- Email: [email protected]
Comparison of the metabolism of [1,2,6,7-3H(N)]cholesteryl oleate, cholesteryl [9,10-3H]oleate, and cholesteryl [1-14C]oleate labeled lipoproteins in the rat
ATHEROSCLEROSIS Atherosclerosis 106 (1994) 203-21 I
Comparison of the metabolism of [ 1,2,6,7-3H(N)]cholesteryl oleate, cholesteryl [9, 10-3H]oleate, and cholesteryl [l- “C]oleate labeled lipoproteins in the rat A.H.M Terpstra Brown University Program-in-Medicine, Division of Nutrition and Metabolism, The Miriam Hospital, 164 Summit Avenue. Providence, RI 02906, USA
(Received 27 May 1993; revision received 16 November
1993; accepted
15 December 1993)
Abstract The intravascular metabolism of sterol labeled [ l,2,6,7-3H(N)]cholesteryl oleate and acyl labeled cholesteryl [9,10-‘Hloleate and cholesteryl [l-‘4C]oleate was compared in the rat, an animal species without plasma cholesteryl ester transfer activity (CETA). In a first series of studies, the metabolism of sterol labeled [l,2,6,7-3H(N)]cholesteryl oleate and acyl labeled cholesteryl [l-‘4C]oleate was compared, and the two tracers had identical plasma clearance rates when incorporated into human low density lipoproteins (LDL). The ‘H sterol labeled cholesteryl ester (CE), into rat Q- and /3however, had a plasma clearance rate lower than the 14C acyl labeled CE when incorporated migrating LDL and human or rat high density lipoproteins (HDL). Unesterifed 3H cholesterol reappeared in the plasma whereas the 14C radioactivity in the plasma remained associated with the CE. In a second set of studies, LDL and HDL were radiolabeled with cholesteryl [9,10-3H]oleate and cholesteryl [l-‘4C]oleate. Large amounts of 3H radioactivity that were dialyzable and not associated with the lipoprotein CE reappeared in the plasma during the kinetic studies. The two tracers had identical plasma disappearance rates when the plasma samples were dialyzed. The results of these studies indicate that the nature of the tracer used to trace lipoprotein CE can affect the estimated kinetic parameters of plasma CE. Key words: Cholesterol;
Cholesteryl
esters; Lipoproteins;
1. Introduction
Plasma cholesteryl ester (CE) metabolism [l] has been subject to renewed interest since the discovery of the CE transfer protein [2] and the re-
Cholesteryl
ester metabolism
cognition that the CE and protein components of plasma lipoproteins are metabolized separately [3-51. Radioactive CE are essential for investigating plasma CE metabolism, and radiolabeled CE can be prepared that contain the radioisotope in
0021-9150/94/%07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0021-9150(93)05201-F
A.H.M Terpstra/ Atherosclerosis106 (1994) 203-211
different locations of the CE molecule. There is evidence, however, that the position of the radioisotope in the CE molecule may affect the plasma disappearance rate of the tracers. It has been reported that the use of sterol labeled [ 1,2,6,7-3H(N)]cholesteryl oleate resulted in lower estimated catabolic rates than when acyl labeled cholesteryl [14C]oleate was used [6]. Plasma CE can be hydrolyzed after uptake by tissues, and the unesterilied cholesterol can be released and reesterified in the plasma by the lecithin:cholesterol acyltransferase (LCAT) enzyme. Recycling of tracer makes it difficult to interpret results of kinetic studies and may lead to underestimation of the kinetic parameters. Comparative studies of the metabolism of [ 1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [‘4C]oleate have been done before in the monkey [6], an animal species with high plasma CETA. Administration of radiolabeled CE incorporated in a specific lipoprotein class resulted, however, in a rapid transfer of the tracer to other lipoprotein classes. This transfer makes it difficult to study the metabolism of a tracer in a particular lipoprotein class. Therefore, in the present study the metabolism of the tracers was compared in the rat, an animal species without plasma CETA [7]. Further, radiolabeled CE can be prepared that contain 3H in the 9,10-position of the fatty acyl moiety (cholesteryl [9,10-3H]oleate). This tracer has also been used in kinetic studies [8] but has not yet been validated. In the present study, the metabolic properties of cholesteryl [9,10-3H]oleate incorporated into lipoproteins were also studied. Thus, the studies reported here were undertaken for the following purposes: (a) to compare the metabolism of [ 1,2,6,7-3H(N)]cholesteryl oleate, cholesteryl (9,l O-3H)oleate, and cholesteryl (lr4C)oleate labeled lipoproteins; (b) to examine whether the source of lipoproteins (human and rat low density lipoproteins (LDL) and high density lipoproteins (HDL)) would affect the magnitude of differences in metabolism of the three tracers when the studies were executed in a species lacking CETA. The hypothesis was that the three tracers had different metabolic properties and that the magnitude of these differences was related to the lipoprotein class that was labeled.
2. Materials and methods 2. I. Radioisotopes Cholesteryl [ 1- “C]oleate (specific activity 2.09 TBq/mol), cholesteryl [9,10-3H]oleate (specific activity 107 TBq/mol), and [1,2,6,7-3H(N)]cholesteryl oleate (specific activity 2.95 PBq/mol) were obtained from New England Nuclear (Boston, MA). The tracers were checked for purity by thin layer chromatography [9]. More than 99%:of the cholesteryl [ I-i4C]oleate and the cholesteryl [9, lo3H]oleate and more than 97% of the [1,2,6,7‘H(N)]cholesteryl oleate migrated together with a cholesteryl oleate standard. 2.2. Preparation of radiolabeled lipoproteins Blood was collected in tubes containing Naz-EDTA as anticoagulant, and benzamidine and aprotinin were added to a final concentration of 20 mmol and 234 000 kallikrein inhibitory units per liter plasma [lo]. Human LDL (1.019 < d < 1.063 kg/l) and HDL (1.080 < d < 1.210 kg/l) were prepared by sequential ultracentrifugation [l l] with a Ti 70 rotor and Quick Seal@ tubes (Beckman Inc., Palo Alto, CA). Rat lipoproteins were first isolated by ultracentrifugation in Quick Seal@ tubes at a background density of 1.21 kg/l. The density of the lipoprotein solution was adjusted to 1.25 kg/l and the HDL (1.070 < d < 1.130 kg/l) were isolated by density gradient ultracentrifugation [ 121. Cynomolgus monkey lipoprotein deficient serum (LPDS) was obtained by ultracentrifugation of serum at a background density of 1.21 kg/l. Rat LDL (1.019 < d < 1.063) were isolated by sequential ultracentrifugation in a Ti 60 rotor. Rat lipoproteins isolated at a density range of 1.019 c d c 1.063 kg/l contain /3migrating particles (LDL which have predominantly apolipoprotein B) and o-migrating particles, HDLi which contain apolipoprotein E as their major component [ 131. The LDL were fractionated by zonal electrophoresis. The LDL and 115 ml of Bio-Gel P-2 were equilibrated with 0.03 M sodium barbital, pH 8.6, and slab electrophoresis was performed at 10°C 300 V, 8-10 mA, for 16 h on an LKB Multiphor. (Y-and flmigrating lipoproteins were identified with a marker lane stained with Sudan Black. Lipopro-
205
A.H.M Terpstra/Atherosclerosis106 (1994) 203-211
teins were eluted with a 0.15 molll NaCl solution
2.4. Analyticalmethods
and concentrated by ultracentrifugation at a background density of 1.063 kg/l. Lipoproteins and LPDS were dialyzed against a solution containing 0.15 mol/l NaCl and 30 pmol/l thimerosal [ 111.Lipoprotein preparations were kept under nitrogen during dialysis and storage. Purity of isolated lipoproteins was assessed by agarose electrophoresis [14]. LCAT activity was inactivated by heating LPDS in a waterbath at 60°C for 10 min [15]. Radiolabeled CE were incorporated into LDL and HDL as described previously [16]. Briefly, tracer was dissolved in 50 ~1 absolute ethanol and added to a mixture of isolated lipoproteins and LPDS from cynomolgus monkeys which have high plasma CETA. Benzamidine (20 mmol/l) and aprotinin (234 000 kallikrein inhibitory units/l) were added as inhibitors of protein degradation [lo] and glutathione (0.65 mol/l) as a lipid antioxidant [17]. The mixture was incubated in a shaking waterbath at 37°C for 24 h under nitrogen, and the radiolabeled lipoproteins were reisolated by density gradient ultracentrifugation.
