82
Biochimica et Biophysics Acta, 1125 (1992) 82-89 0 1992 Elsevier Science Publishers B.V. All rights reserved 0005-2760/92/$05.00
BBALIP 53877
Lipoprotein lipase and hepatic lipase activities are differentially regulated in isolated hepatocytes from neonatal rats Julia Peinado-Onsurbe,
Concepci6 Soler, Maria Soley, Miquel Llobera and Ignasi Ramirez
Departament de Bioquimicu i Fisiologia, Facultut de Biologia, Unicersitat de Barcelona, Barcelona (Spain)
(Received 17 September 1991) (Revised manuscript received 6 December 1991)
Key words: Lipoprotein lipase; Hepatic lipase; Hepatocyte; (Rat) Lipoprotein lipase and hepatic lipase are members of the lipase gene family sharing a high degree of homology in their amino acid sequences and genomic organization. We have recently shown that isolated hepatocytes from neonatal rats express both enzyme activities. We show here that both enzymes are, however, differentially regulated. Our main findings are: (8 fasting induced an increase of the lipoprotein lipase activity but a decrease of the hepatic lipase activity in whole liver, being in both cases the vascular (heparin-releasable) compartment responsible for these variations. (ii) In isolated hepatocytes, secretion of lipoprotein lipase activity was increased by adrenaline, dexamethasone and glucagon but was not affected by epidermal growth factor, insulin or triiodothyronine. On the contrary, secretion of hepatic lipase activity was decreased by adrenaline but was not affected by other hormones. (iii) The effect of adrenaline on lipoprotein lipase activity appeared to involve P-adrenergic receptors, but stimulation of both p- and cu,-receptors seemed to be required for the effect of this hormone on hepatic lipase activity. And (iv), increased secretion of lipoprotein lipase activity was only observed after 3 h of incubation with adrenaline and was blocked by cycloheximide. On the contrary, decreased secretion of hepatic lipase activity was already significant after 90 min of incubation and was not blocked by cycloheximide. We suggest that not only synthesis of both enzymes, but also the posttranslational processing, are under separate control in the neonatal rat liver.
Introduction
Lipoprotein lipase (EC 3.1.1.34) and hepatic lipase are lipolytic enzymes involved in the metabolism of circulating lipoproteins. Lipoprotein lipase hydrolyses triacylglycerols of chylomicra and very low density lipoproteins [l]. The enzyme is widely distributed among many tissues, but expresses high activity in adipose tissue, heart and skeletal muscle and lactating mammary gland [2]. In these tissues the functional fraction of the enzyme is located at the luminal side of endothelial cells but synthesized inside parenchymal cells 131. Hepatic lipase participates in the clearance of chylomicron remnants by the liver [4] and in the conversion of high density lipoprotein subfractions (HDL, into HDL,) contributing to the reverse cholesterol transport to the liver (see Ref. 5 for review). An
Correspondence: I. Ramiez, Unitat de Bioquimica i Biologia Molecular B, Departament de Bioquimica i Fisiologia, Facultat de Biologia, Universitat de Barcelona, Diagonal 645, 08071-Barcelona, Spain.
important fraction of the enzyme protein is also located at the vasculature [6], but the enzyme is synthesized inside hepatocytes [7]. Both enzymes are members of the lipase gene family and share many structural features in common [S]. Mature enzymes are glycoproteins with at least two complex N-linked oligosaccharide chains [9-l 11. In many cell systems it was shown that heparin stimulates the release of either enzyme to the incubation medium [12-141. The levels of tissue lipoprotein lipase activity change in response to a wide number of stimuli. In fasted animals, lipoprotein lipase activity is decreased in adipose tissue but is increased in heart (see Refs. 15, 16 for review). This is the consequence of the difference among tissues in the response to hormones. Thus, in cultured adipocytes insulin increases the secretion of lipoprotein lipase activity by both short-term protein synthesis-independent mechanisms (17) and long-term transcriptional- [lS] and posttrancriptional-dependent mechanisms [19,20]. On the contrary, neither in perfused hearts [21] nor in neonatal heart cells in culture [22] does insulin affect lipoprotein lipase activity. In-
83 sulin, however, restores normal lipase activity in perfused hearts from diabetic rats [23]. Inversely, catecholamines decrease synthesis and increase intracellular degradation of the enzyme in adipose tissue [241, while in neonatal heart cells [25], these hormones increase both synthesis and processing of the enzyme. Regulation of hepatic lipase activity is less understood. It is known that it is decreased in fasted rats [261 but the hormones and the mechanisms involved are still unclear. Activity is decreased in the liver of diabetic rats [27] but in vivo administration of either insulin or glucagon did not regulate the enzyme [281. In cultured hepatocytes, it was described that alteration of the intracelluIar cyclic-AMP levels affected hepatic lipase secretion [29]. However, we have suggested that, in adult rats, catecholamines are involved in the effect of fasting on hepatic lipase activity 1301. In recent years we have shown that neonatal rat hepatocytes express hepatic lipase [3I] and lipoprotein lipase activities [32]. Lipoprotein lipase activity in neonatal liver is under nutritional control [33]. The occurrence of both enzymes in neonatal liver allows the study of the similarities in the processing and regulation of both enzymes which are otherwise structurally related and similarly processed in different cells types. Here we show that both enzymes are independently regulated and evidence is obtained to indicate that their posttranslational processing is also under separate control. Material and Methods Animals Rats of the Wistar strain were used from our own colony, where the animals were fed standard chow diet (65% carbohydrate, 18% protein, 3% fat, 5% fibre, 5% minerals and vitamins, by weight). Newborn rats were either allowed to suckle with their mothers (fed neonates) or 8 h after birth were transferred to an humidified chamber at 28-30°C for 16 h (fasted neonates). Perji.ision ofisolated livers To study the effect of starvation on the vascular (heparin-releasable) lipolytic activities, isolated livers were perfused with heparin as previously described 1341. In summa~, isolated livers were perfused in a retrograde fashion at a flow rate of 0.8 ml/min with a Hepes (20 mM, pH 7.4)containing buffer (buffer A in Ref. 34) at 37°C. After 10 min, the perfusion medium was changed to a heparin (5 U/ml)-supplemented buffer A. The first 1.2 ml were discarded (void volume) and a single 10 ml fraction was collected on ice. To stabilize lipolytic activities, glycerol was immediately added to give a final concentration of 20%. The collected fractions and the perfused livers (homogenized
in 3 ml of buffer B: 10 mM Hepes, I mM dithiothreitoi, 1 mM EDTA, 0.25 M sucrose pH 7.5) were kept, until used to assay lipolytic activities, at -40°C. Hepatocyte isolation and incubation Hepatocytes from fed or fasted neonatal rats were isolated as previously described [32]. These preparations contained about 15% hemopoietic cells but these cells express neither Iipoprotein lipase [32] nor hepatic lipase [31] activities. Isolated hepatocytes (2.5 . lo6 hepatocytes/ml) were incubated in an amino acidsand vitamins-supplemented medium (buffer D in Ref. 32, but without insulin) for up to 3 h at 37°C under O,/CO, (19 : 1) atmosphere. At selected times, 0.5 mI, in duplicate, were taken and the cells were separated of the incubation medium by centrifugation. Both incubation medium and cell homogenates (obtained by sonication in buffer B) 1321were kept at -40°C until used to assay Iipofytic activities. Lipolytic activity assays Lipoprotein lipase activity was determined by the method of Ramirez et al. [35] with minor modifications as described in [32]. The assay mixture contained 0.6 mM glycerol tri[9,1~n)-3HJoleate (12 Ci/mol), 50 mM MgCl,, 0.05% albumin (fatty acid free), 3% serum (preheated 60 min at SOOC),25 mM Pipes (pH 7.5) and 0.02 ml of sample in a final volume of 0.2 ml. The incubation was carried out for 30 min at 25°C. The reaction was te~inated, and the E3H]oleate released was quantified, as previously described [32]. This method is not completely specific for lipoprotein lipase, but we have shown that chicken antiserum to bovine milk lipoprotein lipase (which does not cross-react with hepatic lipase [36]) inhibited 92% of the activity detected in neonatal liver [34] and 98% of the activity in isolated neonatal hepatocytes [32]. Hepatic lipase activity was determined by the method of Ehnholm et al. [37] as previously described [38], but with minor modifications. The assay mixture contained 2.5 mM glycerol tri[9,1~~)-3Hloleate (0.3 Ci/mol), 0.75 M NaCI, 3% albumin (fatty acid free), 50 mM Tris (pH 8.5) and 0.05 ml of sample in a final volume of 0.2 ml. Incubation conditions, reaction termination and quantification of [3H]oleate released were as above for lipoprotein lipase assay. This assay system was found to be specific for the enzyme [61. For both enzymes, one unit of enzyme activity was defined as the amount of enzyme that catalyses the reIease of 1 pmol of oleate per min. Thermal inactivation experiments Lipoprotein lipase and hepatic lipase activities from heparin-perfused neonatal livers were added (l/100 final dilution) to buffer D (see hepatocyte incubation conditions), supplemented or not with heparin (5
84 units/ml), and incubated at 37°C under constant shaking. At selected times a sample was taken and kept in ice-cold water until the incubation was terminated when lipoprotein lipase and hepatic lipase activities were determined in each sample. We previously found that both enzyme activities were stable at 4°C for several hours. From these inactivation experiments, inactivation constants were estimated and used to correct for inactivation both lipoprotein lipase and hepatic lipase activities secreted to the incubation medium. Correction of secreted activities for inactiuation
In agreement with previous reports [13,39,40], in our experimental conditions soluble lipoprotein lipase activity was unstable at 37°C (Fig. 1). The presence of heparin stabilized in part the enzyme activity. Hepatic lipase activity was less unstable at this temperature and also heparin had a protective effect. Data in Fig. 1 were fitted to: (I,
= u,,e
-kl
(1)
where a, is the remaining activity at time t, a,, is the activity at time t = 0 and k is the inactivation constant. Inactivation constant (k) values for lipoprotein lipase and hepatic lipase activities in the presence and in the absence of heparin are shown in Table I. These values may be used to correct for inactivation the amount of secreted activities as follows: In a system where there is a constant secretion of enzyme activity, the amount of apparently secreted
TABLE
Inactivation at 37°C
LIPASE
HEPATIC
constants
of lipoprotein lipase and hepatic lipase actil,itie.r
Data in Fig. 1 follow an exponential the activity at time t, a,, is the inactivation constant. k values were sis after linearization (Lnta, /a,)) = independent experiments.
