- Email: [email protected]
Isolation of sinusoidal and canalicular liver plasma membranes: Effects of frozen storage of human material
Pergamon
0887-2333(93)E0004-Y
Toxic. in Vitro Vol. 8, No. 2, pp. 173-180, 1994 Copyright © 1994 ElsevierScienceLtd Printed in Great Britain. All rights reserved 0887-2333/94 $7.00 + 0.00
ISOLATION OF SINUSOIDAL A N D CANALICULAR LIVER PLASMA MEMBRANES: EFFECTS OF FROZEN STORAGE OF H U M A N MATERIAL* R. J. EDWARDS,C. L. LESLIE and N. H. STACEYt Toxicology Unit, National Institute of Occupational Health and Safety, University of Sydney, NSW, Australia (Received 15 March 1993; revisions received 23 July 1993) Abstract--Sinudoisal and canalicular plasma membranes were prepared from rat and fresh or frozen human liver and characterized using plasma membrane marker enzymes. The distributions of these enzymes were used as a means of comparison between the plasma membrane fractions and homogenates from the rat and fresh and frozen human livers. Compared with their homogenate, all the sinusoidal plasma membrane preparations were poorly enriched with Na+-K+-ATPase (2.6~8.4-fold), a sinusoidal marker enzyme. Activities of Na+-K+-ATPase and Mg2+-ATPase were much greater in rat membranes whereas leucine aminopeptidase and alkaline phosphatase activities were much greater in the human membranes. Consideration of the extent of enrichment showed that Mg2+-ATPase and alkaline phosphatase are better markers in rat canalicular plasma membranes and leucine aminopcptidase in human membranes from both fresh and frozen tissue. Intracellular organdie marker enzymes were not enriched in the plasma membranes, thus indicating no significant contamination. Importantly, the relative enrichments of the marker enzymes showed no difference between fresh and frozen human liver. Absolute activities of the enzymes were decreased, however (except alkaline phosphatase), after frozen storage. This work indicates that human liver tissue may be frozen and stored in liquid nitrogen before plasma membranes are isolated with no effect on the distribution of marker enzymes in the membranes.
INTRODUCTION Hepatocytes, the main cellular component of hepatic tissue, are highly polarized epithelial cells. The plasma membranes of hepatocytes are composed of three main surface domains that have been characterized morphologically, functionally and biochemically (Evans and Enrich, 1989; Meier et aL, 1984). They are termed the sinusoidal, lateral and canalicular domains of the hepatocyte plasma membrane. The sinusoidal front and lateral surface of the plasma membrane is the basolateral pole of the hepatocyte. These two domains are also in physical continuum and exposed to sinusoidal blood and as a result are thought to be functionally equivalent. The canalicular domain, the apical pole of the cell, is physically separated from the other two domains by tight junctional complexes and is functionally distinct (Meier et al., 1984). Bile formation, which occurs primarily from the hepatocyte, is a complex process that involves a number of transport mechanisms in both the sinusoidal and canalicular domains of hepatocyte plasma *The views expressed in this article are those of the author and do not necessarily reflect those of the National Occupational Health and Safety Commission (Worksafe Australia). "['To whom correspondence should be addressed. Abbreviation: UW = University of Wisconsin.
membranes. The transport of endogenous chemicals at these sites plays a critical role in the formation of bile. As a result any interference with these transport processes may affect bile output. These transmembrane transport processes can be studied in a system without the interference of intracellular processes present in whole-cell studies. The development of a number of methods that allow the isolation, and in some cases separation, of the sinusoidal and canalicular plasma membrane domains has enabled these studies to take place (Kessler et al., 1990; Loten and Redshaw-Loten, 1986; Meier et al., 1984; Rosario et aL, 1988). Sinusoidal and canalicular plasma membranes isolated from rat liver have been well characterized with respect to their morphology, function and biochemical content (Blitzer and Donovan, 1984; Changchit et al., 1990; Inoue et al., 1982 and 1983; Meier et al., 1984; Rosario et al., 1988; Wolters et al., 1991). However, few studies have investigated the characteristics of human liver plasma membranes (Novak et al., 1989; Wolters et al., 1991), which would be a preferred alternative to rat tissue. One problem associated with the study of human liver plasma membranes is the difficulty of obtaining a regular supply of human liver tissue. A second problem is that when human tissue does become available, it is often in a quantity that could not be used in just one membrane preparation. 173
174
R.J. EDWARDS et al.
Consequently, tissue is wasted. A method of storing this excess tissue, and of alleviating the problem of irregular supply, needs to be developed. As it is unknown whether cryostorage would affect the tissue or the plasma membranes isolated from the tissue, it was also decided to investigate whether storage of human liver tissue in liquid nitrogen would affect the distribution, activity and/or the relative enrichment of the marker enzymes in the sinusoidal and canalicular plasma membrane fractions. MATERIALS AND METHODS
Materials Magnesium chloride (hexahydrate) [AnalaR grade] was obtained from BDH Chemicals Australia (Victoria, Australia). All other chemicals used were of the highest purity commercially available.
(Brinkman Co.). A 200-#1 aliquot of the homogenate was stored at - 2 0 ° C for enzyme analysis. Ice-cold high-purity water (21.5 ml) was added to each tube of homogenate and mixed gently. The homogenate mix was then transferred to 26.3-ml Beckman centrifuge tubes and centrifuged at 45,000g for 30 min in a Beckman model L8-70 ultracentrifuge. The resulting pellet (P1, Fig. 1), after resuspension in 15 ml buffer, was homogenized in a loose Dounce with 8-10 strokes. The homogenate was transferred to Hitachi tubes and 21.5 ml ice-cold 25.2 mM MgC12 was added to each tube for Mg 2+ precipitation of the sinusoidal plasma membrane fraction. The precipitation was followed by centrifugation at 2500 g for 15 min in a Hitachi H I M A C high-speed centrifuge producing a pellet (P2) and supernatant ($2).
(a) Preparation of sinusoidal plasma membranes.
Isolation of plasma membranes from rat liver
The P2 underwent further homogenization (in a loose Dounce), Mg 2+ precipitation and centrifugation at 2500g for 15 min resulting in a crude sinusoidal pellet (P3*). The supernatant ($3") was used in the preparation of canalicular plasma membranes [see part (b)]. The crude sinusoidal pellet, after resuspension in 15 ml buffer, was homogenized with 8-10 strokes in a loose Dounce. The homogenate was transferred to Hitachi tubes to which 21.5 ml water was added. The tubes were mixed vigorously and centrifuged at 750 g for 15 min. The resulting pellet (P4*) consisted of a red layer, representing mitochondria and other intracellular organelles, and a brown layer, containing crude sinusoidal membranes. The supernatant ($4") and loose brown pellet were transferred to Beckman tubes and centrifuged at 45,000g for 30min. The pellet (P5*), after resuspension in 17 ml 50% sucrose (w/w), was homogenized with a medium fit PotterElvehjeim homogenizer using four strokes. The homogenate was overlaid with 5 ml 41% sucrose (w/w) and 15 ml 38% sucrose (w/w) to form a discontinuous sucrose gradient in Beckman Ultra-Clear centrifuge tubes (38.5 ml). The tubes were centrifuged at 48,000g, w2t = 7.18 × l010, in a SW-28 rotor of the Beckman ultracentrifuge for 139 min. The float layer, containing sinusoidal plasma membranes, was carefully harvested from the top of the discontinuous gradient and pelleted by placing the layer in 25 ml NaHCO 3 (1 mM) and centrifuging at 45,000g for 30 min. This resulted in a purified sinusoidal plasma membrane pellet, which was resuspended in 0.5-1.0ml NaHCO3 (1 mM) and stored at -20°C.
