synthesis of inactive nonsecretable high mannose-type lipoprotein

Vol. 265, No. 3, Issue of January 25, pp. 162%X38,1990 Printed in U. S. A.

Synthesis of Inactive Nonsecretable High Mannose-type Lipoprotein Lipase by Cultured Brown Adipocytes of Combined Lipase-deficient cZcl/cZcZ Mice*

(Received for publication, June 29, 1989)

Hiroshi MasunoS, E. Joan Blanchette-Mackie, Sidney S. Chernick, and Robert 0. Scow From the Endocrinology Section, Laboratory of Cellular and Developmental Biology, National Institutes of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892

Combined lipase deficiency (cld) is a recessive mutation which causes a severe deficiency of lipoprotein lipase and hepatic lipase activities and lethal hyper- triacylglycerolemia within 3 days in newborn mice. The effect of this genetic defect on lipoprotein lipase was studied in primary cultures of brown adipocytes derived from tissue of newborn mice. Cells cultured from cld/cld mice replicated, accumulated triacylglycerol, and differentiated into adipocytes at normal rates. Lipoprotein lipase activity in unaffected cells was detectable on Day 0 of confluence and increased to 1.3 units/mg DNA by Day 6, while that in cldjcld cells was ~4% of that in unaffected cells on Days 4-6. Unaffected cells released 1.2% of their lipase activity in 30 min in the absence of heparin, and 11% in 10 min in the presence of heparin, whereas cld/cld cells released no lipase activity. cld/cld cells contained 2-3 times as much lipoprotein lipase protein as unaffected cells, and released no lipase protein to the medium. Immunofluorescent lipoprotein lipase was not detectable in unaffected adipocytes unless lipase secretion was blocked with monensin, causing retention of the lipase in Golgi. cld/cld adipocytes, in contrast, contained immunofluorescent lipoprotein lipase distributed in a diffuse reticular pattern, indicating retention of lipase in endoplasmic reticulum. Lipoprotein lipase immunoprecipitated from cells incubated l-3 h with [3SS]methionine was digested with or without endogly- cosidase H (endo H) or F, and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Lipoprotein lipase in unaffected cells (Mr = 56,000- 58,000) consisted of three glycosylated forms, of which the most prevalent was endo H-resistant, the next was totally endo H-sensitive, and the least was partially endo H-sensitive. In contrast, lipoprotein lipase in cldl cld cells (Mr = 56,000) consisted of a single, totally endo H-sensitive form. Lipoprotein lipase in both

*Preliminary reports of this study were presented at the 72nd Annual Meeting of the Federation of American Societies for Experi- mental Biology (Masuno, H., Chernick, S. S., and Scow, R. 0. (1988) FASEB J. 2, 1532 (abstr.)) and the Joint Meeting of the American Society for Biochemistry and Molecular Biology and the American Society for Cell Biology (Masuno, H., and Scow, R. 0. (1988) J. Cell Biol. 107, 388 (abstr.); Blanchette-Mackie, E. J., Masuno, H. and Scow, R. 0. (1988) J. Cell Biol. 107, 764 (abstr.)). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Present address: The 2nd Dept. of Medical Biochemistry, School of Medicine, Ehime University, Shigenobu, Onsen-gun, Ehime 791. 02, Japan.

groups of cells contained two oligosaccharide chains. Chromatography studies with heparin-Sepharose in- dicated that at least some of the lipoprotein lipase in cldjcld cells was dimerized. The findings demonstrate that brown adipocytes cultured from cldjcld mice synthesize lipoprotein lipase with two high mannose oligosaccharide chains, but it is inactive and retained in endoplasmic reticulum. Whether the cld mutation affects primarily processing of oligosaccharide chains of lipoprotein lipase in endoplasmic reticulum, transport of the lipase from the reticulum, or some other process, is to be resolved.

Combined lipase deficiency (cld) is a recessive mutation within the T/t complex of chromosome 17 in mice which causes functional deficiencies of lipoprotein lipase and hepatic lipase (1). Lipoprotein lipase is necessary for uptake of triacylglycerol from chylomicrons and very low density lipoproteins in blood (2-4), whereas hepatic lipase converts VLDL remnants to low density lipoproteins (5,6). Mice homozygous for this deficiency (cld/cld) develop extreme hypertriacyl- glycerolemia (10,000 mg of triacylglycerol/dl plasma at 12 h), have tissues that are devoid of fat, and die within 3 days if allowed to suckle (1, 7). Although brown adipose tissue, heart and diaphragm muscle in cld/cld mice have subnormal ((5%) lipoprotein lipase activities, they contain 2-6 times normal amounts of lipoprotein lipase protein and synthesized normal- sized lipoprotein lipase protein (8). Lipoprotein lipase activity in post-heparin plasma in cld/cld mice was <lo% of that in unaffected mice (9).

Lipoprotein lipase is produced by parenchymal cells of tissues and transported to the luminal surfaces of their cap- illaries where it acts (2-4, 10). Studies in cultured 3T3-Ll (11) and 3T3-F442A (12) murine white adipocytes and in freshly isolated guinea pig white adipocytes (13) showed that these cells secrete active lipoprotein lipase soon after it is synthesized, and that active lipase in cells has a short half- life, 20-60 min. The studies in 3T3-Ll white adipocytes indicate that cells that synthesize lipoprotein lipase also contain inactive lipase (11).

Lipoprotein lipase has been described as a noncovalent dimer of identical glycopeptide chains (M, = 55,000) (4, 14). That certain mutations in the T/t complex of chromosome 17 can impair glycosylation of proteins (15) suggested that defective glycosylation of lipoprotein lipase might be part of the cld phenotype. This paper describes a study of the effect of the cld mutation on synthesis, glycosylation, intracellular transport, activity, and secretion of lipoprotein lipase in pri-

1628

by guest on February 11, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Inactive High Mannose-type Lipase in cld/cld Adipocytes

mary cultures of brown adipocytes derived from adipose tissue of newborn mice.

normal of subnormal length tail (1). Thus, suckled mice which were tailless and had creamy pink blood were classified as defective (cld/ cld), whereas suckled mice which had either a normal or subnormal length tail and noncreamy red blood were classified as unaffected controls. Both +/cld heterozygotes and homozygous wild types were included in the unaffected group because it was not possible to identify homozygous wild type mice in litters containing cld/cld mice and there was no gene dosage effect of the cld mutation on lipoprotein lipase (1).

EXPERIMENTAL PROCEDURES

Moterials-L-[35S]Methionine, ENHANCE’“, and endo-P-N-ace- tvlglucosaminidase F were from Du Pont-New England Nuclear. Endo-fl-N-acetylglucosaminidase H was from United States Bio- chemical Corn. Tri19.10-3Hloleovlelvcerol and i4C-methvlated nro- teins for molecular’ weight ‘standards were from Amersham Corp. Nonradioactive molecular weight standards for SDS-PAGE’ were from Bio-Rad. DMEM and methionine-deficient DMEM were from Bio Fluids. Fetal bovine serum was from GIBCO. Leupeptin and pepstatin were from Boehringer Mannheim. Insulin, triiodothyronine, antipain, benzamidine, aprotinin, and octanoic acid were from Sigma. Heparin was from Hynson, Westcott and Dunning, Baltimore, MD. Collagenase (type I) was from Worthington. Crystalline bovine serum albumin was from Miles Laboratories. Rabbit anti-chicken IgG was from Pel-Freez. Chicken antiserum to bovine lipoprotein lipase was kindly given by Dr. Thomas Olivecrona, Department of Physiological Chemistry, University of Umeb, Umei, Sweden. All other chemicals were of highest quality commercially available.