curve: u, = a,, em”, where a, is activity at t = 0 and k is the determined by regression analy- kt ). Results are mean of two
k (min-‘) incubation condition:
- heparin
+ heparin
Lipoprotein lipase Hepatic lipase
1.73~100.29. 10 -’
0.79~10~* 0.12~1o~z
activity (AA,,) is:
in a given period of time (T = t, - t ,>
AA,,
= a2 - a, emh7
(2)
where a2 and a, are the measured activities at times t, and t,, respectively. This is still an underestimation of the actual amount of secreted activity (AA) because it only takes consideration the inactivation of the activity already present at t, but not the inactivation of the activity secreted throughout T. Assuming that the secretion rate is constant throughout T (as it was shown in other cell systems for relatively large periods of time [12,13]), then: AA=rT
(3)
where r is the secretion rate constant. In such system: AA,,
LIPOPROTEIN
I
= /‘r 0
LIPASE
emkr dt
(4)
thus: 1)
AAaI, = G(e-“‘-
(5)
and: r= 0
120180
0
incubation
time
60
60
120180
(emkT-
(6)
I)
combining Eqns. 2, 3 and 6
(mid
Fig. 1. Inactivation of lipoprotein lipase and hepatic lipase activities at 37°C. Neonatal rat liver was perfused with heparin as indicated in the text to obtain lipoprotein lipase and hepatic lipase. The perfusate was immediately diluted in buffer D (see Materials and Methods) supplemented (0) or not (0) with heparin (5 units/ml) and incubated at 37°C. At selected times a sample was taken and kept on ice until the end of the experiment when lipolytic activities were assayed. Results shown are from a representative of two independent experiment.
- (AA,,P
AA=
kT(a,
-a,
(e-k7-
e-l”) 1)
(7)
A detailed example of the use of this mathematical model to correct the amount of secreted lipoprotein lipase and hepatic lipase activities is shown in Table II. Note that the bigger differences between measured and corrected activities were obtained in lipoprotein
85
lipase activity because of the higher inactivation constant. Also, secretion of both enzyme activities (corrected) was almost linear during the 3 h incubation. Eqn. 7 was used in this paper to correct for inactivation the activities measured in the incubation medium. Chemicals
Insulin and glucagon were from Novo Industri, Copenhagen, Denmark. Triiodothyronine, dexamethasone, adrenaline, (+ / - Fisoprenaline, phenylephrine, prazosin and propranolol were purchased from Sigma Chemicals St. Louis, MO, USA. Epidermal growth factor purified from mouse submaxillary glands was a generous gift of Dr. Morley D. Hollenberg, University of Calgary, Calgary, Canada. All other chemicals were obtained as described elsewhere [34,36,38,41]. Results Effect of starvation on lipolytic activities in perfused lillers and isolated hepatocytes
We have previously found that lipoprotein lipase activity increases with fasting in the liver of neonatal rats [33]. We also found that this is the consequence of an increased amount of heparin-releasable (vascular) activity [34]. Here we also show that 93% of the increase in the total lipoprotein lipase activity detected in the whole liver was accounted for the increase in the heparin-releasable pool (Table III>. Neither the residual (heparin non-releasable) activity in the perfused liver, nor the activity in isolated (not incubated) hepatocytes was significantly higher in fasted than in fed neonates. Similar results were described in adult rat heart, where fasting produces an increase in the activity of the whole tissue but not in isolated heart cells [411.
TABLE
Contrary to what was observed with the lipoprotein lipase activity, hepatic lipase activity in the liver of fasted neonates was significantly decreased (Table III>. The heparin-released activity was also significantly decreased. The decrease in the vascular pool of the enzyme activity accounted for about 70% of the decrease in the total activity. The hepatic lipase activity in isolated hepatocytes from fasted neonates was not significantly different to that in fed neonates. These results are similar to what we have found in adult animals [30]. Effect of hormones on lipolytic acticities in isolated hepatocytes
To study the effect of hormones on lipolytic activities, isolated hepatocytes from fed neonates were incubated (3 h) in the presence of heparin (5 U/ml>. Lipoprotein lipase activity within the cells was not affected by the presence of either adrenaline (10 PM), dexamethasone (0.4 mM), epidermal growth factor (10 nM), glucagon (100 nM), insulin (100 nM), or triiodothyronine (10 nM) (Table IV). Adrenaline, dexamethasone and glucagon increased the amount of enzyme activity released to the incubation medium (corrected for inactivation as indicated in methods). At the concentrations used, epidermal growth factor, insulin or triiodothyronine did not affect lipoprotein lipase release. The hepatic lipase activity inside cells was also not affected by the presence of hormones but adrenaline decreased the amount of enzyme activity released to the medium (Table IV>. Any other hormone or factor did not affect the secretion of hepatic lipase activity. It is known that /?-receptors are involved in the regulation of lipoprotein lipase activity by catecholamines in heart mesenchymal cells [25]. On the con-
II
Correction
of lipoprotein lipase and hepatic lipase activities in the incubation
medium for inactivation
Isolated hepatocytes from fed neonates were incubated (2.5. 10h cells/ml) in the presence of heparin (5 units/ml) at 37°C at indicated times a sample was taken and the medium was used to determine lipolytic activities. The measured lipoprotein lipase and hepatic lipase activities were corrected for inactivation as indicated in the text. Results are from a representative experiment. Time tmin)
Measured (mU/lO’
Lipoprotein 0 45 90 180 Hepatic 0 45 90 180
activity cells)
A-‘& (a, - a2 emkT)
AA - kTAA,,/(e-kT-l))
Corrected activity (a, + ZA A) (mU/106 cells)
_ 0.244 0.235 0.339
_ 0.290 0.279 0.473
0.01 0.30 0.58 1.05
0.529 0.468 0.783
0.548 0.486 0.834
0.06 0.61 1.09 1.93
lipase activity (k = 0.79. 10W2) 0.01 0.25 0.41 0.54
lipase activity (k = 0.12. 1O-2) 0.06 0.58 1.02 1.68
86
trary, the inhibitory effect of these hormones on hepatic lipase activity is known to be mediated by Ly,-receptors in adult rat liver [30,42]. We found that the ~-agonist isoprenaline, but not the at-agonist phenylephrine, increased the amount of lipoprotein lipase
TABLE
activity released to the medium by isolated hepatocytes (Table V). Accordingly, the presence of the p-blocker propranolol, but not the cu,-blocker prazosin, prevented the stimulator effect of adrenaline. These resuits indicate that P-receptors are involved in the ac-
111
Effect uf fasting on lipoprotein /iparseand heputic lipase actit.ities in perfused ker
and isolated heputocytes
Livers from fed or fasted neonatal rats were perfused with heparin as indicated in methods. The perfusate was collected in a single 10 ml fraction. After the perfusion, the liver was homogenized. Isolated hepatocytes from either fed or fasted neonates were obtained from collagenase perfused livers as indicated in methods and immediately homogenized by sonication. Lipoprotein lipase and hepatic lipase activities were determined in heparin perfusates and in liver or cell homogenates. Results are means + S.E. of seven to ten animals. Statistical differences between fed and fasted animals were determined by Student’s t-test. a non-significant; * * P < 0.01; * * * P < 0.001. Lipoprotein
lipase
Hepatic
fed Heparin-perfused Eluted Residual Total Isolated
TABLE
bepatocytes
fasted
liver (mU/g) 36 I 43
rt:lO rt2 rt8
152 16 168
(mU/lO” cells) 0.07 rt 0.02
lipase
fed +27 ** * 4” +30 **
fasted
530 190 720
0.08 * 0.02 $’
a.135
+31 +3Y k42
327 102 429
0.01
f 20 * * +23 I’ +30 ***
O.l6*
0.02 ;’
1’~’
Effect of hormones OR lipa~tic acti~~itiesis isolated ~epatocytes from fed neonufe.~ Freshly isolated hepatocytes At the end of the incubation activities. Activity values in Statistical differences versus
were incubated (2.5~10s cells/ml) during 3 h at 37°C in the presence of indicated hormones and heparin (5 U/ml). cells were precipitated by centrifugation and both cells and medium kept at - 30°C until used to determine lipolytic medium were corrected for inactivation as indicated in the text. Results are mean & S.E. of seven experiments. basal activity were determined by the paired t-test. * P < 0.05; * * P < 0.01. Lipoprotein
Basal Adrenaline (10 FM) Dexamethasone (0.4 mM) Epidermal growth factor (10 nM) Glucagon (0.1 FM) insulin (0.1 FM) Triiodothyronine (10 nMf
TABLE
lipase fmU/106
cells)
Hepatic
cells
medium
cells
0.04 f 0.01 0.05 f 0.01 0.05 rt: 0.01 0.01 t 0.01 0.04io.01 0.02 rt 0.02 0.02 I: 0.02
1.01+0.01 1.85+0.38 1.93hO.45 l.OZrtO.31 1.32ri:O.24 1.01 *a,2 1.0710.43
0.07 0.06 0.04 0.08 0.04 0.07 0.08
* ** *
hpase (mU/106
cells) medium
f 0.02 f 0.03 & 0.01 i_ 0.03 i 0.03 i 0.04 + 0.04
1.92 + 0.2 1.29 kO.10 * 2.09 k 0.12 1.99*0.16 t .94 + 0.31 2.27 * 0.16 2.22 * 0.22
V
Effect of catecho~ami~te agonists and antagonists on lipolytic acriuities in isoiated ~epafocytes Cells were isolated and incubated as indicated lipoiytic activities. Activity values were corrected differences versus basal activity were determined
in Table IV. At the end of the incubation the medium was collected and used to determine for inactivation as indicated in the text. results are mean _+S.E. of five experiments. Statistical by the paired t-test. * P < 0.05; * * P i 0.01; * * * P < 0.001.
