isoflavone consumption may affect brain development during the pre-and post-natal periods. This review summarizes the current knowledge on the mechanisms of isoflavone action, particularly in the early stages of brain development by introducing representative human, and animal models, and in vitro studies, and discusses their beneficial and adverse impact on neurobehavior. As a conclusion, the soy product consumption during the pre-and post-natal periods under proper range of dose showed beneficial effects in neurobehavior development, including improvement of anxiety, aggression, hyperactive behavior, and cognition, whereas their adverse effect by taking higher doses cannot be excluded. We also present novel research lines to further assess the effect of soy isoflavone administration during brain development.
Soybean has been part of the human diet for a long time. It contains abundant proteins, dietary fibers, and numerous phytochemicals, including isoflavones [1, 2]. Soybean isoflavones (genistein, daidzein, and daidzein metabolite, S-equol) are known as phytoestrogen due to their structure similarity with endogenous estrogen, 17-β-estradiol (E2) [1, 3]. The isoflavones binding to the estrogen receptors (ERs) leads to the direct binding of ERs to the estrogen-responsive-element (ERE) of the target gene, and then activates the transcription. Isoflavones-ERs complex may also act as a transcription factor that interacts with other transcription factors to activate the other nuclear receptor-mediated transcription, such as vitamin D receptor (VDR) [4-8]. Isoflavones also bind to membrane receptors, such as the G protein-coupled estrogen receptor (GPER1; also known as GPR30) [5, 9]. In addition, isoflavones exhibit considerable antioxidant activity by increasing antioxidant enzyme activity, reducing oxidative stress, and preventing low-density lipoprotein cholesterol oxidation [1, 5]. Furthermore, isoflavones activate mitogen-activated protein kinase (MAPK) and phosphatidylinositide 3-kinase (PI3K) signaling pathways to regulate the target genes [7, 10, 11]. These broad spectrums of isoflavone action may affect physiological conditions of human health, especially during development.
Soybean and its extract have been consumed since the early stage of human evolution, from newborns to elderly. By considering the isoflavones action as discussed above, pre- and post-natal consumption of soy isoflavones, including maternal consumption, might affect the development of various organs, including the brain. The human brain development starts two weeks after the conception. Pre- and post-natal brain development, is influenced by genetic factors, hormones, neurotransmitters, growth factors, and environmental factors such as nutrition (i.e., folic acid) and toxic substances (i.e., alcohol) [12-14]. Particularly, development of many brain regions is regulated by various hormones [12, 15]. Thus, this period is most susceptible to phytoestrogen and endocrine-disrupting chemical exposure [12, 15].
Epidemiological studies have shown that pre- and post-natal exposure to isoflavones affects neurobehavioral development in children. A birth cohort study in which they measured the urinary concentration of isoflavones of the Chinese population showed that moderate levels of isoflavone exposure have neuroprotective effect, showing less anxious/depressed in both boys and girls at 2 and 4 years of age, compared to the lowest exposed groups. In contrast, high-level exposure to isoflavones may be toxic to neurodevelopment in children [16]. In this cohort study, the median urinary concentration of isoflavone (i.e., genistein) is 688.4 ng/mL [16], which is comparable to the Asian countries (640 ng/mL in Japan and 300 ng/mL in Vietnam), but not to Western countries (30–105.4 ng/mL) [17-20]. Another cohort study in Japanese children (Kyushu Okinawa area) showed that maternal isoflavones consumption during pregnancy (the mean of 10–30 mg of isoflavones from daily intake of 100–200 g total soy product) may be protective against hyperactivity problems and lower prevalence of depressive symptoms during pregnancy [21, 22]. However, postnatal exposure to isoflavones may be controversial in inducing the beneficial/toxic endpoints in human health, especially in children. During the postnatal period, the infant received the soy isoflavones from the mother’s daily consumption of soy products through breast milk or direct consumption from soy-based infant formula. The concerns regarding postnatal exposure to isoflavones, especially soy-based infant formula, in the children’s development and human health, have been addressed [23, 24]. It has been reported that soy-based formula-fed infants have higher circulating plasma concentrations of isoflavones compared to endogenous estradiol [25], although neither pediatricians nor pediatric endocrinologists have reported adverse estrogenic effects on sexual development in infant boys or girls fed with soy-based formula [24, 26, 27]. Furthermore, comprehensive literature and epidemiological studies showed no conclusive evidence of adverse effects of soy-based formula fed in nutritional adequacy, on sexual development, neurobehavioral development, cognitive function, immune development, and thyroid disease [23, 24]. In addition, recent studies have shown that in Korean populations, infants fed soy formula had no apparent association with increased risks of epilepsy, ADHD, ASD, and developmental status. In contrast, in the United States population, soy-based infant formula may be contributing to autistic behaviors, and febrile seizures in autistic boys and girls and a concurrent increased risk of epilepsy in these boys [28-30]. Thus, soy food product consumption under normal physiological conditions may be beneficial to our health, although further studies are required to confirm this, especially under acute and high intake conditions. More careful analysis may be required to determine the ambient/toxic dose of isoflavones in brain development during different time points of life.