Solution A, used for aqueous extraction of acetone/ether powders, contained 50 mM NH,/NH&l and 20 fig/ml heparin, pH 8.1. Solution B, used for immunoprecipitation, contained 0.1 M sodium borate, 0.5 M NaCI, 1% Triton X-100, 5 mM EDTA, 10 rg/ml leupeptin, 1 pg/ ml pepstatin, 3.5 pg/ml aprotinin, and 0.1 mM phenylmethylsulfonyl fluoride, pH 8.0. Solution C, used to dissolve immunoprecipitates for SDS-PAGE, contained 0.0625 M Tris, 2% SDS, 5% 2-mercaptoetba- nol, 10% glycerol, and 0.001% bromphenol blue, pH 6.8. Three solutions were used for harvesting and suspending cells: solution D, used when lipoprotein lipase was immunoprecipitated from cell extracts, contained0.2 M Tris, 3% Triton X-100,1% N-lauroylsarcosine, 0.15 M NaCl, and 1 mM phenylmethylsulfonyl fluoride, pH 7.5; solution E, used when acetone/ether powers were prepared from cells, contained 2% bovine serum albumin in solution A; and solution F, used when lipoprotein lipase was applied to a heparin-Sepharose column, contained 25 mM ammonia. 5 mM EDTA. 0.8% Triton X- 100, 0.64% SDS, 1 pg/ml pepstatin, i0 pg/ml leupeptin, and 3.5 fig/ ml aprotinin, pH 8.2. Solution G, used when lipoprotein lipase was eluted from a heparin-Sepharose column, contained 20 mM Tris, 20% glycerol (v/v), and 0.1% Triton X-100 (w/v), pH 7.4.

Standard culture medium contained 10% fetal bovine serum, 17 MM D-pantothenic acid, 33 pM d-biotin, 100 MM ascorbic acid, 100 units/ml penicillin, 0.1 mg/ml streptomycin, and 0.25 pg/ml ampho- tericin B in DMEM. Differentiation medium contained 0.5 nM insulin, 0.5 nM triiodothyronine, and 1 mM octanoic acid in standard culture medium.

Animals-Mice bearing the cld mutation were raised in the animal facilities of the National Institute of Diabetes and Digestive and Kidnev Disease. National Institutes of Health. Bethesda. MD from stock kindly supplied by Dr. Karen Artzt of the Patterson Laboratory, University of Texas, Houston. The cld mutation, which causes combined lipase deficiency, is an autosomal recessive mutation that was extracted from a chromosome bearing mutations at the T/t complex of mouse chromosome 17 (1). Mice homozygous for this defect (cld/ cld) have <5% normal amounts of lipoprotein lipase and hepatic lipase activities (1, 8, 9), and 50 times normal plasma triacylglycerol concentration at 6-8 h and 125 times normal at 12-30 h (1). There was no evidence of a gene dosage effect in the activity levels of lipoprotein lipase or hepatic lipase (1). The plasma triacylglycerol levels of wild type and +/cld heterozygote weanling littermates were virtually identical, about 80 mg/dl (1). It was not feasible to measure plasma triacylglycerol concentrations of individual tissue donors before processing the adipose tissue because of the large number of mice needed for each experiment and the very small amount of plasma that could be obtained from each mouse (<20 ~1). The very high plasma triacylglycerol concentration in suckled >12-h-old cld/cld mice, >lO,OOO mg/dl, gave the blood of such animals a creamy pink color, in contrast to the noncreamy red color of blood of unaffected normolipemic mice. Because of the close linkage of the cld mutation and mutations affecting tail length, about 95% of the mice born with combined lipase deficiency have no tail, and the other 5% have a

’ The abbreviations used are: SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; DMEM, Dulbecco’s modified Ea- gle’s medium containing 25 mM glucose; PBS, Dulbecco’s phosphate- buffered saline.

1629

Culture of Brown. Adipocytes-Primary cultures of brown adipocytes were grown from the stromal-vascular fraction of brown adipose tissue taken from suckled 18-30-h-old cld/cld mice and suckled 12- 24-h-old unaffected mice. Cells were isolated from brown adipose tissue by a modification of the method of Forest et al. (16). Intersca- pular brown adipose tissue was resected under sterile conditions from decapitated mice and cut into small pieces. Although white adipose tissue normally develops around interscapular brown fat pads in mice, it was difficult to detect in <2-day-old cld/cld mice and <l-day-old unaffected mice, even with a dissecting microscope at x 10 magnification. About 9 mg of tissue could be obtained from each cld/cld mice and 13 mg from each unaffected mice. Brown adipose tissue from five cld/cld or three unaffected mice usually provided sufficient cells to seed one 60-mm diameter plate. Tissue fragments, about 40 mg/ml, were incubated in a solution of 2 mg/ml collagenase and 20 mg/ml bovine serum albumin in DMEM, pH 7.5, for 60 min at 37 “C with gentle shal+ng. The digested tissue was filtered through a nylon screen with 100 Frn pore size. The filtrate was centrifuged at 800 X g for 10 min at room temperature. The supernatant was removed and centrifuged at 800 X g for 10 min; this procedure was repeated two more times. The pellets, which consisted of stromal-vascular compo- nents of the tissue, were combined and suspended in differentiation medium. Cells in the suspension were filtered through a nylon screen with 41 Frn pore size, seeded in 5 ml of differentiation medium on 60- mm diameter plates at a density of 2 x lo5 cells/cm*, and incubated at 37 “C in an atmosphere of 5% CO, in air. 2-days later, the culture plates were rinsed 3-5 times with standard medium and replenished with 5 ml of differentiation medium. The differentiation medium was changed every 2 days before confluence and every day after confluence. The cells became confluent on the 4-5th day of culture.

At the end of the experiments, the incubation medium was removed and the plates were rinsed 1-2 times with ice-cold PBS. The cells were scraped from the plates with a plastic cell scraper (Costar) into solution E or D and briefly sonicated at 0 ‘C. Cells were harvested into solution E for preparation of acetone/ether powders and measurement of DNA and triacylglycerol, and into solution D in studies with [35S]methionine.

Assay of Lipoprotein Lipase Actiuity-Lipoprotein lipase activity associated with cells was measured in aqueous extracts of dried- defatted (acetone/ether) powders of cells (17). The extract was made by adding the powder to ice-cold solution A, letting the mixture stand at 0 “C for 60 min, sonicating briefly at 0 “C, centrifuging for 10 min at 1200 X g at 4 ‘C, and decanting the supernatant for assay. Lipo- protein lipase activity in culture medium was measured in aliquots of medium filtered through 0.22-Frn Millex-GS filter units. These assays were begun within 3 min after taking the aliquots.

A stock emulsion containing 5 mCi of tri[9,10-3H]oleoylglycerol, 1.13 mmol of trioleoylglycerol, 60 mg of L-oc-phosphatidylcholine (bovine liver), and 9 ml of glycerol was prepared according to the method of Nilsson-Ehle and Schotz (18). Before assay, 1 volume of the stock emulsion, 19 volumes of 3% bovine serum albumin in 0.2 M Tris-HCl buffer, pH 8.1, and 5 volumes of heat-inactivated fasted rat serum (heated 60 “C for 10 min) were mixed and incubated at 37 “C for 15-30 min. For assay, 100 ~1 of this activated substrate mixture, containing 2.0 &!i of tri[3H]oleoylglycero1, 450 nmol of trioleoylglycerol, and 34 nmol of albumin, was added to 100 ~1 of diluted tissue extract or culture medium, and incubated at 37 “C for 60 min. Fatty acids produced by lipolysis were extracted and measured as described previouslv (17). 1 milliunit of lioolvtic activitv renresents release of i nmol of fatty acid/min. - -

- _

~5SIMethionin.e Zncomoration Studies-Incornoration of 13Slme- thionine into lipoprotein lipase was studied in cultured brown’adipocytes incubated with methionine-deficient DMEM containing 0.5 nM insulin, 0.5 nM triiodothyronine, 17 pM D-pantothenic acid, 33 yM d-biotin, 100 FM ascorbic acid, 1 mM octanoic acid, and 2% fetal bovine serum which had been dialyzed overnight against methionine- deficient DMEM. At the start of the experiment, the culture medium was removed, the plates were rinsed 2 times with PBS, 2 ml of methionine-deficient DMEM was added to each plate, the plates were


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

1630 Inactive High Mannose-type Lipase in cld/cld Adipocytes

returned to the incubator. 30 min later, 180 &i of [Yi]methionine was added to each plate. At the end of the experiment, the medium was removed and the plates were rinsed 2 times with ice-cold PBS. The cells were harvested into 0.5 ml of solution D, sonicated briefly at 0 “C, the suspension was centrifuged at 15,000 x g for 20 min at 4 “C, and the infranatant, the layer below the fatty cake, was removed and stored at -70 “C!.

