Proc. Nati. Acad. Scl. USA Vol. 75, No. 9, pp. 4130-4134, September 1978
Biochemistry
Alterations of hepatic Na+,K+-ATPase and bile flow by estrogen: Effects on liver surface membrane lipid structure and function (membrane viscosity/Triton WR-1339/lipid composition)
ROGER A. DAVIS*t, FRED KERN, JR.t, RADENE SHOWALTERt, EILEEN SUTHERLANDt, MICHAEL SINENSKYt, AND FRANCIS R. SIMONt Deprtments of tMedicine (Gastroenterology) and *Biophysics, University of Colorado Medical School, Denver, Colorado 80262
Communicated by J. Edwin Seegmiller, May 30,1978
ABSTRACT Administration of the synthetic estrogen ethinyl estradiol (17a-ethinyl-1,3,5-estratriene-3,173diol) decreases hepatic Na+,K+-ATPase (ATP phosphohydrolase; EC 3.6.1.3) activity and bile flow to 50% and alters the composition and structure of surface membrane lipid in rats. Although the content of phospholipids was not changed by treatment, free cholesterol (130%) and cholesterol esters (400%) were increased in liver surface membrane fractions. These observations correlate with changes in membrane viscosity, as shown by electron spin resonance probes. Both rotational correlation time, using the isotropic probe methyl (12-nitroxyl)stearate, and the order parameter, determined by the anisotropic probe 5-nitroxylstearic acid, were significantly increased in liver surface membrane fractions from rats treated with ethinyl estradiol. Administration of Triton WR-1339, a nonionic detergent that corrects hepatic and serum lipid changes caused by ethinyl estradiol treatment, restored toward normal elevated membrane lipids and viscosity as well as Na+,K+-ATPase activity and bile flow. Although restoration of normal liver surface membrane structure and function may be due to reversal of abnormal lipid composition, detergents also may directly alter membrane enzyme activity. Addition of Triton WR-1339 in vitro increased Na+,K+-ATPase activity and reduced membrane viscosity of surface membranes from rats treated with ethinyl estradiol. Triton had no effect on either parameter in normal membrane preparations. Studies of membrane structure and function both in vivo and in vitro suggest that alterations in lipid composition may alter Na+,K+-ATPase function and bile flow.
Na+,K+-ATPase (ATP phosphohydrolase, EC 3.6.1.3) is a mammalian surface membrane that is sensitive to the lipid structure of the membrane bilayer (1-10). An important function of Na+,K+-ATPase in the hepatocyte may be the active secretion of sodium into the bile canaliculus, thus driving water across the canalicular membrane (8, 9). This fraction has been called bile salt-independent bile flow, and recent studies have demonstrated a strong correlation between hepatic Na+, K+-ATPase activity and bile flow (10), supporting the hypothesis that this component of bile flow is regulated by the sodium pump. One drug consistently shown to reduce bile salt-independent bile flow is the synthetic estrogen derivative ethinyl estradiol (17a-ethinyl-1,3,5-estratriene-3,17f3-diol) (10-13). The aim of the present study was to examine whether Na+, K+-ATPase activity is reduced after ethinyl estradiol treatment and, if so, what possible mechanisms might be involved. We recently found that ethinyl estradiol significantly increases hepatic cholesterol ester concentrations by activating hepatic microsomal cholesterol acyl-CoA transferase (14). The possibility that altered membrane lipid composition was involved in the mechanism through which ethinyl estradiol The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. § 1734 solely to indicate this fact.
alters hepatic function was therefore examined. We previously found that the inhibition of lipoprotein synthesis (14), bile acid synthesis (15), and bile flow (16) found in ethinyl estradioltreated rats were all restored to normal by administering Triton WR-1339 [an inhibitor of hepatic microsomal cholesterol acyl-CoA transferase (14)]. In the present study, once having demonstrated that ethinyl estradiol does, in fact, inhibit Na+, K+-ATPase activity, we examined if Triton WR-1339 would also restore the activity of this membrane enzyme. We then examined if ethinyl estradiol and its putative pharmacologic antagonist Triton WR-1339 act on surface membrane function by altering surface membrane structure.
MATERIALS AND METHODS Albumin (bovine), adenosine 5'-monophosphate, adenosine 5'-triphosphate, cytochrome c, ouabain, azide, and [ethylene bis(oxyethylene nitrilo)]tetraacetic acid (EGTA) were obtained from Sigma Chemical Company, St. Louis, MO, 3 W-hydroxysteroid dehydrogenase was from Worthington Biochemical Corp., Freehold, NJ. Ethinyl estradiol was a gift from Wyeth Laboratories, Fort Washington, PA, and Triton WR-1339 was a gift from Ray Fall, University of Colorado, Boulder, CO. Male Sprague-Dawley rats weighing 180-220 g (Charles Rivers, Wilmington, MA) had free access to Purina chow and water. Four treatments were used: control, ethinyl estradiol, ethinyl estradiol plus Triton WR-1339, and Triton WR-1339 alone. Control animals were handled similarly to treated animals but received propylene glycol, the vehicle for ethinyl estradiol (5 mg/ml in propylene glycol) at a dosage of 5 mg/kg per day for 5 days. Control and randomly selected treated rats were starved overnight and on the fourth day of treatment were injected intraperitoneally with Triton WR-1339 (62.5 mg/ml dissolved in 0.9% saline) so that each received 22.5 mg/100 g of body weight. Rats were killed by exsanguination 15 hr later. Bile flow was determined as described (16). Liver surface membrane fractions were isolated according to Neville (17) through step 12 as described by Pohl et al. (18). This surface membrane fraction is enriched in bile canalicular fronts, but also contains sinusoidal surface membranes (12, 19). Microsomal fractions were prepared from the supernate of the liver surface membrane nuclear pellet by a modification of the procedure of Evans and Gurd (20), previously described (21). Microsomal fractions were washed by the method of Weihing et al. (22) to remove adsorbed and intracisternal proteins. Abbreviations: ethinyl estradiol, 17a-ethinyl-1,3,5-estratriene3,17,B-diol; ESR, electron spin resonance. * To whom reprint requests should be sent at present address: Department of Medicine M-013D, Division of Metabolic Disease, University of California, San Diego, La Jolla, CA 92093. 4130
Biochemistry:
Davis et al.
