Type C Niemann-Pick Disease - The Journal of Biological Chemistry

Vol. 263,No. 7. Issue of Marcb 5, pp. 3411-3417,1988 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Type C Niemann-Pick Disease LYSOSOMAL ACCUMULATION AND DEFECTIVE INTRACELLULAR MOBILIZATION OF LOW DENSITY LIPOPROTEINCHOLESTEROL* (Received for publication, September 4, 1987)

Jacob SokolS, E. Joan Blanchette-Mackies, Howard S . Kruthl, Nancy K. DwyerQ, LynnM. AmendeQ, Jean D. Butler11, Enid Robinson*, Shutish Patel**, Roscoe 0. Brady$, MarcellaE. ComlyS, Marie T.Vanier$*, and PeterG . Pentchev*§Q From the $Developmental and Metabolic Neurology Branch, National Institute of Neurological and Communicative Disorders and Stroke, the §Endocrinology Section, Laboratory of CeUular and Developmental Biology, National Institute of Diabetes, Digestive and Kidney Diseases, the llLaboratory of Experimental Atherosclerosis, National Heart, Lung and Blood Institute, the I(Human Genetics Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Marylnnd 20892, the $$Laboratoire de Bwchimie, Institut National de la Sante et de la Recherche Medicale U 189, Faculte de Medecine Lyon-Sud, BP 12, F-69921 Oullins Cedex, Frame, and the **NeurologicalService, Veterans Administration Medical Center, Newington, Connecticut 061 11

The intracellular accumulation of unesterified cho- Type C Niemann-Pick disease (2). Although subsequent studlesterol was examined during24 h of low density lip- ies showed both the type A and B disorders to be primary oprotein (LDL) uptake in normal and Niemann-Pick C sphingomyelinase mutations (3-5), no consistent evidence of fibroblasts byfluorescencemicroscopywithfilipin a similar primary lesion in sphingomyelin catabolism has been staining and immunocytochemistry. Perinuclearfluo- reported for Type C Niemann-Pick disease (6). To the conrescence derived from filipin-sterol complexes was ob- trary, recent investigations have suggested that this disorder served in both normal and mutant cells by 2 h. This may, in fact, represent a primary lesion that disturbs critical perinuclear cholesterol staining reached its peak in balances in cholesterol metabolism (7-11). normal cells at 6 h. Subsequent development of fluoCellular cholesterol homeostasis involves a series of interescence during the remaining18 h of LDL incubation grated responses that enable cells to maintain cholesterol was primarily limited to the plasma membrane region levels within a critical range needed for optimal growth and of normal cells. In contrast, mutant cells developed a much more intense perinuclear fluorescence through- development under environmental conditions that include out the entire24 h of LDL uptake with little enhance- both cholesterol excess and deprivation (12). Receptor-mediated uptake and hydrolytic lysosomal processing of LDL’ ment of cholesterol fluorescence staining in the plasma membranes. Direct mass measurements confirmed that in cultured fibroblasts derived from Niemann-Pick C patients internalized LDL cholesterol more readily replenishes are associated with cellular homeostatic responses that are the plasma membrane cholesterol of normal than of uniformily delayed (11). Lipoprotein uptake by the mutant mutant fibroblasts. Perinuclear filipin-cholesterol flu-cells leads to an excessive intracellular accumulation and orescence of both normal and mutant cells was co- storage of cholesterol primarily as unesterified sterol (8).The localized with lysosomes by indirectimmunocytochem- inability of internalized cholesterol to initiate timely regulaical stainingof lysosomal membrane protein. tory responses in these mutantcells could have resulted from Abnormal sequestration of LDL cholesterol in mu- a primary lesion either in the initiation of a regulatory mestant cells within a metabolically latent pool is sup- sage commonly shared by all the affected responses or in the ported by the finding that in vitro esterification of intracellular transport of cholesterol. The present data will cellular cholesterol could be stimulated in mutant but document that a sterol transport error plays a major role in not in normal cell homogenates by extensive disruption the cellular pathology of Niemann-Pick C disease. of the intracellular membranous structures of cells previously cultured withLDL. EXPERIMENTALPROCEDURES Deficient translocation of exogenously derived choMaterials-[l,10-3H]Oleic acid (2-10 Ci/mmol) was obtained from lesterol from lysosomes to other intracellular membrane sites may be responsible for the delayed homeo- Du Pont-New England Nuclear. ATP, CoA, fatty acid-free bovine static responses associated with LDL uptake by mutant serum albumin, and human LDL were purchased from Sigma. Precoated silica gel 60 plates were obtained from Merck. The enzymes Niemann-Pick Type C fibroblasts.