An aliquot of the infranatant was taken for measurement of amount of [35S]methionine incorporated into total protein (precipitated with 10% trichloroacetic acid). lOO-200-/11 aliquots of the infranatant were taken and each was mixed with 1 ml of solution B for immunoprecipitation of lipoprotein lipase. Nonspecific immunoprecipitates were removed by adding first 10 ~1 of 10% normal chicken serum to the chilled sample and adding several h later 50 ~1 of rabbit anti-chicken IgG, allowing the mixture to stand overnight at 4 ‘C!, and removing the precipitates by centrifugation. The supernatant was mixed with 2 pl of either chicken antiserum to bovine lipoprotein lipase or chicken nonimmune serum and the mixture was allowed to stand for 48 h at 4 “C. 100 ~1 of rabbit anti-chicken IgG was added and 16 h later the immunoprecipitate was collected by centrifugation. The immunoprecipitate was washed 5-6 times with 0.1% SDS in solution B, and once with 0.0625 M Tris-HCl buffer at 6.8. The washed precipitates were dissolved in solution C for PAGE in a Laemmli-type system (19) with 10% acrylamide resolving gel and 3% acrylamide stacking gel. i4C-Labeled protein molecular weight mark- ers were run with each gel. Electrophoresis was terminated when the marker dye was within 1 cm of the end of the gel. Gels were stained with Coomassie Blue, destained, impregnated with ENHANCE’“, and dried on a sheet of cellophane with vacuum. Autoradiographs were obtained by exposure of Kodak X-Omat tilm to gels for periods up to 14 days at -70 “C. Radioactive zones were cut from the gels and dissolved in 0.75 ml of 30% hydrogen peroxide at 65 “C overnight. Radioactivity was determined in a liquid scintillation counter.

Puke-Chase Experiments-At the start of experiments, the culture medium was removed, the plates were rinsed 2 times with PBS, 2 ml of methionine-deficient DMEM was added to each plate, and the plates were returned to the incubator. After 30 min, 180 &i of [s5S] methionine was added to each plate and the plates were incubated (pulsed) for 10 min. Following the pulse, the medium was removed, the plates were washed quickly 2 times with PBS, 2 ml of differentiation medium was added to each plate, and the plates were incubated (chased) for O-90 min. After the chase, the medium was removed, the plates were washed 2 times with ice-cold PBS, and the cells were harvested into solution D, as above, for study of ?+labeled lipoprotein lipase.

Enzymatic Deglycosylation of Lipoprotein Lipase-Lipoprotein lipase immunoprecipitated from Y+labeled proteins was washed 5-6 times with 0.1% SDS in solution B and once with 0.1 M sodium citrate, pH 5.5, dissolved with heating for 6 min in 30 ~1 of 1% SDS in 0.1 M sodium citrate, pH 5.5, and centrifuged. For endo H digestion, 10 ~1 of the supernatant was mixed with 4 ~1 of proteinase inhibitor mixture (1 mg/ml leupeptin, 1 mg/ml antipain, 10 mg/ml benzamidine, and 3.5 fig/ml aprotinin) and 110 ~1 of 0.22% 2-mercaptoethanol in 0.1 M sodium citrate, pH 5.5. 6 ~1 of 1 unit/ml endo H solution was then added and the mixture was incubated at 30 “C! for 20-24 h. The reaction was terminated by adding 130 ~1 of double strength solution C and heating at 95 “C for 10 min. For endo F digestion, 10 ~1 of the supernatant was mixed with 4 ~1 of proteinase inhibitor mixture and 70 ~1 of 1.43% Triton X-100 and 1.43% 2-mercaptoethanol in 0.1 M NaHzPOr, pH 6.1. 16 ~1 of 100 units/ml endo F in 0.1 NaH2P0,, pH 6.1, was then added and the mixture was incubated at 30 “C for 20- 24 h. The reaction was terminated by adding 100 ~1 of double strength solution C and heating at 95 “C for 10 min.

Chromatography of 35S-Labeled Protein on Heparin-Sephurose Col- umn-plates of cells were rinsed 2 times with PBS, 2 ml of methionine-deficient DMEM was added to each plate, and the plates were returned to the incubator. After 30 min, 180 &i of [?S]methionine was added to each plate, and the plates were incubated at 37 “C for an additional 2 h. The cells were harvested into 0.5 ml of solution F, sonicated briefly at 0 “C, and centrifuged. 0.9 ml of the combined infranants from two plates was applied to a heparin-Sepharose column (column size 2 ml) equilibrated with solution G. The column was washed with solution G and then eluted first with 0.6 M NaCl in solution G and then with 1.5 M NaCl in solution G (20). Fractions of 0.5 ml were collected. 10 ~1 of each fraction was taken for measurement of [35S]methionine incorporated into total protein (precipitated with 10% trichloroacetic acid) and 200 ~1 was used for immunoprecipitation of lipoprotein lipase.

Radioimmunoassay of Lipoprotein Lipose-Radioimmunoassay of lipoprotein lipase in cells and culture medium was carried out as described previously (8, 11). Culture medium from 3T3-Ll cells was used as the standard and data are expressed as volumes of the standard (S) medium (8, 11).

Chemical Analyses-DNA was measured fluorometrically by the method of Hinegardner (21) using calf thymus DNA as standard. Triacylglycerol was measured fluorometrically using a kit from Sigma (No. 336) for enzymatic determination of triacylglycerol.

Immunolocalization of Lipoprotein Lipase in Cells-A fluorescent double antibody technique was used for immunolocalization of lipoprotein lipase in cells cultured from brown adipose tissue of newborn mice. The primary reactant was antiserum raised in chicken against bovine milk lipoprotein lipase, which cross-reacts with mouse lipoprotein lipase (8, 11). The secondary reactant was rhodamine-labeled affinity-purified rabbit anti-chicken IgG (Jackson Immunoresearch, West Grove, PA). The cells were fixed with 3% paraformaldehyde, free aldehyde groups were quenched (to prevent nonspecific binding of antibodies) with 1.5 mg/ml glycine and 0.75 mM albumin in PBS, and the cells were permeabilized with 0.1% saponin and 0.75 mM albumin in PBS before being incubated with the antisera.

RESULTS

DNA and Triacylglycerol Content of Cells Cultured from Brown Adipose Tissue of Unaffected and cld/cld Mice-Cells cultured from the stromal-vascular fraction of brown adipose tissue of newborn unaffected and cld/cld mice replicated, became confluent within 4-5 days, and accumulated triacylglycerol (Fig. 1). About 70% of the cells contained lipid droplets on Day 4 of confluence (Figs. 2A and 3A). DNA content of the cultures increased 4-fold during the 2 days preceding confluence and about l-fold during the next 6 days,

FIG. 1. Changes in DNA and triacylglycerol content during replication and transformation of cells cultured from brown adipose tissue of unaffected and cld/cld mice. Cells in the stromal-vascular fraction of brown adipose tissue of I-day-old unaffected and cld/c.!d mice were seeded in differentiation medium on 60- mm plates at a density of 2 x lo5 cells/cm’ and grown in 5 ml of differentiation medium. The cells became confluent 4-5 days after plating. Values given are the mean f S.D. of three plates.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Inactive High Mannose-type Lipase in cld/cld Adipocytes 1631

whereas the triacylglycerol content of cells increased about l- fold during the preceding 2 days and 4-fold during the first 4 days after confluence (Fig. 1). The average triacylglycerol content was 41 pmol/mg DNA on Day 4 of confluence. There was no difference between unaffected and cld/cld cells in DNA content, triacylglycerol content, or percentage of cells that contained lipid droplets.

Lipoprotein Lipase Activity in Unaffected and cld/cld Cells- Lipoprotein lipase activity in cells was measured in aqueous extracts of acetone/ether powders of cells (Table I). Lipopro- tein lipase activity in unaffected cells was first detectable on Day 0 of confluence and increased &fold to 1290 milliunits/ mg DNA during the next 6 days. Lipoprotein lipase activity in cld/cld cells also appeared on Day 0 of confluence, but it was 85% less than that in unaffected cells and did not change significantly during the next 6 days. The level of activity on Day 6 was 1% of that in unaffected cells.

Effect of Heparin on Release of Lipoprotein Lipase Activity by Unaffected and cld/cld Cells-Unaffected cells which contained about 510 milliunits/mg DNA of lipoprotein lipase activity on Day 4 of confluence released spontaneously 6.1 milliunits/mg DNA of lipase activity (about 1.2% of total) to the culture medium in 30 min at 37 “C. cld/cld cells, in contrast, released no lipoprotein lipase activity to the medium (data not shown).

The effect of heparin on release of lipoprotein lipase activity was studied in cells cooled on ice (Table II). Heparin (20 pg/ ml) caused 11% of the lipoprotein lipase activity associated with unaffected cells to be released to the medium in 10 min (Table II). A similar effect was observed with a g-fold higher concentration of heparin. Heparin did not cause release of lipoprotein lipase activity from cld/cld cells (data not shown).

