A comprehensive study on bisphenol A and estrogenic activity in the Paraíba do Sul River, São Paulo, Brazil

The introduction of various substances into the environment, originally intended to enhance diverse aspects of the quality of life, has inadvertently contaminated natural resources, such as soil, water, and the atmosphere (Montagner et al. 2019). Despite the reduction in the emissions of some pollutants, such as pesticides, polychlorinated compounds, aromatic hydrocarbons, and heavy metals, in surface waters because of the approval of laws that regulate issues related to pollution, the presence of contaminants of emerging concern (CECs) has become a new challenge (Al-Odaini et al. 2013).

Pollutants that are classified as CECs are natural or artificial substances found in the environment that may cause adverse effects on humans or animals but are not commonly monitored (Madeira et al. 2023). Examples of CECs include pesticides, drugs, hormones, personal care products, and plasticizers (Montagner et al. 2019; Madeira et al. 2023). Generally, conventional domestic effluent treatment systems, consisting of preliminary, primary, and secondary treatments, cannot satisfactorily remove these contaminants (Ribeiro et al. 2015), as these contaminants are found in reduced concentrations of ng L⁻¹ or μg L⁻¹.

Among the CECs are endocrine disruptors, which are compounds that can interfere with the production or action of hormones, and can cause damage to the reproductive and immune systems of higher organisms, especially aquatic organisms. These substances can either block or mimic the activity of natural hormones, which can interfere with the reproductive systems of both humans and animals (Joseph et al. 2013).

Due to the large number of synthetic chemicals currently produced and the fact that the endocrine-disrupting potential of a compound cannot be readily determined by the functional groups present in its molecular structure, the identification and monitoring of endocrine disruptors are highly important (Wang 2016). Moreover, these substances can reach the environment from various anthropogenic sources and reach different environmental matrices (Martini et al. 2021).

The substance 2,2-bis(4-hydroxyphenyl) propane, known as bisphenol A (BPA), is obtained by combining two phenol molecules with one acetone at acidic pH and high temperatures. BPA is used to produce epoxy resins and polycarbonate plastics, phenol resins, polyacrylates and polyesters, and food packaging. It is considered a CEC of interest because it acts as an endocrine disruptor (Ohore & Songhe 2019).

Bisphenols, which mimic several hormones, such as estrogen, can also have health implications in humans, including sex-specific low neurodevelopment, uterine cancer, immune toxicity, neurotoxicity, and interference with cellular pathways (Ohore & Songhe 2019). In vivo experiments demonstrated that exposure to a dose of 100 μg kg⁻¹ of BPA caused high insulin levels, slight insulin resistance, and glucose intolerance in adult male mice, similar to the effects of 17-β-estradiol (E2) (Alonso-Magdalena et al. 2012).

BPA is also important because an estimated 10 million tons are produced annually, making it one of the most significant chemicals produced worldwide (United States Environmental Protection Agency 2023). BPA is used in the industrial production of plastic bottles, can linings, and thermal papers. Its presence in the environment has raised questions about its potential impacts, especially on children's health (Zhou et al. 2017).

The water pollution problem seems to be more acute in developing countries, such as Brazil, since untreated wastewater is frequently released into the environment, and this is the primary source of estrogenic compounds in surface waters (Tue et al. 2023). The Brazilian population (approximately 203 million inhabitants) is concentrated in urban areas (IBGE 2022). The resulting intense land use, insufficient sewage disposal, and inadequate treatment infrastructure negatively affect water quality (Dias et al. 2015).

The number of studies evaluating the occurrence of CECs in Brazil is limited. Nevertheless, some reviews on CECs in Brazil demonstrate the prominent presence of caffeine, drugs, hormones, and BPA in surface waters (Montagner et al. 2019; Starling et al. 2019; Madeira et al. 2023).

