Air pollution, environmental chemicals, and smoking may trigger vitamin D deficiency: Evidence and potential mechanisms

impacts of environmental exposures on the levels of main VD metabolites, and (2) credible engaged mechanisms in VDD because of those exposures. To summarize explanations for these unclear topics, we conducted the present review, using relevant keywords in the PubMed database, to investigate the adverse effects of exposure to air pollution, some environmental che- micals, and smoking on the VD metabolism, and incorporate relevant potential pathways disrupting VD endocrine system (VDES) leading to VDD. Air pollution may lead to the reduction of VD cutaneous production either directly by blocking ultraviolet B photons or indirectly by decreasing outdoor activity. Heavy metals may reduce VD serum levels by increasing renal tubular dysfunction, as well as downregulating the transcription of cyto- chrome P450 mixed-function oxidases (CYPs). Endocrine-disrupting chemicals (EDCs) may inhibit the activity and expression of CYPs, and indirectly cause VDD through weight gain and dysregulation of thyroid hormone, parathyroid hormone, and calcium homeostasis. Smoking through several pathways decreases serum 25(OH)D and 1,25(OH)2D levels, VD intake from diet, and the cutaneous production of VD through skin aging. In sum- mary, disturbance in the cutaneous production of cholecalciferol, decreased intestinal intake of VD, the modulation of genes involved in VD homeostasis, and decreased local production of calcitriol in target tissues are the most likely mechanisms that involve in decreasing the serum VD levels.


Introduction
Vitamin D is provided either through exposure to sunlight, which is the main source or through intake from diet, which is supplementary (Hansdottir et al., 2008;Holick, 2006). To sustain the human health, the presence of a sufficient level of VD is compulsory. Insufficiency and deficiency of VD have been reported in different climates and geographical regions. It is estimated that one billion individuals in the world suffer from VD deficiency (VDD) in all age groups (Holick, 2007;Holick and Chen, 2008). Accordingly, VDD has been identified as one of the important public health concerns globally, and particularly in the Middle East region (Palacios and Gonzalez, 2014). Due to the underlying role of this member of the steroid hormone family in biochemical processes, VD has received a growing attention during the last decades. Along with the understanding of VD metabolism, its related disorders and therapeutic capabilities have been listed as one of the foremost priorities for public health.
Although there is no precise estimation of the global burden of VD deficiency, an array of investigations conducted by Lips and Van Schoor provided a global overview of VD status. Based on studies performed on the national status of VD around the globe, it turned out that VD deficiency and insufficiency is very common in most countries, mainly in the Middle East, India, China, and Mongolia. Lips and Van Schoor also found that at risk groups for VDD include young children with low birth weight, pregnant women, non-western immigrants and older individuals (Lips, 2010;Lips and van Schoor, 2011;Van Schoor and Lips, 2017). Studies have reported that VDD (< 30 nmol/l) prevalence in infants has been 86% in Iran, 61% in India, and 51% in Turkey (Roth et al., 2018). As VDD is considered as a contributing factor to human diseases, studying the interactions between VD metabolism and environmental agents seems to be noteworthy, especially to understand links between environmental exposures and the pathogenesis of chronic disorders. Although adequate exposure to inducible radiation of the sun and adequate intake by dietary sources are key factors in maintaining an acceptable level of VD, there is persuasive evidence demonstrating that there are other environmental factors that interfere in VD endocrine systems (VDES). Therefore, studying the impacts of environmental pollutants on VD metabolism seems necessary.
In 2016, 95% of the world's population lived in areas where ambient particulate matter < 2.5 μm (PM 2.5 ) levels exceeded the World Health Organization guideline value of 10 μg/m 3 (Shaddick et al., 2018). On the other hand, industries disperse thousands of chemicals into environment. Further, considerable fraction of populations are exposed to tobacco smoke. All of these may disrupt biochemical pathways and cause detrimental consequences like VD deficiency (VDD). In addition, we are in need of mechanistic studies to declare how exposure to these exposures disrupt biochemical events and participate in the pathogenesis of disorders through decreasing the levels of active metabolites of VD. In this study, we aimed to provide a mechanistic overview of the aftermaths of exposure to air pollution, some environmental chemicals and tobacco smoke related to dysfunctional VDES accompanied with the declined serum levels of two main metabolites of VD including 25-hydroxyvitamin D (25(OH)D, Calcifediol) and 1α,25-dihydroxyvitamin D (1,25(OH) 2 D, Calcitriol).

Material and methods
Exposure to air pollution, some environmental chemicals, and smoking, which are implicated directly or indirectly in disrupting VDES and decreasing the serum levels of two main metabolites of VD were investigated in this review. Briefly, the PubMed Database search was performed on May 3nd 2018 using a combination of the following keywords: ("Air Pollution", "Heavy Metals", "Endocrine Disruptors", "Halogenated Hydrocarbons", "Bisphenol A" (BPA), "Phthalate",  (D 3 ). The upper process shows the synthesis of vitamin D 2 from ergosterol with the help of UVB radiation in plants and fungi. The lower process displays the photosynthesis of vitamin D 3 from 7-dehydrocholesterol (7-DHC) in keratinocytes-the skin cells responsible for the cutaneous production of vitamin D 3 . "Smoking", "Tobacco Use") AND ("Vitamin D"). Moreover, the reference lists from included and relevant review articles were manually searched. In order to include all available studies till the day, we used Google Scholar as an additional source to fill in any probable gap. Search strategy is presented in online Supplementary File 1.
There are numerous studies that have proposed the exposure to tobacco smoke as one of lifestyle risk factors for VDD; however, we only included those which had specifically focused on the relationship between smoking and the VD levels. Finally, 74 studies were included in the review. Tables 1 and 2 provide a summary of the included articles in this review.

Synthesis of VD
As a member of the fat-soluble group of secosteroids (steroid molecules with a broken ring), VD plays a pivotal role in the metabolism of calcium and phosphate, as well as, other biological processes. As it is shown in the Fig. 1, there are two main types of VD including D 2 (ergocalciferol) and D 3 (cholecalciferol). Both of them could be provided through exogenous sources. The former is originated from the ultraviolet B (UVB) irradiation of the ergosterol in plants and fungi, and the latter is found in the dietary sources, particularly in oily fish, and is produced in the skin as well (the main and endogenous source) (Holick, 2006). Vitamin D 3 can be synthesized in the skin without any enzymatic reaction. It is produced from 7-dehydrocholesterol (7-DHC) via a twostep process in which UV light (wavelengths of 290-315 nm) radiation from the sun photolyzes 7-DHC (provitamin D 3 ) to precholecalciferol (previtamin D 3 ) in the plasma membrane of human skin keratinocytes. Afterward, in a heat-dependent process, previtamin D 3 instantly converts to vitamin D 3 (Holick, 2007). Skin pigmentation and UVB intensity are two significant factors in the rate of vitamin D 3 formation (Holick et al., 1980;MacLaughlin et al., 1982). Thus, all factors which may limit sun exposure (whether periodically or locally) can affect the production of the VD. D 2 , as the first VD analog, is structurally different from D 3 , resulting in its lower affinity to VD-binding protein (DBP), which in turn leads to its faster clearance from the circulation, limited conversion to the active form, and altering catabolism by the enzyme 24-hydroxylase (CYP24A1) (Hollis, 1984;Houghton and Vieth, 2006). As a result, D 2 supplementation does not result in high serum levels of VD as much as D 3 (Viljakainen et al., 2006).

VD metabolism
There are three main steps in the VD metabolism consisting of 25-hydroxylation, 1α-hydroxylation, and 24-hydroxylation performed by cytochrome P450 mixed-function oxidases (CYPs), enzymes located in endoplasmic reticulum (ER) (e.g., CYP2R1) or in the mitochondria (e.g., CYP27A1, CYP27B1, and CYP24A1) (Bikle, 2014). The mentioned types of VD (D 2 and D 3 ) are physiologically inactive and therefore in need of two sequential hydroxylations to turn into the biologically active form (Fig. 2). After transportation of VD by DBP to the liver (the major source of 25(OH)D production), it is hydroxylated by a number of CYPs with 25 hydroxylase activity resulting in the formation of 25(OH)D (Christakos et al., 2010). Since a homozygous mutation of the CYP2R1 gene has been observed in the patients with low circulating levels of 25(OH)D and classic symptoms of VDD, it could be suggested that CYP2R1 is the key enzyme required for 25-hydroxylation of VD (Cheng et al., 2004).
In the second step, 25(OH)D, as the major circulating form of VD and the standard indicator of VD status, is transported to the kidney for the next hydroxylation. 1α-hydroxylase (CYP27B1) is the most abundant enzyme which has the capability to hydroxylate the position of carbon 1 of the A ring leading to the generation of the hormonally active form of VD, 1,25(OH) 2 D. It should be pointed out that the kidney is the major, if not the sole, source of circulating levels of 1,25(OH) 2 D (Bikle, 2014). As it can be seen in the Fig. 2, low levels of calcium and phosphate, as well as the elevated level of parathyroid hormone (PTH) resulted from hypocalcemia stimulate the production of 1,25(OH) 2 D (Murayama et al., 1999). As a result, increased levels of VD and calcium cause diminishing expression of PTH and VD receptor (VDR). VDR as the principal mediator is responsible for cellular effects of VD through natural ligand with 1,25(OH) 2 D. Concisely, it can be said that 1,25(OH) 2 D levels are downregulated via its own production and the level of serum calcium, while PTH induces its renal production (Bikle, 2014;Christakos et al., 2010).
The last but not least step in the metabolism of VD is that it is responsible for the hydroxylation of VD through the CYP24A1 which has 24-hydroxylase activity and results in the water-soluble biologically inactive calcitroic acid. Although the enzyme introduce hydroxyl into both 25(OH)D and 1,25(OH) 2 D, the preferred substrate for 24(OH)ase is 1,25(OH) 2 D (Omdahl et al., 2002;Shinki et al., 1992). Accordingly, the control of 1,25(OH) 2 D level within tissues can be taken into account as the foremost function of 24(OH)ase. This process can occur through the catabolism of 1,25(OH) 2 D to 1,24,25(OH) 3 D or by decreasing the reservoir of 25(OH)D via its catabolism to 24,25(OH) 2 D (Christakos et al., 2010). This process may prevent VD intoxication which could come from more than normal levels of 25(OH) and 1,25(OH) 2 D (Bikle, 2014).

