- Email: [email protected]
Slow biliary elimination of methyl mercury in the marine elasmobranchs, Raja erinacea and Squalus acanthias
TOXICOLOGY
AND
APPLIED
PHARMACOLOGY
S&407-4 15 ( 1986)
Slow Biliary Elimination of Methyl Mercury in the Marine Elasmobranchs, Raja erinacea and Squalus acanthias NAZZARENO
BALLATORI
AND JAMES L. BOYER
Department @‘Medicine and Liver Center, Yale University School of Medicine, New Haven, Connecticut 06510. and Mount Desert Island Biological Laboratory, Salsbury Cove, Maine 04672
Received January 2I, 1986; accepted May 7. I986 Slow Biliary Elimination of Methyl Mercury in the Marine Elasmobranchs, Raja erinacea and Squalus acanthias. BALLATORI, N.. AND BOYER, J. L. (1986). Toxicol. Appl. Pharmacol. 85,407-415. The present study examined the ability of two marine elasmobranchs (Raja erinatea, little skate, and Squahrs acanthias. spiny dogfish shark) to excrete methyl mercury into bile, a major excretory route in mammals. 203Hg-labeled methyl mercury chloride was administered via the caudal vein, and bile collected through exteriorized cannulas in the free swimming fish. Skates and dog&h sharks excreted only a small fraction ofthe *“Hg into bile over a 3-day period: in the skate, the 3-day cumulative excretion (as a % of dose) was 0.44 + 0.10 (n = 4. &SD), 0.7 1 t 0.23 (n = 6), and 1.00 f 0.34 (n = 4) for doses of 1, 5, and 20 Fmol/kg, respectively, while the shark excreted only 0.15 f 0.15% (n = 8) at a dose of 5 pmol/kg. As in mammals. the availability of hepatic and biliary glutathione was a determinant of the biliary excretion of methyl mercury in these species: the administration of sulfobromophthalein. a compound known to inhibit both glutathione and methyl mercury excretion in rats, or of L-buthionine-S,R-sulfoximine, an inhibitor of glutathione biosynthesis, decreased the biliary excretion of both glutathione and mercury in the skate. The slow hepatic excretory process for methyl mercury in the skate and shark was attributed to an inordinately slow rate of bile formation: from 1 to 4 ml/kg.day. An inefficient biliary excretory process in fish may accout in part for the long biological half-times for methyl mercury in marine species. &I 1986 Academic Press. Inc.
Methyl mercury is a major toxic contaminant of the marine biosphere produced by the bacterial methylation of inorganic forms of mercury (Jensen and Jernelov, 1969: Jernelov, 1972; Wood, 1972; Clarkson, 1972). In contrast to inorganic mercury, methyl mercury is avidly accumulated by multicellular aquatic organisms and is substantially concentrated as one moves up the aquatic food chains (Jernelov and Lann, 197 1; Jernelov, 1972; Friberg and Vostal, 1972; Krenkel, 1973). The bioaccumulation factor (methyl mercury in fish/methyl mercury in water) for larger predatory fish is on the order of 1O6 to 1O* (Jernelov and Lann, 197 1; Fri-
berg and Vostal, 1972; Krenkel, 1973). This concentration occurs because methyl mercury is less toxic to simple aquatic organisms with poorly developed central nervous systems, and thus, it can accumulate in aquatic organisms without disturbing the food chains (Jernelov, 1972; Clarkson, 1972; Friberg and Vostal, 1972; Krenkel, 1973). In addition, the biological half-time of methyl mercury in fish is on the order of months to years (Jernelov and Lann, 197 1; Jarvenpaa et al., 1970; Burrows and Krenkel, 1973; Lockhart et al., 1972; Giblin and Massaro, 1973; McKim et al., 1974), and thus, once ingested, it remains essentially for the life of marine animals. Fish
407
004 1-008X/86 $3.00 Copyright 0 1986 by Academic Press, Inc. All n&s of reproduction in any form reserved.
408
BALLATORI
are the primary, if not the sole, source of mercury for most human populations (World Health Organization, 1976) as dramatically demonstrated by the Minamata Bay, Japan, disaster (Kurland et al., 1960; Kutsuna, 1968). The basis for the long biological half-time for methyl mercury in fish is poorly understood. The chemical affinity of methyl mercury for the proteins in fish muscle may be unusually high (Jarvenpaa et al., 1970; Burrows and Krenkel, 1973; Lockhart et al., 1972; Giblin and Massaro, 1973). Also, fish exhibit very slow whole-body elimination rates (Jarvenpaa et al., 1970; Burrows and Krenkel, 1973; Lockhart et al., 1972; Giblin and Massaro, 1973; McKim et al., 1974), suggesting that their total excretory capacity (fecal, urinary, plus secretion across the gills and skin) is minimal. Although the long half-time for methyl mercury in fish precludes the direct determination of its pathways for excretion, indirect evidence obtained in rainbow trout (Giblin and Massaro, 1973) suggests that the main route of methyl mercury’s excretion in trout, as in mammals, is through the feces. Since biliary excretion is the primary pathway contributing to fecal elimination in mammals, a low rate of excretion of mercury into fish bile might account in part for the prolonged biological half-times in these species. Indeed, only small amounts of mercury have been found in gallbladder bile of methyl mercury-treated fish (Giblin and Massaro, 1973; Backstrom, 1969; Guarino et
AND
BOYER
METHODS Male spiny dogfish sharks, 1.96 & 0.16 kg (n = 8). caught by trawl in Frenchman’s Bay. Maine, and male skates, 1 .OO + 0.09 kg (n = 30) netted in Southwest Harbor, Maine, were placed in large tanks equipped with flowing seawater at the Mount Desert Island Biological Laboratory in Salsbury Cove, Maine. Surgical procedures for hepatic bile collections in the free swimming fish were performed within 2 days of capture. as previously described (Boyer et ul., 1976a-c). After the operative procedure. the fish were allowed to swim freely in a large tank that was continuously supplied with 15°C seawater pumped directly from Frenchman’s Bay. As elasmobranchs will not accept food while in captivity, no exogenous food was offered for the remainder of the experiment. After a control bile collection period of 24 hr, 203Hg-labeled methyl mercury chloride was injected into the caudal vein in doses of either 1, 5, or 20 pmol/ kg body wt, in a solution of elasmobranch Ringer’s containing 5 mM Na2C02. Bile was collected in 24-hr intervals through exteriorized cannulas attached to small rubber balloons for an additional 3 days after methyl mercury administration. Bile was collected by briefly restraining the fish and removing the balloons, and its volume was determined gravimetrically, assuming a density of 1 .O g/ml. To ascertain whether any of the biliary excreted ‘03Hg could be lost from the bile collection balloons during the 24-hr collection interval, control skate and shark bile were mixed in vitro with CHx203HgCl, placed in sealed rubber balloons. and the balloons suspended in a continuously stirred 0.5~liter bottle of seawater. No loss of radioactivity from the balloons into the seawater could be detected over a 24-hr interval. In determining the biliary excretion of ‘03Hg, the balloons and their entire contents of bile were counted in a gamma scintillation spectrometer. The 203Hg content of selected tissues was also determined by gamma scintillation counting. The concentration of total ghttathione (reduced. GSH + disulfide, GSSG) in bile and liver tissue was determined as described (Tietze. 1969: Griffith, 1980).
al., 1972).
