- Email: [email protected]
Hepatic and extrahepatic uptake of intravenously injected lipoprotein lipase
Biochimica Elsevier
et Biophysics
513
Acta, 795 (1984) 513-524
BBA 51751
HEPATIC AND EXTRAHEPATIC LIPOPROTEIN LIPASE
UPTAKE OF INTRAVENOUSLY
INJECTED
LARS WALLINDER, OLIVECRONA
THOMAS OLIVECRONA
BENGTSSON-
Department
JONAS PETERSON,
of Physiological
Chemistry,
and GUNILLA
University of Urn&, S-9OI 87 Urn& (Sweden)
(Received April 6th, 1984)
Key words: Lipoprotein
lipase; Enzyme
uptake
Rats were injected intravenously with 12’I-labeled bovine lipoprotein lipase. The lipase disappeared within minutes from the blood due to uptake both in the liver (about 50% of the injected dose) and in extrahepatic tissues. Lipase enzyme activity disappeared in parallel to the ‘=I radioactivity. Thus, there was no inactivation of lipase in the circulating blood. Similar results were obtained when lipoprotein lipase purified from guinea pigs was injected into guinea pigs. Using supradiphragmatic rats we could show that the extrahepatic uptake was saturable and that the amounts of lipase that could be bound far exceeded the amounts of endogenous lipase expected to be present on the endothelium. When the lipase was denatured before injection, its removal in supradiaphragmatlc rats became slower, and in intact rats the fraction, of the uptake that occurred in extrahepatic tissues was much decreased. It is concluded that recognition by the extrahepatic receptors depends on the native conformation of the lipase. The extrahepatic uptake was strongly impeded by injection of heparin prior to injection of the lipase, and the uptake could to a large extent be reversed by injection of heparin after the lipase. Even after 1 h lipase that had been taken up by extrahepatic tissues reappeared immediately in the blood on injection of heparin. This was true both for enzyme activity and for enzyme radioactivity. Thus, internalization-inactivation-degradation occur only slowly in extrahepatic tissues. It is possible that the extrahepatic binding occurs to the enzyme’s physiological receptors. The hepatic uptake was not dependent on the native conformation of the lipase, was less sensitive to heparin, could not be reversed by heparin and was not saturable. The enzyme was not rapidly inactivated after uptake; its activity could be detected in liver homogenates even after 1 h. Degradation to acid-soluble products in the liver was relatively slow; the t,,2 for native lipase was about 1 h. In comparison, in parallel experiments asialofetuin was degraded with a t,,, of about 15 min
Introduction Lipoprotein lipase hydrolyzes acylglycerols in plasma lipoproteins (for review see Refs. 1 and 2). This reaction takes place at the vascular endothelium in some extrahepatic tissue. The enzyme is synthesized in other cells in the tissue and is then transferred to the endothelium. It has been proposed that the lipase is held in place at the endothelial cell surface via interaction with membrane0005-2760/84/$03.00
0 1984 Elsevier Science Publishers B.V.
bound heparan sulfate chains [3]. This model has gained support from studies with cultured endothelial cells [4-61, which have shown that enzymatic degradation of cell surface heparan sulfate impedes binding of lipoprotein lipase to the cells. Little or no lipoprotein lipase activity is found in the circulating blood. One would expect that if lipoprotein lipase was injected into the circulation it would bind to its endothelial receptors. However, in a previous study we found that most of the
514
injected lipase is taken up by the liver and degraded [7]. A minor part was taken up in extrahepatic tissues. Thus, there are receptors in the liver which override the extrahepatic ones in clearing the enzyme from the blood. The hepatic receptors for lipoprotein lipase are not identical with any of the known glycoprotein receptors [7]. Ehnholm and his associates [8] have studied the metabolism of lipoprotein lipase in hepatectomized swine. In their study, endogenous lipoprotein lipase was released into the blood by heparin injection. As the heparin disappeared from the blood, so did the lipase activity, directly demonstrating the existence of extrahepatic binding sites. The aim of the present study was to explore, in more detail, the nature of the hepatic and the extrahepatic uptake of lipoprotein lipase, and the possible differences between the mechanisms involved. To this end we have followed the fate of catalytically active as well as inactive forms of the lipase in intact and in supradiaphragmatic rats. We have also studied the effect of heparin on the uptake mechanisms. Materials and Methods Animals Male Sprague-Dawley rats weighing 200-350 g (AB Anticimex, Stockholm, Sweden) were fed a standard pellet diet (AB Ewos, Siidertalje, Sweden). Male guinea pigs (400-500 g) were from a local laboratory farm. The rats were under light ether anaesthesia during the experiments, while the guinea pigs were given intramuscular injections of Hypnorm Vet.” (AB Leo, Sweden, 1 mg/kg body weight). Intravenous injections were made in the exposed right jugular vein. Blood samples (200 ~1 in 1 ml plastic syringes, one for each blood sample) were withdrawn from the left jugular vein and collected in weighed tubes. In the experiments with heparin a standard dose of 100 ~1 was injected (Lowen, Malmo, Sweden 5000 IE/ml). Immediately following the last blood sample the animals were exsanguinated through the abdominal aorta. The tissues were cut out and weighed and their total radioactivities were determined in an LKB Wallac 1275 Minigamma counter. The data were calculated on the basis of a blood volume of 55 g/kg body weight [7].
For measurement of lipoproteiu lipase activity in the livers, 10% (w/v) homogenates were prepared in 0.025 M ammonia buffer (pH 8.2) containing 5 IU heparin/ml, 1 pg of leupeptin and of pepstatin/ml, 25 IU aprotinine/ml and 5 mM EDTA. The proteinase inhibitors were from Boehringer-Mannheim, Mannheim, F.R.G. .A Polytron homogenizer was used. Supradiaphragmatic rats were prepared according to Bezam-Tarcher and Robinson [9& but without dividing the upper part from the lower part of the body. For these experiments the rats were anesthetized with Hypnorm Vet.a. Enzymes Lipoprotein lipase was purified from bovine milk and from guinea pig milk as previously described [lO,ll]. The enzymes were radiolabeled using the lactoperoxidase/glucose oxidase method. To 2 ml purified bovine lipase (0.3 mg/ml) in 50 mM Tris-HCl (pH 7.4)/1.5 M NaCI/20 mM glucose, was added 10 ~1 of lactoperoxidase (5 mg/n^nl Boehringer-Mannheim), 1 mCi of Na1251 (carrierfree, 0334 New England Nuclear, Dreieich, F.R.G.) and 10 ~1 glucose oxidase (1 mg/ml, Boehringer-Mannheim) in this order. The mixture was incubated for 25 min on ice. The reaction was stopped by adding 6 ml of 10 mM Tris-HCl (pH 7.4)/0.01 M Kl/O.O25 M NaN,. The mixture was applied to a 3 ml heparin-Sepharose column, which had been equilibrated with 10 mM Tris-HCl (pH 7.4)/2 mg/ml bovine serum albumin/O.5 M NaCl. Unreacted iodine was washed away with 30 ml of the equilibration buffer at a flow rate of 1 ml/mm. The enzyme was eluted with a 140 ml linear salt gradient from 0.5 to 2.0 M NaCl in 10 mM Tris-HCl (pH 7.4) with 2 mg bovine serum albumin/ml. 4-ml fractions were collected. Catalytically active lipoprotein lipase eluted at about 1 M NaCi; denatured, inactive lipase eluted at lower salt concentrations. More than 98% of the radioactivity in the pooled peak ‘fractions (eluting at about 1 M NaCl) could be precipitated by trichloroacetic acid. A typical lipase preparation had a specific radioactivity of 1000 cpm/ng as determined by an immunoassay [12]. Guinea pig lipoprotein, lipase was iodinated and purified using the same,procedure. The labeled preparations were divided in portions of 200 ~1 and stored at - 20 0 C. They
515
were used for experiments within 1 month. On longer storage the enzyme gradually changed behavior gaining properties resembling denatured lipase. For some experiments the labeled enzyme was denatured by dialysis against 6 M guanidinium chloride (GdmCl) in 10 mM Tris-Cl (pH 8.5) at 4 o C overnight. The GdmCl was then removed by a 4 h dialysis at 4’C against 0.5 M (NH,),S0,/0.2 M NaCl/lO mM Tris-HCl (pH 7.4). These preparations contained no measurable lipase activity. Prior to injection the labeled enzyme preparations were diluted in an equal volume of titrated rat plasma at 0 o C. 0.25 ml was injected per animal, corresponding to 1-3 pg lipase, (l-3). lo6 cpm. Samples of 0.2 ml from the same syringe were directly injected into weighed counting tubes and their radioactivity was determined. From these samples the injected dose of radioactivity was calculated. Other labeled materials Rat erythrocytes were labeled with Nay CrO, (New England Nuclear) during a 2h incubation at 37 CJC. They were washed three times in 0.15 M NaCl and then used the same day. 1 ml of sedimented erythrocytes contained about lo6 cpm. Erythrocytes corresponding to 150000 cpm were diluted in saline and then injected 2 min prior to injection of lipase. The 51Cr and the lz51 radioactivites in the blood samples were determined in the LKB Minigamma counter. 17% of the 51Cr counts were also detected in the lz51 channel but were subtracted to give the net I*‘1 counts. Fetuin (type IV, Sigma, St. Louis, MO, U.S.A.) was labeled with the lactoperoxidase,’ glucose oxidase method as described above and then separated from unreacted iodine by chromatography on a Sephadex G-25 column in 50 mM Tris-HCl/ 0.15 M NaCl (pH 7.4). To prepare asioalofetuin the labeled protein was subjected to mild acid hydrolysis in 0.05 M H, SO, at 80 o C for 60 min to remove terminal sialic acid residues [13]. After dialysis against 50 mM Tris-HCl/O.lS M NaCl (pH 7.4), 95% of the radioactivity could be precipitated with trichloroacetic acid. The asialofetuin thus obtained had a specific radioactivity of 900 cpm/ng. About 3 pg, diluted in rat citrate plasma, were injected into the rats.
