Introduction

According to the World Health Organization, approximately 800 chemicals worldwide are known or suspected to be capable of interfering with hormone receptors, hormone synthesis, or hormone metabolism and are defined as endocrine-disrupting chemicals (EDCs) (Bergman 2013). EDCs can have adverse effects on the morphology, physiology, growth, development, reproduction or lifespan of an organism, system, or (sub)population and may act as agonists at one hormone receptor, and as antagonists at another (ECHA, EFSA 2018). EDCs disrupt endocrine functions mainly by mimicking natural hormones like estrogens, androgens, or thyroid hormones (Bergman 2013; ECHA, EFSA 2018; Monneret 2017). They represent a heterogeneous group of natural and man-made chemicals that can be found in different environmental sources: agricultural practices (pesticides, insecticides, and fungicides), packaging (plastics, pesticides, and metals in several everyday products, including plastic bottles and food storage containers), industry (solvents, flame retardants, preservatives, emulsifiers, fracking chemicals), consumer products (household chemicals, cosmetics, flame retardants, building materials, children’s toys, electronic devices and media, cookware, water pipes), medical care (resin-based dental sealants and bonding agents, oral contraceptives, biocides, intravenous bags and tubing, disposable gloves, disinfectants), and even meat, fish, and vegetables (Bergman 2013; ECHA, EFSA 2018; ATSDR 2002; Thent et al. 2018; Vasantha et al. 2021).

Plastics and plasticizers (bisphenol A (BPA) and phthalates), pesticides, parabens, perfluoroalkyl and polyfluoroalkyl substances, dioxins, brominated flame retardants, metal, and metalloids are widespread EDCs. People are exposed to EDCs directly through oral and topical routes and indirectly via environmental pollution and food chain (Bergman 2013; ECHA, EFSA 2018; Chin et al. 2018).

The most sensitive window of EDCs exposure for the subsequent effects is during critical periods of development; fetal development and puberty are recognized as such (Bergman 2013). There are lipophilic and non-lipophilic endocrine disruptive chemicals. Non-lipophilic chemicals are transported bound to albumin and stored in the liver in humans (Alcala et al. 2019). The lipophilic chemical structure of some EDCs allows them to cross the blood–placental and blood–brain barriers that lead to prenatal fetus exposure and can be delivered through lactation to infants (Pohjanvirta and Tuomisto 1994; Ünüvar and Büyükgebiz 2012; Tang et al. 2020; Rolfo et al. 2020; Street and Bernasconi 2020). The biological effects of EDCs are carried out by their binding to various receptors in the body affecting body weight, cardiovascular system, glucose metabolism, reproductive function, neurodevelopment and the neuroendocrine system, including the skeletal system as one of the target organs (Bergman 2013; ECHA, EFSA 2018; Street and Bernasconi 2020). Several studies have shown that the action of EDCs is mediated also through the regulation of epigenetic mechanisms such as histone deacetylase activity, mitogen-activated protein kinase activity, and DNA methylation status influencing gene expression by epigenetic mechanisms (Tang et al. 2020; Jones and Baylin 2002; Anway and Skinner 2006; Huen et al. 2016; Lizunkova et al. 2022).

Finally, exposure to endocrine-disrupting chemicals has been associated with the altered regulation of skeletal formation and of the bone remodeling process (Bonefeld-Jørgensen et al. 2007; Agas et al. 2013, 2018). EDCs can act as bone toxicants with implications for bone mass accrual during adolescence (Hu et al. 2019; Carwile et al. 2022) and may induce either increased bone resorption or inhibition of bone deposition (Chin et al. 2018; Smith et al. 2017; Hwang et al. 2013; Toni et al. 2020).

Since EDCs influence biological processes associated with bone health, they may impact skeletal development and the pathogenesis of osteoporosis later in life. A number of studies have been performed to investigate the action of EDCs in the skeleton and its derivatives, but the results are inconsistent.

This narrative review aims to summarize current evidence on the effects of exposure to EDCs on bone from early gestational, at birth and long-term adverse effects.

Methodology

The MEDLINE PubMed, Mendeley, and Scopus databases were searched to identify the relevant studies. The following keywords «EDCs», «endocrine-disrupting chemicals», «bone», and «growth» were used. The selection criteria for the literature research were text availability (full text), article type (books and documents, clinical trial, meta-analysis, randomized controlled trial, review, mini-review, and systematic review), language (English), species (rodents and humans), and child age (birth-18 years), publication date range (2001–2022).

The literature review was conducted in compliance with the recommendations provided in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Figure 1 shows the PRISMA flowchart with identified studies, those included and excluded.

Fig. 1
figure 1

Selection of studies according to PRISMA flow diagram

Full size image

Overall, from the literature search, 159 articles were included in this review for analysis.

Main Mechanisms of Bone Growth and Mechanisms of Action of EDCs

Bone growth and development are a complicated hormone-dependent process (Yaglova and Yaglov 2021; Salhotra et al. 2020). Bone development begins with the replacement of mesenchymal tissue into bone tissue (Fig. 2). This process of bone formation occurs through two basic mechanisms. The first is intramembranous bone formation, when bone forms inside the mesenchymal membrane (ex., skull and jaw); this forms a woven bone, a bone form with randomly organized collagen fibers that are further remodeled into mature lamellar bone, with regular parallel collagen rings and consists of osteocytes (Yaglova and Yaglov 2021; Salhotra et al. 2020; Setiawati and Rahardjo 2018). The second is endochondral bone formation, where bone replaces hyaline cartilage (precursor to bone formation). Lamellar bone is then constantly remodeled by osteoclasts and osteoblasts. Osteoblasts are cells originating from osteoprogenitor cells, forming osteoid, that allows matrix mineralization. Osteoclasts are hemopoietic lineages cells that break down bone matrix through phagocytosis during bone formation and remodeling (Yaglova and Yaglov 2021; Salhotra et al. 2020; Setiawati and Rahardjo 2018).