Total cholesterol was measured enzymatically [20]. Lipoprotein electrophoretograms [14] in 1%
2.3. Lipoprotein kinetic studies Sprague Dawley rats (weighing 260-420 g) that had been fed a chow diet were anesthetized with pentobarbital (5 mg/lOO g body weight), and a silastic catheter (i.d. 0.51 mm, o.d. 0.94 mm, Dow Corning Corporation, Midland, MI) was surgically implanted in the right jugular vein and advanced to the atrium. The catheters were externalized and kept open by a constant infusion of a 0.15 mol/l NaCl solution at a rate of 0.3 ml/h. The animals were equipped with a tethering system and could move freely in their cages. The radiolabeled lipoproteins were administered the following morning, and blood samples were collected into heparinized tubes at appropriate time points. The radioactivity in the plasma samples was expressed as a fraction of the radioactivity in the sample obtained 10 min after tracer infusion. Plasma decay curves were constructed, extrapolated to infinite time, and the apparent fractional catabolic rates (FCR) were calculated as the reciprocal of the area under the curves [18,19].
agarose (Corning Universal Electrophoresis Film, Corning Medical, Palo Alto, CA) were stained with Fat Red 7B Stain. Serum samples were extracted with a mixture of cold (-20°C) ether and methanol (2:1 v/v), and the proteins were sedimented by centrifugation. The lipid extract was dried under a stream of nitrogen and resolubilized in chloroform. An aliquot was applied to plastic thin layer chromatography plates (Eastman Kodak, Rochester, NY) and eluted with a mixture of hexane-ethanol-acetic acid (80:25:2 v/v) [9]. After lipids were visualized by iodine vapor, the sheets were cut and the unesterified and esterified cholesterol bands were assayed for 14C and 3H radioactivity was radioactivity. measured in a Packard Tricarb 4530 scintillation counter (Packard Instrument Company, Downersgrove, IL) using a dual label quench correcting program. Instagel (Packard Instrument Company) was used as scintillation liquid. 2.5. Statistics Differences between FCR of various radiolabeled CE preparations were statistically analyzed with a paired, two-tailed t-test. Correlations were analyzed with the Spearman rank order correlation test. Statistical analyses were done with the SigmaStatTM statistics software. 3. Results 3.1. Tracer validation
Radiolabeled esters were incorporated into lipoproteins as described previously [16], and it has been shown that in vitro labeled rat lipoproteins had metabolic properties identical to in vivo labeled rat lipoproteins. Further, tracer incorporation has been documented by repeat ultracentrifugation, agarose gel electrophoresis, and precipitation with heparin-MnC12 [16]. In the present studies, the rat and human LDL and HDL preparations were subjected to agarose gel electrophoresis and [21]. Between precipitation by heparin-MnC& 91% and 98% of the radioactivity in the rat and
206
A.H.M Terpstra/Atherosclerosis 106 (1994) 203-21 I
Table I Kinetic parameters of cholesteryl [I-‘4C]oleate and [ 1,2,6,7-3H(N)]cholesteryl oleate labeled lipoproteins in rats Infused lipoprotein class
Human LDL 14Cester 3H ester ‘H esters (corrected) Human HDL 14Cester ‘H ester ‘H ester (corrected) Rat P-migrating LDL 14Cester 3H ester 3H ester (corrected) Rat o-migrating LDL 14Cester ‘H ester 3H ester (corrected) Rat HDL 14C ester 3H ester 3H ester (corrected)
Dose administered
FCR (h-‘)
Radioactivity (W)
Cholesterol (rcmol)
87.5 f 3.8 148.6 f 6.4
4.58 f 0.20
71.6 A 0.4 128.7 f 7.3
Ratio 3H/‘4C Infusion sample
IO min samples
0.079 f 0.006 0.076 f 0.005* 0.079 A 0.006
1.70
1.72 f 0.01
3.32 f 0.19
0.244 f 0.022 0.163 f 0.013** 0.178 f 0.015**
1.80
1.78 f 0.05
79.0 f 5.4 206.7 f 14.2
0.30 f 0.02
0.1 I7 ?? 0.052 0.088 f 0.017; 0.095 A 0.018’
2.62
2.68 Z’E 0.02
73.5 f 227.8 f
3.1 9.5
0.26 f 0.01
0.1 IO f 0.007 0.087 f 0.006** 0.094 ?? 0.006**
3.10
3.14 f 0.02
82.9 f 198.2 f
2.8 6.6
4.24 f 0.14
0.190 f 0.038 0. I25 f 0.020; 0.139 ?? 0.021*
2.39
2.56 f 0.02
Data are expressed as mean ?? S.D. of 6 female rats. Differences between the FCR of the [l,2,6,7-3H(N)]cholesteryl oleate and the cholesteryl [I-‘4C]oleate in each experiment were statistically analyzed: *P < 0.01, **P < 0.001. The FCRs were calculated before and after correction for ‘H radioactivity in unesterilied cholesterol.
human LDL and HDL preparations migrated together with the parent lipoproteins on agarose gels. More than 99% of the radioactivity of human LDL preparations was precipitable with heparin-MnC12, whereas between 91% and 97% of the radioactivity in the human HDL preparations remained in the heparin-MnC12 supernatant. 3.2. Kinetic studies with [1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [I-‘4Cjoleate
The sterol labeled [1,2,6,7-3H(N)]cholesteryl oleate and the acyl labeled cholesteryl [l“C]oleate were incorporated into human LDL and HDL and intravenously administered to rats. The plasma FCR of the sterol labeled LDL tracer
was slightly lower than that of the acyl labeled LDL tracer when the two tracers were incorporated into human LDL (Table 1). In addition, unesterified 3H cholesterol appeared in the plasma during the study (Fig. l), whereas all 14C radioactivity remained associated with the CE fraction. The plasma FCRs of the ‘H and 14CCE were identical after correction for unesterilied 3H cholesterol in the plasma (Table 1, Fig. 2). The FCR of the 3H CE was 33% lower than that of the cholesteryl 14C CE when the tracers were incorporated into human HDL (Table 1). The difference was more pronounced than with human LDL, and the proportion of unesteritied 3H cholesterol appearing in the plasma was also higher (Fig. 1). Moreover, the FCR of ‘H CE was
207
A. H. M Terpstra / Atherosclerosis IO6 (1994) 203-211
I
.
I
.
I
20
10 Time
still 27% lower than that of the 14C CE (Table 1, Fig. 2) after correction for plasma unesterified 3H cholesterol. [ 1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [l-‘4C]oleate were also incorporated into rat lipoproteins. Ultracentrifugally isolated rat LDL were separated into (Y- and P-migrating lipoproteins and the metabolism of the CE in the two separate fractions was studied. The FCR of the 3H CE rat LDL was lower than that of the 14C CE (Table l), and the proportion of unesteritied 3H cholesterol appearing in the plasma was higher than when human LDL were used (Fig. 1). The FCR of 3H CE was higher even after correction for unesterified 3H cholesterol (Fig. 2). These findings were true for both the CPand the P-migrating rat LDL (Table 1). There were, however, no differences in clearance rate and metabolic behavior between the LDL fractions. Results with rat HDL were similar to those with human HDL. The plasma clearance rate of the 3H CE was significantly lower than that of the 14C CE (Table 1, Fig. 2) and a considerable proportion of unesterified 3H cholesterol appeared in the
8
30
40
(hours)
Fig. 1. Percentage of plasma ‘H radioactivity in unesteritied cholesterol after administration of [1,2,6,7-‘H(N)]cholesteryl oleate labeled lipoproteins in rats. Values are expressed as mean + S.D. of 6 female rats.
1 .oo
i 0 ..‘i .-c .-:
-w 0
’
0.01’
0 1.00
’
’ 20
10
B
’ 30
’ 40
‘i ‘I\ +,g
0.10 /
\
PIp;z .T
0.10
F
P
LL
rat 0.01’
0
B LDL ’ ’ 10
’ 20
’
Time
(hours)
’ 30
’ 0.01 L 40 0
&
I
.
