- Email: [email protected]
A Mutant of Deleted Variant of Hepatocyte Growth Factor (dHGF) with Alanine Substitution in the N-Terminal Basic Region Has Higher Activityin Vivo
Biochemical and Biophysical Research Communications 254, 363–367 (1999) Article ID bbrc.1998.9950, available online at http://www.idealibrary.com on
A Mutant of Deleted Variant of Hepatocyte Growth Factor (dHGF) with Alanine Substitution in the N-Terminal Basic Region Has Higher Activity in Vivo Masahiko Kinosaki, 1 Kyoji Yamaguchi, Yasushi Yamashita, Yasuhiro Uematsu, Hiromi Aihara, Hiroaki Masunaga, Tomonori Morinaga, and Kanji Higashio Research Institute of Life Science, Snow Brand Milk Products Co. Ltd., Ishibashi-machi, Shimotsuga-gun, Tochigi, Japan
Received November 20, 1998
In a previous study, we generated a mutant of dHGF (deleted variant of hepatocyte growth factor), termed #2, with higher specific activity than dHGF in assays of mitogenic activity on rat hepatocytes and America opossum kidney epithelial cells (OK). In the present study, we examine in vivo hepatotropic and renotropic activities of #2 and its distribution to target tissues, liver and kidney. Administration of #2 to normal rats significantly increased serum levels of total protein, albumin, free-cholesterol, and HDL-cholesterol and liver weight in a dose-dependent manner. Analysis of these parameters suggests that #2 is more potent than dHGF as a hepatotropic factor in vivo. In addition, #2 reduced mortality of mercuric chloride-administered mice and the effect was stronger than that of dHGF. When injected to mice, a larger amount of #2 than dHGF was rapidly distributed to the liver. Sixty minutes after injection, the concentrations of #2 in plasma, liver, and kidney were higher than those of dHGF. These distribution properties and the higher mitogenic activity in vitro may explain why #2 exerts more potent in vivo biological activity than dHGF. © 1999 Academic Press
Hepatocyte growth factor (HGF) (1, 2), also designated as scatter factor (3) or fibroblast-derived tumor cytotoxic factor (4), is a heparin-binding basic protein. The structural analysis based on the amino acid sequence deduced from the cDNA (1, 2) predicts that it consists of 6 major domains: the N-terminal domain 1 To whom correspondence should be addressed at Research Institute of Life Science, Snow Brand Milk Products Co. Ltd., 519 Ishibashi-machi, Shimotsuga-gun, Tochigi 329-0512, Japan. Fax: 81(285)53-1314. E-mail: [email protected]. Abbreviations used: HGF, hepatocyte growth factor; dHGF, deleted variant of hepatocyte growth factor; HDL, high-density lipoprotein; ELISA, enzyme-linked immunosorbent assay; PBS, phosphatebuffered saline; Chaps, 3-[(3-cholamidopropyl)-dimethylammonio]-1propanesulfate.
(including the N-terminal basic region and the hairpin loop structure) and four kringle domains in an a-chain, and a serine protease-like domain in a b-chain. In addition to intact-form HGF, an alternatively spliced variant, which lacks five consecutive amino acid residues (the FLPSS sequence) in the first kringle domain, was isolated and termed dHGF (deleted variant of HGF) (5, 6). Recent studies have revealed that both HGF and dHGF are biologically multifunctional (4, 7–11). They stimulate growth of various epithelial and endothelial cells, while they inhibit growth of tumor cells (4, 9). They act as hepatotropic factors and renotropic factors (12–17, and Masunaga, H. and Aihara, H. unpublished data). It has been reported that dHGF is distinguishable from HGF in biological and immunological properties. dHGF is more potent than HGF in stimulation of DNA synthesis in rat hepatocytes and epithelial cells (6, 18, 19). To study the structure-function relationship, a large number of HGF and dHGF mutants have been prepared (18, 20 –25). Analysis of the mutants demonstrated that some of the residues in the first kringle domain are critical for the biological activity (23, 25) and at least nine residues in the N-terminal region are involved in the binding to heparin (24). We showed that two dHGF mutants are more potent than dHGF in stimulating DNA synthesis in rat hepatocytes and America opossum kidney epithelial cells (OK) but less potent in mouse myeloblastic cells, NFS-60 (25). In the mutants, termed #2 and #27, all basic amino acid residues in the amino acid sequences, 2RKRR and 27KIKTKK, respectively, in the N-terminal basic region of dHGF are replaced with alanine. In this study, we examined in vivo biological potency of one of the mutants, #2. #2 was more potent in enhancing liver functions in normal rats and prevention of mortality caused by mercuric chloride-induced renal failure in mice. Further analysis revealed that #2 was
363
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
present at a higher level than dHGF in target tissues, liver and kidney, 60 min after intravenous injection. MATERIALS AND METHODS dHGF mutants. dHGF and the mutant, #2, were produced by culturing CHO cells transfected with expression vectors containing respective cDNA. They were purified, respectively, to homogeneity (purity: over 95%) from the conditioned media as previously described (25). Assay for hepatic functions. dHGF or #2 dissolved in PBS containing 0.01% Tween 80 was intravenously administered to male Wistar rats (6 weeks old) twice a day for 4 days at doses of 50, 100 and 200 mg/2 ml/kg body wt. Control rats received the same volume of the vehicle (PBS containing 0.01% Tween 80). Serum and plasma samples were collected from the rats. For evaluating both hepatic protein synthesis and cholesterol metabolism, serum levels of total protein, albumin, free-cholesterol and HDL-cholesterol were determined as previously described (14). To evaluate the relative activity of #2 to dHGF in vivo, parallel line assay was performed as previously described (26). Prevention of mercuric chloride-induced renal failure. dHGF, #2 or vehicle was intravenously