Functions and disorders related to VD
Thanks to the extensive studies on the VD metabolism during the last decades, there is growing evidence asserting the significant role of VD inadequacy in the pathogenesis of a wide array of chronic diseases. Even though there is no unanimity on the optimal level of serum 25-hydroxyvitamin D (S-25(OH)D), there is an implicit agreement regarding the definition of VDD. Previous studies reported that S-25(OH) D < 20 ng/ml is considered as VD deficiency, 20 to 30 ng/ml is insufficiency, 30 to 60 ng/ml is a sufficient level, and > 150 ng/ml is intoxication (Holick, 2007(Holick, , 2009. Disturbance in the metabolism of the active form of VD (1,25(OH) 2 D) is considered as a cause of VDD (Clements et al., 1992;Holick, 2007). The conversion of the storage form of VD to its active form stimulates the absorption of intestinal calcium and phosphate, which in turn contribute to the bone health. This function of VD in the homeostasis of calcium and phosphate plays an essential role in musculoskeletal health as its deficiency could lead to osteoporosis, osteomalacia, decreased bone mineral density, and increased risk of fragility fractures (Christodoulou et al., 2013). Potentially, the circulating levels of 1,25(OH) 2 D affect functions of all tissues and cells possessing VDR, particularly active T and B lymphocytes, breast, colon, prostate, and heart skeletal muscle (Holick, 2006;Mathieu and Adorini, 2002;Nagpal et al., 2005). Moreover, it has been observed that aside from kidney, a broad range of tissues, mainly brain (neurons and glial cells), lung (lung epithelial cells), colon (epithelial cells and parasympathetic ganglia), prostate (normal (NP96-5), benign prostatic hyperplasia (BPH) cells), skin (keratinocyte), and breast (normal lobules and ducts tissue), as well as macrophages, have the capability to locally express 25-hydroxyvitamin D-1α-hydroxylase, and produce 1,25(OH) 2 D. Therefore, they regulate the expression of genes involving in their functions, especially cell proliferation and differentiation (Bikle, 2011;Eyles et al., 2005;Hansdottir et al., 2008;Schwartz et al., 1998;Townsend et al., 2005;Zehnder et al., 2001). Thus, in addition to musculoskeletal consequences of VD inadequacy, increasing experimental and epidemiological evidence depicted that VDD can be involved in the pathogenesis and/or progression of chronic illnesses, such as many common cancers (colon, breast, and prostate) autoimmune diseases (multiple sclerosis and type 1 diabetes mellitus, inflammatory bowel disease, systemic lupus erythematosus), hypertension, cardiovascular diseases, and . VD is mainly synthesized in the skin cells and can be found in dietary sources as well. Through a two-stage hydroxylation process, VD firstly converts to 25(OH)D and then is activated to 1,25(OH) 2 D. Liver and kidney are two principal organs where 25(OH)D and 1,25(OH) 2 D are produced, respectively. With the help of DBP and circulating system, 1,25(OH) 2 D is transferred into the target organs possessing VD receptor (VDR) and results in autocrine and endocrine actions. S.E. Mousavi et al. Environment International 122 (2019) 67-90 neurological disorders (Plum and DeLuca, 2010;Wang et al., 2017). Based on the mentioned broad role of VD in human health and adverse health effects derived from its deficiency, numerous efforts have been made to evaluate its analogs in attenuation and treatment of all aforementioned disorders. Antiproliferation, prodifferentiation, immunomodulation, antiangiogenesis, and apoptosis are among the most protective functions that can be exerted by VD supplementations and analogs (Adorini and Penna, 2008;Holick, 2006). Such noncalcemic mechanisms are recruited by VD analogs, especially 1,25(OH) 2 D to treat various types of disorders (Bikle, 2014). It has been shown that VD analogs can be taken into account as preventative and therapeutic agents by which the progression and risk of cancers (Deeb et al., 2007), autoimmune diseases (Adorini and Penna, 2008), heart disorders, and musculoskeletal illnesses would be reduced to a great extent (Plum and DeLuca, 2010;Souberbielle et al., 2010).

Environmental contaminants
Owing to a plethora of studies performed on the detrimental health effects of anthropogenic pollutants, solid evidence has been provided during the last years associating VDD with exposure to environmental contaminants including endocrine-disrupting chemicals (EDCs). Interestingly, both EDCs exposure and VDD are heavily implicated in adverse developmental, neurological, cardiovascular, metabolic and immune effects in humans (Schug et al., 2011;Wang et al., 2017).
During the last decades, there has been an overwhelming urge among researchers to investigate the possible health impacts posed by EDCs and potential EDCs, which are anthropogenic toxic materials in our environment, food, and consumer products. Pesticides, metals, bisphenol A (BPA), phthalates, polycyclic aromatic hydrocarbons (PAHs), and polyhalogenated compounds are on the top of the most damaging EDCs, which damage human health (Frye et al., 2012). According to the definition of U.S. Environmental Protection Agency (EPA), an endocrine-disrupting chemical is "an exogenous agent that interferes with synthesis, secretion, transport, metabolism, binding action, or elimination of natural blood-borne hormones that are present in the body and are responsible for homeostasis, reproduction, and developmental process" (Diamanti-Kandarakis et al., 2009). On the other hand, VD should be considered as a prohormone that can be developed to its hormonal form (1,25(OH) 2 D) through the VD endocrine system (VDES) depicted in Fig. 2. All tissues possessing the enzyme 1α-hydroxylase, especially the kidney, could be taken into consideration as endocrine glands which convert 25(OH)D to the hormonal form of 1,25(OH) 2 D. It is noteworthy that investigators have shown > 36 target organs with VDR as a receptor for steroid hormone 1,25(OH) 2 D (Bouillon et al., 1995;Norman, 2008).
EDCs have been shown to affect the biosynthesis pathways of steroid hormone and thyroid hormone (TH) levels (Boas et al., 2012;Sanderson, 2006). Since VD is similar to steroid hormones in terms of molecular structure and biological functions and its nuclear receptor belong to the same superfamily of proteins as steroid and thyroid hormone receptors (Norman, 2008), it is highly likely that EDCs affect VDES. Consequently, it seems plausible to state that all endocrine systems in general and VDES in particular, can be disrupted by environmental chemical exposures. Such a deductive argument originates from this fact that EDCs activate the same receptors and signaling pathways as hormones and act at low concentrations. They are subjects to the same biological regulatory systems as hormones (Schug et al., 2011).
As indoor and outdoor air pollutants, tobacco smoke, heavy metals and persistent organic pollutants (POPs) can behave like EDCs, we intend to incorporate all human and experimental studies that reported rising prevalence of VDD and disruption of the VDES because of exposure to these mentioned risk factors. To bring convincing associations, interactions between the chemicals and VDES will be mechanistically explained as far as studies provide enough evidence.

