HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 39, 36272-36279, September 27, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/39/36272    most recent M206365200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fjeldstad, K. Articles by Prydz, K. Search for Related Content PubMed PubMed Citation Articles by Fjeldstad, K. Articles by Prydz, K. Social Bookmarking                      What's this?

Sulfation in the Golgi Lumen of Madin-Darby Canine Kidney Cells Is Inhibited by Brefeldin A and Depends on a Factor Present in the Cytoplasm and on Golgi Membranes^*

Katja Fjeldstad, Mona E. Pedersen^§, Tram Thu Vuong, Svein Olav Kolset^¶, Line Mari Nordstrand, and Kristian Prydz^**

From the Department of Biochemistry and ^¶ Institute for Nutrition Research, University of Oslo, Oslo 0316, Norway

Received for publication, June 26, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Madin-Darby canine kidney cells are more resistant than most other cell types to the classical effects of brefeldin A (BFA) treatment, the induction of retrograde transport of Golgi cisternae components to the endoplasmic reticulum. Here we show that sulfation of heparan sulfate proteoglycans (HSPGs), chondroitin sulfate proteoglycans (CSPGs), and proteins in the Golgi apparatus is dramatically reduced by low concentrations of BFA in which Golgi morphology is unaffected and secretion still takes place. BFA treatment seems to reduce sulfation by inhibition of the uptake of adenosine 3'-phosphate 5'-phosphosulfate (PAPS) into the Golgi lumen, and the inhibitory effect of BFA was similar for HSPGs, CSPGs, and proteins. This was different from the effect of chlorate, a well known inhibitor of PAPS synthesis in the cytoplasm. Low concentrations of chlorate (2-5 mM) inhibited sulfation of CSPGs and proteins only, whereas higher concentrations (15-30 mM) were required to inhibit sulfation of HSPGs. Golgi fractions pretreated with BFA had a reduced capacity for the synthesis of glycosaminoglycans (GAGs), but control level capacity could be restored by the addition of cytosol from various sources. This indicates that the PAPS pathway to the Golgi lumen depends on a BFA-sensitive factor that is present both on Golgi membranes and in the cytoplasm.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Sulfation is a frequent Golgi modification found in glycoproteins (1), glycolipids (2), and proteoglycans (3). Intracellular sulfate is provided by uptake via sulfate transporters in the plasma membrane (4, 5). In the cytoplasm, sulfate is enzymatically activated to adenosine 3'-phosphate 5'-phosphosulfate (PAPS),¹ which is translocated through specific transporter molecules (6-8) in the Golgi membrane and utilized as sulfate donor in the Golgi lumen by sulfotransferases (9). Proteoglycans (PGs) are sulfated at various positions along their long linear glycosaminoglycan (GAG) chains that are attached to serines. Reduced (5) or impaired (10) sulfate uptake across the plasma membrane is reported to be the primary defect in patients with mutations in the diastrophic dysplasia (DTD) gene. The resulting reduction in sulfation of PGs in cartilage matrix is the cause of the main clinical features connected with DTD (5). Mutations in the same gene may give complete impairment of sulfate uptake, resulting in achondrogenisis type IB and perinatal death (10). No genetic disorder has been connected with later steps in the sulfation pathway, but the molecular basis is still unknown in the majority of osteochondrodysplasias.

Localization of enzymatic activities to different regions of the Golgi apparatus has been facilitated by use of the fungal isoprenoid metabolite brefeldin A (BFA). In most cell types, secretory transport is inhibited by BFA because treatment with this drug induces retrograde transport of components of the cis-, medial, and trans-Golgi cisternae but not of the trans-Golgi network to the endoplasmic reticulum (11-14). In several cell types, the synthesis of HSPGs may be completed in the presence of BFA although with significantly reduced efficiency, whereas CSPG synthesis is not detected, which indicates that separate enzyme systems in early and late subcompartments of the Golgi complex are involved in HSPG and CSPG synthesis, respectively (15-19).

Two kidney epithelial cell lines, MDCK and PtK, have Golgi stacks that are morphologically resistant to BFA treatment (14, 20, 21). Despite this fact, apical protein secretion is reduced in MDCK cells already at low BFA concentrations, whereas basolateral secretion initially compensates for the reduction in apical secretion and is not inhibited until 20-fold higher BFA concentrations are applied (22, 23).