administered to male ICR mice (body weight: 30 –35 g, n 5 20 per group) twice a day, total 9 times. Six hours after final administration, mercuric chloride (5 mg/kg body wt) was intravenously administered to the mice. Quantification of dHGF or #2 in plasma, liver, and kidney. dHGF or #2 was intravenously injected to mice. Under ether anesthesia, blood was collected from inferior vena cava of the mice 5, 30, or 60 min after the injection. At the same time point, liver was removed from the mice after perfusing it with saline and kidney was also taken out from the mice. The collected blood samples were centrifuged at 3,000 rpm for 10 min (RD-20IV, Tomy Seiko, Japan) and the serum samples were stored at 280°C until use in ELISA. The removed liver or kidney was homogenized in 4 volumes per weight of buffer (20 mM Tris–HCl buffer, pH 7.5, containing 2 M NaCl, 0.1% Tween 80, 1 mM phenylmethylsulfonyl fluoride and 1 mM EDTA). The homogenate was centrifuged at 19,000g for 30 min at 4°C and the supernatant was stored at 280°C until use in ELISA. dHGF or #2 in serum was quantified by a two step sandwich ELISA employing rabbit anti-dHGF polyclonal antibody as previously described (24) with the following modification. Each well in 96-well immunoplates (Maxisorp, Nunc, Denmark) was treated overnight with 100 ml of 10 mg/ml of anti-dHGF antibody in 0.1 M NaHCO 3 at room temperature and non-specific binding sites were blocked with 50% Block Ace (Snow Brand Milk Products, Co., Japan) in PBS at room temperature for 2 h. Each serum sample containing dHGF or #2 was diluted 2-fold with buffer A (50% Block Ace in 0.2 M Tris–HCl, pH 7.3, containing 0.2 M NaCl, 0.1% Tween 20, 0.2% Chaps, 20 mM benzamidine hydrochloride and 10 mM EDTA) and 100 ml of each of the diluted serum samples was added to each well in the plates. dHGF or #2 serially diluted in normal rat serum was used as a standard for ELISA. The plates were allowed to stand at 4°C overnight and subsequently the wells in the plates were washed four times with PBS containing 0.05% Tween 20. Peroxidaseconjugated anti-dHGF antibody diluted with buffer B (0.1 M phosphate buffer, pH 7.0, containing 10% Block Ace, 0.15 M NaCl, 0.1% Tween 20 and 4% rat serum) was added to each well (100 ml/well) and the plates were further incubated at 37°C for 2 h. The antibody bound to each well was quantified as previously described (24). To quantify dHGF or #2 in the supernatant of liver or kidney homogenate, ELISA system was partially modified as follows. The supernatant of each liver or kidney homogenate was diluted 2-fold with buffer A and 100 ml of the solution was added to each well in 96-well immumoplates coated with anti-dHGF polyclonal antibody. The plates were then incubated at room temperature for 2 h. dHGF
FIG. 1. Effect of dHGF and #2 on serum content of HDLcholesterol. dHGF (closed circles) or #2 (open circles) was administered intravenously to rats every 12 h for 4 days. HDL-cholesterol in the serum was determined as described under Materials and Methods. Data represent means 6 se (n 5 8). The parallel line assay was performed as previously described (26).
or #2 in each well was detected using peroxidase-conjugated antidHGF antibody as described above. To quantify dHGF and #2 in the supernatant of liver or kidney homogenate, dHGF and #2 serially diluted in the supernatant of liver or kidney homogenate derived from normal mice were used as standards for ELISA, respectively.
RESULTS Enhancement of hepatic functions and increase in liver weight in normal rats. Purified dHGF or #2 was administered to normal rats and serum levels of total protein, albumin, free-cholesterol and HDL-cholesterol, and liver weight of the rats were determined. Both dHGF and #2 enhanced hepatic functions such as protein synthesis and cholesterol metabolism and increased liver weight in a dose-dependent manner. To estimate the relative potency of #2 to dHGF in enhancing hepatic functions or increasing liver weight, the parallel line assay was used. A representative result of the analysis using a parameter, HDL-cholesterol, is shown in Fig. 1. The analysis showed a linearity in the dose–response relationship and the linearity established a parallelism between #2 and dHGF (Fig. 1). Similar results were obtained by analyzing other parameters except for liver weight (data not shown). The relative potency estimated by the use of each parameter was summarized in Table 1. The results revealed that #2 was approximately 1.5-fold more potent than dHGF in enhancing hepatic functions (Table 1). Further, #2 was significantly more potent (p , 0.05) than dHGF in increasing liver weight at a dose of 200 mg/ kg/day (Table 1). Protective effect on mercuric chloride-induced renal failure in mice. As described previously, #2 is more potent than dHGF in the stimulation of DNA synthesis
364
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Effects of dHGF and #2 on Liver Weight and Concentrations of Serum Proteins and Cholesterols Liver weight (g/100 g) Vehicle dHGF 50 mg/kg 100 mg/kg 200 mg/kg #2 50 mg/kg 100 mg/kg 200 mg/kg Relative potency of #2 a
Total protein (g/dL)
Albumin (g/dL)
Free-cholesterol (mg/dL)
HDL-cholesterol (mg/dL)
3.526 6 0.059
5.4 6 0.3
2.6 6 0.1
15 6 3
32.5 6 6.3
3.610 6 0.208 3.776 6 0.125 3.797 6 0.204
5.7 6 0.2 5.9 6 0.3 6.1 6 0.3
2.9 6 0.1 3.0 6 0.2 3.1 6 0.1
20 6 4 25 6 4 29 6 4
40.9 6 4.2 50.7 6 7.9 57.4 6 7.1
3.659 6 0.175 3.804 6 0.140 4.020 6 0.165
5.9 6 0.2* 6.2 6 0.2 6.5 6 0.1**
3.0 6 0.1 3.1 6 0.1 3.3 6 0.1*
23 6 5 26 6 6 35 6 6*
46.7 6 6.9 55.4 6 8.7 68.3 6 5.6**
Not paralleled
2.12
1.72
1.47
1.66
Note. dHGF or #2 was administered intravenously to rats every 12 h for 4 days. Data represent means 6 SD for 8 rats. Asterisk and double asterisk represent p , 0.05 and p , 0.01 vs rats receiving the same dose of dHGF by Student’s t test, respectively. a Relative potency of #2 to dHGF was calculated by parallel line assay.