Air pollutants
As mentioned above, sun exposure plays an underlying part in VDES. Therefore, every factor that has the potency to restrict the absorption of sunlight by the skin would be capable of dysfunction in VD metabolism fundamentally. Epidemiological and clinical studies represent an association between atmospheric pollution and hypovitaminosis D (Agarwal et al., 2002;Baïz et al., 2012;Calderón-Garcidueñas et al., 2015;Calderón-Garcidueñas et al., 2013;Hosseinpanah et al., 2010;Kelishadi et al., 2014;Manicourt and Devogelaer, 2008). In megacities with a high level of air pollutants, especially ozone, particulate matter (PM), and sulfur dioxide, the UVB photons would be effectively absorbed by these pollutants, diminishing the cutaneous photosynthesis of previtamin D 3 (Gorham et al., 1989;Holick, 1995;Hoseinzadeh et al., 2018) . This could contribute to the modulation of immune responses subsequent to a substantial decrease in the level of VD (Mousavi et al., 2017). In an interesting cross-sectional study, Manicourt and colleagues evaluated interactions between exposure to tropospheric ozone and sunlight affecting the percentage of subjects with VD inadequacy in two groups of postmenopausal women from Brussels or the countryside engaged in outdoor activities. Participants did not differ in mean ages, body mass indices, and VD intakes; nevertheless, urban residents were exposed to ozone levels 3-fold higher than rural inhabitants. Notwithstanding that the urban inhabitants benefited from higher mean sun exposure index (SEI) than rural ones (113 vs. 87; P < 0.001), they had a higher prevalence of 25(OH)D < 30 ng/ml (84% vs. 38%). It should be pointed out that after adjusting for SEI, 25(OH)D was two times higher in rural residents, and after adjusting for 25(OH)D, SEI was three times higher in urban residents (Manicourt and Devogelaer, 2008). In comparison to the countryside, residents in the urban areas with a high level of air pollution may be discouraged from engaging outdoor activities leading to high prevalence of VDD (Bailey et al., 2012). Overall, results of Manicourt et al. should be interpreted with caution as the ozone concentrations are typically higher in rural areas compared to urban areas as nitrogen oxides originated from exhaust emissions scavenge the ozone and reduce it in the urban areas (McConnell et al., 2006). However, some other local conditions in Brussels might be responsible for the reported 3-fold higher ozone concentrations compared with rural areas.
In a recent study performed in Mexico, it was reported that exposure to high concentration of PM 2.5 could lead to S-25(OH)D < 30 ng/ml in 87% of Mexico City metropolitan area normal weight children compared to controls (Calderón-Garcidueñas et al., 2015). In Mexico City, an environment characterized by ozone and PM levels above standard values and decreased UV light, children had an insufficient VD intake and spent less time outdoors than controls (P < 0.001) (Calderón-Garcidueñas et al., 2013). Since it has been shown that air pollution is capable of reducing UV light by 20% on average (Acosta and Evans, 2000) in Mexico City, it could be inferred that in the setting of high concentration of air pollutants, children have gotten less UV light exposure. Furthermore, spending less time outdoor synergistically reduce VD availability. As a matter of fact, an atmospheric situation with a high level of pollutants and dust causes high haze in which the visibility considerably reduces because of the decreased penetration of sunlight. It is proposed that sunlight-blocking haze lessens the exposure to sun's VD inducing radiations. Agarwal et al. compared the VD level of two comparable groups of 9 to 24-month-old toddlers lived in two different areas of India in terms of the level of atmospheric pollution. The mean serum concentration 25(OH)D of children from Mori Gate, well-known for a high level of air pollution, was 12.4 ng/ml, compared with 27.1 ng/ml in those who lived in Gurgaon area with less polluted air (P < 0.001). Moreover, the mean haze score in the Mori Gate area (2.1) was significantly lower (P < 0.05) than in the Gurgaon area (2.7), indicating less solar UVB reaching the ground in Mori Gate (Agarwal et al., 2002).
Through the study of a cohort group in Paris, the relationship between gestational exposure to two urban air pollutants (PM 10 and NO 2 ) and 25(OH)D cord blood serum level was inquired in 375 mother-child pairs. Baïz and colleagues found that maternal exposure to the urban pollutants, in particular during late pregnancy, may contribute to lower VD levels in offspring. After adjustment, log-transformed 25(OH)D declined by 0.15 units (P = 0.05) and 0.41 units (P = 0.04) for an increase of 10 ng/m 3 in NO 2 and PM 10 pregnancy levels, respectively (Baïz et al., 2012). Studies performed in Tehran and Isfahan, two high air-polluted metropolises in Iran, also revealed that there is a negative association between air quality and VD status (Hosseinpanah et al., 2010;Kelishadi et al., 2014). Kelishadi and colleagues conducted a cross-sectional study of 100 children aged 4-10 years from various areas of Isfahan city with different levels of air pollution in order to investigate the association between air quality index (AQI) and serum concentration of 25(OH)D. After adjustment for age, gender, body mass index (BMI), diet and pattern of physical activity, by a multiple linear regression, they demonstrated an inverse association between AQI and 25(OH)D, which properly justified the high prevalence of VDD among Isfahanian children. According to the data related to the dietary habits obtained through validated questionnaires, they reported that dietary intake of VD was not sufficiently low to explain the very low level of S-25(OH)D (Kelishadi et al., 2014). In a similar study, Hosseinpanah et al. compared two groups of healthy women from Tehran (higher-polluted city) and Qazvin (lower-polluted city) concerning VD status. The mean ± SD of S-25(OH)D level was markedly higher in Qazvinian women (18 ± 11) compared with Tehranians (13 ± 7) (P-value < 0.001). The prevalence of severe VDD (25(OH)D < 10 ng/ml) and 25(OH)D of 10-20 ng/ml among women in the high-polluted city were 36% and 54%, respectively, while they were estimated 31% and 32% in women from Qazvin (Hosseinpanah et al., 2010).

Heavy metals
Due to the ubiquity of toxic metals in the environment and their role as putative endocrine disruptors, it is necessary to study those metals that can negatively affect steroid hormones, particularly VD. In this regard, various studies have been demonstrated that heavy metals, especially cadmium (Cd) and lead (Pb), as well as radioactive metals, in particular, uranium (U) and 137 cesium ( 137 Cs) are capable of interfering hormonal systems including VDES (Dyer, 2007).

Cadmium and lead.
Since the kidney and liver as the principal sources of the two main VD metabolites are included target organs for Cd and Pb toxicity, diverse studies have been performed in this domain in order to determine probable associations and mechanisms.
Cd is an environmentally widespread toxic metal with a long biological half-time in organs. The kidney, liver, bone, and cardiovascular system are the main targets for Cd toxicity (Roe, 1993). According to a series of human studies performed in the Cd-polluted areas of Japan, exposure to environmental Cd causes perturbations in VDES (Nogawa et al., 1990;Nogawa et al., 1987;Tsuritani et al., 1992). The most severe form of Cd intoxication can be seen in patients with a bone disease called itai-itai. These patients suffer from a painful bone damage characterized by a combination of osteomalacia and osteoporosis, two VDD-associated disorders. Nogawa et al. found decreased serum 1,25(OH) 2 D levels in itai-itai disease patients and Cd-exposed subjects with kidney damage compared with non-exposed individuals (Nogawa et al., 1987). As mentioned, the kidney is the main producer of circulating 1,25(OH) 2 D; therefore, disturbance in its function can essentially influence biological responses throughout the body via the deficiency of the active form of VD. Reducing activity of enzyme involved in hydroxylating 25(OH)D to 1,25(OH) 2 D has been proposed as a possible mechanism (Alfvén et al., 2000). The first toxic manifestation reported in all human exposure to cadmium is renal tubular dysfunction. There is growing evidence suggesting that kidney damage is the most likely pathway by which an internal VDD could be occurred, which in turn results in bone damage through a reduced level of calcium (Kido et al., 1989). Tsuritani and colleagues suggested that this Cd-exposure-derived consequence could lead to the impaired activation of VD as well as an increase in the PTH level (Tsuritani et al., 1992). VDD is certainly present in individuals with very low levels of total S-25(OH)D accompanied by hyperparathyroidism, hypocalcemia, or low bone mineral density (BMD) (Powe et al., 2013). Among Cd-exposed women, a significant correlation between serum 1,25(OH) 2 D and PTH levels with indices of renal tubular dysfunction has been revealed (Tsuritani et al., 1992). In an experimental study, 30 three-old-months female rats have exposed to cadmium chloride in a dose of 50 mg Cd/l in drinking water for 3 months. Results provided clear evidence that long-term exposure to cadmium chlorine would inhibit renal 1-α-hydroxylase activity, which resulted in reduction of the serum 1,25(OH) 2 D 3 level (Youness et al., 2012). In a similar study, Brzoska and colleagues depicted that low lifetime exposure to CdCl 2 can affect the metabolism of calciotropic hormones, such as 1,25(OH) 2 D in female rats. Exposure to 1 mg Cd/l in drinking water for 24 months, resulted in a significant depression in the serum 25(OH)D and 1,25(OH) 2 D by 50 and 31%, respectively. The 55% decrease in the kidney mitochondrial fraction in the exposed rats bolstered this theory that disorders in VD metabolism caused by exposure to Cd are related to the kidney functional status (Brzóska and Moniuszko-Jakoniuk, 2005). It seems that the decreased renal production of 1,25(OH) 2 D is proportional to the progression of Cd-induced renal tubular dysfunction (Keiko and Minoru, 1991). In agreement with information provided by human and animal studies, WHO suggests that Cd-induced bone effects may be mediated by renal tubular dysfunction, which in turn leads to reduced activation of VD and decreased calcium absorption from the gut (Organization, 2000;Roe, 1993).
The renal damage due to environmental Cd exposure can be lead to elevated urinary levels of DBP as well (Kasuya, 2000). DBP plays a critical role at VDES as it binds the principal VD metabolites (25(OH)D and 1,25(OH) 2 D). DBP acts to bind and transport VD throughout the body. DBP binds 88% of serum 25(OH)D and 85% of serum 1,25(OH) 2 D. DBP is also important for the renal activation of 25(OH)D to 1,25(OH) 2 D (White and Cooke, 2000). Uchida et al. investigated the relationship between urinary DBP levels and markers of renal tubular dysfunction in the residents of a high Cd-polluted area (Cd group) compared with people settled in the low Cd-polluted region (reference group). They observed significantly higher levels of urinary DBP among highly Cd-exposed individuals. Both Cd and reference groups showed remarkable positive correlations between urinary level of DBP and renal tubular dysfunction (Uchida et al., 2007). As urinary loss of DBP may decrease the capacity for reabsorption and activation of VD in proximal tubules, such deterioration of renal tubular function can justify the low level of 1,25(OH) 2 D under exposure to Cd (Berg, 1999). A perturbation in the conversion of 25(OH)D to 24,25(OH) 2 D and 1,25(OH) 2 D has been reported under the situation of increased blood and urinary Cd and blood Pb among smelter workers (Chalkley et al., 1998). In several disorders, striking evidence distinguishes the role of Pb intoxication in the daily intakes of VD and the concentrations of hydroxylated metabolites of cholecalciferol (Anetor et al., 1999;Arbuckle et al., 2016;Dongre et al., 2013;Edelstein et al., 1984;Mazumdar et al., 2017;Rahman et al., 2018;Szabo et al., 1991).
Because of the extensive use of lead in different industries, most of the existing Pb in our environment originates from human activities. Pb is an endocrine modulator in human populations and is considered as one of the risk factors of VDD (Dyer, 2007;Rahman et al., 2018). There is adequate evidence that depicts lead poisoning would have similar effects with Cd on the metabolic pathway of VD, especially in children (Box et al., 1981;Chang et al., 2014;Mahaffey et al., 1982;Rosen et al., 1980;Sorrell et al., 1977). Rosen and colleagues observed a significant negative correlation between blood lead level (BLL) and serum 1,25(OH) 2 D among lead-burdened children. In one to five-year-old children with increased blood concentration of Pb (33-120 μg/dl), serum concentrations of 1,25(OH) 2 D were decreased to levels seen in patients with hypoparathyroidism and metabolic bone disorders. The level of active form of VD turned to the normal level when BLL declined to < 30 μg/dl (Rosen et al., 1980). In a similar study performed by these researchers on 177 1-16-year-old individuals, remarkable negative association (r = −0.88) was obtained between serum 1,25(OH) 2 D levels and the concentrations of blood lead over the entire range of BLL (12 to 120 μg/dl). Adolescents aged between 11 and 16 years had serum 1,25(OH) 2 D levels higher than those observed among children 10 years old or younger (Mahaffey et al., 1982). Such a negative correlation was found between BLL and plasma 25(OH)D below 8.32 ng/ml in Asian children (Box et al., 1981). It has been suggested that further susceptibility of children to Cd and Pb could be attributed to the gradual process of maturity in kidney from fetal period to adulthood (Chalkley et al., 1998). An epidemiological study in China investigated the prevalence of VDD and insufficiency related to BLL. This cross-sectional study performed by Chang et al. confirmed the negative correlation between 25(OH)D and BLL (r = −0.216, P < 0.001). After multivariable adjustment, increasing child age, especially between 8 and 14 years (OR = 18.29; 95% CI 10.14, 32.99; P < 0.001) and BLL (OR = 1.01; 95% CI 1.00, 1.02; P = 0.045) were the significant predictors of 25(OH)D deficiency and insufficiency (Chang et al., 2014). In fact, this study affirmed some relationship between VD status and age-related accumulation of Pb in children.
It is of interest that intake of VD could be modulated by exposure to lead. Through a cohort of 2001 pregnant women, it has been shown that VD intake is negatively associated with maternal blood Cd, Pb, and manganese (Mn), as well as cord blood Pb. Regression analysis demonstrated that higher intake of Ca and VD can be correlated with lower maternal Pb and Cd concentration (Arbuckle et al., 2016). Sorrel et al. reported an interesting chain of correlations among High BLL, 25(OH)D level, and VD intake that may be a rational justification for the decreased serum levels of hydroxylated metabolites of cholecalciferol under exposure to Pb. High BLL (≥60 μg/l) was associated with the declined level of 25(OH)D in lead-burdened children (18 ± 1 ng/ ml) versus controls (32 ± 1 ng/ml). On the other hand, they figured out a positive correlation between VD intake and S-25(OH)D as the high blood Pb group had lower mean daily intakes of VD (210 ± 17 International Unit (IU) vs. 325 ± 20 in controls, P < 0.001) (Sorrell et al., 1977). As a result, a disturbance in the intake of VD can affect its serum and kidney hydroxylated metabolites.
Recently, experimental data verified the destructive role of Pb exposure in the metabolism of VD by affecting the expression of the involved enzymes. For the first time, the impact of Pb on the serum levels of VD metabolites, VD metabolizing enzymes (25-hydroxylase and 1α-hydroxylase), and VDR was observed in Wistar rats exposed to 0.2% Pb-acetate via their dams' drinking water from post-natal day (PND) 1 to 21 and directly in drinking water until PND30. S-25(OH)D markedly dropped at both PND21 and PND30, whereas 1,25(OH) 2 D was decreased (P < 0.05) only at PND21 in the Pb-exposed rats. Additionally, renal 1-α-hydroxylase was decreased by Pb only at PND21 (P < 0.05) but the brain 1α-hydroxylase was not influenced. Hepatic expression of 25-hydroxylase was substantially reduced at PND21 (Rahman et al., 2018). In human studies, a dramatic decrease of 25(OH)D 3 serum level has been reported in jewelry workers compared with controls (P < 0.0001) (Mazumdar et al., 2017), and of 1,25(OH) 2 D 3 serum level in battery manufacture workers (P < 0.01) (Dongre et al., 2013).

