Novel Inhibition of Proteoglycan Synthesis and Exocytosis by ...

2 downloads 0 Views 11MB Size Report
bound 3'S-glycosaminoglycan. These biochemical and morphological studies indicate that DEC treatment of chondrosarcoma chondrocytes alters the transport ...
THE

Val. 260, No. 9,Issue of May 10,pp. 5777-5786,1985 Printed in US.A.

JOURNALOF BIOLOGICAL CHEMISTRY

0 1985 by The American Society of Biological Chemists, Inc.

Novel Inhibition of Proteoglycan Synthesis and Exocytosis by Diethylcarbamazine in the Swarm Rat Chondrocyte* (Received for publication, October 5, 1984)

Richard L. Stevens$, Willis G. Parsons, K. Frank Austen, Ann Hein, and JohnP.Caulfield From the Departmentsof Medicine and Pathology, Harvard Medical School, and the Departmentof Rheumatology and Immunology, Brigham and Women's Hospital, Boston, Massachusetts02115

cell synthesizes and secretes Pretreatment of cultured chondrosarcoma chondro- under serum-free conditions, this cytes at 37 "C for 15 min with 15 mM diethylcarbam- predominantly a M , lo6 species of chondroitin sulfate conazine (DEC)followedby a 60-min pulse with t3'S] taining proteoglycan which is homologous to that in normal sulfate in the presence of DEC resulted in an approxi- bovine, porcine, chick, and human cartilage (1, 14). Because mate 40% inhibition of synthesis anda 75%inhibition of the number of post-translational modifications that occur of secretion of 3'S-proteoglycan. The inhibition was before the chondrocyte proteoglycan is exocytosed and bedose-related and was not due to a decrease in protein cause of the rapid rate of continuous exocytosis of these synthesis. Chondrocytes exposed for 75 min to 15 mM molecules, intracellular proteoglycan intermediates varying in DEC, washed, incubated for 17 h in DEC-free medium, the extentof post-translational modification have been charand thenpulsed with [35S]s~lfate showed no inhibition acterized predominantly by the differential rate of incorpoin the rate of synthesis of proteoglycan or in the per ration of radiolabeled precursors into the secreted molecule. cent of radiolabeled proteoglycans exocytosed into the culture medium, indicating full reversibilityof the in- Pharmacologic agents thatallow transcription and translation hibitory effect. When chondrocytes were incubated for of proteoglycan peptide core and biosynthesis of glycosami75 min with both 1 mM 8-D-xylosideand 15 mM DEC, noglycans but that inhibitexocytosis of the completed molesecretion of 8-D-xyloside-bound 3'S-glycosaminogly- cule are not now available. Diethylcarbamazine (DEC'), an antifilarial agent (15, 16), can was inhibited by more than 70% despite an apthe IgE-Fc-mediated activation-secretion response of proximate &fold increase in intracellular3 6 S - m a ~ r ~inhibits molecules, as compared to cells exposed to 8-D-xyloside monkey and human lung mast cells in tissue fragments (17, alone. Upon removal of DEC, the block in the secretion 18)and mouse bone marrow-derivedmast cells in culture(19), of 8-D-xyloside-bound3'S-glycosaminoglycans was re- as assessed by inhibition of exocytosis of granule markers and versed, although there was a 15-30-min lag in the the arachidonic acid metabolites previously termed slow reinitiation of exocytosis. Light and electron microscopic acting substanceof anaphylaxis andnow known to represent examination of chondrocytes after 75 min of incuba- the sulfidopeptide leukotrienes. DECalso inhibits therelease tion with 15 mM DEC revealed large vacuoles, a dis- of leukotrienes fromcalcium ionophore-activated mouse mastended Golgi apparatus, and a distended endoplasmic tocytoma cells inculture (20) and the generation of slow reticulum which containedelectron dense material. reacting substance of anaphylaxis after the formationof imUpon removal of DEC, the vacuoles disappeared and mune complexes in the peritoneal cavity of rats (21, 22). We distended organelles returned to their normal appear- have now demonstrated thatDEC inhibits proteoglycan synance between 15 and 30 min, coincident with thestart thesis in the cultured chondrosarcoma chondrocyte by disof exocytosis of 35S-proteoglycan and /3-D-xyloside- rupting the vesicular transport of molecules from the endobound 3'S-glycosaminoglycan. These biochemical and plasmic reticulum to theGolgi and from the Golgi to the cell morphological studies indicate that DEC treatment of chondrosarcoma chondrocytes alters the transport of surface. molecules from the endoplasmic reticulum to theGolgi MATERIALSANDMETHODS and the transport of molecules from the Golgi to the cell surface. Dulbecco's modified Eagle's medium with 4.5 g of glucose/liter, penicillin, and streptomycin (Grand Island BiologicalCo., Grand Island, NY); Bes, Hepes, Tes, porcine insulin (25 units/mg), p nitrophenyl-P-D-xyloside,colchicine, monensin, cytochalasin B (Sigma); Zwittergent 3-12 (Calbiochem-Behring);Sephadex G-25 and The Swarm rat chondrosarcoma chondrocyte has been used G-200 and Sepharose CL-PB (Pharmacia, Piscataway, NJ); diethylby a number of investigators as an in vitro and in uiuo model carbamazine citrate (Lederle Laboratories); chondroitinase ABC, 2-

system for studying proteoglycan structure (1-3), synthesis (4-7), packaging (8,9), secretion (10,11), and interaction with The abbreviations used are: DEC, diethylcarbamazine; Bes, N,Nother constitutents of connective tissue(4, 12,13). In culture, bis(2-hydroxyethyl)-2-aminoethanesulfonicacid; ADi-OS, 2-acetamido-2-deoxy-3-O-(~-~-gluco-4-enepyranosyluronic acid) D - g a k -

