Molecular and histological endpoints for developmental reproductive toxicity in Xenopus tropicalis: Levonorgestrel perturbs anti-Müllerian hormone and progesterone receptor expression

Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part C journal homepage: www.elsevier.com/locate/cbpc

Molecular and histological endpoints for developmental reproductive toxicity in Xenopus tropicalis: Levonorgestrel perturbs anti-Müllerian hormone and progesterone receptor expression Moa Säfholm a, Erika Jansson a, Jerker Fick b, Cecilia Berg a,⁎ a b

Uppsala University, Department of Organismal Biology, Norbyvägen 18A, 752 36 Uppsala, Sweden Umeå University, Department of Chemistry, KBC 6A, Linnaeus väg 6, 901 87 Umeå, Sweden

a r t i c l e

i n f o

Article history: Received 25 August 2015 Received in revised form 2 December 2015 Accepted 7 December 2015 Available online 9 December 2015 Keywords: Endocrine disruption Müllerian ducts Sex differentiation Test system Sexual development Amphibians Progestagens

a b s t r a c t There is an increasing concern regarding the risks associated with developmental exposure to endocrine disrupting chemicals and the consequences for reproductive capability. The present study aimed to refine the Xenopus (Silurana) tropicalis test system for developmental reproductive toxicity by characterising molecular and histological features of sexual development, and to explore effects of exposure to the progestagen levonorgestrel (LNG). Larvae were exposed to LNG (0, 3, 30, 300 ng/L) over the first three weeks of development, encompassing the beginning of gonadal differentiation. mRNA levels of amh (anti-Müllerian hormone), amhr2 (amh receptor 2), ipgr (intracellular progesterone receptor), mpgr beta (membrane progesterone receptor beta), and cyp19a1 (cytochrome p450 19a1) were quantified in larvae and juveniles (4 weeks post-metamorphosis). Relative cyp19a1 and amh expression was used as a molecular marker for phenotypic sex of larvae. Gonadal and Müllerian duct development were characterised histologically in juveniles. Compared to controls, LNG exposure increased the expression of amh and ipgr in male larvae. In juveniles, mpgr beta expression was increased in both sexes and amhr2 expression was decreased in males, implying persistent effects of developmental progestagen exposure on amh and pgr expression signalling. No effects of LNG on the gonadal or Müllerian duct development were found, implying that the exposure window was not critical with regard to these endpoints. In juveniles, folliculogenesis had initiated and the Müllerian ducts were larger in females than in males. This new knowledge on sexual development in X. tropicalis is useful in the development of early life-stage endpoints for developmental reproductive toxicity. © 2015 Elsevier Inc. All rights reserved.

1. Introduction There is an increasing concern regarding the risk posed by endocrine disrupting chemicals to the developing endocrine and reproductive systems. Early life chemical perturbation of the development of reproductive organs including the gonads and the Müllerian ducts has been shown to cause reproductive failure later in life in wildlife species as well as in humans (Goyal et al., 2003; Hill and Janz, 2003; Blomqvist et al., 2006; Pettersson et al., 2006; Crain et al., 2008; Gyllenhammar et al., 2009; Kvarnryd et al., 2011). The Müllerian ducts are precursors of the female reproductive tract and they are present in both sexes during early life stages of vertebrates (except teleost fish). In female Abbreviations: Amh, anti-Müllerian hormone; amhr2, anti-Müllerian hormone receptor 2; cyp19a1, cytochrome p450 19a1; LNG, levonorgestrel, NF, Nieuwkoop and Faber; ipgr, intracellular progesterone receptor; mpgr, membrane progesterone receptor; X. tropicalis, Xenopus tropicalis. ⁎ Corresponding author at: Department of Environmental Toxicology, Uppsala University, Norbyvägen 18A, 752 36 Uppsala, Sweden. Tel.: +46 18 471 26 21. E-mail addresses: [email protected] (M. Säfholm), [email protected] (E. Jansson), jerker.fi[email protected] (J. Fick), [email protected] (C. Berg).

http://dx.doi.org/10.1016/j.cbpc.2015.12.001 1532-0456/© 2015 Elsevier Inc. All rights reserved.

mammals, they develop into oviducts, uterus, cervix and vagina whereas in female birds, reptiles and frogs they develop into oviducts (AdkinsRegan, 1987). Evidence suggests that female reproductive disorders observed in wildlife and humans may be symptoms of incorrect differentiation of the ovary and the embryonic Müllerian ducts due to endocrine disruption during early life stages (Crain et al., 2008). However, causal relationships between the female reproductive disorders in humans and wildlife and exposure to environmental chemicals remain to be elucidated (UNEP/WHO, 2013). To determine relationships between chemical exposure and developmental disorders in Müllerian duct-derived tissues appropriate test systems need to be developed. Research on endocrine disrupting effects of chemicals has thus far focused mainly on perturbation of estrogen, androgen and thyroid signalling pathways. The knowledge on chemical effects on progesterone signalling, which is a key regulatory pathway in sexual development and reproductive function, is far less developed. However, recent research has highlighted progestagens (here defined as synthetic or natural progesterone) as potent endocrine disrupting chemicals in the aquatic environment (Arnold et al., 2014; Säfholm et al, 2014; Svensson et al, 2014). Progestagens are widely used in human and

10

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

Fig. 1. Experimental design with regard to levonorgestrel (LNG) exposure period, sampling time points, and endpoints during larval and juvenile stages of Xenopus tropicalis. NF stage = Nieuwkoop and Faber stage (Nieuwkoop and Faber, 1956). amh (anti-Müllerian hormone), amhr2 (anti-Müllerian hormone receptor 2), pgrs (progesterone receptors), and cyp19a1 (cytochrome p450 19 a1).

veterinary medicine (e.g. in contraceptive pills) and there are more than 20 types of progestagens on the global market (according to The American National Institutes of Health). They are constantly released into the environment from sewage treatment plants, agricultural areas via farm animal waste, pharmaceutical industries, and wastewater irrigation. A number of progestagens including levonorgestrel (LNG) have been detected in lakes, rivers, and streams, which may result in exposure of the aquatic wildlife to concentrations ranging from one to a few tens of ng/L (Petrovic et al., 2002; Vulliet et al., 2008; Al-Odaini et al., 2010; Vulliet and Cren-Olive, 2011). Progestagens are potent developmental reproductive toxicants targeting Müllerian duct differentiation in amphibians, birds and mammals, and gonadal development in fish, amphibians, and mammals (Chen et al., 2007; Gray et al., 2001; reviewed in Hayes, 1998; Kvarnryd et al., 2011; Liang et al., 2015; Lorenz et al., 2011; Stoll et al., 1990). The underlying mechanisms for progestagen-induced developmental reproductive toxicity are not fully understood at present. However, it is known that these compounds have a higher affinity to the progesterone receptor (Pgr) than progesterone itself. For instance, the affinity of LNG to the human Pgr is 150–323% compared to that of progesterone (Kumar et al., 2015). As proper Pgr expression is crucial for normal uterine development from the Müllerian ducts (Lydon et al., 1995) disrupted Pgr expression during critical developmental windows is a potential initiating mechanism involved in LNG-induced developmental toxicity. In amphibians, larval LNG exposure resulted in a lack of oviducts and sterility in adult females which might imply Müllerian duct dysgenesis (Kvarnryd et al., 2011). In Xenopus tropicalis, ipgr (intracellular progesterone receptor), and mpgr beta (membrane progesterone receptor beta) are expressed during larval development (Jansson et al., 2015), making them potential targets for progestagen action. The anti-Müllerian hormone (Amh) is required for proper gonadal and Müllerian duct differentiation in vertebrates (Behringer et al., 1994; Josso et al., 2013). In mammals, Amh-deficiency results in inhibited Müllerian duct regression in males and in gonadal abnormalities in both sexes (Behringer et al., 1994; Josso et al., 2013). Embryonic over-expression of Amh resulted in inhibition of Müllerian duct differentiation and masculinisation of the ovaries in mice (Behringer et al., 1990). Furthermore, similar to progesterone, Amh is involved in oogenesis (Nilsson et al., 2011). Given the key role of Amh in sexual development and function, the expressions of amh and its receptor 2 (amhr2) are interesting to explore as potential targets for developmental reproductive toxicants and as early life molecular markers for developmental reproductive toxicity. X. tropicalis represents an excellent model for investigating developmental reproductive toxicity for several reasons (Berg et al., 2009). The organisation and components of the amphibian hypothalamus–