The effect of soy consumption on brain development in human embryos is difficult to study due to the ethical and practical constraints. Alternatively, researchers used various animal models using rodents or other species of different ages to examine their effects on brain development and maturation events. Although the timing of each event in the brain developmental process may differ between humans and rodents, the composition of cell subsets and signaling pathways are essentially the same, allowing us to use animal models to examine the effect of various substances [31]. In animal models studies, isoflavones caused neurobehavioral improvements or disorders depending on the dose of administration. Low-dose perinatal exposure to genistein (1–5 mg/kg/day) showed an improvement in spatial learning and memory, reduced aggressive behavior, and induced defensive behavior. On the other hand, high-dose perinatal exposure to genistein (100–300 mg/kg/day) showed an increase in anxiety and aggressive behavior, induced higher offensive behavior, and reduced social behavior [32-36]. In addition, the exposure of 5 mg/kg/day genistein during gestation and lactation period impaired spatial learning in male rats, but not in rats exposed only during gestation or lactation [37]. In contrast, the rats that were exposed to rich phytoestrogen diets (20–30 g/day daily food intake of Phyto-rich pellet; that approximately contains 420 μg/g of total isoflavones) during perinatal and postnatal periods reduced the anxiety in both male and female rats and also increased the visual and spatial memory in female rats [38, 39]. Therefore, dose variation of isoflavones and the timing of exposure might differently affect the neurobehavioral development in animal models. Furthermore, the combination or single (i.e., genistein only) exposure to isoflavones compound might also result in different outcomes. Further studies to examine the differential exposure levels of isoflavones at specific timing during brain development are still needed.
During brain development, various subset of the cells is involved. Thus, it is complicated to examine the effect of exposure of specific chemicals in specific subset of cells and/or cell-cell interactions. However, various studies have tried to clarify such action by combining in vitro studies and in vivo animal models. Isoflavones affect the proliferation, migration, differentiation, neuritogenesis, and synaptogenesis of neurons and glial cells [11, 14, 40-44]. As a result, isoflavone exposure during development affects the neurobehavioral and cognitive function in offspring [12, 21, 22, 32, 38, 45]. This review article aimed to integrate the current knowledge on the mechanism of action of isoflavones during the early stages of brain development in representative brain regions. In addition, we introduce a novel pathway of isoflavone action in brain cells and propose new lines of research to assess isoflavone action during brain development.
Isoflavone is a secondary plant metabolite biogenetically derived from a 2-phenylchroman flavonoid skeleton. Isoflavones can present as aglycons or glycosides, primarily in β-D-glycoside [1, 5]. The aglycone form of isoflavones is biologically active. Genistein, daidzein (the aglycones), and S-equol (daidzein metabolite) are the most extensively investigated isoflavones, mainly from soybean and its products. Naringenin is the precursor of genistein. The precursor of daidzein is flavanone liquiritigenin; meanwhile, S-equol is a daidzein metabolite produced by intestinal bacteria in approximately 50% of humans [1, 17, 18].
Isoflavones are produced in soybean (Glycine max), red clover (Trifolium pratense), white clover (Trifolium repens), and alfalfa (Medicago sativa). In addition to the three representative soy isoflavones mentioned before, there are several other isoflavones such as biochanin A, formononetin, and coumestrol. The isoflavone content of soybean, red clover, white clover, and alfalfa is 1.2–5, 10–25, 0.5–0.6, and 0.05–0.3 mg/g dry weight, respectively [1, 2, 46]. The isoflavone concentrations depend greatly on the variety, growing locations and conditions, climatic conditions, and preservation methods [1, 3, 5]. Soybean isoflavones are the primary source of isoflavones consumed by humans and established into various foodstuffs (Table 1) [47]. Although red clover has higher isoflavones content, in this review, we focus on soybean isoflavones since it is mainly consumed during development from pregnancy until adulthood.
The isoflavones content of various soybean products
Food Items | Total Isoflavones (mg/100 g) |
---|---|
Soybean, raw, green, mature | 128.83 |
Miso | 41.45 |
Natto | 82.29 |
Soy flour | 172.55 |
Soy protein drink | 81.65 |
Soy protein isolate | 91.05 |
Soy sauce | 1.18 |
Soybean chips | 54.16 |
Soymilk, iced | 4.71 |
Tempeh | 60.61 |
Tofu | 22.73 |
ERα and ERβ are nuclear superfamily steroid hormone receptors. In the absence of endogenous E2, the ERs are associated with heat-shock proteins in a transcriptionally inactive state. Endogenous E2 binding to ER induces a conformational change, promoting homodimerization in specific tissues leading to nuclear translocation. In the nucleus, the endogenous E2-ER complex binds to the estrogen response element (ERE) in the promoter region of the target genes, then recruits the coactivator to regulate target gene transcription in a ligand-dependent manner. This classical mechanism is known as genomic estrogen action [48-50]. Isoflavones are widely known as phytoestrogens that can bind and activate both ERα and ERβ, as shown in Fig. 1 [1, 5, 51, 52]. The isoflavones-ER binding was examined under the condition with free endogenous E2 [52]. In addition to phytoestrogen, isoflavones are also classified as selective estrogen receptor modulators, implying that isoflavones have tissue-selective effects [5, 53]. Therefore, isoflavones binding to ER also affect gene transcription, modulating estrogen-responsive gene expression. The DNA microarray analysis to examine the effects of genistein in the developing rat uterus shows that genistein alters the expression of a significant number of genes 6 to 8 times more than endogenous E2. Genistein affected 227 genes, most of which were downregulated [54, 55]. The abilities of isoflavones in modulating of different estrogen-responsive genes may also depend on the relative ratio of ERα and ERβ present in various cell types.
