Introduction

Endocrine-disrupting compounds (EDCs) are exogenous environmental chemicals which are ubiquitous in water, air, food, and industrial products that can interfere with the normal hormone function of humans or animals, thus posing a potential threat to the environment and human health. The endocrine-disrupting activity of organic compounds (polycyclic aromatic hydrocarbons, polychlorinated biphenyls, organochlorine pesticides, alkylphenol, etc.) has been widely studied (Oehlmann et al. 2000; Staniszewska et al. 2014; Yilmaz et al. 2020; Hou et al. 2021). Nonylphenol (4-NP) and octylphenol (4-t-OP) are alkylphenols (APs) and primary derivatives of alkyl phenol ethoxylates (APEs). APEs had an extensive application area such as industrial formulations (paper, leather, tannery, textile, oil industries, and metal working fluids), antifoams, detergents, dispersants, emulsifiers, paint ingredients, pesticide adjuvants, and personal care products (Cox 1996; Vazquez-Duhalt et al. 2006; DEFRA 2008). Until recently, 4-NP and 4-t-OP were widely used as intermediates in the production of phenolic resins and alkyl phenol ethoxylates in European Union (EU) and the USA (Van Miller and Staples 2005; Vazquez-Duhalt et al. 2006). Since 2000, 4-t-OP and 4-NP have been added in the priority dangerous substances list in line with the Directive 2000/60/EC (European Commission 2000). In 2003, a reduction policy for 4-NP has also been implemented in EU, and in January 2005, a restriction has been imposed on the sale and use of products containing more than 0.1% NP (European Commission 2003). Only a few countries (e.g., Asian countries) continue to use APEs today (David et al. 2009). Bisphenol A (BPA) is a high production volume chemical. During the production of resins and polymers (polycarbonate, epoxy, polysulfone, polyacrylate, polyetherimide, unsaturated polyester, and phenolic), BPA acts as a crucial intermediate. As BPA takes part during the production process of many materials and products such as bottles, coatings, pipes, dental sealants, food packaging, nail polishes, and flame-retardant materials, high human exposure is inevitable on a daily basis (Maragou et al. 2006, 2008; Shelby 2008; Ballesteros-Gómez et al. 2009). In 2010, EU banned usage of BPA in plastic bottles in order to protect infant and public health (European Commission 2010). On the other hand, the Food and Drug Administration (FDA) changed regulations and outlaw the use of BPA-based polycarbonate resins in feeding bottles and sippy containers in 2012. Furthermore, baby food packages were no longer able to contain BPA-based epoxy resins since 2013 (US Food and Drug Administration 2013). APs and BPA have been detected in marine, fresh water, wastewater treatment plants (WWTPs), sediments, drinking water, and rivers in worldwide studies (Ying et al. 2002; Soares et al. 2008; Stasinakis et al. 2008; David et al. 2009; Gatidou et al. 2010; Hawker et al. 2011; Zhang et al. 2011; Niu et al. 2015), and bioaccumulation studies have shown that APs and BPA can accumulate in environmental matrixes (in various species of algae, plants, invertebrates, and freshwater fish) even at low doses (Ahel et al. 1993; Snyder et al. 2001; Ying et al. 2002; Soares et al. 2008; Mortazavi et al. 2013; Staniszewska et al. 2014). Studies conducted in many countries have shown that APs and BPA can also be found in human tissues such as human blood, urine, breast milk, amniotic fluid, follicular fluid, placental tissue, semen, umbilical cord blood, fetal serum, and adipose tissues (Vandenberg et al. 2010; Li et al. 2013). According to the studies, APs and BPA exposure even at low concentrations may cause cancer, cardiovascular disease, diabetes, reproductive, teratogenic, and development toxicities in individuals who are exposed to them (Sugiura-Ogasawara et al. 2005; Berkowitz 2006; Bredhult et al. 2007; Tachibana et al. 2007; Soares et al. 2008; Lang et al. 2008; Barrett 2010). As phenols are mainly excreted in urine and feces through glucuronidation and sulfation, urine is considered the most suitable matrix for biomonitoring of phenols (Dekant and Völkel 2008; Wu et al. 2010). Therefore, biomonitoring of BPA, 4-NP, and 4-t-OP in urine samples is highly precious to assess correlations between exposure and adverse health effects. During a biomonitoring study, applications of highly sensitive analytical methods improve the reliability and efficacy of the study. There are many studies that measured urinary BPA levels in Turkey. To authors knowledge, there has been no study which simultaneously measured urinary BPA, 4-NP, and 4-t-OP levels in humans (Battal et al. 2014a, b; Buluş et al. 2016; Ince et al. 2018; Durmaz et al. 2018; Sayıcı et al. 2019; Akgül et al. 2019; Çok et al. 2020; Ayar et al. 2021). Thus, the main objectives of the present study were as follows: (i) to determine levels of three EDCs (BPA, 4-NP, and 4-t-OP) in urine of population living in Mersin Region, Turkey; (ii) to study factors influencing the BPA, 4-NP, and 4-t-OP levels; and (iii) to estimate exposure and risk assessment for BPA, 4-NP, and 4-t-OP for population of interest.