The procedure used to isolate plasma membranes from rat liver was based on the one described by Rosario et al. (1988). The complete method was as follows. The rats were anaesthetized using ether, and their livers (weighing 8-10 g) removed and minced thoroughly. All procedures were performed on ice. Minced liver portions (l.8g), placed in 45.5-ml Hitachi centrifuge tubes, were homogenized in 15 ml buffer (300mM mannitol, 5mM EGTA, 15mM Tris-HCI, 0.1 mM phenylmethane sulfonyl fluoride at pH 7.4) for 45 sec at 4800 rpm using a polytron
The supernatant from the first Mg 2+ precipitation, $2, was centrifuged at 45,000g for 30 min to obtain a crude canalicular fraction (P3). The P3 was resuspended with the $3", from the sinusoidal plasma membrane preparation pathway, and centrifuged at 45,000 g for 30 min to obtain a greater crude canalicular fraction (P4). The P4 was resuspended in 15 ml buffer and homogenized using a tight fit PotterElvehjeim homogenizer (Wheaton) with five strokes. To the homogenate 21.5 ml MgCl2 (25.2mM) was
Human livers Human livers were obtained from three organ donors from whom most of the liver was used for transplantation. Livers were stored in University of Wisconsin (UW) solution, on ice, until the start of an isolation procedure or until storage in liquid nitrogen. Livers were in UW for no longer than 2 hr. Use of human tissue was approved by the Ethics Review Committee of the Central Sydney Health Service and with the co-operation of the Royal Prince Alfred Hospital.
Freezing and thawing of human liver tissue Human livers were divided into 10-g portions, drained of UW solution and securely wrapped in aluminium foil. Each parcel of liver was then frozen and stored in liquid nitrogen. Before each isolation procedure a frozen portion of human liver was thawed at room temperature and then placed on ice ready for the isolation procedure.
Animals Male PVG/c hooded rats (200--250g), obtained from Blackburn Animal House (University of Sydney, Australia), were used in all experiments. The rats were housed in cages of not more than five per cage with free access to food and water. The rats were fasted overnight before the isolation procedure.
(b) Preparation of canalicular plasma membranes.
Liver plasma membranes: frozen storage
175
MINCED LIVER
Polytron homogenization
[ HOMOGENATE ]
a45,000g x 30 min
Mg 2÷ precipitation (pptn) 2500g x 15 min
Mg 2+ pptn 2500g x 15 min
45,000g x 30 min
+$3
Water "cut" 750g x 15 min
45,000g
30 min
Discard
45,000g x 30 min
Mg 2+ pptn 2500g x 20 min
(~
Discontinuous sucrose gradient (50%/41%/38%) 48,000g x 139 min
Floatlayer on 38% sucrose surface
I
Discard
45,000g x 30 min
C A N A L I C U L A R PLASMA MEMBRANES
45,000g x 30 min
Discard SINUSO1DAL PLASMA MEMBRANES Fig. 1. Method for the preparation of plasma membranes from rat or human liver, a: centrifugal speed (g) and time (min); b: S = supernatant; c: P = pellet; *S or P from the sinusoidal plasma membrane preparation pathway.
176
R . J. E D W A R D S
et al.
18-16-.~,_, 14
_
~
12-
,,~ 8
10--
÷1
80
B
[]
Rat
1
Human
4--
60
I
~ "~ o
50
O ~
40
+
10
A¢1
0
H
s
C
8-
I
o
.~.
6--
2--
H :~
10
nn S
-~
50--
o = ~'~
40
H
S
C
=., =-
30
o
20--
o
~
--
10--
A
c
c
3.0--
-
•~8
S
D
C
•~ ' Z
H
60 --
10--
~
5_
70
~1.
Z
D
A
F 2.5--
-
2.0-~'~.
9...
4
2 < Z
7"1
6 -
0
-
-
nn
A s
S
om.
1.5
oo E O
1.0
o
/ IA IA /A /A /A /A
0.5
,',1
0.0 C
,A ,4 /4 /4
3_
H
S
c
Fig. 2. Marker enzyme activities in homogenate, sinusoidal and canalicular plasma membrane fractions prepared from rat and fresh human livers. (A) Na+-K+-ATPase; (B) Mg2+-ATPase; (C) leucine aminopeptidase; (D) alkaline phosphatase; (E) NADPH cytochrome c reductase; (F) succinate cytochrome c reductase. (H = homogenate; S = sinusoidal fraction; C = canalicular fraction). Note different ordinate scale of each graph. Each column represents mean value of triplicate experiments + SE. added, for precipitation o f any c o n t a m i n a t i n g sinusoidal plasma m e m b r a n e s , followed by centrifugation at 2 5 0 0 g for 2 0 m i n . The s u p e r n a t a n t ($5) was transferred to B e c k m a n tubes a n d centrifuged at 4 5 , 0 0 0 g for 30 m i n a n d the resulting pellet resuspended in 0 . 5 - 1 . 0 m l N a H C O 3 (1 mM). This pellet consisted o f purified canalicular plasma m e m b r a n e fraction, a n d was stored at - 2 0 ° C .
Isolation of plasma membranes from human liver Portions of h u m a n liver (weighing approximately 10 g) were cleared as extensively as possible o f vascular a n d connective tissue. Isolation of sinusoidal and canalicular m e m b r a n e s from the tissue was achieved by following the same isolation procedure as t h a t for m e m b r a n e isolation from rat liver.