Antiserum against hepatic hpase Hepatic lipase was purified from heparin perfusates of rat livers [14]. The lipase in perfusates from 10 rats was precipitated by an equal volume of saturated ammonium sulfate and then solubilized in 10 mM Tris-HCl (pH 7.4)/20% glycerol/ 0.2% Triton X-100. This solution was applied to a heparin-Sepharose column and eluted with a gradient of NaCl in the same buffer. The peak fractions of activity were combined, diluted and then applied to a column of immobilized heparin, which had been modified by N-desulfation followed by acetylation [ll]. The lipase was eluted with a gradient of NaCl and the material thus obtained was used for immunization of a rabbit. The antiserum obtained inhibited rat hepatic lipase activity both in liver homogenates and in postheparin plasma but did not inhibit rat or bovine lipoprotein lipase activity. On Western blotting of SDS-polyacrylamide gels the antiserum reacted with two components in the purified hepatic lipase preparation. Assay procedure Lipoprotein lipase activity was assayed, using 3H-labeled Intralipid@ (a kind gift from AB Vitrum, Stockholm, Sweden). To 10 pl Intralipid (100 mg triacylglycerol/ml) were added 40 ~1 heat-inactivated human serum and 100 ~1 of a mixture containing 12% bovine serum albumin/30 IU heparin/0.2 M NaC1/0.3 M Tris-HCl (pH 8.5). The total volume was adjusted to 200 ~1 with distilled water. The incubations were carried out for 30-120 min in a 25 “C water bath. The incubation was stopped and the fatty acids were extracted as described previously [15]. The assay was linear with time and with the amount of enzyme in the range used in the experiments. For measurements of lipoprotein lipase activity in liver homogenates and in post-heparin plasma the hepatic lipase activity was inhibited by a prior incubation of the samples for 2 h on ice with an equal volume of anti-hepatic lipase antiserum. No hepatic lipase activity could be detected in these samples even when using an assay optimized for hepatic lipase [16]. The data on lipoprotein lipase activity in liver homogenates are given as the difference between the activities in rats which had been injected with lipoprotein lipase and the activities in control rats.
516
latter were hardly detectable and never amounted to more than 5% of the activities in the injected rats.
The
Results
To determine whether the ‘251-labeled lipdprotein lipase behaved in the same way as catalytically active unlabeled lipase, we injected a trace amount of labeled and a relatively large amount of unlabeled lipase (0.2 mg) intravenously into rats (Fig. 1). Both radioactivity and lipase activity were rapidly cleared from the blood; less than 5% of the injected doses remained after 10 min. The disap-
20
40
60
Time (min) Fig. 1. Disappearance of lipoprotein lipase enzyme activity (O,O, A) and ‘251-radioactivity (0, 1, A) after intravenous injection to rats and reappearance of active enzyme in the blood after injection of heparin. A mixture of unlabeled lipoprotein lipase (200 sg) and 1251-labeled lipase was injected into three rats. Injections of heparin were made at the times indicated by the arrows. The blood samples were collected in heparinized tubes and their radioactivity was determined. They were then centrifuged and the plasma lipoprotein lipase activity was determined after inhibition of the hepatic lipase activity with antiserum (see Methods). The endogenous lipoprotein lipase activity released by heparin was determined in corresponding rats which had not been given bovine lipase. It amounted to 15% of the injected dose of bovine lipase activity and was subtracted from the values presented in the figure.
pearance curves were almost identical, demonstrating that the iodination had not altered the behavior of the enzyme. As shown previously 171, injection of heparin caused reappearance in the blood of radioactively labeled lipase. In the experiment in Fig. 1 about 35% of the injected dose reappeared on injection of heparin PO min after injection of the lipase. A corresponding amount of catalytic activity reappeared (Fig. 1). Thus, no inactivation of the enzyme had occurred on uptake into the heparin-releasable pool. This was true even at longer times after the injection; heparin at 60 min released 13% of the radioactivity and 15% of the catalytic activity. Before injection of heparin little or no endogenous lipoprotein lipase was present in the blood of the rats. The lipoprottin lipase activity released by heparin corresponded to about 15% of the injected dose of bovine lipase. This activity was subtracted from the values in Fig. 1. For practical reasons we used here a heterologous experimental model; bovine lipoprotein lipase was injected into rats. In a few experiments we have used a homologous model with ‘251-1abeled lipoprotein lipase purified from guinea pig milk injected into guinea pigs. Fig. 2 shows that the results were very similar to those obtained with bovine lipase in rats. 50% of the injected dose disappeared within I min. At PO min less than I5% of the radioactivity remained in the blood. 20% of the injected dose reappeared in the blood on heparin injection after 15 min. Similar results were obtained when guinea pig lipoprotein lipase was injected into rats (Fig. 2). Thus, we conclude that the rapid clearing of lipoprotein lipase from the blood is not because it is recognized as a foreign protein by the rats but for some other physiological reason. To determine whether the native lipase conformation was required for recognition by the uptake mechanism(s) i2’ I-labeled lipase was denatured with 6 M GdmCl. The GdmCl was removed by dialysis (see Methods) and the material was then injected into rats. In all experiments the denatured lipase (Fig. 3a) disappeared more rapidly from the blood than the native lipase did (Fig. 3b). Hegarin injected at different times after the lipase caused reappearance of less of the denatured lipase than of the native lipase. At 15 min 35% of the injected native lipase reappeared, while only 5% of the
517
5
10
15
Time (mid
Fig. 2. Disappearance curves for 1251-labeled guinea pig lipoprotein lipase after intravenous injection into guinea pigs (0, W) and into rats (0,O). Two animals of each species were injected. The arrows indicate injection of heparin into one animal of each species.
denatured lipase reappeared. At 30 min hardly measurable amounts of denatured lipase reappeared, while 30% of the native lipase was still available for release by heparin. Thus, in contrast to native lipase most of the .denatured lipase had been taken up by mechanisms that were not reversed by heparin. The effect of injection of heparin immediately before injection of 1251-labeled lipase is shown in Fig. 4. After an equilibration phase the disappearance of both native and denatured lipase was considerably slower than it was in the absence of heparin (compare with Fig. 3). The half-times calculated from the second slower parts of the curves were 20 min and 10 min, respectively, for native and denatured lipase in the presence of heparin. 60% of the uptake of native lipase and 85% of the uptake of denatured lipase occurred in the liver (Table I). When heparin was injected before the lipase a higher proportion of native lipase was taken up in the liver (72.8 f 5.6%) than in the absence of heparin. The proportion of the denatured lipase taken up by the liver was not significantly altered by heparin. Considerable amounts of native lipase were
b) a)
Time hid
I
I
I
10
20
30
Time (mid
Fig. 3. Disappearance in rats of i.v. injected ‘251-labeled bovine lipoprotein lipase. (a) Lipase denatured by GdmCl as described in Methods. (b) Native lipase. Heparin was injected 3, 5, 10, 1.5 and 30 min after the injection of the lipase.
L
relative contribution by the heart to the tota”l uptake was small (about 1.5%) but this uptake showed some conspicious features. It was reduced by more than 50% by heparin and the proportional uptake of denatured lipase was only about one-tenth that of native lipase. Similar effects were seen also in other extrahepatic tissues. Taken together they explain why more of the uptake occurred in the liver in the presence of heparin and why more of the denatured lipase was directed to the liver. The results described above suggested that there are at least two routes of uptake for lipoprotein lipase from the circulation; one into the liver and another one into extrahepatic tissues. To study the extrahepatic uptake directly we injected lipase into supradiaphragmatic rats. Fig. 5 shows that the disappearance of native lipase from the circulation of the upper part of the body was slower than the disappearance in the intact rat. In the experiments on supradiaphragmatic rats we noted that there was a decrease of radioactivity per volume of blood of injected 1251-labeled albumin and of 5’Cr-labeled erythrocytes during the course of the experiment (Fig. 5). The effect was most marked during the first S-10 min after the injection. The lipase radioactivity per volume of blood decreased more rapidly than the 51Cr-radioactivity, demonstrating a true disappearance of the lipase from the circulation into the tissues (Fig. 5). In all smbsequent experiments on supradiaphragmatic rats the
I
_I-
10
5
Time
(min)
Fig. 4. Effect of heparin injected prior to injection of native (0) or denatured (0) ‘251-Iabeled bovine lipoprotein lipase into rats. The heparin was injected 2 min before the enzyme.
