Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Methoxychlor is a chlorinated hydrocarbon insecticide that is used worldwide in both agriculture and the home. Although the pure, parent compound has not been reported to have any hormonal activities, it has been found that its primary metabolites, the mono- and bis-hydroxy methoxychlor, are estrogenic (Bulger et al., 1978; Kupfer et al., 1990; Walters et al., 1993). In a previous study, we found that the administration of methoxychlor to rats also inhibited the hepatic, microsomal conversion of the prohormone thyroxine (T4), to the active hormone thyronine (T3; Zhou et al., 1995), indicating that methoxychlor may interfere with this step in normal thyroid hormone metabolism. Our initial study suggested that inhibition resulted from binding of one of its metabolites to a 56-kDa protein that had been identified as the iodothyronine 5'-deiodinase, form I (ID-I; Boado et al., 1988; Santini et al., 1992; Zhou et al., 1995). This protein is 98% homologous to the thiol:protein disulfide oxidoreductase, form Q5 (ERp55; E.C.1.8.4.2; Boado et al., 1988). This protein also is known as protein disulfide isomerase and thyroxine-binding protein (Holtzman, 1998). Yet, at the time there was a controversy in the literature whether this protein is truly ID-I.

The deiodination of T4 is catalyzed by a family of enzymes, ID-I-III (E.C.1-11-1-8). The primary form found in the liver is a microsomal enzyme that has been designated form I. This enzyme catalyzes both the synthesis of T3 from T4 and the degradation of T3 into inactive metabolites (Berry et al., 1993). Two groups have suggested that ID-I was ERp55 (Edman et al., 1985; Boado et al., 1988; Goswami and Rosenberg, 1990; Sakane and Chopra, 1990; Srivastava et al., 1991; Santini et al., 1992). This protein also is known as protein disulfide isomerase (PDI) and thiol:protein disulfide oxidoreductase, form Q5 (TPDO-Q5) (Holtzman, 1998). The identification of ERp55 as ID-I was based on two observations. First, in partial purification studies ID-I activity appeared to copurify with ERp55 (Goswami and Rosenberg, 1990; Sakane and Chopra, 1990). Second, Mol et al. (1984) reported that the T3 affinity label n-bromoacetyl-3,3'5-triiodothyronine (BrAcT3) bound to ERp55. This reagent also was found to inhibit ID-I activity.

Yet, Schoenmakers et al. (1989) found that [I¹²⁵]BrAcT3 also labeled a second microsomal protein with a molecular mass of 27 kDa. Furthermore, they reported that although substrates and inhibitors of ID-I blocked the labeling of the 27-kDa protein they had no effect on the labeling of the ERp55. Furthermore, although membrane-bound ERp55 is insensitive to trypsin digestion, treatment of microsomes with this protease abolished both the labeling of the 27-kDa protein and the ID-I activity, but had no effect on the labeling of ERp55. Similarly, rat pancreatic microsomes had neither ID-I activity nor labeled the 27-kDa protein, but had high concentrations of ERp55 that was extensively labeled by the [I¹²⁵]BrAcT3. Finally, purified ERp55 had no ID-I activity.

The 27-kDa protein was subsequently cloned and sequenced (Mandel et al., 1992) and numerous studies have indicated that it catalyzes ID-I activity (Berry et al., 1990; 1991a,b; Kohrle et al., 1990; Tagami et al., 1991). Because methoxychlor did not bind to the 27-kDa protein these data would suggest that our original interpretation of the biochemical basis for the decrease in ID-I activity observed after methoxychlor administration was in error and that this insecticide does not block the conversion of T4 to T3 through adduct formation with ID-I (Zhou et al., 1995).