* This study was supported inpart by Grants AI-22531, AM-00775, tose; ADi-4S, 2-acetamido-2-deoxy-3-O-(~-~-gluco-4-enepyranosylAM-35907,AM-20580,AM-35984, HL-17382, and RR-05669 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. $ T o whom correspondence should be addressed The SeeleyG. Mudd Building, Room608,250LongwoodAvenue, Boston, MA 02115.

uronic acid) 4-O-sUlfO-D-galactose;ADi-GS, 2-acetamido-2-deoxy-3~-(~-D-g~uco-4-enepyranosy~uronic acid) 6-~-sulfo-D-galactose;ADidiSo, 2-acetamido-2-deoxy-3-O-(2-O-sulfo-~-~-gluco-4-enepyranosyluronic acid) 6-O-sulfo-~-galactose;ADi-diSE, 2-acetarnido-2-deoxy3-O-(~-D-gluco-4-enepyranosyluronic acid) 4,6-di-O-sulfo-~-galactose; Hepes, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonicacid HPLC, high performance liquid chromatography; Tes, N-tris[hydroxymethyl]methyl-2-amino-ethanesulfonic acid.

5777

5778

DEC Inhibition of Exocytosis of Proteoglycan

acetamido-2-deoxy-3-O-(~-D-gluco-4-enepyranosyluronic acid) D-galactose (ADi-OS), 2-acetamido-2-deoxy-3-O-(~-~-gluco-4-enepyranosyluronic acid) 4-O-sulfo-~-galactose(ADi-4S), and 2-acetamido-2deoxy-3-0-(~-D-gluco-4-enepyranosyluronic acid) 6-0-sulfo-D-galactose (ADi-6s) (Miles Laboratories, Elkhart, IN); Partisil-10-PAC aminocyano HPLC column (Whatman); [35S]sulfate(4000 Ci/mmol), [3H]glycine (44 Ci/mmol), and ~-3-[3,4,5-~H]leucine (100 Ci/mmol) (New England Nuclear) were obtained as noted. Rat chondrosarcoma proteoglycan carrier was purified as described previously (14). 2Acetamido-2-deoxy-3-O-(2-O-sulfo-~-~-gluco-4-enepyranosyluronic acid) 6-0-sulfo-D-galactose (ADi-diSD) and 2-acetamido-2-deoxy-3O-(~-~-gluco-4-enepyranosyluronic acid) 4,6-di-O-sulfo-D-galactose (ADi-diSE) were prepared from shark and squid chondroitin sulfate, respectively (23, 24). Cell Culture andp5SlSulfate Incorporation upon Exposure to DEC-Swarm rat chondrocytes were isolated by a trypsin-collagenase treatment of the tumor (4) and were cultured at a density of 1.5 X lo6 cells/35-mm culture dish in 1.5ml of Dulbecco'shighglucose modified Eagle's medium containing 0.25% bovine serum albumin, 1 pg/ml porcine insulin, 100 pg/ml streptomycin, 100 units/ml penicillin, 15 mM Hepes, 10 mM Tes, 10 mM Bes, pH 7.2 (DHG+),as described previously (5). DEC citrate was dissolved in DHG+ at a Control A Control B C concentration of 200 mM, neutralized to pH 7.3 with 6 N NaOH, and serially diluted in DHG'. 0-D-Xylosidewas made 200 mM in dimethyl Day 2 Day1 15 mM DEC sulfoxide and diluted in DHG+ to 1 mM. Samples of chondrocytes FIG. 2. [a6S]Sulfate incorporation and secretion of 35S-macwere incubated for 15 min at 37 "C with or without 1 mM exogenous romolecules by chondrocytes after DEC treatment in the presglycosaminoglycan acceptor, p-nitrophenyl-P-D-xyloside(25,26), and either with 2-20 mM DEC or with 15-20 mM sodium citrate as a ence or absence of drug. Control cells were radiolabeled on day 1 and day 2, respectively. Replicate cells were exposed to 15 mM DEC control. At this concentration of p-D-xyloside, approximately 85% of for 15 min and radiolabeled for 60 min in medium containing DEC the total35S-glycosaminoglycansare polymerized onto the exogenous ( A ) ,15 mM DEC for 75 min and radiolabeled in DEC-free medium acceptor, rather than the peptide core (25). Each cell culture was ( B ) ,and 15 mM DEC for 75 min, washed, cultured overnight, and then incubated for 60 min with [35S]sulfate,and the culture medium radiolabeled on day 2 in DEC-free medium ( C ) . 36S-Proteoglycan was removed by aspiration. The cell layers were extracted for 60 min present in the culture medium (hatched region of each bar) and that at room temperature with 1.5 ml of 1%Zwittergent 3-12 in 4 M remaining cell-associated (stippled regionof each bar) were deterguanidine HCl, 0.1 M sodium sulfate, and 0.1 M Tris-HC1, pH 7.5 mined. The height of the bar indicates total [35S]sulfateincorporation, (Extraction Buffer) (14). Carrier proteoglycan, purified from the and theper cent of total 35S-proteoglycanexocytosed into the medium tumor (I), was added to each medium and cell extract sample (1 mg/ is indicated at the top of each bar. Data are based on the means of ml final concentration), and the 35S-macromoleculespresent were four experiments, each conducted in duplicate, except for the data quantitated by Sephadex (2-25 chromatography under dissociative depicted in bar graph B, which were done twice. conditions. Characterization of Chondrocyte Proteoglycan-The hydrodynamic sizes of both the secreted and cell-associated 35S-proteoglycansfrom each culture were estimated by Sepharose CL-2B chromatography. Samples (100 pl) of media and cell extracts were mixed with 200 pg of purified chondrosarcoma proteoglycan monomer, made 4 M in guanidine HCI, and then were applied to and eluted from replicate 18 Sepharose CL-2B columns (0.6 X 100 cm) previously equilibrated with 4 M guanidine HCl, 0.1 M sodium sulfate, 0.1 M Tris-HC1, pH 16 7.5 (14). Glycosaminoglycans were liberated from the proteoglycan 14 -

12

-

10 -

864-

D€C COfVC€#TRRT/ON fmMl

FIG.1. Dose-related effect of DEC on synthesis and secretion of 36S-macromoleculesby chondrocytes. 35S-Proteoglycans exocytosed into themedium (O),those remaining cell-associated (O), and their sum, which indicates total [35S]sulfateincorporation (MI, are plotted as counts/min; the obtained data at each concentration of DEC are expressed as the mean of duplicate cultures with the range of values indicated.

