Exposure to Bisphenol AF during embryonic development causes feminization in male chickens Mimmi Wänn

(1)

Exposure

to

Bisphenol

AF

during

embryonic

development

causes

feminization

in

male

chickens

Mimmi

Wänn

(2)

Abstract

The exposure to endocrine disrupting chemicals exerts an imminent threat to wildlife and mankind, causing adverse effects, as for instance malformations and functional impairment of the reproductive system. Bisphenol A (BPA) has been extensively used in the plastic industry for decades and is known to hold appreciated characters as great flexibility and resistance. Leakage of BPA has unfortunately been detected and both human and wildlife get

continuously exposed to low concentrations of this xenoestrogen in our everyday life. Studies investigating the estrogen-like influence of BPA exposure in birds, demonstrated altered development in the reproductive organs during sexual differentiation. Today there are

numerous bisphenols on the market, serving as BPA alternatives due to their similar structure and function. Bisphenol AF (BPAF) is one of many BPA analogues that was chosen for further evaluation in a survey by Kemikalieinspektionen in 2011.

The major aim of this study was to examine the impact of BPAF during the embryonic development of the chicken embryo and focused on gonadal differentiation endpoints. Injection into the air chamber of the incubated eggs was executed during embryonic

development and examination performed prior to anticipated hatching. Causation of increased mortality induced by BPAF was statistically significant, established at 70 μg/g egg and 175 μg/g egg. Male right:left testes area ratio was statistically significant, affected by treatment dose 70 μg/g egg, as well as liver weight and liver:body weight ratio, hepatosomatic index. No effects of treatment were found in females. Gross morphology revealed an increasing frequency of feminization in male testes, which was statistically significant at 175 μg/g egg. Feminization was characterised by ovarian-like appearance of the left testis. Histological analysis only confirmed ovotestis in one individual, the same individual showing pronounced characteristics of ovotestis in the gross morphology assessment. This individual displayed features of ovarian-like cortex with abundant clusters of oocyte-like germ cells, and lacunae in the medulla. The results from our study indicate a similar estrogenic treatment effect of BPAF, as for BPA, during early-life exposure to the developing chicken embryo. Hence human and wildlife health concerns should be further addressed regarding potential exposure of BPA alternatives in our everyday life.

Keywords: Bisphenol AF; ovotestis; feminization; sex differentiation; endocrine disruptors;

(5)

Svensk sammanfattning

Exponeringen för hormonstörande ämnen utgör ett överhängande hot mot mänskligheten och djurlivet. Dessa ämnen orsakar skadliga effekter, som exempelvis missbildningar och

funktionell nedsättning av reproduktionsorganen. Bisfenol A (BPA) har sedan flera årtionden använts i stor utsträckning av plastindustrin och är känt för egenskaper så som hög flexibilitet och tålighet. Olyckligtvis har läckage av BPA upptäckts i vår vardag och både människor och djur blir kontinuerligt utsatta för låga koncentrationer av denna xenoöstrogen. Studier, som har undersökt östrogenlik inverkan vid exponering av BPA hos fåglar, uppvisade förändrad utveckling av reproduktionsorganen under könsdifferentieringen. Idag finns det ett flertal bisfenoler på marknaden, vilka tjänar som alternativ till BPA tack vare liknande funktion och strukturell uppbyggnad. Bisfenol AF (BPAF) är en av många BPA-analoger som blev utvalda för vidare undersökning om dess toxikologiska effekter i en kartläggning av

Kemikalieinspektionen 2011.

Syftet med denna studie var att undersöka inverkan av BPAF under embryonalutvecklingen hos kycklingembryo och fokuserade på gonaddifferentiering som process och slutpunkt. De befruktade hönsäggen injicerades in i luftkammaren och embryot undersöktes före förväntad kläckning. BPAF inducerade ökad mortalitet, som var statistiskt signifikant vid 70 μg/g ägg och 175 μg/g ägg. Förhållandet mellan hanlig höger:vänster testikelarea, levervikt och förhållandet mellan lever:kroppsvikt (hepatosomatiskt index) påverkades vid 70 μg/g ägg. Inga behandlingseffekter återfanns hos honor. Morfologisk undersökning avslöjade en ökande feminiseringsfrekvens av testiklar hos hanar, vilket var statistiskt signifikant vid 175 μg/g ägg. Feminiseringen karakteriserades av ovarielikt utseende hos den vänstra testikeln. Histologisk analys konfirmerade endast ovotestis hos en individ, samma individ som uppvisade markanta karaktärsdrag av ovotestis vid den morfologiska bedömningen. Denna individ visade

egenskaper av ovarielikt kortex med rikligt förekommande kluster av oocytlika könsceller och lakuner i medullan. Resultaten från vår studie ger indikationer om liknande, som för BPA, östrogena behandlingseffekter orsakade av BPAF vid exponering under de tidiga

(6)

Introduction

The following paragraph is supported by information from the assessment: “State of the

Science of endocrine disrupting chemicals – 2012”, carried out by the United Nations