In the present studies we quantified the extent of methyl mercury excretion into hepatic bile in two marine elasmobranchs, and determined several factors that regulate methyl mercury excretion into bile in these species. The spiny dogfish and the little skate were chosen as the experimental animals since techniques for extended bile collections have been developed in these species (Boyer et al., 1976a-c).
RESULTS Free swimming cannulated small skates and dogfish sharks excreted a minute fraction of the administered radiolabeled mercury into bile over a 3-day period (Fig. 1). In the skate, biliary excretion of mercury was dose dependent: the 3-day cumulative excretion expressed as a percentage of the administered
METHYL SKATE
MERCURY
IN ELASMOBRANCH
DOGFISH
DAYS
FIG. 1. Bile flow. mercury excretion into bile, and glutathione excretion into bile in the little skate (Ruju erinatea) and in the spiny dogfish shark (Squalus acanthias). Bile was collected into rubber balloons in free swimming cannulated skates and sharks for a total of 4 days. After a control bile collection period of 24 hr, 203Hg-labeled methyl mercury chloride was injected into the caudal vein in doses of I (n = 4), 5 (n = 6), and 20 (n = 4) Fmol/ kg body wt in the skate. and at a dose of 5 pmoI/kg in the shark (n = 8). The CH3 203HgC1was dissolved in elasmobranch Ringer’s containing 5 mM NaZC03. Values are means + standard deviations.
dose was 0.44 + 0.10, 0.71 + 0.23, and 1.00 ? 0.34 for doses of 1,5, and 20 gmol/kg body wt, respectively. Dogfish sharks excreted an even smaller percentage of the administered mercury into bile: the 3-day cumulative biliary excretion was 0.15 ? 0.15% at a dose of 5 pmol/kg body wt (Fig. 1). By comparison, guinea pigs or rabbits given the same dose of mercury can excrete 0.15% of the dose into bile in only 5 hr, while the rat can excrete greater than 3% in 5 hr (Vostal and Clarkson, 1973). Although the overall rate of excretion into bile was low in the skate and shark, the
BILE
409
concentration of radiolabeled mercury in bile (skate = 4- 10 PM, shark = l-2 PM at a dose of 5 pmol/kg) was similar to that found in mammalian bile after an equimolar dose of methyl mercury. Thus, the slow rate of biliary methyl mercury elimination in skate and shark was attributed to their unusually low rate of bile formation (Fig. 1). For the skate, bile flow rates averaged 3.1 + 1.4 ml/kg.day on the first day after cannulation and decreased to a value of 1.8 it 0.6 ml/kg.day on the fourth day. For the dogfish shark, bile flow was 2.14 -+ 0.7 1 and 0.83 + 0.33 ml/kg. day on the first and fourth day after cannulation, respectively. This low rate of bile formation has been described previously in both the free swimming fish and in the isolated perfused skate liver, and is at least one to two orders of magnitude lower than that observed in most mammalian species, including man (Boyer et al., 1976a-c; Reed et al., 1982a,b). That the low biliary excretion of mercury in the skate and shark is not related to an impaired hepatic uptake for methyl mercury is shown in Table 1. Greater than 10% of the administered ‘03Hg was found in the liver of both skates and sharks at 3 days after its administration. Table 1 also lists the distribution of lo3Hg in a number of other organs and tissues. Skeletal muscle. liver. kidney, and whole blood accounted for the major fraction of the administered dose of mercury. Of the tissues analyzed for ‘03Hg, the highest concentrations of mercury (Io3Hg/g tissue) were found in the heart, rectal gland, kidney, and liver of both the shark and the skate. Since the ‘03Hg contents of the head, fins, gills, skin, intestines, intestinal contents, and bone were not determined, and since only representative samples of certain tissues were analyzed for their 203Hg content, a complete quantitative account of the overall distribution of the administered dose of methyl mercury cannot be performed. The reason for the higher normalized biliary excretion rates observed at higher doses of methyl mercury in the skate (Fig. 1) is not
BALLATORI
AND
BOYER
known. The tissue distribution of *03Hg was also dependent on the dose (Table 1). In particular, the percentage of administered ‘03Hg in the kidney and rectal gland was over twofold greater at the lowest as compared to the higher doses. Although these observations are consistent with saturation of a high-affinity/ low-capacity compartment for methyl mercury in the skate, no direct evidence is presently available for the existence of such a compartment. The ability of the skate to excrete radiolabeled mercury into bile at a higher concentration and rate than the dogfish shark may be related to differences in the biliary excretion of glutathione (Fig. 1). as previously shown for the rat (Refsvik and Norseth, 1975: Ballatori and Clarkson, 1982, 1983, 1985b). Hepatic levels of glutathione in the skate and shark were 0.62 + 0.43 (n = 14) and 0.36 + 0.12 (n = 8) ymol glutathione equivalents/ g, respectively. However, glutathione was detectable only in skate bile; the concentration of glutathione in bile was 0.69 -t 0.50 and 0.76 f 0.83 /*mol glutathione equivalents/ml on the first and fourth days of bile collection, respectively. Most (>80%) of the glutathione in excreted skate bile was in the disulfide form, GSSG. However, because of the long bile collection interval, 24 hr, and because of the lack of an antioxidant in the bile collection balloons, it was impossible to distinguish whether the skate actually secreted GSSG into bile or whether it was the product of the postsecretory oxidation of biliary GSH. Gel filtration of bile on Sephadex G-75 showed that a major fraction of the 203Hg in freshly collected skate bile eluted in the low molecular weight region (less than approximately 10,000 Da; V,/ V,, > 2) (Fig. 2). In contrast. most of the mercury in freshly collected shark bile eluted in the void volume. When the individual fractions from the Sephadex G-75 column were analyzed for their glutathione content, glutathione was noted at a VJ V, of 2.8 to 3.5 for skate bile. However, glutathione was not detectable in the G-75 fractions from shark bile. Most of the mercury
METHYL
MERCURY
IN ELASMOBRANCH DOGFISH
SKATE I
I
I
I5
,
I
I
411
BILE I
I
BILE
FIG. 2. Distribution of “‘Hg in Sephadex G-75 fractions of bile and liver cytosol from the little skate (Rajn erinacea) and spiny dogfish shark (Squahsacanthias). Bile collected during the frrst 24-hr interval after the administration of 5 pmol of CH 3203HgCl/kg was applied directly to the 0.9 x 60-cm column, in a volume of0.3 to 0.4 ml. Liver cytosol. obtained by homogenizing liver tissue in 3 to 5 vol ofelasmobranch Ringer’s and centrifuging at 105,OOOg for 1 hr. was applied to the column in a volume of 0.3 ml. Gel filtration was carried out at room temperature (16-23°C) at a flow rate of 23 ml/hr. One-milliliter fractions were collected. The eluant was 0.1 M Tris-HCl buffer, pH 7.4, containing 1 mM EDTA and 0.02% sodium azide. Radioactivity in each fraction was determined by gamma scintillation spectrometry. The void volume was defined by the elution of blue Dextran. and the elution volume of glutathione was determined by analyzing the individual fractions for their glutathione content. Glutathione itself, as well as the *“Hg from a mixture of CH3203HgCl and glutathione, eluted from the G-75 column at a VJV, of 2.8 to 3.5. Although glutathione was noted at a VJVa of 2.8 to 3.5 for the liver cytosols of both skate and shark, and in bile from the skate, no glutathione was found in the Sephadex G-75 fractions of shark bile.