Proc. Natl. Acad. Sci. USA 75 (1978)
Na+,K+-ATPase activity was determined in liver surface membrane fractions and homogenates according to the method of Ismail-Beigi and Edelman with minor modifications (12,23). A detailed account of the procedure used has been reported (9). Under the conditions used hepatic Na+,K+-ATPase activity is linear to protein concentrations of 500 ,ug and for 7.5 min. Enzyme activity is expressed as micromoles of inorganic phosphate released per milligram of protein per hour. Glucose-6-phosphatase (24) and cytochromes P-450 and b5 (25) were determined in liver surface membrane fractions as described. Proteins were determined by the method of Lowry et al. (26), with bovine serum albumin as standard. After extraction (27) of the membrane preparations, free and esterified cholesterol were quantitated by gas-liquid chromatography as described (16). A portion of the CHC13 extracts was used for phospholipid (28) analysis. Electron Spin Resonance (ESR) Spin Label Probe Studies. Rotational correlation times of the isotropically moving probe, methyl (12-nitroxyl)stearate (Syva Corp., Palo Alto, CA) were determined in lipid films at 37"C essentially as described (29, 30). In a previous study it was found that r correlates with true viscosity (31). In all cases, the mass ratio of lipid to probe exceeded 40:1. Order parameters of the anisotropically moving probe 5nitroxylstearic acid (Syva Corp., Palo Alto, CA) were determined in membrane preparations at 370C by the method of Hubbell and McConnell (31). The probe molecule was incorporated into membrane fractions by drying a hexane solution of probe under nitrogen in a test tube, adding the membrane suspension in buffer, and stirring vigorously for 2 min. RESULTS Effect of Ethinyl Estradiol on Hepatic Na+,K+-ATPase Activity. Administration of ethinyl estradiol to rats daily for 5 days reduced Na+,K+-ATPase activity 50% (P < 0.001) in both liver homogenates and surface membrane fractions (Fig. 1). Similar reductions in activity were observed when either potassium or sodium was removed from the assay, indicating that ouabain inhibition of total ATPase activity measures enzyme activity. Effect of Ethinyl Estradiol and Triton VR-1339 on Lipid Composition of Liver Surface Membrane Fraction. Since ethinyl estradiol treatment alters hepatic and serum lipids (14) and Na+,K+-ATPase is a lipid-dependent enzyme, the composition of liver surface membranes was determined. Although 3 4-0
60
a 0z
.
40
i0E +l -
as0
6
C
EE
C
EE
FIG. 1. Na+, K+-ATPase activity in (Left) liver homogenate and (Right) surface membrane fractions from control (C) and ethinyl estradiol-treated (EE) rats. Values (mean ± 1 SEM) represent the mean of six animals in each group. Statistically significant differences are shown between EE (ethinyl estradiol) and C (control) values (P < 0.001).
4131
total phospholipid concentrations were unaltered, administration of ethinyl estradiol increased the concentrations of free cholesterol (1.3-fold) and cholesterol esters (4-fold) compared to the control (Table 1). Reflecting the change in cholesterol, the molar ratio of cholesterol to phospholipid was significantly increased. Although cholesterol esters represent less than 3% of the total membrane lipids, their presence in surface membrane preparations was a consistent finding. Administration of Triton WR-1339 restores serum and hepatic lipid changes produced by ethinyl estradiol treatment (14, 16). In liver surface membrane fractions Triton also returns the elevated levels of free and esterified cholesterol to control levels (Table 1). In contrast, Triton WR-1339 did not significantly change lipid composition in rats not treated with ethinyl estradiol. To examine the possibility that changes in Na+,K+-ATPase activity and lipid composition result from altered recovery of these components in ethinyl estradiol-treated animals, we quantitated the distribution of protein, enzyme markers and lipids. Ethinyl estradiol did not alter recovery of protein in liver surface membrane fractions or the relative enrichment of Na+,K+-ATPase. Microsomal enzymes cytochromes P-450 and b5 and glucose-6-phosphatase were not detected in any of the individual surface membrane functions, indicating minimal microsomal contamination. In addition, increased cholesterol in the surface membranes of ethinyl estradiol-treated rats was not due to adherent lipoproteins since no serum lipoprotein contaminants were detected by either double immunodiffusion or immunoelectrophoresis. Effect of Ethinyl Estradiol and Triton VR-1339 on Rotational Correlation Times (r) and Molecular Ordering Parameter (s) in Surface Membrane Fractions. Two independent ESR spin probe methods were used to determine the possible alterations of surface membrane structure caused by the different treatments. Rotational correlation time (r) was measured in lipid extracts of surface membrane fractions with the spin label, methyl (12-nitroxyl)stearate, which preferentially associates with the fluid domains of the membrane (32). Molecular ordering parameter (s) was measured with the oriented spin label (5-nitroxylstearic acid) in total membrane fractions (31). The spectra from both probes demonstrated immobilization in the membrane preparations of ethinyl estradioltreated rats. Both T (60%, P < 0.005) and s (5%, P < 0.005) were increased in surface membranes from rats treated with ethinyl estradiol compared to controls (Fig. 2). When administered to ethinyl estradiol-treated rats, Triton WR-1339 restored r and s values to values not significantly different from those of the control. Since the ESR data were obtained at 37°C, it is likely that the changes in membrane fluidity represent significant differences at physiologic temperature. Effect of Triton WR-1339 on Na+,K+-ATPase Activity and Basal Bile Flow. If decreased liver surface membrane functions result from altered viscosity, it should follow that restoration of membrane lipid composition and viscosity to normal should also be associated with correction of abnormal Na+,K+-ATPase and bile flow. As shown in Table 2, administration of Triton WR-1339 restored hepatic Na+,K+-ATPase activity and bile flow to the control levels. This effect is limited to ethinyl estradiol-treated animals, for Triton had no effect on untreated rats. Certain detergents increase Na+,K+-ATPase activity in vitro (33). Therefore, it seemed possible that correction of reduced Na+,K+-ATPase activity by Triton WR-1339 may result from a direct action of the detergent on the membrane. To examine this possibility, we added Triton WR-1339 in various concentrations to membranes from untreated and ethinyl estradiol-
4132
Biochemistry:
Davis et al.
Proc. Natl. Acad. Sci. USA 75 (1978) Table 1. Lipid composition of surface membrane fractions
Molar ratio,
cholesterol/
jug of lipid/mg of membrane protein Treatment group Control Ethinyl estradiol Ethinyl estradiol + Triton WR-1339 Control + Triton WR-1339 *
Phospholipid
Cholesterol
Cholesterol ester
phospholipid
360 + 51 380 + 18
90 ± 8 4* 120 V
4+1 , 16 ± 3*
0.48 i 0.04 0.61 + 0.03*
383 + 10
100 i 6
7I2
0.51 ± 0.03
352 1 33
80 1 6
6±1
0.44 I 0.04
Expressed as the mean ± SEM. Number of animals in each group = 6.