10
Time
I
.
20
I
30
I 40
(hours)
Fig. 2. Plasma disappearance curves of (0) [l,2,6,7-3H (N)]cholesteryl oleate and (0) cholesteryl [I-‘4C]oleate labeled lipoproteins in rats. The curves for [I ,2,6,7-3H(N)]cholesteryl oleate were corrected for plasma unesteritied 3H cholesterol. Values are expressed as mean f SD. of 6 female rats.
208
A. H.M Terpstra / Atherosclerosis
106 (1994) 203-211
Table 2 Kinetic parameters of cholesteryl [I- 14C]oleate and cholesteryl [9,10-‘Hloleate labeled lipoproteins in rats Labeled lipoprotein class
Dose administered
Human LDL 14Cester ‘H ester Human HDL 14C ester ‘H ester Rat HDL 14Cester ‘H ester Rat HDL 14Cester ‘H ester ‘H ester(corrected)
FCR (h-l)
Radioactivity (kI4)
Cholesterol (pmol)
71.0 f 3.5 316.1 f 15.7
10.62 f 0.53
69.0 f 1.9 323.2 zt 8.9
6.94 zt 0.19
Ratio ‘H/14C Infusion sample
IO min samples
0.055 f 0.004 0.051 f 0.003*
4.45
4.40 f 0.05
0.190 A 0.016 0.127 f 0.009,
4.67
4.73
?? 0.03
71.4 * 4.4 165.4 f 10.2
1.44 ?? 0.04
0.174 * 0.005 0.101 * 0.005*
2.32
2.35 f 0.02
102.6 zt 4.6 433.3 f 19.3
2.22 f 0.10
0.164 f 0.045 0.109 zt 0.018’ 0.167 z’z0.048
4.22
4.15 * 0.02
Data are expressed as mean f SD. of 5 rats (6 rats in the study with human LDL). Female rats were used in the experiments with human LDL and rat HDL and male rats in the study with human HDL. Differences between the FCR of the cholesteryl [9,10-3H]oleate and the cholesteryl [I- “C]oleate in each experiment were statistically analyzed: *P < 0.001. In one study, FCR were also calculated after correction for plasma 3H radioactivity that was dialyzable.
1.00 x .e > .<
0.10
0 ._
1
-0
2 0 ._
.e
LDL
human
0.01 I
1
l.OO?,
.
human f
10
0
c
.
.
20
,
.
I 0.011
1
30
,
.
,
.
0
40 1.00
.
HDL
’
10
.
’
’
’
’ 40
30
20
: .\ 0.10 T
h, A:\
0.10 '?*-•
1 *1*.
. rat HDL I
0.01
0
*
10
I
,
20
Time
(hours)
I 30
* 40
0.01 l
0
’
’
10
3
Time
’
20
’
’
30
3
J
40
(hours)
Fig. 3. Plasma disappearance curves of (0) cholesteryl [9,10-3H]oleate and (0) cholesteryl [I-‘4C]oleate labeled lipoproteins in rats. The curves of cholesteryl [9,10-‘Hloleate ester were corrected for dialyzable 3H radioactivity in only one study with rat HDL (A). Values are expressed as mean f S.D. of 6 female rats (human LDL studies), 5 female rats (rat LDL and HDL studies), and 5 male rats (human HDL studies).
A. H.M Terpstra /Atherosclerosis
209
106 (1994) 203-21 I
plasma (Fig. 1). The 14Cradioactivity in the plasma remained associated with the CE fraction in all experiments.
clearance rates of the two tracers were similar after dialysis of the plasma samples (Table 2, Fig. 3). 4. Discussion
3.3. Kinetics cholesteryl
of cholesteryl [I-‘“CJoleate
[9,10-:‘H]oleate
and
In the next set of studies, the metabolism of fatty acyl chain labeled CE containing either 14Cin the l-position or ‘H in the 9,10-position was studied. The total ‘H radioactivity disappeared from the plasma at a lower rate than the 14Cradioactivity, and this difference was more pronounced when the two tracers were incorporated in HDL than in LDL (Table 2, Fig. 3). In addition, 3H radioactivity in the plasma became non-precipitable with a 100/otrichloroacetic acid (TCA) solution. All 14C in the plasma remained TCA precipitable during the kinetic studies. Considerable amounts of ‘H radioactivity in the plasma samples could be removed by dialysis and the ‘H radioactivity left was TCA precipitable and migrated together with a CE standard on thin layer chromatography plates. These observations indicated that administration of cholesteryl [9, 10-3H]oleate labeled lipoproteins resulted in the generation of a 3H labeled compound or compounds that were dialyzable and not associated with the lipoprotein CE. In a subsequent study, the two tracers were incorporated into rat HDL, and all the plasma samples collected during the kinetic study were dialyzed. A considerable amount of plasma 3H radioactivity became dialyzable, more than 60% at 36 h after tracer administration (Fig. 4). The
60
60 40 20 0
0
10 Time
20
30
40
(hours)
Fig. 4. Percentage of dialyzable 3H radioactivity in plasma after administration of cholesteryl [9,10-3H]oleate labeled rat HDL to rats. Values are expressed as mean * S.D. of 5 female rats.
The use of radiolabeled CE is a prerequisite for the study of lipoprotein CE metabolism. Radiolabeled CE can be prepared that contain the radioisotope in different locations of the CE molecule. The results of the present studies in rats, however, indicated that the use of these differently radiolabeled CE can result in different estimated kinetic parameters for CE metabolism. Lipoprotein [1,2,6,7-3H(N)]cholesteryl oleate containing the radioisotope in the sterol moiety disappeared at a lower rate from rat plasma than cholesteryl [ l“C]oleate containing the label in the acyl moiety. A considerable amount of unesterified 3H labeled cholesterol reappeared in the plasma, and even after correction for plasma 3H radioactivity in unesteritied cholesterol the [ 1,2,6,7-3H(N)]cholesteryl oleate had a lower plasma disappearance rate than the cholesteryl [ 1-‘4C]oleate. This observation suggests that recycling of 3H labeled cholesterol had taken place. The [ 1,2,6,7-3H(N)] cholesteryl oleate was probably hydrolyzed after uptake by tissues, released in the plasma, and reesterified by the LCAT enzyme. These findings in rats, which lack CETA, are in agreement with studies in monkeys, which have high CETA. Thirty years ago, Portman and Sugano [22] noted that after administration of sterol labeled lipoprotein CE of oleic, palmitic, or linolenic acid to cebus monkeys, the labeled cholesterol redistributed to other plasma CE. The specific activities of all the various species of plasma CE became similar after l-3 days. Similar results have been reported by Thomas and Rude1 [6]. They used doubly labeled cholesteryl oleate, labeled with 3H in the sterol moiety and with 14C in the l-position of the oleoyl component. The doubly labeled ester was incorporated into HDL and administered to African green monkeys. It was observed that plasma CE other than cholesteryl oleate became radiolabeled with 3H cholesterol and that this redistribution gradually took place over time. Studies by Goldberg et al. [23] have also shown that unesteritied radiolabeled
210
cholesterol was generated in the plasma of rhesus monkeys when LDL or HDL containing sterol labeled CE were infused. Thus, recycling of sterol labeled lipoprotein CE leads to labeling of CE species other than the originally labeled CE species. Further, this process of recycling may also result in labeling of lipoprotein classes other than the originally labeled lipoprotein classes. LCAT acts mainly on HDL, and radiolabeled CE generated by LCAT will reappear in the HDL. Newly labeled CE may to some extent also arise from acyl-coenzyme A:cholesterolacyltransferase (ACAT) activity in the liver and then be excreted in the VLDL which can be converted into LDL. Thus, recycled CE may ultimately reappear in all the lipoprotein classes even in the absence of CETA in the plasma. Kinetic studies with cholesteryl [9,10-3H]oleate resulted in the generation of a 3H containing compound or compounds that were no longer associated with the CE. On the other hand, in kinetic studies with cholesteryl [l- 14C]oleate, all the radioactivity remained in the CE fraction during the time course of the kinetic studies. These different metabolic properties of the two tracers may be related to the position of the radioisotope in the oleate molecule or to a different metabolic fate of the radioactive carbon and hydrogen atoms. Plasma CE can be hydrolyzed after tissue uptake, and the 14C atom might be catabolized to a product that does not reappear in the plasma or that is cleared rapidly, such as 14C02. The 3H products, possibly 3H,0, may remain in the plasma or may be cleared slowly. The present study does not identify these 3H products but indicates that they can be removed by dialysis; the FCR of cholesteryl [9,10-3H]oleate and cholesteryl [l-14C]oleate were similar after correction for dialyzable 3H radioactivity in the plasma. Thus, cholesteryl [9,103H]oleate can be used to trace lipoprotein CE provided that a correction is made for plasma 3H radioactivity that is not associated with the CE and that can be removed by dialysis. In