pituitary–gonadal axis are very similar to those in higher vertebrates (reviewed in Kloas and Lutz, 2006). The genome of X. tropicalis is sequenced (Hellsten et al., 2010), facilitating gene expression analysis. Being water-dwelling throughout life X. tropicalis is a suitable model in experimental aquatic toxicology. It allows a more comprehensive and detailed analysis of reproductive organ development compared with commonly used teleost fish models that lack Müllerian ducts. The differentiation of the Müllerian ducts and gonads in X. tropicalis is very sensitive to endocrine disruption (Pettersson et al., 2006; Gyllenhammar et al, 2009; Porter et al., 2011). X. tropicalis has a generation time of about 4–6 months which provides unique possibilities for life-cycle studies compared to other frog species. Life-cycle studies in X. tropicalis have revealed that exposure to endocrine disrupting chemicals during sex differentiation can result in reproductive failure in the adult frog (Pettersson et al., 2006; Gyllenhammar et al., 2009; Kvarnryd et al., 2011). However, as life-cycle studies are very timeand resource consuming the characterisation of sensitive early lifestage endpoints/biomarkers for developmental reproductive toxicity would increase the usefulness of the X. tropicalis test system. Currently there is no general marker of genetic sex available for X. tropicalis. Molecular tools to determine phenotypic sex before the gonads are morphologically different therefore need to be developed. Cytochrome p450 19a1 (cyp19a1 or aromatase, catalysing the biosynthesis of estrogens from androgens) has been proposed as a female marker as the expression in the ovary is higher than that of the testis during sex differentiation (Duarte-Guterman and Trudeau, 2011; Navarro-Martín

Gonads

MD

MD

K

K

Fig. 2. Schematic illustration of the localisation of the histological section (dotted line) through the urogenital complex of juvenile Xenopus tropicalis. MD = Müllerian duct. K = kidney.

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

11

Table 1 Mean (S.D.) measured concentrations of levonorgestrel during the exposure period. Measured values represent the concentration in water samples taken from replicate test tanks, before water exchange if nothing else is stated. Time point for sampling

3 ng/L exposure (n = 4)

30 ng/L exposure (n = 4)

300 ng/L exposure (n = 3)

Zero sample (before exposure start) Week 1 Week 2 Week 3 Week 3 (after water exchange) All samples between weeks 5 and 10

bLOQ bLOQ bLOQ bLOQ 3 (0.4)d ng/L bLOQ

66 (13)a ng/L 9 (3)a ng/L 7b ng/L 5 (4)c ng/L 23 (8) ng/L bLOQ

337 (93)a ng/L 112 (22) ng/L 222 (52) ng/L 32 (6) ng/L 174 (9)a ng/L bLOQ

a b c d

One sample could not be analysed due to technical failure. The LNG level in three samples was below the limit of quantification (LOQ), e.g. 1 ng/L. The LNG level in two samples was below LOQ. The LNG level in one sample was below LOQ.

et al., 2012). Recently, amh mRNA expression was shown to be associated to the testes during gonadal differentiation in amphibians, i.e. in Pleurodeles waltl (Al-Asaad et al., 2013), Rana rugosa (Kodama et al., 2015) and X. tropicalis (Jansson et al., 2015). It was therefore suggested that using amh expression as a testicular marker in combination with cyp19a1 expression as an ovarian marker would provide a robust tool to determine phenotypic sex during gonadal differentiation in X. tropicalis (Jansson et al., 2015). The robustness of cyp19a1 and amh as markers of gonadal sex during and after exposure to progestagens and other endocrine disrupting chemicals remains to be elucidated. The overall aims of the present study was to further develop and refine the X. tropicalis test system for developmental reproductive toxicity. The specific objectives were to a) Determine effects of larval exposure to LNG on mRNA expression of amh, amhr2, ipgr, mpgr beta and cyp19a1, and on histological features of gonadal and Müllerian duct development, b) Explore histological and molecular responses as early life biomarkers for developmental reproductive toxicity, and c) Contribute baseline data for ontogenetic mRNA expression of genes involved in sexual development, i.e., amh, amhr2, ipgr, mpgr beta and cyp19a1.

before water exchange eight times during the exposure period (with one or two weeks apart). At one time point, after three weeks of exposure, water samples were taken after water exchange. At metamorphosis, exposure was discontinued and the remaining metamorphosed frogs were held unexposed in 20-L tanks in a flow-through system for 4 weeks. The animals were kept in copper-free tap water at a mean conductivity of about 500 μS/cm, a temperature of 27 ± 0.5 °C and with a 12:12 h light:dark cycle. Larvae and juveniles were fed Sera micron (Sera, Heinsberg, Germany), Sera Vipan Baby (Sera, Heinsberg, Germany), and Aquatic nature tropical Excel S (Aquatic Nature, Roeselare, Belgium). All animal experiments were approved by the Uppsala Local Animal Ethics Committee, and were performed in accordance with EU Directive 2010/63/EU. 2.2. Animal sampling The experimental design with regard to exposure period, sampling time points, and endpoints is presented in Fig. 1. Larvae were sampled for mRNA analyses of the urogenital complex (containing kidneys, gonads and Müllerian ducts) at NF 50 (3 weeks after start of exposure) and at NF 56 (5 weeks after start of exposure). NF 50 is at the beginning of gonadal differentiation, and at NF 56 is during gonadal differentiation. The number of individuals subjected to mRNA analysis at NF 50 was 25–40/treatment group, and at NF 56 it was 18–60/treatment group. Juveniles at 4 weeks post-metamorphosis were sampled for histological analysis of gonadal and Müllerian duct morphology (16–32/ treatment group) and for mRNA analyses (28–30/treatment group). As X. tropicalis displays a high variability in larval developmental rate, i.e. the time to metamorphosis (Pettersson et al., 2006), the present sampling scheme was designed to obtain both age- and stage matched groups. Therefore, in order to determine the age of the juveniles at 4 weeks post-metamorphosis, the newly metamorphosed frogs from all tank replicates within each treatment were allocated to 4 new tanks depending on their time to metamorphosis. Hence, for each treatment the first, second, third and fourth group consisted of individuals that reached NF 66 at day 1–7, 8–18, 19–29, and 30–46, respectively, after the first individual to reach this stage.