Immunoinhibition of Lipoprotein Lipase Activity in Unaf- fected CeEls-Antiserum raised in chicken against bovine milk lipoprotein lipase reacts with mouse lipoprotein lipase (8, 11). We studied the effect of this antiserum on lipoprotein lipase activity in unaffected cells. 50 ~1 of the aqueous extract of acetone/ether powders prepared from unaffected cells on Day 4 of confluence was incubated 2 h with 50 ~1 of either diluted chicken antiserum to bovine lipoprotein lipase or diluted chicken nonimmune serum at 0 “C, and then assayed for lipase activity in the presence of serum-activated substrate. We found that antiserum to bovine lipoprotein lipase inhibited lipoprotein lipase activity of unaffected cells in a dose-de- pendent manner, with 5 ~1 of antiserum producing 95% inhi- bition, while nonimmune serum had no effect. The antiserum also inhibited lipase activity released spontaneously to the medium by unaffected cells. These results indicate that chicken antiserum to bovine lipoprotein lipase cross-reacted

with lipoprotein lipase of cells cultured from brown adipose tissue of newborn unaffected mice, and that lipoprotein lipase accounted for most of the lipolytic activity measured in extracts of acetone/ether powders of these cells.

Lipoprotein Lipase Protein in Unaffected and cld/cld Cells- Lipoprotein lipase protein in cells was measured by radioimmunoassay (8, ll), using chicken antiserum to bovine lipoprotein lipase, in the aqueous extract of acetone/ether powders of unaffected and cl&/& cells (Table I). Lipoprotein lipase protein appeared in unaffected and cld/cld cells at confluence, and remained constant in amount in unaffected cells for 6 days and in cld/cld cells for 4 days. The amount in cld/cld cells was l-2 times higher than that in unaffected cells on Days O-4.

Lipoprotein lipase protein was released spontaneously to the culture medium by unaffected cells, starting at confluence, reaching a maximum on Days 2-4, and decreasing slightly on Day 6 (Table I). The amount released per 24 h was equivalent to 4-8 times the amount of lipoprotein lipase protein in the cells. In contrast, cld/cld cells released no lipoprotein lipase protein to the medium even though they contained l-2 times more lipase protein than unaffected cells.

Immunolocalization of Lipoprotein Lipase in Unaffected and cld/cld Brown Adipocytes-Lipoprotein lipase was localized in cells with a fluorescent double labeled antibody technique, using antiserum against bovine lipoprotein lipase and rhodamine-conjugated second antibodies. Lipoprotein lipase could not be localized in cells cultured from brown adipose tissue of unaffected mice unless the cells were treated with monensin to inhibit transport/secretion of lipoprotein lipase (22). In monensin-treated unaffected cells, the lipase was located mostly in the Golgi complex (Fig. 2B), confirming observations of Vannier et al. (22). In contrast, most of the cells cultured from tissue of cld/cld mice contained immunolabeled lipoprotein lipase and the lipase was distributed in a diffuse reticular pattern, suggesting it was located mostly in endoplasmic reticulum (Fig. 3, B and C). Monensin had no effect on amount or distribution of lipoprotein lipase protein in cld/ cld cells. Lipoprotein lipase was found in both groups in cells that contained lipid droplets. On the basis of these findings, we conclude that the principle cells involved in synthesis and processing of lipoprotein lipase in these studies were brown adipocytes.

Incorporation of p5S]Methionine into Lipoprotein Lipase by Unaffected and cld/cld Brown Adipocytes-Unaffected and cld/cld cells were incubated with [35S]methionine-deficient DMEM for l-3 h. The amount of 35S in total proteins in these cells increased at similar rates for 2 h (Fig. 4). 35S-Labeled lipoprotein lipase in the infranatant of extracts of these cells

TABLE I

Changes in amount of lipoprotein lipax activity and protein in cells and lipase protein in medium during replication and transformation of cells cultured from brown adipose tissue of unaffected and cld/cld mice

Lipoprotein lipase activity and protein in cells were measured in aqueous extracts of acetone/ether powders of the cells. Lipase protein was measured in medium in which cells had been grown during the preceding 24 h. Values given are the mean f SD. for three plates.

Day of COdlUenCe

Lipoprotein lipase activity in cells

Unaffected ckilcld

Lipoprotein lipase protein

In cells In medium at 24 h

Unaffected cldlcki Unaffected cldlcid

-2 0 2 4 6

milliunits/mg DNA ~1 S medium/mg DNA

0 0 0 0 0 0 144 + 9 21 k 12 440 + 80 990 z!z 240 1730 + 160 0 195 r 10 71 f 26 430 + 40 1310 zk 440 3020 + 270 0 911 + 136 29 + 4 560 + 50 1350 -t 390 2770 + 310 0

1290 + 233 12 + 4 500 + 50 610 + 110 1760 + 150 0


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

1632 Inactive High Mannose-type Lipase in cldlcld Adipocytes

TABLE II E/feet of heparin on release of lipoprotein lipase actioity from cells

cultured from brouln adipose tissue of unaffected mice The culture medium was removed from plates of cells on Day 4 of

confluence and replaced with 3 ml of differentiation medium. The plates were returned to the incubator for 30 min, removed from the incubator, and placed on crushed ice. After standing for 2 min, 1 ml of the medium was removed, passed through a 0.22.pm filter and assayed immediately for lipoprot.ein lipase activity. Heparin was then added to the remaining medium (2 ml) in sufficient quantity to give a final concentration of 20 fig/ml. 10 min after addition of heparin, medium was taken for assay of lipoprotein lipase activity. The plates were immediately drained and rinsed with ice-cold PBS, and the cells were harvested in 1.2 ml of solution E, sonicated briefly at 0 “C, processed into acetone/ether powder, and assayed for lipoprotein lipase activity. DNA content of control cells was 51.8 f 1.1 pg/plate and that of heparin-treated cells was 53.5 + 0.8 pg/plate. The values given are the mean + SD. of 3-4 plat.es.

Lipoprotein lipase activity

Group

Control Heparin

In medium Remaining Released in Initial After 10 min in cells 10 min

microunits/ml milliunits/m,g DNA

110 + 10 200 + 25 501 + 25 1.5 c 1.0 110 f 20 1890 + 220 491 + 9 64.2 + 6.8

A B

FIG. 2. Light micrographs showing lipid droplets and im-

The rate of migration of the lipoprotein lipase band did not change with duration of incubation of the cells (Fig. 5B). ““S- Labeled lipoprotein lipase from unaffected cells migrated as a broad band with M, = 56,000-58,000, suggesting it could consist of several forms of lipoprotein lipase. In contrast, YS- labeled lipoprotein lipase from cld/cld cells migrated as a thinner band with M, = 56,000. The results indicate that each group of cells synthesized the same form(s) of lipoprotein lipase for at least 3 h.

was immunoprecipitated with antiserum to lipoprotein lipase and resolved on SDS-PAGE (Fig. 5, A and B). The samples applied to each lane in Fig. 5A contained the same amount of cell extract whereas those in Fig. 5B contained the same amount of radioactivity. Radioactivity in the band corre- sponding to lipoprotein lipase increased with incubation time, but not linearly, and was about 60% larger in cld/cld cells than in unaffected cells at 2 h (Fig. 4). The ratio of radioactivity in lipoprotein lipase to that in total protein at 1 h was 0.014% in unaffected cells and 0.021% in cld/cld cells. The smaller amount of ““S-labeled lipoprotein lipase in unaffected cells probably resulted from more rapid turnover of newly synthesized lipase in unaffected cells.

munofluorescence of lipoprotein lipase in Golgi region of unaffected brown adipocytes treated with monensin. Cells on Day 2 of confluence were incubated 2 h with 15 pM cycloheximide + 3 pg/ ml heparin in differentiation medium, the medium was changed, and the cells were incubated 90 min with 1 PM monensin (to block transport of lipoprotein lipase from the Golgi) before they were taken for morphological study. Lipoprotein lipase was detected immunocytochemically with a fluorescent double antibody technique, using chicken antisera to bovine lipoprotein lipase and rhodamine-labeled affinity-purified rabbit antibodies against chicken IgG. A, phase micrograph showing numerous brown adipocytes (A) containing mu- tilocular lipid droplets (LD). My, myocte; N, nucleus. Magnification X 235. B, immunofluorescence of lipoprotein lipase in the perinuclear Golgi region of monensin-treated adipocytes, the same cells as those shown in A. Magnification X 235.

presence of tunicamycin.’ The lipoprotein lipase component with M, = 36,000-37,000 was mostly endo H-resistant, but small amounts were digested to products with M, = 35,500 and 34,000-34,500.