A.H.M Terpstra / Atherosclerosis 104 (1994) 203-211
place. Thus, the metabolism of tracers in a particular lipoprotein class could be studied. Others have used the monkey, a species with high CETA, to study the properties of [1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [ l- “C]oleate labeled HDL [6]. They observed a rapid transfer of labeled HDL CE to other lipoprotein classes, and after 90 min the specific activity of LDL and HDL had become identical. Plasma lipoprotein CE represented a single pool, and the disappearance rate of labeled CE from the plasma was independent of the lipoprotein class that was originally labeled. In the rat, there was a considerable difference in plasma disappearance rates of the CE in the various lipoprotein classes. Human LDL CE had lower plasma disappearance rates than human HDL CE and the same was true for rat LDL and HDL CE (Tables 1 and 2). There were, however, also differences in FCR between experiments in which the same tracers and lipoprotein classes were used. For example, the FCR in the experiment with cholesteryl [ l- “C]oleate labeled human LDL in Table 1 were somewhat higher than the FCR in the experiment with cholesteryl [l14C]oleate labeled human LDL in Table 2. These differences may be attributed to differences in lipoprotein preparations and rats. Further, the magnitude of the differences in FCR of the various tracers was also related to the lipoprotein class that was labeled. For example, there was no difference in FCR between [ 1,2,6,73H(N)]cholesteryl oleate and cholesteryl [l14C]oleate when the two tracers were incorporated into human LDL (Table, 1). Human LDL had low turnover rates, which might explain the fact that only small amounts of unesterified 3H cholesterol reappeared in the plasma (Fig. 1) and that no significant amounts of recycled 3H cholesteryl esters were generated (Fig. 2). On the other hand, the FCR of the [ 1,2,6,7-3H(N)]cholesteryl oleate labeled HDL was 27% lower than the FCR of the cholesteryl [l-‘4C]oleate labeled HDL. The HDL had considerably higher turnover rates than the LDL, and also larger amounts of unesterified 3H cholesterol reappeared in the plasma. As a consequence, more recycled 3H CE could probably be generated by the LCAT enzyme. There was a significant positive correlation between the FCR
A.H.M Terpstra / Atherosclerosis 106 (1994) 203-211
of cholesteryl [l-14C]oleate and the differences in FCR of the cholesteryl [l-14C]oleate and [1,2,6,73H(N)]cholesteryl oleate (corrected for unesteritied plasma cholesterol) (r = 0.957, P < 0.005, n = 30, all individual data from the experiments in Table 1). Similarly, there was a significant positive correlation between the FCR of cholesteryl [l“C]oleate and the differences in FCR of the cholesteryl [ l- “C]oleate and cholesteryl [9, lo3H]oleate (not corrected for dialyzable plasma 3H radioactivity) (r = 0.883, P < 0.005, n = 21, all individual data from the experiments in Table 2). Thus, a higher turnover rate of the lipoproteins was associated with accumulation of more plasma 3H radioactivity that was not associated with the CE.
211
9
10
II
bit plasma, B&him. Biophys. Acta, 409 (1975) 393. Ha, Y.C., Calve& G.D., McIntosh, G.H. and Barter, P.J., A physiological role for the esterified cholesterol transfer protein: in vivo studies in rabbits and pigs. Metabolism, 30 (1981) 380. Stein, Y., Dabach, Y., Hollander, G., Halparin, G. and Stein, O., Metabolism of HDL-cholesteryl esters in the rat, studied by a non-hydrolyzable analog, cholesteryl linoleyl ether, B&him. Biophys. Acta, 752 (1983) 98. Glass, C., Pittman, R.C., Civen, M. and Steinberg, D., Uptake of high-density lipoprotein-associated apoprotein A-I and cholesterol esters by I6 tissues of the rat in vivo and by adrenal cells and hepatocytes in vitro, J. Biol. Chem., 260 (1985) 744 Thomas, M.S. and Rudel, L.L., Intravascular metabolism of lipoprotein cholesteryl esters in African Green monkeys: differential fate of doubly labeled cholesteryl oleate, J. Lipid Res., 28 (1987) 572. Ha, Y.C. and Barter, P.J.. Differences in plasma cholesteryl ester transfer activity in sixteen vertebrate species, Comp. Biochem. Physiol., 71B (1982) 265. Goldberg, D.I., Beltz, W.F. and Pittman, R.C., Evaluation of pathways for the cellular uptake of high density lipoprotein cholesterol esters in rabbits, J. Clin. Invest., 87 (1991) 331.
20 (1967) 1057. Cardin, A.D., Witt, K.R., Chao, J., Margolis, H.S., Donaldson, V.H. and Jackson R.L., Degradation of apoprotein B-100 of human plasma low density lipoproteins by tissue and plasma kallikreins, J. Biol. Chem., 259 (1984) 8522. Lindgren, F.T., Preparative ultracentrifugal laboratory procedures and suggestions for lipoprotein analysis. In:
I2
Perkins, E.G. (Ed.), Analysis of Lipids and Lipoproteins, American Oil Chemist’s Society, Champaign, IL, 1975, pp. 204-224. Terpstra, A.H.M. and Pels, A.E., Isolation of plasma li-
I3
poproteins by a combination of differential and density gradient ultracentrifugation, Fresenius Z. Anal. Chem., 330 (1988) 149. Weisgraber, K.H., Mahley, R.W. and Assmann, G., The
14
rat arginine-rich apoprotein and its redistribution following injection of iodinated lipoproteins into normal and hypercholesterolemic rats, Atherosclerosis, 28 (1977) 121. Noble, R.P., Electrophoretic separation of plasma lipo-
5. References Goodman, D.S., Cholesterol ester metabolism, Physiol. Rev., 45 (1965) 747. Zilversmit, D.B., Hughes, L.B. and Balmer, J., Stimulation of cholesterol ester exchange by lipoprotein-free rab-
Nichaman, M.Z., Sweeley, C.C. and Olson, R.E., Plasma fatty acids in normolipemic and hyperlipemic subjects during fasting and after linoleate feeding, J. Clin. Nutr.,
I6
proteins in agarose gel, J. Lipid Res., 9 (1968) 693. Glomset, J.A., The plasma lecithin:cholesterol acyltransferase, J. Lipid Res., 9 (1968) 1557. Terpstra, A.H.M., Nicolosi, R.J. and Herbert, P.N., In
I7
vitro incorporation of radiolabeled cholesteryl esters into high and low density lipoproteins, J. Lipid Res., 30 (1989) 1663. Lee, D.M., Malondialdehyde formation in stored plasma,
I5
I8 19
20
21
22
23
Biochem. Biophys. Res. Commun., 95 (1980) 1663. Shipley, R.A. and Clark, R.E., Tracer Methods for In Vivo Kinetics, Academic Press, New York, 1972. Terpstra, A.H.M., Determination of the kinetic parameters of the cholesteryl esters and proteins of plasma lipoproteins, Math. Comput. Modelling, I2 (1989) 601. Allain, C.C., Poon, L.S., Chen, C.S.G., Richmond, W. and Fu, P.C., Enzymatic determination of total serum cholesterol, Clin. Chem., 20 (1974) 470. Burnstein, M., Scholnick, H.R. and Martin, R.J., Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions, J. Lipid Res., 11 (1970) 583. Portman, O.W. and Sugano, M., The hydrolysis and fatty acid exchange of cholesterol esters administered in lipoproteins to cebus monkeys, Arch. Biochem. Biophys., 102 (1963) 335. Goldberg, I.R., Rosenfeld, R.S., Paul, 1. and Leeman, B., Generation of plasma free cholesterol from circulating lipoprotein-associated cholesteryl ester, Am. J. Physiol., 250 (1986) E265.
