Separate mechanisms for the uptake of high and low density lipoproteins by mouse adrenal gland in vivo.

The adrenal gland of the mouse exhibits uptake mechanisms for plasma high density lipoprotein (HDL) and low density lipoprotein (LDL). To study this uptake, we lowered the endogenous plasma lipoprotein level in mice by administering 4-aminopyrazolopyrimidine and then injected the animals with tracer amounts of human ““1-HDL or ‘““I-LDL intravenously. Uptake of ‘“‘I-HDL and ““I-LDL in the adrenal gland was demonstrable within 2 min after injection, and the content of radioactivity reached a steady state within 30 min. The adrenal gland accumulated 20-fold more lZ51 radioactivity/mg of tissue than lung or kidney. Moreover, the adrenal took up 50to 200-fold more lZ51-HDL or lZ51LDL than ‘251-albumin. Adrenal uptake of both lipoproteins was reduced when adrenocorticotrophic hormone secretion was suppressed by dexamethasone. Uptake of either LDL or HDL raised the level of cholesteryl esters in the adrenal gland and suppressed the activity of microsomal3-hydroxy-3-methylglutaryl coenzyme A reductase. The uptake mechanisms were saturable in that unlabeled lipoproteins competed with the ‘““I-lipoproteins for uptake. In cross-competition experiments, unlabeled LDL competed more effectively than unlabeled HDL for ‘““I-LDL uptake; conversely, unlabeled HDL competed more effectively than unlabeled LDL for lZ51-HDL uptake. These data suggest that two different lipoprotein uptake systems supply cholesterol to the adrenal gland of the mouse, one using LDL and another using HDL.

The adrenal gland of the mouse exhibits uptake mechanisms for plasma high density lipoprotein (HDL) and low density lipoprotein (LDL). To study this uptake, we lowered the endogenous plasma lipoprotein level in mice by administering 4-aminopyrazolopyrimidine and then injected the animals with tracer amounts of human ""1-HDL or '""I-LDL intravenously. Uptake of '"'I-HDL and ""I-LDL in the adrenal gland was demonstrable within 2 min after injection, and the content of radioactivity reached a steady state within 30 min. The adrenal gland accumulated 20-fold more lZ51 radioactivity/mg of tissue than lung or kidney. Moreover, the adrenal took up 50-to 200-fold more lZ51-HDL or lZ51-LDL than '251-albumin.
Adrenal uptake of both lipoproteins was reduced when adrenocorticotrophic hormone secretion was suppressed by dexamethasone. Uptake of either LDL or HDL raised the level of cholesteryl esters in the adrenal gland and suppressed the activity of microsomal3-hydroxy-3-methylglutaryl coenzyme A reductase.
The uptake mechanisms were saturable in that unlabeled lipoproteins competed with the '""I-lipoproteins for uptake. In cross-competition experiments, unlabeled LDL competed more effectively than unlabeled HDL for '""I-LDL uptake; conversely, unlabeled HDL competed more effectively than unlabeled LDL for lZ51-HDL uptake. These data suggest that two different lipoprotein uptake systems supply cholesterol to the adrenal gland of the mouse, one using LDL and another using HDL.
Recent studies have focused attention on the mechanism by which the adrenal gland obtains cholesterol from plasma lipoproteins for use in steroid hormone synthesis (reviewed in Ref. 1). Cultured steroid-secreting mouse adrenal tumor cells (Y-l clone) (2) and cultured adult bovine adrenal cells (3) possess surface receptors for plasma low density lipoprotein. Binding of LDL' to these receptors leads to the uptake of the * This research was supported by United States Public Health Service Grant PO-1.HL-20948 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. intact lipoprotein by adsorptive endocytosis and hydrolysis of its protein and cholesteryl ester components in cellular lysosomes (l-4). The unesterified cholesterol released from the hydrolysis of LDL supplies the substrate for steroid synthesis and suppresses de nouo cholesterol synthesis by suppressing the activity of the rate-controlling enzyme 3-hydroxy-3-methylglutaryl coenzyme A reductase (2, 3). Excess cholesterol entering the adrenal cells through the receptor-mediated uptake of LDL is re-esterified and stored as cholesteryl esters (2, 3). LDL receptor activity in adrenal cells is enhanced by adrenocorticotrophic hormone and is suppressed when the cells have obtained sufficient cholesterol from the uptake of LDL (l-3). The LDL receptor in the cultured mouse and bovine adrenal cells does not bind high density lipoprotein, the other major cholesterol-carrying lipoprotein of plasma, and hence HDL is unable to supply cholesterol to these cells (2, 3).