Here we report a novel effect of BFA in MDCK cells. At low concentrations of BFA in which the Golgi apparatus is morphologically intact, the sulfation of proteins and PGs was dramatically reduced. This effect seemed to be caused by a reduction in the uptake of PAPS into the lumen of the Golgi apparatus. Lowering the cytoplasmic PAPS level by chlorate also resulted in reduced PAPS levels in the Golgi lumen but gave a different reduction profile in the incorporation of sulfate into CSPG, HSPG, and proteins. The reconstitution in vitro of glycosaminoglycan (GAG) chain synthesis by incubating MDCK Golgi fractions with UDP sugars, [³⁵S]PAPS, Mg²⁺, and Mn²⁺ was impaired when the Golgi fractions were pretreated with BFA. This impairment was overcome by the addition of cytosol from pig brain, rat liver, or MDCK cells. Our results indicate that sulfation within the Golgi depends on a factor present both in the cytoplasm and on Golgi membranes.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Bovine serum albumin, N-ethylmaleimide, -aminocaproic acid, phenylmethylsulfonyl fluoride, 2,4-dinitrophenylalanine, Dextran Blue, Triton X-100, and UDP sugars were all from Sigma. Chondroitin ABC lyase (EC 4.2.2.4.) was purchased from Seikagaku Kogyo Co. (Tokyo, Japan). Protein A-Sepharose 4B, DEAE-Sephacel, and Sephadex G-50 Fine and Superfine were obtained from Amersham Biosciences. [³⁵S]Sulfate, [³H]glucosamine, ¹⁴C-labeled molecular mass standards and Amplify were from Amersham Biosciences. [³⁵S]PAPS was obtained from Marco Maccarana (University of Uppsala, Uppsala, Sweden) or PerkinElmer Life Sciences. Brefeldin A was purchased from Epicentre Technologies (Madison, WI). NOVEX 4-20% Precast Tris glycine gels were purchased from NOVEX (Encinitas, CA). Sodium chlorate and other inorganic chemicals were purchased from Merck (Darmstadt, Germany) with the exception of guanidine hydrochloride, which was purchased from (Fluka, Buchs, Switzerland). All tissue culture plastics were purchased from Costar Europe (VC Badhoevedorp, Holland). DME medium, fetal calf serum (FCS), L-glutamine, and penicillin/streptomycin were purchased from B.I. Bio-Whittaker (Verviers, Belgium). Sulfate-free RPMI 1640 and DME media without Arg, Cys, Gln, Leu, Met, glucose, inositol, and phosphate and the necessary additives were purchased from Invitrogen.

Cell Culture-- MDCKII cells were grown as described previously (24). The cells were established on polycarbonate filters (pore size 0.4 µm; diameter 24.5 mm, Costar Transwell) at a density of 10⁶ cells/filter unless otherwise described. The cells were used for labeling 3-4 days later. An evaluation of filter-grown epithelial monolayers and [³⁵S]sulfate labeling in sulfate-free medium was carried out as described previously (25). Labeling with [³H]glucosamine was carried out in DME medium without Arg, Cys, Gln, Leu, Met, glucose, inositol, and phosphate supplemented with all additives with the exception of glucose.

[³⁵S]Sulfate-labeled Macromolecules-- After metabolic labeling with [³⁵S]sulfate or [³H]glucosamine, the medium fractions were harvested, eventual free cells were removed by centrifugation, and an equal volume of 8 M guanidine, 4% Triton X-100, 0.1 M sodium acetate buffer, pH 6.0, was added. The cell fraction was washed twice (5 min each) with ice-cold phosphate-buffered saline and solubilized in 4 M guanidine, 2% Triton X-100, 0.05 M sodium acetate buffer, pH 6.0. To measure the level of [³⁵S]sulfate incorporated into macromolecules, 1 ml of each fraction was applied to a 4-ml column of Sephadex G-50 Fine in 0.05 M Tris/HCl, pH 8.0, 0.15 M NaCl. The first 1 ml of elute after application was discarded, the next 1.5 ml was collected, and an aliquot of this elute was counted for radioactivity in the scintillation counter. Free [³⁵S] remained associated with the column, and the exchange of guanidine with Tris buffer made further analysis possible.

Ion-exchange Chromatography-- ³⁵S-Labeled macromolecules from cell and medium fractions obtained by Sephadex G-50 Fine chromatography were subjected to preparative ion-exchange chromatography. The samples were loaded onto columns (4-ml wet gel volume) in 0.05 M Tris/HCl, 0.15 M NaCl, pH 8.0, and washed with the same buffer. A gradient extending from 0.15 to 1.5 M NaCl in 0.05 M Tris/HCl then was applied. Fractions of 2 ml were collected from the start of the chromatography. Aliquots of each fraction were counted for radioactivity in a scintillation counter (1900 TR, Packard, Downers Grove, IL) after the addition of Ultima Gold AB scintillation fluid (Packard). Fractions containing peaks of ³⁵S-labeled material were pooled and dialyzed against water at 4 °C with a mixture of protease inhibitors containing 10 mM EDTA, 1 mM -aminocaproic acid, 1 mM N-ethylmaleimide, and 1 mM phenylmethylsulfonyl fluoride. Dialyzed samples were frozen until further analysis.