Comparison of the metabolism of [ 1,2,6,7-3H(N)]cholesteryl oleate, cholesteryl [9, 10-3H]oleate, and cholesteryl [l- “C]oleate labeled lipoproteins in the rat A.H.M Terpstra Brown University Program-in-Medicine, Division of Nutrition and Metabolism, The Miriam Hospital, 164 Summit Avenue. Providence, RI 02906, USA
(Received 27 May 1993; revision received 16 November
1993; accepted
15 December 1993)
Abstract The intravascular metabolism of sterol labeled [ l,2,6,7-3H(N)]cholesteryl oleate and acyl labeled cholesteryl [9,10-‘Hloleate and cholesteryl [l-‘4C]oleate was compared in the rat, an animal species without plasma cholesteryl ester transfer activity (CETA). In a first series of studies, the metabolism of sterol labeled [l,2,6,7-3H(N)]cholesteryl oleate and acyl labeled cholesteryl [l-‘4C]oleate was compared, and the two tracers had identical plasma clearance rates when incorporated into human low density lipoproteins (LDL). The ‘H sterol labeled cholesteryl ester (CE), into rat Q- and /3however, had a plasma clearance rate lower than the 14C acyl labeled CE when incorporated migrating LDL and human or rat high density lipoproteins (HDL). Unesterifed 3H cholesterol reappeared in the plasma whereas the 14C radioactivity in the plasma remained associated with the CE. In a second set of studies, LDL and HDL were radiolabeled with cholesteryl [9,10-3H]oleate and cholesteryl [l-‘4C]oleate. Large amounts of 3H radioactivity that were dialyzable and not associated with the lipoprotein CE reappeared in the plasma during the kinetic studies. The two tracers had identical plasma disappearance rates when the plasma samples were dialyzed. The results of these studies indicate that the nature of the tracer used to trace lipoprotein CE can affect the estimated kinetic parameters of plasma CE. Key words: Cholesterol;
Cholesteryl
esters; Lipoproteins;
1. Introduction
Plasma cholesteryl ester (CE) metabolism [l] has been subject to renewed interest since the discovery of the CE transfer protein [2] and the re-
Cholesteryl
ester metabolism
cognition that the CE and protein components of plasma lipoproteins are metabolized separately [3-51. Radioactive CE are essential for investigating plasma CE metabolism, and radiolabeled CE can be prepared that contain the radioisotope in
0021-9150/94/%07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0021-9150(93)05201-F
A.H.M Terpstra/ Atherosclerosis106 (1994) 203-211
different locations of the CE molecule. There is evidence, however, that the position of the radioisotope in the CE molecule may affect the plasma disappearance rate of the tracers. It has been reported that the use of sterol labeled [ 1,2,6,7-3H(N)]cholesteryl oleate resulted in lower estimated catabolic rates than when acyl labeled cholesteryl [14C]oleate was used [6]. Plasma CE can be hydrolyzed after uptake by tissues, and the unesterilied cholesterol can be released and reesterified in the plasma by the lecithin:cholesterol acyltransferase (LCAT) enzyme. Recycling of tracer makes it difficult to interpret results of kinetic studies and may lead to underestimation of the kinetic parameters. Comparative studies of the metabolism of [ 1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [‘4C]oleate have been done before in the monkey [6], an animal species with high plasma CETA. Administration of radiolabeled CE incorporated in a specific lipoprotein class resulted, however, in a rapid transfer of the tracer to other lipoprotein classes. This transfer makes it difficult to study the metabolism of a tracer in a particular lipoprotein class. Therefore, in the present study the metabolism of the tracers was compared in the rat, an animal species without plasma CETA [7]. Further, radiolabeled CE can be prepared that contain 3H in the 9,10-position of the fatty acyl moiety (cholesteryl [9,10-3H]oleate). This tracer has also been used in kinetic studies [8] but has not yet been validated. In the present study, the metabolic properties of cholesteryl [9,10-3H]oleate incorporated into lipoproteins were also studied. Thus, the studies reported here were undertaken for the following purposes: (a) to compare the metabolism of [ 1,2,6,7-3H(N)]cholesteryl oleate, cholesteryl (9,l O-3H)oleate, and cholesteryl (lr4C)oleate labeled lipoproteins; (b) to examine whether the source of lipoproteins (human and rat low density lipoproteins (LDL) and high density lipoproteins (HDL)) would affect the magnitude of differences in metabolism of the three tracers when the studies were executed in a species lacking CETA. The hypothesis was that the three tracers had different metabolic properties and that the magnitude of these differences was related to the lipoprotein class that was labeled.
2. Materials and methods 2. I. Radioisotopes Cholesteryl [ 1- “C]oleate (specific activity 2.09 TBq/mol), cholesteryl [9,10-3H]oleate (specific activity 107 TBq/mol), and [1,2,6,7-3H(N)]cholesteryl oleate (specific activity 2.95 PBq/mol) were obtained from New England Nuclear (Boston, MA). The tracers were checked for purity by thin layer chromatography [9]. More than 99%:of the cholesteryl [ I-i4C]oleate and the cholesteryl [9, lo3H]oleate and more than 97% of the [1,2,6,7‘H(N)]cholesteryl oleate migrated together with a cholesteryl oleate standard. 2.2. Preparation of radiolabeled lipoproteins Blood was collected in tubes containing Naz-EDTA as anticoagulant, and benzamidine and aprotinin were added to a final concentration of 20 mmol and 234 000 kallikrein inhibitory units per liter plasma [lo]. Human LDL (1.019 < d < 1.063 kg/l) and HDL (1.080 < d < 1.210 kg/l) were prepared by sequential ultracentrifugation [l l] with a Ti 70 rotor and Quick Seal@ tubes (Beckman Inc., Palo Alto, CA). Rat lipoproteins were first isolated by ultracentrifugation in Quick Seal@ tubes at a background density of 1.21 kg/l. The density of the lipoprotein solution was adjusted to 1.25 kg/l and the HDL (1.070 < d < 1.130 kg/l) were isolated by density gradient ultracentrifugation [ 121. Cynomolgus monkey lipoprotein deficient serum (LPDS) was obtained by ultracentrifugation of serum at a background density of 1.21 kg/l. Rat LDL (1.019 < d < 1.063) were isolated by sequential ultracentrifugation in a Ti 60 rotor. Rat lipoproteins isolated at a density range of 1.019 c d c 1.063 kg/l contain /3migrating particles (LDL which have predominantly apolipoprotein B) and o-migrating particles, HDLi which contain apolipoprotein E as their major component [ 131. The LDL were fractionated by zonal electrophoresis. The LDL and 115 ml of Bio-Gel P-2 were equilibrated with 0.03 M sodium barbital, pH 8.6, and slab electrophoresis was performed at 10°C 300 V, 8-10 mA, for 16 h on an LKB Multiphor. (Y-and flmigrating lipoproteins were identified with a marker lane stained with Sudan Black. Lipopro-
205
A.H.M Terpstra/Atherosclerosis106 (1994) 203-211
teins were eluted with a 0.15 molll NaCl solution
2.4. Analyticalmethods
and concentrated by ultracentrifugation at a background density of 1.063 kg/l. Lipoproteins and LPDS were dialyzed against a solution containing 0.15 mol/l NaCl and 30 pmol/l thimerosal [ 111.Lipoprotein preparations were kept under nitrogen during dialysis and storage. Purity of isolated lipoproteins was assessed by agarose electrophoresis [14]. LCAT activity was inactivated by heating LPDS in a waterbath at 60°C for 10 min [15]. Radiolabeled CE were incorporated into LDL and HDL as described previously [16]. Briefly, tracer was dissolved in 50 ~1 absolute ethanol and added to a mixture of isolated lipoproteins and LPDS from cynomolgus monkeys which have high plasma CETA. Benzamidine (20 mmol/l) and aprotinin (234 000 kallikrein inhibitory units/l) were added as inhibitors of protein degradation [lo] and glutathione (0.65 mol/l) as a lipid antioxidant [17]. The mixture was incubated in a shaking waterbath at 37°C for 24 h under nitrogen, and the radiolabeled lipoproteins were reisolated by density gradient ultracentrifugation.