Study of lipoprotein-mediated regulation of cholesterol metabolism in the adrenal gland of intact animals has recently been made possible by the use of 4-aminopyrazolopyrimidine, a purine analog that blocks lipoprotein secretion from the liver (5, 6). Within 24 h after the administration of 4-APP to rats, the plasma cholesterol level falls by more than 90%. Deprived of exogenous cholesterol, the adrenal gland rapidly becomes depleted of cholesteryl esters and then develops a marked elevation in cholesterol synthesis that is mediated by an increase in HMG-CoA reductase activity (7-9). The subsequent intravenous infusion of human LDL restores the content of adrenal cholesteryl esters and suppresses the elevated level of HMG-CoA reductase activity and cholesterol synthesis (7-9). In contrast to the results with the cultured mouse and bovine adrenal cells, the uptake process for cholesterol in the adrenal gland of the 4-APP-treated rat was not restricted to LDL. The administration of human or rat HDL also restored adrenal cholesteryl esters and suppressed HMG-CoA reductase activity and cholesterol synthesis (7-9). Moreover, HDL appeared more potent than LDL in that lower plasma cholesterol levels were required to produce a given increment in adrenal sterol content when HDL was infused as compared with LDL (7-9).
The above results raised the question as to whether the adrenal gland of the 4-APP-treated rat was taking up cholesterol by binding lipoproteins at receptor sites like those of the cultured adrenal cells. If so, was the same binding system recognizing both LDL and HDL, or were two different binding mechanisms involved? In the current studies, we have begun to answer these questions by measuring the uptake of '?labeled lipoproteins in the adrenal gland of 4-APP-treated mice. The choice of mice was dictated by two considerations: 1) the small size of the mouse permits the use of smaller amounts of '""I-labeled lipoproteins than are required for intravenous infusion in rats, and 2) cultured mouse adrenal cells (Y-l tumor clone) are known to possess specific LDL receptors. The current results demonstrate that the adrenal gland of the 4-APP-treated mouse exhibits saturable uptake processes for both '""I-LDL and '"'I-HDL and that these two uptake processes appear to be distinct from one another. and stripped of adherent fatty and fibrous tissue. The two adrenal glands from each mouse were then placed in a tube and their content of IL'1 radioactivity was determined in a well-type scintillation counter. After counting, the glands were weighed and then homogenized in 0.5 ml of ice cold water with a Dounce homogenizer.
The whole homogenates were incubated for 1 h with 10% trichloroacetic acid, after which the resulting precipitate was counted for determination of its "'I-radioactivity. In the experiments described in Fig. 4 and Table I, the body organs were subjected to electrophoresis in SDS-polyacrylamide gels using the system described by Laemmli (13) with minor modifications. A 15% acrylamide slab gel (15 x 15 x 0.15 cm) was overlaid with a 5% acrylamide stacking gel (2 x 15 x 0.15 cm) that contained 30% glycerol. All buffers contained 2 mM sodium EDTA. Samples for electrophoresis were prepared as follows. Lipid was extracted with chloroform/methanol by the method of Bligh and Dyer (14). The upper aqueous phase, which contained the precipitated protein and greater than 98% of the total IL31 radioactivity for all samples, was blown to dryness under a stream of air at room temperature. Each residue was then resuspended in 200 ~1 of sample buffer containing 50 mM Tris-chloride (pH 6.8), lo%> glycerol, 2 IIIM sodium EDTA, 1% 2.mercaptoethanol, 1% SDS, and 0.008% bromphenol blue, after which the solution was heated for 1 h at 37", followed by 3 min at 100°C. The gels were subjected to electrophoresis at a constant current of 40 mA per slab for 3 h at room temperature.