taken up also by extrahepatic tissues; about 40%. This radioactivity was distributed all over the body. The uptake in the heart is shown in Table IA. The
TABLE
I
TISSUE DISTRIBUTION INTO RATS
OF RADIOACTIVITY
10 MIN
AFTER
INJECTION
OF ‘25i-LABELED
LIPOPROTEIN
LIPASE
In B, heparin was injected 2 min before injection of the lipase. The lipase was denatured with GdmCl as described in Methods. due to the blood conten! of the “Cr-labeled erythrocytes were injected into the rats to permit subtraction of the “‘I-radioactivity organs. All data are means of five experiments & S.D. Tissue
% of the total uptake
of ‘Z51-labeled lipase per g tissue
per organ
A. Without Liver Heart
native
denatured
native
denatured
60.0 k2.7 1.45 + 0.53
85.0 14.3 0.13 + 0.02
4.03 1.0.16 1.41 kO.51
6.34iO.53 0.14+0.03
12.8 +5.3 0.69 + 0.07
81.2 k3.8 0.23 rtO.03
5.28 + 0.39 0.69 to.09
6.32 + 2.3 0.23 kO.04
heparin
B. With heparin Liver Heart
519
I
1
I
I
I
I
5
10
/
10
5 Time
I
Time (min)
(mid
Fig. 5. Disappearance curves for ‘251-labeled bovine lipoprotein lipase (A), “Cr-labeled rat erythrocytes (0) and ‘251-labeled human serum albumin (0) after intravenous injection into supradiaphragmatic rats. The blood volume used for the calculations was that obtained from the dilution of the “Cr-radioactivity in a sample taken 30 s after the injection.
disappearance of lipoprotein lipase was corrected for the apparent disappearance of 5’Cr-radioactivity. Enzyme activity and enzyme radioactivity disappeared in parallel also in the supradiaphragmatic rat (Fig. 6). No endogenous lipoprotein lipase activity could be demonstrated in blood. Thus, the lipoprotein lipase activity measured in this experiment derived only from the injected lipase preparation. In previous experiments with intact rats we were not able to saturate the uptake mechanisms with the amounts of lipase which for practical reasons we could inject, 300 yg/rat (Ref. 7; see also Fig. 1). With the supradiaphragmatic rats, however, there was a definite dose-dependency for the rate of uptake (Fig. 7). With the lowest dose used, 12 peg lipase, more than 60% disappeared within 10 min, whereas when 200 pg of lipase was
Fig. 6. Disappearance of lipoprotein lipase enzyme activity (0) and radioactivity (0) after i.v. injection into supradiaphragmatic rats. In this experiment a relatively large dose of the ‘251-labeled lipase preparation (2.9.106 cpm), but no additional unlabeled lipase, was injected. The blood samples were treated and assayed as in Fig. 1 but without antiserum. No endogenous lipase activity was detected in blood samples from corresponding rats which had not been injected.
injected less than 30% left the blood within that time. Heparin injected before the enzyme in supradiaphragmatic rats slowed down the disappearance considerably (Fig. 8). Heparin injected at 2 and 20 min after the enzyme caused reappearance of about 50% of the radioactivity that had been taken up (Fig. 8). The amount that could be released by heparin was similar to the difference in the uptake with or without heparin injected before the enzyme. Denatured lipase disappeared from the circulation also in supradiaphragmatic rats (Fig. 9). After a short equilibration phase the disappearance was, however, quite slow. Injection of heparin after 5 min caused release of about l/3 of the radioactivity that had been taken up at this time (Fig. 9). In the supradiaphragmatic rats we did not find
520
Time
(min)
20
10 Time
30
(mini
Fig. 7. Saturation of the uptake mechanism(s) in supradiaphragmatic rats by injection of large amounts of lipoprotein hpase. 0, 12 gg ‘251-labeled lipoprotein lipase only. 0, 12 pg labeled lipase+ 200 pg unlabeled lipase. The total volume injected in both cases was 0.7 ml.
Fig. 8. Effects of heparin injected before and after ‘251-labeIed lipoprotein lipase in supradiaphragmatic rats. Heparin was injected 2 mm before the hpase (A). Heparin was injected 160 s (0) or 20 min (0) after the lipase.
any single organ which dominated the uptake; there was a low but significant radioactivity in all tissue samples. The contribution by heart and lungs to the total uptake of radioactivity was 3-4fold higher in the supradiaphragmatic rats (Table II) than in intact rats (Table I). Heparin injected before the enzyme decreased this proportion in both organs by 70-80%. The uptake of denatured lipase in the heart was lower than the uptake of native lipase (Table II). The uptake in lungs was not affected by denaturation of the lipase. Heparin decreased the uptake of denatured lipase in both organs. From the results described above it was evident that some of the injected lipoprotein lipase bound to sites from which it could be released in active form by heparin even after hours, but some of the enzyme was taken up irreversibly. Perfusion of rat livers with heparin-containing medium 15 min after injection of native lipase caused release of less than 10% of the radioactivity which had been
taken up in the liver at this time (not shown). Thus, a main part of the heparin-irreversible uptake occurred in the liver. The radioactivity in the livers decreased with time (Fig. 10) indicating degradation of the lipase. In support of this, acidsoluble radioactivity was found in the blood 20 min after the injection (not shown). For comparison similar experiments were carried out with ‘251-asialofetuin (Fig. 10). This protein is known to be taken up by the galactose-receptor in the liver and to be rapidly degraded by lysosomal proteinases [17]. The radioactivity in the liver as a function of time after injection is plotted in Fig. 10. Two features are noteworthy. First, the curves for asialofetuin and for denatured lipoprotein lipase extrapolate to close to lOO%, indicating that these proteins were taken up almost quantitatively into the liver. In contrast, the curve for native lipoprotein lipase extrapolates to about 60%, in agreement with the other results which demonstrate that substantial uptake occurs in extrahe-
521
I
I
60 I
I
I
5
10
15
Time (min)
Fig. 9. Effect of denaturation on the disappearance of lz51labeled lipoprotein lipase in supradiaphragmatic rats. The lipase was denatured with GdmCl as described in Methods and then injected (0). In another supradiaphragmatic rat heparin was injected 6.5 min after the denatured lipase (0).
patic tissues. Secondly, the rate of degradation was quite different for the two proteins. The half-life for asialofetuin in the liver was only about 15 min, whereas that for native lipoprotein lipase was about 1 h. The degradation rate for denatured lipase was intermediate. Measurements of lipoprotein lipase
TABLE
120
160
Time (min)
Fig. 10. Disappearance of ‘251-radioactivity from the liver following uptake of intravenously injected labeled asialofetuin (m), native lipoprotein hpase (0) or lipase denatured with GdmCl (0). Each data point represents the mean from two or three rats. Inset: the same data plotted in a line/log graph.
