Monocrotophos pesticide decreases the plasma levels of total 3,3',5-triiodo-l-thyronine and alters the expression of genes associated with the thyroidal axis in female goldfish (Carassius auratus).

Our recent study showed that monocrotophos (MCP) pesticide disrupted the hypothalamic-pituitary-thyroid (HPT) axis in male goldfish (Carassius auratus); however, the effects of MCP on the thyroid system in female goldfish are remain unclear. In the present study, plasma thyroid hormone (TH) and thyroid-stimulating hormone (TSH) levels were evaluated in female goldfish exposed to 0.01, 0.10, and 1.00 mg/L of 40% MCP-based pesticide for 21 days in a semi-static exposure system. Expression profiles of HPT axis-responsive genes, including transthyretin (ttr), deiodinases (d1, d2, and d3), tshβ, thyrotropin-releasing hormone (trh), and corticotrophin-releasing hormone (crh), were determined. The results indicated that MCP decreased the plasma levels of total 3,3',5-triiodo-l-thyronine (TT3) and the ratio of TT3 to total 3,3',5,5'-l-thyroxine (TT4), and induced alternative expression of TH-related genes. Exposure to 0.01 and 0.10 mg/L MCP pesticide resulted in the up-regulation of ttr mRNA. The reduction of plasma TT3 levels was partly attributed to an increase in the metabolism of T3 in the liver, as revealed by the highly elevated hepatic d1 and d3 mRNA levels in the MCP treatment groups, and the expression of hepatic d3 showed a negative correlation with the plasma TT3/TT4 levels in females. Moreover, the plasma TSH levels were lower in females exposed to 0.01 and 0.10 mg/L MCP pesticide, whereas the up-regulation of tshβ mRNA levels was compensated by the decreased plasma TT3 levels. These results indicated that MCP had the potential to influence several pathways of HPT axis homeostasis in female goldfish.

Introduction

In fish, growth and reproduction are, at least partly, under the control of thyroid hormones (THs), 3,3′,5,5′-l-thyroxine (T4) and 3,3′,5-triiodo-l-thyronine (T3) [1], [2]. Considering that reproduction in females from the early development of follicles to oocyte maturation and ovulation involves a significant investment of energy to support the related physiological functions, the reproductive-related allocation of energy is particularly important during ovarian development in oviparous species, including fish. Since THs have profound effects on energy metabolism, for example, they are the major regulators of oxidative energy metabolism at the level of the mitochondria in teleost fish [3], they should invariably be involved in the multifactorial regulation of metabolism associated with reproduction. They might also act as direct modulators of the reproductive cycle. A previous study on goldfish (Carassius auratus) has shown that T4 could act synergistically with gonadotropin to influence ovarian development by increasing ovarian sensitivity to gonadotropic stimulation [4]. Moreover, the expression of steroidogenic enzymes and steroid receptors in goldfish could be modulated by THs [5].

Monocrotophos (MCP; CAS number, 6923-22-4), an organophosphorus pesticide, has been banned in developed countries due to its high toxicity; however, it is still used extensively in agricultural practices in developing countries such as Pakistan, China, and India, where high pesticide residue levels have led to excessive MCP levels in the environment. For example, the concentrations of MCP detected in water sources in China and rain water in India were 0.165 and 4 µg/L, respectively [6], [7]. Anjum and Malik [8] reported the presence of organophosphorus pesticide in the industrial wastewater around Lucknow, India, and the concentration of MCP was determined to be 8.32±3.9 ng/mL. Our recent study showed that the MCP pesticide exhibited thyroid-disrupting effects by interfering with the thyroidal axis, resulting in decreased plasma total T3 (TT3) levels in male goldfish [9]. Further, a gender difference in the thyroid system in response to metolachlor has been reported by Jin et al. [10], and the authors considered that endogenous sex hormones might modify the response of the thyroid system to the environmental chemical. In our previous study, MCP was found to interfere with the reproductive axis via several pathways, thereby inducing increases in plasma 17β-estradiol (E2) levels and decreases in testosterone (T) levels in both male and female goldfish [11], [12]. Balanced plasma TH levels are known to be crucial for normal reproductive function; in mammals, both hyperthyroidism and hypothyroidism were found to result in reproductive impairment and lower fertility [13]. Considering gender differences in the response of the thyroid endocrine system in fish exposed to environmental pollutants, the thyroid disruption effects of MCP in females might be speculated to be similar to those found in males. The effects of MCP on the thyroid endocrine system in females, however, remain unclear.