Liver plasma membranes: frozen storage
Enzyme assays and protein determination A n u m b e r o f m a r k e r enzymes were used to test for m e m b r a n e separation. T h e enzyme N a + - K +-ATPase was used as a m a r k e r for the sinusoidal fraction of
177
plasma m e m b r a n e s since its activity is k n o w n to be restricted to the basolateral plasma m e m b r a n e surface o f hepatocytes (Blitzer a n d Boyer, 1978). The activity of N a + - K + - A T P a s e was measured by the m e t h o d o f Schoner et al. (1967). T h r e e m a r k e r
16
6
-
A
B
14
5 m
10 - -
[]
Fresh
[]
Frozen
'~
4
<~ ~
~
3
//
~
// // // // // // // //// //
4
Z
1
o H
S
C
H
10 --
D
I
50
8 -
~2 "=
6
,5~
4
'~'~
C
60
J_
C ..>
S
o = ~'~ eu
40
~
30
,~-~ ~
20
<
10
v, Ii /i
o
2
ii
m
.3 H
lln
Z i/ // I¢
Z
m
i/ I
,
s
H
3.0
m
..>
F
2.5
8 --
"~ 2.0 ~8
~
:
I
1.5
i ,.o
1i
~ z
S
o
0.5
~ 0.0 S C H S C Fig. 3. Marker enzyme activities in homogenate, sinusoidal and canalicular plasma membrane fractions prepared from fresh human and frozen human livers. (A) Na+-K+-ATPase; (B) Mg2+-ATPase; (C) leucine aminopeptidase; (D) alkaline phosphatase; (E) NADPH cytochrome c reductase; (F) succinate cytochrome c reductase. (H = homogenate; S = sinusoidal fraction; C = canalicular fraction). Note different ordinate scale of each graph. Each column represents mean value of triplicate experiments + SE.
178
R.J. EDWARDSet al. Table 1. Relative enrichment* of marker enzyme activities in sinusoidal and canalicular membrane fractions prepared from rat, fresh human and frozen human liver Rat Marker enzyme Na+-K+-ATPase Mg2+-ATPase Leucine aminopeptidase Alkaline phosphatase N A D P H cytochrome c reductase Succinate cytochrome c reductase
Fresh human
Thawed human
SPM
CPM
SPM
CPM
SPM
CPM
8.4 6.5
5.8 21.8
2.6 1.1
4.5 2.3
2.7 I.I
5.4 2.4
2.6 7.2
7.2 28.0
5.2 3.4
15.0 10.2
4.1 3.3
15.6 7.9
0.7
0.4
1.5
1.4
1.0
0.7
0.6
0.6
1.6
0.2
1.1
0.2
CPM = canalicular plasma membranes SPM = sinusoidal plasma membranes *Relative enrichment is expressed as specific activity of membrane/specific activity of homogenate. Values represent ratio of the means of specific activities where n = 3.
enzymes were used to identify the canalicular plasma membrane: Mg2+-ATPase, leucine aminopeptidase and alkaline phosphatase. These enzymes are known to be predominantly active in the canalicular plasma membrane but not exclusively located at that site (Evans, 1970; Wisher and Evans, 1975). Mg2+-ATPase activity was measured by the method of Schoner et al. (1967). Alkaline phosphatase activity, using p-nitrophenyl phosphate as substrate, was measured according to Bessey et al. (1946). Leucine aminopeptidase activity was measured by the method of Haase et al. 0978). Contamination with mitochondria and microsomes was measured by N A D P H cytochrome c reductase and succinate eytochrome c reductase activities, respectively (Sottocasa et al., 1967). Protein was estimated by the method of Lowry et aL (1951) using bovine serum albumin as standard. All enzyme activities were measured at 37°C following overnight storage of the plasma membrane fractions at 20oc. -
RESULTS
In membranes prepared from rat livers the activity of Na+-K+-ATPase was highest in the sinusoidal fraction (Fig. 2A). Activities were less in preparations from human liver, Mg2+-ATPase activity (Fig. 2B) was greatest in the canalicular membrane fractions from rat liver especially when compared with human liver values. Figure 2C shows that leucine aminopeptidase activity was greatest in the canalicular membrane fractions prepared from both rat and human livers, with activity being two-fold higher in the latter. Activity of the third canalicular marker enzyme, alkaline phosphatase (Fig. 2D), was greatest in the canalicular fr~/ctions of membranes prepared from both rat and human livers. As with the enzyme leucine aminopeptidase, alkaline phosphatase activity was much greater in the membranes prepared from human tissue.
During the isolation procedure it was possible for the membrane fractions to be contaminated with other cell components, for example mitochondria and microsomes. N A D P H cytochrome c reductase was used as a marker enzyme for mitochondria (Fig. 2E), and succinate cytochrome c reductase for mtcrosomes (Fig. 2F). N A D P H cytochrome c reductase activity was higher in the homogenate, sinusoidal and canalicular fractions prepared from human livers than in the corresponding fractions prepared from rat livers. Succinate cytochrome c reductase activity in membrane fractions prepared from both rat and human livers was low. The values show that there were no increases to suggest significant contamination with other membranes. A comparison of the activities of the marker enzymes in the membrane fractions prepared from fresh and frozen human liver tissues showed that there was a decrease in most of the enzymes' activities in the thawed tissue membranes (Fig. 3A-F), except for alkaline phosphatase (Fig. 3D). Table l presents the relative enrichments of the marker enzymes' activities measured in the sinusoidal and canalicular fractions of both rat and human tissue (fresh and frozen). Relative enrichment is defined as the specific activity of a given enzyme in a membrane fraction divided by the specific activity of this enzyme in the homogenate. The sinusoidal marker enzyme Na ÷-K ÷-ATPase showed greater enrichment in the sinusoidal fraction for rat liver, while fresh and frozen human liver showed a small but greater enrichment in the canalicular membranes. The canalicular marker enzymes, Mg2+-ATPase and alkaline phosphatase, showed greater enrichment in rat canalicular fractions, while leucine aminopeptidase enrichment was greater in human canalicular membrane fractions. The relative enrichments for membranes from fresh and frozen human liver tissue showed remarkable similarity for all marker enzymes including those of other organelles.
