Role of Lysosomal Acid Lipase in the Metabolism of Plasma Low Density Lipoprotein OBSERVATIONS IN CULTURED FIBROBLASTS FROM A PATIENT WITH CHOLESTERYL ESTER STORAGE DISEASE*

The hydrolysis of cholesteryl esters contained in plasma low density lipoprotein was reduced in cultured fibroblasts derived from a patient with cholesteryl ester storage disease, an inborn error of metabolism in which lysosomal acid lipase activity is deficient. While these mutant cells showed a normal ability to bind low density lipoprotein at its high affinity cell surface receptor site, to take up the bound lipoprotein through endocytosis, and to hydrolyze the protein component of the lipoprotein in lysosomes, their defective lysosomal hydrolysis of the cholesteryl ester component of the lipoprotein led to the accumulation within the cell of unhydrolyzed cholesteryl esters, the fatty acid distribution of which resembled that of plasma lipoprotein. When the cholesteryl ester storage disease cells were incubated with low density lipoprotein, the reduced r&e of liberation of free cholesterol by these mutant cells was associated with a delay in the occurrence of two lipoprotein-mediated regulatory events, suppression of 3-hydroxy-3-methylglutaryl coenzyme A reductase activity, and activation of

The hydrolysis of cholesteryl esters contained in plasma low density lipoprotein was reduced in cultured fibroblasts derived from a patient with cholesteryl ester storage disease, an inborn error of metabolism in which lysosomal acid lipase activity is deficient. While these mutant cells showed a normal ability to bind low density lipoprotein at its high affinity cell surface receptor site, to take up the bound lipoprotein through endocytosis, and to hydrolyze the protein component of the lipoprotein in lysosomes, their defective lysosomal hydrolysis of the cholesteryl ester component of the lipoprotein led to the accumulation within the cell of unhydrolyzed cholesteryl esters, the fatty acid distribution of which resembled that of plasma lipoprotein.
When the cholesteryl ester storage disease cells were incubated with low density lipoprotein, the reduced r&e of liberation of free cholesterol by these mutant cells was associated with a delay in the occurrence of two lipoprotein-mediated regulatory events, suppression of 3-hydroxy-3-methylglutaryl coenzyme A reductase activity, and activation of endogenous cholesteryl ester formation.
In contrast to their defective hydrolysis of exogenously derived lipoprotein-bound cholesteryl esters, the cholesteryl ester storage disease cells showed a normal rate of hydrolysis of cholesteryl esters that had been synthesized within the cell. These data lend support to the concept that in cultured human fibroblasts cholesteryl esters entering the cell bound to low density lipoprotein are hydrolyzed within the lysosome and that one of the functions of this intracellular organelle is to supply the cell with free cholesterol.
Cultured human fibroblasts possess a specific mechanism for the net uptake of cholesterol derived from plasma low density lipoprotein (LDL),' the major cholesterol-carrying protein in human blood (l-6). Previous data suggest that this mechanism involves the following sequences of events: (a) LDL binds to a specific receptor on the cell surface (1, 2); (b) the surfacebound LDL becomes incorporated into endocytotic vesicles; (c) the internal endocytotic vesicles containing bound LDL fuse with lysosomes (4, 5); (d) the cholesteryl ester and protein components of LDL are hydrolyzed by lysosomal enzymes to products, including free cholesterol and amino acids (2, 4, 5); and (e) the liberated free cholesterol is transferred from lysosomes to cellular membranes (6). The resultant accumulation of cholesterol within the cell regulates two events in cellular cholesterol metabolism: (a) cholesteryl ester formation is stimulated through an activation of a membrane-bound fatty acyl-CoA:cholesterol acyltransferase (7, 8) and (b) cholesterol synthesis is reduced through a suppression of the activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase) (9). Formulation of this model of LDL metabolism in human fibroblasts has been facilitated by a comparison of the behavior of normal fibroblasts with that of mutant cells from subjects with the homozygous form of familial hypercholesterolemia, which lack the cell surface LDL receptor and thus fail to manifest all of the metabolic consequences of high affinity LDL binding (l-9).