Fig. 2
figure 2

The major stages of bone development. Bone development is divided into several stages: a) osteogenic cells (stem cells) differentiate into chondrocytes and form a hyaline cartilage model; b) matrix calcification; primary ossification centers with capillaries and periosteum develop from the hyaline cartilage model; c) blood vessels invade by the periosteal bud the internal cavities; spongy bone formation; d) ossification continues accompanied by remodeling inside the medullary cavity with changes occurring in the epiphyseal regions as well; these regions are referred to as a secondary ossification center; e) ossification continues, epiphyseal centers begin to calcify and hyaline cartilage persists at the epiphyseal (growth) plates between the diaphysis and epiphysis until adolescence. This allows epiphyseal growth of long bones

Full size image

Bone growth consists of two processes: interstitial epiphyseal growth and appositional growth. During interstitial epiphyseal growth (bone elongation), the growth plate with the zonal organization of endochondral ossification allows bones to lengthen without epiphyseal growth plates enlarging zones. On the epiphyseal side, cartilage is formed; on the diaphyseal side, cartilage is ossified, and the diaphysis then grows in length. During appositional growth, osteoclasts resorb old bone that lines the medullary cavity, while osteoblasts, via intramembranous ossification, produce new bone tissue (Yaglova and Yaglov 2021; Salhotra et al. 2020; Setiawati and Rahardjo 2018).

Bone remodeling is a key dynamic process that requires osteoclasts, derived from the hematopoietic lineage and responsible for bone resorption, osteoblasts derived from the mesenchymal lineage and responsible for bone formation, and osteocytes derived from terminally differentiated osteoblasts permanently entombed in the bone matrix. (Bergman 2013; Chin et al. 2018; Salhotra et al. 2020; Setiawati and Rahardjo 2018; Florencio-Silva et al. 2015). The bone remodeling process is regulated by several hormones and by local factors, among these: the parathyroid hormone (Yaglova and Yaglov 2021; Florencio-Silva et al. 2015; Hodsman et al. 2003), thyroid hormones (triiodothyronine, thyroxine, calcitonin) (Yaglova and Yaglov 2021), bone morphogenetic proteins (Agas et al. 2018; Salhotra et al. 2020), prostaglandins (Agas et al. 2013, 2018; Toni et al. 2020; Sabbieti et al. 2009, 2008), testosterone, estrogen and progesterone, somatotropin and somatostatin, and adrenal cortex hormones (Yaglova and Yaglov 2021).

Growth hormone (GH) is a hormone produced by the pituitary somatotroph cells and nonpituitary tissues that act directly and indirectly on the epiphyseal plates of long bones and also has a metabolic function. One of the growth hormone predominant actions is hepatic synthesis and secretion of insulin growth factor-1 (IGF-1), and its synthesis by osteoblasts and chondrocytes, suggesting local synthesis contributes also to the regulation of statural growth (LeRoith 2008). The IGF system is important in mammalian cell growth, proliferation, and development and consists of two IGF ligands, insulin-like growth factor-1 (IGF-1) and insulin-like growth factor-2 (IGF-2), two IGF receptors (IGF-1 receptor (IGF-1R) and IGF-2 receptor (IGF-2R)), and six IGF binding proteins (IGFBPs) and several binding protein proteases that regulate IGF bioactivity (Renehan et al. 2006; Bach et al. 2005). GH increases the expression through the signaling pathways of both IGF-1 and IGFBP-3 besides through its own receptor, and by regulating IGFBP-4 and -5 of alkaline phosphatase and type I procollagen mRNA expression favoring bone formation (Blum et al. 2018). IGFBPs modulate IGF-I bioavailability regulating in turn skeletal, homeostasis (Kawai and Rosen 2012). For instance, these proteins may inhibit the actions of IGF-1 by preventing access to the IGF-1R in bone (Kawai and Rosen 2012). In addition, bone formation is regulated through the activation of the IGF-1-GSK-3β-RUNX2-mediated signaling pathway and runt-related transcription factor-2 (RUNX2) regulation and release (Toni et al. 2020). IGF-2 shares biochemical and biological properties with IGF-1 and is important in skeletal development, but its synthesis is not GH-dependent (Giustina et al. 2008).

Several studies showed that microRNAs (miRNA) have an important role in the regulation of the GH/IGF-1 axis and the IGF system, bone homeostasis, and, in particular, osteoclast formation (Scheller et al. 2011; Catellani et al. 2022). Estrogen and xenoestrogens can modulate miRNA expression, some pri-miRNA genes contain estrogen response elements (EREs) in their promoters and can be regulated by the activity of estrogen receptors (ERα and ERβ) (Sabry et al. 2019). Low-dose EDCs can upregulate enzyme aromatase (CYP19) mRNA, increase CYP19 activity, significantly increase CYP19-induced biosynthesis of the breast carcinogen 17β-E2, and increase ERα-positive breast cell proliferation (Williams and Darbre 2019).

Steroids are implicated in bone growth, mineralization, and remodeling. Steroids can suppress autophagy to cause osteoblasts and osteocytes apoptosis, enhance osteoclast activity, and decrease osteoprotegerin synthesis, and they play the main role in bone mineral metabolism for the attainment of peak bone mass during puberty and maintaining bone mass and strength during adulthood (Yaglova and Yaglov 2021; Mitra 2011).

EDCs can bind nuclear receptors (NRs), especially the androgen and estrogen receptors (ERs), the glucocorticoid receptor (GR), the mineralocorticoid receptor, the thyroid hormone receptors, the retinoid X receptor (RXR) and the peroxisome proliferator-activated receptors (PPARs) (Agas et al. 2018; Swedenborg et al. 2009; Alavian-Ghavanini and Rüegg 2018). The nuclear receptors together with glucocorticoid receptors and the aryl hydrocarbon receptor (AhR) regulate DNA methylation at a specific gene locus through interactions with DNA methyltransferases (DNMTs) (Agas et al. 2018; Kouzmenko et al. 2010). The direct or indirect EDC–receptor interactions lead to different epigenetic alterations such as changes in DNA methylation and histone acetylation (Agas et al. 2018; Swedenborg et al. 2009). EDCs may alter DNA methylation patterns in an ER-dependent way (Alavian-Ghavanini and Rüegg 2018; Kitraki et al. 2015) or through interaction with the AhR (Alavian-Ghavanini and Rüegg 2018; Romagnolo et al. 2016). The EDC vinclozolin, a fungicide, and androgen receptor antagonist were shown to induce a marked increase in DNA methyltransferases mRNA expression in vivo through an AR-mediated pathway (Alavian-Ghavanini and Rüegg 2018). The receptor activator of NF-kappaB (RANK) is critical for differentiation and activation of osteoclasts and RUNX2 is important as a key transcription factor for osteoblastogenesis (Hwang et al. 2013; Xin et al. 2018; Shoucri et al. 2017). By reducing the RANK ligand or its receptor availability, EDCs may induce a net loss in RANK/RANKL signaling, ultimately impairing osteoclast activity. Also, EDCs bound ERα capable to stimulate and repress of RUNX2 activity in human and rodent osteoblasts (Xin et al. 2018). Histone post-translational modifications, including methylation and acetylation, are epigenetic marks capable of influencing gene expression through alterations in chromatin accessibility, for bone development and homeostasis, but studies on how EDCs affect histone post-translational modifications are yet limited (Xin et al. 2018; Shoucri et al. 2017). Estrogen deficiency can be caused by inflammation and oxidative stress through increased TNF-α, IL-6, TLR-2, TLR-4, changes in NF-κB, and increased ROS production which all lead to increased bone resorption (Mohamad et al. 2020).