Significant difference from control values at P < 0.05.
treated rats. Addition of Triton WR-1339 in vitro increased activity of Na+,K+-ATPase as a function of the amount of detergent added to membrane fractions obtained from ethinyl estradiol-treated rats until a maximal value of activity was obtained that was not significantly different from control (Fig. 3). In contrast, addition of Triton WR-1339 to surface membrane fractions from untreated rats did not change enzyme activity. Activation was not due to solubilization, for Na+,K+ATPase activity was recovered only in the pellet after centrifugation (104,000 X g for 30 min) at each concentration of detergent. In addition, at the concentrations used neither protein
membrane lipids were selectively released from the membrane fractions, nor did Triton WR-1339 affect the inorganic phosphorus determination. In further experiments, s was determined in duplicate samples of each assay. Addition of Triton WR-1339 (6.25 ,ug) to surface membrane fractions of ethinyl estradiol-treated rats decreased s from 0.730 + 0.006 to the control value of 0.700 + 0.006. Additional Triton WR-1339 did not alter s further. Furthermore, Triton WR-1339 added to surface membrane preparations from untreated rats changed neither s nor Na+,K+-ATPase activity.
nor
3
DISCUSSION Current views of membrane structure suggest their essential feature is a bilayer containing a polar shell consisting primarily of phospholipids and free cholesterol. This lipid environment significantly influences the activity of several membrane enzymes (34-38). The molecular mechanisms by which the lipid composition of the membrane bilayer influences enzymes probably involves protein conformational changes due to the physical state of the immediate or boundary lipids and/or restrictions of molecular motion necessary for formation of the transition state. Lipid viscosity is apparently rate determining for the membrane-bound enzymes, polyisoprenoid alcohol phosphokinase (29) and cytochrome b5 (37). Although the molecular mechanisms for Na+,K+-ATPase function are not fully elucidated, Kyte (3) suggests that conformational changes are required both for exposing cation sites and for forming a channel through the membrane bilayer through which cations are transported. Previous studies (7, 36, 39) also suggest an important role for membrane viscosity in controlling the activity of Na+,K+-ATPase in that there is a good correlation between membrane lipid phase transitions and Arrhenius activation energies. Indeed, Kimelberg and Papahadjopoulis have speculated that membrane cholesterol accumulation may impair Na+,K+-ATPase activity through changes in membrane structure (7). The results of the present study support these previous studies and suggest that the synthetic estrogen, ethinyl estradiol, decreases hepatic Na+,K+-ATPase activity through changes in surface membrane lipid structure.
C
z;
Na+,K+-ATPase activity in ethinyl estradiol-treated rats is reduced to 50% of control in liver surface membrane fractions. Since the surface membrane from hepatocytes is a mosaic of three distinct surfaces, and Na+,K+-ATPase may be preferentially localized to one site, it is possible that ethinyl estradiol treatment alters enzyme activity by changing the relative re-
0. 7
E
00
0
0.65,
0~~~~~~~~
Table 2. Effect of ethinyl estradiol and Triton WR-1339 on hepatic Na+,K+-ATPase and basal bile flow
0
Treatment
0
0.6
C
EE
EE+
C+
Triton Triton
C
EE
EE+
C+
Triton Triton
FIG. 2. ESR analysis of surface membrane structure. (Left) Rotational correlation time (r); (Right) order parameter (s). Measurements were determined from the paramagnetic spectra as described in text. Results are expression of duplicate determinations from three different animals in each treatment group. C = control (propylene glycol); EE = ethinyl estradiol, EE + Triton = ethinyl estradiol + Triton WR-1339; C + Triton = untreated + Triton Wr1339. Values are expressed as the mean + SEM. (*) P < 0.005, compared to control determinations.
Na+,K+-ATPase*
Basal bile flowt 7.6 ± 0.2 5.4 ± 0.6t
Control 44.3 i 3.0 21.9 ± 2.0t Ethinyl estradiol Ethinyl estradiol + Triton 48.9 ± 1.6 7.8 + 0.4 Triton alone 48.3 ± 2.0 8.0 ± 0.6 Values are expressed as mean + SEM for duplicate determinations in six different animals in each group. Na+,K+-ATPase was measured in liver surface membrane fractions as described in text. * tmol Pi/hr per mg of membrane protein. t ,l/min per 100 g of body weight. Pp < 0.001.
Biochemistry:
10
Proc. Natl. Acad. Sci. USA 75 (1978)
Davis et al.
100 Triton added, ,g
50
130
625
FIG. 3. Effect of adding Triton WR-1339 to surface membrane fractions on Na+,K+-ATPase activity in vitro. Triton WR-1339 in 0.9% saline was added at various concentrations to surface membrane fractions (190 yg of protein per tube) of control (O--- -0) and ethinyl estradiol-treated (0-0) rats. Each point represents the mean of duplicate determinations of Na+,K+-ATPase. Duplicate values agreed within 10% of each other. Lines were drawn by the method of least squares.
covery of the membrane fraction. This possibility is unlikely, for recovery of surface membrane protein is unaltered and the relative enrichment of Na+,K+-ATPase was equal for all groups
examined. We conclude that the in vitro changes in surface membrane lipid composition, lipid structure, and Na+,K+ATPase reflect the in vivo surface membrane, a conclusion supported by recent studies of Reichen and Paumgartner (40). This suggestion is supported by restoration to normal of surface membrane lipid composition and structure, Na+,K+-ATPase activity, and bile flow in ethinyl estradiol-treated rats given Triton WR-1339.