Lipoprotein (mu/lo6 Basal Isoprenaline (1 FM) Phenylephrine (1 PM) Adrenaline (10 PM) i propranolol f 1 p MI f prazosin t 1 ,u MI Isoprenaline c pheny~ephrine
1.14*0,13 1.78 f 0.22 1.37 f 0.27 1.80+0.28 1.24 * 0.26 1.98iO.05 1.74 * 0.55
Hepatic
lipase in medium cells)
(“rot 100 155+14 11X*20 158+ 14 108rt 5 176&20 150*20
lipase in medium
(mu/IO6
* * * *
1.78 + 0.30 1.49kO.20 1.44 + 0.25 1.07 * 0.21 1.2O~bO.17 1.51+0.16 1.09 f 0.09
cells)
(%) 100 86k 83+ 59rf68+ 85+ 61+
6 9 2”“” 10 * 2** 7**
87 tion of adrenaline. Isoprenaline or phenylephrine alone did not reproduce the effect of adrenaline on the secretion of hepatic lipase activity to the incubation medium (Table V). The presence of propranolol or prazosin prevented only in part the inhibitory effect of adrenaline. These results indicate that both types of receptors are partially involved in the effect of adrenaline on hepatic lipase activity. In keeping with this hypothesis, when both isoprenaline and phenylephrine were added together the amount of enzyme secreted was as decreased as when adrenaline was added alone (Table VI. The time-course of the effect of adrenaline, dexamethasone and glucagon on both enzyme activities is shown in Fig. 2. It was found that, while the amount of lipoprotein lipase activity secreted to the medium was only significantly increased after 3 h of incubation with either hormone, the effect of adrenaline on the release of hepatic lipase activity was clearly significant 90 min after the addition of the hormone. Neither dexamethasone nor glucagon affected the release of hepatic lipase activity to the medium during the whole incubation period studied. To determine if the effect of hormones was dependent on protein synthesis, isolated hepatocytes were incubated in the presence of 0.35 mM cycloheximide and either adrenaline (10 PM), dexamethasone (0.4 mMI, or glucagon (0.1 PM). It was found that cycloheximide decreased significantly the secretion to the medium of both lipoprotein lipase and hepatic lipase activities (Fig. 3). In the presence of cycloheximide neither hormone was able to increase the amount of lipoprotein lipase activity secreted to the medium (Fig.
LIPOPROTEIN
LIPASE
2.5 1
2.0 ** 1.5
'Z ,.i ,;. '1
1.0
2.0 1.5 1.0
t* 0.5
0.5
I il 0
60
0
120180
incubation
time
60
120180
(min)
Fig. 2. Effect of hormones on secretion of lipoprotein Iipase and hepatic Iipase activities. Isolated hepatocytes were incubated as indicated in Table III, in the absence (0) or in the presence of 10 PM adrenaline CO), 0.4 mM dexamethasone (0) or 0.1 FM glucagon (A ), at indicated times a sample was taken, centrifuged and the medium used to determine Iipolytic activities. Results (corrected for inactivation as indicated in the text) are mean+S.E. of four experiments. Statistical differences versus basal activity were determined by the paired t-test. * P < 0.05; * * P < 0.01.
LIPOPROTEIN
pxo.05
cycloheximide
-
;:.: :::: d LIPASE
I
HEPATIC
I
LIPASE
;$; :;;:: :.:.: .:.:.
I
I
p-co.01
+
+
Fig. 3. Effect of cycloheximide on the secretion of lipoprotein lipase and hepatic lipase activities. Isolated hepatocytes were incubated (3 h) as indicated in Table III in the absence or in the presence of cycloheximide (0.35 mM) and in the absence (open bars) or in the presence of 10 PM adrenaline (filled bars), 0.4 mM dexamethasone (dotted bars) or 0.1 PM glucagon (dashed bars). At the end of the incubation the medium was obtained and used to determine Iipolytic activities. Results (corrected for inactivation as indicated in the text) are mean & S.E. of three experiments. Statistical differences versus basal activity were determined by the paired t-test. * P < 0.05; * * P < 0.01.
3). On the contrary, adrenaline was still able to decrease the amount of hepatic lipase activity secreted (Fig. 3). Discussion
Lipoprotein lipase activity in neonatal rat hepatocytes is regulated by hormones. In adult animals the effect of hormones on the enzyme activity varies from one tissue to another. Our results indicate that lipoprotein lipase is regulated in neonatal liver as it is in the heart of adult rats. Thus, it is increased in the liver of fasted neonates and the vascular pool accounts for most of the whole tissue increase. As in heart, insulin has no acute effect on the enzyme activity (it should be noted however that insulin is necessary to maintain high levels of enzyme activity in the heart, since streptozotocin-induced diabetes produces a decrease of the activity in both whole tissue [23] and isolated cardiomyocytes 1431). We have found that adrenaline and glucagon are able to increase the secretion of enzyme activity by isolated hepatocytes. The effect of adrenaline would be mediated by P-adrenergic receptors since P-blocker propranolol prevented, and the P-agonist isoprenaline reproduced, the effect of the catecholamine. These results suggest that lipoprotein lipase is regulated in hepatocytes by cyclic-AMP-dependent mechanism. It is known that both catecholamines and glucagon increase the activity released by heparin in perfused hearts from adult rats [21] and increase the secretion of enzyme activity from mesenchymal heart cells in culture [25].
88
Cyclic-AMP analogues also increase the secretion of lipoprotein lipase by cultured heart cells [25]. We also found that dexamethasone increased the secretion of lipoprotein Iipase activity from isolated hepatocytes, Glucocorticoids are also known to be major regulators of lipoprotein lipase activity in adult heart [44]. Two independent evidences indicate that regulation of lipoprotein lipase in neonatal hepatocytes appears to involve the synthesis of new enzyme protein: fi) The. amount of secreted activity was only increased after 3 h of incubation in the presence of hormones, and (ii) in the presence of cyclohexirnide, adrenaline, dexamethasone or glucagon did not increase the amount of secreted activity. Thus, we suggest that in neonatal rat hepatocytes lipoprotein lipase is regulated mainly at the transcriptional level. This is in agreement with our recent observation that the increase of the lipoprotein lipase activity in the liver of fasted neonates correlates with an increase in the content of lipoprotein lipase mRNA in the tissue (Peinado-Onsurbe, J., Ramirez, f. and Ltobera, M., unpublished data). In the human lipoprotein lipase gene 5’flanking region, consensus sequences for both glucocorticoid- and cyclic-AMP-responsivc elements have been tentatively identified [45], In adipose tissue and heart of adult rats 1461 and in adipose tissue of young adult guinea pigs [47], recent publications indicate that nutritional regulation of the enzyme also involves ~osttranslational events. Tn neonatal heart cells posttranslational processing of the enzyme is also under hormonal control [ZS]. Since those hormones that increased the secretion of lipoprotein lipase activity through a protein synthesis-dependent mechanisms did not affect the intracellular levels of activity, we suggest that secretion is limited by the rate of enzyme synthesis. This would explain the differences between fed and fasted rats in the amount of heparin-releasable activity in perfused livers and the lack of difference in the cell-associated ~residua~~ activity. Our results indicate that regulation of hepatic lipase activity is well developed in neonates since, as in adults, it is decreased in fasted animals and the heparin-reIeasabe compartment accounts for most of this effect. Also, in isolated hepatocytes the secretion of the enzyme does not appear to be affected by pancreatic, thyroid, or adrenocortical hormones or by epidermal growth factor. Both in neonates and in adults, adrenaline decreases the secretion of the enzyme. We have suggested that in adult rats catechohrmines are the hormones responsibfes for the effect of fasting on hepatic lipase activity !30]. But, while in adult animals the involvement of a,-adrenergic receptors is well established in the action of catecholamines [30,421, our results indicate that in neonates both LY,+and P-receptors are involved. In fact other effects of catecholamines that in ad&t rats are mediated by ar,-receptors,
as glycogen phosphorylase activation, in neonates are mediated by @-adrenergic receptors [48]. This is probably the consequence of the different proportion of both receptor types in adult and neonatal animals. Thus, while in adult males the liver contains most@ cy,-receptors, P-receptors account for about 50% of the total catecholamines binding sites in neonates [49,50]. In humans, both lipoprotein Iipase and hepatic lipase genes have nine exons similar in size and coding for homologous domains of their respective proteins [45,51-533. Lipoprotein lipase ‘gene contains an additional exon (number 10) which is transcribed but not translated [45,Sl]. In both genes consensus sequences for both glucocorticoid- and Cyclic-AMP-responsive elements have been tentatively identified [45,533. Despite similarities between hepatic lipase and lipoprotein lipase, the regulation of both enzymes in neonatal liver is clearly differentiated. The most striking difference is the effect of fasting that increases lipoprotein lipase but decreases hepatic hpase activities. Probably, this is the consequence of the different effect of hormones on each enzyme at the hepatocyte level. For example, adrenaline increased the secretion of lipoprotein lipase activity but decreased the release of hepatic lipase activity. Our results further indicate that, at least in our experimental conditions, enzyme synthesis is required for the acute reguiation of lipoprotein lipase but not for hepatic fipase. Whether this is the consequence of a different transcriptional control of each enzyme in this tissue have to be established. Both enzymes contain at least two complex N-linked oligosaccharide chains [9-111 and glycosylation appears to be required for enzyme secretion [7,9,1 I,%-56]. Thus, it would be thought that they might follow a common pathway in a cell type expressing both enzymes simultaneously. Our results suggest that the posttranslational processing appears to be particular for each enzyme. Evidence for this comes from the obse~atio~ that, in the presence of the protein synthesis inhibitor cycloheximide, adrenaline was able to selectively reduce the secretion of hepatic lipase activity. This effect may be due to a selective inhibition of the terminal glycosylation of the enzyme and/or to an increase in the intracellular degradation. In conclusion, neonatal rat hepatocytes may be a suitable non-transformed cell model to study the differences in the processing of lipoprotein lipase and hepatic lipase.