Total cholesterol was measured enzymatically [20]. Lipoprotein electrophoretograms [14] in 1%
2.3. Lipoprotein kinetic studies Sprague Dawley rats (weighing 260-420 g) that had been fed a chow diet were anesthetized with pentobarbital (5 mg/lOO g body weight), and a silastic catheter (i.d. 0.51 mm, o.d. 0.94 mm, Dow Corning Corporation, Midland, MI) was surgically implanted in the right jugular vein and advanced to the atrium. The catheters were externalized and kept open by a constant infusion of a 0.15 mol/l NaCl solution at a rate of 0.3 ml/h. The animals were equipped with a tethering system and could move freely in their cages. The radiolabeled lipoproteins were administered the following morning, and blood samples were collected into heparinized tubes at appropriate time points. The radioactivity in the plasma samples was expressed as a fraction of the radioactivity in the sample obtained 10 min after tracer infusion. Plasma decay curves were constructed, extrapolated to infinite time, and the apparent fractional catabolic rates (FCR) were calculated as the reciprocal of the area under the curves [18,19].
agarose (Corning Universal Electrophoresis Film, Corning Medical, Palo Alto, CA) were stained with Fat Red 7B Stain. Serum samples were extracted with a mixture of cold (-20°C) ether and methanol (2:1 v/v), and the proteins were sedimented by centrifugation. The lipid extract was dried under a stream of nitrogen and resolubilized in chloroform. An aliquot was applied to plastic thin layer chromatography plates (Eastman Kodak, Rochester, NY) and eluted with a mixture of hexane-ethanol-acetic acid (80:25:2 v/v) [9]. After lipids were visualized by iodine vapor, the sheets were cut and the unesterified and esterified cholesterol bands were assayed for 14C and 3H radioactivity was radioactivity. measured in a Packard Tricarb 4530 scintillation counter (Packard Instrument Company, Downersgrove, IL) using a dual label quench correcting program. Instagel (Packard Instrument Company) was used as scintillation liquid. 2.5. Statistics Differences between FCR of various radiolabeled CE preparations were statistically analyzed with a paired, two-tailed t-test. Correlations were analyzed with the Spearman rank order correlation test. Statistical analyses were done with the SigmaStatTM statistics software. 3. Results 3.1. Tracer validation
Radiolabeled esters were incorporated into lipoproteins as described previously [16], and it has been shown that in vitro labeled rat lipoproteins had metabolic properties identical to in vivo labeled rat lipoproteins. Further, tracer incorporation has been documented by repeat ultracentrifugation, agarose gel electrophoresis, and precipitation with heparin-MnC12 [16]. In the present studies, the rat and human LDL and HDL preparations were subjected to agarose gel electrophoresis and [21]. Between precipitation by heparin-MnC& 91% and 98% of the radioactivity in the rat and
206
A.H.M Terpstra/Atherosclerosis 106 (1994) 203-21 I
Table I Kinetic parameters of cholesteryl [I-‘4C]oleate and [ 1,2,6,7-3H(N)]cholesteryl oleate labeled lipoproteins in rats Infused lipoprotein class
Human LDL 14Cester 3H ester ‘H esters (corrected) Human HDL 14Cester ‘H ester ‘H ester (corrected) Rat P-migrating LDL 14Cester 3H ester 3H ester (corrected) Rat o-migrating LDL 14Cester ‘H ester 3H ester (corrected) Rat HDL 14C ester 3H ester 3H ester (corrected)
Dose administered
FCR (h-‘)
Radioactivity (W)
Cholesterol (rcmol)
87.5 f 3.8 148.6 f 6.4
4.58 f 0.20
71.6 A 0.4 128.7 f 7.3
Ratio 3H/‘4C Infusion sample
IO min samples
0.079 f 0.006 0.076 f 0.005* 0.079 A 0.006
1.70
1.72 f 0.01
3.32 f 0.19
0.244 f 0.022 0.163 f 0.013** 0.178 f 0.015**
1.80
1.78 f 0.05
79.0 f 5.4 206.7 f 14.2
0.30 f 0.02
0.1 I7 ?? 0.052 0.088 f 0.017; 0.095 A 0.018’
2.62
2.68 Z’E 0.02
73.5 f 227.8 f
3.1 9.5
0.26 f 0.01
0.1 IO f 0.007 0.087 f 0.006** 0.094 ?? 0.006**
3.10
3.14 f 0.02
82.9 f 198.2 f
2.8 6.6
4.24 f 0.14
0.190 f 0.038 0. I25 f 0.020; 0.139 ?? 0.021*
2.39
2.56 f 0.02
Data are expressed as mean ?? S.D. of 6 female rats. Differences between the FCR of the [l,2,6,7-3H(N)]cholesteryl oleate and the cholesteryl [I-‘4C]oleate in each experiment were statistically analyzed: *P < 0.01, **P < 0.001. The FCRs were calculated before and after correction for ‘H radioactivity in unesterilied cholesterol.
human LDL and HDL preparations migrated together with the parent lipoproteins on agarose gels. More than 99% of the radioactivity of human LDL preparations was precipitable with heparin-MnC12, whereas between 91% and 97% of the radioactivity in the human HDL preparations remained in the heparin-MnC12 supernatant. 3.2. Kinetic studies with [1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [I-‘4Cjoleate
The sterol labeled [1,2,6,7-3H(N)]cholesteryl oleate and the acyl labeled cholesteryl [l“C]oleate were incorporated into human LDL and HDL and intravenously administered to rats. The plasma FCR of the sterol labeled LDL tracer
was slightly lower than that of the acyl labeled LDL tracer when the two tracers were incorporated into human LDL (Table 1). In addition, unesterified 3H cholesterol appeared in the plasma during the study (Fig. l), whereas all 14C radioactivity remained associated with the CE fraction. The plasma FCRs of the ‘H and 14CCE were identical after correction for unesterilied 3H cholesterol in the plasma (Table 1, Fig. 2). The FCR of the 3H CE was 33% lower than that of the cholesteryl 14C CE when the tracers were incorporated into human HDL (Table 1). The difference was more pronounced than with human LDL, and the proportion of unesteritied 3H cholesterol appearing in the plasma was also higher (Fig. 1). Moreover, the FCR of ‘H CE was
207
A. H. M Terpstra / Atherosclerosis IO6 (1994) 203-211
I
.
I
.
I
20
10 Time
still 27% lower than that of the 14C CE (Table 1, Fig. 2) after correction for plasma unesterified 3H cholesterol. [ 1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [l-‘4C]oleate were also incorporated into rat lipoproteins. Ultracentrifugally isolated rat LDL were separated into (Y- and P-migrating lipoproteins and the metabolism of the CE in the two separate fractions was studied. The FCR of the 3H CE rat LDL was lower than that of the 14C CE (Table l), and the proportion of unesteritied 3H cholesterol appearing in the plasma was higher than when human LDL were used (Fig. 1). The FCR of 3H CE was higher even after correction for unesterified 3H cholesterol (Fig. 2). These findings were true for both the CPand the P-migrating rat LDL (Table 1). There were, however, no differences in clearance rate and metabolic behavior between the LDL fractions. Results with rat HDL were similar to those with human HDL. The plasma clearance rate of the 3H CE was significantly lower than that of the 14C CE (Table 1, Fig. 2) and a considerable proportion of unesterified 3H cholesterol appeared in the
8
30
40
(hours)
Fig. 1. Percentage of plasma ‘H radioactivity in unesteritied cholesterol after administration of [1,2,6,7-‘H(N)]cholesteryl oleate labeled lipoproteins in rats. Values are expressed as mean + S.D. of 6 female rats.
1 .oo
i 0 ..‘i .-c .-:
-w 0
’
0.01’
0 1.00
’
’ 20
10
B
’ 30
’ 40
‘i ‘I\ +,g
0.10 /
\
PIp;z .T
0.10
F
P
LL
rat 0.01’
0
B LDL ’ ’ 10
’ 20
’
Time
(hours)
’ 30
’ 0.01 L 40 0
&
I
.
10
Time
I
.