Thus, in this study we have examined the possible biochemical basis for the decreased ID-I activity observed in animals receiving methoxychlor. Our data indicate that this inhibition is due to decreased transcription of the mRNA for the 27-kDa protein. Because methoxychlor has significant stereochemical similarities to T4 (Zhou et al., 1995), our data raise the possibility that decreased transcription is due to binding of the parent compound to nuclear thyroid-binding proteins, thereby altering their interaction with the promoter region of the gene encoding ID-I.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. NADPH, glucose 6-phosphate, glucose-6-phosphate dehydrogenase (type IV), methoxychlor, Tris, thyroxine, and thyronine were purchased from Sigma Chemical Co. (St. Louis, MO). Polyethylene glycol 6000 (PEG) was purchased from Fluka (Ronkonkoma, NY). Polyclonal antibodies to the two TPDO isoforms, ERp55 and ERp57, were developed in laying hens as previously described (Srivastava et al., 1991; Chen et al., 1996). Antibodies to the 27-kDa protein were prepared against a synthetic peptide based on the implied sequence of the protein (Berry et al., 1994). The antibodies to ID-I were kindly provided by Dr. P. Reed Larsen (Brigham and Women's Hospital, Boston, MA).

Enzyme Preparation. Animals used in these studies were 150 to 200 g, fed male CD rats from Harlan Laboratories (Haslett, MI). They received either methoxychlor in corn oil or vehicle alone (0.8 ml) by gavage. The animals were sacrificed with CO₂ and their livers were removed, weighed, and minced in cold KCl-Tris (150 mM-50 mM; pH 7.2). The individual livers were separately homogenized in KCl-Tris (4 ml/g of liver) and the homogenates were centrifuged for 20 min at 10,000g. The supernatants were centrifuged for 1 h at 100,000g. The pellets were washed and resuspended in KCl-Tris and recentrifuged for 1 h at 100,000g. The final microsomal pellets were suspended in KCl-Tris (0.5 ml/g of liver). The microsomal suspensions were aliquoted and stored at 80°C until analyzed. In control studies, we determined that freezing at this temperature did not affect the ID-I activity.

Determination of ID-I Activity. T3 and T4 were dissolved in NaOH (0.1 M) and diluted in 5% BSA. Dithiothreitol (DTT) was freshly dissolved in potassium phosphate buffer (150 mM; pH 7.22). PEG was dissolved in water to give a 50% solution. Incubations were performed in potassium phosphate buffer (150 mM; pH 7.4) containing hepatic microsomes (200 µg of protein), DTT (5 mM), and T4 (6.4 µM) in a final volume of 1 ml. The samples were incubated for 1 h at 37°C in a shaking water bath. The reaction was terminated by the addition of 160 µl of PEG solution. The samples were centrifuged for 20 min at 14,000 rpm to remove precipitated proteins. The concentrations of T3 were determined in the supernatant by an immunoassay (Incstar, Stillwater, MN). To correct for the cross-reactivity of the assay with T4, a separate zero time control was determined for each T4 concentration. The assay conditions were linear with respect to time and the concentrations of microsomal protein, DTT, and T4.

Immunoelectrophoresis. Protein SDS-polyacrylamide gel electrophoresis was performed by the method of Laemmli (1970) with a 3% stacking gel and a 12% running gel. The proteins were transferred from the gel to polyvinylidene difluoride membranes in 3-cyclohexylamino-propanesulfonic acid buffer as described by Towbin et al. (1979). The membranes were treated with antibody followed by either goat anti-chicken IgY or anti-rabbit IgG conjugated to alkaline phosphatase as appropriate. The indicator dye was a combination of nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate (Bio-Rad, Richmond, CA). The membranes were laminated between cellulose sheets. The densities of the stained bands were determined on a flat bed scanner (UMAX, Taiwan). The scans were analyzed with NIH Image (version 1.54) on a Power PC (Apple Computer, Cupertino, CA). The blank for each band was determined from the adjacent, unreacted area in the lane. In control studies we found that the immunoassays were linear with respect to the concentrations of the respective proteins and that the band intensities were stable indefinitely after the membranes had been laminated.

