Inhibition of the Copper Incorporation into Ceruloplasmin Leads to the Deficiency in Serum Ceruloplasmin Activity in Long-Evans Cinnamon Mutant Rat*

Although ceruloplasmin is known to be a copper- transporting protein, little is known about the bio- chemical mechanisms of copper incorporation into ceruloplasmin during the biosynthesis. We have exam- ined various levels of ceruloplasmin biosynthesis in the Long-Evans Cinnamon (LEC) rat, which possesses a mutation causing the deficiency in serum ceruloplas- min activity associated with excess hepatic copper accumulation. Southern and Northern blot analyses re-vealed that the gene and mRNA encoding ceruloplas- min resided normally in LEC rat liver. Western blot analysis showed a normal level of ceruloplasmin in LEC rat serum. Following metabolic labeling of hepatocytes with “Cu, no radioactive copper was detected in the ceruloplasmin fraction in LEC rat hepatocytes using Sephadex G-75 column chromatography, indi- cating that copper incorporation into ceruloplasmin is deficient in the LEC rat. Furthermore, LEC rat hepa- tocytes incubated with 64Cu also showed a reduction in the efficiency of copper transport from cytosolic to noncytosolic fractions and a reduced copper efflux from the hepatocytes, indicating that LEC rat hepatocytes possess an abnormality in copper metabolism. These results suggest that biosynthesis in LEC rat liver. Our results suggest that an abnormality of the copper delivery mechanism inhibits the incorporation of copper atoms into the ceruloplasmin molecule in the liver, and as a result, the ceruloplas-The

Copper is an integral enzyme cofactor essential for a variety of processes in homeostasis such as electron transport, amino acid metabolism, connective tissue biosynthesis, pigment formation, and neurotransmitter and hormone production (14). Ceruloplasmin provides an attractive model for studying the mechanisms of copper protein biosynthesis. The complete amino acid and nucleotide sequences of ceruloplasmin have been determined in both the human and the rat (3,13,15,16), and the cis-acting element in the 5"flanking region of the ceruloplasmin gene has been recently characterized (17). Copper is incorporated into newly synthesized ceruloplasmin within hepatocytes (18), and turnover data indicate that very little copper exchanges from the protein in the circulation (19,20). About 10% of circulating ceruloplasmin occurs as apoprotein, presumably synthesized and secreted from the liver without copper incorporation (21). The biosynthesis and secretion of apo-and holoceruloplasmin from a human cell line occur at identical rates (18). However, little is known about the molecular structure essential for copper incorporation into ceruloplasmin during biosynthesis and the biochemistry involved in this process.
It has been reported that the Long-Evans Cinnamon (LEC)' mutant rat spontaneously develops a necrotizing hepatic injury and liver cancer (22,23). Recently,Li et al. (24) reported that the LEC rat exhibits an abnormal accumulation of hepatic copper and a marked decrease in serum ceruloplasmin activity, and they proposed the hypothesis that the cytotoxicity of excessively accumulated hepatic copper is likely to cause necrotizing hepatic injury. However, the molecular mechanisms of the deficiency in ceruloplasmin activity associated with hepatic copper accumulation in the LEC rat have not been clarified.
In the present study, we have examined various levels of ceruloplasmin biosynthesis in LEC rat liver. Our results suggest that an abnormality of the copper delivery mechanism inhibits the incorporation of copper atoms into the ceruloplasmin molecule in the liver, and as a result, the ceruloplas-The abbreviations used are: LEC, Long-Evans Cinnamon; PBS, phosphate-buffered saline; kb, kilobase pair(s).

8965
Inhibition of Copper Incorporation into Ceruloplasmin min without copper atoms cannot exert its enzymatic activity in the serum of the LEC rat. In addition, the abnormality probably causes the inhibition of biliary copper excretion. This inhibition together with the inhibition of copper exclusion by ceruloplasmin seem to lead to copper accumulation in LEC rat liver. Our present report suggests that the copper incorporation into ceruloplasmin and the biliary copper excretion require an identical intracellular biochemical process responsible for the copper delivery. The LEC rat should be an interesting animal which is primarily deficient in a process responsible for the copper deliveries into a site of copper incorporation into the ceruloplasmin and biliary copper excretion pathway. Thus, the LEC rat is a good and unique animal model for studying the specific biochemical process responsible for copper delivery.