Reduced Na+,K+-ATPase activity may also result from decreased function of a fixed number of enzyme units or may be caused by a decreased number of enzyme units. Since in both isolated cell membranes (Fig. 3) and in vivo (Fig. 1) Triton WR-1339 increased the Na+,K+-ATPase activity of ethinyl estradiol-treated rats to control levels and not further, and since there was no effect in untreated rats, the number of enzyme units of Na+,K+-ATPase is probably not changed by ethinyl estradiol treatment. Furthermore, Triton WR-1339 activation of Na+,K+-ATPase in ethinyl estradiol-treated rats was not inhibited by cycloheximide, indicating that de novo protein synthesis is not required (unpublished observations). In addition, preliminary studies with [3H]ouabain and [y-32P]ATP binding to surface membrane fractions indicate no change in the number of enzyme units in liver surface membrane fractions from ethinyl estradiol-treated rats (unpublished observations). These observations are similar to those reported in studies of erythrocyte membranes from cholesterol-treated guinea pigs (40, 41), in which Na+,K+-ATPase activity is reduced but the number of [3H]ouabain binding sites is unaltered. The findings of significant increases in both s and y in liver surface membrane fractions indicate that ethinyl estradiol alters the lipid structure of this membrane fraction by decreasing its fluidity. The exact mechanism by which ethinyl estradiol alters the structure of liver surface membrane lipids is not settled by the present experiments. However, in model membrane studies, Davis and Sinensky (42) have shown that increased concentrations of esterified cholesterol, similar to those found in surface membrane fractions from ethinyl estradiol-treated rats, cause increased lipid viscosity. Triton may reduce membrane viscosity and increase
4133
Na+,K+-ATPase activity through two possible mechanisms in ethinyl estradiol-treated rats. First, Triton inhibits the activation of hepatic microsomal cholesterol acyl-CoA transferase caused by ethinyl estradiol and restores hepatic lipid composition to normal (14). Second, Triton may directly affect the structure and function of abnormal membranes, possibly by its inseftion into the lipid bilayer. Whatever the mechanism, these data suggest that increased membrane viscosity caused by ethinyl estradiol treatment is reduced to normal by Triton WR-1339 through its effect on the surface membrane structure. Changes in membrane structure are closely correlated with sodium pump activity and bile flow, suggesting that the physical state of membrane lipids plays an important role in regulating transport across membranes. These observations may explain the reductions in organic anion transport previously reported for ethinyl estradiol treatment in human beings and rats (43). We thank Patricia Coan for her excellent technical assistance and Drs. D. Steinberg, I. M. Arias, and J. C. Williams for their critical review of the manuscript. This work was supported by Grants AM 15851 and AM 12626 from the National Institutes of Health and by American Cancer Society Grant BC 219. 1. Skou, J. C. (1965) Physiol. Rev. 45,596-617. 2. Uesugi, S., Dulak, N. C., Dixon, J. F., Hexum, T. D., Dahl, J. L., Perdue, J. F. & Hokin, L. E. (1971) J. Biol. Chem. 246, 531543. 3. Kyte, J. (1975) J. Biol. Chem. 250, 7443-7449. 4. Keith, A. D., Sharnoff, M. & Cohn, G. E. (1973) Biochim. Biophys. Acta 300, 379-419. 5. Hokin, L. E. & Hexum, J. D. (1972) Arch. Biochem. Biophys. 151, 453-463. 6. Tanaka, R. & Strickland, K. P. (1965) Arch. Biochem. Biophys. 111,583-592. 7. Kimelberg, H. K. & Papahadjopoulis, D. (1974) J. Biol. Chem. 249, 1071-1080. 8. Erlinger, S., Dhumeaux, D. & Benhamou, J. P. (1969) Nature (London) 223, 1276-1277. 9. Boyer, J. L. (1971) Am. J. Physiol. 221, 1156-1163. 10. Simon, F. R., Sutherland, E. & Accatino, L. (1977) J. Clin. Invest. 59, 849-861. 11. Davis, R. A. & Kern, F., Jr. (1975) Gastroenterology 70, 1130-1135. 12. Simon, F. R. & Arias, I. M. (1973) J. Clin. Invest. 52, 765775. 13. Gumucio, J. J. & Valdevieso, V. D. (1971) Gastroentrology 61, 339-344. 14. Davis, R. A., Showalter, R. & Kern, F., Jr. (1978) Biochem. J. 174, 45-51. 15. Davis, R. A. & Kern, F. (1977) Gastroenterology 72, 1045. 16. Davis, R. A. & Kern, F., Jr. (1977) in Bile Acid Metabolism in Health and Disease, eds. Paumgartner, G. & Stiehl, A. (MTP Press Ltd., London), pp. 25-33. 17. Neville, D. M., Jr. (1968) Biochim. Biophys. Acta 154, 540552. 18. Pohl, S. L., Bimbaumer, R. & Rodbell, M. (1971) J. Biol. Chem. 246, 1849-1856. 19. Neville, D. M. (1975) in Methods in Membrane Biology, ed. Korn, I. D. (Plenum, New York), Vol. 3, pp. 1-49. 20. Evans, W. H. & Gurd, J. W. (1971) Biochem. J. 125,615-624. 21. Accatino, L. & Simon, F. R. (1976) J. Clin. Invest. 57, 496508. 22. Weihing, R. R., Manganiello, V. C., Chin, R. & Phillips, A. H. (1972) Biochemistry 11, 3128-3135. 23. Ismail-Beigi, F. & Edelman, I. D. (1971) J. Gen. Physiol. 57,
710-722.
24. de Duve, C., Pressman, B. C., Gianetto, R., Wattiaux, R. & Appelmans, F. (1955) Biochem. J. 60, 604-617. 25. Ojnura, T. & Sato, R. (1964) J. Biol. Chem. 239, 2370-2378. 26. Lowry, 0. H., Rosebrough, N. D., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, 265-275.
4134
Biochemistry: Davis et al.
27. Folch, J., Lees, M. & Sloan-Stanley, G. H. (1957) J. Biol. Chem. 226,497-509. 28. Bartlett, G. R. (1959) J. Biol. Chem. 234, 466-470. 29. Sinensky, M. (1974) Proc. Natl. Acad. Sci. USA 71,522-525. 30. Gennis, R. B., Sinensky, M. & Strominger, J. L. (1976) J. Biol. Chem. 251, 1270-1276. 31. Hubbell, W. L. & McConnell, H. M. (1971) J. Am. Chem. Soc. 93,314-329. 32. Oldfield, K., Keough, M. & Chapman, D. (1972) FEBS Lett. 20, 344-346. 33. Schwartz, A., Lindenmayer, G. E. & Allen, J. C. (1976) Pharmacol. Rev. 27,3-134. 34. Coleman, R. (1973) Biochim. Biophys. Acta 300, 1-30. 35. Farias, R. N., Bloj, B., Morero, R. D., Sifieriz, F. & Trucco, R. E. (1975) Biochim. Biophys. Acta 415,231-251.
Proc. Nati. Acad. Sci. USA 75 (1978) 36. Kimelberg, H. K. (1976) Molec. Cell Biochem. 10, 171-190. 37. Rogers, M. J. & Strittmatter, P. (1974) J. Biol. Chem. 249, 895-900. 38. Solomonson, L. P., Liepkains, V. A. & Spector, A. A. (1976) Biochemistry 15, 892-897. 39. Kroes, J. & Ostwald, R. (1971) Biochim. Biophys. Acta 249, 647-650. 40. Reichen, J. & Paumgartner, G. (1977) J. Clin. Invest. 52,429434. 41. Kroes, J., Ostwald, R. & Keith, A. (1972) Biochim. Biophys. Acta
274,71-74. 42. Davis, R. A. & Sinensky, M. (1977) Fed. Proc. Fed. Am. Soc. Exp. Biol. 37, 2290. 43. Adlercreutz, H. & Tenhunen, R. (1970) Am. J. Med. 49, 630648.
