ARCHIVES

Vol.

OF BIOCHEMISTRY

298, No. 2, November

AND

BIOPHYSICS

1, pp. 371-379,1992

Effects of MgCI, on the Release and Recycling of Heparan Sulfate Proteoglycans in a Rat Parathyroid Cell Line Y asuhiro

Takeuchi,’

Bone Research Branch, Received

April

Masaki National

Y anagishita,2 Institute

13, 1992, and in revised

form

of

June

and Vincent

0 1992

Academic

Press,

Inc.

’ Present address: Fourth Department of Tokyo, School of Medicine, 3-28-6 Japan. ’ To whom correspondence should 0003-9861/92 $5.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

of Internal Medicine, Mejirodai, Bunkyoku, be addressed.

Institutes

of

Health, Bethesda,

Maryland

20892

23, 1992

Divalent cations, such as Mgz+, Baz+, and Co2+, are known to mimic the effects of Ca2+ in parathyroid cells, but it is not clear whether the mechanism of their action is the same as that of Ca2’. We have shown that extracellular Ca” concentration ([Ca2’],) regulates the distribution and recycling of cell-surface heparan sulfate (HS) proteoglycans in a rat parathyroid cell line; at normal to high [Ca2’], (e.g., 2 mM) HS proteoglycans are primarily localized intracellularly, while at low [Ca”], (0.05 mM) they are translocated to the cell surface and rapidly recycle (Takeuchi, Y., Sakaguchi, K., Yanagishita, M., Aurbach, G. D., and Hascall, V. C., 1990, J. Biol. Chem. 265, 13661-13668). We now show that a high concentration of Mg2+ (8 mM) reduces the amount of recycling HS proteoglycans in low [Ca2’],. However, the primary effects of high Ca2+ and high Mg2+ on the recycling HS proteoglycans are different. High [Ca2+], causes translocation of HS proteoglycans to intracellular compartments, while high Mg2+ stimulates cleavage of their core proteins and subsequent shedding of HS proteoglycans into the medium, thereby depleting the recycling moleof cules. However, high Mg 2+ does not induce shedding HS proteoglycans in high [Ca2’],. The effects of Ba2+ and Co2+ were similar to those of M8+, but Sr2’ showed no significant effects on HS proteoglycan translocation. Otherwise, 8 mM Mg 2+ did not alter biosynthesis or intracellular catabolism of HS proteoglycans. These observations suggest that the recycling of HS proteoglycans in parathyroid cells is sensitive only to [Ca2’], , whereas several other divalent cations can deplete the recycling HS proteoglycans by a distinctly different mechanism. Thus, the mechanism by which Ca” regulates the amounts of the recycling HS proteoglycans may be more physiological and play a functional role in parathyroid CellS.

C. Hascall

Dental Research, National

University Tokyo 112,

Sensing extracellular Ca2+ concentration ([Ca”‘],) is a principal function for parathyroid glands, since secretion of parathyroid hormone (PTH)3 is regulated by changes of [Ca”],. Sensing mechanisms that have been postulated include receptors for divalent cations (1) and Ca2+channels (2) on the cell surface. Both of these would increase intracellular Ca2+ in response to an elevation of [Ca2+], (1, 3). However, the relationship between an increase of [Ca2+], and the suppression of secretion of PTH is not known and is novel because increases of intracellular Ca2+ and physiological levels of [Ca*‘], are generally potent stimulators for secretion and exocytosis in many other systems (4, 5). A rat parathyroid cell line established by Sakaguchi et al. (6) retains certain characteristics typical of parathyroid cells in primary cultures, most importantly secretion of parathyroid hormone-related proteins in response to decreases of [Ca2+], (7). We have reported that the distribution of heparan sulfate (HS) proteoglycans between the cell surface and an intracellular compartment in this cell line is regulated by [Ca2+], (8) and that a major portion of the HS proteoglycans recycle between these compartments when [Ca2’], is lowered (9). Further, the cells respond rapidly to changes of [Ca2’], by redistribution of HS proteoglycans (T,,, of -5 min) (9). Therefore, the distribution and recycling capacity of HS proteoglycans is a convenient parameter for studying mechanisms of the responsiveness of these cells to changes of [Ca2’],. Although the effects of divalent cations other than Ca2+ on functions of parathyroid cells have been reported to be similar to those of Ca2+, several different modes of action have been suggested (1, 10). Cell-surface HS proteoglycans are widely distributed throughout animal tissues. Their ubiquitous cell-surface 3 Abbreviations used: PTH, parathyroid TPA, phorbol 12-myristate 13-acetate; pyl)dimethylammonio]propanesulfonate; sulfate-polyacrylamide gel electrophoresis; inositol triphosphate.

hormone; HS, heparan sulfate; Chaps, 3-[(3-cholamidoproSDS-PAGE, sodium dodecyl PI, phosphatidylinositol; IP$,

371 Inc. reserved.

372

TAKEUCHI,

YANAGISHITA,

localization and the capacity of HS chains to interact with numerous molecules, including growth factors, cytokines, extracellular matrix proteins, enzymes, and protease inhibitors strongly suggest that HS proteoglycans are involved in critical cell functions such as cell-cell and cell-extracellular matrix interactions (11-14). In this report, we describe the effects of several divalent and recycling cations, especially Mg2+, on the distribution of HS proteoglycans in the rat parathyroid cell line. EXPERIMENTAL