Summary of isoflavone’s actions. Isoflavones can affect cell function mainly through nuclear ER shown in the red box. After entering the cell, isoflavones bind to ER in the nucleus and directly mediate gene transcription by binding to the specific response element or interacting with other transcription factors to influence transcriptional activity. Isoflavone also affects the transcriptional activity by binding or interacting with the other nuclear receptors (NR), such as TR and PPAR. In addition, isoflavones also bind to GPER1 and activate protein kinase pathways that regulate transcription and other mechanisms that may affect F-actin rearrangement and neurotransmitter release. Furthermore, isoflavones bind and modulate the membrane ER (mER) that induces changes in intracellular calcium or interact with mGluR1 and NMDR to induce synaptic plasticity. Isoflavones may also directly interact or modulate mGluR1 and NMDR to induce neurotransmitter release and synaptic plasticity (created with BioRender.com).
In some cells, such as breast cancer cells, the relative binding affinity of isoflavones is greater for ERβ than for ERα, while E2 binds to both receptors with approximately equal affinity [5, 56]. Genistein exhibits a >20–fold higher affinity for ERβ than ERα [57, 58], whereas daidzein and its metabolite S-equol exhibit ERβ selectivity [58]. The potency of isoflavones in activating ER-mediated transcription has been studied using a reporter gene assay with ERα or ERβ expression vectors. In MCF–7 cells, the EC50 range of isoflavones was 10–6–10–7 M and 10–9–10–7 M for ERα and ERβ, respectively [59, 60]. On the other hand, in MCF–7SH cells that exclusively express ERα, but not ERβ, genistein activates endogenous ERα, and a 10–8 M dose is sufficient to achieve substantial induction [61]. Furthermore, fold induction of transcription by genistein in the reporter assay in the HeLa cell is higher in Gal–ERα than Gal–ERβ [61]. The differential binding properties of isoflavones ERα and ERβ result in a differential potency to modulate estrogen action in each cell subset depending on the relative expression ratio of ERs.
In addition to the full-length protein of ERα (66-kDa, designated ER66), splice variants of ERα have been described in the 46 kDa protein (ER46, truncation at amino-terminal) and the 36 kDa protein (ER36, truncation at amino and carboxy-terminal) [58, 62-64]. These ERs located in the plasma membrane are known as membrane ER (mER), which mediated physiological changes through calcium mobilization [62]. Several reports have shown that isoflavones, especially genistein and daidzein, can rapidly increase intracellular Ca2+ via mER [62-65]. For example, in the ER46 expression vector of HEK293, the IC50 for genistein is 2.53 nM, daidzein, 1.11 μM, and E2, 17.58 pM, whereas ER36 showed no saturable specific binding [64]. In addition, a macromolecular conjugate of genistein demonstrated specific mER binding and colocalization with a macromolecular conjugate of E2 by immunofluorescence in a vascular smooth muscle cell line and cultured human osteoblasts [65]. Taken together, in cells that express ER and mER, there is a possibility that isoflavones may exert their action through nuclear ER and mER signaling crosstalk, depending on the affinity and specificity of isoflavone binding to the receptors.
G protein-coupled estrogen receptor 1 (GPER1)E2 and other steroids mediate rapid cellular and physiological responses via membrane receptors. In addition to nuclear ERs, the study of the orphan G protein-coupled estrogen receptor (GPR30 or G protein-coupled ER, GPER1), a member of the 7-transmembrane G protein-coupled receptors, has been identified and clarified as an authentic estrogen receptor [66]. Upon binding to GPER1, endogenous E2 activates several intracellular signal transduction pathways. It regulates various cell functions, such as apoptosis, autophagy, proliferation, and differentiation, via a wide variety of signaling pathways, including Ras/ERK [67, 68], PI3K/Akt [68-70], receptor tyrosine kinase [58], PLC-mediated pathway [71], and cAMP-mediated pathway [58]. Furthermore, GPER1 induces rapid cellular effects, including cAMP production, intracellular calcium mobilization, and activation of kinases, such as ERK and PI3K, and ion channels and endothelial nitric oxide synthase (eNOS) [58]. The GPER1 also mediates actin polymerization through the SRC-1 and PI3K/mTORC2 pathways [70]. In addition, the GPER1 acts via the PLCβ-PKC and Rho/ROCK/LIMK/cofilin pathways to regulate F-actin cytoskeleton assembly, thereby enhancing TAZ nuclear localization and activation, leading to increased cell migration and invasion [71]. These various GPER1-activated pathways that regulate diverse cell functions indicate the profound implications of GPER1 in physiological and pathophysiological conditions. Each signal transduction pathway of GPER1 may play a distinct role in cell function. However, in cells that express ER and GPER1, there is certainly a possibility of crosstalk or squelching between receptors. Further study is necessary to clarify such crosstalk.