Biochemistry
Alterations of hepatic Na+,K+-ATPase and bile flow by estrogen: Effects on liver surface membrane lipid structure and function (membrane viscosity/Triton WR-1339/lipid composition)
ROGER A. DAVIS*t, FRED KERN, JR.t, RADENE SHOWALTERt, EILEEN SUTHERLANDt, MICHAEL SINENSKYt, AND FRANCIS R. SIMONt Deprtments of tMedicine (Gastroenterology) and *Biophysics, University of Colorado Medical School, Denver, Colorado 80262
Communicated by J. Edwin Seegmiller, May 30,1978
ABSTRACT Administration of the synthetic estrogen ethinyl estradiol (17a-ethinyl-1,3,5-estratriene-3,173diol) decreases hepatic Na+,K+-ATPase (ATP phosphohydrolase; EC 3.6.1.3) activity and bile flow to 50% and alters the composition and structure of surface membrane lipid in rats. Although the content of phospholipids was not changed by treatment, free cholesterol (130%) and cholesterol esters (400%) were increased in liver surface membrane fractions. These observations correlate with changes in membrane viscosity, as shown by electron spin resonance probes. Both rotational correlation time, using the isotropic probe methyl (12-nitroxyl)stearate, and the order parameter, determined by the anisotropic probe 5-nitroxylstearic acid, were significantly increased in liver surface membrane fractions from rats treated with ethinyl estradiol. Administration of Triton WR-1339, a nonionic detergent that corrects hepatic and serum lipid changes caused by ethinyl estradiol treatment, restored toward normal elevated membrane lipids and viscosity as well as Na+,K+-ATPase activity and bile flow. Although restoration of normal liver surface membrane structure and function may be due to reversal of abnormal lipid composition, detergents also may directly alter membrane enzyme activity. Addition of Triton WR-1339 in vitro increased Na+,K+-ATPase activity and reduced membrane viscosity of surface membranes from rats treated with ethinyl estradiol. Triton had no effect on either parameter in normal membrane preparations. Studies of membrane structure and function both in vivo and in vitro suggest that alterations in lipid composition may alter Na+,K+-ATPase function and bile flow.
Na+,K+-ATPase (ATP phosphohydrolase, EC 3.6.1.3) is a mammalian surface membrane that is sensitive to the lipid structure of the membrane bilayer (1-10). An important function of Na+,K+-ATPase in the hepatocyte may be the active secretion of sodium into the bile canaliculus, thus driving water across the canalicular membrane (8, 9). This fraction has been called bile salt-independent bile flow, and recent studies have demonstrated a strong correlation between hepatic Na+, K+-ATPase activity and bile flow (10), supporting the hypothesis that this component of bile flow is regulated by the sodium pump. One drug consistently shown to reduce bile salt-independent bile flow is the synthetic estrogen derivative ethinyl estradiol (17a-ethinyl-1,3,5-estratriene-3,17f3-diol) (10-13). The aim of the present study was to examine whether Na+, K+-ATPase activity is reduced after ethinyl estradiol treatment and, if so, what possible mechanisms might be involved. We recently found that ethinyl estradiol significantly increases hepatic cholesterol ester concentrations by activating hepatic microsomal cholesterol acyl-CoA transferase (14). The possibility that altered membrane lipid composition was involved in the mechanism through which ethinyl estradiol The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. § 1734 solely to indicate this fact.
alters hepatic function was therefore examined. We previously found that the inhibition of lipoprotein synthesis (14), bile acid synthesis (15), and bile flow (16) found in ethinyl estradioltreated rats were all restored to normal by administering Triton WR-1339 [an inhibitor of hepatic microsomal cholesterol acyl-CoA transferase (14)]. In the present study, once having demonstrated that ethinyl estradiol does, in fact, inhibit Na+, K+-ATPase activity, we examined if Triton WR-1339 would also restore the activity of this membrane enzyme. We then examined if ethinyl estradiol and its putative pharmacologic antagonist Triton WR-1339 act on surface membrane function by altering surface membrane structure.
MATERIALS AND METHODS Albumin (bovine), adenosine 5'-monophosphate, adenosine 5'-triphosphate, cytochrome c, ouabain, azide, and [ethylene bis(oxyethylene nitrilo)]tetraacetic acid (EGTA) were obtained from Sigma Chemical Company, St. Louis, MO, 3 W-hydroxysteroid dehydrogenase was from Worthington Biochemical Corp., Freehold, NJ. Ethinyl estradiol was a gift from Wyeth Laboratories, Fort Washington, PA, and Triton WR-1339 was a gift from Ray Fall, University of Colorado, Boulder, CO. Male Sprague-Dawley rats weighing 180-220 g (Charles Rivers, Wilmington, MA) had free access to Purina chow and water. Four treatments were used: control, ethinyl estradiol, ethinyl estradiol plus Triton WR-1339, and Triton WR-1339 alone. Control animals were handled similarly to treated animals but received propylene glycol, the vehicle for ethinyl estradiol (5 mg/ml in propylene glycol) at a dosage of 5 mg/kg per day for 5 days. Control and randomly selected treated rats were starved overnight and on the fourth day of treatment were injected intraperitoneally with Triton WR-1339 (62.5 mg/ml dissolved in 0.9% saline) so that each received 22.5 mg/100 g of body weight. Rats were killed by exsanguination 15 hr later. Bile flow was determined as described (16). Liver surface membrane fractions were isolated according to Neville (17) through step 12 as described by Pohl et al. (18). This surface membrane fraction is enriched in bile canalicular fronts, but also contains sinusoidal surface membranes (12, 19). Microsomal fractions were prepared from the supernate of the liver surface membrane nuclear pellet by a modification of the procedure of Evans and Gurd (20), previously described (21). Microsomal fractions were washed by the method of Weihing et al. (22) to remove adsorbed and intracisternal proteins. Abbreviations: ethinyl estradiol, 17a-ethinyl-1,3,5-estratriene3,17,B-diol; ESR, electron spin resonance. * To whom reprint requests should be sent at present address: Department of Medicine M-013D, Division of Metabolic Disease, University of California, San Diego, La Jolla, CA 92093. 4130
Biochemistry:
Davis et al.