20
I
30
I 40
(hours)
Fig. 2. Plasma disappearance curves of (0) [l,2,6,7-3H (N)]cholesteryl oleate and (0) cholesteryl [I-‘4C]oleate labeled lipoproteins in rats. The curves for [I ,2,6,7-3H(N)]cholesteryl oleate were corrected for plasma unesteritied 3H cholesterol. Values are expressed as mean f SD. of 6 female rats.
208
A. H.M Terpstra / Atherosclerosis
106 (1994) 203-211
Table 2 Kinetic parameters of cholesteryl [I- 14C]oleate and cholesteryl [9,10-‘Hloleate labeled lipoproteins in rats Labeled lipoprotein class
Dose administered
Human LDL 14Cester ‘H ester Human HDL 14C ester ‘H ester Rat HDL 14Cester ‘H ester Rat HDL 14Cester ‘H ester ‘H ester(corrected)
FCR (h-l)
Radioactivity (kI4)
Cholesterol (pmol)
71.0 f 3.5 316.1 f 15.7
10.62 f 0.53
69.0 f 1.9 323.2 zt 8.9
6.94 zt 0.19
Ratio ‘H/14C Infusion sample
IO min samples
0.055 f 0.004 0.051 f 0.003*
4.45
4.40 f 0.05
0.190 A 0.016 0.127 f 0.009,
4.67
4.73
?? 0.03
71.4 * 4.4 165.4 f 10.2
1.44 ?? 0.04
0.174 * 0.005 0.101 * 0.005*
2.32
2.35 f 0.02
102.6 zt 4.6 433.3 f 19.3
2.22 f 0.10
0.164 f 0.045 0.109 zt 0.018’ 0.167 z’z0.048
4.22
4.15 * 0.02
Data are expressed as mean f SD. of 5 rats (6 rats in the study with human LDL). Female rats were used in the experiments with human LDL and rat HDL and male rats in the study with human HDL. Differences between the FCR of the cholesteryl [9,10-3H]oleate and the cholesteryl [I- “C]oleate in each experiment were statistically analyzed: *P < 0.001. In one study, FCR were also calculated after correction for plasma 3H radioactivity that was dialyzable.
1.00 x .e > .<
0.10
0 ._
1
-0
2 0 ._
.e
LDL
human
0.01 I
1
l.OO?,
.
human f
10
0
c
.
.
20
,
.
I 0.011
1
30
,
.
,
.
0
40 1.00
.
HDL
’
10
.
’
’
’
’ 40
30
20
: .\ 0.10 T
h, A:\
0.10 '?*-•
1 *1*.
. rat HDL I
0.01
0
*
10
I
,
20
Time
(hours)
I 30
* 40
0.01 l
0
’
’
10
3
Time
’
20
’
’
30
3
J
40
(hours)
Fig. 3. Plasma disappearance curves of (0) cholesteryl [9,10-3H]oleate and (0) cholesteryl [I-‘4C]oleate labeled lipoproteins in rats. The curves of cholesteryl [9,10-‘Hloleate ester were corrected for dialyzable 3H radioactivity in only one study with rat HDL (A). Values are expressed as mean f S.D. of 6 female rats (human LDL studies), 5 female rats (rat LDL and HDL studies), and 5 male rats (human HDL studies).
A. H.M Terpstra /Atherosclerosis
209
106 (1994) 203-21 I
plasma (Fig. 1). The 14Cradioactivity in the plasma remained associated with the CE fraction in all experiments.
clearance rates of the two tracers were similar after dialysis of the plasma samples (Table 2, Fig. 3). 4. Discussion
3.3. Kinetics cholesteryl
of cholesteryl [I-‘“CJoleate
[9,10-:‘H]oleate
and
In the next set of studies, the metabolism of fatty acyl chain labeled CE containing either 14Cin the l-position or ‘H in the 9,10-position was studied. The total ‘H radioactivity disappeared from the plasma at a lower rate than the 14Cradioactivity, and this difference was more pronounced when the two tracers were incorporated in HDL than in LDL (Table 2, Fig. 3). In addition, 3H radioactivity in the plasma became non-precipitable with a 100/otrichloroacetic acid (TCA) solution. All 14C in the plasma remained TCA precipitable during the kinetic studies. Considerable amounts of ‘H radioactivity in the plasma samples could be removed by dialysis and the ‘H radioactivity left was TCA precipitable and migrated together with a CE standard on thin layer chromatography plates. These observations indicated that administration of cholesteryl [9, 10-3H]oleate labeled lipoproteins resulted in the generation of a 3H labeled compound or compounds that were dialyzable and not associated with the lipoprotein CE. In a subsequent study, the two tracers were incorporated into rat HDL, and all the plasma samples collected during the kinetic study were dialyzed. A considerable amount of plasma 3H radioactivity became dialyzable, more than 60% at 36 h after tracer administration (Fig. 4). The
60
60 40 20 0
0
10 Time
20
30
40
(hours)
Fig. 4. Percentage of dialyzable 3H radioactivity in plasma after administration of cholesteryl [9,10-3H]oleate labeled rat HDL to rats. Values are expressed as mean * S.D. of 5 female rats.
The use of radiolabeled CE is a prerequisite for the study of lipoprotein CE metabolism. Radiolabeled CE can be prepared that contain the radioisotope in different locations of the CE molecule. The results of the present studies in rats, however, indicated that the use of these differently radiolabeled CE can result in different estimated kinetic parameters for CE metabolism. Lipoprotein [1,2,6,7-3H(N)]cholesteryl oleate containing the radioisotope in the sterol moiety disappeared at a lower rate from rat plasma than cholesteryl [ l“C]oleate containing the label in the acyl moiety. A considerable amount of unesterified 3H labeled cholesterol reappeared in the plasma, and even after correction for plasma 3H radioactivity in unesteritied cholesterol the [ 1,2,6,7-3H(N)]cholesteryl oleate had a lower plasma disappearance rate than the cholesteryl [ 1-‘4C]oleate. This observation suggests that recycling of 3H labeled cholesterol had taken place. The [ 1,2,6,7-3H(N)] cholesteryl oleate was probably hydrolyzed after uptake by tissues, released in the plasma, and reesterified by the LCAT enzyme. These findings in rats, which lack CETA, are in agreement with studies in monkeys, which have high CETA. Thirty years ago, Portman and Sugano [22] noted that after administration of sterol labeled lipoprotein CE of oleic, palmitic, or linolenic acid to cebus monkeys, the labeled cholesterol redistributed to other plasma CE. The specific activities of all the various species of plasma CE became similar after l-3 days. Similar results have been reported by Thomas and Rude1 [6]. They used doubly labeled cholesteryl oleate, labeled with 3H in the sterol moiety and with 14C in the l-position of the oleoyl component. The doubly labeled ester was incorporated into HDL and administered to African green monkeys. It was observed that plasma CE other than cholesteryl oleate became radiolabeled with 3H cholesterol and that this redistribution gradually took place over time. Studies by Goldberg et al. [23] have also shown that unesteritied radiolabeled
210
cholesterol was generated in the plasma of rhesus monkeys when LDL or HDL containing sterol labeled CE were infused. Thus, recycling of sterol labeled lipoprotein CE leads to labeling of CE species other than the originally labeled CE species. Further, this process of recycling may also result in labeling of lipoprotein classes other than the originally labeled lipoprotein classes. LCAT acts mainly on HDL, and radiolabeled CE generated by LCAT will reappear in the HDL. Newly labeled CE may to some extent also arise from acyl-coenzyme A:cholesterolacyltransferase (ACAT) activity in the liver and then be excreted in the VLDL which can be converted into LDL. Thus, recycled CE may ultimately reappear in all the lipoprotein classes even in the absence of CETA in the plasma. Kinetic studies with cholesteryl [9,10-3H]oleate resulted in the generation of a 3H containing compound or compounds that were no longer associated with the CE. On the other hand, in kinetic studies with cholesteryl [l- 14C]oleate, all the radioactivity remained in the CE fraction during the time course of the kinetic studies. These different metabolic properties of the two tracers may be related to the position of the radioisotope in the oleate molecule or to a different metabolic fate of the radioactive carbon and hydrogen atoms. Plasma CE can be hydrolyzed after tissue uptake, and the 14C atom might be catabolized to a product that does not reappear in the plasma or that is cleared rapidly, such as 14C02. The 3H products, possibly 3H,0, may remain in the plasma or may be cleared slowly. The present study does not identify these 3H products but indicates that they can be removed by dialysis; the FCR of cholesteryl [9,10-3H]oleate and cholesteryl [l-14C]oleate were similar after correction for dialyzable 3H radioactivity in the plasma. Thus, cholesteryl [9,103H]oleate can be used to trace lipoprotein CE provided that a correction is made for plasma 3H radioactivity that is not associated with the CE and that can be removed by dialysis. In