Type C Niemann-Pickdisease is a humanautosomal-recessive lipid storage disorder (1).Certain clinical, morphological, and biochemical similarities with type A and B NiemannPick diseases prompted the classification of this disorder as * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 5s To whom correspondence should be addressed.

used in the direct fluorometric assay of cholesterol masses, cholesterol oxidase, and horseradish peroxidase were purchased from Boehringer Mannheim. Lipoprotein-deficient fetal bovine serum (LPDS) was prepared by Biomedical Technologies, Boston, from KBr serum solutions by ultracentrifugation as described (8).ITS (insulin/transferrin/selenium) was obtained from Collaborative Research, Inc., Bedford, MA. Cell Cultures-Normal and mutant Type C Niemann-Pick fihroblasts represented established secondary cell lines derived from superficial skin biopsies of normal volunteers and confirmed patients of the Developmental and Metabolic Neurology Branch of the National Institutes of Health. Cell cultures were maintained in Eagle’s The abbreviations used are: LDL, low density lipoprotein; LPDS, bovine lipoprotein-deficient serum.

341 1

3412

Type C Niernann-Pick Disease

minimal essential medium supplemented with 10% (v/v) complete The remaining cell suspensions were extracted with 4 ml of chlorofetal bovine serum, 2 mM L-glutamine, 100 units of penicillin, and form/methanol (2:l v:v) and the lower lipid-containing phase evapo100 pg streptomycin/ml in humidified 95% air and 5%CO, at 37 "C. rated under a nitrogen stream. Lipid residues were taken up in 0.20 Cells were harvested by washing monolayers three times with phos- ml of isopropanol and the unesterified cholesterol measured with an phate-buffered saline (PBS) and subsequent treatment with 0.05% enzymatic fluorescence assay (17). Cholesterol levels in non-oxidasetrypsin (Sigma) for 5 min a t 37"C.Specific experimental culture treated cells equaled total cellular free sterol. Cholesterol levels remanipulations and conditions are described in the appropriate leg- maining in oxidase-treated cells represented intracellular free sterol. ends. Free cholesterol specifically associated with the plasma membrane Fimrescent Cytochemical and Immunocytochemical Staining of was calculated from the totalminus the intracellular levels. Cholesterol and Lysosomes-Cells were seeded and cultured directly In Vitro Esterification of CeUular Cholesterol-Freshly harvested on microscopic slide chambers (Lab Tek). In experiments designed and washed cell pellets (5 mg of protein) were suspended in 1 ml of to measure only unesterified cholesterol, cell mololayers were fixed 250 mM sucrose and 10 mM Tris-HC1, pH 7.4. The cell suspensions with 10%phosphate-buffered formalin and subsequently stained with were divided into separate aliquots of 0.20 ml. Some of the suspen0.05 mg/ml of filipin (generously supplied by The Upjohn Co.) for 60 sions were centrifuged and the cell pellets frozen in liquid nitrogen min as described previously (8).Fluorescence of stained preparations for 1h. The frozen cell pellets were subsequently taken up in 0.20 ml was photographed with excitation from a 100-watt mercury arc lamp of 10 mM Tris-HC1, pH 7.4, and frozen and thawed an additional passed through UG-1 filter and emission viewed through a 510-nm three times in liquid nitrogen and at37 'C at 5-min intervals. These filter using a 60-8 exposure. Concurrent filipin-cholesterol staining lysed cells were further disrupted by vigorous homogenization for 1 and rhodamine-labeled anti-lysosomal antibody fluorescence detec- min at 0 "C in a small glass-fritted homogenizing tube with a motortion were carried out as follows. Cells in the plastic slide chambers ized tight-fitting glass pestle at lo00 rpm. Other portions of the fresh were washed with PBS, fixed in 3% paraformaldehyde for 30 min at cell suspensions in the isotonic sucrose buffer were placed in a small room temperature, and washed three times with PBS. All subsequent NZcavitation chamber (Kontes) at 40 p.s.i. for 5 min at 4 'C. These steps were carried out in a 10% solution of normal fetal calf serum in partially lysed cell suspensions were further homogenized gently in a PBS with 0.05 mg/ml of filipin. The use of filipin in all incubation smooth-surface glass homogenizing tube fitted with a loose Teflon solutions served to permeabilize the cells to theantibody preparations pestle a t 100 r.p.m. for 30s a t 0 "C. It has previously been shown that and to fluorescently label unesterified cholesterol. The primary an- such controlled disruption allows cell-free extracts to essentially tibody was rat antibody specificallydirected against human lysosomal retain intact subcellular organelles (18).Aliquots (0.010 ml and 50 pg membrane protein and was a generous gift of Dr. J. W. Chen, of protein) of the respective total cell-free extracts were incubated in Department of Pharmacology and Experimental Therapeutics, Johns 0.19 ml of 250 mM sucrose, 2 mM dithiothreitol, 5 mM KF, and 10 Hopkins University School of Medicine, Baltimore, MD (13). Cells mM Tris-HC1, pH 7.4, containing 6 mM ATP, 0.6 