Any disruption of molecular mediators of bone formation and remodeling has an effect on bone strength, architecture, and density parameters such as bone mineral density (BMD) and bone mineral content (BMC), bending force, hardness, plasticity of bones, and so on (Agas et al. 2013, 2018; Carwile et al. 2022; Smith et al. 2017; Lejonklou et al. 2016; Miettinen et al. 2005; Hermsen et al. 2008; Finnilä et al. 2010). As EDCs may affect the modeling and remodeling of bone structure, they can contribute to an achievement of lower bone mass which is the main determinant of early osteoporosis in offspring (Miki et al. 2016).

In Vitro Evidence of Adverse Effects of EDCs on Bone

Several experiments have shown how single EDCs can cause disruption in bone cells causing changes in metabolism, and morphological, and functional changes in bone maturation (Bergman 2013). Xenoestrogens exert an effect by binding α and β estrogen receptors on osteoblasts and osteoclasts. EDCs as xenoestrogens influence bone structure by disrupting differentiation, downstream signaling pathways of receptors, ultimately remodeling, and modifying the equilibrium between bone resorption and formation (Fig. 3) (Agas et al. 2018; Yaglova and Yaglov 2021; Salhotra et al. 2020; Bolli et al. 2008; Ronis et al. 2020; García-Recio et al. 2022; Talia et al. 2021).

Fig. 3
figure 3

In vitro effects of endocrine-disrupting chemicals on bone. Endocrine-disrupting chemicals inhibit the differentiation and maturation of osteoblasts and osteocytes. The arrows indicate that an EDC can increase or decrease processes having effects on osteogenic cells. Single endocrine disrupting chemicals via binding estrogen receptors can inhibit the expression of RUNX2, OXS, Wnt, BMP and ALP, decrease TRAP and M-CSF, can activate aryl-hydrocarbon receptors and PPARγ and modify cytoplasmic fibroblast growth factor-2 translocation into the nucleus. Arrows identify the influence of endocrine-disrupting chemicals. Abbreviations: EDCs — endocrine disrupting chemicals; PFAS — perfluoroalkyl substances; PCB — polychlorinated biphenyl; PPARγ — peroxisome proliferator-activated receptor gamma; OSX — Osterix; BMP — bone morphogenetic protein; RUNX2 — Runtrelated transcription factor 2; Wnt — Wnt/b-catenin signaling pathway; ALP — alkaline phosphatase; FGF-2 — fibroblast growth factor-2; AhR — aryl-hydrocarbon receptor; TRAP — tartrate resistant acid phosphatase; M-CSF — macrophage colony-stimulating factor

Full size image

Bisphenol A, in particular, can bind to estrogen receptors present in bone and cartilage (Agas et al. 2018; Spelsberg et al. 1999; Richmond et al. 2000). Specifically, in vitro experiments showed that Bisphenol A (BPA) treatment induces apoptosis and inhibits osteoclast maturation (Toni et al. 2020; Miki et al. 2016), but studies are yet limited. BPA is structurally similar to endogenous 17β-oestradiol (E2), and via binding both estrogen receptors (ER) α and β (Bolli et al. 2008) can inhibit the expression of RUNX2 and Osterix (OSX), which activates genes during differentiation of pre-osteoblasts into mature osteoblasts and osteocytes, and the Wnt/b-catenin signaling pathway in the MC3T3-E1 mouse cell line. RUNX2 and OSX are key for differentiation as they reduce the expression of the markers of differentiation alkaline phosphatase (ALP), collagen-1 (COL-1), osteocalcin (OCN), bone morphogenetic protein 2 (BMP-2), and bone morphogenetic protein 7 (BMP-7) with an overall negative effect not only on the maturation of osteoblasts but also on their functional capacity (Toni et al. 2020; García-Recio et al. 2022). Furthermore, García-Recio et al. (2022) described a negative effect of BPA on bone matrix formation; in particular, they described changes in the expression of CD54 and CD80, where reduced expression was associated with osteoblast differentiation and elevated expression with suppression of osteoblasts differentiation (García-Recio et al. 2022). BPA had a negative effect also on alkaline phosphatase activity and bone matrix formation, and on osteoblasts mineralization (García-Recio et al. 2022). These same authors, in a further in vitro study, showed that BPA treatment caused a significant reduction in the expression of COL-1, ALP, OCN, RUNX2, OSX, BMP-2, and BMP-7, modifying human osteoblast differentiation, antigen, and gene expression in primary cultures resulting in an inhibition of growth (García-Recio et al. 2022).

Bisphenol A may impact the IGF system also by promoting the expression of IGF-1R and inducing IGF-1 expression (Talia et al. 2021; Murata and Kang 2018). In addition, bisphenol A can activate the human steroid xenobiotic receptor (SXR) and cause xenobiotic receptor (SXR)-mediated transcription on the SXR response elements by recruiting SRC-1 to the ligand-binding domain of SXR in osteoblast-like MG-63 cells, in fetal bone tissue (Miki et al. 2016; Ichikawa et al. 2006). A further study by Urano et al. (2009) defined a nucleotide polymorphism in the SXR gene that was significantly associated with high bone mineral density in primary osteoblasts (Urano et al. 2009).

Dioxin has been described in an in vitro study to be able to suppress bone deposition by inhibiting osteoblastogenesis through the dependent aryl hydrocarbon receptor (AhR), the extracellular signal`lated kinases activation, and by Wnt/b-catenin signaling suppression (Iqbal et al. 2015) affecting the IGF system also by inhibiting IGF-1, IGF-2, and IGF-1R mRNA expression (Talia et al. 2021; Liu et al. 1992; Croutch et al. 2005a).