This work was supported by grants CPB88-0203 to M.L. and PB88-0191 to M.S.) from DGICYT, Ministerio de Education y Ciencia, Spain, and by CIRIT, General&at de Catalunva. C.S. is the recipient of a
89 F.P.I. fellowship from Ministerio de EducaciBn y Ciencia, Spain. References 1 Robinson, D.S. (1970) Compr. Biochem. 18, 51-116. 2 Cryer, A. (1981) Int. J. Biochem. 13, 525-541. 3 Camps, L., Reina, M., Llobera, M., Vilaro, S. and Olivecrona, T. (1990) Am. J. Physiol. 258, C673-C681. 4 Landin, B., Nilsson, A., Twu, JS. and Schotz, M.C. (19841 J. Lipid Res. 25, 559-563. 5 Simard, G. and Perret, 8. (1990) Ann. Biol. Clin. 48, 61-76. 6 Vilaro, S., Ramirez, I., Bengtsson-Olivecrona, G., Olivecrona, ‘I’. and Llobera, M. (1988) Biochim. Biophys. Acta 959, 106-117. 7 Laposata, E.A., Laboda, H.M. and Strauss III, J.F. (1987) J. Biol. Chem. 2625333-5338. 8 Doohttle, M.H., Ben-Zeev, O., Davis, R.C., Kirchgessner, T. and Schotz, M.C. (1990) in Obesity: Towards a Molecular Approach. (Bray, GA., Ricquier, D. and Spiegelman, B.M., eds.), pp. 237246, Wiley-Liss, New York. 9 Cisar, L. and Bensadoun, A. (1987) Biochim. Biophys. Acta 927, 305-314. 10 Semb, H. and Olivecrona, T. (1989) J. Bioi. Chem. 264,4195-4200. 11 Vannier, C. and Ailhaud, G. (1989) J. Biol. Chem. 264, 1320613216. 12 Stewart, J.E. and Schotz, MC. (1974) J. Biol. Chem. 249,904-907. 13 Chohan, P. and Cryer, A. (1980) Biochem. J. 186, 873-879. 14 Jansen, H., Kalkman, C., Zonnenveld, A.J. and Hiilsmann, W.C. (1979) FEBS Lett. 98, 299-302. 15 Eckel, R.H. (1987) in Lipoprotein lipase. (Borensztajn, J., ed.), pp. 79-132, Evener Publishing, Chicago. 16 Borensztajn, J. (19871 in Lipoprotein lipase. (Borensztajn, J., ed.1, pp. 133-148, Evener Publishing, Chicago. 17 Chan, B.L., Lisanti, M.P., Rodriguez-Boulan, E. and Saltiel, A.R. (1988) Science 241, 1670-1672. 18 Ong, J.M., Kirchgessner. T.G., Schotz, MC. and Kern, P.A. (1988) J. Biol. Chem. 263, 12933-12938. 19 Spooner, P.M., Chernick, S.S., Garrison, M.M. and Scow, R.O. (1979) J. Biol. Chem. 254, 1~21-1~29. 20 Semenkovich, C.F., Wims, M., Noe, L., Etienne, J. and Chart, L. (1989) J. Biol. Chem. 264, 9030-9038. 21 Stam, H. and Hiilsmann, W.C. (1984) Biochim. Biophys. Acta 794, 72-82. 22 Chajek-Shaul, T., Halimi, O., Ben-Naim, M., Stein, 0. and Stein, Y. (1989) Biochim. Biophys. Acta 1014, 178-183. 23 O’Looney, P., van der Maten, M. and Vahouny, G.V. (19831 J. Biol. Chem. 258, 12994-13001. 24 Ball, K.L., Speake, B.K. and Robinson, D.S. (1986) Biochim. Biophys. Acta 877, 399-405. 25 Friedman, G., Chajek-Shaul, T., Stein, O., Noe, L., Etienne, J. and Stein, Y. (1986) Biochim. Biophys. Acta 877, 112-120. 26 Elkels, R.S. and Hambley, J. (1977) Diabetes 26, 58-60. 27 Jansen, H. and Hiilsmann, W.C. (1975) Biochim. Biophys. Acta 398, 333-346. 28 Knauer, T.E., Woods, J.A., Lamb, R.G. and Fallon, H.J. (1982) J. Lipid Res. 23, 631-637.
29 Leitersdorf, E., Stein, 0. and Stein, Y. (1984) Biochim. Biophys. Acta 794, 261-268. 30 Peinado-Onsurbe, J., Soler, C., Galan, X., Poveda, B., Soley, M., Llobera, M. and Ramirez, 1. (1991) End~rinology 129, 25992696. 31 Burgaya, F., Peinado, J., Llobera, M. and Ramirez, I. (1989) Biosci. Rep. 9, 559-564. 32 Burgaya, F., Peinado, J., Vilarci, S., Llobera, M. and Ramirez, I. (1989) Biochem. J. 259, 159-166. 33 Grinberg, D.R., Ramirez, I., Vilaro, S., Reina, M., Llobera, M. and Herrera, E. (1985) Biochim. Biophys. Acta 833, 217-222. 34 Peinado, J., Martinez, J.V., Vilanova, J., Robert, M.Q., Llobera, M. and Ramirez, I. (1990) Biochim. Biophys. Acta 1043, 7-11. 35 Ramirez, I., Kryski, A.J., Ben-Zeev, O., Schotz, M.C. and Severson, D.L. (1985) Biochem. J. 232, 229-236. 36 Olivecrona, T. and Bengtsson, G. (1983) Biochim. Biophys. Acta 752,38-45. 37 Ehnholm, C., Shaw, W., Greten, H. and Brown, W.V. (1975) J. Biol. Chem. 250, 6756-6764. 38 Vilaro, S., Reina, M., Ramirez, I. and Llobera, M. (1986) Biochem. J. 236, 273-278. 39 Vannier, C. and Ailhaud, G. (1986) Biochim. Biophys. Acta 875, 324-333. 40 Severson, D.L., Lee, M. and Carroll, R. (1988) Mol. Cell. Biochem. 79, 17-24. 41 Choan, P. and Cryer, A. (1979) B&hem. J. 181, 83-93. 42 Shoondetwoerd, K., Hiilsmann, W.C. and Jansen, H. (19841 Biochim. Biophys. Acta 795, 481-486. 43 Ramirez, I. and Severson, D.L. (1986) Biochem. J. 238, 233-238. 44 Pedersen, M.E., Wolf, L.E. and Schotz, M.C. (1981) Biochim. Biophys. Acta 666, 191-197. 45 Kirchgessner, T.G., Chaut, J.C., Heinzmann, C., Etienne, J., Guilhot, S., Svenson, K., Ameis, D., Pilon, C., D’Auriol, L., Andalibi, A., Schotz, M.C., Caliber& F. and Lusis, A.J. (1989) Proc. Natl. Acad. Sci. USA 86, 9647-9651. 46 Doolittle, M.H., Ben-Zeev, O., Elovson, J., Martin, D. and Kirchgessner, T.G. (1990) J. Biol. Chem. 265, 4570-4577. 47 Semb, H. and Olivecrona, T. (1989) B&hem. J. 262, 505-511. 48 Noguchi, A., Jett, P.A. and Gold, A.H. (19851 Am. J. Physiol. 248, E560-E566. 49 Bendeck, J.L. and Noguchi, A. (1985) Pediatr. Res. 19, 862-868. 50 Studer, R.K. and Ganas, L. (1989) Am. J. Physiol. 256, E392E400. 51 Deeb, S.S. and Peng, R. (1989) Biochemistry 28, 4131-4135. 52 Cai, S.J., Wong, D.M., Chen, S.H. and Chan, L. (1989) Biochemistry 28, 8966-8971. 53 Ameis, D., Stahnke, G., Kobayashi, J., McLean, J., Lee, G., Buscher, M., Schotz, M.C. and Will, H. (1990) J. Biol. Chem. 265, 6552-6555. 54 Friedman, G., Chajek-Shaul, T., Etienne, J., Stein, 0. and Stein, Y. (1988) Biochim. Biophys. Acta 960,455-457. 55 Semenkovich, CF., Luo, CC, Nakanishi, M.K., Chen, S.H., Smith, L.C. and Chan, L. (1990) J. Biol. Chem. 265, 5429-5433. 56 Stahnke, G., Davis, R.C., Doolittle, M.H., Wong, II., Schotz, M.C. and Will, H. (1991) J. Lipid Res. 32, 477-484.