Northern Blotting of mRNA Encoding the 27-kDa Protein. Fresh livers were rapidly homogenized with a Polytron homogenizer and the RNA was extracted according to the RNAgents total RNA isolation system protocol and stored at 80°C in RNA storage buffer. A 1% agarose gel was run in Northern running buffer according to the procedure of Ausubel et al. (1996). The RNA samples (8 µg) were loaded in a final volume of 20 µl of Northern sample buffer. Electrophoresis was performed at 170 V for 4 h. The RNAs from the gel were transferred overnight onto nylon membranes in NaCl-sodium citrate [3-0.3 M; 20× standard saline citrate (SSC)]. The membranes were washed with 2× SSC for 20 min and photographed by UV shadowing to document the presence of the RNA. The RNA was cross-linked to the membrane with a UV Stratalinker 1800 (Stratagene).

Other Assays. The microsomal protein concentrations were determined by the bicinchoninic acid assay (Pierce, Rockford, IL). Statistical significance was estimated by Student's t test. A two-tailed P value of less than .05 was taken as significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

In our previous studies, we had found that although there was no decrease in ID-I activity 4 to 6 h after the administration of methoxychlor (200 mg/kg/day), there was a 25% decline after 4 days of treatment (Zhou et al., 1995). Furthermore, the addition of methoxychlor (0-160 nM) to control incubations had no effect on the ID-I activity (data not shown).

In this study, we observed that after the administration of methoxychlor (200 mg/kg/day) for 14 days, the ID-I activity decreased 48% (controls = 2.5 ± 0.22 pmol of T3/mg of protein/min versus treated = 1.3 ± 0.12 pmol of T3/mg of protein/min; P < .05). But there was also a decrease in the average liver weight (controls = 15.7 ± 0.4 g versus treated = 13.0 ± 1.0 g), which was associated with a comparable decrease in body weight (controls = 266 ± 11 g versus treated = 217 ± 20 g). As a result, the total activity per liver showed an even greater decline (controls = 0.30 ± 0.01 nmol of T3/min/liver versus treated = 0.12 ± 0.02 nmol of T3/min/liver; P < .001). These decreases in both body and liver weights suggested that this high dosage of methoxychlor represented a significant stress to the animals.

We next examined the effect of daily treatment for 14 days with varying doses of methoxychlor (0-100 mg/kg/day). There were small decreases in body (controls = 254 ± 11 g versus methoxychlor 100 mg/kg/day = 225 ± 10 g) and liver weights (controls = 15.3 ± 0.4 g versus methoxychlor 100 mg/kg/day = 13.0 ± 0.5 g; P < .02) and the microsomal protein/g of liver (controls = 22.4 ± 0.4 mg/g of liver versus methoxychlor 100 mg/kg/day = 19.4 ± 0.4 mg/g of liver; P < .01). But there was a more marked decrease in the ID-I specific activity. This decrease was linear in log (dose) versus ID-I (r² = 0.99178; Fig. 1). At the highest dosage of methoxychlor used in this study (100 mg/kg/day), the ID-I specific activity decreased 42% (controls = 2.96 ± 0.13 pmol/min/mg of protein versus methoxychlor 100 mg/kg/day = 1.72 ± 0.11 pmol/min/mg of protein; P < .005; Table 1).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Effect in male rats of 14 days of various dosages of oral methoxychlor administration on ID-I activity per milligram of microsomal protein. The values are the mean ± S.E. of three animals. The assay procedure is described in the text.

The decrease in the ID-I activity was associated with a concomitant decrease in the content of the 27-kDa protein (Fig. 2). At dosages of methoxychlor of 11 to 100 mg/kg/day the specific content of ID-I decreased by 50 to 60% (Fig. 2; Table 1). At 100 mg/kg/day the liver showed a comparable decline in the ratio of the content of the mRNA for the 27-kDa protein to the 18S RNA (control = 2.2 ± 0.27 × 10³ versus treated = 0.99 ± 0.09 × 10³; P < .01; Table 1).

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effect in male rats of various dosages of 14 days of oral methoxychlor administration on content per milligram of microsomal protein of the 27-kDa thyroid-binding protein. The values are the mean ± S.E. of three animals. The assay procedure is described in the text.