Polychlorinated biphenyl (PCB)-126 exposure was associated with decreased serum osteocalcin, no change in bone resorption markers, and with a significant reduction in serum calcium levels associated with an increase in parathyroid hormone, which represents an appropriate response to the calcium reduction. Osteoblastogenesis suppression was hypothesized to be blunted by the AhR antagonist in the presence of PCB-126 (Ronis et al. 2020). PCBs have been shown to be able to disrupt the estrogen-relatedγ39 receptor that negatively regulates BMP-2-induced osteoblast differentiation and bone formation (Hu et al. 2019; Kiess et al. 2021) and can contribute to an increased risk of osteoporosis in the exposed population.

Perfluoroalkyl substances (PFAS) and phthalates can affect bone homeostasis through the activation of peroxisome proliferator-activated receptor gamma (PPARγ) (Yamamoto et al. 2015; Szilagyi et al. 2020), by suppressing osteoblast formation and acting as androgen receptor antagonists possibly inhibiting androgen-mediated osteoblastogenesis (Smith et al. 2017; Kjeldsen and Bonefeld-Jørgensen 2013; Engel et al. 2017) or by directly intercalating into bone (Bogdanska et al. 2014; Koskela et al. 2017; Pérez et al. 2013). In human endometrial stromal cells, PFOS can decrease the IGFBP-1 transcript (Yang et al. 2016).

Di-2-ethylhexyl phthalate (DEHP) treatment (100 µM) has been described to decrease the expression of RUNX2 and of its transcriptional coactivator with PDZ-binding motif (TAZ) at the protein level, leading to a decreased collagen matrix secretion and mineralization, besides decreased ALP mRNA expression suggesting impaired osteoblast differentiation and maturation (Bhat et al. 2013). Benzyl butyl phthalate (BBP) and di-n-butyl phthalate (DBP) are estrogen-mimicking compounds that modify the cytoplasmic fibroblast growth factor-2 (FGF-2) translocation into the nucleus in rat osteoblasts (Menghi et al. 2001). In particular, the exposure to 10–9 to10−6 M of BBP and DBP induces FGF-2 perinuclear accumulation with subsequent translocation into the nucleus (Menghi et al. 2001) that may modify gene transcription (Hu et al. 2019; LeRoith 2008; Menghi et al. 2001; Jiang et al. 2015).

In vitro studies have reported that alkylphenols influence bone architecture through the downregulation of critical factors, such as tartrate-resistant acid phosphatase (TRAP) and macrophage colony-stimulating factor (M-CSF) leading to changes in bone matrix organization, in osteoblast and osteoclast differentiation by inhibiting osteoclast formation and affecting stromal cells and osteoclast-supporting cells (Hagiwara et al. 2008).

Butyltin trichloride is used as a polyvinyl chloride stabilizer, wood preservative, and as biocide for cooling systems. In vitro exposure to tributyltin chloride (TBT) (10–10 to 10–7) suppressed the expression levels of alkaline phosphatase and osteocalcin, markers of bone differentiation, and interfered with calcium signaling and deposition in rat calvarial osteoblast-like (ROB) cells. Thus, the observed delayed ossification of the fetal skeleton could be ascribed to the changes induced by TBT on important differentiating markers and signaling cascades (Tsukamoto et al. 2004).

Finally, EDCs have immunomodulatory effects also and may induce either increased bone resorption or inhibit bone deposition and lead to disruption of skeletal homeostatic preservation (Agas et al. 2018; Hu et al. 2019; Yaglova and Yaglov 2021; Ronis et al. 2020; Yang et al. 2016; Hwang et al. 2017; Ramajayam et al. 2007).

Table 1 summarizes the findings of the effects of EDCs in bone in in vitro studies.

Table 1 Current findings relative to the effects of single endocrine disruptors in vitro
Full size table

In Vivo Evidence of Adverse Effects of EDCs on Bone (Rodent Models)

Many studies have shown the effects of EDCs on bone development and structure in rodents (Table 2).

Table 2 In vivo evidence of adverse effects of EDCs on bone in rodent models
Full size table

Bisphenol A is one of the most common and most studied EDCs as it is transferred across the placenta of rat dams to the fetus (Moors et al. 2006). Chemek and Nevoral (2019) reported that low-dose BPA exposure potentially alters the epigenetic pattern in the testes, including increased IGF-2 gene methylation and decreased protein lysine acetylation (Mao et al. 2015; Chemek and Nevoral 2019). The in vivo murine studies have shown that BPA inhibits osteoblast proliferation and induces apoptosis, ALP synthesis, calcium nodule formation, and the expression of RUNX2, OSX, and β catenin genes, which are markers of bone differentiation (Rolfo et al. 2020; García-Recio et al. 2022; Zhang et al. 2021). Kim et al. (2001) observed delayed ossification of fetal rat skeletons after administration by gavage of the BPA chemical to mated females from days 1 to 20 of gestation at high doses such as 100, 300, and 1,000 mg/kg/day (daily dose volume 10 ml/kg body weight) (Kim et al. 2001). However, other studies showed that lower doses (0.1 and 1% w/w) of this chemical in diet, for 5 months, prevented bone loss in female mice lacking the aromatase gene CYP19 (Toda et al. 2002), and thus lacking estrogens, confirming the pro-estrogenic effect of BPA. In particular, in that study, the BPA diet completely reversed the loss of femoral trabecular bone observed in the CYP19 knockout mice, by increasing femoral bone mineral density in a dose-dependent manner (1% w/w BPA administration was more effective than 0.1% w/w) without altering femoral bone density in wild-type mice (Toda et al. 2002). In rodents, bisphenol A exposure (50 μg/kg BPA/day diluted in 0.1% ethanol for 7 days) during early pregnancy led to growth retardation (Müller et al. 2018). Low-dose bisphenol A exposure (0.5 µg/kg BW/day) in pregnant rats reduced bone length and size in the male rat offspring, determining specifically femoral bone shortening and decreased trabecular area (Lind et al. 2017; Pelch et al. 2012) and higher exposure doses (10–25 mg/kg/day) showed sexual dimorphic changes on bone geometrics and increased femoral length in both male and female offspring (Hu et al. 2019; Lejonklou et al. 2016). The most recent study by Wang et al. (2021) showed that exposure to 10 μg/kg per day of BPA throughout pregnancy delayed bone development and bone mass was reduced in the female rat offspring, and these effects were mediated by the suppression of the osteogenic function via the ERβ/HDAC5/TGFβ signaling pathway (Wang et al. 2021).