Biochimica et Biophysics Acta, 1125 (1992) 82-89 0 1992 Elsevier Science Publishers B.V. All rights reserved 0005-2760/92/$05.00
BBALIP 53877
Lipoprotein lipase and hepatic lipase activities are differentially regulated in isolated hepatocytes from neonatal rats Julia Peinado-Onsurbe,
Concepci6 Soler, Maria Soley, Miquel Llobera and Ignasi Ramirez
Departament de Bioquimicu i Fisiologia, Facultut de Biologia, Unicersitat de Barcelona, Barcelona (Spain)
(Received 17 September 1991) (Revised manuscript received 6 December 1991)
Key words: Lipoprotein lipase; Hepatic lipase; Hepatocyte; (Rat) Lipoprotein lipase and hepatic lipase are members of the lipase gene family sharing a high degree of homology in their amino acid sequences and genomic organization. We have recently shown that isolated hepatocytes from neonatal rats express both enzyme activities. We show here that both enzymes are, however, differentially regulated. Our main findings are: (8 fasting induced an increase of the lipoprotein lipase activity but a decrease of the hepatic lipase activity in whole liver, being in both cases the vascular (heparin-releasable) compartment responsible for these variations. (ii) In isolated hepatocytes, secretion of lipoprotein lipase activity was increased by adrenaline, dexamethasone and glucagon but was not affected by epidermal growth factor, insulin or triiodothyronine. On the contrary, secretion of hepatic lipase activity was decreased by adrenaline but was not affected by other hormones. (iii) The effect of adrenaline on lipoprotein lipase activity appeared to involve P-adrenergic receptors, but stimulation of both p- and cu,-receptors seemed to be required for the effect of this hormone on hepatic lipase activity. And (iv), increased secretion of lipoprotein lipase activity was only observed after 3 h of incubation with adrenaline and was blocked by cycloheximide. On the contrary, decreased secretion of hepatic lipase activity was already significant after 90 min of incubation and was not blocked by cycloheximide. We suggest that not only synthesis of both enzymes, but also the posttranslational processing, are under separate control in the neonatal rat liver.
Introduction
Lipoprotein lipase (EC 3.1.1.34) and hepatic lipase are lipolytic enzymes involved in the metabolism of circulating lipoproteins. Lipoprotein lipase hydrolyses triacylglycerols of chylomicra and very low density lipoproteins [l]. The enzyme is widely distributed among many tissues, but expresses high activity in adipose tissue, heart and skeletal muscle and lactating mammary gland [2]. In these tissues the functional fraction of the enzyme is located at the luminal side of endothelial cells but synthesized inside parenchymal cells 131. Hepatic lipase participates in the clearance of chylomicron remnants by the liver [4] and in the conversion of high density lipoprotein subfractions (HDL, into HDL,) contributing to the reverse cholesterol transport to the liver (see Ref. 5 for review). An
Correspondence: I. Ramiez, Unitat de Bioquimica i Biologia Molecular B, Departament de Bioquimica i Fisiologia, Facultat de Biologia, Universitat de Barcelona, Diagonal 645, 08071-Barcelona, Spain.
important fraction of the enzyme protein is also located at the vasculature [6], but the enzyme is synthesized inside hepatocytes [7]. Both enzymes are members of the lipase gene family and share many structural features in common [S]. Mature enzymes are glycoproteins with at least two complex N-linked oligosaccharide chains [9-l 11. In many cell systems it was shown that heparin stimulates the release of either enzyme to the incubation medium [12-141. The levels of tissue lipoprotein lipase activity change in response to a wide number of stimuli. In fasted animals, lipoprotein lipase activity is decreased in adipose tissue but is increased in heart (see Refs. 15, 16 for review). This is the consequence of the difference among tissues in the response to hormones. Thus, in cultured adipocytes insulin increases the secretion of lipoprotein lipase activity by both short-term protein synthesis-independent mechanisms (17) and long-term transcriptional- [lS] and posttrancriptional-dependent mechanisms [19,20]. On the contrary, neither in perfused hearts [21] nor in neonatal heart cells in culture [22] does insulin affect lipoprotein lipase activity. In-
83 sulin, however, restores normal lipase activity in perfused hearts from diabetic rats [23]. Inversely, catecholamines decrease synthesis and increase intracellular degradation of the enzyme in adipose tissue [241, while in neonatal heart cells [25], these hormones increase both synthesis and processing of the enzyme. Regulation of hepatic lipase activity is less understood. It is known that it is decreased in fasted rats [261 but the hormones and the mechanisms involved are still unclear. Activity is decreased in the liver of diabetic rats [27] but in vivo administration of either insulin or glucagon did not regulate the enzyme [281. In cultured hepatocytes, it was described that alteration of the intracelluIar cyclic-AMP levels affected hepatic lipase secretion [29]. However, we have suggested that, in adult rats, catecholamines are involved in the effect of fasting on hepatic lipase activity 1301. In recent years we have shown that neonatal rat hepatocytes express hepatic lipase [3I] and lipoprotein lipase activities [32]. Lipoprotein lipase activity in neonatal liver is under nutritional control [33]. The occurrence of both enzymes in neonatal liver allows the study of the similarities in the processing and regulation of both enzymes which are otherwise structurally related and similarly processed in different cells types. Here we show that both enzymes are independently regulated and evidence is obtained to indicate that their posttranslational processing is also under separate control. Material and Methods Animals Rats of the Wistar strain were used from our own colony, where the animals were fed standard chow diet (65% carbohydrate, 18% protein, 3% fat, 5% fibre, 5% minerals and vitamins, by weight). Newborn rats were either allowed to suckle with their mothers (fed neonates) or 8 h after birth were transferred to an humidified chamber at 28-30°C for 16 h (fasted neonates). Perji.ision ofisolated livers To study the effect of starvation on the vascular (heparin-releasable) lipolytic activities, isolated livers were perfused with heparin as previously described 1341. In summa~, isolated livers were perfused in a retrograde fashion at a flow rate of 0.8 ml/min with a Hepes (20 mM, pH 7.4)containing buffer (buffer A in Ref. 34) at 37°C. After 10 min, the perfusion medium was changed to a heparin (5 U/ml)-supplemented buffer A. The first 1.2 ml were discarded (void volume) and a single 10 ml fraction was collected on ice. To stabilize lipolytic activities, glycerol was immediately added to give a final concentration of 20%. The collected fractions and the perfused livers (homogenized
in 3 ml of buffer B: 10 mM Hepes, I mM dithiothreitoi, 1 mM EDTA, 0.25 M sucrose pH 7.5) were kept, until used to assay lipolytic activities, at -40°C. Hepatocyte isolation and incubation Hepatocytes from fed or fasted neonatal rats were isolated as previously described [32]. These preparations contained about 15% hemopoietic cells but these cells express neither Iipoprotein lipase [32] nor hepatic lipase [31] activities. Isolated hepatocytes (2.5 . lo6 hepatocytes/ml) were incubated in an amino acidsand vitamins-supplemented medium (buffer D in Ref. 32, but without insulin) for up to 3 h at 37°C under O,/CO, (19 : 1) atmosphere. At selected times, 0.5 mI, in duplicate, were taken and the cells were separated of the incubation medium by centrifugation. Both incubation medium and cell homogenates (obtained by sonication in buffer B) 1321were kept at -40°C until used to assay Iipofytic activities. Lipolytic activity assays Lipoprotein lipase activity was determined by the method of Ramirez et al. [35] with minor modifications as described in [32]. The assay mixture contained 0.6 mM glycerol tri[9,1~n)-3HJoleate (12 Ci/mol), 50 mM MgCl,, 0.05% albumin (fatty acid free), 3% serum (preheated 60 min at SOOC),25 mM Pipes (pH 7.5) and 0.02 ml of sample in a final volume of 0.2 ml. The incubation was carried out for 30 min at 25°C. The reaction was te~inated, and the E3H]oleate released was quantified, as previously described [32]. This method is not completely specific for lipoprotein lipase, but we have shown that chicken antiserum to bovine milk lipoprotein lipase (which does not cross-react with hepatic lipase [36]) inhibited 92% of the activity detected in neonatal liver [34] and 98% of the activity in isolated neonatal hepatocytes [32]. Hepatic lipase activity was determined by the method of Ehnholm et al. [37] as previously described [38], but with minor modifications. The assay mixture contained 2.5 mM glycerol tri[9,1~~)-3Hloleate (0.3 Ci/mol), 0.75 M NaCI, 3% albumin (fatty acid free), 50 mM Tris (pH 8.5) and 0.05 ml of sample in a final volume of 0.2 ml. Incubation conditions, reaction termination and quantification of [3H]oleate released were as above for lipoprotein lipase assay. This assay system was found to be specific for the enzyme [61. For both enzymes, one unit of enzyme activity was defined as the amount of enzyme that catalyses the reIease of 1 pmol of oleate per min. Thermal inactivation experiments Lipoprotein lipase and hepatic lipase activities from heparin-perfused neonatal livers were added (l/100 final dilution) to buffer D (see hepatocyte incubation conditions), supplemented or not with heparin (5
84 units/ml), and incubated at 37°C under constant shaking. At selected times a sample was taken and kept in ice-cold water until the incubation was terminated when lipoprotein lipase and hepatic lipase activities were determined in each sample. We previously found that both enzyme activities were stable at 4°C for several hours. From these inactivation experiments, inactivation constants were estimated and used to correct for inactivation both lipoprotein lipase and hepatic lipase activities secreted to the incubation medium. Correction of secreted activities for inactiuation
In agreement with previous reports [13,39,40], in our experimental conditions soluble lipoprotein lipase activity was unstable at 37°C (Fig. 1). The presence of heparin stabilized in part the enzyme activity. Hepatic lipase activity was less unstable at this temperature and also heparin had a protective effect. Data in Fig. 1 were fitted to: (I,
= u,,e
-kl
(1)
where a, is the remaining activity at time t, a,, is the activity at time t = 0 and k is the inactivation constant. Inactivation constant (k) values for lipoprotein lipase and hepatic lipase activities in the presence and in the absence of heparin are shown in Table I. These values may be used to correct for inactivation the amount of secreted activities as follows: In a system where there is a constant secretion of enzyme activity, the amount of apparently secreted
TABLE
Inactivation at 37°C
LIPASE
HEPATIC
constants
of lipoprotein lipase and hepatic lipase actil,itie.r
Data in Fig. 1 follow an exponential the activity at time t, a,, is the inactivation constant. k values were sis after linearization (Lnta, /a,)) = independent experiments.