FIG.3. Dose-related effects of DEC on synthesis and secretion of 35S-macromoleculesin chondrocytes cultured with 1 mM b-D-xyloside. 35S-Glycosaminoglycansexocytosed into the methose remaining cell-associated (O), and their sum, which dium ,).( indicates total [35S]sulfateincorporation (M), are plotted as counts/ min; the obtained data at each concentration of DEC are expressed as themean of duplicate cultures, with the range of values indicated.

5779

DEC Inhibitionof Exocytosis of Proteoglycan

88 3

99 a\o

5040-

3020-

01

io-

O J

0-

I

to+

I

1

10-~

30-3

MOLA R

120

TIME lminufesl

I

10-5

trnmni

FIG. 4. Exocytosis of s6S-macromolecules in chondrocytes treated with &D-xyloside with (0)or without (0)DEC followed by a chase in DEC-free medium. The obtained data at each time point are expressed as the mean of duplicate cultures with the range of values indicated. The bur indicates the amount of 35Smacromolecules exocytosed from cells continuously incubated in DEC-containing media. by an overnight treatment with 0.5 M NaOH at 4 "C, and their hydrodynamic sizes were estimated by the method of Wasteson (27) using filtration on Sephadex G-200. The disaccharide composition of the chondroitin sulfate chainswas assessed with [3H]GlcNH~-labeled proteoglycan obtained from cells radiolabeled for 1 h with 33 pCi/ml [3H]GlcNHzand extracted as detailed above. The cell-associated [3H] GlcNH2-labeled material was centrifuged for 48 h under dissociative conditions of 4 M guanidine HCI at the startingdensity of 1.55 g/ml with CsCl (28). The bottom third of each CsCl gradient was dialyzed against 1 M sodium acetate andthen against 0.1 M NH4HC03, lyophilized, and resuspended in enriched Tris buffer containing 0.01 M NaF (29,30), andincubated a t 37 "C for 60 min with chondroitinase ABC (0.001 units/pg of carrier) (29). The chondroitinase digests were made 80% in ethanol by the addition of 4 volumes of 4 "C absolute ethanol (31) and wereheld for 1 h at 4 "C. The samples were centrifuged at 8000 X g for 5 min to separate the chondroitinase ABC-generated disaccharides present in the supernatantfrom precipitable undigested proteincontaminants.These supernatants were dried over N, and the unsaturated disaccharides were chromatographed on a Whatman Partisil-10-PAC aminocyano HPLC column at 1 ml/min with a solvent composed of 65% acetonitrile/methanol (3:1, v/v) and 35% 0.5 M ammonium acetate/acetic acid, pH 5.3 (31). Carrier disaccharides were detected by monitoring optical density at 232 nm, while radiolabeled disaccharides were quantitated by @scintillation counting of 0.5-ml samples of collected fractions. The HPLC column was standardized with unsaturated unsulfated (ADiOS), monosulfated (ADi-4S, ADi-6S), and disulfated ( ADi-diSD, ADidiSE) disaccharides (31). Microscopic Analysis of Chondrocytes-Two million cells in 35-mm culture dishes were fixed by adding 2 mi of room temperature Karnovsky's mixed aldehydes (32) to the culture medium. After 30-60 min, half of the fixation mixture was removed and the cells were scraped from the dish with a Teflon policeman. Approximately 200 MI of the cell suspension was pelleted for 5 min in a Beckman Microfuge B. The pellets were removed from the tubes, post-fixed with Os01 for 1.5 h at 4 "C,and stained in block with uranyl acetate for 2h at room temperature. Dehydration and embedding were routine. Thick sections (0.3 l m ) were cut on glass knives and stained with azure II/methylene blue. Thin sections were stained ongrid with uranyl acetate for 10 min and with lead citrate for 1min and examined in a JEOL lOOC electron microscope. Six culture dishes of @-Dxyloside-treated chondrocytes, 12 dishes of cells treated with 8-Dxyloside and DEC for 15 min to 3.5 h, and 20 dishes treated with pD-xyloside and DEC for 75 min and then cultured for 5-120 rnin in the absence of DEC were examined. RESULTS

Effect ofDEC on the Rate of p5S]Sulfate Incorporation and Exocytosis of 35S-Macromo.!ecules-In a representative experiment, controlchondrocyte cultures incubated for 1h at 37 "C

16

"I

I'

12-

CY JOCHALASIN

-

01

I

I

I

I

IO-^

10"

I

10

pLCg'mJ

MONENSIN

MOLA/? FIG. 5. Effect of colchicine, cytochalasin B, and monensin on synthesis and secretion of s6S-macromolecules from O-Dxyloside-treated cells. The molecules exocytosed (0)and those remaining cell-associated (0)are indicated as counts/min, with their sum (M) being total incorporation. In these representative experiments, proteoglycan synthesis and exocytosis were investigated a t each drug concentration in single cultures for colchicine and cytochalasin B and in duplicate cultures for monensin. For monensintreated cells, the data are expressed as the mean, with the range of values indicated.