Uranium and cesium.
U is an alpha-emitting heavy metal occurring naturally in the earth's crust. Mining, energy industries, and nuclear accidents are the main risk of exposure to depleted and enriched uranium (DU and EU, respectively). Unfortunately, during the last decades, environmental concentrations of U have been increasing because of the growing use of this radionuclide in civil and military applications. Ingestion, skin penetration, and inhalation are the ways by which U can enter the body. Although a wide range of organs can be targeted by chemical and radiological toxic impacts of U, its major health effect is chemical kidney toxicity (ENDS, 2011).
In the case of U, according to the information provided by in vivo studies, it also appears that disrupting of renal production of VD is the underlying mechanism that interferes with VDES (Tissandié et al., 2008;Tissandie et al., 2007;Tissandie et al., 2006b;Yan et al., 2011). To investigate the impacts of DU on vitamin D 3 metabolism, Tissandie and colleagues chronically exposed rats to DU through drinking water at 40 ml/l for 9 months. For the first time, this study demonstrated that DU decreased plasma level of 1,25(OH) 2 D 3 and VD receptor expression in the kidney, resulting in the modulation of CYP24A1 expression and VD target genes in calcium homeostasis. In addition to a small decrease of 25(OH)D 3 in plasma, low level of CYP27A1 (the gene encoding 25(OH)D 3 ) has been observed in the brain of DU-exposed rats compared to the controls (Tissandie et al., 2007). In a similar study, the significant reduction of 1α-hydroxylase in the kidney of rats has been reported under chronic exposure to DU (Yan et al., 2011). Likewise, the chronic exposure to EU influences both mRNA and protein expression of renal nuclear receptors involved in the metabolism of VD (Tissandié et al., 2008). Consistent with findings obtained from experimental investigations on rats, epidemiological studies on the effects of DU exposure on a cohort of Gulf War veterans described elevated urinary levels of calcium and altered serum levels of PTH and 1,25(OH) 2 D 3 among DUexposed soldiers (McDiarmid et al., 2011;McDiarmid et al., 2008). Hence, disturbance in renal functions related to VDES is expectable under short-or long-term exposure to environmental DU and EU.
The explosion of Chernobyl nuclear power plant was a disaster that released 137 Cs in the environment and imposed serious health consequences on the populations of contaminated areas, mainly through the food consumption. This radioactive metal is formed in the fission nuclear of fissionable isotopes in nuclear reactors and weapons. Chronic contamination by 137 Cs can impair the liver metabolism of cholesterol, a precursor for the biosynthesis of steroid hormones, such as VD (Souidi et al., 2006). Tissandie et al. designed a research to explore the biological effects of chronic exposure to a post-accidental dose of 137 Cs on VD metabolism in liver, kidney, and brain. For this purpose, they encountered rats to the radioactive metal through drinking water for 3 months at a dose of 6500 Bq/l (approximately 150 Bq/rat/day), a similar concentration ingested by the people residing in contaminated territories in the former Union of Soviet Socialist Republics countries. Besides an insignificant decline in 25(OH)D 3 and a drastic decrease in 1,25(OH) 2 D 3 (53%, P = 0.02) in plasma, CYP2R1 (the key gene encoding 25(OH)D 3 ) mRNA level (20%, P < 0.05) has exhibited a noticeable drop in plasma level and brain. Nonetheless, the expression level of CYP2R1 in the liver was increased by 40% (P < 0.05). CYP27B1 mRNA level was declined in the brain by 20% (P < 0.05) as well (Tissandie et al., 2006a). This research group continued their metabolic studies on 137 Cs. As children are known to be the most susceptible group for VD metabolism disorders, effects of 137 Cs on VD biosynthetic pathway were investigated in newborn rats. The experiments were performed in 21-day-old male offspring of dams exposed to 137 Cs under similar conditions with the previous study but during the lactation period. Declined expression levels of CYP2R1 and CYP27B1 (−26 and −39%, respectively, P < 0.01) were reported in liver and kidney (Tissandie et al., 2009). Further, a significant increase of 1,25(OH) 2 D 3 and an unaffected level of 25(OH)D 3 (similar to the previous study) has been reported in that study.