curve: u, = a,, em”, where a, is activity at t = 0 and k is the determined by regression analy- kt ). Results are mean of two
k (min-‘) incubation condition:
- heparin
+ heparin
Lipoprotein lipase Hepatic lipase
1.73~100.29. 10 -’
0.79~10~* 0.12~1o~z
activity (AA,,) is:
in a given period of time (T = t, - t ,>
AA,,
= a2 - a, emh7
(2)
where a2 and a, are the measured activities at times t, and t,, respectively. This is still an underestimation of the actual amount of secreted activity (AA) because it only takes consideration the inactivation of the activity already present at t, but not the inactivation of the activity secreted throughout T. Assuming that the secretion rate is constant throughout T (as it was shown in other cell systems for relatively large periods of time [12,13]), then: AA=rT
(3)
where r is the secretion rate constant. In such system: AA,,
LIPOPROTEIN
I
= /‘r 0
LIPASE
emkr dt
(4)
thus: 1)
AAaI, = G(e-“‘-
(5)
and: r= 0
120180
0
incubation
time
60
60
120180
(emkT-
(6)
I)
combining Eqns. 2, 3 and 6
(mid
Fig. 1. Inactivation of lipoprotein lipase and hepatic lipase activities at 37°C. Neonatal rat liver was perfused with heparin as indicated in the text to obtain lipoprotein lipase and hepatic lipase. The perfusate was immediately diluted in buffer D (see Materials and Methods) supplemented (0) or not (0) with heparin (5 units/ml) and incubated at 37°C. At selected times a sample was taken and kept on ice until the end of the experiment when lipolytic activities were assayed. Results shown are from a representative of two independent experiment.
- (AA,,P
AA=
kT(a,
-a,
(e-k7-
e-l”) 1)
(7)
A detailed example of the use of this mathematical model to correct the amount of secreted lipoprotein lipase and hepatic lipase activities is shown in Table II. Note that the bigger differences between measured and corrected activities were obtained in lipoprotein
85
lipase activity because of the higher inactivation constant. Also, secretion of both enzyme activities (corrected) was almost linear during the 3 h incubation. Eqn. 7 was used in this paper to correct for inactivation the activities measured in the incubation medium. Chemicals
Insulin and glucagon were from Novo Industri, Copenhagen, Denmark. Triiodothyronine, dexamethasone, adrenaline, (+ / - Fisoprenaline, phenylephrine, prazosin and propranolol were purchased from Sigma Chemicals St. Louis, MO, USA. Epidermal growth factor purified from mouse submaxillary glands was a generous gift of Dr. Morley D. Hollenberg, University of Calgary, Calgary, Canada. All other chemicals were obtained as described elsewhere [34,36,38,41]. Results Effect of starvation on lipolytic activities in perfused lillers and isolated hepatocytes
We have previously found that lipoprotein lipase activity increases with fasting in the liver of neonatal rats [33]. We also found that this is the consequence of an increased amount of heparin-releasable (vascular) activity [34]. Here we also show that 93% of the increase in the total lipoprotein lipase activity detected in the whole liver was accounted for the increase in the heparin-releasable pool (Table III>. Neither the residual (heparin non-releasable) activity in the perfused liver, nor the activity in isolated (not incubated) hepatocytes was significantly higher in fasted than in fed neonates. Similar results were described in adult rat heart, where fasting produces an increase in the activity of the whole tissue but not in isolated heart cells [411.
TABLE
Contrary to what was observed with the lipoprotein lipase activity, hepatic lipase activity in the liver of fasted neonates was significantly decreased (Table III>. The heparin-released activity was also significantly decreased. The decrease in the vascular pool of the enzyme activity accounted for about 70% of the decrease in the total activity. The hepatic lipase activity in isolated hepatocytes from fasted neonates was not significantly different to that in fed neonates. These results are similar to what we have found in adult animals [30]. Effect of hormones on lipolytic acticities in isolated hepatocytes
To study the effect of hormones on lipolytic activities, isolated hepatocytes from fed neonates were incubated (3 h) in the presence of heparin (5 U/ml>. Lipoprotein lipase activity within the cells was not affected by the presence of either adrenaline (10 PM), dexamethasone (0.4 mM), epidermal growth factor (10 nM), glucagon (100 nM), insulin (100 nM), or triiodothyronine (10 nM) (Table IV). Adrenaline, dexamethasone and glucagon increased the amount of enzyme activity released to the incubation medium (corrected for inactivation as indicated in methods). At the concentrations used, epidermal growth factor, insulin or triiodothyronine did not affect lipoprotein lipase release. The hepatic lipase activity inside cells was also not affected by the presence of hormones but adrenaline decreased the amount of enzyme activity released to the medium (Table IV>. Any other hormone or factor did not affect the secretion of hepatic lipase activity. It is known that /?-receptors are involved in the regulation of lipoprotein lipase activity by catecholamines in heart mesenchymal cells [25]. On the con-
II
Correction
of lipoprotein lipase and hepatic lipase activities in the incubation
medium for inactivation
Isolated hepatocytes from fed neonates were incubated (2.5. 10h cells/ml) in the presence of heparin (5 units/ml) at 37°C at indicated times a sample was taken and the medium was used to determine lipolytic activities. The measured lipoprotein lipase and hepatic lipase activities were corrected for inactivation as indicated in the text. Results are from a representative experiment. Time tmin)
Measured (mU/lO’
Lipoprotein 0 45 90 180 Hepatic 0 45 90 180
activity cells)
A-‘& (a, - a2 emkT)
AA - kTAA,,/(e-kT-l))
Corrected activity (a, + ZA A) (mU/106 cells)
_ 0.244 0.235 0.339
_ 0.290 0.279 0.473
0.01 0.30 0.58 1.05
0.529 0.468 0.783
0.548 0.486 0.834
0.06 0.61 1.09 1.93
lipase activity (k = 0.79. 10W2) 0.01 0.25 0.41 0.54
lipase activity (k = 0.12. 1O-2) 0.06 0.58 1.02 1.68
86
trary, the inhibitory effect of these hormones on hepatic lipase activity is known to be mediated by Ly,-receptors in adult rat liver [30,42]. We found that the ~-agonist isoprenaline, but not the at-agonist phenylephrine, increased the amount of lipoprotein lipase
TABLE
activity released to the medium by isolated hepatocytes (Table V). Accordingly, the presence of the p-blocker propranolol, but not the cu,-blocker prazosin, prevented the stimulator effect of adrenaline. These resuits indicate that P-receptors are involved in the ac-
111
Effect uf fasting on lipoprotein /iparseand heputic lipase actit.ities in perfused ker
and isolated heputocytes
Livers from fed or fasted neonatal rats were perfused with heparin as indicated in methods. The perfusate was collected in a single 10 ml fraction. After the perfusion, the liver was homogenized. Isolated hepatocytes from either fed or fasted neonates were obtained from collagenase perfused livers as indicated in methods and immediately homogenized by sonication. Lipoprotein lipase and hepatic lipase activities were determined in heparin perfusates and in liver or cell homogenates. Results are means + S.E. of seven to ten animals. Statistical differences between fed and fasted animals were determined by Student’s t-test. a non-significant; * * P < 0.01; * * * P < 0.001. Lipoprotein
lipase
Hepatic
fed Heparin-perfused Eluted Residual Total Isolated
TABLE
bepatocytes
fasted
liver (mU/g) 36 I 43
rt:lO rt2 rt8
152 16 168
(mU/lO” cells) 0.07 rt 0.02
lipase
fed +27 ** * 4” +30 **
fasted
530 190 720
0.08 * 0.02 $’
a.135
+31 +3Y k42
327 102 429
0.01
f 20 * * +23 I’ +30 ***
O.l6*
0.02 ;’
1’~’
Effect of hormones OR lipa~tic acti~~itiesis isolated ~epatocytes from fed neonufe.~ Freshly isolated hepatocytes At the end of the incubation activities. Activity values in Statistical differences versus
were incubated (2.5~10s cells/ml) during 3 h at 37°C in the presence of indicated hormones and heparin (5 U/ml). cells were precipitated by centrifugation and both cells and medium kept at - 30°C until used to determine lipolytic medium were corrected for inactivation as indicated in the text. Results are mean & S.E. of seven experiments. basal activity were determined by the paired t-test. * P < 0.05; * * P < 0.01. Lipoprotein
Basal Adrenaline (10 FM) Dexamethasone (0.4 mM) Epidermal growth factor (10 nM) Glucagon (0.1 FM) insulin (0.1 FM) Triiodothyronine (10 nMf
TABLE
lipase fmU/106
cells)
Hepatic
cells
medium
cells
0.04 f 0.01 0.05 f 0.01 0.05 rt: 0.01 0.01 t 0.01 0.04io.01 0.02 rt 0.02 0.02 I: 0.02
1.01+0.01 1.85+0.38 1.93hO.45 l.OZrtO.31 1.32ri:O.24 1.01 *a,2 1.0710.43
0.07 0.06 0.04 0.08 0.04 0.07 0.08
* ** *
hpase (mU/106
cells) medium
f 0.02 f 0.03 & 0.01 i_ 0.03 i 0.03 i 0.04 + 0.04
1.92 + 0.2 1.29 kO.10 * 2.09 k 0.12 1.99*0.16 t .94 + 0.31 2.27 * 0.16 2.22 * 0.22
V
Effect of catecho~ami~te agonists and antagonists on lipolytic acriuities in isoiated ~epafocytes Cells were isolated and incubated as indicated lipoiytic activities. Activity values were corrected differences versus basal activity were determined
in Table IV. At the end of the incubation the medium was collected and used to determine for inactivation as indicated in the text. results are mean _+S.E. of five experiments. Statistical by the paired t-test. * P < 0.05; * * P i 0.01; * * * P < 0.001.