in America opossum kidney epithelial cells (OK) (25). To evaluate the renotropic effect of #2 in comparison with dHGF, preventive effect of dHGF and #2 on mortality caused by mercuric chloride administration was examined. dHGF, #2 or vehicle was administered to mice (n 5 20 per group) prior to intravenous administration of a lethal dose of mercuric chloride. In vehicletreated group, no mice was alive 7 days after mercuric chloride administration. In contrast, 60 and 50% of the mice treated with #2 at doses of 50 mg/mouse/day and 25 mg/mouse/day, respectively, survived for 14 days after the administration (Fig. 2A). Higher doses of dHGF were required to obtain comparable effects with those by #2 (100 mg/mouse/day for 66% and 50 mg/ mouse/day for 50%) (Fig. 2B). These results indicate that #2 is approximately twofold more potent than dHGF in reduction of mortality caused by renal failure induced with mercuric chloride. Levels of dHGF and #2 in plasma, liver and kidney. ICR mice were given a single intravenous injection of dHGF or #2 at a dose of 25 mg/mouse. Five, 30, and 60 min after the injection, the plasma samples were collected and the concentration of dHGF or #2 in the plasma was determined by ELISA as described under Materials and Methods. The concentrations of dHGF and #2 in the plasma at 5 min were 572 6 239 and 305 6 163 ng/ml, respectively. The concentration of dHGF in the plasma decreased rapidly to 26.9 6 6.1 and 6.4060.47 ng/ml within 30 and 60 min, respectively, but that of #2 decreased slowly to 230 6 33 and 96.9 6 22.3 ng/ml (Fig. 3A). #2 was significantly higher than dHGF in hepatic concentration at each of the time points of 5, 30 and 60 min (Fig. 3B). The concentration of #2 in the kidney was much lower than that of dHGF at 5 min (810 6 215 ng/g vs 3674 6 704 ng/g). But beyond 5 min after the injection, #2 in the kidney
FIG. 2. Prevention of mercuric chloride-induced renal failure by dHGF and #2; (A) Survival rate of #2-preadministered mice. (B) Survival rate of dHGF-preadministered mice. Vehicle, dHGF or #2 was administered to mice twice a day. After final administration, mercuric chloride was given to the mice. Crosses, vehicle; square, 100 mg protein/mouse/day; circles, 50 mg protein/mouse/day; diamonds, 25 mg protein/mouse/day; and triangles, 12.5 mg protein/ mouse/day.
365
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Time profiles of dHGF or #2 in plasma (A), liver (B) and kidney (C) after intravenous administration; dHGF (gray bars) or #2 (black bars) was administered to mice. At indicated time point, samples of the plasma, the liver and the kidney were prepared from the mice as described under Materials and Methods. The concentration of dHGF or #2 in the samples was quantified by ELISA. Asterisk and tandem asterisk represent p , 0.05 and p , 0.01 vs group received same dose of dHGF by Student’s t test, respectively.
decreased only slightly. The concentration of #2 in the kidney at 60 min was higher than that of dHGF, which rapidly decreased in a time dependent manner (Fig. 3C). DISCUSSION As with HGF, dHGF is a potent mitogen for a variety of cell types such as hepatocytes, epithelial cells, endothelial cells and myeloblastic cells (6, 19). dHGF is, however, more potent than HGF in stimulation of DNA synthesis in rat hepatocytes (6, 18, 19), America opossum kidney epithelial cells (OK) and pig kidney epithelial cells, LLC-PK1 (19). In vivo, dHGF has been found to ameliorate disordered hepatic protein synthesis in rats with liver failure, and to exert preventive effects on various liver injuries and anti-fibrogenic effect on liver fibrosis in rats (14, 17, 27). In addition, dHGF has preventive effect on renal failure induced with mercuric chloride (Masunaga, and Aihara, unpublished data). In the previous study, we demonstrated that dHGF mutants, #2 and #27, which have alanine substitution in the N-terminal basic region, show a decrease (approximately 3-fold) in dose required for 50% maximal cell proliferative response (ED 50 ) compared with dHGF in stimulation of DNA synthesis in rat hepatocytes (25). The mutants are also more potent than dHGF in stimulation of DNA synthesis in America opossum kidney epithelial cells (OK) (25). In this study, we examined in vivo biological potency of one of the mutants, #2. After administration to normal rats, serum levels of total protein, albumin, free-
cholesterol and HDL-cholesterol, all of which reflect hepatic functions (28, 29), and liver weight were measured. To accurately estimate the relative activity of #2 to dHGF in vivo, the parallel line assay was applied. #2 was approximately 1.5-fold more potent than dHGF in enhancing hepatic functions, and was significantly higher than dHGF in increasing liver weight (Table 1). These results suggest that the higher mitogenic activity of #2 on rat hepatocytes in vitro is at least partly responsible for the more potent in vivo efficacy in the enhancement of hepatic functions and the increase in liver weight. Administration of nephrotoxic reagents, such as cyclosporin A and mercuric chloride, causes a renal failure. Such a renal failure is accompanied with destruction of renal tubular epithelial cells, leading to impaired glomerular filtration (30). HGF stimulates DNA synthesis in renal tubular cells after renal injuries, and prevents the onset of acute renal dysfunction (15). When administered prior to the induction of renal failure with mercuric chloride in mice, #2 was approximately 2-fold more potent than dHGF in reduction of the mortality (Fig. 2). The enhanced mitogenic activity of #2 on kidney epithelial cells in vitro is likely to be one of the reasons why it shows higher efficacy in the prevention of the renal failure than dHGF. The concentrations of #2 in the plasma and the kidney were lower than that of dHGF at a time point of 5 min after the injection, but decreased more slowly and became higher at 30 and 60 min, respectively (Fig. 3A and 3C). In contrast, the concentration of #2 in the liver was higher than that of dHGF at each of the time points after the injection (Fig. 3B). These observations
366
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
suggest that the higher hepatic level and the prolonged retention in kidney also contribute to more potent in vivo biological activity of #2 in these target tissues. Both dHGF and HGF have affinity for heparin, and bind to heparin-like molecules which exist on cell surface and in extracellular matrix (4, 31, 32). Because #2 is lower in affinity for heparin than dHGF (25), #2 may have tendency to stay longer in the circulation without being trapped by vascular cell surface. Liu et al. previously reported that the liver is the major clearance organ and the internalization of HGF takes place not only via the signaling receptors (c-Met), but also via the low-affinity binding sites, presumably cell surface heparin-like molecules (33). The reduced heparinbinding ability could also diminish the binding of #2 to cell surface heparin-like molecules eventually preventing internalization into hepatocytes. This may account for the observation that #2 was sustained at a higher level than dHGF in plasma, liver and kidney 60 min after the injection (Fig. 3). In conclusion, #2, a dHGF mutant with alanine substitution in the N-terminal domain, was found to be more potent than dHGF in stimulation of hepatic functions, increase in liver weight and prevention of renal failure induced with mercuric chloride. Of numerous HGF or dHGF mutants prepared so far, #2 is the first example with enhanced hepatotropic and renotropic activities both in vitro and in vivo. The results in this study suggest that #2 has advantages over dHGF for the treatment of patients with liver or renal diseases.