Liver plasma membranes: frozen storage DISCUSSION
179
Wolters et al. (1991), which may account for some of the differences in the human data. Alkaline phosphaEnzymatic characterization, using six marker entase activity was much greater in human homogenzymes, was used as a method of comparing liver ates compared with rat homogenates. This may be plasma membranes from rats and humans and, more due to the fact that in the human liver this enzyme is importantly, for fresh v. frozen human liver. The located in at least two different sites on the hepatosinusoidal marker enzyme Na+-K+-ATPase was cyte, whereas in rat livers it is located at only one site poorly enriched in the sinusoidal plasma membranes (Wolters et al., 1991). Na+-K+-ATPase and Mg 2--from all three liver sources--rat, and fresh and frozen ATPase activities were greater in rat membranes, human tissue. Overall, the activities and enrichments whereas leucine aminopeptidase and alkaline of the marker enzymes for the canalicular fraction phosphatase activities were greater in the human showed suitable enrichment. The plasma membrane membranes. However, enrichment of leucine fractions were not enriched with enzymes from other aminopeptidase was greater in human membranes intracellular organelles. Comparison of plasma mem- and alkaline phosphatase in rat membranes. These branes from rat and fresh human liver sources results indicate that Mg2+-ATPase would be the highlighted some differences. However, numerous preferred marker in rat plasma membranes and similarities were also observed. Lower activities of leucine aminopeptidase in human plasma memmost marker enzymes were measured in the plasma branes. Alkaline phosphatase is a useful second membranes from frozen human liver compared with marker for both species. those from fresh human liver. However, in general, The comparison of enzymatic characteristics of the enrichments of the marker enzymes were the membranes prepared from fresh and frozen human same. There were greater variations in the human livers showed that each enzyme, except alkaline phosplasma membrane preparations than in rat prep- phatase, was lower in the homogenate and the plasma arations, and this may be explained by the greater membrane fractions prepared from frozen human number of non-controllable factors present in human liver. The intracellular organelle marker enzymes tissue studies, that is, sex, age, diet and so on. activities were also reduced in the frozen liver plasma Na+-K÷-ATPase activity in rat sinusoidal plasma membranes. However, even when absolute levels were membranes was enriched eight-fold over homogenate decreased, relative enrichments stayed the same after in this study. The enrichment was much lower than frozen storage. This indicates that the loss of activity the 20-35-fold enrichment of Na+-K+-ATPase ac- was evenly distributed across the surface of the tivity in sinusoidal membranes reported in the litera- human hepatocyte and did not affect specific plasma ture (Blitzer and Donovan, 1984; Changchit et al., membrane domains. 1990; Inoue et al., 1982; Meier et al., 1984; Rosario In conclusion, the activities and distributions of et al., 1988; Wolters et al., 1991). In this study, some membrane marker enzymes were different in although Na÷-K÷-ATPase activity in the sinusoidal plasma membranes from rat liver compared with fraction was the same as that reported by Meier et al. plasma membranes from human liver. More impor(1984) and Inoue et al. (1982), activity of the enzyme tantly, it has been shown that distributions and in the homogenate was much higher than they re- relative enrichments of marker enzymes were unported, which would account for the lower relative changed in plasma membranes from human liver enrichments. Despite the same method of plasma after frozen storage. Thus, it should be acceptable to membrane isolation being used in this study and by use human liver samples that have been stored in Rosario et al. (1988), different marker enzyme activi- liquid nitrogen as a source of tissue for plasma ties and enrichments were obtained. In the Rosario membrane isolation. Increasing the availability and et al. (1988) study Na+-K+-ATPase activities were use of human tissues for in vitro studies has advanmuch greater than those reported by Meier et al. tages for both reduced use of laboratory animals and (1984) and Inoue et al. (1982) or in our study. Despite relevance of the tissue used to the ultimate species of the high activities reported by Rosario et al. (1988) concern. the relative enrichment of Na+-K÷-ATPase was the same as that reported by others (Inoue et al., 1982; Acknowledgement--This work was supported in part by a Meier et al., 1984). The precise reasons for these National Health and Medical Research Council grant. inconsistencies from various laboratories remain to be determined. REFERENCES The activities of the three canalicular marker enBessey O. A., Lowry O. H. and Brock M. J. (1946) A zymes (Mg2+-ATPase, leucine aminopeptidase and method for the rapid determination of alkaline phosphaalkaline phosphatase) were well enriched in the tase with 5 cm3 of serum. Journal o f Cell Biology 164, 321 329. canalicular fraction, although some were lower than those published elsewhere (Inoue et al., 1983; Meier Blitzer B. L. and Boyer J. L. (1978) Cytochemical localization of (Na +-K ÷ )ATPase in the rat hepatocyte. Journal et al., 1984; Rosario et al., 1988; Wolters et al., 1991). o f Clinical Investigation 62, 1101-1104. It should be noted that different methods of plasma Blitzer B. L. and Donovan C. B. (1984) A new method for the rapid isolation of basolateral plasma membrane membrane isolation were used in our study and by
180
R.J. EDWARDS et al.
vesicles from rat liver. Journal of Biological Chemistry 259, 9295-9301. Changchit A., Durham S. and Vore M. (1990) Characterization of [3H]estradiol- 17fl-(//-o-glucuronide) binding sites in basolateral and canalicular liver plasma membranes. Biochemical Pharmacology 40, 1219-1225. Evans W. H. (1970) Fractionation of liver plasma membranes prepared by zonal centrifugation. Biochemical Journal 166, 833-842. Evans W. H. and Enrich C. (1989) The hepatocyte's plasma membrane domains. Interrelations with the endocytic compartment. In Hepatic Transport in Organic Substances. Edited by E. Petzinger, P. K.-H. Hinne and H. Sies. pp. 35-44. Springer-Verlag, Berlin. Haase W., Schafer A., Murer H. and Kinne R. (1978) Studies on the orientation of brush-border membrane vesicles. Biochemical Journal 172, 57~2. InDue M., Kinne R., Tran T. and Arias I. M. (1982) Taurocholate transport by rat liver sinusoidal membrane vesicles: evidence of sodium cotransport. Hepatology 2, 572-579. InDue M., Kinne R., Tran T., Biempica L. and Arias 1. M. (1983) Rat liver canalicular membrane vesicles. Journal of Biological Chemistry 258, 5183-5188. Kessler F., Bennardini F., Bachs O., Serratosa J., James P., Caride A. J., Gazzoti P., Penniston J. T. and Carofoli E. (1990) Partial purification and characterization of the Ca2+-pumping ATPase of the liver plasma membrane. Journal of Biological Chemistry 265, 16012-19019. Loten E. G. and Redshaw-Loten J. C. (1986) Preparation of rat liver plasma membranes in a high yield. Analytical Biochemistry 154, 183-185. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall
R. J. (1951) Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 265-275. Meier P. J., Sztul E. S., Reuben A. and Boyer J. L. (1984) Structural and functional polarity of canalicular and basolateral plasma membrane vesicles isolated in high yield from rat liver. Journal of Cell Biology 98, 991-1000. Novak D. A., Ryckman F. C. and Suchy F. J. (1989) Taurocholate transport by basolateral plasma membrane vesicles isolated from human liver. Hepatology 10, 447-453. Rosario J., Sutherland E., Zaccaro L. and Simon F. R. (1988) Ethinylestradiol administration selectively alters liver sinusoidal membrane lipid fluidity and protein composition. Biochemistry 27, 3939-3946. Schoner W., Von Ilberg C., Kramer R. and Seubert W. (1967) On the mechanism of Na +- and K+-stimulated hydrolysis of adenosine triphosphate. EuropeanJournal oJ Biochemistry 1, 334-343. Sottocasa G. L., Kuylenstierna B., Ernster L. and Bergstrand A. (1967) An electron-transport system associated with the outer membrane of liver mitochondria: a biochemical and morphology study. Journal of Cell Biology 32, 415-438. Wisher M. H. and Evans W. H. (1975) Functional polarity of the rat hepatocyte surface membrane. Isolation and characterization of plasma membrane subfractions from the blood-sinusoidal, bile-canalicular and contiguous surfaces of the hepatocyte. BiochemicalJournal 146, 376-388. Wolters H., Spiering M., Gerding A., Slooff M. J. H., Kuipers F., Hardonk M. J. and Vonk R. J. (1991) Isolation and characterization of canalicular and basolateral plasma membrane fractions from human liver. Biochimica et Biophysica Acta 1069, 61~9.