Materials and methods

Study population and sample collection

To conduct this study, 103 authentic urine samples were collected from both male, female, and children living in the Mersin Region (latitude: 36° 47′ 42.94″ N; longitude: 34° 37′ 4.51″ E), Turkey, between the years 2015 and 2016. Mersin is a large city and a port on the Mediterranean coast of southern Turkey (Fig. 1). Prior to the sample collection, all participants were asked to fill in a questionnaire to gather information about sociodemographic data such as age, sex, occupation, diet, smoking, alcohol consumption, their dietary habits, family history of diseases, and study-specific data such as exposure to BPA, 4-NP, and 4-t-OP in their daily life, home, and school environment. Urine samples were collected in 125-mL glass (PTFE-lined) screw cap cultures that had been previously cleaned with hexane and divided into aliquots. They were subsequently kept at −20 °C until analysis. To avoid any possible contamination from environmental sources of free BPA, glassware was preferred where possible. Urine samples were collected based on a formal consent form approved by Mersin University Clinical Research Ethical Committees in Mersin, Turkey (document number 2011/53).

Fig. 1
figure 1

Sampling area, Mersin Region in Turkey (Vikipedi)

Full size image

Measurements of urinary concentrations of BPA, 4-NP, and 4-t-OP

Chemicals and reagents

During the study, only analytical grade chemicals were used. Artificial urine, BPA (CAS: 80-05-7), 4-NP (CAS: 84852-15-3), 4-t-OP (CAS: 140-66-9), and creatinine (CAS: 60-27-5) standards were obtained from Sigma-Aldrich (St Louis, MO, USA). The isotope-labeled internal standards 13C12-BPA (BPA-IS, 99%; CAS: 263261-65-0), 13C12-NP (NP-IS, 99%; CAS: 140-40-5), 13C6-OP (OP-IS, 99%; CAS: 1173020-24-0), and creatinine-d3 were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). HPLC-grade acetonitrile and methanol (KGaA, Darmstadt, Germany) were utilized. Ammonium acetate was purchased from J. T. Baker (Phillipsburg, NJ, USA). For precipitation/dilution, enzyme β-glucuronidase from Helix pomatia was used. For all aqueous solutions, the water used was deionized water (18.2 MΩ) treated with Millipore (Simplicity, 185) Milli-Q water purification system (Elga Labwater Veolia, Anthony, France).

Chromatographic and mass spectrometric conditions

An Agilent 1200 Series 6460 triple quadrupole LC-MS/MS with Jet Stream (ESI-MS/MS; Waghausel-Wiesental, Germany) equipped with a binary pump, a vacuum degasser, low carryover autosampler, a thermostatted column compartment, and a MassHunter data system was used to identify and quantify the compounds under consideration. Twenty microliters of the extract was injected onto an Agilent Pursuit 3 pentafluorophenyl propyl (PFP) column (100×3.0mm, 3μm particle size). The column temperature was kept at 20°C. The gradient mobile phases A and B consisted of 2 mM ammonium acetate in water and methanol, respectively. MS 2 Segment was used for analyses. To avoid contamination of the MS source, the diverter valve was switched to waste position. The tandem mass spectrometry (MS-MS) was run with negative electrospray ionization in the multiple reaction monitoring (MRM) mode. Nitrogen was used as both curtain and collision gas. T Optimal conditions for ESI operation were as follows: sheath gas and auxiliary gas (N2, 99.995%) flow rates 11 L/min; sheath gas at 200°C; auxiliary gas temperature 150°C; 3500V negative capillary voltage; and 300V negative nozzle voltage. The peak areas of analytes and their IS (13C12-BPA, 13C12-NP, and 13C6-OP) were studied using the MassHunter software version 4.0.