MATERIALS AND METHODS
Animals-Inbred strains of LEC, F344, and WKAH rats are bred under specific pathogen-free conditions in the Institute for Animal Experimentation, University of Tokushima School of Medicine, which is coded as Tj (Tokushima Japan). F344/Tj and WKAH/Tj rats were used as a control.
Measurement of Copper Content and Ceruloplasmin Activity-Liver or fractions eluted from column chromatography were treated with a mixture of nitric, perchloric, and sulfuric acids. Copper concentrations were determined with an inductively coupled argon plasma emission spectrophotometer, model ICAP-750N (Nippon Jarrell-Ash Co., Kyoto, Japan). The ceruloplasmin activities were measured as an oxidase activity by the method of Schosinsky et al. (25). Serum or fractions eluted from column chromatography (50 pl) were incubated in 1 ml of 0.1 M acetate buffer (pH 5.0) containing 7.88 mM odianisidine &hydrochloride for 5 and 15 min at 30 "C, and the absorption at 540 nm was measured. The oxidase activity was obtained by subtracting the absorbance at 5 min from that of 15 min.
Probe-The probe used in this study was the cDNA clone, phCpl, which contains DNA encoding amino acid residues 202-1046 of the ceruloplasmin (16). This DNA was kindly provided by Dr. Mac-Gillivray (University of British Columbia, Vancouver, Canada). The clone, phCpl, was labeled with [w3'P]dCTP (Amersham International, Amersham, U. K.) using a nick translation kit (Amersham International). The specific activity of the probe was 1.5 X 10' dpm/ a . Southern Blot Analysis-High molecular weight DNA was extracted from rat liver and digested with restriction enzymes BgnI, EcoRI, EcoRV, and Hind111 (Toyobo, Kyoto, Japan). Digested DNAs (IO pg) were electrophoresed, transferred to Biodyne A nylon membranes (Pall, Glen Cove, NY), and hybridized according to the supplier's instructions (Pall). Filters were washed twice in 2 X SSC (1 X SSC is 0.15 M NaCl, 0.015 M sodium citrate (pH 7.4)) containing 0.1% SDS at room temperature for 30 min and 0.5 X SSC containing 0.1% SDS at 65 "C for 30 min and were exposed to XAR-5 x-ray film (Eastman Kodak) with intensifying screens at -70 "C overnight.
Northern Blot Analysis-RNA was prepared from the guanidine thiocyanate (5.5 M) extracts by cesium chloride (5 M ) density gradient centrifugation as described previously (26) with slight modifications. An aliquot (10 rg) of RNA was treated with formaldehyde, subjected to 1% agarose gel electrophoresis, and then transferred to the nylon membrane. The conditions of hybridization, washing, and exposure were the same as the procedure described for Southern blot analysis.
To determine the ratio of the steady-state levels of ceruloplasmin to that of &actin mRNAs, signal intensities were examined with a laser densitometer (Ultrascan XL, Pharmacia, Uppsala, Sweden).
Western Blot Analysis-Sera were collected from LEC and normal rats. An aliquot of sera was subjected to SDS-polyacrylamide (7%) gel electrophoresis as described previously (27). After transferring to the nitrocellulose membrane, the blot was blocked with phosphatebuffered saline (PBS) (pH 7.0) containing 2% BSA and 0.05% Tween 80 and incubated with rabbit anti-rat ceruloplasmin (18). After the blot was washed three times with PBS containing 0.05% Tween 80, immune complexes on the blot were detected with a Vectastain ABC kit (Vector Laboratories Inc., Burlingame, CA). For treatment of sera with endoglycosidase H, 5-pl aliquots of sera were treated with 50 milliunits of endoglycosidase H in 25 mM sodium acetate (pH 5.5) for 16 h at 37 "C. For endoglycosidase F treatment, 5-pl aliquots of sera were treated with 0.5 unit of endoglycosidase F in 50 mM sodium acetate (pH 5.0),20 mM EDTA, 0.1% SDS, 0.75% Nonidet P-40, and 1% (v/v) 2-mercaptoethanol for 16 h at 37 "C. Following enzymatic digests the aliquots were analyzed as described above.