8.
9. 10. 11.
12. 13.
14. 15. 16. 17.
18.
19.
20.
21.
ACKNOWLEDGMENTS We thank Fumie Kobayashi, Toshiko Satake, Reiko Tadokoro, and Maki Okada for their excellent assistance.
REFERENCES
22. 23. 24.
1. Nakamura, T., Nishizawa, T., Hagiya, M., Seki, T., Shimonishi, M., Sugimura, A., Tashiro, K., and Shimizu, S. (1989) Nature 342, 440 – 443. 2. Miyazawa, K., Tsubouchi, H., Naka, D., Takahashi, K., Okigaki, M., Arakaki, N., Nakayama, H., Hirano, S., Sakiyama, O., Takahashi, K., Gohda, E., Daikuhara, Y., and Kitamura, N. (1989) Biochem. Biophys. Res. Commun. 163, 967–973. 3. Stoker, M. E., Gherardi, E., Perryman, M., and Gray, J. (1987) Nature 327, 239 –242. 4. Higashio, K., Shima, N., Goto, M., Itagaki, Y., Nagao, M., Yasuda, H., and Morinaga, T. (1990) Biochem. Biophys. Res. Commun. 170, 397– 404. 5. Seki, T., Ihara, J., Sugiyama, A., Shimonishi, M., Nishizawa, T., Asami, O., Hagiya, M., Nakamura, T., and Shimizu, S. (1990) Biochem. Biophys. Res. Commun. 172, 321–327. 6. Shima, N., Nagao, M., Ogaki, F., Tsuda, E., Murakami, A., and Higashio, K. (1991) Biochem. Biophys. Res. Commun. 180, 1151– 1158. 7. Gohda, H., Tsubouchi, H., Nakayama, H., Hirano, S., Sakiyama,
25. 26. 27. 28. 29. 30. 31. 32. 33.
367
O., Takahashi, K., Miyazaki, H., Hashimoto, S., and Daikuhara, Y. (1988) J. Clin. Invest. 81, 414 – 419. Igawa, T., Kanda, S., Kanetake, H., Saitoh, Y., Ichihara, A., Tomita, Y., and Nakamura, T. (1991) Biochem. Biophys. Res. Commun. 174, 831– 838. Shima, N., Itagaki, Y., Nagao, M., Yasuda, H., Morinaga, T., and Higashio, K. (1991) Cell Biol. Int. Rep. 15, 397– 408. Gherardi, E., and Stoker, M. (1990) Nature 346, 228. Morimoto, A., Okamura, K., Hamanaka, R., Sato, Y., Shima, N., Higashio, K., and Kuwano, M. (1991) Biochem. Biophys. Res. Commun. 179, 1042–1049. Ishiki, Y., Ohnishi, H., Muto, Y., Matsumoto, K., and Nakamura, T. (1992) Hepatology 16, 1227–1235. Fujiwara, K., Nagoshi, S., Ohno, A., Hirata, K., Ohta, Y., Mochida, S., Tomiya, T., Higashio, K., and Kurokawa, K. (1993) Hepatology 18, 1443–1449. Masunaga, H., Fujise, N., Shiota, A., Yamashita, Y., Yasuda, H., and Higashio, K. (1996) J. Pharm. Pharmacol. 48, 876 – 879. Kawaida, K., Matsumoto, K., Shimazu, H., and Nakamura, T. (1994) Proc. Natl. Acad. Sci. USA 91, 4357– 4361. Amaike, H., Matsumoto, K., Oka, T., and Nakamura, T. (1996) Cytokine 8, 387–394. Masunaga, H., Fujise, N., Shiota, A., Ogawa, H. Sato, Y., Imai, E., Yasuda, H., and Higashio, K. (1998) Eur. J. Pharmacol. 342, 267–279. Matsumoto, K., Takehara, T., Inoue, H., Hagiya, M., Shimizu, S., and Nakamura, T. (1991) Biochem. Biophys. Res. Commun. 181, 691– 699. Shima, N., Tsuda, E., Goto, M., Yano, K., Hayasaka, H., Ueda, M., and Higashio, K. (1994) Biochem. Biophys. Res. Commun. 200, 808 – 815. Hartmann, G., Naldini, K., Weidner, K. M., Sachs, M., Vigna, E., Comoglio, P. M., and Birchmeier, W. (1992) Proc. Natl. Acad. Sci. USA 89, 11574 –11578. Lokker, N. A., Mark, M. R., Luis, E. A., Bennett, G. L., Robbins, K. A., Baker, J. B., and Godwski, P. J. (1992) EMBO J. 11, 2503–2510. Okigaki, M., Komada, M., Uehara, Y. Miyazawa, K., and Kitamura, N. (1992) Biochemistry 31, 9555–9561. Lokker, N. A., Presta, L. G., and Godowski, P. J. (1994) Protein Eng. 7, 895–903. Kinosaki, M., Yamaguchi, K., Murakami, A., Ueda, M., Morinaga, T., and Higashio, K. (1998) Biochim. Biophys. Acta 1384, 93–102. Kinosaki, M., Yamaguchi, K., Murakami, A., Morinaga, T., Ueda, M., and Higashio K. (1998) FEBS Lett. 434, 165–170. Sakata, S., and Atassi, M. Z. (1980) Mol. Immunol. 17, 139 –142. Yasuda, H., Imai, E., Shiota, A., Fujise, N., Morinaga, T., and Higashio, K. (1997) Hepatology 24, 636 – 642. Calanda, S., Martin, M. J., and McIntyre, N. (1971) Eur. J. Clin. Invest. 1, 352–360. Brown, M. S., and Goldstein, J. L. (1986) Science 232, 34 – 47. Prestayko, W. A., D’Aoust, C. J., Issel, F. B., and Grook, T. S. (1979) Cancer Treat. Rev. 6, 17–37. Matsumoto, A., and Yamamoto, N. (1991) Biochem. Biophys. Res. Commun. 174, 90 –95. Kato, Y., Liu, K. L., Nakamura, T., and Sugiyama, Y. (1994) Hepatology 20, 417– 424. Liu, K. L., Kato, Y., Narukawa, M., Kim, D. C., Hanano, M., Higuchi, O., Nakamura, T., and Sugiyama, Y. (1992) Am. J. Physiol. 263, 642– 649.