0887-2333(93)E0004-Y
Toxic. in Vitro Vol. 8, No. 2, pp. 173-180, 1994 Copyright © 1994 ElsevierScienceLtd Printed in Great Britain. All rights reserved 0887-2333/94 $7.00 + 0.00
ISOLATION OF SINUSOIDAL A N D CANALICULAR LIVER PLASMA MEMBRANES: EFFECTS OF FROZEN STORAGE OF H U M A N MATERIAL* R. J. EDWARDS,C. L. LESLIE and N. H. STACEYt Toxicology Unit, National Institute of Occupational Health and Safety, University of Sydney, NSW, Australia (Received 15 March 1993; revisions received 23 July 1993) Abstract--Sinudoisal and canalicular plasma membranes were prepared from rat and fresh or frozen human liver and characterized using plasma membrane marker enzymes. The distributions of these enzymes were used as a means of comparison between the plasma membrane fractions and homogenates from the rat and fresh and frozen human livers. Compared with their homogenate, all the sinusoidal plasma membrane preparations were poorly enriched with Na+-K+-ATPase (2.6~8.4-fold), a sinusoidal marker enzyme. Activities of Na+-K+-ATPase and Mg2+-ATPase were much greater in rat membranes whereas leucine aminopeptidase and alkaline phosphatase activities were much greater in the human membranes. Consideration of the extent of enrichment showed that Mg2+-ATPase and alkaline phosphatase are better markers in rat canalicular plasma membranes and leucine aminopcptidase in human membranes from both fresh and frozen tissue. Intracellular organdie marker enzymes were not enriched in the plasma membranes, thus indicating no significant contamination. Importantly, the relative enrichments of the marker enzymes showed no difference between fresh and frozen human liver. Absolute activities of the enzymes were decreased, however (except alkaline phosphatase), after frozen storage. This work indicates that human liver tissue may be frozen and stored in liquid nitrogen before plasma membranes are isolated with no effect on the distribution of marker enzymes in the membranes.
INTRODUCTION Hepatocytes, the main cellular component of hepatic tissue, are highly polarized epithelial cells. The plasma membranes of hepatocytes are composed of three main surface domains that have been characterized morphologically, functionally and biochemically (Evans and Enrich, 1989; Meier et aL, 1984). They are termed the sinusoidal, lateral and canalicular domains of the hepatocyte plasma membrane. The sinusoidal front and lateral surface of the plasma membrane is the basolateral pole of the hepatocyte. These two domains are also in physical continuum and exposed to sinusoidal blood and as a result are thought to be functionally equivalent. The canalicular domain, the apical pole of the cell, is physically separated from the other two domains by tight junctional complexes and is functionally distinct (Meier et al., 1984). Bile formation, which occurs primarily from the hepatocyte, is a complex process that involves a number of transport mechanisms in both the sinusoidal and canalicular domains of hepatocyte plasma *The views expressed in this article are those of the author and do not necessarily reflect those of the National Occupational Health and Safety Commission (Worksafe Australia). "['To whom correspondence should be addressed. Abbreviation: UW = University of Wisconsin.
membranes. The transport of endogenous chemicals at these sites plays a critical role in the formation of bile. As a result any interference with these transport processes may affect bile output. These transmembrane transport processes can be studied in a system without the interference of intracellular processes present in whole-cell studies. The development of a number of methods that allow the isolation, and in some cases separation, of the sinusoidal and canalicular plasma membrane domains has enabled these studies to take place (Kessler et al., 1990; Loten and Redshaw-Loten, 1986; Meier et al., 1984; Rosario et aL, 1988). Sinusoidal and canalicular plasma membranes isolated from rat liver have been well characterized with respect to their morphology, function and biochemical content (Blitzer and Donovan, 1984; Changchit et al., 1990; Inoue et al., 1982 and 1983; Meier et al., 1984; Rosario et al., 1988; Wolters et al., 1991). However, few studies have investigated the characteristics of human liver plasma membranes (Novak et al., 1989; Wolters et al., 1991), which would be a preferred alternative to rat tissue. One problem associated with the study of human liver plasma membranes is the difficulty of obtaining a regular supply of human liver tissue. A second problem is that when human tissue does become available, it is often in a quantity that could not be used in just one membrane preparation. 173
174
R.J. EDWARDS et al.
Consequently, tissue is wasted. A method of storing this excess tissue, and of alleviating the problem of irregular supply, needs to be developed. As it is unknown whether cryostorage would affect the tissue or the plasma membranes isolated from the tissue, it was also decided to investigate whether storage of human liver tissue in liquid nitrogen would affect the distribution, activity and/or the relative enrichment of the marker enzymes in the sinusoidal and canalicular plasma membrane fractions. MATERIALS AND METHODS
Materials Magnesium chloride (hexahydrate) [AnalaR grade] was obtained from BDH Chemicals Australia (Victoria, Australia). All other chemicals used were of the highest purity commercially available.
(Brinkman Co.). A 200-#1 aliquot of the homogenate was stored at - 2 0 ° C for enzyme analysis. Ice-cold high-purity water (21.5 ml) was added to each tube of homogenate and mixed gently. The homogenate mix was then transferred to 26.3-ml Beckman centrifuge tubes and centrifuged at 45,000g for 30 min in a Beckman model L8-70 ultracentrifuge. The resulting pellet (P1, Fig. 1), after resuspension in 15 ml buffer, was homogenized in a loose Dounce with 8-10 strokes. The homogenate was transferred to Hitachi tubes and 21.5 ml ice-cold 25.2 mM MgC12 was added to each tube for Mg 2+ precipitation of the sinusoidal plasma membrane fraction. The precipitation was followed by centrifugation at 2500 g for 15 min in a Hitachi H I M A C high-speed centrifuge producing a pellet (P2) and supernatant ($2).
(a) Preparation of sinusoidal plasma membranes.
Isolation of plasma membranes from rat liver
The P2 underwent further homogenization (in a loose Dounce), Mg 2+ precipitation and centrifugation at 2500g for 15 min resulting in a crude sinusoidal pellet (P3*). The supernatant ($3") was used in the preparation of canalicular plasma membranes [see part (b)]. The crude sinusoidal pellet, after resuspension in 15 ml buffer, was homogenized with 8-10 strokes in a loose Dounce. The homogenate was transferred to Hitachi tubes to which 21.5 ml water was added. The tubes were mixed vigorously and centrifuged at 750 g for 15 min. The resulting pellet (P4*) consisted of a red layer, representing mitochondria and other intracellular organelles, and a brown layer, containing crude sinusoidal membranes. The supernatant ($4") and loose brown pellet were transferred to Beckman tubes and centrifuged at 45,000g for 30min. The pellet (P5*), after resuspension in 17 ml 50% sucrose (w/w), was homogenized with a medium fit PotterElvehjeim homogenizer using four strokes. The homogenate was overlaid with 5 ml 41% sucrose (w/w) and 15 ml 38% sucrose (w/w) to form a discontinuous sucrose gradient in Beckman Ultra-Clear centrifuge tubes (38.5 ml). The tubes were centrifuged at 48,000g, w2t = 7.18 × l010, in a SW-28 rotor of the Beckman ultracentrifuge for 139 min. The float layer, containing sinusoidal plasma membranes, was carefully harvested from the top of the discontinuous gradient and pelleted by placing the layer in 25 ml NaHCO 3 (1 mM) and centrifuging at 45,000g for 30 min. This resulted in a purified sinusoidal plasma membrane pellet, which was resuspended in 0.5-1.0ml NaHCO3 (1 mM) and stored at -20°C.