with [35S]~ulfate incorporated 6.75 X lo5 cpm into newly synthesized proteoglycan, and 57% of these radiolabeled molecules were exocytosed into the culture medium(Fig. l). Incubation of replicate cultures of chondrocytes with DEC (2-16 mM) 15 min before and during the 1-h radiolabeling period resulted in an inhibition of total [35S]sulfateincorporation intoproteoglycan over the dose range of 4-16 mM DEC with a maximal inhibition of 63% at 16 mM DEC. The amount of 35S-proteoglycansecreted into the culture medium was even more reduced incrementally upon exposure to increasing concentrations of the drug; only 14% of the total 35S-proteoglycans were exocytosedinto the culturemedium of cells treated with 16 mM DEC, and 50% inhibition in exocytosis occurred

5780

DEC Inhibition of Exocytosis of Proteoglycan

FIG. 6. Light micrographs of 0.3thick sections. P-D-Xylosidetreated chondrocytes with large nuclei ( n )and nucleoli (nu)are seen in linear array or as single cells and contain a few small vesicles (a). Cells treated with PD-xyloside and 15 mM DEC for 75 min ( b )or 3.5 h (c) are vacuolated, and larger vacuoles are seen at thelater time point (arrows).When the cells were incubated with P-D-xyloside and DEC for 75 min and then in DEC-free medium for 30 min (d), the large vacuoles seen in b are absent and thecells resemble those in a. Magnification X 700. pm

at a dose of 9 mM DEC. In three separate experiments, the concentrations of DEC that resulted in a 50% inhibition (ID50) in the synthesis of 35S-proteoglycanwere 14, 17, and 22 mM (extrapolated), respectively, while the ID5, values of DEC for inhibition of exocytosis of 35S-proteoglycanwere 9, 9, and 10 mM, respectively. No difference in thedegree of inhibition of proteoglycan exocytosis was obtained if DEC was neutralized with either Na+, K+, or NH: (n = 2) before exposure to the cells. No inhibition of [35S]sulfateincorporation or inhibition of exocytosis of 35S-proteoglycan was observed with cells treated with 15-20 mM sodium citrate (n = 4). Protein synthesis by the chondrocytes that had been exposed to 15 mM DEC for 75 min was similar to that by cells maintained in medium alone as assessed by the incorporation of 158,700 & 3,000 cpm (mean & S.D., n = 3) and 131,900 +: 7,400cpm' (mean & S.D., n = 3) of [3H]glycine,respectively, and 17,700 and 21,000 cpm of [3H]leucine (n = l ) , respectively, into protein. Thus, DEC attenuated stepsin proteoglycan synthesis and/or secretion. The possibility that DEC was preferentially inhibiting secretion rather thansynthesis of proteoglycan was assessed by radiolabeling the chondrocytes for either 5,10,or 30 min with 200 pCi/ml [35S]sulfateand then incubating them with fresh

culture medium containing or lacking 15 mM DEC for 3 h at 37 "C. In a representative experiment done in triplicate of cells radiolabeled for 5, 10, and 30 min, means of 67, 72, and 78%, respectively, of the totalradiolabeled proteoglycans were exocytosed intothe culture medium from the non-DECtreated cells, while only 42, 53, and 60%, respectively, of the newly synthesized proteoglycans were recovered in the culture medium from DEC-treated cells. In order to assess the toxicity of DEC, chondrocytes were incubated at 37 "C for 15 min with or without 15 mM DEC, washed twice with 2 ml of DHG+, and radiolabeled in the presence or absence of DEC. No decrease in the rate of [35S] sulfate incorporation or the degree of exocytosis of radiolabeled proteoglycan was observed compared to non-DECtreated cells when cells were exposed to DEC for 15 min, washed, and radiolabeled for 60 min in DEC-free medium (n = 4) (data not shown). Proteoglycan synthesis was inhibited 42 & 8%(mean k S.D.) in chondrocytes incubated for 75 min with15 mM DEC, washed, and radiolabeled in DEC-free medium, but theper cent of 35S-proteoglycanbeing exocytosed was similar to thatof day 1and day 2 non-DEC-treated cells (Fig. 2). Cells exposed to 15 mM DEC for 15 min and then radiolabeled for 60 min in the presence of DEC were also 42

Inhibition DEC

of Exocytosis of Proteoglycan

5781

FIG. 7. Electron micrographof a 8-D-xyloside-treated chondrosarcoma chondrocyte. The endoplasmic reticulum (er) is seen on the right, and thecisternae are parallel. The Golgi apparatus has cis and trans faces. The vesicles (v) on the trans face contain electron dense material. Many of the lysosomes (ly) contain fragments of membrane. rn, mitochondrion; cu, coated vesicle. Magnification X 26,000.

FIG. 8. Atypical mitochondria from P-D-xyloside-treated chondrocytes. On the left, crystalline patterns are present in two mitochondria (arrows). On the right, the cristae are arranged concentrically. Magnification X 62,000 (left) and 48,000 (right).

proteoglycan exocytosed into theculture medium weresimilar to those values for non-DEC-treated cells (Fig. 2), indicating the per cent of molecules being exocytosed. The value for the full reversibility of the effect of DEC on incorporation and per cent of exocytosed 35S-macromoleculeswas reduced from secretion of 35S-macromolecules. 45 f 5% (mean f S.D.) to 16 f 4% (mean f S.D.) as compared Effect of DEC on p5S]Sulfate Incorporation and Secretion to cells treated with 15 mM DEC for 75 min and thenwashed of 35S-Macromoleculesin P-D-Xyloside-treated Chondrocytesand radiolabeled in DEC-free medium. When chondrocytes In a representative experiment, incubation of chondrocytes treated with DEC for 75 min were washed and cultured with the glycosaminoglycan acceptor 0-D-xyloside resulted in overnight in DEC-free medium prior to radiolabeling, both an increase in the incorporation of total [35S]sulfate into the rate of [35S]sulfateincorporation and the per cent of 35S- macromolecules from 6.75 x lo5 cpm (Fig. 1) to 1.81 X lo6 f 9% (mean f S.D.) inhibited in proteoglycan synthesis but, in addition, exhibited a 64 +: 9% (mean & S.D.) decrease in