Proc. Natl. Acad. Sci. USA 75 (1978)
Na+,K+-ATPase activity was determined in liver surface membrane fractions and homogenates according to the method of Ismail-Beigi and Edelman with minor modifications (12,23). A detailed account of the procedure used has been reported (9). Under the conditions used hepatic Na+,K+-ATPase activity is linear to protein concentrations of 500 ,ug and for 7.5 min. Enzyme activity is expressed as micromoles of inorganic phosphate released per milligram of protein per hour. Glucose-6-phosphatase (24) and cytochromes P-450 and b5 (25) were determined in liver surface membrane fractions as described. Proteins were determined by the method of Lowry et al. (26), with bovine serum albumin as standard. After extraction (27) of the membrane preparations, free and esterified cholesterol were quantitated by gas-liquid chromatography as described (16). A portion of the CHC13 extracts was used for phospholipid (28) analysis. Electron Spin Resonance (ESR) Spin Label Probe Studies. Rotational correlation times of the isotropically moving probe, methyl (12-nitroxyl)stearate (Syva Corp., Palo Alto, CA) were determined in lipid films at 37"C essentially as described (29, 30). In a previous study it was found that r correlates with true viscosity (31). In all cases, the mass ratio of lipid to probe exceeded 40:1. Order parameters of the anisotropically moving probe 5nitroxylstearic acid (Syva Corp., Palo Alto, CA) were determined in membrane preparations at 370C by the method of Hubbell and McConnell (31). The probe molecule was incorporated into membrane fractions by drying a hexane solution of probe under nitrogen in a test tube, adding the membrane suspension in buffer, and stirring vigorously for 2 min. RESULTS Effect of Ethinyl Estradiol on Hepatic Na+,K+-ATPase Activity. Administration of ethinyl estradiol to rats daily for 5 days reduced Na+,K+-ATPase activity 50% (P < 0.001) in both liver homogenates and surface membrane fractions (Fig. 1). Similar reductions in activity were observed when either potassium or sodium was removed from the assay, indicating that ouabain inhibition of total ATPase activity measures enzyme activity. Effect of Ethinyl Estradiol and Triton VR-1339 on Lipid Composition of Liver Surface Membrane Fraction. Since ethinyl estradiol treatment alters hepatic and serum lipids (14) and Na+,K+-ATPase is a lipid-dependent enzyme, the composition of liver surface membranes was determined. Although 3 4-0
60
a 0z
.
40
i0E +l -
as0
6
C
EE
C
EE
FIG. 1. Na+, K+-ATPase activity in (Left) liver homogenate and (Right) surface membrane fractions from control (C) and ethinyl estradiol-treated (EE) rats. Values (mean ± 1 SEM) represent the mean of six animals in each group. Statistically significant differences are shown between EE (ethinyl estradiol) and C (control) values (P < 0.001).
4131
total phospholipid concentrations were unaltered, administration of ethinyl estradiol increased the concentrations of free cholesterol (1.3-fold) and cholesterol esters (4-fold) compared to the control (Table 1). Reflecting the change in cholesterol, the molar ratio of cholesterol to phospholipid was significantly increased. Although cholesterol esters represent less than 3% of the total membrane lipids, their presence in surface membrane preparations was a consistent finding. Administration of Triton WR-1339 restores serum and hepatic lipid changes produced by ethinyl estradiol treatment (14, 16). In liver surface membrane fractions Triton also returns the elevated levels of free and esterified cholesterol to control levels (Table 1). In contrast, Triton WR-1339 did not significantly change lipid composition in rats not treated with ethinyl estradiol. To examine the possibility that changes in Na+,K+-ATPase activity and lipid composition result from altered recovery of these components in ethinyl estradiol-treated animals, we quantitated the distribution of protein, enzyme markers and lipids. Ethinyl estradiol did not alter recovery of protein in liver surface membrane fractions or the relative enrichment of Na+,K+-ATPase. Microsomal enzymes cytochromes P-450 and b5 and glucose-6-phosphatase were not detected in any of the individual surface membrane functions, indicating minimal microsomal contamination. In addition, increased cholesterol in the surface membranes of ethinyl estradiol-treated rats was not due to adherent lipoproteins since no serum lipoprotein contaminants were detected by either double immunodiffusion or immunoelectrophoresis. Effect of Ethinyl Estradiol and Triton VR-1339 on Rotational Correlation Times (r) and Molecular Ordering Parameter (s) in Surface Membrane Fractions. Two independent ESR spin probe methods were used to determine the possible alterations of surface membrane structure caused by the different treatments. Rotational correlation time (r) was measured in lipid extracts of surface membrane fractions with the spin label, methyl (12-nitroxyl)stearate, which preferentially associates with the fluid domains of the membrane (32). Molecular ordering parameter (s) was measured with the oriented spin label (5-nitroxylstearic acid) in total membrane fractions (31). The spectra from both probes demonstrated immobilization in the membrane preparations of ethinyl estradioltreated rats. Both T (60%, P < 0.005) and s (5%, P < 0.005) were increased in surface membranes from rats treated with ethinyl estradiol compared to controls (Fig. 2). When administered to ethinyl estradiol-treated rats, Triton WR-1339 restored r and s values to values not significantly different from those of the control. Since the ESR data were obtained at 37°C, it is likely that the changes in membrane fluidity represent significant differences at physiologic temperature. Effect of Triton WR-1339 on Na+,K+-ATPase Activity and Basal Bile Flow. If decreased liver surface membrane functions result from altered viscosity, it should follow that restoration of membrane lipid composition and viscosity to normal should also be associated with correction of abnormal Na+,K+-ATPase and bile flow. As shown in Table 2, administration of Triton WR-1339 restored hepatic Na+,K+-ATPase activity and bile flow to the control levels. This effect is limited to ethinyl estradiol-treated animals, for Triton had no effect on untreated rats. Certain detergents increase Na+,K+-ATPase activity in vitro (33). Therefore, it seemed possible that correction of reduced Na+,K+-ATPase activity by Triton WR-1339 may result from a direct action of the detergent on the membrane. To examine this possibility, we added Triton WR-1339 in various concentrations to membranes from untreated and ethinyl estradiol-
4132
Biochemistry:
Davis et al.
Proc. Natl. Acad. Sci. USA 75 (1978) Table 1. Lipid composition of surface membrane fractions
Molar ratio,
cholesterol/
jug of lipid/mg of membrane protein Treatment group Control Ethinyl estradiol Ethinyl estradiol + Triton WR-1339 Control + Triton WR-1339 *
Phospholipid
Cholesterol
Cholesterol ester
phospholipid
360 + 51 380 + 18
90 ± 8 4* 120 V
4+1 , 16 ± 3*
0.48 i 0.04 0.61 + 0.03*
383 + 10
100 i 6
7I2
0.51 ± 0.03
352 1 33
80 1 6
6±1
0.44 I 0.04
Expressed as the mean ± SEM. Number of animals in each group = 6.
