A novel method for rapid and quantitative detection of bisphenol A in urine*

1Institute of Nursing and Midwifery, Medical University of Gdańsk, Gdańsk, Poland; 2Department of Analytical Chemistry, Faculty of Chemistry, Gdańsk University of Technology, Gdańsk, Poland; 3Faculty of Natural Sciences and Technology, Karkonosze College in Jelenia Góra, Jelenia Góra, Poland; 4Department of Toxicology and Health Risk Assessment, National Institute of Public Health – National Institute of Hygiene, Warsaw, Poland; 5DetoxED Ltd, Gdańsk, Poland


INTRODUCTION
Bisphenol A (BPA) is one of numerous endocrine disrupting chemicals (EDCs) or the so-called endocrine disruptors (EDs). The United States Environmental Protection Agency defines ED as an "exogenous agent that interferes with the production, release, transport, metabolism, binding, action, or elimination of natural hor-mones in the body responsible for the maintenance of homeostasis and the regulation of developmental processes" (Kavlock et al., 1996). Similarly to some steroid hormones, the structure of BPA contains aromatic rings; hence, it can interact with a variety of receptors for hormones and in consequence lead to hormonal disruption and increase the risk of various adverse health effects (Di Donato et al., 2017;Sifakis et al., 2017).
Widespread use of BPA-containing products, together with its ability to migrate from these materials, may explain the detectable levels of BPA in laboratory specimens that include serum, urine, amniotic fluid, milk, placenta, and neonatal blood (Vandenberg et al., 2010;Mendonca et al., 2014;Lee et al., 2018). Since humans are permanently exposed, there is an indisputable need for reducing the exposure to BPA (Gore et al., 2015).
In addition, special attention should be given to pregnant women, as elevated concentrations of EDs in the amniotic fluid may disrupt the endocrine balance crucial for proper development of the gonads and can increase the risk of congenital anomalies of the urogenital tract (Braun, 2017;Pergialiotis et al., 2018). Permanent exposure may also alter epigenetic mechanisms, resulting in a higher risk of alterations in the offsprings even in the third generation (Mileva et al., 2014, Rutkowska et al., 2016Ideta-Otsuka et al., 2017).
Despite the high relevance of biomonitoring and efforts aimed at reducing the exposure, no low-cost and easily applied method for rapid determination of BPA levels in a patient's body has been yet developed. Currently available methods for quantitative detection of EDs can be divided into three groups. The first consists of highly sensitive chromatographic techniques, such as gas chromatography (GC) or high-performance liquid chromatography (HPLC); often coupled with mass spectrometry, tandem mass spectrometry, or solid-phase microextraction (Sun et al., 2016;Jurek et al., 2017;Jurek et al., 2018;Dreolin et al., 2019).
The second group consists of bioassays used to determine the endocrine disruption potential of chemicals, their damaging effect on DNA (genotoxicity), and generation of oxidative stress. It includes various methods and the most commonly used are CALUX (Chemically Activated LUciferase gene eXpression), E-SCREEN, and YES (Yeast Estrogen Screen) (Park et al., 2008;Valitalo et al., 2016;Aneck-Hahn et al, 2018;. The third group includes enzyme immunoassays used to visualize and quantify antigens, such as ELISA (enzymelinked immunosorbent assay) (Lee et al., 2017). Vol. 67, No 3/2020 409-415 https://doi.org/10.18388/abp.2020_5368 All of the above-mentioned techniques require access to a fully equipped specialized laboratory and involve experienced chemists; hence, they are also expensive and time-consuming.
Another method that can be used for EDs determination is immunochromatography. Using it in a lateral flow rapid test makes it a low-cost and simple-to-use method that can easily be performed outside a laboratory, even by the patients themselves in the outpatients' clinics or at home.
Although analytical methods are being developed with greater simplicity and efficiency, higher selectivity and sensitivity, lower sample and solvent consumption, and greater automation routine, analysis of BPA levels in patients still poses great challenges (Sun et al., 2016;Dhanjai et al., 2018), especially since the serum/whole blood concentrations may often not predict the total body burden.
Here, we describe an engineered fast and inexpensive lateral flow assay (LFA) of our design for BPA detection in urine samples. This is a first powerful analytical tool that can be ready-to-use for biomonitoring without the need for an access to a laboratory and scientific equipment. Its unique advantage is the possibility to determine if the individual exposure to BPA is low, medium, or high, which may be of key importance from the public health perspective and human biological monitoring.