As for isoflavones, they can also exert their action by interacting with the GPER1. Activation of the GPER1 by isoflavones triggers cell signaling pathways and growth factor receptor crosstalk in vitro [58, 66, 72]. Our previous studies showed that isoflavones activate the PI3K/FAK/Akt/RhoA/Rac1/Cdc42 signaling pathway through GPER1, resulting in increased glial cell migration [11, 14]. Knockdown of GPER1 significantly reduced isoflavones-activated PI3K/FAK/Akt/RhoA/Rac1/Cdc42 pathways, indicating the direct action of isoflavones through GPER1 [11]. Activation of the GPER1 pathway also activates c-fos promoters and increases c-fos, cyclin A, and cyclin D1, leading to an increase in thyroid cancer cell proliferation [67]. Furthermore, isoflavones also reduce lipopolysaccharide-induced NO production by activating GPER1 in glial cells [73]. In PC12 neuronal cells, genistein exposure induces GPER1-mediated acetylcholinesterase (AChE) expression through p–CREB (cAMP) cyclic AMP responding element (CRE) binding by CREB (cAMP response element-binding protein) located on the ACHE gene promoter [74]. These results indicate that isoflavones activate a wide variety of intracellular signal transduction pathways via GPER1 in a cell subset-dependent manner.
Other possible mechanisms of isoflavone actionIsoflavones have been known to bind and modulate other nuclear and membrane receptors. Isoflavones, especially genistein and daidzein, bind to the thyroid hormone (TH: thyroxine, T4, and triiodothyronine, T3) receptors (TR) α and β, increasing the TR–cofactor interaction leading to activation of TR–mediated transcription in a dose-dependent manner in [40]. Knockdown of ER did not reduce the isoflavones-activated TR-mediated transcription, indicating the possible direct binding of isoflavones to TR [40]. Dietary soy protein isolates and isoflavone supplements in Sprague–Dawley rats increased TRβ1 protein expression in the liver. In contrast, high doses of isoflavone supplement reduced the hepatic TRα1 protein and had no significant effect on TRβ1. Although soy protein isolate did not affect total T3 and T4 levels, higher doses of isoflavones showed increased T4 levels in female rats [75]. Isoflavones also bind and activate aryl hydrocarbon receptor (AhR) to modulate AhR–responsive genes with EC50 ranges from 10 nM to 100 nM in cultured neuronal cells [76], breast cancer and liver cells [77], and GH3 cells [78]. In addition, genistein partially antagonizes the androgen receptor (AR) activity in a tissue-specific and AR target gene-specific manner in male mice (effective in prostate, testes, and brain, but not in skeletal muscles and lung). It also acts as a weak agonist in the brain and prostate tissues of mice [79]. Genistein and S-equol also upregulate pregnane X receptor (PXR)–mediated transcription and PXR–mediated CYP3A4 mRNA expression in mouse hepatocytes [80]. Isoflavones also modulate peroxisome proliferator-activated receptors (PPARs), a lipid–regulating nuclear receptors. Isoflavones decrease mRNA levels of PPARγ during adipogenesis in human primary bone marrow stromal cells [81], upregulating PPARγ–mediated transcriptional activity to enhance the adipocyte differentiation [82], and activate PPARα transcriptional activity to enhance gene expression involved in fatty acid catabolism [83]. Such a wide variety of actions indicate that these compounds probably act through several different signaling pathways.
Due to the variety of actions of isoflavones, crosstalk between receptors through isoflavone signaling has also been reported. Crosstalk between members of nuclear receptors or nuclear and membrane receptors can induce multiple possible modes of gene regulation, leading to a greater and more flexible array of transcriptional responses. In the brain, this gene regulation may affect the proliferation, migration, and differentiation of various cell types and synaptic plasticity, leading to the coordination of behavioral responses of the organism [11, 14, 84-89]. High isoflavones doses (10–100 μM), especially genistein, have been reported to act as protein tyrosine kinase inhibitors affecting synaptic plasticity [90-93]. Genistein at a dose of 40 μM reduced both the metabotrophic glutamate receptor (mGluR)–mediated inward current and the [Ca2+]i signal by (S)-3,5-dihydroxyphenylglycine (DHPG, mGluR1 agonist) in dopamine neurons [90]. The activation of mGluR1 by DHPG leads to potentiating N-methyl-D-aspartate (NMDA) current that involves tyrosine kinase activation in CA3 pyramidal cells [93]. The DHPG–induced potentiation of NMDA current was significantly reduced after applying 30 μM of genistein [93]. In addition, genistein and daidzein directly inhibited the NMDA–activated voltage-dependent current in hippocampal slice culture [94]. It has also been reported that mGluR1 interact directly with membrane ERα, which activates PLC to generate IP3 (inositol triphosphate) and stimulate Ca2+ release via the IP3R located on the endoplasmic reticulum in CA1 pyramidal cells [95-97]. This result indicates that isoflavones can mediate intracellular and membrane signaling pathways in the brain. The possible action of isoflavone in the developing brain is discussed further in the next chapter.
As discussed above, isoflavones exposure during prenatal and/or postnatal periods may affect the brain development in both humans and animal models through various signaling pathways, especially ER (Table 2). In addition to ER, TH also plays important roles in brain development through TR–dependent or –independent signaling pathways [98-100]. THs regulate various developmental processes from neuronal and glial proliferation to differentiation and neuronal migration in definitive brain regions [100, 101]. They also regulate the formation of neuronal cytoarchitecture and synaptogenesis [102]. Thus, TH deficiency can change neuronal development and function. In addition, neuronal excitability and neurotransmitter transportation are also affected [103]. Consequently, abnormal motor coordination, decreased locomotor activity, and increased anxiety has been observed in hypothyroid patients and experimental animal models [103]. As discussed above, because isoflavones may also modulate TRs action to activate the transcription of target genes, isoflavones might affect brain development through TRs, at least in part.