mM CoA, 15 mM were incubated with the primary antibody at a 1:4 dilution for 30 MgCI, and 0.40 mM [3H]oleate (370 dmp/pmol in 14% fatty acid-free min. The cells were washed free of unbound primary antibody and bovine serum albumin). Incubations were carried out for 2 h at 37 "C incubated with affinity-purified goat anti-rat IgG conjugated to rbo- and thelipids subsequently extracted with chloroform/methanol (2:l damine (Jackson Labs, Avondale, PA) a t a dilution of 1:40 or 30 min. v:v). The level of [3H]oleateincorporated into cellular cholesterol to Cells were washed, mounted in para-phenylenediamineglycerol,and form chole~teryl-[~H]oleate was measured by thin layer chromatogviewed with a Leitz fluorescence microscope using an excitation filter raphy as described previously (8). (band pass 350-410) for filipin and (band pass 530-560) for rhodamine. Control for the immunocytochemical study was the replaceRESULTS ment of the specific primary monoclonal antibody to human lysosomal membrane protein with a nonspecific monoclonal antibody to Fluorescence Microscopic Studies of the IntracellularStorage mouse lysosomolmembrane protein and subsequent treatment of the of Unesterified Cholesterol-LDL uptake was monitored over cells with rhodamine-conjugated antibody. No rhodamine fluoresC cence was noted with the human cells. Controls for discrete visuali- aperiod of 24 h in normal and mutant Niemann-Pick zation of fluorescent signal were: 1)cells exposed to filipin alone and fibroblasts conditioned by an extended prior period of lipoviewed to the rhodamine (band pass 530-560) showed no signal; 2) protein deprivation. At indicated intervals, cell cultures were cells exposed to specific primary antibodies and second rhodamine- washed, fixed, and stained with filipin to follow the intracelconjugated IgG using saponin as the permeabilizing agent and viewed lular depositionof unesterified LDL-derived cholesterol. (Fig. at thefilipin (band pass350-410) showed no signal. 1). Prior to the uptake of LDL only low levels of filipin Determination of Unesterified Cholesterol in Plasma Membranesa minimal Advantage was taken of the observation of Lange(14, 15) that staining were found intheculturesindicating cholesterol oxidase efficiently and selectively oxidizes only plasma cellular content of unesterified cholesterol. Following 2 h of membrane cholesterol when intact cell suspensions are first treated LDL uptake, a distinctive perinuclear filipin staining develwith glutaraldehyde. Freshly harvested cell pellets (5 mg of protein) oped in both cell lines which was somewhat more intense in were suspended in 1 ml of PBS and aliquoted into separate 0.20-ml the Niemann-Pick C cells. After 6 h of LDL uptake, perinusamples. Cells were pelleted and resuspended in 0.20 mlof PBS & clear fluorescence in the mutant cells had increased to levels 1%glutaraldehyde. Following gentle mixing and incubation in an ice bath for 10-15 min, cells were pelleted by low speed centrifugation which were now significantly higher than those in normal and washed three times with 1 ml of 310 mM sucrose and 0.5 mM cells. Between 6 and 24 h of LDL uptake,a further significant phosphate buffer, pH 7.5, at 10 'C. The individual cell pellets were increase of perinuclear filipin-cholesterol staining was noted subsequently suspended in 0.20 ml of this buffer in the presence of in mutant but not in normal cells. The current fluorescent 2.5 mg/ml of cholesterol oxidase (6000 units/gm; Breuibacterium sp, studies expand our earlier observationsof excessive unesterBeckman Instruments). Cell suspensions were incubated for 45 min at 10 "C with gentle agitation. Cells were pelleted, washed, and ified cholesterol storage in mutant cells (9,10) and clearly suspended in 0.20 ml of PBS. Cell suspensions not exposed to gluta- now document the temporal and sequential fashion of LDL raldehyde were used to measure protein by the method of Lowry (16). cholesterol accumulation in NP-C cells. Flc. 1. Fluorescent filipin staining of the sequential intracellular accumulation of LDL-derived cholesterol in normal and Niemann-Pick C fibroblasts. Confluent flasks of normal and mutantcell cultures were placed on McCoy's medium with penicillin (100 units/ml) and streptomycin (100 mg/ml) + 1%glutamine, 0.1% ITS, 0.2% bovine serum albumin and no serum for 8 days. Cells were harvested and seeded in tissue culture wells (pretreated with 2.5 pg of fibronectin/cm') at a density of 2.0 X 10' cells/9.5-cm2chamber in 4 ml of McCoy's medium and 0.2% bovine serum albumin. These particular conditions of cholesterol deprivation were found to maximize most consistently a low base-line level of cholesterol in the cultured cells. The Niemann-Pick C phenotype of abnormal cholesterol homeostatic responses is fully expressed under these particular culture conditions (data not presented). Cells were cultured for 2 days and subsequently incubated in fresh medium f LDL (50 pg/ml). Cell monolayers were washed at the indicated intervals with PBS and subsequently fixed and stained with filipin for cholesterol as described in the experimental procedures. Magnification X 60.