THs are synthesized and secreted by the thyroid follicles under the control of the HPT axis. In teleosts, the thyrotropin-releasing hormone (TRH) and/or corticotrophin-releasing hormone (CRH), released from the hypothalamus, coordinate the HPT axis function by controlling the release of thyrotropin (TSH) from the pituitary, which could stimulate TH synthesis and release [1], [14]. Most of the plasma THs in fish are bound to transthyretin (TTR), a specific TH transport protein in teleosts [15], and only free hormones can enter target cells to elicit a response. In the liver and some other peripheral tissues, three types of deiodinases (type I, D1; type II, D2; and type III, D3) are known to control the conversion of T4 to the more physiologically active T3 or the production of metabolically inactive counterparts [16]. Such complex regulatory pathways are involved in thyroid homeostasis. Therefore, environmental chemicals can act at multiple stages in the HPT axis.

In the present study, following administration of 40% MCP pesticide to female goldfish, plasma TH levels, including TT3, total T4 (TT4), and the physiologically relevant free T3 (FT3) and free T4 (FT4) levels, were determined, and the changes in endocrine-mediated responses along the HPT axis were evaluated, including the regulation of TRH and CRH, synthesis and secretion of TSH, and expression of deiodinases (d1, d2, and d3) and transthyretin (ttr) genes.

Materials and Methods

Pesticide

MCP pesticide (3-hydroxyl-N-methyl-cis-crotonamidedimethyl phosphate, 40% water-soluble preparation) was purchased from the Qingdao pesticide factory in China. The concentration on the label was 40%, which was consistent with the actual concentration determined by gas chromatography (40±0.1%) [17]. The half-life of MCP is approximately 66 days at pH 7.0 and 20°C [18].

Fish exposure and sample protocols

Sexually mature goldfish (C. auratus) with mature and fully developed gonads (8.7±0.8 cm standard length; 21.2±4.1 g wet weight; 6.8±0.3% gonadal somatic index) were obtained from a local dealer in Qingdao, PR China, and sampled in late spring. Fish were handled according to the National Institute of Health Guidelines for Handling and Care of Experimental Animals. The animal utilization protocol was approved by the Institutional Animal Care and Use Committee of the Ocean University of China. Fish were maintained in a 70-L aquarium containing 50-L dechlorinated tap water at ambient temperature (18±2°C) with dissolved oxygen content of 7.0±0.1 mg/L and were fed small shrimp daily.

After the female goldfish were acclimated to laboratory conditions for 2 weeks, they were exposed to nominal 0.01, 0.10, and 1.00 mg/L 40% MCP pesticide (4, 40, and 400 µg/L equivalent levels of pure MCP), the concentrations of which were 1/10,000; 1/1,000; and 1/100 of the 96-h LC50 [11], respectively. The experiments were conducted in the same 70-L aquarium containing 50-L dechlorinated tap water by using a semi-static toxicity test (20-L water renewal daily to constantly maintain the MCP concentration). Each group of fish was exposed in three aquaria (seven fish/tank) and a control (dechlorinated tap water) was included in the exposure design. Less than 10% mortality was observed in all the treatments during the experiment, and abnormal behaviors such as unusual swimming pattern or bumping against the tank were observed in the group exposed to the highest dose.

After 21 days of exposure, all fish were anesthetized with 75 mg/L MS-222 (Sigma, St. Louis, MO, USA) and sampled between 9∶00 and 11∶00 h to avoid the possible influences of diurnal fluctuations in hormone levels on the results [19]. Fish from each group were divided into two subgroups at the time of sampling. One subgroup of 9–10 fish was sampled to collect 0.6–0.8 mL blood for TH measurements, whereas the remaining fish were used to collect blood with the same volume for TSH and cortisol measurements. Blood was collected from the caudal vein by using chilled heparinized syringes and maintained on ice. Plasma samples were obtained after centrifugation at 1000×g for 10 min and stored at −80°C until the hormone assay could be performed. The liver, pituitary, and hypothalamus tissues were dissected (n = 9), frozen in liquid nitrogen, and then stored at −80°C for the quantification of gene expression by real-time polymerase chain reaction (PCR).