Several researchers have reported negative associations between prenatal exposure to dioxin and dioxin metabolites administration and IGF-2 mRNA expression and protein levels (Wu et al. 2004; Ding et al. 2018; Ma et al. 2015; Zhang et al. 2019), but Wang et al. (2011) reported an increase in both mRNA expression and protein (Wang et al. 2021, 2011). Prenatal exposure to 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) in mice during the preimplantation stage was related with decreased IGF-2 mRNA expression in a whole-body analysis of embryos (Wu et al. 2004). Interestingly, a transgenerational study reported that both in F1 male mice exposed prenatally to a dioxin single dose (10 μg/kg body weight) on gestational day 15 (GD15) (direct prenatal dioxin exposure) and in F3 male descendants (ancestral exposure), decreased placental IGF-2 content was observed (Ding et al. 2018). Prenatal exposure to TCDD (0.8 μg/kg body weight/day, from gestational day 8 to 14) also decreased both the transcript and protein levels of IGF-2 in adult livers of F1 rats (Ma et al. 2015). This is opposite to the findings of Wang et al. (Wang et al. 2011) (10 μg/kg body weight single TCDD exposure on gestational day 10) who described increased IGF-2 in fetal livers (Wang et al. 2011). Different levels of exposure and timing besides different approaches to analyses may have contributed to these different results. Korkalainen et al. (2009) reported that dioxin exposure diminished bone growth through aryl hydrocarbon receptor (AhR) in rodents (Korkalainen et al. 2009; Turan 2021). In adult rats exposed to dioxin, both chronically and with a single large dose, circulating IGF-1 levels were found to be significantly decreased (Croutch et al. 2005b; Linden et al. 2014). Moreover, dioxin treatment induced hepatic IGFBP-1 mRNA expression (Minami et al. 2008). TCDD treatment in mice induced IGFBP-6 in the thymus and bone marrow, at both the transcriptional and protein levels, with the finding being confirmed in vitro using a thymoma cell line (EL-4) (Park et al. 2001, 2003). Administration of 0.03, 0.1, 0.3, or 1 l g/kg TCDD at different time points in rodents significantly decreased tibia and femur length and cross-sectional areas of the cortex and determined a smaller endosteal and periosteal circumference in both these long bones (Hu et al. 2019; Miettinen et al. 2005).

Few in vivo studies have explored the effects associated with polychlorinated biphenyl administration. Rats prenatally exposed to PCB-126 present reduced circulating levels of both IGF-1 and IGF-2 (Talia et al. 2021; Ahmed et al. 2018). Skeletal toxicity after exposure to PCB-126 is a result of disruption of calcium homeostasis and of the GH-IGF-1 axis and involves once again direct AhR-mediated effects on bone formation and decreases long bone length and diameter with increasing trabecular thickness and volume (Ronis et al. 2020).

Recent studies have reported that exposure to PFOS, administered orally to rodents at 1, 5, or 10 mg/kg by gavage, was associated with the downregulation of IGF-1 mRNA expression levels, key for longitudinal bone growth, and of IGF-1R mRNA expression in the liver and testes (Ohlsson et al. 2009; Wan et al. 2011). Exposure to 0.3 mg·kg−1·day−1 PFOA mixed with food in pregnant rats led to increased femoral periosteal area, with a parallel decrease in the mineral density of tibias in offspring. Researchers reported also that PFOA decreased osteoblast activity secondary to increased vitamin D levels and altered vitamin D-responsive genes (Turan 2021; Nisio et al. 2020), accumulated in bone and consequently exerted persistent deleterious effects on bone geometry and mineral density (Hu et al. 2019; Koskela et al. 2017).

In vivo studies have also reported an increased number of skeletal malformations, such as deformity of the thoracic vertebrae and fusion of the vertebral arches, in rat fetuses exposed to phthalates and phthalate metabolites (Agas et al. 2018; Sabbieti et al. 2008; Ema et al. 1993). More specifically, DBP exposure upregulated mRNA expression for IGF-1, IGF-2, IGF-1R, and IGF binding protein 5 (IGFBP5) genes and for matrix metalloproteinase (Agas et al. 2018; Talia et al. 2021; Bowman et al. 2005). Liu et al. (2018) observed elevated G protein-coupled receptor 54 (GPR54) and gonadotropin-releasing hormone (GnRH) protein expression in the hypothalamus and increased release of Kiss-1 mRNA, allowing for stimulation of GnRH secretion with increased IGF-1 and GH in rats treated with 500 mg/kg/d and 1000 mg/kg/d DEHP (Liu et al. 2018). Negative associations have been consistently reported among DEHP and the percentage of methylated CpG sites in IGF-2R in rodents’ primordial germ cells (Li et al. 2014).

There are few studies on the effects of alkylphenols’ on bones. Hagiwara et al. (2008) reported that exposure to 0.1 mg/kg p-n-nonylphenol (NP) or p-n-octylphenol (OP) treatment accelerated the ossification of the sternum bone in the fetuses, as compared to the control group of rodents. The NP and OP were further confirmed to inhibit the formation of osteoclasts in vitro. But alkylphenols did not affect the proliferation and differentiation of rat calvarial osteoblast-like cells (Hagiwara et al. 2008).

Transplacental tributyltin chloride exposure at the doses of 10 or 20 mg/kg, from 0 to 19 gestation days in rodents, was associated with reduced ossification in the fetuses. The 20 mg/kg concentration caused misaligned sternum or sternoschisis with delayed ossification of the fetal pelvic girdle, skull, and limbs, while doses of 10 mg/kg reduced bone formation of the sternum (Talia et al. 2021; Tsukamoto et al. 2004; Adeeko et al. 2003; Sarpa et al. 2007). Injection of 1 mg of TBT or MBT/kg body weight to pregnant mice was associated with delayed ossification of the supraoccipital bone, metacarpals, and metatarsals (Tsukamoto et al. 2004). In a further study, the rodent TBT-treated group exhibited a decreased osteogenic capacity of the stromal stem cells and increased adipogenic differentiation capacities (Kirchner et al. 2010). High concentrations of TBT resulted in decreased ALP mRNA levels (culture day 6) and decreased OCN, which led to bone homeostasis impairment (Tsukamoto et al. 2004; Koskela et al. 2012). However, the findings by Carvalho Rangel Resgala et al.(2019) were different as these authors reported that TBT treatment (100 ng/kg/day TBT for 15 days) led to higher bone mineral density and observed increased osteocalcin, alkaline phosphatase, and urinary deoxipyridinolines excretion ratios, as markers of bone resorption and on microCT scans or with scanning electron microscopy detected lesions in the spinal bones (Resgala et al. 2019).