SDS-PAGE-- Samples with ³⁵S-labeled or ³H-labeled PGs and proteins were boiled in sample buffer with 1% SDS and applied to precast 4-20% gradient NuPAGE gels from Novex. Standards used were ¹⁴C-labeled rainbow standards from Amersham. After electrophoresis, the gels were fixed, treated with Amplify, dried, and subjected to autoradiography using Fuji Medical x-ray film (Tokyo, Japan).

Immune Precipitations-- MDCKII cells were labeled with [³⁵S]sulfate (0.1 mCi/ml) or [³H]glucosamine (0.2 mCi/ml) for 20 h after which the apical and basolateral media and the cell fractions were harvested. The cell fraction was solubilized directly into 0.05 M Tris/HCl, pH 7.5, with 1% Nonidet P-40, 2 mM EDTA, 150 mM NaCl, 35 µg/ml phenylmethylsulfonyl fluoride. The three fractions obtained were incubated overnight at 4 °C with a rabbit antiserum against mouse Perlecan (Dr. J. R. Hassell, Shriners Hospital, Tampa, FL). The fractions had been supplemented with 5 mM MgSO₄ or 5 mM glucose to reduce unspecific binding of free [³⁵S]sulfate and [³H]glucosamine, respectively. The samples were subsequently incubated with protein A-Sepharose prewashed with phosphate-buffered saline containing 1% bovine serum albumin. The samples were washed and finally run on 4-20% Novex gradient gels. After electrophoresis, bands were visualized by autoradiography.

Treatment with Chondroitin ABC Lyase and HNO₂-- Samples from media and cell fractions of ~10,000 cpm were incubated with 0.01 units of enzyme as described previously (25). The degraded material represented chondroitin/dermatan sulfate, and the samples were compared with untreated samples by SDS-PAGE by gel filtration chromatography on Superose-6 columns (Amersham Biosciences). The amount of HSPG was determined by degradation with nitrous acid at pH 1.5, as described by Shively and Conrad (26).

NaOH Treatment-- NaOH treatment releases GAG, CS and HS chains from their protein cores by -elimination. Samples (100-800 µl) of medium and cell fractions (in guanidine) were added to one-tenth of the sample volume of 5 M NaOH and incubated overnight at room temperature. The incubation was terminated by the addition of 5 M HCl to pH 7.0-8.0. The lengths of the free GAG chains were analyzed by Sepharose Cl-6B column chromatography. The elution volumes and K_av coefficients were determined relatively to the elution of the V_o marker Dextran Blue and the V_t marker 2,4-dinitrofenylalanin or K₂CrO₄.

Preparation of Golgi Fractions-- Golgi-enriched subcellular fractions from control and BFA-treated MDCKII cells were prepared as described previously (14, 24). Cells were grown to confluency in 75-cm² plastic flasks (Costar) in DME medium (24). In each experiment, two flasks were treated for 60 min with 1 µg/ml BFA in minimum Eagle's medium (Invitrogen) with 10 mM HEPES, pH 7.4, whereas two flasks were incubated in minimum Eagle's medium with 10 mM HEPES, pH 7.4, alone. After homogenization, a post-nuclear supernatant was made and 840 µl of this supernatant was mixed with 660 µl of 2 M sucrose, 10 mM CsCl, 1 mM HEPES, and this mixture was applied above a 6-ml gradient in SW 41 tubes as described previously (24). Two more layers were applied; first 3 ml of 0.9 M sucrose, 1 mM HEPES, and finally 2 ml of homogenization buffer (0.3 M sucrose, 3 mM imidazole, pH 7.4). Golgi components could be recovered from the interphase between the two latter layers after centrifugation (33,000 rpm, 4.5 h in a Beckman SW 41 rotor).

Incubation of Golgi Fractions with PAPS-- Incubation of Golgi fractions with [³⁵S]PAPS was performed according to the translocation assay described by Brändli et al. (27). Variable amounts of Golgi protein were incubated for 30 min at 37 °C with 25 × 10⁵ dpm [³⁵S]PAPS in incubation buffer: 0.25 M sucrose, 150 mM KCl, 1 mM MgCl₂, 10 mM Tris/HCl, pH 7.5, in a total volume of 1 ml. The reaction was stopped by the addition of 2 ml of ice-cold incubation buffer, and Golgi vesicles were sedimented by centrifugation (40,000 rpm/60 min) TFT 65.13 rotor (Kontron). The supernatant was removed, and the pellet was carefully resuspended and repelleted by centrifugation. The pellet was subsequently dissolved in 50 mM Tris/HCl, pH 7.5, 1% Nonidet P-40, 2 mM EDTA, 150 mM NaCl, 35 µg/ml phenylmethylsulfonyl fluoride, and the radioactivity incorporated in the pellet was determined in a 1900 TR scintillation counter (Packard) after the addition of Ultima Gold AB scintillation fluid (Packard). Protein was measured with the Biuret method or with the Lowry method as modified by Bensadoun and Weinstein (28).