A Mutant of Deleted Variant of Hepatocyte Growth Factor (dHGF) with Alanine Substitution in the N-Terminal Basic Region Has Higher Activity in Vivo Masahiko Kinosaki, 1 Kyoji Yamaguchi, Yasushi Yamashita, Yasuhiro Uematsu, Hiromi Aihara, Hiroaki Masunaga, Tomonori Morinaga, and Kanji Higashio Research Institute of Life Science, Snow Brand Milk Products Co. Ltd., Ishibashi-machi, Shimotsuga-gun, Tochigi, Japan
Received November 20, 1998
In a previous study, we generated a mutant of dHGF (deleted variant of hepatocyte growth factor), termed #2, with higher specific activity than dHGF in assays of mitogenic activity on rat hepatocytes and America opossum kidney epithelial cells (OK). In the present study, we examine in vivo hepatotropic and renotropic activities of #2 and its distribution to target tissues, liver and kidney. Administration of #2 to normal rats significantly increased serum levels of total protein, albumin, free-cholesterol, and HDL-cholesterol and liver weight in a dose-dependent manner. Analysis of these parameters suggests that #2 is more potent than dHGF as a hepatotropic factor in vivo. In addition, #2 reduced mortality of mercuric chloride-administered mice and the effect was stronger than that of dHGF. When injected to mice, a larger amount of #2 than dHGF was rapidly distributed to the liver. Sixty minutes after injection, the concentrations of #2 in plasma, liver, and kidney were higher than those of dHGF. These distribution properties and the higher mitogenic activity in vitro may explain why #2 exerts more potent in vivo biological activity than dHGF. © 1999 Academic Press
Hepatocyte growth factor (HGF) (1, 2), also designated as scatter factor (3) or fibroblast-derived tumor cytotoxic factor (4), is a heparin-binding basic protein. The structural analysis based on the amino acid sequence deduced from the cDNA (1, 2) predicts that it consists of 6 major domains: the N-terminal domain 1 To whom correspondence should be addressed at Research Institute of Life Science, Snow Brand Milk Products Co. Ltd., 519 Ishibashi-machi, Shimotsuga-gun, Tochigi 329-0512, Japan. Fax: 81(285)53-1314. E-mail: [email protected]. Abbreviations used: HGF, hepatocyte growth factor; dHGF, deleted variant of hepatocyte growth factor; HDL, high-density lipoprotein; ELISA, enzyme-linked immunosorbent assay; PBS, phosphatebuffered saline; Chaps, 3-[(3-cholamidopropyl)-dimethylammonio]-1propanesulfate.
(including the N-terminal basic region and the hairpin loop structure) and four kringle domains in an a-chain, and a serine protease-like domain in a b-chain. In addition to intact-form HGF, an alternatively spliced variant, which lacks five consecutive amino acid residues (the FLPSS sequence) in the first kringle domain, was isolated and termed dHGF (deleted variant of HGF) (5, 6). Recent studies have revealed that both HGF and dHGF are biologically multifunctional (4, 7–11). They stimulate growth of various epithelial and endothelial cells, while they inhibit growth of tumor cells (4, 9). They act as hepatotropic factors and renotropic factors (12–17, and Masunaga, H. and Aihara, H. unpublished data). It has been reported that dHGF is distinguishable from HGF in biological and immunological properties. dHGF is more potent than HGF in stimulation of DNA synthesis in rat hepatocytes and epithelial cells (6, 18, 19). To study the structure-function relationship, a large number of HGF and dHGF mutants have been prepared (18, 20 –25). Analysis of the mutants demonstrated that some of the residues in the first kringle domain are critical for the biological activity (23, 25) and at least nine residues in the N-terminal region are involved in the binding to heparin (24). We showed that two dHGF mutants are more potent than dHGF in stimulating DNA synthesis in rat hepatocytes and America opossum kidney epithelial cells (OK) but less potent in mouse myeloblastic cells, NFS-60 (25). In the mutants, termed #2 and #27, all basic amino acid residues in the amino acid sequences, 2RKRR and 27KIKTKK, respectively, in the N-terminal basic region of dHGF are replaced with alanine. In this study, we examined in vivo biological potency of one of the mutants, #2. #2 was more potent in enhancing liver functions in normal rats and prevention of mortality caused by mercuric chloride-induced renal failure in mice. Further analysis revealed that #2 was
363
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
present at a higher level than dHGF in target tissues, liver and kidney, 60 min after intravenous injection. MATERIALS AND METHODS dHGF mutants. dHGF and the mutant, #2, were produced by culturing CHO cells transfected with expression vectors containing respective cDNA. They were purified, respectively, to homogeneity (purity: over 95%) from the conditioned media as previously described (25). Assay for hepatic functions. dHGF or #2 dissolved in PBS containing 0.01% Tween 80 was intravenously administered to male Wistar rats (6 weeks old) twice a day for 4 days at doses of 50, 100 and 200 mg/2 ml/kg body wt. Control rats received the same volume of the vehicle (PBS containing 0.01% Tween 80). Serum and plasma samples were collected from the rats. For evaluating both hepatic protein synthesis and cholesterol metabolism, serum levels of total protein, albumin, free-cholesterol and HDL-cholesterol were determined as previously described (14). To evaluate the relative activity of #2 to dHGF in vivo, parallel line assay was performed as previously described (26). Prevention of mercuric chloride-induced renal failure. dHGF, #2 or vehicle was intravenously administered to male ICR mice (body weight: 30 –35 g, n 5 20 per group) twice a day, total 9 times. Six hours after final administration, mercuric chloride (5 