Only 29.2% of all generated domestic sewage is treated in Brazil (SNIS 2022). Most of the rest is released untreated into rivers, which are the population's primary water sources. Regarding the monitoring of CECs in Brazilian wastewater treatment facilities, studies have shown that conventional (up to the secondary treatment stage) treatment plants, even when they do exist, are inadequate for removing these compounds (Starling et al. 2019). Regarding BPA contamination in freshwaters used as a source of drinking water after treatment in Brazil, studies show its occurrence, especially in the most industrialized and urbanized states. In São Paulo, BPA concentrations have been reported by Martini et al. (2021) and Madeira et al. (2023), ranging from 6.5 to 1,300 ng L⁻¹ and 13.1 to 895 ng L⁻¹, respectively. In Rio de Janeiro, studies by Lopes et al. (2016) and Sabino et al. (2021) documented BPA levels ranging from 1.37 to 39.86 μg L⁻¹ and 0.03 to 1.16 μg L⁻¹, respectively. In Minas Gerais, Corrêa et al. (2021) and Ramos et al. (2021) reported concentrations ranging from 5.8 to 1587.8 ng L⁻¹ and 0.45 to 3.01 μg L⁻¹, respectively.

Although endocrine disruptors have harmful effects on organisms' health, Brazil has no specific regulation for their presence in water for human consumption (Starling et al. 2019). The difficulty of removing chemical substances such as emerging contaminants, including BPA, from sewage and water treatment plants represents a significant barrier to controlling their dissemination in the aquatic environment.

The conventional drinking water treatment process involves coagulation, flocculation, sedimentation, filtration, and disinfection stages and has low BPA removal rates. In Brazil, studies have demonstrated BPA removal rates in the range of 30% for conventional treatment (Lima et al. 2017). Although Teixeira et al. (2021) verified lower concentrations of BPA in treated water after disinfection than those in raw water, chlorination could generate chlorinated derivatives of BPA that could also cause health problems (Plattard et al. 2021). Considering that approximately 75% of the water that is treated in Brazil is treated by conventional treatment (IBGE 2017), BPA is likely to be found in drinking water for which the source is contaminated with BPA.

The Paraíba do Sul River, formed by the confluence of the Paraitinga and Paraibuna Rivers in the state of São Paulo, extends over 1,150 km, providing a critical water resource to 184 municipalities across São Paulo, Minas Gerais, and Rio de Janeiro. The basin sustains over 5.2 million residents (IBGE 2022) and is a key economic hub, contributing approximately 13% of Brazil's gross domestic product and supplying 85% of the water to the Rio de Janeiro metropolitan area, which supports over 12 million inhabitants (OECD 2017). There is also a recent water transfer from the Jaguari River (a Paraíba do Sul tributary) to the dams of the Cantareira System, which is responsible for the water supply for almost 9 million people in the São Paulo metropolitan area (ANA [sense data]).

Despite its economic and social significance, the Paraíba do Sul River basin is beset by severe environmental challenges. Water quality has significantly deteriorated, largely due to inadequate waste management practices, widespread deforestation, unregulated extraction of mineral resources, and the excessive application of pesticides. Compounding these issues, approximately 1 billion liters of untreated domestic sewage are discharged daily into the river, underscoring the urgent need for enhanced sanitation infrastructure (CEIVAP [sense data]). Given the basin's crucial role in supplying drinking water to Brazil's largest metropolitan regions and the ongoing conflicts over water use (ANA 2016), there is a pressing need for further studies to evaluate and improve water quality within this essential watershed.

Research on CECs in the Paraíba do Sul River Basin is sparse, with limited studies indicating BPA contamination, particularly in the São Paulo section. Notably, a study found BPA in 58% of water samples from key cities along the river (Souza et al. 2011). However, there is a critical need for comprehensive research to assess the extent of BPA contamination, its seasonal dynamics, and correlations with other water quality parameters. Such research is vital for informing policies and protecting both public health and the sustainability of this essential water resource.

Hence, this study aimed to investigate BPA contamination, specifically within the São Paulo State stretch of the Paraíba do Sul River. The primary objective is to comprehensively assess the extent of BPA pollution in this region. Additionally, we aim to analyze the level of estrogenic activity to gain insights into potential endocrine-disrupting effects. Through this investigation, we seek to elucidate seasonal variations in BPA concentrations and identify possible sources of contamination contributing to its presence in the river.