activity in homogenates of livers taken out 15 min after injection of 300 pg unlabeled lipase together with 1251-labeled lipase showed that the lipase which had been taken up in the liver was still active. While 64% of the injected radioactivity was in the liver, 52% of the injected lipoprotein lipase activity was found there. When the livers were not taken out until 60 min after the injection, 30% of
II
CONTENT OF RADIOACTIVITY IN HEARTS SUPRADIAPHRAGMATIC RATS
AND LUNGS
15 MIN AFTER
INJECTION
OF “‘I-LABELED
LIPASE INTO
Procedure as in Table I. Data from two rats are shown. A. Percent of the total amount of ‘s51-labeled lipoprotein lipase which had been cleared from the blood. B. As in A, but expressed per g tissue instead of per total organ. Tissue
Without heparin Heart Lungs
Native lipase
Denatured lipase
A
B
A
B
4.2,4.1 13.6, 14.7
4.9, 4.9 8.8, 9.1
1.4,1.7 11.7, 14.7
1.2, 1.5 7.0, 8.9
0.8, 1.2 6.1, 3.6
1.0, 1.0 3.9,1.7
0.5, 0.5 6.6, 10.3
0.6, 0.4 4.5, 7.3
With heparin Heart Lungs
522
the radioactivity remained but only 13% of the lipase activity. This indicated that the inactivation of the lipase preceded its total proteolytic degradation. In these experiments the endogenous hepatic lipase activity was inhibited by an antiserum raised against hepatic lipase purified from heparin perfusates of rat livers. All data given are means of data from two rats. Discussion This study demonstrates a rapid uptake of *251labeled bovine lipoprotein lipase from the circulating blood not only by the liver but also by extrahepatic tissues. This was already evident in a previous study [7], but we have now studied the uptake in more detail. It was found that the distribution of the lipase between hepatic and extrahepatic tissues was strongly dependent on whether the enzyme was in its native (catalytically active) conformation or not. In this study we have recromatographed the labeled preparations of lipase on heparin-Sepharose and used only the fraction which bound with high affinity and which co-eluted with enzyme activity during gradient elution of the column. Disappearance curves for enzyme activity and for enzyme radioactivity followed each other closely, demonstrating that the purified labeled lipase behaved in the same way as native lipase. In a homologous system with guinea pig lipoprotein lipase injected into guinea pigs, similar disappearance curves were obtained. Thus, the rapid removal from blood of bovine lipoprotein lipase injected to rats was not due to the fact that a Iipase from a different species was injected and we conclude that this system can be used as a convenient model for studies on the rate of removal of circulating lipoprotein lipase. 40-50% of the uptake of native lipoprotein lipase occurred in extrahepatic tissues. A major part of this enzyme could be released back to the circulation in its active form by heparin. Even 1 h after injection of the lipase about one-third of the lipase which had been taken up reappeared in active form. This suggests that this pool of lipase was not rapidly degraded and that little redistribution from this pool occurred, since the amount of heparin-releasable lipase decreased only slowly. It seems probable that heparin-releasable binding to
the enzyme’s physiological receptors occurred. Binding of lipoprotein lipase to cultures of endothelial cells has been demonstrated by several groups [4-6,18,19]. When bound to the cells the lipase is relatively stable [4-41. 50% or more of the bound activity can rapidly be released by Aepari-F, [4,5,19]. Apparently little internalization and degradation of the enzyme occurs. The receptor at the cell surface for lipoprotein lipase is thought to be heparan sulfate [a]. Specific degradation of this glycosaminoglycan impedes binding of lipase to the cells 1431 and releases lipase that has already bound to the cells [4]. Heparan sulfate proteoglycan is probably present on most cells but its detailed structure is different in different tissues [20-23]. Endothelial cell heparan sulfate has a more heparin-like structure than heparan sulfates of most other cell types [24]. Lipoprotein lipase was shown to bind more avidly to heparin than to heparan sulfate from human aorta; the difference in affinity was estimated to be 40-fold [Z]. No direct studies on lipase binding to heparan sulfate isolated from endothelial cells have been reported. The general abundance of heparin-like glycosaminoglycans on cells may explain why circulating lipoprotein lipase was bound in most tissues and why the binding was not directly related to the physiological presence of lipoprotein lipase activity in the tissues. The Steins’ and their collaborators [I&19! have demonstrated binding also to cells other ehan endothelial cells, e.g., skin fibroblasts, heart muscle cells and preadipocytes. We do not know from our in vivo studies to what extent cells other than the endothelial cells participate in the extrahepatic binding of lipoprotein lipase. Since the binding occurs within minutes and is rapidly reversed by heparin the cells probably must be in direct contact with the blood. The extrahepatic uptake was not completely blocked, but only delayed, by injection of heparin before the enzyme. Furthermore, not all of the bound lipase could be released by heparin. This indicated binding also to sites other than the heparin-like ones. Similar findings have been reported in studies with ceils in tissue culture [18]. Thus, there may be at least two mecbanisms for binding of lipoprotein lipase to extrahepatic cells. The extrahepatic binding of lipoprotein lipase
523
was dose-dependent but the amounts needed to saturate the uptake far exceeded the amounts of lipase expected to be normally present on the endothelium in the rat. Not even 200 pg injected lipoprotein lipase blocked the uptake in supradiaphragmatic rats, though the amount of endothelial lipase in a whole rat should be about 20 pg, estimated from plasma post-heparin lipoprotein lipase activity, assuming the same specific activity as that of bovine lipoprotein lipase. A high binding capacity has also been observed with cultured endothelial cells; several pg of lipoprotein lipase could bind to a single culture plate [4,5]. The hepatic uptake of native lipase amounted to 50-60%. This uptake is not saturable [7]. Since the extrahepatic uptake is saturable this means that when high doses of lipase are injected a larger proportion is taken up in the liver. Enzymatically active lipase could be measured in liver homogenates even 30-60 min after injection. Thus, there was no immediate inactivation of the lipase after uptake. Radioactivity in the liver decreased with a half-time of about 1 h. The loss of lipase activity was more rapid, indicating that inactivation precedes total degradation of the lipase. Both inactivation and degradation of native lipase were significantly slower then the degradation of asialofetuin, which is taken up by the well-known galactose receptor [17]. Lipoprotein lipase is not taken up by any of the known glycoprotein receptors in the liver [7]. At this stage we do not know the nature of the uptake mechanism for lipoprotein lipase and we do not know which route of transport the lipase follows in the liver cells. On perfusion of the livers with or without heparin only small amounts of lipase radioactivity appeared in the perfusates, indicating that the uptake was virtually irreversible. Whether this is due to rapid internalization of most of the lipase into the cells or to a high-affinity binding to a receptor on the outside of the cells which is not reversed by heparin cannot be deduced from our data. If the enzyme is internalized it is interesting that it stays active for some time in the cells though it is a relatively labile enzyme. Treatment of lipoprotein lipase with GdmCl leads to monomerization and irreversible loss of enzyme activity. When the GdmCl concentration is reduced by dilution or dialysis the lipase regains
some structure but does not reattain its native catalytically active conformation (Osborne et al., unpublished data). Denatured lipase binds to heparin but with lower affinity than the native lipase (Bengtsson-Olivecrona, unpublished data). Denatured lipase was rapidly cleared from the blood, but 85% of the uptake occurred in the liver, i.e., little of the denatured lipase was taken up in extrahepatic tissues. The clearance of denatured lipase in supradiaphragmatic rats was slow. Apparently the extrahepatic uptake of circulating lipoprotein lipase was largely dependent on the native conformation of the lipase. Only small amounts of denatured lipase reappeared in the circulation after injection of heparin. Thus, most of the denatured lipase had been taken up by the liver in a heparin-irreversible manner. In this respect the hepatic uptake of denatured lipase resembles that for native lipase, which is also not readily reversible by heparin. We do not know whether the hepatic uptake of native and that of denatured lipoprotein lipase follow the same route. The degradation of denatured lipase in the liver was more rapid than that of native lipase; the estimated half-times were 30 min for the denatured lipase versus 60 min for native lipase. This may indicate a difference in the cellular fate of the denatured form but could also be due to differences in the rate of intracellular handling along the same route. Acknowledgements We thank our secretary, Ms. Marianne Lundberg, for help during preparation of the manuscript. This work was supported by the Swedish Medical Research Council 13X-00727 and the Medical Faculty, Umel University. References Quinn, E., Shirai, K. and Jackson, R.L. (1983) Prog. Lipid Res. 22, 35-78 Hamosh, M. and Hamosh, P. (1983) Mol. Aspects Med. 6, 199-289 Olivecrona, T., Bengtsson, G., Marklund, S.-E., Lindahl, U. and Htik, H. (1977) Fed. Proc. 36, 60-65 Shimada, K., Gill, P.-J., Silbert, J.E., Douglas, W.H.J. and Fanburg, B.L. (1981) J. Clin. Invest. 68, 995-1002. Cheng, C.-F., Oosta, G.M., Bensadoun, A. and Rosenberg, R.D. (1981) J. Biol. Chem. 256, 12893-12898
524
6 Brown, W.V., Wang-Iverson, P. and Paterniti, J.R. (1981) in Chemistry and Biology of Heparin (Lundblad, R.L., Brown, W.V., Mann, K.G. and Roberts, H.R., eds.), pp, 175-185, Elsevier/North Holland 7 Wallinder, L., Bengtsson, G. and Olivecrona, T. (1979) Biochim. Biophys. Acta 575, 166-173 8 Ehnholm, C., Schriider, T., Kuusi, T., Bang, B., Kinnunen, P., Kahma, K. and Lempinen, M. (1980) Biophys. Acta 617, 141-149. 9 Bezman-Tarcher, A. and Robinson, D.S. (1965) Proc. R. Sot. (London) B, 162,406-410 10 Bengtsson, G. and Olivecrona, T. (1977) Biochem. J. 167, 109-119 11 Wallinder, L., Bengtsson, G. and Olivecrona, T. (1982) Biochim. Biophys. Acta 711, 107-113 12 Olivecrona, T. and Bengtsson, G. (1983) Biochim. Biophys. Acta 152, 38-45 13 Graham, E.R.B. (1972) in Glycoproteins (Gottschalk, A., ed.), 2nd Edn., pp. 717-731, Elsevier, Amsterdam 14 Kuusi, T., Kinnunen, P.K.J., Ehnholm, C. and Nikkil’a, E.A. (1979) FEBS Lett. 98, 314-318 15 Belfrage, P. and Vaughan, M. (1969) J. Lipid Res. 10, 341-344