Incubation of Golgi Fractions with UDP Sugars and PAPS-- Golgi fractions were isolated as described previously (14, 24) but in larger scale. Cells were grown to confluency in 500-cm² square tissue culture dishes (Corning/Costar) in DME medium with 5% FCS (24). In each experiment, one plate was treated overnight (20 h) with 2 µg/ml BFA in DME medium with FCS, whereas one plate was incubated in DME medium without BFA. After homogenization, a post-nuclear supernatant was made and 8.4 ml of this supernatant was mixed with 6.6 ml of 2 M sucrose, 10 mM CsCl, 1 mM HEPES, and this mixture was applied above a 10-ml gradient in SW 28 tubes (as described in previous paragraph). Two more layers were applied; first 5 ml of 0.9 M sucrose, 1 mM HEPES, and finally 5 ml of homogenization buffer (0.3 M sucrose, 3 mM imidazole, pH 7.4). Golgi components could be recovered from the interphase between the two latter layers after centrifugation (28,000 rpm, 4.5 h in a Beckman SW 28 rotor).

The Golgi fraction was recovered as described previously (14, 24) and incubated with UDP-glucuronic acid, UDP-N-acetylglucosamine, N-acetylgalactosamine, MgCl₂, MnCl₂, and [³⁵S]PAPS for 2 h at 37 °C. The concentrations are indicated in each experiment. ³⁵S-Labeled macromolecules were isolated by Sephadex G-50 Fine as described above and determined by scintillation counting and, in some cases, analyzed by SDS-PAGE. Pig brain cytosol was isolated as described previously (29).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Inhibition of Sulfation by Chlorate-- Sulfation involves the transfer of sulfate from PAPS to substrates for sulfotransferases located both in the lumen of the Golgi apparatus and in the cytosol. Glycolipids, glycoproteins, and PGs in transit to the cell surface are substrates for sulfotransferases localized to different regions of the Golgi apparatus. Sulfation within the Golgi requires efficient uptake of PAPS from its site of synthesis in the cytosol into the Golgi lumen. This uptake is mediated via PAPS transporters within the Golgi membrane (6, 22, 23).

In our study, we have inhibited the incorporation of sulfate into proteins and PGs in two ways. One approach has been to inhibit the synthesis of PAPS in the cytoplasm by the addition of chlorate (30-32). Low concentrations of chlorate (2-5 mM) reduced the sulfation of proteins and CSPGs, whereas the sulfation of HSPGs was unaffected at these chlorate concentrations but could be inhibited at higher concentrations of chlorate (15-30 mM). Fig. 1 shows the pattern of incorporation of [³⁵S]sulfate into macromolecules in the apical medium, the basolateral medium, and the cell fraction after the treatment of MDCK cells with different concentrations of chlorate during the labeling period. PGs are seen as broad bands in the high molecular weight region of the gel, whereas sulfated proteins are seen as more distinct bands with lower molecular weights. Clearly, protein sulfation is reduced by 2 mM chlorate and is essentially undetectable with 5 mM chlorate in all three fractions. Sulfation of PGs is still efficient at 30 mM chlorate although somewhat reduced. To determine whether sulfation of CSPGs and HSPGs was affected differently by chlorate treatment, the PGs were separated from proteins by DEAE ion-exchange chromatography and subjected to chondroitinase ABC or HNO₂ treatment to degrade CSPGs or HSPGs, respectively. Fig. 2 shows the result from the apical medium. With increasing concentrations of chlorate, the relative fraction of PGs (i.e. HSPG) is increasing, and already with 5 mM chlorate, almost all of the PG is HSPG and very little is CSPG. At all of the higher concentrations of chlorate, only HSPGs were detected (data not shown). The results for the cell fractions were similar to those for the apical medium, whereas the basolateral medium contained essentially only HSPG (data not shown) (33). These results indicate that in MDCK cells, HSPG sulfation, which generally is completed before the trans-Golgi network, may operate at a lower cellular PAPS concentration than sulfation events (protein and CS) that take place in the trans-Golgi network (15-19). Although PAPS K_m values for MDCK cell sulfotransferases have not been determined, such K_m values determined for sulfotransferases involved in HS synthesis in other cell types are considerably lower (0.2-2.4 µM) (34-35) than those involved in CS synthesis (40 µM for 6-sulfation) (36-37). Chlorate acts as a specific inhibitor of sulfation, because the incorporation of [³H]glucosamine into macromolecules was unaffected in the presence of 5 and 20 mM chlorate (Fig. 3).