Materials. Chemicals. Phosphate-Buffered Saline
(PBS), Bovine Serum Albumin (BSA), β-glucuronidase enzyme, as well as the BPA standard and all LC chromatography reagents were ordered from Sigma Aldrich (USA). Innova Bioscience Latex Conjugation Kit was from Expedeon (formerly Innova BioScience, United Kingdom).
Antibodies. Rabbit polyclonal antibodies to BPA, and goat polyclonal antibodies to the rabbit antibodies were from Antibodies-online GmbH (Germany).
LFA constructions. All membranes: nitrocellulose, glass and cotton used in LFA were purchased from Ahlstrom-Munksjö (Finland). Color latex beads were from Expedeon (formerly Innova BioScience, United Kingdom).
Urine samples. Forty healthy volunteers (age range 2-40 years old) including males, females (also pregnant), and children had mid-stream urine samples collected into glass vials (previously autoclaved and sterilized) after an overnight fast. All samples were frozen at -70°C until analysis.
Preparation of the lateral flow strips and test run. The lateral flow strips were created as a combination of different types of membranes, polyclonal antibodies to BPA -both adsorbed on membranes and on color carrier beads, and antibodies to the anti-BPA antibodies. A urine sample was being applied onto the sample pad (1) and then it flowed through an overlapping conjugate pad (2) onto which color carrier beads, coated with polyclonal antibodies to BPA, had been placed earlier.
Further, the sample flowed through a membrane within which the urine sample and the color carrier beads, including those bound to the BPA present in the sample, can freely migrate (3). The membrane contains three indication areas covered with immobilized polyclonal antibodies to BPA (5) and one dedicated control area covered with immobilized polyclonal antibodies to antibodies to BPA, i.e. the control line (6). The first BPA indication area (the closest to the sample pad) detects a concentration that corresponds to low BPA exposure. Appearance of color in the first two areas indicates intermediate BPA exposure, whereas its appearance in the three consecutive BPA indication areas corresponds to high BPA exposure. A control area (the control line) contains antibodies to anti-BPA antibodies and the appearance of color in these lines upon application of the biological material indicates a proper test functioning. Migration of the sample and the unbound latex beads coated with antibodies stops at the termination area of the test (4) -the absorbent pad, where the adsorption of the sample occurs. The backing card, free from BPA and its derivatives (7), is an additional element that supports the membranes and facilitates the performance of the test outside a laboratory; however, it was skipped during laboratory validation.
A nitrocellulose membrane of 240-280 µm in thickness was selected for the membrane of sample migration (3). Rabbit polyclonal antibodies to BPA, previously placed in PBS with 2% BSA, were applied onto the membrane. The amount of the antibodies corresponded to the detection of BPA concentrations of the maximum values of 2 ng/mL for the first, 3 ng/mL for the second, and 10 ng/mL for the third indication area. Goat polyclonal antibodies to rabbit antibodies to BPA were applied at the end of the membrane to form the control line of the test. A glass fiber membrane was selected for the sample pad (1) and the conjugate pad (2), whereas cotton fiber membrane was selected for the absorbent pad (4). The membrane was impregnated with PBS with 2% BSA and dried for 24 hours with subsequent application of 40 µL of 1% solution of latex beads coated with rabbit antibodies to BPA (Expedeon, formerly Innova BioScience, United Kingdom).
Conjugation of color carries beads to antibodies. The process of conjugation was conducted by using the reagents and protocol according to the Innova Bioscience Latex Conjugation Kit manual. Briefly, the stock of polyclonal antibodies to BPA was diluted to 0.1 mg/ mL and then added to 400 nm blue latex nanoparticles. After 15 minutes of incubation at room temperature, the conjugation reaction was quenched. Then, the samples were centrifuged at 10 000 rpm for 6 minutes. The conjugation buffer was removed, and the pellet was resuspended in a buffer with the addition of 0.1% BSA.
Determination of BPA concentration in urine samples and estimation of individual exposure. A small volume of urine was poured into a container made from materials free from BPA and its derivatives. Then, 2 mL were transferred into a glass receptacle included in the test kit with a micropipette, up to the level marked with a horizontal line visible on the side wall of the receptacle. ß-glucuronidase enzyme was sprayed onto the bottom and walls of the receptacle in order to catalyze hydrolysis of BPA glucuronide from urine; hence, to obtain the free form of BPA. The enzyme activity was at least 20 000 units per gram, and it was added at a concentration of 20 µL per 1 mL of urine. Afterwards, 100 µL of the sample collected from the receptacle was applied on the sample pad and flowed to the area with immobilized color carrier beads where free BPA was bound by antibodies on the color latex beads. The urine sample, the unbound BPA, the beads with bound BPA, and the beads that did not bind BPA migrated onto the nitrocellulose membrane. According to the parameters of the membrane, the average migration rate was 150 s/4 cm.
A scheme of the procedure of urinary BPA determination using LFA is shown in Fig. 1.
The BPA quantification results using lateral flow strips were verified and validated using liquid chromatography-tandem mass spectrometry (LC-MS/MS) (Shimadzu LCMS-8050, Japan). Each determination of BPA concentration in the urine sample was conducted in triplicate. The final results of BPA quantification in urine samples were expressed as an arithmetic mean of three measurements and the uncertainty of measurement; the procedure has been already published elsewhere (Wilczewska et al., 2016).