CETESB, the São Paulo State Environmental Agency, has operated the Surface Water Quality Monitoring Program since the 1970s. This program involves testing the chemical, microbiological, hydrobiological, and toxicological aspects of surface water samples collected from various locations across the state of São Paulo, Brazil (Roubicek et al. 2020). Currently, it comprises more than 500 sampling sites from different state river basins. The analyzed and interpreted data are available in Portuguese to the general public in annual reports on their website. CETESB uses all the results to prioritize pollution control actions, identify degraded sites, and verify the conditions for aquatic life protection and other water uses, such as irrigation, livestock, and recreation. As part of the water quality evaluation, CETESB calculates, among other indices, the water quality index (WQI) for each site.

The quality variables, which are part of the WQI calculation, mainly reflect the contamination of water bodies by the release of domestic sewage. It comprises the following parameters, including w_i, the limit of detection (LOD), and the limit of quantification (LOQ): Escherichia coli (w_i = 0.15, LOQ = 1 Colony Formation Unit (CFU) (100 mL)⁻¹), biochemical oxygen demand (BOD) (w_i = 0.10, LOD = 1 mg L⁻¹, LOQ = 2 mg L⁻¹), total nitrogen (w_i = 0.10, LOD = 0.1 mg L⁻¹, LOQ = 0.5 mg L⁻¹), total phosphorus (w_i = 0.10, LOD = 0.003 mg L⁻¹, LOQ = 0.02 mg L⁻¹), temperature (w_i = 0.10, LOD = 0.1 °C, LOQ = 1 °C), pH (w_i = 0.12, LOD = 0.03, LOQ = 0.1), turbidity (w_i = 0.08, LOD = 1.0 Nephelometric Turbidity Unit (NTU), LOQ = 5 NTU), total solids (w_i = 0.08, LOD = 10 mg L⁻¹, LOQ = 50 mg L⁻¹), and dissolved oxygen (DO) (w_i = 0.17, LOD = 0.1 mg L⁻¹, LOQ = 0.5 mg L⁻¹) (CETESB 2023). The analytical methods and curves for the WQI are described in the Supplementary Material. As the WQI variables were collected simultaneously as the sample collections for BPA quantification and estrogenic activity essay, this study used the WQI calculated for correlation purposes for the sites studied.

The São Paulo stretch of the Paraíba do Sul River comprises its formation at the confluence of the Paraitinga and Paraibuna Rivers to the Rio de Janeiro State border. With more than 2 million inhabitants in 39 municipalities (IBGE 2022), this is the first stretch of the Paraíba do Sul River before it enters the states of Rio de Janeiro and Minas Gerais.

Water samples were collected using 1 L amber flasks for each analysis, BPA quantification, and estrogenic activity assay and preserved according to Standard Methods for the Examination of Water and Wastewater (APHA 2023). For comparison purposes, the sampling campaigns for the estrogenic activity assay and BPA analysis were run simultaneously with the sampling of the CETESB's Surface Water Quality Monitoring Program. Each sample collection obtained all the complementary parameters for WQI calculations.

Samples for BPA analysis and estrogenic activity assays were previously concentrated using solid-phase extraction.

Samples were extracted using an automated system (SPE DEX 4790, Horizon Technology, Salem, NH) coupled with Oasis HLB cartridges obtained from Waters (Milford, MA, USA) in accordance with U.S. EPA Method 1694 (United States Environmental Protection Agency 2008). The cartridges were conditioned with 15 mL of methanol and 10 mL of water. About 1 L of water sample was extracted. The cartridge was dried for 20 min using nitrogen, and the analytes were eluted with 15 mL of methanol. The eluate was evaporated using a Genevac EZ-2 evaporator, and the sample was reconstituted in 1 mL of 10% dimethyl sulfoxide (in water) for the BLYES assay.

For BPA analysis, the sample was reconstituted in 1 mL of methanol for subsequent analysis by liquid chromatography coupled with sequential mass spectrometry.