The representative in vivo and in vitro studies investigating the effects of isoflavones during brain development
Model of Study | Type of Isoflavones | Doses | Durations | Findings | Reference |
---|---|---|---|---|---|
In vitro: Primary culture of mouse cerebral cortex astrocytes |
Genistein, daidzein, and S-equol | 1–100 nM | DIV 7–9 | Isoflavones exert their action via the GPER1 to activate the PI3K/FAK/Akt/RhoA/Rac1/Cdc42 signaling pathway, resulting in increased glial cell migration. | [11] |
In vitro: Primary culture of mouse cerebellar cells |
S-equol | 1–100 nM | DIV 1–17 | S-equol augmented the dendrite arborization of Purkinje cells and increased astrocyte proliferation and migration through ER and GPER1. | [14] |
In vivo: Rat |
Genistein | 1 or 10 mg/kg/day | GD1–P14 | Perinatal exposure to genistein improved spatial learning and memory of rat offspring but impaired the passive avoidance learning and memory. | [32] |
In vivo: Mouse |
Genistein | 0, 5 or 300 mg/kg | GD1–P21 | The low dose of genistein exposure reduced aggressive behaviors and increased the defensive behaviors in male offspring. | [35] |
In vivo: Mouse |
Genistein | 5 and 100 μg/g of BW | GD11–P8 | Perinatal exposure to genistein increased anxiety and nitric-oxide synthase-positive cells in the amygdala in male offspring. | [36] |
In vivo: Rat |
Genistein | 5 mg/kg | GD1–P21 | Genistein exposure during gestation and lactation periods impaired spatial learning and decreased the body mass in the male offspring. | [37] |
In vitro: Primary culture of mouse cerebellar cells |
Genistein, daidzein, and S-equol | 1–100 nM | DIV 7–17 | Isoflavone-augmented estradiol mediated dendrite arborization in Purkinje cells and induced neurite outgrowth via ERα signaling pathway. | [44] |
In vitro: Mouse brain slice cultures |
Genistein and daidzein | 1 μM | DIV1-14 | Isoflavone exposure modulated the hypothalamic oxytocin neurons in the supraoptic nucleus of male mice and the paraventricular hypothalamic of female mice. | [45] |
Another note regarding isoflavone’s effects on brain development is epigenetic changes. It is widely known that dietary nutrition alters epigenetic changes in brain development that impact gene expression profiles with long-term consequences on health outcomes. Isoflavones alter DNA methylation and histone modification [104-110]. Genistein acts as a DNA methyltransferases (DNMT) inhibitor and reduces DNMT expression levels [108, 109, 111]. DNMT is a conserved set of DNA-modifying enzymes that have a central role in epigenetic gene regulation and maintain DNA methylation patterns in the developing embryo [112]. Genistein also alters methylation patterns in mouse brains after feeding with a diet containing 300 mg genistein/kg for four weeks [110]. In this chapter, we discuss the possible actions of isoflavones on various cellular events (proliferation, migration, differentiation, and synaptogenesis) during brain development.
ProliferationCell proliferation is an early step during brain development after the external form of the brain is established. Neurons and glial progenitors are produced and proliferate from the gestational to early postnatal periods [104, 113]. Soybean isoflavones affect neuron and glial proliferation. Daily intraperitoneal injection of 50 mg/kg daidzein for 13 days induces metabolic alterations (low HDL-cholesterol levels, glucose tolerance, insulin, adiponectin, and testosterone, and high leptin and endogenous E2 levels) associated with cell proliferation enhancement and reduction of apoptosis and gliosis in response to a high–fat diet in the rat hippocampus [42]. In vitro studies showed genistein and daidzein exposures with a range of 20–2000 nM significantly promoted cell proliferation and viability in fetal rat hippocampal neuronal cell lines through activation of brain-derived neurotrophic factor (BDNF) signaling via Trk receptors [114]. In addition, isoflavones exposure ranging from 1–100 nM also increased cell proliferation in neurons and glia through the crosstalk between ER and GPER1 [14, 44]. These results indicate that isoflavones affect neuron and glia proliferation via various signaling pathways, including membrane and nuclear receptors.
Cell migrationAfter neuronal progenitor cells proliferate and differentiate, they migrate from their sites of origin to different brain areas where they connect with other neurons [104]. The earliest migrating cells occupy the deepest cortical layer, whereas subsequent migrations pass through previously formed layers to form outer layers [15]. Endogenous E2 and its receptors play an essential role during neuronal migration. Although there is no neuronal deficit reported in ERα knockout mice, ERβ knockout mice showed neuronal deficits in corticogenesis at late developmental stages due to abnormalities in radial glia migration [115]. Moreover, GPER1 inhibition significantly impairs migration in ventricular-subventricular (V-SVZ)–derived cells [116]. These results suggest that ER and GPER1 modulation by isoflavones may affect neuronal migration during development.
In addition to neurons, astrocytes are most likely to migrate to their destination shortly after birth in the ventricular zone or subventricular zone (SVZ); astrocytes in the cortical gray matter migrate along with radial glia processes. In contrast, white matter astrocytes migrate along developing neuron axons [117, 118]. Our previous study showed that isoflavones accelerated astrocytes migration via GPER1 activating the PI3K/FAK/Akt/RhoA/Rac1/Cdc42 signaling pathway [11]. Thus, isoflavones also affect brain development by modulating glial cell migration.