Environment Programme (UNEP) and the World Health Organization (WHO) (2013). Human health, growth and reproduction are dependent on a prosperous endocrine system. Endocrine disrupting compounds (EDCs) are of great concern today as we learn more about their impact to human and wildlife. The incidence of EDC-related disorders shows an increasing trend. Hormone synthesis, conversion and receptors are known targets for EDCs. Children, especially, are considered to be more frequently exposed than adults due to their regular hand-to-mouth activity. The embryonic period includes critical developmental stages, in both human and wildlife, that appear to be the most sensitive ones to endocrine disruption. A prenatal exposure of an embryo to EDCs, can cause adverse toxic effects like structural and functional abnormalities. Functional impairment of reproductive replicability in both male and female is a known effect of EDCs. Malformations, eggshell thinning and mortality are proven consequences in the bird embryo due to exposure of anthropogenic chemicals polluting the environment. EDCs are defined as exogenous compounds causing adverse health effects when disrupting and altering the endogenous endocrine balance. EDCs might not only affect the organism itself, but it can also cause adverse effects to the progeny. Commonly, hormones and EDCs display a non-linear dose response (sigmoidal or even more complex) and action of endocrine disruption is known not only to affect one but several targets at the time. Androgen, thyroid hormone and estrogen are the most commonly studied hormone systems, though it is now understood that other endocrine-systems should be considered as potential targets to endocrine disruptors as well. There are several ways in which EDCs can affect the endocrine balance. For instance, EDCs can imitate the sex hormones derived from the hypothalamic-pituitary-gonadal axis and hijack the endocrine system by binding to the active site of e.g. estrogen receptors.

This study will focus on the estrogen hormone system. The estrogen signalling pathway might be disrupted due to xenoestrogens, foreign substances to the body that bind to estrogen

receptors (Yu et al. 2018). Bisphenol A (BPA) is one of the chemicals exerting estrogenic properties, exhibiting reproductive toxicity in avian models (Berg et al. 2001).

Background

Bisphenol A in our everyday life

(7)

long durability and great flexibility. Epoxy resin is used in dental restorative and in lining of metal food cans. Polycarbonate is a type of transparent plastic with exceptional durability and impact resistance and was put on the market in 1957. It is often used in e.g. CD album covers, medical devices and aircraft windows. BPA is also known to be used in flame retardants, plastic bottles, receipt rolls, lunch boxes and children’s toys (SOU 2014). Polycarbonate was earlier used in production of baby bottles, but is now prohibited in all European countries since 2011 (Kemikalieinspektionen 2016a). The European commission has also decided to prohibit the use of BPA in cash register receipts and tickets by the beginning of 2020 (Kemikalieinspektionen 2016b). The European Food Safety Authority (EFSA) re-evaluated their earlier standings of the Tolerable Daily Intake (TDI) safety level of BPA (EFSA 2015a). The TDI decreased from 50 to 4 µg/kg of bw/day due to new findings in more recent studies. Cocktail effect, toxicological endpoint and uncertainties of data were brought up as factors for even further limitations of the safety level (EFSA 2015b).

Why hazardous?

Depending on usage BPA has been shown to leak, due to no covalent bond existing to the product in question (Singh et al. 2016). For instance, Brotons et al. (1995) presented clear evidence that BPA was found to leak from the plastic lining and contaminated the content of metal food cans. Another study, presented by Olea et al. (1996), showed that BPA was found in the saliva from patients that had received dental repair with epoxy resin. Trying to find out what substance affected their study, a research group in the 1990s found that BPA was leaking from the autoclave and causing effects with estrogenic properties. It was due to these findings that further studies looked into the endocrine disrupting properties and effects of BPA, and it has since then been heavily debated regarding use and regulation (SOU 2014). Even though several studies indicate that low dose exposure to BPA during the embryonic development alters the meiotic progression (Yu et al. 2018), the seriousness of overall consequences in humans exposed to low doses is still debated.

Since 2016 there is a new classification for BPA within a higher level of hazard (GHS08, can

harm the fertility or the unborn child (Kemikalieinspektionen 2018)). It has now a

harmonized classification meaning that substances, like BPA, that are known to cause more severe effects to human and environment will be under the same terms of regulation, that apply to all European countries. The harmonization can facilitate control and regulations of substances that are often exported and imported, for easier follow up. In the case of BPA, the classification was updated due to several animal studies proving the negative effects in fertility and reproductive toxicity of the substance, even in low concentrations. Even though the effects are not tested in humans, they are said to have a presumed impact. Regulations concerning these questions about registration, authorisation, restriction and evaluation can be found in REACH(Europaparlamentet och Europeiska Unionens Råd 2006). It is the

committee of Echa (European Chemicals Agency) together with the committee of REACH that decide about new regulations for substances on the European market

(8)

Effects in humans

Bisphenols are suspected to have several negative effects in humans. To our knowledge, there are still limited epidemiological studies to support any relation of BPA or bisphenol AF (BPAF) to human disorders (Bergman et al. 2013). Disturbed reproductive cycle, reduced fertility, endometriosis, reduced number of sperms, an altered weight of both the uterus, testes and prostate are just some of the effects of bisphenols as seen in laboratory studies utilising mammalian or non-mammalian model organisms (Kemikalieinspektionen 2011).

Survey and analysis of alternatives to Bisphenol A

As part of achieving the national environmental quality objectives of Sweden

(Naturvårdsverket 2019), The Swedish Chemicals Agency (KemI) presented a contribution to the action plan for a toxic-free everyday environment, better protection of the children

(Kemikalieinspektionen 2014). The survey (Kemikalieinspektionen 2011) presented a list of potential analogous bisphenols, where over 200 substances with similar chemical structure as BPA could be identified. The report also accounted for that 37 of these 200 substances where expected to have endocrine disrupting effects, predicted via data simulation. But due to lack of sufficient information regarding risk assessment and usage of these BPA analogues, they are not yet regulated and harmonized in the same way as BPA. Still only a pre-registration of the BPA-replacements is required by Echa, making them hard to trace in the European market. Not even all substances need to be pre-registered, making it even harder to draw any conclusions of their whereabouts.