The electrophoresis buffer contained 0.05 M Tris, 0.384 M glycine, 2 mM sodium EDTA, and 0.1% SDS at pH 8.3. The gels were fixed and stained with Coomassie Blue G (15), dried onto paper, and exposed for 24 h to Kodak X-Omat R fiim. Following autoradiography, the labeled bands were cut out and counted for radioactivity in a well-type scintillation counter. Gels were calibrated with the following molecular weight (M,)  to the range of 5 mg/dl, a 95% fall in the adrenal content of esterified cholesterol to about 5 pg of sterol/mg of protein, no change in the adrenal content of free cholesterol, and a 20-fold increase in the activity of adrenal HMG-CoA reductase to the range of 3.5 nmol. mini' . mg of protein'.
All of these changes were virtually identical to those previously observed in the adrenal gland of 4-APP-treated rats (7). The addition of ACTH to the 4-APP treatment regimen did not significantly affect the magnitude of the changes observed, but did lessen variability between animals. Thus, in all experiments a standard protocol was used in which the mice were treated with 4-APP plus ACTH for 48 h prior to the experiment. Fig. 1 shows an experiment in which mice were treated with 4-APP plus ACTH for 48 h. The mice were then given various amounts of human HDL as a bolus intravenously and were killed 12 h later. The HDL infusion raised the plasma cholesterol level (Fig. IA), suppressed adrenal HMG-CoA reductase activity (Fig. lB), and restored the adrenal content of esterified cholesterol (Fig. 1C). Fig. 2 shows a similar experiment in which mice were infused with various amounts of human LDL instead of HDL. As with the HDL infusion, the infusion of LDL raised the plasma cholesterol level ( Fig. 2A), suppressed adrenal HMG-CoA reductase activity (Fig. 2B), and restored the adrenal content of esterified cholesterol (Fig. 2C). The LDL-mediated cholesterol uptake process in the adrenal gland was less efficient than the HDL-mediated uptake process in that the injection of larger amounts of cholesterol and the attainment of higher plasma cholesterol levels were necessary to achieve a given effect on the adrenal gland when LDL was used as opposed to HDL.
The data in Figs. 1 and 2 raised the possibility that the adrenal gland of the 4-APP-treated mouse possessed mechanisms for taking up the cholesterol of HDL and LDL. To study these uptake processes in more detail, we turned to the use of lipoproteins labeled with "'1 in the protein component. Figure 3 shows an experiment in which mice iYere treated for 48 h with 4-APP plus ACTH and were then given an intravenous bolus of 2.6 x 10" cpm of human ""I-HDL or ""I-LDL. Groups of mice were killed at various times after the injection and the adrenal content of "'1 radioactivity was measured. After the infusion of either lipoprotein, adrenal lz51 radioactivity rose rapidly and reached a steady state by 15 min. During this steady state, the two adrenal glands of a single mouse contained 2.3% of the administered ""I-HDL radioactivity and 1% of the administered "' I-LDL radioactivity.
In both cases, more than 95% of the adrenal '""I-radioactivity was precipitable by 10% trichloroacetic acid. To test the specificity of the adrenal uptake process, we measured the adrenal content of ""I-radioactivity after infusion of human '""I-LDL, ""I-HDL, or ""I-albumin to mice treated with various combinations of 4-APP, ACTH, and dexamethasone acetate (Fig. 4). For comparative purposes, we also measured the content of radioactivity in two other extrahepatic tissues, kidney and lung. In the adrenal gland, treatment with 4-APP enhanced the uptake of ""I-HDL by 20-fold (Fig. 4A). The addition of ACTH gave a small additional stimulation.
In the presence of exogenous ACTH, suppression of endogenous ACTH secretion with dexamethasone had little effect on ""I-HDL uptake. However, in the absence of exogenous ACTH, the administration of dexamethasone decreased the uptake of ""I-HDL by 80% (Fig. 4A). In other experiments not shown, we observed that ACTH itself in the dllo adrenal gland show a Sold higher amount af "'I-HDL ur '""I-_r .8-3w LDL uptake as compared with '""I-albumin. No such differ-;:s j3~zho kdl ence existed for kidney or Iung, in which the contents of the : IbcI "T-lipoproteins and '""I-aIbumin were comparable. 1) The data of Fig. 4 and Table I demonstrate that the adrena * 9:; J gland of the 4-APP-treated mouse has selective uptake syskg" terns for "'I-LDL and "'I-HDL.