A critical aspect of the above model involves the role of the lysosome in hydrolyzing the cholesteryl ester and protein components of LDL so as to make free cholesterol available to the cell. This requirement for lysosomes has been inferred from studies in which lysosomal hydrolytic activity has been blocked in intact cells through the use of inhibitors such as chloroquine (4, 5). Monolayers of fibroblasts treated with chloroquine are unable to hydrolyze the LDL taken up from the medium and as a consequence their cellular cholesteryl ester formation is not activated and HMG-CoA reductase activity is not suppressed (4).
To investigate further the role of the lysosome in cellular LDL metabolism, the present studies were conducted utilizing mutant fibroblasts that are deficient in lysosomal acid lipase activity (10-12). Extracts of these mutant cells, which were derived from a subject with cholesteryl ester storage disease (12), manifest a 95% reduction in their ability to hydrolyze the cholesteryl esters contained in LDL. Comparison of LDL metabolism in monolayers of the mutant cells and normal cells has confirmed that: (a) lysosomal acid lipase plays a major role in hydrolyzing cholesteryl esters that enter the cell bound to LDL and (b) the reciprocal regulation of cellular cholesteryl ester formation and cholesterol synthesis initiated by LDL binding to its receptor is dependent upon the subsequent liberation of free cholesterol from LDL by the action of this lysosomal enzyme. Tissue culture supplies, thin layer chromatographic materials, and reagents for assays were obtained from sources as previously reported (2, 6, 9).

Cells
The normal human fibroblast strain used in this study was the same strain (D. S.) that has been used in our previous studies (l-9, 13). The regulation of lipoprotein and cholesterol metabolism in these cells is typical of that in 20 other normal fibroblast strains that have been studied in our laboratory and that were derived from skin biopsies of healthy adults and children. The fibroblasts from the subject with cholesteryl ester storage disease were derived from a 12-year-old girl (J. R.); these cells have been previously shown to be severely deficient in lysosomal acid lipase activity (12). All cells were grown in monolayer and were used between the 5th and 20th passage.

Lipoproteins
Human LDL (density 1.019 to 1.063 g/ml) and human LPDS (density >1.215 g/ml) were obtained from the plasma of healthy subjects and prepared by differential ultracentrifugation (9). Lipoprotein-deficient fetal calf serum was prepared as previously described (14). The concentration of LDL is expressed in terms of its protein content. The mass ratio of total cholesterol to protein content in LDL was 1.6:1, and 70% of the total cholesterol was in an esterified form. '*?-labeled LDL (specific activity, 400 to 600 cpm/ng of protein) was prepared as previously described (1,2). For experiments, the '*'I-LDL was diluted with native LDL to give the final specific activity indicated in the legends. (900 x g, 3 min, 4"), the pooled cell pellet was resuspended in 5 ml of the same buffer, washed once more in the same manner, and the final pellet suspended in 1.5 ml of water. The extract was sonicated using a Branson Sonifier with a microprobe (4, 5) and then centrifuged (2500 x g, 20 min, 4"); the resulting supernatant was used for enzyme assays.

RESULTS
Cell-free extracts of fibroblasts from a patient with cholesteryl ester storage disease were severely deficient in their ability to hydrolyze [BHlcholesteryl linoleate bound to LDL (Fig. 1). At pH 4, the degree of reduction in hydrolytic activity toward this lipoprotein-bound substrate was similar to the 95% reduction in hydrolytic activity previously observed in this cell strain as rapidly as did the normal cell extracts at acid pH (Fig. 2B).
To determine the effect of the lysosomal acid lipase deficiency on LDL metabolism in intact cells, monolayers of fibroblasts were incubated with LDL labeled either in its protein moiety with '*'I or in its cholesteryl ester portion with [r'H]cholesteryl linoleate (Fig. 3). These experiments were initiated after cells had been incubated for 48 hours in the absence of LDL so as to give a lower basal cellular cholesteryl ester content (see below). Parallel monolayers were incubated either in the absence or presence of chloroquine, an inhibitor of lysosomal degradative processes (19)(20)(21) that has previously been shown to block the hydrolysis of both the protein and cholesteryl ester components of LDL in fibroblasts (4,5). In the absence of chloroquine, the metabolism of the protein component of LDL was similar in the normal and mutant cells ( Fig.   3A and C). In both cell strains, the cellular content of '*?-LDL reached a plateau by 2 hours and this level remained constant throughout the subsequent 8 hours (Fig. 3A). As previously reported (2, 4), the steady state content of "'I-LDL reflects a dynamic balance between equal rates of cellular uptake and degradation of the lipoprotein.