A component with M, = 36,000-37,000 was also immunoprecipitated with antiserum to lipoprotein lipase from extracts of unaffected and cld/cld cells (labeled B in Fig. 5A). This suggested that proteolytic cleavage of lipoprotein lipase molecules occurred before or during incubation of the sample with antiserum.

Digestion with Endo H and Endo F of ““S-Labeled Lipopro- tein Lipase from Unaffected and cld/cld Adipocytes-Cells were incubated with [““Slmethionine for 2 h at 37 “C, and “‘S- labeled lipoprotein lipase was immunoprecipitated from extracts of the cells, redissolved in SDS solution, digested 24 h with endo H or endo F, and resolved on SDS-PAGE.

Endo F digestion of lipoprotein lipase from unaffected cells yielded a large amount of product with M, = 51,000-52,000 and a small amount with M, = 53,000-54,000 (Fig. 6). This indicates that the endo H-resistant form noted above, with M, = 57,000-58,000, had complex-type oligosaccharides. The component with M, = 36,000-37,000 was converted by endo F to two products with M, = 35,500 and 34,000-34,500,

Lipoprotein lipase from cld/cld adipocytes, in contrast, was totally endo H-sensitive, yielding a product with M, = 51,000- 52,000 (Fig. 6). The component with M, = 36,000-37,000 from

indicating that this component also contained complex-type

these cells was also totally endo H-sensitive. It is concluded

oligosaccharides.

that the endo H-sensitive lipase in cld/cld cells contained only high mannose-type oligosaccharide chains because the lipase was retained in endoplasmic reticulum.

The studies made with endo H showed that unaffected cells contained at least three forms of lipoprotein lipase (Fig. 6). One form was endo H-resistant with M, = 57,000-58,000, the second was partially endo H-sensitive, yielding a product with M, = 53,000-54,000, and the third was totally endo H-sensitive, yielding a product with M, = 51,000-52,000. The latter product had the same M, as the unglycosylated form of lipoprotein lipase synthesized by brown adipocytes in the

Endo F digestion of lipoprotein lipase from cld/cld adipocytes yielded a single product with M, = 51,000-52,000, as would be expected from the studies with endo H (Fig. 6). The component with M, = 36,000-37,000 derived from cld/cld cells was also totally digested by Endo F to a product with M, = 34,000-34,500.

Time Course of Endo H Digestion of Pulse-labeled Lipopro- tein Lipase from Unaffected and cld/cld Adipocytes-To deter- mine how many N-linked oligosaccharide chains might be attached to lipoprotein lipase in unaffected cells and in cld/ cld cells, ““S-labeled lipoprotein lipase was immunoprecipi-

L H. Masuno, E. J. Blanchette-Mackie, and R. 0. Scow, manuscript in preparation.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


FIG. 3. Light micrographs showing lipid droplets and immunofluorescence of lipoprotein lipase in endoplasmic reticulum of cld/cld brown adipocytes. Cells on Day 2 of confluence were used in this study. Lipoprotein lipase was detected immunocytochemically with a fluorescent double antibody technique, using chicken antisera to bovine lipoprotein lipase and rhodamine-labeled affinity-purified rabbit antibodies against chicken IgG. A, a section of cultured c/d/& cells on Day 2 of confluence showing numerous adipocytes containing multilocular lipid droplets (HI). These cells were fixed with aldehyde and 0~0, before being sectioned. Magnification X 310. I3 and C, immunofluorescence of lipoprotein lipase in cld/cld adipocytes. The reticular pattern of distribution of immunofluorescence indicates that lipoprotein lipase was located in endoplasmic reticulum (ER) in these cells. The thin line of immunofluorescence surrounding nuclei (arrowhead) indicates that lipase was also present in the nuclear envelope. LD, lipid droplet; N, nucleus. Magnification: B, x 620; C, X 310.

tated from extracts of such cells incubated 10 min with [%] methionine, redissolved in SDS solution, digested O-24 h with endo H, and resolved by SDS-PAGE.

““S-Labeled lipoprotein lipase from both unaffected and cld/cld cells was completely digested by endo H within 60 min (Fig. 7). Two digestion products appeared at 20 min, and the product with slower mobility disappeared at 3 h, while the other product was still present at 24 h. The conversion of high mannose-type lipoprotein lipase from each group of cells by endo H to a transient intermediate product and a single final product demonstrates that both groups of cells synthesized lipoprotein lipase that had at least two oligosaccharide chains.

Fate of Pulse-labeled Lipoprotein Lipase in Unaffected and cld/cld Adipocytes-Unaffected and cld/cld cells were incubated 10 min with [““S]methionine and then O-90 min with [““Slmethionine-free medium. 3”S-Labeled lipoprotein lipase was immunoprecipitated from extracts of the cells, redissolved in SDS solution, digested 24 h with endo H, and resolved on SDS-PAGE.

When :%-labeled lipoprotein lipase was chased in unaffected cells, its band on SDS-PAGE became broader and then fainter with time up to 60 min, and was absent at 90 min (Fig. 8). The broadening of the band indicates that lipoprotein lipase was being modified during the chase. Endo H digestion of ““S-labeled lipoprotein lipase collected at different times shows that at 0 min (after a lo-min incubation with [““S] methionine) the lipase molecules were totally endo H-sensitive, at 10 min most were totally endo H-sensitive and some were resistant, and at 30-60 min most were totally endo H-

resistant, some were partially endo H-sensitive, and less were totally endo H-sensitive. These findings demonstrate that newly synthesized lipoprotein lipase in unaffected adipocytes was rapidly glycosylated with high mannose-type oligosaccharides, these oligosaccharides were processed to complex-type units within 30 min, and the lipase was secreted/degraded within 90 min.

“‘S-Labeled lipoprotein lipase in cld/cld cells, however, did not change in either amount or M,, and endo H digestion of lipase collected at different times showed that the lipase molecules retained their high mannose-type oligosaccharides throughout the 90-min chase (Fig. 8). These findings indicate that newly synthesized lipoprotein lipase in cld/cld adipocytes was also rapidly glycosylated with high mannose-type oligosaccharides, but the oligosaccharides remained high mannose- type, and the lipase was retained in the cells for at least 90 min.

Chromatography on Heparin-Sepharose of Lipoprotein Lip- ase Activity of Unaffected Brown Adipocytes-Lipoprotein lipase activity in extracts made with solution F from unaffected cells on Day 3 of confluence was chromatographed on heparin-Sepharose (Fig. 9). About two-thirds of the lipoprotein lipase activity applied to the column was bound by heparin-Sepharose, and one-third of that bound was eluted with 0.6 M NaCl and two-thirds was eluted with 1.5 M NaCl in solution G (Table III). The presence of lipoprotein lipase activity in the void volume probably resulted from overloading the column. The active form of lipoprotein lipase is a dimer (23), and the dimeric form binds with higher affinity to heparin than the inactive monomeric form (24). These find-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Inactive High Mannose-type Lipase in cld/cld Adipocytes

en I----

% in Total Protein

12 c 35S in Lipoprotein Lipase

Hours

FIG. 4. [?S]Methionine in total protein and lipoprotein lipase of unaffected and cld/cZd cells. The culture medium was removed from plates of cells on Day 3 of confluence and replaced with 2 ml of methionine-deficient DMEM and 30 min later 180 PCi of [“Slmethionine was added to each plate. The cells were then incubated for 1 or 2 h, harvested into 0.5 ml of solution D, sonicated briefly at 0 “C, and centrifuged. Aliquots of the infranatants were taken for measurement of [“Slmethionine in total protein and lipoprotein lipase as described under “Experimental Procedures.” The dots and cwcles represent data collected from single plates.

ings indicate that dimeric lipoprotein lipase of mouse brown adipocytes bound to heparin-Sepharose, and some was eluted with 0.6 M NaCl and the rest with 1.5 M NaCl.