View larger version (52K): [in this window] [in a new window]   Fig. 1.   Effect of chlorate on the incorporation of [35S]sulfate into proteoglycans and proteins in media and cells. Filter-grown MDCKII cells were incubated for 22 h with 0.1 mCi/ml [35S]sulfate and in the absence or presence of increasing concentrations of sodium chlorate (as indicated) in sulfate-free medium before apical and basolateral media and cell fractions were harvested, macromolecules were purified by Sephadex G-50 Fine gel filtration, and samples were analyzed by SDS-PAGE. Loaded samples correspond to equal volume fractions of media and cell lysates. The figure shows one representative of three experiments.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effect of chlorate on the synthesis of chondroitin sulfate and heparan sulfate. Filter-grown MDCKII cells were incubated for 22 h with 0.1 mCi/ml [35S]sulfate without chlorate or in the presence of 2 or 5 mM chlorate in sulfate-free medium. At the end of the incubation period, the media were harvested and macromolecules were isolated by Sephadex G-50 Fine gel filtration before PGs were separated from proteins by DEAE ion-exchange chromatography. Pooled and dialyzed PG fractions were analyzed by Superose 6 chromatography. Untreated samples of media and cells and samples treated with nitrous acid or chondroitinase ABC were eluted in 0.05 M Tris/HCl, pH 8.0, 0.15 M NaCl, 0.5% Triton X-100. The figure shows the result for the apical medium. One experiment representative of three is shown. , untreated sample; , chondroitinase ABC-treated sample; ×, nitrous acid-treated sample.

View larger version (75K): [in this window] [in a new window]   Fig. 3.   Incorporation of [3H]glucosamine into macromolecules is unaffected by chlorate. MDCKII cells were labeled for 22 h in glucose-free medium with 2% FCS and 200 µCi of [3H]glucosamine added basolaterally to each filter. Apical (1 ml) and basolateral (2 ml) media were harvested, and the cell layer was solubilized. Labeled macromolecules were isolated on Sephadex G-50 Fine columns and analyzed by SDS-PAGE and autoradiography. One of two similar experiments is shown. The migration distances of the protein standards are indicated.

Inhibition of Sulfation by Brefeldin A-- BFA has been shown to inhibit apical transport of glycoproteins in MDCK cells (22-23) without reducing the basolateral counterpart. Filter-grown MDCK cells were labeled with [³⁵S]sulfate in the absence or presence of various concentrations of BFA (0.5, 1.0, 2.0, 3.0, and 5.0 µg/ml). After 20 h, macromolecules in the media and the cell fraction were separated from free [³⁵S]sulfate. Radioactively labeled molecules in the eluted fractions were analyzed by SDS-PAGE. BFA treatment resulted in a decrease in sulfate-labeled macromolecules in both the apical medium, the basolateral medium, and the cell fraction (Fig. 4). The reduction, however, was first observed for the apical medium. This effect could be a combination of reduced sulfation and reduced apical secretion.

View larger version (52K): [in this window] [in a new window]   Fig. 4.   Effect of BFA on the incorporation of [35S]sulfate into proteoglycans. Macromolecules eluted from Sephadex G-50 Fine columns were analyzed by SDS-PAGE. Equivalent fractions of media and cell lysates were treated as described under "Experimental Procedures." The figure shows apical and basolateral media and cell fractions from untreated MDCK cells or from cells treated with 0.5-5.0 µg/ml BFA. The BFA treatment started 30 min before and lasted throughout the 22-h labeling period with 0.1 mCi/ml [35S]sulfate in sulfate-free medium supplemented with 2% FCS.

We next wanted to investigate whether the effect of BFA was limited to the sulfation steps in GAG synthesis. To answer this question, we studied the incorporation of [³H]glucosamine and [³⁵S]sulfate into the GAG chains of one particular PG, Perlecan, which is secreted to both sides of MDCK monolayers (25). By immune precipitation and SDS-PAGE, we could show that while the incorporation of [³⁵S]sulfate into the GAG chains of Perlecan was blocked by 2 µg/ml BFA (Fig. 5B), the incorporation of [³H]glucosamine was essentially unaffected (Fig. 5A). An analysis of HS-GAG chains have shown that these chains are of the same length in PGs from chlorate-treated cells (32), BFA-treated cells, and control cells (data not shown) (38).

An analysis by DEAE ion-exchange chromatography where sulfated proteins are eluted at low salt concentrations while PGs are eluted at higher salt concentrations verified the impression from SDS-PAGE that lower concentrations of chlorate inhibit the sulfation of proteins more than PGs (data not shown). Selective degradation of HSPG and CSPG GAG chains followed by gel filtration showed that the sulfation of CSPG was more reduced than the sulfation of HSPG (Table I).