Today there is an ongoing phase out of BPA. KemI wish to establish a dialogue with

manufacturers to stress the need of keeping BPA-replacements, with unwanted properties, off the market and to find suitable replacements to dangerous compounds. KemI also need help from the research community in evaluating the toxicological effects of the compounds as a proactive chemical control measure. Amongst these 37 analogous BPA-replacements with endocrine disrupting functions, BPAF can be found. BPAF was one of the bisphenols that was chosen for further evaluation and monitoring by KemI (Kemikalieinspektionen 2011).

Bisphenol AF

BPAF, a derivative of BPA with structural similarity, is used as a monomer in several

specialised polymers, as polycarbonate (Kemikalieinspektionen 2011). BPAF is also used as a precipitation agent and a crosslinker for fluoroelastomers (National Toxicology Program et al. 2008). Both BPA and BPAF have two phenol functioning groups but, as seen in Figure 1 and 2, BPAF has replaced the methyl hydrogen groups with fluorines instead, hence the

(9)

Registration of chemicals in REACH is divided into low-volume products (LVP, 10-1000 tonnes/year/manufacturer/importer) and high-volume products (HVP, >1000

tonnes/year/manufacturer/importer). For chemicals produced in a volume of less than 100 tonnes/year, REACH does not require registration until later, regardless of the potency of the chemical. BPAF, and many other PFASs, is classified as a low-volume product, which creates challenges in finding explicit answers of the extent of use and its whereabouts (Sternbeck et

al. 2014).

A recent study by Moreman et al. (2017) stated that BPAF is the most potent BPA-alternative of the tested chemicals in their study. Zebrafish embryo-larvae was used as a model organism to exposure of BPS, BPF, BPA and BPAF. The estrogenic mechanisms, target tissue, relative potency and toxicity were assessed and compared between the bisphenols.

Figure 1. 2D structure of BPAF (PubChem 2019a) Figure 2. 2D structure of BPA (PubChem 2019b).

The domestic fowl as a model organism

The Greek philosopher Aristotle (384–322 BC) interpreted, to our knowledge, the first description of the chicken embryo as a model organism in the writings of his work Historia

Animalium (Aristotle et al. 2002). When conducting this experiment fertilised eggs from

(10)

Validation of the test method in avian embryos for gonadal and developmental endpoints when exposed to EDCs has been recently achieved by Jessl et al. (2018). The study

investigated the natural variability in untreated and solvent-treated control groups of the test system and developed a standardised protocol for this type of research. The same study

declared, as stated earlier by Berg et al. (1999), the avian embryo as a hopeful replacement for mammalian animal testing in evaluating endocrine disrupting chemicals.

Another benefit of using the avian model organism is the short embryonic developmental period up until hatching. As for the chicken, hatching occurs at embryonic day 20-21 (E20-21) (Hamburger & Hamilton 1992).

Sexual differentiation

The two gonads are ventrally developed on the embryonic kidney and start to show at about E3. Sex differentiation is initiated at about E6.5 where the gonads start to display

morphological differences (Lambeth & Smith 2012). At first both Müllerian and Wolffian duct are present in both sexes, but at the onset of sex differentiation the Müllerian ducts (MD) regress in males and the Wolffian ducts (WD) regress in females (Romanoff 1960). In males the testes are formed by E9.5 and the WD further develops into the vas deferens and

epididymis by the presence of testosterone. In females only the left gonad continues to develop into an ovary, whilst the right gonad regresses to a rudimentary organ. Also, the left MD develops into an oviduct in females, whilst the right duct regresses, in the same manner as the right gonad. The anti-Müllerian hormone, AMH, is also responsible for the regression of the MDs in both males (both left and right) and females (only right) (Lambeth & Smith 2012).

Gene expression in sex differentiation and the importance of estrogen

The review article, Sex determination: insights from the chicken, written by Smith and Sinclair (2004) underlies the following statements of this paragraph.

As for mammals, the sex differentiation in birds relies on the inheritance of sex chromosomes. Contrary to mammals the sex chromosomes in birds are homogametic, ZZ, for males and heterogametic, ZW, for females. The sex steroid hormone estrogen is crucial for the development of ovary. There is a female-specific expression due to the cytochrome P450 enzyme, aromatase. This enzyme is expressed in the medulla of both the left and right gonad and is responsible for catalysing the synthesis of estrogen at the time for sex differentiation, hence crucial for the ovarian development. Male chickens express a gene named SOX9, that triggers the establishment of testes. Research has still not established which genes that really are the master sex-determinants in birds.

(11)

The impact of BPA exposure on sexual differentiation and molecular mechanism Toxic insults of estrogenic EDCs, e.g. BPA, during the sexual differentiation are known to end up in malformations of the reproductive organs in the chicken embryo, first established in avian models testing the estrogen-like effects of BPA by Berg et al. (2001). BPA is known to possess estrogenic activity and in the same study by Berg et al. a dose-response relationship in BPA treatment to the embryos was established. BPA was also proven to affect the sex differentiation by inducing feminization of the left testis in males, hence named ovotestis. The study also focused on the endpoints of gross morphology abnormalities and presence of oocyte-like cells in the testicular cortex, visible at microscopic level in histological analysis.