Type C Niemann-Pick Disease

3413

0

e

Y

C C

E Q,

z

E

E r

E r

r

0

hl

hl

m E

b

Z

mc. 1.

3414

.l!ype

C Niemann-Pick Disease

Rhodamine Immuno-Staining of Lysosomes

Filipin Staining for Unesterified Cholesterol

I m m

FIG. 2 .C w kdhti~ Of m ~a a d ~holeoterolmd in 4 md NIenuanPick C ceh cultured with LDL. Normal and mutant cells were cultured and prepared for LDL uptake as described in the legend to Fig. 1. Cells were incubated with LDL (50 d m l )for 24 h and subsequently washed, fixed, and prepnred for cytochemical staining of unesteri6ed cholesterol with filipin and immunocytochemical stainiq of lyeoeomea with rhodamine-labled antibodiesas described under 'Experimental pmceduree."Magnification X 94.

Plasma Membmes-In order to directly quantitate relative distributions of unesterified cholesterol in the mpective intracellular and plasma membrane domainsof cultured fibroblasts, advantage was taken of the selectiveoxidation by cholesterol oxidaseof plasma membrane cholesterol in intact. glutaraldehyde-treatedcells (14,151. In cells grown with 10% lipoprotein-deficientserum for 4 days, comparable levelsand distributions of unesterified cholesterol were seen in normal and mutant cells with somewhat higher intracellular levels found in the mutant cells (Table I). Following 24 h of LDL uptake, total cellular unesterified cholesterol rose in mutant cells to levels substantially higher than those found in normal cells.However, in associationwith this internalization of excessive LDL cholesterol by mutant cells, there was less enrichment of the plasma membraneswith cholesterol when compared to normal cells. Both the relative maas measurementsas well as fluorescent filipin staining show plasma membrane cholesterol levelsto be substantially lower (1646% of total) than reported by others for cultured fibroblasts (>90% of total) (14,151. These former studies were carried out with confluent cells cultured with LPDS for 24 h. The present experiments were specscaUy desiied to study intracellular distribution of cholesterol inextensively sterol-depleted, sparsely populated, and actively growing cultures during the active phase of LDL uptake. It is likely that specific culture conditions play a major role in determining the disposition of cellular cholesterol between intracellular and plasma membranepools. Comparative Esterifications of InternaliEed Cholesterol in Normal and Mutant CeU Preparations-This documented excessive lysosomalstorage and tardy intracellular mobilization of cholesterol in Niemann-Pick C fibroblasts suggested that LDL uptake by the mutant cells resulted in aequeatration of exogenously derived cholesterolwithin a metabolicallysilent or trapped pool. In order to explore this possibility, advantage was taken of the reported in oitro modulations of cholesterol * J. Sokol, E. J. Blanchette-Mackie,H. S. Kruth. N. K. m e r , L. ester formation through direct alterations of cholesterol M. Amende, J. D. Butler, E. Robinson, S. Patel. R 0. Brady, M. E. within membranes that contain acyl-CoA.cholestero1 acyltransferase (19,201, the enzyme responsible forintracellular Comly, M. T. Vanier. and P. G. Pentchev. unpublisheddata.