The proteolytic degradation of LDL during the incubation period was evidenced by the constant rate of release of [?]iodotyrosine to the culture medium (Fig. 3C). Chloroquine severely inhibited the proteolytic degradation of '*'I-LDL in both cell strains (Fig. 30). While this agent had no effect on the initial rate of "'I-LDL uptake as measured at 2 hours, the continuing uptake in the absence of degradation led to a progressive increase in the cellular content of '*?-LDL over the subsequent 8-hour interval (Fig. 3B).
Whereas in the normal cells the metabolism of the choles- began to be formed but at a rate that was only about one-third that of the normal cells (Fig. 3G). As in the normal cells, the hydrolysis of [ BH]cholesteryl linoleate in the mutant cells was inhibited by chloroquine (Fig. 3H), suggesting that the low level of hydrolytic activity in these intact mutant cells was due to the residual acid lipase activity demonstrable in UitFO (Figs.  1 and 2). Overall, the metabolism of the LDL-bound cholesteryl linoleate in the untreated mutant cells bore a partial resemblance to that observed in the normal cells treated with chloroquine.
However, unlike the situation in chloroquinetreated normal cells in which acid lipase activity was completely inhibited, the residual acid lipase activity in untreated mutant cells was sufficient to allow an appreciable rate of The cells were grown under standard conditions except that the growth medium was switched to 5% human LPDS on Day 5. After 48 hours in LPDS (Day 7), one group of cell monolayers received 2 ml of growth medium containing 5% human LPDS, 10 pg of protein/ml of lZ"I-LDL (116 cpm/ng of protein), and either no chloroquine (.,A) or 50 pM chloroquine (0,A). After incubation at 37" for the indicated time, the amount of cellular binding (A and B) and proteolytic degradation (C and D) of the ""I-LDL  (Fig. 3E).
The degree of reduction in cholesteryl ester hydrolytic activity in the mutant cells was sufficiently pronounced to permit an analysis of the relationship between hydrolysis of the cholesteryl esters in LDL and the subsequent LDL-mediated regulatory events in cellular cholesterol metabolism.   was achieved at approximately 10 &ml, a value similar to that previously reported for half-maximal binding of LDL to its receptor (1,2). Thus, the saturating concentration for hydrolysis of ['H]CL-LDL was much lower in intact cells than in cell-free extracts (Fig. 2), suggesting that in the intact cell LDL binding to its receptor was rate-limiting for cholesteryl ester hydrolysis. At the 4-hour time point studied and at saturating LDL levels, a 7-fold difference in hydrolysis of [SH]CL-LDL was noted between the normal and mutant cells. At a similar early time point, the maximal rate of endogenous cholesteryl ester formation in the mutant cells was also reduced approxi- A 6 HOURS mately 7-fold at saturating LDL levels (Fig. 6), again suggesting that the rate of endogenous cholesteryl ester formation was dictated by the rate of hydrolysis of exogenous cholesteryl esters contained in LDL. The LDL-mediated suppression of HMG-CoA reductase activity could also be correlated with the hydrolysis of the cholesteryl esters in LDL (Fig. 7). Thus, at 6 hours, a time when hydrolysis of [BH]CL-LDL in the mutant cells was markedly reduced (Fig. 4B), there was relatively little suppression of HMG-CoA reductase activity in these mutant cells (Fig.  7A). However, at 24 hours after considerable free [JH]cholesterol had been formed from LDL (Fig. 4B), HMG-CoA reductase activity in the mutant cells was almost completely suppressed (Fig. 7B). In contrast to their retarded response to LDL, the mutant cells promptly suppressed HMG-CoA reductase activity and developed enhanced cholesteryl esterification capacity in a manner identical with normal cells when treated with 25-hydroxycholesterol (data not shown). We have previously reported that incubation of normal fibroblasts with LDL resulted in a marked increase in the cellular content of cholesteryl esters (6). The current data suggest that this net accumulation is due to a hydrolysis of the incoming LDL-bound cholesteryl esters by a