Regarding the molecular mechanism, estrogen plays a pivotal role with the associated estrogen receptors, ERa and ERβ, to whom estrogen binds and activates. Mattsson et al. (2011) clarified how xenoestrogens, e.g. BPA, induce alterations in the early stages of sexual differentiation in chicken embryos. The same study established that the causation to this disruption was mediated by ERa, concluded when exposing chicken embryos to the ERa agonist PPT, which is known to induce similar effects as BPA and other xenoestrogens.

Ovarian histology

In the female ovary there is normally a clear ovarian cortex area, blending together with the inner medulla. Dense irregular and highly vascularized connective tissue characterize the medulla. On the contrary, the cortex is characterized by fibrocytes and collagen. It is in the thickened core-cortex of the ovary in which oocyte germ cells are found, that further develop into ovarian follicles, seen later on at different developmental stages throughout the ovarian cortex (Eroschenko & Fiore 2013).At time for hatching, the oocyte germ cells are in meiotic prophase. The oocyte is distinguished by the rounded shape and large nucleus (Berg et al. 2001).

Testicular histology

Supporting information about the testicular histology, and the statements following this paragraph, can be found in “diFiore's atlas of histology with functional correlations” written by Eroschenko and Fiore (2013). Each testis is enclosed by a capsule, the tunica albuginea, consisting of dense connective tissue in the adult chicken and loose irregular connective tissue in the developing embryo, which constitutes the thin testicular cortex. At time for hatching the tunica albuginea is not fully developed. Instead there is a thin layer of epithelium surrounding the testes. Internal to the tunica albuginea (thin epithelium in embryo) there is loose

connective tissue. Further in to the centre of the testis, the medulla, there is interstitial

(12)

Histology of ovotestis

Previous studies define the ovotestis with an ovarian-like cortex and medulla (varying appearance), with or without few seminiferous cords. Presence of oocyte-like germ cells are identified in the ovarian-like core cortex (Berg et al. 1998).

Aim and objective

The aim of this project is to study if and how BPAF can affect the gonadal differentiation in chicken embryos, and more specifically to look at the male left testis to see if the testis is feminized or, altered in another way, at the histological level. Signs of feminization include presence of ovarian-like cortex, oocyte-like germ cells, and an ovarian-like medulla without or with few seminiferous cords (Berg et al. 2001). The relevance of this study is based upon earlier research accomplished by the bird group at the department of Environmental

Toxicology at EBC (Evolutionsbiologiskt centrum), Uppsala University, Sweden. Some of these earlier studies (Berg et al. 1998, 1999, 2001, Mattsson et al. 2011) were focused on the effect of several EDCs, e.g. BPA, to gonadal differentiation in the chicken embryo model. Hence it is relevant to look into analogous compounds like BPAF, that KemI has put to light in their survey of bisphenols (Kemikalieinspektionen 2011). Due to the use of chicken embryos as a model organism, the results are also to some extent applicable to wildlife avian species, when assessing potency of environmental pollutants and endocrine toxicology. Birds are one of the top predators, hence the relevance of the results of this study (Berg et al. 1999).

Materials and methods

Ethics

Regarding ethical approval the animal experiment in this study has been approved by the Uppsala Ethical Committees for Research on Animals (permit no C 90/15). The facilities and egg incubators have been approved by the Swedish Board of Agriculture.

Incubation

(13)

Figure 2. Candling of egg for viable embryo. A; fertilised but dead embryo. B; fertilised egg with viable

embryo. Visible growth of blastodisc at the centre of the egg and growth of branching blood vessels. Treatment

In this study the samples were divided into five groups depending on administered dose. The administered dose, group name and number of injected eggs is presented in Table 1.

Table 1. Treatment groups1_.

Dose BPAF (µg/g egg) Group name Injected eggs (n)

0 G0 22

0.7 G0.7 23

7 G7 24

70 G70 23

175 G175 23

1. _{Dose and corresponding group names, used throughout the report. Total number} of injected eggs in each treatment group.

The eggs were divided into two batches, for which the incubations were started at two consecutive days. This was done to keep the number of embryos at a manageable level at the time for dissection. Each batch was divided into the five treatment groups (Table 1).

Injections with Hamilton syringe and disposable needles were performed at day four post incubation, E4, following candling. The substance used was bisphenol AF (BPAF; CAS: 1478-61-1, ≥97% purity; Sigma-Aldrich, St. Louis, MO, USA). Prior to injection, BPAF was dissolved in DMSO (Dimethyl sulfoxide). As seen in Figure 3, the site of injection was established by placing a mark on the eggshell at the centre of the air chamber at the blunt end, visible whilst candling for fertilisation. The injection site was wiped with ethanol. Further, the mark was manually pierced by a dental drill, leaving a small opening for the needle to pass through. The droplet was deposited onto the inner shell membrane in the air chamber. Doses were set based on results from a pilot study, performed by the research group (the bird

group). The volume of injection for each treatment was 20 µl/egg and the following doses of

BPAF (µg BPAF/g egg) were administered; 0 (G0, vehicle control), 0.7 (G0.7), 7 (G7), 70 (G70) and 175 (G175).Post injection the egg was immediately placed horizontally, and the

(14)

In a pilot study the same research group noted that the incidence of mortality increased after injection with DMSO if the egg was placed vertically instead of horizontally. Presumably this was due to the high concentration of DMSO in close contact with the chorioallantoic

membrane surrounding the delicate embryo, which floates towards the air chamber if the egg is put in a vertical position. Figure 4 visualises the inside of the egg and the normal position of the embryo in relation to the air chamber (air pocket). Following labelling by dose and deposit of injection solution the eggs were immediately put back into the incubator.