PROCEDURES

Materials. Guanidine HCl and urea were purchased from Life Technologies/BRL. [?S]Sulfate (-40 Ci/mg), D-[6-3H(N)]glucosamine (40 Ci/mmol), and L-[3,4,5-3H(N)]leucine (160 Ci/mmol) were purchased from Du Pant/New England Nuclear; Sephadex G-50 (fine), Q-Sepharose, and prepacked Superose 6 from Pharmacia/LKB Biotechnology Inc.; MgCls, CoCl,, BaC&, CoClz, SrCls, phorbol 12-myristate 13-acetate (TPA), diphenylcarbamyl chloride-treated trypsin, and soybean trypsin inhibitor from Sigma. The culture medium used was a mixture (1:l) of Coon’s modified Ham’s F-12 and Dulbecco’s modified Eagle’s minimum essential medium (both without Ca*+ and Mgs+ and obtained from the NIH media unit) supplemented with 5% calf serum (Biofluids), 20 mM N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid (Hepes), 100 units/ml penicillin, and 100 pg/ml streptomycin sulfate. CaClz and MgClr were added to the media at the indicated final concentrations. The concentration of Mgs+ was 0.5 mM unless otherwise specified. Free Car+ concentrations in the final media with 5% calf serum were determined with a calcium electrode (Orion). [Gas’], in medium for cell growth and passage was 0.7 mM. In this paper, medium containing 2 mM [Gas’], was utilized as “high Ca*+” medium and 0.05 mM [Car+], as “low Gas+” medium unless otherwise indicated. Other reagents were obtained from commercial sources. Overnight labeling of cell cultures and chase protocols. Cells were cultured in 6-well (Falcon) or 12-well plastic plates (Costar) with 2 or 1 ml of growth medium as described previously (9). Cultures were labeled at confluence with 10 &i/ml [35S]sulfate, 100 &i/ml [aH]leucine, or 100 &i/ml [3H]glucosamine in medium with the indicated [Ca”]. and 5% calf serum for 24 h. After labeling, cultures were washed three or four times over 30 min with 0.15 M NaCl, 20 mM Hepes, pH 7.4, containing the same concentrations of Car+ and Mgs+ as the labeling medium (20 mM Hepes buffer). They were then chased at 37°C for the indicated periods in the serum-free medium containing various concentrations of divalent cations (Mgs+, Bas+, Co’*, or S?‘) for 60 min at 37°C unless otherwise indicated. Unincorporated radioactivity in the medium was less than 2% of the total incorporated activity in macromolecules after the washing procedure. Short-term pulse-labeling of cell cultures and chase protocols. After washing confluent cultures with 20 mM Hepes buffer at 37”C, cells were pulse-labeled with 200 &i/ml [?S]sulfate for 10 min at 37°C in low sulfate (-0.1 mM) medium and washed five times within 3 min with 20 mM Hepes buffer and the same concentration of Ca2+ as in the labeling medium, but without isotopes and supplemented with 0.5 mM MgSOI. Cultures were then chased in serum-free medium with the same [Car’], as for the pulse-labeling for the times indicated. Cell Measurement of radioactivity incorporated into macromolecules. layer extracts, media, and trypsin-solubilized samples in 4 M guanidine HCl solutions were chromatographed on Sephadex G-50 columns (bed vol 4 ml) equilibrated with 8 M urea, 50 mM sodium acetate, 0.15 M NaCI, 0.5% (w/v) Triton X-100, pH 6.0 (8 M urea buffer) (9). Aliquots of excluded fractions were measured for radioactivity to estimate the total radioactivity incorporated into macromolecules. Cellular localization of HS proteoglycans. For the determination of trypsin-accessible proteoglycans, cultures were digested with 200 or 20 @g/ml trypsin at 37°C for 2 or 15 min, respectively, in serum-free medium

AND

HASCALL

containing the same [Car+], as for the chase at the indicated periods of time. Trypsin-accessible and trypsin-inaccessible materials were separately collected and analyzed for radiolabeled proteoglycans as described previously (9). Heparitinase digestion. Macromolecules metabolically labeled with [35S]sulfate and/or [sH]glucosamine or [sH]leucine from cell-extracts and media were prepared in 8 M urea, 0.30 M NaCl, 0.05 M Na acetate, 0.5% Triton X-100, pH 6.0, by Sephadex G-50 chromatography as described above, and applied to Q-Sepharose columns (bed vol 0.2 or 1 ml). Each column was washed with 2.5 bed vol of 8 M urea, 0.30 M NaCl, 0.05 M Na acetate, 0.5% Triton X-100, pH 6.0, and then with 2.5 bed vol of the same buffer with 0.5% Chaps instead of Triton X-100. Molecules containing glycosaminoglycans were eluted in 4 M guanidine HCl, 0.1 M Na acetate, pH 6.0, with 0.5% Chaps, and lyophilized following dialysis first against 0.5 M NaCl and then distilled water. This procedure decreased detergents in samples to a level which did not interfere with heparitinase activity. Lyophilized samples were dissolved in 0.1 M Trisacetate, pH 7.3, containing 50 mM EDTA and a mixture of protease inhibitors (8) and then digested with 10 mIU/ml heparitinase at 37°C for 2 h. Octyl-Sepharose CL-II3 chromatography. Samples prepared as described in the previous section were applied to octyl-Sepharose CL-4B columns (2 ml bed vol) and eluted with 4 M guanidine HCl buffer with a gradient (O-0.5%) of Triton X-100 (15). SDS-polyacrybmide gel electrophoresis. SDS-polyacrylamide gel electrophoresis was done as described by Laemmli (16) using precast mini gels (4-20% gradient or 7.5% acrylamide, Daiichi Fine Chemicals, Tokyo). Prestained proteins with known molecular weights (Life Technologies/BRL) were run as molecular weight standards. After electrophoresis, gels were fixed in 40% methanol/lo% acetic acid and dried. The radioactivity was detected by fluorography using Enlightning (DuPont/NEN).