Fractionation of Serum Proteins on a Sepharose CL-4B Column-Sera were collected and pooled from four male rats at 1 month of age. Nine ml of the serum was applied to a Sepharose CL-4B column equilibrated with PBS. The column was eluted at 20 ml/h by using a peristatic pump (model P-1, Pharmacia), and 5-ml fractions were collected. The fractions were assayed for copper content and ceruloplasmin oxidase activity.
Hepatocyte Isolation-Male rats at 6 weeks of age, weighing 150-200 g, were used for hepatocyte isolation. Hepatocytes were isolated as reported previously (28). Briefly, the livers were perfused with a washing solution containing 0.5 mM EGTA and then with a collagenase-containing buffer. The isolated cells were washed with Eagle's minimum essential medium to remove nonparenchymal cells, suspended in Dulbecco's modified Eagle's medium containing 5% fetal calf serum and 30 mg/liter of kanamycin, and then approximately 2 X lo6 cells were placed in collagen-coated dishes (15 X 60 mm). After culture for 3 h, loosely attached and dead cells were removed, and fresh medium was added. The cells were incubated in a CO, incubator overnight and used on the following day. This method gave a yield of 80-95% viable cells by the trypan blue exclusion criterion.
"CU Accumulation in Hepatocytes-Cultured hepatocytes were washed three times with Hanks' balanced salt solution, and Earle's balanced salt solution containing 10% fetal calf serum was added. The accumulation of copper by hepatocytes was initiated by the addition of 5 pCi of "Cu to each plate. Cells were incubated at 37 "C for various periods up to 16 h. After incubation, the cells were washed three times with PBS containing 10 mM EDTA. Cells were then scraped off the dish surface with a rubber policeman following the addition of 1 ml of ice-cold PBS. They were disrupted by freezing and thawing several times, and a 100-pl aliquot of the lysed cell suspension was removed for protein analysis. To measure copper accumulation in cytosolic and noncytosolic fractions, a 900-pl aliquot of the suspension was centrifuged at 100,000 X g for 10 min, and the supernatant and pellet were counted with a y-counter. The "CU as Cu(CH3C00), in 5% CH3COOH was obtained from the Japan Atomic Energy Research Institute (Ibaragi, Japan) at an initial specific activity of 200 mCi/mg and was used within 2 days. y-Counting was performed with a y-counter (Aloka model ARC-361, Tokyo, Japan) with automatic decay correction and background subtraction. Protein content was assayed by BCA protein assay kit (Pierce Chemical Co.).
"CU Efflux from Hepatocytes-Hepatocytes were loaded with "Cu for 6 h as described above. At the end of incubation, radioactive medium was removed, and the cells were washed three times with Hanks' balanced salt solution. Fresh, nonradioactive medium was then added to each plate, and the incubation was continued at 37 "C for an additional 5, 15, and 60 min and 16 h. The cells were assayed for remaining intracellular '%u.
Fractionation of Hepatocyte Lysates on a Sephadex G-75 Column-For column chromatography, the hepatocytes were removed from the plates after exposure to "CU overnight. The cells pooled from five plates were resuspended in 1.5 ml of isolation buffer consisting of 100 mM KC1, 20 mM Hepes, 1 mM dithiothreitol, and 0.1% mebutamate (pH 7.3). They were frozen and thawed several times and centrifuged at 5,000 X g for 10 min to get the cytoplasmic fraction. The supernatant was applied to a Sephadex G-75 column equilibrated with the isolation buffer. The column was eluted at 10 ml/h by using a peristatic pump (model P-1, Pharmacia) at 4 "C, and 0.8 ml of each fraction was collected. Blue dextran (2,000 kDa) and cytochrome c (12.4 kDa) were used as molecular mass markers.