““S-Labeled Lipoprotein Lipase in Fractions Isolated by Chromatography on Heparin-Sepharose from Extracts of Un- affected and cld/cld Brown Adipocytes-Unaffected and cld/ cld cells were incubated 2 h with [‘“S]methionine on Day 3 of confluence, and YS-labeled proteins were extracted from the cells with solution F and chromatographed on heparin-Seph- arose (Fig. 10). About 60% of ““S-labeled protein applied to the column was bound by heparin-Sepharose, and 97% of that bound was eluted with 0.6 M NaCl and 3% was eluted with 1.5 M NaCl in solution G (Table III, Fig. 10). The fraction eluted from heparin-Sepharose with 1.5 M NaCl had a ratio of lipoprotein lipase activity to ““S-labeled total protein that was >25 times that in the other fractions. There was no difference in distribution between ““S-labeled total protein of unaffected cells and that of cld/cld cells.

““S-Labeled lipoprotein lipase was immunoprecipitated from the void volume (fractions 3-5), the fraction eluted with 0.6 M NaCl (fractions 35-37), and the fraction eluted with 1.5 M NaCl (fractions 64-66), redissolved in SDS solution, and resolved on SDS-PAGE. The autoradiographs in Fig. 11 demonstrate that :‘“S-labeled lipoprotein lipase from both unaffected and cld/cld adipocytes was collected in the void volume and in fractions eluted with 0.6 M NaCl and 1.5 M NaCl. Most of the ““S-labeled lipoprotein lipase from unaffected adipocytes that bound to heparin-Sepharose was eluted with 1.5 M NaCl, whereas most of that from cld/cld adipocytes was eluted with 0.6 M NaCl. Although the findings in Fig. 9 indicate that dimeric mouse lipoprotein lipase can be eluted from heparin- Sepharose with 0.6 M NaCl, recent studies of the metabolism of bovine lipoprotein lipase by perfused rat liver indicate that inactive/monomeric lipase can bind to heparin-Sepharose and

Unaffected c/d/c/d

A ’ , I I Hours of lncubat,on

30-

- I a

I

-A

-5

B Hours of Incubation

r+lr+l+ u s B

FIN. 5. Radioautographs demonstrating incorporation of [““S]methionine into lipoprotein lipase by unaffected and cld/ cld brown adipocytes. The culture medium was removed from plates of cells on Day 3 of confluence and replaced with 2 ml of methionine deficient DMEM, and 30 min later 180 /*Ci of [%‘S] methionine was added to each plate. The cells were then incubated 1-3 h at 37 “C, harvested into 0.5 ml of solution D, sonicated briefly at 0 “C, and centrifuged. Aliquots of the infranantant were treated with either chicken antiserum to lipoprotein lipase (AS) or chicken nonimmune serum (NS), the immunoprecipitates were dissolved in solution C with heating, and “S-labeled proteins in the precipitates were resolved by PAGE. A, the same volume of cell extract was applied to each lane: A, location of lipoprotein lipase; B, location of degraded lipoprotein lipase. B, the same amount of radioactivity was applied to each lane: M, of lipoprotein lipase in unaffected cells was 56,000-58,000 and that in cld/cld cells was 56,000.

Unaffected cld/cld I I Yl I I LLI

5M9 k- -5759 k -53.54k - 51-52 k

Sk- -51.52 k

36-37 k- -36.37k 36-37 k- -35.5 k -34-34.5 k

FIG. 6. Diaestion with endo H and

-3434.5 k

S-labeled lipoprotein kpase extracted from unaffected and cld/cld brown adipocytes. The culture medium was removed from plates of cells on Day 2 of confluence and replaced with 2 ml of methionine- deficient DMEM, and 30 min later 180 PCi of [““Slmethionine was added to each plate. The cells were then incubated for 2 h at 37 “C, harvested into 0.5 ml of solution D, sonicated briefly at 0 “C, and centrifuged. ““S-Labeled lipoprotein lipase in the infranatant was immunoprecipitated with chicken antiserum to bovine lipoprotein lipase, the immunoprecipitate was dissolved in 30 ~1 of 1% SDS in 0.1 M sodium citrate, pH 5.5, with heating, and aliquots of dissolved lipase were digested 20-24 h at 30 “C with endo H or endo F as described under “Experimental Procedures.” The products of diges- tions were resolved by SDS-PAGE and autoradiographed.

be eluted with 0.5-0.6 M NaCl (25). The presence of “S- labeled lipoprotein lipase of cld/cld cells in fractions eluted with 1.5 M NaCl from heparin-Sepharose indicates that at least some of the lipoprotein lipase synthesized by cld/cld brown adipocytes was dimerized.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


Unaffected Cells

Diaestion Time

IO’ 5 10’ 20’ 40’ lh 3h 6h 24hl

cld/cld Cells

Digestion Time

lo’ 5’ 10’ 20’ 40’ lh 3h 6h 24h 1 - - - cij. _“L -- -- _ - --.-

FIG. 7. Time course of endo H-digestion of lipoprotein lipase extracted from unaffected and cld/cld brown adipocytes. The culture medium was removed from plates of cells on Day 3 of confluence and replaced with 2 ml of methionine-deficient DMEM. 30 min later, 180 PCi of [“‘Slmethionine was added to each plate and the cells were incubated for 10 min at 37 “C. The cells were then harvested into 0.2 ml of solution D, sonicated briefly at 0 “C, and centrifuged. “‘S-Labeled lipoprotein lipase in the infranatant was immunoprecipitated with chicken antiserum to bovine lipoprotein lipase. The immunoprecipitates from 12 plates were combined, dissolved in 100 ~1 of 1% SDS in 0.1 M sodium citrate, pH 5.5, by heating, and digested with endo H for O-24 h at 30 “C, and the products of digestion were resolved by SDS-PAGE and autoradiographed.

Unaffected Cells

Chase Time

Endo H- Digestion

cld/cld Cells

Chase Time

Endo H- ‘A &, & A e; Digestion _

FIG. 8. Chase of [?3]methionine incorporated into lipoprotein lipase by unaffected and cld/cld brown adipocytes. The culture medium was removed from plates of cells on Day 3 of confluence and replaced with 2 ml of methionine-deficient DMEM. 30 min later 180 FCi of [“‘Slmethionine was added to each plate and the cells were incubated (pulsed) for 10 min at 37 “C. The medium was then removed, the plates were washed 2 times with PBS, 2 ml of differen- tiating medium was added, and the cells were incubated (chased) for O-90 min. The cells were harvested into 0.2 ml of solution D, sonicated briefly at 0 “C, and centrifuged. ?S-Labeled hpoprotein lipase in the infranatant was immunoprecipitated with chicken antiserum to bovine lipoprotein lipase. The immunoprecipitates from two plates were combined, dissolved in 30 ~1 of 1% SDS in 0.1 M sodium citrate, pH 5.5, with heating, and digested with endo H for 20-24 h at 30 “C. The products were resolved by SDS-PAGE and autoradiographed.

DISCUSSION

The present study shows that primary culture of brown adipocytes derived from tissue of newborn mice can be used to study genetic factors that regulate synthesis and secretion of lipoprotein lipase. Brown adipocytes cultured from unaffected mice synthesized and secreted active lipoprotein lipase, whereas cells from cld/cld mice synthesized lipoprotein lipase that was normal in size but was neither active nor secreted.

Cells grown from the stromal-vascular fraction of brown adipose tissue of l-day-old unaffected mice replicated, became confluent within 4-5 days, and transformed into brown adipocytes with rapid accumulation of intracellular triacylglyc-

1 I I I

Fraction Number

FIG. 9. Chromatography on heparin-Sepharose of lipoprotein lipase (LPL) activity extracted from unaffected brown adipocytes. Cells on Day 3 of confluence were harvested from two plates with 0.5 ml of solution F, sonicated briefly at 0 “C, and centrifuged. 0.9 ml of the infranatant was applied to a heparin- Sepharose column equilibrated with solution G (column volume = 2 ml). The column was washed with 10 ml of solution G, and lipoprotein lipase activity was sequentially eluted from the column with 10 ml of 0.6 M NaCl in solution G and 10 ml of 1.5 M NaCl in solution G. Fractions of 0.5 ml were collected and assayed for lipoprotein lipase activity.

erol, up to 41 pmol/mg DNA by Day 4 of confluence. These cells contained a small amount of lipoprotein lipase activity at confluence and increased their lipase activity 8-fold during the next 6 days. Similar observations were made in brown adipocytes cultured from the stromal-vascular fraction of brown adipose tissue of 21-day-old mice (16) and from BFC- 1 mouse preadipocytes (26). It was difficult to maintain brown adipocytes in monolayer culture after Day 6 of confluence because fat-laden cells often floated and were lost when the medium was changed.