16 Hernell, O., Egehud, T. and Blivecrona, T. (1975) Biochim. Biophys. Acta 381, 233-241 17 Ashwell, G. and Harford, J. (1982) Annu. Rev. Biochem 51,53?-534 18 Friedman, G., Chajek-Shaul, T., Olivecrona, T., Stein, 0. and Stein, Y. (1982) Biochim. Biophys. Acta 7il, 114-122 19 Chajek-Shaul, T., Friedman, G., Stein, O., Olivecrona, T. and Stein, Y. (1982) Biochim. Biophys. Acta 712, 200-210 20 Oldberg, A., Kjellen, L. and Hook, M. (1979) J. Biol. Chem. 254, 8505-8510 21 Oohira, A., Wight, T.N. and Bomstein, P. (1983) J. Bioi. Chem. 258, 2014-2021 22 Carlstedt, I., Ciister, L., MaImstrom, A. and Fransscn, L.-A. (1983) J. Biol. Chem. 258, 11629-11635 23 Rapraeger, A.C. and Bernfield, M. (1983) J. Bid Chem. 258, 3632-3636 24 Gamse, G., Fromme, H.G. and Kresse, H. (1978) Biochim. Biophys. Acta 535, 544-550 25 Bengtsson, G., Olivecrona, T., Hook, RI., Riesenfeld, J. and Lindahl, U. (1980) Biochem. J. 189, 625-633
et Biophysics
513
Acta, 795 (1984) 513-524
BBA 51751
HEPATIC AND EXTRAHEPATIC LIPOPROTEIN LIPASE
UPTAKE OF INTRAVENOUSLY
INJECTED
LARS WALLINDER, OLIVECRONA
THOMAS OLIVECRONA
BENGTSSON-
Department
JONAS PETERSON,
of Physiological
Chemistry,
and GUNILLA
University of Urn&, S-9OI 87 Urn& (Sweden)
(Received April 6th, 1984)
Key words: Lipoprotein
lipase; Enzyme
uptake
Rats were injected intravenously with 12’I-labeled bovine lipoprotein lipase. The lipase disappeared within minutes from the blood due to uptake both in the liver (about 50% of the injected dose) and in extrahepatic tissues. Lipase enzyme activity disappeared in parallel to the ‘=I radioactivity. Thus, there was no inactivation of lipase in the circulating blood. Similar results were obtained when lipoprotein lipase purified from guinea pigs was injected into guinea pigs. Using supradiphragmatic rats we could show that the extrahepatic uptake was saturable and that the amounts of lipase that could be bound far exceeded the amounts of endogenous lipase expected to be present on the endothelium. When the lipase was denatured before injection, its removal in supradiaphragmatlc rats became slower, and in intact rats the fraction, of the uptake that occurred in extrahepatic tissues was much decreased. It is concluded that recognition by the extrahepatic receptors depends on the native conformation of the lipase. The extrahepatic uptake was strongly impeded by injection of heparin prior to injection of the lipase, and the uptake could to a large extent be reversed by injection of heparin after the lipase. Even after 1 h lipase that had been taken up by extrahepatic tissues reappeared immediately in the blood on injection of heparin. This was true both for enzyme activity and for enzyme radioactivity. Thus, internalization-inactivation-degradation occur only slowly in extrahepatic tissues. It is possible that the extrahepatic binding occurs to the enzyme’s physiological receptors. The hepatic uptake was not dependent on the native conformation of the lipase, was less sensitive to heparin, could not be reversed by heparin and was not saturable. The enzyme was not rapidly inactivated after uptake; its activity could be detected in liver homogenates even after 1 h. Degradation to acid-soluble products in the liver was relatively slow; the t,,2 for native lipase was about 1 h. In comparison, in parallel experiments asialofetuin was degraded with a t,,, of about 15 min
Introduction Lipoprotein lipase hydrolyzes acylglycerols in plasma lipoproteins (for review see Refs. 1 and 2). This reaction takes place at the vascular endothelium in some extrahepatic tissue. The enzyme is synthesized in other cells in the tissue and is then transferred to the endothelium. It has been proposed that the lipase is held in place at the endothelial cell surface via interaction with membrane0005-2760/84/$03.00
0 1984 Elsevier Science Publishers B.V.
bound heparan sulfate chains [3]. This model has gained support from studies with cultured endothelial cells [4-61, which have shown that enzymatic degradation of cell surface heparan sulfate impedes binding of lipoprotein lipase to the cells. Little or no lipoprotein lipase activity is found in the circulating blood. One would expect that if lipoprotein lipase was injected into the circulation it would bind to its endothelial receptors. However, in a previous study we found that most of the
514
injected lipase is taken up by the liver and degraded [7]. A minor part was taken up in extrahepatic tissues. Thus, there are receptors in the liver which override the extrahepatic ones in clearing the enzyme from the blood. The hepatic receptors for lipoprotein lipase are not identical with any of the known glycoprotein receptors [7]. Ehnholm and his associates [8] have studied the metabolism of lipoprotein lipase in hepatectomized swine. In their study, endogenous lipoprotein lipase was released into the blood by heparin injection. As the heparin disappeared from the blood, so did the lipase activity, directly demonstrating the existence of extrahepatic binding sites. The aim of the present study was to explore, in more detail, the nature of the hepatic and the extrahepatic uptake of lipoprotein lipase, and the possible differences between the mechanisms involved. To this end we have followed the fate of catalytically active as well as inactive forms of the lipase in intact and in supradiaphragmatic rats. We have also studied the effect of heparin on the uptake mechanisms. Materials and Methods Animals Male Sprague-Dawley rats weighing 200-350 g (AB Anticimex, Stockholm, Sweden) were fed a standard pellet diet (AB Ewos, Siidertalje, Sweden). Male guinea pigs (400-500 g) were from a local laboratory farm. The rats were under light ether anaesthesia during the experiments, while the guinea pigs were given intramuscular injections of Hypnorm Vet.” (AB Leo, Sweden, 1 mg/kg body weight). Intravenous injections were made in the exposed right jugular vein. Blood samples (200 ~1 in 1 ml plastic syringes, one for each blood sample) were withdrawn from the left jugular vein and collected in weighed tubes. In the experiments with heparin a standard dose of 100 ~1 was injected (Lowen, Malmo, Sweden 5000 IE/ml). Immediately following the last blood sample the animals were exsanguinated through the abdominal aorta. The tissues were cut out and weighed and their total radioactivities were determined in an LKB Wallac 1275 Minigamma counter. The data were calculated on the basis of a blood volume of 55 g/kg body weight [7].
For measurement of lipoproteiu lipase activity in the livers, 10% (w/v) homogenates were prepared in 0.025 M ammonia buffer (pH 8.2) containing 5 IU heparin/ml, 1 pg of leupeptin and of pepstatin/ml, 25 IU aprotinine/ml and 5 mM EDTA. The proteinase inhibitors were from Boehringer-Mannheim, Mannheim, F.R.G. .A Polytron homogenizer was used. Supradiaphragmatic rats were prepared according to Bezam-Tarcher and Robinson [9& but without dividing the upper part from the lower part of the body. For these experiments the rats were anesthetized with Hypnorm Vet.a. Enzymes Lipoprotein lipase was purified from bovine milk and from guinea pig milk as previously described [lO,ll]. The enzymes were radiolabeled using the lactoperoxidase/glucose oxidase method. To 2 ml purified bovine lipase (0.3 mg/ml) in 50 mM Tris-HCl (pH 7.4)/1.5 M NaCI/20 mM glucose, was added 10 ~1 of lactoperoxidase (5 mg/n^nl Boehringer-Mannheim), 1 mCi of Na1251 (carrierfree, 0334 New England Nuclear, Dreieich, F.R.G.) and 10 ~1 glucose oxidase (1 mg/ml, Boehringer-Mannheim) in this order. The mixture was incubated for 25 min on ice. The reaction was stopped by adding 6 ml of 10 mM Tris-HCl (pH 7.4)/0.01 M Kl/O.O25 M NaN,. The mixture was applied to a 3 ml heparin-Sepharose column, which had been equilibrated with 10 mM Tris-HCl (pH 7.4)/2 mg/ml bovine serum albumin/O.5 M NaCl. Unreacted iodine was washed away with 30 ml of the equilibration buffer at a flow rate of 1 ml/mm. The enzyme was eluted with a 140 ml linear salt gradient from 0.5 to 2.0 M NaCl in 10 mM Tris-HCl (pH 7.4) with 2 mg bovine serum albumin/ml. 4-ml fractions were collected. Catalytically active lipoprotein lipase eluted at about 1 M NaCi; denatured, inactive lipase eluted at lower salt concentrations. More than 98% of the radioactivity in the pooled peak ‘fractions (eluting at about 1 M NaCl) could be precipitated by trichloroacetic acid. A typical lipase preparation had a specific radioactivity of 1000 cpm/ng as determined by an immunoassay [12]. Guinea pig lipoprotein, lipase was iodinated and purified using the same,procedure. The labeled preparations were divided in portions of 200 ~1 and stored at - 20 0 C. They
515
were used for experiments within 1 month. On longer storage the enzyme gradually changed behavior gaining properties resembling denatured lipase. For some experiments the labeled enzyme was denatured by dialysis against 6 M guanidinium chloride (GdmCl) in 10 mM Tris-Cl (pH 8.5) at 4 o C overnight. The GdmCl was then removed by a 4 h dialysis at 4’C against 0.5 M (NH,),S0,/0.2 M NaCl/lO mM Tris-HCl (pH 7.4). These preparations contained no measurable lipase activity. Prior to injection the labeled enzyme preparations were diluted in an equal volume of titrated rat plasma at 0 o C. 0.25 ml was injected per animal, corresponding to 1-3 pg lipase, (l-3). lo6 cpm. Samples of 0.2 ml from the same syringe were directly injected into weighed counting tubes and their radioactivity was determined. From these samples the injected dose of radioactivity was calculated. Other labeled materials Rat erythrocytes were labeled with Nay CrO, (New England Nuclear) during a 2h incubation at 37 CJC. They were washed three times in 0.15 M NaCl and then used the same day. 1 ml of sedimented erythrocytes contained about lo6 cpm. Erythrocytes corresponding to 150000 cpm were diluted in saline and then injected 2 min prior to injection of lipase. The 51Cr and the lz51 radioactivites in the blood samples were determined in the LKB Minigamma counter. 17% of the 51Cr counts were also detected in the lz51 channel but were subtracted to give the net I*‘1 counts. Fetuin (type IV, Sigma, St. Louis, MO, U.S.A.) was labeled with the lactoperoxidase,’ glucose oxidase method as described above and then separated from unreacted iodine by chromatography on a Sephadex G-25 column in 50 mM Tris-HCl/ 0.15 M NaCl (pH 7.4). To prepare asioalofetuin the labeled protein was subjected to mild acid hydrolysis in 0.05 M H, SO, at 80 o C for 60 min to remove terminal sialic acid residues [13]. After dialysis against 50 mM Tris-HCl/O.lS M NaCl (pH 7.4), 95% of the radioactivity could be precipitated with trichloroacetic acid. The asialofetuin thus obtained had a specific radioactivity of 900 cpm/ng. About 3 pg, diluted in rat citrate plasma, were injected into the rats.