The perinuclear storage depots for the unesterifkd cholesterol were identified as lysosomes (Fig.2). Filipin fluorescence in the perinuclear region showed a very high degree of identity with a second separate fluorescent signal (rhodamine antibody complex) targeted specifically to the lysosomes. Previous observations of mutant cells had indicated that excessive perinuclear accumulation of LDL cholesterol was contrasted witha deficient cholesterol replenishment of plasmamembraneswhencompared to normal cells (10). These preliminary observations could now be confirmedwith cell cultures that were extensively deprived of medium cholesterol prior to their uptake of LDL (Fig. 3). This extensive prior cholesterol depletion enhances the differences in intensity of filipin staining in LDL-treated and nontreated fibroblasts. Presumably, this pretreatment minimizes endogenous cellular cholesterol levels prior to lipoprotein loading. LDL uptake by normal cells was shown to be associated not only with notable perinuclear filipin staining butalso with a fainter but discemable development of filipin staining at the outer plasma membrane regionof the cells (Fig. 3). In comparably treated Niemann-Pick C fibroblasts,a significantlylower level of filipin fluorescence staining was noted in the plasma membrane even though very intense filipin-cholesterol staining formed within the perinuclear region of the mutant cells. Although these photomicrographs are made at asingle plane and show a somewhatdiffuse intracellular fluorescence at the outer boundaries of the cholesterol-loaded cells,application of through focus analysis at many planes in the cell allowed one to clearly identify a peripheral plasma membrane fluorescent line. Other ongoing studies with filipin at the electron microscopiclevel indicate that filipin-cholesterol complexes are present in the plasma membrane of normal LDL-loaded cells? Independent biochemical verification of specificplasma membrane cholesterol enrichment is presented in the next section. Relative Mass Measurement of Unesterified Cholesterol in

Type C Niemnn-Pick Disease

3415

NL

NP-C

FIG. 3. Fluoreecent filipin etaining of the differential intracellular distribution of LDL-derived cholesterol in normal and Niemann-Pick Type C fibroblasts. Stock normal and mutant cells were harvested and seeded in 25-cm' flasks at a density of 1X 106 cells in 5 ml of McCoy's medium containing 1% glutamine, 100 units/mlpenicillin, and 100 pg/d streptomycin and 5%LPDS. Medium was replaced every 4 days for 12 days. Cultures were harvested and seeded at a 1:2 dilution in 25-cm' flasks in 5 ml of above complete culture medium. After 24 h of c utu l re,medium wasexchanged with fresh medium in which LPDS was exchangedwith 0.2% bovine serum albumin. After 5 days of culture, cells were trypsinized and seeded in tissue culture chambers (pretreated with fibronectin, 1 pg/cm') at a density of 30,000 cell49.5 cm' of chamber area in 4 ml of above McCoy's medium

containing 0.2% bovine serum albumin. Following 2 days of culture, the cells were treated with fresh medium f LDL ( 5 0 &ml) for 24 h. Cell monolayers were washed three times with 4 ml of PBS. Fixation and the specific 6lipin staining techniques for cholesterol are described under "Experimental Procedures." Magnification X 114.

cholesterol estefllcation (21). In mutant Niemann-Pick C cells, the availability of cholesterol for interaction with acylCoAcholesterol acyltransferase was consideredpotentially latent because of possible topologicalhinderances which could be envisioned to block the translocation of sterol to the catalytic site of acyl-CoAcholesterol acyltransferase.Following in vivo uptake of LDL, subsequent in vitro esterification of internalized cholesterol could be modulated in cell-free extracts of mutant, but not normal, cells by regulating the extent of secondary organelledisruption (Table II). Following LDL uptake, in vitro synthesis of cholesterol [%]oleate from endogenous cholesterolstores could be varied in mutant cellfree extracts from levels below((20%) to above (150%) those of comparably treated control cell extracts by controlliig the extent of subcellular organelle disruption. It should also be noted that the relative in vitro deficiency of cholesterol esterification observed in mutant cell-free extracts with intact subcellular organelles corresponded to the relative deficiency of cholesterol esterification o b g e ~ e din situ with intact Niemann-Pick Ccells. In principle, activation of in uitro cholesterol esterification secondary to a disruption of intracellular membranous struc-

tures may just as readily reflect latency on the part of acylCoAcholesterol acyltransferase as it could the lack of availablecholesterol.Acyl-CoAcholesterolacyltransferase has been shownto reside normallyon the cytoplasmic sideof the rough endoplasmic reticulum(22). With regard to the latency of cholesterol esterification in Niemann-Pick C fibroblasts, the evidence strongly favors the existence of a sequestered and metabolically unavailable pool ofexogenously derived cholesterol rather than atopologicallymisplacedacylCoAcholesterol acyltransferase enzyme: (a) the phenotypic abnormalities presented by the Niemann-Pick C mutation reflect not only deficient acyl-CoAcholesterol acyltransferase catalysis but also deficient down-regulationof two other cholesterol-regulated proteins, the LDL receptor and hydroxymethylglutaryl-CoAreductase (ll), (b) the histochemical findings clearlyshow abnormal sterol accumulation, (c) normal orientation of acyl-CoAcholesterolacyltransferaseon the cytosolic sideof the endoplasmic reticulum inmutant cells is supported by the finding that acyl-CoAcholesterolacyltransferase of mutant cells was as susceptible, in gently disrupted cell homogenates, to proteolytic inactivation by added proteases as the enzyme of normal cellextracts (data not shown);