Current Evidence of Adverse Effects of EDCs on Growth and Bone in Humans in Childhood

Prenatal and early postnatal periods have been shown to be the most vulnerable periods of life for endocrine disruption since organs and systems are developing and changing more intensively (Huen et al. 2016; Smith et al. 2017; Anway et al. 2005). Prenatal exposure to EDCs may affect the fetus directly or dysregulate placental functions, insulin, glucocorticoid, estrogenic, and thyroid hormonal pathways, epigenetic regulation of gene expression, and inflammatory pathways (Jones and Baylin 2002). There are few epidemiological studies that have focused on the impact of epigenetic changes owing to prenatal exposures to EDCs within the context of growth, metabolic programming, and bone health. Kim et al. (2018) showed that in utero exposure to EDCs can perturb normal biological functions via the epigenome (Kim et al. 2018). In particular, they showed that dichlorodiphenyltrichloroethane (DDTs), including p,p′-DDT, can influence the methylation of two key imprinted genes, i.e., IGF-2/H19, in the placenta, showing a significant association between a 1% increase in placental long interspersed nuclear element-1 (LINE-1) methylation with a 0.19 cm decrease in birth length (β =  − 0.19; 95% CI =  − 0.35 to − 0.03). However, placental methylation of neither IGF-2 nor H19 was significantly associated with the birth outcome parameters measured (Kim et al. 2018). Although this study showed many limitations as the limited number of subjects in the population study, the heterogeneous nature of cells in the placenta, and potential changes in the exposure to POPs throughout gestation that were not taken into consideration (Kim et al. 2018), the findings are consistent with another study which observed hypomethylation of LINE-1, IGF-2, and PPARA after exposure to increasing phthalate concentrations (Montrose et al. 2018), and with the Goodrich et al. (2016) ELEMENT study where phthalate exposure in the third trimester of pregnancy altered H19 methylation in children. H19 is important for the regulation of body composition and growth (Goodrich et al. 2016). With regard to bisphenol A, exposure during human embryonic body differentiation was confirmed to be associated with downregulation of IGF-1 and IGF-1R mRNA expression with subsequent changes in fetal development (Huang et al. 2017).

Body mass index (BMI) is an another factor that has an effect on bone health as it can affect bones through a variety of mechanisms, including body weight, fat volume, bone formation/resorption, and proinflammatory cytokines together with bone marrow microenvironment (Hou et al. 2020). Extremely low BMI and oppositely obesity have a direct negative influence on bone marrow adipose tissue with subsequent significant clinical effects on bone density and strength. Additional visceral fat may contribute to bone loss. Possibly because both osteoblasts and fat cells are derived from the mesenchymal stem cells in the bone marrow, an increase in adipogenic differentiation could potentially result in a reduction of osteogenic differentiation of these mesenchymal stem cells (Hou et al. 2020). Different studies showed that prenatal exposure to EDCs is associated with increased BMI in childhood which can indirectly affect bone health (Harley et al. 2017).

Figure 4 summarizes the adverse effects exercised by single EDC during fetal life on mechanisms involved with future bone health.

Fig. 4
figure 4

Adverse effects of endocrine-disrupting chemicals during pregnancy on bone health. Circles identify subgroups of effects all influencing bone formation/growth in the fetus. Arrows in the circles show the influence of endocrine-disrupting chemicals on processes. Abbreviations: EDCs — endocrine disrupting chemicals; IGF-1 — insulin growth factor-1; IGF-1R — insulin growth factor-1 receptor; miRNA — microRNA

Full size image

Bisphenol

High maternal bisphenol S urine concentrations have been directly associated with a larger head circumference and higher birth weight and lower risk of being small for gestational age at birth, especially when exposure occurs mainly during the first trimester of pregnancy (Sol et al.2021). Similar findings were reported in a further study that described an association between high bisphenol A concentrations in children’s urine and increased BMI z-score at 4 years of age, whereas prenatal exposure (high BPA concentrations in mother’s urine) was negatively associated with BMI and measures of adiposity in girls and positively in boys, but with no other growth parameters (Vafeiadi et al. 2015). Recent Snijder et al. and Ish et al. studies, which used a job exposure matrix to identify EDCs, reported no relationships between bisphenols and fetal or infant growth (Snijder et al. 2012; Ish et al. 2022).

Specifically for bone, a 10-year population-based prospective cohort study reported that maternal serum bisphenol S levels in the first trimester of gestation were associated with lower childhood bone mineral density (BMD) and area-adjusted bone mineral content (aBMC) in 10-year-old children. However, there were no associations between maternal serum total and specific bisphenol concentrations and bone measures at 6 years of age, and no associations between BPA and bone measurements at both 6 and 10 years (Zwol-Janssens et al. 2020).

Dioxins and Organochlorine Pesticides

Exposure to dioxins and dioxin-like compounds is known to be associated with increased IGF-1 circulating levels in children and decreased IGF-1 serum concentrations in adults (Talia et al. 2021). Placental weight and birth weight are positively associated with IGF-1 and IGFBP-3 levels at birth and later in childhood (Talia et al. 2021). Interestingly, studies of prenatal exposure to high levels of polychlorinated biphenyls showed low IGFBP-3 levels in 8-year-old girls and high IGF-1 levels in 5-year-old boys (Su et al. 2010, 2015). These results were partially confirmed in further longitudinal cohort study, in which pre- and early postnatal exposure to dioxin-like compounds were associated with increased IGF-1 levels in 3-month-old infants, accelerated longitudinal growth, and weight gain (Wohlfahrt-Veje et al. 2014). But in the more recent CHECK cohort study by Kim et al. (2018), no significant association was observed between p,p′‑dichloro‑diphenyl dichloroethylene (p,p′DDE) or PCB levels and placental LINE-1, a retrotransposon and marker of methylation (Kim et al. 2018), and a further significant positive association was described with placental IGF-2 methylation levels.

Elevated prenatal exposure to Σ19-organochlorine pesticides (Σ19OCPs) was associated significantly with increased IGF-2 methylation and decreased H19 methylation in the placenta. The CHECK cohort that consisted of 109 placental samples and maternal blood during delivery found a positive association between exposure to Σ6DDTs and birth length and a suggestive inverse association between Σ6DDTs and LINE-1 methylation (β =  − 0.63; 95% CI =  − 1.36 to 0.11) in the placenta. This study did not show an association between placental DNA methylation and birth weight but defined a significant relationship with reduced birth length (p < 0.05) (Kim et al. 2018).