Dissection, gross morphology and sampling

Prior to dissection (Batch 1: E17, Batch 2: E16), each egg was candled again, and dead embryos removed. Post injection mortality rates were noted. The sex differentiation is not yet completed prior to hatching at E21, but for this study the difference of sex is sufficient enough at that time. Excision of testes and ovaries was performed at E19, with the intention to wait for as long as possible for the gonads to develop in ovo. Eggs were chosen by random selection, unwittingly to treatment. The embryo was extracted from the egg, euthanized by decapitation and weighed (body weight only, not external yolk). The liver was excised, and weight recorded. Gender was determined based on visual assessment; presence of left ovary and MD in females, and two testes in males. The right MD length in females was measured. Furthermore, the gonads were photographed in situ, dissected and stored in phosphate buffered formalin (4% formaldehyde in 0.1 M phosphate buffer, pH 7.4; v/v). Thereafter, gonad area in both sexes was calculated by image analysis.

Image analysis of gonadal histology

The gonads were dehydrated with ethanol (70%, 95%, and absolute ethanol; v/v). Technovit 7100 (Heraeus Kulzer, Hanau, Germany) was used for embedding of tissue (left ovary from female control and left testis in males) prior to sectioning. With a thickness of 2 μm, cross-sections were taken at three different levels, separated by ~300 µm. Sections were stained

Figure 3. Marked site of injection, placed at

the centre of the air chamber.

Figure 4. Schematic illustration showing the embryo

(15)

with Hematoxylin and Eosin and once again dehydrated with ethanol (95% and absolute ethanol; v/v) and finally Xylene prior to mounting. Superfrost glass slides were used for mounting sections. For the histological image analysis, sections at all levels were

photographed with 10x magnification, using a digital camera (Leica DFC550) connected to an upright light microscope (Leica Leitz DMRXE). The equipment was operated by Leica

Application Suite, LAS, version 4.2.0 (Leica Microsystems, Wetzlar, Germany). On average, two overlapping pictures were taken to cover each section of the gonads. To merge and align the overlapping pictures the software Hugin, version 2019.0.0.a369cbe55179 (1989, 1991 Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA), was used. Histological analysis was used to identify presence, if any, of oocyte-like germ cells, ovarian-like cortex and an ovarian-like medulla with or without few seminiferous cords. Potential findings were compared to both male and female control. Condensed

chromatin, and thus a large nucleus, distinguish the female oocyte germ cells (Berg et al. 2001).

Statistics

Prior to further statistical analysis of continuous data comparing control to treatment groups, the D’Agostino & Pearson omnibus K2 test was used to test for normality. To establish any statistical difference regarding sex differences, treatment effects, or sex-treatment

interactions, two-way analysis of variance (ANOVA) and Bonferroni's multiple comparisons post-hoc tests were used. For normally distributed data with no differences in variance, one-way ANOVA and Dunnett's multiple comparisons post-hoc tests were applied to males and females together, if differences in sex-treatment interaction could not be found. For data that was not normally distributed or where differences in variance were found, Kruskal-Wallis one-way ANOVA were used instead, followed by Dunn’s multiple comparisons post-hoc tests. The workflow, as stated above, was applied to the following data; body weight, liver weight, hepatosomatic index (HSI, liver weight:body weight ratio), male right:left testis area ratio and female right MD duct length. Fisher’s exact test (one-tailed, predicted beforehand) was applied to analysis of mortality and frequencies of ovotestis in males. P-values smaller than 0.05 represented statistical significance. Software utilised to calculate the statistical analyses was GraphPad Prism version 5.01 (GraphPad Software Inc,San Diego,CA, USA).

Results

Body weight and mortality rate

Body weight showed no sex differences (data not shown) and was not affected by treatment, as seen in Figure 5A. Figure 5B displays the cumulated mortality (%) of embryos at

(16)

Figure 5. Body weight, mortality, liver weight and HSI (liver:body weight ratio). Since no sex differences

were found, males and females were analysed together. Body weight (A) shows no effect to treatment. The mortality (B) was significantly increased in G70 and G175. Liver weight (C) and HSI (D) were significantly reduced in G70. ***Treatment-effect (p<0.001).

Liver weight and HSI

Both liver weight (Figure 5C) and HSI (Figure 5D) revealed an evident outlier in G7. This indicates that the liver of the embryo in question not only was smaller than the others, but also much smaller in relation to the mean liver:body weight ratio. The liver weight and HSI were not affected by BPAF in G0.7 and G7 compared to G0, but treatment effect was present in G70 (p<0.001). There is no treatment effect in G175 compared to controls. The mean for both liver weight and HSI in G175 clearly differs from G70 and resembles, to a higher degree, the mean of the controls.

Figure 6. Female right MD duct length (A) and Male R:L testes area ratio (B). BPAF does not affect right

(17)

Female right Müllerian duct length

Figure 6A displays the graphical results of the female right Müllerian duct length, which was not affected by treatment in any of the groups. For G70 and G175 it is worth mentioning that the number of replicates is small, thus making it harder to detect any difference, due to low statistical power.

Male right:left testes area ratio

The male right:left testes area ratio was neither affected by BPAF at any dose below G70, nor was it affected to G175. Even though the replicates are few in G175 (hence low statistical power) the mean is approximately the same as for G70, indicating that the area ratio of the testes may be affected and decreased here as well. G70 displays a difference to the control, as seen in Figure 6B (p<0.05).