Brown adipocytes from newborn unaffected mice released O.Oli%/min of their lipoprotein lipase activity spontaneously, and 11% in 10 min when exposed to heparin in ice-cold medium. 3T3-Ll white adipocytes released about O.l%/min spontaneously and 40% in 10 min in response to heparin under similar conditions (11). Because intracellular move- ment of lipase would be inhibited by the cold medium, these findings indicate that 11% of active lipoprotein lipase in brown adipocytes and 40% in 3T3-Ll adipocytes was located at or near the surface of the cells, readily accessible to heparin. Spontaneous release of lipoprotein lipase was undetectable in Ob17 white adipocytes (22), O.l%/min in chick white adipocytes (27), O.S%/min in guinea pig white adipocytes (13), and 1.3%/min in 3T3-F442A white adipocytes (12). Prolonged exposure to heparin greatly increased the release of lipase activity to the medium and decreased the amount degraded intracellularly in Ob17 (22), chick (27), guinea pig (13), and 3T3-F442A (12) adipocytes. Intracellular degradation ac- counts for a major part of turnover of active/newly synthesized lipoprotein lipase in cultured/isolated adipocytes, espe- cially in the absence of heparin (11-13, 22, 27).

Unaffected brown adipocytes contained three forms of glycosylated lipoprotein lipase; the most prevalent was totally endo H-resistant, the next was totally endo H-sensitive, and the least was partially endo H-sensitive. A4, of the endo H- resistant form was 57,000-58,000, and that of the totally endo H-sensitive forms after digestion was 51,000-52,000, the same as that of unglycosylated lipase synthesized in the presence of tunicamycin.’ This difference in M, reflects the finding that lipoprotein lipase of cultured brown adipocytes had two oligosaccharide chains. Similar values for M, of glycosylated and unglycosylated forms of the lipase have been observed in


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


TABLE III Comparison of lipoprotein lipase actiuity with amount of newly synthesized protein in three fractions isolated by

chromatography on heparin-Sepharose from extracts of unaffected cells Data on lipoprotein lipase activity are from studies presented in Fig. 11 and data on incorporation of [““S]

methionine into total nrotein are from studies uresented in Fig. 10.

Fraction

A. Not bound by heparin-Sepharose B. Bound by heparin-Sepharose and eluted

with 0.6 M NaCl C. Bound by heparin-Sepharose and eluted

with 1.5 M NaCl


mi11iunits 3.9 2.6

4.5

Total “S- labeled protein

cpm x 1P 41 61

2


‘“S-Labeled protein

milliunits/cpm X 10’

9.5 4.3

262

FIG 10. Chromatography on heparin-Sepharose of ?S-labeled urotein extracted from unaffected and cldlcld cells. The

Fraction Number

medium was removed from plates of cells on Day 3 of confluence and the plates were washed 2 times with PBS, replenished with 2 ml of methionine-deficient DMEM, and returned to the incubator. 30 min later 180 ~1 of [%]methionine was added to each plate and the cells were incubated for 2 h at 37 “C, harvested into 0.5 ml of solution F, sonicated briefly at 0 “C, and centrifuged. 0.9 ml of the combined infranatants from 2 plates, containing 1.3 X 10’ cpm, was applied to a heparin-Sepharose column equilibrated with solution G (column size = 2 ml). The column was washed with 15 ml of solution G and proteins were eluted sequentially with 15 ml of 0.6 M NaCl in solution G and 10 ml of 1.5 M NaCl in solution G. Fractions of 0.5 ml were collected. 10 ~1 of each fraction was taken for measurement of the amount of [%]methionine incorporated into total protein (precipitated with 10% trichloroacetic acid (TCA)). See Fig. 11 for additional studies of some of these fractions.

Ob17 (22), 3T3-Ll (ll), 3T3-F442A (12), and rat adipocytes (28), and for M, of glycosylated lipoprotein lipase in guinea pig adipocytes (29). The values for M, of unglycosylated lipase agree well with those predicted by nucleotide sequence studies of mouse (30) and guinea pig lipoprotein lipase (31). Endo H digestion studies showed that mature lipoprotein lipase of 3T3-F442A cells also had two oligosaccharide chains (32), whereas that of guinea pig adipocytes had three chains (29).

Studies in unaffected brown adipocytes pulsed 10 min with [“‘S]methionine showed that all of the lipase was totally endo H-sensitive at the end of the pulse, some was endo H-resistant 10 min later, and most was endo H-resistant at 30-60 min after the pulse. Only traces of radioactive lipase were present in the cells at 60 min and none at 90 min. Spontaneous release of “5S-labeled lipase to the medium was not detectable. Im- munolocalization studies did not detect lipoprotein lipase in unaffected cells unless intracellular transport was blocked by

Fraction Number w

x 10-3 4 5 3536 37 6465 66

Fraction Number

FIG. 11. Autoradiographs demonstrating 36S-labeled lipoprotein lipase in specific fractions isolated by chromatography on heparin-Sepharose from extracts of unaffected and cZd/cld brown adipocytes. 200-~1 aliquots of fractions isolated by chromatography on heparin-Sepharose in the experiments described in Fig. 10 were treated with antiserum to lipoprotein lipase, the immunoprecipitates were dissolved in solution C with heating, and ‘%-labeled proteins in the precipitates were resolved by SDS-PAGE.

monensin, reflecting fast turnover of lipoprotein lipase in these cells. Studies in 3T3-F442A adipocytes (32) produced similar results, with core glycosylation within 10 min and processing of lipase to the mature form, with two complex- type chains, within 60 min. 3T3-F442 cells released 20% of the newly synthesized lipase to the medium and the rest was degraded intracellularly. A different pattern of glycosylation and processing of lipoprotein lipase occurred in guinea pig adipocytes (29). Some of the lipase molecules were processed to the mature form, with two complex and one endo H- sensitive type chains, and secreted within 40 min, whereas other lipase molecules remained fully endo H-sensitive for at least 2 h and were not released to the medium.

Our findings in unaffected brown adipocytes support models proposed by Semb and Olivecrona (29) and by Vannier


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


and Ailhaud (32) for glycosylation and processing of lipoprotein lipase in adipocytes. Accordingly, core glycosylation of the lipase occurs in endoplasmic reticulum and high mannose- type oligosaccharides are processed in Golgi to complex chains before the lipase is secreted by the cells. The rapid diversion of lipase by heparin from intracellular degradation to secretion in cultured cells (22, 29, 32) suggests that lipoprotein lipase molecules in transit from Golgi may reside temporarily in transport/secretory vesicles before being secreted or degraded (10, 32).

Cells cultured from brown adipose tissue of l-day-old cld/ cld mice replicated, became confluent, transformed into brown adipocytes, and accumulated triacylglycerol at the same rates as cells of unaffected mice. cld/cld cells had 2-3 times as much lipoprotein lipase protein as unaffected cells, yet contained ~5% as much lipoprotein lipase activity and released no lipase activity to the medium. cld/cld cells readily incorporated [35S] methionine into lipoprotein lipase and total protein. Lipopro- tein lipase accounted for 0.021% of total protein synthesis in cld/cld cells and 0.014% in unaffected cells. Lipoprotein lipase synthesized in cld/cld cells was immediately glycosylated with two high mannose oligosaccharides and at least some was dimerized, but it was retained unprocessed in the cells for at least 90 min. M, of glycosylated lipoprotein lipase of cld/cld cells was 56,000, 2,000 less than that of the endo H-resistant lipase of unaffected cells, whereas M, of endo H-digested cld/ cld lipase was normal, 51,000-52,000. Immunofluorescence (Fig. 3) and electron microscopic immunoperoxidase (33) lo- calization studies demonstrated that lipoprotein lipase was retained in cld/cld adipocytes primarily in the lumen of rough endoplasmic reticulum.

Dimerization of lipoprotein lipase in cld/cld cells probably occurred in endoplasmic reticulum because the lipase was confined to the reticulum in these cells. Dimerization of lipoprotein lipase in guinea pig adipocytes also occurred in endoplasmic reticulum (29). These conclusions agree with reports that most oligomeric proteins are assembled in endoplasmic reticulum (34). In 3T3-F442A adipocytes, glycosylated lipoprotein lipase molecules did not appear to form dimers until after they were processed to endo H-resistant forms, suggesting that dimerization occurred in Golgi in these cells (32).