RESULTS

Extracellular Mg2’ stimulates the shedding of HS proteoglycans into the medium. An increase in [Ca2+], to 2 mM decreased the rate of release of HS proteoglycans associated with rat parathyroid cells from -20 to - 13% during 60 min of chase (Table I). Among other divalent cations tested in the presence of low [Ca’+],, none decreased the release of proteoglycans. In contrast, Mg2+, Ba2+, and Co2+significantly increased the shedding of HS proteoglycans in low [Ca2’], while Sr2+ up to 16 mM had no significant effect. The presence of high [Ca2’], prevented the effects of 8 mM Mg2+ and 4 mM Co2+. These observations suggest that the action of Mg2f or Co2+ is not competitive with that of Ca2+and that there is a population of HS proteoglycans whose release is stimulated by Mg2+, Ba2+, or Co2’, but only in low [Ca2’],. After labeling with [35S]sulfate in low [Ca2’], for 24 h, cultures were chased in either 0.5 or 8 mM Mg2+ in low [Ca2’], medium (Fig. 1). The release of HS proteoglycans was stimulated by 8 mM Mg2+ during the first hour, and thereafter gave a rate similar to that in the control with 0.5 mM Mg2+ through 3 h of chase. Other cultures were labeled with [35S]sulfate for 24 h in low [Ca2’], and chased for 60 min in the same medium with different Mg2+ concentrations. The effect of Mg2+ continued to increase up to the highest concentration tested (16 mM) with a sharp increase in release between 1 and 4 mM (Fig. 2).

MgClz

AND

HEPARAN

TABLE Effects of Divalent HS Proteoglycans

Divalent

cation

CaClz

Cations in Rat

Concentration bM) 0” 0.05 2

CaClz

SULFATE

PROTEOGLYCANS

I on the Shedding of Parathyroid Cells % of 3SS-HSPG released into medium 24.1 + l.Ob 19.6 + 0.7 12.6

k 0.3 + + + + k 5 f t

+ 1.3 + 0.6

(0.05 mM)

+ CoCl,

1

+ SrC12

4 8 16

44.6 48.5 27.2 30.2 51.9 52.7 18.2 18.0

8 4

15.6 14.2

+ MgClz

+ BaClz

CaCl, (2 mM) + MgCll + CoCl*

8 16 8 16

0.8 0.8 0.9 0.5 0.3 0.5 0.2 0.3

Note. Cultures were labeled with [35S]sulfate for 24 h in low [Ca”]. and treated with the indicated concentrations of divalent cations for 60 min at 37” C. Released 35S-macromolecules during the treatment period were measured. ’ Medium without CaCl* and containing 0.5 mM EGTA. b Mean + SE (n = 3).

Both HS proteoglycans and HS chains are released into the medium (17). Therefore, the media from an experiment as in Fig. 1 were fractionated on Superose 6 to estimate the proportions of each. An example of a Superose 6 analysis is shown in Fig. 3, and the data for the proportions at each chase time are shown in Fig. 4. The data indicate that Mg ‘+ stimulated the release of HS proteoglycans but did not affect the generation and the release of HS chains. The contents of free [35S]sulfate in media and cell layer extracts, which result from HS proteoglycan degradation in lysosomes, were measured with ion chromatography (17). Mg2+ did not affect the generation of free [35S]sulfate (data not shown). These observations indicate that Mg2+ did not alter the degradation processes that (a) produce and release HS chains into the medium, and (b) generate free [35S]sulfate intracellularly. Thus, the excess HS proteoglycans released in the presence of 8 mM Mg2+ do not contribute directly to either of these degradation pathways. The molecular sizes of the intact HS proteoglycans released into the medium could not be distinguished from those of the cell-associated HS proteoglycans by Superose 6 chromatography or by SDS-PAGE using gradient (420%) polyacrylamide gels. However, the core proteins of the two species of HS proteoglycans in the medium, referred to as HS-PGl, and HS-PG2, (180 and 64 kDa), were significantly smaller than those in the cell layer, referred to as HS-PGl, and HS-PG2, (8) (200 and 70 kDa; Fig. 5A, lanes 2, 3, and 5). Cultures were labeled

IN

RAT

PARATHYROID

373

CELLS

with [35S]sulfate for 24 h in low [Ca’+], and chased for 60 min in either 0.5 or 8 mM Mg2+. HS proteoglycans released into the medium during 60 min of chase in the presence of 8 mM Mg2+ had no other core protein bands than those also observed in the low [Ca”], medium with 0.5 InM Mg2+ (Fig. 5A, lanes 2 and 3). HS-PGl, and HSPG2, also contained small amounts of core proteins which have molecular sizes similar to those of secreted HS-PGl, and HS-PG2, as minor components (more apparent in Figs. 5B and 7B). Separate cultures were labeled with [3H]leucine for 24 h in low [Ca2’], and chased for 1 h in low [Ca2+], media with 0.5 or 8 mM MgCl,. Proteoglycans were isolated from the cell layer, treated with heparitinase, and analyzed on SDS-PAGE with and without @mercaptoethanol (Fig. 5B). Treatment with 8 mM Mg2+ decreased all core proteins associated with the cells. &Mercaptoethanol treatment did not show any effect on the mobility of the bands. Thus, the enhanced release of HS proteoglycans of smaller core protein size into the medium in the presence of 8 mM Mg2+ is probably due to enhanced proteolytic cleavage. Although Mg2+ stimulated the release of both HSPGl, and HS-PG2, (Fig. 5A, lanes 2 and 3), [35S]sulfate labeling experiments shown in this paper are mainly con-