Frequencies of ovotestis in males

Figure 7 displays the frequency (%) of ovotestis, assessed visually at dissection and by subsequent image evaluation. The results indicate increasing frequencies of ovotestis to increasing dose BPAF. The effects appear to be dose-dependent. Statistical significance, of increased frequency of ovotestis, is only present in G175 (p<0.05).

Table 2. Ovotestis frequency.

Group Frequency ovotestis (n) Frequency (%)

G0 0/9 0 G0.7 1/13 8 G7 1/9 11 G70 3/8 38 G175 3/4* 75* * Treatment-effect (p<0.05).

(18)

Representative sample images are presented in Figure 8. Figure 8A shows a male control with two symmetrical and elongated testes. Contrary to the male control, the gonads in the female control (Figure 8D) develop asymmetrically; the right gonad regresses and the left gonad develops into an ovary. Normally, the ovary is more irregular in shape than a testis, appearing thinner and wider. The ovary is, together with the characteristic shape, distinguished by a more or less rough granular surface. Tendencies of granular surface can also be seen in the left testis of the male control group.

Figure 8B represents one of the assessed ovotestes. When compared to both male and female control it is clear that these testes are feminized, due to the diminished right testis and

indication of ovary-like shape of the left testis. The evident granular surface of the left ovotestis is also visible. To show the full range of effect in the treated specimens, there is another treated male as reference with no observed effect of ovotestis, seen in Figure 8C. These testes look very much like the testes of the male control and not at all similar to the ovary of the female control, giving credit to the assessment of the ovotestis in Figure 8B.

Figure 8. In situ photographs of gonads prior to excision. Male control [A]. Assessed as ovotestis in treated

male [B]. As reference; treated male with no remark of ovotestis [C]. Female control [D].

A

B

(19)

Histology

Figure 9C shows a representative female control, where boundaries of ovarian cortex (Co) and medulla (Me) are indicated. There is interstitial connective tissue surrounding the seminiferous tubule (testicular cords), indicated by arrows, in the male control (Figure 9A) and in the male reference (Figure 9B). The treated male reference, with no signs of ovotestis, is the corresponding embryo to Figure 8C. Lumen of tubule is indicated by asterisk. Medulla (Me) and testicular cortex (Co) are shown in Figure 9A and 9B, respectively.

Figure 9. Histological sections at 10x magnification (hematoxylin-eosin staining). Corresponding embryos to

Figure 8A, C and D. Male control with medullary region (Me) and white arrow pointing at interstitial connective tissue (Int) surrounding the tubules [A]. For comparison; treated male with no remark of ovotestis. Arrow pointing at interstitial connective tissue (Int) surrounding the tubules, testes cortex marked with Co [B]. Female control with ovarian cortex (Co), medulla (M) and marked region of lacunae (*) [C]. Scale bar: 200 µm.

Only one individual differed from the control group. The histological section in Figure 10 (corresponding embryo to Figure 8B) shows features of distinct ovotestis appearance.

(20)

Occasional findings of oocyte clusters, embedded in the testicular cortex, were resembled in other individuals. Varying density of seminiferous tubule structure was also noted. However, these observations were found across all treatment groups and were not considered treatment-related.

Figure 10. Histological sections of ovotestis (hematoxylin-eosin staining). Corresponding embryo to Figure 8B. [A] shows the whole ovotestis, at 10x magnification, indicating cortex-area (Co), medulla (M), interstitial

connective tissue (Int), lumen of seminiferous tubule (white *) and lacunae (*). Scale bar: 200 µm. [B] presents a 20x magnification and [C] presents a 40x magnification of a smaller section from [A]. White arrows pointing at oocyte-like germ cells. Scale bar at 100 µm and 50 µm, respectively.

Discussion

The endocrine disrupting action of BPA has been proved to cause adverse effects and

anomalies to both function and structure in the reproductive system of the developing chicken embryo. This study investigated if and how exposure of the xenoestrogen BPAF, a BPA alternative, affected the sex differentiation in the chicken embryo. Injection of substance was conducted at E4 to ensure exposure during the early sex differentiation. Feminization of the left testis was revealed by gross morphology assessment in all treatment groups except the control. Occasional findings of ovotestis could be noted in histological analysis.

(21)

respectively. The following general toxicity samplings were significant at G70: liver weight, HSI and male right:left testes area ratio. However, not finding any significance in the other treatment groups does not mean that the effect is non-existent. EDCs, as BPAF, are known to show a complex non-linear dose response (Bergman et al. 2013), meaning that there is a critical window of concentration in which treatment effect appears, not necessarily corresponding to the highest dose (Kemikalieinspektionen 2011). In this study the high mortality rate left only a few numbers of individuals to observe, hence the low statistical power. G70 appears to be most toxic as indicated by statistical significance of treatment effect causing mortality, but G175 presents the highest mortality rate, indeed indicating a

considerable toxic effect. The lowest observed adverse effect level (LOAEL) in this study was G70, observed in all the parameters with significant treatment effect, except frequencies of ovotestis. When BPA was tested in the study by Berg et al. (2001), the treatment doses were 67 µg BPA/g egg and 200 µg BPA/g egg. Treatment effects were only statistically significant in 200 µg BPA/g egg, indicating that the threshold value lies somewhere in between, which corresponds to the results obtained in this study.