mg/kg body wt) was intravenously administered to the mice. Quantification of dHGF or #2 in plasma, liver, and kidney. dHGF or #2 was intravenously injected to mice. Under ether anesthesia, blood was collected from inferior vena cava of the mice 5, 30, or 60 min after the injection. At the same time point, liver was removed from the mice after perfusing it with saline and kidney was also taken out from the mice. The collected blood samples were centrifuged at 3,000 rpm for 10 min (RD-20IV, Tomy Seiko, Japan) and the serum samples were stored at 280°C until use in ELISA. The removed liver or kidney was homogenized in 4 volumes per weight of buffer (20 mM Tris–HCl buffer, pH 7.5, containing 2 M NaCl, 0.1% Tween 80, 1 mM phenylmethylsulfonyl fluoride and 1 mM EDTA). The homogenate was centrifuged at 19,000g for 30 min at 4°C and the supernatant was stored at 280°C until use in ELISA. dHGF or #2 in serum was quantified by a two step sandwich ELISA employing rabbit anti-dHGF polyclonal antibody as previously described (24) with the following modification. Each well in 96-well immunoplates (Maxisorp, Nunc, Denmark) was treated overnight with 100 ml of 10 mg/ml of anti-dHGF antibody in 0.1 M NaHCO 3 at room temperature and non-specific binding sites were blocked with 50% Block Ace (Snow Brand Milk Products, Co., Japan) in PBS at room temperature for 2 h. Each serum sample containing dHGF or #2 was diluted 2-fold with buffer A (50% Block Ace in 0.2 M Tris–HCl, pH 7.3, containing 0.2 M NaCl, 0.1% Tween 20, 0.2% Chaps, 20 mM benzamidine hydrochloride and 10 mM EDTA) and 100 ml of each of the diluted serum samples was added to each well in the plates. dHGF or #2 serially diluted in normal rat serum was used as a standard for ELISA. The plates were allowed to stand at 4°C overnight and subsequently the wells in the plates were washed four times with PBS containing 0.05% Tween 20. Peroxidaseconjugated anti-dHGF antibody diluted with buffer B (0.1 M phosphate buffer, pH 7.0, containing 10% Block Ace, 0.15 M NaCl, 0.1% Tween 20 and 4% rat serum) was added to each well (100 ml/well) and the plates were further incubated at 37°C for 2 h. The antibody bound to each well was quantified as previously described (24). To quantify dHGF or #2 in the supernatant of liver or kidney homogenate, ELISA system was partially modified as follows. The supernatant of each liver or kidney homogenate was diluted 2-fold with buffer A and 100 ml of the solution was added to each well in 96-well immumoplates coated with anti-dHGF polyclonal antibody. The plates were then incubated at room temperature for 2 h. dHGF
FIG. 1. Effect of dHGF and #2 on serum content of HDLcholesterol. dHGF (closed circles) or #2 (open circles) was administered intravenously to rats every 12 h for 4 days. HDL-cholesterol in the serum was determined as described under Materials and Methods. Data represent means 6 se (n 5 8). The parallel line assay was performed as previously described (26).
or #2 in each well was detected using peroxidase-conjugated antidHGF antibody as described above. To quantify dHGF and #2 in the supernatant of liver or kidney homogenate, dHGF and #2 serially diluted in the supernatant of liver or kidney homogenate derived from normal mice were used as standards for ELISA, respectively.
RESULTS Enhancement of hepatic functions and increase in liver weight in normal rats. Purified dHGF or #2 was administered to normal rats and serum levels of total protein, albumin, free-cholesterol and HDL-cholesterol, and liver weight of the rats were determined. Both dHGF and #2 enhanced hepatic functions such as protein synthesis and cholesterol metabolism and increased liver weight in a dose-dependent manner. To estimate the relative potency of #2 to dHGF in enhancing hepatic functions or increasing liver weight, the parallel line assay was used. A representative result of the analysis using a parameter, HDL-cholesterol, is shown in Fig. 1. The analysis showed a linearity in the dose–response relationship and the linearity established a parallelism between #2 and dHGF (Fig. 1). Similar results were obtained by analyzing other parameters except for liver weight (data not shown). The relative potency estimated by the use of each parameter was summarized in Table 1. The results revealed that #2 was approximately 1.5-fold more potent than dHGF in enhancing hepatic functions (Table 1). Further, #2 was significantly more potent (p , 0.05) than dHGF in increasing liver weight at a dose of 200 mg/ kg/day (Table 1). Protective effect on mercuric chloride-induced renal failure in mice. As described previously, #2 is more potent than dHGF in the stimulation of DNA synthesis
364
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Effects of dHGF and #2 on Liver Weight and Concentrations of Serum Proteins and Cholesterols Liver weight (g/100 g) Vehicle dHGF 50 mg/kg 100 mg/kg 200 mg/kg #2 50 mg/kg 100 mg/kg 200 mg/kg Relative potency of #2 a
Total protein (g/dL)
Albumin (g/dL)
Free-cholesterol (mg/dL)
HDL-cholesterol (mg/dL)
3.526 6 0.059
5.4 6 0.3
2.6 6 0.1
15 6 3
32.5 6 6.3
3.610 6 0.208 3.776 6 0.125 3.797 6 0.204
5.7 6 0.2 5.9 6 0.3 6.1 6 0.3
2.9 6 0.1 3.0 6 0.2 3.1 6 0.1
20 6 4 25 6 4 29 6 4
40.9 6 4.2 50.7 6 7.9 57.4 6 7.1
3.659 6 0.175 3.804 6 0.140 4.020 6 0.165
5.9 6 0.2* 6.2 6 0.2 6.5 6 0.1**
3.0 6 0.1 3.1 6 0.1 3.3 6 0.1*
23 6 5 26 6 6 35 6 6*
46.7 6 6.9 55.4 6 8.7 68.3 6 5.6**
Not paralleled
2.12
1.72
1.47
1.66
Note. dHGF or #2 was administered intravenously to rats every 12 h for 4 days. Data represent means 6 SD for 8 rats. Asterisk and double asterisk represent p , 0.05 and p , 0.01 vs rats receiving the same dose of dHGF by Student’s t test, respectively. a Relative potency of #2 to dHGF was calculated by parallel line assay.