The procedure used to isolate plasma membranes from rat liver was based on the one described by Rosario et al. (1988). The complete method was as follows. The rats were anaesthetized using ether, and their livers (weighing 8-10 g) removed and minced thoroughly. All procedures were performed on ice. Minced liver portions (l.8g), placed in 45.5-ml Hitachi centrifuge tubes, were homogenized in 15 ml buffer (300mM mannitol, 5mM EGTA, 15mM Tris-HCI, 0.1 mM phenylmethane sulfonyl fluoride at pH 7.4) for 45 sec at 4800 rpm using a polytron
The supernatant from the first Mg 2+ precipitation, $2, was centrifuged at 45,000g for 30 min to obtain a crude canalicular fraction (P3). The P3 was resuspended with the $3", from the sinusoidal plasma membrane preparation pathway, and centrifuged at 45,000 g for 30 min to obtain a greater crude canalicular fraction (P4). The P4 was resuspended in 15 ml buffer and homogenized using a tight fit PotterElvehjeim homogenizer (Wheaton) with five strokes. To the homogenate 21.5 ml MgCl2 (25.2mM) was
Human livers Human livers were obtained from three organ donors from whom most of the liver was used for transplantation. Livers were stored in University of Wisconsin (UW) solution, on ice, until the start of an isolation procedure or until storage in liquid nitrogen. Livers were in UW for no longer than 2 hr. Use of human tissue was approved by the Ethics Review Committee of the Central Sydney Health Service and with the co-operation of the Royal Prince Alfred Hospital.
Freezing and thawing of human liver tissue Human livers were divided into 10-g portions, drained of UW solution and securely wrapped in aluminium foil. Each parcel of liver was then frozen and stored in liquid nitrogen. Before each isolation procedure a frozen portion of human liver was thawed at room temperature and then placed on ice ready for the isolation procedure.
Animals Male PVG/c hooded rats (200--250g), obtained from Blackburn Animal House (University of Sydney, Australia), were used in all experiments. The rats were housed in cages of not more than five per cage with free access to food and water. The rats were fasted overnight before the isolation procedure.
(b) Preparation of canalicular plasma membranes.
Liver plasma membranes: frozen storage
175
MINCED LIVER
Polytron homogenization
[ HOMOGENATE ]
a45,000g x 30 min
Mg 2÷ precipitation (pptn) 2500g x 15 min
Mg 2+ pptn 2500g x 15 min
45,000g x 30 min
+$3
Water "cut" 750g x 15 min
45,000g
30 min
Discard
45,000g x 30 min
Mg 2+ pptn 2500g x 20 min
(~
Discontinuous sucrose gradient (50%/41%/38%) 48,000g x 139 min
Floatlayer on 38% sucrose surface
I
Discard
45,000g x 30 min
C A N A L I C U L A R PLASMA MEMBRANES
45,000g x 30 min
Discard SINUSO1DAL PLASMA MEMBRANES Fig. 1. Method for the preparation of plasma membranes from rat or human liver, a: centrifugal speed (g) and time (min); b: S = supernatant; c: P = pellet; *S or P from the sinusoidal plasma membrane preparation pathway.
176
R . J. E D W A R D S
et al.
18-16-.~,_, 14
_
~
12-
,,~ 8
10--
÷1
80
B
[]
Rat
1
Human
4--
60
I
~ "~ o
50
O ~
40
+
10
A¢1
0
H
s
C
8-
I
o
.~.
6--
2--
H :~
10
nn S
-~
50--
o = ~'~
40
H
S
C
=., =-
30
o
20--
o
~
--
10--
A
c
c
3.0--
-
•~8
S
D
C
•~ ' Z
H
60 --
10--
~
5_
70
~1.
Z
D
A
F 2.5--
-
2.0-~'~.
9...
4
2 < Z
7"1
6 -
0
-
-
nn
A s
S
om.
1.5
oo E O
1.0
o
/ IA IA /A /A /A /A
0.5
,',1
0.0 C
,A ,4 /4 /4
3_
H
S
c
Fig. 2. Marker enzyme activities in homogenate, sinusoidal and canalicular plasma membrane fractions prepared from rat and fresh human livers. (A) Na+-K+-ATPase; (B) Mg2+-ATPase; (C) leucine aminopeptidase; (D) alkaline phosphatase; (E) NADPH cytochrome c reductase; (F) succinate cytochrome c reductase. (H = homogenate; S = sinusoidal fraction; C = canalicular fraction). Note different ordinate scale of each graph. Each column represents mean value of triplicate experiments + SE. added, for precipitation o f any c o n t a m i n a t i n g sinusoidal plasma m e m b r a n e s , followed by centrifugation at 2 5 0 0 g for 2 0 m i n . The s u p e r n a t a n t ($5) was transferred to B e c k m a n tubes a n d centrifuged at 4 5 , 0 0 0 g for 30 m i n a n d the resulting pellet resuspended in 0 . 5 - 1 . 0 m l N a H C O 3 (1 mM). This pellet consisted o f purified canalicular plasma m e m b r a n e fraction, a n d was stored at - 2 0 ° C .
Isolation of plasma membranes from human liver Portions of h u m a n liver (weighing approximately 10 g) were cleared as extensively as possible o f vascular a n d connective tissue. Isolation of sinusoidal and canalicular m e m b r a n e s from the tissue was achieved by following the same isolation procedure as t h a t for m e m b r a n e isolation from rat liver.