in liver cytosol from both sharks and skates eluted in the void volume (Fig. 2). Similar ‘03Hg elution profiles to those shown in Fig. 2 were obtained when bile collected on Days 2 and 3 after methyl mercury administration were applied to the Sephadex G-75 columns as well as when freshly collected shark or skate bile and liver cytosols were supplemented in vitro with CH3’03HgCl. In addition, sulfobromophthalein (BSP), a compound known to inhibit both glutathione and methyl mercury excretion in rats (Ballatori and Clarkson, 1985a), and L-buthionineS,R-sulfoximine (L-BSO), an inhibitor of glutathione biosynthesis (Griffith and Meister, 1979), were able to inhibit glutathione and mercury excretion in the skate (Table 2). Al-
though BSP is selectively cleared by the skate liver, it is excreted into bile much more slowly than in mammals (Boyer et al., 1976a-c), presumably accounting for the prolonged inhibition of glutathione and methyl mercury excretion into skate bile (Table 2). In addition to inhibiting glutathione efflux into bile, BSP administration was also associated with a markedly increased hepatic tissue concentration of glutathione (Table 2). Since the increase in hepatic glutathione levels is much larger than can be accounted for by diminished efflux into bile, BSP must have also inhibited either the intracellular utilization or efflux of glutathione from the liver into blood plasma. L-Buthionine-S,R-sulfoximine, on the other hand, inhibited hepatic
BALLATORI
AND
BOYER
glutathione synthesis (Table 2). Since reductions in the biliary excretion of glutathione and ‘03Hg were not evident until the second day after L-BSO administration, the rate of hepatic glutathione turnover in this species might also be relatively slow. Additional experiments are needed to establish this possibility. DISCUSSION The selective bioconcentration of methyl mercury by aquatic organisms has serious environmental as well as human health consequences (Jernelov, 1972; Krenkel, 1973: Friberg and Vostal, 1972; Kurland et a/., 1960; Kutsuna, 1968). The reasons for the long biological half-time of methyl mercury in fish are not known, but may be related to (1) inefficient excretory mechanism (biliaryfecal, urinary, or secretion across the gills), (2) an unusually high affinity of methyl mercury for the proteins in fish muscle, or (3) a combination of these factors. Because of the long biological half-time of methyl mercury in fish, it has been extremely difficult to assess the relative contributions of the various excretory routes in the whole-body elimination of the metal by fish. Studies by Giblin and Massaro ( 1973) showed that the biological half-time for methyl mercury in rainbow trout is greater than 200 days, and that the main route of its excretion in trout, as in mammals, is through the feces. These investigators also noted that only a small amount of mercury appears in gallbladder bile of methyl mercury-treated fish, suggesting that the slow fecal elimination is the result of slow biliary excretion of methyl mercury. However, since the concentration of methyl mercury in gallbladder bile is a poor index of the rate, or extent? of mercury excretion into bile, the overall contribution ofthis excretory mechanism in the elimination of methyl mercury by trout could not be assessed.
METHYL
MERCURY
IN ELASMOBRANCH
In the present study we quantified the biliary excretion of methyl mercury in two elasmobranch species by directly collecting hepatic bile in the free swimming fish. Our results indicate that both skate and dogfish shark excrete mercury into bile slowly, when compared to the rate of biliary excretion by other animals. The inefficient hepatic excretory process was primarily related to the inordinately slow production of bile by these fish (Fig. 1 and Boyer et al., 1976a-c; Reed et al., 1982a,b). Since biliary excretion is a major determinant of fecal methyl mercury elimination, a slow biliary excretory process would retard whole-body elimination of the metal, and favor its accumulation in fish muscle. In a preliminary study, Guarino et al. ( 1972) found that urinary excretion was also a minor pathway of methyl mercury elimination in the dogfish shark. During the first 4 hr after the intravenous administration of 0.5 pmol [‘4C]CH3HgCl/kg, 0.10% of the administered dose was excreted in urine, with an additional 0.19% excreted in the subsequent 20 hr. The amount of radiolabel in the kidney was 5.9 and 2.9% of the administered dose at 4 and 24 hr, respectively (Guarino et al., 1972). Since more mercury is delivered to the kidney, and therefore presumably into urine, after its intravenous administration, when compared to its oral or intramuscular administration in fish (Backstrom, 1969), the values given above probably represent maximal rates of urinary excretion by the dogfish. Thus, as with the biliary route, urinary excretion appears to be inefficient in elasmobranchs. These slow excretory processes may explain the avid retention of methyl mercury by certain fish species. Whether fish muscle also possess an unusually high chemical affinity for methyl mercury, as has been suggested (Burrows and Krenkel, 1973: Giblin and Massaro, 1973; Jarvenpaa et al., 1970: Lockhart et al., 1972), remains to be determined. In the present studies, sequestration of methyl mercury by fish muscle did not ap-
BILE
413
pear to be the rate-limiting factor in the elimination of this metal by elasmobranchs. Although both the skate and shark excreted mercury into bile slowly, there were significant differences in the hepatic handling of methyl mercury between these elasmobranchs that were partly attributed to differences in biliary glutathione excretion (Fig. I), as previously shown for the rat (Refsvik and Norseth, 1975; Ballatori and Clarkson, 1982, 1983, 1985b). The skate. which had a higher hepatic tissue concentration of glutathione than the shark, and excreted a substantial amount of glutathione into bile (Fig. l), also excreted radiolabeled mercury into bile at a higher concentration and rate than the dogfish shark. On gel filtration, most of the ‘03Hg in skate bile eluted in the low molecular weight region, where glutathione would elute, while most of the mercury in shark bile which had undetectable amounts of glutathione, eluted near the void volume (Fig. 2). In addition, treatment of the skate with agents known to decrease biliary glutathione levels in rats (BSP; Ballatori and Clarkson, 1983, 1985a; L-BSO, Griffith and Meister, 1979) produced parallel changes in the biliary excretion of mercury and of glutathione (Table 2). Taken together, these observations indicate that in addition to the rate of bile formation, the rate of glutathione excretion is a key determinant ofthe biliary transport of methyl mercury in elasmobranchs. The reason for the presence of these relatively high concentrations of glutathione in skate bile, or for its absence in shark bile, is not known. That L-BSO and BSP elicited similar changes in biliary glutathione in the skate as they do in mammals suggests the presence of similar glutathione biosynthetic and transport processes in elasmobranchs. Of interest, the organic anion BSP not only inhibited glutathione efflux into skate bile, but it markedly increased hepatic tissue concentrations of the tripeptide (Table 2). The increase in hepatic glutathione is much larger than can be explained on the basis of dimin-
414
BALLATORI
ished efflux into bile. In the rat, BSP is believed to decrease glutathione secretion into bile by inhibiting the canalicular membrane transport system (Ballatori and Clarkson, 1983, 1985a). However, most (80%) of the glutathione released by the rat liver goes into plasma across a sinusoidal membrane glutathione transport system that is similar, if not identical, to that on the canalicular membrane (Inoue et al., 1984). It has not been established whether BSP also inhibits the sinusoidal elllux of glutathione in rat liver. The present findings in the skate, however, indicate that BSP may be inhibiting the efflux of glutathione across both the canalicular and sinusoidal plasma membrane, to effect a net increase in hepatic glutathione levels. The prolonged inhibition of glutathione efflux in the skate is most likely due to the continual presence of high concentrations of BSP in the skate liver, which are in turn maintained by the slow clearance (mainly in the unconjugated form; Boyer et al., 1976a-c) of the dye into bile. ACKNOWLEDGMENTS We thank Annie Boyer and Christin Ciresi for excellent technical assistance. Supported by grants AM3 1493. AM25636 and AM34989 from the U.S. Public Health Service. Support for N.B. was provided by a postdoctora1 fellowship from the U.S. Public Health Service, AM07497, and a Research Fellowship Award from the American Liver Foundation.