View this table: [in this window] [in a new window]   Table I Effect of chlorate on [35S]sulfate incorporation into macromolecules Filter-grown MDCKII cells were incubated with 0.1 mCi/ml [35S]SO42 in the absence or presence of various concentrations of sodium chlorate for 22 h as described under "Experimental Procedures." Macromolecules containing [35S]SO42 eluted from Sephadex G-50 Fine columns were first determined by scintillation counting. Equivalent fractions of media and cell lysates were treated with HNO2, chondroitinase ABC, or not as described under "Experimental Procedures" and separated by SDS-PAGE. Quantitation was performed by phosphorfluorimaging scanning. This table shows apical and basolateral media and cell fractions from untreated MDCKII cells or from cells treated with various concentrations of chlorate. The reduction in incorporation of [35S]SO42 into proteins and proteoglycans at increasing chlorate concentrations is expressed as the percent of the respective controls. The values are means ±range based on three separate experiments. The percent of CS relative to HS is the means ±range calculated on the basis of two separate sets of experiments.

The sulfation of proteins and PGs was reduced to a similar extent for all BFA concentrations, and the CS/HS sulfation ratio was also constant. Thus, sulfation has been inhibited in two different ways but with differential effects. Chlorate is known to inhibit the formation of PAPS in the cytoplasm (39), which in turn also reduces the PAPS concentration in the Golgi lumen. We have looked for a substrate for a cytosolic sulfotransferase in MDCK cells without success, but in another epithelial cell line, CaCo-2, we have found that the sulfation of lithocholic acid in the cytoplasm was insensitive to BFA, whereas the sulfation of PGs in the Golgi apparatus was blocked (40). For this reason and because most of the previously reported effects of BFA have been effects on the Golgi apparatus, we postulated that the effect of BFA on sulfation in MDCK cells could have to do with the uptake of PAPS into the Golgi lumen.

View larger version (61K): [in this window] [in a new window]   Fig. 5.   Effect of BFA on incorporation of [3H]glucosamine and [35S]sulfate into the proteoglycan Perlecan. MDCKII cells were metabolically labeled with [3H]glucosamine or [35S]sulfate for 22 h as described under "Experimental Procedures." Half of the filters were treated with 2 µg/ml BFA for 30 min before the metabolic labeling was started. BFA was present throughout the incubation period. Samples of media and cell lysates were immune-precipitated with a rabbit antiserum to mouse Perlecan as described under "Experimental Procedures" and analyzed by SDS-PAGE. Panel A shows the results with [3H]glucosamine labeling, and panel B shows the results with [35S]sulfate labeling.

To investigate whether this hypothesis was correct, we isolated Golgi membrane fractions from control cells and BFA-treated cells. The isolated fractions had the same protein concentration, and the content of galactosyltransferase using ovalbumin and UDP-galactose as substrates is also unchanged (14). The Golgi fractions were incubated with [³⁵S]PAPS before the vesicles were washed and pelleted to determine the uptake of PAPS during the incubation period. Fig. 6 clearly shows that BFA-treated vesicles contained significantly less PAPS than control vesicles and that this difference might explain the reduced sulfation of macromolecules in the presence of BFA.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Uptake of [35S]PAPS into Golgi vesicles from BFA-treated and control MDCK cells. MDCKII cells were treated with BFA (1 µg/ml) for 2 h before treated and non-treated cells were scraped off of the plastic surface (75-cm2 bottles), pelleted, and homogenized as described under "Experimental Procedures." Golgi fractions were prepared by centrifugation of post-nuclear supernatants in sucrose step gradients as described previously (14). Aliquots of Golgi fractions were incubated with [35S]PAPS, washed, and analyzed with respect to PAPS uptake as described under "Experimental Procedures." Uptake into vesicles from untreated cells () and from BFA-treated cells () is shown for an experiment representative of three in which each point represents the mean of duplicate samples. The protein concentration in Golgi fractions from BFA-treated and untreated cells was very similar. The uptake of PAPS is expressed as counts/min/milligram of protein in Golgi fractions as a function of increasing fraction volume.

Inhibition of Sulfation in an in Vitro Assay-- We have previously developed a GAG synthesizing in vitro assay in which Golgi fractions were incubated with cytosol (from rat liver, pig brain, or MDCK cells), [³⁵S]sulfate, an ATP-regenerating system, Mg²⁺, and Mn²⁺. The cytosol catalyzed [³⁵S]PAPS formation and provided UDP sugars for GAG chain synthesis. Elongation of GAG chains was presumably taking place on templates present in the Golgi fraction. In this system, the pretreatment of MDCK cells with BFA before the isolation of Golgi fractions had no inhibitory effect on GAG sulfation. To overcome the demand for the addition of cytosol, we developed a more defined system where Golgi fractions are incubated with UDP sugars, Mg²⁺, Mn²⁺, and [³⁵S]PAPS. The dominating ³⁵S-labeled macromolecular product was of heparan sulfate nature (Fig. 7A). In this system did the pretreatment of the MDCK cells with BFA before the isolation of Golgi fractions reduce GAG sulfation dramatically (Fig. 7B) on average to 32% of control values. The addition of cytosol to this system gave similar GAG sulfation levels for control and BFA pretreated Golgi fractions (Fig. 7B). Only the results for pig brain cytosol are shown, although rat liver and MDCK cytosol also had similar effects.