U

I

I

1

2

J

3

Chase(h)

FIG. 1. Effect of MgCl* on the release of HS proteoglycans synthesized by rat parathyroid cells. Cultures were labeled with [35S]sulfate for 24 h in medium containing low [Ca”], and 0.5 mM MgClz followed by incubation in medium containing low [Ca*‘],, and either 0.5 or 8 mM MgC&. %S-HS proteoglycans released into the media during the indicated periods of time were measured and are expressed as percentages of the total 35S-HS proteoglycans at time 0 (bottom). The top shows the differences in the amounts of s5S-HS proteoglycans released in the media containing 0.5 and 8 mM MgClz.

374

TAKEUCHI.

YANAGISHITA,

CaCI, 0.05 mM

AND

HASCALL

A __--

-.--

--A

,A---

B

0

4

8

12

Free [35S]sulfate

16

FIG. 2. Effect of MgClz concentration on the release of HS proteoglycans in rat parathyroid cells. Cultures were labeled with [%]sulfate for 24 h in medium with low [Ca”], and were then treated with various concentrations of MgCIZ for 60 min. The 35S-HS proteoglycans released into medium and those associated with cells were measured. Data are means f SE (n = 3) and are expressed as percentages of the total s5Sradioactivity incorporated into HS proteoglycans.

0.5 mM MgCI, -----8 mM MgCI, 2

cerned with HS-PG2 species because -90% of the total incorporation of [35S]sulfate into proteoglycans is initially associated with HS-PG2, (8). Mg2+ reduces the recycling compartment of HS proteogZycans. Cultures were labeled for 24 h with [35S]sulfate in low [Ca’+],, and were then chased in the presence of either 8 or 0.5 mM Mg 2f . At the indicated times cultures were digested for 2 or 15 min with 200 or 20 pug/mltrypsin, respectively, to measure the size of the recycling compartment of HS proteoglycans (9) (Fig. 6A). Trypsin digestion for 2 min removes HS proteoglycans that are

0.5 mM MgCI,

s

Superose 6

cn 0

FIG. 3. Superose 6 profiles for %-labeled macromolecules in medium and cell layer. %-labeled HS proteoglycans isolated from the cell layer and medium of a culture pulse labeled for 24 h were analyzed by Superose 6 chromatography in 4 M guanidine HCI buffer. The cell extract contained mostly HS proteoglycan (& = 0.16), while the medium compartment contained HS chains (& = 0.4 - 0.8) as well.

4 6 Chase (h)

8

FIG. 4. Effect of 8 mM MgC& on the generation of HS chains and free sulfate. Cultures were labeled with [35S]sulfate for 24 h in medium with low [Ca”], and were then chased in the presence of 0.5 mM (solid lines) or 8 mM MgCl* (dashed lines). The %-labeled HS chains released into medium and [%]sulfate generated by intracellular degradation of %-labeled HS proteoglycans during the chase were measured as described elsewhere (17). A, 35S-Labeled macromolecules in media were fractionated into HS proteoglycans (A) and HS chains (A). B, The total [?S]sulfate values are shown for cells and medium combined. The free [%]sulfate was measured by ion chromatography of the included fractions from Sephadex G-25 analyses (17). Data are averages of duplicate cultures.

on the cell surface as well as those that appear on the cell surface during the 2 min digestion, and digestion for 15 min depletes almost all of the HS proteoglycans in the recycling compartment, because the average cycling time of HS proteoglycans is -9 min (9). The data in Fig. 6 indicate that the recycling compartment was almost abolished within 15 min in the presence of 8 mM Mg2+ since there was little difference between the 2- and 15min digestions (Fig. 6B). Cultures in low [Ca2’], were labeled with [35S]sulfate for 10 min and chased for 60 min before 2- and 15-min trypsin treatments (Table II). The presence of 8 mM Mg2+ stimulated the release of 35S-labeled, newly synthesized HS proteoglycans into the medium and reduced the size of the recycling pool. The transport of the newly synthesized molecules to the cell surface or to the recycling compartment was not affected; i.e., 80% of them were either released into the medium or trypsin-accessible in each case. The net effect of Mg2+, then, was to release HS

MgClz

AND

HEPARAN

SULFATE

Medium Medium hW

MgCI,

HS’ase

PROTEOGLYCANS

Cell

8

8

0.5

a

a

-

+

+

-

+

[%]sulfate [3H]glcN

IN

RAT

PARATHYROID

label

375

CELLS

Cell Mg treatment

--

p -mercaptoethanol

-

++ +

-

+

[3H]lewlabel HS’ase

VW ___

200

Wa) HS-PG2c --t

___

97.4

___

68 43

97.4 ___

29

68 4;:

1

2

3

4

5

B

FIG. 5. Analysis of HS proteoglycans and their core proteins by SDS-polyacrylamide gel electrophoresis. A, HS proteoglycans labeled with [?S]sulfate and [3H]glucosamine as precursors for 24 h in low [Ca”], were isolated from medium containing 8 mM MgCl* (lanes 1 and 2), 0.5 mM MgCl, (lane 3), or from the cell layers chased in the presence of 8 mM MgCl* (lanes 4 and 5). Isolated proteoglycans were analyzed by SDSpolyacrylamide gradient (4-20%) gel electrophoresis followed by fluorography, before (1 and 4) and after (2, 3, and 5) heparitinase digestion. Radioactive materials at the top of the gel after heparitinase digestion (lanes 2, 3, and 5) represent mostly chondroitin sulfate proteoglycans, which can be digested by additional chondroitinase digestion (data not shown). B, HS proteoglycans metabolically labeled with [3H]leucine for 20 h in low [Ca”], were extracted and isolated from cell layers after 1 h chase in medium containing 0.5 mM (lanes 1 and 2) or 8 mM (lanes 3 and 4) MgC12. Samples were analyzed by SDS-polyacrylamide (7.5%) gel electrophoresis in the presence or absence of @-mercaptoethanol after heparitinase digestion.