In summary, unaffected mouse brown adipocytes, 3T3- F442A white adipocytes (12), and guinea pig white adipocytes (29) synthesize lipoprotein lipase which contains endo H- resistant (complex) oligosaccharide chains and is dimeric, active, and secreted. cld/cld brown adipocytes also synthesize fully glycosylated lipoprotein lipase, some of which is dimerized, but it contains only endo H-sensitive-type oligosaccharides and is inactive and not secreted. A form of endo H- sensitive-type lipoprotein lipase which is both active and secreted is synthesized in guinea pig adipocytes treated with deoxymannojirimycin or methyldeoxynojirimycin (29), suggesting that trimming and processing of lipoprotein lipase may not be necessary for activity or secretion of the lipase. Endo H-sensitive glycoproteins are often thought to contain only high mannose oligosaccharide chains (29, 32, 35, 36). However, some endo H-sensitive glycoproteins contain hybrid chains (37-40), which are formed in the medial compartment of Golgi (40-42). The type of oligosaccharide chains present in endo H-sensitive lipoprotein lipase in guinea pig adipocytes treated with glycosidase inhibitors was not determined (29). The lipase synthesized in cld/cld adipocytes is considered to be high mannose-type because the lipase was endo H-sensitive and retained in endoplasmic reticulum. Whether the cld mutation affects primarily transport of lipoprotein lipase from

endoplasmic reticulum to Golgi, trimming and glycosylation of oligosaccharide chains of the lipase, or some other process is to be resolved.

Acknowledgments-We thank Alfred Spaeth, Janice Allen, and David D. Soergel, Jr. for expert technical assistance.

1.

2.

3.

4.

5.

6.

I.

8.

9.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19. 20.

21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

REFERENCES

Paterniti, J. R., Jr., Brown, W. V., Ginsberg, H. N., and Artzt, K. (1983) Science 221, 167-169

Robinson, D. S. (1970) in Comprehensive Biochemistry (Florkin, M., and Stotz, E. H., eds) Vol. 18, pp. 51-116, Elsevier, Am- sterdam

Scow, R. O., Blanchette-Mackie, E. J., and Smith, L. C. (1976) Circ. Res. 39, 149-162

Garfinkel, A. S., and Schotz, M. C. (1987) in Lipoproteins (Gotto, A. M., Jr., ed) pp. 335-357, Elsevier Science Publishers B. V., Amsterdam

Goldberg, I. J., Le, N.-A., Paterniti, J. R., Jr., Ginsberg, H. N., Lindgren, F. T., and Brown, W. V. (1982) J. Clin. Znuest. 70, 1184-1192

Kinnunen, P. K. J. (1984) in Lipuses (Borgstrom, B., and Brock- man, H. L., eds) pp. 307-328, Elsevier Science Publishing Co., Inc., New York

Blanchette-Mackie, E. J., Wetzel, M. G., Chernick, S. S., Pater- niti, J. R., Jr., Brown, W. V., and Scow, R. 0. (1986) Lab. Invest. 55,347-362

Olivecrona, T., Chernick, S. S., Bengtsson-Olivecrona, G., Pater- niti, J. R., Jr., Brown, W. V., and Scow, R. 0. (1985) J. Biol. Chem. 260,2552-2557

Olivecrona, T., Bengtsson-Olivecrona, G., Chernick, S. S., and Scow, R. 0. (1986) Biochim. Biophys. Acta 876, 243-248

Blanchette-Mackie, E. J.. Masuno, H., Dwver, N. K.. Olivecrona, T., and Scow, R. 0. (1989) Am. J. Physibl. i56, E818-E828

Olivecrona. T.. Chernick. S. S.. Ben&son-Olivecrona. G.. Gar- rison, M:, and Scow, R: 0. (1987) ;I. Biol. Chem. 26i, 10748- 10759

Vannier, C., Deslex. S., Pradines-Figueres, A., and Ailhaud, G. (1989) J. kol. Chem. 264, 13199-13205

Semb. H.. and Olivecrona T. (1987) Biochim. Bionhvs. Acta 921. 104-115

_ I

Olivecrona, T., and Ben&son-Olivecrona, G. (1987) in Lipopro- tein Lipase (Bornsztajn, J., ed) pp. 15-58, Evener Publishers, Chicago

Cheng, C. C., and Bennett, D. (1980) Cell 19, 537-543 Forest. C., Doslio, A., Casteilla, L.. Ricquier, D., and Ailhaud, G.

(1987) &xp. cell Res 168, 233%i46 - Snooner. P. M.. Chernick. S. S.. Garrison. M. M.. and Scow. R.

^O. (1979) J. Zkol. Chem.‘254,‘10021-10629 Nilsson-Ehle, P., and Schotz, M. C. (1976) J. Lipid Res. 17,536-

541 Laemmli, U. K. (1970) Nature 227,680-685 Semb, H., and Olivecrona, T. (1986) Biochim. Biophys. Acta 878,

330-337 Hinegardner, R. T. (1971) Anal. Biochem. 39, 197-201 Vannier, C., Amri, E.-Z., Etienne, J., Negrel, R., and Ailhaud, G.

(1985) J. Biol. Chem. 260,4424-4431 Osborne. J. C.. Jr.. Bern&son-Olivecrona. G.. Lee. N. S.. and

Olivecrona, f. (1985) Z%ochemistry 24, 560515611 Bengtsson-Olivecrona, G., and Olivecrona, T. (1985) Biochem. J.

226,409-413 Chajek-Shaul, T., Friedman, G., Ziv, E., Bar-On, H., and Bengts-

son-Olivecrona, G. (1988) Biochim. Biophys. Actu 963, 183- 191

Forest, C., Doglio, A., Ricquier, D., and Ailhaud, G. (1987) Exp. CellRes. 168,218-232

Cupp, M., Bensadoun, A., and Melford, K. (1987) J. Biol. Chem. 262,6383-6388

Ong, J. M., and Kern, P. A. (1989) J. Biol. Chem. 264, 3177- 3182

Semb, H., and Olivecrona, T. (1989) J. Biol. Chem. 264, 4195- 4200

Kirchgessner, T. G., Svenson, K. L., Lusis, A. J., and Schotz, M. C. (1987) J. Biol. Chem. 262, 8463-8466

Enerback, Semb, H., Bengtsson-Olivecrona, G., Carlsson, P., Hermansson, M.-L., Olivecrona, T., and Bjursell, G. (1987) Gene (Am&.) 58, l-12


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


32. Vannier, C., and Ailhaud, G. (1989) J. Biol. Chem. 264, 13206- 38. Arumugham, R. G., and Tanzer, M. L. (1983) J. Biol. Chem. 258, 13216 11883-11889

33. Bianchette-Mackie, E. J., Masuno, H., and Scow, R. 0. (1988) J. 39. Kobata, A. (1984) in B~OZO~Y of Carbohydrates (Ginsberg, V., and CeZZBioZ. 107, 764a

34. Rose, J. K., and Doms, R. W. (1988) Annu. Reu. Cell Biol. 4, 257- Robbins, P. W., eds) Vol. 2, pp. 88-161, John Wiley & Son,

288 New York

35. Cisar, L. A., and Bensadoun, A. (1987) Biochim. Biophys. Acta 40. Hubbard, S. C. (1988) J. Biol. Chem. 263,19303-19317

927,305-314 41. Kornfeld, R., and Kornfeld, S. (1985) Annu. lieu. Biochem. 54,

36. Doolittle, M. H., Wong, H., Davis, R. C., and Schotz, M. C. (1987) 631-664 J. Lipid Res. 28, 1326-1334 42. Elbein, A. D. (1989) in Biology of Carbohydrates (Ginsberg, V.,

37. Tai, T., Yamashita, K., and Kobata, A. (1977) Biochem. Biophys. and Robbins, P. W., eds) Vol. 3, Jai Press Inc., Greenwich, CT, Res. Comma 67,434-441 in press


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

H Masuno, E J Blanchette-Mackie, S S Chernick and R O Scowcultured brown adipocytes of combined lipase-deficient cld/cld mice.

Synthesis of inactive nonsecretable high mannose-type lipoprotein lipase by

1990, 265:1628-1638.J. Biol. Chem.

http://www.jbc.org/content/265/3/1628Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/265/3/1628.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/content/265/3/1628

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;265/3/1628&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/265/3/1628

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=265/3/1628&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/265/3/1628

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/265/3/1628.full.html#ref-list-1

http://www.jbc.org/

synthesis of inactive nonsecretable high mannose-type lipoprotein

Documents