Hormone assay

The plasma levels of TT3, TT4, FT3, FT4, TSH, and cortisol were measured using radioimmunoassay (RIA) by using commercially available kits following the protocols provided by the manufacturers (Beijing North Institute of Biological Technology, Beijing, China). The RIA kits for human TT3, TT4, FT3, FT4, TSH, and cortisol were validated for use with goldfish samples by showing parallelism between a series of diluted and spiked samples in relation to the standard curve included with the assay kits. Standards and samples were added to test tubes in duplicate. The assay detection limits were 0.05 ng/mL for T3, 2 ng/mL for T4, 0.06 pg/mL for FT3, 0.23 pg/mL for FT4, 0.1 µIU/mL for TSH, and 1 ng/mL for cortisol. The inter- and intra-assay coefficients of variation for all the above-mentioned hormones were <10% and <15%, respectively.

Gene expression analysis

Total RNA from each tissue was isolated using the phenolic reagent TRIzol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. The extracted RNA was measured by spectrometry at OD260/280 before treatment with DNase I (Promega, Madison, WI, USA). Equal amounts of RNA (1 µg) were reverse-transcribed into cDNA in 20-µL reactions containing 10 pmol oligo(dT)20, 4 µL 5× RT Buffer, 2 µL dNTP mixture, 10 U RNase inhibitor, and 1 µL ReverTra Ace (Toyobo, Tokyo, Japan). Reverse transcription reactions were conducted in a Bio-Rad DNA Thermal Cycler (Hercules, CA, USA) at 42°C for 20 min and terminated for 5 min at 85°C.

Oligonucleotide primers were designed for the specific amplification of tshβ, trh, crh, ttr, d1, d2, d3, β-actin, and 18S rRNA by using the Primer Premier 5.0 software (PREMIER Biosoft Int., Palo Alto, USA) and the sequences for goldfish available from the GenBank database (Table 1). The amplifications were performed using an Eppendorf MasterCycler ep RealPlex 4 (Eppendorf, Wesseling-Berzdorf, Germany). Parallel PCR analyses were conducted to amplify the cDNA of the target and reference genes. Real-time PCR was performed in 20-µL reaction mixtures containing 1× SYBR Premix Ex Taq, 0.4 µM of each primer, 0.4 µL of ROX reference Dye (Takara Bio Inc., Shiga, Japan), and 4 µL of first-strand cDNA (template). The thermal profile was 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and 60°C for 30 s. Amplification of a single product was ensured by performing a melting curve analysis by using the PCR products obtained at the end of each PCR run. In addition, 2% agarose gel electrophoresis of the PCR products was performed to confirm the presence of single amplicons having the predicted sizes (data not shown). The β-actin and 18S rRNA transcripts were used as housekeeping genes to standardize the results and eliminate variations in mRNA and cDNA quantity and quality. Neither the β-actin nor 18S rRNA levels were affected by any of the experimental conditions used in the study. For each reaction, the relative target gene mRNA expression levels were normalized to the geometric mean of β-actin and 18S rRNA expression levels by using the formula 2−ΔΔCt and plotted on a logarithmic scale [20].

Table 1

Nucleotide sequences of the primers used for real-time polymerase chain reaction and product sizes.