REFERENCES BACKSTROM, J. (1969). Distribution studies of mercuric pesticides in quail and some fresh-water fishes. Acta Pharmacd Toxicol. 27(Suppl. 3), l-103. BALLATORI, N., AND CLARKSON, T. W. (I 982). Developmental changes in the biliary excretion of methylmercury and ghttathione. Science 216,6 l-63. BALLATORI. N., AND CLARKSON, T. W. (I 983). Biliary transport of glutathione and methylmercury. Amer. J. Physiol. 244, G435-G44 1. BALLATORI. N., AND CLARKSON, T. W. (1985a). Sulfobromophthalein inhibition of glutathione and meth-
AND BOYER ylmercury secretion into bile. Amer. J. Physiol. 248, G238-G245. BALLATORI, N., AND CLARKSON. T. W. (1985b). Biliary secretion of glutathione and of glutathione-metal complexes. Fundam. Appl. Toxicol. 5,8 16-83 1. BOYER, J. L.. SCHWARZ, J., AND SMITH, N. (1976a). Biliary excretion in elasmobranchs. I. Bile collection and composition. Amer. J. Physiol. 230,970-973. BOYER, J. L., SCHWARZ,J., AND SMITH, N. (1976b). Biliary excretion in elasmobranchs. II. Hepatic uptake and biliary excretion of organic anions. Amer. J. Physiol. 230,974-98 1. BOYER, J. L., SCHWARZ, J.. AND SMITH, N. (1976~). Selective hepatic uptake and biliary excretion of %-sulfobromophthalein in marine elasmobranchs. Ga.struenterology70,254-256. BURGER, J. W. ( 1967). Some parameters for the dogfish, Squalus acanthias. Mt. Desert Island Biol. Lab. Bull. 7,5-9.
BURROWS, W. D., AND KRENKEL, P. R. (1973). Studies on uptake and loss of methylmercuryby bluegills. Environ. Sri. Technol. 7, I 127- 1130. CLARKSON, T. W. (I 972). Recent advances in the toxicology of mercury with emphasis on the alkylmercurials. CRC Crit. Rev. Toxicol. 2, 203-234. FRIBERG, L., AND VOSTAL. J. (1972). Mercury in the Environment. Chemical Rubber Co. Press, Cleveland, Ohio. GIBLIN, F. J., AND MASSARO, E. J. (1973). Pharmacodynamics of methylmercury in the rainbow trout (Sulmo gairdneri): Tissue uptake, distribution, and excretion. To-&l. Appl. Pharmacol. 24, 8 l-9 1. GRIF’F~TH,0. W. (1980). Determination of glutathione and glutathione disultide using glutathione reductase and 2-vinylpyridine. Anal. Biochem. 106,207-2 12. GRIFRTH, 0. W.. AND MEISTER, A. (1979). Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocystine sulfoximine). J. Biol. Chem. 254,7558-7560. GUARINO, A. M., BRILEY. P. M., ANDERSON, J. B.. KINTER. M. A., SCHNEIDERMAN. S., KLIPP, L. D.. AND ADAMSON, R. H. (1972). Renal and hepatic excretion of foreign compounds by Squalus acanthias. Mt. Desert Island Biol. Lab. Bull. 12,4 l-43. INOUE, M.. KINNE, R.. TRAN. T., AND ARIAS, 1. M. (1984). Glutathione transport across hepatocyte plasma membranes: Analysis using isolated rat-liver sinusoidal-membrane vesicles. Eur. J. Biochem. 138, 491-495.
JARVENPAA, T., TILLANDER, M., AND MIETTINEN, J. K. (1970). Methylmercury: Half-time of elimination in flounder, pike and eel. Suom. Kemistii. B43,439-442. JENSEN, S.. AND JERNELOV. A. (1969). Biological methylation of mercury in aquatic organisms. Nature (London) 223,753-754.
METHYL
MERCURY
IN ELASMOBRANCH
JERNELOV, A. (1972). Mercury and food chains. In Environmental Mercury Contamination (R. Hartung and B. 0. Dinman, eds.), pp. 174-I 77. Ann Arbor Sci. Publ., Ann Arbor, Mich. JERNELOV, A.. AND LANN, H. (197 1). Mercury accumulation in food chains. Oikos 22,403-406. KRENKEL, P. A. (1973). Mercury Environmental Considerations, Part 1, CRC Critical Reviews in Environmental Control, pp. 303-373. CRC Press, Cleveland, Ohio. KURLAND, L. T., FARO, S. W., AND SIEDLER. H. (1960). Minamata disease: The outbreak ofa neurologic disorder in Minamata, Japan and its relationship to the ingestion of seafood contaminated by mercuric compounds. A’orld Neural. 1,370-39 1. KUTSUNA, S. (1968). Minamata Disease. Study Group of Minamata Disease, Kumamoto University, Japan. LOCKHART, W. L.. UTHE, J. F., KENNEY, A. R., AND MEHRLE, P. M. (1972). Methylmercury in Northern Pike (Esox lucius): Distribution. elimination. and some biochemical characteristics of contaminated fish. J. Fish. Res. BoardCanad. 29, 1519-1523. MCKIM, J. M., CHRISTENSEN, G. M.. TUCKER, J. H., BENOIT, D. A., AND LEWIS. M. J. (1974). Effects of pollution on freshwater fish. J. U’ater Polk. Control FcJd.46, 1562- 1568.
BILE
415
REED, J. S., SMITH, N. D., AND BOYER, J. L. (1982a). Determinants of biliary secretion in isolated perfused skate liver. Amer. J. Phvsiol. 242, G3 19-G325. REED, J. S., SMITH, N. D., AND BOYER, J. L. (198213). Hemodynamic effects on oxygen consumption and bile flow in isolated skate liver. Amer. J. PhJaiol. 242, G313-G318. REFSVIK. T., AND NORSETH, T. (1975). Methyl mercuric compounds in rat bile. ‘4cta Pharmacol. To,yicol. 36, 67-78.
THORSON, T. ( 1958). Measurement of the fluid compartments of four species of marine Chondrichthyes. Ph.vsiol. Zool. 31, 16-23. TIETZE, F. (1969). Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: Applications to mammalian blood and other tissues. Anal. Biochem. 27,502-522. VOSTAL. J. J., AND CLARKSON, T. W. ( 1973). Mercury as an environmental hazard. J. Occup. Med. 15,649656.
WOOD, J. M. ( 1972). A progress report on mercury. Environment 14,33-39. World Health Organization (1976). Environmental Health Criteria. 1, Mercury. World Health Organization. Geneva.