A.H.M Terpstra / Atherosclerosis 104 (1994) 203-211
place. Thus, the metabolism of tracers in a particular lipoprotein class could be studied. Others have used the monkey, a species with high CETA, to study the properties of [1,2,6,7-3H(N)]cholesteryl oleate and cholesteryl [ l- “C]oleate labeled HDL [6]. They observed a rapid transfer of labeled HDL CE to other lipoprotein classes, and after 90 min the specific activity of LDL and HDL had become identical. Plasma lipoprotein CE represented a single pool, and the disappearance rate of labeled CE from the plasma was independent of the lipoprotein class that was originally labeled. In the rat, there was a considerable difference in plasma disappearance rates of the CE in the various lipoprotein classes. Human LDL CE had lower plasma disappearance rates than human HDL CE and the same was true for rat LDL and HDL CE (Tables 1 and 2). There were, however, also differences in FCR between experiments in which the same tracers and lipoprotein classes were used. For example, the FCR in the experiment with cholesteryl [ l- “C]oleate labeled human LDL in Table 1 were somewhat higher than the FCR in the experiment with cholesteryl [l14C]oleate labeled human LDL in Table 2. These differences may be attributed to differences in lipoprotein preparations and rats. Further, the magnitude of the differences in FCR of the various tracers was also related to the lipoprotein class that was labeled. For example, there was no difference in FCR between [ 1,2,6,73H(N)]cholesteryl oleate and cholesteryl [l14C]oleate when the two tracers were incorporated into human LDL (Table, 1). Human LDL had low turnover rates, which might explain the fact that only small amounts of unesterified 3H cholesterol reappeared in the plasma (Fig. 1) and that no significant amounts of recycled 3H cholesteryl esters were generated (Fig. 2). On the other hand, the FCR of the [ 1,2,6,7-3H(N)]cholesteryl oleate labeled HDL was 27% lower than the FCR of the cholesteryl [l-‘4C]oleate labeled HDL. The HDL had considerably higher turnover rates than the LDL, and also larger amounts of unesterified 3H cholesterol reappeared in the plasma. As a consequence, more recycled 3H CE could probably be generated by the LCAT enzyme. There was a significant positive correlation between the FCR
A.H.M Terpstra / Atherosclerosis 106 (1994) 203-211
of cholesteryl [l-14C]oleate and the differences in FCR of the cholesteryl [l-14C]oleate and [1,2,6,73H(N)]cholesteryl oleate (corrected for unesteritied plasma cholesterol) (r = 0.957, P < 0.005, n = 30, all individual data from the experiments in Table 1). Similarly, there was a significant positive correlation between the FCR of cholesteryl [l“C]oleate and the differences in FCR of the cholesteryl [ l- “C]oleate and cholesteryl [9, lo3H]oleate (not corrected for dialyzable plasma 3H radioactivity) (r = 0.883, P < 0.005, n = 21, all individual data from the experiments in Table 2). Thus, a higher turnover rate of the lipoproteins was associated with accumulation of more plasma 3H radioactivity that was not associated with the CE.
211
9
10
II
bit plasma, B&him. Biophys. Acta, 409 (1975) 393. Ha, Y.C., Calve& G.D., McIntosh, G.H. and Barter, P.J., A physiological role for the esterified cholesterol transfer protein: in vivo studies in rabbits and pigs. Metabolism, 30 (1981) 380. Stein, Y., Dabach, Y., Hollander, G., Halparin, G. and Stein, O., Metabolism of HDL-cholesteryl esters in the rat, studied by a non-hydrolyzable analog, cholesteryl linoleyl ether, B&him. Biophys. Acta, 752 (1983) 98. Glass, C., Pittman, R.C., Civen, M. and Steinberg, D., Uptake of high-density lipoprotein-associated apoprotein A-I and cholesterol esters by I6 tissues of the rat in vivo and by adrenal cells and hepatocytes in vitro, J. Biol. Chem., 260 (1985) 744 Thomas, M.S. and Rudel, L.L., Intravascular metabolism of lipoprotein cholesteryl esters in African Green monkeys: differential fate of doubly labeled cholesteryl oleate, J. Lipid Res., 28 (1987) 572. Ha, Y.C. and Barter, P.J.. Differences in plasma cholesteryl ester transfer activity in sixteen vertebrate species, Comp. Biochem. Physiol., 71B (1982) 265. Goldberg, D.I., Beltz, W.F. and Pittman, R.C., Evaluation of pathways for the cellular uptake of high density lipoprotein cholesterol esters in rabbits, J. Clin. Invest., 87 (1991) 331.
20 (1967) 1057. Cardin, A.D., Witt, K.R., Chao, J., Margolis, H.S., Donaldson, V.H. and Jackson R.L., Degradation of apoprotein B-100 of human plasma low density lipoproteins by tissue and plasma kallikreins, J. Biol. Chem., 259 (1984) 8522. Lindgren, F.T., Preparative ultracentrifugal laboratory procedures and suggestions for lipoprotein analysis. In:
I2
Perkins, E.G. (Ed.), Analysis of Lipids and Lipoproteins, American Oil Chemist’s Society, Champaign, IL, 1975, pp. 204-224. Terpstra, A.H.M. and Pels, A.E., Isolation of plasma li-
I3
poproteins by a combination of differential and density gradient ultracentrifugation, Fresenius Z. Anal. Chem., 330 (1988) 149. Weisgraber, K.H., Mahley, R.W. and Assmann, G., The
14
rat arginine-rich apoprotein and its redistribution following injection of iodinated lipoproteins into normal and hypercholesterolemic rats, Atherosclerosis, 28 (1977) 121. Noble, R.P., Electrophoretic separation of plasma lipo-
5. References Goodman, D.S., Cholesterol ester metabolism, Physiol. Rev., 45 (1965) 747. Zilversmit, D.B., Hughes, L.B. and Balmer, J., Stimulation of cholesterol ester exchange by lipoprotein-free rab-
Nichaman, M.Z., Sweeley, C.C. and Olson, R.E., Plasma fatty acids in normolipemic and hyperlipemic subjects during fasting and after linoleate feeding, J. Clin. Nutr.,
I6
proteins in agarose gel, J. Lipid Res., 9 (1968) 693. Glomset, J.A., The plasma lecithin:cholesterol acyltransferase, J. Lipid Res., 9 (1968) 1557. Terpstra, A.H.M., Nicolosi, R.J. and Herbert, P.N., In
I7
vitro incorporation of radiolabeled cholesteryl esters into high and low density lipoproteins, J. Lipid Res., 30 (1989) 1663. Lee, D.M., Malondialdehyde formation in stored plasma,
I5
I8 19
20
21
22
23
Biochem. Biophys. Res. Commun., 95 (1980) 1663. Shipley, R.A. and Clark, R.E., Tracer Methods for In Vivo Kinetics, Academic Press, New York, 1972. Terpstra, A.H.M., Determination of the kinetic parameters of the cholesteryl esters and proteins of plasma lipoproteins, Math. Comput. Modelling, I2 (1989) 601. Allain, C.C., Poon, L.S., Chen, C.S.G., Richmond, W. and Fu, P.C., Enzymatic determination of total serum cholesterol, Clin. Chem., 20 (1974) 470. Burnstein, M., Scholnick, H.R. and Martin, R.J., Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions, J. Lipid Res., 11 (1970) 583. Portman, O.W. and Sugano, M., The hydrolysis and fatty acid exchange of cholesterol esters administered in lipoproteins to cebus monkeys, Arch. Biochem. Biophys., 102 (1963) 335. Goldberg, I.R., Rosenfeld, R.S., Paul, 1. and Leeman, B., Generation of plasma free cholesterol from circulating lipoprotein-associated cholesteryl ester, Am. J. Physiol., 250 (1986) E265.