Lipoprotein (mu/lo6 Basal Isoprenaline (1 FM) Phenylephrine (1 PM) Adrenaline (10 PM) i propranolol f 1 p MI f prazosin t 1 ,u MI Isoprenaline c pheny~ephrine
1.14*0,13 1.78 f 0.22 1.37 f 0.27 1.80+0.28 1.24 * 0.26 1.98iO.05 1.74 * 0.55
Hepatic
lipase in medium cells)
(“rot 100 155+14 11X*20 158+ 14 108rt 5 176&20 150*20
lipase in medium
(mu/IO6
* * * *
1.78 + 0.30 1.49kO.20 1.44 + 0.25 1.07 * 0.21 1.2O~bO.17 1.51+0.16 1.09 f 0.09
cells)
(%) 100 86k 83+ 59rf68+ 85+ 61+
6 9 2”“” 10 * 2** 7**
87 tion of adrenaline. Isoprenaline or phenylephrine alone did not reproduce the effect of adrenaline on the secretion of hepatic lipase activity to the incubation medium (Table V). The presence of propranolol or prazosin prevented only in part the inhibitory effect of adrenaline. These results indicate that both types of receptors are partially involved in the effect of adrenaline on hepatic lipase activity. In keeping with this hypothesis, when both isoprenaline and phenylephrine were added together the amount of enzyme secreted was as decreased as when adrenaline was added alone (Table VI. The time-course of the effect of adrenaline, dexamethasone and glucagon on both enzyme activities is shown in Fig. 2. It was found that, while the amount of lipoprotein lipase activity secreted to the medium was only significantly increased after 3 h of incubation with either hormone, the effect of adrenaline on the release of hepatic lipase activity was clearly significant 90 min after the addition of the hormone. Neither dexamethasone nor glucagon affected the release of hepatic lipase activity to the medium during the whole incubation period studied. To determine if the effect of hormones was dependent on protein synthesis, isolated hepatocytes were incubated in the presence of 0.35 mM cycloheximide and either adrenaline (10 PM), dexamethasone (0.4 mMI, or glucagon (0.1 PM). It was found that cycloheximide decreased significantly the secretion to the medium of both lipoprotein lipase and hepatic lipase activities (Fig. 3). In the presence of cycloheximide neither hormone was able to increase the amount of lipoprotein lipase activity secreted to the medium (Fig.
LIPOPROTEIN
LIPASE
2.5 1
2.0 ** 1.5
'Z ,.i ,;. '1
1.0
2.0 1.5 1.0
t* 0.5
0.5
I il 0
60
0
120180
incubation
time
60
120180
(min)
Fig. 2. Effect of hormones on secretion of lipoprotein Iipase and hepatic Iipase activities. Isolated hepatocytes were incubated as indicated in Table III, in the absence (0) or in the presence of 10 PM adrenaline CO), 0.4 mM dexamethasone (0) or 0.1 FM glucagon (A ), at indicated times a sample was taken, centrifuged and the medium used to determine Iipolytic activities. Results (corrected for inactivation as indicated in the text) are mean+S.E. of four experiments. Statistical differences versus basal activity were determined by the paired t-test. * P < 0.05; * * P < 0.01.
LIPOPROTEIN
pxo.05
cycloheximide
-
;:.: :::: d LIPASE
I
HEPATIC
I
LIPASE
;$; :;;:: :.:.: .:.:.
I
I
p-co.01
+
+
Fig. 3. Effect of cycloheximide on the secretion of lipoprotein lipase and hepatic lipase activities. Isolated hepatocytes were incubated (3 h) as indicated in Table III in the absence or in the presence of cycloheximide (0.35 mM) and in the absence (open bars) or in the presence of 10 PM adrenaline (filled bars), 0.4 mM dexamethasone (dotted bars) or 0.1 PM glucagon (dashed bars). At the end of the incubation the medium was obtained and used to determine Iipolytic activities. Results (corrected for inactivation as indicated in the text) are mean & S.E. of three experiments. Statistical differences versus basal activity were determined by the paired t-test. * P < 0.05; * * P < 0.01.
3). On the contrary, adrenaline was still able to decrease the amount of hepatic lipase activity secreted (Fig. 3). Discussion
Lipoprotein lipase activity in neonatal rat hepatocytes is regulated by hormones. In adult animals the effect of hormones on the enzyme activity varies from one tissue to another. Our results indicate that lipoprotein lipase is regulated in neonatal liver as it is in the heart of adult rats. Thus, it is increased in the liver of fasted neonates and the vascular pool accounts for most of the whole tissue increase. As in heart, insulin has no acute effect on the enzyme activity (it should be noted however that insulin is necessary to maintain high levels of enzyme activity in the heart, since streptozotocin-induced diabetes produces a decrease of the activity in both whole tissue [23] and isolated cardiomyocytes 1431). We have found that adrenaline and glucagon are able to increase the secretion of enzyme activity by isolated hepatocytes. The effect of adrenaline would be mediated by P-adrenergic receptors since P-blocker propranolol prevented, and the P-agonist isoprenaline reproduced, the effect of the catecholamine. These results suggest that lipoprotein lipase is regulated in hepatocytes by cyclic-AMP-dependent mechanism. It is known that both catecholamines and glucagon increase the activity released by heparin in perfused hearts from adult rats [21] and increase the secretion of enzyme activity from mesenchymal heart cells in culture [25].
88
Cyclic-AMP analogues also increase the secretion of lipoprotein lipase by cultured heart cells [25]. We also found that dexamethasone increased the secretion of lipoprotein Iipase activity from isolated hepatocytes, Glucocorticoids are also known to be major regulators of lipoprotein lipase activity in adult heart [44]. Two independent evidences indicate that regulation of lipoprotein lipase in neonatal hepatocytes appears to involve the synthesis of new enzyme protein: fi) The. amount of secreted activity was only increased after 3 h of incubation in the presence of hormones, and (ii) in the presence of cyclohexirnide, adrenaline, dexamethasone or glucagon did not increase the amount of secreted activity. Thus, we suggest that in neonatal rat hepatocytes lipoprotein lipase is regulated mainly at the transcriptional level. This is in agreement with our recent observation that the increase of the lipoprotein lipase activity in the liver of fasted neonates correlates with an increase in the content of lipoprotein lipase mRNA in the tissue (Peinado-Onsurbe, J., Ramirez, f. and Ltobera, M., unpublished data). In the human lipoprotein lipase gene 5’flanking region, consensus sequences for both glucocorticoid- and cyclic-AMP-responsivc elements have been tentatively identified [45], In adipose tissue and heart of adult rats 1461 and in adipose tissue of young adult guinea pigs [47], recent publications indicate that nutritional regulation of the enzyme also involves ~osttranslational events. Tn neonatal heart cells posttranslational processing of the enzyme is also under hormonal control [ZS]. Since those hormones that increased the secretion of lipoprotein lipase activity through a protein synthesis-dependent mechanisms did not affect the intracellular levels of activity, we suggest that secretion is limited by the rate of enzyme synthesis. This would explain the differences between fed and fasted rats in the amount of heparin-releasable activity in perfused livers and the lack of difference in the cell-associated ~residua~~ activity. Our results indicate that regulation of hepatic lipase activity is well developed in neonates since, as in adults, it is decreased in fasted animals and the heparin-reIeasabe compartment accounts for most of this effect. Also, in isolated hepatocytes the secretion of the enzyme does not appear to be affected by pancreatic, thyroid, or adrenocortical hormones or by epidermal growth factor. Both in neonates and in adults, adrenaline decreases the secretion of the enzyme. We have suggested that in adult rats catechohrmines are the hormones responsibfes for the effect of fasting on hepatic lipase activity !30]. But, while in adult animals the involvement of a,-adrenergic receptors is well established in the action of catecholamines [30,421, our results indicate that in neonates both LY,+and P-receptors are involved. In fact other effects of catecholamines that in ad&t rats are mediated by ar,-receptors,
as glycogen phosphorylase activation, in neonates are mediated by @-adrenergic receptors [48]. This is probably the consequence of the different proportion of both receptor types in adult and neonatal animals. Thus, while in adult males the liver contains most@ cy,-receptors, P-receptors account for about 50% of the total catecholamines binding sites in neonates [49,50]. In humans, both lipoprotein Iipase and hepatic lipase genes have nine exons similar in size and coding for homologous domains of their respective proteins [45,51-533. Lipoprotein lipase ‘gene contains an additional exon (number 10) which is transcribed but not translated [45,Sl]. In both genes consensus sequences for both glucocorticoid- and Cyclic-AMP-responsive elements have been tentatively identified [45,533. Despite similarities between hepatic lipase and lipoprotein lipase, the regulation of both enzymes in neonatal liver is clearly differentiated. The most striking difference is the effect of fasting that increases lipoprotein lipase but decreases hepatic hpase activities. Probably, this is the consequence of the different effect of hormones on each enzyme at the hepatocyte level. For example, adrenaline increased the secretion of lipoprotein lipase activity but decreased the release of hepatic lipase activity. Our results further indicate that, at least in our experimental conditions, enzyme synthesis is required for the acute reguiation of lipoprotein lipase but not for hepatic fipase. Whether this is the consequence of a different transcriptional control of each enzyme in this tissue have to be established. Both enzymes contain at least two complex N-linked oligosaccharide chains [9-111 and glycosylation appears to be required for enzyme secretion [7,9,1 I,%-56]. Thus, it would be thought that they might follow a common pathway in a cell type expressing both enzymes simultaneously. Our results suggest that the posttranslational processing appears to be particular for each enzyme. Evidence for this comes from the obse~atio~ that, in the presence of the protein synthesis inhibitor cycloheximide, adrenaline was able to selectively reduce the secretion of hepatic lipase activity. This effect may be due to a selective inhibition of the terminal glycosylation of the enzyme and/or to an increase in the intracellular degradation. In conclusion, neonatal rat hepatocytes may be a suitable non-transformed cell model to study the differences in the processing of lipoprotein lipase and hepatic lipase.