Levels of Hepatic Copper Content and Serum Ceruloplasmin
Activity in the LEC Rat- Table I shows the mean f S.E. of copper concentrations in liver and the levels of serum ceruloplasmin activity in six male animals of WKAH, (LEC x WKAH)FI, and LEC rats. The copper concentrations were approximately 40-fold higher in the LEC rat than in the normal rat. (LEC X WKAH)Fl rats showed a normal level of hepatic copper concentration. The difference between LEC and WKAH or (LEC X WKAH)F1 rats was statistically significant. Since the difference between WKAH and (LEC X WKAH)Fl rats was not found, it was indicated that the  Genomic DNA was digested with four kinds of restriction endonucleases and electrophoresed in 0.7% agarose gel. After being transferred to nylon membranes, the blot was hybridized with 32P-labeled ceruloplasmin cDNA probe. Lanes 1 , 3 , 5 , and 7, normal rat; lanes 2,4,6, and 8, LEC rat; lanes 1 and 2, genomic DNA digested with BglII; lanes 3 and 4, genomic DNA digested with EcoRV; lanes 5 and 6, genomic DNA digested with HindIII; lanes 7 and 8, genomic DNA digested with EcoRI. The length (in kb) of the DNA is shown on the right side of each panel. The result shown is from a representative experiment repeated for a t least three separate DNAs prepared from different animals in both normal and LEC rats.
abnormal copper accumulation in the LEC rat liver is inherited as a recessive trait. The levels of serum ceruloplasmin activity were about 40-fold lower in the LEC rat than in the normal rat. The levels of serum ceruloplasmin activity in (LEC x WKAH)FI rats showed an almost intermediate value between those of LEC and normal rats. The differences among the three strains of rats were statistically significant. Since the value of the serum ceruloplasmin activity level in the F1 rat was intermediate between those of WKAH and LEC rats, it was suggested that the deficiency in serum ceruloplasmin activity in the LEC rat is inherited as a codominant trait. Southern Blot Analysis of the Ceruloplasmin Gene-The deficiency in ceruloplasmin activity could be explained by either a deletion of gene, a defective transcription, a defective translation, or a defect in a post-translational modification mechanism. To test their possibilities, first the ceruloplasmin gene in LEC and normal rats was analyzed by Southern blot hybridization. The hybridization pattern of the genomic DNA digested by restriction endonucleases with a probe for ceruloplasmin is shown in Fig. 1. In both strains, the sizes of the fragments were 5.2 and 9.2 kb for BglII, 2.3 and 3.9 kb for EcoRI, 6.0 and 9.4 kb for EcoRV, and 2.4 and 9.0 kb for HindIII. The results of Southern blot analysis indicated that the ceruloplasmin gene exists normally on the LEC rat genome, excluding the possibility of deletion of the ceruloplasmin gene because of a deficiency in serum ceruloplasmin activity. Among 15 restriction enzymes tested, a restriction fragment length polymorphism between LEC and F344 was detected by using KpnI (29).
Detection of Ceruloplasmin mRNA in the LEG Rat by Northern Blot Analysis-RNAs prepared from liver of LEC and normal rats at 1 and 5 months of age were subjected to Northern blot analysis. The autoradiograph of Northern blot hybridization is shown in Fig. 2. There were no qualitative differences in the ceruloplasmin mRNA between LEC and normal rats. The molecular size of the mRNA was 4,800 nucleotides in the LEC rat, consistent with that of the normal rat. No quantitative difference between LEC and normal rats was observed. The ratio of the steady-state levels of ceruloplasmin to that of P-actin mRNAs in LEC rats at 1 month of age (0.397 f 0.002 in three separate RNA preparations from three individual rats) was indistinguishable from that of agematched normal rats (0.411 f 0.016 in three separate RNA preparations from three individual rats) by Student's t test. Moreover, the ratio of the steady-state levels at 5 months of age was also indistinguishable between normal and LEC rats by Student's t test (0.110 f 0.008 and 0.093 f 0.015 in three separate RNA preparations of normal and LEC rats, respectively). Detection of normal levels of ceruloplasmin mRNA in the LEC rat indicated that the expression process from the ceruloplasmin gene to the mRNA for ceruloplasmin occurs normally.