Gene	GenBank accession no.	Primer sequence (5′–3′)	Amplicon size (bp)
tshβ	AB003584	F: TGGCTGTCAACACCACCATCTG	94
		R: CCCCTTTGAACCAGGAAACGAG
trh	AB179819	F: GAACAAGAGTCGCTGTAAGGGAG	102
		R: GGATGCTGTGGAAAACAATGAAC
crh	AF098629	F: CCAGAATTATCTCCGATCCCCAG	109
		R: CCCAACTTTCCCCTCCAACAG
ttr	EU313781	F: TTCAGATGCTCACTGCCCTTTG	103
		R: CCCACCTTGTTCTTGGCGATAG
d1	EU313785	F: GGCAGATTTCCTTGTTGTTTACC	112
	EF190702	R: CGCAGCCAGTCTCTCCTCC
d2	EU313786	F: CAAACTCCAAGGTGGTGAAGGT	86
	EF190703	R: GTCCAGCAGATGGCACTCGT
d3	EF190704	F: GCTGTTCCTGGATTTCCTGTGC	163
	EU313787	R: GAAGAAGTCCAGCTTGTGTCCGT
β-actin	AB039726	F: GAAACTGGAAAGGGAGGTAGC	115
		R: CTGTGAGGGCAGAGTGGTAGA
18S rRNA	AF047349	F: AGAAACGGCTACCACATCCAAG	169
		R: GCACCAGATTTGCCCTCCAG

Goldfish contain duplicate genes encoding for D1, D2, and D3, and the primer pairs amplify the two genes.

Statistical analysis

The experimental data were presented as the means ± standard deviations, and the differences between the control and each exposure group were evaluated using one-way analysis of variance (ANOVA) followed by Tukey’s test. Before parametric analysis, assumptions of normality and homogeneity of variances were assessed using probability plots and normality tests. Pearson’s correlation coefficient was used to calculate the relationship between the expression of certain genes involved in the HPT axis and the plasma TH levels. Values were considered significant when 0.01

Results

Effects of MCP pesticide on plasma TT3 and TT4 levels and the ratio of TT3 to TT4

The TT3 content in the plasma of the control female goldfish was 1.31±0.29 ng/mL, whereas that in the fish exposed to 0.10 and 1.00 mg/L MCP pesticide was significantly decreased by 30% to 0.92±0.23 ng/mL (0.01MCP treatment, the TT4 content was 7.21±1.55 ng/mL in the control group, and there was no significant change in the TT4 content of any of the MCP pesticide-treated groups (Fig. 1B). There was a dose-related decrease in the ratio of TT3 to TT4 after MCP treatment (P<0.01; Fig. 1C).

Figure 1

Quantification of plasma total 3,3′,5-triiodo-l-thyronine (T3) and total 3,3′,5,5′-l-thyroxine (T4) levels in female goldfish exposed to 0, 0.01, 0.10, and 1.00 mg/L 40% monocrotophos (MCP) pesticide.

(designated C, MCP0.01, MCP0.10, and MCP1.00, respectively). The data are presented as the means ± standard deviations (n = 9). Asterisks indicate statistically significant differences from the control group (*0.01

Effects of MCP pesticide on plasma FT3 and FT4 levels and hepatic ttr mRNA expression

The basal FT3 levels in the plasma of the control female goldfish were 2.39±0.39 pg/mL; they were significantly reduced by 32%, 32%, and 50% in the 0.01, 0.10, and 1.00 mg/L MCP pesticide-treated groups, respectively (P<0.01; Fig. 2A). Plasma FT4 levels in the control females were 2.10±0.80 pg/mL; they were significantly increased to 4.71±0.85 pg/mL in the 0.01 mg/L MCP pesticide-treated group (P<0.01) but decreased to 1.26 pg/mL in the 1.00 mg/L MCP pesticide-treated group (P<0.01; Fig. 2B). Furthermore, the mRNA expression of hepatic ttr was significantly higher after treatment with 0.01 and 0.10 mg/L of MCP pesticide (P<0.01), whereas there was no significant difference in the hepatic ttr mRNA expression between the group treated with the highest dose and the control group (Fig. 2C).

Figure 2

Quantification of plasma free 3,3′,5-triiodo-l-thyronine (T3) and free 3,3′,5,5′-l-thyroxine (T4) contents and the relative mRNA expression levels of hepatic transthyretin (ttr) in female goldfish exposed to 0, 0.01, 0.10, and 1.00 mg/L 40% monocrotophos (MCP) pesticide.