lysosomal acid lipase, followed by a re-esterification of the liberated free cholesterol within the cell. To further assess the role of the lysosomal acid lipase in this process, the relative fatty acid composition of the cholesteryl esters that accumulated in fibroblasts incubated with LDL was compared with that of LDL itself. Table I shows that the predominant fatty acid in the cholesteryl esters of LDL was linoleate (l&2), which comprised 48% of the total. This predominance of linoleate in the cholesteryl esters of plasma lipoproteins is well known and has been attributed to the formation of these esters by the plasma enzyme, 1ecithin:cholesterol acyltransferase, which preferentially transfers polyunsaturated fatty acids from position 2 of lecithin to cholesterol (22). In contrast to its Cells were seeded (Day 0) at a concentration of 3 x 10' cells/lOO-mm Petri dish in 7 ml of growth medium containing 10% fetal calf serum. On Day 3, the medium was replaced with 7 ml of fresh growth medium containing 10% fetal calf serum. On Day 6 each cell monolayer was washed with 7 ml of phosphate-buffered saline after which were added 7 ml of growth medium containing 5% human LPDS. After 24 hours in LPDS (Day 7), the medium was replaced with 5 ml of fresh growth medium containing 5% human LPDS and the indicated concentration of LDL. After incubation at 37" for 30 hours, cell monolayers were washed, harvested, pooled, and the cellular content of cholesterol and cholesteryl esters was determined as described under "Experimental Procedure." Each data point represents the value obtained from three dishes. abundance in LDL, cholesteryl linoleate comprised only 12% of cells. On the other hand, when cells were incubated in the the cholesteryl esters that accumulated in normal cells after absence of all exogenous cholesteryl esters and cellular choincubation with LDL. The distribution of the fatty acids in lesteryl ester formation was stimulated by the presence of cellular cholesteryl esters resembled the distribution in cellular 25-hydroxycholesterol (8), the endogenously synthesized chotriglycerides, suggesting that both classes of esters were syn-lesteryl esters in both the normal and mutant cells had a high thesized within the cell and that their fatty acids had origi-ratio of mono-unsaturated to di-unsaturated fatty acids, a nated from a common precursor pool. These data lend further finding that further indicates that the mutant cells have no support to the concept that when LDL is taken up by cells its primary defect in their ability to synthesize cholesteryl esters. cholesteryl esters are hydrolyzed and the liberated free choles-Two types of studies were conducted to determine whether, terol is re-esterified within the cell.
in addition to hydrolyzing exogenously derived cholesteryl If, as suggested by the previous data, this hydrolysis occurs esters, the lysosomal acid lipase also participated in the in lysosomes, then the inhibition of lysosomal hydrolysis with hydrolysis of endogenously synthesized cholesteryl esters. In chloroquine should result in the cellular accumulation of the first set of experiments, normal and mutant cells were cholesteryl esters, the fatty acid composition of which resemincubated with LDL in the presence of ["C]oleate so as to label bles that of the incoming plasma LDL rather than that of the fatty acid moiety of endogenously synthesized cholesteryl endogenous synthesis. As shown in Table II, chloroquine had esters (Fig. 9A). The cells were then switched to medium that no effect on the total amount of cholesteryl esters that contained no LDL and were further incubated either in the accumulated when cells were incubated with LDL. However, presence or absence of chloroquine. As expected from the data this agent prevented the shift in the relative distribution of in Fig. 6, the content of cholesteryl ["Cloleate at the beginning mono-unsaturated and di-unsaturated fatty acids contained in of the second incubation (zero time in Fig. 9A) was approxicholesteryl esters so that the resultant cellular pattern resem-mately 6-fold higher in the normal cells than in the mutant bled that of plasma LDL.