3416


( d ) in viuo and in vitro cholesterol esterification was normal or even somewhat elevated in mutant cells not cultured with

LDL (Table 11). DISCUSSION

Type C Niemann-Pick disease appears to representa newly defined and unique cholesterol storage disorder. The pathogenic abnormalities include major disruptions of intracellular cholesterolprocessing. Inmutantfibroblasts,extracellular LDL is carried by receptor-mediated endocytosis to lysosomes where apparently normal proteolytic and lipolytic processing of the exogenous lipoprotein initially takes place (11).However, the subsequent intracellular fate of the lysosomal cholesterol and the normal cellular responses to cholesterol uptake are compromised by the mutation. A translocation of exogenously derived cholesterol from lysosomes appears deficient. The internalized and sequesteredcholesterol of Niemann-Pick C fibroblasts fails to initiate the prompt homeoTABLE I Cellular distribution of unesterifiedcholesterol in normal and Niemann-Pick C fibroblasts Stock normal and mutant Niemann-Pick C fibroblasts were harvested and seeded a t a density of 9 X 10’ cells in 850-cm2 roller culture bottles with 100 ml of Eagle’s minimal essential medium, 2 mM glutamine, and 10% fetal bovine serum for 2 days. The culture medium was replaced with fresh McCoy’s medium with 10% lipoprotein-deficient human serum and 2 mM glutamine for 4 days. This medium wasreplaced with fresh medium 50 pg/ml of LDL protein/ ml and monolayers incubated an additional 24 h. Roller bottles were rinsed three times with 20 ml of PBS and subsequently harvested with 10 ml of 0.05%trypsin in PBS for 5 min at 37 “C. Cell suspensions were subsequently pelleted at 700 X g for 5 min and cells washed three times with 10 ml of PBS. Cell pellets (4-5 mg of protein) were suspended in 1.0 mlof PBS andkept on ice for subsequent analytical procedures described under “Experimental Procedures.” Each data point is the average of two separate cell cultures.

+

Cell culture

LDL

Unesterified cholesterol levels % total

Total in cell Plasmamembrane nmolfmg protein cell

50 pgfrnlf.24 h

Normal (2) Mutant (2) + Normal (2) + Mutant (2) a No additional LDL.

55 73 123 167

9 13 57 28

16 18 46 17

static responses that serve to control and to balance intracellular cholesterol levels in normal cells. There is associated with the Niemann-Pick C mutation a tardy down-regulation of the LDL receptor, a delayed suppression of 3-hydroxy-3methylglutaryl-coenzyme A reductase, and a defective stimulation of acyl-CoAcholesterol acyltransferaseexpression (11). As would be predicted, these delayed metabolic responses lead to excessive intracellular accumulationof unesterified cholesterol which is primarily stored inlysosomes (Fig. 2). This defective lysosomal translocation of cholesterol is not only associated with delayed homeostatic responses but also with an impaired enrichment of cholesterol in the plasma membranes of mutant cells (Fig. 3 and Table I). Abnormal intracellular cholesterol sequestrationcan also be inferred from the finding that additional extensive organelle disruption in cell-free extracts greatly enhances the in vitro availability of cellular cholesterol for esterification in cell-free extracts of mutant fibroblasts cultured with LDL (Table11). The molecular basis for the abnormal lysosomal sequestration of LDL-derived cholesterol in Niemann-Pick C disease is not known. LDL-cholesterolreleasedin lysosomes is thought to reach the endoplasmic reticulum and the Golgi apparatus (23). Saturation of a limited sterol pool within the endoplasmicreticulumpresumably initiatesthenumerous cellular homeostatic responses that enable normal cells to regulate intracellular cholesterol levels. The components of this cholesterol transport process from lysosomes to the endoplasmic reticulum are not known. It has been speculated that active vesicular or carrier-mediated transport may be involved (23). It is likely that Niemann-Pick C disease will prove to be a useful pathological model for elucidating additional steps of intracellular cholesterol processing. Earlier documentationof induced or genetic pertubations of the LDL pathway at the lysosomal stephasbeenlimitedtothe observations that blocked hydrolysis of LDL cholesterol esters leads to lysosomal accumulation of unhydrolyzed esters and retarded homeostatic responses (24-27). The Niemann-Pick C mutation clearly affects astep subsequent tohydrolytic lysosomal processing (11).Accumulation of unesterified cholesterol within the lysosomes of mutant cellsbegins to exceed the levels ‘found in normal cells as early as 2 h after initiation of LDL uptake, and by 24 h an extensive lysosomal cholesterol pool