(designated C, MCP0.01, MCP0.10, and MCP1.00, respectively). For panel C, fold change (Y axis) represents the expression of the target gene mRNA relative to that of the control group (equals 1 by definition). The data are presented as the means ± standard deviations (n = 9). Asterisks indicate statistically significant differences from the control group (*0.01

Effects of MCP pesticide on d1, d2, and d3 mRNA expression levels in the liver, brain, and kidney

The d1 mRNA levels in the liver were significantly increased by 0.72- and 2.15-fold and the hepatic d2 mRNA expression was significantly up-regulated by 1.30- and 0.84-fold after treatment with 0.01 and 0.10 mg/L MCP pesticide, respectively (P<0.01), relative to those in the control (Fig. 3A and 3B). Hepatic d3 mRNA levels were significantly higher in all the MCP-treated groups, especially in the 0.01 and 0.10 mg/L MCP pesticide-treated groups (2.66- and 4.50-fold higher, respectively), than those in the control (P<0.01; Fig. 3C).

Figure 3

Relative mRNA expression levels of types I, II, and III deiodinases (d1, d2, and d3) in the liver, brain, and kidney of female goldfish exposed to 0, 0.01, 0.10, and 1.00 mg/L 40% monocrotophos (MCP) pesticide.

(designated C, MCP0.01, MCP0.10, and MCP1.00, respectively). Fold change (Y axis) represents the expression of the target gene mRNA relative to that of the control group (equals 1 by definition). The data are presented as the means ± standard deviations (n = 9). Asterisks indicate statistically significant differences from the control group (*0.01

With regard to the gene transcription level of deiodinase in the brain (Fig. 3D–3F), there was no significant difference in the d1 gene transcription in any of the MCP-exposed groups. MCP pesticide exposure significantly stimulated d2 gene transcription in the 0.01 mg/L and 0.10 mg/L groups (P<0.01) and d3 gene transcription in the 0.10 mg/L group (P<0.01). However, the transcription of both d2 and d3 genes was significantly inhibited after exposure to 1.00 mg/L MCP pesticide (P<0.01).

Transcription of the d1 gene in the kidney was significantly up-regulated after exposure to 0.10 mg/L MCP (P<0.01; Fig. 3G). The transcription of d2 was significantly down-regulated after exposure to 0.01 and 1.00 mg/L MCP pesticide, but was stimulated in the 0.10 mg/L group (0.01MCP-exposed group (4.0-fold, P<0.01; Fig. 3I).

Effects of MCP pesticide on pituitary tshβ mRNA expression and plasma TSH levels

The mRNA expression of pituitary tshβ was significantly up-regulated in both the 0.10 and 1.00 mg/L MCP pesticide-treated groups (P<0.01; Fig. 4A). In contrast, the plasma TSH levels in the 0.10 and 1.00 mg/L MCP pesticide-treated groups were markedly decreased to 0.85±0.16 µIU/mL (P<0.01) and 0.91±0.14 µIU/mL (0.01

Figure 4

Relative mRNA expression levels of pituitary thyroid-stimulating hormone β subunit (tshβ) and quantification of plasma TSH content in female goldfish exposed to 0, 0.01, 0.10, and 1.00 mg/L 40% monocrotophos (MCP) pesticide.

(designated C, MCP0.01, MCP0.10, and MCP1.00, respectively). For panel A, fold change (Y axis) represents the expression of the target gene mRNA relative to that of the control group (equals 1 by definition). The data are presented as the means ± standard deviations (n = 9). Asterisks indicate statistically significant differences from the control group (*0.01

Effects of MCP pesticide on hypothalamic trh and crh mRNA and plasma cortisol levels

The expression of hypothalamic trh was significantly down-regulated after treatment with the MCP pesticide (Fig. 5A). The transcription of the crh gene was significantly stimulated after treatment with 0.10 mg/L MCP pesticide (P<0.01), but inhibited in the 1.00 mg/L MCP pesticide-treated group (0.01cortisol content was 142.01±24.11 ng/mL in the control female goldfish and was significantly decreased to 74.53±18.26 ng/mL and 89.88±22.05 ng/mL in the 0.10 and 1.00 mg/L MCP pesticide-treated groups, respectively (P<0.01; Fig. 5C).