cells. However, after the removal of LDL the relative rate of Since the cells from the patient with cholesteryl ester storage hydrolysis was similar in both cell strains, one-half of the disease bore a partial resemblance to chloroquine-treated  Table III, after incubation with LDL the after formation of endogenous cholesteryl ["Cloleate was ratio of mono-unsaturated to di-unsaturated fatty acids in stimulated by 25hydroxycholesterol (8) (Fig. 9B). Under these cholesteryl esters in the mutant cells resembled more closely conditions, the cellular content of cholesteryl ["Cloleate was that of plasma LDL than did the cellular esters of the normal similar in both the normal and mutant cells, and again the cholesteryl esters were hydrolyzed at the same rate in both cell  After a further incubation at 37" for 6 hours, each monolayer was washed twice as described in Experiment A, after which were added 2 ml of fresh growth medium containing 10% lipoprotein-deficient fetal calf serum. At the indicated time, duplicate dishes of cells were harvested and the cellular content of cholesteryl ["C]oleate was determined by thin layer chromatography. At zero time, the cellular content of total cholesteryl esters (determined by gas-liquid chroma-may not be critical to the hydrolysis of cholesteryl esters synthesized within the cell.

DISCUSSION
The experiments described in this paper support the hypothesis that in human fibroblasts the lysosome is the cellular organelle that hydrolyzes the cholesteryl esters of exogenous LDL and thus provides the cell with free cholesterol, one of its most important constituents. Mutant fibroblasts derived from a patient with the autosomal recessive disorder cholesteryl ester storage disease, which have been reported previously to be severely deficient in lysosomal acid lipase activity (lo-12), were shown in the current studies to manifest a reduced ability to hydrolyze the cholesteryl esters contained in LDL. The metabolic defect in these mutant cells was characterized in several different ways. First, using [W]CL-LDL as a substrate, it was demonstrated that hydrolysis of the ['Hlcholesteryl linoleate by intact monolayers was slower in the mutant cells than in normal cells and, moreover, that this hydrolysis was not maximal in the mutant cells until they had accumulated a 3-fold higher intracellular level of the radioactive substrate as compared with normal cells. Second, when incubated with native LDL the mutant cells in the steady state contained a 3-fold larger amount of unhydrolyzed cholesteryl linoleate than did normal cells (cf. the di-unsaturated fatty acyl cholesteryl ester content data in Table III). Third, when LDL was added to cells previously incubated in the absence of lipoproteins, the metabolic consequences that depend upon the liberation of free cholesterol from LDL (i.e. stimulation of endogenous cholesteryl ester formation and suppression of HMG-CoA reductase activity) were delayed in the mutant cells to an extent that was proportional to their slower rate of cholesteryl ester hydrolysis. An interesting finding emerged when the relative rates of hydrolysis of [W]CL-LDL were compared in normal and mutant cells using cell-free extracts incubated at acid pH and intact cell monolayers. As previously reported (12) mutant extracts was less than one-twentieth that of the normal extracts, the intact mutant cells showed rates of cholesteryl ester hydrolysis that were nearly one-third that of the normal cells. Similar disparities between enzyme activities measured in uitro and in intact cells have been noted in other human genetic disorders involving incomplete enzyme deficiencies (23). In the case of the mutant cells, part of this disparity might be explained by the 3-fold higher substrate concentration (i.e. [Wlcholesteryl linoleate) that was present within the intact cells during the time of maximal hydrolysis. Whatever the mechanism for the observed hydrolysis of [*H]CL-LDL in the intact mutant cells, this residual enzyme activity,like the normal, is believed to have occurred within lysosomes since it could be abolished by treatment of the cells with chloroquine.
While the lysosomal acid lipase appears to be required for the hydrolysis of exogenous cholesteryl esters bound to LDL, the data suggest that endogenously formed cholesteryl esters are hydrolyzed by a different mechanism. This conclusion is based on the observations that: (a) the rate of hydrolysis of endogenously synthesized cholesteryl ["Cloleate was the same in the normal and mutant cells and (b) that this hydrolysis was not inhibited by chloroquine. Although the experiments in Fig.  1 using an exogenous substrate (i.e. [%I]CL-LDL) did not demonstrate any hydrolysis at neutral pH, it is possible that such enzyme activity may be demonstrable in uitro by the use of a more physiologic, endogenously synthesized cholesteryl ester substrate.