View larger version (54K): [in this window] [in a new window]   Fig. 7.   In vitro synthesis of glycosaminoglycans: effect of BFA and cytosol. MDCK Golgi fractions, control or pretreated for 20 h with 2 µg/ml BFA, were isolated as described under "Experimental Procedures" and incubated for 2 h at 37 °C. A, control Golgi fractions in the presence of UDP-N-acetylglucosamine (25 µM), UDP-glucuronic acid (25 µM), UDP-N-acetylgalactosamine (25 mM), 15 mM MnCl2, 15 mM MgCl2, and 50 or 100 µM [35S]PAPS. B, control Golgi fractions and fractions from BFA-pretreated MDCK cells were incubated in the presence of UDP-N-acetylglucosamine (25 µM), UDP-glucuronic acid (25 µM), UDP-N-acetylgalactosamine (25 µM), 15 mM MnCl2, 15 mM MgCl2, and 50 µM [35S]PAPS in the absence or presence of 150 µl of pig brain cytosol (total volume 550 µl in all incubations). Labeled macromolecules were separated from unincorporated [35S]PAPS in both A and B. Samples in A were treated with HNO2 and chondroitinase ABC, degrading heparan sulfate, and chondroitin sulfate, respectively, before analysis on SDS-PAGE gels as shown in Figs. 1 and 4. Error bars represent a range of GAG synthesis with BFA-treated Golgi fractions compared with controls from three different experiments with 2 or 3 parallels. The average reduction without cytosol was 32%.

Sulfation Is Also Inhibited by BFA in PtK1 Cells-- In addition to MDCK cells, PtK1 cells have also been reported to have Golgi structures resistant to BFA treatment (20). A 22-h incubation of filter-grown PtK1 cells with [³⁵S]sulfate in the absence or presence of BFA (5 µg/ml) resulted in decreased levels of ³⁵S-labeled macromolecules in the apical medium (48.6% decrease), in the basolateral medium (73.4% decrease), and in the cell fraction (66.1% decrease). At this BFA concentration, the Golgi apparatus is still observable with intact morphology (20). This indicates that the mechanism by which BFA reduces sulfation in MDCK cells is a general one.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this paper, we have investigated the consequences of reducing the level of active sulfate PAPS in the Golgi lumen in two different ways. First, we reduced the cytoplasmic PAPS concentration by the inhibition of PAPS synthesis with chlorate (39). Low concentrations of chlorate reduced the sulfation of proteins and CSPG but did not affect the sulfation of HSPG to the same extent. This differential effect of chlorate could be attributed to lower K_m values for sulfotransferases involved in HSPG synthesis than for those involved in protein and CSPG sulfation. Although no K_m values have been determined for MDCK cell sulfotransferases, the values determined for other cell types fit with this pattern (34-37). An alternative explanation could be a compartmentalized utilization of PAPS within the Golgi apparatus. Protein and CSPG sulfation are events that generally take place in the trans-Golgi network, whereas HSPG sulfation takes place in Golgi cisternae (41). It cannot be excluded that different mechanisms with different K_m values are responsible for the uptake of PAPS into the different subregions of the Golgi apparatus.

The second way in which we have inhibited sulfation events in the Golgi apparatus is by BFA treatment. The Golgi apparatus is morphologically resistant to BFA in MDCK cells (14, 21, 24) with a functional secretory pathway. Still, at low BFA concentrations, sulfation was dramatically reduced at concentrations in which UDP sugars were imported into the Golgi lumen and polymerized to GAG chains. Thus, this is the first report where the effect of BFA on sulfation in the Golgi apparatus is uncoupled from the dramatic effects of BFA on Golgi morphology.

PAPS is also the sulfate donor to cytosolic sulfotransferases. To localize the intracellular target of BFA, one could start with monitoring a cytosolic sulfotransferase. Unfortunately, we were unable to identify a proper substrate for a cytosolic sulfotransferase in MDCK cells. However, in another epithelial cell line, CaCo-2, cytosolic sulfation of lithocholic acid was not inhibited by BFA (38), whereas PG sulfation was reduced. Although the morphology of the Golgi apparatus is more sensitive to BFA in CaCo-2 cells than in MDCK cells, this result indicates that the cytoplasmic PAPS level is not reduced in BFA-treated cells. The isolation of Golgi vesicles from control and BFA-treated MDCK cells revealed that BFA treatment reduced the uptake of PAPS into vesicles of the Golgi fraction by ~50%. This finding represents an average reduction for trans-Golgi network and Golgi cisternae and could be larger or smaller in each of these two Golgi subregions. Thus, the final consequence of both chlorate and BFA treatment seems to be a reduced concentration of PAPS in the lumen of the Golgi apparatus, but the effect on the sulfotransferase substrates is variable.