Iszatt et al. (2015) described an increase in infant growth associated with prenatal p,p′-DDE and a decrease associated with postnatal polychlorinated biphenyl 153 (PCB-153) exposure (Iszatt et al. 2015). A similar relationship was described by Cock et al. also, between low exposure to p,p´-DDE and growth during the first year of life (Iszatt et al. 2015). For PCB-153 exposure, changes in length were significant during the first 3 months of postnatal life in both sexes, but not thereafter (Cock et al. 2014).

In the prospective study by Vafeiadi et al. (RHEA, 2015), consisting of a cohort of 689 mother–child pairs, where prenatal and neonatal serum samples were obtained and a 4-year observation period was carried out for children, hexachlorobenzene (HCB) concentrations were found to be positively but not significantly associated with rapid growth during the first 6 months of life (Vafeiadi et al. 2015). HCB exposure was associated with a higher BMI z-score that showed no significant association with HCB, p,p′-DDE, and PCBs exposure at the end of follow-up (Vafeiadi et al. 2015). A further study confirmed the association of HCB exposure with BMI z-score in preschool children (Karlsen et al. 2017). Previous studies, however, found no associations between PCB or p,p′-DDE exposures and length and growth in infants independent of maternal race and child sex (Pan et al. 2010).

Height was significantly different depending on PCDD/Fs exposure levels at 2 (p = 0.0029) and 5 years of age (p = 0.0276) in boys and girls (Su et al. 2010). Considering the male and female subgroups, height, weight, chronological age, and bone age were significantly different depending on the exposure levels to PCDD/Fs at 2 years in females, whereas no differences were reported in males. Children who were exposed to PCDD/Fs had higher levels of T3, and IGF-1 at 2 years of age being significantly higher in females than in males. Furthermore, at age 5 yr, the most exposed had higher serum IGF-I, prevailing in the males with respect to females (Ish et al. 2022). A further study did not detect any effect of exposure to dioxin on children’s anthropometric measurements (Delvaux et al. 2014).

Phthalates and Their Metabolites

Since in vivo studies have shown that exposure to phthalates has an effect on longitudinal bone growth and bone remodeling, we examined the studies related with prenatal and postnatal exposure to phthalates. Boas et al. (2010) described a negative correlation with both IGF-1 and IGFBP-3 serum concentrations (Boas et al. 2010). Tsai et al. (2016) described a negative association of measurements of exposure to DEHP exposure with IGF-1, height and weight, but no association with IGFBP-3 (Tsai et al. 2016) whereas a negative association among some phthalate metabolites (MEP and MMP) and IGF-1 and IGFBP-3 levels was confirmed further by a recent study by (Huang et al. 2020).

One study proved that when a phthalate metabolite mixture was considered, the effect of exposure during pregnancy on abdominal circumference and femur length was greater than single phthalate exposure. All phthalate metabolites were modestly and inversely associated with birth size in single-pollutant analyses. It was also shown that a one-quartile increase in exposure in the pregnancy-averaged phthalate metabolite mixture was inversely associated with birth length (p = 0.14). At birth, male infants were 0.5 cm shorter for each one-quartile increase compared to girls (Stevens et al. 2022). This is in agreement with the findings by Botton et al. (2016) where MBzP levels during the third trimester of pregnancy contributed the most to the effects of the phthalate metabolite mixture on birth length (Botton et al. 2016). Snijder et al. (2012) also found that fetal length was negatively associated with phthalate and pesticide exposure (Snijder et al. 2012) whereas a very recent study found no associations with growth (Ish et al. 2022).

Several studies have evaluated the associations of exposure to phthalate metabolites with BMI. Cock et al. described that low exposure was associated with higher BMI values, and high exposure levels were associated with greater head circumference (Cock et al. 2014). DEHP exposure, especially MECPP and MEHHP, possibly affected the growth parameters during the first year of life (Cock et al. 2014).

A study that included outcomes over a 20-year period found a negative association between changes in height z-score at 2 years of age, and maternal phthalate metabolite concentrations for MHBP and MCiOP in maternal serum samples at 18 and 34 weeks of gestation. Before 2 years of age, participants who had detectable serum MHBP or MCiOP levels grew, respectively, on average of 0.73 cm and 0.66 cm less compared to those children who had undetectable levels. Between 2 and 10 years of age, significant positive relationships were detected between height z-score and higher exposure to MBzP and MECPP. Participants with detectable levels of MBzP were 1.19 cm taller than participants with undetectable levels of MBzP. Participants in the middle and upper tertiles of exposure to MECPP were 0.99 cm taller than those in the lowest tertile, respectively. Weak positive associations between some of the phthalate metabolites and height z-score were detected during childhood. However, up to 20 years of age, associations between maternal phthalates and dual-energy X-ray absorptiometry outcomes were not defined (Berman et al. 2021).

Maternal exposure to MCNP and MCPP measured in urine has been associated with increased fetal femoral length in the second and third trimesters and with increased length at birth. After birth, MBzP exposure was positively associated with height in the two first years of life; however, after 5 years of follow-up, maternal urinary concentrations of MEP were positively associated with weight growth velocity from 2 years onwards, with weight at 3 and 5 years and BMI at 5 years of age (Botton et al. 2016). Interestingly, the most recent study by Ish et al. (2022) did not find evidence of associations between any exposure to EDCs and femur length (Ish et al. 2022). The ten-year follow-up study by van Zwol-Janssens et al. (2020) described no association between exposure to phthalates and BMD and adjusted BMC both in 6- and 10-year-old children (Zwol-Janssens et al. 2020).

Perfluoroalkyl Substances (PFAS)

Exposure to PFAS has also been associated with lower levels of IGF-1 and sex hormones in 2292 children aged 6–9 years (Lopez-Espinosa et al. 2016).