5782

DEC Inhibition of Exocytosis of Proteoglycan

FIG. 9. Chondrocytes treated with 0-D-xyloside and 15 mM DEC for 35 min. Note the vacuoles (ua) that are present near the Golgi apparatus (g).The endoplasmic reticulum (er)appears normal. Magnification X 9600.

cpm (Fig. 3) and an increase in the per cent of exocytosed molecules from 57% (Fig. 1) to 79% (Fig. 3) as compared to non-xyloside-treated cells which were radiolabeled for the same time period. This observation reflects the capacity of the cells to synthesize chondroitin sulfateglycosaminoglycans onto both theexogenous glycosaminoglycan acceptor and the endogenous peptide core and tomore efficiently exocytose the smaller P-D-xyloside-bound chondroitin sulfate (25). Exposure of the cells to P-D-xyloside and increments of DEC resulted in a dose-related decrease in the rateof [35S]sulfate incorporation into macromolecules and a comparably greater decrease in exocytosis such that anincrease incell-associated macromolecules occurred at concentrations of DEC greater than 6 mM (Fig. 3). Inthreeseparate experimentswith chondrocytes treated with 1mM P-D-xyloside and incremental doses of DEC, the IDSovalues for proteoglycan/glycosaminoglycan synthesis were 21, 21, and 34 mM DEC (extrapolated), respectively, while the ID50values for proteoglycan/glycosaminoglycan secretion were 9,9, and10 mM. The 50% effective concentration (EC50)of DEC for increasing cell-associated radiolabeled molecules was 9 f 1 mM (mean f S.D., n = 3); with a 3-fold increase occurring in cells treated with 14-16 mM DEC (Fig. 3). Thus, DEC preferentially inhibited secretion of proteoglycan and P-D-xyloside-boundglycosaminoglycan while allowing polymerization of glycosaminoglycan. The kinetics of the reversibility of the DEC inhibition of exocytosis was examined in cells treated for 75 min a t 37 "C with medium containing 1mM 6-D-xylosidewith and without 15 mM DEC and radiolabeled for the last 60 min of the

FIG. 10. Chondrocyte treated with 0-D-xyloside and 15 mM DEC for 75 min. The vacuoles (ua) are larger than those seen in Fig. 9, and both the endoplasmic reticulum (er) and tram face of the Golgi apparatus ( t )are dilated. The endoplasmic reticulum contains electron dense material, but the Golgi is lucent. Magnification x 11,000.

incubation. After two rapid washes of the radiolabeled monolayers at room temperature, the chondrocyte cultures were incubated in DEC-free medium andthe amounts of 35Smacromolecules exocytosed during the following 1-120 min were determined. In a representative experiment, non-DECtreated cells rapidly secreted intracellular 35S-macromolecules into the chase medium (Fig. 4). In contrast, after removal of the DEC from the chondrocytes, there was a 15-min delay in onset of exocytosis of 35S-macromoleculesand the rate of exocytosis was retarded (Fig. 4). For three separate experiments, the exocytosis time for one-half of the preformed 35Smacromolecules (tnh)from non-DEC-treated cells was 8 f 3 min (mean f S.D.), whereas DEC-treated cells exhibited a 15 f 5-min delay in the onset of exocytosis and a subsequent tu exocytosis time of 100 f 20 min (mean f S.D.). Characterization of the Proteoglycans Synthesized by DECtreated Cells-The 35S-proteoglycans produced by DECtreated cells were of similar hydrodynamic size to those produced by non-DEC-treated cells, as indicated by mean KeV values of 0.27 and 0.29 (n = 4) on Sepharose CL-2B, respectively, for cell-associated proteoglycan and 0.27 for proteoglycan exocytosed in the control culture. The M , of the chondroitin [35S]sulfateside chains as assessed by Sephadex G200 chromatography was 25,000 in both control and DECtreated cells. As assessed by HPLC analysis of chondroitinase ABC digests of density gradient-purified [3H]GlcNH2-labeled proteoglycan (n = 3), 81 f 2% (mean f S.D.) and 19 & 2% (mean f S.D.) of the radiolabeled disaccharides were ADi-4S and ADi-OS, respectively, in the control culture, and 66 f 3% and 34 f 3% (mean f S.D.) of the radiolabeled disaccharides were ADi-4S and ADi-OS, respectively, in the DEC-treated cells.

DEC Inhibition of Exocytosis of Proteoglycan

5783

FIG. 11. Chondrocyte treated for 3.5 h with P-D-xyloside and 15 mM DEC. Note that the trans face of the Golgi is dilated but the cis face is relatively normal. Thevacuoles (ua) contain scant amountsof electron dense material (arrows). Magnification X 25,000.

FIG. 12. Endoplasmic reticulum from a cell treated for 75 min with 8-D-xyloside and 15 mM DEC. Note the dilated cisternaefilled with electron dense material (arrows).Magnification X 24,000.