Standard solutions, calibration standards

To procure working solutions of analytes, methanol was used in the dilution process of the stock solutions (to achieve a concentration in the range of 5–500 ng/mL). Stock solutions of the BPA (10 ng/ml), BPA-IS (100 ng/ml), 4-NP (500 ng/ml), NP-IS (10 ng/ml), 4-t-OP (500 ng/ml), and OP-IS (100 ng/ml) were prepared in methanol and were stored at −20°C under dark conditions till further steps. The validation of the method used in this study was described in our previous study (Battal et al. 2021). Calibration curves (n=8) were developed by spiking the artificial urine with established concentrations of the analytes (5, 20, 50, 100, 200, 300, 400, 500 ng/mL). To prepare calibration curves, urine samples and solvent-fortified with a known amount of analytes were used. Each sample was spiked with IS. The limit of detections (LOD) is the lowest amounts of BPA, 4-NP, and 4-t-OP in one sample that bioanalytical procedure can reliably differentiate from background noise. Limit of quantifications (LOQ) is similar to LOD except it is a range of the upper and lower bounds that can be quantified with a predetermined acceptance level of precision, accuracy, and specificity. The LODs and LOQs of the study were calculated with S/N 3 and 10, respectively.

Sample preparation

Urine samples from healthy volunteers were collected for determining the spot urine concentrations of BPA, 4-NP, and 4-t-OP. Urine samples (500 μL) were incubation 1 h at 37 °C with enzyme (β-glucuronidase from Helix pomatia). Following incubation, the samples were purified by protein precipitation/dilution with 500 μL of acetonitrile and 50 μL of the IS (BPA, 4-NP, and 4-t-OP). After protein precipitation, samples were centrifuged at 3000 rpm at 25 °C for 15 min. Briefly, 13C12-BPA, 13C12-NP, and 13C6-OP were used as an internal standard and were added to the samples (Fig. 2).

Fig. 2
figure 2

Sample preparation for BPA, 4-NP, and 4-t-OP analysis with LC-MS/MS

Full size image

Analysis of creatinine in urine

To evaluate the influence of creatinine adjustment on the total variance of spot authentic urine samples, urine creatinine levels were analyzed by using a modified method which was developed and validated for creatinine analyses by Park et al. (2008). In brief, Milli-Q water (1000-fold) was used to dilute an aliquot of urine (10 μL). After the addition of 100 μL (5 mg/L) of creatinine-d3, LC-ESI-MS/MS was operated in electrospray positive ionization mode in order to analyze creatinine. The SRM transitions for creatinine and creatinine-d3 were monitored as 114.1 > 86.1 and 117.2 > 89.2, respectively. The column used was Agilent Zorbax SB-C18 chromatographic column (3 × 50 mm, 3.5 μm particle sizes). One microliter of the extract was injected onto the column. 2mM ammonium acetate was found in both mobile phases A (water) and B (methanol). Using isocratic condition (80% B), the analysis for creatinine was successfully done. The calculations were done regarding the calibration curve, and the data were corrected for recovery using an internal standardization method. Creatinine was quantified in accordance with the calibration curve obtained after insertion of known amounts of creatinine (0.5, 1, 1.5, 2.0, 2.5, 3.0, 6.0 mg/mL) and creatinine-d3 (IS) (5 mg/mL) to Milli-Q water. Calibration curve was constructed from seven data points using MassHunter 4.0 with R2 > 0.9999 for creatinine.