2. Materials and methods 2.1. Animals and LNG exposure X. tropicalis larvae were obtained by mating adults (Xenopus1, Dexter, MI, USA) as described previously (Pettersson et al., 2006). Larvae from three pairs of frogs were exposed to LNG (purity ≥ 99%, CAS: 797-63-7, Sigma–Aldrich, St. Louis, MO, USA) from NF 47–48 until completed metamorphosis (NF 66). Quadruplicate tanks (15 L, Ferplast, Vicenza, Italy), initially containing 90 larvae each, were used for all treatments. The nominal LNG concentrations in the exposure tanks were 0, 3.0, 30, and 300 ng/L. The exposure to 3 ng/L is considered environmentally relevant. Acetone was used as a solvent and its concentration was 0.0002% in all tanks, including controls. Larvae were exposed under semi-static conditions, with half the test solution being renewed three times a week during the first three weeks and thereafter five times a week. For chemical analysis, water samples were taken

Table 2 mRNA expression levels in the urogenital complex of putative female (F) and putative male (M) Xenopus tropicalis at the Nieuwkoop and Faber (NF) developmental stage 49–52 after developmental levonorgestrel exposure (nominal concentrations). The data is expressed as mean (S.D.) fold change to the control group = E−Ct sample A / mean E−Ct control sample. amha

amhr2a

ipgra

mpgr betaa

cyp19a1a

Treatment (ng/L)

n F

M

F

M

F

M

F

M

F

M

F

M

Control 3 30 300

13 22 20 11

24 18 17 14

1.00 (0.54) 1.38 (0.95) 0.95 (0.40) 1.35 (0.90)

1.00 (0.71) 0.86 (0.58) 1.32 (1.13) 1.94 (1.29)⁎

1.00 (0.36) 1.00 (0.36) 1.22 (0.41) 1.23 (0.39)

1.00 (0.50) 0.70 (0.30) 1.08 (0.64) 1.35 (0.51)

1.00 (0.69) 0.83 (0.51) 0.83 (0.56) 0.72 (0.43)

1.00 (0.54) 1.00 (0.72) 0.88 (0.71) 1.01 (0.51)

1.00 (0.57) 0.80 (0.66) 1.00 (0.84) 1.07 (1.12)

1.00 (0.67) 0.58 (0.49) 0.68 (0.68) 0.84 (0.28)

1.00 (0.63) 0.73 (0.46) 0.72 (0.45) 0.98 (0.45)

1.00 (0.85) 0.92 (0.66) 1.18 (0.91) 1.19 (0.73)

a Full gene names: amh (anti-Müllerian hormone), amhr2 (anti-Müllerian hormone receptor 2), ipgr (intracellular progesterone receptor), mpgr (membrane progesterone receptor), cyp19a1 (cytochrome p450 19 a1). ⁎ Significantly different from the control group, p b 0.05 (ANCOVA with the exact NF stage as covariate, Tukey post-hoc test).

12

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

Table 3 mRNA expression levels in the urogenital complex of putative female (F) and putative male (M) Xenopus tropicalis at the Nieuwkoop and Faber (NF) developmental stage 55–57.5 after developmental levonorgestrel exposure (nominal concentrations). The data is expressed as mean (S.D.) fold change to the control group = E−Ct sample A / mean E−Ct control sample. amha

amhr2a

ipgra

mpgr betaa

cyp19a1a

Treatment (ng/L)

n F

M

F

M

F

M

F

M

F

M

F

M

Control 3 30 300

30 8 11 9

30 10 8 9

1.00 (0.37) 0.82 (0.49) 1.20 (1.07) 1.83 (1.83)

1.00 (0.68) 1.25 (1.01) 0.97 (0.69) 0.95 (0.74)

1.00 (0.36) 0.87 (0.62) 0.75 (0.65) 0.89 (0.54)

1.00 (0.59) 1.00 (0.61) 1.10 (0.61) 1.00 (0.69)

1.00 (0.35) 0.88 (0.33) 0.81 (0.46) 0.71 (0.37)

1.00 (0.53) 1.64 (1.19)⁎ 0.85 (0.30) 1.17 (0.66)

1.00 (0.69) 1.00 (0.54) 0.94 (0.87) 0.88 (0.68)

1.00 (0.57) 0.99 (0.55) 0.77 (0.53) 0.74 (0.36)

1.00 (0.50) 0.74 (0.49) 0.58 (0.36) 0.94 (0.48)

1.00 (0.84) 1.73 (1.62) 1.36 (1.39) 0.88 (0.83)

a Full gene names: amh (anti-Müllerian hormone), amhr2 (anti-Müllerian hormone receptor 2), ipgr (intracellular progesterone receptor), mpgr (membrane progesterone receptor), cyp19a1 (cytochrome p450 19 a1). ⁎ Significantly different from the control group, p b 0.05 (ANCOVA with the exact NF stage as covariate followed by Tukey post-hoc test).

2.3. Chemical analysis

2.5. Molecular sex determination

LNG concentrations were determined by chemical analysis using an in-line Solid Phase Extraction (SPE) column coupled to liquid chromatography-tandem mass spectrometry (LC-MS/MS). Details of the chemical analysis and the in-line extraction, including chemicals used, chromatographic details and selected reaction monitoring (SRM) transitions, are described in Kvarnryd et al. (2011). In the present study, however, labelled LNG-D6 was used as internal standard and the SRM data used were tube lens 101, 319.1 → 251.3, collision energy 20 V.

As there is no marker available to determine genetic sex in X. tropicalis we used mRNA expression of cyp19a1 and amh as molecular tools to determine putative phenotypic sex (Jansson et al., 2015). Individuals at NF 50 and 4 weeks post-metamorphosis were classified as putative females if their relative expression level of cyp19a1 was high and as putative males if it was low. The putative sex was verified by plotting the expression of cyp19a1 against that of amh regardless of exposure groups. The individuals sampled at NF 56 were sexed based on the two groups obtained by plotting cyp19a1 expression against amh expression since both genes showed highly significant sexual dimorphism at this stage.

2.4. RNA isolation and analysis Detailed descriptions of total RNA extraction, RNA integrity check, cDNA synthesis, and real-time quantitative polymerase chain reaction (qPCR) analysis are given in Supplementary Information. Briefly, total RNA was extracted from the urogenital complex using Aurum Total RNA Mini Kit (Bio-Rad Laboratories Inc., Hercules, CA, USA) and quantified in NanoDrop 2000c Spectrophotometer (Thermo Scientific, NanoDrop Products, Wilmington, DE, USA). cDNA was synthesised from total RNA with the iScript cDNA Synthesis Kit (Bio-Rad Laboratories Inc., Hercules, CA, USA) using an iCycler (Bio-Rad Laboratories Inc., Hercules, CA, USA). Gene-specific primers for analysing mRNA levels of amh, amhr2, ipgr, mpgr beta (paqr8) and cyp19a1 (Supplemental Table S1) were designed by Jansson et al. (2015). Each sample was run in a qPCR reaction mixture containing iQ SYBRGreen Supermix (Bio-Rad Laboratories Inc., Hercules, CA, USA), forward and reverse primers and cDNA within a Rotor-Gene 6000 real-time DNA amplification system (Qiagen, Hilden, Germany). The repeatability between qPCR runs is given in Supplemental Table S2. Data for NF 50, 56 and 4 weeks post-metamorphosis samples are presented as relative gene expression to the mean values of all control samples = E−Ct sample A/ mean E− Ct control sample. The data are also presented as E−Ct × 108 to show absolute differences between sex and life-stages.