EDCs
Rapidly increasing evidence suggests that environmental pollutions, particularly POPs possess an offensive capability to disrupt endocrine systems. Biological and environmental accumulation, lipophilicity, hydrophobicity, and resistance to environmental degradation are common traits of POPs, which include polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and organochlorine (OC) pesticides (Council, 2006). As mentioned above, since hormonally active VD metabolite and PTH have a vital interrelated role in the optimal balance of phosphorous and calcium for regulating bone mineralization, disrupting VD homeostasis may affect skeletal development and bone metabolism (Fig. 2). In fact, 1,25(OH) 2 Dsimulated intestinal absorption and renal reabsorption of phosphate and calcium induces bone mineralization. Lowering the active metabolite of VD bring about disturbances in bone mineralization and subsequently progressive bone loss (Holick, 2006). There are a couple of field and animal investigations showing the perturbations in VDES and decreased serum levels of 25(OH)D and 1,25(OH) 2 D under exposure to POPs, especially PCBs and Dichlorodiphenyltrichloroethane (DDT) (Alvarez-Lloret et al., 2009;Fletcher et al., 2005;Ju et al., 2012;Lilienthal et al., 2000;Routti et al., 2008).
In the first study of PCB-induced effects on vitamin D 3 metabolites, serum concentrations of 25(OH)D and 1,25(OH) 2 D were evaluated in rat dams and offspring after exposure to a mixture of PCB reconstituted based on the pattern of congeners found in human breast milk. Dosedependent reductions in serum levels of both VD metabolites were observed in both dams and newborns. It should be added that even at the lowest level of exposure (5 mg PCB/kg diet), reductions of 1,25(OH) 2 D were seen in dams (Lilienthal et al., 2000). Also, it was shown that embryonic exposure to 2,2′,3,3′,4-pentachlorobiphenyl (Arclor1254) have the capability to induce developmental deficits in the zebrafish skeleton subsequent to changes in PTH and VDR expression (Ju et al., 2012). These outcomes suggest that exposure to POPs could play an effective role in the VD metabolism. Non-maternal exposure to POPs generates similar impacts. Through an experiment on Baltic seals, Routti and colleagues perceived that thyroid system and VD are initial targets for PCBs and DDT to disturb bone homeostasis. Gray seals that were exposed to high levels of these chemicals have shown decreased levels of 1,25(OH) 2 D and thyroid hormones, which were negatively related to hepatic POPs. In accordance with the incorporated information, the authors concluded that contaminant-mediated VD and thyroid disruption might explain the bone lesions observed in the seals (Routti et al., 2008). Similar findings were realized in the case of exposure to dioxin-like PCBs. Using two cutting-edge techniques, namely Fourier transform infrared spectrometry (FTIR), and transmission electron microscopy (TEM), lower degree of mineralization (−8.5%; P < 0.05) and undesirable changes in size and crystallinity were detected in the vertebrae of rats exposed to 3,3′,4,4′,5-pentachlorobiphenyl (PCB126) compared with controls (Alvarez-Lloret et al., 2009). According to Norman (Norman, 2008), correct bone mineralization depends on VD-provided balance of calcium and phosphorous; therefore, lower levels of thyroid hormones, and VD (−21%; P < 0.005) seen in the rats exposed to PCB126 would potentially contribute in bone dysfunction (Alvarez-Lloret et al., 2009).
Recent epidemiological studies indicate a correlation between reduction of serum VD levels and exposure to POPs (Morales et al., 2013;Yang et al., 2012). In a Spanish population-based cohort, cross-sectional association of S-25(OH)D 3 (measured by gas chromatography) with the serum concentration of eight POPs (measured by high-performance liquid chromatography) were examined in 2031 pregnant women. Multivariate regression models disclosed an inverse linear relationship between PCB180 and S-25(OH)D 3 in which higher PCB180 levels were associated with lower 25(OH)D 3 level: quartile Q4 vs. quartile Q1, coefficient = −1.59, 95% CI −3.27, 0.08, P trend = 0.060. A nonmonotonic inverse association was also discovered between the sum of predominant PCB congeners (PCB 180,153 and 138) (Morales et al., 2013). The role of POPs in the pathogenesis of chronically metabolic disorders has been ascribed to their endocrine disrupting features. Importantly, some of the mentioned diseases have been associated with VDD. The first human study that linked the OC pesticides and S-25(OH)D in human was performed by Yang and colleagues. They sought the association between seven OC pesticides and S-25(OH)D in the 1275 subjects aged > 20 years. Of measured pesticides, p,p′-DDT (β = −0.022, P < 0.01), p,p′-DDE (β = −0.018, P = 0.04), and β-HCH (β = −0.022, P = 0.02) showed significant inverse associations with S-25(OH)D levels . Based on the found inverse association between PCBs and OC pesticides, it can be inferred that background exposure to these chemicals may lead to lower level of serum VD and probably VDD.
There is scarce human evidence of the destructive role of two main endocrine disrupting agents, namely phthalate and BPA, in VDES (Erden et al., 2014;Johns et al., 2017;Johns et al., 2016). Phthalate and BPA are broadly utilized for the production of a vast range of industrial and consumer products. These environmental chemicals have been extensively detected in the U.S. population. Johns et al. assessed the relationship between phthalate and BPA and VD by data obtained from a cross-sectional study with 4724 adults (aged ≥20 years). They reported an inverse association between total S-25(OH)D and urinary levels of di (2-ethylhexyl) phthalate (DEHP) metabolites in the overall population and urinary BPA only in women (Johns et al., 2016). This finding was consistent with the remarked negative correlation between serum BPA and S-25(OH)D reported among 128 patients with obstructive sleep apnea syndrome (OSAS) (Erden et al., 2014). Johns and colleagues evaluated the relationship between urinary phthalate metabolites and BPA and circulating total 25(OH)D in a large prospective cohort of pregnant women with 477 participants. Repeated measures of urinary DEPH metabolites and mono-3-carboxypropyl phthalate (MCPP) were negatively correlated with total S-25(OH)D levels. A nonsignificant inverse association was detected in the overall population analysis between urinary and total 25(OH)D (Johns et al., 2017).
Carbon tetrachloride (CCl 4 ), as a hepatotoxic solvent, is capable of disrupting VDES. For the first time, Nussler et al. reported some disturbances in the expression of genes involved in the VD metabolism. The hepatic expression levels of CYP2R1, CYP27A1, and VD-binding protein GC have significantly reduced in C27B16/N mice after 6-week treatment with CCl 4 . They have also reported a meaningful increase in 7-dehydrocholesterol reductase (DHCR7) gene expression in the liver (Nussler et al., 2014). Since there is a direct correlation between the expression of genes CYP2R1, CYP27A1, DHCR7 and GC with S-25(OH) D levels in healthy populations (Bu et al., 2010;Wang et al., 2010), it can be inferred that exposure to CCl 4 is capable of significantly decreasing VD levels.

Contributions of environmental contaminations in most likely mechanisms toward VDD
According to the conducted studies on the role of exposure to environmental pollutants in VDD until now (summarized in Table 1), whether human or animal, a variety of mechanisms have been proposed to interfere in VDES resulting in VD deficiency.
Sunlight exposure accounts for > 90% of VD production in human beings (Hoseinzadeh et al., 2018). Despite the stated evidence, air pollution might be an ignored risk factor while it is important for dysregulating VDES at the early stage. As shown in the Fig. 3, air pollutants could change the VD status directly and indirectly. Outdoor air pollutants, especially ozone and PMs can limit UVB photons reaching the ground through absorption of these sunlight's radiations. Thus, they can affect the cutaneous production of precholecalciferol and finally vitamin D 3 . Moreover, air pollution in certain situations may discourage people to engage in daily outdoor activities. This can synergistically, and indirectly decrease the cutaneous production of VD.
Increases of hypovitaminosis D in populations, while benefiting from enough dietary intakes of VD, makes environmental risk factors, such as air pollution a culprit. Overall, increases in air pollutants concentrations, may diminish sunlight rays reaching the skin and consequently lessen the cutaneous production of vitamin D 3 . Recent studies reported that usage of sunscreens with higher sun protection factor (SPF) in a regular basis could affect cutaneous production of VD among those individuals that are living in air-polluted areas with sunny climates (Faurschou et al., 2012;Libon et al., 2017).
Other studies have reported that exposure to Cd and Pb could lead to damages to kidney, especially renal tubular dysfunction and consequently impair bone homeostasis (Dongre et al., 2013; Keiko and Table 1 The effects of environmental contaminants on VDES.  In Tehran: 7 ± 13 In Gazvin: 11 ± 18 Living in a polluted area plays a significant independent role in VD deficiency. Cd and Pb interact with renal mitochondrial hydroxylases of the vitamin D 3 endocrine complex. Tsuritani et al. (1992)/ Japan/human 30 males and 44 females in the exposed group and 24 men and 48 women in the control group/ 50 years and above Environmental Cd

Manicourt and Devogelaer
The exposed group were from Cd-polluted area and control group from the non-Cdpolluted area Serum VD decreased in the exposed group, significantly in women. Renal damage due to Cd exposure induces the impairment of VD metabolism. Keiko and Minoru (1991)/ Japan/human 11 women/60-73 years Environmental Cd -The levels of S-25(OH)D were low-normal with a mean of 13 ± 5.0 ng/ml in 11 subjects.
Cd initially disturbs hydroxylation from 25(OH)D to 24,25(OH) 2 D and then disturbs hydroxylation from 25(OH)D to 1,25(OH) 2 D. Nogawa et al. (1987)/ Japan/human 5 itai-itai patients 36 (10 men and 26 women) exposed, and 17 (6 men and 11 women) non-exposed individuals/over 59 years Environmental Cd Exposed group and itai-itai patients were from Cdpolluted area and control group from the non-Cdpolluted area Serum 1,25(OH) 2 D levels were lower in itaiitai disease patients and Cd-exposed subjects than in non-exposed subjects.
Cd induces kidney damage, which in turn leads to a disturbance in VD and parathyroid hormone metabolism.

Rahman et al. (2018)/
Kuwait/animal Newborn Wistar rat pups Pb-exposed: 37 Chronic DU exposure could induce renal damages and inhibit the synthesis of the biologically active form of VD. Tissandié et al. (2008)/ animal/France 20 Sprague-Dawley male rats/ 12 weeks old EU-exposed group: 10 Control group: 10 Rats were exposed to EU in their drinking water.