Figure 5

Relative mRNA expression levels of thyrotropin-releasing hormone (trh) and corticotrophin-releasing hormone (crh) in the hypothalamus glands and quantification of plasma cortisol content in female goldfish exposed to 0, 0.01, 0.10, and 1.00 mg/L 40% monocrotophos (MCP) pesticide.

(designated C, MCP0.01, MCP0.10, and MCP1.00, respectively). For panels A and B, fold change (Y axis) represents the expression of the target gene mRNA relative to that of the control group (equals 1 by definition). The data are presented as the means ± standard deviations (n = 9). Asterisks indicate statistically significant differences from the control group (*0.01

The correlation coefficient for plasma THs levels and the expression of certain genes related to the HPT axis

The correlation coefficient for plasma THs levels and the expression of certain genes associated with the HPT axis are shown in Table 2. Significant correlations of the ratio of TT3 to TT4 with hepatic d3 expression, TT3 with pituitary tshβ expression, and FT4 with hepatic ttr expression were found.

Table 2

Pearson correlation coefficients for the plasma thyroid hormone levels and the expression of certain genes related to the hypothalamus-pituitary-thyroid axis.

Paired samples	Correlation coefficient
TT3/TT4 vs. hepatic d1	−.189
TT3/TT4 vs. hepatic d2	−.232
TT3/TT4 vs. hepatic d3	−.384*
TT3 vs. pituitary tshβ	−.469*
TT3 vs. hypothalamus trh	.356
TT3 vs. hypothalamus crh	.025
FT3 vs. hepatic ttr	−.069
FT4 vs. hepatic ttr	.396*

*Statistically significant at 0.01

Discussion

An increasing number of studies have reported that groups of pesticides, including acetochlor, amitrole, and metolachlor, have the potential to influence several steps in HPT axis homeostasis and to induce THs disturbance in adult fish, particularly with respect to sex differences occurring in response to chemical-induced thyroid system disruption [10], [21], [22]. For example, Li et al. [21] showed that TH-related genes such as malic enzyme and sodium iodide symporter were significantly down-regulated in the brains of the rare minnow Gobiocypris rarus, and that the expression of these genes in females was more sensitive to acetochlor than that in males. Recently, we found that a 21-d exposure to MCP pesticide caused significant decreases in plasma TT3 levels and TT3-to-TT4 ratios in male goldfish [9]; however, whether similar effects occur in females is not clear.

In the present study, although the plasma levels of TT4 remained unchanged, those of TT3, FT4, and FT3 in female goldfish were significantly altered after a 21-day exposure to MCP pesticide, suggesting a failed adaptation and auto-regulation of THs homeostasis. Almost all THs circulating in the plasma are bound to transporter proteins, and the equilibrium of TH binding to the plasma proteins determines the concentration of free THs within the plasma [23]. TTR, which is primarily a secretory product of the liver, has been proposed to be the major TH-carrier protein that binds THs and transports them to target tissues in fish [24]–[26]. In our study, TTR gene expression was up-regulated after exposure to 0.01 and 0.10 mg/L MCP pesticide. This up-regulation might have resulted in higher TTR mRNAs and thus TTR proteins, leading to decreases in plasma FT3 and/or FT4 levels. Moreover, the expression of ttr showed a positive correlation with plasma FT4 levels in females. Notably, less than 1% of plasma TT4 is free with 99% reversibly bound to plasma proteins [23]. Indeed, changes of such small amount of FT4 contents may not represent a dynamic circulating TT4 reservoir and vice verse. For example, in brown trout (Salmo trutta) fed diets enriched with β–Tetrabromoethylcyclohexane for 56 days, there was no significant difference among treatments in FT4, but TT4 was significantly reduced in the high dose group relative to the control [27]. In the 0.01 mg/L MCP group in particular, one possible explanation for the increased plasma FT4 levels could be that feedback systems attempt to respond to the reduction in plasma TT3 levels. However, the stimulatory effects of 0.01 mg/L MCP pesticide on plasma FT4 levels in females were not observed in males [9], indicating that the thyroidal system in female goldfish is more sensitive to MCP than that of males. In the group treated with the highest dose of MCP pesticide, the total TH level was more readily responsive to the changes in free hormone levels, since TTR gene expression was not stimulated in this group.