Detection of Serum Ceruloplasmin in the LEC Rat by Western Blot Analysis-Sera taken from LEC and normal rats at 1 and 5 months of age were subjected to Western blot analysis (Fig. 3). Both 135-kDa fragment and 115-kDa proteolytic fragment were identified with rabbit anti-rat ceruloplasmin in LEC as well as in normal rats, consistent with the results reported previously (18). The results of Western blot analysis indicated that the expression process from the mRNA to single peptide for ceruloplasmin occurs normally in LEC rat agarose gel. After being transferred to nylon membranes, the blot was first hybridized with "P-labeled ceruloplasmin cDNA probe and then rehybridized with 32P-labeled p-actin cDNA probe. Lane 1, normal rat at the age of 1 month; lane 2, LEC rat a t the age of 1 month; lane 3, normal rat a t the age of 5 months; lane 4, LEC rat at the age of 5 months. The arrow indicates the mRNA for ceruloplasmin. The result shown is from a representative experiment repeated for three separate RNAs prepared from different animals in both normal and LEC rats.

Inhibition of Copper Incorporation into Ceruloplasmin
liver and that the produced ceruloplasmin is normally secreted into LEC rat serum. Furthermore, the treatment of sera with endoglycosidase H or F resulted in no change in the mobility in the electrophoresis, indicating that the serum ceruloplasmin of both LEC and normal rats was resistant to these endoglycosidases (Fig. 3). In the treatment by endoglycosidase H or F, no difference in the glycosylation was a t least found between LEC and normal rat serum ceruloplasmin. Copper atoms bound on the ceruloplasmin molecule appear to be essential for the oxidase activity of serum ceruloplasmin. Therefore, we propose the mechanism that the ceruloplasmin in LEC rat serum includes no copper and, as a result, cannot exert oxidase activity.
Identification of Serum Ceruloplasmin in the LEC Rat as Apoceruloplasmin-To depict that the ceruloplasmin in the LEC rat serum includes no copper, the serum was chromatographed on a Sepharose CL-4B column (Fig. 4). One peak containing ceruloplasmin oxidase activity was detected in normal rat and coeluted with copper atoms, indicating that the peak contains holoceruloplasmin (copper atoms-ceruloplasmin molecule complex). In contrast, in the LEC rat, no Western blot analysis of the serum ceruloplasmin. Sera were electrophoresed in SDS-polyacrylamide (7%) gel. After being transferred to the nitrocellulose membrane, the blot was incubated with rabbit anti-rat ceruloplasmin serum. Immune complexes were detected with a Vectastain ABC kit. Lanes I , 5, and 7, normal rat at the age of 1 month; lanes 2,6, and 8, LEC rat at the age of 1 month; lane 3, normal rat at the age of 5 months; lane 4, LEC rat at the age of 5 months; lanes 1-4, untreated serum; lanes 5 and 6, serum treated with endoglycosidase H; lanes 7 and 8, serum treated with endoglycosidase F. The result shown is from a representative experiment repeated for a t least three separate sera prepared from different animals in both normal and LEC rats. peak showing ceruloplasmin activity was detected, and only a very small amount of copper atoms was detected in fractions corresponding to the peak containing ceruloplasmin in normal rat. This result indicates that the ceruloplasmin in the LEC rat serum exists as apoceruloplasmin (ceruloplasmin molecule including no copper atoms), which, therefore, cannot exert the oxidase activity. We thought that the occurrence of apoceruloplasmin in LEC rat serum was caused by a defect in either the transportation of holoceruloplasmin from liver to serum or formation of a copper-ceruloplasmin complex in liver. Therefore, we examined copper binding to ceruloplasmin by fractionating copper-binding proteins in 64Cu-loaded hepatocytes.