Maisonet et al. (2012) found that girls born to mothers with higher serum PFAS concentrations during pregnancy were smaller at birth and heavier at 20 months with respect to controls. And the PFHxS metabolite had a statistically significant association with reduced birth length (Maisonet et al. 2012). Consistently, a statistically significant association between maternal PFHxS and a decreased birth length was found by Lee et al. (2013). Although the authors described that this effect disappeared after adjustment for gestational age in the crude model, it became significant again after adjusting for both gestational and maternal age (Lee et al. 2013). A lower birth length was found also in the HOME (2021) study (Buckley et al. 2021), but not in the study by Cock et al. (Wohlfahrt-Veje et al. 2014). In a two-year follow-up study, Shoaff et al.(2018) did not detect any significant associations between child growth and PFAS exposure, except for the association between PFHxS and rapid growth (Shoaff et al. 2018). Birth length and head circumference seemed to have no relationships with PFAAs exposure either and no associations were found between maternal PFAA levels and childhood growth during the first 5 years of life (Gyllenhammar et al. 2018). Louis et al. (2018), however, described in Hispanic women, a negative association between PBDE #28 levels in mothers’ serum and child length. In black mothers, serum PFOA, PFOSA, and PBDE #153 levels were negatively associated with child length, and in White women, a positive association between PFOA, PFOSA, and child length was further detected. In this study, PFDA was the only EDC associated with upper arm length, and some PFASs, specifically N-MeFOSA, PFDA, PFHpA, PFHxS, PFNA, PFOA, and PFUnDA, were found to be associated with reductions (cm) in upper thigh length. These findings apparently were most predominant among caucasic versus other ethnic neonates (Buck Louis et al. 2018).

The NHANES study reported that PFAS serum levels and urinary phthalate biomarkers may be associated with reduced bone mineral density in 12–19-year-old adolescents. In this study, by using Bayesian kernel machine regression, the authors estimated that higher concentrations of PFOA, PFOS, MBP, and MiBP were related with lower aBMD Z-scores in boys, whereas in girls higher concentrations of PFOA and MiBP were related with higher aBMD Z-scores (Smith et al. 2017). Similarly, Project Viva showed that higher PFAS plasma concentrations were associated with lower aBMD z-scores in covariate-adjusted models, with the strongest associations holding for PFOA, PFOS, and PFDA. The strongest significant association was found between PFAS exposure and aBMD z-score in girls (Cluett et al. 2019).

The HOME study associated maternal serum PFOA, PFNA, and PFAS mixture concentrations with lower BMC z-scores in children in particular at the hip and forearm. The associations of PFOA in maternal serum concentrations with bone outcome z-scores were generally stronger among males than females (Buckley et al. 2021). Similar results were observed in the ODENSE (2022) study, where children with higher prenatal PFAS exposure and with higher PFAS serum concentrations at 18 months of age had lower BMC and BMD Z-scores at 7 years of age. All associations between PFAS and BMC were significant, whereas only PFNA and PFDA were significantly associated with reduced BMD. Furthermore, adjustment for child sex, height Z-score at age 7 years, and maternal education analyses showed that the effect of 18 months of exposure appeared to affect more bone mineral content than bone mineral density (Højsager et al. 2022). Perfluorononanoic acid also had an association with reduced bone mineral density in the Khalil et al. (2018) study in school-age children, but this association was statistically insignificant after adjusting for multiple testing (Khalil et al. 2018).

The ALSPAC study followed up girls for 17 years established that prenatal PFOS, PFOA, PFHxS, and PFNA concentrations in maternal serum collected during pregnancy were negatively correlated with bone size and mass in a confounder-adjusted analysis for age, gestational age, and maternal education. Interestingly, hip BMD had no significant association with prenatal PFAS exposure. Conversely, PFNA was positively associated with area-adjusted bone mineral content (Jeddy et al. 2018).

Polycyclic Aromatic Hydrocarbons (PAHs)

Although the lack of epidemiological studies does not give the possibility to describe PAHs impact on the IGF system, a recent study reported a negative association between exposure to PAHs and IGF-1 plasma levels and height (Zeng et al. 2020).

Mothers occupationally exposed to PAHs and phthalates showed significantly lower fetal weight growth rates (average decline in SD per gestational week: 0.01660 for PAHs and 0.01691 for phthalates) compared with non-exposed mothers (Snijder et al. 2012). PAHs have been reported to be associated negatively with children’s birth length; children with higher exposure were shorter compared to lower exposed children (Jedrychowski et al. 2004). At variance with these findings, Perera et al. found associations between PAHs, birth weight (β = –0.08, p = 0.02), and head circumference (β = –0.02, p = 0.04) in African-Americans and no associations with other growth measurements (Perera et al. 2003).

Table 3 and table 4 summarize the effects of endocrine-disrupting chemical exposure on children’s growth and bone health. Table 3 reports the known effects of mixtures and Table 4 reports the effects of single chemicals.

Table 3 Effects of prenatal exposure to endocrine-disrupting chemical mixtures on children’s growth
Full size table
Table 4 Effects of endocrine disruptive chemical prenatal exposure on children’s bone health
Full size table

Conclusions and Further Perspectives

Endocrine-disrupting chemicals are commonly found in our environment. Studies suggest that they have an effect on growth and on bone growth, homeostasis, and remodeling, and thus can negatively affect bone health. This prompts the necessity for further knowledge and awareness.

This review provides an overview of the current knowledge on how prenatal endocrine-disrupting chemicals exposure can affect bone growth, length, structure, mineral content, and development. Many in vitro and in vivo studies have shown how single EDCs can negatively affect these processes but the mechanisms are not fully understood. Furthermore, human data are yet scarce but there are many studies that do suggest that prenatal exposure affects longitudinal growth, BMI, and mineral content, thus suggesting that this will have consequences into adult life.

Results have been sometimes inconsistent across studies possibly due to differences in sample size, study design, study populations, life stage, sex, ethnicity, data collecting, data analysis approaches, and/or strategies for attaining data on exposure.

Many of the potential exposure–response relationships described here have not been entirely understood and need further investigation on how EDCs affect bone tissue.

Further studies on the effects of endocrine-disrupting chemicals in children would be relevant and needed to determine their influence on linear growth and skeletal disorders to preserve bone health and prevent diseases and complications in adulthood such as osteoporosis. Longitudinal studies starting from pregnancy would be of particular relevance to study the effects of exposure throughout life. Future research studies will need to understand how different levels of exposure to EDCs can influence bone development before and after birth, considering that the effect of mixtures on BMD is still controversial and poorly understood although more adherent to real life. Another required field of investigation is endocrine disruptor-driven epigenetic changes and their effect on bone remodeling processes and pathways required for growth plate growth, as well as risk factor for patients having growth hormone deficiency, being this latter of key importance for longitudinal bone growth and bone quality. There is a need for preventing exposure to EDCs during pregnancy and early life and for further understanding of the underlying molecular mechanisms to develop possible prevention and therapeutic strategies.