The effects of BPAF in male reproductive system show similar indication of feminization as in previous studies testing BPA in chicken (Berg et al. 2001, Jessl et al. 2018, Yu et al. 2018). The alterations included transformation of the left testis into ovary-like structures, ovotestis, characterised by a thickened testicular cortex area and presence of oocyte-like germ cells and occasional lacunae. Berg et al. (2001) presented convincing results of treatment effect and an increased frequency of ovotestis. During treatment-blinded assessment in this study a limited number of ovotestes were retrieved in every treatment group except for the control. All ovotestes but one were classified as mild transformation. The obvious ovotestis showed convincing characteristics in both gross morphology and histological analysis. Corresponding gross morphology assessment of the mild ovotestes could not be confirmed by histological analysis, indicating limited tissue transformation or perhaps too subtle changes to be observed at the histological level.

Using the chicken embryo as a model organism is one of the primary strengths to the research in this paper. Not only is the chicken embryo relatively cost-friendly, easily handled and with a short embryonic period up until hatching. It is also well studied and a good replacement to animal studies utilising mammals (Berg et al. 1999, Jessl et al. 2018).

(22)

became really prominent, as well as the importance of comparison to the control group, so as not to make any inaccurate decisions.

BPAF is, unlike BPA, an organofluorine, which is known not to degrade that easily due to the very strong carbon-fluorine bonds (Kemikalieinspektionen 2019). One can speculate how the differences in chemical composition between the compounds possibly affect the uptake and receptivity in the chicken embryo. Above all, BPAF ought to be the one more difficult to metabolize due to the strong carbon-fluorine bonds, hence having a slower elimination rate than BPA. In a zebrafish study, Moreman et al. (2017) stated BPAF as the more potent one when compared to several other bisphenols with estrogenic properties, including BPA.

BPAF, and several other BPA-analogous compounds, is classified as a low volume product. A consequence to this system of classification is that regardless of potency, regulation according to REACH is weaker compared to high volume products (Sternbeck et al. 2014). Concerning BPAF, one could reason that these regulations are not in favour for the general public, since BPAF is shown to be more potent than BPA. As BPA is continuously replaced with BPA alternatives, it is reasonable to think that the use of BPAF will increase.

For further research I find it interesting to elucidate the treatment effect of more frequent dose intervals in between 7, 70 and 175 µg BPAF/g egg. In particular the medium dose range is of interest, adding even more replicates to begin with, now that we are aware of the high

mortality rate. Perhaps it is possible to find a closer range of doses in where mortality decreases compared to G175 and where estrogenic effect, as ovotestis, can be more clearly observed both in gross morphology and in histological evaluation.

It would be interesting to test for potential impact on sex differentiation, as an endpoint of estrogenic activity from BPAF, in the chicken-relative Japanese quail (Coturnix japonica). C.

japonica is also commonly used as an avian model organism when investigating the potential

toxic impact of estrogenic EDCs. Berg et al. (2001) declared the quail to be less suited when looking at the endpoint of increased frequency of ovotestis, due to ovotestis retrieved in the controls of quail exposed to BPA. Yet quail appeared to be more susceptible to BPA and prominent than chicken, when screening for anomalies of the MDs, instead of the testes (Berg

et al. 2001). Thus, I am curious to find out if the results coincide concerning BPAF.

Due to the ongoing replacement of BPA and to feedback to the survey of bisphenol analogues (Kemikalieinspektionen 2011), it is of great importance to further study BPA alternatives, especially the ones with estrogenic properties that impart on sexual differentiation. We need to procure a deeper understanding to assess the adequacy concerning these substances. It is important to learn more about bioavailability, bioaccumulation and persistency in relation to potential route dependency and susceptibility in different developmental stages in life regarding BPAF, as well as other xenoestrogens. Especially, fetal and childhood exposure to BPAF should be prevented. In conclusion, developmental exposure to BPAF causes

(23)

Author contribution

Participants of the research group performed project administration, planning of experimental design, treatment, dissection, photography of the gonads, gross morphology sampling and data curation. The research group included the following persons; Anna Mattsson, Björn Brunström, Maria Jönsson, Anna Mentor and master student Zeinab Issa. The author of this paper, Mimmi Wänn, carried out literature search, contributed to the exposure of treatment and to sexing by visual means of gonads, histological analysis, statistical account and final writing.

Acknowledgements

(24)

References

Aristotle, Balme DM, Gotthelf A. 2002. Historia animalium. Cambridge University Press, Cambridge.

Berg C, Halldin K, Brunström B. 2001. Effects of bisphenol A and tetrabromobisphenol A on sex organ development in quail and chicken embryos. Environmental Toxicology and

Chemistry 20: 2836–2840.

Berg C, Halldin K, Brunstrom B, Brandt I. 1998. Methods for studying xenoestrogenic effects in birds. Toxicology Letters 102–103: 671–676.

Berg C, Halldin K, Fridolfsson A-K, Brandt I, Brunstrom B. 1999. The avian egg as a test system for endocrine disrupters: effects of diethylstilbestrol and ethynylestradiol on sex organ development. The Science of the Total Environment 233: 57–66.

Bergman Å, United Nations Environment Programme, World Health Organization. 2013. State of the science of endocrine disrupting chemicals - 2012 an assessment of the state of the science of endocrine disruptors. WHO : UNEP, Geneva.

Brotons JA, Olea-Serrano MF, Villalobos M, Pedraza V, Olea N. 1995. Xenoestrogens Released from Lacquer Coatings in Food Cans. Environmental Health Perspectives 103: 608– 612.

EFSA. 2015a. No consumer health risk from bisphenol A exposure. WWW document 21 January 2015: http://www.efsa.europa.eu/en/press/news/150121. Accessed 14 May 2019.