Statistical analysis

Normality assumption was evaluated with Kolmogorov-Smirnov and Shapiro-Wilk test, and Levene test was used to assess homogeneity of variances. One-way ANOVA, Welch test, or Kruskal-Wallis test was used to compare groups in terms of numerical variables, as appropriate according to the assumptions. BPA, 4-NP, and 4-t-OP levels were analyzed by Mann-Whitney U or Kruskal-Wallis (followed by Dunn post hoc test) test, in group comparisons. Spearman rank correlation was used to examine correlation between BPA, 4-NP, and 4-t-OP levels and numerical variables. Categorical variables were analyzed with Pearson chi-square, Fisher’s exact, and Fisher-Freeman-Halton test, according to the expected count rule and number of groups compared. Categorical data were summarized as frequency and percentage, while mean ± standard deviation or median, interquartile range, and minimum-maximum values were used to summarize quantitative data. Besides, BPA, 4-NP, and 4-t-OP levels were summarized by calculating the arithmetic mean, standard deviation, geometric mean, minimum, maximum, median, and percentiles (25th, 50th, 75th, 95th). Statistical analyses were done with SPSS v.22 and Statistica 8.0 statistical packages, and significance level was considered as 0.05.

Risk assessment

Exposure and risk assessment have been calculated based on the detected urine levels of BPA, 4-NP, and 4-t-OP in the Turkish population. The hazard quotient (HQ), as a risk measurement, can be estimated by calculating the EDI (estimated daily intake) by reverse dosimetry (Katsikantami et al. 2019). Following this approach, the EDI has been calculated according to Eq. (1):

$$\mathrm{EDI}\ \left(\upmu \mathrm{g}/\mathrm{kg}\ \mathrm{bw}\ \mathrm{day}\right)=\left(\mathrm{C}\ \left(\upmu \mathrm{g}/\mathrm{L}\right)\ {\mathrm{V}}_{\mathrm{urine}}\ \left(\mathrm{L}/\mathrm{day}\right)\right)/\mathrm{F}\ \mathrm{BW}\ \left(\mathrm{kg}\right)\Big)$$

where C is the obtained concentration of BPA, 4-NP, and 4-t-OP, Vurine is the total urinary volume excreted in 24 h, F is the compound’s urinary excretion factor (1), and BW is the body weight for each participant. The volume of urine for 24 h was 1.60 L/day for male, 1.20 L/day for female, and 0.4 L/day (ages 3–4 years) and 0.5 L/day (ages 5–6 years) for children (Dirtu et al. 2013; Çok et al. 2020).

The risk assessment was estimated depending on the reference values available for each compound. The EFSA (2015) established a reference dose/tolerable daily intake (TDI) of 4 μg/kg bw/day for BPA (European Food Safety Authority 2015). A TDI for 4-NP is suggested to 5 μg/kg bw/day (Nielsen and Ostergaard 1999). TDI of 4-t-OP is suggested as 0.067 ng/kg bw/day and 33.3 ng/kg bw/day for men and women, respectively (Jonsson 2006). Consequently, the risk assessment for the of BPA, 4-NP, and 4-t-OP was estimated using the HQ as a risk descriptor, which was calculated according to Eq. (2):

$$\mathrm{HQ}=\left(\mathrm{EDI}\ \mathrm{calculated};95\mathrm{th}\right)/\left(\mathrm{ADI}\right)$$

The calculated EDI was considered safe if HQ ˂ 1 (EFSA 2015; Katsikantami et al., 2019).

Results

General information on the study population

In this study, 103 urine specimens were analyzed, 38 (36.89%) were males, 47 (45.63%) were females, and 18 (17.47%) were children. The age of subjects ranged from 3 to 51 years (Table 1), totally. The ages of the participants were as follows: males (21–47 years), females (19–23 years), and children (3–15 years).

Table 1 The characteristics of the study groups.
Full size table

BPA, 4-NP, and 4-t-OP urine levels

Information about method development and validation were obtained from our previous validation study (Battal et al. 2021). The linear ranges for BPA, 4-NP, and 4-t-OP were 5.0–500 ng/mL. The peak areas of analytes and their IS (13C12-BPA, 13C12-NP, and 13C6-OP) were determined with R2≥ 0.9997, R2 ≥ 0.9995, and R2 ≥ 0.9997 for BPA, 4-NP, 4-t-OP, and creatinine, respectively. The LOQs for BPA, 4-NP, and 4-t-OP were established at 0.001, 0.007, and 0.005 ng/mL, respectively. The BPA, 4-NP, and 4-t-OP concentrations were in the range of <LOQ-1.3634, <LOQ-0.0827, and <LOQ-0.1647 μg/g creatinine, respectively. The BPA, 4-NP, and 4-t-OP levels quantified are shown in Table 2. The range of concentrations for BPA, 4-NP, and 4-t-OP in males and females were given in Table 3.