To determine whether these +x6 * i$i ( uptake systems were saturable and to determine whether Y-there was cross-competition between the two lipoproteins, we 0 injected mice with '"'I-LDL or "'I-HDL in the presence of 'r-increasing concentrat.ions of unlabeled LDL or HDL. Fig. 5A $6, FL 60 shows that the uptake of 10 pg of protein of """I-LDL HDL protein, which produced a plasma cholesterol level of 24 mg/dl. In contra& the injection of 5 mg of protein of LDL, m&ml of bovine serum albumin. All of the animals were killed 12 h after the injection, and the six adrenal glands from the three mice in which produced a plasma cholesterol level of 57~1 mg/dl, was each group were pooled. All measurements were made as described required to produce a 60% inhibition of lzSI-HDL uptake. uptake into the adrenal gland after the injection, and the six adrenal glands from the three mice in (Fig. 6B). each grtrup were pooled. All measurements were made as described .g presence ofendogenous ACTH. The uptake of '"'I-LDL in the g 30 adrenal gland followed a pattern similar to that of '""I-HDL, 0 ,"z except that the maximal uptake of '%I-LDL radioactivity was A!,5 about half as great as with '""I-HDL (Fig. 4B).

G 2 20
In untreated animals, the adrenal uptake of IXgI-albumin x"g was significantly lower than that for either '"'I-LDL or "'I-E * HDL (45 2 9 cpm/mg for albumin as compared to 229 + 18 lj 10 and 215 & 35 cpmJmg for LDL and HDL, respectively) and it = failed to rise with 4-APP or ACTH treatment (Fig. 4C). 5 Similarly, in untreated animals the uptake values for ""'T-LDL a 0 and '""I-HDL in the kidney and lung were much lower than 0 IO x1 30 for the adrenal gland (Fig. 4  Each mouse was then injected intravenously with 15 kg of protein (2.4 X IO" cpm) of either "'I-HDL (A), '"'I-LDL (B), or "'I-albumin (C). The animals were killed 60 min after injection, the organs were perfused with a solution of sodium chloride and albumin via the inferior vena cava, and the tissue content of ""I radioactivity was determined as described under "Experimental Procedures." Each point represents the average of values obtained from five mice (A and B) or three mice (C). The bra&& represent 1 S.E. Data on the plasma cholesterol level, the plasma content of ""I radioactivity, and the weight of the various tissues are shown in Table  I Each mouse was then injected intravenously with '""I-HDL (20 pg of protein, 1.6 X IO" cpm) together with one of the following unlabeled lipoproteins: 0, none; 0, HDL, 5 mg of protein; or a, LDL, 5 mg of protein.
The animals were killed at the indicated time after injection, and the content of lriI radioactivity in plasma (A)   Each mouse was then injected intravenously with "'I-HDL (20 pg of protein, 2.6 x IO" cpm). The animals were killed at the indicated time after injection, and the content of "'1 radioactivity in plasma (0) and in the adrenal gland (0) was determined as described under "Experimental Procedures." Each point represents the average of values obtained from three or four mice. The "loo%, value" was 1,720 & 140 cpm/pl for plasma (0) and 6,200 + 1,100 cpm/gland for the adrenal gland (0). The mean plasma cholesterol level in the mice treated with 4-AI'P and ACTH was 10 mg/dl (range, 0 to 24).
curve of "'I radioactivity from the adrenal gland was parallel to the plasma disappearance curve, but was delayed by about 60 min.