2.6. Preparation of histological sections The urogenital complex of 4-week old individuals was excised and fixed in formaldehyde (4% in phosphate buffer). The tissue was dehydrated in increasing concentrations of ethanol and embedded in hydroxyethyl methacrylate (Leica Historesin, Germany). For gonadal and Müllerian ducts analysis, three transverse sections (2 μm) were taken at one level through the anterior part of the gonads and Müllerian ducts (Fig. 2) and stained with haematoxylin–eosin as previously described (Berg, 2012). 2.7. Histological sex determination and gonadal maturity In 4 week-old juveniles, gonads containing ovarian cavity and oocytes were scored as ovaries, and gonads with spermatogonia and with medulla, lacking a cavity, were scored as testes. Ovaries were evaluated histologically with respect to oocyte maturation. The oocytes in one cross-section per ovary (Fig. 2) were classified into two stages: pre-follicular (early meiotic prophase) or follicular stage, using the criteria described by Hausen and Riebesell (1991). The proportion of follicular oocytes per ovary was estimated and converted into a score

Table 4 mRNA expression levels in the urogenital complex (containing gonads, Müllerian ducts and kidneys) of putative female (F) and putative male (M) Xenopus tropicalis at 4 weeks post-metamorphosis after developmental levonorgestrel exposure (nominal concentrations). The data is expressed as mean (S.D.) fold change to the control group = E−Ct sample A / mean E−Ct control sample . amha

amhr2a

ipgra

mpgr betaa

cyp19a1a

Treatment (ng/L)

n F

M

F

M

F

M

F

M

F

M

F

M

Control 3 30 300

10 21 22 22

18 9 8 8

1.00 (0.67) 1.28 (0.75) 1.23 (0.63) 1.60 (0.79)

1.00 (0.40) 0.58 (0.28) 1.05 (0.58) 0.88 (0.46)

1.00 (0.56) 0.78 (0.43) 0.99 (0.53) 1.18 (0.54)

1.00 (0.43) 0.58 (0.30)⁎ 0.82 (0.40) 0.90 (0.51)

1.0 (0.67) 0.91 (0.33) 1.19 (0.46) 1.23 (0.48)

1.00 (0.43) 0.94 (0.29) 0.98 (0.47) 1.76 (0.86)

1.00 (0.55) 1.62 (0.74)# 1.36 (0.57) 1.19 (0.55)

1.00 (0.37) 0.99 (0.64) 1.03 (0.53) 2.00 (0.79)⁎⁎

1.00 (0.92) 1.04 (0.92) 1.25 (0.99) 1.54 (0.99)

1.00 (0.90) 0.95 (1–09) 0.94 (0.68) 0.87(1.05)

a Full gene names: amh (anti-Müllerian hormone), amhr2 (anti-Müllerian hormone receptor 2), ipgr (intracellular progesterone receptor), mpgr (membrane progesterone receptor), cyp19a1 (cytochrome p450 19 a1). ⁎ Significantly different from the control group, p b 0.05 (Kruskal–Wallis test followed by Dunn's post-hoc test). ⁎⁎ Significantly different from control group, p b 0.01 (Kruskal–Wallis test followed by Dunn's post-hoc test). # Different from the control group, p = 0.05 (one-way ANOVA followed by Tukey post-hoc test).

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

13

number, 0–4. Ovaries lacking follicular oocytes were scored as 0 and those with the highest proportion of follicular oocytes were scored as 4. Testis maturation was evaluated in one cross-section through the urogenital complex from 4 week-old juveniles (Fig. 2). Both testes were classified into one out of four stages reflecting the developmental stage of the most mature germ cell i.e.: primary spermatogonia, early secondary spermatogonia, late secondary spermatogonia, or primary spermatocytes. All histological evaluations were conducted by one analyst using coded slides. 2.8. Müllerian duct histology In female and male 4 week-old juveniles, the two Müllerian ducts were evaluated with regard to developmental stage (1–5) using the criteria established by (Jansson et al., 2015). The developmental stages were: 1 — a small rounded bulge of irregularly packed mesenchymal cells at the lateral side of the kidney, 2 — a small bud protruding from the kidney, 3 — a distinct structure attached to the lateral side of the kidney, 4 — a distinct tubular structure without a cavity, and 5 — a distinct tubular structure with a cavity lined by elongated epithelial cells. If the two Müllerian ducts in one individual were scored differently, a mean score was calculated. All histological evaluations were conducted by one analyst using coded slides. 2.9. Statistics Statistical tests were performed using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA, USA) and Statistica 12.5 (StatSoft, Tulsa, OK, USA). Survival rates, sex ratio, developmental stages of Müllerian ducts and testis, and frequencies of ovaries with follicular oocytes were compared using Fisher's exact test. The treatment groups were compared with respect to time to metamorphosis and number of follicular oocytes in the ovaries using Kruskal–Wallis test followed by Dunn's multiple comparison test. For the larval stages the treatment groups were compared with respect to mRNA expression levels of the studied genes per sex for each sampling time-point using analysis of covariance (ANCOVA), with the exact NF-stage as covariate, followed by the Tukey post-hoc test if no interaction between the treatment groups and the covariate was observed, otherwise multiple regression was used. For all endpoints, data from all replicate tanks within each exposure group were compared statistically and if no difference was found the data were pooled. At 4 weeks post-metamorphosis, the treatment groups were compared with respect to mRNA expression levels of the studied genes per sex using one-way ANOVA and Tukey's post-hoc

Fig. 3. Maturation stages of germ cells in histological slides of testicles from juvenile male Xenopus tropicalis at 4 weeks post-metamorphosis, after exposure to 0, 3, 30, or 300 ng levonorgestrel/L (nominal concentrations). PS, primary spermatogonia; ES, early secondary spermatogonia; LS, late secondary spermatogonia; PC, primary spermatocytes. The sample sizes are shown above the bars.

Fig. 4. Ovarian maturity presented as score numbers (1–4) based on estimated proportion of follicular oocytes in histological slides of ovaries from juvenile female Xenopus tropicalis, 4 weeks post-metamorphosis. Ovaries lacking follicular oocytes were scored as 0 and those with the highest number of follicular oocytes were scored as 4. The intervals (1–7, 8–18, 19–29, and 30–46 days) represent the time (in number of days) to completion of metamorphosis after the first individual had metamorphosed. n = 22, 4, 4, and 16 for the groups that metamorphosed on day 1–7, 8–18, 19–29, and 30–46, respectively. Asterisks (*) indicate significant difference compared with the group that metamorphosed on day 1–7 (* p b 0.05, *** p b 0.001, Kruskal–Wallis test with Dunn's post-hoc-test).

test if the data passed the Shapiro–Wilk normality test, otherwise Kruskal–Walllis test was used followed by Dunn's multiple comparison test. The change in mRNA expression of the genes of interest over time in the control group was analysed per putative sex using ANOVA and Tukey's post-hoc test. Sex differences in mRNA expression of the studied genes in control animals were analysed with unpaired t-test.

3. Results 3.1. Larval development and sampling Mean (S.D.) survival rates during the larval period were 62 (3), 49 (16), 83 (8), 69 (15) % in the control, 3, 30 and 300 ng/L groups, respectively. One replicate tank in the 300 ng/L group was excluded from the study due to high mortality. Mean (S.D.) survival rates postmetamorphosis were 54 (34), 55 (34), 35 (16) 44 (28) % in the control, 3, 30 and 300 ng/L groups, respectively. No significant treatment related effects on survival were observed.

Fig. 5. Frequencies of small (stage 1–2) and large (stage 3–5) Müllerian ducts in female and male Xenopus tropicalis, at 4 weeks post-metamorphosis. The stage of Müllerian duct development in histological cross-sections through the urogenital complex was scored 1–5 using the following criteria: 1 — a small rounded bulge of irregularly packed mesenchymal cells at the lateral side of the kidney, 2 — a small bud protruding from the kidney, 3 — a distinct structure attached to the lateral side of the kidney, 4 — a distinct tubular structure without a cavity, and 5 — a distinct tubular structure with a cavity lined by elongated epithelial cells. Sample size for females = 43, sample size for males = 45. Asterisks (*) indicate significant difference compared with the female frequency (*** p b 0.001, Fischer's exact test).