Antiserum against hepatic hpase Hepatic lipase was purified from heparin perfusates of rat livers [14]. The lipase in perfusates from 10 rats was precipitated by an equal volume of saturated ammonium sulfate and then solubilized in 10 mM Tris-HCl (pH 7.4)/20% glycerol/ 0.2% Triton X-100. This solution was applied to a heparin-Sepharose column and eluted with a gradient of NaCl in the same buffer. The peak fractions of activity were combined, diluted and then applied to a column of immobilized heparin, which had been modified by N-desulfation followed by acetylation [ll]. The lipase was eluted with a gradient of NaCl and the material thus obtained was used for immunization of a rabbit. The antiserum obtained inhibited rat hepatic lipase activity both in liver homogenates and in postheparin plasma but did not inhibit rat or bovine lipoprotein lipase activity. On Western blotting of SDS-polyacrylamide gels the antiserum reacted with two components in the purified hepatic lipase preparation. Assay procedure Lipoprotein lipase activity was assayed, using 3H-labeled Intralipid@ (a kind gift from AB Vitrum, Stockholm, Sweden). To 10 pl Intralipid (100 mg triacylglycerol/ml) were added 40 ~1 heat-inactivated human serum and 100 ~1 of a mixture containing 12% bovine serum albumin/30 IU heparin/0.2 M NaC1/0.3 M Tris-HCl (pH 8.5). The total volume was adjusted to 200 ~1 with distilled water. The incubations were carried out for 30-120 min in a 25 “C water bath. The incubation was stopped and the fatty acids were extracted as described previously [15]. The assay was linear with time and with the amount of enzyme in the range used in the experiments. For measurements of lipoprotein lipase activity in liver homogenates and in post-heparin plasma the hepatic lipase activity was inhibited by a prior incubation of the samples for 2 h on ice with an equal volume of anti-hepatic lipase antiserum. No hepatic lipase activity could be detected in these samples even when using an assay optimized for hepatic lipase [16]. The data on lipoprotein lipase activity in liver homogenates are given as the difference between the activities in rats which had been injected with lipoprotein lipase and the activities in control rats.
516
latter were hardly detectable and never amounted to more than 5% of the activities in the injected rats.
The
Results
To determine whether the ‘251-labeled lipdprotein lipase behaved in the same way as catalytically active unlabeled lipase, we injected a trace amount of labeled and a relatively large amount of unlabeled lipase (0.2 mg) intravenously into rats (Fig. 1). Both radioactivity and lipase activity were rapidly cleared from the blood; less than 5% of the injected doses remained after 10 min. The disap-
20
40
60
Time (min) Fig. 1. Disappearance of lipoprotein lipase enzyme activity (O,O, A) and ‘251-radioactivity (0, 1, A) after intravenous injection to rats and reappearance of active enzyme in the blood after injection of heparin. A mixture of unlabeled lipoprotein lipase (200 sg) and 1251-labeled lipase was injected into three rats. Injections of heparin were made at the times indicated by the arrows. The blood samples were collected in heparinized tubes and their radioactivity was determined. They were then centrifuged and the plasma lipoprotein lipase activity was determined after inhibition of the hepatic lipase activity with antiserum (see Methods). The endogenous lipoprotein lipase activity released by heparin was determined in corresponding rats which had not been given bovine lipase. It amounted to 15% of the injected dose of bovine lipase activity and was subtracted from the values presented in the figure.
pearance curves were almost identical, demonstrating that the iodination had not altered the behavior of the enzyme. As shown previously 171, injection of heparin caused reappearance in the blood of radioactively labeled lipase. In the experiment in Fig. 1 about 35% of the injected dose reappeared on injection of heparin PO min after injection of the lipase. A corresponding amount of catalytic activity reappeared (Fig. 1). Thus, no inactivation of the enzyme had occurred on uptake into the heparin-releasable pool. This was true even at longer times after the injection; heparin at 60 min released 13% of the radioactivity and 15% of the catalytic activity. Before injection of heparin little or no endogenous lipoprotein lipase was present in the blood of the rats. The lipoprottin lipase activity released by heparin corresponded to about 15% of the injected dose of bovine lipase. This activity was subtracted from the values in Fig. 1. For practical reasons we used here a heterologous experimental model; bovine lipoprotein lipase was injected into rats. In a few experiments we have used a homologous model with ‘251-1abeled lipoprotein lipase purified from guinea pig milk injected into guinea pigs. Fig. 2 shows that the results were very similar to those obtained with bovine lipase in rats. 50% of the injected dose disappeared within I min. At PO min less than I5% of the radioactivity remained in the blood. 20% of the injected dose reappeared in the blood on heparin injection after 15 min. Similar results were obtained when guinea pig lipoprotein lipase was injected into rats (Fig. 2). Thus, we conclude that the rapid clearing of lipoprotein lipase from the blood is not because it is recognized as a foreign protein by the rats but for some other physiological reason. To determine whether the native lipase conformation was required for recognition by the uptake mechanism(s) i2’ I-labeled lipase was denatured with 6 M GdmCl. The GdmCl was removed by dialysis (see Methods) and the material was then injected into rats. In all experiments the denatured lipase (Fig. 3a) disappeared more rapidly from the blood than the native lipase did (Fig. 3b). Hegarin injected at different times after the lipase caused reappearance of less of the denatured lipase than of the native lipase. At 15 min 35% of the injected native lipase reappeared, while only 5% of the
517
5
10
15
Time (mid
Fig. 2. Disappearance curves for 1251-labeled guinea pig lipoprotein lipase after intravenous injection into guinea pigs (0, W) and into rats (0,O). Two animals of each species were injected. The arrows indicate injection of heparin into one animal of each species.
denatured lipase reappeared. At 30 min hardly measurable amounts of denatured lipase reappeared, while 30% of the native lipase was still available for release by heparin. Thus, in contrast to native lipase most of the .denatured lipase had been taken up by mechanisms that were not reversed by heparin. The effect of injection of heparin immediately before injection of 1251-labeled lipase is shown in Fig. 4. After an equilibration phase the disappearance of both native and denatured lipase was considerably slower than it was in the absence of heparin (compare with Fig. 3). The half-times calculated from the second slower parts of the curves were 20 min and 10 min, respectively, for native and denatured lipase in the presence of heparin. 60% of the uptake of native lipase and 85% of the uptake of denatured lipase occurred in the liver (Table I). When heparin was injected before the lipase a higher proportion of native lipase was taken up in the liver (72.8 f 5.6%) than in the absence of heparin. The proportion of the denatured lipase taken up by the liver was not significantly altered by heparin. Considerable amounts of native lipase were
b) a)
Time hid
I
I
I
10
20
30
Time (mid
Fig. 3. Disappearance in rats of i.v. injected ‘251-labeled bovine lipoprotein lipase. (a) Lipase denatured by GdmCl as described in Methods. (b) Native lipase. Heparin was injected 3, 5, 10, 1.5 and 30 min after the injection of the lipase.
L
relative contribution by the heart to the tota”l uptake was small (about 1.5%) but this uptake showed some conspicious features. It was reduced by more than 50% by heparin and the proportional uptake of denatured lipase was only about one-tenth that of native lipase. Similar effects were seen also in other extrahepatic tissues. Taken together they explain why more of the uptake occurred in the liver in the presence of heparin and why more of the denatured lipase was directed to the liver. The results described above suggested that there are at least two routes of uptake for lipoprotein lipase from the circulation; one into the liver and another one into extrahepatic tissues. To study the extrahepatic uptake directly we injected lipase into supradiaphragmatic rats. Fig. 5 shows that the disappearance of native lipase from the circulation of the upper part of the body was slower than the disappearance in the intact rat. In the experiments on supradiaphragmatic rats we noted that there was a decrease of radioactivity per volume of blood of injected 1251-labeled albumin and of 5’Cr-labeled erythrocytes during the course of the experiment (Fig. 5). The effect was most marked during the first S-10 min after the injection. The lipase radioactivity per volume of blood decreased more rapidly than the 51Cr-radioactivity, demonstrating a true disappearance of the lipase from the circulation into the tissues (Fig. 5). In all smbsequent experiments on supradiaphragmatic rats the
I
_I-
10
5
Time
(min)
Fig. 4. Effect of heparin injected prior to injection of native (0) or denatured (0) ‘251-Iabeled bovine lipoprotein lipase into rats. The heparin was injected 2 min before the enzyme.