Among the thyroid follicle secretions, T4 is the predominant circulating hormone in the blood of fish, and T3 appears to be produced largely by enzymatic deiodination of T4 in the peripheral tissues [28]. Consequently, plasma T3 levels mostly decline due to a drop in thyroidal T4 production and secretion and/or changes in the peripheral TH metabolism [29]. Our finding that decreases in plasma TT3 levels along with relatively stable plasma TT4 levels suggested possible changes in peripheral TH deiodination or metabolism. Iodothyronine deiodinases play a crucial role in the mechanism of TH biotransformation in extra-thyroidal tissues. Conversion from T4 to T3 is mediated by outer-ring deiodination (ORD). Both inner-ring deiodination (IRD) and ORD are involved in the inactivation of T4 to 3,3′,5′-triiodo-l-thyronine (rT3) and T3 and rT3 to 3,3′-diiodo-l-thyronine (3,3′-T2) [16], [30]. Three types of deiodinases have been identified in teleosts: type I (D1) has both ORD and IRD activities, whereas types II (D2) and III (D3) have ORD and IRD activity, respectively [31]–[34]. At the end of the 21-day exposure to 0.01 and 0.10 mg/L MCP pesticide, the transcription of all three types of deiodinases was stimulated in the liver: changes in the expression of d1 and d3 mRNA were higher than those of d2 mRNA. Further, in the 1.00 mg/L treatment, only hepatic d3 mRNA levels were significantly up-regulated. Van der Geyten et al. [35] showed that the changes in hepatic deiodinases activities tend to be consistent with those of their mRNA levels, indicating pre-translated regulation of the hepatic deiodinases. Both D1 and D3 are involved in the metabolism of THs. D3 catalyzes the degradation of T4 and T3 to inactive metabolites and could protect tissues from an excess of biologically active TH, namely, T3 [36]. Although D1 also has the ability to degrade T3, it is less effective than D3, and its preferred substrate is rT3 [37]. Therefore, the stimulated enzymatic metabolism of T3 in the liver could cause a reduction in the levels of circulating TT3. Consistently, a significant negative correlation between hepatic d3 mRNA expression and the ratio of TT3 to TT4 was observed. It is also worth noting that, in the highest dose group, alterations of gene expression of the three types of deiodinases in females were not consistent with those in males in which hepatic d1 and d2 mRNA levels were significantly down-regulated, whereas d3 mRNA levels were not affected [9].

Although liver is considered to be the main peripheral source of circulating T3, TH deiodination also occurs in other extra-thyroidal tissues such as brain and kidney; however, the available T3 derived from these tissues is primarily utilized by the same tissues itself [38]. Exposure to MCP pesticide had no effect on the d1 mRNA expression in the brain, but stimulated the transcription of d1 gene in the kidney of female goldfish. The functional roles of D1 might indicate why the patterns of d1 mRNA regulation varied among tissues after exposure to MCP. Recent studies have suggested that the major role of D1 might be to clear rT3 and sulfated iodothyronines from circulation. Indeed, it functions as a scavenger enzyme to remove inactive iodothyronines and recycle iodine within the organisms [39]–[41]. Thus, 0.10 mg/L MCP pesticide might enhance the metabolism of THs by up-regulating the expression of d1 mRNA in the kidney. Further, the specific patterns of deiodinase-mediated gene regulation show tissue-specific variations, probably in accordance with the distinct thyroid status and tissue-specific requirements for available THs [42]. In the brain, the transcriptions of d2 and d3 gene were stimulated after treatment with 0.01 and 0.10 mg/L MCP but inhibited by 1.00 mg/L MCP pesticide, whereas, in the kidney, the transcriptions of d2 and d3 gene were stimulated after treatment with 0.10 mg/L MCP but inhibited by treatment with 0.01 and 1.00 mg/L MCP pesticide. Such expression profiles of d2 and d3 changing in a parallel way indicates that intracellular T3 levels are tightly controlled [43].