proteoglycans, which are normally in the recycling pool, into the medium. of Mg’+-releasable HS proteoglyCharacterization cans. Cell-associated HS proteoglycans labeled with [35S]sulfate for 24 h were separated into three fractions by octyl-Sepharose CL-4B hydrophobic chromatography (Fig. 7A). HS proteoglycans labeled with [3H]glucosamine as a precursor were analyzed similarly by octyl-Sepharose CL-4B (data not shown) and each peak was analyzed by SDS-polyacrylamide gel electrophoresis with or without heparitinase treatment (Fig. 7B). The unbound HS proteoglycans had smaller core proteins, whose molecular sizes were identical to those of the HS-PGl, and HSPG2,. Those in bound fractions that eluted with lessthan 0.17% of Triton X-100 (peak Bl) contained HS-PGl, and HS-PG2, with intact core proteins and only a small amount of HS proteoglycans with smaller core proteins. Those in tightly bound fractions that eluted at 0.2% Triton X-100 (fraction B2) were almost exclusively HS-PGl,. More than 90% of the HS proteoglycans in the medium did not bind to octyl-Sepharose (Fig. 7A). The label in Bl-fractions decreased when low [Ca’+], cultures were chased in the presence of 8 mM Mg2+ for 60 min after 24 h labeling (Table III), resulting in an

increased release into the medium. Unbound fractions in the cell layer were not significantly altered by Mg’+, suggesting that enhanced proteolysis, perhaps removing hydrophobic peptides from the core proteins of the HS proteoglycans in the Bl fraction, was required for the stimulated release of proteoglycans by 8 mM Mg2+. Changes of [Ca”], did not affect the bound fractions. Increasing [Ca”], from low to high during the chase reduced the release of HS proteoglycans and reciprocally increased the unbound fraction in the cell layer. Lowering [Ca’+], from high to low during the chase changed the distribution of HS proteoglycans in medium and unbound fractions in the cell layer in the opposite direction. When cells labeled with [35S]sulfate for 24 h were treated with 8 mM Mg2+ in low [Ca’+], at 4°C to prevent endocytosis, - 12% of HS proteoglycans were released, compared with only 2-3% in the low [Ca’+], medium with 0.5 mM Mg2+ at 4°C or in the Ca’+-free medium containing 0.5 mM EGTA at 4°C (Table IV). This indicates that Mg2+ still enhances release of HS proteoglycan in low [Ca2’], and further distinguishes the mechanism by which Mg2+ stimulates the release of HS proteoglycans from the effects of Ca2+ on HS proteoglycan parameters. The effects of CoCl, on the release

376

TAKEUCHI,

30tr

YANAGISHITA,

AND

labeling cells in low [Ca’+], with [35S]sulfate stimulated the release of HS proteoglycans and reduced the size of the recycling compartment as determined by trypsin digestion (Table V). In high [Cazfle, TPA did not change the trypsin-accessibility of HS proteoglycans and was similar to Mg2+ in this respect. However, TPA stimulated the release of HS proteoglycans even in high [Ca’+], much more than did Mg2+ (Tables V and VI). Cell-associated HS proteoglycans were fractionated by octyl-Sepharose CL-4B after treatment with 1 PM TPA for 60 min (Table VI). Treatment with TPA decreased both unbound and Bl-fractions in either low or high [Ca2+le, while 8 mM Mg2+ did not affect the distribution of HS proteoglycans in high [Ca2’],. Further, the release of HS proteoglycans stimulated by TPA was higher than that induced by 8 mM Mg2+ in low [Ca2+le (Tables V and VI); 8 mM Mg2+ had no additive effects to those of TPA (Table V). These observations suggest that the actions of Mg2+ are dependent on the distribution of HS proteoglycans to stimulate the release of HS proteoglycans in the recycling compartment. Further, they suggest that TPA stimulates the release of HS proteoglycans in the recycling compartment as well as those in compartments which do not communicate with the cell surface in high [Ca”],.

B-

32 2

E

(min) FIG. 6. Effect of 8 mM MgClr on the recycling of HS proteoglycans. A, cultures were labeled with [%J]sulfate for 24 h in medium containing low [Ca”], and were then washed and chased in medium containing 0.5 or 8 mM MgClz. The amounts of “S-labeled HS proteoglycans released into medium during the chase (A) 8 mM MgClx; A 0.5 mM MgCl, and those accessible to trypsin (curves with bars) were measured. The r coordinates and horizontal length of the bar indicate timing and duration (2 or 15 min) of trypsin treatment. The y coordinates indicate amounts of %-HS proteoglycans released by the enzyme treatment. Filled and unfilled bars indicate MgCIp concentrations in the chase medium, either 0.5 or 8 mM, respectively. Data are averages of duplicate cultures and expressed as percentages of the total radioactivity in HS proteoglycans at time 0. B, differences between trypsin-accessible “S-HS proteoglycans for 2- and 15 min-digestions in cultures chased in 8 mM MgClz are shown.