Significant difference from control values at P < 0.05.
treated rats. Addition of Triton WR-1339 in vitro increased activity of Na+,K+-ATPase as a function of the amount of detergent added to membrane fractions obtained from ethinyl estradiol-treated rats until a maximal value of activity was obtained that was not significantly different from control (Fig. 3). In contrast, addition of Triton WR-1339 to surface membrane fractions from untreated rats did not change enzyme activity. Activation was not due to solubilization, for Na+,K+ATPase activity was recovered only in the pellet after centrifugation (104,000 X g for 30 min) at each concentration of detergent. In addition, at the concentrations used neither protein
membrane lipids were selectively released from the membrane fractions, nor did Triton WR-1339 affect the inorganic phosphorus determination. In further experiments, s was determined in duplicate samples of each assay. Addition of Triton WR-1339 (6.25 ,ug) to surface membrane fractions of ethinyl estradiol-treated rats decreased s from 0.730 + 0.006 to the control value of 0.700 + 0.006. Additional Triton WR-1339 did not alter s further. Furthermore, Triton WR-1339 added to surface membrane preparations from untreated rats changed neither s nor Na+,K+-ATPase activity.
nor
3
DISCUSSION Current views of membrane structure suggest their essential feature is a bilayer containing a polar shell consisting primarily of phospholipids and free cholesterol. This lipid environment significantly influences the activity of several membrane enzymes (34-38). The molecular mechanisms by which the lipid composition of the membrane bilayer influences enzymes probably involves protein conformational changes due to the physical state of the immediate or boundary lipids and/or restrictions of molecular motion necessary for formation of the transition state. Lipid viscosity is apparently rate determining for the membrane-bound enzymes, polyisoprenoid alcohol phosphokinase (29) and cytochrome b5 (37). Although the molecular mechanisms for Na+,K+-ATPase function are not fully elucidated, Kyte (3) suggests that conformational changes are required both for exposing cation sites and for forming a channel through the membrane bilayer through which cations are transported. Previous studies (7, 36, 39) also suggest an important role for membrane viscosity in controlling the activity of Na+,K+-ATPase in that there is a good correlation between membrane lipid phase transitions and Arrhenius activation energies. Indeed, Kimelberg and Papahadjopoulis have speculated that membrane cholesterol accumulation may impair Na+,K+-ATPase activity through changes in membrane structure (7). The results of the present study support these previous studies and suggest that the synthetic estrogen, ethinyl estradiol, decreases hepatic Na+,K+-ATPase activity through changes in surface membrane lipid structure.
C
z;
Na+,K+-ATPase activity in ethinyl estradiol-treated rats is reduced to 50% of control in liver surface membrane fractions. Since the surface membrane from hepatocytes is a mosaic of three distinct surfaces, and Na+,K+-ATPase may be preferentially localized to one site, it is possible that ethinyl estradiol treatment alters enzyme activity by changing the relative re-
0. 7
E
00
0
0.65,
0~~~~~~~~
Table 2. Effect of ethinyl estradiol and Triton WR-1339 on hepatic Na+,K+-ATPase and basal bile flow
0
Treatment
0
0.6
C
EE
EE+
C+
Triton Triton
C
EE
EE+
C+
Triton Triton
FIG. 2. ESR analysis of surface membrane structure. (Left) Rotational correlation time (r); (Right) order parameter (s). Measurements were determined from the paramagnetic spectra as described in text. Results are expression of duplicate determinations from three different animals in each treatment group. C = control (propylene glycol); EE = ethinyl estradiol, EE + Triton = ethinyl estradiol + Triton WR-1339; C + Triton = untreated + Triton Wr1339. Values are expressed as the mean + SEM. (*) P < 0.005, compared to control determinations.
Na+,K+-ATPase*
Basal bile flowt 7.6 ± 0.2 5.4 ± 0.6t
Control 44.3 i 3.0 21.9 ± 2.0t Ethinyl estradiol Ethinyl estradiol + Triton 48.9 ± 1.6 7.8 + 0.4 Triton alone 48.3 ± 2.0 8.0 ± 0.6 Values are expressed as mean + SEM for duplicate determinations in six different animals in each group. Na+,K+-ATPase was measured in liver surface membrane fractions as described in text. * tmol Pi/hr per mg of membrane protein. t ,l/min per 100 g of body weight. Pp < 0.001.
Biochemistry:
10
Proc. Natl. Acad. Sci. USA 75 (1978)
Davis et al.
100 Triton added, ,g
50
130
625
FIG. 3. Effect of adding Triton WR-1339 to surface membrane fractions on Na+,K+-ATPase activity in vitro. Triton WR-1339 in 0.9% saline was added at various concentrations to surface membrane fractions (190 yg of protein per tube) of control (O--- -0) and ethinyl estradiol-treated (0-0) rats. Each point represents the mean of duplicate determinations of Na+,K+-ATPase. Duplicate values agreed within 10% of each other. Lines were drawn by the method of least squares.
covery of the membrane fraction. This possibility is unlikely, for recovery of surface membrane protein is unaltered and the relative enrichment of Na+,K+-ATPase was equal for all groups
examined. We conclude that the in vitro changes in surface membrane lipid composition, lipid structure, and Na+,K+ATPase reflect the in vivo surface membrane, a conclusion supported by recent studies of Reichen and Paumgartner (40). This suggestion is supported by restoration to normal of surface membrane lipid composition and structure, Na+,K+-ATPase activity, and bile flow in ethinyl estradiol-treated rats given Triton WR-1339.