The 1200 chromatographic system from Agilent (Santa Clara, CA, USA) connected to a 6410 triple quadrupole mass spectrometer from Agilent (Santa Clara, CA, USA) was used for liquid chromatography–tandem mass spectrometry analysis. Precisely 10 μL of the sample was injected into a ZORBAX SB-C18 column (30 mm × 2.1 mm, 3.5 μm particle size) from Agilent (Santa Clara, CA, USA) at 30 °C. The mobile phase used for the analysis consisted of 0.01% (v/v) ammonium hydroxide in water and methanol, which was previously filtered through a 0.2 μm porous membrane at a flow rate of 0.3 mL min⁻¹. The following gradient was used: 0 min 30%, 3 min 70%, 6 min 90%, 12 min 90%, and 13 min 30% methanol. The chromatographic column effluent was directed to the mass spectrometer through the electrospray ionization source, and it was monitored by selected reaction monitoring. The ionization source was operated in negative ion mode. The instrumental limits of detection and quantification were obtained by the signal-to-noise ratio (SNR) method, which compares the analytical signal of the standards with the baseline noise. The LOD was determined with an SNR of 3:1, and the LOQ was determined with an SNR of 10:1, with values of 5 and 10 μg L⁻¹, respectively. Santos et al. (2022) and Madeira et al. (2023) describe the quality data and recovery tests.

Estrogenic activity was determined using Saccharomyces cerevisiae BLYES strains (Bioluminescent Yeast Estrogen Screen). Yeast cells were cultured, and the assay was carried out as described by Sanseverino et al. (2005). S. cerevisiae cells do not naturally have receptors for estrogens or androgens. This BLYES yeast strain is genetically modified to respond to estrogenic agents by the insertion of the gene to express the human estrogen receptor (hER-α). In this system, the hER-α produced binds to estrogen response elements, which have been inserted into a strong promoter on a plasmid. This promoter controls the expression of the luxA and luxB genes, which were originally obtained from luminescent bacteria, was also inserted into the plasmid and are responsible for luminescence production. Yeast can emit light when in contact with a substance that binds to hER, activating the promoter (Sanseverino et al. 2005).

The sample extracts and a 17-β-estradiol (E2) standard solution were serially diluted in dimethylsulfoxide, and 100 μL was transferred to a 96-well microtiter plate. Next, 100 μL of seeded yeast medium was added to the wells. The plates were then sealed and shaken on a vortex for approximately 20 s. After incubation for 4 h at 30 °C, the luminescence was read using a plate reader (Victor X3). In each test, water, dimethylsulfoxide and E2 (serially diluted) were used as blank solvents and positive controls.

The results were expressed quantitatively through the equivalent estrogenic activity (ng L⁻¹ of estradiol-equivalent (EEQ)) using effect curves produced with the tested concentrations of 17-β-estradiol (CETESB 2020). The LOD and LOQ are 0.03 and 0.1 ng L⁻¹ of estradiol-equivalent or EEQ, respectively.

Dias et al. (2015) evaluated the estrogenic activity at the Paraíba do Sul River in the Rio de Janeiro State stretch. According to these authors, the levels of estrogenic activity in the water samples from the Paraíba do Sul River appear to rise as the river passes through the municipalities. This behavior is also observed in our evaluation, in which the three last municipalities present higher results than the three first cities. An increase in estrogenic activity is observed after the most populated cities, decreasing after the town of Tremembé, where the least populated cities are located (Table 1).

The highest concentration of BPA (116.9 ng L⁻¹) was detected at Queluz. Despite being a small city with no relevant industrial activity, it is located downstream of industrialized municipalities with lower levels of wastewater treatment in the region, such as Guaratinguetá and Cruzeiro, which present remaining loads of organic matter of 4.9 and 4.2 tons of BOD_5,20 per day, respectively (CETESB 2023).

The Tremembé site showed positive results in 100% of the samples, ranging from 29.5 to 47.7 ng L⁻¹, probably due to its location downstream of São José dos Campos, which is the region's most populated, urbanized, and industrialized municipality. Although the municipality treats almost 94.4% of its urban sewage, the remaining loads of organic matter are 7.2 tons of BOD_5,20 per day (Table 1).

In 2022, the remaining loads of organic matter in sewage in the São Paulo stretch of the Paraíba do Sul River were estimated to be 37 tons of BOD_5,20 per day (CETESB 2023).