Effect of Colchicine, Cytochalasin B, and Monensin on Synthesis and Exocytosis of Proteoglycan from the Swarm Rat Chondrosarcoma Chondrocyte-Because colchicine, cytochalasin B, and monensin have each been reported td inhibit synthesis and/or exocytosis of chondrocyte proteoglycan, samples of P-D-xyloside-treated cells were separately exposed toto M colchicine, loF3to 10 pg/ml cytochalasin B, and IO-' to M monensin for 75 min at 37 "C, radiolabeled for the last 60 min of this period, washed, and analyzed for [35S]sulfateincorporation and exocytosis of 35S-glycosaminoglycans within 60 min. At the highest concentrations examined, colchicine and cytochalasin B had only minimal effects

on [35S]sulfateincorporation and exocytosis of 35S-macromolecules in this representative experiment (Fig. 5) and in a separate study. In contrast, in a separate experiment done in duplicate, exposure to lo-' to 1O"j M monensin resulted in a substantial dose-related decrease in total [35S]sulfateincorporation, cell-associated 35S-macromolecules,and theper cent of 35S-macromoleculesthat were exocytosed. At 1O"j M monensin, [35S]~ulfate incorporation was inhibited by 80% with only 2% of the total 35S-macromoleculesbeing exocytosed (Fig. 5). In three separate experiments with chondrocytes exposed to 1O"j M monensin for 75 min, synthesis of 35smacromolecules was inhibited 80, 93, and 87%, with only 2,

5784

Inhibition DEC

of Exocytosis of Proteoglycan

FIG. 13. Golgi apparatus (g)from a cell incubated for 75 min in 8-D-xyloside and 15 mM DEC and then in DEC-free medium for 30 min. The cisternae are flattened, and the structure resembles that seen in untreated cells (see Fig. 7). Many coated (cu) and uncoated ( u ) vesicles are present. cp, coated pit. Magnification X 24,000.

0.2, and 0%, respectively, of the newly synthesized 35S-mac- removal, the cells contained a vacuole population similar to romolecules being exocytosed. No intracellular accumulation that seen before DEC removal. However, by 30 min, the of P-D-xyloside-boundchondroitin sulfatewas observed in the vacuoles were absent from the cytoplasm so that by light microscopy the cells were indistinguishable from cells that monensin-treated cultures. Morphology of Chondrosarcoma Chondrocytes-The mor- were never exposed to DEC (Fig. 6d). The dilated cisternae phology of (3-D-xyloside-treated Swarm rat chondrosarcoma of the trans face of the Golgi apparatusandthe dilated chondrocytes was the same as that reported by others for endoplasmic reticulum also returned to theirnormal flattened non-P-D-xyloside-treated cells (33, 34) and will be briefly appearance within 30 min to 1h (Fig. 13). Fusion of the large described here. Cells scraped from the dishes were arranged vacuoles with the plasma membrane was not observed in cells in linear groups or singly and were often flattened on one examined a t 5-min intervals between 5 and 30 min after DEC side, measuring approximately 22 & 7 pm in length and 12 f removal. However, coated pits were seen on the vacuole mem5 pm in height (n = 37). The cells contained a single nucleus, branes of the recovering cells (Fig. 13). In addition, small 20a prominent nucleolus (Fig. 6a), and a rough endoplasmic 50-nm vesicles and coated vesicles were observed in the trans reticulum composed of flattened cisternae that were often Golgi area of the recovering cells (Fig. 13). arranged in parallel and contained electron dense material DISCUSSION (Fig. 7). The Golgi apparatus consisted of four to six parallel cisternae withthe cis face having fewer vesicles than the trans Incubation of Swarm rat chondrosarcoma chondrocytes in face (Fig. 7). These lattervesicles ranged in size from approx- vitro for 75 min with DEC resulted in a substantial doseimately 20 to 50 nm and were partially filled with electron dependent decrease in the rateof synthesis of 35S-proteoglydense material (Fig. 7). Coated vesicles were present on the can and an even greater relative decrease in secretion of trans Golgi face (Fig. 7), and both coated pits and vesicles radiolabeled proteoglycan (Fig. l), as assessed by mean were seen at the plasma membrane. The mitochondria occa- values of 18 and 9 mM DEC, respectively. The complete sionally containedcrystallinestructures, assumed toroid reversal of the DEC-induced inhibitory effects upon removal shapes, and had cristae arranged concentrically (Fig. 8). The of the drug (Fig. 2) indicated that theviability of DEC-treated lysosomes oftencontainedmembranefragments (Fig. 7). cells was not reduced. The failure of DEC to reduce overall Other cytoplasmic organelles such as tubules and filaments protein synthesis as assessed by either [3H]leucine or [3H] were unremarkable. glycine incorporation into macromolecules suggested an acAfter 15-35 min of incubation in DEC, dilated vacuoles up tion which preferentially inhibited a post-translational step to 1pm in diameter were seen near the Golgi apparatus (Fig. involved in proteoglycan synthesis and/or secretion. p-D9). The number and size of these vacuoles increased with time Xyloside-treated chondrocytes were therefore used to examine of incubation in DEC so that at 75 min most cells had 2-3- whether DEC inhibited proteoglycan synthesis and secretion pm vacuoles scattered through the cytoplasm (Fig. 6b). These by disrupting the Golgi, the intracellular site where glycosylcells were 26 & 10 pm long and 13 f 5 pm high (n = 36) at and sulfotransferases participate in glycosaminoglycan biothis time and thus were similar in size to non-DEC-treated synthesis. Upon exposure of cells to P-D-xyloside and increcells. At 3.5 h, very large (-5-9 pm) vacuoles were seen in mental concentrations of DEC, the rate of synthesis of 35Smany cells (Fig. 6c). The vacuoles contained scant amounts glycosaminoglycansonto theexogenous acceptor was reduced, of electron dense material a t all time points (Figs. 9-11). The and the rateof secretion of these p-D-xyloside-bound glycosGolgi cisternae on the trans face were dilated and electron aminoglycan chains was decreased to a relatively greater lucent (Figs. 10 and 11). Furthermore, the cisternae of the extent (Fig. 3). The mean IDSofor inhibition of synthesis and endoplasmic reticulum were dilated, particularly at the later secretion of 35S-glycosaminoglycanoccurred at 25 and 9 mM, time points, and filled with electron dense material (Figs. 10 respectively. However, since /3-D-xyloside-treated cells accuand 12). mulated 35S-glycosaminoglycansintracellularly inresponse to Cells treated with DEC for 75 min, washed, andthen DEC (Fig. 3), the siteof glycosaminoglycan synthesis within incubated in DEC-free medium for 30 min were 24 f 9 pm the Golgiwas not totallydisrupted.Furthermore, as cells long and 10 & 3 pm high (n = 20). Fifteen minutes after DEC treated with DEC alonedid not accumulate 35S-proteoglycan