The observed decreases in body and liver weights might suggest that, even at the lowest dosages of methoxychlor, the decrease in ID-I activity could be due to a drug-induced stress. To evaluate this possibility, we next examined the microsomal contents of two stress-responsive proteins, ERp55 (Fig. 3) and ERp57 (Fig. 4; Srivastava et al., 1991; Holtzman, 1998). These proteins were determined in the same microsomal preparations that had been used to determine the ID-I activities. Although ERp55 was unaffected by the administration of methoxychlor, ERp57 showed a significant decrease at the lowest dosage and then a significant increase at the higher dosages (Figs. 3 and 4). Because ERp55 was essentially unaffected by the treatment, these data would suggest that the changes in ID-I or its mRNA were not due to a drug-induced stress.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Effect in male rats of 14 days of various dosages of oral methoxychlor administration on content per milligram of microsomal protein of ERp55, the TPDO, form Q5. The values are the mean ± S.E. of three animals. The assay procedure is described in the text.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Effect in male rats of 14 days of various dosages of oral methoxychlor administration on content per milligram of microsomal protein of ERp57 the TPDO, form Q2. The values are the mean ± S.E. of three animals. The assay procedure is described in the text.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

This study clearly indicates that the administration of methoxychlor decreased the 27-kDa protein with a comparable decrease in the ID-I activity. These data are consistent with the widely accepted concept that ID-I activity in hepatic microsomes is catalyzed by this protein. Furthermore, this decrease in both the ID-I activity and the 27-kDa protein was proportional to the decrease in mRNA for the protein. This finding would suggest that the administration of methoxychlor causes a decrease in either the transcription or the stability of the mRNA encoding ID-I rather than a decrease in translation or a direct inhibition of the enzyme. The chemical similarity of this insecticide to T4 might suggest a possible mechanism for this decrease in transcription (Zhou et al., 1995).

The 27-kDa protein is a product of the dio 1 gene (Berry et al., 1993). Other studies have shown that T3 up-regulates mRNA transcription for this protein through a direct action on the gene, without requiring de novo protein synthesis (Menjo et al., 1993). Because ID-I catalyzes both the synthesis and degradation of T3 through inner ring deiodination, the increased transcription of ID-I by T3 could serve as a counter-regulatory mechanism to maintain constant T3 levels. (Berry and Larsen, 1993; Farwell and Braverman, 1996). As we previously observed, there is a great similarity between the conformations of methoxychlor and T4 so that both may bind to the thyroid hormone-binding proteins (Zhou et al., 1995). This similarity would suggest that the decrease in transcription of the 27-kDa protein may result from a direct competition of methoxychlor for the T3-binding sites on a nuclear receptor.

Finally, the question arises as to what led to the initial misidentification of the ERp55 variant as ID-I. One possibility is that the variant associates with the 27-kDa protein. The latter protein may be the catalytic subunit and the ERp55 variant could serve to maintain this subunit in the active state. This association seems likely because ERp55 serves such a role in other systems, such as prolyl 4-hydroxylase (Berg et al., 1979; Koivu and Myllyla, 1986; Holtzman, 1998). Hence, during the early stages of purification the two subunits may have copurified while the activity could then have been lost as the 56-kDa protein was brought to homogeneity and the 27-kDa subunit was separated from the complex. A similar problem has arisen with ERp57, which has been erroneously reported to have a variety of activities that resulted from contamination of supposedly purified preparations with other microsomal proteins (Srivastava et al., 1991, 1993; Chen et al., 1996; Holtzman, 1998). This persistent contamination is probably due to the tendency of these proteins, which are also chaperons, to tenaciously bind to other proteins.

In conclusion, our data would suggest that another possible adverse environmental effect of methoxychlor may be to interfere with thyroid hormone metabolism. Because this insecticide is widely used in many areas of the world, this inhibition could have significant deleterious effects on both human and animal populations. In support of this concept, preliminary data from our laboratory have suggested that it may interfere with tadpole metamorphosis at concentrations that have been found in the environment (Morrell et al., 1998).