Concentration of 40 mg/l
(1 mg/rat day)/9 months VDR and RXRα showed a low expression in the kidney.
Chronic exposure to EU affects both mRNA and protein expressions of renal nuclear receptors involved in VD metabolism. Tissandie et al. (2007)/ animal/France 20 Sprague-Dawley male rats/ 12 weeks old DU-exposed group: 10 Control group: 10 Rats were exposed to DU in their drinking water.

Concentration of 40 mg/l
(1 mg/rat day)/9 months 1. In exposed rats, 1,25(OH) 2 D 3 plasma level was significantly decreased. 2. In the kidney, a decreased gene expression was observed for CYP24A1, VDR, and RXRα. 3. In the brain, lower levels of messengers were observed for CYP27A1.
DU affects both the VD active form 1,25(OH) 2 D 3 level and the VD receptor expression.
(continued on next page) S.E. Mousavi et al. Environment International 122 (2019)  (continued on next page) S.E. Mousavi et al. Environment International 122 (2019)  S.E. Mousavi et al. Environment International 122 (2019) 67-90 Minoru, 1991Nogawa et al., 1987). Elevated urinary levels of DBP, accumulation of heavy metal in the kidney, and decreased level of 1,25(OH) 2 D confirm such disturbances (Chalkley et al., 1998;Kido et al., 1989;Uchida et al., 2007). In addition, experimental studies have shown that the impaired renal activation of VD originates from the dysregulation of renal 1α-hydroxylase (Youness et al., 2012). Proximal tubules including their mitochondria are the only sites of 1,25(OH) 2 D biosynthesis in the kidney (Morgan, 2001). Mitochondria hydroxylases are in charge of the renal conversion of 25(OH)D to 1,25(OH) 2 D (Youness et al., 2012). Since Cd and Pd exposure are engaged in malfunctioning mitochondria, the interaction of these toxic metals with the resident hydroxylases in renal mitochondria can be an involved mechanism in VDD (Alfvén et al., 2000;Chalkley et al., 1998;Dongre et al., 2013;Mazumdar et al., 2017;Youness et al., 2012). Altogether, the renal tubular dysfunction and in particular the inhibition of the renal 1α-hydroxylase enzyme are the most proposed pathways by which Cd and Pb contribute to VDD (Fig. 3).
In the case of radionuclides U and 137 Cs, based on conclusions deduced from in vivo studies in Table 1, exposure to these toxic metals can affect the expression of CYPs enzymes that deal with VD metabolism in liver and kidney, which results in the dysregulation of mineral homeostasis. Such a chain of events could modulate the expression of target genes by affecting the level of vitamin D 3 active form (Tissandie et al., 2006a, b). In fact, disrupting the renal production of VD is an indirect pathway that affects bone development and maintenance. The sustainable and well-balanced production of VD depends on the optimal function of VDES shown in the Fig. 2. In fact, losing the balance between involved factors in VDES influences the serum level of VD metabolites. In target organs of VD including kidney, bone, and intestine, 1,25(OH) 2 D binds to VDR and forms a heterodimeric complex with the retinoic acid X receptor alpha (RXRα), the receptor for the 9-cis retinoic acid. This complex acts as a ligand-activated transcription factor by altering the transcription rates of target genes, such as CYPs responsible for VD metabolism (Tissandie et al., 2007;Tissandie et al., 2006b). VDR activation by 1,25(OH) 2 D is a key factor in stimulating the intestinal absorption of calcium and phosphorus, the renal reabsorption of calcium, and other bone-related homeostatic processes. On the other hand, the decreased expression of VDR and RXRα (the principal regulators of CYPs related to the VD metabolism) under exposure to U and 137 Cs in animal studies has been reported (Tissandié et al., 2008;Tissandie et al., 2009;Tissandie et al., 2007). Thus, the results obtained by experimental studies can be interpreted that the reduced serum level of 1,25(OH) 2 D originated from the decreased activity of CYP27B1 and the decreased expression of receptors involved in VD metabolism are two main factors that modulate ligand-activated transcription factor of genes encoding proteins involved in VDES and finally end in VDD. Such procedure is the case with Cd and Pb.
Having background exposure to POPs could be a determinant factor in the pathogenesis of disorders that disrupt endocrine systems. As shown in Table 1, bone-related parameters, as well as thyroid and VD homeostasis, might be disturbed by exposure to EDCs. Vitamin D 3 , as a secosteroid, is structurally similar to steroids, such as cholesterol, testosterone, and cortisol. Influence of PCBs on the activity and expression of CYPs isomers, including those that catalyze the metabolism of steroid hormone have been reported. Thus, the disruption of hepatic hydroxylation of cholecalciferol to 25(OH)D 3 by the metabolites of PCBs containing hydroxyl group is plausible and susceptible to VDD (Morales et al., 2013). Theoretically, the inhibition of activity and expression of CYPs involved in the metabolism of steroid and thyroid hormones through exposure to EDCs depicts that CYPs interrelated to VDES could be probable targets for these chemicals by which POPs exposure-derived VDD observed in human and experimental models can be explained. As an indirect mechanism shown in Fig. 3, disrupting the homeostasis of TH, PTH, and calcium under exposure to POPs leads to chaos in bone metabolism and VDES and consequently the depression of VD level (Alvarez-Lloret et al., 2009;Ju et al., 2012;Routti et al., 2008). Perturbations in endogenous hormonal regulation stemmed from exposure to EDCs could result in weight gain (Elobeid and Allison, 2008;Tang-Péronard et al., 2011) which in turn participates in the hypovitaminosis D (Barrea et al., 2017;Johns et al., 2017;Pereira-Santos et al., 2015).

Fig. 3.
Possible biochemical pathways that air pollution, toxic metals, and EDCs may lead to VD deficiency (VDD). Decreasing outdoor activity and blocking UVB photons are the two mechanisms that air pollution may lead to the reduction of VD cutaneous production. Increasing renal tubular dysfunction as well as downregulating the transcription of CYPs are major pathways that heavy metals may initiate to bring about the reduction of VD serum level. EDCs may inhibit the activity and expression of CYPs. They may indirectly cause VDD through weight gain and the dysregulation of TH, PTH, and calcium homeostasis.