The kinetic data of Figs. 6 and 7 are compatible with a mechanism in which the adrenal content of '""I-HDL is in constant and rapid flux, with the ""I-HDL having a residence time in the adrenal gland of less than 60 min. To test this hypothesis in another way, we injected animals with 15 pg protein of '251-HDL, allowed it to be taken up by the adrenal gland for 15 min when a near-steady state had been reached, and then injected a bolus of unlabeled HDL (5 mg of protein) (Fig. 8A). By diluting the specific radioactivity of the circulating '251-HDL, the unlabeled HDL prevented further entry of IL?-HDL into the gland. Within 15 min after the infusion of the unlabeled HDL, the adrenal content of 1251 radioactivity had declined to the same level that it assumed when the unlabeled HDL was injected simultaneously with the labeled HDL at zero time (Fig. 8A). A similar rapid decline in the adrenal content of 1251-LDL occurred when unlabeled LDL was injected 15 min after the injection of ""I-LDL (Fig. 8B). The data of Fig. 5 suggested that HDL did not compete effectively with ""I-LDL for uptake by the adrenal. This lack of cross-competition allowed us to perform an experiment to determine whether prior uptake of unlabeled HDL could metabolically suppress the uptake of '""I-LDL (Fig. 9). In one part of this experiment, mice were injected with 5 mg of protein of unlabeled LDL. After varying intervals, the mice were injected with 10 pg of protein of ""I-LDL and killed 30 min later (Fig. 9A). As expected, the unlabeled LDL competed directly with the lL"I-LDL and hence the adrenal uptake of ""I-LDL was low at each time point. In the second part of this experiment, the mice were injected with 5 mg of protein of unlabeled HDL instead of unlabeled LDL (Fig. 9B). The unlabeled HDL produced only a slight reduction of ""I-LDL uptake when the ""I-LDL was injected simultaneously with the unlabeled HDL at zero time or when the '""I-LDL was injected 1 h after the injection of unlabeled HDL. However, when the ""I-LDL was injected 4 or 8 h after the unlabeled HDL, the uptake of ""I-LDL was reduced by 75%. These data suggest that the uptake of lipoproteins in the adrenal may be regulated by the cholesterol content and that the cholesterol derived from the prior uptake of unlabeled HDL may have suppressed the uptake of ""I-LDL.
Whereas human LDL contains only one major protein and ACTH (2 units) at 12-h intervals. Experiment A, each mouse was then injected intravenously with '""I-HDL (15 pg of protein, 2.6 x 10" cpm) in the absence (0) or presence (0) of unlabeled HDL (5 mg of protein). The mice receiving ""I-HDL alone (0) were divided into two groups, one of which received an intravenous injection of unlabeled HDL (5 mg of protein) 15 min after the initial injection of '?-HDL (A). All animals were killed at the indicated time, and the content of '? radioactivity in the adrenal gland was determined as described under "Experimental Procedures." Experiment B, this experiment was exactly the same as that of Experiment A except that ""I-LDL (15 pg of protein, 2.6 x 10" cpm) and unlabeled LDL (5 mg of protein) were used as indicated.
Each point represents the average of values obtained from three mice. The brackets represent 1 S.E. The mean plasma cholesterol level in the mice treated with 4-APP and ACTH was 3.6 mg/dl (range, 0 to 7.7). component, human HDL contains several proteins, of which the major ones are apoproteins . To determine whether both of these apoproteins were being taken up by the adrenal gland, we administered 38 pg of '""I-HDL in the absence or presence of a 170-fold excess of unlabeled HDL to 4-APP-treated animals and killed the animals after 15 min. The adrenal glands were homogenized and an aliquot of the whole homogenate was reduced with 2-mercaptoethanol and subjected to SDS-polyacrylamide gel electrophoresis. For comparative purposes, a sample of the initial ""I-HDL and a sample of the plasma from each of the animals were also reduced with 2-mercaptoethanol and subjected to electrophoresis. Electrophoresis was also performed on one sample of '""I-HDL that was not reduced with 2-mercaptoethanol. The slab gel was subjected to autoradiography ( Fig. 10) after which the bands corresponding to A-I and A-II were cut out and their content of lZ51 radioactivity was determined in a scintillation counter. In the sample of '""I-HDL that was not reduced with 2-mercaptoethanol, the two major radiolabeled proteins migrated to positions corresponding to molecular weights of 29,000 and 17,000, which are similar to the reported molecular weights of   (Lane 1, Fig. 10). When the ""I-HDL was reduced with 2-mercaptoethanol, the A-II band disappeared and a new band at a molecular weight of 8,500 appeared (Lanes 2 and 3, Fig. 10). A-II is known to be reduced to a molecular weight of 8,500 in the presence of 2mercaptoethanol, whereas A-I is unaffected (18-20). By scintillation counting, the ratio of radioactivity in A-I to A-II was 1.3, and these two bands accounted for 75% of the lZ51 radioactivity on the gel.