DEC Inhibition of Exocytosis of Proteoglycan

5785

intracellularly (Fig. l),it seemed likely that newly translated between the Golgi apparatus and theplasma membrane, as is proteoglycan peptide core was also not reaching the Golgi. the case in other cells of this type (41, 42). A second vesicle Thus, DEC seemed to be altering vesicular transport of pro- shuttle system exists between the endoplasmic reticulum and teoglycan peptide core from the endoplasmic reticulum to the the Golgi (42). The morphological lesions produced by DEC Golgi and vesicular transport of proteoglycan peptide core indicate damage to these two shuttle systems. In the case of containing chondroitin sulfate glycosaminoglycans from the the endoplasmic reticulum, protein is synthesized but cannot enter the Golgi; hence, the cisternae become dilated with Golgi to theplasma membrane. DEC preferentially inhibited the secretion rather than the electron dense material (Fig. 12). The cisternae of the trans synthesis of proteoglycan, as demonstrated by the fact that face of the Golgi apparatus dilate more rapidly than the cells radiolabeled for 5 min with [35S]sulfatein the absence of endoplasmic reticulum, and large vacuoles form throughout DEC exocytosed only 42% of the preformed radiolabeled the cell (Figs. 6 and 9-11). However, the dilated Golgi and proteoglycan into the culture medium after 3 h of incubation vacuoles contain littleelectron dense material, and since proteins are not reaching these compartments they presumin the presence of DEC, as compared to 67% of the total proteoglycan after incubation in DEC-free medium. Since the ably contain DEC, salt, sugars, or ions that raise the osmotic pressure. The vacuole membranes arepresumably drawn from tlh for exocytosis of 35S-proteoglycan from the Golgi to the surrounding extracellular matrix and from the matrix to the the vesicle system that shuttles between the Golgi and the culture medium for these cells has been reported to be ap- plasma membrane. Vacuole formation could cause a decrease proximately 7 and 60 min, respectively, at 37 "C (4, lo), the in the plasma membrane surface area, although the overall ability of DEC to alter exocytosis of preformed proteoglycan cell dimensions remain constant. Furthermore, vesicles must indicates that proteoglycan transport is altered within 8 min be fusing with the vacuoles or vacuoles must fuse with one of exposure to DEC. Kinetic experiments with p-D-xyloside- another, since the vacuole diameter increases throughout the treated cells revealed that upon removal of DEC, exocytosis entire time that thedrug is present. The removal of DEC appears to re-establish the vesicle of preformed P-D-xyloside-bound chains resumed after a 15min lag, although the rateof exocytosis was noticeably slower shuttle system because the vacuoles do not fuse with the than that for non-DEC-treated cells (Fig. 4). If chondrocytes plasma membranebut gradually disappear between 15 and 30 were incubated in DEC-containingmedium, washed, and then min after drug removal (Fig. 64.This period corresponds to pulsed with [35S]sulfate in DEC-free medium, proteoglycan the initiation of exocytosis of biosynthetically labeled molesynthesis remained inhibited, but not the per cent of 35S- cules (Fig. 4). Coated vesicles may be involved in the disasproteoglycan being exocytosed from the cells (Fig. 2). Thus, sembly of the vacuoles, since numerous coated pits are seen the vesicular transport of peptide core to the siteof glycosy- on the vacuole membranes during the recovery period (Fig. lation in the Golgi appears to recover more slowly than the 13). Furthermore, coated vesicles are seen at the Golgi and post-Golgi vesicular transport of completed proteoglycan to plasma membrane of normal cells (Fig. 7), which may indicate the surface of the cell. The [35S]sulfate-labeled and t3H] a role in normal post-Golgi transport for these structures. GlcNH,-labeled proteoglycans produced by the control and REFERENCES 15 mM DEC-treated cells in which suppression of total 35Sincorporation and secretion of 35S-macromoleculeswere in1. Oegema, T. R., Jr., Hascall, V. C., and Dziewiatkowski, D. D. (1975) J . Biol. Chem. 250,6151-6159 hibited 61 -+ 2% (mean f S.D., n = 3) and 71 k 1%(mean & 2. Lohmander, L. S.,De Luca,S., Nilsson, B., Hascall, V. C., Caputo, S.D., n = 3) revealed similar sized proteoglycans possessing C. B., Kimura, J. H., and Heinegkrd, D. (1980) J . Biol. Chem. similar sized chondroitin sulfate side chains. The only differ255,6084-6091 ence noted was that the glycosaminoglycan chains of the 3. Oegema, T. R., Jr., Kraft, E. L., Jourdian, G. W., and vanValen, proteoglycans from the DEC-treatedcultures were somewhat T. R. (1984) J. Biol. Chem. 259, 1720-1726 undersulfatedrelative to those from the controlcultures. 4. Kimura, J . H., Hardingham, T. E., Hascall, V. C., and Solush, M. (1979) J. Biol. Chem. 254,2600-2609 These results again suggest that DEC does not substantially 5. Stevens, R. L., Nissley, S. P., Kimura, J. H., Rechler, M. M., disrupt the region of the Golgi where chondroitin sulfate is Caplan, A. I., and Hascall, V. C. (1981) J. Biol. Chem. 256, synthesized, but rather altersvesicular transport. 2045-2052 The pharmacologic agents colchicine and cytochalasin B, 6. Thonar, E. J.-M. A., Lohmander, L. S., Kimura, J. H., Fellini, S. which have been reported to alter synthesis and exocytosis of A., Yanagishita, M., and Hascall, V. C. (1983) J. Biol. Chem. proteoglycan in chondrocytes (35, 36) by disrupting microtu258,11564-1 1570 7. Mitchell, D., and Hardingham, T. (1982) Biochem. J . 202, 387bules (37) and microfilaments (38), respectively, had only 395 minimal effects on synthesis and exocytosis of p-D-xyloside8. Kimura, J. H., Thonar, E. J.-M., Hascall, V. C., Reiner, A., and bound 35S-glycosaminoglycans.Monensin,acationic ionoPoole, A. R. (1981) J. Biol. Chem. 256, 7890-7897 phore that inhibits late biosynthetic events in the Golgi (39), 9. Mitchell, D., and Hardingham, T., (1982) Biochem. J . 202, 249inhibited exocytosis without intracellular accumulation of 6254 D-xyloside-bound chondroitin sulfate (Fig. 5). Thus, monen- 10. Stevens, R. L., Schwartz, L. B., Austen, K. F., Lohmander, L. S., and Kimura, J . H. (1982) J. Bid. Chem. 257,5745-5750 sin disrupts the /3-D-xyloside-inducedglycosaminoglycan polymerization function of the Golgi complex, a finding in 11. Mitchell, D., and Hardingham, T. (1981) Biochem. J . 196, 521529 agreement with reports by other investigators (9, 40). In 12. Caterson, B., and Baker, J. R. (1979) J . Biol. Chem. 254, 2394contrast, DEC inhibited secretionbut induced an intracellular 2399 a nMason, R. M., Kimura, J. H., and Hascall, V. C. (1982) J . Biol. accumulation of p-D-xyloside-bound 3 5 S - g l y ~ ~ ~ a m i n ~ g l y c13. (Fig. 3), indicating a site of action distinct and separate from Chem. 257, 2236-2245 14. Stevens, R. L., and Hascall, V. C. (1981) J . Biol. Chem. 256, that of monensin. 2053-2058 The Swarm rat chondrosarcoma chondrocyte is a continuously secreting cell with a characteristically prominent Golgi 15. Santiago-Stevenson, D., Oliver-Gonzalez, J., and Hewitt, R. I. (1948) Ann. N . Y. Acad. Sci. 5 0 , 161-170 and endoplasmic reticulum but no large secretory granules 16. Hawking, F. (1979) Adu. Pharrnacol. Chemother. 16, 129-193 (33, 34). Instead, transport of secretory products is presum- 17. Orange, R. p., Austen, W. G., and Austen, K. F. (1971) J . Exp. ably accomplished by a system of small vesicles that shuttle Med. 134, 136-148