Table 2 Urinary concentration of BPA, 4-NP, and 4-t-OP Turkish population (μg/g creatinine; n=103).
Full size table
Table 3 Concentration of BPA, 4-NP, and 4-t-OP among different groups (μg/g creatinine; n=103).
Full size table

Factors of influencing on BPA, 4-NP, and 4-t-OP levels

Association between BPA, 4-NP, and 4-t-OP levels in urine with data included in the questionnaire (sociodemographic, exposure to EDCs sources, lifestyle) of the participants is shown in Table 4. Age and BMI of the participants negatively correlates with BPA and 4-NPA levels. Similarly, there is a negative correlation between years of residency and the chemicals of interest.

Table 4 Descriptive analysis of metabolite levels in urine according to the lifestyle, sociodemographic characteristics, and food consumption habits of the participants.
Full size table

In addition, fast food consumption habits (1 meal/month) show a significant difference only for BPA.

Exposure and risk assessment

Using urinary levels of BPA, 4-NP, and 4-t-OP exposure and risk assessments for the population were carried out. The EDI for the BPA, 4-NP, and 4-t-OP was calculated as (geometric mean) 0.095 μg/kg bw/day; 0.041 μg /kg bw/day; and 0.091 μg /kg bw/day, respectively (Table 5).

Table 5 Estimated daily intake of BPA, 4-NP, and 4-t-OP in Turkish populations (μg/kg bw/day).
Full size table

Recommended TDI value of BPA is 0.4 μg/kg bw/day (European Food Safety Authority 2015). Institute of Food Safety and Toxicology Danish Veterinary and Food Administration has set a TDI value of 0.005 mg/kg bw/day for 4-NP (Nielsen and Ostergaard 1999). Furthermore, TDI value of 4-t-OP was suggested as 0.067 ng/kg bw/day and 33.3 ng/kg bw/day for men and women, respectively (Jonsson 2006). The 95th percentile daily intakes of BPA, 4-NP, and 4-t-OP were obtained in the present study (Table 6). These values indicate that BPA and 4-NP exposure levels show no risk for Turkish population. On the other hand, 4-t-OP exposure possesses a risk.

Table 6 Estimated daily intakes, tolerable daily intake, and hazard quotients of BPA, 4-NP, and 4-t-OP.
Full size table

Discussion

Due to their extensive and worldwide use, types of EDCs, BPA, 4-NP, and 4-t-OP are widely distributed in the environment and human beings who are exposed to these chemicals through drinking water, contaminated food, plastics, and detergents (Huang et al. 2016; Acir and Guenther 2018). Since people are likely to be exposed to these chemicals in their daily life, studies in which EDCs are measured in blood, urine, and breast milk have a place in the literature. This study is the first of its kind which simultaneously measure BPA, 4-NP, and 4-t-OP levels in Turkish population. There are only a few number of studies that measures urinary BPA, 4-NP, and 4-t-OP levels concurrently. Table 7 summarizes these studies. According to the table, the highest GM for BPA (5.55 ng/mL) concentration was in Zhou et al. (2013) study, and the lowest BPA value (0.5–3.9 ng/g) was in that of Van Overmeire et al. (2019). Our BPA and 4-t-OP values have the second lowest values among these studies. In addition, the lowest 4-NP level was obtained in our study (0.0177 μg/g creatinine), and the 4-NP levels of other studies are higher than our result. Due to the interferences from large amounts of endogenous compounds (e.g., proteins and phospholipids) in the matrix to be analyzed, the sample pre-treatment steps for trace analysis in the biological media consists important and often laborious steps. However, our study offers a great advantage for sample pre-treatment step of the analysis of BPA, 4-t-OP, and 4-NP chemicals in urine.