Distribution of -Cu among Hepatic Copper-binding Proteins-Rat hepatocytes were incubated with 64Cu overnight and chromatographed on a Sephadex G-75 column. Two major peaks (I and 11) containing 64Cu activity were detected in normal rat hepatocytes (Fig. 5). Peaks I and I1 were eluted at the void volume and near 10-kDa position, respectively. From the molecular mass, peaks I and I1 were thought to correspond to major copper-binding proteins, ceruloplasmin and metallothionein, respectively. Peak I was confirmed to be ceruloplasmin by Western blot analysis with anti-rat ceruloplasmin (Fig. 5, inset). In contrast to normal rat hepatocytes, LEC rat hepatocytes showed no @Cu activity in the ceruloplasmin fraction. Since the total cytoplasmic fraction of LEC rat liver contains a ceruloplasmin level equal to that of control rats when analyzed with Western blot (data not shown), it was indicated that the deficiency in serum ceruloplasmin activity was caused by the nonformation of the copper-ceruloplasmin complex, namely the inhibition of copper incorporation into ceruloplasmin in LEC rat hepatocytes. The lack of formation of the copper-ceruloplasmin complex could be explained by either an abnormality of copper delivery into ceruloplasmin or a mutation in the ceruloplasmin molecule in LEC rat hepatocytes. However, the possibility of a mutation in the ceruloplasmin molecule was excluded by the data in genetic analysis (see "Discussion"). A question is raised as to whether the abnormal accumulation of hepatic copper (Table I)

FIG. 5. Distribution of "Cu within hepatic copper-binding proteins in hepatocytes.
Hepatocytes were incubated with 5 pCi of W u overnight. The cells were suspended in isolation buffer and disrupted as described under "Materials and Methods." The cytoplasmic fraction of normal (open circles) and LEC (closed circles) rat hepatocytes was applied to a Sephadex G-75 column and eluted with the isolation buffer at a flow rate of 10 ml/h. 0.8 ml of each fraction was collected and counted for T u . Western blot analysis showed the presence of ceruloplasmin in fractions 8-10. The arrow indicates the position of cytochrome c. The arrowheads indicate the positions of peaks I and 11. The experiments were repeated for three separate hepatocytes prepared from different animals in both normal and LEC rats, and this is representative of the three similar results. Total "2u content value in a mixture of fractions 8-10 was significantly different between normal and LEC rats (12.76 k 5.04 and 0.41 & 0.15 ng in the three preparations of normal and LEC rats, respectively) ( p < 0.05 by Student's t test).
ceruloplasmin. We thought that a putative defect in LEC rat liver might affect the other copper metabolism pathways. Therefore, we examined the copper metabolism, namely uptake, compartmentalization, and efflux in LEC rat hepatocytes, by loading them with 64Cu. @Cu Accumulation by Cytosolic and Noncytosolic Fractions in LEC Rat Hepatocytes-When the total copper contents in isolated hepatocytes were measured, they were about 170 and 330 pmol/mg of protein in cytosolic and noncytosolic fractions, respectively. On the contrary, the total copper content in LEC rat hepatocytes showed about 6,500 and 2,100 pmol/ mg of protein in the cytosolic and noncytosolic fractions, respectively. The ratio of the cytosolic copper content to the noncytosolic copper content in the LEC rat was 3:1, whereas that in the normal rat was 1:2. The uptake of copper into the cytosol largely precedes the copper uptake into other noncytosolic compartments (30), and in hepatocytes loaded with excess copper most of the excess copper is accumulated in the noncytosolic compartment, which includes nuclear, mitchondrial, and microsomal fractions (31, 32). Therefore, we thought of the possibility that in LEC rat hepatocytes the copper cannot be transported from the cytosol into noncytosolic compartments. To assess this possibility, the accumulation of radioactive copper in cytosolic and noncytosolic fractions of hepatocytes was examined using primary-cultured hepatocytes. 