taken up also by extrahepatic tissues; about 40%. This radioactivity was distributed all over the body. The uptake in the heart is shown in Table IA. The
TABLE
I
TISSUE DISTRIBUTION INTO RATS
OF RADIOACTIVITY
10 MIN
AFTER
INJECTION
OF ‘25i-LABELED
LIPOPROTEIN
LIPASE
In B, heparin was injected 2 min before injection of the lipase. The lipase was denatured with GdmCl as described in Methods. due to the blood conten! of the “Cr-labeled erythrocytes were injected into the rats to permit subtraction of the “‘I-radioactivity organs. All data are means of five experiments & S.D. Tissue
% of the total uptake
of ‘Z51-labeled lipase per g tissue
per organ
A. Without Liver Heart
native
denatured
native
denatured
60.0 k2.7 1.45 + 0.53
85.0 14.3 0.13 + 0.02
4.03 1.0.16 1.41 kO.51
6.34iO.53 0.14+0.03
12.8 +5.3 0.69 + 0.07
81.2 k3.8 0.23 rtO.03
5.28 + 0.39 0.69 to.09
6.32 + 2.3 0.23 kO.04
heparin
B. With heparin Liver Heart
519
I
1
I
I
I
I
5
10
/
10
5 Time
I
Time (min)
(mid
Fig. 5. Disappearance curves for ‘251-labeled bovine lipoprotein lipase (A), “Cr-labeled rat erythrocytes (0) and ‘251-labeled human serum albumin (0) after intravenous injection into supradiaphragmatic rats. The blood volume used for the calculations was that obtained from the dilution of the “Cr-radioactivity in a sample taken 30 s after the injection.
disappearance of lipoprotein lipase was corrected for the apparent disappearance of 5’Cr-radioactivity. Enzyme activity and enzyme radioactivity disappeared in parallel also in the supradiaphragmatic rat (Fig. 6). No endogenous lipoprotein lipase activity could be demonstrated in blood. Thus, the lipoprotein lipase activity measured in this experiment derived only from the injected lipase preparation. In previous experiments with intact rats we were not able to saturate the uptake mechanisms with the amounts of lipase which for practical reasons we could inject, 300 yg/rat (Ref. 7; see also Fig. 1). With the supradiaphragmatic rats, however, there was a definite dose-dependency for the rate of uptake (Fig. 7). With the lowest dose used, 12 peg lipase, more than 60% disappeared within 10 min, whereas when 200 pg of lipase was
Fig. 6. Disappearance of lipoprotein lipase enzyme activity (0) and radioactivity (0) after i.v. injection into supradiaphragmatic rats. In this experiment a relatively large dose of the ‘251-labeled lipase preparation (2.9.106 cpm), but no additional unlabeled lipase, was injected. The blood samples were treated and assayed as in Fig. 1 but without antiserum. No endogenous lipase activity was detected in blood samples from corresponding rats which had not been injected.
injected less than 30% left the blood within that time. Heparin injected before the enzyme in supradiaphragmatic rats slowed down the disappearance considerably (Fig. 8). Heparin injected at 2 and 20 min after the enzyme caused reappearance of about 50% of the radioactivity that had been taken up (Fig. 8). The amount that could be released by heparin was similar to the difference in the uptake with or without heparin injected before the enzyme. Denatured lipase disappeared from the circulation also in supradiaphragmatic rats (Fig. 9). After a short equilibration phase the disappearance was, however, quite slow. Injection of heparin after 5 min caused release of about l/3 of the radioactivity that had been taken up at this time (Fig. 9). In the supradiaphragmatic rats we did not find
520
Time
(min)
20
10 Time
30
(mini
Fig. 7. Saturation of the uptake mechanism(s) in supradiaphragmatic rats by injection of large amounts of lipoprotein hpase. 0, 12 gg ‘251-labeled lipoprotein lipase only. 0, 12 pg labeled lipase+ 200 pg unlabeled lipase. The total volume injected in both cases was 0.7 ml.
Fig. 8. Effects of heparin injected before and after ‘251-labeIed lipoprotein lipase in supradiaphragmatic rats. Heparin was injected 2 mm before the hpase (A). Heparin was injected 160 s (0) or 20 min (0) after the lipase.
any single organ which dominated the uptake; there was a low but significant radioactivity in all tissue samples. The contribution by heart and lungs to the total uptake of radioactivity was 3-4fold higher in the supradiaphragmatic rats (Table II) than in intact rats (Table I). Heparin injected before the enzyme decreased this proportion in both organs by 70-80%. The uptake of denatured lipase in the heart was lower than the uptake of native lipase (Table II). The uptake in lungs was not affected by denaturation of the lipase. Heparin decreased the uptake of denatured lipase in both organs. From the results described above it was evident that some of the injected lipoprotein lipase bound to sites from which it could be released in active form by heparin even after hours, but some of the enzyme was taken up irreversibly. Perfusion of rat livers with heparin-containing medium 15 min after injection of native lipase caused release of less than 10% of the radioactivity which had been
taken up in the liver at this time (not shown). Thus, a main part of the heparin-irreversible uptake occurred in the liver. The radioactivity in the livers decreased with time (Fig. 10) indicating degradation of the lipase. In support of this, acidsoluble radioactivity was found in the blood 20 min after the injection (not shown). For comparison similar experiments were carried out with ‘251-asialofetuin (Fig. 10). This protein is known to be taken up by the galactose-receptor in the liver and to be rapidly degraded by lysosomal proteinases [17]. The radioactivity in the liver as a function of time after injection is plotted in Fig. 10. Two features are noteworthy. First, the curves for asialofetuin and for denatured lipoprotein lipase extrapolate to close to lOO%, indicating that these proteins were taken up almost quantitatively into the liver. In contrast, the curve for native lipoprotein lipase extrapolates to about 60%, in agreement with the other results which demonstrate that substantial uptake occurs in extrahe-
521
I
I
60 I
I
I
5
10
15
Time (min)
Fig. 9. Effect of denaturation on the disappearance of lz51labeled lipoprotein lipase in supradiaphragmatic rats. The lipase was denatured with GdmCl as described in Methods and then injected (0). In another supradiaphragmatic rat heparin was injected 6.5 min after the denatured lipase (0).
patic tissues. Secondly, the rate of degradation was quite different for the two proteins. The half-life for asialofetuin in the liver was only about 15 min, whereas that for native lipoprotein lipase was about 1 h. The degradation rate for denatured lipase was intermediate. Measurements of lipoprotein lipase
TABLE
120
160
Time (min)
Fig. 10. Disappearance of ‘251-radioactivity from the liver following uptake of intravenously injected labeled asialofetuin (m), native lipoprotein hpase (0) or lipase denatured with GdmCl (0). Each data point represents the mean from two or three rats. Inset: the same data plotted in a line/log graph.