in America opossum kidney epithelial cells (OK) (25). To evaluate the renotropic effect of #2 in comparison with dHGF, preventive effect of dHGF and #2 on mortality caused by mercuric chloride administration was examined. dHGF, #2 or vehicle was administered to mice (n 5 20 per group) prior to intravenous administration of a lethal dose of mercuric chloride. In vehicletreated group, no mice was alive 7 days after mercuric chloride administration. In contrast, 60 and 50% of the mice treated with #2 at doses of 50 mg/mouse/day and 25 mg/mouse/day, respectively, survived for 14 days after the administration (Fig. 2A). Higher doses of dHGF were required to obtain comparable effects with those by #2 (100 mg/mouse/day for 66% and 50 mg/ mouse/day for 50%) (Fig. 2B). These results indicate that #2 is approximately twofold more potent than dHGF in reduction of mortality caused by renal failure induced with mercuric chloride. Levels of dHGF and #2 in plasma, liver and kidney. ICR mice were given a single intravenous injection of dHGF or #2 at a dose of 25 mg/mouse. Five, 30, and 60 min after the injection, the plasma samples were collected and the concentration of dHGF or #2 in the plasma was determined by ELISA as described under Materials and Methods. The concentrations of dHGF and #2 in the plasma at 5 min were 572 6 239 and 305 6 163 ng/ml, respectively. The concentration of dHGF in the plasma decreased rapidly to 26.9 6 6.1 and 6.4060.47 ng/ml within 30 and 60 min, respectively, but that of #2 decreased slowly to 230 6 33 and 96.9 6 22.3 ng/ml (Fig. 3A). #2 was significantly higher than dHGF in hepatic concentration at each of the time points of 5, 30 and 60 min (Fig. 3B). The concentration of #2 in the kidney was much lower than that of dHGF at 5 min (810 6 215 ng/g vs 3674 6 704 ng/g). But beyond 5 min after the injection, #2 in the kidney
FIG. 2. Prevention of mercuric chloride-induced renal failure by dHGF and #2; (A) Survival rate of #2-preadministered mice. (B) Survival rate of dHGF-preadministered mice. Vehicle, dHGF or #2 was administered to mice twice a day. After final administration, mercuric chloride was given to the mice. Crosses, vehicle; square, 100 mg protein/mouse/day; circles, 50 mg protein/mouse/day; diamonds, 25 mg protein/mouse/day; and triangles, 12.5 mg protein/ mouse/day.
365
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Time profiles of dHGF or #2 in plasma (A), liver (B) and kidney (C) after intravenous administration; dHGF (gray bars) or #2 (black bars) was administered to mice. At indicated time point, samples of the plasma, the liver and the kidney were prepared from the mice as described under Materials and Methods. The concentration of dHGF or #2 in the samples was quantified by ELISA. Asterisk and tandem asterisk represent p , 0.05 and p , 0.01 vs group received same dose of dHGF by Student’s t test, respectively.
decreased only slightly. The concentration of #2 in the kidney at 60 min was higher than that of dHGF, which rapidly decreased in a time dependent manner (Fig. 3C). DISCUSSION As with HGF, dHGF is a potent mitogen for a variety of cell types such as hepatocytes, epithelial cells, endothelial cells and myeloblastic cells (6, 19). dHGF is, however, more potent than HGF in stimulation of DNA synthesis in rat hepatocytes (6, 18, 19), America opossum kidney epithelial cells (OK) and pig kidney epithelial cells, LLC-PK1 (19). In vivo, dHGF has been found to ameliorate disordered hepatic protein synthesis in rats with liver failure, and to exert preventive effects on various liver injuries and anti-fibrogenic effect on liver fibrosis in rats (14, 17, 27). In addition, dHGF has preventive effect on renal failure induced with mercuric chloride (Masunaga, and Aihara, unpublished data). In the previous study, we demonstrated that dHGF mutants, #2 and #27, which have alanine substitution in the N-terminal basic region, show a decrease (approximately 3-fold) in dose required for 50% maximal cell proliferative response (ED 50 ) compared with dHGF in stimulation of DNA synthesis in rat hepatocytes (25). The mutants are also more potent than dHGF in stimulation of DNA synthesis in America opossum kidney epithelial cells (OK) (25). In this study, we examined in vivo biological potency of one of the mutants, #2. After administration to normal rats, serum levels of total protein, albumin, free-
cholesterol and HDL-cholesterol, all of which reflect hepatic functions (28, 29), and liver weight were measured. To accurately estimate the relative activity of #2 to dHGF in vivo, the parallel line assay was applied. #2 was approximately 1.5-fold more potent than dHGF in enhancing hepatic functions, and was significantly higher than dHGF in increasing liver weight (Table 1). These results suggest that the higher mitogenic activity of #2 on rat hepatocytes in vitro is at least partly responsible for the more potent in vivo efficacy in the enhancement of hepatic functions and the increase in liver weight. Administration of nephrotoxic reagents, such as cyclosporin A and mercuric chloride, causes a renal failure. Such a renal failure is accompanied with destruction of renal tubular epithelial cells, leading to impaired glomerular filtration (30). HGF stimulates DNA synthesis in renal tubular cells after renal injuries, and prevents the onset of acute renal dysfunction (15). When administered prior to the induction of renal failure with mercuric chloride in mice, #2 was approximately 2-fold more potent than dHGF in reduction of the mortality (Fig. 2). The enhanced mitogenic activity of #2 on kidney epithelial cells in vitro is likely to be one of the reasons why it shows higher efficacy in the prevention of the renal failure than dHGF. The concentrations of #2 in the plasma and the kidney were lower than that of dHGF at a time point of 5 min after the injection, but decreased more slowly and became higher at 30 and 60 min, respectively (Fig. 3A and 3C). In contrast, the concentration of #2 in the liver was higher than that of dHGF at each of the time points after the injection (Fig. 3B). These observations
366
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
suggest that the higher hepatic level and the prolonged retention in kidney also contribute to more potent in vivo biological activity of #2 in these target tissues. Both dHGF and HGF have affinity for heparin, and bind to heparin-like molecules which exist on cell surface and in extracellular matrix (4, 31, 32). Because #2 is lower in affinity for heparin than dHGF (25), #2 may have tendency to stay longer in the circulation without being trapped by vascular cell surface. Liu et al. previously reported that the liver is the major clearance organ and the internalization of HGF takes place not only via the signaling receptors (c-Met), but also via the low-affinity binding sites, presumably cell surface heparin-like molecules (33). The reduced heparinbinding ability could also diminish the binding of #2 to cell surface heparin-like molecules eventually preventing internalization into hepatocytes. This may account for the observation that #2 was sustained at a higher level than dHGF in plasma, liver and kidney 60 min after the injection (Fig. 3). In conclusion, #2, a dHGF mutant with alanine substitution in the N-terminal domain, was found to be more potent than dHGF in stimulation of hepatic functions, increase in liver weight and prevention of renal failure induced with mercuric chloride. Of numerous HGF or dHGF mutants prepared so far, #2 is the first example with enhanced hepatotropic and renotropic activities both in vitro and in vivo. The results in this study suggest that #2 has advantages over dHGF for the treatment of patients with liver or renal diseases.