DISCUSSION

It has been postulated that Ca2+-receptors, which also have an affinity for other divalent cations, or Ca2+-channels are present on the surface of parathyroid cells as a mechanism to increase intracellular Ca2+ in response to the elevation of [Ca2’], (1,2). However, the mechanisms by which divalent cations modulate biological functions of parathyroid cells, especially the regulated release of PTH by the change of [Ca2+],, are presently uncertain (18). The causal relationship between the increased intracellular Ca2+evoked by an elevation of [Ca2’], in parathyroid cells in vitro and the subsequent inhibition of the secretion of PTH (1-3) is paradoxical to the general concept for the regulation of exocytosis (5). This is a puzzle because the secretion of PTH is actually stimulated by the increase of Ca2+ in permeabilized parathyroid cells

of HS proteoglycans at 4°C were similar to those of MgClz (Table IV). Effect of phorbol ester on the release and recycling of HS proteoglycans. Treatment with 0.1 or 1 pM TPA after

TABLE Effect

of MgCls

on Distribution

of Newly

HASCALL

II

Synthesized

HS

Proteoglycans

in Rat

Parathyroid

Cells

Trypsin-accessible Chase condition

0.05mM

Release of 36S-HSPG into medium

2 min

15 min

34.8 It_ 2.1 40.9 f 2.5

62.9 f 0.9 51.0 + 1.2

A Trypsin-accessible (15 min - 2 min)

Cat&

+ 0.5 mM MgClz f8 mM MgClz

18.0 + 0.7 30.2 + 0.4

Note. Cells were labeled with [“S]sulfate for 10 min and chased for 60 min in the presence of 0.5 or 8 mM MgClz by digestion with trypsin for 2 or 15 min. Values are percentages of the total %-HS proteoglycan and represent and n = 3 for trypsin-accessible samples).

28.1 10.1 in low [Ca’+], medium followed means + SE (n = 6 for medium

MgClz

AND 8

HEPARAN

r

Inbound

SULFATE

PROTEOGLYCANS

IN

RAT

PARATHYROID

377

CELLS

Bl 0.5

. I-

UB

heparitinase

u.4

o@

0.3

-8

-

81

+

-

82

+

-

+

WW -200

A 0.2

,=s ti

0.1

:., ‘

0.0

Fraction

FIG. 7. Fractionation of HS proteoglycans by [Ca”], medium. A, ?S-HS proteoglycans isolated CL-4B column with a gradient of Triton X-100 were analyzed by SDS-polyacrylamide gradient top of gel after heparitinase digestion represent digestion (data not shown).

of Cell-Associated

Low Ca Low Ca Low Ca High Ca High Ca

Low Ca Low Ca + 8 mM MgCl, High Ca High Ca Low Ca

43

Activation of protein kinase C by a phorbol ester has been shown to stimulate PTH secretion regardless of [Ca2’], (24). Our experiments also indicated that a phorbol ester stimulates shedding of HS proteoglycans into the medium in rat parathyroid cell cultures irrespective of the [Ca”‘], tested. Kobayashi et al. reported that decreases of [Ca2+],, which did not activate PI turnover, stimulated protein kinase C activity associated with plasma membranes of parathyroid cells (25). This is in contrast to the situation found in other secretory cells where stimulation of the PI turnover results in activation of protein kinase C and eventually secretory processes. These observations suggest that, in parathyroid cells, activation of protein kinase C, either by lowered [Ca2+], or by stimulation with a

HS

III

Proteoglycans

by Octyl-Sepharose

CL-4B

Cell Chase

-

octyl-Sepharose CL-4B chromatography. Cultures were labeled for 24 h with [35S]sulfate in low from the cell layer (solid line) and medium (dashed line) were analyzed by an octyl-Sepharose (O-0.5%). B, samples labeled with [3H]glucosamine as a precursor and pooled as indicated in A (4-20%) gel electrophoresis before and after heparitinase digestion. Radioactive materials at the mostly chondroitin sulfate proteoglycans, which can be digested by additional chondroitinase

TABLE

Labeling

68

Number

(19) as is seen in other secretory systems. Thus, an intact plasma membrane is required for an appropriate relationship between intracellular Ca2+ and the secretory process of PTH. Increases of [ Ca2+], also stimulate phosphatidylinositol (PI) turnover in parathyroid cells, resulting in an elevation of inositol triphosphate (IP,) and diacylglycerol (20, 21). IPB stimulates release of Cazf stored in microsomes resulting in transient elevation of intracellular Ca2+ and of diacylglycerol, which stimulates activity of a protein kinase C (22). Generally, activated PI turnover, subsequent increase of Cazt, and stimulation of protein kinase C enhance the secretory processes (23). This fact is another puzzle for the control of PTH secretion in parathyroid cells where low [Ca2’], stimulates the secretion process.

Fractionation

97.4

~

-.,‘I.

B

A

-

Medium

54.2

9.7 8.6 26.8 23.0

+ 0.3*

13.4 -c 1.4 9.7 + 0.4

24.1 * 1.5

Note. Cultures were labeled with [a5S]sulfate for 24 h in indicated. MgCl, concentration was 0.5 mM except the one were purified by Q-Sepharose anion-exchange chromatography gradient of Triton X-100. Bl, peak eluted between 0.02 and n Mean rt SE (n = 3). * Statistically significantly different from low Ca/0.5 mM

+ + f -t

layer Bl

Unbound

29.7 + 1.3O

Chromatography

0.6 0.2 0.9 0.8

12.2 + 0.1

52.8 32.8 51.5 58.7 57.2

IL 1.7

+ O.l* 2 2.0 + 0.6 k 1.6

Medium + cell unbound

B2 7.8 4.4 8.4 8.5 6.5

5 zk + + 31

0.7 0.4 2.3 0.5 0.3

39.4 62.8 40.2 32.8 36.3

L + f + f

1.4 0.3* 0.6 0.5 1.6

0.05 mM (low) or 2 mM (high) [Ca”], and chased for 60 min in various media as indicated otherwise. 35S-HS proteoglycans associated with cell layers after the chase and fractionated into three peaks (see Fig. 7A) by octyl-Sepharose CL-4B with a 0.17% Triton X-100; B2, peak eluted at 0.2% Triton X-100. MgC12

group

(P < 0.005).