14

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

Fig. 6. Graphs presenting the relationship between cytochrome p450 19a1 (cyp19a1) and anti-Müllerian hormone (amh) mRNA expression levels in Xenopus tropicalis individuals regardless of treatment group sampled at (A) Nieuwkoop and Faber (NF) developmental stage 50, (B) NF 56 and (C) at 4 weeks post-metamorphosis (logarithmic scale).

Mean (S.D.) number of days to complete metamorphosis was 52 (13), 54 (14), and 51 (14), in the control, 3, and 30 ng/L groups, respectively. In the 300 ng/L group, the mean (S.D.) time to metamorphosis was 61 (10) days which was significantly longer than that of the control. To obtain groups of individuals that were both age- and stage matched, the sampling at each developmental stage occurred over a limited period of time. Hence, the chronological age of the individuals at NF stage 49–52 was 21–22 days post-hatching. The age of the individuals at NF 55–57 was 27–34 days post-hatching. The age of the juvenile frogs sampled 4 weeks post-metamorphosis for mRNA analyses was 80–93 days post-hatching for all exposure groups. The first, second, third, and fourth group of juvenile frogs used for histological analysis were sampled within 7, 10, 10, and 16 days, respectively (their age was 68–109 days post-hatching).

3.2. LNG exposure and water quality The results of the LNG analysis are shown in Table 1. LNG was not detected in the control tanks. The measured LNG concentration in all samples taken between week 5 and 10 was below quantification limit (1 ng/L). Hence, exposure was ascertained up to the time point for sampling of the NF 50 larvae, which occurred after three weeks. The sampling of NF 56 larvae occurred after five weeks i.e. after the exposure levels had dropped. The mean concentrations of ammonia and nitrite in the exposure tanks and juvenile tanks are presented in Supplemental Table S3.

3.3. Effects of LNG on ontogenetic mRNA expression of the studied genes At NF 50, the amh mRNA level in putative males was significantly increased in the 300 ng/L group compared to the control (Table 2). At NF 56, ipgr mRNA level in putative males was increased in the 3 ng/L group compared to the controls (Table 3). In putative females, no significant difference between the treatment groups in mRNA expression level of the studied genes was observed at the larval stages (Table 2 and 3). At 4 weeks post-metamorphosis, mpgr beta was up-regulated in both putative females and males compared to the controls (Table 4). In the males, the mRNA expression of amhr2 was significantly decreased in the 3 ng/L group compared to the control (Table 4).

3.4. Gonadal histology The results from the evaluation of testicular histology are presented in Fig. 3. No significant difference was found between the control and 30 ng/L group with respect to maturation stage of the testicular germ cells (Fig. 3). The data on testicular histology for the 3- and 300 ng/L groups were not subjected to statistical analysis due to small sample sizes. The histological evaluation of the ovaries showed that the developmental stage of the germ cells ranged from oogonium to follicular oocyte. Follicular oocytes were absent in only three females out of 46 (one from the control group and two from the 30 ng/L group). The proportion of follicular oocytes in the ovary at 4 weeks post-metamorphosis increased with increasing time for the individual to complete metamorphosis

Fig. 7. Sex ratios in Xenopus tropicalis based on mRNA expression of cyp19a1 and amh in the urogenital complex of larvae at (A) NF stage 50 and (B) NF 56, or (C) based on gonadal histology in juveniles (4 weeks post-metamorphosis) after exposure to levonorgestrel at the concentrations 3, 30 or 300 ng/L (nominal). The sample sizes are shown above the bars. NF = Nieuwkoop and Faber developmental stage.

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

(Fig. 4). No differences between the LNG exposure groups and the control with respect to maturation stage of oocytes (early meiotic prophase or follicular stage) were revealed (Supplemental Table S4). 3.5. Müllerian duct histology The Müllerian ducts were significantly more developed (p b 0.001) in females compared to males (Fig. 5). The size of the Müllerian ducts did not depend on time to metamorphosis. No treatment related effect on the size of the Müllerian ducts was found in females or in males.

15

3.6. Molecular and histological sex determination At all investigated stages the amh and cyp19a1 mRNA expression data clustered into two groups, one group with low amh expression and high cyp19a1 expression and one with high amh expression and low cyp19a1 expression (Fig. 6A–C). At 4 weeks post-metamorphosis, the cyp19a1 level was on average 43 times higher in putative females than in putative males compared to the average difference of 1.2 times between putative male and putative female amh levels in pooled exposure groups.

Fig. 8. mRNA expression of cyp19a1, amh, amhr2, mpgr beta and ipgr in the urogenital complex at different life-stages of control Xenopus tropicalis putative females (broken line) and males (dotted line). Data are presented as mean E−Ct × 108. The sample sizes of putative female/putative male were: 13/24 for NF 50, 30/30 for NF 56 and 10/18 for 4 weeks post-metamorphosis (4 w. post-meta.). NF = Nieuwkoop and Faber developmental stage. The mRNA levels at the different sampling time points were compared using ANOVA and the p-values are presented in the graph. Differences between putative females and putative males in mRNA levels were tested using an unpaired t-test. Asterisks (*) indicate a significant difference compared to putative males, ** = p b 0.01, *** = p b 0.001, **** = p b 0.0001. Full gene names: amh (anti-Müllerian hormone), amhr2 (anti-Müllerian hormone receptor 2), ipgr (intracellular progesterone receptor), mpgr (membrane progesterone receptor), cyp19a1 (cytochrome p450 19 a1).

16

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

The sex ratios based on mRNA expression analysis at NF 50, NF 56 and on gonadal histology in 4 week-old juveniles are given in Fig. 7. No significant differences between the treatment groups with respect to sex ratios were observed at any time point. 3.7. Ontogenetic mRNA expression of the studied genes In the control animals, the cyp19a1 expression was female-biased and amh expression was male-biased at all sampled life-stages (Fig. 8). Expression of amhr2 mRNA was higher in putative females than in putative males at NF 56 and 4 weeks post-metamorphosis. No sex difference in the mpgr beta and ipgr levels was observed. The expression levels of amh, amhr2, ipgr and cyp19a1 were significantly increased at 4 weeks post-metamorphosis compared with the larvae stages (NF 50 and NF 56), in control animals of both sexes. In addition, expression of amh and ipgr increased from NF 50 to NF 56 in putative males (Fig. 8). mpgr beta expression level was significantly decreased in putative males between the first and last sampling point and in putative females between the second and the last sampling points (Fig. 8). 4. Discussion The overall aim of the present study was to further develop and refine the X. tropicalis test system for developmental reproductive toxicity by characterising early life-stage molecular and histological endpoints in control larvae and larvae exposed to LNG. The chemical analysis revealed that at week three after water exchange, the measured LNG concentrations were close to nominal, but thereafter the measured LNG concentrations of all samples were below the quantification limit. Exposure was therefore confirmed only for the early larval development, from NF stage 47–48 until about NF stage 50. This developmental period encompasses the beginning of gonadal sex differentiation and the ontogenetic expression of genes regulating gonadal sex and Müllerian duct development in X. tropicalis (El Jamil et al., 2008; Jansson et al., 2015). Hence, the exposure period encompasses sensitive stages of gonadal differentiation in Xenopus (Villalpando and Merchant-Larios, 1990). LNG exposure modulated the mRNA expression of amh, amhr2, ipgr and mpgr beta in a concentration-, sex- and developmental stagespecific manner. In putative male larvae, being in the beginning of the gonadal differentiation period, LNG exposure caused an increased expression level of amh (NF stage 50), and of ipgr (NF stage 56), compared to controls. At 4 weeks post-metamorphosis (long after the exposure was discontinued), LNG exposure resulted in increased mpgr beta levels in both sexes, and in decreased amhr2 expression in males, compared to the controls. The mRNA expressions of amh, amhr2 and pgrs during the sexual development have previously been localised to the gonadal component of the urogenital complex in X. tropicalis (Jansson et al., 2015). The present results therefore suggest persistent effects of the LNG exposure on progesterone and Amh signalling pathways in the developing gonads in X. tropicalis. In zebrafish whole-body homogenates (based on pooled male and female juveniles), amh expression was not affected by exposure to norgestrel (4, 34 and 77 ng/L) or progesterone (4 and 33 ng/L), but was reduced after exposure to 63 ng progesterone/L, which could reflect the significantly decreased proportion of males in this exposure group (Liang et al., 2015). Increased pgr expression levels following short-term progesterone exposure (2, 20 and 200 ng/L) have been demonstrated in zebrafish whole-embryo homogenates (Zucchi et al., 2012). Likewise, increased pgr expression was measured in whole-body homogenates of zebrafish juveniles exposed to progesterone (63 ng/L) or norgestrel (77 ng/L) (Liang et al., 2015). In X. tropicalis, the hepatic mRNA expression level of ipgr was not altered after long-term (7 weeks) larval exposure to LNG (17 ng/L) (Säfholm et al., 2015). Altered gonadotropin mRNA expression has been suggested to be involved in the mode of action of LNG in both fish and amphibians, however, effects were only seen after exposure to high concentrations (312–3120 ng/L) and not at lower concentrations (3–31 ng/L) (Lorenz et al., 2011; Kroupova et al.,