Reduced Na+,K+-ATPase activity may also result from decreased function of a fixed number of enzyme units or may be caused by a decreased number of enzyme units. Since in both isolated cell membranes (Fig. 3) and in vivo (Fig. 1) Triton WR-1339 increased the Na+,K+-ATPase activity of ethinyl estradiol-treated rats to control levels and not further, and since there was no effect in untreated rats, the number of enzyme units of Na+,K+-ATPase is probably not changed by ethinyl estradiol treatment. Furthermore, Triton WR-1339 activation of Na+,K+-ATPase in ethinyl estradiol-treated rats was not inhibited by cycloheximide, indicating that de novo protein synthesis is not required (unpublished observations). In addition, preliminary studies with [3H]ouabain and [y-32P]ATP binding to surface membrane fractions indicate no change in the number of enzyme units in liver surface membrane fractions from ethinyl estradiol-treated rats (unpublished observations). These observations are similar to those reported in studies of erythrocyte membranes from cholesterol-treated guinea pigs (40, 41), in which Na+,K+-ATPase activity is reduced but the number of [3H]ouabain binding sites is unaltered. The findings of significant increases in both s and y in liver surface membrane fractions indicate that ethinyl estradiol alters the lipid structure of this membrane fraction by decreasing its fluidity. The exact mechanism by which ethinyl estradiol alters the structure of liver surface membrane lipids is not settled by the present experiments. However, in model membrane studies, Davis and Sinensky (42) have shown that increased concentrations of esterified cholesterol, similar to those found in surface membrane fractions from ethinyl estradiol-treated rats, cause increased lipid viscosity. Triton may reduce membrane viscosity and increase
4133
Na+,K+-ATPase activity through two possible mechanisms in ethinyl estradiol-treated rats. First, Triton inhibits the activation of hepatic microsomal cholesterol acyl-CoA transferase caused by ethinyl estradiol and restores hepatic lipid composition to normal (14). Second, Triton may directly affect the structure and function of abnormal membranes, possibly by its inseftion into the lipid bilayer. Whatever the mechanism, these data suggest that increased membrane viscosity caused by ethinyl estradiol treatment is reduced to normal by Triton WR-1339 through its effect on the surface membrane structure. Changes in membrane structure are closely correlated with sodium pump activity and bile flow, suggesting that the physical state of membrane lipids plays an important role in regulating transport across membranes. These observations may explain the reductions in organic anion transport previously reported for ethinyl estradiol treatment in human beings and rats (43). We thank Patricia Coan for her excellent technical assistance and Drs. D. Steinberg, I. M. Arias, and J. C. Williams for their critical review of the manuscript. This work was supported by Grants AM 15851 and AM 12626 from the National Institutes of Health and by American Cancer Society Grant BC 219. 1. Skou, J. C. (1965) Physiol. Rev. 45,596-617. 2. Uesugi, S., Dulak, N. C., Dixon, J. F., Hexum, T. D., Dahl, J. L., Perdue, J. F. & Hokin, L. E. (1971) J. Biol. Chem. 246, 531543. 3. Kyte, J. (1975) J. Biol. Chem. 250, 7443-7449. 4. Keith, A. D., Sharnoff, M. & Cohn, G. E. (1973) Biochim. Biophys. Acta 300, 379-419. 5. Hokin, L. E. & Hexum, J. D. (1972) Arch. Biochem. Biophys. 151, 453-463. 6. Tanaka, R. & Strickland, K. P. (1965) Arch. Biochem. Biophys. 111,583-592. 7. Kimelberg, H. K. & Papahadjopoulis, D. (1974) J. Biol. Chem. 249, 1071-1080. 8. Erlinger, S., Dhumeaux, D. & Benhamou, J. P. (1969) Nature (London) 223, 1276-1277. 9. Boyer, J. L. (1971) Am. J. Physiol. 221, 1156-1163. 10. Simon, F. R., Sutherland, E. & Accatino, L. (1977) J. Clin. Invest. 59, 849-861. 11. Davis, R. A. & Kern, F., Jr. (1975) Gastroenterology 70, 1130-1135. 12. Simon, F. R. & Arias, I. M. (1973) J. Clin. Invest. 52, 765775. 13. Gumucio, J. J. & Valdevieso, V. D. (1971) Gastroentrology 61, 339-344. 14. Davis, R. A., Showalter, R. & Kern, F., Jr. (1978) Biochem. J. 174, 45-51. 15. Davis, R. A. & Kern, F. (1977) Gastroenterology 72, 1045. 16. Davis, R. A. & Kern, F., Jr. (1977) in Bile Acid Metabolism in Health and Disease, eds. Paumgartner, G. & Stiehl, A. (MTP Press Ltd., London), pp. 25-33. 17. Neville, D. M., Jr. (1968) Biochim. Biophys. Acta 154, 540552. 18. Pohl, S. L., Bimbaumer, R. & Rodbell, M. (1971) J. Biol. Chem. 246, 1849-1856. 19. Neville, D. M. (1975) in Methods in Membrane Biology, ed. Korn, I. D. (Plenum, New York), Vol. 3, pp. 1-49. 20. Evans, W. H. & Gurd, J. W. (1971) Biochem. J. 125,615-624. 21. Accatino, L. & Simon, F. R. (1976) J. Clin. Invest. 57, 496508. 22. Weihing, R. R., Manganiello, V. C., Chin, R. & Phillips, A. H. (1972) Biochemistry 11, 3128-3135. 23. Ismail-Beigi, F. & Edelman, I. D. (1971) J. Gen. Physiol. 57,
710-722.
24. de Duve, C., Pressman, B. C., Gianetto, R., Wattiaux, R. & Appelmans, F. (1955) Biochem. J. 60, 604-617. 25. Ojnura, T. & Sato, R. (1964) J. Biol. Chem. 239, 2370-2378. 26. Lowry, 0. H., Rosebrough, N. D., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, 265-275.
4134
Biochemistry: Davis et al.
27. Folch, J., Lees, M. & Sloan-Stanley, G. H. (1957) J. Biol. Chem. 226,497-509. 28. Bartlett, G. R. (1959) J. Biol. Chem. 234, 466-470. 29. Sinensky, M. (1974) Proc. Natl. Acad. Sci. USA 71,522-525. 30. Gennis, R. B., Sinensky, M. & Strominger, J. L. (1976) J. Biol. Chem. 251, 1270-1276. 31. Hubbell, W. L. & McConnell, H. M. (1971) J. Am. Chem. Soc. 93,314-329. 32. Oldfield, K., Keough, M. & Chapman, D. (1972) FEBS Lett. 20, 344-346. 33. Schwartz, A., Lindenmayer, G. E. & Allen, J. C. (1976) Pharmacol. Rev. 27,3-134. 34. Coleman, R. (1973) Biochim. Biophys. Acta 300, 1-30. 35. Farias, R. N., Bloj, B., Morero, R. D., Sifieriz, F. & Trucco, R. E. (1975) Biochim. Biophys. Acta 415,231-251.
Proc. Nati. Acad. Sci. USA 75 (1978) 36. Kimelberg, H. K. (1976) Molec. Cell Biochem. 10, 171-190. 37. Rogers, M. J. & Strittmatter, P. (1974) J. Biol. Chem. 249, 895-900. 38. Solomonson, L. P., Liepkains, V. A. & Spector, A. A. (1976) Biochemistry 15, 892-897. 39. Kroes, J. & Ostwald, R. (1971) Biochim. Biophys. Acta 249, 647-650. 40. Reichen, J. & Paumgartner, G. (1977) J. Clin. Invest. 52,429434. 41. Kroes, J., Ostwald, R. & Keith, A. (1972) Biochim. Biophys. Acta
274,71-74. 42. Davis, R. A. & Sinensky, M. (1977) Fed. Proc. Fed. Am. Soc. Exp. Biol. 37, 2290. 43. Adlercreutz, H. & Tenhunen, R. (1970) Am. J. Med. 49, 630648.