5786

Inhibition DEC

of Exocytosis of Proteoglycan

18. Ishizaka, T., Ishizaka, K., Orange, R. P., and Austen, K. F. (1971) J.Immunol. 106, 1267-1273 19. Razin, E., Romeo, L. C., Krilis, S., Liu, F.-T., Lewis, R. A., Corey, E. J., and Austen, K. F. (1984) J. Immunol. 133,938-945 20. Mathews, W. R., and Murphy, R. C. (1982) Biochem. Phurmacol. 31,2129-2132 21. Orange, R. P., Valentine, M. D., and Austen, K. F. (1968) Proc. SOC.Exp. Biol. Med. 127, 127-132 22. Orange, R. P., and Austen, K. F. (1968) Proc. SOC.Exp. Biol. Med. 129,836-841 23. Kawai, Y., Seno, N., and Anno, S. (1966) J. Biochem. (Tokyo) 60,317-321 24. Suzuki. S.. Saito. H.. Yamagata. T.. Anno. K.. Seno. N.. Kawai. Y.,and Furuhashi,’T. (19E8) J. Biol. Chem. 243, 1543-1550 ’ 25. Schwartz, N. B. (1977) J. Biol. Chem. 252, 6316-6321 26. Okayama, M., Kimata, K., and Suzuki, S. (1973) J. Biochem. (Tokyo) 74, 1069-1073 27. Wasteson, A. (1971) J. Chromatogr. 59, 87-97 28. Hascall, V. C., and Sajdera, S. W. (1969) J. Biol. Chem. 244, 2384-2396 29. Saito, H., Yamagata, T., and Suzuki, S. (1968) J. Biol. Chem. 243, 1536-1542

30. Handley, C. J., and Lowther, D. A. (1979) Biochim. Biophys. Acta 582,234-245 31. Seldin, D. C., Seno, N., Austen, K. F., and Stevens, R. L. (1984) Anal. Biochem. 141, 291-300 32. Karnovsky, M. J. (1965) J. Cell Biol. 27, 137a 33. Hascall, G . K. (1980) Anat. Rec. 198, 135-146 34. Hascall, G . K., and Kimura, J. H. (1981) Anat. Rec. 200, 287292 35. Lohmander, S., Madsen, K., and Hinek, A. (1979) Arch. Biochem. Biophy~.192, 148-157 36. Madsen, K., Holmstrom, S., and Ostrowski, K. (1983) Exp. Cell Res. 148,493-501 37. Lacy, P. E., Howell, S. L., Young, D. A., and Fink, J. (1968) Nature (Lond.) 219, 1177-1179 38. Sanger, J. W., and Holtzer, H. (1972) Proc. Natl. Acud. Sci. U. S. A. 69, 253-257 39. Tartakoff, A,, Vassalli, P., and DBtraz, M. (1978) J. Cell Biol. 79, 694-707 40. Nishimoto, S. K., Kajiwara, T., and Tanzer, M. L. (1982) J.Biol. Chem. 257,10558-10561 41. Farquhar, M. G., and Palade, G . E. (1981) J. Cell Biol. 91, 77S103s 42. Palade, G . E. (1975) Science (Wash. D.C.) 189,347-358