THs are also regulated by TSH and TRH/CRH, but can interfere with the synthesis and release of TSH and TRH/CRH via feedback mechanisms. This regulatory system plays an important role in maintaining the homeostatic balance along the HPT axis [44], [45]. In the present study, the tshβ mRNA levels were up-regulated and showed a negative correlation with plasma TT3 levels. The T3 feedback mechanism is known to control TSH expression in certain teleosts, and Sohn et al. [46] provided evidence that T3 acted directly on the pituitary and inhibited tshβ gene expression in goldfish, probably via the negative T3-responsive elements in the tshβ gene. Hence, the decreased plasma TT3 levels would feedback to stimulate the transcription of tshβ, and similar negative feedback has also been observed in males [9]. TSH is a glycoprotein secreted by the pituitary thyrotrope cells. In our previous study, Bing [47] reported direct damage of the MCP pesticide to the structure of thyrotrope cells in adenohypophysis. The partly dissolved nuclear membrane, prominent dilated rough endoplasmic reticulum, and slightly dissolved mitochondrial cristae indicated a reduction on the hormone secretion yield of thyrotrope cells. Accordingly, in the current study, although pituitary tshβ mRNA levels were elevated, the reduced plasma TSH levels in the 0.10 and 1.00 mg/L MCP treatments were probably resulted from the diminished TSH protein synthesis and secretion of thyrotrope cells. In goldfish, both TRH and CRH might be involved in the regulation of the pituitary-thyroid axis, and TRH might act as a multifunctional hypophysiotropic factor. For instance, TRH is a potent stimulator of alpha melanocyte-stimulating hormone release from the pars intermedia cells in various species of teleost fish, including goldfish [48], and is also involved in the regulation of feeding and locomotor behaviors [49]. Therefore, the up-regulated trh mRNA expression levels caused by the MCP pesticide would not be the only response of the HPT axis. Instead, CRH might act as a TSH-releasing factor in lower vertebrates [14], and might also be mediated and/or intensified by a concomitant increase in corticosteroids [50]. It is also worth noting that CRH plays important roles in mediating the hypothalamic-pituitary-interrenal axis (HPI axis) to produce cortisol in response to stress, and that cortisol could also tightly control the release of CRH via a negative feedback loop [51], [52]. The reductions in plasma cortisol levels in the 0.10 mg/L group would be the result of MCP interference with cortisol synthesis and metabolism at some level. A recent study in our lab revealed that 0.10 mg/L MCP pesticide decreased interrenal synthesis of cortisol and further promoted the metabolism of cortisol, resulting in lower overall cortisol levels and a diminished stress capacity in female zebrafish (Danio rerio) [53]. In goldfish, whether CRH acts on thyroid activity either directly or via adrenal steroids has yet to be determined; the decreased plasma TT3 and cortisol levels would also signal the up-regulation of crh gene expression, as observed in the 0.10 mg/L group. On the other hand, in the 1.00 mg/L MCP pesticide group, both plasma cortisol levels and hypothalamic crh mRNA levels were decreased, indicating that the high concentration of MCP pesticide disrupted the feedback mechanisms along the HPI axis. Some investigators have suggested that hyperactivity of the cortisol-producing cells associated with severe stress might lead to an inter-renal exhaustion with very low levels of cortisol [54], [55].

Our results showed that the MCP pesticide disturbed the THs homeostasis and interfered with the transport and conversion of THs, synthesis and secretion of pituitary TSHs, and regulation of hypothalamic TRH/CRH in female goldfish. Similar to the findings in males, MCP pesticide decreased plasma TT3 and FT3 levels with relatively stable plasma TT4 levels, and profiles of the relative abundance of deiodinase transcripts were observed in the liver, brain, and kidneys in female goldfish; however, 0.01 mg/L MCP pesticide merely enhanced the plasma FT4 levels in females, and gender differences existed in the hepatic deiodinase transcripts at the highest dose. Although the reproductive and thyroidal endocrine systems have been reported to be related to the neuroendocrine control of many complex functions, including growth, metabolism, and reproduction in teleosts [1], [2], [5], and many chemicals, including polychlorinated biphenyl and perfluorooctanoic acid, could simultaneously cause a variety of effects along both endocrine systems [56]–[59], further studies unraveling the interaction between the MCP-induced reproductive axis disruption and thyroidal axis disruption are necessary to explain gender differences observed in the thyroid system response to MCP exposure in goldfish.