The results presented in Figure 4(b) show that there is no evident seasonal influence on BPA concentrations because May (autumn) and August (winter) are dry seasons (less than 100 mm of rain monthly), and February (summer) and October (spring) are rainy seasons (more than 100 mm of rain monthly). The highest concentrations of BPA occurred in the May and October campaigns. Czarczyńska-Goślińska et al. (2017) also did not find a correlation between BPA concentrations and seasonal influence in their study in Poland, with concentrations between 5 and 95 ng L⁻¹, similar to the Paraíba do Sul River, which has concentrations ranging between 11.1 and 116.9 ng L⁻¹.

Montagner et al. (2019) evaluated the presence of 58 emerging contaminants in the state of São Paulo between 2006 and 2015. BPA was quantified in almost 67% of samples analyzed in a wide range of concentrations, between 2 and 13,016 ng L⁻¹. According to the authors, the highest BPA levels were observed at sampling sites influenced by industrial and densely populated areas. In their study, Martini et al. (2021) also concluded that regions with more industrial activity have considerably higher BPA concentrations.

Table 2 compares the highest levels of BPA contamination quantified in the Paraíba do Sul River with those in other regions of Brazil and with those of international papers that evaluated BPA contamination in surface waters between 2015 and 2021. The analyses show that the BPA levels reported in this study are similar to those in most other monitored Brazilian rivers. The Paraíba do Sul River also shows BPA contamination levels similar to those of the South Platte River in the United States (Bai et al. 2018) and the Hogsmill River in the United Kingdom (Wilkinson et al. 2017).

Studies have shown that BPA binding to the estrogen receptors ERα and ERβ is low, with an affinity 10,000-fold lower than that of E2 for both ER types. However, it was demonstrated that BPA can elicit estrogen-like effects with the same potency as 17-β-estradiol (E2) because classical ERs interact with other transcription factors (Nadal et al. 2018). According to Nadal et al. (2018), the evidence of different pathways presented challenges the concept of BPA as a weak estrogen with no effects at low doses (1 nM). This is particularly significant because in vivo animal studies have demonstrated that BPA can interfere with endocrine signaling pathways at low doses during fetal, neonatal, and perinatal periods, as well as in adulthood (Alonso-Magdalena et al. 2012).

Since BPA has different action pathways than the h-ERa receptor used in estrogenic activity, it has a relative potency of only 2.26 x 10⁻⁷ compared to E2 in the BLYES assay (Sanseverino et al. 2009). Applying this potency factor, it is possible to calculate the expected estrogenic activity as a function of the measured BPA concentration. This factor was used to calculate BPA concentrations to determine BPA's contribution to estrogen activity in the BLYES assay. The maximum quantified concentration of BPA (116.9 ng L⁻¹) contributed to the estrogenic activity at 2.64 × 10⁻⁵ ng L⁻¹ of estradiol-equivalent. Compared with the 0.62 ng L⁻¹ estradiol-equivalent quantified in the BLYES assay, the BPA contribution to the BLYES assay corresponded to only 0.004% of the total estrogenic activity. The relative potency of BPA cannot fully explain the biological response, which indicates that the BLYES assay could have detected other substances, including natural and synthetic hormones found in surface water that were not chemically analyzed in this study (Di Dea Bergamasco et al. 2011).

Montagner et al. (2019) assessed the risk of surface waters using the criteria of the chronic toxicity standard, which for BPA is 240 ng.L⁻¹ (Swiss Centre for Applied Ecotoxicology 2016), to evaluate the effects of concentration on aquatic organisms. As all the results of the Sao Paulo state stretch of the Paraíba do Sul River in this study were below this criterion, it demonstrates no expected risk for water use.

Due to the importance of the Paraíba do Sul River as a water supply source in Brazil's most urbanized and industrialized states, it is necessary to continue monitoring BPA and other CEC to identify other endocrine disruptors that can contribute to the levels of estrogenic activity in the Paraíba do Sul River.

All results from the quality monitoring of the sites studied are presented in the Supplementary Material. CONAMA Resolution 357/2005 is a Brazilian law that classifies water bodies according to their uses and establishes water quality standards. According to this classification, water bodies that receive wastewater discharge should retain their characteristics and quality standards. The Paraíba do Sul River is classified as a Class 2 river (it can be used for drinking water after conventional treatment, it should protect aquatic life, and it can be used for recreation, irrigation, aquaculture, and fishing), and is expected to meet the quality standards for the multiple uses for which it is intended (CONAMA 2005).