RESULTS
The colored indication area in LFA pointed to BPA detection in the urine sample. The concentrations of detected BPA sum up; thus, the appearance of color in one, two, or three BPA indication area(s), starting from the sample pad, corresponds to a level of up to 2 ng/ mL (low BPA concentration), up to 5 ng/mL (moderate BPA concentration), and up to 15 ng/mL (high BPA concentration), respectively. Proper functioning of the test (migration of carrier beads correctly coated with antibodies to BPA) was confirmed by the appearance of blue color in the control area. Figure 2 Fig. 3 were obtained for the same samples of urine using LC-MS/MS during assay validation to confirm the results of the assay. Addition of β-glucuronidase to urine samples improved the LFA sensitivity. An example of chromatographic analysis results for the content of free BPA in the urine samples from healthy women (aged 21 to 24) before and after addition of β-glucuronidase (incubation with 20 µL per 1 mL of urine) are summarized in Table 1. Table 2 presents concentrations of BPA in all urine samples measured via chromatography after addition of the enzyme, divided into levels with clear and unclear LFA results. LFA could give unclear/false results for the lowest concentrations of BPA detected in the samples. Although addition of β-glucuronidase has increased the level of free BPA in the sample and the sensitivity of the assay, there was still a risk for troubles in detection of the lowest BPA concentrations. Nevertheless, we believe that the uncertainty of some of the LFA results may be explained by manual application of antibodies on the membrane.

DISCUSSION
Constant human exposure to BPA may increase the risk of developing endocrine and metabolic disorders, obesity, diabetes mellitus, cardiovascular disease, infertility, and endocrine-dependent tumors in both, women and men (Street et al., 2018). In 2015 EFSA suggested   Although BPA has been removed from most consumer products and human exposure should be dropping among the population in the future, it is still present in the daily lifestyle; thus, biomonitoring is of human and environmental benefit. Table 3 presents a comparison of advantages and drawbacks of the currently known and Semi-quantitative.
The LFA technology presented here makes it possible to conduct a rapid diagnostic test without access to a central laboratory and it can be performed either by medical staff or by the patient at home. The addition of β-glucuronidase improves the sensitivity of detection, as it leads to a release of free BPA from glucuronide complexes in the urine. It is estimated that in the European Union only, exposure to EDs may substantially contribute to diseases and dysfunction across individual life courses, with costs of hundreds of billions of EUR per year. Without a doubt, its prevention can benefit the economy, but most importantly-the human life (Trasande et al., 2015). Thus, our invention may be a powerful and helpful analytical device for biomonitoring, as well as a concept to be used for preparing other LFAs for EDs detection; and therefore, it may help lowering the risk of some lifestyle diseases that have been linked with exposure to BPA (Gerona et al., 2020).