TABLE I1 Accessibility of cellular cholesterol to in vitroand in vivo esterification innormal and Niemann-Pick C fibroblnsts Stock cell cultures were harvested and seeded at 9 X 10‘ cells in 850-cm2roller bottles and at 3 X lo6 cells in 25cm2flasks in 100 and 7 ml, respectively, of Eagle’s minimal essential medium + 10% fetal bovine serum for 2 days. The cultures were depleted of cellular cholesterol by culturing in McCoy’s medium + 10% LPDS for 4 days. Medium was replaced with fresh medium f 50 pg LDL protein/ml for 12 h. To the smaller 25-cm2culture flasks, 0.012 ml of 6 mM [3H]oleate (200 dpm/pmol) in 14% acid-free bovine serum albumin was added for the last 2 h of the incubation. These particular cultureswere washed, harvested, and subsequently analyzed for in vivocholesterol [3H]oleate formation by lipid extraction and thin layer chromatography (8). The larger cell cultures were also incubated f LDL (50 pg/ml) for 12 h and subsequently washed and harvested as described in Table I. These cell suspensions were analyzed for unesterified cholesterol levels and invitro cholesterol [3H]oleate formation as described under “Experimental Procedures.” The determinations represent the average of twoseparate cell cultures. Cholesterol [3H]oleatesynthesis Cell culture

LDL co-culture 50 ~ f m l f 1 h 2

Normal (2) Mutant (2) 400 + Normal (2) + Mutant (2) No additional LDL.