Differentiation, axogenesis, and dendritogenesisOnce neurons migrate to reach its destination, it generally proceeds along one of two paths: It can differentiate into a mature neuron, complete with axogenesis and dendritogenesis, or it can retract through apoptosis. The differentiation processes were associated with the growth and elongation of axons and dendrites [15, 119]. Neurons are highly polarized cells with distinct subcellular compartments that consist of one or multiple dendrites that arise from the cell body and a single axon. Both extrinsic and intrinsic factors contribute to the initial polarization that influences the neuronal growth of axons and dendrites during development [119, 120]. The actin and microtubule dynamics, BDNF, nerve growth factor (NGF), neurotrophin (NT-3), neuron-glia cell adhesion molecules, drug abuse, and environmental chemical exposure affect axonal or dendritic growth, guidance, and branching [119-122]. E2 and ER effects on neuronal differentiation, axogenesis, and dendritogenesis have been widely reported. E2 regulates axonal and neuronal growth through various signaling pathways through highly expressed ER and GPER1 [123-125]. Although the cellular and molecular mechanisms that guide neuronal differentiation and morphogenesis have not yet been fully known, extensive studies over the past decade have begun to shed light on the molecular mechanisms that orchestrate growth, arborization, and dendrite and axon guidance.
Our previous study showed that isoflavone exposure induced neurite outgrowth with increased ER-responsive genes in Neuro-2A cells mainly through ERα [44]. Deletion of ERα significantly reduced genistein-induced neurite outgrowth, whereas the deletion of ERβ and GPER1 did not affect the neurite outgrowth [44]. However, in the presence of glial cells, isoflavones increased the dendrite arborization of Purkinje cells in primary cerebellar cultures through both ERα and ERβ [44]. On the other hand, we also found that daidzein and S-equol activated ER- and GPER1-mediated pathway to augment the dendrite arborization of Purkinje cells [14, 44]. Genistein exposure also induced the differentiation and neuritogenesis in human cortical neuronal cells-A1 [126] and PC12 neuronal cells through the activation of Na+/K+/2Cl– cotransporter isoform 1 (NKCC1) [127]. These results indicate that isoflavones modulate neuronal differentiation and morphogenesis through multiple signal transduction pathways in a cell subset-dependent manner.
SynaptogenesisNeurons communicate through synapses, particularly chemical synapses, which consist of the presynaptic terminal at axons and postsynaptic terminals at dendrite or soma of neurons [15, 119]. In the late 1990s, the tripartite synapse concept was introduced, in which astrocytes were shown to function as an integral element for synaptic function. They establish bidirectional communication with the neuronal components of the synapse mediated by neurotransmitters and gliotransmitters that influence neuronal and synaptic plasticity [128-131]. Synaptic plasticity plays a critical role during development, allowing the formation of precise neural connectivity. At the beginning of synaptogenesis, massive overproduction of synapse is followed by gradual reduction (pruning). As with synapse production, the timing of synapse pruning is dependent on the area of the brain in which it occurs [15, 132]. E2 is a potent modulator that regulates synaptic patterning, including an increase or decrease in the density and/or number of spines, depending on the brain region. E2 can also modulate dendritic spine formation in the adult brain with different mechanisms in the neonatal brain. E2-induced spines are transient in adulthood, whereas, in the neonate, the spine density formed within the first few days after birth will persist to adulthood [125].
Although there is no clear evidence regarding the effects of isoflavones on synaptogenesis during brain development, several in vivo and in vitro studies have been reported the synaptic plasticity modulation by isoflavones. The soy-free diet supplemented with genistein and daidzein showed increased spine density on apical dendrites of pyramidal neurons in the hippocampus CA1 region of ovariectomized rats [41]. S-equol prevented HIV–1 Tat+ cocaine synaptic loss via a selective ERβ dependent mechanism in cortical neurons [133]. As discussed above, isoflavones also modulate mGluR1 and NMDA receptor action, which play an important role in glutamatergic synapses [93, 94]. Isoflavones also affect several other neurotransmitter systems, including dopaminergic [134-137], cholinergic [138-140], GABAergic [141-144], glycinergic [145-147], and glutamatergic [93, 141, 148-150] systems. These results may indicate that isoflavone administration during synaptogenesis affects this process, although it is unclear whether administration can cause a beneficial or adverse effect.
Brain development is the most sensitive process that may be affected by various environmental insults including the endocrine-disrupting chemicals (EDCs), to affect the structure and function of the brain throughout the life [151, 152]. As discussed above, since isoflavones can affect various hormone system, isoflavones can be considered as a potent EDC [152]. Undoubtedly, the safety concerns regarding soy isoflavones and soy infant formula exposure during development have been widely expressed. A few epidemiological studies have been conducted including newborns consuming soy formula. The studies showed that soy formula did not exert adverse hormonal effects in children or affect pubertal development [56]. However, synthetic counterparts of isoflavones are increasingly viewed with caution. In animal study, perinatal exposure of genistein affects the normal development of anxiety and aggressive behavior in male mice and postnatal exposure of genistein irreversibly alter the control of reproduction, energy metabolism and abolish sexual dimorphism in a specific hypothalamic dopaminergic system [36, 153]. Regardless of the health benefits or toxic effects of soy isoflavones exposure during development, the awareness should be increased while consuming soy-based food and soy infant formula, particularly for pregnant women, infant, and hypothyroid patients.