Liver plasma membranes: frozen storage
Enzyme assays and protein determination A n u m b e r o f m a r k e r enzymes were used to test for m e m b r a n e separation. T h e enzyme N a + - K +-ATPase was used as a m a r k e r for the sinusoidal fraction of
177
plasma m e m b r a n e s since its activity is k n o w n to be restricted to the basolateral plasma m e m b r a n e surface o f hepatocytes (Blitzer a n d Boyer, 1978). The activity of N a + - K + - A T P a s e was measured by the m e t h o d o f Schoner et al. (1967). T h r e e m a r k e r
16
6
-
A
B
14
5 m
10 - -
[]
Fresh
[]
Frozen
'~
4
<~ ~
~
3
//
~
// // // // // // // //// //
4
Z
1
o H
S
C
H
10 --
D
I
50
8 -
~2 "=
6
,5~
4
'~'~
C
60
J_
C ..>
S
o = ~'~ eu
40
~
30
,~-~ ~
20
<
10
v, Ii /i
o
2
ii
m
.3 H
lln
Z i/ // I¢
Z
m
i/ I
,
s
H
3.0
m
..>
F
2.5
8 --
"~ 2.0 ~8
~
:
I
1.5
i ,.o
1i
~ z
S
o
0.5
~ 0.0 S C H S C Fig. 3. Marker enzyme activities in homogenate, sinusoidal and canalicular plasma membrane fractions prepared from fresh human and frozen human livers. (A) Na+-K+-ATPase; (B) Mg2+-ATPase; (C) leucine aminopeptidase; (D) alkaline phosphatase; (E) NADPH cytochrome c reductase; (F) succinate cytochrome c reductase. (H = homogenate; S = sinusoidal fraction; C = canalicular fraction). Note different ordinate scale of each graph. Each column represents mean value of triplicate experiments + SE.
178
R.J. EDWARDSet al. Table 1. Relative enrichment* of marker enzyme activities in sinusoidal and canalicular membrane fractions prepared from rat, fresh human and frozen human liver Rat Marker enzyme Na+-K+-ATPase Mg2+-ATPase Leucine aminopeptidase Alkaline phosphatase N A D P H cytochrome c reductase Succinate cytochrome c reductase
Fresh human
Thawed human
SPM
CPM
SPM
CPM
SPM
CPM
8.4 6.5
5.8 21.8
2.6 1.1
4.5 2.3
2.7 I.I
5.4 2.4
2.6 7.2
7.2 28.0
5.2 3.4
15.0 10.2
4.1 3.3
15.6 7.9
0.7
0.4
1.5
1.4
1.0
0.7
0.6
0.6
1.6
0.2
1.1
0.2
CPM = canalicular plasma membranes SPM = sinusoidal plasma membranes *Relative enrichment is expressed as specific activity of membrane/specific activity of homogenate. Values represent ratio of the means of specific activities where n = 3.
enzymes were used to identify the canalicular plasma membrane: Mg2+-ATPase, leucine aminopeptidase and alkaline phosphatase. These enzymes are known to be predominantly active in the canalicular plasma membrane but not exclusively located at that site (Evans, 1970; Wisher and Evans, 1975). Mg2+-ATPase activity was measured by the method of Schoner et al. (1967). Alkaline phosphatase activity, using p-nitrophenyl phosphate as substrate, was measured according to Bessey et al. (1946). Leucine aminopeptidase activity was measured by the method of Haase et al. 0978). Contamination with mitochondria and microsomes was measured by N A D P H cytochrome c reductase and succinate eytochrome c reductase activities, respectively (Sottocasa et al., 1967). Protein was estimated by the method of Lowry et aL (1951) using bovine serum albumin as standard. All enzyme activities were measured at 37°C following overnight storage of the plasma membrane fractions at 20oc. -
RESULTS
In membranes prepared from rat livers the activity of Na+-K+-ATPase was highest in the sinusoidal fraction (Fig. 2A). Activities were less in preparations from human liver, Mg2+-ATPase activity (Fig. 2B) was greatest in the canalicular membrane fractions from rat liver especially when compared with human liver values. Figure 2C shows that leucine aminopeptidase activity was greatest in the canalicular membrane fractions prepared from both rat and human livers, with activity being two-fold higher in the latter. Activity of the third canalicular marker enzyme, alkaline phosphatase (Fig. 2D), was greatest in the canalicular fr~/ctions of membranes prepared from both rat and human livers. As with the enzyme leucine aminopeptidase, alkaline phosphatase activity was much greater in the membranes prepared from human tissue.
During the isolation procedure it was possible for the membrane fractions to be contaminated with other cell components, for example mitochondria and microsomes. N A D P H cytochrome c reductase was used as a marker enzyme for mitochondria (Fig. 2E), and succinate cytochrome c reductase for mtcrosomes (Fig. 2F). N A D P H cytochrome c reductase activity was higher in the homogenate, sinusoidal and canalicular fractions prepared from human livers than in the corresponding fractions prepared from rat livers. Succinate cytochrome c reductase activity in membrane fractions prepared from both rat and human livers was low. The values show that there were no increases to suggest significant contamination with other membranes. A comparison of the activities of the marker enzymes in the membrane fractions prepared from fresh and frozen human liver tissues showed that there was a decrease in most of the enzymes' activities in the thawed tissue membranes (Fig. 3A-F), except for alkaline phosphatase (Fig. 3D). Table l presents the relative enrichments of the marker enzymes' activities measured in the sinusoidal and canalicular fractions of both rat and human tissue (fresh and frozen). Relative enrichment is defined as the specific activity of a given enzyme in a membrane fraction divided by the specific activity of this enzyme in the homogenate. The sinusoidal marker enzyme Na ÷-K ÷-ATPase showed greater enrichment in the sinusoidal fraction for rat liver, while fresh and frozen human liver showed a small but greater enrichment in the canalicular membranes. The canalicular marker enzymes, Mg2+-ATPase and alkaline phosphatase, showed greater enrichment in rat canalicular fractions, while leucine aminopeptidase enrichment was greater in human canalicular membrane fractions. The relative enrichments for membranes from fresh and frozen human liver tissue showed remarkable similarity for all marker enzymes including those of other organelles.