AND
APPLIED
PHARMACOLOGY
S&407-4 15 ( 1986)
Slow Biliary Elimination of Methyl Mercury in the Marine Elasmobranchs, Raja erinacea and Squalus acanthias NAZZARENO
BALLATORI
AND JAMES L. BOYER
Department @‘Medicine and Liver Center, Yale University School of Medicine, New Haven, Connecticut 06510. and Mount Desert Island Biological Laboratory, Salsbury Cove, Maine 04672
Received January 2I, 1986; accepted May 7. I986 Slow Biliary Elimination of Methyl Mercury in the Marine Elasmobranchs, Raja erinacea and Squalus acanthias. BALLATORI, N.. AND BOYER, J. L. (1986). Toxicol. Appl. Pharmacol. 85,407-415. The present study examined the ability of two marine elasmobranchs (Raja erinatea, little skate, and Squahrs acanthias. spiny dogfish shark) to excrete methyl mercury into bile, a major excretory route in mammals. 203Hg-labeled methyl mercury chloride was administered via the caudal vein, and bile collected through exteriorized cannulas in the free swimming fish. Skates and dog&h sharks excreted only a small fraction ofthe *“Hg into bile over a 3-day period: in the skate, the 3-day cumulative excretion (as a % of dose) was 0.44 + 0.10 (n = 4. &SD), 0.7 1 t 0.23 (n = 6), and 1.00 f 0.34 (n = 4) for doses of 1, 5, and 20 Fmol/kg, respectively, while the shark excreted only 0.15 f 0.15% (n = 8) at a dose of 5 pmol/kg. As in mammals. the availability of hepatic and biliary glutathione was a determinant of the biliary excretion of methyl mercury in these species: the administration of sulfobromophthalein. a compound known to inhibit both glutathione and methyl mercury excretion in rats, or of L-buthionine-S,R-sulfoximine, an inhibitor of glutathione biosynthesis, decreased the biliary excretion of both glutathione and mercury in the skate. The slow hepatic excretory process for methyl mercury in the skate and shark was attributed to an inordinately slow rate of bile formation: from 1 to 4 ml/kg.day. An inefficient biliary excretory process in fish may accout in part for the long biological half-times for methyl mercury in marine species. &I 1986 Academic Press. Inc.
Methyl mercury is a major toxic contaminant of the marine biosphere produced by the bacterial methylation of inorganic forms of mercury (Jensen and Jernelov, 1969: Jernelov, 1972; Wood, 1972; Clarkson, 1972). In contrast to inorganic mercury, methyl mercury is avidly accumulated by multicellular aquatic organisms and is substantially concentrated as one moves up the aquatic food chains (Jernelov and Lann, 197 1; Jernelov, 1972; Friberg and Vostal, 1972; Krenkel, 1973). The bioaccumulation factor (methyl mercury in fish/methyl mercury in water) for larger predatory fish is on the order of 1O6 to 1O* (Jernelov and Lann, 197 1; Fri-
berg and Vostal, 1972; Krenkel, 1973). This concentration occurs because methyl mercury is less toxic to simple aquatic organisms with poorly developed central nervous systems, and thus, it can accumulate in aquatic organisms without disturbing the food chains (Jernelov, 1972; Clarkson, 1972; Friberg and Vostal, 1972; Krenkel, 1973). In addition, the biological half-time of methyl mercury in fish is on the order of months to years (Jernelov and Lann, 197 1; Jarvenpaa et al., 1970; Burrows and Krenkel, 1973; Lockhart et al., 1972; Giblin and Massaro, 1973; McKim et al., 1974), and thus, once ingested, it remains essentially for the life of marine animals. Fish
407
004 1-008X/86 $3.00 Copyright 0 1986 by Academic Press, Inc. All n&s of reproduction in any form reserved.
408
BALLATORI
are the primary, if not the sole, source of mercury for most human populations (World Health Organization, 1976) as dramatically demonstrated by the Minamata Bay, Japan, disaster (Kurland et al., 1960; Kutsuna, 1968). The basis for the long biological half-time for methyl mercury in fish is poorly understood. The chemical affinity of methyl mercury for the proteins in fish muscle may be unusually high (Jarvenpaa et al., 1970; Burrows and Krenkel, 1973; Lockhart et al., 1972; Giblin and Massaro, 1973). Also, fish exhibit very slow whole-body elimination rates (Jarvenpaa et al., 1970; Burrows and Krenkel, 1973; Lockhart et al., 1972; Giblin and Massaro, 1973; McKim et al., 1974), suggesting that their total excretory capacity (fecal, urinary, plus secretion across the gills and skin) is minimal. Although the long half-time for methyl mercury in fish precludes the direct determination of its pathways for excretion, indirect evidence obtained in rainbow trout (Giblin and Massaro, 1973) suggests that the main route of methyl mercury’s excretion in trout, as in mammals, is through the feces. Since biliary excretion is the primary pathway contributing to fecal elimination in mammals, a low rate of excretion of mercury into fish bile might account in part for the prolonged biological half-times in these species. Indeed, only small amounts of mercury have been found in gallbladder bile of methyl mercury-treated fish (Giblin and Massaro, 1973; Backstrom, 1969; Guarino et
AND
BOYER
METHODS Male spiny dogfish sharks, 1.96 & 0.16 kg (n = 8). caught by trawl in Frenchman’s Bay. Maine, and male skates, 1 .OO + 0.09 kg (n = 30) netted in Southwest Harbor, Maine, were placed in large tanks equipped with flowing seawater at the Mount Desert Island Biological Laboratory in Salsbury Cove, Maine. Surgical procedures for hepatic bile collections in the free swimming fish were performed within 2 days of capture. as previously described (Boyer et ul., 1976a-c). After the operative procedure. the fish were allowed to swim freely in a large tank that was continuously supplied with 15°C seawater pumped directly from Frenchman’s Bay. As elasmobranchs will not accept food while in captivity, no exogenous food was offered for the remainder of the experiment. After a control bile collection period of 24 hr, 203Hg-labeled methyl mercury chloride was injected into the caudal vein in doses of either 1, 5, or 20 pmol/ kg body wt, in a solution of elasmobranch Ringer’s containing 5 mM Na2C02. Bile was collected in 24-hr intervals through exteriorized cannulas attached to small rubber balloons for an additional 3 days after methyl mercury administration. Bile was collected by briefly restraining the fish and removing the balloons, and its volume was determined gravimetrically, assuming a density of 1 .O g/ml. To ascertain whether any of the biliary excreted ‘03Hg could be lost from the bile collection balloons during the 24-hr collection interval, control skate and shark bile were mixed in vitro with CHx203HgCl, placed in sealed rubber balloons. and the balloons suspended in a continuously stirred 0.5~liter bottle of seawater. No loss of radioactivity from the balloons into the seawater could be detected over a 24-hr interval. In determining the biliary excretion of ‘03Hg, the balloons and their entire contents of bile were counted in a gamma scintillation spectrometer. The 203Hg content of selected tissues was also determined by gamma scintillation counting. The concentration of total ghttathione (reduced. GSH + disulfide, GSSG) in bile and liver tissue was determined as described (Tietze. 1969: Griffith, 1980).
al., 1972).