The electrophoretic pattern of the lZ51 radioactivity in the plasma of the mice injected with ""I-HDL was qualitatively (Lane 5, Fig. 10) and quantitatively similar to that of the starting '""I-HDL preparation.
There was no change in the pattern of plasma radioactivity when the animal received unlabeled HDL together with the ""1-HDL (Lane 4, Fig. 10). In the animal that received only 12"1-HDL, the two adrenal glands contained a total of 150,000 cpm. The electrophoretic pattern of the adrenal radioactivity (Lane 6, Fig. 10) was similar to that of the infused ""I-HDL.
The A-I to A-II ratio was 1.5 and the sum of the radioactivity in A-I  and ACTH (4 units) at 24-h intervals.
Each mouse was then injected intravenously with ""I-HDL (38 pg, 12 x 10" cpm) either alone or together with 6.4 mg of unlabeled HDL.
The animals were killed 15 min after injection, after which plasma and adrenal glands were removed.
The samples were prepared for electrophoresis as described under "Experimental Procedures." 72% of the total radioactivity on the gel. In the mouse that received 12"1-HDL plus unlabeled HDL, the total content of radioactivity in the adrenal gland was reduced by 90%. Only a faint trace of A-I and A-II could be visualized by electrophoresis (Lane 7, Fig. 10). These data demonstrate that at the 15-min time point, the '""I-radioactivity that entered the adrenal gland as a result of the saturable uptake process was indistinguishable from that of the infused 12"1-HDL.

DISCUSSION
In the current experiments, the adrenal gland of the 4-APPtreated mouse was observed to take up '""I-labeled human LDL and HDL with selectivity, speed, and saturability.
The selectivity was evident in two ways. First, when different organs of the 4-APP-treated mouse were compared, the adrenal gland was found to accumulate 20-fold more iL"I-LDL and HDL/mg of tissue than kidney or lung. Second, when various proteins were compared, the adrenal gland was found to accumulate much more "'I-LDL or lL'I-HDL (0.5 to 2% of injected dose) than ""I-albumin (0.01% of injected dose) (Figs. 3 and 4 and Table I).
The adrenal lipoprotein uptake process was extremely rapid. Selective uptake was demonstrated within 2 min after injection and the content of ""I-HDL or ""I-LDL in the gland reached a maximum within 15 to 30 min. The rate-limiting step in the uptake process was saturable as evidenced by the ability of unlabeled lipoproteins to compete with 'L"I-lipoproteins for uptake. This saturable step appeared to be different for LDL and HDL. Thus, the uptake of ""I-HDL was inhibited more effectively by unlabeled HDL than LDL. Conversely, '251-LDL uptake was more effectively competed by unlabeled LDL than by HDL. Based on analogy with lipoprotein metabolism in cultured fibroblasts and adrenal cells (l-4), it seems likely that the saturable rate-limiting step involves binding of the '251-lipoproteins to cell surface receptors. The cross-competition experiments raise the possibility that two different receptors are involved, one for LDL and one for HDL.
The uptake of ""I-LDL and of ""I-HDL appeared to be under metabolic regulation in the adrenal. Thus, uptake of both lipoproteins was suppressed when ACTH secretion was suppressed with dexamethasone, and this suppression of uptake was prevented by exogenous ACTH. A similar enhancement of '""I-LDL uptake by ACTH occurs in cultured mouse and bovine adrenal cells (l-3).
The uptake systems for both "'1-LDL and "'1-HDL appeared to be functional in mouse adrenal even in the absence of 4-APP treatment as indicated by the 5-fold higher uptake of the two lipoproteins as compared with ""I-albumin in untreated animals (Table I). The adrenal uptake of ""I-lipoproteins increased markedly when the animals were treated with 4-APP. This enhancement was due in part to the reduction of endogenous unlabeled lipoproteins so that the labeled lZ"I-LDL and ""I-HDL were no longer diluted with unlabeled material. In addition, it is possible that the number of adrenal lipoprotein binding sites also increased in the 4-APP-treated mice in a manner analogous to the increase in LDL receptors that occurs in cultured fibroblasts and adrenal cells that are deprived of exogenous lipoproteins (l-4). The current types of studies cannot evaluate this latter possibility.