Liver plasma membranes: frozen storage DISCUSSION
179
Wolters et al. (1991), which may account for some of the differences in the human data. Alkaline phosphaEnzymatic characterization, using six marker entase activity was much greater in human homogenzymes, was used as a method of comparing liver ates compared with rat homogenates. This may be plasma membranes from rats and humans and, more due to the fact that in the human liver this enzyme is importantly, for fresh v. frozen human liver. The located in at least two different sites on the hepatosinusoidal marker enzyme Na+-K+-ATPase was cyte, whereas in rat livers it is located at only one site poorly enriched in the sinusoidal plasma membranes (Wolters et al., 1991). Na+-K+-ATPase and Mg 2--from all three liver sources--rat, and fresh and frozen ATPase activities were greater in rat membranes, human tissue. Overall, the activities and enrichments whereas leucine aminopeptidase and alkaline of the marker enzymes for the canalicular fraction phosphatase activities were greater in the human showed suitable enrichment. The plasma membrane membranes. However, enrichment of leucine fractions were not enriched with enzymes from other aminopeptidase was greater in human membranes intracellular organelles. Comparison of plasma mem- and alkaline phosphatase in rat membranes. These branes from rat and fresh human liver sources results indicate that Mg2+-ATPase would be the highlighted some differences. However, numerous preferred marker in rat plasma membranes and similarities were also observed. Lower activities of leucine aminopeptidase in human plasma memmost marker enzymes were measured in the plasma branes. Alkaline phosphatase is a useful second membranes from frozen human liver compared with marker for both species. those from fresh human liver. However, in general, The comparison of enzymatic characteristics of the enrichments of the marker enzymes were the membranes prepared from fresh and frozen human same. There were greater variations in the human livers showed that each enzyme, except alkaline phosplasma membrane preparations than in rat prep- phatase, was lower in the homogenate and the plasma arations, and this may be explained by the greater membrane fractions prepared from frozen human number of non-controllable factors present in human liver. The intracellular organelle marker enzymes tissue studies, that is, sex, age, diet and so on. activities were also reduced in the frozen liver plasma Na+-K÷-ATPase activity in rat sinusoidal plasma membranes. However, even when absolute levels were membranes was enriched eight-fold over homogenate decreased, relative enrichments stayed the same after in this study. The enrichment was much lower than frozen storage. This indicates that the loss of activity the 20-35-fold enrichment of Na+-K+-ATPase ac- was evenly distributed across the surface of the tivity in sinusoidal membranes reported in the litera- human hepatocyte and did not affect specific plasma ture (Blitzer and Donovan, 1984; Changchit et al., membrane domains. 1990; Inoue et al., 1982; Meier et al., 1984; Rosario In conclusion, the activities and distributions of et al., 1988; Wolters et al., 1991). In this study, some membrane marker enzymes were different in although Na÷-K÷-ATPase activity in the sinusoidal plasma membranes from rat liver compared with fraction was the same as that reported by Meier et al. plasma membranes from human liver. More impor(1984) and Inoue et al. (1982), activity of the enzyme tantly, it has been shown that distributions and in the homogenate was much higher than they re- relative enrichments of marker enzymes were unported, which would account for the lower relative changed in plasma membranes from human liver enrichments. Despite the same method of plasma after frozen storage. Thus, it should be acceptable to membrane isolation being used in this study and by use human liver samples that have been stored in Rosario et al. (1988), different marker enzyme activi- liquid nitrogen as a source of tissue for plasma ties and enrichments were obtained. In the Rosario membrane isolation. Increasing the availability and et al. (1988) study Na+-K+-ATPase activities were use of human tissues for in vitro studies has advanmuch greater than those reported by Meier et al. tages for both reduced use of laboratory animals and (1984) and Inoue et al. (1982) or in our study. Despite relevance of the tissue used to the ultimate species of the high activities reported by Rosario et al. (1988) concern. the relative enrichment of Na+-K÷-ATPase was the same as that reported by others (Inoue et al., 1982; Acknowledgement--This work was supported in part by a Meier et al., 1984). The precise reasons for these National Health and Medical Research Council grant. inconsistencies from various laboratories remain to be determined. REFERENCES The activities of the three canalicular marker enBessey O. A., Lowry O. H. and Brock M. J. (1946) A zymes (Mg2+-ATPase, leucine aminopeptidase and method for the rapid determination of alkaline phosphaalkaline phosphatase) were well enriched in the tase with 5 cm3 of serum. Journal o f Cell Biology 164, 321 329. canalicular fraction, although some were lower than those published elsewhere (Inoue et al., 1983; Meier Blitzer B. L. and Boyer J. L. (1978) Cytochemical localization of (Na +-K ÷ )ATPase in the rat hepatocyte. Journal et al., 1984; Rosario et al., 1988; Wolters et al., 1991). o f Clinical Investigation 62, 1101-1104. It should be noted that different methods of plasma Blitzer B. L. and Donovan C. B. (1984) A new method for the rapid isolation of basolateral plasma membrane membrane isolation were used in our study and by
180
R.J. EDWARDS et al.
vesicles from rat liver. Journal of Biological Chemistry 259, 9295-9301. Changchit A., Durham S. and Vore M. (1990) Characterization of [3H]estradiol- 17fl-(//-o-glucuronide) binding sites in basolateral and canalicular liver plasma membranes. Biochemical Pharmacology 40, 1219-1225. Evans W. H. (1970) Fractionation of liver plasma membranes prepared by zonal centrifugation. Biochemical Journal 166, 833-842. Evans W. H. and Enrich C. (1989) The hepatocyte's plasma membrane domains. Interrelations with the endocytic compartment. In Hepatic Transport in Organic Substances. Edited by E. Petzinger, P. K.-H. Hinne and H. Sies. pp. 35-44. Springer-Verlag, Berlin. Haase W., Schafer A., Murer H. and Kinne R. (1978) Studies on the orientation of brush-border membrane vesicles. Biochemical Journal 172, 57~2. InDue M., Kinne R., Tran T. and Arias I. M. (1982) Taurocholate transport by rat liver sinusoidal membrane vesicles: evidence of sodium cotransport. Hepatology 2, 572-579. InDue M., Kinne R., Tran T., Biempica L. and Arias 1. M. (1983) Rat liver canalicular membrane vesicles. Journal of Biological Chemistry 258, 5183-5188. Kessler F., Bennardini F., Bachs O., Serratosa J., James P., Caride A. J., Gazzoti P., Penniston J. T. and Carofoli E. (1990) Partial purification and characterization of the Ca2+-pumping ATPase of the liver plasma membrane. Journal of Biological Chemistry 265, 16012-19019. Loten E. G. and Redshaw-Loten J. C. (1986) Preparation of rat liver plasma membranes in a high yield. Analytical Biochemistry 154, 183-185. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall
R. J. (1951) Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 265-275. Meier P. J., Sztul E. S., Reuben A. and Boyer J. L. (1984) Structural and functional polarity of canalicular and basolateral plasma membrane vesicles isolated in high yield from rat liver. Journal of Cell Biology 98, 991-1000. Novak D. A., Ryckman F. C. and Suchy F. J. (1989) Taurocholate transport by basolateral plasma membrane vesicles isolated from human liver. Hepatology 10, 447-453. Rosario J., Sutherland E., Zaccaro L. and Simon F. R. (1988) Ethinylestradiol administration selectively alters liver sinusoidal membrane lipid fluidity and protein composition. Biochemistry 27, 3939-3946. Schoner W., Von Ilberg C., Kramer R. and Seubert W. (1967) On the mechanism of Na +- and K+-stimulated hydrolysis of adenosine triphosphate. EuropeanJournal oJ Biochemistry 1, 334-343. Sottocasa G. L., Kuylenstierna B., Ernster L. and Bergstrand A. (1967) An electron-transport system associated with the outer membrane of liver mitochondria: a biochemical and morphology study. Journal of Cell Biology 32, 415-438. Wisher M. H. and Evans W. H. (1975) Functional polarity of the rat hepatocyte surface membrane. Isolation and characterization of plasma membrane subfractions from the blood-sinusoidal, bile-canalicular and contiguous surfaces of the hepatocyte. BiochemicalJournal 146, 376-388. Wolters H., Spiering M., Gerding A., Slooff M. J. H., Kuipers F., Hardonk M. J. and Vonk R. J. (1991) Isolation and characterization of canalicular and basolateral plasma membrane fractions from human liver. Biochimica et Biophysica Acta 1069, 61~9.