In the present studies we quantified the extent of methyl mercury excretion into hepatic bile in two marine elasmobranchs, and determined several factors that regulate methyl mercury excretion into bile in these species. The spiny dogfish and the little skate were chosen as the experimental animals since techniques for extended bile collections have been developed in these species (Boyer et al., 1976a-c).
RESULTS Free swimming cannulated small skates and dogfish sharks excreted a minute fraction of the administered radiolabeled mercury into bile over a 3-day period (Fig. 1). In the skate, biliary excretion of mercury was dose dependent: the 3-day cumulative excretion expressed as a percentage of the administered
METHYL SKATE
MERCURY
IN ELASMOBRANCH
DOGFISH
DAYS
FIG. 1. Bile flow. mercury excretion into bile, and glutathione excretion into bile in the little skate (Ruju erinatea) and in the spiny dogfish shark (Squalus acanthias). Bile was collected into rubber balloons in free swimming cannulated skates and sharks for a total of 4 days. After a control bile collection period of 24 hr, 203Hg-labeled methyl mercury chloride was injected into the caudal vein in doses of I (n = 4), 5 (n = 6), and 20 (n = 4) Fmol/ kg body wt in the skate. and at a dose of 5 pmoI/kg in the shark (n = 8). The CH3 203HgC1was dissolved in elasmobranch Ringer’s containing 5 mM NaZC03. Values are means + standard deviations.
dose was 0.44 + 0.10, 0.71 + 0.23, and 1.00 ? 0.34 for doses of 1,5, and 20 gmol/kg body wt, respectively. Dogfish sharks excreted an even smaller percentage of the administered mercury into bile: the 3-day cumulative biliary excretion was 0.15 ? 0.15% at a dose of 5 pmol/kg body wt (Fig. 1). By comparison, guinea pigs or rabbits given the same dose of mercury can excrete 0.15% of the dose into bile in only 5 hr, while the rat can excrete greater than 3% in 5 hr (Vostal and Clarkson, 1973). Although the overall rate of excretion into bile was low in the skate and shark, the
BILE
409
concentration of radiolabeled mercury in bile (skate = 4- 10 PM, shark = l-2 PM at a dose of 5 pmol/kg) was similar to that found in mammalian bile after an equimolar dose of methyl mercury. Thus, the slow rate of biliary methyl mercury elimination in skate and shark was attributed to their unusually low rate of bile formation (Fig. 1). For the skate, bile flow rates averaged 3.1 + 1.4 ml/kg.day on the first day after cannulation and decreased to a value of 1.8 it 0.6 ml/kg.day on the fourth day. For the dogfish shark, bile flow was 2.14 -+ 0.7 1 and 0.83 + 0.33 ml/kg. day on the first and fourth day after cannulation, respectively. This low rate of bile formation has been described previously in both the free swimming fish and in the isolated perfused skate liver, and is at least one to two orders of magnitude lower than that observed in most mammalian species, including man (Boyer et al., 1976a-c; Reed et al., 1982a,b). That the low biliary excretion of mercury in the skate and shark is not related to an impaired hepatic uptake for methyl mercury is shown in Table 1. Greater than 10% of the administered ‘03Hg was found in the liver of both skates and sharks at 3 days after its administration. Table 1 also lists the distribution of lo3Hg in a number of other organs and tissues. Skeletal muscle. liver. kidney, and whole blood accounted for the major fraction of the administered dose of mercury. Of the tissues analyzed for ‘03Hg, the highest concentrations of mercury (Io3Hg/g tissue) were found in the heart, rectal gland, kidney, and liver of both the shark and the skate. Since the ‘03Hg contents of the head, fins, gills, skin, intestines, intestinal contents, and bone were not determined, and since only representative samples of certain tissues were analyzed for their 203Hg content, a complete quantitative account of the overall distribution of the administered dose of methyl mercury cannot be performed. The reason for the higher normalized biliary excretion rates observed at higher doses of methyl mercury in the skate (Fig. 1) is not
BALLATORI
AND
BOYER
known. The tissue distribution of *03Hg was also dependent on the dose (Table 1). In particular, the percentage of administered ‘03Hg in the kidney and rectal gland was over twofold greater at the lowest as compared to the higher doses. Although these observations are consistent with saturation of a high-affinity/ low-capacity compartment for methyl mercury in the skate, no direct evidence is presently available for the existence of such a compartment. The ability of the skate to excrete radiolabeled mercury into bile at a higher concentration and rate than the dogfish shark may be related to differences in the biliary excretion of glutathione (Fig. 1). as previously shown for the rat (Refsvik and Norseth, 1975: Ballatori and Clarkson, 1982, 1983, 1985b). Hepatic levels of glutathione in the skate and shark were 0.62 + 0.43 (n = 14) and 0.36 + 0.12 (n = 8) ymol glutathione equivalents/ g, respectively. However, glutathione was detectable only in skate bile; the concentration of glutathione in bile was 0.69 -t 0.50 and 0.76 f 0.83 /*mol glutathione equivalents/ml on the first and fourth days of bile collection, respectively. Most (>80%) of the glutathione in excreted skate bile was in the disulfide form, GSSG. However, because of the long bile collection interval, 24 hr, and because of the lack of an antioxidant in the bile collection balloons, it was impossible to distinguish whether the skate actually secreted GSSG into bile or whether it was the product of the postsecretory oxidation of biliary GSH. Gel filtration of bile on Sephadex G-75 showed that a major fraction of the 203Hg in freshly collected skate bile eluted in the low molecular weight region (less than approximately 10,000 Da; V,/ V,, > 2) (Fig. 2). In contrast. most of the mercury in freshly collected shark bile eluted in the void volume. When the individual fractions from the Sephadex G-75 column were analyzed for their glutathione content, glutathione was noted at a VJ V, of 2.8 to 3.5 for skate bile. However, glutathione was not detectable in the G-75 fractions from shark bile. Most of the mercury
METHYL
MERCURY
IN ELASMOBRANCH DOGFISH
SKATE I
I
I
I5
,
I
I
411
BILE I
I
BILE
FIG. 2. Distribution of “‘Hg in Sephadex G-75 fractions of bile and liver cytosol from the little skate (Rajn erinacea) and spiny dogfish shark (Squahsacanthias). Bile collected during the frrst 24-hr interval after the administration of 5 pmol of CH 3203HgCl/kg was applied directly to the 0.9 x 60-cm column, in a volume of0.3 to 0.4 ml. Liver cytosol. obtained by homogenizing liver tissue in 3 to 5 vol ofelasmobranch Ringer’s and centrifuging at 105,OOOg for 1 hr. was applied to the column in a volume of 0.3 ml. Gel filtration was carried out at room temperature (16-23°C) at a flow rate of 23 ml/hr. One-milliliter fractions were collected. The eluant was 0.1 M Tris-HCl buffer, pH 7.4, containing 1 mM EDTA and 0.02% sodium azide. Radioactivity in each fraction was determined by gamma scintillation spectrometry. The void volume was defined by the elution of blue Dextran. and the elution volume of glutathione was determined by analyzing the individual fractions for their glutathione content. Glutathione itself, as well as the *“Hg from a mixture of CH3203HgCl and glutathione, eluted from the G-75 column at a VJV, of 2.8 to 3.5. Although glutathione was noted at a VJVa of 2.8 to 3.5 for the liver cytosols of both skate and shark, and in bile from the skate, no glutathione was found in the Sephadex G-75 fractions of shark bile.