Tobacco smoke
Tobacco smoke is a mixture of hazardous chemicals implicated in the pathogenesis of a wide range of diseases. Owing to the presence of toxicologically harmful compounds, such as PAHs, aldehydes, and DDT, the smoke is carcinogen, neurotoxic, and endocrine disrupter (Diamanti-Kandarakis et al., 2009;Smith and Hansch, 2000). As mentioned above, a large number of studies have been conducted on the association of smoking and VDD where it has been one of the determinants of VDD. In this review, we focused on the effect of all types of smoking exposure on VDD, whether active or passive. Growing evidence affirms the disruptive role of this notorious blend in VDES. The incidence and prevalence of hypovitaminosis D among Norwegian adults (Larose et al., 2014), Belgium pregnant women (Vandevijvere et al., 2012), north European elderly women (Andersen et al., 2005), middle-aged and elderly Chinese individuals (Zhen et al., 2015), south Europeans (Cutillas-Marco et al., 2012) young Finnish adults (Lamberg-Allardt et al., 2001), young and middle-aged Greek males (Kassi et al., 2015), elderly Italian women (Isaia et al., 2003) and Spanish newborns (Díaz-Gómez et al., 2007) have been attributed to the exposure to tobacco smoke as an endocrine-disrupting risk factor. Unlike air pollutants, which principally have a significant effect on the photosynthesis of cholecalciferol, based on epidemiological, in vitro and in vivo studies, it appears that smoking effect could be attributed to the enzymatic and hormonal production of the subsequent metabolites.
An increased risk of hypovitaminosis D has been detected among smokers versus non-smokers (odds ratio: 1.8; 95% confidence interval: 1.00-3.35). Smoking also has been correlated negatively with the serum level of PTH (r = −0.24; P < 0.001) (Cutillas-Marco et al., 2012). There are strong evidences asserting the suppressive role of tobacco smoke in the production of PTH, cholecalciferol, and calcitriol (Brot et al., 1999;Cutillas-Marco et al., 2012;Jorde et al., 2005;Need et al., 2002;Supervia et al., 2006). In light of findings emanated from studies performed in this domain, the decrease of serum level of VD is proportionate to depression in the PTH level in smokers (Need et al., 2002). In fact, the key point is chaos in the interrelated regulation of VD and PTH levels, which results in the disruption of the VD metabolism. Hence, the dysfunctional vitamin D-PTH axis has been proposed as a common impairment, which occurs under tobacco smoke exposure. By using recorded data in the Tromsø study, the serum level of PTH was evaluated in 7896 subjects consisting of smokers and non-smokers. After correcting for confounding variables, the serum PTH levels were significantly lower in smokers compared with non-smokers (3.1 ± 1.4 vs 3.6 ± 1.9 pmol/l in males; 3.1 ± 1.5 vs 3.6 ± 1.8 pmol/l in females (P, 0.001)) (Jorde et al., 2005). Reduced levels of iPTH among smokers have been described in young and old individuals (Gudmundsson et al., 1987;Landin-Wilhelmsen et al., 1995). Moreover, lower intake of VD, decreased levels of S-25(OH)D, and decreased calcium absorption among smokers. Interestingly, one year after smoking cessation, serum levels of PTH have been similar to never smokers (Jorde et al., 2005). Similar results have been obtained in a cross-sectional study conducted by Brot and colleagues. They assessed the effect of smoking on serum VD metabolites and PTH in a cohort of 510 healthy Danish perimenopausal women. In comparison to nonsmokers, current smokers (50% of the cohort) had significantly reduced levels of serum 25(OH)D 3 (P = 0.02), 1,25(OH) 2 D 3 (P = 0.001), and PTH (P < 0.001). They further reported a disruption in the vitamin-PTH system among smokers, which was not explicable via lifestyle factors other than smoking; it was considered a crystal clear explanation for the harmful effect of smoking on osteoporosis in smokers (Brot et al., 1999). It is of interest that S-25(OH)D and osteocalcin have shown an inverse relationship with the number of smoked cigarettes per day in premenopausal women, which may accelerate the rate of bone loss (Hermann et al., 2000). Recently, a Chinese cohort with 612 older men reported a dose-response pattern between lower serum concentrations of VD and greater number of smoked cigarettes per day, longer smoking duration, and more pack-years (Jiang et al., 2016). Such results support the inverse association between VDD and smoking. However, further studies are needed to confirm such pattern between smoking and serum levels of PTH.
Although the observed inverse association has been significantly reported among women in general and older ones in particular, the mentioned relationship between exposure to tobacco smoke and VDD is not confined to a specific age group or particular gender. Kassi et al. carried out a research on the relationship between the prevalence of VDD and lifestyle parameters including smoking in healthy young and middle-aged men. They observed a strong correlation between 25(OH) D 3 and smoking in participants (P < 0.001). By the help of a multinomial logistic regression model, they showed that the probability of having vitamin D 3 inadequacy has increased in smokers up to 58% and 63% for 20-29 year and 40-50-year age subgroups, respectively, compared with non-smokers (Kassi et al., 2015). Maternal exposure to tobacco smoke during pregnancy can affect VD level and subsequently calcium metabolism in mothers and infants (Banihosseini et al., 2013;Díaz-Gómez et al., 2007;Khuri-Bulos et al., 2014;Lawlor et al., 2013). Diaz-Gomez and colleagues evaluated the effects of smoking on the vitamin D-PTH system during the perinatal period. In their cohort study with 61 mothers and newborns, declined serum levels of PTH in both mothers and neonates, as well as a significantly lower levels of 25(OH) D in neonates were observed (Díaz-Gómez et al., 2007). These findings have been also seen in another prospective cohort study where 7.6% of mothers reported a history of smoking in the gestational period and 72.4% of all mothers reported passive exposure to tobacco smoke during pregnancy. Their results showed that the prevalence of severe VDD in newborns was associated with maternal smoke exposure (Khuri-Bulos et al., 2014). In addition, those mothers who have smoked during their pregnancy had lower 25(OH)D levels across all their trimesters compared with non-smokers (Lawlor et al., 2013). In a most recent study, Chinellato and colleagues studies the association between S-25(OH)D level and parental smoking in 152 children who were suffering from asthma. Children with both nonsmoking parents benefited from a significantly higher serum level of 25(OH)D than those with both smoking parents (median of 20.5 ng/ml vs median of 14.5 ng/ml; P < 0.001) (Chinellato et al., 2018).
In addition to the depression of serum levels of VD metabolites, it has been shown that exposure to tobacco smoke can influence the intake of VD as well. A population-based survey of 2319 women conducted by Morabia et al. in Switzerland showed that tobacco smoke could cause the low intake of VD through changing the dietary taste (Morabia et al., 2000). Furthermore, VD malabsorption might be a possible hypothesis that needs experimental investigations to be testified. It is worth noting that after a long time (5 years) smoking cessation, similar intake of VD and calcium has been reported in both former and never smokers (Morabia et al., 2000). On the other hand, the impact of smoking quitting on maintaining or even increasing serum VD levels described in some studies confirms the crucial importance of tobacco smoke exposure in dysregulating VDES (Gilman et al., 2006;Jiang et al., 2016). Jiang and colleagues revealed that longer duration of smoking cessation (> 20 years) could result in higher VD levels in former smokers compared with current smokers (P for trend = 0.04) (Jiang et al., 2016). A related point to consider is the corrective action of smoking quitting in relation to the calcitriol-PTH axis. Indeed, undesirable changes imposed on vitamin D-PTH system by smoking could be reversible by smoking ceases (Need et al., 2002).
Accumulating evidence suggests the effect of smoking on the deterioration of different disorders interrelated to VD levels. It means exposure to tobacco smoke influences the state of a disease through the alteration of VD metabolism. Gestational diabetes risk (Dodds et al., 2016), activating Crohn's disease (Gilman et al., 2006;Jørgensen et al., 2013), high rates of depression (Ren et al., 2016), the risk of myocardial infarction (Deleskog et al., 2012), and increased markers of inflammation in HIV-infected individuals (Legeai et al., 2013) might be Table 2 The effects of smoking on VDES. 2. An additive interaction was detected between smoking status and 25(OH)D.
The study supports the inverse association of VD status with gestational diabetes risk, particularly among women who smoke during pregnancy. Kassi et al. (2015)/Greece/ human 181 men/20-50 years old Passive smoking exposure: 54 Controls with no exposure: 54 1. The mean level of 25(OH)D in maternal serum was 9.28 ± 5.19 ng/ml in exposed and 5.26 ± 10.75 ng/ml in non-exposed group (P > 0.05).
2. The mean concentration of 25(OH)D in cord serum was 10.83 ± 6.68 ng/ml in the exposed and 11.05 ± 4.99 ng/ml in the non-exposed group (P > 0.05).
The serum VD level was not significantly different in mothers and infants between two groups, but it was lower in the exposed group. Crohn's disease patients who smoked had lower VD levels (51 nmol/l) than patients who did not smoke (76 nmol/l), P < 0.01.
Patients who smoked had lower 25(OH)D levels than patients who did not smoke, independent of disease activity. Legeai et al. (2013) (continued on next page)  associated with lower VD levels triggered by smoking. Significant and independent association between smoking and 25(OH)D concentrations has been reported in current-smoking patients (Deleskog et al., 2012;Jørgensen et al., 2013;Legeai et al., 2013). Through a case-control study, Dodds et al. revealed an inverse correlation between VD status and gestational diabetes, especially among women who smoked during pregnancy (Dodds et al., 2016). This inverse association was also reported among two cohorts of HIV-infected individuals between tobacco use and severe VDD [25(OH)D < 10 ng/ml] (Legeai et al., 2013;Wasserman and Rubin, 2010). In two studies exploring the relationship between VD status and smoking in patients with acute ischemic stroke and myocardial infarction, VD concentrations were significantly lower in smokers compared with non-smokers.