Except for the municipality of Santa Branca, all sites showed E. coli results above the quality standard. The highest E. coli and BOD results occurred in Aparecida, which was expected because this municipality does not have a sewage treatment facility. It is possible to observe the increasing load of phosphorus along the river, with the Tremembé, Aparecida, and Queluz sites presenting all the results above the regulated quality standard. Concerning DO, the worst results occurred in the stretch between São José dos Campos and Aparecida.

These results demonstrate that releasing untreated sewage or effluents with inadequate treatment is a problem along the São Paulo stretch of the Paraíba do Sul River.

According to Chidiac et al. (2023), the WQI is one of the most commonly used tools for describing water quality by reducing the bulk of information (physical, chemical, and biological factors) into a single value ranging between 0 (poor quality) and 100 (best quality). The calculated WQI for each campaign and the WQI mean value for 2022 are shown in Table 3.

Table 3

WQI of the São Paulo stretch of the Paraíba do Sul River in 2022 (CETESB 2023)

View Large

February and October represent the rainy seasons, during which we observed WQI results classified as ‘fair’, and the Tremembé and Aparecida sites presented the worst results among all the campaigns. Comparing the WQI in 2022 with the WQI indices of the previous 5 years (Table S2—Supplementary Material), the WQI 2022 values were lower, probably because the rainfall intensity was up to 10% greater in 2022 than the historical average (CETESB 2023).

The linear correlation (R²) between the WQI mean values for 2022 and the BPA and estrogenic activity mean concentrations at each site were 0.4282 and 0.9157, respectively [Figure 7(c) and 7(d)]. As the WQI parameters primarily indicate water body contamination from domestic sewage discharge, it is possible to infer that industrial wastewaters constitute an important share of the contributing factors to BPA contamination since the region is highly industrialized, including several chemical and petrochemical, metallurgical, paper, and food industries. However, estrogenic activity is strongly correlated with inadequate disposal of domestic sewage, especially because other natural estrogens like 17β-estradiol (E2), 17α-ethinylestradiol (EE2), estrone (E1), and estriol (E2) are found in surface waters with domestic sewage contamination (Di Dea Bergamasco et al. 2011; Dias et al. 2015).

This study underscores the pressing issue of water pollution, mainly focusing on the emerging contaminant BPA and estrogenic activity in the Paraíba do Sul River in São Paulo, Brazil. This study quantified BPA concentrations (11.1−116.9 ng L⁻¹) and assessed estrogenic activity (0.12–1.36 ng L⁻¹ of estradiol-equivalent) using the BLYES assay, both of which showed positive results in 50% of the samples. However, the absence of a significant correlation (R² = 0.01783) between BPA concentrations and estrogenic activity suggests the presence of other compounds contributing to estrogenic effects, emphasizing the need for a comprehensive approach to monitoring water quality. Furthermore, the water quality evaluation using the WQI revealed compromised water quality at the studied sites, particularly during the rainy season. The correlation between the WQI and estrogenic activity (R² = 0.9157) suggested that untreated domestic sewage disposal could be an important source of endocrine-disrupting compounds. Concerning BPA contamination, its correlation with the WQI (R² = 0.4282) demonstrates that it is necessary to continue studies to track the sources of pollution, especially in this highly urbanized and industrialized region. This research underscores the importance of continuous monitoring, regulatory measures, and public awareness to preserve water resources and protect public health. The findings contribute to the growing knowledge of emerging contaminants, urging decision-makers and environmental agencies to take proactive steps to safeguard water resources and public health.

The authors acknowledge the following institutions that endorsed this study: CETESB – São Paulo State Environmental Agency of CETESB - São Paulo State Environmental Agency, São Paulo State University (UNESP), and University of Campinas (UNICAMP).

Part of the funding support was provided by the National Institute of Advanced Analytical Sciences and Technologies (INCTAA, CNPq Grant N. 465768/2014-8) and the São Paulo Research Foundation (FAPESP Grant N. 2014/50951-4).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.