As discussed above, Isoflavones affect morphogenesis and functional development in various brain cells through several different pathways (Table 2). Consequently, it may alter the morphological and functional organization of various brain regions. This chapter discusses the possible effect of isoflavones in several representative brain regions.
CerebellumCurrent evidence showed that cerebellum plays an essential role for motor learning and cognitive functions through the integrating and processing motor and sensory information [154-157]. The cerebellum consists of various cell types, which are integrated into a cytoarchitecture array of stripes and zones such as Purkinje cells, stellate cells, granule cells, astrocytes, basket cells, etc., [156]. The effects of isoflavones in the cerebellum have been well documented. Isoflavone exposure prevents low-potassium-dependent apoptosis of cerebellar granule cells by preventing glucose oxidation and mitochondrial coupling, reducing cytochrome c release, and preventing adenine nucleotide translocator impairment and opening of the mitochondrial permeability transition pore [158].
Our previous studies have shown that in vitro isoflavones exposure in primary cerebellar culture caused dendrite arborization of Purkinje cells through both ER and GPER1 [14, 44]. The primary cerebellar cells contain not only neurons but also glial cells. Thus, it is difficult to separate the action of isoflavones in each subset of cells. In addition, isoflavones also affects the cerebellar astrocytes proliferation and stimulates Ca2+ membrane fluxes via Na+/Ca2+ exchange [14, 44, 159]. These results indicate that isoflavones affect various cell subsets in the cerebellum.
HippocampusThe hippocampus is located in the medial regions of the telencephalon. It consists of the dentate gyrus and hippocampus proper, divided into three pyramidal subregions, including CA1 (cornu ammonis 1), CA2, and CA3 [160, 161]. It plays an essential role in informing short- and long-term memory and spatial navigation [161, 162]. Soy isoflavone action in this region has attracted attention, especially in learning and cognitive effects. The animals and cell culture models have been performed to elucidate the mechanism of isoflavones’ action in learning and memory function in the hippocampus. The perinatal exposure of genistein showed improved spatial learning and memory assessed by the MAZE test in both male and female rats in offspring [32]. Soy isoflavones induced neuroprotective effects on scopolamine-induced amnesia in mice and upregulated the phosphorylation levels of ERK, CREB, and BDNF in the hippocampus [163]. Isoflavones increased BDNF expression in ovariectomized rats fed with an isoflavone-rich diet, which led to reduced E2–induced anxiety and depressive-like behavior [164]. The intraperitoneal administration of daidzein (50 mg/kg) for 13 days enhanced adult hippocampal neurogenesis in middle-aged female mice, including increased proliferation and dendrite arborization in the dentate gyrus [165]. Genistein exposure in rat hippocampal neurons inhibited protein tyrosine kinase leading to reduced orthovanadate-mediated intracellular Ca2+ stores that modulate mGluR and NMDA receptor responses [92]. These results indicated that an appropriate amount of isoflavone administration during development might improve neural function in the hippocampus. Although administration of isoflavone supplements is not recommended, it may improve cognitive performance. Further study is required to define effective and/or toxic doses of isoflavones.
Cerebral cortexThe cerebral cortex, or neocortex, is the part of the brain’s outer surface that computes higher-level processes such as consciousness, thought, emotion, reasoning, language, sensory, and cognitive functions. Its complex organization is coordinated by cell proliferation, migration, and differentiation to assemble functional circuits [166]. Cortical layering arises during embryonic development in an inside-out manner as forebrain progenitors proliferate and generate distinct waves of interneurons and projection neurons. During the cerebral cortex development, the interplay between intrinsic intracellular signaling and extrinsic factors is critical for the proper development of the cerebral cortex [167]. The extrinsic factors, including peripheral hormones and nutrients and environmental insults such as viral infection, toxins, or EDCs, may also greatly affect cerebral cortex development [167, 168]. Isoflavones have been widely reported to have neuroprotective effects in the cerebral cortex against the rotenone [169, 170], glutamate [126], peptide beta-amyloid, cocaine [133], or cerebral ischemia [171, 172]. In addition, isoflavones exposure improved the cognitive performance of senescence-accelerated mice and increased the activities of acetylcholinesterase, superoxide dismutase, and glutathione peroxidase in the cerebral cortex [173]. Genistein ameliorates the cognitive deficits in chronic sleep deprivation-induced cognitive dysfunction mice by activating the antioxidant element nuclear erythroid-2-related factor 2 (Nrf2) and its downstream targets in the cortex [174]. Genistein exposure also activated insulin receptor substrate 1 (IRS1) and ER46 in the cerebral cortex of ovariectomized aged female rats [175]. Furthermore, soy isoflavone supplementation may enhance men’s spatial working memory and cognitive processes [176]. Although isoflavones’ mechanisms of action on cortical function have not yet been fully understood, previous studies indicate that the actions are exerted through nuclear and membrane-associated receptors [175]. In addition, protective action against oxidative stress may also be involved [173]. Although further study is required, isoflavones may contain various beneficial actions on higher brain functions controlled by the cerebral cortex. Needless to say, an appropriate dose of isoflavone supplemantaion needs to determine to avoid adverse effects.