64Cu accumulation in the noncytosolic fraction of LEC rat hepatocytes was about 2-fold lower than that of normal rat hepatocytes, whereas W u accumulation in the cytosolic fraction of LEC rat hepatocytes was indistinguishable from that of normal rat hepatocytes (Fig. 6). Student's t test confirmed that at the loading time of 16 h, the value of noncytosolic T u accumulated was significantly different be- FIG. 6. '%u accumulation in cytosolic and noncytosolic fractions of hepatocytes. Hepatocytes were incubated with 5 pCi of %u in Earle's balanced salt solution plus 10% fetal calf serum at 37 "C. At the indicated times, the cells were washed and processed as described under "Materials and Methods" followed by the determination of MCu accumulation in cytosolic (panel a ) and noncytosolic (panel b) fractions of normal (open circles) and LEC (closed circles) rat hepatocytes. Each point and bar represents the mean & S.E. of triplicate observations. The experiments were repeated for two and three separate hepatocytes prepared from different animals in normal and LEC rats, respectively, and this is representative of the two similar results for control rats and three for LEC rats. tween normal and the LEC rats (182.2 f 13.6 and 69.2 f 13.6 pmol/mg in combined data from two and three separate experiments using different hepatocyte preparations of normal and LEC rats, respectively), whereas the value of cytosolic 64Cu accumulated was indistinguishable (17.8 f 0.6 and 20.8 f 3.0 pmol/mg in combined data from two and three separate experiments using different hepatocyte preparations of normal and LEC rats, respectively). Therefore, these results suggest that the efficiency of copper transport from the cytosol to noncytosolic compartments is reduced in the LEC rat compared with the normal rat and that as a result, a large part of the abnormally accumulated copper exists in the cytosolic fraction in LEC rat hepatocytes.
Efflux of 64Cu from Preloaded Hepatocytes-Hepatocytes were preloaded with 64Cu for 6 h, and then the cells were recultured in 64Cu-free media. As shown in Fig. 7,65% of the preloaded ' %u was released from normal rat hepatocytes a t 960 min after reculturing in 64Cu-free media. In contrast, LEC rat hepatocytes exhibited a reduced efflux of 64Cu from hepatocytes. Student's t test confirmed that the value of the fraction of Y!u retained was significantly different between normal and LEC rats at 960 min after reculturing (36.7 f 9.1 and 85.3 & 10.2% in three separate hepatocyte preparations of normal and LEC rats, respectively). This result, together with the data showing the reduction in the efficiency of copper transport from the cytosolic to noncytosolic fractions, indicated an abnormality of the copper delivery mechanism in LEC rat hepatocytes. We therefore postulate the hypothesis that the abnormality in copper delivery into the ceruloplasmin molecule is a primary cause for the deficiency in the serum ceruloplasmin activity. In addition, the abnormality seems to cause an inhibition of biliary copper excretion and, as a consequence, abnormal copper accumulation in LEC rat hepatocytes (see "Discussion").

DISCUSSION
Our present study demonstrates that the deficiency in serum ceruloplasmin activity in the LEC mutant rat is caused by the inhibition of the incorporation of copper atoms into ceruloplasmin, which is essential to its catalytic activity. The question arises: what is the primary defect for the inhibition of copper incorporation into ceruloplasmin which leads to the deficiency in serum ceruloplasmin activity? Indeed, the ceruloplasmin gene existed normally on the LEC rat genome, and the expression process to protein occurred normally in LEC rat liver. We considered the possibility that the ceruloplasmin molecule might possess a mutation by which the ceruloplasmin molecule cannot bind copper. However, we have obtained the result that the ceruloplasmin gene is not the gene causing the deficiency in serum ceruloplasmin activity by genetic analysis using restriction fragment length polymorphism of the ceruloplasmin gene detected with KpnI (29). From our present data in 64Cu distribution among copper-binding proteins together with the result of the genetic study, we have suggested that the inhibition of the copper incorporation into ceruloplasmin is because of an abnormality of copper delivery into ceruloplasmin. The LEC rat might lack a putative factor essential to the transportation of the copper atom to the cellular sites where copper is incorporated into ceruloplasmin. In the brindled mouse, which is an inborn error of copper metabolism and an animal model of Menkes disease (33), a 48-kDa cytosolic copper-binding protein has been suggested as a protein responsible for the primary defect in abnormal copper accumulation in kidney (34). Hepatocytes may possess a similar protein involved in intracellular copper delivery. Metallothionein levels are elevated as the copper-bound form in LEC rat liver (35 and Fig. 5). It should be noted here that the elevated metallothionein is supposed to be a secondary effect accompanied by the elevated copper levels rather than the primary defect.