2014). The results from the present study showing perturbed amh and pgr signalling in the developing reproductive organs suggest that these pathways could be involved in the mode of action of LNG-induced developmental reproductive toxicity in amphibians. Further characterisation of these molecular responses is needed to elucidate their usefulness as early life biomarkers for developmental reproductive toxicity. Sexing NF 50 larvae by plotting cyp19a1 against amh mRNA expression gave the same results as sexing based only on relative cyp19a1 expression levels. However, the former method is likely more robust as it is based on two markers for gonadal sex (one ovarian and one testicular marker) instead of one. LNG exposure caused an increased amh expression level in putative males but this did not affect the sexing as amh expression was higher in males than in females also in the control group. The results therefore suggest that amh mRNA expression is a robust molecular marker for phenotypic sex in X. tropicalis larval also after progestagen exposure. No effects of the LNG exposure on sex ratio, female or male germ cell maturation or on Müllerian duct development were observed. Previous research in X. tropicalis has shown that LNG exposure (156 ng/L) during the whole larval period resulted in a drastically decreased percentage of follicular oocytes and oviductal agenesis as determined in adults (Kvarnryd et al., 2011). Hence, the present results imply that the restricted window of LNG exposure was not critical with regard to disrupting ovarian or Müllerian duct development in X. tropicalis. The explanation is likely that the exposure period ended before folliculogenesis and Müllerian duct development started. In male amphibians, developmental exposure to LNG at high (μg/L) concentrations can inhibit testicular development (Lorenz et al., 2011). However, at lower concentrations (19 and 156 ng/L) no effects of larval LNG exposure on sex ratio, testicular development or spermatogenesis have been demonstrated, neither at metamorphosis, nor in adults (Kvarnryd et al., 2011). The basal mRNA expression levels of amh, amhr2, ipgr and cyp19a1 increased over the studied developmental period in X. tropicalis. In contrast, the mpgr beta level decreased over the development, indicating different functions of the two pgrs during amphibian development. Sexual dimorphism was observed from NF 50 (at the beginning of the gonadal differentiation period) in the mRNA expression levels of cyp19a1 and amh and from NF 56 for amhr2. These results are in accordance with previous studies reporting an up-regulation of amh mRNA and protein expression in male amphibians during gonadal differentiation (Al-Asaad et al., 2013; Piprek et al., 2013; Jansson et al., 2015; Kodama et al., 2015), suggesting a role of Amh in testicular differentiation. Previous studies suggest that Amh may be involved in testicular differentiation also in fish, birds and mammals (Rashedi et al., 1983; Oreal et al., 1998; Rodriguez-Mari et al., 2005; Hattori et al., 2012; Cutting et al., 2013). The female-biased amhr2 expression during gonadal differentiation is in accordance with the results by Jansson et al. (2015) and suggests that Amh action on testicular differentiation involves another mechanism than up-regulation of amhr2 expression. The histological evaluation of reproductive organ development in 4-week old juveniles showed that the majority of females had ovaries containing follicular oocytes, and the number of these oocytes depended on the time to complete metamorphosis. The longer the larval period lasted, the more developed was the ovary (i.e. more follicular oocytes) at 4 weeks post-metamorphosis (68–109 days post-hatching). This is consistent with results in Xenopus laevis, showing that the number of follicular oocytes in the ovary depends on time to complete metamorphosis (Wolf et al., 2010). Furthermore, the Müllerian ducts were significantly larger in female than in male juveniles. The reason for this sexual dimorphism is currently not known as information on ontogenetic development of the Müllerian ducts in amphibians is scarce. It has been shown that the Müllerian ducts can be detected histologically at NF 64 in both sexes (Jansson et al., 2015) and that the oviducts reach their final size after about 7 months in X. tropicalis (Olmstead et al., 2009). The timing of Müllerian duct regression in male X. tropicalis is not known, however, no regression in males at 4 weeks post-