Table 7 Other studies examining BPA, 4-NP, and 4-t-OP exposures in populations.
Full size table

Nowadays, the number of published LC-MS/MS methods is constantly increasing with the development in analytical methodology of LC-MS/MS bioanalysis of BPA, 4-NP, and 4-t-OP as expected, especially in terms of sensitivity. According to the table, the LOD values that were obtained for BPA, 4-NP, and 4-t-OP in our validation study are more sensitive than other studies (Battal et al. 2021). While in our previous study (Battal et al. 2014a), both free and conjugated BPA values were measured, in this study, BPA values indicate total BPA which includes both BPA values. Possible correlations between concentrations compounds of interest and the factors affecting their existence were investigated. In accordance with the data presented in Table 4, age and BMI of participants negatively correlated with BPA and 4-t-OP exposure. Furthermore, a negative correlation was observed between bottled water usage and BPA exposure. In relation to the results of the questionnaire, only fast food consumption habits 1 meal/month for BPA (BPA, p=0.004) is associated with BPA concentrations. Apart from this parameter, no significant difference was observed with any other parameter in the questionnaire. Regarding the urinary levels of BPA, 4-NP, and 4-t-OP, an exposure and risk assessment for the population were carried out for those compounds. The BPA obtained in our study does not exceed the TDI (0.4 μg/kg bw/day) established by EFSA. According to calculations, EDI values for BPA and 4-NP were found to be below the limit set by authorities. Thus, HQ values of these chemicals were below 1 which is the safe limit. However, the situation is different for 4-t-OP, and the calculated HQ value was found to be much higher than 1. Along with other studies conducted on risk assessment of BPA, the results have similar findings with our study and that BPA does not pose a risk. Low-dose exposure to such chemicals is important, as studies have reported that potential endocrine disruptors for instance BPA often follow a non-monotonic dose-response curve and may cause greater effects at lower doses (Wolstenholme et al. 2011; Vandenberg et al. 2012; Beausoleil et al. 2013).

In contrast to BPA, articles concentrating on 4-NP and 4-t-OP exposure assessment and epidemiology studies on human population are limited. Our study has an important place in the literature as BPA, 4-NP, and 4-t-OP were analyzed in a single injection and risk assessment was also carried out. In line with the result, 4-t-OP exposure can be risky for the population. 4-t-OP are used as an intermediate for production of resins, non-ionic surfactants, and rubber additives and in manufacturing of antioxidants, fuel oil stabilizers, adhesives, dyes, fungicides, and bactericides.

It is predicted that the reason for high HQ value of 4-t-OP might be due to the lack of information on exposure limits specific to the 4-t-OP. Further studies are needed to investigate exposure limits of 4-t-OP. Even though there are plenty of worldwide studies conducted on BPA exposure, the number of studies based on BPA exposure of Turkish population is very limited (Table 8). Apart from one study, these studies of interest were all conducted on children. According to the table, the most intensive BPA exposure of control group (7.72±1.74 μg/g Cr) was found in the study carried out by Akgül et al. In our present study, the obtained findings (0.0079 μg/g Cr) are far below compared to other studies. Furthermore, Çok et al. (2020) calculated the EDI values of children exposure to BPA, and their geometric mean value was 0.035 μg/g Cr. In this study, the EDI value was calculated as 0.095 μg/g Cr. Even EDI values are higher in this study, this should be taken into consideration that our study population does not only include only children but also adults.

Table 8 BPA exposure studies conducted on Turkish population.
Full size table

Conclusion

Three endocrine-disrupting compounds (BPA, 4-NP, and 4-t-OP) were found in the urine from Turkish populations. BPA was the most frequently detected bisphenol, with concentrations ranging from values from <LOQ −0.1363 μg/g creatinine. For 4-NP and 4-t-OP were found <LOQ −0.0827 and <LOQ −0.1647 μg/g creatinine, respectively. The biomonitoring data was interpreted in a risk assessment context. Because the hazard quotient (HQ) value for 4-t-OP was above 1 for both genders, a high priority for risk management actions would be required for this substance in this study population. In addition, BPA and 4-NP did not exceed HQ. Exposure assessment and biomonitoring of BPA, 4-NP, and 4-t-OP are vital for protecting public health all around the globe. Thus, this study will be pioneer to other exposure and biomonitoring studies conducted in Turkish population. Larger and diverse group of participants might be selected from people who are highly exposed to these chemicals to aid the understanding of possible health outcomes.