activity in homogenates of livers taken out 15 min after injection of 300 pg unlabeled lipase together with 1251-labeled lipase showed that the lipase which had been taken up in the liver was still active. While 64% of the injected radioactivity was in the liver, 52% of the injected lipoprotein lipase activity was found there. When the livers were not taken out until 60 min after the injection, 30% of
II
CONTENT OF RADIOACTIVITY IN HEARTS SUPRADIAPHRAGMATIC RATS
AND LUNGS
15 MIN AFTER
INJECTION
OF “‘I-LABELED
LIPASE INTO
Procedure as in Table I. Data from two rats are shown. A. Percent of the total amount of ‘s51-labeled lipoprotein lipase which had been cleared from the blood. B. As in A, but expressed per g tissue instead of per total organ. Tissue
Without heparin Heart Lungs
Native lipase
Denatured lipase
A
B
A
B
4.2,4.1 13.6, 14.7
4.9, 4.9 8.8, 9.1
1.4,1.7 11.7, 14.7
1.2, 1.5 7.0, 8.9
0.8, 1.2 6.1, 3.6
1.0, 1.0 3.9,1.7
0.5, 0.5 6.6, 10.3
0.6, 0.4 4.5, 7.3
With heparin Heart Lungs
522
the radioactivity remained but only 13% of the lipase activity. This indicated that the inactivation of the lipase preceded its total proteolytic degradation. In these experiments the endogenous hepatic lipase activity was inhibited by an antiserum raised against hepatic lipase purified from heparin perfusates of rat livers. All data given are means of data from two rats. Discussion This study demonstrates a rapid uptake of *251labeled bovine lipoprotein lipase from the circulating blood not only by the liver but also by extrahepatic tissues. This was already evident in a previous study [7], but we have now studied the uptake in more detail. It was found that the distribution of the lipase between hepatic and extrahepatic tissues was strongly dependent on whether the enzyme was in its native (catalytically active) conformation or not. In this study we have recromatographed the labeled preparations of lipase on heparin-Sepharose and used only the fraction which bound with high affinity and which co-eluted with enzyme activity during gradient elution of the column. Disappearance curves for enzyme activity and for enzyme radioactivity followed each other closely, demonstrating that the purified labeled lipase behaved in the same way as native lipase. In a homologous system with guinea pig lipoprotein lipase injected into guinea pigs, similar disappearance curves were obtained. Thus, the rapid removal from blood of bovine lipoprotein lipase injected to rats was not due to the fact that a Iipase from a different species was injected and we conclude that this system can be used as a convenient model for studies on the rate of removal of circulating lipoprotein lipase. 40-50% of the uptake of native lipoprotein lipase occurred in extrahepatic tissues. A major part of this enzyme could be released back to the circulation in its active form by heparin. Even 1 h after injection of the lipase about one-third of the lipase which had been taken up reappeared in active form. This suggests that this pool of lipase was not rapidly degraded and that little redistribution from this pool occurred, since the amount of heparin-releasable lipase decreased only slowly. It seems probable that heparin-releasable binding to
the enzyme’s physiological receptors occurred. Binding of lipoprotein lipase to cultures of endothelial cells has been demonstrated by several groups [4-6,18,19]. When bound to the cells the lipase is relatively stable [4-41. 50% or more of the bound activity can rapidly be released by Aepari-F, [4,5,19]. Apparently little internalization and degradation of the enzyme occurs. The receptor at the cell surface for lipoprotein lipase is thought to be heparan sulfate [a]. Specific degradation of this glycosaminoglycan impedes binding of lipase to the cells 1431 and releases lipase that has already bound to the cells [4]. Heparan sulfate proteoglycan is probably present on most cells but its detailed structure is different in different tissues [20-23]. Endothelial cell heparan sulfate has a more heparin-like structure than heparan sulfates of most other cell types [24]. Lipoprotein lipase was shown to bind more avidly to heparin than to heparan sulfate from human aorta; the difference in affinity was estimated to be 40-fold [Z]. No direct studies on lipase binding to heparan sulfate isolated from endothelial cells have been reported. The general abundance of heparin-like glycosaminoglycans on cells may explain why circulating lipoprotein lipase was bound in most tissues and why the binding was not directly related to the physiological presence of lipoprotein lipase activity in the tissues. The Steins’ and their collaborators [I&19! have demonstrated binding also to cells other ehan endothelial cells, e.g., skin fibroblasts, heart muscle cells and preadipocytes. We do not know from our in vivo studies to what extent cells other than the endothelial cells participate in the extrahepatic binding of lipoprotein lipase. Since the binding occurs within minutes and is rapidly reversed by heparin the cells probably must be in direct contact with the blood. The extrahepatic uptake was not completely blocked, but only delayed, by injection of heparin before the enzyme. Furthermore, not all of the bound lipase could be released by heparin. This indicated binding also to sites other than the heparin-like ones. Similar findings have been reported in studies with ceils in tissue culture [18]. Thus, there may be at least two mecbanisms for binding of lipoprotein lipase to extrahepatic cells. The extrahepatic binding of lipoprotein lipase
523
was dose-dependent but the amounts needed to saturate the uptake far exceeded the amounts of lipase expected to be normally present on the endothelium in the rat. Not even 200 pg injected lipoprotein lipase blocked the uptake in supradiaphragmatic rats, though the amount of endothelial lipase in a whole rat should be about 20 pg, estimated from plasma post-heparin lipoprotein lipase activity, assuming the same specific activity as that of bovine lipoprotein lipase. A high binding capacity has also been observed with cultured endothelial cells; several pg of lipoprotein lipase could bind to a single culture plate [4,5]. The hepatic uptake of native lipase amounted to 50-60%. This uptake is not saturable [7]. Since the extrahepatic uptake is saturable this means that when high doses of lipase are injected a larger proportion is taken up in the liver. Enzymatically active lipase could be measured in liver homogenates even 30-60 min after injection. Thus, there was no immediate inactivation of the lipase after uptake. Radioactivity in the liver decreased with a half-time of about 1 h. The loss of lipase activity was more rapid, indicating that inactivation precedes total degradation of the lipase. Both inactivation and degradation of native lipase were significantly slower then the degradation of asialofetuin, which is taken up by the well-known galactose receptor [17]. Lipoprotein lipase is not taken up by any of the known glycoprotein receptors in the liver [7]. At this stage we do not know the nature of the uptake mechanism for lipoprotein lipase and we do not know which route of transport the lipase follows in the liver cells. On perfusion of the livers with or without heparin only small amounts of lipase radioactivity appeared in the perfusates, indicating that the uptake was virtually irreversible. Whether this is due to rapid internalization of most of the lipase into the cells or to a high-affinity binding to a receptor on the outside of the cells which is not reversed by heparin cannot be deduced from our data. If the enzyme is internalized it is interesting that it stays active for some time in the cells though it is a relatively labile enzyme. Treatment of lipoprotein lipase with GdmCl leads to monomerization and irreversible loss of enzyme activity. When the GdmCl concentration is reduced by dilution or dialysis the lipase regains
some structure but does not reattain its native catalytically active conformation (Osborne et al., unpublished data). Denatured lipase binds to heparin but with lower affinity than the native lipase (Bengtsson-Olivecrona, unpublished data). Denatured lipase was rapidly cleared from the blood, but 85% of the uptake occurred in the liver, i.e., little of the denatured lipase was taken up in extrahepatic tissues. The clearance of denatured lipase in supradiaphragmatic rats was slow. Apparently the extrahepatic uptake of circulating lipoprotein lipase was largely dependent on the native conformation of the lipase. Only small amounts of denatured lipase reappeared in the circulation after injection of heparin. Thus, most of the denatured lipase had been taken up by the liver in a heparin-irreversible manner. In this respect the hepatic uptake of denatured lipase resembles that for native lipase, which is also not readily reversible by heparin. We do not know whether the hepatic uptake of native and that of denatured lipoprotein lipase follow the same route. The degradation of denatured lipase in the liver was more rapid than that of native lipase; the estimated half-times were 30 min for the denatured lipase versus 60 min for native lipase. This may indicate a difference in the cellular fate of the denatured form but could also be due to differences in the rate of intracellular handling along the same route. Acknowledgements We thank our secretary, Ms. Marianne Lundberg, for help during preparation of the manuscript. This work was supported by the Swedish Medical Research Council 13X-00727 and the Medical Faculty, Umel University. References Quinn, E., Shirai, K. and Jackson, R.L. (1983) Prog. Lipid Res. 22, 35-78 Hamosh, M. and Hamosh, P. (1983) Mol. Aspects Med. 6, 199-289 Olivecrona, T., Bengtsson, G., Marklund, S.-E., Lindahl, U. and Htik, H. (1977) Fed. Proc. 36, 60-65 Shimada, K., Gill, P.-J., Silbert, J.E., Douglas, W.H.J. and Fanburg, B.L. (1981) J. Clin. Invest. 68, 995-1002. Cheng, C.-F., Oosta, G.M., Bensadoun, A. and Rosenberg, R.D. (1981) J. Biol. Chem. 256, 12893-12898
524
6 Brown, W.V., Wang-Iverson, P. and Paterniti, J.R. (1981) in Chemistry and Biology of Heparin (Lundblad, R.L., Brown, W.V., Mann, K.G. and Roberts, H.R., eds.), pp, 175-185, Elsevier/North Holland 7 Wallinder, L., Bengtsson, G. and Olivecrona, T. (1979) Biochim. Biophys. Acta 575, 166-173 8 Ehnholm, C., Schriider, T., Kuusi, T., Bang, B., Kinnunen, P., Kahma, K. and Lempinen, M. (1980) Biophys. Acta 617, 141-149. 9 Bezman-Tarcher, A. and Robinson, D.S. (1965) Proc. R. Sot. (London) B, 162,406-410 10 Bengtsson, G. and Olivecrona, T. (1977) Biochem. J. 167, 109-119 11 Wallinder, L., Bengtsson, G. and Olivecrona, T. (1982) Biochim. Biophys. Acta 711, 107-113 12 Olivecrona, T. and Bengtsson, G. (1983) Biochim. Biophys. Acta 152, 38-45 13 Graham, E.R.B. (1972) in Glycoproteins (Gottschalk, A., ed.), 2nd Edn., pp. 717-731, Elsevier, Amsterdam 14 Kuusi, T., Kinnunen, P.K.J., Ehnholm, C. and Nikkil’a, E.A. (1979) FEBS Lett. 98, 314-318 15 Belfrage, P. and Vaughan, M. (1969) J. Lipid Res. 10, 341-344
16 Hernell, O., Egehud, T. and Blivecrona, T. (1975) Biochim. Biophys. Acta 381, 233-241 17 Ashwell, G. and Harford, J. (1982) Annu. Rev. Biochem 51,53?-534 18 Friedman, G., Chajek-Shaul, T., Olivecrona, T., Stein, 0. and Stein, Y. (1982) Biochim. Biophys. Acta 7il, 114-122 19 Chajek-Shaul, T., Friedman, G., Stein, O., Olivecrona, T. and Stein, Y. (1982) Biochim. Biophys. Acta 712, 200-210 20 Oldberg, A., Kjellen, L. and Hook, M. (1979) J. Biol. Chem. 254, 8505-8510 21 Oohira, A., Wight, T.N. and Bomstein, P. (1983) J. Bioi. Chem. 258, 2014-2021 22 Carlstedt, I., Ciister, L., MaImstrom, A. and Fransscn, L.-A. (1983) J. Biol. Chem. 258, 11629-11635 23 Rapraeger, A.C. and Bernfield, M. (1983) J. Bid Chem. 258, 3632-3636 24 Gamse, G., Fromme, H.G. and Kresse, H. (1978) Biochim. Biophys. Acta 535, 544-550 25 Bengtsson, G., Olivecrona, T., Hook, RI., Riesenfeld, J. and Lindahl, U. (1980) Biochem. J. 189, 625-633