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

metamorphosis was observed in a previous study (Jansson et al., 2015). At metamorphosis the majority of the oocytes are in pre-follicular stages (Kvarnryd et al., 2011) and the Müllerian ducts are generally very small, rendering any effects on folliculogenesis and Müllerian duct development difficult to detect at that stage of development. The present results suggest that Müllerian duct and ovary histology in 4-week old juveniles are suitable early life-stage endpoints for developmental reproductive toxicity in X. tropicalis. 5. Conclusions The present study aimed to refine the X. tropicalis test system for developmental reproductive toxicity by characterising ontogenetic development of gonads, Müllerian ducts and mRNA expression of genes involved in sexual development and by exploring these features as endpoints for progestagen exposure. Persistent effects of developmental LNG exposure on amh and pgr expression were found, suggesting that these signalling pathways might be involved in the developmental toxicity of this compound. The restricted early window of LNG exposure was, however, not critical with regard to disturbing gonadal and Müllerian duct development. The sexually dimorphic mRNA levels of cyp19a1 and amh observed throughout the study period imply that they are robust markers for gonadal sex during sex differentiation in X. tropicalis also after exposure to a progestogen. The characterisation of ontogenetic gonadal and Müllerian duct development contributes new knowledge on sexual development in X. tropicalis, suggesting that Müllerian duct and ovary histology in 4-week old juveniles are suitable early life-stage endpoints for developmental reproductive toxicity in X. tropicalis. Acknowledgement We would like to thank Professor Ingvar Brandt for valuable comments on the manuscript. We are also thankful to Andreas Eriksson and Margareta Mattsson for excellent technical assistance. This work was supported by the Research Council Formas (2011-949-20138-48), Carl Tryggers Foundation CTS (12:48), and MistraPharma (2004-147), a research programme supported by the Swedish Foundation for Strategic Environmental Research (Mistra). The funding agencies were not involved in the study design. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.cbpc.2015.12.001. References Adkins-Regan, E., 1987. Hormones and sexual differentiation. In: Norris, D.O., Jones, R.E. (Eds.), Hormones and Reproduction in Fishes, Amphibians, and Reptiles. Plenum Press, New York, p. 4. Al-Asaad, I., Chardard, D., di Clemente, N., Picard, J.Y., Dumond, H., Chesnel, A., Flament, S., 2013. Müllerian inhibiting substance in the caudate amphibian Pleurodeles waltl. Endocrinology 154, 3931–3936. Al-Odaini, N.A., Zakaria, M.P., Yaziz, M.I., Surif, S., 2010. Multi-residue analytical method for human pharmaceuticals and synthetic hormones in river water and sewage effluents by solid-phase extraction and liquid chromatography–tandem mass spectrometry. J. Chromatogr. A 1217, 6791–6806. Arnold, K.E., Brown, A.R., Ankley, G.T., Sumpter, J.P., 2014. Medicating the environment: assessing risks of pharmaceuticals to wildlife and ecosystems. Philos. Trans. R. Soc. B 369. Behringer, R.R., Cate, R.L., Froelick, G.J., Palmiter, R.D., Brinster, R.L., 1990. Abnormal sexual development in transgenic mice chronically expressing mullerian inhibiting substance. Nature 345, 167–170. Behringer, R.R., Finegold, M.J., Cate, R.L., 1994. Mullerian-inhibiting substance function during mammalian sexual development. Cell 79, 415–425. Berg, C., 2012. An amphibian model for studies of developmental reproductive toxicity. In: Harris, C., Hansen, J.M. (Eds.), Developmental Toxicology: Methods and Protocols. Humana Press, pp. 73–83. Berg, C., Gyllenhammar, I., Kvarnryd, M., 2009. Xenopus tropicalis as a test system for developmental and reproductive toxicity. J. Toxicol. Environ. Health A 72, 219–225.

17

Blomqvist, A., Berg, C., Holm, L., Brandt, I., Ridderstråle, Y., Brunström, B., 2006. Defective reproductive organ morphology and function in domestic Rooster embryonically exposed to o,p′-DDT or ethynylestradiol. Biol. Reprod. 74, 481–486. Chen, Y., Jefferson, W.N., Newbold, R.R., Padilla-Banks, E., Pepling, M.E., 2007. Estradiol, progesterone, and genistein inhibit oocyte nest breakdown and primordial follicle assembly in the neonatal mouse ovary in vitro and in vivo. Endocrinology 148, 3580–3590. Crain, D.A., Janssen, S.J., Edwards, T.M., Heindel, J., Ho, S., Hunt, P., Iguchi, T., Juul, A., McLachlan, J.A., Schwartz, J., Skakkebaek, N., Soto, A.M., Swan, S., Walker, C., Woodruff, T.K., Woodruff, T.J., Giudice, L.C., Guillette Jr., L.J., 2008. Female reproductive disorders: the roles of endocrine-disrupting compounds and developmental timing. Fertil. Steril. 90, 911–940. Cutting, A., Chue, J., Smith, C.A., 2013. Just how conserved is vertebrate sex determination? Dev. Dyn. 242, 380–387. Duarte-Guterman, P., Trudeau, V.L., 2011. Transcript profiles and triiodothyronine regulation of sex steroid- and thyroid hormone-related genes in the gonad–mesonephros complex of Silurana tropicalis. Mol. Cell. Endocrinol. 331, 143–149. El Jamil, A., Magre, S., Mazabraud, A., Penrad-Mobayed, M., 2008. Early aspects of gonadal sex differentiation in Xenopus tropicalis with reference to an antero-posterior gradient. J. Exp. Zool. 309A, 127–137. Goyal, H.O., Robateau, A., Braden, T.D., Williams, C.S., Srivastava, K.K., Ali, K., 2003. Neonatal estrogen exposure of male rats alters reproductive functions at adulthood. Biol. Reprod. 68, 2081–2091. Gray, C.A., Bazer, F.W., Spencer, T.E., 2001. Effects of neonatal progestin exposure on female reproductive tract structure and function in the adult ewe. Biol. Reprod. 64, 797–804. Gyllenhammar, I., Holm, L., Eklund, R., Berg, C., 2009. Reproductive toxicity in Xenopus tropicalis after developmental exposure to environmental concentrations of ethynylestradiol. Aquat. Toxicol. 91, 171–178. Hattori, R.S., Murai, Y., Oura, M., Masuda, S., Majhi, S.K., Sakamoto, T., Fernandino, J.I., Somoza, G.M., Yokota, M., Strussmann, C.A., 2012. A Y-linked anti-Müllerian hormone duplication takes over a critical role in sex determination. Proc. Natl. Acad. Sci. U. S. A. 109, 2955–2959. Hausen, P., Riebesell, M., 1991. The Early Development of Xenopus laevis, an Atlas of the Histology. Verlag der Zeitschrift für Naurforschung, Tubingen, Germany. Hayes, T.B., 1998. Sex determination and primary sex differentiation in amphibians: genetic and developmental mechanisms. J. Exp. Zool. 281, 373–399. Hellsten, U., Harland, R.M., Gilchrist, M.J., Hendrix, D., Jurka, J., Kapitonov, V., Ovcharenko, I., Putnam, N.H., Shu, S., Taher, L., Blitz, I.L., Blumberg, B., Dichmann, D.S., Dubchak, I., Amaya, E., Detter, J.C., Fletcher, R., Gerhard, D.S., Goodstein, D., Graves, T., Grigoriev, I.V., Grimwood, J., Kawashima, T., Lindquist, E., Lucas, S.M., Mead, P.E., Mitros, T., Ogino, H., Ohta, Y., Poliakov, A.V., Pollet, N., Robert, J., Salamov, A., Sater, A.K., Schmutz, J., Terry, A., Vize, P.D., Warren, W.C., Wells, D., Wills, A., Wilson, R.K., Zimmerman, L.B., Zorn, A.M., Grainger, R., Grammer, T., Khokha, M.K., Richardson, P.M., Rokhsar, D.S., 2010. The genome of the Western clawed frog Xenopus tropicalis. Science 328, 633–636. Hill, R.L., Janz, D.M., 2003. Developmental estrogenic exposure in zebrafish (Danio rerio): I. Effects on sex ratio and breeding success. Aquat. Toxicol. 63, 417–429. Jansson, E., Mattsson, A., Goldstone, J., Berg, C., 2015. Sex-dependent expression of antiMüllerian hormone (amh) and amh receptor 2 during sex organ differentiation and characterization of Müllerian duct development in Xenopus tropicalis (Submitted) . Josso, N., Rey, R.A., Picard, J.Y., 2013. Anti-mullerian hormone: a valuable addition to the toolbox of the pediatric endocrinologist. Int. J. Endocrinol. 2013, 674105. Kloas, W., Lutz, I., 2006. Amphibians as model to study endocrine disrupters. J. Chromatogr. A 1130, 16–27. Kodama, M., Suda, M., Sakamoto, D., Iwasaki, T., Matsuo, Y., Uno, Y., Matsuda, Y., Nakamura, Y., Maekawa, S., Katsum, Y., Nakamuram, M., 2015. Molecular cloning and characterization of anti-Mullerian hormone (AMH) from the Japanese wrinkled frog, Rana rugosa. Endocrinology 156, 1914–1923, en20132053. Kroupova, H.K., Trubiroha, A., Lorenz, C., Contardo-Jara, V., Lutz, I., Grabic, R., Kocour, M., Kloas, W., 2014. The progestin levonorgestrel disrupts gonadotropin expression and sex steroid levels in pubertal roach (Rutilus rutilus). Aquat. Toxicol. 154, 154–162. Kumar, V., Johnson, A.C., Trubiroha, A., Tumová, J., Ihara, M., Grabic, R., Kloas, W., Tanaka, H., Kroupová, H.K., 2015. The challenge presented by progestins in ecotoxicological research: a critical review. Environ. Sci. Technol. 49, 2625–2638. Kvarnryd, M., Grabic, R., Brandt, I., Berg, C., 2011. Early life progestin exposure causes arrested oocyte development, oviductal agenesis and sterility in adult Xenopus tropicalis frogs. Aquat. Toxicol. 103, 18–24. Liang, Y.Q., Huang, G.Y., Liu, S.S., Zhao, J.L., Yang, Y.Y., Chen, X.W., Tian, F., Jiang, Y.X., Ying, G.G., 2015. Long-term exposure to environmentally relevant concentrations of progesterone and norgestrel affects sex differentiation in zebrafish (Danio rerio). Aquat. Toxicol. 160, 172–179. Lorenz, C., Contardo-Jara, V., Trubiroha, A., Kruger, A., Viehmann, V., Wiegand, C., Pflugmacher, S., Nützmann, G., Lutz, I., Kloas, W., 2011. The synthetic gestagen Levonorgestrel disrupts sexual development in Xenopus laevis by affecting gene expression of pituitary gonadotropins and gonadal steroidogenic enzymes. Toxicol. Sci. 124, 311–319. Lydon, J.P., DeMayo, F.J., Funk, C.R., Mani, S.K., Hughes, A.R., Montgomery Jr., C.A., Shyamala, G., Conneely, O.M., O'Malley, B.W., 1995. Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev. 9, 2266–2278. Navarro-Martín, L., Velasco-Santamaría, Y.M., Duarte-Guterman, P., Robertson, C., Lanctôt, C., Pauli, B., Trudeau, V.L., 2012. Sexing frogs by real-time PCR: Using aromatase (cyp19) as an early ovarian differentiation marker. Sex. Dev. 6, 303–315. Nieuwkoop, P., Faber, J., 1956. Normal Table of Xenopus laevis (Daudin). North-Holland Publishing Company, Amsterdam. Nilsson, E.E., Schindler, R., Savenkova, M.I., Skinner, M.K., 2011. Inhibitory actions of antiMullerian hormone (AMH) on ovarian primordial follicle assembly. PLoS One 6, e20087.

