br Materials and methods br Results br
Materials and methods
Discussion The effect of progestogen treatment showed some tendencies affecting the body length and weight, however these data were not verifiable statistically. This observation could be explained with the relatively short exposure period and/or the relatively low number of specimens used for the analyses. At the same time, the hormone exposure resulted in significant changes in the somatic indices of several organs. The liver and the kidney showed significant growth in the treated groups already at 10ng/L concentration which was not affected further by increasing progestogen concentration. In PF-3758309 to that the gonadosomatic indices showed concentration dependent decrease which was significant at the highest hormone concentration (500ng/L). But in this test, we originally used mixed gender for treatments because it was impossible to separate the male and female fish based on phenotype, so this might have affected our observations. However, similarly to our results, kidney hypertrophy were also observed due to ≥40ng/L LNG treatment on female sticklebacks in a previous study (Svensson et al., 2013). VTG genes are present in both female and male fish, but are normally only expressed in mature females, which have high enough plasma 17ß-estradiol levels. VTG is suitable biomarker to detect the estrogenic or androgenic substances (i.e. 17-alpha-ethynylestradiol, 17-beta-trenbolone, dihydrotestosterone) in fish where its level shows increases to estrogens and decreases to androgens (Miracle et al., 2006, Runnalls et al., 2015, Shilling and Williams, 2000). But unfortunately, VTG shows an alternating level depending on the type of progestogen and the applied concentrations. For example, using 55ng/L DRO increases the VTG level, but 55ng/L DRO coadministered with 3.5ng/L PRG decreases it in female zebrafish. DRO treatment at the concentration of 553ng/L decreases but 553ng/L DRO and 33ng/L PRG applied together increases the VTG level (Zucchi et al., 2014). Furthermore, according to other studies 0.5 and 5ng/L LNG had no effect but 25ng/L did already cause a significant increase, while 100ng/L concentration significantly decreased the plasma VTG level in fathead minnow (Runnalls et al., 2015, Runnalls et al., 2013). We observed that the mixture of DRO, LNG and PRG induced an increased VTG level in the liver of female roach but its significant increase was observed only in the 500ng/L group. Based on the above, we suggested that due to variable effects of progestogens the VTG cannot be used as indicator to reveal progestogens contamination. There are several factors that are targeted by both the steroids and the DJ-1 chaperon protein. Both DJ-1 and NRF2 were found to regulate androgen receptors (Niki et al., 2003, Schultz et al., 2014, Takahashi et al., 2001). DJ-1 inhibits PSF, which normally has transcriptional silencing activity and thereby stimulates the neuronal apoptosis (Xu et al., 2005). Furthermore, it is observed that PSF negatively regulates the progesterone receptor (Dong et al., 2005). In addition, DJ-1 has a neuroprotective effect due to its ability to stabilize the NRF2 protein, the antioxidant transcriptional master regulator and thereby prevents cell apoptosis induced by oxidative stress (Clements et al., 2006). Due to the progestogen treatment the DJ-1 protein level increased significantly in the brain and liver of the fish independent from its gender, although the concentration in the brain was substantially lower than in the liver. We also observed a parallel decrease in the serum cholesterol and LDL levels in hormone treated animals. The results suggest first the effects of DJ-1 in cholesterol homeostasis of fish. In DJ-1-knockdown cells and DJ-1-knockout mice a reduced expression of the LDLR-gene was observed and increase in the cholesterol and LDL serum level was reported (Dong et al., 2005, Yamaguchi et al., 2012). It is suggested therefore that the decrease in the serum LDL level induced by mixture of DRO, LNG and PRG and DJ-1 was mediated through stimulation of the expression of LDLR, which facilitates reuptake subsequently. We hypothesized that progestogens might also directly affect the expression of DJ-1 protein and thereby indirectly stimulate LDLRs (see Fig. S1). It is well known that serum cholesterol, LDL-, HDL-cholesterol and LDLR levels are not effected remarkably by gender in vertebrates. Because we hypothesized based on our observations that DJ-1 potentially affected directly the LDLRs, and indirectly influenced the serum cholesterol, LDL-, HDL-cholesterol levels, therefore, the mixed gender is also suitable in this type of experiments. Thus, the DJ-1 protein could be a more applicable indicator than VTG in progestogen contamination where mainly female specimens are needed. However, the usability of DJ-1 as potent biomarker for environmental progestogen contaminations in freshwater ecosystems should also be confirmed using other species in the future.