Our data in the 64Cu efflux experiment indicate that the abnormality of the copper delivery mechanism in LEC rat hepatocytes also causes a reduced copper efflux from hepatocytes, leading to the excess hepatic copper accumulation. The major route of copper efflux from hepatocytes is thought to be copper excretion into bile. Based on our previous data that the two anomalies, the deficiency in serum ceruloplasmin activity and the hepatic copper accumulation, are correlated completely (29), copper delivery into the biliary copper excretion pathway appears to be impaired in the LEC rat through the primary defect associated with the abnormality of copper delivery into ceruloplasmin. Furthermore, from the data of 64Cu accumulation by hepatocytes, this primary defect in the LEC rat also seems to cause an inhibition of the copper transportation from the cytosol to noncytosolic compartments. We assume that the major noncytosolic compartments should be the dense polysome fraction on the rough endoplasmic reticulum and lysosomes. The possibility of the rough endoplasmic reticulum is partly supported by evidence that the copper atoms are incorporated only into newly synthesized ceruloplasmin and that de nouo synthesis of the ceruloplasmin appears to occur in a dense polysome fraction on the rough endoplasmic reticulum (18). The possibility of lysosomes is partly supported by evidence that the pathway for biliary copper efflux involves the lysosomes (36).
The kinetic parameter Vmax for copper uptake in LEC rat hepatocytes was significantly lower than that in normal rat, whereas the kinetic parameter K, was the same in LEC and normal rats (data not shown). This indicates that in the LEC rat an intracellular putative copper carrier protein, to which copper binds initially after the passive copper entrance, has an identical affinity for copper but decreases copper binding sites as compared with the normal rat. The apparent lower Vmax of copper uptake for hepatocytes in the LEC rat would be explained by the occupation of copper binding sites on a putative cytosolic factor, since the rate in copper transportation from cytosolic to noncytosolic fractions was reduced in LEC rat liver. However, a reduction of Vmax might be explained by the fact that the capacity of the copper binding in the noncytosolic fraction has been filled in LEC rat hepatocytes. Therefore, it seems important that we present evidence that the reduction in the copper transport rate is not a secondary effect of the accumulated copper in noncytosolic fraction. The total copper content in LEC rat fetal liver was shown to be lower (1,000 pmol/mg of protein) than that of normal rat fetus (7,700 pmol/mg of protein) (data not shown), probably reflecting the lower serum copper content in the pregnant LEC rat than in a normal pregnant rat. The ratio of the cytosolic to noncytosolic copper content in the LEC rat fetus is similar to that of the adult, in spite of the total hepatic copper content in the fetus being smaller than that in the normal rat fetus. These data suggest that the reduction in the copper transport rate is a primary inherent defect in the LEC rat irrespective of the copper accumulation.
Based on the results in our present study together with others (24), we consider that the pathogenesis of the hepatic disorder in the LEC rat is very similar to that of human Wilson's disease (37-39). First, the deficiency in serum ceruloplasmin activity in the LEC rat is caused by the inhibition of copper incorporation into ceruloplasmin, and the biliary copper excretion is also inhibited in the LEC rat. These data are consistent with the data reported for Wilson's disease (40). Second, dissociation of the ceruloplasmin gene from a putative gene causing the deficiency in serum ceruloplasmin activity in the LEC rat (29) is consistent with the data reported for Wilson's disease, namely, the ceruloplasmin gene is located on chromosome 3, whereas a putative Wilson's disease gene is located on chromosome 13 (41-43). Further detailed genetic and molecular analyses in the LEC rat may elucidate the pathogenesis of human Wilson's disease as well as a specific biochemical process responsible for the intracellular copper delivery.