18

M. Säfholm et al. / Comparative Biochemistry and Physiology, Part C 181–182 (2016) 9–18

Olmstead, A.W., Korte, J.J., Woodis, K.K., Bennett, B.A., Ostazeski, S., Degitz, S.J., 2009. Reproductive maturation of the tropical clawed frog: Xenopus tropicalis. Gen. Comp. Endocrinol. 160, 117–123. Oreal, E., Pieau, C., Mattei, M.G., Josso, N., Picard, J.Y., Carre-Eusebe, D., Magre, S., 1998. Early expression of AMH in chicken embryonic gonads precedes testicular SOX9 expression. Dev. Dyn. 212, 522–532. Petrovic, M., Sole, M., Lopez de Alda, M.J., Barcelo, D., 2002. Endocrine disruptors in sewage treatment plants, receiving river waters, and sediments: integration of chemical analysis and biological effects on feral carp. Environ. Toxicol. Chem. 21, 2146–2156. Pettersson, I., Arukwe, A., Lundstedt-Enkel, K., Mortensen, A.S., Berg, C., 2006. Persistent sex-reversal and oviducal agenesis in adult Xenopus (Silurana) tropicalis frogs following larval exposure to the environmental pollutant ethynylestradiol. Aquat. Toxicol. 79, 356–365. Piprek, R.P., Pecio, A., Laskowska-Kaszub, K., Kubiak, J.Z., Szymura, J.M., 2013. Sexual dimorphism of AMH, DMRT1 and RSPO1 localization in the developing gonads of six anuran species. Int. J. Dev. Biol. 57, 891–895. Porter, K.L., Olmstead, A.W., Kumsher, D.M., Dennis, W.E., Sprando, R.L., Holcombe, G.W., Korte, J.J., Lindberg-Livingston, A., Degitz, S.J., 2011. Effects of 4-tert-octylphenol on Xenopus tropicalis in a long term exposure. Aquat. Toxicol. 103, 159–169. Rashedi, M., Maraud, R., Stoll, R., 1983. Development of the testes in female domestic fowls submitted to an experimental sex reversal during embryonic life. Biol. Reprod. 29, 1221–1227. Rodriguez-Mari, A., Yan, Y.L., BreMiller, R.A., Wilson, C., Canestro, C., Postlethwait, J.H., 2005. Characterization and expression pattern of zebrafish anti-Mullerian hormone (amh) relative to sox9a, sox9b, and cyp19ala, during gonad development. Gene Expr. Patterns 5, 655–667. Säfholm, M., Ribbenstedt, A., Fick, J., Berg, C., 2014. Risks of hormonally active pharmaceuticals to amphibians: a growing concern regarding progestagens. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 369, 20130577.

Säfholm, M., Jansson, E., Fick, J., Berg, C., 2015. Mixture effects of levonorgestrel and ethinylestradiol: estrogenic biomarkers and hormone receptor mRNA expression during sexual programming. Aquat. Toxicol. 161, 146–153. Stoll, R., Faucounau, N., Maraud, R., 1990. Action of estradiol on Mullerian duct regression induced by treatment with norethindrone of female chick embryos. Gen. Comp. Endocrinol. 80, 101–106. Svensson, J., Fick, J., Brandt, I., Brunström, B., 2014. Environmental concentrations of an androgenic progestin disrupts the seasonal breeding cycle in male three-spined stickleback (Gasterosteus aculeatus). Aquat. Toxicol. 147, 84–91. UNEP, United Nations Environment Programme/WHO, World Health Organisation. 2013. Becher, G., Bergman Å., Bjerregaard, P., Bornman, R., Brandt, I., Heindel, J., Iguchi, T., Jobling, S., Kidd, K., Kortenkamp, A., Muir, D., Ochieng, R., Skakkebaek, N., Toppari, J., Woodruff, T. J., Zoeller, T. In State of the science of endocrine disrupting chemicals — 2012. Villalpando, I., Merchant-Larios, H., 1990. Determination of the sensitive stages for gonadal sex-reversal in Xenopus laevis tadpoles. Int. J. Dev. Biol. 34, 281–285. Vulliet, E., Cren-Olive, C., 2011. Screening of pharmaceuticals and hormones at the regional scale, in surface and groundwaters intended to human consumption. Environ. Pollut. 159, 2929–2934. Vulliet, E., Wiest, L., Baudot, R., Grenier-Loustalot, M.F., 2008. Multi-residue analysis of steroids at sub-ng/L levels in surface and ground-waters using liquid chromatography coupled to tandem mass spectrometry. J. Chromatogr. A 1210, 84–91. Wolf, J.C., Lutz, I., Kloas, W., Springer, T.A., Holden, L.R., Krueger, H.O., Hosmer, A.J., 2010. Effects of 17 beta-estradiol exposure on Xenopus laevis gonadal histopathology. Environ. Toxicol. Chem. 29, 1091–1105. Zucchi, S., Castiglioni, S., Fent, K., 2012. Progestins and antiprogestins affect gene expression in early development in zebrafish (Danio rerio) at environmental concentrations. Environ. Sci. Technol. 46, 5183–5192.