Cellular levels of unesterified Disrupted cholesterol Intact

In vitro I n vivo organelles

nmolfmg protein pmol

5

20 30 540 40 90

organelles

PHloleate incorporatedfmgf2 h

7 6 2000 500

0

10 30

80

530


3417

has formed (Fig. 1). Previous studies have shown that the 4. Schneider, p. B., and Kennedy, E. p. (1967) J. Lipid Res. 8,202209 total cellular accumulation of LDL cholesterol in these mu5. Sloan, H. R., Uhlendorf, B. W., Kanfer, J. N., Brady, R. O., and tant cells does not exceed that of normal cells during the first Fredrickson, D. S. (1969) Biochem. Biophys. Res. Commun. 3 4 , 6 h of lipoprotein uptake (8).Consequently, excessive storage 582-587 of cholesterol within lysosomes of the affected cells at this 6. Vanier, M. T., Rousson, R., Zeitouni, R., Pentchev, P. G., and early phase of lipoprotein uptake would suggest a delay in the Louisot, P. (1986) in Enzymesin Lipid Metabolism, Part II translocation of cholesterol from lysosomes to furtherintra(Freysz, L., Dreyfus, H., Massarelli, R., and Galt, S., eds) pp. 791-802, Plenum Press, New York cellular sites of distribution rather than over active endocytic 7. Pentchev, P. G., Comly,M. E., Kruth, H. S., Vanier, M. T., uptake. Wenger, D.A., Patel, S., and Brady, R. 0. (1985) Proc. Natl. The relationship between lysosomal storage and deficient Acad. Sei. U. S. A. 8 2 , 8247-8251 intracellular mobilization of cholesterol in Niemann-Pick c 8. Kruth, H. S., Comly, M. E., Butler, J. D., Vanier, M. T., Fink, J. fibroblasts suggests several possible disruptive mechanisms. K., Wenger, D. A., Patel, S., and Pentchev, P. G. (1986) J.Biol. Excessive lysosomal cholesterol accumulation may represent Chem. 2 6 1 , 16769-16774 a primary lesion at the ~ysosome itself. The diffusion of 9. Pentchev, P. G., Kruth, H. S., Comb, M. E., Butler, J. D., Vanier, M. T., Wenger, D. A., and Patel, S. (1986) J. Biol. Chem. 2 6 1 , hydrolyzed metabolites from lysosomes is often under the 16775-16780 control of Carrier-Imdiated Processes (28). Lysosomal ~ c u - io. Butler, J. D., Comly, M. E., Kruth, H. S., Vanier, M. T., Fillingmulation of cystine in cystinosis (29) and sialic acid in Salla Katz, M., Fink, J., Barton, N., Weintroub, H., Quirk, J. M., disease (30) are examples of blocked translocations of metaTokoro, T., Marshall, D. C., Brady, R. O., and Pentchev, P. G. (1987) Proc. Natl. Acad. Sci. U. S. A. 8 4 , 556-560 bolic products from loaded lysosomes. P. G., Comly, M. E., Kruth, H. S., Tokoro, T., Butler, On the otherhand, it is also possible that lysosomal choles- 11. Pentchev, J., Sokol, J., Filling-Katz, M., Quirk, J. M., Marshall, D. C., terol accumulation simply reflects the capacity and availabilPatel, S., Vanier, M. T., and Brady, R. 0. (1987) FASEB J. 1, ity of lysosomes to store cholesterol when they arecalled upon 40-45 to do so because of some more distal primary block. Potential 12. Brown, M.S., and Goldstein, J. L. (1986) Science 232,34-47 primary post-lysosomal abnormalities could include deficient 13. Chen, J. w., Pan, w., D’Souza, M. p., and August, J. T. (1985) Arch. Biochem. Biophys. 239,574-586 sterol carrier proteins or lesions in membrane interactions 14. Lange, Y., and Ramos, B. V. (1983) J. Bid. Chem. 258, 15130which normally serve to transport cholesterol to specific tar15134 get sites. A Partial and temporaw manifestation of excessive 15. Lange, Y., and Matthies, H. J. G . (1984) J. Biol. Chem. 2 5 9 , cholesterol storage and deficient sterol transport in hetero14624-14630 zygous mutant Niemann-Pick C fibroblasts during only the 16. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 early active phase of LDL uptake (8) tends to favor the possibility that these cells were temporarily oversaturated and 17. Gamble, Vaughan, M.9 Kruth, H. s.,and Avigan, J. (1978) J. Lipid Res. 19,1068-1070 partially deficient in some carrier or receptor-mediated trans- 18. Rome, L. H., Gamin, A. J., Allietta, M.M., and Neufeld, E. F. location process. It should also be noted that even with (1979) Cell 1 7 , 143-153 homozygous mutant fibroblasts, errors of cholesterol process- 19. Gavigan, S. J. P., and Knight, B. L. (1983) Biochem. J. 216,93100 ing reflect delayed or tardy cellular responses rather than absolute deficiencies (11). Whether the recovery toward nor- 20. Mitropoulos, K. A., Venkatesan, s.,Vrettakou, s. s.,Reeves, B. E. A., and Gallagher, J. J. (1984) Biochim. Biophys. Acta.7 9 2 , mal responses represents a “leaky” mutation or secondary 227-237 pathways Of lysosomal processing remains to be 21. Spector, A. A., Mathur, S. N., and Kaduce, T. L. (1979) Prog. established. The heterogeniety in the clinical presentations of Lipid. Res. 1 8 , 31-53 Niemann-Pick C patients (6) and thevariability noted in the 22. Hashimoto, S., and Fogelman, A. M. (1980) J. BWZ. Chem. 2 5 5 , 8678-8684 cholesterol processing deficiencies of different Niemann-Pick L., Brown, M. S., Goldstein, J. L., Garcia-Seura, L. M., and c cell lines (6, 7, 10) suggeststhat such considerations may 23. Orci, Anderson, R. G. W. (1984) Cell 3 6 , 835-845 be pertinent for a Of the and 24. Goldstein, J. L., Brunschede, G. Y., and Brown, M. S. (1975) J. molecular pathogenesis of this disease. Biol. Chem. 2 5 0 , 7854-7862 w.9

Acknowledgment-We wish to thank Priscilla Gaeta for typing the manuscript.

REFERENCES 1. Bra& R. 0. (1983) in The Metabolic Basis of Inherited Disease (Stanbury, J. B., Wyngaarden, J. B., Fredrickson, D. S., Goldstein, J. L., and Brown, M. s.,eds) pp. 831-841, McGraw-Hill Publications, New York 2. Crocker, A. C. (1961) J. Neurochem. 7,69-80 3. Brady, R. 0.. Kanfer, J. N., Mock, M. B., and Fredrickson, D. S. (1966) Proc. Natl. Acad. Sci. U. S. A. 55, 366-369 759-762

25. Brown, M. S., Dana, S. E., and Goldstein, J. L. (1975) Proc. Natl. Acad.Sci. U. S. A . 7 2 , 2925-2929 26. Goldstein, J. L., Dana, S. E., Faust, J. R., Beaudet, A. L., and Brown, M. S. (1975) J. Biol. Chem. 250,8487-8495 27. Brown, M. S., Sobhani, M. K., Brunschede, G. Y., and Goldstein, J. L. (1976) J. Biol. Chem. 251,3277-3286 28. Bernar, J., Tietze, F., Kohn, L. D., Bernardini, I., Harper, G. S., Grollman, E. F., and Gahl, W. A. (1986) J. Biol. Chem. 2 6 1 , 17107-17112 29. Gahl, W. A., Bashan, N., Tietze, F., Bernardini, I., and Shulman, J. D. (1982) Science 2 1 7 , 1263-1265 30. Renlund, M., Tietze, F., and Gahl, W. A. (1988) Science 2 3 2 ,