This work was supported by grants CPB88-0203 to M.L. and PB88-0191 to M.S.) from DGICYT, Ministerio de Education y Ciencia, Spain, and by CIRIT, General&at de Catalunva. C.S. is the recipient of a
89 F.P.I. fellowship from Ministerio de EducaciBn y Ciencia, Spain. References 1 Robinson, D.S. (1970) Compr. Biochem. 18, 51-116. 2 Cryer, A. (1981) Int. J. Biochem. 13, 525-541. 3 Camps, L., Reina, M., Llobera, M., Vilaro, S. and Olivecrona, T. (1990) Am. J. Physiol. 258, C673-C681. 4 Landin, B., Nilsson, A., Twu, JS. and Schotz, M.C. (19841 J. Lipid Res. 25, 559-563. 5 Simard, G. and Perret, 8. (1990) Ann. Biol. Clin. 48, 61-76. 6 Vilaro, S., Ramirez, I., Bengtsson-Olivecrona, G., Olivecrona, ‘I’. and Llobera, M. (1988) Biochim. Biophys. Acta 959, 106-117. 7 Laposata, E.A., Laboda, H.M. and Strauss III, J.F. (1987) J. Biol. Chem. 2625333-5338. 8 Doohttle, M.H., Ben-Zeev, O., Davis, R.C., Kirchgessner, T. and Schotz, M.C. (1990) in Obesity: Towards a Molecular Approach. (Bray, GA., Ricquier, D. and Spiegelman, B.M., eds.), pp. 237246, Wiley-Liss, New York. 9 Cisar, L. and Bensadoun, A. (1987) Biochim. Biophys. Acta 927, 305-314. 10 Semb, H. and Olivecrona, T. (1989) J. Bioi. Chem. 264,4195-4200. 11 Vannier, C. and Ailhaud, G. (1989) J. Biol. Chem. 264, 1320613216. 12 Stewart, J.E. and Schotz, MC. (1974) J. Biol. Chem. 249,904-907. 13 Chohan, P. and Cryer, A. (1980) Biochem. J. 186, 873-879. 14 Jansen, H., Kalkman, C., Zonnenveld, A.J. and Hiilsmann, W.C. (1979) FEBS Lett. 98, 299-302. 15 Eckel, R.H. (1987) in Lipoprotein lipase. (Borensztajn, J., ed.), pp. 79-132, Evener Publishing, Chicago. 16 Borensztajn, J. (19871 in Lipoprotein lipase. (Borensztajn, J., ed.1, pp. 133-148, Evener Publishing, Chicago. 17 Chan, B.L., Lisanti, M.P., Rodriguez-Boulan, E. and Saltiel, A.R. (1988) Science 241, 1670-1672. 18 Ong, J.M., Kirchgessner. T.G., Schotz, MC. and Kern, P.A. (1988) J. Biol. Chem. 263, 12933-12938. 19 Spooner, P.M., Chernick, S.S., Garrison, M.M. and Scow, R.O. (1979) J. Biol. Chem. 254, 1~21-1~29. 20 Semenkovich, C.F., Wims, M., Noe, L., Etienne, J. and Chart, L. (1989) J. Biol. Chem. 264, 9030-9038. 21 Stam, H. and Hiilsmann, W.C. (1984) Biochim. Biophys. Acta 794, 72-82. 22 Chajek-Shaul, T., Halimi, O., Ben-Naim, M., Stein, 0. and Stein, Y. (1989) Biochim. Biophys. Acta 1014, 178-183. 23 O’Looney, P., van der Maten, M. and Vahouny, G.V. (19831 J. Biol. Chem. 258, 12994-13001. 24 Ball, K.L., Speake, B.K. and Robinson, D.S. (1986) Biochim. Biophys. Acta 877, 399-405. 25 Friedman, G., Chajek-Shaul, T., Stein, O., Noe, L., Etienne, J. and Stein, Y. (1986) Biochim. Biophys. Acta 877, 112-120. 26 Elkels, R.S. and Hambley, J. (1977) Diabetes 26, 58-60. 27 Jansen, H. and Hiilsmann, W.C. (1975) Biochim. Biophys. Acta 398, 333-346. 28 Knauer, T.E., Woods, J.A., Lamb, R.G. and Fallon, H.J. (1982) J. Lipid Res. 23, 631-637.
29 Leitersdorf, E., Stein, 0. and Stein, Y. (1984) Biochim. Biophys. Acta 794, 261-268. 30 Peinado-Onsurbe, J., Soler, C., Galan, X., Poveda, B., Soley, M., Llobera, M. and Ramirez, 1. (1991) End~rinology 129, 25992696. 31 Burgaya, F., Peinado, J., Llobera, M. and Ramirez, I. (1989) Biosci. Rep. 9, 559-564. 32 Burgaya, F., Peinado, J., Vilarci, S., Llobera, M. and Ramirez, I. (1989) Biochem. J. 259, 159-166. 33 Grinberg, D.R., Ramirez, I., Vilaro, S., Reina, M., Llobera, M. and Herrera, E. (1985) Biochim. Biophys. Acta 833, 217-222. 34 Peinado, J., Martinez, J.V., Vilanova, J., Robert, M.Q., Llobera, M. and Ramirez, I. (1990) Biochim. Biophys. Acta 1043, 7-11. 35 Ramirez, I., Kryski, A.J., Ben-Zeev, O., Schotz, M.C. and Severson, D.L. (1985) Biochem. J. 232, 229-236. 36 Olivecrona, T. and Bengtsson, G. (1983) Biochim. Biophys. Acta 752,38-45. 37 Ehnholm, C., Shaw, W., Greten, H. and Brown, W.V. (1975) J. Biol. Chem. 250, 6756-6764. 38 Vilaro, S., Reina, M., Ramirez, I. and Llobera, M. (1986) Biochem. J. 236, 273-278. 39 Vannier, C. and Ailhaud, G. (1986) Biochim. Biophys. Acta 875, 324-333. 40 Severson, D.L., Lee, M. and Carroll, R. (1988) Mol. Cell. Biochem. 79, 17-24. 41 Choan, P. and Cryer, A. (1979) B&hem. J. 181, 83-93. 42 Shoondetwoerd, K., Hiilsmann, W.C. and Jansen, H. (19841 Biochim. Biophys. Acta 795, 481-486. 43 Ramirez, I. and Severson, D.L. (1986) Biochem. J. 238, 233-238. 44 Pedersen, M.E., Wolf, L.E. and Schotz, M.C. (1981) Biochim. Biophys. Acta 666, 191-197. 45 Kirchgessner, T.G., Chaut, J.C., Heinzmann, C., Etienne, J., Guilhot, S., Svenson, K., Ameis, D., Pilon, C., D’Auriol, L., Andalibi, A., Schotz, M.C., Caliber& F. and Lusis, A.J. (1989) Proc. Natl. Acad. Sci. USA 86, 9647-9651. 46 Doolittle, M.H., Ben-Zeev, O., Elovson, J., Martin, D. and Kirchgessner, T.G. (1990) J. Biol. Chem. 265, 4570-4577. 47 Semb, H. and Olivecrona, T. (1989) B&hem. J. 262, 505-511. 48 Noguchi, A., Jett, P.A. and Gold, A.H. (19851 Am. J. Physiol. 248, E560-E566. 49 Bendeck, J.L. and Noguchi, A. (1985) Pediatr. Res. 19, 862-868. 50 Studer, R.K. and Ganas, L. (1989) Am. J. Physiol. 256, E392E400. 51 Deeb, S.S. and Peng, R. (1989) Biochemistry 28, 4131-4135. 52 Cai, S.J., Wong, D.M., Chen, S.H. and Chan, L. (1989) Biochemistry 28, 8966-8971. 53 Ameis, D., Stahnke, G., Kobayashi, J., McLean, J., Lee, G., Buscher, M., Schotz, M.C. and Will, H. (1990) J. Biol. Chem. 265, 6552-6555. 54 Friedman, G., Chajek-Shaul, T., Etienne, J., Stein, 0. and Stein, Y. (1988) Biochim. Biophys. Acta 960,455-457. 55 Semenkovich, CF., Luo, CC, Nakanishi, M.K., Chen, S.H., Smith, L.C. and Chan, L. (1990) J. Biol. Chem. 265, 5429-5433. 56 Stahnke, G., Davis, R.C., Doolittle, M.H., Wong, II., Schotz, M.C. and Will, H. (1991) J. Lipid Res. 32, 477-484.