The detailed molecular mechanism for the effect of BFA on sulfation in MDCK cells remains to be elucidated. In most cell types, BFA induces a dissociation of proteins associated with the cytoplasmic side of Golgi membranes. However, the PAPS translocase is believed to be a membrane-multispanning protein like other Golgi transporters (6-8). However, no DNA or protein sequences have been published for the PAPS translocase, and no antibodies are available. Thus, the question of whether BFA has a direct effect on PAPS translocation across the Golgi membrane is difficult to address. Additional Golgi proteins could influence the sulfation pathway, but no candidate proteins have been reported. An intriguing model for the regulation of the PAPS concentration in the Golgi lumen has been postulated by Pasyk and Foskett (42). These authors discuss the possibility that the content of PAPS in the Golgi lumen is not only dependent on the uptake of PAPS into the Golgi apparatus but also on a regulated release mechanism. According to these authors (42), hypersulfated PGs found in cystic fibrosis patients may be a result of a block in the release of PAPS from the Golgi apparatus in cystic fibrosis patients.

Pretreatment of MDCK cells with BFA before the isolation of Golgi fractions also reduces GAG sulfation in a defined in vitro system with no requirement for the addition of cytosol. A back addition of cytosol abolishes this effect of BFA, indicating that a factor present both on Golgi membranes and in the cytoplasm is required for efficient sulfation. The disassociation from the Golgi membrane upon BFA treatment is compatible with previous reported effects of BFA. Further studies of this factor might shed more light on the molecular mechanisms involved in the PAPS pathway from the cytoplasm to the lumen of the Golgi apparatus.

	ACKNOWLEDGEMENTS

We thank Marco Maccarana for the donation of [³⁵S]PAPS and Samawal Atta for skillfull technical assistance.

	FOOTNOTES

^* This work was supported by the Norwegian Cancer Society, Novo Nordisk Fonden, Freia-fondet, Blix-fondet, and the Norwegian Research Council.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Previously published under the name Katja Svennevig.

^§ Present address: The Norwegian School of Veterinary Science, P.O. Box 8146, Dep, 0033 Oslo, Norway.

Present address: Dept. of Microbiology, The National Hospital of Norway, 0027 Oslo, Norway.

^** To whom correspondence should be addressed: Dept. of Biochemistry, University of Oslo, Box 1041, Blindern, 0316 Oslo, Norway. Tel.: 4722856753; Fax: 4722854443; E-mail: kristian.prydz@biokjemi.uio.no.

Published, JBC Papers in Press, July 22, 2002, DOI 10.1074/jbc.M206365200

	ABBREVIATIONS

The abbreviations used are: PAPS, adenosine 3'-phosphate 5'-phosphosulfate; BFA, brefeldin A; PG, proteoglycans; GAG, glycosaminoglycan; CSPG, chondroitin sulfate proteoglycan; HSPG, heparan sulfate proteoglycan; MDCK, Madin-Darby canine kidney; DME medium, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; PtK, potorous tridactylis kidney.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				G. Dick, F. Grondahl, and K. Prydz Overexpression of the 3'-Phosphoadenosine 5'-Phosphosulfate (PAPS) Transporter 1 Increases Sulfation of Chondroitin Sulfate in the Apical Pathway of MDCK II Cells Glycobiology, January 1, 2008; 18(1): 53 - 65. [Abstract] [Full Text] [PDF]

				T. T. Vuong, K. Prydz, and H. Tveit Differences in the apical and basolateral pathways for glycosaminoglycan biosynthesis in Madin-Darby canine kidney cells Glycobiology, April 1, 2006; 16(4): 326 - 332. [Abstract] [Full Text] [PDF]

				K. S. Jones, C. Petrow-Sadowski, D. C. Bertolette, Y. Huang, and F. W. Ruscetti Heparan Sulfate Proteoglycans Mediate Attachment and Entry of Human T-Cell Leukemia Virus Type 1 Virions into CD4+ T Cells J. Virol., October 15, 2005; 79(20): 12692 - 12702. [Abstract] [Full Text] [PDF]

				H. Tveit, G. Dick, V. Skibeli, and K. Prydz A Proteoglycan Undergoes Different Modifications en Route to the Apical and Basolateral Surfaces of Madin-Darby Canine Kidney Cells J. Biol. Chem., August 19, 2005; 280(33): 29596 - 29603. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/39/36272    most recent M206365200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fjeldstad, K. Articles by Prydz, K. Search for Related Content PubMed PubMed Citation Articles by Fjeldstad, K. Articles by Prydz, K. Social Bookmarking                      What's this?