8.
9. 10. 11.
12. 13.
14. 15. 16. 17.
18.
19.
20.
21.
ACKNOWLEDGMENTS We thank Fumie Kobayashi, Toshiko Satake, Reiko Tadokoro, and Maki Okada for their excellent assistance.
REFERENCES
22. 23. 24.
1. Nakamura, T., Nishizawa, T., Hagiya, M., Seki, T., Shimonishi, M., Sugimura, A., Tashiro, K., and Shimizu, S. (1989) Nature 342, 440 – 443. 2. Miyazawa, K., Tsubouchi, H., Naka, D., Takahashi, K., Okigaki, M., Arakaki, N., Nakayama, H., Hirano, S., Sakiyama, O., Takahashi, K., Gohda, E., Daikuhara, Y., and Kitamura, N. (1989) Biochem. Biophys. Res. Commun. 163, 967–973. 3. Stoker, M. E., Gherardi, E., Perryman, M., and Gray, J. (1987) Nature 327, 239 –242. 4. Higashio, K., Shima, N., Goto, M., Itagaki, Y., Nagao, M., Yasuda, H., and Morinaga, T. (1990) Biochem. Biophys. Res. Commun. 170, 397– 404. 5. Seki, T., Ihara, J., Sugiyama, A., Shimonishi, M., Nishizawa, T., Asami, O., Hagiya, M., Nakamura, T., and Shimizu, S. (1990) Biochem. Biophys. Res. Commun. 172, 321–327. 6. Shima, N., Nagao, M., Ogaki, F., Tsuda, E., Murakami, A., and Higashio, K. (1991) Biochem. Biophys. Res. Commun. 180, 1151– 1158. 7. Gohda, H., Tsubouchi, H., Nakayama, H., Hirano, S., Sakiyama,
25. 26. 27. 28. 29. 30. 31. 32. 33.
367
O., Takahashi, K., Miyazaki, H., Hashimoto, S., and Daikuhara, Y. (1988) J. Clin. Invest. 81, 414 – 419. Igawa, T., Kanda, S., Kanetake, H., Saitoh, Y., Ichihara, A., Tomita, Y., and Nakamura, T. (1991) Biochem. Biophys. Res. Commun. 174, 831– 838. Shima, N., Itagaki, Y., Nagao, M., Yasuda, H., Morinaga, T., and Higashio, K. (1991) Cell Biol. Int. Rep. 15, 397– 408. Gherardi, E., and Stoker, M. (1990) Nature 346, 228. Morimoto, A., Okamura, K., Hamanaka, R., Sato, Y., Shima, N., Higashio, K., and Kuwano, M. (1991) Biochem. Biophys. Res. Commun. 179, 1042–1049. Ishiki, Y., Ohnishi, H., Muto, Y., Matsumoto, K., and Nakamura, T. (1992) Hepatology 16, 1227–1235. Fujiwara, K., Nagoshi, S., Ohno, A., Hirata, K., Ohta, Y., Mochida, S., Tomiya, T., Higashio, K., and Kurokawa, K. (1993) Hepatology 18, 1443–1449. Masunaga, H., Fujise, N., Shiota, A., Yamashita, Y., Yasuda, H., and Higashio, K. (1996) J. Pharm. Pharmacol. 48, 876 – 879. Kawaida, K., Matsumoto, K., Shimazu, H., and Nakamura, T. (1994) Proc. Natl. Acad. Sci. USA 91, 4357– 4361. Amaike, H., Matsumoto, K., Oka, T., and Nakamura, T. (1996) Cytokine 8, 387–394. Masunaga, H., Fujise, N., Shiota, A., Ogawa, H. Sato, Y., Imai, E., Yasuda, H., and Higashio, K. (1998) Eur. J. Pharmacol. 342, 267–279. Matsumoto, K., Takehara, T., Inoue, H., Hagiya, M., Shimizu, S., and Nakamura, T. (1991) Biochem. Biophys. Res. Commun. 181, 691– 699. Shima, N., Tsuda, E., Goto, M., Yano, K., Hayasaka, H., Ueda, M., and Higashio, K. (1994) Biochem. Biophys. Res. Commun. 200, 808 – 815. Hartmann, G., Naldini, K., Weidner, K. M., Sachs, M., Vigna, E., Comoglio, P. M., and Birchmeier, W. (1992) Proc. Natl. Acad. Sci. USA 89, 11574 –11578. Lokker, N. A., Mark, M. R., Luis, E. A., Bennett, G. L., Robbins, K. A., Baker, J. B., and Godwski, P. J. (1992) EMBO J. 11, 2503–2510. Okigaki, M., Komada, M., Uehara, Y. Miyazawa, K., and Kitamura, N. (1992) Biochemistry 31, 9555–9561. Lokker, N. A., Presta, L. G., and Godowski, P. J. (1994) Protein Eng. 7, 895–903. Kinosaki, M., Yamaguchi, K., Murakami, A., Ueda, M., Morinaga, T., and Higashio, K. (1998) Biochim. Biophys. Acta 1384, 93–102. Kinosaki, M., Yamaguchi, K., Murakami, A., Morinaga, T., Ueda, M., and Higashio K. (1998) FEBS Lett. 434, 165–170. Sakata, S., and Atassi, M. Z. (1980) Mol. Immunol. 17, 139 –142. Yasuda, H., Imai, E., Shiota, A., Fujise, N., Morinaga, T., and Higashio, K. (1997) Hepatology 24, 636 – 642. Calanda, S., Martin, M. J., and McIntyre, N. (1971) Eur. J. Clin. Invest. 1, 352–360. Brown, M. S., and Goldstein, J. L. (1986) Science 232, 34 – 47. Prestayko, W. A., D’Aoust, C. J., Issel, F. B., and Grook, T. S. (1979) Cancer Treat. Rev. 6, 17–37. Matsumoto, A., and Yamamoto, N. (1991) Biochem. Biophys. Res. Commun. 174, 90 –95. Kato, Y., Liu, K. L., Nakamura, T., and Sugiyama, Y. (1994) Hepatology 20, 417– 424. Liu, K. L., Kato, Y., Narukawa, M., Kim, D. C., Hanano, M., Higuchi, O., Nakamura, T., and Sugiyama, Y. (1992) Am. J. Physiol. 263, 642– 649.