Whether the in viva adrenal uptake processes involve internalization of the intact lipoprotein or whether they represent only binding to surface structures cannot yet be determined. The turnover experiments of Figs. 6 and 7 suggest that, at least for HDL, the bound '"'I-lipoprotein is in rapid flux. Whether the '""I-HDL leaves the gland as intact protein or whether it is degraded by the cells in a manner similar to that demonstrated for LDL in cultured cells is not yet known. The pulse-chase experiments of Fig. 9 in which the injection of tracer amounts of "'I-lipoproteins was followed by a large dose of unlabeled lipoproteins suggested that ""I-LDL and ""I-HDL could leave the adrenal gland within 15 min. This seems too rapid to be accounted for by degradation and suggests that a large portion of the ""I-lipoproteins is initially bound to a surface from which it can be displaced by the unlabeled lipoproteins.
Is the LDL uptake process in the adrenal of the 4-APPtreated mouse mediated by the same LDL receptor that is demonstable in cultured mouse adrenal cells? Several lines of evidence address this point. In experiments to be published elsewhere, we have been able to demonstrate high affinity binding of ""I-LDL to membranes prepared from homoge-nates of normal mouse adrenal glands.' This binding site is similar to the one demonstrated in homogenates of fresh bovine adrenal glands (21) and is similar to the functional LDL receptor in cultured mouse and bovine adrenal cells (2, 3). HDL does not compete for ""I-LDL binding to this membrane binding site in vitro, ' an observation that correlates with its reduced ability to compete with ""I-LDL uptake in vivo.
Although indirect, these data are suggestive that LDL uptake in the adrenal in vivo is mediated by the same LDL receptor that is demonstrable in vitro. How then can the in viva I"'I-HDL uptake process be explained? As discussed above, the in vivo cross-competition experiments suggest that HDL uptake is mediated by a different receptor than the one for LDL. This putative HDL receptor has not yet been detected in any in vitro system. For example, in experiments not shown, we have found that the same ""I-HDL preparation that is taken up with great avidity by the adrenal gland of the 4-APP-treated mouse is not taken up by cultured mouse Y-l adrenal cells, cultured bovine adrenal cells, or cultured human fibroblasts.
Moreover, the same HDL preparation that raises the cholesterol content of the mouse adrenal in vivo and suppresses its HMG CoA reductase activity has no such effect on any of the cultured cell systems. Finally, although high affinity ""I-LDL binding can be demonstrated with isolated adrenal membranes, in vitro binding of ""I-HDL cannot yet be demonstrated.' Several possible mechanisms might explain the differences between the metabolism of HDL in the adrenal in vivo and in vitro.
First, the adrenal gland in vivo may express an HDL receptor that is not expressed by isolated adrenal cells in tissue culture. This HDL receptor in vivo need not be present on adrenal cells per se. It might be present on neighboring cells, such as endothelial or sinusoidal lining cells. Such a receptor could supply cholesterol to adrenal cells by a mechanism analogous to the delivery of fatty acids to adipose cells by lipoprotein lipase bound to the capillary endothelium (22). Finally, the metabolism of HDL in vivo appears to be complex in that its apoproteins can exchange with those of other lipoproteins (23). Thus, even though uptake of ""I-HDL can be demonstrated as early as 2 min after injection into 4-APPtreated mice, it remains possible that immediately upon infusion the injected lipoprotein is modified into a form that can bind to an adrenal receptor. Such binding would not be detected in vitro since the HDL would have no opportunity to be modified.
As discussed above, even if it recognized modified HDL, the putative HDL receptor is not likely to be the LDL receptor, since LDL is not an effective competitor for '""I-HDL uptake.
' 11. Chappell, P. T. Kovanen, J. L. Goldstein, and M. S. Brown, manuscript in preparation. In summary, while the current in vivo data are not yet definitive as to mechanisms, they suggest that the adrenal gland of the 4-APP-treated mouse expresses two types of lipoprotein receptors. One of these is analogous to the LDL receptor previously demonstrated in cultured adrenal cells and other in vitro systems (l-4)' and the other is an HDL receptor whose counterpart has not yet been demonstrated in vitro.