378

TAKEUCHI. TABLE

Release of HS Proteoglycans

Treatment

medium

CaCI, 0 mMn 0.05 mM 4mM call, (0.05mM) + MgCl* (8 mM) +CoCl, (1 mM)

YANAGISHITA,

AND

HASCALL

IV

TABLE

into Medium

at 4°C

Effect of Phorbol Myristate of HS Proteoglycans

V

Acetate (TPA) on the Recycling in Rat Parathyroid Cells

Released ?3-HSPG (% of the total ?S-HSPG)

Trypsin Treatment

Medium

1.7 + O.lb 2.7 f 1.5 1.4 + 0.0 12.3 f 1.2 7.1 AI 0.5

Note. Cells were labeled with [?S]sulfate for 24 h in low [Ca”], and kept at 4°C for 60 min in the presence of the indicated concentrations of divalent cations after washing cells with medium at 4°C. Amounts of 3r’S-labeled macromolecules released during 60 min at 4°C were measured. ’ Medium without CaClz and containing 0.5 mM EGTA. b Average + standard deviation (n = 2).

phorbol ester, appears to be involved in the stimulation of PTH secretion, but that PI turnover activated by the elevation of [Ca2+], is not a second messenger system leading to the activation of protein kinase C. Thus, modulation of secretory processes by increased intracellular Ca2+ in parathyroid cells operates in a different manner from other secretory or exocytotic systems. We have shown previously that the size of the recycling compartment and the rate of shedding of HS proteoglycans are regulated by [Ca2+], (9). The amounts of HS proteoglycans in the recycling compartment decrease and the rate of their release is suppressed as [Ca2’], increases. When the effects of other divalent cations on these phenomena were studied, surprisingly, many divalent cations, such as Mg2+, Ba2+, and Co’+, had an effect opposite to that seen for Ca2+ in terms of the release of HS proteoglycans. While 8 MM Mgzf reduced the recycling pool of HS proteoglycans as does increased Ca2+, the mechanism clearly differs in that Mg2+ stimulated the rapid release of HS proteoglycans, mainly derived from the recycling compartment. Mg2+ appears to stimulate the cleavage of core proteins on or near the cell surface. Core proteins with the same molecular size as those in medium containing 8 mM MgCl, are present within the cells in small quantities, suggesting that the cleavage is cell-mediated and not an extracellular process. Effects of Mg2+ ion are less specific because Ba2’ and Co2+ can substitute for Mg2+. Divalent cations other than Ca2+ have an inhibitory effect on the secretion of PTH (26), suggesting that the recycling of HS proteoglycans, but not their release, could be relevant to the secretory processesof parathyroid cells. Whatever the mechanism is, HS proteoglycans in the divalent cation-sensitive compartment, i.e., the recycling compartment, are transported from the Golgi complex to the recycling compartment in low [Ca2’],, where they re-

2 min

Experiment CaClr 0.05 mM + TPA 0.1 PM CaCiz 4 mM + TPA 0.1 PM

16.9 35.4 8.3 23.6

2 + + f

1 49.1 35.4 14.9 19.7

0.9 0.6 0.9 3.1

Experiment CaCir 4 mM + MgCla

+ + + +

0.5 0.9 1.4 0.6

56.6 + 0.2 37.2 ?I 1.0 ND ND

2

15.6 f 1.6” Experiment

CaCi, 0.05 mM + TPA 1 aM + TPA + MgClz

15 min

ND

11.7 + 0.0”

ND ND

24.3 + 0.8 20.3 + 0.3

3

57.0 + 0.8 62.7 k 0.6

Note. Cells were labeled with [36S]sulfate for 24 h in low or high [Ca*+ Je and treated with the indicated concentrations of TPA and/or 8 mM MgC& for 60 min at 37’C followed by digestion with trypsin for 2 or 15 min. ’ Values represent mean + SE (n = 3 except those marked (I, where n = 2). ND; not determined.

main for no more than 4 h with gradual release of some into the medium and internalization and degradation in lysosomes of the rest (9,17). The recycling compartment is not activated in high [Ca2+], and is depleted in high

TABLE

VI

Effect of Phorbol Myristate Acetate (TPA) on the Release of HS Proteoglycans Cell layer Labeling medium

Chase

Medium

Unbound

(% of the total Low Ca Low Ca Low Ca High Ca High Ca High Ca

Low 8 mM 1 PM High 8 mM 1 PM

Ca MgClz TPA Ca MgCl? TPA

10.8 39.8 65.3 6.9 8.9 41.7

+- 0.4” + 2.0 f 1.6 _+ 0.9 f 0.4 + 2.9

37.7 29.3 12.6 43.7 41.7 23.2

Bl 36S-HSPG)

+ 2.3 5 1.2 3~ 0.1 zk 2.7 f 4.4 f 2.5

43.5 22.9 14.4 41.3 41.4 27.1

zk 2.4 f 1.8 3~ 1.6 f 3.6 + 4.7 iz 3.6

Note. Cultures were labeled with [?S]sulfate for 24 h in low or high [Ca’*]. and incubated for 60 min in the presence or absence of 8 mM MgClz or 10-s M TPA. [Ca”], was constant through each experiment. ?S-HS proteoglycans associated with the cell layer were purified by QSepharose anion-exchange chromatography and fractionated by octylSepharose CL-4B with a gradient of Triton X-100 into three peaks (see Fig. 7). Bl, Peak eluted between 0.02 and 0.17% Triton X-100. o Mean f SE (n = 3).