in liver cytosol from both sharks and skates eluted in the void volume (Fig. 2). Similar ‘03Hg elution profiles to those shown in Fig. 2 were obtained when bile collected on Days 2 and 3 after methyl mercury administration were applied to the Sephadex G-75 columns as well as when freshly collected shark or skate bile and liver cytosols were supplemented in vitro with CH3’03HgCl. In addition, sulfobromophthalein (BSP), a compound known to inhibit both glutathione and methyl mercury excretion in rats (Ballatori and Clarkson, 1985a), and L-buthionineS,R-sulfoximine (L-BSO), an inhibitor of glutathione biosynthesis (Griffith and Meister, 1979), were able to inhibit glutathione and mercury excretion in the skate (Table 2). Al-
though BSP is selectively cleared by the skate liver, it is excreted into bile much more slowly than in mammals (Boyer et al., 1976a-c), presumably accounting for the prolonged inhibition of glutathione and methyl mercury excretion into skate bile (Table 2). In addition to inhibiting glutathione efflux into bile, BSP administration was also associated with a markedly increased hepatic tissue concentration of glutathione (Table 2). Since the increase in hepatic glutathione levels is much larger than can be accounted for by diminished efflux into bile, BSP must have also inhibited either the intracellular utilization or efflux of glutathione from the liver into blood plasma. L-Buthionine-S,R-sulfoximine, on the other hand, inhibited hepatic
BALLATORI
AND
BOYER
glutathione synthesis (Table 2). Since reductions in the biliary excretion of glutathione and ‘03Hg were not evident until the second day after L-BSO administration, the rate of hepatic glutathione turnover in this species might also be relatively slow. Additional experiments are needed to establish this possibility. DISCUSSION The selective bioconcentration of methyl mercury by aquatic organisms has serious environmental as well as human health consequences (Jernelov, 1972; Krenkel, 1973: Friberg and Vostal, 1972; Kurland et a/., 1960; Kutsuna, 1968). The reasons for the long biological half-time of methyl mercury in fish are not known, but may be related to (1) inefficient excretory mechanism (biliaryfecal, urinary, or secretion across the gills), (2) an unusually high affinity of methyl mercury for the proteins in fish muscle, or (3) a combination of these factors. Because of the long biological half-time of methyl mercury in fish, it has been extremely difficult to assess the relative contributions of the various excretory routes in the whole-body elimination of the metal by fish. Studies by Giblin and Massaro ( 1973) showed that the biological half-time for methyl mercury in rainbow trout is greater than 200 days, and that the main route of its excretion in trout, as in mammals, is through the feces. These investigators also noted that only a small amount of mercury appears in gallbladder bile of methyl mercury-treated fish, suggesting that the slow fecal elimination is the result of slow biliary excretion of methyl mercury. However, since the concentration of methyl mercury in gallbladder bile is a poor index of the rate, or extent? of mercury excretion into bile, the overall contribution ofthis excretory mechanism in the elimination of methyl mercury by trout could not be assessed.
METHYL
MERCURY
IN ELASMOBRANCH
In the present study we quantified the biliary excretion of methyl mercury in two elasmobranch species by directly collecting hepatic bile in the free swimming fish. Our results indicate that both skate and dogfish shark excrete mercury into bile slowly, when compared to the rate of biliary excretion by other animals. The inefficient hepatic excretory process was primarily related to the inordinately slow production of bile by these fish (Fig. 1 and Boyer et al., 1976a-c; Reed et al., 1982a,b). Since biliary excretion is a major determinant of fecal methyl mercury elimination, a slow biliary excretory process would retard whole-body elimination of the metal, and favor its accumulation in fish muscle. In a preliminary study, Guarino et al. ( 1972) found that urinary excretion was also a minor pathway of methyl mercury elimination in the dogfish shark. During the first 4 hr after the intravenous administration of 0.5 pmol [‘4C]CH3HgCl/kg, 0.10% of the administered dose was excreted in urine, with an additional 0.19% excreted in the subsequent 20 hr. The amount of radiolabel in the kidney was 5.9 and 2.9% of the administered dose at 4 and 24 hr, respectively (Guarino et al., 1972). Since more mercury is delivered to the kidney, and therefore presumably into urine, after its intravenous administration, when compared to its oral or intramuscular administration in fish (Backstrom, 1969), the values given above probably represent maximal rates of urinary excretion by the dogfish. Thus, as with the biliary route, urinary excretion appears to be inefficient in elasmobranchs. These slow excretory processes may explain the avid retention of methyl mercury by certain fish species. Whether fish muscle also possess an unusually high chemical affinity for methyl mercury, as has been suggested (Burrows and Krenkel, 1973: Giblin and Massaro, 1973; Jarvenpaa et al., 1970: Lockhart et al., 1972), remains to be determined. In the present studies, sequestration of methyl mercury by fish muscle did not ap-
BILE
413
pear to be the rate-limiting factor in the elimination of this metal by elasmobranchs. Although both the skate and shark excreted mercury into bile slowly, there were significant differences in the hepatic handling of methyl mercury between these elasmobranchs that were partly attributed to differences in biliary glutathione excretion (Fig. I), as previously shown for the rat (Refsvik and Norseth, 1975; Ballatori and Clarkson, 1982, 1983, 1985b). The skate. which had a higher hepatic tissue concentration of glutathione than the shark, and excreted a substantial amount of glutathione into bile (Fig. l), also excreted radiolabeled mercury into bile at a higher concentration and rate than the dogfish shark. On gel filtration, most of the ‘03Hg in skate bile eluted in the low molecular weight region, where glutathione would elute, while most of the mercury in shark bile which had undetectable amounts of glutathione, eluted near the void volume (Fig. 2). In addition, treatment of the skate with agents known to decrease biliary glutathione levels in rats (BSP; Ballatori and Clarkson, 1983, 1985a; L-BSO, Griffith and Meister, 1979) produced parallel changes in the biliary excretion of mercury and of glutathione (Table 2). Taken together, these observations indicate that in addition to the rate of bile formation, the rate of glutathione excretion is a key determinant ofthe biliary transport of methyl mercury in elasmobranchs. The reason for the presence of these relatively high concentrations of glutathione in skate bile, or for its absence in shark bile, is not known. That L-BSO and BSP elicited similar changes in biliary glutathione in the skate as they do in mammals suggests the presence of similar glutathione biosynthetic and transport processes in elasmobranchs. Of interest, the organic anion BSP not only inhibited glutathione efflux into skate bile, but it markedly increased hepatic tissue concentrations of the tripeptide (Table 2). The increase in hepatic glutathione is much larger than can be explained on the basis of dimin-
414
BALLATORI
ished efflux into bile. In the rat, BSP is believed to decrease glutathione secretion into bile by inhibiting the canalicular membrane transport system (Ballatori and Clarkson, 1983, 1985a). However, most (80%) of the glutathione released by the rat liver goes into plasma across a sinusoidal membrane glutathione transport system that is similar, if not identical, to that on the canalicular membrane (Inoue et al., 1984). It has not been established whether BSP also inhibits the sinusoidal elllux of glutathione in rat liver. The present findings in the skate, however, indicate that BSP may be inhibiting the efflux of glutathione across both the canalicular and sinusoidal plasma membrane, to effect a net increase in hepatic glutathione levels. The prolonged inhibition of glutathione efflux in the skate is most likely due to the continual presence of high concentrations of BSP in the skate liver, which are in turn maintained by the slow clearance (mainly in the unconjugated form; Boyer et al., 1976a-c) of the dye into bile. ACKNOWLEDGMENTS We thank Annie Boyer and Christin Ciresi for excellent technical assistance. Supported by grants AM3 1493. AM25636 and AM34989 from the U.S. Public Health Service. Support for N.B. was provided by a postdoctora1 fellowship from the U.S. Public Health Service, AM07497, and a Research Fellowship Award from the American Liver Foundation.