Tobacco smoke can potentially target almost all points of VDES
Although the mechanisms through which smoking interrupts the VD metabolism is poorly defined and still remain unknown, on the ground of the foregoing evidence summarized in the Table 2, making some highly likely assumptions is presumably by the help of a chain of experimental-based deductive reasoning. The sufficiency of serum VD level plays a decisive role in the bone health subsequent to calcium intake and serum parathyroid hormone levels. Under normal conditions, serum 25(OH)D and 1,25(OH) 2 D levels are negatively correlated with the serum level of PTH (Cheng et al., 2003;Holick, 2006). Through a cross-sectional study with 944 healthy participants divided into three groups of age (30-45 years, 50-65 years, and 70-85 years), calcium intake (< 800 mg/d, 800-1200 mg/d, and < 1200 mg/d), and S-25(OH)D level (< 10 ng/ml, 10-18 ng/ml, and < 18 ng/ml), Steingrimsdottir et al. evaluated the importance of high levels of 25(OH)D and calcium intake in serum intact PTH (iPTH). After adjustment for relevant factors, such as smoking, it was observed that serum PTH was lowest in the group with a serum 25(OH)D level of > 18 ng/ml but highest in the group with an S-25(OH) D level of < 10 ng/ml. They revealed that at low S-25(OH)D level (< 10 ng/ ml), the calcium intake of < 800 mg/d vs > 1200 mg/d was significantly associated with higher serum PTH (P = 0.04) (Steingrimsdottir et al., 2005). Therefore, they reasoned that VD adequacy is much more important than high calcium intake in maintaining the optimum level of serum PTH. Accordingly, in response to the decreased level of VD, an increase in PTH level is expected; however, under exposure to tobacco smoke serum levels of both 25(OH)D and PTH have been lowered markedly (Brot et al., 1999;Cutillas-Marco et al., 2012;Jorde et al., 2005;Need et al., 2002;Supervia et al., 2006). Hypoparathyroidism has been discovered in both mother and their newborns under maternal smoking (Díaz-Gómez et al., 2007). It has been speculated that the dropped level of PTH in individuals exposed to tobacco smoke may be as a result of a small unmeasurable change in serum ionized calcium, a reduction in secretion or a rise in degradation of the hormone (Brot et al., 1999;Need et al., 2002). Due to the presence of endocrine disrupters in tobacco smoke, it is expected to see smoking-derived deleterious impacts on the VDES similar to the case regarding the reduction of vitamin D-PTH axis. Notably, in comparison to nonsmokers, smokers suffer from a significant decline in BMD caused by the impairment of calcium absorption, which in turns is attributed to the imposed destructive changes on vitamin D-PTH system (Brot et al., 1999;Jorde et al., 2005;Need et al., 2002). The association between lower BMD and smoking along with the declined concentration of VD have been described in current and former smokers (Lorentzon et al., 2006;Szulc et al., 2002). In line with the abovementioned studies, recent observations have demonstrated that lower concentrations of steroid hormones, as well as 25(OH)D, are associated with passive and active cigarette smoking (Lorentzon et al., 2006;Shinkov et al., 2015;Soldin et al., 2011;Szulc et al., 2002). There is further evidence based on a cross-sectional study of 405 postmenopausal women where poor calcium absorption in smokers has been associated with declined level of PTH (which leads to decreased renal hydroxylation of the storage form of VD to active structure) (Need et al., 2002).
Acrolein is an environmentally ubiquitous volatile organic compound (VOC) (Amini et al., 2017). This unsaturated aldehyde is responsible for a considerable part of cellular and molecular adverse effects of tobacco smoke. Through an in vitro study, the effects of cigarette smoking extract (CSE) and acrolein on the conversion of 25(OH)D 3 to 1,25(OH) 2 D 3 were separately studied by Hansdottir and colleagues. They found that both of them had the capability to significantly reduce the generation of active vitamin D 3 in the airway epithelial cells. It is of note that the effects of CSE and acrolein were reversed by the enzyme aldehyde dehydrogenase (Hansdottir et al., 2010). This enzyme catalyzes the oxidation of aldehydes by addition of oxygen rather than by removal of hydrogen-that is, it converts aldehydes to carboxylic acids (Marchitti et al., 2008). The reduction of 1α-hydroxylase under exposure to CSE was observed as well. Consequently, lower level of 1,25(OH) 2 D 3 could contribute to the overexpression of inflammatory factors in smokers (Hansdottir et al., 2010). In a similar investigation, Mulligan et al. genetically scrutinized the effect of cigarette smoke (CS) on the levels of vitamin D 3 and its conversion to bioactive form in patients with chronic rhinosinusitis. They found that CS could impair the capability of epithelial cells in the conversion of 25(OH)D 3 to 1,25(OH) 2 D 3 by the downregulation of CYP27B1 expression. CS exposure also was associated with declined circulating level of 25(OH)D 3 (Mulligan et al., 2014). In vivo and in vitro exposure to other components of tobacco smoke such as nicotine and Benzo[a]Pyrene (BaP) can decrease body storage of 25(OH)D and 1,25(OH) 2 D (Fung et al., 1999;Fung et al., 1998;Iwaniec et al., 2002;Matsunawa et al., 2009). By using an array of research, Fung et al. demonstrated that long-and short-term treatment with nicotine, the principal and addictive alkaloid in tobacco, significantly decreased S-25(OH)D in female rats (Fung et al., 1999;Fung et al., 1998;Iwaniec et al., 2002). Interestingly, such negative correlation between VD serum concentrations (< 20 ng/ml) and serum cotinine, the main metabolite of nicotine, has been reported among American females with different ethnic backgrounds. Based on the results, female active-smokers have the lowest mean VD concentration and higher prevalence of VD deficiency and inadequacy within the overall population in the U.S. (Manavi et al., 2015). 24-Hydroxylase is a member of the cytochrome P450 superfamily of enzymes encoded by the CYP24A1 gene involved in the metabolism of VD. It catalyzes reactions including 24-hydroxylation of 25(OH)D and 1,25(OH) 2 D 3 resulting in VDD. SNP rs4809957, located in the 3′ untranslated region of CYP24A1 at 20q13.2, interacts with smoking dose (Dong et al., 2012). Moreover, BaP, a PAH found in tobacco smoke, increases the catabolism of 1,25(OH) 2 D 3 in human monocyte/macrophage-derived THP-1 cells by enhanced expression of CYP24A1. In other words, BaP enhances the production of 25-hydroxyvitamin D 3 24 hydroxylase (CYP24A1), which leads to inactive metabolites of 24,25(OH) 2 D and 1,24,25(OH) 3 D, and consequently decreased serum level of VD (Fig. 4) (Matsunawa et al., 2009). Indeed, dysregulation of genes that encode those enzymes involved in VD metabolism would be one of the possible mechanisms engaged in tobacco smoke-originated VDD.
Skin plays the most important role in the provision and metabolism of VD. Smoking (Ernster et al., 1995) and solar ultraviolet radiation (Pillai et al., 2005) have been known as extrinsic factors inducing an aging effect on human skin. Facial wrinkling is a marker of skin aging. Lopez Hernandez et al. discovered strong evidence of accelerated skin aging among smokers. They found a statistically remarkable effect of smoking habit (OR = 3.1; 95% CI = 1.28-7.76; P = 0.008), sun exposure (OR = 1.50; 95% CI = 1.25-1.80; P = 0.05), and age (OR = 1.18; 95% CI = 1.13-1.23; P = 0.024) on facial wrinkling (López et al., 1995). Increasing skin aging can diminish the capability of skin in the conversion of 7-DHC to precholecalciferol substantially (Holick, 1995). Using a cross-sectional study of 299 never smokers, 551 former smokers and 286 current smokers, a positive association between pack-years smoking and facial wrinkle score has been revealed.
The relative risk of moderate/severe wrinkling was estimated for current men smokers as 2.3 (95% confidence interval CI = 1.2, 4.2) compared to women never smokers 3.1 (95% CI = 1.6, 5.9) (Ernster et al., 1995). Mechanistically, the induction of matrix metalloproteinases (MMPs) mediates photo aging. Concerning tobacco smoke, there is adequate evidence putting an emphasis on the reinforcing role of smoking in skin aging through activation of MMPs (Holick, 1995;Lahmann et al., 2001). A significant increase of MMP-1 mRNA has been reported in the skin of smokers compared with non-smokers, whereas no difference has been seen for the tissue inhibitor of MMP-1 (Lahmann et al., 2001). It should be pointed out that smoker patients with chronic obstructive pulmonary disease (COPD) are at high risk of VDD because they are more susceptible to skin aging (Janssens et al., 2010). Overall, since smokers are more prone to skin aging, it is anticipated that smoking-derived skin aging causes VDD through the disturbance in the cutaneous production of vitamin D 3 .
The declined level of serum 1,25(OH) 2 D observed in smokers might also be as a result of cadmium accumulation in the kidneys (Brot et al., 1999;Kido et al., 1989). Since the tobacco plant has a considerable capability to concentrate heavy metals in general (Cd and Pb in particular), tobacco smoke is one of the common and rich sources of Cd and Pb exposure (Lugon-Moulin et al., 2006). Accordingly, inhibiting VD activation mediated by Cd and Pb poisoning can be one of the mechanisms involved in VDD. Impaired kidney function is the central core of this pathway. Indeed, there are elevated serum cadmium and lead levels in smokers, which results in the deterioration of renal tubular function and glomerular dysfunction (Cooper, 2006).
Putting experimental and epidemiological findings together, as shown in Fig. 4, skin aging, simultaneous decrease of VD and PTH, disturbance in the intake of dietary VD, renal tubular dysfunction, and dysregulation of CYPs genes related to the metabolism of VD are most possible pathways that interfere with VDES via exposure to tobacco smoke. In Fig. 4, we have integrated all mentioned mechanisms in order to provide a comprehensive view of a wide, and to some extent independent, mechanisms that might lead to VDD. Accordingly, there is an internationally emerging consensus among researchers corroborating the claim that smoking enhances the risk of VDD and severe VDD. Nonetheless, further investigation in the genetic analysis of suspected CYPs enzymes is needed to clearly ascertain metabolic mechanisms responsible for the reported smoking-derived VDD.

Summary and conclusion
This review integrated various pieces of evidence that exposure to air pollution, environmental chemicals, and smoking (with endocrine disrupting properties) can negatively interfere VDES and in most cases lead to VDD. To the best of our knowledge, the possible mechanisms that may be more likely disturbed and induce low serum levels of VD include: (1) decreased intestinal intake of VD, (2) decreased cutaneous production of cholecalciferol, (3) the modulation of genes involved in VD homeostasis, particularly CYPs, and (4) decreased local production of calcitriol in target tissues. Nevertheless, we still need mechanistic studies to explain precisely the biochemical pathways for mentioned chemicals. Therefore, we are in need of conducting more epidemiologic and experimental investigations on those sorts of environmental contaminants, which have the potency to cause VDD, so that involved mechanisms would be delineated more clearly.
Based on the presented evidence and potential mechanisms, monitoring serum level of VD in individuals with high exposure to the mentioned stimuli, and consuming higher amount of VD fortified foods and supplements by people settled in polluted areas are recommended in order to minimize the detrimental impacts on VDES. It is expected that the correction of VD status result into two vital achievements. First, it ameliorates anti-inflammatory and anti-oxidative capabilities in order to effectively deal with inflammatory-oxidative conditions originated from environmental toxicants. Second, such action averts the dramatic decrease of serum VD under conditions of exposure to environmental contaminants, and maintains related biochemical Fig. 4. Potential mechanisms related to the disruption of VDES by tobacco smoke. The positive and negative signs on the end of the arrows imply increasing and decreasing the targets, respectively. Red arrows originated from the cigarette result in decreasing serum 25(OH)D and 1,25(OH) 2 D levels, VD intake from diet, and the cutaneous production of VD. These four items determine the VD level of individuals. From a systematic point of view, cigarette smoke leads to depression of VD levels in humans. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) pathways subsequent to keeping VDES well balanced. Accordingly, since pregnant women, on the one hand, are susceptible to VDD, and on the other hand, susceptible to environmental pollutants, recent investigations recommend a VD screening program and VD supplementation for those pregnant women who need it. These actions potentially could lead to a substantial decrease in related adverse pregnancy outcomes, especially in environmentally polluted areas (Holick, 2018;Rostami et al., 2018).

Declarations of interest
None.