MgCl,

AND

HEPARAN

SULFATE

PROTEOGLYCANS

concentrations of extracellular Mg2+ only in low [Ca2’],. Therefore, it is postulated that the recycling of HS proteoglycans is directly related to the activated state of parathyroid cells in response to the decrease of [Ca2+J,. We cannot propose, at this stage, the exact mechanism by which the recycling HS proteoglycans is regulated under lowered [Ca2’], conditions, and what the signal is. Proteoglycans are common components in secretory vesicles, such as those in presynaptic regions containing neurotransmitters (27) or in hemopoietic cells (28), and it has been reported that HS proteoglycans associated with presynaptic vesicles might recycle in conjunction with the release of neurotransmitters (29). Taken together, the recycling of HS proteoglycans in parathyroid cells could be related to general secretory processes. If this is the case, the depletion of HS proteoglycans in the recycling compartment by high concentrations of divalent cations other than Ca2+ may reduce the efficiency of the secretory processes by an indirect mechanism.

E. F., and Scarpa,

2. Fitzpatrick, (1988) Proc. 3. Shoback, D. (1984) Proc. 4. Douglas, 5. Ozawa,

Broadus, 115.

RAT

PARATHYROID

A. E. (1989)

Biochem.

A. (1987)

J. Biol.

Chem.

262,

L. A., Chin, H., Nirenberg, M., and Aurbach, Natl. Acad. Sci. USA 85, 2115-2119. M., Thatcher, J., Leombruno, R., and Brown, Nutl. Acad. Sci. USA 81, 3113-3117.

W. W. (1974) S., and Sand,

&o&em. 0. (1986)

Sot. Symp. Physiol.

6. Sakaguchi, K., Santora, A., Zimering, G. D., and Brandi, M. L. (1987) Proc. 3269-3273.

39,

Reu. 66, M., N&l.

5188G. D. E. M.

l-28. 887-952.

Curcio, Acad.

F., Aurbach, Sci. USA 84,

7. Ikeda, K., Weir, E. C., Sakaguchi, K., Burits, W. J., Zimering, M., Mangin, M., Dreyer, B. E., Brandi, M. L., Aurbach, G. D., and

379

CELLS Biophys.

Res. Commun.

162,

108-

8. Yanagishita, M., Brandi, M. L., and Sakaguchi, K. (1989) J. Biol. Chem. 264,15714-15720. 9. Takeuchi, Y., Sakaguchi, K., Yanagishita, M., Aurbach, G. D., and Hascall, V. C. (1990) J. Biol. Chem. 265, 13661-13668. 10. Brown, E. M., Enyedi, P., LeBoff, M. S., Rotberg, J., Preston, and Chen, C. (1987) FEBS I&t. 218,113-118. 11. Yanagishita, M., and Hascall, V. C. (1992) J. Biol. Chem. 267,94519454. 12. Kjellen, 475.

L., and Lindahl,

13. Gallagher, 236,313-325. 14. Fransson, Eds.), pp. 15. Yanagishita, Enzymol. 16. Laemmli,

J. T., Lyon,

U. (1991)

Annu.

M., and Steward,

Reu. Biochem. W. P. (1986)

60,

J.,

443-

&o&em.

J.

L.-A. (1989) in Heparin (Lane, D. A., and Lindahl, U., 115-134, CRC Press, Boca Raton, FL. M., Midura, R. J., and Hascall, V. C. (1987) Meth. 138, 279-289. U. K. (1970) Nature 227, 680-685.

17. Takeuchi, Y., Yanagishita, M., and Hascall, V. C. (1992) Chem. 267, 14677-14684. 18. Brown, E. M. (1991) Physiol. Reu. 71, 371-411. 19. Oetting, M., LeBoff, M. S., Levy, S., Swiston, C., and Brown, E. M. (1987) Endocrinology

REFERENCES 1. Nemeth, 5196.

IN

L., Preston,

J. Biol.

J., Chen,

121, 1571-1576.

20. Shoback, D. M., Membreno, L. A., and McGhee, docrinology 123, 382-389. 21. Kifor, O., and Brown, E. M. (1988) Endocrinology Y. (1986) Science 233, 305-312. 22. Nishizuka,

J. G. (1988)

En-

123,2723-2729.

23. Berridge, M. J., and Irvine, R. F. (1989) Nature 341, 197-205. 24. Nemeth, E. F., Wallace, J., and Scarpa, A. (1986) J. Biol. Chem. 26 1,2668-2674. 25. Kobayashi, N., Russell, J., Lettieri, D., and Sherwood, L. M. (1988) Proc. Natl. Acad. Sci. USA 65,4857-4860. 26. Brown, 27. Kiene,

E. M. (1982) Mineral Electrolyte Metab. 8, 130-150. M. L., and Stadler, H. (1987) EMBO J. 6,2209-2215.

28. Stevens, 29. Stadler,

L. (1986) Ciba Found. Symp. 124,272-280. H., and Kiene, M. L. (1987) EMBO J. 6,2217-2221.