HypothalamusThe hypothalamus is one of the most critical brain regions involved in vital physiological functions, such as feeding behavior, energy expenditure, thermoregulation, stress responsiveness, and reproduction [177, 178]. The development of hypothalamic neuronal networks is established from the prenatal period, then undergoes further extensive development and fine-tuning postnatally. In addition, the hypothalamus contains various neurons that produce several types of neuropeptides, neurohormones, and neurotransmitters [177-180]. Thus, the peripheral hormones and nutrients function as key regulators of prenatal and postnatal development of the hypothalamus.
Isoflavones have been reported to affect hypothalamic function. Isoflavone exposure as a phytoestrogen and nutritional supplement may change hormone status and neuronal plasticity. Dietary isoflavones decreased body and adipose tissue weight but increased food and water intake with a higher hypothalamic neuropeptide Y (NPY) concentration and lower plasma leptin and insulin levels in Long-Evans male rats [39]. In addition, isoflavones also increased thyroid hormone (T3), increased uncoupling protein-1 (UCP-1) mRNA levels in brown adipose tissue, increased glucose, and reduced insulin levels [39]. Perinatal exposure of daidzein (200 mg/kg) increased the ERα expression level in the arcuate nucleus and other parts of the brain, including the bed nucleus of the stria terminalis, medial preoptic, and central amygdaloid nuclei, enhanced masculinization in male mice, improved spatial learning and memory, but with higher levels anxiety and aggression behaviors could be induced [181].
Furthermore, the slice cultures of the neonatal mouse hypothalamus showed that isoflavones modulate oxytocin neurons in a brain region-specific and sexually dimorphic manner. Isoflavone exposure induced prominent vacuolation of oxytocin neurons more frequently in male than in female mice and in the supraoptic nucleus than in the paraventricular hypothalamic nucleus [45]. Dietary exposure of genistein ranges from 5–500 ppm in rats during gestational and lactational periods results in an increased volume of the calbindin D28k-labeled in the sexually dimorphic nucleus of the medial preoptic hypothalamus in male offspring [182]. Daidzein treatment in orchidectomized rats showed higher levels of corticotropin-releasing hormone and increased numbers of adrenocorticotropic hormone (ACTH) cells, plasma ACTH, and serum corticosterone concentrations [183]. Isoflavone exposure also alters hypothalamic-pituitary-adrenal axis function by reduced basal corticosterone secretion, increased mRNA mineralocorticoid and glucocorticoid receptor expression, and decreased mRNA expression of corticotrophin-releasing factor receptor 1 (CRFR1) following photoperiod alteration in hypothalamic male mice [184]. These results indicate that isoflavone exposure altered the signaling pathway in the hypothalamus and induced physiological and behavioral changes. It may be controversial whether such changes are beneficial or adverse. Further study may be required.
In conclusion, the accumulated evidence presented here indicates that isoflavones such as genistein, daidzein, and S-equol profoundly affect brain development. Isoflavones can induce and modulate various signaling pathways through ER, GPER1, or other receptors that trigger proliferation, migration, differentiation, and synaptogenesis in neurons or glial cells. Isoflavones’ positive or negative neurodevelopment effects depend on the timing, dose, and duration of isoflavones exposure/treatment. Timing plays an essential role because of the nature of brain development. The dose of isoflavones is also important due to biphasic effects. Despite an increasing number of studies, there is still a long road to travel to know the biological effects of isoflavones and their impact on brain development. Phytoestrogens isoflavones are a promising substance that may provide new ideas of physiological regulations and therapeutic interventions during brain development.
Additionally, it is uncertain whether the results from in vitro and in vivo mice models can be extrapolated to humans. The differences in development timing, structure, and degrees of protein expression in human and mouse brains might induce differential responses. We expect that more precisely targeted therapeutic approaches will be developed to eliminate the adverse effects and potentiate the beneficial effects of isoflavones, especially during development. Nevertheless, it may become a promising supplement to enhance brain development or improve cognitive function.
While reports on the health benefits of isoflavones are abundant in the literature, there are also studies regarding the concern of isoflavones’ adverse effect on growth and development and hormonal balance in the brain. The complex structure and various brain cell types led researchers to clarify the mechanism of action of isoflavones in each cell subset and brain region. Furthermore, cell-cell interactions, such as neuron-glia interactions, have led to a new understanding of migration, differentiation, and synaptic plasticity that has significant implications for knowledge of the brain development process or neurodevelopmental disorder. Further studies to build novel tools and models to assess the roles and functions of isoflavones exposure in a specific subset of cells or cell-cell crosstalk in different brain regions remain a major goal in this field. Combining in vitro, brain organoids, in vivo, and gene-editing technology becomes comprehension tools to define isoflavone’s cell and molecular basis of action. In addition, optogenetic and chemogenetic methodology will provide us with selective activation of specific cells types (neurons, astrocytes, or microglia), a situation that will facilitate new information in cell metabolism, gene expression, neuro– or glio–transmission, cell-cell interaction, and neural network following isoflavone exposure during brain development.
Conceptualization, W.A. and N.K. Methodology, W.A. Investigation, W.A. Writing original draft, W.A. Writing review and editing, W.A. and N.K. Supervision, N.K. All authors have read and agreed to the published version of the manuscript.
This work was supported in part by Grants-in-Aid for Scientific Research (nos. 18H03379 to NK, and 18J23449 to WA) from the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT).
We thank Noriko Sekie for her support in preparing the manuscript.
NK is a member of Endocrine Journal’s Editorial Board.
NK received research funding support from Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan. The remaining author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.