EFSA. 2015b. Report on the two-phase public consultation on the draft EFSA scientific opinion on bisphenol A (BPA). EFSA Supporting Publications 12: 740E.

Eroschenko VP, Fiore MSH di. 2013. diFiore’s atlas of histology with functional correlations, 12th ed. Wolters Kluwer Health/Lippincott Williams & Wilkins, Philadelphia.

Europaparlamentet och Europeiska Unionens Råd. 2006. REACH. WWW document 18 December 2006:

https://eur-lex.europa.eu/legal-content/SV/TXT/PDF/?uri=CELEX:32006R1907&from=EN.

Guioli S, Lovell-Badge R. 2007. PITX2 controls asymmetric gonadal development in both sexes of the chick and can rescue the degeneration of the right ovary. Development 134: 4199–4208.

(25)

Jessl L, Scheider J, Oehlmann J. 2018. The domestic fowl (Gallus gallus domesticus) embryo as an alternative for mammalian experiments – Validation of a test method for the detection of endocrine disrupting chemicals. Chemosphere 196: 502–513.

Kemikalieinspektionen. 2016a. Sänkt gränsvärde för bisfenol A i leksaker. WWW document 15 November 2016: https://www.kemi.se/nyheter-fran-kemikalieinspektionen/2016/sankt-gransvarde-for-bisfenol-a-i-leksaker/. Accessed 14 May 2019.

Kemikalieinspektionen. 2016b. Förbud mot bisfenol A i kvitton och biljetter. WWW document 19 December 2016:

https://www.kemi.se/nyheter-fran-kemikalieinspektionen/2016/forbud-mot-bisfenol-a-i-kvitton-och-biljetter/. Accessed 14 May 2019.

Kemikalieinspektionen. 2018. Klassificering och märkning enligt CLP. WWW document April 2018: https://www.kemi.se/global/planscher/ver-3-kemi-plansch-clp-a2-webb.pdf.

Kemikalieinspektionen. 2011. Rapport 5/17. Bisfenoler - en kartläggning och analys. 177.

Kemikalieinspektionen. 2014. Rapport 5/14. Handlingsplan för en giftfri vardag 2015-2020. Skydda barnen bättre.

Kemikalieinspektionen. 2015. Rapport 6/15. Förekomst och användning av högfluorerade ämnen och alternativ. 120.

Kemikalieinspektionen. 2019. Highly fluorinated substances. WWW document 7 February 2019: https://www.kemi.se/en/chemical-substances-and-materials/highly-fluorinated-substances. Accessed 28 May 2019.

Konieczna A, Rutkowska A, Rachoń D. 2015. HEALTH RISK OF EXPOSURE TO BISPHENOL A (BPA). Roczniki Państwowego Zakładu Higieny 66: 5–11.

Lambeth LS, Smith CA. 2012. Disorders of Sexual Development in Poultry. Sexual Development 6: 96–103.

Mattsson A, Olsson JA, Brunström B. 2011. Activation of estrogen receptor alpha disrupts differentiation of the reproductive organs in chicken embryos. General and Comparative Endocrinology 172: 251–259.

(26)

National Toxicology Program, National Institute of Environmental Health Sciences, National Institutes of Health, U.S. Department of Health and Human Services, Research Triangle Park, NC. 2008. Chemical Information Profile for Bisphenol AF [CAS No. 1478-61-1]. WWW document September 2008:

https://ntp.niehs.nih.gov/ntp/htdocs/chem_background/exsumpdf/bisphenolaf_093008_508.pd f.

Naturvårdsverket. 2019. Sveriges miljömål - Sveriges miljömål. WWW document 2019: http://www.sverigesmiljomal.se/miljomalen/. Accessed 14 May 2019.

O’Hagan D. 2008. Understanding organofluorine chemistry. An introduction to the C–F bond. Chem Soc Rev 37: 308–319.

Olea N, Pulgar R, Pérez P, Olea-Serrano F, Rivas A, Novillo-Fertrell A, Pedraza V, Soto AM, Sonnenschein C. 1996. Estrogenicity of Resin-Based Composites and Sealants Used in

Dentistry. Environmental Health Perspectives 104: 298–305.

PubChem. 2019a. Bisphenol AF. WWW document 25 May 2019:

https://pubchem.ncbi.nlm.nih.gov/compound/73864. Accessed 14 May 2019.

PubChem. 2019b. Bisphenol A. WWW document 25 May 2019:

https://pubchem.ncbi.nlm.nih.gov/compound/6623. Accessed 14 May 2019.

Romanoff AL. 1960. The Avian embryo: Structural and Functional Development. Macmillan, New York.

Singh RP, Shafeeque CM, Sharma SK, Singh R, Kannan M, Sastry KVH, Raghunandanan S, Mohan J, Azeez PA. 2016. Effects of bisphenol-A on male reproductive success in adult Kadaknath chicken. Ecotoxicology and Environmental Safety 128: 61–66.

Smith CA, Sinclair AH. 2004. Sex determination: insights from the chicken. BioEssays 26: 120–132.

SOU (ed). 2014. Bisfenol A: kartläggning och strategi för minskad exponering: betänkande. Fritze, Stockholm.

Sternbeck J, Fäldt J, Ribé V. 2014. Bisfenolanaloger- användning och förekomst. 54.

Yu M, Xu Y, Li M, Li D, Lu Y, Yu D, Du W. 2018. Bisphenol A accelerates meiotic