REFERENCES BACKSTROM, J. (1969). Distribution studies of mercuric pesticides in quail and some fresh-water fishes. Acta Pharmacd Toxicol. 27(Suppl. 3), l-103. BALLATORI, N., AND CLARKSON, T. W. (I 982). Developmental changes in the biliary excretion of methylmercury and ghttathione. Science 216,6 l-63. BALLATORI. N., AND CLARKSON, T. W. (I 983). Biliary transport of glutathione and methylmercury. Amer. J. Physiol. 244, G435-G44 1. BALLATORI. N., AND CLARKSON, T. W. (1985a). Sulfobromophthalein inhibition of glutathione and meth-
AND BOYER ylmercury secretion into bile. Amer. J. Physiol. 248, G238-G245. BALLATORI, N., AND CLARKSON. T. W. (1985b). Biliary secretion of glutathione and of glutathione-metal complexes. Fundam. Appl. Toxicol. 5,8 16-83 1. BOYER, J. L.. SCHWARZ, J., AND SMITH, N. (1976a). Biliary excretion in elasmobranchs. I. Bile collection and composition. Amer. J. Physiol. 230,970-973. BOYER, J. L., SCHWARZ,J., AND SMITH, N. (1976b). Biliary excretion in elasmobranchs. II. Hepatic uptake and biliary excretion of organic anions. Amer. J. Physiol. 230,974-98 1. BOYER, J. L., SCHWARZ, J.. AND SMITH, N. (1976~). Selective hepatic uptake and biliary excretion of %-sulfobromophthalein in marine elasmobranchs. Ga.struenterology70,254-256. BURGER, J. W. ( 1967). Some parameters for the dogfish, Squalus acanthias. Mt. Desert Island Biol. Lab. Bull. 7,5-9.
BURROWS, W. D., AND KRENKEL, P. R. (1973). Studies on uptake and loss of methylmercuryby bluegills. Environ. Sri. Technol. 7, I 127- 1130. CLARKSON, T. W. (I 972). Recent advances in the toxicology of mercury with emphasis on the alkylmercurials. CRC Crit. Rev. Toxicol. 2, 203-234. FRIBERG, L., AND VOSTAL. J. (1972). Mercury in the Environment. Chemical Rubber Co. Press, Cleveland, Ohio. GIBLIN, F. J., AND MASSARO, E. J. (1973). Pharmacodynamics of methylmercury in the rainbow trout (Sulmo gairdneri): Tissue uptake, distribution, and excretion. To-&l. Appl. Pharmacol. 24, 8 l-9 1. GRIF’F~TH,0. W. (1980). Determination of glutathione and glutathione disultide using glutathione reductase and 2-vinylpyridine. Anal. Biochem. 106,207-2 12. GRIFRTH, 0. W.. AND MEISTER, A. (1979). Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocystine sulfoximine). J. Biol. Chem. 254,7558-7560. GUARINO, A. M., BRILEY. P. M., ANDERSON, J. B.. KINTER. M. A., SCHNEIDERMAN. S., KLIPP, L. D.. AND ADAMSON, R. H. (1972). Renal and hepatic excretion of foreign compounds by Squalus acanthias. Mt. Desert Island Biol. Lab. Bull. 12,4 l-43. INOUE, M.. KINNE, R.. TRAN. T., AND ARIAS, 1. M. (1984). Glutathione transport across hepatocyte plasma membranes: Analysis using isolated rat-liver sinusoidal-membrane vesicles. Eur. J. Biochem. 138, 491-495.
JARVENPAA, T., TILLANDER, M., AND MIETTINEN, J. K. (1970). Methylmercury: Half-time of elimination in flounder, pike and eel. Suom. Kemistii. B43,439-442. JENSEN, S.. AND JERNELOV. A. (1969). Biological methylation of mercury in aquatic organisms. Nature (London) 223,753-754.
METHYL
MERCURY
IN ELASMOBRANCH
JERNELOV, A. (1972). Mercury and food chains. In Environmental Mercury Contamination (R. Hartung and B. 0. Dinman, eds.), pp. 174-I 77. Ann Arbor Sci. Publ., Ann Arbor, Mich. JERNELOV, A.. AND LANN, H. (197 1). Mercury accumulation in food chains. Oikos 22,403-406. KRENKEL, P. A. (1973). Mercury Environmental Considerations, Part 1, CRC Critical Reviews in Environmental Control, pp. 303-373. CRC Press, Cleveland, Ohio. KURLAND, L. T., FARO, S. W., AND SIEDLER. H. (1960). Minamata disease: The outbreak ofa neurologic disorder in Minamata, Japan and its relationship to the ingestion of seafood contaminated by mercuric compounds. A’orld Neural. 1,370-39 1. KUTSUNA, S. (1968). Minamata Disease. Study Group of Minamata Disease, Kumamoto University, Japan. LOCKHART, W. L.. UTHE, J. F., KENNEY, A. R., AND MEHRLE, P. M. (1972). Methylmercury in Northern Pike (Esox lucius): Distribution. elimination. and some biochemical characteristics of contaminated fish. J. Fish. Res. BoardCanad. 29, 1519-1523. MCKIM, J. M., CHRISTENSEN, G. M.. TUCKER, J. H., BENOIT, D. A., AND LEWIS. M. J. (1974). Effects of pollution on freshwater fish. J. U’ater Polk. Control FcJd.46, 1562- 1568.
BILE
415
REED, J. S., SMITH, N. D., AND BOYER, J. L. (1982a). Determinants of biliary secretion in isolated perfused skate liver. Amer. J. Phvsiol. 242, G3 19-G325. REED, J. S., SMITH, N. D., AND BOYER, J. L. (198213). Hemodynamic effects on oxygen consumption and bile flow in isolated skate liver. Amer. J. PhJaiol. 242, G313-G318. REFSVIK. T., AND NORSETH, T. (1975). Methyl mercuric compounds in rat bile. ‘4cta Pharmacol. To,yicol. 36, 67-78.
THORSON, T. ( 1958). Measurement of the fluid compartments of four species of marine Chondrichthyes. Ph.vsiol. Zool. 31, 16-23. TIETZE, F. (1969). Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: Applications to mammalian blood and other tissues. Anal. Biochem. 27,502-522. VOSTAL. J. J., AND CLARKSON, T. W. ( 1973). Mercury as an environmental hazard. J. Occup. Med. 15,649656.
WOOD, J. M. ( 1972). A progress report on mercury. Environment 14,33-39. World Health Organization (1976). Environmental Health Criteria. 1, Mercury. World Health Organization. Geneva.